Coinage Metal Hydrides: Reactive Intermediates in Catalysis and Significance to Nanoparticle Synthesis

Home , Bifrustum, Copper hydride, Formic acid, Miroslav Radman

Athanasios Zavras ORCID: 0000-0003-2797-9303

Submitted in total fulfillment of the requirements of the degree of Doctor of Philosophy

May 2019

School of Chemistry The University of Melbourne

Abstract

The coinage metal hydrides of copper, silver and gold have applications in catalysis and nanoparticle synthesis. Coinage metal hydrides are key intermediates in the chemical transformations of a range of substrates including fine chemical syntheses and chemical storage of hydrogen. Ranging from mononuclear coinage metal hydrides to clusters and nanoparticles, a fundamental understanding of their atomic and molecular interactions is invaluable in developing innovative solutions to practical problems. The reactive sites can be identified using a range of spectroscopic methods allowing the “tuning” and/or “reshaping” of the reactive site by ligands to control the reactivity. Mass spectrometry provides a means to identify coinage metal hydrides in solution and further allows isolation of discrete coinage metal hydrides that can be: (i) characterised, for example by spectroscopic methods, (ii) reacted with neutral substrates, or (iii) fragmented to generate reactive intermediates in the gas phase.

The use of borohydride in nanoparticle synthesis is well-known. Chapter 2 describes a mass spectrometry directed synthesis to afford the first isolable silver hydride borohydride cluster,

[Ag3(μ3-H)(μ3-BH4)L3]BF4 (L =bis(diphenylphosphino)methane), structurally characterised by

X-ray crystallography. Gas-phase experiments and DFT calculations reveal ligand (L) loss

+ from [Ag3(H)(BH4)L3] results in the loss of BH3 and a geometry change of the cluster to yield

+ [Ag3(H)(BH4)Ln] (n = 1 or 2). This work reveals links between silver hydride/borohydride and silver hydride nanoclusters adding to our understanding of silver nanoparticle synthesis using borohydride salts.

Chapter 3 examines that the reactivity of CO2 with the binuclear silver hydride cation core,

+ [Ag2H] , can be controlled by design. Reshaping the geometry and reaction environment of

+ [Ag2H] using a range of phosphine ligands (bis(diphenylphosphino)methane, 1,2- bis(diphenylphosphino)benzene and bis(diphenylphosphino)ethane) allows “tuning” of the active site’s reactivity toward formic acid to produce H2. Gas-phase ion-molecule reactions, collision-induced dissociation, infrared and ultraviolet action spectroscopy and computational

i chemistry link structure to reactivity and mechanism. The gas-phase studies were then translated to solution-phase studies using NMR to show that H2 could be produced from solutions comprising well-defined ratios of ligand, AgBF4, NaO2CH and HO2CH at near ambient temperature.

Chapter 4 further developed the concept of altering the reactive site by changing the binuclear

+ metal centres of the [LAg2H] core to compare all six possible combinations of copper silver

+ + + + + + and gold i.e. [LAg2H] , [LCu2H] , [LAu2H] , [LCuAgH] , [LCuAuH] and [LAgAuH] in the gas phase. DFT calculations, gas-phase ion-molecule reactions and gas-phase energy-resolved collision-induced dissociation showed both metal centres play a role in the reaction with formic acid. One metal site functions as an “anchor” for an oxygen of formic acid or formate while the other facilitates the dehydrogenation step resulting in the formation of H2. It was found that the copper homobinuclear species performed best overall.

+ Attempts to isolate the reactive intermediate [LAg2(O2CH)] by using a range of bisphosphine ligands resulted in the isolation of an unusual co-crystal in the case of L = dcpm as described in Chapter 5. Single crystal X-ray diffraction of crystals suitable for crystallographic analysis

2 revealed two discrete tetranuclear silver clusters [(μ2-dcpm)Ag2(μ2-O2CH)(η -NO3)]2·[(μ2- dcpm)2Ag4(μ2-NO3)4]. The solution-phase studies, tracked by NMR, show that H2 could be produced from solutions comprising well-defined ratios of ligand, AgBF4, NaO2CH and HO2CH

+ at 65⁰C. Gas-phase studies indicate that while the tetranuclear cluster [L2Ag4(O2CH)3] undergoes sequential decarboxylation reactions, none of the resultant hydrides react with

+ formic acid. These results highlight important role of the binuclear hydride [LAg2(H)] in the catalytic decarboxylation of formic acid.

- Hydrido cuprate [CuH2] has been explored for its applications in hydrogen storage. Chapter 6

- indicates two chemically induced routes for the liberation of hydrogen when [CuH2] is reacted with various chemical substrates. One path occurs via homocoupling of both hydride ligands giving the substrate-coordinated copper, the other by protonation with acids.

ii The mechanisms for nanocluster to nanoparticle transformation are explored in chapter 7.

Isolable [Ag3(μ3-H)(BH4)(dppm)3]BF4 is proposed as a starting material in the isolation of the

[{Cl@Ag12}@Ag48(dppm)12] nanoparticle. The mechanism of decomposition for [Ag3(μ3-

H)(BH4)(dppm)3]BF4, known to occur via liberation of H2 (Chapter 2), is suggested to reduce silver(I) to silver(0) thus triggering the formation of the nanoparticle.

iii iv Declaration

This is to certify that

i. the thesis comprises only my original work towards the PhD except where indicated in the Preface,

ii. due acknowledgement has been made in the text to all other material used,

iii. the thesis is less than 100 000 words in length, exclusive of tables, maps, bibliographies and appendices.

Athanasios Zavras

May 2019

vi Preface

I certify that I have contributed > 50% of the content in each of the publications included in this thesis, am the “primary author”, have written the “initial draft” and edited the revisions. The following are peer review publications included and the role of the co-authors:

Chapter 2

Zavras, A., Ariafard, A., Khairallah, G. N., White, J. M., Mulder, R. J., Canty, A. J., O'Hair, R.

A. J. Synthesis, structure and gas-phase reactivity of the mixed silver hydride borohydride

Ph nanocluster [Ag3(μ3-H)(μ3-BH4)L 3]BF4 (LPh = bis(diphenylphosphino)methane). Nanoscale,

2015, 7 (43), 18129-18137. DOI: 10.1039/c5nr05690j

Ariafard, A.: Assistance with the theoretical calculations and advisory role

Khairallah, G. N.: Advisory role

White, J. M.: X-ray crystallography of crystalline material and advisory role

Mulder, R. J.: Assistance with the NMR

Canty, A. J.: Advisory role

O'Hair, R. A. J.: Advisory role

Chapter 3

Zavras, A., Khairallah, G. N., Krstic, M., Girod, M., Daly, S., Antoine, R., Maitre, P., Mulder, R.

J., Alexander, S.-A., Bonacic-Koutecky, V., Dugourd, P., O’Hair, R. A. J. Ligand-induced

+ substrate steering and reshaping of [Ag2(H)] scaffold for selective CO2 extrusion from formic acid. Nat. Commun. 2016, 7, 11746. DOI: 10.1038/ncomms11746

Khairallah, G. N.: Assistance with the IRMPD and UVPD and advisory role

Krstic, M.: Assistance with the theoretical calculations

Girod, M.: Assistance with the UVPD

vii Daly, S.: Assistance with the UVPD

Antoine, R.: Assistance with the UVPD

Maitre, P.: Assistance with the IRMPD

Mulder, R. J.: Assistance with the NMR

Alexander, S.-A.: Assistance with the NMR

Bonacic-Koutecky, V.: Advisory role

Dugourd, P.: Advisory role

O’Hair, R. A. J.: Advisory role

Chapter 4

Zavras, A., Krstic, M., Dugourd, P., Bonacic-Koutecky, V., O’Hair, R. A. J. Selectivity Effects in Bimetallic Catalysis: Role of the Metal Sites in the Decomposition of Formic Acid into H2 and

+ CO2 by the Coinage Metal Binuclear Complexes [dppmMM'(H)] . Chem. Cat. Chem., 2017, 9,

1298-1302

Krstic, M.: Assistance with the theoretical calculations

Dugourd, P.: Advisory role

Bonacic-Koutecky, V.: Advisory role

O’Hair, R. A. J.: Advisory role

Chapter 5

2 Zavras, A., White, J. M., O’Hair, R. A. J. An unusual co-crystal [(μ2-dcpm)Ag2(μ2-O2CH)(η -

NO3)]2·[(μ2-dcpm)2Ag4(μ2-NO3)4] and its connection to the selective decarboxylation of formic acid in the gas phase. Dalton Trans., 2016, 45, 19408-19415

White, J. M.: X-ray crystallography of crystalline material and advisory

role

O’Hair, R. A. J.: Advisory role

viii Chapter 6

Zavras, A., Ghari, H., Ariafard, A., Canty, A. J., O’Hair, R. A. J. Gas-Phase Ion-Molecule

– – Reactions of Copper Hydride Anions [CuH2] and [Cu2H3] . Inorg. Chem., 2017, 56 (5), 2387-

2399

Ghari, H.: Assistance with the theoretical calculations

Ariafard, A.: Advisory role

Canty, A. J.: Advisory role

O’Hair, R. A. J.: Advisory role

Chapter 7

Zavras, A. Mravak, A., Buzancic, M., White, J. M., Bonacic-Koutecky, V., O’Hair, R. A. J.

Structure of the Ligated Ag60 Nanoparticle [{Cl@Ag12}@Ag48(dppm)12] (where dppm = bis(diphenylphosphino)methane). CJCP, 2019 Manuscript ID CJCP1812285 Accepted

Mravak, A.: Assistance with the theoretical calculations

Buzancic, M.: Assistance with the theoretical calculations

White, J. M.: X-ray crystallography of crystalline material and advisory

role

Bonacic-Koutecky, V.: Advisory role

O’Hair, R. A. J.: Advisory role

ix x xi

xii xiii

xiv xv

xvi xvii

xviii xix

All co-authors must complete this ·rorm. By signing below co-authors agree to the listed publication being included in the student's thesis and that the student contributed greater than 50% of the content of the publication and is the "primary author" ie. the student was responsible primarily for the planning, execution and preparation of the work for publication.

In cases where all members of a large consortium are listed as authors of a publication, only those that actively collaborated with the student on material contained within the thesis should complete this form. This form is to be used in conjunction with the Declaration for a thesis with publication form.

Students must submit this form, along with the Declaration for thesis with publicationform, when the thesis is submitted to the Thesis Examination System: https:lltes.app.unimefb.edu.au/

Further information on this policy and the requirements is available at:

h e a e l •• ,. I _ Q[8_c/r��f!.8(C :.'-!r.!'!1._ /b_;.f!.cl.'!.·. l!.IPr. p�(ir1J;-py_�l!?�sJ�(l�_e_si�:"Ji'.�:PY�_ �cfi�!9.'!. ... . •' ...., ...... , :" :·

PUBLIC,ATION DJ;.TAILS·(/o he complet d b; the �tudent) · . :· . . A. ',. � e ' . . � Full title Gas-Phase Ion-Molecule Reactions of Copper Hydride Anions [CuH2]· and [Cu2H3]· '.

Authors Zavras, A., Ghari, 1-1., Ariafard, A., Canly, A. J., O'Hair, R. A. J.

, Student's contribution (%) 60 f,., •. , .•...•. I Journal or book name lnorg. Chem.

Volume/page numbers 56/2387-2399

\ Status Published Date published: 10 February2017 ' ......

. ) . - . -

B. CO-AUTHOR'S Df;Cl'.ARATION (to j)e completedb y the c;ollaborator) , . . ' ' , . ' , � - ' ) , , I authorise the inclusion of this publication In the student's thesis and certify that: • the declaration made by the student on the Declaration for a thesis with publication form correctly reflects the extent of the student's contribution to this work; • the student contributed greater than 50% of the coi:itent of the publication and is the "primary author" ie. the student was responsible primarily for the planning, execution and preparation of the work for p�)) l�c��i- <>.n. :. I ...... , ... . , ___ ...... ___ ...... _ . , ...... , .? ·1 Co-author's name Co-author's signatu Dale (ddlmmlyy) . � - ! Ghari, H. GhuVI HusseiY\. � l 13 /03/ .20lj r · · I i' Ariafard, A. · · ;- _\ �/()7$/2olqi : .. . ·-' -.. � . . . Canty, A. J. lj l 2010312019 ; : � (] O'Hair, R. A. J. µ·_·_� l rn10312019

1e University of Melbourne �ICOS Provider Number. 00116K Last Updated 19 October2017

xx xxi

xxii Acknowledgements

“If I have seen further than others, it is by standing upon the shoulders of giants.”

- Sir Isaac Newton

A postgraduate degree, of Doctor of Philosophy, is a remarkable adventure. Contributing new discoveries to the scientific community through peer-reviewed publications and oral or poster presentations at conferences has been very rewarding. I’ve had the pleasure of interacting with wonderful people united by their common interests in science. My peers and colleagues have provided extensive support to ensure the continuous progress of my projects and I have also invested in progressing the work of others in a collaborative, supporting and nurturing environment. These experiences have made my time during my studies a pleasure and I am left with immeasurable fulfillment. Many contributed directly to the work presented in this thesis and indirectly in supporting me during this journey.

The most valuable contributor of my PhD has been my primary supervisor Prof. Richard A. J.

O’Hair. His enthusiasm, knowledge and personal drive are second to none. His work ethic is unrelenting and truly admirable and his passion for science is truly contagious. If you bring

Richard some interesting data or a mass spectrum showing exciting new chemistry, you’ll often observe a genuine reaction of joy and can almost visualise the ideas forming in his mind as he processes the information before proceeding to discuss possible avenues to explore. He inspired and motivated me to continue finding new results while simultaneously helping navigate projects towards their completion. The O’Hair research group at Bio21 has been an intricate part of this journey and I’d like to thank all members past and present.

I’m truly grateful for the contribution of my secondary supervisor, Dr. George N. Khairallah.

George is a beacon of light. I came to the University of Melbourne because I was so impressed with his lecture series on mass spectrometry in the 3rd year of my undergraduate degree at La

Trobe University. I admired that George went above and beyond for all his students and could see in him a genuine and gentle soul. He has always provided kind and caring advice on all matters. He is a committed individual that you can absolutely rely on. xxiii

My advisory committee members comprised Prof. Johnathan M. White and Prof. Paul S.

Donnelly. They provided concise feedback and rigorous discussions on a regular basis and were always there to support me. Johnathan is a crystallographic genius and Paul is an expert in synthetic inorganic synthesis. I always enjoyed walking with Johnathan, first thing in the morning, to analyse a crystal on the X-ray diffractometer. Both the White and Donnelly groups students contributed their expertise and friendship of which I am grateful.

I would like to thank the co-authors that contributed to the work described in this thesis comprising the groups and collaborators of: Prof. Phillippe Dugourd and Prof. Vlasta Bonacic-

Koutecky; Allan J. Canty and Dr. Alireza Ariarfard; Dr. Roger J. Mulder from CSIRO. Staff and students at the University of Melbourne with an honourable mention to Prof. Brendan

Abrahams, Thermo Fisher engineers and many more.

I would like to thank you, the reader. You may be a student about to embark on your own research adventure, a colleague, a family member, a friend, someone with an interest in science or perhaps someone that has stumbled across this thesis by accident. In any case, I hope you enjoy the material presented and that you can find what you are looking for. Whether that be inspiration, validation or simply to reminisce as I will be doing from time to time.

My everlasting gratitude lies with my wife Dr. Stefanie-Ann Zavras (née Alexander), who has co-authored two publications with me. She has been my greatest motivation and my best friend. The past few years have been challenging in many ways and Stefanie has brought light to my life. I’m comforted to know that she always has my best interests at heart. She’s a kind a loving person that I will cherish forever. I look forward to every moment that we share and am excited for the future ahead as we journey together.

xxiv Lastly my eternal gratitude is to my parents, Panagiotis and Maria Zavras, who have been the biggest influence in my life, forging me into the person I am today. The past few years have been extremely challenging, however we have been there for each other to offer support and love to overcome the hardest of times. They have taught me respect, patience, humility, persistence and love among other qualities. It is because of them that I have been fortunate enough to embrace every opportunity in life. This message is for them:

“Αγαπητή μαμά και μπαμπά, σας ευχαριστώ για όλα όσα έχετε κάνει για μένα στη ζωή μου.

Δεν θα μπορούσα να το έκανα χωρίς τη βοήθεια και την αγάπη σας. Σας εκτιμώ και σας αγαπώ

για πάντα. Αυτή η εργασία είναι για εσάς.”

Με όλη μου την αγάπη

Αθανάσιος Ζάβρας

xxv

xxvi Abbreviations

3D three dimensional

Å angstrom

AcOH acetic acid

aq. aqueous

AR analytical reagent

atm. atmosphere

au arbitrary unit/s

br broad (NMR)

˚C degree Celsius

calc. calculated

CDCl3 deuterated chloroform

CD3OD deuterated methanol

CID collision-induced dissociation

CPC cooperative patent classification

δ delta

d doublet

Da Dalton

dd doublet of doublets

DFT density functional theory

DMF N,N-dimethylformamide

DMSO dimethyl sulfoxide

DPCb carborane diphosphine

dt doublet of triplets (NMR)

Ea activation energy

ECP effective core potential

EI electron ionisation

em emission

eq. equivalent/s

xxvii

ERCID energy-resolved collision-induced dissociation

ESI electrospray ionisation

Et ethyl

Et3N triethylamine

Et2O diethyl ether

EtOAc ethyl acetate

EtOH ethanol

ex excitation

FAB fast atom bombardment

FEL free electron laser

FTICR Fourier-transform ion cyclotron resonance

g gram/s

GHG greenhouse gas

HPLC high performance liquid chromatography

hr hour/s

HRMS high resolution mass spectrometry/spectrum

HSQC heteronuclear single quantum coherence

Hz hertz (NMR)

IMR ion-molecule reaction

in vacuo under vacuum

in situ in solution

iPr isopropyl

IR infrared

IRMPD infrared multiple photon dissociation

J coupling constant (NMR)

L litre

Lit. Literature

LIT linear ion trap

LTQ linear triple-quadrupole

µ micro

xxviii m milli, mass or multiplet (NMR)

M concentration in moles per litre

M+● radical cation

max maxima/maximum

Me methyl

min minute/s

MHz megahertz (NMR)

mm millimetre

mol mole/s

MRI magnetic resonance imaging

MS mass spectrometry

m/z mass to charge ratio

n nano

NHC N-heterocyclic carbene

NMR nuclear magnetic resonance

ORTEP Oak Ridge thermal ellipsoid plot

p para

PD photodissociation

PES potential energy surface

Ph phenyl

ppm parts per million (NMR)

q quartet (NMR)

RECP relativistic effective core potential

rt room temperature

s singlet (NMR)

sat. saturated

t tertiary

TDDFT time-dependent density functional theory

THF tetrahydrofuran

TOF time of flight

xxix

TS transition state

UV ultraviolet

VT variable temperature (NMR)

xxx List of Publications

Publications 1 – 18 (PhD candidature)

1. Zavras, A. Mravak, A., Buzancic, M., White, J. M., Bonacic-Koutecky, V., O’Hair, R. A. J.

Structure of the Ligated Ag60 Nanoparticle [{Cl@Ag12}@Ag48(dppm)12] (where dppm =

bis(diphenylphosphino)methane). CJCP, 2019, 32 (2), 182-186.

2. Daly, S., Choi, C. M., Zavras, A., Kristic, M., Chirot, F., Connell, T. U., Williams, S. J.,

Donnelly, P. S., Antoine, R., Giuliani, A., Bonacic, V.-K., Dugourd, P., O’Hair, R. A. J. Gas-

Phase Structural and Optical Properties of Homo- and Heterobimetallic Rhombic

+ Dodecahedral Nanoclusters [Ag14-nCun(C≡CtBu)12X] (X = Cl and Br): Ion Mobility, VUV

and UV Spectroscopy, and DFT Calculations. J. Phys. Chem. C., 2017, 121 (20), 10719-

10727.

3. Zavras, A., Krstic, M., Dugourd, P., Bonacic-Koutecky, V., O’Hair, R. A. J. Selectivity

Effects in Bimetallic Catalysis: Role of the Metal Sites in the Decomposition of Formic Acid

+ into H2 and CO2 by the Coinage Metal Binuclear Complexes [dppmMM'(H)] . Chem. Cat.

Chem., 2017, 9, 1298-1302.

4. Zavras, A., Ghari, H., Ariafard, A., Canty, A. J., O’Hair, R. A. J. Gas-Phase Ion-Molecule

– – Reactions of Copper Hydride Anions [CuH2] and [Cu2H3] . Inorg. Chem., 2017, 56 (5),

2387-2399.

5. Spillane, S., Sharma, R., Zavras, A., Mulder, R., Ohlin, C. A., Goerigk, L., O’Hair, R. A. J.,

Ritchie, C. Non-Aqueous Microwave-Assisted Syntheses of Deca- and Hexa-

Molybdovanadates. Angew. Chem. Int. Ed., 2017, 56, 8568-8572.

6. Moore, P. W., Hooker, J. P., Zavras, A., Khairallah, G. N., Krenske, E. H., Bernhardt, P.

V., Quach, G., Moore, E. G., O’Hair, R. A. J., Williams, C. M. Hydroxyl Radical via

Collision-Induced Dissociation of Trimethylammonium Benzyl Alcohols. Aust. J. Chem.,

2017, 70, 297-406.

2 7. Zavras, A., White, J. M., O’Hair, R. A. J. An unusual co-crystal [(μ2-dcpm)Ag2(μ2-O2CH)(η -

NO3)]2·[(μ2-dcpm)2Ag4(μ2-NO3)4] and its connection to the selective decarboxylation of

formic acid in the gas phase. Dalton Trans., 2016, 45, 19408-19415.

xxxi

8. Krstic, M., Zavras, A., Khairallah, G. N., Dugourd, P., Bonacic-Koutecky, V., O’Hair, R. A.

J. ESI/MS Investigation of Routes to the Formation of Silver Hydride Nanocluster Dications

2+ 2+ [AgxHx-2Ly] and Gas-phase Unimolecular Chemistry of [Ag10H8L6] . IJMS, 2016, 413, 97-

105.

9. Zavras, A., Khairallah, G. N., Krstic, M., Girod, M., Daly, S., Antoine, R., Maitre, P., Mulder,

R. J., Alexander, S.-A., Bonacic-Koutecky, V., Dugourd, P., O’Hair, R. A. J. Ligand-

+ induced substrate steering and reshaping of [Ag2(H)] scaffold for selective CO2 extrusion

from formic acid. Nat. Commun. 2016, 7, 11746. DOI: 10.1038/ncomms11746.

10. Canale, V., Robinson, R., Zavras, A., Khairallah, G. N., d’Alessandro, N., Yates, B. F.,

- O’Hair, R. A. J. Two Spin-State Reactivity in the Activation and Cleavage of CO2 by [ReO2]

. J. Phys. Chem. Lett. 2016, 7 (10), 1934-1937 DOI: 10.1021/acs.jpclett.6b00754.

11. Barzegar Amiri Olia, M., Zavras, A., Schiesser, C. H., Alexander, S.-A. Blue 'turn-on'

fluorescent probes for the direct detection of free radicals and nitric oxide in Pseudomonas

aeruginosa biofilms. Org. and Biomol. Chem., 2016, 14, 2272-2281. DOI:

10.1039/c5ob02441b.

12. Vikse, K.L., Zavras, A., Thomas, T.H., Ariafard, A., Khairallah, G. N., Canty, A. J., Yates,

B. F., O'Hair, R.A.J. Prying open a Reactive Site for Allylic Arylation by Phosphine-Ligated

Geminally Diaurated Aryl Complexes. Organometallics, 2015, 34(13), 3255-3263. DOI:

10.1021/acs.organomet.5b00287.

13. Canale, V., Zavras, A., Khairallah, G. N., D'Alessandro, N., O'Hair, R. A. J. Gas-phase

- reactions of the rhenium oxide anions, [ReOx] (x = 2-4) with the neutral organic substrates

methane, ethene, methanol and acetic acid. EJMS, 2015, 21 (3), 557-567. DOI:

10.1255/ejms.1332.

14. Feketeová, L., Postler, J., Zavras, A., Scheier, P., Denifl, S., O'Hair, R. A. J.

Decomposition of nitroimidazole ions: Experiment and theory. Phys. Chem. Chem. Phys.

2015, 17 (19), 12598-12607. DOI: 10.1039/c5cp01014d.

15. Daly, S., Krstić, M., Giuliani, A., Antoine, R., Nahon, L., Zavras, A., Khairallah, G. N.,

Bonačić-Koutecký, V., Dugourd, P., O'Hair, R. A. J. Gas-phase VUV photoionisation and

2+ photofragmentation of the silver deuteride nanocluster [Ag10D8L6] (L =

xxxii bis(diphenylphosphino)methane). A joint experimental and theoretical study. Phys. Chem.

Chem. Phys., 2015, 17 (39), 25772-25777. DOI: 10.1039/c5cp01160d.

16. Clark, A. J., Zavras, A., Khairallah, G. N., O'Hair, R. A. J. Bis(dimethylphosphino)methane-

ligated silver chloride, cyanide and hydride cluster cations: Synthesis and gas-phase

unimolecular reactivity. IJMS, 2015, 378, art. no. 15239, 86-94. DOI:

10.1016/j.ijms.2014.07.015.

17. Zavras, A., Ariafard, A., Khairallah, G. N., White, J. M., Mulder, R. J., Canty, A. J., O'Hair,

R. A. J. Synthesis, structure and gas-phase reactivity of the mixed silver hydride

Ph borohydride nanocluster [Ag3(μ3-H)(μ3-BH4)L 3]BF4 (LPh =

bis(diphenylphosphino)methane). Nanoscale, 2015, 7 (43), 18129-18137. DOI:

10.1039/c5nr05690j.

18. Vonci, M., Akhlaghi Bagherjeri, F., Hall, P. D., Gable, R. W., Zavras, A., O'Hair, R. A. J.,

Liu, Y., Zhang, J., Field, M. R., Taylor, M. B., Du Plessis, J., Bryant, G., Riley, M., Sorace,

L., Aparicio, P. A., Lõpez, X., Poblet, J. M., Ritchie, C., Boskovic, C. Modular Molecules:

Site-Selective Metal Substitution, Photoreduction, and Chirality in Polyoxometalate

Hybrids. Chemistry - A European Journal, 2014, 20 (43), 14102-14111. DOI:

10.1002/chem.201403222.

Publications 19 – 23 (MPhil)

19. Zavras, A., Khairallah, G. N., Connell, T. U., White, J. M., Edwards, A. J., Mulder, R. J.,

Donnelly, P. S., OHair, R. A. J. Synthesis, structural characterization, and gas-phase

unimolecular reactivity of the silver hydride nanocluster [Ag3((PPh2)2CH2)3(μ3-H)](BF4)2.

Inorg. Chem., 2014, 53 (14), 7429-7437. DOI: 10.1021/ic5007499.

20. Zavras, A., Khairallah, G. N., O'Hair, R. A. J. Gas phase formation, structure and reactivity

of gold cluster ions. Structure and Bonding, 2014, 162, 139-230. DOI:

10.1007/430_2014_140.

21. Girod, M., Krstić, M., Antoine, R., MacAleese, L., Lemoine, J., Zavras, A., Khairallah, G.

N., Bonacic-Koutecky, V., Dugourd, P., O'Hair, R. A. J. Formation and characterisation of

+ the silver hydride nanocluster cation [Ag3H2((Ph2P)2CH2)] and its release of hydrogen.

xxxiii

Chemistry - A European Journal, 2014, 20 (50), 16626-16633. DOI:

10.1002/chem.201404110.

22. Zavras, A., Khairallah, G. N., Connell, T. U., White, J. M., Edwards, A. J., Donnelly, P. S.,

O'Hair, R. A.J. Synthesis, structure and gas-phase reactivity of a silver hydride complex

[Ag3{(PPh2)2CH2}3(μ3-H)(μ3-Cl)]BF4. Angew. Chem. Int. Ed., 2013, 52 (32), 8391-8394.

DOI: 10.1002/anie.201302436.

23. Zavras, A., Khairallah, G. N., O'Hair, R. A. J. Bis(diphenylphosphino)methane ligated gold

cluster cations: Synthesis and gas-phase unimolecular reactivity. IJMS, 2013, 354-355,

242-248. DOI: 10.1016/j.ijms.2013.05.034.

Publication 24 (BSc(hons))

24. Zavras, A., Fry, J. A., Beavers, C. M., Talbo, G. H., Richards, A. F. 2-

Pyridylmethylphosphonic acid: A flexible, multi-dentate ligand for metal phosphonates.

Cryst. Eng. Comm., 2011, 13 (10), 3551-3561. DOI: 10.1039/c0ce00969e.

xxxiv Table of Contents

Abstract ...... i Declaration ...... v Preface ...... vii Acknowledgements ...... xxiii Abbreviations ...... xxvii List of Publications ...... xxxi Table of Contents ...... xxxv List of Figures ...... xxxix List of Schemes ...... xlvii List of Tables...... xlix 1 Introduction ...... 1

1.1 Coinage Metal Hydrides ...... 3 1.1.1 History ...... 3 1.1.2 Synthesis ...... 5 1.1.3 Structures ...... 10 1.1.3.1 Copper hydrides ...... 10 1.1.3.1.1 Mononuclear complexes ...... 10 1.1.3.1.2 Binuclear complexes ...... 11 1.1.3.1.3 Trinuclear complexes ...... 12 1.1.3.1.4 Tetranuclear complexes ...... 15 1.1.3.1.5 Pentanuclear complexes ...... 15 1.1.3.1.6 Hexanuclear complexes ...... 15 1.1.3.1.7 Heptanuclear complexes ...... 17 1.1.3.1.8 Octanuclear complexes ...... 18 1.1.3.1.9 Larger nuclearity clusters ...... 20 1.1.3.2 Silver hydrides...... 25 1.1.3.2.1 Mononuclear complexes ...... 25 1.1.3.2.2 Binuclear complexes ...... 26 1.1.3.2.3 Trinuclear complexes ...... 27 1.1.3.2.4 Tetranuclear complexes ...... 29 1.1.3.2.5 Pentanuclear complexes ...... 29 1.1.3.2.6 Hexanuclear complexes ...... 29 1.1.3.2.7 Heptanuclear complexes ...... 30

xxxv

1.1.3.2.8 Octanuclear complexes ...... 32 1.1.3.2.9 Large nuclearity clusters ...... 33 1.1.3.3 Gold hydrides ...... 33 1.1.3.4 Heterometallic coinage metal hydrides ...... 37 1.1.4 Bonding and Reaction Environment ...... 38 1.1.4.1 Ligands ...... 38 1.1.4.1.1 Hydride ...... 38 1.1.4.1.2 Phosphines ...... 39 1.1.4.2 Entatic Effect ...... 41 1.1.5 Reactivity ...... 44 1.1.5.1 Copper hydrides ...... 45 1.1.5.2 Silver hydrides ...... 46 1.1.5.3 Gold hydrides ...... 47 1.1.6 Hydrogen storage ...... 49

1.1.7 CO2 Mitigation ...... 51 1.1.8 Links between Nanoclusters and Nanostructured Materials ...... 52 1.1.8.1 Morphology of clusters and nanoparticles ...... 53 1.1.8.2 Clusters as “building blocks” or precursors to nanoparticles...... 53 1.1.8.3 Transformation of nanoparticles into nanoclusters ...... 54 1.2 Characterisation Methods Used in this Thesis ...... 55 1.2.1 NMR Spectroscopy ...... 55 1.2.2 Crystallography ...... 56 1.2.2.1 X-ray diffraction ...... 56 1.2.2.2 Neutron diffraction ...... 57 1.2.3 Density Functional Theory ...... 57

1.3 Mass spectrometry ...... 58 1.3.1 Ionisation Methods ...... 59 1.3.1.1 Electrospray ionisation (ESI) ...... 61 1.3.2 Mass Analysers ...... 63 1.3.2.1 Linear Ion Trap ...... 65 1.3.2.2 Fourier-Transform Ion Cyclotron Resonance ...... 66 1.3.2.3 3D Quadrupole Ion Trap ...... 67 1.3.3 Gas-phase Chemistry ...... 68 1.3.3.1 Collision-Induced Dissociation ...... 69 1.3.3.2 Ion-Molecule Reactions ...... 70 1.4 Thesis Scope ...... 71 1.5 References ...... 72 xxxvi 2 Synthesis, structure and gas-phase reactivity of the mixed silver hydride Ph borohydride nanocluster [Ag3(μ3-H)(μ3-BH4)L 3]BF4 (LPh = bis(diphenylphosphino)methane) ...... 93 + 3 Ligand-induced substrate steering and reshaping of [Ag2(H)] scaffold for

selective CO2 extrusion from formic acid ...... 107 4 Selectivity Effects in Bimetallic Catalysis: Role of the Metal Sites in the

Decomposition of Formic Acid into H2 and CO2 by the Coinage Metal Binuclear Complexes [dppmMM’(H)]+ ...... 117 2 5 An unusual co-crystal [(μ2-dcpm)Ag2(μ2-O2CH)(η -NO3)]2·[(μ2-

dcpm)2Ag4(μ2-NO3)4] and its connection to the selective decarboxylation of formic acid in the gas phase ...... 125 − 6 Gas-Phase Ion−Molecule Reactions of Copper Hydride Anions [CuH2] and − [Cu2H3] ...... 135

7 Structure of the Ligated Ag60 Nanoparticle [{Cl@Ag12}@Ag48(dppm)12] (where dppm = bis(diphenylphosphino)methane) ...... 151 8 Conclusions ...... 159 9 Appendices ...... 163

9.1 Appendix A - Supplementary material for Chapter 2 ...... 165 9.2 Appendix B - Supplementary material for Chapter 3 ...... 213 9.3 Appendix C - Supplementary material for Chapter 4 ...... 247 9.4 Appendix D - Supplementary material for Chapter 5 ...... 285 9.5 Appendix E - Supplementary material for Chapter 6 ...... 321 9.6 Appendix F - Supplementary material for Chapter 7 ...... 367

xxxvii xxxviii List of Figures

Figure 1.1 Crystal structure of copper hydride. Grey spheres represent copper 3

positions. Black spheres represent hydrogen (or deuterium)

positions for the ordered model (Wurtzite structure‐type). On the

right it is shown that face sharing tetrahedra cannot be occupied

simultaneously because the H–H (or D–D) distance will be too

short. Bond lengths are given in pm.

2 Figure 1.2 The ORTEP view of (a) [(μ-H)2Cu2(η -tripod)2]; Cu-P2 = 2.229 (2), 11

Cu-P3 = 2.229 (2) A, Cu-H = 1.66 (8), Cu-H' = 1.81 (8); P2-Cu-P3

= 101.3 (1), P2-Cu-H = 113 (3), P3-Cu-H = 116 (3), H-Cu-H' = 94

(3), Cu-H-Cu' = 86 (3). Figure reproduced from reference.81 (b)

- [{(IDipp)Cu}2(μ-H)]BF4 wherein the BF4 counterion is omitted for

clarity. Cu1–Cu2 2.5331(15), C1–Cu1 = 1.873(9), Cu1–H1 =

1.45(2), H1–Cu2 = 1.45(2), Cu2–C28 = 1.908(8); C1-Cu1-H1

167.4(16), Cu1-H1-Cu2 122(3), H1-Cu2-C28 174.4(16), C1-Cu1-

Cu2 163.2(3), Cu1-Cu2-C28 156.1(3). Distances in (Å) and angles

in (°).

Figure 1.3 (a) Ball-and-stick diagram of [Cu3H(dppm)3(OAc)2]. All hydrogen 13

atoms (except the hydride ligand) and solvent molecules are

omitted for clarity. Colour legend: Cu = green; H = magenta; P =

yellow-orange; O = red; C = gray wireframe. (b) ORTEP-3 derived

+ structure of cluster cation of [Cu3(μ3-H)(μ3-BH4)(dppa)3] (BF4) with

30% ellipsoid possibilities. Phenyl hydrogen atoms, amine

− hydrogen atoms, and anion (BF4 ) are omitted for clarity. Distances

in (Å): Cu1−Cu3 = 2.6164(5), Cu1−Cu2 = 2.6706(5), Cu2−Cu3 =

2.6785(5), Cu1−B1 = 2.584(2), Cu2−B1 = 2.598(2), Cu3−B1 =

2.549(2), Cu1−P5 = 2.2624(6). The noncoordinating BF4 counter-

anion and disordered CH2Cl2 solvent molecules are not shown.

xxxix

Figure 1.4 (a) Perspective view of the cation of [Cu3H(dppm)3(O2CH)]PF6 14

89 highlighting the C3 symmetry of the formate due to disorder, and

(b) The core of [dppm3Cu3(µ3-H)( µ2,µ1-S2CH)](BF4) highlighting

the dithioformate bonding to the Cu3 core.

Figure 1.5 (a) Molecular structure of the H6Cu6P6 core of the H6Cu6[P(p- 16

tolyl)3]6 cluster. Thermal ellipsoids are at the 50% probability

level.11 Average pertinent distances (showed in Å): Cu-Cu(long) =

2.68(4), Cu-Cu(short) = 2.52(3), Cu-H = 1.76(3). Primed atoms are

related to unprimed ones by a crystallographic inversion. The

drawing shows one of the two independent clusters in the unit cell.

(b) Two views of the [HCuP]6 core of Stryker’s reagent. The

hydrides appear to be triply bridging and edge bridging, left and

right view, respectively.

Figure 1.6 (a) Molecular structure of the [Cu4(μ4-H)(μ3-Cu)3{S2C(aza-15- 18

crown-5)}6]; H atoms omitted for clarity except the interstitial

hydride. (b) Cu7 framework displays an elongated tricapped

tetrahedron from X-ray. (c) Cu7H core drawing from neutron

diffraction data.

i Figure 1.7 X-ray diffraction of [Cu8(H){S2P(O Pr)2}6]PF6 (a) Discrete structure. 18

(b) “Cu8(μ-H)” core.

2 31 1 Figure 1.8 (a) VT H NMR of [Cu8D6dppm5](PF6)2, and (b) VT P{ H} NMR of 19

[Cu8H6dppm5](PF6)2.

Figure 1.9 (a) The side view of the racemic Cu18 kernels (top cupola: green; 21

bottom cupola: orange), divided by mirror symmetry (s); (b) The top

view of two racemic Cu18 kernels and the superimposed bottom

cupolas (orange). (c) The side view of the racemic Cu16 kernels

(top cupola: green; bottom cupola: orange), divided by mirror

symmetry (s); (d) The top view of two racemic Cu16 kernels and the

superimposed bottom cupolas (orange). The drawings of bonds

between top and bottom cupolae are not shown.

xl i Figure 1.10 Structure of [Cu20(H)11{S2P(O Pr)2}9]. (a) X-ray structure and (b) 22

neutron diffraction structure. (Color code: Cu, blue; Cu−Cu edges,

cyan; S, yellow; P, gray; H, pink.).

Figure 1.11 X-ray (a−c) and neutron (d) structure of [Cu32(H)20{S2P(OiPr)2}12]. 22

(a) Hexacapped rhombohedron (pink) of 14 Cu atoms sandwiched

between two triangular cupola (cyan) of (2 × 9) Cu atoms, (b)

Elongated triangular gyrobicupola of Cu18 motif, (c) Twelve ligands

in four spherical rows (w, x, y, z) around metal core, (d) Position of

20 hydrides. (Colour code: Cu, cyan and magenta; S, yellow; P,

green; H, pink.).

Figure 1.12 (a) The cubic [Cu28(H)15{S2CNPr2}12] cation. (b) The 24

rhombicuboctahedral framework of Cu24. (c) Each of the twelve

square faces of the Cu24 core capped by a dtc ligand. (d) The Cu24

core is enclosed by a truncated octahedron of 24 sulfur ligands. (e)

Three-coordinate hydrides (µ3-H) on eight triangular faces of the

Cu24 core. (f) Six truncating hydrides inside six square faces of the

Cu24 core form a bridging octahedron motif; the core is

concentrically arranged around the interstitial hydride (µ4-H). (g)

The core assembly:

H@Cu4(tetrahedron)@H6(octahedron)@Cu24(rhombicuboctahedr

on)@H8(cube) @S24(truncated octahedron) reminiscent of a

Chinese puzzle ball (inset). Carbon gray, copper cyan, hydrogen

red, nitrogen pink, sulfur yellow.

Figure 1.13 (a) Molecular structure of (IPr**)AgH in the solid state. Ellipsoids 26

are set at 50% probability; solvent molecules and hydrogen atoms

other than the hydride are omitted for clarity. C1–Ag1 = 2.108(6) Å.

(b) Molecular structure of (IPr**)AgH[B(C6F5)3] in the solid state.

Ellipsoids are set at 50% probability; solvent molecules and

hydrogen atoms other than the hydride have been omitted for

clarity. Selected distances and angles: C1–Ag1 = 2.095(2), Ag1–

xli

H1 = 1.81(3), H1–B1 = 1.21(3); C1-Ag1-H1 = 173.5(9), Ag1-H1-B1

= 159.7(13).

+ Figure 1.14 Solid-state structure showing the cation {[(SIDipp)Ag]2(µ-H)} , 50% 27

probability ellipsoids. Anion and co-crystallised solvent omitted for

clarity. Selected bond lengths (Å) and angles (⁰): Ag(1)–C(1) =

2.108(3); Ag(2)–C(28) = 2.104(3); Ag(1)–Ag(2) = 2.8087(4); Ag(1)–

H(1) = 1.69(4); Ag(2)–H(1) = 1.71(3); C(1)–Ag(1)–Ag(2) =

158.88(8); Ag(1)–Ag(2)–C(28) = 155.46(8); C(1)–Ag(1)–H(1) =

166(1); C(28)–Ag(2)–H(1) = 170(1).

+ Figure 1.15 ORTEP-3 representations of (a) [Ag3((PPh2)2CH2)3(µ3-H)(µ3-Cl)] ; 28

+ (b) the distorted trigonal bipyramidal “Ag3(µ3-H)(µ3-Cl)” core, with

the phenyl groups are omitted for clarity. Displacement ellipsoids

set at the 20% probability level for both (a) and (b); (c)

2+ [Ag3((PPh2)2CH2)3(µ3-H)] with hydrogen atoms and solvent

omitted for clarity; (d) the silver core with the ligand (phenyl rings

and hydrogen atoms of the ligand omitted for clarity). Displacement

ellipsoids set at the 50% probability level for both (c) and (d). The

location of the deuterium is unambiguously proven by neutron

diffraction.

Figure 1.16 Ball-and-stick diagram of [Ag6H4(dppm)4(OAc)2]. All hydrogen 29

atoms and solvent molecules are omitted for clarity. Colour legend:

Purple = silver; Orange = phosphorus; Red = oxygen; The gray

wireframe represents the carbon backbone.

i Figure 1.17 (a) Ag7H core of [Ag7(μ4-H){Se2P(O Pr)2}6] disordered in two 31

orientations (50% each). (b) Molecular structure of [Ag7(μ4-

i H){Se2P(O Pr)2}6] (30% thermal ellipsoid) with isopropyl groups

omitted for clarity. (c) Molecular structure of [Ag7(μ4-

H){S2P(OEt)2}6] (30% thermal ellipsoid) with ethyl groups omitted

for clarity (d) The Ag7H core of [Ag7(μ4-H){S2P(OEt)2}6] in

disordered in two orientations (top 75%, bottom 25%).

xlii i + Figure 1.18 (a) “Ag8H” core of [Ag4(μ4-H)(μ3-Ag)4{Se2P(O Pr)2}6] . (b) Molecular 32

i + structure of [Ag4(μ4-H)(μ3-Ag)4{Se2P(O Pr)2}6] (50% thermal

ellipsoid). Isopropyl groups omitted for clarity.

Figure 1.19 A perspective view of the Ag11H core of 1H. Selected bond lengths 33

(Å) and angles (⁰): Ag(1)-Ag(3) = 2.8262(8); Ag(2)-Ag(3) =

2.8187(8); Ag(3)-Ag(3a) = 2.9623(8); Ag(1)-Ag(4) = 2.8775(5);

Ag(2)-Ag(5) = 2.9309(6); Ag(3)-Ag(4) = 3.1094(8); Ag(3)-Ag(5) =

3.1249(8); H(1)-Ag(2) = 1.9(2); H(1)-Ag(3) = 1.75(4).

Figure 1.20 (a) Solid state structure of [(IPr)AuH]. (b) Solid-state structure of 34

+ f − {[(6Dipp)Au]2(μ-H)} [BAr 4] ellipsoids set at 50% probability. For

clarity, anion and H atoms except hydride are omitted, and iPr

groups are shown in wireframe view. The hydride was located in

the difference Fourier map and refined isotropically. Selected bond

lengths and angles: Au(1)−Au(2) = 2.7571(3) Å; Au(1)−C(33) =

2.040(5) Å; Au(2)−C(3) = 2.049(5) Å; Au(1)−Au(2)−C(3) =

165.7(1)°; Au(2)−Au(1)−C(33) = 164.6°.

Figure 1.21 Crystal structure of [(C^N^C)AuH] with selected bond distances (Å) 35

and angles (°) for: Au1–H1 = 1.64(5), Au1–N1 = 2.019(3), Au1–C9

= 2.060(3), Au1–C18 = 2.058(4); N1-Au1-C9 = 81.06(13), N1-Au1-

C18 = 80.79(13), C9-Au1-H1 = 98.2(19), C18-Au1-H1 = 100.0(19),

N1-Au1-H1 = 179.2(19), C9-Au1-C18 = 161.85(14).

3+ Figure 1.22 DFT optimised structure and energy diagrams of: (a) [Au9(PMe3)8] 36

+ and (b) [Au9H(PMe3)8] . Yellow, blue, and white balls represent Au,

P, and H atoms, respectively. Methyl groups are depicted as sticks.

Kohn−Sham orbitals are depicted with isodensity values at 0.025e

levels.

Figure 1.23 Solid-state structure of a cationic Au(III)-hydride. Counter-anion 37

and hydrogen atoms are omitted for clarity.

Figure 1.24 X-ray structures heterobimetallic coinage metal clusters 37

highlighting key bond lengths and angles: (a) [{{ArNC(CR3)}2CH}-

xliii

Cu-H-Au(IPr)]; (Ar = Mes, R = CF3): Au–Cu = 2.6376(5), Au–C =

2.026(3); C-Au-Cu = 161.86(10); (b) [{{ArNC(CR3)}2CH}-Cu-H-

Au(IPr)]; (Ar = C6F5, R = CH3): Au–Cu = 2.5910(6), Au–C =

2.025(4); C-Au-Cu = 153.00(11); and (c)

[((ArNC(CR3)(O(CR3))CH)(PPh3)-Cu-H-Au(IPr)]; (Ar = C6F5, R =

CH3): Au–Cu = 2.7525(7), Au-C = 2.043(8); C-Au-Cu = 153.4(4);

Cu–P = 2.2401(11). Selected bond lengths [⁰] and angles [Å].

Figure 1.25 Hydride binding modes for a selection of metal-hydrides. 39

Figure 1.26 A graphic representation of the cone angle to describe the steric 40

effects of a phosphine ligand coordinated to a metal centre.

Figure 1.27 A generic representation of a single-step metal-catalysed reaction 42

whereby a metal-ligand complex (L) bound to a reactant (R) is

transformed into a product (P). The plot shows the reaction

coordinate (x-axis) vs the energy (y-axis) for the reaction without

entatic effect (red dotted line) compared to the reaction with an

entatic effect. Ea = activation energy, Ea(e) = activation energy with

entatic effect, L = metal-ligand complex, Le = metal-ligand complex

with entatic effect, R = reactant, P = product; Red dotted line =

energy profile, Black solid line = energy profile.

Figure 1.28 Representative energy profile for the reaction of iodobenzene with: 43

+ + (a) [Au(PR3)2] , and (b) [(DPCb)Au] .

Cluster to nanoparticle transformation highlighting the similarities in 53

Figure 1.29 their morphology between (a) the rhombohedral core of 14 copper

i atoms (in purple) for [Cu32(H)20{S2P(O Pr)2}12], and (b) the rhombus

shaped nanoparticles yielded from the reaction of

i [Cu32(H)20{S2P(O Pr)2}12] with excess borohydride.

Figure 1.30 The basic components of a mass spectrometer. 59

Figure 1.31 ESI in the positive ion mode. 62

Figure 1.32 Linear ion trap quadrupole rod assembly. 65

xliv Figure 1.33 A Penning trap cross-section showing: (a) ion excitation by the 66

excitation electrodes (green), and (b) detection of ions by detection

electrodes (red).

Figure 1.34 Cross-sectional view of the Thermo Finnigan quadrupole ion trap 67

mass analyser.

Figure 1.35 A diagram showing the ion-molecule reaction setup used in this 70

thesis.

xlv

xlvi List of Schemes

Scheme 1.1 A generalised summary for the stepwise synthesis of coinage metal 6

hydride complexes. A = Charged Ligands; L = Neutral Ligands; M

= Cu, Ag or Au; Y = Counter Ions.

Scheme 1.2 The structures of some common ligands used to synthesis coinage 7

metal hydride clusters.

Scheme 1.3 Diverse methods for the preparation of [HCu(PR3)]6. Scheme 8

reproduced from reference.

Scheme 1.4 Preparative hydrogenolysis of ligand-protected silver alkoxide and 9

silver fluoride complexes. t-Pent-O = 2,2-dimethylpropoxide.

Scheme reproduced from reference.

Scheme 1.5 Synthesis of a stable gold hydride. 9

Scheme 1.6 A selection of coordination modes for: monodentate tertiary 40

phosphines (a) and (b), and; bisphosphine ligand (c)-(h).

Scheme 1.7 1,4-hydrosilylation reaction of 2-cyclohexen-1-one by 44

[Cu3(H)L3(OAc)2], [Ag6(H)4L4(OAc)2] and [CuH(PPh3)]6.

Scheme 1.8 elected reactions of copper hydride mediated organic 45

transformations for: (a) reduction of ketones (b) enantisoelective

reduction of ketones (c) reduction of alkenyl ketones (d) reduction

of enones (e) reduction of akynes (f) reductive aldol

addition/lactonisation (g) anti-Markovnikov hydroamination of

terminal alkenes; and (h) enantioselective reductive aldol reaction.

Scheme 1.9 Silver hydride mediated organic transformations for: (a) 46

hydrogenation of aldehydes in water (b) hydrogenation of alkynes.

Scheme 1.10 Catalytic production of propene by reaction of methane in the 47

- + presence of ethene over silver(I)-exchanged zeolite (ZO Agn ).

Scheme 1.11 Gold hydride mediated organic transformations for: (a) 48

enantioselective hydrogenation (b) alkene hydrogenation (c)

hydrogenation of alkynes.

xlvii

Scheme 1.12 Proposed Mechanism for gold(I)-catalysed dehydrogenative 48

silylation.

Scheme 1.13 The decomposition pathways of formic acid. 49

Scheme 1.14 Proposed catalytic cycle for the dehydrogenation of formic acid. 50

Structures and DH/DG energies (kcalmol@1) of the catalytic cycle

computed at the M06/6-31G**(SDD) level of theory, including

solvent effects (IEFPCM, solvent=formic acid).

Scheme 1.15 Reactions of selected coinage metal hydrides with carbon dioxide. 51

Scheme 1.16 Formation of silver nanoparticles from a silver hydride nanocluster. 54

Scheme 1.17 Overview of the gas-phase chemistry of coinage metal hydrides: (a) 69

condensed phase coinage metal hydrides identification, and (b)

gas-phase chemistry after ionisation.

xlviii

List of Tables

Table 1.1 Various ionisation methods. 60

Table 1.2 A comparison of selected performance characteristics of various 64

mass analysers.361

xlix

1 Introduction

“The puzzle of finding out how things work is the main drive, the drive towards fundamental understanding… To discover new, totally unexpected processes, compounds, and effects is the basis of fundamental research… A favourite example is the invention of the laser: Charles

Townes was interested in developing a high frequency amplifier and the laser was the result.

He didn’t foresee the use of lasers in eye surgery. You could give all the money in the world to eye surgeons in 1950 but they would not have developed a laser – they couldn’t do it. So, there’s a very nice example of why fundamental science should be supported and go hand-in- hand with applied research.”1

- Sir Harold Kroto 1939 - 2016

A myriad of hypothesis-driven research has led to innovative advances in modern medicine, agriculture, materials science and the like. Current and foreseeable challenges such as climate change, disease and the exhaustion of commodities demand an understanding of the fundamental scientific principles that describe or predict these phenomena. Fundamental research, often driven by curiosity, is targeted at improving our current understanding of scientific theories and the prediction of natural phenomena. Other than providing intellectually stimulating research, it provides a foundation of principles for the applied sciences that ultimately drive innovative solutions to real-world problems.

In 1897, J. J. Thomson, an eminent figure in the development of mass spectrometry (MS), performed a series of curiosity-driven experiments, using cathode ray tubes, designed to study electric discharges. By refining the apparatus of Wilhelm Wien, who first analysed anode rays by magnetic deflection in 1898,2 Thomson constructed the first mass spectrometer (then called a parabola spectrograph) in 1912 and observed a range of ions from gaseous precursors.

Notably, he observed the isotopes of neon and with remarkable foresight foreshadowed the birth of MS in analytical chemistry, noting in his book “I have described at some length the application of Positive Rays to chemical analysis; one of the main reasons for writing this book was the hope that it might induce others, and especially chemists, to try this method of analysis. I feel sure that there are many problems in chemistry, which could be solved with far

greater ease by this than any other method. The method is surprisingly sensitive — more so than even that of spectrum analysis, requires an infinitesimal amount of material, and does not require this to be specially purified; the technique is not difficult if appliances for producing high vacua are available.”3

In 1985 another curiosity-driven discovery, highlighting the power of MS, was the discovery of the fullerenes. Experiments aimed at understanding how long-chain carbon molecules are formed in interstellar space were conducted. The vaporisation of graphite, by laser irradiation, produced remarkably stable carbon clusters.4 The identity of the clusters was revealed by MS.

The observation of C60 as the predominate species led to the proposal of the buckminsterfullerene structure for which Robert Curl, Harlod Kroto and Richard Smalley were awarded the 1996 Nobel Prize in chemistry. This work inspired a range of researchers to develop approaches to synthesise bulk samples of C60 to investigate its structure, properties and reactivity. A great deal of research has followed, with recent highlights being the

5 spectroscopic identification of some interstellar bands to the C60 cation and the development of fullerene derivatives as key materials in photovoltaic devices.6-9

MS proves to be more than an analytical method to identify ions based on their mass-to-charge ratio (m/z) and isotope distributions. MS based methods are capable of: (i) directing the condensed phase synthesis of new material by observing and identifying prominent ions in solution and their development over time, (ii) exploring the fundamental structure of ions by means of energising said ions and rationalising their fragmentation, and (iii) investigating the reactivity of discrete ions, based on their mass-selected m/z, with neutral molecules, ions or photons leading to new and exciting reactivities that can be translated into real-world applications. These topics, among others, will be explored in greater detail in later sections of this introduction.

This thesis presents a MS directed approach for the investigation of coinage metal hydrides as reactive intermediates in catalysis and to explore their significance to nanoparticle synthesis. A powerful suite of MS investigative tools comprising multi-faceted methods to

2 explore the structure, properties and reactivity of such compounds has been used. Gas-phase studies have been used to direct the condensed-phase synthesis, where possible, in attempts to develop practical applications from fundamental discoveries. The formulae shown throughout this thesis are in accordance with the description in their respective references.

1.1 Coinage Metal Hydrides

1.1.1 History

The earliest scientific report of an isolable coinage metal hydride dates back to 1844 wherein

Wurtz noted that the reduction of cupric sulfate with a solution of hypophosphorus acid resulted in a red-brown pyrophoric precipitate.10 Careful experiments noting its decomposition and stoichiometric evolution of hydrogen gas upon reaction with hydrochloric acid concluded that this substance was copper hydride. The nature of the substance was debated in an exchange of publications between Wurtz and Berthelot.11-14 Further structural characterisation of the isolable binary copper hydride/deuteride had to wait more than 80 years until a series of reports on its analysis via X-ray powder diffraction,15 X-ray and neutron diffraction,16-17 which revealed the Wurtzite structure shown in Figure 1.1.

Figure 1.1 Crystal structure of copper hydride. Grey spheres represent copper positions. Black spheres represent hydrogen (or deuterium) positions for the ordered model (Wurtzite structure‐type). On the right it is shown that face sharing tetrahedra cannot be occupied simultaneously because the H–H (or D–D) distance will be too short. Bond lengths are given in pm. Figure reproduced from reference.17

In 1952,18 Wiberg and Henle reported that when a solution of cuprous iodide in pyridine was reacted with a solution of LiAlH4 in diethyl ether mixed with pyridine a product identical to that prepared by Wurtz was formed. This synthesis was later investigated in greater detail by Dilts and Shriver who designed experiments to more thoroughly characterise CuH prepared in organic media.19 However, evidence of copper-hydrogen bonds by infrared and 1H NMR spectroscopies proved elusive in light of poor solubility and the high nuclear spin and quadrupole moment of copper respectively.

In 1897, Bartlett and Rice isolated and characterised the first silver hydride which was formed by reacting a dilute solution of silver nitrate with excess dilute phosphinic acid resulting in the solution becoming wine-red in colour, then black and finally depositing as a black flocculent precipitate. The precipitate was thoroughly dried and carefully weighed before and after ignition in a porcelain crucible. The difference in mass suggested the dried precipitate prior to ignition was 99.289 % silver and thus the mass loss was attributed to a single hydride.20

Berthelot proposed silver hydride as a product in the reactions of hydrogen with silver,21 however it wasn’t until 1926 that Hulthen and Zumstein reacted metal vapours of silver in a carbon tube and H2 to observe the absorption spectrum of silver hydride in the ultraviolet spectral region,22 with similar studies to published between 1927-1943.23-28 Pietsch and

Seurferling noted the formation of a metallic hydride of silver by treating a silver plate with a stream of hydrogen wherein after 2 hours a fine white powdered layer of solid salt-like silver hydride was formed.29 Although several scientific reports of silver hydride under extreme conditions have appeared, silver hydride has eluded isolation from reactions carried out under similar conditions to those described above for copper hydride.

Gold hydride has proven most elusive compared to its copper and silver counterparts. One of the earliest scientific reports describing gold hydride was in 1915 by Mostowitsch and

Pletneff.30 They investigated the volatility of gold in air and other gases in a Heraeus furnace and noted that “…apparently above 1200⁰C an unstable compound is formed…” which was proposed as Au2H2. In 1926 Hulthen and Zumstein reported the first absorption spectra for

AuH in the ultraviolet under extreme experimental conditions by heating the vapor of the metal

to about 1700⁰C in a stream of hydrogen gas.31 Similar studies to investigate the spectral bands and electronic states of gold hydride followed,24, 32-37 some of which searched for gold isotopes other than the solitary naturally occurring 195Au,38 or to investigate hydride/deuteride isotope effects.39 It wasn’t until the work of Wiberg that the existence of gold hydride as an isolable complex was reported.40 Wiberg proposed the formation of complexes of gold(III) hydrides at low temperatures from the reaction of gold trichloride with lithium hydride to produce AuH3 and lithium chloride.

Recently the applications of heteroleptic coinage metal clusters, have shown promise in catalysis, hydrogen storage and environmental mitigation of carbon dioxide. Heteroleptic coinage metal hydrides are represented by the general formula MxHyLz (M = Cu, Ag or Au; H

= hydride; L = a ligand; x, y and z ≥ 1) and homoleptic coinage metal hydrides are represented by the general formula of MxHy. To better understand how they are synthesised and used in practical applications, it is useful to appreciate the underlying chemical principles that govern their synthesis, structure and properties. Several in-depth reviews on coinage metal hydrides have been published, for example: in 2016 Dhayal et al41 reviewed polynuclear copper hydrides; Jordan et al42 focused on the condensed-phase synthesis, structure and reactivity of coinage metal hydride clusters in general, and; in 2018 Liu et al43 described atomically precise nanoclusters and their applications. The following sections of this introduction will focus on these key topics: the formation of various coinage metal hydride clusters as isolable compounds and their structural characterisations; the transformation of organic substrates by coinage metal hydride clusters; applications in hydrogen storage and CO2 mitigation; inspirations from nature in relation to designing the active site of catalysts involving coinage metal hydrides, and; methods relevant to the work undertaken in this thesis.

1.1.2 Synthesis

The most common method of synthesising heteroleptic coinage metal hydrides is the bottom- up or self-assembly approach, Scheme 1.1.

Scheme 1.1 A generalised summary for the stepwise synthesis of coinage metal hydride complexes. A

= Charged Ligands; L = Neutral Ligands; M = Cu, Ag or Au; Y = Counter Ions.

This typically occurs in three phases: (i) Precursor assembly: metal ions and ligands are mixed in the appropriate stoichiometry, usually in a solvent, to give an ionic precursor complex. The ligands can be charged (e.g. thiolates, carboxylates, halides etc.) or neutral (e.g. N- heterocyclic ligands (NHCs), phosphines, crown ethers etc.). The ligands play a crucial role in arresting cluster/particle growth by preventing the formation of nanoparticles or metallic mirrors commonly formed from reducible metal ions in the presence of reducing agents such as hydrides.44-48

The coinage metals are usually provided in the form of a metal salt such as Cu(CH3CN)4BF4,

Ph3PAuCl, AgNO3 etc.; (ii) Hydride source: wherein reduction or hydride transfer occurs,

- yielding the coinage metal hydride complex. Borohydride (BH4 ) is well-known in the formation of monodisperse gold nanoparticles49 and has frequently been used as a hydride source in the formation of heteroleptic coinage metal hydrides, although the mechanism is not well

6 understood and is often referred to as a “black box”;50-51 (iii) Post-complex treatment/processing: including further reaction with the same or another hydride source, ligand etching, halide treatment etc. The structures of some common ligands used to synthesise these coinage metal hydride complexes and their abbreviations are given in

Scheme 1.2.

Scheme 1.2 The structures of some common ligands used to synthesis coinage metal hydride clusters.

52 The hexameric copper hydride [HCu(PR3)]6 known as Stryker’s reagent or Osborn’s complex53-54 is a well-known and well-studied example wherein diverse synthetic approaches

have been reported using different hydride sources ranging from superhydride (LiEt3BH), H2 gas, hydrosilanes, and borohydrides, see Scheme 1.3.

41 Scheme 1.3 Diverse methods for the preparation of [HCu(PR3)]6. Scheme reproduced from reference.

Other bottom-up methods resulting in the formation of hydrides include hydrogenolysis, which has been successfully used in the isolation of ligated silver hydride compounds. For example,

Tate et al. synthesised a dinuclear µ2-silver ligand-protected alkoxide which was reacted with hydrogen at increased pressure and room temperate for 4 days. The resulting hydrogenolysis yielded the alcohol and a ligand-protected dinuclear silver µ2-hydride as the tetrafluoroborate salt i.e. [(LAg)2(µ2-H)][BF4], see Scheme 1.4. Similar hydrogenolysis reactions were explored

55 wherein ligand-protected silver fluoride salts were reacted to form [(LAg)2(µ2-H)][HF2].

Heteroleptic gold hydrides have long eluded their isolation as stable compounds until the work of Tsui et al.56 who used the NHC 1,3-bis(2,6-diisopropylphenyl)imidazole-2-ylidene (IPr) because of its known ability to stabilise unusual gold species57-60 and its previous application

61 in stabilising the dimeric copper hydride [(IPr)CuH]2. The ligand stabilised gold (I) hydride,

[(IPr)AuH], was generated from either one of two pathways starting from the known gold chloride compound [(IPr)AuCl] reacting: (i) directly with superhydride LiBEt3H (or the deuteride analogue, LiBEt3D), or (ii) in a stepwise manner first transforming [(IPr)AuCl] to [(IPr)AuOtBu] by reacting the chloride with NaOtBu in dichloromethane or benzene and subsequent reaction

8 of the isolated [(IPr)AuOtBu] with trimethoxysilane in dichloromethane or benzene for 1.5 hours at room temperature to yield the hydride [(IPr)AuH], see Scheme 1.5.

Scheme 1.4 Preparative hydrogenolysis of ligand-protected silver alkoxide and silver fluoride complexes. t-Pent-O = 2,2-dimethylpropoxide. Scheme reproduced from reference.55

Scheme 1.5 Synthesis of a stable gold hydride.56

Careful consideration of the reaction conditions, choice in ligand(s), counterions, solvent etc. all play a critical role in the resulting product(s). The choice of reaction conditions is critical in acquiring isolable diverse architectural assemblies that can be further investigated in the condensed phase. The next section of this thesis will explore key papers reporting on the structures of isolable coinage metal hydrides.

1.1.3 Structures

Structural information concerning the spatial arrangement and connectivity of atoms within a coinage metal hydride can provide useful information regarding the properties and the potential reactivity of the compound. For example: (i) the number, bonding and arrangement of metal atoms can suggest metal-metal bonding (metallophilicity)62-63 with the potential of optical properties,64-74 for example luminescent compounds (ii) the strength of the metal- hydride bond may have the potential, albeit limited, to explain trends in reactivity,75-76 and (iii) steric and electronic contributions of ligands77 may have the potential, albeit limited, to “tune” the reactivity of the active site.78-79 Analytical methods that provide structural information include crystallographic techniques, spectroscopic techniques and spectrometric techniques and will be covered in sections 1.2 and 1.3. The remaining parts of this section will discuss key coinage metal hydride structures primarily for which crystallographic data has been recorded.

1.1.3.1 Copper hydrides

Copper hydrides have been widely studied by crystallographic methods. Diverse molecular architectures have been described, ranging from simple mononuclear compounds to more complex clusters or nanoparticles. This section provides a summary of selected copper hydrides that have been structurally characterised by X-ray and/or neutron crystallography.

1.1.3.1.1 Mononuclear complexes

Mononuclear homo- and heteroleptic copper complexes have eluded structural characterisation via X-ray and/or neutron crystallography. In 2017 Romero et al. reported the spectroscopic characterisation of [(IPr**)CuH] (where IPr** = 1,3-bis[2,6-bis[di(4-tertbutylphenyl)-methyl]-4-methylphenyl]imid-azol-2-ylidene), was found to exist at equilibrium with its dimer in solution.80 The results supported the hypothesis that the dissociation of copper hydride aggregates occurs in solution and may be responsible for the apparent dearth of isolable heteroleptic copper terminal-hydrides.

10 1.1.3.1.2 Binuclear complexes

The earliest example of a dimeric copper hydride was first reported by Goeden et al. in 1986.81

The tridentate phosphine ligand tripod (CH3C(CH2PPh2)3), with a bidentate chelate binding

2 mode to the copper(I) atoms, was used to generate the [(μ-H)2Cu2(η -tripod)2] complex comprising two bridging hydrides, see Figure 1.2 (a).

(a)

(b)

2 Figure 1.2 The ORTEP view of (a) [(μ-H)2Cu2(η -tripod)2]; Cu-P2 = 2.229 (2), Cu-P3 = 2.229 (2) A, Cu-

H = 1.66 (8), Cu-H' = 1.81 (8); P2-Cu-P3 = 101.3 (1), P2-Cu-H = 113 (3), P3-Cu-H = 116 (3), H-Cu-H' =

81 94 (3), Cu-H-Cu' = 86 (3). Figure reproduced from reference. (b) [{(IDipp)Cu}2(μ-H)]BF4 wherein the

- BF4 counterion is omitted for clarity. Cu1–Cu2 2.5331(15), C1–Cu1 = 1.873(9), Cu1–H1 = 1.45(2), H1–

Cu2 = 1.45(2), Cu2–C28 = 1.908(8); C1-Cu1-H1 167.4(16), Cu1-H1-Cu2 122(3), H1-Cu2-C28 174.4(16),

C1-Cu1-Cu2 163.2(3), Cu1-Cu2-C28 156.1(3). Distances in (Å) and angles in (°). Figure reproduced

from reference.82

The reaction of hydrogen gas with (CuOtBu)4 in the presence of stoichiometric amounts of ligand in THF yielded yellow cubic crystals that were characterised by X-ray crystallography.

The X-ray diffraction data shows the unit cell of HCu(tripod) to be composed of centrosymmetric dimers. One of three phosphorus atoms from each of the tripod ligand donor arms is not coordinated to the copper atom which is bonded by the remaining two contiguous bidentate chelating phosphorus atoms of each the ligands. In 2013 Wyss et al. reported a

+ + triangular {Cu2H} , (better described as {Cu2(μ-H)} to highlight the bridging hydride), core supported by NHC ligands as its tetrafluoroborate salt, see Figure 1.2 (b).82 The complex was prepared by the reaction of a siloxide-bridged dicopper tetrafluoroborate with pinacolborane

4,4,5,5-tetramethyl-1,3,2-dioxaborolane. A short intermetallic distance of 2.5331(15) Å and

2.5354(15) Å for the copper atoms are much less than twice the van der Waals radius of copper suggesting that cuprophillic interactions are involved.83 The tetrafluoroborate counter anion is outside the coordination sphere. The hydride shows significant anionic character,

+ within the positively charged {Cu2H} core, with a charge of -0.123 determined by natural population analysis.84

1.1.3.1.3 Trinuclear complexes

In 2013 Eberhart et al. reported a trinuclear copper hydride cluster prepared by the reaction of dppbz with CuCl and KOtBu in toluene. Hydrogen gas was bubbled through to give a precipitate which was extracted and dried. The dried precipitate was then dissolved in a minimum amount of toluene and layered with acetonitrile to afford yellow crystals suitable for

X-ray crystallography. The hydride ligands are μ2 in the triangular copper plane and the Cu-

Cu distances were reported to be 2.58 Å.85 Mao et al. prepared the trinuclear clusters

- - - - 66 [Cu3(dcpm)3(μ3-H)]Y2 (Y = ClO4 , BF4 , PF6 , CF3SO3 ) by the reaction of [Cu2(dcpm)2]Y2, with

86 2+ a saturated methanol solution of KOH. The triangular Cu3 core of the [Cu3(dcpm)3(μ3-H)] ion is supported by three bulky dcpm ligands each of which bridge one Cu-Cu edge. The Cu-

Cu distances range from 2.879(1)-2.885(2) Å. Due to the poor quality of the crystals and the inherent difficulty of locating the hydride by X-ray crystallography, 1H NMR was used to interrogate the arrangement of atoms around the hydride. The observed septet at ca. 2.2 ppm, which collapsed to a singlet in 31P-decoupled 1H NMR experiments, is consistent with a

12 structural arrangement in which the 6 phosphorus atoms from three dcpm ligands surround the trinuclear copper core to which the hydride is triply bridging.

(a) (b)

Figure 1.3 (a) Ball-and-stick diagram of [Cu3H(dppm)3(OAc)2]. All hydrogen atoms (except the hydride ligand) and solvent molecules are omitted for clarity. Colour legend: Cu = green; H = magenta; P =

87 yellow-orange; O = red; C = gray wireframe. (b) ORTEP-3 derived structure of cluster cation of [Cu3(μ3-

+ H)(μ3-BH4)(dppa)3] (BF4) with 30% ellipsoid possibilities. Phenyl hydrogen atoms, amine hydrogen

− atoms, and anion (BF4 ) are omitted for clarity. Distances in (Å): Cu1−Cu3 = 2.6164(5), Cu1−Cu2 =

2.6706(5), Cu2−Cu3 = 2.6785(5), Cu1−B1 = 2.584(2), Cu2−B1 = 2.598(2), Cu3−B1 = 2.549(2), Cu1−P5

= 2.2624(6). The noncoordinating BF4 counter-anion and disordered CH2Cl2 solvent molecules are not shown. Figure reproduced from reference.88

Acetato coordinated trinuclear copper hydride clusters were reported by Cook et al. in 2016.87

The reaction of Cu(OAc) with Ph2SiH2 and dppm (1:1:0.5) in benzene afforded the trinuclear copper cluster with a triply bridging hydride centered to one face of the three copper atoms and bidentate and monodentate acetate ligands. The 1H NMR spectrum recorded a septet centred at 3.1 ppm assigned to the hydride. In 2016 the first MS directed synthesis of a trinuclear copper hydride cluster supported by bis(diphenylphosphino)amine (dppa) ligands was reported.88 Electrospray ionisation (ESI) MS analysis of acetonitrile solutions comprising

[Cu(CH3CN)4]BF4, dppa and NaBH4 revealed an intense peak corresponding to [Cu3(μ3-H)(μ3-

+ BH4)(dppa)3] , see Figure 1.3 (b). This prompted the isolation and characterisation of the cation as the tetrafluoroborate salt. Structural characterisation revealed a triangular trinuclear

copper cluster wherein a μ3-BH4 and μ3-H are coordinated to opposing faces of the copper core.

(a) (b)

Figure 1.4 (a) Perspective view of the cation of [Cu3H(dppm)3(O2CH)]PF6 highlighting the C3 symmetry

89 of the formate due to disorder, and (b) The core of [dppm3Cu3(µ3-H)( µ2,µ1-S2CH)](BF4) highlighting the

90 dithioformate bonding to the Cu3 core. Figure reproduced from reference.

Recently there have been reports on the formation of trinuclear complexes by reaction of copper hydride complexes with substrates. In 2017 Nakamae et al. showed a stepwise

89 approach to the formation of [Cu3H(dppm)3(O2CH)]PF6. Firstly, the well-known Stryker’s reagent (section 1.1.3.1.6) [CuH(PPh3)]6 was reacted with dppm and [Cu(CH3CN)4]PF6 to yield

[Cu8(µ-H)6(µ-dppm)5](PF6)2 (section 1.1.3.1.8). Further reaction of [Cu8(µ-H)6(µ-dppm)5](PF6)2

+ with CO2 afforded isolable [Cu3H(dppm)3(O2CH)]PF6. The cation [Cu3H(dppm)3(O2CH)] comprises an equilateral Cu3 core, a µ3-hydride and three dppm ligands similar to those

88 described above. Ma et al. described the reaction of [dppm3Cu3(µ3-H)(µ3-BH4)](BF4), supra,

90 with carbon disulphide to yield isolable [dppm3Cu3(µ3-H)(µ2,µ1-S2CH)](BF4). Carbon disulphide reacts at the borohydride “site” to give a µ2,µ1-dithioformate coordinated to the Cu3 triangular core with a planar geometry. A single sulfur of the dithioformate is coordinated to a single Cu of the Cu3 core with a bond distance of 2.3568(12) Å. The remaining sulfur is coordinated to two copper ions with a bond distance of 2.3985(12) and 2.4620(12) Å respectively, see Figure 1.4 (b).

1.1.3.1.4 Tetranuclear complexes

+ The two tetrahedral copper hydride cations [Cu4(μ4-H)(μ2-X)2(PPh2Py)4] (X = Cl, Br;

91 Py = pyridyl) were synthesised exhibiting a common tetrahedral unit, “Cu4(μ4-H)”. Doubly bridging halides (X=Cl-, Br-) are located at opposite edges of the tetrahedron with the P and N atoms of each of the four ligands bridging the remaining four edges of the tetrahedron.

1.1.3.1.5 Pentanuclear complexes

The pentanuclear complex [HCu(PPh3)]5 was derived from Stryker’s reagent (see section

1.1.3.1.6) as a product of partial hydrolysis.92 The five triphenylphosphine ligands of the pentamer form shorter Cu-P bonds, compared to Stryker’s reagent, ranging between 2.16(1) to 2.21(1) Å. The complex forms a distorted trigonal-bipyramidal metal array of [HCu(PPh3)]5.

The array is distorted from the ideal trigonal-bipyramid in that: (i) the Cu-Cu bond lengths differ in each of the pyramidal subunits and (ii) the equatorial Cu3 arrangement forms an isosceles arrangement rather than an equilateral triangle. The authors, faced with the challenge of uncertain hydride locations (due to the use of X-ray rather than neutron diffraction), hypothesise that differentiation of one of the six faces of the bipyramid would arise due to the absence of a hydride at that face. However, no face of the bipyramid presents itself as a candidate to invoke such perturbations. A further hypothesis is provided that the cluster may

- in fact be [(PPh3)5Cu5H6] , however no counterion was located. Consequently, questions remain regarding the true structure of the proposed pentamer [HCu(PPh3)]5.

1.1.3.1.6 Hexanuclear complexes

Arguably one of the most well-studied copper hydride reagents is Stryker’s reagent52 (also known as Osburn’s complex).53-54 First described by Bezman et al. in 197153-54 as a hexameric copper hydride coordinated to six triphenylphosphine ligands of the formula [PPh3CuH]6. This was the first truly stoichiometric copper hydride complex since the reports of Wurtzite in 1844.

In 1989 Albert et al. reported an oxo-bridged analogue of Stryker’s reagent.92 The hydride

source used was potassium tri-sec-butylborohydride (k-selectride). The structures described reported the oxo-bridged hydrolysis product, [(Ph3P)6Cu6H4O]·THF.

(a) (b)

Figure 1.5 (a) Molecular structure of the H6Cu6P6 core of the H6Cu6[P(p- tolyl)3]6 cluster. Thermal ellipsoids are at the 50% probability level.11 Average pertinent distances (showed in Å): Cu-Cu(long) =

2.68(4), Cu-Cu(short) = 2.52(3), Cu-H = 1.76(3). Primed atoms are related to unprimed ones by a crystallographic inversion. The drawing shows one of the two independent clusters in the unit cell.93 (b)

Two views of the [HCuP]6 core of Stryker’s reagent. The hydrides appear to be triply bridging and edge bridging, left and right view, respectively. Figure reproduced from reference.94

The triphenylphosphine derivative of Stryker’s reagent was found to crystallise into thin flat plates and was found unsuitable for neutron diffraction studies.53-54 Stevens et al. applied a lateral solution to this problem by substituting the triphenylphosphine for a structurally similar analogue, P(p-tolyl)3, which afforded crystals in a prismatic habitat that were suitable for

93 neutron diffraction studies, see Figure 1.5 (a). Neutron scattering of H6Cu6[P(p- tolyl)3]6 revealed the six hydride ligands to be situated at face-bridging positions rather than edge- bridging positions. The positions of the hydrides in Stryker’s reagent ([PPh3CuH]6) remained elusive until Bennett et al. used a combination of inelastic neutron scattering and infrared spectroscopy, combined with ab initio calculations, to identify the spectral features that contain substantial contributions from C-H stretching and bending modes.94 The hydrides were found to be best described as edge bridging rather than face bridging, see Figure 1.5 (b).

+ In 2017 Liu et al. reported the oxidation of [PPh3CuH]6 with decamethylferrocenium (Cp*2Fe )

95 + yielded novel cationic copper hydride complexes. [PPh3CuH]6 is first oxidised with Cp*2Fe

16 ●+ to give the radical cation [(Ph3P)6Cu6H6] . Further oxidation yields the dication,

2+ + + [(Ph3P)6Cu6H6] , which loses a proton to yield [(Ph3P)6Cu6H5] . The [(Ph3P)6Cu6H5] complex could also be synthesised by reacting the hexamer, [PPh3CuH]6, with MeOTf (1 eq.) in benzene. Subsequently, isolable [(Ph3P)6Cu6H5]OTf was characterised by X-ray diffraction.

The asymmetric unit displayed two Cu6 cluster cations wherein each complex displays two tetrahedra sharing an edge. The Cu-Cu distances differ only slightly.

A hexanuclear cluster was described by Kohn et al. in 2003 whereby transformation of [Cu3(μ3-

Cl)(μ2-Cl)L2]2(BF4)2 to [Cu6(μ6-H)(μ3-Cl)(μ2-Cl)3L3]BF4 (where L = N,N’,N’’-trimethyl-1,3,5- triazacyclohexane).96 It is proposed that C-H activation of the ligand occurs on the surface of the “Cu3” face of the precursor cluster subsequently leading to the rearrangement of the cluster with a 6-coordinate hydride in the centre of a copper octahedron. The Cu-Cu distances are similar in each of the two cluster units. All copper atoms of [Cu6(μ6-H)(μ3-Cl)(μ2-Cl)3L3]BF4 showed similar distances to the centre hydride of 1.85(3) Å.

1.1.3.1.7 Heptanuclear complexes

+ The heptanuclear cluster, [(Ph3P)7Cu7H6] , was prepared by reacting [PPh3CuH]6 with

95 1 [Cp*2Fe][B(C6F5)4] in CH2Cl. The H NMR at 250 K revealed an octet indicating seven equivalent phosphine atoms, while X-ray crystallographic analysis of suitable crystals revealed that the heptanuclear complex [(Ph3P)7Cu7H6](OTf) consists of a [PPh3CuH]6 unit capped by

+ a “CuPPh3” unit that occupies one of the larger Cu3 faces of the cluster core.

The octanuclear cluster [Cu8(H){S2CR}6](PF6), described below, reacted with one eq. of

97 [NH4][BH4] to produce isolable [Cu7(H){S2CR}6] as a neutral cluster. Neutron diffraction studies revealed a tricapped distorted tetrahedron in which the central site was occupied by a

4-coordinate hydride, see Figure 1.6 (a) & (b). The mean Cu-H distance, as determined by neutron diffraction, was 1.86(2) Å, see Figure 1.6 (c).

(a) (b) (c)

Figure 1.6 (a) Molecular structure of the [Cu4(μ4-H)(μ3-Cu)3{S2C(aza-15-crown-5)}6]; H atoms omitted for clarity except the interstitial hydride. (b) Cu7 framework displays an elongated tricapped tetrahedron from

97 X-ray. (c) Cu7H core drawing from neutron diffraction data. Figure reproduced from reference.

1.1.3.1.8 Octanuclear complexes

In 1985, Lemmen et al. used X-ray crystallography to show that H8Cu8[Ph2P(CH2)3PPh2] consists of a Cu8 dodecahedron decorated by four phosphine ligands, 1,3- bis(diphenylphosphino)propane (dppp).98 The hydride ligands were not clearly resolved from the X-ray data and their locations were proposed based on similar structures.

(a) (b)

i Figure 1.7 X-ray diffraction of [Cu8(H){S2P(OPr)2}6]PF6 (a) Discrete structure. (b) “Cu8(μ-H)” core. Figure reproduced from reference.99

i + 99 In 2009 Liu et al. reported a cationic octanuclear complex [Cu4(H)(μ3-Cu)4{S2P(O Pr)2}6] .

Four copper atoms per cube are fully occupied in a tetracapped tetrahedron, see Figure 1.7.

The average Cu-H distances reported was 1.84 Å.

2 31 1 Figure 1.8 (a) VT H NMR of [Cu8D6dppm5](PF6)2, and (b) VT P{ H} NMR of [Cu8H6dppm5](PF6)2.

Figure reproduced from reference.89

Recently, Tanase and co-workers showed that the [Cu8H6dppm5](PF6)2 cluster could be

89 formed directly from Stryker's reagent (see section 1.1.3.1.6). X-ray analysis revealed a Cu8 trans-bicapped octahedron wherein the Cu8P10 fragment is described as having a distorted C2

100 symmetry compared to the [Cu8H8(dppp)4] possessing a S4 distorted Cu8 dodecahedron.

The crystals were too small for neutron diffraction to determine the location of the hydrides.

Thus, the structural analysis was complimented by theoretical calculations which used the X- ray data for the coordinates of the larger atoms of the Cu8P10 framework. Variable temperature

(VT) NMR measurements revealed fluxional characteristics in solution where the hydrides and

2 ligands scramble around the Cu8 structure. The VT H NMR of the deuterated analogue

[Cu8D6dppm5](PF6)2 shows changes in the broad singlet at δ 2.68 ppm at 20⁰C , which collapses and then separates into three distinct signals as the temperature is decreased, see

Figure 1.8 (a).89 Similarly, the VT 31P{1H} NMR showed the evolution of distinct signals as the temperature was decreased, see Figure 1.8 (b).89 The authors propose various models wherein the phosphorus atoms of the dppm ligands switch binding sites on the Cu8 framework.

They further assume that the hydrides are coupled to site-exchange processes as dppm moves about the Cu8 framework.

1.1.3.1.9 Larger nuclearity clusters

In 2015 Nguyen et al. demonstrated that Stryker’s reagent ([(Ph3P)CuH]6) can be used as a suitable hydride source for the synthesis of larger clusters. The reaction of Stryker’s reagent with 1,10-phenanthroline (phen) in the presence of a halide or a pseudo-halide gave the

2+ 1 dication [Cu14H12(phen)6(PPh3)4] . The presence of 12 hydride ligands was supported by H

NMR spectroscopy and ESI-MS.101

102 The air and moisture stable copper hydride, [Cu18H7L10I] was reported in 2014. The

- phosphinothiolate ligands (L = [SC6H4PPh2] ) were used to form the complex [{CuI(LH)}2] wherein the {Cu2(I)2} unit is planar with a crystallographic inversion centre. The complex,

[{CuI(LH)}2], was then reacted with 15 eq. of sodium borohydride to give the polyhydrido cluster

[Cu18H7L10I]. X-ray crystallography was used to identify the structure of the compounds. The location of the hydrides was estimated by density functional theory (DFT) calculations.

Hydrogen gas evolution was noted under mild conditions such as thermolysis at near ambient temperature and UV-light irradiation.

The nanoclusters [Cu18H16(dppe)6](BF4)(Cl) and [Cu16H14(dppa)6](BF4)2 were synthesised and structurally characterised by X-ray and neutron diffraction revealing bifrustum and frustrum

103 2+ metal-core architectures. [Cu18H16(dppe)6] exhibits two frustrum cupolae and

2+ [Cu16H14(dppa)6] exhibits a Cu9 frustum cupola and heptanuclear distorted hexagonal-shape base, see Figure 1.9.

Figure 1.9 (a) The side view of the racemic Cu18 kernels (top cupola: green; bottom cupola: orange), divided by mirror symmetry (s); (b) The top view of two racemic Cu18 kernels and the superimposed bottom cupolas (orange). (c) The side view of the racemic Cu16 kernels (top cupola: green; bottom cupola: orange), divided by mirror symmetry (s); (d) The top view of two racemic Cu16 kernels and the superimposed bottom cupolas (orange). The drawings of bonds between top and bottom cupolae are not shown. Figure reproduced from reference.103

A Cu20 cluster was isolated wherein a Cu18 cage adopts an elongated triangular orthobicupola

104-106 (ETO) that encapsulate the remaining Cu2. The encapsulated Cu2 fragment reported short Cu-Cu distances of 2.307(1) Å. The ETO structure is composed of 12 quadrilateral faces and 8 triangular faces exhibiting D3h symmetry. Nine dithiophosphate (dtp) ligands in various coordination modes encase the cluster core, see Figure 1.10 (a). Neutron diffraction studies identified the precise location of the 11 hydrides, which display three distinct coordination modes, see Figure 1.10 (b).

(a) (b)

i Figure 1.10 Structure of [Cu20(H)11{S2P(OPr)2}9]. (a) X-ray structure and (b) neutron diffraction structure.

(Color code: Cu, blue; Cu−Cu edges, cyan; S, yellow; P, gray; H, pink.). Figures reproduced from references.104-106

(a) (b) (c) (d)

Figure 1.11 X-ray (a−c) and neutron (d) structure of [Cu32(H)20{S2P(OiPr)2}12]. (a) Hexacapped rhombohedron (pink) of 14 Cu atoms sandwiched between two triangular cupola (cyan) of (2 × 9) Cu atoms, (b) Elongated triangular gyrobicupola of Cu18 motif, (c) Twelve ligands in four spherical rows (w, x, y, z) around metal core, (d) Position of 20 hydrides. (Colour code: Cu, cyan and magenta; S, yellow;

P, green; H, pink.). Figure reproduced from reference.107

An extended form of the Cu20 cluster, described earlier, was prepared from the reaction of a

- copper(I) salt, dtp and [BH4] in a ratio of 8:3:12 under inert and cooled conditions as reported

i 107 by Dhayal in 2015 to yield [Cu32(H)20{S2P(O Pr)2}12], see Figure 1.11. The Cu-H distances, as determined by neutron diffraction, ranged between 1.640(70) to 2.121(8) Å. The cluster exhibits both air and moisture stability.

Ligand-induced transformation of [Cu25H22(PPh3)12]Cl with Ph2Phen and a chloride source was reported by Nguyen et al.108 The process resulted in the isolation and structural characterisation of [Cu29Cl4H22(Ph2phen)12]Cl. The ligand exchange reactions are thought to template the growth of the cluster and thus resemble the Ostwald ripening process.109-110

Furthermore, Cu(OAc) and CuCl were shown to react with Ph2SiH2 and PPh3 to yield

111 [Cu25H22(PPh3)12]Cl which features an icosahedral Cu13 core. The reaction also produced

[Cu18H17(PPh3)10]Cl which was similarly characterised.

In 2014 Edwards et al. reported two copper hydride clusters, [Cu28(H)15(S2CNR)12]PF6 (NR =

n N Pr2 or aza-15-crown-5), comprising 15 hydride and 28 copper atoms. The rhombicuboactahedral copper polyhdrido complexes were described as “Chinese puzzle molecules” drawing an analogy from Chinese puzzle balls, carved from the same piece of material and consisting of several concentric spheres, each of which rotate freely.112. The two compounds were characterised by X-ray and neutron diffraction wherein each cluster was found to have an isostructural core. The Cu-Cu distances were recorded as ranging from

n 2.603(2) to 2.824(1) Å [Cu28(H)15(S2CNR)12]PF6 (NR = N Pr2 or aza-15-crown-5). The hydrides exhibit a range of coordination modes and Cu-H lengths ranging from 1.56(5) to 2012(3) Å, see Figure 1.12 (a)-(g).

(b) (c) (d)

(e)

(a) (g)

(f)

Figure 1.12 (a) The cubic [Cu28(H)15{S2CNPr2}12] cation. (b) The rhombicuboctahedral framework of

Cu24. (c) Each of the twelve square faces of the Cu24 core capped by a dtc ligand. (d) The Cu24 core is enclosed by a truncated octahedron of 24 sulfur ligands. (e) Three-coordinate hydrides (µ3-H) on eight triangular faces of the Cu24 core. (f) Six truncating hydrides inside six square faces of the Cu24 core form a bridging octahedron motif; the core is concentrically arranged around the interstitial hydride (µ4-H). (g)

The core assembly: H@Cu4(tetrahedron)@H6(octahedron)@Cu24(rhombicuboctahedron)@H8(cube)

@S24(truncated octahedron) reminiscent of a Chinese puzzle ball (inset). Carbon gray, copper cyan, hydrogen red, nitrogen pink, sulfur yellow. Figure reproduced from reference.112

24 1.1.3.2 Silver hydrides

In the late 19th century, and almost 60 years after Wurtz’s discovery, Bartlett20 and Webster113-

115 first proposed the formation of silver hydride. Structural characterisation of the first isolable silver hydride compound, see section 1.1.3.2.1, remained elusive for almost 60 years.

1.1.3.2.1 Mononuclear complexes

The first isolable mononuclear silver hydride was reported by Romero et al. in 2017.80 The authors hypothesised that the bulky ligand 1,3-bis[2,6-bis[di(4-tertbutylphenyl)-methyl]-4- methylphenyl]imid-azol-2-ylidene (IPr**) would provide sufficient bulk to stabilise the monomeric hydride.116 Three key steps allowed the isolation of the hydride: (i) the synthesis of (IPr**)AgCl by the reaction of IPr**HCl imidazolium with Ag2O in DCM; (ii) isolation of

(IPr**)AgCl and subsequent reaction with KOPh in DCM to give the phenoxide (IPr**)AgOPh;

(iii) addition of pinacolborane (1 eq.) at -40⁰C to a toluene solution of (IPr**)AgOPh resulting in an immediate colour change. The sample was analysed by 1H and 13C{1H} NMR spectroscopy which showed a pair of doublets indicating the presence of the monomeric hydride coordinated to the two spin active silver nuclei 107Ag and 109Ag with a relative abundance of ca. 51.8% and 48.2% respectively.117 The solution was kept in the freezer overnight to afford clear, colourless needles of (IPr**)AgH suitable for X-ray crystallography to give (IPr**)AgH, see Figure 1.13 (a). (IPr**)AgH crystallises in a chiral space group with two molecules in the asymmetric unit. No intermolecular interaction is observed with the nearest atoms of adjacent molecules being 3.74 Å apart. The silver atom exhibits positional disorder therefore the hydride could not be accurately located and was fixed in the molecular structure.

However, the addition of tris(pentafluorphenyl)borane to a solution of (IPr**)AgH yields isolable

(IPr**)AgH[B(C6F5)3] wherein the hydride was located and isotropically refined, giving a Ag-H

- distance of 1.81(3) Å, see Figure 1.13 (b). The B(C6F5)3 appear to be incorporated into the complex via a 3-c, 2e- Ag-H-B bond.

(a) (b)

Figure 1.13 (a) Molecular structure of (IPr**)AgH in the solid state. Ellipsoids are set at 50% probability; solvent molecules and hydrogen atoms other than the hydride are omitted for clarity. C1–Ag1 = 2.108(6)

Å. (b) Molecular structure of (IPr**)AgH[B(C6F5)3] in the solid state. Ellipsoids are set at 50% probability; solvent molecules and hydrogen atoms other than the hydride have been omitted for clarity. Selected distances and angles: C1–Ag1 = 2.095(2), Ag1–H1 = 1.81(3), H1–B1 = 1.21(3); C1-Ag1-H1 = 173.5(9),

Ag1-H1-B1 = 159.7(13). Figure reproduced from reference.80

1.1.3.2.2 Binuclear complexes

+ In 2013 Tate et al. showed that the [Ag2H] core is isolable by use of the N-heterocyclic carbene 1,3-bis(2,6-diisoprpylphenyl)imidazoline-2-ylidene (SIDipp).118 To obtain the

+ binuclear silver hydride they reacted salts of {[(SIDipp)Ag]2(µ-Ot-Bu)} with phenylsilane in

THF. The solution was analysed by 1H NMR spectroscopy, revealing a triplet of triplets cantered at δ -1.13 ppm, suggesting the formation of a bridging binuclear silver hydride. The pattern arises via superimposition of the three resonances arising from the two spin active isotopologues of silver, namely [107Ag-H-107Ag] ↔ [107Ag-H-109Ag] ↔ [109Ag-H-109Ag].117 Slow diffusion of pentane into this solution afforded crystals suitable which were characterised by

X-ray crystallography, see Figure 1.14.

+ Figure 1.14 Solid-state structure showing the cation {[(SIDipp)Ag]2(µ-H)} , 50% probability ellipsoids.

Anion and co-crystallised solvent omitted for clarity. Selected bond lengths (Å) and angles (⁰): Ag(1)–

C(1) = 2.108(3); Ag(2)–C(28) = 2.104(3); Ag(1)–Ag(2) = 2.8087(4); Ag(1)–H(1) = 1.69(4); Ag(2)–H(1) =

1.71(3); C(1)–Ag(1)–Ag(2) = 158.88(8); Ag(1)–Ag(2)–C(28) = 155.46(8); C(1)–Ag(1)–H(1) = 166(1);

C(28)–Ag(2)–H(1) = 170(1). Figure reproduced from reference.118

+ Crystallographic analysis of {[(SIDipp)Ag]2(µ-H)} revealed short silver-silver distance of

2.8087(4) Å compared to the sum of the van der Waals radii for two silver atoms i.e. 3.44 Å.83

Furthermore, the distance between each of the silver atoms to the hydride are 1.69(4) Å and

1.71(3) Å. These bond distances suggest three-centre, two electron bonding in describing the

+ 119-122 + [Ag2H] core. The reactivity of {[(SIDipp)Ag]2(µ-H)} toward carbon dioxide in an

Umpolung reaction will be described in greater detail in section 1.1.7. The “M2(µ-H)” core will be described and explored in greater detail in the subsequent chapters of this thesis.

1.1.3.2.3 Trinuclear complexes

Previously, we have successfully isolated two trinuclear silver hydride clusters

[Ag3((PPh2)2CH2)3(µ3-H)(µ3-Cl)]BF4 and [Ag3((PPh2)2CH2)3(µ3-H)](BF4)2 via a “MS directed approach” to yield isolable crystalline material suitable for crystallographic analysis, see Figure

1.15.123-124 Both clusters feature a triangular silver core. A silver atom is located at each of the vertices of the triangle and one of each of the three dppm ligands, at each edge, bridging two

silver atoms. The hydride is triply bridging at the face of Ag3 core and approaches a trigonal-

2+ planar coordination mode for the dication [Ag3((PPh2)2CH2)3(µ3-H)] and lies 0.31 Å from the

+ plane of the core, compared to 0.91 Å for [Ag3((PPh2)2CH2)3(µ3-H)(µ3-Cl)] .

(a) (c)

(b) (d)

+ Figure 1.15 ORTEP-3 representations of (a) [Ag3((PPh2)2CH2)3(µ3-H)(µ3-Cl)] ; (b) the distorted trigonal

+ bipyramidal “Ag3(µ3-H)(µ3-Cl)” core, with the phenyl groups are omitted for clarity. Displacement

2+ ellipsoids set at the 20% probability level for both (a) and (b); (c) [Ag3((PPh2)2CH2)3(µ3-H)] with hydrogen atoms and solvent omitted for clarity; (d) the silver core with the ligand (phenyl rings and hydrogen atoms of the ligand omitted for clarity). Displacement ellipsoids set at the 50% probability level for both (c) and (d). The location of the deuterium is unambiguously proven by neutron diffraction. Figures reproduced from references.123-124

Chapter 2 of this thesis applies a “MS directed approach” in isolating the related borohydride containing trinuclear cluster [Ag3((PPh2)2CH2)3(µ3-H)(µ3-BH4)]BF4.

28 1.1.3.2.4 Tetranuclear complexes

There appear to be no reports of isolable tetranuclear silver hydride complexes that have been structurally characterised in the condensed-phase. However, zeolites have been proposed to

125 support stable Ag4H2 clusters.

1.1.3.2.5 Pentanuclear complexes

There appear to be no reports of discrete isolable pentanuclear silver hydride complexes at the date of this literature review.

1.1.3.2.6 Hexanuclear complexes

In 2016 Cook et al. reported the synthesis of the polyhydrido [Ag6H4(dppm)4(OAc)2] and its characterisation by X-ray crystallography, see Figure 1.16. 87

Figure 1.16 Ball-and-stick diagram of [Ag6H4(dppm)4(OAc)2]. All hydrogen atoms and solvent molecules are omitted for clarity. Colour legend: Purple = silver; Orange = phosphorus; Red = oxygen; The gray wireframe represents the carbon backbone. Figure reproduced from reference.87

The cluster was synthesised by the addition of dppm and Ph2SiH2 (0.5 eq.) to a slurry of silver(I) acetate in benzene. Workup of the solution afforded [Ag6H4(dppm)4(OAc)2] as a

colourless crystalline solid in 47% yield. The silver atoms occupy the vertices of an octahedron with the 4 dppm ligands coordinated around the Ag4 square plane. The remaining two silver atoms located at the opposing apices and are each coordinated to an acetato ligand. The acetato ligands coordinate via κ1 and κ2 bridging modes. The four hydride ligands were not located in the Fourier map and are proposed to occupy the four trigonal faces opposing the dppm ligands with a triply bridging binding mode.

1.1.3.2.7 Heptanuclear complexes

In 2013 Liu et al. reported the condensed phase structure of two heptanuclear silver hydride

i 126 complexes [Ag7(μ4-H){Se2P(O Pr)2}6] and [Ag7(μ4-H){S2P(OEt)2}6], see Figure 1.17. The clusters were synthesised by converting either of the precursor octanuclear clusters

i [Ag8(H)(Se2P Pr2)6](PF6) or [Ag8(H)(S2P(OEt)2)6](PF6) through the reaction with sodium borohydride (1 eq.). The heptanuclear clusters can be further reacted with sodium borohydride to yield silver nanoparticles of ca. 30 nm or they can regenerate the octanuclear clusters by

i reacting with Ag(CH3CN)4PF6 (1 eq.). Both clusters, [Ag7(μ4-H){Se2P(O Pr)2}6] and [Ag7(μ4-

H){S2P(OEt)2}6], exhibit a tricapped tetrahedral heptanuclear silver structure with the hydride located at the centre. The hydride displays µ4 bonding to the Ag4 core. This core is further decorated by three face capping atoms to give the Ag7 skeleton. This is further enclosed within an icosahedral structure composed of the twelve S/Se atoms contributed by the six dialkyl dichalcogenphosphate ligands.

(a) (b)

i Figure 1.17 (a) Ag7H core of [Ag7(μ4-H){Se2P(OPr)2}6] disordered in two orientations (50% each). (b)

i Molecular structure of [Ag7(μ4-H){Se2P(OPr)2}6] (30% thermal ellipsoid) with isopropyl groups omitted for

clarity. (c) Molecular structure of [Ag7(μ4-H){S2P(OEt)2}6] (30% thermal ellipsoid) with ethyl groups omitted for clarity (d) The Ag7H core of [Ag7(μ4-H){S2P(OEt)2}6] in disordered in two orientations (top

75%, bottom 25%). Figure reproduced from reference.126

31 1.1.3.2.8 Octanuclear complexes

In 2010 Liu et al. were first to report on isolable silver hydride complexes.127 The reaction of

i Ag(CH3CN)4PF6 (8 eq.) with NH4Se2P(OR)2 (6 eq.; R= Pr or Et) further reacted with sodium borohydride (1 eq.) afforded crystals suitable for X-ray diffraction, see Figure 1.18 for the

i structure of [Ag4(μ4-H)(μ3-Ag)4{Se2P(O Pr)2}6](PF6). The presence of the hydride was confirmed by both 1H and 109Ag NMR spectroscopies. The 1H NMR of the complexes display a nonet resonance at chemical shift 3.14 ppm attributed to the hydride. The octanuclear complexes could further be synthesised by the reaction of extended chain polymer

128 [Ag5{S2P(OEt)2}4(PF6)}]n with sodium borohydride.

(a) (b)

i + Figure 1.18 (a) “Ag8H” core of [Ag4(μ4-H)(μ3-Ag)4{Se2P(OPr)2}6] . (b) Molecular structure of [Ag4(μ4-

i + H)(μ3-Ag)4{Se2P(OPr)2}6] (50% thermal ellipsoid). Isopropyl groups omitted for clarity. Figure reproduced from reference.127

In 2011 Liao et al. described another octanuclear silver hydride with a similar Ag8H core, as

97 described vide supra. The “Ag8H” core is surrounded by six 1,1-dicyanoethylene-2,2- dithiolate (i-MNT) ligands giving a highly charged anionic complex as the t-butyl ammonium salt [Bu4N]5[Ag8(H){S2CC(CN)2}6].

32 1.1.3.2.9 Large nuclearity clusters

In 2011 Liu et al. described a hexacapped trigonal bipyramidal undecanuclear silver hydride complex stabilised by dithiocarbamate ligands, [Ag11(H)(S2CNPr2)9](NO3), see Figure 1.19 for

129 the Ag11H core. An acetonitrile solution at -20⁰ of Na(S2CNPr2), AgNO3 and NaBH4 in a

9:11:1 molar ratio afforded crystals suitable for X-ray crystallography. The central hydride is described as having a μ5 bridging mode.

Figure 1.19 A perspective view of the Ag11H core of 1H. Selected bond lengths (Å) and angles (⁰): Ag(1)-

Ag(3) = 2.8262(8); Ag(2)-Ag(3) = 2.8187(8); Ag(3)-Ag(3a) = 2.9623(8); Ag(1)-Ag(4) = 2.8775(5); Ag(2)-

Ag(5) = 2.9309(6); Ag(3)-Ag(4) = 3.1094(8); Ag(3)-Ag(5) = 3.1249(8); H(1)-Ag(2) = 1.9(2); H(1)-Ag(3) =

1.75(4). Figure reproduced from reference.129

1.1.3.3 Gold hydrides

Reports of stable gold hydrides are rare compared to copper and silver hydrides. Indeed, the commonly used reaction of metal salts with borohydrides in the presence of ligands typically gives rise to “metallic” gold nanoclusters.130-133 Homometallic gold hydride complexes with a nuclearity of > 2 have proven elusive. In 2008 Tsui et al. described the first stable monomeric gold hydride [(IPr)AuH], see Figure 1.20 (a).56 The coordination of the hydride ligand was reinforced by 1H NMR spectroscopy displaying a broad singlet at a chemical shift of 3.38 ppm.

Furthermore, the IR spectroscopy displayed a sharp intense band at 1976 cm-1 indicative of a terminal hydride. Similar monomeric gold hydrides have been reported.134-136

(a)

(b)

56 Figure 1.20 (a) Solid state structure of [(IPr)AuH]. (b) Solid-state structure of {[(6Dipp)Au]2(μ-

+ f − H)} [BAr 4] ellipsoids set at 50% probability. For clarity, anion and H atoms except hydride are omitted, and iPr groups are shown in wireframe view. The hydride was located in the difference Fourier map and refined isotropically. Selected bond lengths and angles: Au(1)−Au(2) = 2.7571(3) Å; Au(1)−C(33) =

2.040(5) Å; Au(2)−C(3) = 2.049(5) Å; Au(1)−Au(2)−C(3) = 165.7(1)°; Au(2)−Au(1)−C(33) = 164.6°. Figure reproduced from reference.137

137 Phillips et al. reported on the synthesis of “Au2H”, see Figure 1.20 (b). The triangular Au2H core was stabilised using various expanded-ring N-heterocyclic carbenes. In each case the

Au2H core was supported by two ligands, each coordinated to a gold centre.

More work needs to be done to synthesise Au(III) hydrides.138 In 2012 Rosca et al. described a gold(III) hydride stabilised by the doubly cyclometallated 2,6-bis(4’-tert-butylphenyl)pyridine

(C^N^C)* ligand.135 The reaction of [(C^N^C)*AuOH]139 with superhydride in toluene at -78⁰C occurred rapidly within 15 min and gave a yellow crystalline material. Structural

34 characterisation of the material by 1H NMR revealed a broad singlet resonance at δ -6.51 in

CD2Cl assigned to the hydride. The structure of the complex was confirmed by X-ray crystallography. Recently, Rocchigiani et al. reported a series of spectroscopically observed

140 gold(III) hydrides. Reactions of (C^C)AuH(PR3) (R = Me, Ph, p-tolyl; C^C = 4,4′-di-tert- butylbiphenyl-2,2′-diyl) with superhydride afforded the first examples of phosphino gold(III) hydrides. Leading on from this work in 2016 Pintus et al. reported a series of gold(III) hydride complexes.141

Figure 1.21 Crystal structure of [(C^N^C)AuH] with selected bond distances (Å) and angles (°) for: Au1–

H1 = 1.64(5), Au1–N1 = 2.019(3), Au1–C9 = 2.060(3), Au1–C18 = 2.058(4); N1-Au1-C9 = 81.06(13),

N1-Au1-C18 = 80.79(13), C9-Au1-H1 = 98.2(19), C18-Au1-H1 = 100.0(19), N1-Au1-H1 = 179.2(19), C9-

Au1-C18 = 161.85(14). Figure reproduced from reference.141

In 2018 Takano et al. observed an isolable hydride-doped gold superatom characterised by

142 3+143 MS and NMR spectroscopy. The reaction of [Au9(PPh3)8] with 1 mol eq. of NaBH4 was monitored in situ by ESI-MS in the positive ion mode. After the addition of NaBH4 the mass

3+ spectra showed that the peak for [Au9(PPh3)] disappeared almost completely and that a new

2+ dominant peak was observed assigned as [Au9H(PPh3)8] . The assignment was validated by using NaBD4. ESI-MS monitoring revealed that the hydride did not undergo exchange with

2+ 1 31 1 protic solvent, thus leading to characterisation of [Au9H(PPh3)8] via H and P{ H} NMR. The

1 2+ H of [Au9H(PPh3)8] gave a multiplet cantered at 15.1 ppm assigned to the hydride. This was

31 1 2+ confirmed when the peak collapsed to a singlet for the P{ H} NMR. The structure of [Au9H] core was probed using DFT calculations and substituting PPh3 for PMe3 ligands to spare

3+144 computational power. DFT suggests the [Au9(PMe3)8] adopts an Au9 core with a crown

3+ motif as experimentally observed for [Au9(PPh3)8] , see Figure 1.22 (a). The DFT optimised

2+ structure of [Au9H(PMe3)8] shows that the hydride is directly bonded to the centre gold atom with an Au-H length of 1.69 Å, see Figure 1.22 (b).

3+ + Figure 1.22 DFT optimised structure and energy diagrams of: (a) [Au9(PMe3)8] and (b) [Au9H(PMe3)8] .

Yellow, blue, and white balls represent Au, P, and H atoms, respectively. Methyl groups are depicted as sticks. Kohn−Sham orbitals are depicted with isodensity values at 0.025e levels. Figure reproduced from reference.142

In 2016 Kleinhans et al. reported T-shaped Au(I) complexes that upon protonation by TfOH or

145 CF3CO2H in THF yielded the cationic Au(III)-hydride. Structural characterisation by X-ray diffraction, see Figure 1.23, reveal Au-N distance of 2.051(2) Å. The hydride appears in the 1H

NMR at δ -8.34 ppm. FT-IR report the v(Au-H) absorption at 2197 cm-1.

36 Figure 1.23 Solid-state structure of a cationic Au(III)-hydride. Counter-anion and hydrogen atoms are

omitted for clarity. Figure reproduced from reference.145

1.1.3.4 Heterometallic coinage metal hydrides

Recently, in 2017, there have been reports of heterobimetallic coinage metal complexes.

Hicken et al. reported a series of heterobimetallic complexes comprising a three-centre two- electron [Au-H-Cu]+ core.146 The reaction of fluorinated copper(I) β-diketiminate complexes

with [AuH(IPr)] in toluene afforded various heterobimetallic complexes structurally

characterised by X-ray crystallography, see Figure 1.24..

(a) (b) (c)

Figure 1.24 X-ray structures heterobimetallic coinage metal clusters highlighting key bond lengths and

angles: (a) [{{ArNC(CR3)}2CH}-Cu-H-Au(IPr)]; (Ar = Mes, R = CF3): Au–Cu = 2.6376(5), Au–C = 2.026(3);

C-Au-Cu = 161.86(10); (b) [{{ArNC(CR3)}2CH}-Cu-H-Au(IPr)]; (Ar = C6F5, R = CH3): Au–Cu = 2.5910(6),

Au–C = 2.025(4); C-Au-Cu = 153.00(11); and (c) [((ArNC(CR3)(O(CR3))CH)(PPh3)-Cu-H-Au(IPr)]; (Ar =

C6F5, R = CH3): Au–Cu = 2.7525(7), Au-C = 2.043(8); C-Au-Cu = 153.4(4); Cu–P = 2.2401(11). Selected bond lengths [⁰] and angles [Å]. Figure reproduced from reference.146

1.1.4 Bonding and Reaction Environment

An understanding of bonding is critical in optimising the reaction environment of coinage metal hydride complexes and clusters. Metal-metal147-150 and metal-ligand bonds are contributing factors that control the types of coinage metal hydride species formed and their modes and rates of reactivity. Gray and Sadighi have published a comprehensive review on the noncovalent and covalent metal-metal interactions of the Group 11 metals.62 Metal-ligand bonds are critical since the choice of ligand can modify the electronic and steric properties at the active site. Furthermore, ligands thermodynamically control metal cluster nucleation in solution and can potentially allow the synthesis of monodisperse, atomically precise clusters.

1.1.4.1 Ligands

Ligands provide a means to control the growth of nanoclusters and their reaction environment.151 A ligand comprises at least one electron pair for coordination to a metal ion and can be as simple as the hydride ion, whose bonding modes are discussed in section

1.1.4.1.1. Larger ligands are used to synthesise atomically precise clusters, and some common types used for the preparation of coinage metal hydride complexes and clusters were given in Scheme 1.2. Since phosphines ligands are used in the work reported in this thesis, they are described below in section 1.1.4.1.1. Although a wide array of other ligands (thiols, amines, selenides, N-heterocyclic carbenes, etc.) are commonly used to prepare transition metal complexes and clusters their binding modes are not covered here.

1.1.4.1.1 Hydride

Hydrides can form metal-hydride bonds75 with coinage metals adopting various binding modes, see Figure 1.25, and forming ionic, covalent or interstitial bonds. Typical sources of hydride are: borohydride, superhydride, hydrogen (via heterolytic dissociation) etc. Hydricity

(ΔG°H-) is used to describe the free energy required for cleaving the M-H bond which when used with other thermodynamic data enables predictions on the thermodynamics of reactions of metal-hydrides involving X-H bond formation with concomitant M-H bond breaking.152-154

38 Selected literature reports for the neutral M-H bond dissociation energies of the coinage metals have been described wherein, (Au-H at 70.6 kcal/mol)155 > (Cu-H at 63 ± 3 kcal/mol)156 > (Ag-

H at 48 ± 2 kcal/mol)157 although there is debate regarding the precise values.76, 158 A simple thermochemical cycle has been used to estimate the hydride anion affinities of the monoatomic cations, D(M+-H-),159 as: (Au+ at 265 kcal/mol) > (Cu+ at 223 kcal/mol) > (Ag+ at

209 kcal/mol). For more information a detailed review describing the thermodynamic hydricity of coinage metal hydrides is given by Appel and co-workers.154 It must be noted that the predictive capabilities of hydricity has inherent limitations for more complex systems. For example, hydricity cannot readily predict the reactivity of bimetallic heteroleptic coinage metal hydrides [dppmMM’H]+ with formic acid, see Chapter 4, where competing effects exist such as the anchoring of oxygen (from formic acid or formate) to the second metal site.

Figure 1.25 Hydride binding modes for a selection of metal-hydrides.

1.1.4.1.2 Phosphines

Phosphine ligands are ideal candidates for the synthesis of coinage metal hydrides complexes/clusters as they possess a diverse range of properties and many are commercially available or can readily be synthesised.160 For coinage metal complexes in a +1 oxidation state, monodentate tertiary phosphines, PR3, typically take up one or two coordination sites at a single metal centre (see Scheme 1.6 a and b). Triphenylphosphine (PPh3) or its derivatives are commonly used since they are easier to handle than the pyrophoric trimethylphosphine

(PMe3) and triethylphosphine (PEt3). Bidentate bisphosphine ligands are popular ligands since they offer two potential bonding sites and thus exhibit richer binding modes as shown in

Scheme 1.6 c-h. The most commonly used bisphosphine ligands are the bis(diphenylphosphino)alkanes with methylene linker sizes ranging from n = 1 to 6.

Coordination complexes of silver(I) with tertiary phosphine and related ligands have been reviewed.161

Scheme 1.6 A selection of coordination modes for: monodentate tertiary phosphines (a) and (b), and; bisphosphine ligand (c)-(h).

Figure 1.26 A graphic representation of the cone angle to describe the steric effects of a phosphine ligand coordinated to a metal centre.

Systematic changes to the steric and electronic effects of phosphines are useful in optimising their desired properties, for example reactivity, selectivity or stability. The steric effects of the

R groups of tertiary phosphines, PR3, are typically described by the Tolman cone angle, see

Figure 1.26 (a).77 The electronic effect can be adjusted by modifying the R groups to adjust the donor/acceptor strength of the ligand, for example by having electron withdrawing or electron donating groups. Similarly, bidentate ligands can be sterically described by their

162 natural bite angle (βn) first described by Casey et al, see Figure 1.26 (b). The bite angle

effect,163 small bite angle,164 and rational design165 of bisphosphine ligands for use with metal catalysts have been reviewed. The ligand database continues to expand for monodentate166 and bidentate167 phosphine ligands.

1.1.4.2 Entatic Effect

The entatic effect is defined as “a state of an atom or group, which due to its binding in a protein, has its geometric or electronic condition adapted for function”.168 Valee and Williams first coined the term when exploring the link between the modification of specific amino acid side chains around the active site of metalloenzymes and the resultant effect on its chemical and catalytic properties.169 The modified amino acid side chains facilitate a state of entasis whereby the active site is “energised” bringing the energy of the enzyme-substrate complex closer to that of the unimolecular transition state. This allows for increased reactivity by lowering the activation energy relative to an energised enzyme-substrate complex.

Comba et al. have subsequently used entasis to describe the reactivity of coordination compounds.170-171 Stable compounds are said to fit well together. Structural distortion(s) of bond distances and coordination geometries resulting in deviations from the lowest energy state can be described as misfit within a metal complex. This leads to thermodynamic and kinetic instability whereby the misfit is higher in energy than other more stable geometries.

Achieving these distortions can be implemented by ligand choices/design, e.g. rigid ligands, that are capable of pre-organising the metal centre environment to accommodate reactivity.

Figure 1.27 shows a simplified energy diagram highlighting the concept of the entatic effect for a single-step catalysed reaction between a metal-ligand complex and a substrate which is transformed into a product. The reaction energy is plotted against the reaction coordinate to describe the entatic effect from an energetic standpoint. The substrate approaches the metal- ligand complex (L) and binds to form a loosely bound complex. An activation energy barrier

(Ea) must be overcome for the reaction to transform the reactant into a product (P). The reaction rate can be further increased when an entatic effect is observed. The red dotted line represents the reaction when the entatic effect is at play. A metal-ligand complex may be

41 “energised” by adapting the reaction environment around the metal centre. This may be achieved by pre-organising the metal centre adopting energetic “strain” by using rigid ligands which deviate from the lowest energy geometric arrangements around the metal centre.

Figure 1.27 A generic representation of a single-step metal-catalysed reaction whereby a metal-ligand complex (L) bound to a reactant (R) is transformed into a product (P). The plot shows the reaction coordinate (x-axis) vs the energy (y-axis) for the reaction without entatic effect (red dotted line) compared to the reaction with an entatic effect. Ea = activation energy, Ea(e) = activation energy with entatic effect,

L = metal-ligand complex, Le = metal-ligand complex with entatic effect, R = reactant, P = product; Red dotted line = energy profile, Black solid line = energy profile.

A nice example of how the entatic effect can have a profound effect on coinage metal reactivity

+ involves mononuclear phosphine gold(I) complexes such as Au(PPh3)2 , which prefer a linear

2-coordinate geometry and are unreactive towards aryl iodides. When both phosphines are replaced by a carborane diphosphine (DPCb) ligand, the resultant complex is now reactive.172

This is achieved by pre-organising the metal centre environment from the stable gold(I) geometry where P-Au-P would be 180⁰ (no reaction is observed at up to 120⁰C when PPh3 is

42 used), to a geometry that is energised by the use of DPCb ligands to give a P-Au-P angle of ca. 90⁰, thereby inducing C-I bond activation in aryl iodides, see Figure 1.28. The entatic effect was used in this thesis to design a silver hydride catalyst as discussed in Chapter 3.

+ Figure 1.28 Representative energy profile for the reaction of iodobenzene with: (a) [Au(PR3)2] , and (b)

[(DPCb)Au]+.

43 1.1.5 Reactivity

This section will focus on key papers highlighting the chemical reactivity of coinage metal hydrides in the transformation of organic substrates. A large proportion of scientific reports on the reactivity of coinage metal hydrides are skewed toward copper hydrides. This is likely due to the elusive nature of silver and gold hydrides, which eluded isolation until the past decade.56,

127 This contrast with reports on the isolation of copper hydride compounds which date back as early as 1844.10 A brief overview of selected organic transformations mediated by coinage metal hydrides will be presented herein, for an in-depth review see Jordan et al.42

In 2016 Cook et al. reported on the synthesis and structural characterisation of

[Cu3(H)L3(OAc)2] and [Ag6(H)4L4(OAc)2], see Figure 1.3 (a) and Figure 1.16 respectively.

Given the precedent of coinage metal hydrides as intermediates in the hydrosilylation of ketones the authors screened reactions for each of [Cu3(H)L3(OAc)2], [Ag6(H)4L4(OAc)2] and

- Stryker’s reagent (0.05 H eq.) in the presence of 1.5 eq. Ph2SiH2 to 2-cyclohexen-1-one, see

Scheme 1.7. Reaction with [Cu3(H)L3(OAc)2] was found to yield 79% total conversion with 76% conversion to the silyl enol (A) and 3% conversion to the 1,2-hydrosilylation product (C) at 24 h. The reaction with [Ag6(H)4L4(OAc)2] yields 96% total conversion with 57% to A and 10% to

C at 24 h. Stryker’s reagent however yields >99% conversion in <0.25 h with 79% conversion to the silyl enol.

Scheme 1.7 1,4-hydrosilylation reaction of 2-cyclohexen-1-one by [Cu3(H)L3(OAc)2], [Ag6(H)4L4(OAc)2] and [CuH(PPh3)]6.

44 1.1.5.1 Copper hydrides

Copper hydrides have become versatile reagents in the organic chemist “toolkit” for organic synthesis, with a focus on reductive transformations of organic substrates.173-174 In the past decade, reports expanding the range of functional groups amenable to reduction have emerged together with improvements in the selectivity of known transformations. The hydrogenation175 and hydrosilylation176 reactions of copper hydrides with carbonyl compounds were first reported by Stryker and Lipshutz respectively. For a summary of selected reactions highlighting organic transformations mediated by copper hydrides see Scheme 1.8, Jordan et al. provides a comprehensive review which covers the literature before 2016.

Scheme 1.8 Selected reactions of copper hydride mediated organic transformations for: (a) reduction of ketones177 (b) enantisoelective reduction of ketones178-180 (c) reduction of alkenyl ketones181 (d) reduction of enones182 (e) reduction of akynes183-187 (f) reductive aldol addition/lactonisation138, 176 (g) anti-

Markovnikov hydroamination of terminal alkenes;188 and (h) enantioselective reductive aldol reaction.189

For a complete summary of the literature before ca. 2016 refer to the review by Jordan et al.42

1.1.5.2 Silver hydrides

Compared to copper hydride, reports on silver hydride as intermediates in organic transformations have been rare. The catalytic hydrogenation of aldehydes, see Scheme 1.9 a, in water is predicted to occur through the formation of a silver hydride intermediate, generated by the heterolytic activation of H2 with silver salts which allows insertion of the aldehyde to form silver alkoxide.190 The terminal alcohol is then released by heterolytic reaction with H2, regenerating silver hydride. Similar reactions using sodium formate as a hydride source have been reported.191 The catalytic hydrogenation of alkynes by a heterobimetallic silver-ruthenium catalyst was proposed to occur via formation of a silver hydride which drives isomerisation, however no experimental supporting evidence was provided.192

Scheme 1.9 Silver hydride mediated organic transformations for: (a) hydrogenation of aldehydes in water

(b) hydrogenation of alkynes.

Silver hydrides have been revealed as reactive intermediates in the active sites of zeolites.

Baba et al. discovered that small silver clusters supported on porous aluminosilicates form silver hydride upon activation with hydrogen or methane.193-202 The presence of silver hydride is confirmed by 1H magic angle spinning (MAS) NMR at a chemical shift of -1.8 ppm.201

Scheme 1.10 shows the catalytic production of propene by reaction of methane in the presence of ethene over silver(I)-exchanged zeolite wherein silver hydride is formed by activation with methane.203

46 Scheme 1.10 Catalytic production of propene by reaction of methane in the presence of ethene over

- + 203 silver(I)-exchanged zeolite (ZO Agn ).

1.1.5.3 Gold hydrides

Few reports have provided unequivocal evidence of gold hydrides as intermediates in organic transformations. It is predicted however that Au(I) and Au(III) complexes give rise to gold hydride intermediates via hydrogenation pre-catalysts. Computational studies204 suggest that gold hydride, formed by heterolytic cleavage of H2, is a key intermediate in the hydrogenation of alkenes and imines, see Scheme 1.11 a.205 Hydrogenation of diethyl itaconate by Schiff base Au(III) complexes propose gold(III) hydride, see Scheme 1.11 b, as a key intermediate is supported by computational studies although there is no experimental evidence.206 In 2008

Tsui et al. reported on the reactions of a stable monomeric gold(I) hydride complex with alkynes.56 This inspired the work of Pintus et al. who in 2016 reported on the stereo- and regioselective alkyne hydroauration with a gold(III) hydride, see Scheme 1.11 c.141 In 2009

Hajime et al. reported a gold(I) hydride species as a key intermediate in the gold(I)-catalysed dehydrogenative alcohol silylation, see Scheme 1.12.207 A stable gold(I) hydride intermediate was generated by the σ-bond metathesis of the gold(I) chloride species with R3SiH yielding

R3SiCl used for the silylation of alcohols (R’OH), see Scheme 1.12. Chlorination of the hydride intermediate regenerated the gold(I) chloride species and resulted in the evolution of H2, closing the catalytic cycle.

Scheme 1.11 Gold hydride mediated organic transformations for: (a) enantioselective hydrogenation (b) alkene hydrogenation (c) hydrogenation of alkynes.

Scheme 1.12 Proposed Mechanism for gold(I)-catalysed dehydrogenative silylation. Scheme reproduced from reference.207

48 1.1.6 Hydrogen storage

The development of viable alternatives to fossil fuels is paramount for two main reasons: (i) sources are finite; and (ii) combustion generates greenhouse gases such as CO2 that further damage our atmosphere. A much cleaner fuel is hydrogen, which only produces water upon combustion. Unfortunately, unlike liquid alkane fuels, hydrogen is a gas and thus presents challenges for storage in transportation applications. Chemical and physical solutions for hydrogen storage present opportunities for development of a sustainable alternative to liquid alkane fuels.208 Metal hydrides are ideal candidates because of their safe operating pressures and reduced volume compared to pressurised hydrogen gas.209-213 Dihydridometallates also have the capacity for hydrogen storage applications.214 One potential issue with stoichiometric metal hydrides is that the mass of the metal exceeds that of the H atom(s) so that many metal hydrides fail the first of the five hydrogen storage “commandments” of a high storage capacity

(minimum 6.5 wt. %). Thus, there has been growing interest in carbon-based hydrogen storage molecules from which H2 can be released via the use of a (metal) catalyst. An ideal candidate chemical storage of hydrogen formic acid.179, 215-222

Key to the use of formic acid is its selective catalysed decomposition, since in the absence of a catalyst, formic acid decomposes at high temperatures via two pathways: (i) dehydration, and (ii) decarboxylation. A further complication is equilibrium between these two decomposition pathways is known as the water-gas shift reaction, see reaction 3 of Scheme

1.13.223

Scheme 1.13 The decomposition pathways of formic acid.

49 Base metals have been shown to catalyze both dehydration and decarboxylation either directly or indirectly.224-227 In contrast, the noble metals (surfaces,216, 228-230 clusters,231-232 complexes179, 220, 233-236 and alloys237-239) are known for the selective decomposition of formic acid by dehydrogenation to generate hydrogen.216 Carbon dioxide and formic acid together may be a potential combination for the production of environmentally friendly fuels.240 The selective decarboxylation of formic acid has been reported for complexes/clusters copper, silver and gold.241 In 2017 Kumar et al. reported on a highly reactive (N^C^C)gold(III)-hydride complex arising from the β-hydride elimination of its formate precursor, see Scheme 1.14.242

The (N^C^C)gold(III)-hydride is predicted to be highly reactive and undergoes rapid dehydrogenation of formic acid to regenerate the formate.

Scheme 1.14 Proposed catalytic cycle for the dehydrogenation of formic acid. Structures and DH/DG energies (kcalmol@1) of the catalytic cycle computed at the M06/6-31G**(SDD) level of theory, including solvent effects (IEFPCM, solvent=formic acid). Scheme reproduced from reference.242

1.1.7 CO2 Mitigation

A steady rise in the use of fossil fuels has led to increased amounts of carbon dioxide in the atmosphere.243 Carbon dioxide plays a critical role as a greenhouse gas and there is growing interest in mitigating emissions to prevent and/or reverse climate change.244-245 Methods to use carbon dioxide as a chemical feedstock in chemical industry have been investigated as a

246-249 feasible means of CO2 mitigation. A viable strategy for CO2 mitigation may be its reduction into formic acid derivatives.250-256 A critical step is the addition of the hydride anion to CO2 to yield a metal formate.

Scheme 1.15 Reactions of selected coinage metal hydrides with carbon dioxide.

In 2013 Tate et al. described the reaction {[(SIDipp)Ag]2(µ-H)}BF4, see Figure 1.14, in

13 deuterated dichloromethane (CD2Cl2), with an atmosphere of C-labeled CO2, see Scheme

1.15 a.118 The reaction was monitored by 13C NMR wherein the evolution of a doublet resonance at δ 167.9 ppm was consistent with a formate bridging “Ag2(O2CH)” core. After four days, 5% conversion of CO2 to formate was achieved however decomposition of the silver hydride salt was reported. In 2017 Nakamae et al. described a similar reaction for the fixation

51 of carbon dioxide. Here, the authors reacted [Cu8H6dppm5](PF6)2, see 1.1.3.1.8, in acetonitrile at room temperate with an atmosphere of CO2 for 15 minutes. From this reaction solution was isolated as [Cu3H(dppm)3(O2CH)]PF6 in 49% yield. Furthermore, both [Cu8H6dppm5](PF6)2

[Cu3H(dppm)3(O2CH)]PF6 were found to be active precatalysts for the hydrosilylation of CO2 with a series of tertiary hydrosilanes (Me2PhSiH, MePh2SiH and Ph3SiH) to yield R3SiO2CH.

1.1.8 Links between Nanoclusters and Nanostructured Materials

Nanoparticles exist in a variety of morphologies including: nanotubes,257-260 nanorods,261-264 nanowires,265-267 nanosheets,268-270 etc. Both nanoparticles and metal clusters have emerged as materials with practical applications in catalysis, quantum computing, sensing, imaging etc.271-274 The two represent a bridge between the solid and molecular states, however the distinction between clusters and nanoparticles remains somewhat ambiguous, with the former considered as a “monodisperse” species defined by chemical composition and structure and the latter being particles of less precise characterisation, often consisting of a size distribution

(i.e. “polydisperse”).275 It should be noted that the term cluster was first introduced by Boyle in

1661.276 and has since been used by chemists,277 physicists278 and mass spectrometrists.279

Various definitions of a cluster exist in the literature,280-282 here we will define a cluster as having metal atoms (M) for Mn where n ≥ 2 (when n is small, e.g. 2-5 the literature may refer to these clusters as complexes). More work is needed at understanding how atomically precise clusters and nanoparticles can be synthesised. Indeed, the reproducibility of nanoparticle synthesis requires an in-depth understanding of the growth and reaction dynamics that ultimately define the synthesis of atomically precise particles.283 Understanding the link between atoms, clusters/complexes and nanoparticles is crucial in striving towards atomic precision and reproducibility.284-289

In the subsequent sections the following links between clusters and nanoparticles are briefly explored: (1) the relationship between the morphology of clusters and nanoparticles; (2) clusters as “building blocks” or precursors to nanoparticles; (3) decomposition of nanoparticles into nanoclusters.

52 1.1.8.1 Morphology of clusters and nanoparticles

Clusters and nanoparticles can have similar morphologies. In 2015 Dhayal et al. reported on the synthesis of rhombus-shaped copper nanoparticles from the reaction of

i 107 [Cu32(H)20{S2P(O Pr)2}12] with excess borohydride. The cluster exhibits a hexacapped pseudo-rhombohedral core of 14 copper atoms which exhibits structural similarities to rhombus-shaped nanoparticles, see Figure 1.29..

(a) (b)

Figure 1.29 Cluster to nanoparticle transformation highlighting the similarities in their morphology

i between (a) the rhombohedral core of 14 copper atoms (in purple) for [Cu32(H)20{S2P(OPr)2}12], and (b)

i the rhombus shaped nanoparticles yielded from the reaction of [Cu32(H)20{S2P(OPr)2}12] with excess borohydride. Scheme adapted from reference.107

1.1.8.2 Clusters as “building blocks” or precursors to nanoparticles.

In 2018, Li et al. reported the first and smallest triangular bifrustum copper nanoclusters, see

Figure 1.9. Bifrustrum-type structures might act as seeds for the synthesis of larger nanoparticles.290-291 Therefore, there is potential for these novel nanocluster structures to find applications as a “stepping stone” toward the synthesis of novel nanoparticles.

Indeed, there are examples of clusters being transformed to nanoparticles. Thus in 2013 Liu et al. reported on the synthesis of monodisperse silver nanoparticles, with an average size of

30 nm.126 The synthesis involved the reaction of the discrete heptanuclear silver hydride cluster, [Ag7(μ4-H){S2P(OEt)2}6] (see section 1.1.3.2.7 for its structure), with excess sodium borohydride. Monitoring of the reaction via 31P NMR suggested the presence of free ligand is

292 likely due to the formation of the known Ag7L4 kernel. The authors propose that further rapid reductions of the Ag7L4 kernels leads to formation of the silver nanoparticles (see Scheme

1.16). This work highlights the possibility of isolating monodisperse nanoparticles from atomically precise nanoclusters, however opportunities to determine the exact mechanistic pathways remain elusive and further development is necessary. Similarly,

i [Cu32(H)20{S2P(O Pr)2}12], see Figure 1.11, was synthesised from the reaction of

i [Cu20(H)11{S2P(O Pr)2}9], see Figure 1.10, with excess borohydride.

Scheme 1.16 Formation of silver nanoparticles from a silver hydride nanocluster.

1.1.8.3 Transformation of nanoparticles into nanoclusters

Nanoparticles can change their composition as a function of time and additives. Aging in solution tends to broaden the size dispersion and increase the average size (Ostwald ripening), but additives can have the opposite effect to produce a narrower size dispersion and decrease in average size via etching processes.293

The complex reaction events in solution are nicely demonstrated in the 2018 paper of Suber et al. The authors described the synthesis of Ag35(dodecanethiolate)16 from the reaction of the

Ag-dodecanethiolate complex in dichloromethane with triethylamine and excess sodium borohydride.294 It was reported that stopping the reaction 2 hours after the addition of sodium

54 borohydride leads to the formation of the Ag35 nanoparticle. However, when the sodium borohydride reaction time was extended to 4 hours Ag15(dodecanethiolate)11 was isolated.

This demonstrates the importance of reaction time, and conditions in general, in selectively obtaining atomically precise clusters. Similar nanoparticle-to-cluster transformations were observed in 2013 by Qu et al. The authors described the spontaneous transformation of polyethyleneimine protected silver nanoparticles to nanoclusters.295

1.2 Characterisation Methods Used in this Thesis

A range of characterisation methods can be suitable in examining the structure and reactivity of coinage metal hydrides. This section explores key analytical methods that have been used throughout this thesis and gives a brief understanding of the underlying principles, modes of action and suitability.

1.2.1 NMR Spectroscopy

Nuclear magnetic resonance (NMR) was first discovered in 1946 by Edward Purcell, Robert

Pound and Henry Torrey.296 In 1952 Edward Mills Purcell and Felix Bloch were awarded the

Nobel Prize in physics for their development of NMR spectroscopy. In brief, the principle of

NMR spectroscopy relies on spin active nuclei and provides valuable information on the structural and electronic properties regarding the connectivity of atoms within a molecule.117,

297-300 Of particular interest is NMR for Hydride301 and Silver117 nuclei.

Compared to MS, larger amounts of sample are needed, however NMR spectroscopy allows for the non-destructive analysis of a sample. Often, and mostly serendipitously, when a sample is left in an NMR tube over time crystalline material may develop suitable for crystallographic studies.

55 1.2.2 Crystallography

Crystallisation of a coinage metal hydrides allows structural characterisation via X-ray crystallography and/or Neutron diffraction. These methods are dependent on obtaining suitable, and often stable, crystalline material for analysis. Many different methods for obtaining suitable crystals for analysis appear in the literature.302-303 Once a crystal is obtained, containing highly ordered and repetitive units of the target molecule, the sample is treated with a suitable wavelength of either X-ray or neutron beams for analysis. The diffraction patterns are measured and mathematically translated into meaningful data that describes the molecule in 3D.304 Accurate and precise structural information on hydrides is crucially important to understanding the reactivity of coinage metal hydrides. The following sections briefly discuss and compare x-ray crystallography and neutron diffraction for locating hydrides, for further reading see: Bragg’s paper304 for a historical context on coherent and incoherent scattering from a crystal lattice, and; Bacon and Haar305 for a modern introduction to X-ray and neutron diffraction.

1.2.2.1 X-ray diffraction

Wilhelm Röntgen first discovered X-rays in 1895 which would lead to an analytical method for determination the spatial arrangement of individual atoms in crystalline material.306 Almost two decades later, Max von Laue proposed that X-rays could be used on crystalline material to obtain a diffraction pattern relating to the structure of the material.307-309 The suitability of the method is a direct result of the comparable order of magnitude shared by the wavelength of

X-rays and the smallest interatomic distances. X-ray scattering occurs as a result of the electron density of the atoms, which can be beneficial when requiring to differentiate atoms based on their varying electron densities.

X-ray crystallography works exceptionally well for heavier elements since they have greater electron density. It is well known that X-ray data provides accurate and precise structural information, however, in certain molecules hydrides often prove elusive because they have the least electron density of all ions. The issue is compounded when the hydride is near

56 heavier elements, although X-ray data can be used to locate terminal hydrides.310 Methods are currently being developed to improve X-ray crystallography and its ability to precisely and accurately locate hydrides.311 Furthermore, computational methods can use the spatial arrangement of the heavier elements with confidence to save computational time when theoretically proposing the hydride location(s).

1.2.2.2 Neutron diffraction

Although neutron diffraction was not used in this thesis a brief summary is provided considering its importance in unequivocally locating hydride atoms. In 1920 Ernst Rutherford first proposed the existence of neutrons.312 Twelve years later, James Chadwick discovered the neutron after a series of experiments.313 Wollan and Shull used neutron diffraction to determine the structure of ice, sodium hydride and sodium deuteride.314-318

Neutron diffraction is particularly valuable in locating elusive hydrides.319 Their precise location is achieved because neutron diffraction utilises a neutron beam, generated from a nuclear reactor, that interacts with the nucleus of the molecule, rather than the electron density.

However, compared to crystals suitable for X-ray diffraction those for neutron diffraction must be appreciably larger and are commonly approximately 1 mm3 in size.320

1.2.3 Density Functional Theory

Density functional theory (DFT) is used to examine the structures and energetics of a reaction and is useful in complementing experimental data.321 The basis of DFT arises from the two

Hohenberg-Kohn theorems in solving the many-body Schrodinger equation for the compound of interest.322 For heavier elements effective core potentials (ECP) are used in order to reduce the computational demand eliminating the need for core basis functions. These pseudopotentials may be used to describe relativistic effects323-324 using relativistic effective core potentials (RECP). The relativistic effects of silver hydride325 and gold hydride326 have been examined.

57 For catalytic reactions a potential energy surface (PES) allows examination of a series of local minima and corresponding transition states (TS). The overall exothermicity or endothermicity of the reaction is based on the difference in total energy between the total energy of the reactants and the total energy of the products. The activation energy for each transition state can also be determined to provide information regarding the activation barriers for corresponding transition states.

DFT is particularly useful in examining the optical properties of compounds. Time-dependant density functional theory (TDDFT) is used for the prediction of the absorption and emission spectra.327

1.3 Mass spectrometry

The study of ions based on their mass-to charge ratio (m/z) is the fundamental principle of

MS.2, 328-329 All modern mass spectrometers consist of four components, see Figure 1.30, each with their specific role in generating mass spectra from given sample. The ionisation source is used to generate and/or transfer ions into the gas-phase. The ions are then separated according to their m/z by the mass analyser. An ion detector measures the m/z of the ions generating usable signal which is transformed by a computer into a mass spectrum, typically as a 2D graphical representation of the ions m/z (x-axis) vs their abundance (y-axis). The computer can: (i) control the instrumentation, (ii) acquire and process data, and (iii) facilitate analysis of experimental results, via for example database searching. Although MS typically cannot recover analysed sample (with the exception of ion soft landing MS330) when compared to other non-destructive analytical methods, such as NMR and IR spectroscopy which can require milligrams of analyte, MS allows for much greater sensitivity and can perform well when reaction products are scarce.

The next sections will focus on describing the underlying principles of the most relevant methods as used in this thesis.

58 Figure 1.30 The basic components of a mass spectrometer.

1.3.1 Ionisation Methods

Mass spectrometric analysis of a sample requires that it be introduced into the ionisation source, where gas phase ions are formed, which can then be extracted via electrostatic lenses or ion funnels into the mass analyser for mass analysis. Over the years, a range of ionisation methods have been developed that use different physical and chemical processes to generate ions.328 The most commonly used ionisation methods are listed in Table 1.1, which highlights their suitability in the study of inorganic/organometallic compounds.

59 Table 1.1 Various ionisation methods.

Suitability for Energy transfer Typical Mass Inorganic / Ionisation Method (soft/hard Refs Range for Analytes Organometallic ionisation) Ions

Electron ionisation (EI) High (hard) <1000 Da LIMITED 331-332

Chemical Ionisation (CI) Low (soft) <1000 Da LIMITED 333

Fast Atom Bombardment Low (soft) 200-2000 m/z YES 334 (FAB)

Matrix-assisted Laser Low (soft) <100000 m/z YES 335-337 Desorption Ionisation (MALDI)

Electrospray Ionisation Low (soft) High mass range YES 338-342 (ESI) due to multiply charged ions. Mostly < 2000 m/z

Electron ionisation (EI), one of the first ionisation methods developed, is widely used in the analysis of volatile organic compounds.343 It involves bombarding a gaseous analyte, M, with

70eV electrons, which produces radical cations, M+●, also known as a molecular ion. Since the ionisation process imparts excessive energy on the analyte, fragment ions are formed from

M+●. The resultant mass spectrum is thus a combination of molecular ions and fragment ions and provides a unique “fingerprint” for a compound. Although this fingerprint cannot be predicted a priori for a given compound, analytes in mixtures can often be identified by comparing their EI mass spectrum with those in databases. Since EI is a harsh ionisation process, in some instances no molecular ion is observed. Thus, chemical ionisation (CI) was later developed as a softer ionisation technique with less fragmentation. Again, the analyte is volatilised, but now instead of being bombarded with electrons, it undergoes ionisation via ion- molecule reactions with reagent ions formed from reagent gases. However, its application to the study of inorganic/organometallic compounds is still limited in that the compounds must be relatively volatile. Fast atom bombardment (FAB) revolutionised the field of MS, since it does not require heating a sample to get it into the gas-phase. Instead, the analyte is embedded in a viscous liquid matrix (typically glycerol) and then bombarded with a beam of fast inert gas

atoms (argon or neon) or a beam of ions (Cs+ ions; liquid secondary ion MS344). The analyte is both desorbed as well as ionised during the FAB process and is suitable for thermally unstable compounds. FAB-MS was widely used for the analysis of inorganic and organometallic compounds.345-346 Drawbacks from FAB include the high chemical background noise and the fact that the selection of the ideal matrix was often sample dependent. Matrix- assisted laser desorption ionisation (MALDI) provides a soft ionisation and is ideal for large molecular weight ions, however it is mostly limited to TOF instruments and can produce high chemical background noise at low m/z values. Nonetheless, it has found application in the analysis of inorganic and organometallic compounds.347-348 Softer techniques, like electrospray ionisation (ESI) allow direct transfer of ions from solution to the gas-phase or via processes involving desorption or evaporation. The suitability of ESI-MS for the study of metal complexes and clusters in solution has been well established. For example, Hudgens et al. have extensively studied gold nanoclusters using ESI-MS to explore properties such as ligand exchange reactions, the formation of monodisperse clusters with varied reaction conditions etc.48, 349-357 Since this ESI is widely employed in the work described in this thesis, it will be described in more detail in the next section.

1.3.1.1 Electrospray ionisation (ESI)

Although electrospray has been known to transfer ions from solution to the gas-phase based on the pioneering work of Dole, a key break through was the development of an electrospray ionisation source that could be coupled to a mass spectrometer to generate mass spectra.

Thus in 1984, Yamashita and Fenn described the in vacuo production of positively and negatively charged cluster ions from solutions comprising various solvents and metal salts.338,

342 ESI is considered a soft ionisation method that preserves structures of ions as they are transferred to the gas-phase. The experiments of Fenn et al. preserved the structure of ions generated from thermally labile species and allowed large molecules to be captured within narrower m/z isolation windows due to the observation of multiply charged ions. ESI-MS has also proved to be well suited to study inorganic and organometallic compounds.358-360

The ionisation process of ESI begins with the sample containing the analyte dissolved in a suitable solvent being passed through a capillary with a weak flux (ca. 1-10 µL min-1) and is nebulised at atmospheric pressure, as it exits the charged ESI needle, into a fine aerosol of highly charged droplets, see Figure 1.31. The solvent contained in the highly charged microdroplets is evaporated by a controlled flow of inert gas (typically nitrogen). Heat is also used to evaporate the solvent. As evaporation of the solvent increases the charge density to a point where ion repulsion is at the same order as the surface tension a “Coulomb explosion” can occur to produce many smaller “daughter” microdroplets which can cascade until bare analyte ions result. The polarity can be switched between positive or negative ions. The ions are then focussed and transmitted by the ion optics region (not described) toward the mass analyser.

Figure 1.31 ESI in the positive ion mode.

The performance of ESI can be affected by: liquid surface tension, droplet size, solvent volatility, ion solvation strength and surface charge. Performance often decreases for high surface tension, large droplets having poor volatility, showing strong ion solvation and/or low surface charge. ESI allows for excellent detection limits with relatively low background noise.

In-source fragmentation can be controlled by adjusting the cone voltage. It is highly suitable for charged, polar or basic compounds. It is also compatible with MS/MS methods wherein mass analysers comprising ion trap mass spectrometers are a popular choice.

1.3.2 Mass Analysers

The “heart” of the mass spectrometer is the mass analyser(s), which separates out ions based on their mass to charge (m/z) ratios prior to being “counted” by the ion detector. Mass analysers can also be used to mass select an ion of a specific m/z range from a mixture of ions to “purify” it for subsequent MS/MS studies using CID, IMRs, PD etc. A range of mass analysers have been developed and each of them uses different approaches to separate ions resulting in different performance characteristics.

Notable characteristics of a mass analyser relevant to its performance are the: (i) mass resolving power, (ii) mass accuracy, (iii) mass range, (iv) linear dynamic range, and (v) precision.361 The mass resolving power describes the ability of the analyser to produce distinct signals for two ions. Mass accuracy is an indication of how close the m/z produced value produced by the analyser is to the theoretically calculated m/z and is often expressed in ppm.

The mass range discloses the limits of the m/z for which the analyser can measure ions, typically described by the upper mass limit. The linear dynamic range describes the relationship over which the ion signal is linear with the concentration of analyte. The precision describes the reproducibility with which ion abundances can be determined. Table 1.2 provides a summary to compare the performance characteristics of TOF, LIT and FTICR mass analysers.

Table 1.2 A comparison of selected performance characteristics of various mass analysers.361

TOF LIT FTICR

Mass resolving power 103 - 104 103 - 104 104 - 106

Mass accuracy (ppm) 5 - 50 50 - 100 1 - 5

Mass range >105 1.5x105 >104

Linear dynamic range 102 - 106 102 - 105 102 - 105

Precision 0.1 - 1.0% 0.2 - 5.0% 0.3 - 5.0%

m/z (resonance m/z (resonance Principle of Separation Velocity frequency) frequency)

Tandem mass spectrometry (MS/MS) refers to any method that involves two or more stages of mass analysis. Each stage is either in conjunction with a fragmentation process or an ion- molecule reaction that results in a change in the mass or the charge of the mass selected precursor ion.362-364 MS/MS can occur in two distinct ways: (i) MS/MS in space, achieved by coupling two mass analysers, or (ii) MS/MS in time, achieved by an ion storage component.

A typical MS/MS in space instrument has two mass analysers, for example a TOF/TOF mass spectrometer. Achieving MSn spectra where n > 2, n analysers must be combined. This results in increased costs of coupling more than two mass analysers and practical limitations wherein ions transmitted at each subsequent step are lost leading to diminishing signals.

MS/MS in time instruments differ in the ability to program the mass analyser to carry out successive MSn stages in the same mass analyzer. This greatly increases the practical maximum stages of MSn to ca. 8 and the proportion of ions at each stage is high. Suitable mass analysers to perform MS/MS in time experiments include quadrupole ion trap (QIT) and ion cyclotron resonance (ICR) instruments.

Mass analysers can be combined to provide powerful “hybrid” mass spectrometers which take advantage of the strengths of each of the mass analysers. One such example of a hybrid mass

64 spectrometer if the LTQ-FTICR instrument used in this thesis, which combines the ruggedness, versatility and MSn capabilities of the ion trap mass spectrometer with the accurate mass measurement of an Ion Cyclotron Resonance analyser. In the next section the two mass analysers associated with that of the LTQ-FTICR hybrid instrument are discussed.

1.3.2.1 Linear Ion Trap

The linear ion trap is based on a four-rod quadrupole capped by front and back lenses that repel the ions maintained inside the rods, see Figure 1.32.365 The radial dimension confines ions by means of a quadrupolar field and the axial dimension by an electric field to create an ion trapping potential within the multipole.366 Collisional cooling occurs with an inert gas

(typically helium) once the ions are inside the trap and simultaneously oscillate in the xy plane while flying along the z axis owing to an RF only potential applied to the rods.

Figure 1.32 Linear ion trap quadrupole rod assembly.

LITs have a high ion trapping capacity, 10-fold cf. 3D ion traps (vide infra) and can contain a large number of ions before space charge effects occur. They also have a high trapping efficiency of more than 50% cf. 5% for 3D ion traps. Ions can be mass selected by adjusting parameters such as the RF voltage and applying multi-frequency resonance ejection waveforms to maintain ions with the desired m/z range and eliminating all other ions outside of the range. The ions are then stabilised by collisional cooling with the inert gas.

Subsequently, the trapped/purified ions may be maintained within the trap for further MSn experiments involving CID, IMRs, UVPD, IRMPD or the ions may be axially ejected through the back lens for example into a tandem analyser such as an FT-ICR.

1.3.2.2 Fourier-Transform Ion Cyclotron Resonance

It is well-established that trajectories of ions are curved in a magnetic field. A low ion velocity coupled with an intense magnetic field decreases the radius of the trajectory. This principle of ions being trapped in a circular trajectory within a magnetic field form the basis of the ion cyclotron or Penning trap which was first applied by Sommer et al.367-368 An electromagnetic wave that has the same frequency as an ion in the cyclotron allows resonance absorption of said wave. The ion subsequently translates absorbed energy and increases its kinetic energy causing an increase in the radius of its trajectory, see Figure 1.33 (a). Detection electrodes which are at opposite ends to each other record an “image current” of all ions circulating in the cell, see Figure 1.33 (b). Fourier transform mass spectrometry (FTMS), first described by

Comisarow and Marshall,369-370 yields a frequency spectrum which is readily converted into a mass spectrum using the known relationship between the mass and the frequency.

(a) (b)

Figure 1.33 A Penning trap cross-section showing: (a) ion excitation by the excitation electrodes (green), and (b) detection of ions by detection electrodes (red).

The Penning trap is also capable of utilising various ion activation methods. Two such methods are the use of electron beams on a mass-selected ion, called electron capture dissociation

(ECD)371-373 or by laser beams, photodissociation (PD) or infrared multiphoton dissociation

(IRMPD). The dissociation by ECD, which is most suitable to multiply charged ions, is driven

by radical ion chemistry and can provide very different results to other activation methods. This is because the electron capture by the multiply charged ion involved the fragmentation of an odd-electron cation. The non-ergodic nature of ECD does not fragment at the weakest bonds as preferential sites of fragmentation. It is at or surrounding the positive charge sites, where the electron is captured, that fragmentation occurs. Ion activation methods by laser beams will be discussed in later sections.

1.3.2.3 3D Quadrupole Ion Trap

The quadrupole ion trap374-376 comprises two metal hyperbolic electrodes arranged with the foci facing each other and a hyperbolic ring electrode located between the entrance and exit endcap electrodes, see Figure 1.34.

Figure 1.34 Cross-sectional view of the Thermo Finnigan quadrupole ion trap mass analyser.

A small hole in the centre of each endcap electrodes permits passage of ions into or out of the mass analyser. Helium damping gas enters via a nipple on the exit endcap electrode acting to

67 dampen the motion of the trapped ions and increases instrument resolution.377-378 AC voltages can be applied to the endcap and ring electrodes to fragment, eject and trap ions based on their m/z. The 3D quadrupole ion trap is highly suited for fundamental gas-phase studies of chemical reactions, particularly metal mediated chemistry.379 These instruments can be modified to investigate ion-molecule reactions (see section 1.3.3.2 Ion-Molecule Reactions).

Ions are trapped at an effective ion temperature of 310 ± 20 K, due to the helium bah gas, making them suitable for gas phase reactions under near thermal conditions.380-381 Kinetic and thermodynamic data for fragmentation/dissociation pathways of ions may also be examined.382-383

1.3.3 Gas-phase Chemistry

The gas-phase reactions of ions within a mass spectrometer allow for identification of the relationship between the parent and product ions.364 Coinage metal hydride ions from an analyte are identified and signify the species in the condensed phase when soft ionisation techniques are used, Scheme 1.17(a). After the generation of gas-phase ions a suite of techniques may be available to investigate their structure and reactivity, Scheme 1.17(b).

Mass selected ions may undergo: (i) unimolecular dissociation wherein fragmentation (e.g. via

CID) provides structural information regarding the governing processes such as ligand loss, core fission or ligand activation. Subsequently, new ions can be generated for further fragmentation (ii) bimolecular processes to investigate the reactivity of the ion (IMR), and (iii) ion-photon interactions to investigate allows for the gas-phase spectroscopy of the ion for example IRMPD384-389 and UVPD390-392. There are many ion interactions that provide useful information regarding the gas-phase chemistry of ions,393 the following will focus on key methods applied throughout this thesis.

68 Scheme 1.17 Overview of the gas-phase chemistry of coinage metal hydrides: (a) condensed phase coinage metal hydrides identification, and (b) gas-phase chemistry after ionisation.

1.3.3.1 Collision-Induced Dissociation

Collision induced dissociation (CID) assists in determining the structure of an ion since soft ionisation techniques, like ESI, transfer intact molecular ions and thus tend to produce few fragment ions. In the case of ion trap mass spectrometers, the process of CID occurs when the trajectory of the mass selected parent ions is increased resulting in bimolecular collisions with the inert dampening/bath gas (e.g. helium) that lead to an increase in internal energy of

69 the parent ions. The energy is rapidly redistributed across the ions vibrational modes (3N – 6 where N = non-linear atoms). Unimolecular dissociation follows and results in the fragmentation of the parent ion into: (i) complementary ion pairs, or (ii) a product ion and neutral fragments. The CID of larger clusters can produce new ions that may be elusive in their characterisation in the condensed phase.

1.3.3.2 Ion-Molecule Reactions

Ion-molecule reactions are valuable as an analytical394 and structural395 tool when investigating ions.396 It allows the quantitative measure of the reaction rate for an ion with a neutral reagent and has found applications in examining a variety of metal-mediated catalysis.397-401

Figure 1.35 A diagram showing the ion-molecule reaction setup used in this thesis.

The mass spectrometer setup and instrument modifications for ion-molecule reactions used in this thesis has been described in detail elsewhere.399-400 In brief, helium is passed through a stainless-steel tube and mixes with thermally vaporised neutral reagent, the flow rate of both are carefully measured. A small amount of the helium/reagent mixture is carried via a

70 restriction capillary to the ion trap while the remaining is diverted to an exhaust, see Figure

1.35 for an example of the experimental setup used in this thesis. There is a large excess of neutral reagent to the parent/reactant ion and consequently pseudo first-order kinetics apply.

If the reaction of the neutral with the parent/reactant ion produces a charged species this is observed as a product ion. The reaction delay of neutral to ion is varied over ca. 10 time points and the branching ratios of the parent to product ions are measured and extrapolated determine the pseudo first-order kinetics of the reaction.

1.4 Thesis Scope

This work will focus on the use of a combination of powerful gas-phase techniques and density functional theory to reveal the structure and properties of key intermediates in the reactions of coinage metal hydride ions with neutral reagents. An emphasis will be placed on the selective dehydrogenation of formic acid by ligated coinage metal hydride catalysts. The role of the metal(s) and ligand(s) will be examined in order to optimise the reactivity of each step in the catalytic cycle. It is also an aim of this work to translate the gas-phase studies into the condensed phase and to track key intermediates in the reactivity and growth of coinage metal hydrides.

71 1.5 References

1. Cardellini, L., J. Chem. Educ. 2005, 82 (5), 751. 2. De Hoffmann, E.; Stroobant, V., Mass spectrometry: principles and applications. John Wiley & Sons: 2007. 3. Thomson, J. J., Proc. Royal Soc. Lond. A 1913, 89 (607), 1-20. 4. Kroto, H. W.; Heath, J. R.; O'Brien, S. C.; Curl, R. F.; Smalley, R. E., Nature 1985, 318 (6042), 162-163. 5. Campbell, E. K.; Holz, M.; Gerlich, D.; Maier, J. P., Nature 2015, 523, 32-323. 6. Matsuo, Y., Fullerene Derivatives for Organic Solar Cells. In Chemical Science of π- Electron Systems, Akasaka, T.; Osuka, A.; Fukuzumi, S.; Kandori, H.; Aso, Y., Eds. Springer Japan: Tokyo, 2015; pp 559-573. 7. Mi, D.; Kim, J.-H.; Kim, H. U.; Xu, F.; Hwang, D.-H., J. Nanosci. Nanotechnol. 2014, 14 (2), 1064-1084. 8. Seyler, H.; Wong, W. W. H.; Jones, D. J.; Holmes, A. B., J. Org. Chem. 2011, 76 (9), 3551- 3556. 9. Wong, W. W. H.; Banal, J. L.; Geraghty, P. B.; Hong, Q.; Zhang, B.; Holmes, A. B.; Jones, D. J., Chem. Rec. 2015, 15 (6), 1006-1020. 10. Wurtz, A., Comptes Rendus Acad. Sci. 1844, 18, 702 - 704. 11. Berthelot, Comptes rendus hebdomadaires des séances de l'Académie des sciences 1879, 89, 1005-1011. 12. Wurtz, A., Comptes Rendus Acad. Sci. 1879, 89, 1066-1068. 13. Berthelot, Comptes rendus hebdomadaires des séances de l'Académie des sciences 1879, 89, 1097-1099. 14. Wurtz, A., Comptes Rendus Acad. Sci. 1880, 90, 22-24. 15. Muller, H.; Bradley, A. J., J. Chem. Soc. 1926, 129 (0), 1669-1673. 16. Goedkoop, J. A.; Andresen, A. F., Acta Cryst. 1955, 8 (2), 118-119. 17. Auer, H.; Kohlmann, H., Z. anorg. allg. Chem. 2014, 640, 3159-3165. 18. Wiberg, E.; Henle, W., Z. Naturforsch., B: J. Chem. Sci. 1952, 7b, 250. 19. Dilts, J. A.; Shriver, D. F., J. Am. Chem. Soc. 1968, 90 (21), 5769-5772. 20. Bartlett, E. J.; Rice, W. F., Amer. Chem. J. 1897, 19, 49-52. 21. Berthelot, Comptes rendus hebdomadaires des séances de l'Académie des sciences 1900, 131, 1169-1170. 22. Hulthen, E.; Hulthen; Hulthèn, E.; Zumstein, R. V., Phys. Rev. 1926, 28 (1), 13-24. 23. Watson, W. W.; Perkins, B., Jr., Phys. Rev. 1927, 30, 592-7.

72 24. Bengtsson, E.; Hulthen, E., Trans. Faraday Soc. 1929, 25, 751-757. 25. Bengtsson, E., Nature 1931, 127, 14. 26. Mecke, R., Z. Phys. 1931, 72, 155-162. 27. Bengtsson, E.; Olsson, E., Z. Phys. 1931, 72, 163-176. 28. Gero, L.; Schmid, R. F., Z. Phys. 1943, 121, 459-487. 29. Pietsch, E.; Seuferling, F., Naturwissenschaften 1931, 19, 573-574. 30. Mostowitsch, W.; Pletneff, W., Zh. Russ. Metall. O-va. 1915, 410-31. 31. Hulthen, E.; Zumstein, R. V., Phys. Rev. 1926, 28, 13-24. 32. Farkas, A., Z. physik. Chem. 1929, 5 (Abt. B), 467-75. 33. Guggenheimer, K. M., Proc. Phys. Soc. 1946, 58, 456-468. 34. Ershov, A. D., Uchenye Zapiski Astrakhan. Gosudarst. Pedagog. Inst. 1955, 4, 125-30. 35. Varshni, Y. P.; Mitra, S. S.; Shukla, R. C., Indian J. Phys. 1959, 33, 473-82. 36. Lippincott, E. R.; Steele, D.; Caldwell, P., J. Chem. Phys. 1961, 35, 123-141. 37. Ringstrom, U., Nature 1963, 198, 981. 38. Imanishi, S., Nature 1935, 136, 476. 39. Heimer, T., Naturwissenschaften 1936, 24, 78. 40. Wiberg, E.; Neumaier, H., Inorg. Nucl. Chem. Lett. 1965, 1 (2), 35-7. 41. Dhayal, R. S.; van Zyl, W. E.; Liu, C. W., Acc. Chem. Res. 2016, 49 (1), 86-95. 42. Jordan, A. J.; Lalic, G.; Sadighi, J. P., Chem. Rev. 2016, 116 (15), 8318-8372. 43. Liu, X.; Astruc, D., Coord. Chem. Rev. 2018, 359, 112-126. 44. Ligand-Stabilized Metal Clusters: Structure, Bonding, Fluxionality, and the Metallic State. In Advances in Chemical Physics. 45. Bai, C.; Liu, M., Angew. Chem. Int. Ed. 2013, 52 (10), 2678-2683. 46. Castleman, A. W., J. Phys. Chem. Lett. 2011, 2 (9), 1062-1069. 47. Daniel, M. C.; Astruc, D., Chem. Rev. 2004, 104 (1), 293-346. 48. Pettibone, J. M.; Osborn, W. A.; Rykaczewski, K.; Talin, A. A.; Bonevich, J. E.; Hudgens, J. W.; Allendorf, M. D., Nanoscale 2013, 5 (14), 6558-6566. 49. Brust, M.; Walker, M.; Bethell, D.; Schiffrin, D. J.; Whyman, R., J. Chem. Soc., Chem. Commun. 1994, (7), 801-802. 50. Wuithschick, M.; Paul, B.; Bienert, R.; Sarfraz, A.; Vainio, U.; Sztucki, M.; Kraehnert, R.; Strasser, P.; Rademann, K.; Emmerling, F.; Polte, J., Chem. Mater. 2013, 25 (23), 4679-4689. 51. Sobol, O.; Gadot, E.; Wang, Y.; Weinstock, I. A.; Meshi, L., Inorg. Chem. 2015, 54 (22), 10521-10523.

73 52. Mahoney, W. S.; Brestensky, D. M.; Stryker, J. M., J. Am. Chem. Soc. 1988, 110 (1), 291- 293. 53. Bezman, S. A.; Churchill, M. R.; Osborn, J. A.; Wormald, J., J. Am. Chem. Soc. 1971, 93 (8), 2063-2065. 54. Churchill, M. R.; Bezman, S. A.; Osborn, J. A.; Wormald, J., Inorg. Chem. 1972, 11 (8), 1818-1825. 55. Tate, B. K.; Nguyen, J. T.; Bacsa, J.; Sadighi, J. P., Chem.: Eur. J. 2015, 21 (28), 10160- 10169. 56. Tsui, E. Y.; Müller, P.; Sadighi, J. P., Angew. Chem., Int. Ed. 2008, 47 (46), 8937-8940. 57. Baker, M. V.; Barnard, P. J.; Brayshaw, S. K.; Hickey, J. L.; Skelton, B. W.; White, A. H., Dalton Trans. 2005, (1), 37-43. 58. Laitar, D. S.; Müller, P.; Gray, T. G.; Sadighi, J. P., Organometallics 2005, 24 (19), 4503- 4505. 59. de Frémont, P.; Singh, R.; Stevens, E. D.; Petersen, J. L.; Nolan, S. P., Organometallics 2007, 26 (6), 1376-1385. 60. Lin, I. J. B.; Vasam, C. S., Can. J. Chem. 2005, 83 (6-7), 812-825. 61. Mankad, N. P.; Laitar, D. S.; Sadighi, J. P., Organometallics 2004, 23 (14), 3369-3371. 62. Gray, T. G.; Sadighi, J. P., Group 11 Metal-Metal Bonds. In Molecular Metal-Metal Bonds: Compounds, Synthesis, Properties, 2015; pp 397-428. 63. Schmidbaur, H.; Schier, A., Chem. Soc. Rev. 2012, 41 (1), 370-412. 64. Catalano, V. J.; Kar, H. M.; Garnas, J., Angew. Chem., Int. Ed. 1999, 38 (13/14), 1979- 1982. 65. Chaudhari, K.; Xavier, P. L.; Pradeep, T., ACS Nano 2011, 5 (11), 8816-8827. 66. Che, C.-M.; Mao, Z.; Miskowski, V. M.; Tse, M.-C.; Chan, C.-K.; Cheung, K.-K.; Phillips, D. L.; Leung, K.-H., Angew. Chem., Int. Ed. 2000, 39 (22), 4084-4088. 67. Che, C.-M.; Tse, M.-C.; Chan, M. C. W.; Cheung, K.-K.; Phillips, D. L.; Leung, K.-H., J. Am. Chem. Soc. 2000, 122 (11), 2464-2468. 68. Chin, P. T. K.; van, d. L. M.; van, H. E. J.; Barendregt, A.; Rood, M. T. M.; Koster, A. J.; van, L. F. W. B.; de, M. D. C.; Heck, A. J. R.; Meijerink, A., Nanotechnology 2013, 24 (7), 075703/1-075703/7. 69. Jiang, Y.; Wang, Y.-T.; Ma, Z.-G.; Li, Z.-H.; Wei, Q.-H.; Chen, G.-N., Organometallics 2013, 32 (17), 4919-4926. 70. Lavenn, C.; Albrieux, F.; Tuel, A.; Demessence, A., J. Colloid Interface Sci. 2014, 418, 234-239. 71. Rao, T. U. B.; Pradeep, T., Angew. Chem., Int. Ed. 2010, 49 (23), 3925-3929.

72. Wang, C.-F.; Peng, S.-M.; Chan, C.-K.; Che, C.-M., Polyhedron 1996, 15 (11), 1853-8. 73. Yam, V. W.-W.; Lo, K. K.-W.; Wang, C.-R.; Cheung, K.-K., Inorg. Chem. 1996, 35 (18), 5116-5117. 74. Zhou, R.; Shi, M.; Chen, X.; Wang, M.; Chen, H., Chem.: Eur. J. 2009, 15 (19), 4944- 4951. 75. Pearson, R. G., Chem. Rev. 1985, 85 (1), 41-49. 76. Martinho Simoes, J., Chem. Rev. 1990, 90 (4), 629-688. 77. Tolman, C. A., Chem. Rev. 1977, 77 (3), 313-348. 78. Gorin, D. J.; Sherry, B. D.; Toste, F. D., Chem. Rev. 2008, 108 (8), 3351-3378. 79. Xu, Y.; Mingos, D. M.; Brown, J. M., Chem. Commun. 2008, (2), 199-201. 80. Romero, E. A.; Olsen, P. M.; Jazzar, R.; Soleilhavoup, M.; Gembicky, M.; Bertrand, G., Angew. Chem., Int. Ed. 2017, 56 (14), 4024-4027. 81. Goeden, G. V.; Huffman, J. C.; Caulton, K. G., Inorg. Chem. 1986, 25 (15), 2484-2485. 82. Wyss, C.; Tate, B.; Bacsa, J.; Gray, T.; Sadighi, J., Angew. Chem., Int. Ed. 2013, 52 (49), 12920-12923. 83. Bondi, A., J. Phys. Chem. 1964, 68 (3), 441-451. 84. Reed, A. E.; Weinstock, R. B.; Weinhold, F., J. Phys. Chem. 1985, 83 (2), 735-746. 85. Eberhart, M. S.; Norton, J. R.; Zuzek, A.; Sattler, W.; Ruccolo, S., J. Am. Chem. Soc. 2013, 135 (46), 17262-17265. 86. Mao, Z.; Huang, J.-S.; Che, C.-M.; Zhu, N.; Leung, S. K.-Y.; Zhou, Z.-Y., J. Am. Chem. Soc. 2005, 127 (13), 4562-4563. 87. Cook, A. W.; Nguyen, T. A. D.; Buratto, W. R.; Wu, G.; Hayton, T. W., Inorg. Chem. 2016, 55 (23), 12435-12440. 88. Li, J.; White, J. M.; Mulder, R. J.; Reid, G. E.; Donnelly, P. S.; O'Hair, R. A. J., Inorg. Chem. 2016, 55 (19), 9858-9868. 89. Kanako, N.; Miho, T.; Bunsho, K.; Takayuki, N.; Yasuyuki, U.; Tomoaki, T., Chem.: Eur. J. 2017, 23 (40), 9457-9461. 90. Ma, H. Z.; Li, J.; Canty, A. J.; O'Hair, R. A. J., Dalton Trans. 2017, 46 (43), 14995-15003. 91. Nie, H.-H.; Han, Y.-Z.; Tang, Z.; Yang, S.-Y.; Teo, B. K., J. Clust. Sci. 2018. 92. Albert, C. F.; Healy, P. C.; Kildea, J. D.; Raston, C. L.; Skelton, B. W.; White, A. H., Inorg. Chem. 1989, 28 (7), 1300-1306. 93. Stevens, R. C.; McLean, M. R.; Bau, R.; Koetzle, T. F., J. Am. Chem. Soc. 1989, 111 (9), 3472-3. 94. Bennett, E. L.; Murphy, P. J.; Imberti, S.; Parker, S. F., Inorg. Chem. 2014, 53 (6), 2963- 2967.

95. Liu, S.; Eberhart, M. S.; Norton, J. R.; Yin, X.; Neary, M. C.; Paley, D. W., J. Am. Chem. Soc. 2017, 139 (23), 7685-7688. 96. Köhn, R. D.; Pan, Z.; Mahon, M. F.; Kociok-Köhn, G., ChemComm 2003, 9 (11), 1272- 1273. 97. Liao, P. K.; Liu, K. G.; Fang, C. S.; Liu, C. W.; Fackler, J. P., Jr.; Wu, Y. Y., Inorg. Chem. 2011, 50 (17), 8410-8417. 98. Lemmen, T. H.; Folting, K.; Huffman, J. C.; Caulton, K. G., J. Am. Chem. Soc. 1985, 107 (25), 7774-7775. 99. Liu, C. W.; Sarkar, B.; Huang, Y. J.; Liao, P. K.; Wang, J. C.; Saillard, J. Y.; Kahlal, S., J. Am. Chem. Soc. 2009, 131 (31), 11222-11233. 100. Goeden, G. V.; Caulton, K. G., J. Am. Chem. Soc. 1981, 103 (24), 7354-7535. 101. Nguyen, T.-A. D.; Goldsmith, B. R.; Zaman, H. T.; Wu, G.; Peters, B.; Hayton, T. W., Chem.: Eur. J. 2015, 21 (14), 5341-5344. 102. Huertos, M. A.; Cano, I.; Bandeira, N. A. G.; Benet-Buchholz, J.; Bo, C.; van Leeuwen, P. W. N. M., Chem.: Eur. J. 2014, 20 (49), 16121-16127. 103. Li, J.; Ma, H. Z.; Reid, G. E.; Edwards, A. J.; Hong, Y.; White, J. M.; Mulder, R. J.; O'Hair, R. A. J., Chem.: Eur. J. 2018, 24 (9), 2070-2074. 104. Dhayal, R. S.; Liao, J.-H.; Lin, Y.-R.; Liao, P.-K.; Kahlal, S.; Saillard, J.-Y.; Liu, C. W., J. Am. Chem. Soc. 2013, 135 (12), 4704-4707. 105. Liao, J.-H.; Dhayal, R. S.; Wang, X.; Kahlal, S.; Saillard, J.-Y.; Liu, C. W., Inorg. Chem. 2014, 53 (20), 11140-11145. 106. Dhayal, R. S.; Liao, J.-H.; Wang, X.; Liu, Y.-C.; Chiang, M.-H.; Kahlal, S.; Saillard, J.-Y.; Liu, C. W., Angew. Chem., Int. Ed. 2015, 54 (46), 13604-13608. 107. Dhayal, R. S.; Liao, J. H.; Kahlal, S.; Wang, X.; Liu, Y. C.; Chiang, M. H.; Van Zyl, W. E.; Saillard, J. Y.; Liu, C. W., Chem.: Eur. J. 2015, 21 (23), 8369-8374. 108. Nguyen, T.-A. D.; Jones, Z. R.; Leto, D. F.; Wu, G.; Scott, S. L.; Hayton, T. W., Chem. Mater. 2016, 28 (22), 8385-8390. 109. Hansen, T. W.; DeLaRiva, A. T.; Challa, S. R.; Datye, A. K., Acc. Chem. Res. 2013, 46 (8), 1720-1730. 110. Wang, F.; Richards, V. N.; Shields, S. P.; Buhro, W. E., Chem. Mater. 2014, 26 (1), 5-21. 111. Nguyen, T.-A. D.; Jones, Z. R.; Goldsmith, B. R.; Buratto, W. R.; Wu, G.; Scott, S. L.; Hayton, T. W., J. Am. Chem. Soc. 2015, 137 (41), 13319-13324. 112. Edwards, A. J.; Dhayal, R. S.; Liao, P.-K.; Liao, J.-H.; Chiang, M.-H.; Piltz, R. O.; Kahlal, S.; Saillard, J.-Y.; Liu, C. W., Angew. Chem., Int. Ed. 2014, 53 (28), 7214-7218. 113. Webster, A. H.; Halpern, J., J. Phys. Chem. 1956, 60 (3), 280-285.

76 114. Webster, A. H.; Halpern, J., J. Phys. Chem. 1957, 61, 1245-1248. 115. Webster, A. H.; Halpern, J., J. Phys. Chem. 1957, 61 (9), 1239-1245. 116. W., H. M.; T., H. W.; Frank, R.; F., S. B., Angew. Chem., Int. Ed. 2015, 54 (35), 10331- 10335. 117. Penner, G. H.; Liu, X., Prog. Nucl. Magn. Reson. Spectrosc. 2006, 49 (2), 151-167. 118. Tate, B. K.; Wyss, C. M.; Bacsa, J.; Kluge, K.; Gelbaum, L.; Sadighi, J. P., Chem. Sci. 2013, 4 (8), 3068-3074. 119. DeKock, R. L.; Van Zee, R. D.; Ziegler, T., Inorg. Chem. 1987, 26 (4), 563-567. 120. Mitrić, R.; Petersen, J.; Kulesza, A.; Röhr, M. I. S.; Bonačić-Koutecký, V.; Brunet, C.; Antoine, R.; Dugourd, P.; Broyer, M.; Ohair, R. A. J., J. Phys. Chem. Lett. 2011, 2 (6), 548-552. 121. Zhao, S.; Liu, Z.-P.; Li, Z.-H.; Wang, W.-N.; Fan, K.-N., J. Phys. Chem. A 2006, 110 (40), 11537-11542. 122. Gaspar, R.; Gáspár, R.; Tamássy Lentei, I., Acta Phys. Acad. Sci. Hung. 1981, 50 (4), 343- 347. 123. Zavras, A.; Khairallah, G. N.; Connell, T. U.; White, J. M.; Edwards, A. J.; Donnelly, P. S.; O'Hair, R. A. J., Angew. Chem., Int. Ed. 2013, 52 (32), 8391-8394. 124. Zavras, A.; Khairallah, G. N.; Connell, T. U.; White, J. M.; Edwards, A. J.; Mulder, R. J.; Donnelly, P. S.; Ohair, R. A. J., Inorg. Chem. 2014, 53 (14), 7429-7437. 125. Chiodo, S. G.; Mineva, T., J. Phys. Chem. C. 2016, 120 (8), 4471-4480. 126. Liu, C. W.; Lin, Y.-R.; Fang, C.-S.; Latouche, C.; Kahlal, S.; Saillard, J.-Y., Inorg. Chem. 2013, 52 (4), 2070-2077. 127. Liu, C. W.; Chang, H.-W.; Sarkar, B.; Saillard, J.-Y.; Kahlal, S.; Wu, Y.-Y., Inorg. Chem. 2010, 49 (2), 468-475. 128. Liu, C. W.; Chang, H.-W.; Fang, C.-S.; Sarkar, B.; Wang, J.-C., ChemComm 2010, 46, 4571-4573. 129. Liu, C. W.; Liao, P.-K.; Fang, C.-S.; Saillard, J.-Y.; Kahlal, S.; Wang, J.-C., Chem. Commun. 2011, 47 (20), 5831-5833. 130. Mingos, D. M. P., Gold Bull. 1984, 17 (1), 5-12. 131. Mingos, D. M. P., Gold Clusters, Colloids and Nanoparticles I. Springer International Publishing: 2014. 132. Mingos, D. M. P., Gold Clusters, Colloids and Nanoparticles II. Springer International Publishing: 2014. 133. Zhao, P.; Li, N.; Astruc, D., Coord. Chem. Rev. 2013, 257 (3–4), 638-665. 134. Rosca, D. A.; Wright, J. A.; Hughes, D. L.; Bochmann, M., Nat. Commun. 2013, 4, 2167.

77 135. Rosca, D.-A.; Smith, D. A.; Hughes, D. L.; Bochmann, M., Angew. Chem., Int. Ed. 2012, 51 (42), 10643-10646. 136. Daugherty, N. T.; Bacsa, J.; Sadighi, J. P., Organometallics 2017, 36 (17), 3171-3174. 137. Phillips, N.; Dodson, T.; Tirfoin, R.; Bates, J. I.; Aldridge, S., Chem.: Eur. J. 2014, 20 (50), 16721-16731. 138. Hashmi, A. S. K., Angew. Chem., Int. Ed. 2012, 51 (52), 12935-12936. 139. Roşca, D.-A.; Smith, D.; Bochmann, M., Chemical communications (1996) 2012, 48 (58), 7247-7249. 140. Rocchigiani, L.; Fernandez-Cestau, J.; Chambrier, I.; Hrobárik, P.; Bochmann, M., J. Am. Chem. Soc. 2018, 140 (26), 8287-8302. 141. Pintus, A.; Rocchigiani, L.; Fernandez-Cestau, J.; Budzelaar, P. H. M.; Bochmann, M., Angew. Chem. 2016, 128 (40), 12509-12512. 142. Takano, S.; Hirai, H.; Muramatsu, S.; Tsukuda, T., J. Am. Chem. Soc. 2018, 140 (27), 8380-8383. 143. Bos, W.; Bour, J. J.; Steggerda, J. J., Inorg. Chem. 1985, 24 (25), 4298-4301. 144. Yamazoe, S.; Matsuo, S.; Muramatsu, S.; Takano, S.; Nitta, K.; Tsukuda, T., Inorg. Chem. 2017, 56 (14), 8319-8325. 145. Kleinhans, G.; Hansmann, M. M.; Guisado-Barrios, G.; Liles, D. C.; Bertrand, G.; Bezuidenhout, D. I., J. Am. Chem. Soc. 2016, 138 (49), 15873-15876. 146. Hicken, A.; White, A. J. P.; Crimmin, M. R., Angew. Chem., Int. Ed. 2017, 56 (47), 15127- 15130. 147. Pyykkö, P., Chem. Rev. 1997, 97 (3), 597-636. 148. Wong, K. M. C.; Au, V. K. M.; Yam, V. W. W., 8.03 - Noncovalent Metal–Metal Interactions. In Comprehensive Inorganic Chemistry II (Second Edition), Reedijk, J.; Poeppelmeier, K., Eds. Elsevier: Amsterdam, 2013; pp 59-130. 149. Powers, I. G.; Uyeda, C., ACS Catal. 2017, 7 (2), 936-958. 150. Dai, Y.; Wang, Y.; Liu, B.; Yang, Y., Small 2015, 11 (3), 268-289. 151. Lundgren, R. J.; Stradiotto, M., Key Concepts in Ligand Design. In Ligand Design in Metal Chemistry, 2016. 152. Waldie, K. M.; Ostericher, A. L.; Reineke, M. H.; Sasayama, A. F.; Kubiak, C. P., ACS Catalysis 2018, 8 (2), 1313-1324. 153. DuBois, D. L.; Berning, D. E., Appl. Organomet. Chem. 2000, 14 (12), 860-862. 154. Wiedner, E. S.; Chambers, M. B.; Pitman, C. L.; Bullock, R. M.; Miller, A. J. M.; Appel, A. M., Chem. Rev. 2016, 116 (15), 8655-8692. 155. Weil, D. A.; Wilkins, C. L., J. Am. Chem. Soc. 1985, 107 (25), 7316-7320.

78 156. Fisher, E. R.; Armentrout, P. B., J. Phys. Chem. 1990, 94 (4), 1674-1683. 157. Chen, Y.-M.; Armentrout, P. B., J. Phys. Chem. 1995, 99 (29), 11424-11431. 158. Kant, A.; Moon, K. A., High Temp. Sci. 1979, 11 (1), 55-62. 159. Simoes, J. A. M.; Beauchamp, J. L., Chem. Rev. 1990, 90 (4), 629-688. 160. Bootharaju, M. S.; Dey, R.; Gevers, L. E.; Hedhili, M. N.; Basset, J. M.; Bakr, O. M., J. Am. Chem. Soc. 2016, 138 (42), 13770-13773. 161. Meijboom, R.; Bowen, R. J.; Berners-Price, S. J., Coord. Chem. Rev. 2009, 253 (3-4), 325-342. 162. Casey, C. P.; Whiteker, G. T., Isr. J. Chem. 1990, 30 (4), 299-304. 163. Freixa, Z.; van Leeuwen, P. W. N. M., Dalton Trans. 2003, (10), 1890-1901. 164. Mansell, S. M., Dalton Trans. 2017, 46 (44), 15157-15174. 165. Gillespie, J. A.; Dodds, D. L.; Kamer, P. C. J., Dalton Trans. 2010, 39 (11), 2751-2764. 166. Jover, J.; Fey, N.; Harvey, J. N.; Lloyd-Jones, G. C.; Orpen, A. G.; Owen-Smith, G. J. J.; Murray, P.; Hose, D. R. J.; Osborne, R.; Purdie, M., Organometallics 2010, 29 (23), 6245-6258. 167. Jover, J.; Fey, N.; Harvey, J. N.; Lloyd-Jones, G. C.; Orpen, A. G.; Owen-Smith, G. J. J.; Murray, P.; Hose, D. R. J.; Osborne, R.; Purdie, M., Organometallics 2012, 31 (15), 5302- 5306. 168. . 169. Vallee, B. L.; Williams, R. J., Proc. Natl. Acad. Sci. U.S.A. 1968, 59 (2), 498-505. 170. Comba, P., Coord. Chem. Rev. 2000, 200-202, 217-245. 171. Comba, P.; Schiek, W., Coord. Chem. Rev. 2003, 238-239, 21-29. 172. Joost, M.; Zeineddine, A.; Estévez, L.; Mallet−Ladeira, S.; Miqueu, K.; Amgoune, A.; Bourissou, D., J. Am. Chem. Soc. 2014, 136 (42), 14654-14657. 173. Deutsch, C.; Krause, N.; Lipshutz, B. H., Chem. Rev. 2008, 108 (8), 2916-2927. 174. Frieman, B. A.; Lipshutz, B. H. In Copper hydride in a bottle: Asymmetric hydrosilylations in a hurry, American Chemical Society: 2005; pp ORGN-731. 175. Chen, J.-X.; Daeuble, J. F.; Brestensky, D. M.; Stryker, J. M., Tetrahedron 2000, 56 (15), 2153-2166. 176. Lipshutz, B. H.; Chrisman, W.; Noson, K., J. Organomet. Chem. 2001, 624 (1), 367-371. 177. Zhou, J.-N.; Fang, Q.; Hu, Y.-H.; Yang, L.-Y.; Wu, F.-F.; Xie, L.-J.; Wu, J.; Li, S., Org. Biomol. Chem. 2014, 12 (6), 1009-1017. 178. Shimizu, H.; Igarashi, D.; Kuriyama, W.; Yusa, Y.; Sayo, N.; Saito, T., Org. Lett. 2007, 9 (9), 1655-1657.

79 179. Loges, B.; Boddien, A.; Gartner, F.; Junge, H.; Beller, M., Top. Catal. 2010, 53 (13-14), 902-914. 180. Krabbe, S. W.; Hatcher, M. A.; Bowman, R. K.; Mitchell, M. B.; McClure, M. S.; Johnson, J. S., Org. Lett. 2013, 15 (17), 4560-4563. 181. Moser, R.; Boskovic, Z. V.; Crowe, C. S.; Lipshutz, B. H., J. Am. Chem. Soc. 2010, 132 (23), 7852-7853. 182. Waidmann, C. R.; “Pete” Silks, L. A.; Wu, R.; Gordon, J. C., Catal. Sci. Technol. 2013, 3 (5), 1240-1245. 183. Semba, K.; Fujihara, T.; Xu, T.; Terao, J.; Tsuji, Y., Adv. Synth. Catal. 2012, 354 (8), 1542- 1550. 184. Whittaker, A. M.; Lalic, G., Org. Lett. 2013, 15 (5), 1112-1115. 185. Cao, H.; Chen, T.; Zhou, Y.; Han, D.; Yin, S.-F.; Han, L.-B., Adv. Synth. Catal. 2014, 356 (4), 765-769. 186. Pape, F.; Thiel, N. O.; Teichert, J. F., Chem.: Eur. J. 2015, 21 (45), 15934-15938. 187. Wakamatsu, T.; Nagao, K.; Ohmiya, H.; Sawamura, M., Organometallics 2016, 35 (10), 1354-1357. 188. Zhu, S.; Niljianskul, N.; Buchwald, S. L., J. Am. Chem. Soc. 2013, 135 (42), 15746-15749. 189. Lipshutz, B. H.; Amorelli, B.; Unger, J. B., J. Am. Chem. Soc. 2008, 130 (44), 14378- 14379. 190. Jia, Z.; Zhou, F.; Liu, M.; Li, X.; Chan, A. S. C.; Li, C.-J., Angew. Chem., Int. Ed. 2013, 52 (45), 11871-11874. 191. Liu, M.; Zhou, F.; Jia, Z.; Li, C.-J., Org. Chem. Front. 2014, 1 (2), 161-166. 192. Karunananda, M. K.; Mankad, N. P., J. Am. Chem. Soc. 2015, 137 (46), 14598-14601. 193. Baba, T., Catal. Surv. Asia 2005, 9, 147-154. 194. Baba, T., Zeoraito 2008, 25 (4), 133-146. 195. Baba, T.; Abe, Y., Applied Catalysis A: General 2003, 250 (2), 265-270. 196. Baba, T.; Abe, Y.; Nomoto, K.; Inazu, K.; Echizen, T.; Ishikawa, A.; Murai, K., J. Phys. Chem. B 2005, 109 (9), 4263-4268. 197. Baba, T.; Inazu, K., Chem. Lett. 2006, 35 (2), 142-147. 198. Baba, T.; Iwase, Y.; Inazu, K.; Masih, D.; Matsumoto, A., Microporous and Mesoporous Materials 2007, 101 (1-2), 142-147. 199. Baba, T.; Komatsu, N.; Sawada, H.; Yamaguchi, Y.; Takahashi, T.; Sugisawa, H.; Ono, Y., Langmuir 1999, 15 (23), 7894-7896. 200. Baba, T.; Sawada, H., Phys. Chem. Chem. Phys. 2002, 4 (15), 3919-3923.

80 201. Baba, T.; Sawada, H.; Takahashi, T.; Abe, M., Applied Catalysis A: General 2002, 231 (1- 2), 55-63. 202. Baba, T.; Tohjo, Y.; Takahashi, T.; Sawada, H.; Ono, Y., Catalysis Today 2001, 66 (1), 81- 89. 203. Inazu, K.; Koyama, T.; Miyaji, A.; Baba, T., J. Jpn. Pet. Inst. 2008, 51 (4), 205-216. 204. Comas-Vives, A.; Ujaque, G., J. Am. Chem. Soc. 2013, 135 (4), 1295-1305. 205. González-Arellano, C.; Corma, A.; Iglesias, M.; Sánchez, F., ChemComm 2005, (27), 3451-3453. 206. Comas-Vives, A.; González-Arellano, C.; Corma, A.; Iglesias, M.; Sánchez, F.; Ujaque, G., J. Am. Chem. Soc. 2006, 128 (14), 4756-4765. 207. Ito, H.; Saito, T.; Miyahara, T.; Zhong, C.; Sawamura, M., Organometallics 2009, 28 (16), 4829-4840. 208. Eberle, U.; Felderhoff, M.; Schüth, F., Angew. Chem., Int. Ed. 2009, 48 (36), 6608- 6630. 209. Chandra, D.; Reilly, J. J.; Chellappa, R., JOM 2006, 58 (2), 26-32. 210. Grochala, W.; Edwards, P. P., Chem. Rev. 2004, 104 (3), 1283-1315. 211. Mohtadi, R.; Orimo, S.-i., Nat. Rev. Mat. 2016, 2, 16091. 212. Schneemann, A.; White, J. L.; Kang, S.; Jeong, S.; Wan, L. F.; Cho, E. S.; Heo, T. W.; Prendergast, D.; Urban, J. J.; Wood, B. C.; Allendorf, M. D.; Stavila, V., Chem. Rev. 2018, 118 (22), 10775-10839. 213. Fukuzumi, S.; Suenobu, T., Dalton Trans. 2013, 42 (1), 18-28. 214. Andrews, L.; Wang, X., J. Am. Chem. Soc. 2003, 125 (38), 11751 - 11760. 215. Boddien, A.; Junge, H., Nat. Nanotechnol. 2011, 6 (5), 265-266. 216. Li, Z.; Xu, Q., Acc. Chem. Res. 2017, 50 (6), 1449-1458. 217. Mellmann, D.; Sponholz, P.; Junge, H.; Beller, M., Chem. Soc. Rev. 2016, 45 (14), 3954- 3988. 218. Muller, K.; Brooks, K.; Autrey, T., Energy Fuels 2017, 31 (11), 12603-12611. 219. Onishi, N.; Iguchi, M.; Yang, X.; Kanega, R.; Kawanami, H.; Xu, Q.; Himeda, Y., Adv. Energy Mater. 2018, Ahead of Print. 220. Singh, A. K.; Singh, S.; Kumar, A., Catal. Sci. Technol. 2016, 6 (1), 12-40. 221. Wang, X.; Meng, Q.; Gao, L.; Jin, Z.; Ge, J.; Liu, C.; Xing, W., Int. J. Hydrogen Energy 2018, 43 (14), 7055-7071. 222. Zhong, S.; Xu, Q., Bull. Chem. Soc. Jpn. 2018, 91 (11), 1606-1617. 223. Grasemann, M.; Laurenczy, G., Energy Environ. Sci. 2012, 5 (8), 8171-8181. 224. Kim, K.; Barteau, M., Langmuir 1988, 4 (4), 945-953.

81 225. Mars, P.; Scholten, J.; Zwietering, P., The catalytic decomposition of formic acid. In Advances in Catalysis, Elsevier: 1963; Vol. 14, pp 35-113. 226. Larsson, R.; Jamróz, M. H.; Borowiak, M. A., J. Mol. Catal. Chem. 1998, 129 (1), 41-51. 227. Trillo, J.; Munuera, G.; Criado, J., Catal. Rev. 1972, 7 (1), 51-86. 228. Bulushev, D. A.; Beloshapkin, S.; Ross, J. R. H., Catal. Today 2010, 154 (1-2), 7-12. 229. Jiang, Y.; Lu, Y.; Han, D.; Zhang, Q.; Niu, L., Nanotechnology 2012, 23 (10), 105609/1- 105609/9. 230. Tingey, H. C.; Hinshelwood, C. N., J. Chem. Soc., Trans., 1922, 121 (0), 1668-1676. 231. Ojeda, M.; Iglesia, E., Angew. Chem., Int. Ed. 2009, 48 (26), 4800-3. 232. Bi, Q.-Y.; Du, X.-L.; Liu, Y.-M.; Cao, Y.; He, H.-Y.; Fan, K.-N., J. Am. Chem. Soc. 2012, 134 (21), 8926-8933. 233. Banerjee, R.; Das, K.; Das, A.; Dasgupta, S., Inorg. Chem. 1989, 28 (3), 585-588. 234. Czaun, M.; Goeppert, A.; May, R.; Haiges, R.; Prakash, G. K. S.; Olah, G. A., ChemSusChem 2011, 4 (9), 1241-1248. 235. Fink, C.; Montandon-Clerc, M.; Laurenczy, G., Chimia 2015, 69 (12), 746-752. 236. Gao, Y.; Kuncheria, J.; Puddephatt, R. J.; Yap, G. P. A., Chem. Commun. 1998, (21), 2365-2366. 237. Cho, J.; Lee, S.; Han, J.; Yoon, S. P.; Nam, S. W.; Choi, S. H.; Lee, K. Y.; Ham, H. C., J. Phys. Chem. C 2014, 22553-22560. 238. Yurderi, M.; Bulut, A.; Zahmakiran, M.; Kaya, M., Appl. Catal., B 2014, 160-161, 514-524. 239. Zhang, S.; Metin, O.; Su, D.; Sun, S., Angew. Chem., Int. Ed. 2013, 52 (13), 3681-3684. 240. Enthaler, S.; von Langermann, J.; Schmidt, T., Energy Environ. Sci. 2010, 3 (9), 1207- 1217. 241. Krstić, M.; Jin, Q.; Khairallah, G. N.; O'Hair, R. A. J.; Bonačić-Koutecký, V., ChemCatChem 2018, 10 (5), 1173-1177. 242. Kumar, R.; Krieger, J.-P.; Gómez-Bengoa, E.; Fox, T.; Linden, A.; Nevado, C., Angew. Chem., Int. Ed. 2017, 56 (42), 12862-12865. 243. Anastasi, C.; Hudson, R.; Simpson, V. J., Energy Policy 1990, 18 (10), 936-944. 244. Artz, J.; Mueller, T. E.; Thenert, K.; Kleinekorte, J.; Meys, R.; Sternberg, A.; Bardow, A.; Leitner, W., Chem. Rev. 2018, 118 (2), 434-504. 245. Pigeon, J., Energy Procedia 2017, 114, 7333-7342. 246. Ola, O.; Mercedes Maroto-Valer, M.; Mackintosh, S., Energy Procedia 2013, 37, 6704- 6709. 247. Otto, A.; Grube, T.; Schiebahn, S.; Stolten, D., Energy Environ. Sci. 2015, 8 (11), 3283- 3297.

82 248. Song, Q.-W.; Zhou, Z.-H.; He, L.-N., Green Chem. 2017, 19 (16), 3707-3728. 249. Bui, M.; Adjiman, C. S.; Bardow, A.; Anthony, E. J.; Boston, A.; Brown, S.; Fennell, P. S.; Fuss, S.; Galindo, A.; Hackett, L. A.; Hallett, J. P.; Herzog, H. J.; Jackson, G.; Kemper, J.; Krevor, S.; Maitland, G. C.; Matuszewski, M.; Metcalfe, I. S.; Petit, C.; Puxty, G.; Reimer, J.; Reiner, D. M.; Rubin, E. S.; Scott, S. A.; Shah, N.; Smit, B.; Trusler, J. P. M.; Webley, P.; Wilcox, J.; Mac Dowell, N., Energy Environ.Sci. 2018, 11 (5), 1062-1176. 250. Sattler, W.; Parkin, G., J. Am. Chem. Soc. 2012, 134 (42), 17462-17465. 251. Hull, J. F.; Himeda, Y.; Wang, W.-H.; Hashiguchi, B.; Periana, R.; Szalda, D. J.; Muckerman, J. T.; Fujita, E., Nat. Chem. 2012, 4, 383-388. 252. Yadav, R. K.; Baeg, J.-O.; Oh, G. H.; Park, N.-J.; Kong, K.-j.; Kim, J.; Hwang, D. W.; Biswas, S. K., J. Am. Chem. Soc. 2012, 134 (28), 11455-11461. 253. Li, C. W.; Kanan, M. W., J. Am. Chem. Soc. 2012, 134 (17), 7231-7234. 254. Das Neves Gomes, C.; Jacquet, O.; Villiers, C.; Thuéry, P.; Ephritikhine, M.; Cantat, T., Angew. Chem., Int. Ed. 2012, 51 (1), 187-190. 255. Schmeier, T. J.; Dobereiner, G. E.; Crabtree, R. H.; Hazari, N., J. Am. Chem. Soc. 2011, 133 (24), 9274-9277. 256. Tanaka, R.; Yamashita, M.; Nozaki, K., J. Am. Chem. Soc. 2009, 131 (40), 14168-14169. 257. Duan, Y.; Zhang, J.; Xu, K., Sci. China Phys. Mech. Astron. 2014, 57 (4), 644-651. 258. Kumar, A.; Doradla, P.; Narkhede, M.; Li, L.; Samuelson, L. A.; Giles, R. H.; Kumar, J., RSC Adv. 2014, 4 (69), 36671-36674. 259. Bridges, C. R.; DiCarmine, P. M.; Seferos, D. S., Chem. Mater. 2012, 24 (6), 963-965. 260. Navyatha, B.; Kumar, R.; Nara, S., J. Environ. Eng. (N. Y.) 2016, 4 (1), 924-931. 261. Mardiansyah, D.; Triyana, K.; Sosiati, H.; Harsojo, Aip Conf Proc 2016, 1755 (1), 150019. 262. Rekha, C. R.; Nayar, V. U.; Gopchandran, K. G., J. Sci. Adv. Mater.Dev. 2018, 3 (2), 196- 205. 263. Jakab, A.; Rosman, C.; Khalavka, Y.; Becker, J.; Trügler, A.; Hohenester, U.; Sönnichsen, C., ACS Nano 2011, 5 (9), 6880-6885. 264. Pérez-Juste, J.; Pastoriza-Santos, I.; Liz-Marzán, L. M.; Mulvaney, P., Coord. Chem. Rev. 2005, 249 (17), 1870-1901. 265. Zhao, S.; Han, F.; Li, J.; Meng, X.; Huang, W.; Cao, D.; Zhang, G.; Sun, R.; Wong, C.-P., Small 2018, 14 (26), 1800047. 266. Zhang, P.; Wyman, I.; Hu, J.; Lin, S.; Zhong, Z.; Tu, Y.; Huang, Z.; Wei, Y., Mater. Sci. Eng. C 2017, 223, 1-23.

267. Vargas, J. A.; Petkov, V.; Nouh, E. S. A.; Ramamoorthy, R. K.; Lacroix, L.-M.; Poteau, R.; Viau, G.; Lecante, P.; Arenal, R., ACS Nano 2018, 12 (9), 9521-9531. 268. He, Y.; Wu, X.; Lu, G.; Shi, G., Mater. Chem. Phys 2006, 98 (1), 178-182. 269. Li, Z.; Liu, Z.; Zhang, J.; Han, B.; Du, J.; Gao, Y.; Jiang, T., J. Phys. Chem. B. 2005, 109 (30), 14445-14448. 270. Dai, L.; Qin, Q.; Wang, P.; Zhao, X.; Hu, C.; Liu, P.; Qin, R.; Chen, M.; Ou, D.; Xu, C.; Mo, S.; Wu, B.; Fu, G.; Zhang, P.; Zheng, N., Sci. Adv. 2017, 3 (9), e1701069/1-e1701069/9. 271. Lu, Y.; Chen, W., Chem. Soc. Rev. 2012, 41 (9), 3594-3623. 272. Schmid, G.; Bäumle, M.; Geerkens, M.; Heim, I.; Osemann, C.; Sawitowski, T., Chem. Soc. Rev. 1999, 28 (3), 179-185. 273. Jyoti Borah, B.; Pollov Sarmah, P., Catal. Rev. 2015, 57 (3), 257-305. 274. Widegren, J. A.; Finke, R. G., J. Mol. Catal. Chem. A 2003, 191 (2), 187-207. 275. Schmid, G.; Fenske, D., Philos. Trans. Royal Soc. A 2010, 368 (1915), 1207-1210. 276. Boyle, R., The sceptical chymist: or Chymico-physical doubts & paradoxes, touching the spagyrist's principles commonly call'd hypostatical, as they are wont to be propos'd and defended by the generality of alchymists. Whereunto is praemis'd part of another discourse relating to the same subject. London, Printed by J. Cadwell for J. Crooke, 1661.: 1661. 277. The IUPAC inorganic chemistry definition of cluster is: A number of metal centres grouped close together which can have direct metal bonding interactions or interactions through a bridging ligand, but are not necessarily held together by these interactions. See: IUPAC. Compendium of Chemical Terminology, 2nd ed. (the "Gold Book"). Compiled by A. D. McNaught and A. Wilkinson. Blackwell Scientific Publications, Oxford (1997). XML on-line corrected version: http://goldbook.iupac.org (2006-) created by M. Nic, J. Jirat, B. Kosata; updates compiled by A. Jenkins. ISBN 0-9678550-9-8. doi:10.1351/goldbook. 278. The IUPAP do not appear to have a formal definition of a cluster. Johnston (ref. 611) suggests the following “physics” definition of clusters: I will take the term cluster to mean an aggregate of a countable number (2 – 10n, where n can be as high as 6 or 7) of particles (i.e. atoms or molecules). The constituent particles may be identical, leading to homo-atomic (or homo-molecular) clusters , Aa, or they can be two or more different species – leading to hetero-atomic (hetero-molecular) clusters AaBb. 279. The old IUPAC mass spectrometry definition of a cluster is: An ion formed by the combination of more ions or atoms or molecules of a chemical species often in

84 association with a second species. See “Recommendations for nomenclature and symbolism for mass spectroscopy (including an appendix of terms used in vacuum technology”, J.F.J. Todd, “Recommendations for nomenclature and symbolism for mass spectroscopy (including an appendix of terms used in vacuum technology)”, Pure App. Chem., 63, 1541 (1991). doi:10.1351/pac199163101541. 280. Eleanor, C. Atom cluster. https://ezp.lib.unimelb.edu.au/login?url=https://search.ebscohost.com/login.aspx? direct=true&db=edsasc&AN=edsasc.059850&scope=site. 281. Kreibig, U.; Vollmer, M., Optical properties of metal clusters / Uwe Kreibig, Michael Vollmer. Berlin ; New York : Springer, c1995.: 1995. 282. Ott, L. S.; Finke, R. G., Chemistry of Materials 2008, 20 (7), 2592-2601. 283. Cook, A. W.; Hayton, T. W., Acc. Chem. Res. 2018, 51 (10), 2456-2464. 284. Chakraborty, I.; Pradeep, T., Chem. Rev. 2017, 117 (12), 8208-8271. 285. Femoni, C.; Iapalucci, M. C.; Ruggieri, S.; Zacchini, S., Acc. Chem. Res. 2018, 51 (11), 2748-2755. 286. Higaki, T.; Li, Q.; Zhou, M.; Zhao, S.; Li, Y.; Li, S.; Jin, R., Acc. Chem. Res. 2018, 51 (11), 2764-2773. 287. Ghosh, A.; Mohammed, O. F.; Bakr, O. M., Acc. Chem. Res. 2018, 51 (12), 3094-3103. 288. Hossain, S.; Niihori, Y.; Nair, L. V.; Kumar, B.; Kurashige, W.; Negishi, Y., Acc. Chem. Res. 2018, 51 (12), 3114-3124. 289. Pei, Y.; Wang, P.; Ma, Z.; Xiong, L., Acc. Chem. Res. 2019, 52 (1), 23-33. 290. Tsuji, M.; Miyamae, N.; Lim, S.; Kimura, K.; Zhang, X.; Hikino, S.; Nishio, M., Cryst. Growth Des. 2006, 6 (8), 1801-1807. 291. Jin, R.; Zeng, C.; Zhou, M.; Chen, Y., Chem. Rev. 2016, 116 (18), 10346-10413. 292. Wu, Z.; Lanni, E.; Chen, W.; Bier, M. E.; Ly, D.; Jin, R., J. Am. Chem. Soc. 2009, 131 (46), 16672-16674. 293. Wilcoxon, J. P.; Provencio, P., J. Phys. Chem. B 2003, 107 (47), 12949-12957. 294. Suber, L.; Imperatori, P.; Pilloni, L.; Caschera, D.; Angelini, N.; Mezzi, A.; Kaciulis, S.; Iadecola, A.; Joseph, B.; Campi, G., Nanoscale 2018, 10 (16), 7472-7483. 295. Qu, F.; Li, N. B.; Luo, H. Q., J. Phys. Chem. C 2013, 117 (7), 3548-3555. 296. Purcell, E. M.; Torrey, H. C.; Pound, R. V., Phys. Rev. 1946, 69 (1-2), 37-38. 297. Macomber, R. S., A complete introduction to modern NMR spectroscopy. Wiley: 1998. 298. Abraham, R. J.; Fisher, J.; Loftus, P., Introduction to NMR spectroscopy. Wiley: 1988. 299. Akitt, J. W.; Mann, B. E., NMR and Chemistry: An introduction to modern NMR spectroscopy, Fourth Edition. Taylor & Francis: 2000.

85 300. Lambert, J. B.; Mazzola, E.; Ridge, C. D., Nuclear Magnetic Resonance Spectroscopy: An Introduction to Principles, Applications, and Experimental Methods. Wiley: 2019. 301. Ruiz-Morales, Y.; Schreckenbach, G.; Ziegler, T., Organometallics 1996, 15 (19), 3920- 3923. 302. Hulliger, J., Angew. Chem., Int. Ed. 1994, 33 (2), 143-162. 303. Sun, C.; Xue, D., CrystEngComm 2016, 18 (8), 1262-1272. 304. Bragg, W. L., Proc. Cambridge Philos. Soc. 1913, 17, 43-57. 305. Bacon, G. E.; Haar, D. T., X-Ray and Neutron Diffraction: The Commonwealth and International Library: Selected Readings in Physics. Elsevier Science: 2013. 306. RÖNTGEN, W. C., Science 1896, 3 (59), 227-231. 307. von Laue, M., J. Chem. Soc., Abstr. 1917, 112 (II), 166. 308. von Laue, M., Ber. Dtsch. Chem. Ges. 1917, 50, 8-20. 309. von Laue, M., Naturwissenschaften 1920, 8, 968-71. 310. Ibers, J. A., Location of Terminal Hydride Ligands in Transition Metal Hydrides. In Transition Metal Hydrides, AMERICAN CHEMICAL SOCIETY: 1978; Vol. 167, pp 26-35. 311. Woińska, M.; Grabowsky, S.; Dominiak, P. M.; Woźniak, K.; Jayatilaka, D., Sci. Adv. 2016, 2 (5), e1600192-e1600192. 312. Rutherford, E., Proc. Royal Soc. Lond. A 1920, 97 (686), 374-400. 313. Chadwick, J., Nature 1932, 129, 312. 314. Shull, C. G.; Wollan, E. O., Science 1948, 108 (2795), 69-75. 315. Shull, C. G.; Wollan, E. O., Naturwissenschaften 1949, 36, 291-296. 316. Shull, C. G.; Wollan, E. O.; Morton, G. A.; Davidson, W. L., Phys. Rev. 1948, 73, 842- 847. 317. Wollan, E. O.; Davidson, W. L.; Shull, C. G., Phys. Rev. 1949, 75, 1348-1352. 318. Wollan, E. O.; Shull, C. G., Phys. Rev. 1948, 73, 830-841. 319. Asano, H.; Hirabayashi, M., Neutron Diffraction Studies of Metal Hydrides. In Metal Hydrides, Bambakidis, G., Ed. Springer US: Boston, MA, 1981; pp 81-103. 320. Piccoli, P. M. B.; Koetzle, T. F.; Schultz, A. J., Comm. Inorg. Chem. 2007, 28 (1-2), 3-38. 321. Roithová, J.; Schröder, D., Coord. Chem. Rev. 2009, 253 (5), 666-677. 322. Hohenberg, P.; Kohn, W., Phys. Rev. 1964, 136 (3B), B864-B871. 323. Pyykkö, P., Chem. Rev. 1988, 88 (3), 563-594. 324. Almlofl, J.; Gropen, O., Relativistic effects in chemistry. In Reviews in computational chemistry, 1996; Vol. 8, p 203. 325. Hess, B. A.; Chandra, P., Phys. Scr. 1987, 36 (3), 412-415. 326. Jansen, G.; Hess, B. A., Z. Phys. D 1989, 13, 363-375.

86 327. Runge, E.; Gross, E. K. U., Phys. Rev. Lett. 1984, 52 (12), 997-1000. 328. Henderson, W.; McIndoe, J. S., Mass spectrometry of inorganic, coordination and organometallic compounds : tools -- techniques -- tips / William Henderson and J. Scott McIndoe. Chichester, England ; Hoboken, NJ : J. Wiley, c2005.: 2005. 329. Gross, J. H.; Roepstorff, P., Mass Spectrometry: A Textbook. Springer Berlin Heidelberg: 2011. 330. Johnson, G. E.; Hu, Q.; Laskin, J., Soft landing of complex molecules on surfaces. 2011; Vol. 4, pp 83-104. 331. Bleakney, W., Phys. Rev. 1929, 34, 157-160. 332. Nier, A. O., Rev. Sci. Instrum. 1947, 18, 398-411. 333. Harrison, A., Chemical Ionization Mass Spectrometry. CRC Press: 1983; p 156 pp. 334. Barber, M.; Bordoli, R. S.; Sedgwick, R. D.; Tyler, A. N., J. Chem. Soc., Chem. Commun. 1981, (7), 325-327. 335. Karas, M.; Bachmann, D.; Bahr, U.; Hillenkamp, F., Int. J. Mass Spectrom. Ion Processes 1987, 78, 53-68. 336. Karas, M.; Hillenkamp, F., Anal. Chem. 1988, 60 (20), 2299-2301. 337. Hillenkamp, F.; Karas, M., Methods Enzymol. 1990, 193, 280-95. 338. Yamashita, M.; Fenn, J. B., J. Phys. Chem. 1984, 88 (20), 4451-4459. 339. Loo, J. A.; Udseth, H. R.; Smith, R. D., Anal. Biochem. 1989, 179 (2), 404-412. 340. Cole, R. B.; Editor, Electrospray Ionization Mass Spectrometry: Fundamentals, Instrumentation, and Applications. Wiely: 1997; p 577 pp. 341. Ikonomou, M. G.; Blades, A. T.; Kebarle, P., Anal. Chem. 1990, 62 (9), 957-67. 342. Yamashita, M.; Fenn, J. B., J. Phys. Chem. 1984, 88 (20), 4671-4675. 343. Vetter, W., Biological Mass Spectrometry 1994, 23 (6), 379-379. 344. Aberth, W.; Straub, K. M.; Burlingame, A. L., Anal. Chem. 1982, 54 (12), 2029-2034. 345. Miller, J. M., Fast-Atom Bombardment Mass Spectrometry and Related Techniques. In Advances in Inorganic Chemistry, Emeleus, H. J.; Sharpe, A. G., Eds. Academic Press: 1984; Vol. 28, pp 1-27. 346. Bruce, M. I.; Liddell, M. J., Appl. Organomet. Chem. 1987, 1 (3), 191-226. 347. Wyatt, M. F., J. Mass Spectrom. 2011, 46 (7), 712-719. 348. Bailey, G. A.; Fogg, D. E., ACS Catal. 2016, 6 (8), 4962-4971. 349. Pettibone, J. M.; Hudgens, J. W., Small 2012, 8 (5), 715-725. 350. Pettibone, J. M.; Hudgens, J. W., Phys. Chem. Chem. Phys. 2012, 14 (12), 4142-4154. 351. Pettibone, J. M.; Hudgens, J. W., ACS Nano 2011, 5 (4), 2989-3002.

352. Hudgens, J. W.; Pettibone, J. M.; Senftle, T. P.; Bratton, R. N., Inorg. Chem. 2011, 50 (20), 10178-10189. 353. Pettibone, J. M.; Hudgens, J. W., J. Phys. Chem. Lett. 2010, 1 (17), 2536-2540. 354. Hudgens, J. W.; Senftle, T. P.; Bratton, R. N.; Bergeron, D. E., Formation and degradation of diphosphine-protected gold clusters. In Unpublished Work, 2010. 355. Bergeron, D. E.; Coskuner, O.; Hudgens, J. W.; Gonzalez, C. A., J. Phys. Chem. C 2008, 112 (33), 12808-12814. 356. Golightly, J. S.; Gao, L.; Castleman Jr, A. W.; Bergeron, D. E.; Hudgens, J. W.; Magyar, R. J.; Gonzalez, C. A., J. Phys. Chem. C 2007, 111 (40), 14625-14627. 357. Bergeron, D. E.; Hudgens, J. W., Journal of Physical Chemistry C 2007, 111 (23), 8195- 8201. 358. Gatlin, C. L.; Turecek, F. In Electrospray ionization of inorganic and organometallic complexes, Wiley: 1997; pp 527-570. 359. Jirasko, R.; Holcapek, M., Mass Spectrom. Rev. 2011, 30 (6), 1013-1036. 360. Di Marco, V. B.; Bombi, G. G., Mass Spectrom. Rev. 2006, 25 (3), 347-379. 361. McLuckey, S. A.; Wells, J. M., Chem. Rev. 2001, 101 (2), 571-606. 362. Cooks, R. G.; Beynon, J. H.; Caprioli, R. M.; Lester, G. R., Metastable Ions. Elsevier: 1973; p 296 pp. 363. McLafferty, F. W.; Editor, Tandem Mass Spectrometry. John Wiley and Sons: 1983; p 506 pp. 364. Busch, K. L.; Glish, G. L.; McLuckey, S. A., Mass Spectrometry/Mass Spectrometry: Techniques and Applications of Tandem Mass Spectrometry. VCH Publishers, Inc.: 1988; p 333 pp. 365. Douglas, D. J.; Frank, A. J.; Mao, D., Mass Spectrom. Rev. 2005, 24 (1), 1-29. 366. Mao, D.; Douglas, D. J., J. Am. Soc. Mass Spectrom. 2003, 14 (2), 85-94. 367. Sommer, H.; Thomas, H. A., Phys. Rev. 1950, 78, 806. 368. Sommer, H.; Thomas, H. A.; Hipple, J. A., Phys. Rev. 1951, 82, 697-702. 369. Comisarow, M. B.; Marshall, A. G., Chem. Phys. Lett. 1974, 26 (4), 489-490. 370. Comisarow, M. B.; Marshall, A. G., Chem. Phys. Lett. 1974, 25 (2), 282-283. 371. Feketeová, L.; O'Hair, R. A. J., Rapid Commun. Mass Spectrom. 2009, 23 (1), 60-64. 372. Zubarev, R. A.; Kelleher, N. L.; McLafferty, F. W., J. Am. Chem. Soc. 1998, 120 (13), 3265-3266. 373. Zubarev, R. A.; Haselmann, K. F.; Budnik, B.; Kjeldsen, F.; Jensen, F., Eur. J. Mass Spectrom. 2002, 8 (5), 337-349. 374. Paul, W., Angew. Chem., Int. Ed. 1990, 29 (7), 739-748.

88 375. Paul, W.; Steinwedel, H., Notizen: Ein neues Massenspektrometer ohne Magnetfeld. In Z. Nat. A, 1953; Vol. 8, p 448. 376. Paul, W.; Steinwedel, H. Separation and indication of ions with different specific charges. DE944900, 1956. 377. Louris, J. N.; Amy, J. W.; Ridley, T. Y.; Cooks, R. G., Int. J. Mass Spectrom. Ion Processes 1989, 88 (2), 97-111. 378. McLuckey, S. A.; Van Berkel, G. J.; Glish, G. L., J. Am. Chem. Soc. 1990, 112 (14), 5668- 5670. 379. O'Hair, R. A. J., Chem. Commun. 2006, (14), 1469-1481. 380. Gronert, S., J. Am. Soc. Mass Spectrom. 1998, 9 (8), 845-848. 381. Flores, A. E.; Gronert, S., J. Am. Chem. Soc. 1999, 121 (11), 2627-2628. 382. Colorado, A.; Brodbelt, J., J. Am. Soc. Mass Spectrom. 1996, 7 (11), 1116-1125. 383. Falvo, F.; Fiebig, L.; Dreiocker, F.; Wang, R.; Armentrout, P. B.; Schäfer, M., Int. J. Mass Spectrom. 2012, 330-332, 124-133. 384. Woodin, R. L.; Bomse, D. S.; Beauchamp, J. L., J. Am. Chem. Soc. 1978, 100 (10), 3248- 3250. 385. Okumura, M.; Yeh, L. I.; Lee, Y. T., J. Chem. Phys. 1985, 83 (7), 3705. 386. Jos, O.; André, J. A. v. R.; Gerard, M.; Gert von, H., Astrophys. J. 2000, 542 (1), 404. 387. Lemaire, J.; Boissel, P.; Heninger, M.; Mauclaire, G.; Bellec, G.; Mestdagh, H.; Simon, A.; Caer, S. L.; Ortega, J. M.; Glotin, F.; Maitre, P., Phys. Rev. Lett. 2002, 89 (27), 273002. 388. Bakker, J. M.; Sinha, R. K.; Besson, T.; Brugnara, M.; Tosi, P.; Salpin, J.-Y.; Maître, P., J. Phys. Chem. A 2008, 112 (48), 12393-12400. 389. Asmis, K. R.; Fielicke, A.; von Helden, G.; Meijer, G., Chapter 8 Vibrational spectroscopy of gas-phase clusters and complexes. 2007; Vol. 12, p 327-375. 390. Duncan, M. A., Annu. Rev. Phys. Chem. 1997, 48 (1), 69-93. 391. Barrow, R. F.; Crozet, P., Annual Reports on the Progress of Chemistry - Section C 1992, 89, 353-471. 392. Brodbelt, J., J. Am. Soc. Mass Spectrom. 2011, 22 (2), 197-206. 393. McLuckey, S.; Mentinova, M., J. Am. Chem. Soc. 2011, 22 (1), 3-12. 394. Brodbelt, J. S., Mass Spectrom. Rev. 1997, 16 (2), 91-110. 395. Green, M. K.; Lebrilla, C. B., Mass Spectrom. Rev. 1997, 16 (2), 53-71. 396. Osburn, S.; Ryzhov, V., Anal. Chem. 2013, 85 (2), 769-778. 397. Zhao, Y.-X.; Liu, Q.-Y.; Zhang, M.-Q.; He, S.-G., Dalton Trans. 2016, 45 (28), 11471-11495. 398. Li, X.-N.; Zou, X.-P.; He, S.-G., Chin. J. Catal. 2017, 38 (9), 1515-1527.

399. Waters, T.; O'Hair, R. A. J.; Wedd, A. G., J. Am. Chem. Soc. 2003, 125 (11), 3384-3396. 400. Robinson, P. S. D.; Khairallah, G. N.; Da Silva, G.; Lioe, H.; O'Hair, R. A. J., Angew. Chem., Int. Ed. 2012, 51 (16), 3812-3817. 401. Böhme, D. K.; Schwarz, H., Angew. Chem. Int. Ed. 2005, 44 (16), 2336-2354.

90 91 92 2 Synthesis, structure and gas- phase reactivity of the mixed silver hydride borohydride

nanocluster [Ag3(μ3-H)(μ3- Ph BH4)L 3]BF4 (LPh = bis(diphenylphosphino)methan e)

93 94 Addendum to Chapter 2

The published paper reports at page 18133 lines 2-3 of the right-hand column :

Ph + “[Ag3(H)(BH4)L ] (m/z 723), formed via sequential ligand losses (eqn (4) and (6)),…”

This is a typographical error and should read:

Ph + “[Ag3(H)(BH4)L ] (m/z 723), formed via sequential ligand losses (eqn (5) and (7)),…”.

95 96 Nanoscale

View Article Online PAPER View Journal | View Issue

Synthesis, structure and gas-phase reactivity of the mixed silver hydride borohydride nanocluster Cite this: Nanoscale, 2015, 7, 18129 Ph Ph [Ag3(µ3-H)(µ3-BH4)L 3]BF4 (L = bis(diphenylphosphino)methane)†

Athanasios Zavras,a Alireza Ariafard,a,b,c George N. Khairallah,a Jonathan M. White,a Roger J. Mulder,d Allan J. Cantyb and Richard A. J. O’Hair*a

Borohydrides react with silver salts to give products that span multiple scales ranging from discrete mono- + nuclear compounds through to silver nanoparticles and colloids. The cluster cations [Ag3(H)(BH4)L3] are observed upon electrospray ionization mass spectrometry of solutions containing sodium borohydride, Me silver(I) tetraﬂuoroborate and bis(dimethylphosphino)methane (L ) or bis(diphenylphosphino)methane Ph Ph (L ). By adding NaBH4 to an acetonitrile solution of AgBF4 and L , cooled to ca. −10 °C, we have been Ph able to isolate the ﬁrst mixed silver hydride borohydride nanocluster, [Ag3(µ3-H)(µ3-BH4)L 3]BF4, and structurally characterise it via X-ray crystallography. Combined gas-phase experiments (LMe and LPh) and Me + DFT calculations (L ) reveal how loss of a ligand from the cationic complexes [Ag3(H)(BH4)L3] provides a + Received 21st August 2015, change in geometry that facilitates subsequent loss of BH3 to produce the dihydride clusters, [Ag3(H)2Ln] Accepted 5th October 2015 (n = 1 and 2). Together with the results of previous studies (Girod et al., Chem. – Eur. J., 2014, 20, 16626), DOI: 10.1039/c5nr05690j this provides a direct link between mixed silver hydride/borohydride nanoclusters, silver hydride nano- www.rsc.org/nanoscale clusters, and silver nanoclusters.

2–4 Introduction via hydrogen and BH3 loss (eqn (2)). Hydrogen can also be liberated via reaction with methanol (eqn (3)). Ab initio calcu-

The chemistry of alkali metal borohydrides reacting with lations suggested that the decomposition of AgBH4 should silver salts has a rich history of over 60 years1 and continues yield silver hydride (eqn (4)).23 Finally, there have been several

Published on 16 October 2015. Downloaded by The University of Melbourne Libraries 01/02/2016 05:50:18. to yield a variety of silver containing products, including reports that discrete mononuclear complexes of AgBH4 could – – – silver borohydrides,2 4 silver hydrides5 10 and reduced silver be isolated when phosphine ligands were employed,2 4 ulti- – species.11 16 These products span multiple scales ranging mately leading to the X-ray crystallographic characterization of 2–4 4 from discrete mononuclear compounds through to ligand (Ph2MeP)3AgBH4 (Scheme 1(a)). – capped silver nanoclusters,9 17 silver nanoparticles, and colloids − 18–21 AgClO4 þ LiBH4 ! AgBH4 þ LiClO4 ð1Þ capped with various species including BH4 (Scheme 1). ,‡ In 1952 Wiberg and Henle reported that at −80 °C silver per- AgBH4 ! Ag þ 0:5H2 þ BH3 ð2Þ chlorate reacts with lithium borohydride to yield AgBH4 (eqn (1)), which decomposes when the temperature rises to −30 °C þ ! þ : þ ð Þ ð Þ AgBH4 3MeOH Ag 3 5H2 B OMe 3 3

AgBH ! AgH þ BH ð4Þ aSchool of Chemistry and Bio21 Institute of Molecular Science and Biotechnology, 4 3 The University of Melbourne, Melbourne, Victoria 3010, Australia. In the last few years, several new silver nanoclusters have E-mail: [email protected] been isolated from reaction mixtures containing silver salts, bThe School of Physical Sciences, University of Tasmania, Private Bag 75, Hobart, Tasmania 7001, Australia sodium borohydride and either anionic ligands or neutral cDepartment of Chemistry, Faculty of Science, Central Tehran Branch, Islamic Azad ligands and structurally characterised via X-ray crystallography.

University, Shahrak Gharb, Tehran, Iran Structural reports include (i) silver hydride clusters [Ag2(µ2-H)L2]- dCSIRO Manufacturing, Bayview Avenue, Clayton, VIC 3168, Australia 8 BF4 (where L = 1,3-bis(2,6-diisopropylphenyl)imidazolin- ‡There are other classes of silver nanoclusters that are formed without the use Ph 9 22 2-ylidene (Scheme 1(b)), [Ag3(µ3-H)(µ3-Cl)L 3]BF4 and of borohydride salts. Ph 10 Ph †Electronic supplementary information (ESI) available. See DOI: 10.1039/ [Ag3(µ3-H)L 3](BF4)2 (where L = bis(diphenylphosphino)- i 18 c5nr05690j methane), [{Ag7(µ4-H)(E2P(OR)2}6](R= Pr, E = Se),

This journal is © The Royal Society of Chemistry 2015 Nanoscale,2015,7,18129–18137 | 18129 97 View Article Online

Paper Nanoscale

− Scheme 1 Reactions of BH4 with silver salts to give products spanning multiple scales. (a)–(f) represent discrete isolated species whose structures have been determined via X-ray crystallography. The scale indicates ionic interactions or the outermost diameter of each silver cluster not inclusive of ligands. Counterions and hydrogen atoms have been omitted for clarity.

+5 + 6,7 [Ag8H(S2P(OEt)2)6] ,and[Ag11(H)(S2P(OEt)2)9] (Scheme 1(c)), immediate colour change from colourless to light yellow for 12 Ph Me (ii) the silver clusters Ag14(SC6H3F2)12(PPh3)8 (Scheme 1(d)), both L and L . ESI/MS analysis 5 minutes after the addition 2− Ph Ag16(dppe)4(SC6H3F2)14, {Ag32(dppe)5(SC6H4CF3)24} (dppe = of NaBH4 are shown in Fig. 1. The bulkier L ligand yields the 14 Ph + 1,2-bis(diphenylphosphino) ethane, Scheme 1(e)), abundant peak [Ag3(H)(BH4)L 3] m/z 1493 (Fig. 1a), con- 4− 13 [Ag44(SAr)30] (where ArS is an arylsulfide, Scheme 1(f)), firmed by high resolution mass spectrometry experiments (ESI i + 15 Me [Ag21{S2P(O Pr)2}12] , Ag29(BDT)12(PPh3)4 (BDT = 1,3-benzene- Fig. S1a and b†). In contrast the L yields a mixture of silver 16 − 17 Me 2+ Me + dithiol) and [Ag25(SR)18] . Importantly, Liu et al. have clusters assigned as [Ag3(H)L 3] m/z 366, [Ag3(H)(BH4)L 3]

shown that [{Ag7(µ4-H)(E2P(OR)2}6] (E = Se, S) are precursors to further growth into silver nanoparticles.24 “ ” Published on 16 October 2015. Downloaded by The University of Melbourne Libraries 01/02/2016 05:50:18. Here we report the mass spectrometry directed synthesis

of the first mixed silver hydride/borohydride cluster [Ag3(µ3-H)- Ph 25–27 (µ3-BH4)L 3]BF4, and its structural characterization by X-ray crystallography. DFT calculations indicate that the loss of a ligand (L) under collision induced dissociation (CID) conditions results in a change in cluster geometry that facili- − tates decomposition of the ligated BH4 via loss of BH3 (cf. eqn (2)).

Results and discussion

Electrospray ionization mass spectrometry (ESI/MS) was used to monitor the identity of cationic silver clusters formed upon mixing silver(I) tetrafluoroborate with diphosphine ligands in cooled acetonitrile solutions and subsequently treating with

excess NaBH4. The ligands bis(diphenylphosphino)methane (LPh) and bis(dimethylphosphino)methane (LMe) were added Fig. 1 Full LTQ ESI/MS for solution phase synthesis of silver hydride clusters protected by diphosphine ligands: (a) LPh; (b) LMe. Solutions to an acetonitrile solution of AgBF4 (1.9 mg, 5 mM) with a containing condensed phase silver clusters were diluted to 50 µM in ligand-to-metal ratio of 1 : 1 giving a clear solution. Addition of acetonitrile. Spectra were recorded 5 minutes after the addition of m z 15 equivalents of NaBH4 to each of these solutions gave an NaBH4. The most intense peak in the cluster represents the / value.

18130 | Nanoscale,2015,7,18129–18137 This journal is © The Royal Society of Chemistry 2015 98 View Article Online

Nanoscale Paper

Me + m/z 746 (Fig. 1b, ESI Fig. S2†) and [Ag3(BH4)2(MeCN)L 3] m/z change in solution, even at −15 °C, precluded the overnight or 801. Since clusters formed from LPh were less prone to longer acquisition of a 109Ag NMR spectrum. decomposition reactions in solution compared to those X-ray crystallography was used to determine the structure of Me 28 Ph Ph + formed from L , the synthesis of crystalline material suit- [Ag3(H)(BH4)L 3]BF4.§ The cation [Ag3(H)(BH4)L 3] which able for characterization was pursued for LPh. has crystallographic 3-fold symmetry, Fig. 2, consists of a trinuclear core with silver(I) ions occupying the vertices of an Structural characterization of (1) by ESI/MS, NMR and equilateral triangle. IR spectroscopy and X-ray crystallography The Ag(1)–Ag(1) distances that connect the edges of the tri- The crystals of (1) formed in the bulk synthesis were first ana- angle are 2.9100(3) Å. The hydride H which lies on a crystallo- lysed via ESI/MS in both the positive and negative ion mode. graphic 3-fold axis is 0.96 Å displaced from the plane defined The former gave an almost identical mass spectrum to that by the triangular silver(I) core and is coordinated to all silver(I)

shown in Fig. 1a (data not shown), while the latter gave an ions as a µ3-bridging ligand with a Ag(1)–H distance of − abundant signal due to the BF4 counter ion (ESI Fig. S3†). 1.93(3) Å and Ag(1)–H–Ag(1) angle of 97.5(3)°. Relative to the − IR spectroscopy confirmed the presence of both BF4 and − BH4 (ESI Fig. S4†). We next attempted to characterise 1 via various NMR experi- – † ments (ESI Fig. S5 S16 ). 1 was dissolved into cold CD3CN to produce a saturated solution and this solution immediately introduced into the pre-cooled NMR probe at −15 °C. The 1H NMR spectrum collected at −15 °C displayed a very broad

multiplet centred at 0.5 ppm, attributed to coordinated BH4, from which no fine structure could be resolved (ESI Fig. S5†). This may be due to (i) the fluxional nature of the binding of

BH4, (ii) the complex splitting patterns due to spin-active nuclei and isotopomers of silver (107/109Ag) and boron (10/11B) and (iii) the influence of the quadrupolar 10B nucleus. This signal collapsed into a broad singlet at 0.5 ppm under 11B-decoupling (ESI Fig. S6 and S7†). There was no apparent change to the signals upon 31P-decoupling (ESI Fig. S8 and S9†). The 1H NMR spectrum further displayed a broad multiplet centred at 4.5 ppm attributed to the coordinated hydride (ESI Fig. S5†), the peaks of which sharpened slightly under 31P-decoupling (ESI Fig. S8 and S9†). The 1H-decoupled 31P NMR spectrum displayed a broad peak at 0.56 ppm, shifted downfield from the free ligand resonance at 23.1 ppm (ESI Fig. S10†). The 1H-decoupled 11B NMR spectrum displays Published on 16 October 2015. Downloaded by The University of Melbourne Libraries 01/02/2016 05:50:18. – two resonances at 0.57 and 41.49 ppm attributable to BF4 and † 19 11 BH4, respectively (ESI Fig. S11 ), as confirmed via the F- B and 1H-11B HSQC NMR experiments (ESI Fig. S12 and S13† respectively). The 1H-decoupled 19F NMR spectrum displays a − – resonance at 150.16 ppm corresponding to BF4 (ESI Fig. S14†). The 1H-decoupled 13C NMR spectrum displays resonances attributable to coordinated phosphine ligand (ESI Fig. S15†). Heating the sample from −15 °C to +25 °C in the NMR probe enabled the collection of 1H data at various tempera-

tures. The most obvious change in the spectra with time at Fig. 2 ORTEP-3 representations of: (a) the cation present in [Ag3(µ3-H)- Ph −15 °C and then upon heating was the increase in the inten- (µ3-BH4)L 3]BF4 and, (b) the trinuclear silver hydride/borohydride ‘ ’ sity of the singlet at 4.56 ppm which is attributable to dis- Ag3(µ3-H)(µ3-BH4) core, where phenyl rings are omitted for clarity. Dis- placement ellipsoids are set at the 50% probability level. Ag(1)–Ag(1) solved H2 and the corresponding reduction in the intensity of – – – † 2.9100(3), Ag(1) P(1) 2.4483(5), Ag(1) P(2) 2.4486(5), Ag(1) H 1.93(3), the coordinated hydride signal (ESI Fig. S16 ). Ag(1)–H(1a) 2.17(3), B(1)–H(1a) 1.10(3), B(1)–H(1b) 1.07(6). This series of experiments required the preparation of several diﬀerent samples, as it was noted that 1 appears to

undergo decomposition/reactions in these highly concentrated Ph §The crystallographic information file for [Ag3(µ3-H)(µ3-BH4)L 3]BF4 has been solutions, ultimately resulting in precipitation of a black deposited at the Cambridge Crystallographic Data Centre and assigned the code: material after approximately 3 hours at 25 °C. This rapid CCDC 1419573.

This journal is © The Royal Society of Chemistry 2015 Nanoscale,2015,7,18129–18137 | 18131 99 View Article Online

Paper Nanoscale

+ µ3-hydride H, the opposing face of the trinuclear core has a µ3 Unimolecular gas-phase chemistry of [Ag3(H)(BH4)L3] borohydride with distorted tetrahedral geometry, the boron Given that AgBH4 is known to undergo thermal decomposition lies on the 3-fold axis of symmetry and three symmetry related 1,23 reactions that liberate BH3 (eqn (2)), we were interested in hydrides (H1a) coordinate to the three silver atoms. Related examining whether such reactions occur in the gas phase for µ3-borohydride binding to the metal triangle of Fe(CO)3 frag- isolated, stoichiometrically well defined cluster cations. Thus, ments has been reported for the trinuclear cluster [Fe3(µ-H)- CID was carried out in a 2D linear ion trap to probe the low- 27 – − (µ3-BH4)(CO)9]. The boron hydrogen bond lengths for BH4 energy fragmentation pathways of [Ag (H)(BH )L ]+, where the – – 3 4 3 are: B(1) H(1a) 1.10(3) Å and B(1) (H1b) 1.07(6) Å. There are cluster identity has been confirmed by high resolution mass − – two types of bond angles in the BH4 tetrahedron: (i) H(1a) spectrometry (HRMS). – – – B(1) H(1a) 110.1(16)° and (ii) H(1a) B(1) H(1b) 108.9(16)°. The Mass selection and subsequent CID of [Ag (H)(BH )LPh ]+ – – 3 4 3 tetrahedral face H(1a) H(1a) H(1a), is parallel to the larger (m/z 1493, HRMS (ESI Fig. S1a and b†)) to ca. 50% relative Ag(1)–Ag(1)–Ag(1) plane. Each of the three hydrogen atoms of Ph + − intensity yields [Ag3(H)(BH4)L 2] (m/z 1109, HRMS (ESI BH4 , H(1a), are individually coordinated to one silver(I) ion Fig. S1c and d†)) via neutral ligand loss (eqn (5)) as the main where H(1a)–Ag(1) is 2.17(3) Å. The core is surrounded by Ph + fragmentation channel. [Ag3(H)2L 2] (m/z 1095, HRMS (ESI three µ2-bridging phosphine ligands which coordinate to two Fig. S17a and b†)), [Ag (H)(BH )LPh]+ (m/z 723, (ESI Fig. S1e – 3 4 silver(I) ions through P(1) and P(2): Ag(1) P(1) is 2.4483(5) Å and f†)) and [Ag (H) LPh]+ (m/z 709, (ESI Fig. S17c and d†)); – 3 2 and Ag(1) P(2) is 2.4486(5) Å. The distance between P(1) and [Ag (H)LPh]+ (m/z 601) and; [AgLPh]+ (m/z 491) are all also – – 2 P(2) within the chelate ring is 3.041 Å and the P(1) C(20) P(2) observed. angle is 110.8(1)°. P(1) is below the plane of the trinuclear Me + The main fragmentation channel of [Ag3(H)(BH4)L 3] (m/z core and both C(20) and P(2) are above the plane with C(20) 747, HRMS (ESI Fig. S2†)) involves neutral ligand loss (eqn (5)) above P(2). Me + to form [Ag3(H)(BH4)L 2] (m/z 610, Fig. 3b). Other ions The phenyl rings extending from the phosphorus atoms Me + Me + observed include [Ag3(H)2L 2] (m/z 596), [Ag3(H)(BH4)L ] adopt one of three distinct geometrical conformations. Both Me + (m/z 475) and [Ag3(H)2L ] (m/z 461). P(1) and P(2) have one phenyl ring towards the µ3-hydride. Energy-resolved CID (ERCID) was used in a 3D ion trap to A pseudoequatorial phenyl ring in regards to the Ag3 triagnular determine whether the product ions in Fig. 3a are due to plane extends from P(1). One phenyl ring from each P(2) Ph + primary fragmentation pathways of [Ag3(H)(BH4)L 3] or extends toward the µ3-borohydride. secondary fragmentation of primary fragment ions (ESI Ph + Ph 2+ The structures of [Ag3(µ3-H)(µ3-BH4)L 3] , [Ag3(µ3-D)L 3] Fig. S18†). The onset of ligand loss (eqn (5)) begins at ca. 0.6 V Ph + (ref. 10) and [Ag3(µ3-H)(µ3-Cl)L 3] (ref. 9) are compared in and continues to steadily increase up until 0.8 V (ESI Table 1. All structures consist of a trinuclear silver(I) core and Fig. S18†). An increase in the collision voltage beyond this Ph 2+ maintain a triangular geometry. The dication [Ag3(µ3-D)L 3] Ph + point results in the consumption of [Ag3(H)(BH4)L 2] (m/z has the longest Ag–Ag and Ag–P interactions at 3.1193 and Ph + 1109) and the increase of [Ag3(H)2L 2] (m/z 1095). These 2.4632 Å respectively. All hydrides coordinate as µ3-bridging results suggest that the primary product ion upon CID of – ligands where the longest Ag H interaction exists in [Ag3(µ3-H)- [Ag (H)(BH )LPh ]+ arises from ligand loss and that ions of Ph + 3 4 3 (µ3-Cl)L 3] at 1.91(2) Å. − The BF4 counterion is disordered over two crystallographic Published on 16 October 2015. Downloaded by The University of Melbourne Libraries 01/02/2016 05:50:18. special positions, a −3 site with normal 1/6 occupancy and on a 3-fold axis with 50% the normal occupancy of 1/3, with the same position occupied the other 50% by a molecule of acetonitrile. The relative occupancy at these special positions were supported by SQUEEZE calculations29 which were not used to remove these disordered components. The Ag bound hydride

and BH4 hydrides were located on Fourier diﬀerence maps and refined isotropically without restraint.

Table 1 A comparison of selected bond lengths (Å) and angles (°) with Ph 9 estimated standard deviations in parentheses for [Ag3(µ3-D)L 3](BF4)2, Ph 10 Ph [Ag3(µ3-H)(µ3-Cl)L 3]BF4, and [Ag3(µ3-H)(µ3-BH4)L 3]BF4

[Ag3(µ3-D)- [Ag3(µ3-H)- [Ag3(µ3-H)- Ph 2+ Ph + Ph + L 3] (µ3-Cl)L 3] (µ3-BH4)L 3]

Ag–Ag 3.1193 2.8988(2) 2.9100(3) Ph + m z Me + Ag-(H/D) 1.83 1.91(2) 1.93(3) Fig. 3 LTQ CID of (a) [Ag3(H)(BH4)L 3] , / 1493; (b) [Ag3(H)(BH4)L 3] , m z m z Ag-(Cl/BH4) NA 2.859(1) 2.17(3) / 747. The most intense peak in the cluster is represented by the / Ag–P 2.4632 2.4421(9) 2.4486(5) value. *Refers to the mass-selected precursor ion.

18132 | Nanoscale,2015,7,18129–18137 This journal is © The Royal Society of Chemistry 2015 100 View Article Online

Nanoscale Paper

+ lower m/z are from subsequent secondary fragmentation of Unimolecular gas-phase chemistry of [Ag3(H)(BH4)L] Ph + [Ag3(H)2L 2] (m/z 1095). Ph + [Ag3(H)(BH4)L ] (m/z 723), formed via sequential ligand þ þ losses (eqn (4) and (6)), was mass selected and allowed to ½Ag3ðHÞðBH4ÞL3 !½Ag3ðHÞðBH4ÞL2 þ L ð5Þ undergo CID. The sole fragmentation channel observed is due

to the loss of neutral BH3 (eqn (8)) (ESI Fig. S20†). Unimolecular gas-phase chemistry of [Ag (H)(BH )L ]+ 3 4 2 þ þ ½Ag ðHÞðBH ÞL !½Ag ðHÞ L þ BH ð8Þ The primary product ions formed via ligand loss (eqn (5), 3 4 3 2 3 Fig. 3) were mass selected and subjected to CID in the 2D ion + trap (Fig. 4). Mass selection and subsequent CID of [Ag3(H)- Unimolecular gas-phase chemistry of [Ag3(H)2Ln] Ph + (BH4)L 2] (m/z 1109, Fig. 4a) to ca. 30% relative intensity The dihydride clusters [Ag (H) LPh ]+ where n = 2, 1 were sub- Ph + 3 2 n yields [Ag3(H)2L 2] (m/z 1095) via neutral borane loss jected to ERCID using a 3D ion trap (ESI Fig. S21 and S22†). Ph + (eqn (5)) as a minor fragmentation channel and [Ag3(H)2L ] The major primary fragmentation of [Ag (H) LPh ]+ (m/z 1095) Ph + 3 2 2 (m/z 709) as the main fragmentation channel; (ii) [Ag2(H)L ] occurs via ligand loss (eqn (9)), with an onset requiring Ph + (m/z 601) and; (iii) [AgL ] (m/z 491) are all also observed. The ca. 0.6 V. In contrast, The major primary fragmentation of Me + main fragmentation channel upon CID of [Ag3(H)2L 2] Ph + [Ag3(H)2L ] (m/z 708) involves AgH loss, as previously (m/z 596, Fig. 4b) involves neutral borane loss (eqn (6)). Other described,32 with the onset of fragmentation occurring at Me + ions observed include [Ag3(H)(BH4)L ] (m/z 475), and ca. 0.4 V. Me + [Ag3(H)2L ] (m/z 461). ½ ð Þ þ !½ ð Þ þ þ ð Þ Once again, ERCID was used in a 3D ion trap to determine Ag3 H 2L2 Ag3 H 2L L 9 which of the product ions observed in Fig. 4a were primary þ þ † Ph + ½Ag3ðHÞ L !½Ag2ðHÞL þ AgH ð10Þ (ESI Fig. S19 ). [Ag3(H)2L 2] (m/z 1095) begins to appear at 2 Ph + 0.5 V upon CID [Ag3(H)(BH4)L 2] (m/z 1109), which corresponds to BH loss (eqn (6)). A minor primary fragmentation − 3 Computational study of BH4 decomposition triggered via channel assigned to neutral ligand loss (eqn (7)) is observed at Me + ligand loss in the clusters [Ag3(H)(BH4)L n] (n =1–3) around 0.6 V. Although decomposition reactions of co- To better understand how the number of diphosphine ligands, ordinated ligands in metal complexes and clusters have been + 30,31 n, in the clusters, [Ag3(H)(BH4)Ln] , influence the competition well studied in the gas-phase, this appears to be the first − between decomposition of the ligated BH (eqn (11)) versus experimental report on the gas-phase decomposition of a co- 4 − loss of a ligand (eqn (12)), we turned to DFT calculations to ordinated BH4 ligand via BH3 loss. This reaction is related to examine the structures and energetics of the reactants and pro- that described by Wiberg and Henle (eqn (2)).1 ducts of eqn (11) and (12) for the case of clusters containing ½ ð Þð Þ þ !½ ð Þ þ þ ð Þ Me Me + Ag3 H BH4 L2 Ag3 H 2L2 BH3 6 L ligands. The initial geometry for [Ag3(H)(BH4)L 3] was that related to the core structure from the X-ray structure for 1. ½ ð Þð Þ þ !½ ð Þð Þ þ þ ð Þ Ag3 H BH4 L2 Ag3 H BH4 L L 7 Thus changing the phosphine substitutent from Ph to Me has little eﬀect on the core structure. To calculate fragment ion structures, either BH or LMe was removed and the resultant Published on 16 October 2015. Downloaded by The University of Melbourne Libraries 01/02/2016 05:50:18. 3 fragment was allowed to fully optimise.¶k ½ ð Þð Þ þ !½ ð Þ þ þ ð Þ Ag3 H BH4 Ln Ag3 H 2Ln BH3 11 þ þ ½Ag3ðHÞðBH4ÞLn !½Ag3ðHÞðBH4ÞLn1 þ L ð12Þ

− + To understand why BH4 in [Ag3(H)(BH4)L3] bridges all three Ag atoms via three separate two-electron, two-centre

¶While there is no significant change in the DFT calculated gas-phase structure Ph + of [Ag3(H)(BH4)L 3] compared to the X-ray structure, detailed discussions of the calculations are limited to the LMe systems, as the Ag–P bond energies in Ph + [Ag3(H)(BH4)L 3] are overestimated using the M06 functional. The same is also true when the B3LYP-D3BJ functional is applied. For example, the CID results Ph + show that loss of the phosphine ligand from [Ag3(H)(BH4)L 3] is preferred to

the BH3 loss while both the M06 and B3LYP-D3BJ functionals predict a reverse trend. This inconsistency is likely to arise from the overestimation of the magni- Ph + tude of dispersive interactions in [Ag3(H)(BH4)L 3] due to the presence of the Ph + Me + Fig. 4 LTQ CID of: (a) [Ag3(H)(BH4)L 2] , m/z 1109); (b) [Ag3(H)(BH4)L 2] , aromatic rings. m z m z / 610. The most intense peak in the cluster is represented by the / kWe have attempted to locate transition states for the BH3 loss reactions. In all value. *Refers to the mass-selected precursor ion. #Refers to back- cases we have been unable to locate a transition state and attempts led to the

ground noise. loss of BH3.

This journal is © The Royal Society of Chemistry 2015 Nanoscale,2015,7,18129–18137 | 18133 101 View Article Online

Paper Nanoscale

(2e,2c) bonds, the structure and bonding of this cluster was – analysed based on simple electron counting rules.33 37 As expected, a transition metal centre with a d10 electron configuration has four empty orbitals. Two of these four are always available for ligand coordination to give linear com- + plexes [ML2] . However, the availability of the other two orbitals depends on the identity of the L ligands. For example, if – – + the L M L bond angle in [ML2] using the bidentate ligands is forced to be bent, the two extra orbitals become available and + as a result the tetrahedral complex [ML4] is formed. Also, the monodentate L ligands with relatively weak σ-donor abilities increase the possibility of all four orbitals on the metal centre being available. + In [Ag3(H)(BH4)L3] , the presence of the three bidentate phosphine ligands render all the four empty orbitals on three Ag centres susceptible to coordination. In such a case, the cluster has 3×4=12available orbitals. Six of these twelve orbitals are occupied by the phosphine ligands. Three of them are

involved in interaction with the hydride ligand (a µ3-bridging ligand) via a four-centre two-electrons bonding mode. Finally, the last three orbitals on Ag centres overlap with three filled − − B–H σ orbitals of BH4 , leading to coordination of BH4 in + µ3-form. The 50 valence electron [Ag3(H)(BH4)L3] cluster is not expected to have direct metal–metal interactions, consist- 10 ent with other related M3L6 clusters where M has a d electron Fig. 5 DFT calculated structures and energetics for the competition 35 Me configuration. This is highlighted by an examination of its between BH3 loss and L ligand loss. Hydrogen atoms on the L ligands HOMO, which suﬀers from the Ag–Ag anti-bonding inter- are omitted for clarity.

actions derived from the silver dxz orbitals (Fig. 6). The short – – Ag Ag bond distances in [Ag3(H)(BH4)L3] (2.971 2.993 Å) can be mainly rationalised by the presence of the hydride ligand that creates the four-centre two-electron bonds with the Ag centres. A similar metal–metal bond distance was also − observed by Harvey et al. in the [Pd3(H2PCH2PH2)3(CO)(H)] cluster (2.932 Å), where the corresponding anti-bonding skeletal molecular orbitals are also fully occupied.38 With regards to the unimolecular fragmentation chemistry, Published on 16 October 2015. Downloaded by The University of Melbourne Libraries 01/02/2016 05:50:18. the calculations indicate that the first loss of the ligand, Me + L (eqn (12)), from [Ag (H)(BH )L ] results in transferring the Me + 3 4 3 Fig. 6 Depiction of the DFT calculated HOMO of [Ag3(H)(BH4)L 3] . hydride to Ag2 and causes this centre to adopt a mainly linear structure with a very weak interaction with Ag1 (Fig. 5). The presence of the very strong σ donating hydride ligand on the 2 2 Me + Ag centre makes the two empty orbitals on Ag less available fore, each Ag(I) centre in [Ag3(H)(BH4)L ] experiences a − and thus does not allow BH4 to strongly interact with them. linear two coordinated environment. − 1 3 In this case, BH4 is only able to interact with the Ag and Ag Our calculations show that, in excellent agreement with the 2 centres to give a µ -coordination mode (Fig. 5). If we ignore the CID data, the BH3 loss becomes easier as the number of Me + weak interactions between Ag centres in [Ag3(H)(BH4)L 2] , ligands, L, decreases. The loss of the ligand, L, increases the the Ag1 and Ag3 centres can be considered as three coordinate electron deficiency of the metal centres, leading to the stronger Me + 2 − centres. For [Ag3(H)(BH4)L 2] , the Ag –H σ orbital interacts coordination of BH4 to the Ag centres, as evident from the 3 – – with one of the empty orbitals on Ag and creates a 3-centre shorter Ag H(BH4) and longer B H bond distances in [Ag3(H)- Me + 2-electron bond between two silver(I) ions via aµ2-bridging (BH4)L ] (Fig. 5). In other words, ligand loss enhances the hydride ligand. acidity of Ag centres and makes them more prone to compete Me From loss of a second L ligand, bonding in the product with BH3 for hydride abstraction. The better the competition, Me + [Ag3(H)(BH4)L ] can be viewed as interaction of the linear the easier the BH3 loss. 3 − Me 2+ complex [BH4–Ag –H] with [Ag2L ] (Fig. 5). In this cluster, In contrast to BH3 loss (eqn (11)) neutral ligand loss 3 2 Me + ﬃ an Ag –H σ orbital interacts with an empty orbital on Ag , and (eqn (12)) from [Ag3(H)(BH4)L ] is more di cult than that from − 3 1 – Me + Me + BH4 bridges Ag to Ag through three of its B H bonds. There- [Ag3(H)(BH4)L 3] and [Ag3(H)(BH4)L 2] , supported by the

18134 | Nanoscale,2015,7,18129–18137 This journal is © The Royal Society of Chemistry 2015 102 View Article Online

Nanoscale Paper

DFT calculations. This diﬀerence can be rationalised in terms silver nanocluster containing a “captured” borohydride anion, 39 − of the molecular orbital approach. In general, the HOMO of and may have relevance to binding of BH4 to silver nano- 10 20,21 40 d complexes (MLn) with a coordination number greater than particle surfaces, or bulk silver metal surfaces. Given

two (n >2)suﬀers from a slight anti-bonding interaction that nanoclusters such as [{Ag7(µ4-H)(E2P(OR)2}6] are pre- between L and M, leading to weakening of the M–L bonds. cursors to further growth into silver nanoparticles,21 it will be

However, this anti-bonding interaction disappears in linear interesting to establish whether the nanoclusters [Ag3(µ3-H)- 10 Ph 9 Ph 10 d -ML2 complexes, causing the M–L bonds in ML2 to be (µ3-Cl)L 3]BF4, [Ag3(µ3-H)L 3](BF4)2, and [Ag3(µ3-H)- Ph much stronger than those in ML3 and ML4. As mentioned (µ3-BH4)L 3]BF4 can further grow into silver nanoparticles. Me + Me + 32 above, clusters [Ag3(H)(BH4)L 3] and [Ag3(H)(BH4)L 2] The current and previous gas-phase experiments and DFT + have the Ag centres not present as two-coordinate, and thus calculations on [Ag3(H)2−x(BH4)xLn] clusters provide a direct the loss of the ligand, L, from these clusters is relatively easy. link between mixed hydride/borohydride silver clusters (x = 1), Me + 32 By contrast, all the Ag centres in [Ag3(H)(BH4)L ] are mainly dihydride silver clusters (x =0) and silver clusters via dis- two-coordinate, thereby not having the relevant anti-bonding crete unimolecular reactions occurring for isolated clusters interaction, forming very strong M–L bonds. (Scheme 2). Thus CID triggers loss of the ligand, L (eqn (12)),

resulting in a change in the binding mode(s) of the H and BH4 ligands (Fig. 5). Perhaps related reactions occur at the surfaces Conclusions of silver nanoparticles, which might drive the development of catalysts for hydrogen storage applications.41 The sodium borohydride induced reduction of silver(I) salts to form nanoparticles has been described as a “black-box” synthesis.19 While it is now well established that there are – Experimental diﬀerent growth stages,18 20 the actual molecular species associated during growth to nanoparticles and the mecha- Synthesis of solution phase silver clusters for MS analyses nisms for growth are not fully understood. By studying the for- Silver(I) tetrafluoroborate (1.9 mg, 0.010 mmol) and bis(di- mation and reactions of small ligand protected nanoclusters, phenylphosphino)methane (3.8 mg, 0.010 mmol) in 20 mL we are able to better understand the fundamental interactions acetonitrile were added to a 25 mL Quickfit Erlenmeyer flask between silver salts and borohydride. equipped with a magnetic stirrer and stopper. The solution − We have previously shown that BH4 is a source of hydride was cooled to ca.−10 °C by immersing the reaction flask in an Ph 2+ Ph + for [Ag3(µ3-H)L 3] (ref. 10) and [Ag3(µ3-H)(µ3-Cl)L 3] (ref. 9) ice/water bath above the solvent level. All reagents were kept in at ambient conditions, however at −10 °C the decomposition the dark and flasks covered in foil. Sodium borohydride − of BH4 can be prevented and coordination to silver(I) ions can (5.7 mg, 0.150 mmol) was added as a powder and the solution Ph occur to yield [Ag3(µ3-H)(µ3-BH4)L 3]BF4. This is the first changed colour from clear to light yellow.

Ph Synthesis of crystalline [Ag3(H)(BH4)(L )3]BF4 (1)

Silver(I) tetrafluoroborate (194 mg, 1.0 mmol) and bis(diphenylphosphino)methane (384 mg, 1.0 mmol) in 100 mL aceto- Published on 16 October 2015. Downloaded by The University of Melbourne Libraries 01/02/2016 05:50:18. nitrile were added to a 250 mL Quickfit round bottomed flask equipped with a magnetic stirrer and stopper. The solution was cooled to −10 °C by immersing the reaction flask in an ice/water bath above the solvent level. All reagents were kept in the dark and flasks covered in foil. Sodium borohydride (57.0 mg, 1.50 mmol), was added as a powder and the solution changed colour from clear to light yellow over ca. 5 minutes. The solution was filtered after stirring for 3 hours and frozen solid by immersing the flask in liquid nitrogen. While frozen, 100 mL of diethylether was added and the flask moved to the fridge. After 72 hours crystalline material was formed and characterised by X-ray crystallography.

NMR spectroscopy experiments The NMR experiments were performed on a Bruker Avance 400 NMR spectrometer (400.13 MHz 1H frequency) equipped with Scheme 2 Direct link established between mixed hydride/borohydride a 5 mm triple resonance broadband probe (BB/2H–1H/19F). clusters (Black), dihydride clusters (Blue) and “all metal” clusters (Red) based on gas-phase unimolecular fragmentation reactions of mass Solutions for analysis by NMR were prepared by dissolving Ph selected clusters reported here using CID and in ref. 32 using laser- [Ag3(µ3-H)(µ3-BH4)L 3]BF4 in 0.6 ml of deuteroacetonitrile. induced dissociation (LID). NMR experiments were performed with the sample held at

This journal is © The Royal Society of Chemistry 2015 Nanoscale,2015,7,18129–18137 | 18135 103 View Article Online

Paper Nanoscale

temperatures between −15 °C and +25 °C (±0.1 °C). Chemical structure was solved by direct methods and diﬀerence Fourier shifts for 1H experiments are referenced to the residual proto- synthesis.43 The thermal ellipsoid plot was generated using 11 44 45 nated solvent signal (CD2HCN, δ 1.94 ppm); B externally the program ORTEP-3 integrated within the WINGX suite 13 − referenced to BF3·OEt2 capillary in CD3CN; C referenced to of programs. The BF4 counterion was disordered over two δ 19 − the solvent signal (CD3CN, 1.39 ppm); F externally refer- crystallographic special positions, a 3 site with normal 1/6 31 enced to a CFCl3 in CD3CN (δ 0.00 ppm); P externally refer- occupancy and on a 3-fold axis with 50% the normal occu-

enced to a 85% H3PO4 capillary in CDCl3 (δ 0.00 ppm). One- pancy of 1/3, with the same position occupied the other 50%

and two-dimensional NMR experiments were acquired using by a molecule of acetonitrile. The Ag bound hydride and BH4 standard Bruker library pulse sequences. hydrides were located on Fourier diﬀerence maps and refined isotropically without restraint. Mass spectrometry Crystal data for 1:C75H71B2F4P6Ag3·(0.25 CH3CN) M = Mass spectra were recorded using a Finnigan hybrid linear 1589.63, T = 130.0(2) K, λ = 1.54184 Å, cubic, space group Paˉ3, 3 −3 quadrupole (LTQ) Fourier transform ion cyclotron resonance a = 24.1922(1) Å, V = 14158.79(18) Å , Z =8,Dc = 1.491 mg M − (FTICR) mass spectrometer. The silver clusters prepared in the μ(Cu-Kα) 8.296 mm 1, F(000) 6428, crystal size 0.17 × 0.16 × solution phase were diluted to 50 µM and introduced into 0.09 mm. 67033 reflections measured, 4996 independent

the mass spectrometer via a syringe pump set at a flow rate of reflections (Rint = 0.0442), the final R was 0.0279 [I >3(I) 4853 − 5 µL min 1 to the ESI capillary. The ESI conditions used, for data] and wR(F) (all data) was 0.0735. optimum intensity of the target ions, typically were: spray – voltage, 4.2 5.0 kV, capillary temperature, 250 °C, nitrogen Density functional theory sheath gas pressure, 5 (arbitrary units), capillary voltage 25 V, – Computational details.46 53 Gaussian 09 46 was used to fully tube lens voltage 15 V. Selected ions were transferred to the optimise all the structures reported in this paper at the M06 FTICR cell for accurate mass measurement with the use of level of density functional theory.47,48 The effective-core poten- selected ion monitoring (SIM) and selected reaction monitor- tial of Hay and Wadt with a double-ξ valence basis set ing (SRM) to obtain the most reliable results. The unimolecu- (LANL2DZ) was chosen to describe Ag. The 6-31G(d) basis set lar fragmentation of silver clusters was examined via CID. The was used for other atoms. Polarization functions were also mass-selected precursor ion was depleted to 10–20% using a added for Ag (ξf = 1.611). This basis set combination will be normalised collision energy typically between 20–25% and a referred to as BS1. To further refine the energies obtained mass selection window of 15 Th to isolate the full range of iso- from the M06/BS1 calculations, we carried out single-point topes due to boron and silver isotopes. energy calculations for all of the structures with a larger basis Energy resolved CID experiments were carried out using a set (BS2) at the M06 level of theory. BS2 utilises the def2-TZVP Finnigan 3D ion trap (LCQ) mass spectrometer. The method of basis set on all atoms. Effective core potentials including Broadbelt was adapted.42 The silver clusters were diluted to scalar relativistic effects were used for silver atom. We have 50 µM and introduced into the mass spectrometer via asyringe − used the corrected potential energies obtained from the M06/ pump set at 5 µL min 1 through a Finnigan ESI source. The BS2//M06/BS1 calculations throughout the paper unless other- source conditions used for optimum intensity of the target wise stated. ions were: spray voltage 4.5–5.1 kV, capillary temperature Published on 16 October 2015. Downloaded by The University of Melbourne Libraries 01/02/2016 05:50:18. 200 °C, nitrogen sheath gas pressure, 50 (arbitrary units), capillary voltage 30 V, tube lens voltage −55 V. The mass- selected precursor ion was isolated with a mass selection Acknowledgements window of 15 Th. The normalised collision energy (NCE) was increased incrementally by 1.0% typically starting from a NCE We thank the ARC for financial support via grants DP1096134 where no fragmentation is observed, until reaching the NCE (to GNK) and DP150101388 (to RAJO and AJC). The authors required for depleting the precursor ion to <5% relative inten- gratefully acknowledge the generous allocation of computing sity. The NCE was converted to an amplitude of the resonance time from the University of Tasmania and the National Com- excitation RF voltage (tick amp) as described in the ESI.† The puting Infrastructure. We thank Assoc. Prof. Paul Donnelly for relative intensity of precursor and product ions were plotted as useful discussions. a function of the increasing amplitude to determine: (i) the onset of precursor fragmentation and (ii) the assignment of product ions as primary or secondary fragments of the mass- Notes and references selected silver cluster. 1 E. Wiberg and W. Henle, Z. Naturforsch., B: J. Chem. Sci., Crystallography 1952, 7, 575–576. Intensity data for compound 1 was collected on an Oxford 2 J. C. Bommer and K. W. Morse, Inorg. Chem., 1980, 19, Diffraction SuperNova CCD diffractometer using Cu-Kα radi- 587–593. ation, the temperature during data collection was maintained 3 F. Cariati and L. Naldini, Gazz. Chim. Ital., 1965, 95, 201– at 130.0(1) using an Oxford Cryostream cooling device. The 205.

18136 | Nanoscale,2015,7,18129–18137 This journal is © The Royal Society of Chemistry 2015 104 View Article Online

Nanoscale Paper

4 E. B. Lobkovskii, M. Y. Antipin, A. P. Borisov, 28 A. J. Clark, A. Zavras, G. N. Khairallah and R. A. J. O’Hair, V. D. Makhaev, K. N. Semenenko and Y. T. Struchkov, Int. J. Mass Spectrom., 2015, 378,86–94. Koord. Khim., 1981, 7, 307–310. 29 P. Van der Sluis and A. L. Spek, Acta Crystallogr., Sect. A: 5 C. W. Liu, H.-W. Chang, C.-S. Fang, B. Sarkar and Found. Crystallogr., 1990, A46, 194–201. J.-C. Wang, Chem. Commun., 2010, 46, 4571–4573. 30 R. A. J. O’Hair, Gas Phase Ligand Fragmentation to 6 C. W. Liu, H.-W. Chang, B. Sarkar, J.-Y. Saillard, S. Kahlal Unmask Reactive Metallic Species, in Reactive Intermediates. and Y.-Y. Wu, Inorg. Chem., 2010, 49, 468–475. MS Investigations in Solution, ed. L. S. Santos, Wiley-VCH, 7 C. W. Liu, P.-K. Liao, C.-S. Fang, J.-Y. Saillard, S. Kahlal and Weinheim, 2010, ch. 6, pp.199–227, ISBN: 978-3-527-32351-7. J.-C. Wang, Chem. Commun., 2011, 47, 5831–5833. 31 R. A. J. O’HairandN.J.Rijs,Acc. Chem. Res., 2015, 48, 329–340. 8 B. K. Tate, C. M. Wyss, J. Bacsa, K. Kluge, L. Gelbaum and 32 M. Girod, M. Krstić, R. Antoine, L. MacAleese, J. Lemoine, J. P. Sadighi, Chem. Sci., 2013, 4, 3068–3074. A. Zavras, G. N. Khairallah, V. Bonačić-Koutecký, P. Dugourd 9 A. Zavras, G. N. Khairallah, T. U. Connell, J. M. White, and R. A. J. O’Hair, Chem. – Eur. J.,2014,20, 16626–16633. A. J. Edwards, P. S. Donnelly and R. A. J. O’Hair, Angew. 33 D. M. P. Mingos, T. Slee and Z. Lin, Chem. Rev., 1990, 90, Chem., Int. Ed., 2013, 52, 8391–8394. 383–402. 10 A. Zavras, G. N. Khairallah, T. U. Connell, J. M. White, 34 D. M. P. Mingos and D. J. Wales, Introduction to Cluster A. J. Edwards, R. J. Mulder, P. S. Donnelly and Chemistry, Prentice Hall, 1990. R. A. J. O’Hair, Inorg. Chem., 2014, 53, 7429–7437. 35 D. G. Evans and D. M. P. Mingos, J. Organomet. Chem., 11 A. Desireddy, B. E. Conn, J. Guo, B. Yoon, R. N. Barnett, 1982, 240, 321–327. B. M. Monahan, K. Kirschbaum, W. P. Griﬃth,R.L.Whetten, 36 D. M. P. Mingos, J. Organomet. Chem., 2004, 689, 4420– U. Landman and T. P. Bigioni, Nature, 2013, 501,399–402. 4436. 12 H. Yang, J. Lei, B. Wu, Y. Wang, M. Zhou, A. Xia, L. Zheng 37 F. K. Sheong, W.-J. Chen and Z. Lin, J. Organomet. Chem., and N. Zheng, Chem. Commun., 2013, 49, 300–302. 2015, 792,93–101. 13 H. Yang, Y. Wang, H. Huang, L. Gell, L. Lehtovaara, 38 C. Cugnet, D. Lucas, E. Collange, B. Hanquet, A. Vallat, S. Malola, H. Hakkinen and N. Zheng, Nat. Commun., 2013, Y. Mugnier, A. Soldera and P. D. Harvey, Chem. – Eur. J., 4, 2422. 2007, 13, 5338–5346. 14 H. Yang, Y. Wang and N. Zheng, Nanoscale, 2013, 5, 2674– 39 H. Batebi, F. Zarkoob, K. Daraei, B. F. Yates and A. Ariafard, 2677. J. Organomet. Chem., 2013, 748,89–97. 15 R. S. Dhayal, J.-H. Liao, Y.-C. Liu, M.-H. Chiang, S. Kahlal, 40 M. C. Sison Escaño, E. Gyenge, R. Lacdao Arevalo and J.-Y. Saillard and C. W. Liu, Angew. Chem., Int. Ed., 2015, 54, H. Kasai, J. Phys. Chem. C, 2011, 115, 19883–19889. 3702–3706. 41 W. Grochala and P. P. Edwards, Chem. Rev., 2004, 104, 16 L. G. AbdulHalim, M. S. Bootharaju, Q. Tang, S. del Gobbo, 1283–1315. R. G. AbdulHalim, M. Eddaoudi, D.-e. Jiang and 42 A. Colorado and J. Brodbelt, J. Am. Soc. Mass Spectrom., O. M. Bakr, J. Am. Chem. Soc., 2015, 137, 11970–11975. 1996, 7, 1116–1125. 17 C. P. Joshi, M. S. Bootharaju, M. J. Alhilaly and O. M. Bakr, 43 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystal- J. Am. Chem. Soc., 2015, 137, 11578–11581. logr., 2008, 64, 112–122. 18 J. Polte, X. Tuaev, M. Wuithschick, A. Fischer, 44 L. J. Farrugia, J. Appl. Crystallogr., 1997, 30, 565. Published on 16 October 2015. Downloaded by The University of Melbourne Libraries 01/02/2016 05:50:18. A. F. Thuenemann, K. Rademann, R. Kraehnert and 45 L. J. Farrugia, J. Appl. Crystallogr., 1999, 32, 837–838. F. Emmerling, ACS Nano, 2012, 6, 5791–5802. 46 M. J. Frisch, et al., Gaussian 09, revision D.01, Gaussian, 19 M. Wuithschick, B. Paul, R. Bienert, A. Sarfraz, U. Vainio, Inc., Wallingford, CT, 2009. For the complete reference see M. Sztucki, R. Kraehnert, P. Strasser, K. Rademann, the ESI.† F. Emmerling and J. Polte, Chem. Mater., 2013, 25, 4679– 47 Y. Zhao and D. G. Truhlar, Acc. Chem. Res., 2008, 41, 157– 4689. 167. 20 D. L. Van Hyning and C. F. Zukowski, Langmuir, 1998, 14, 48 P. C. Hariharan and J. A. Pople, Theor. Chim. Acta, 1973, 28, 7034–7046. 213–222. 21 J.-S. Seo, D.-M. Son, H. Lee, J. Kim and Y. Kim, Bull. Korean 49 A. W. Ehlers, M. Böhme, S. Dapprich, A. Gobbi, Chem. Soc., 2009, 30, 2651–2654. A. Höllwarth, V. Jonas, K. F. Köhler, R. Stegmann, 22 Q.-Q. Xu, X.-Y. Dong, R.-W. Huang, B. Li, S.-Q. Zang and A. Veldkamp and G. Frenking, Chem. Phys. Lett., 1993, 208, T. C. W. Mak, Nanoscale, 2015, 7, 1650–1654. 111–114. 23 D. G. Musaev and K. Morokuma, Organometallics, 1995, 14, 50 A. Höllwarth, M. Böhme, S. Dapprich, A. W. Ehlers, 3327–3334. A. Gobbi, V. Jonas, K. F. Köhler, R. Stegmenn, A. Veldkamp, 24 C. W. Liu, Y.-R. Lin, C.-S. Fang, C. Latouche, S. Kahlal and G. Frenking and R. Stegmann, Chem. Phys. Lett., 1993, 208, J.-Y. Saillard, Inorg. Chem., 2013, 52, 2070–2077. 237–240. 25 V. D. Makhaev, Russ. Chem. Rev., 2000, 69, 727–746. 51 K. Fukui, J. Phys. Chem., 1970, 74, 4161–4163. 26 M. Besora and A. Lledós, Struct. Bonding, 2008, 130,149–202. 52 K. Fukui, Acc. Chem. Res., 1981, 14, 363–368. 27 J. Vites, C. Eigenbrot and T. P. Fehlner, J. Am. Chem. Soc., 53 S. Grimme, S. Ehrlich and L. Goerigk, J. Comput. Chem., 1984, 106, 4633–4635. 2011, 32, 1456–1465.

This journal is © The Royal Society of Chemistry 2015 Nanoscale,2015,7,18129–18137 | 18137 105 106 3 Ligand-induced substrate steering and reshaping of + [Ag2(H)] scaffold for selective

CO2 extrusion from formic acid

107 108 ARTICLE

Received 14 Dec 2015 | Accepted 26 Apr 2016 | Published 6 Jun 2016 DOI: 10.1038/ncomms11746 OPEN Ligand-induced substrate steering and reshaping of þ [Ag2(H)] scaffold for selective CO2 extrusion from formic acid

Athanasios Zavras1,2, George N. Khairallah1,2, Marjan Krstic´3, Marion Girod4, Steven Daly5, Rodolphe Antoine5, Philippe Maitre6, Roger J. Mulder7, Stefanie-Ann Alexander1,2, Vlasta Bonacˇic´-Koutecky´3,8, Philippe Dugourd5 & Richard A.J. O’Hair1,2

Metalloenzymes preorganize the reaction environment to steer substrate(s) along the required reaction coordinate. Here, we show that phosphine ligands selectively facilitate þ protonation of binuclear silver hydride cations, [LAg2(H)] by optimizing the geometry of the active site. This is a key step in the selective, catalysed extrusion of carbon dioxide from formic acid, HO2CH, with important applications (for example, hydrogen storage). Gas-phase ion-molecule reactions, collision-induced dissociation (CID), infrared and ultraviolet action spectroscopy and computational chemistry link structure to reactivity and mechanism. þ þ [Ag2(H)] and [Ph3PAg2(H)] react with formic acid yielding Lewis adducts, while þ [(Ph3P)2Ag2(H)] is unreactive. Using bis(diphenylphosphino)methane (dppm) reshapes þ the geometry of the binuclear Ag2(H) scaffold, triggering reactivity towards formic acid, to þ þ produce [dppmAg2(O2CH)] and H2. Decarboxylation of [dppmAg2(O2CH)] via CID þ regenerates [dppmAg2(H)] . These gas-phase insights inspired variable temperature NMR studies that show CO2 and H2 production at 70 °C from solutions containing dppm, AgBF4,

NaO2CH and HO2CH.

1 School of Chemistry and Bio21 Molecular Science and Biotechnology Institute, The University of Melbourne, 30 Flemington Road, Parkville, Victoria 3010, Australia. 2 ARC Centre of Excellence for Free Radical Chemistry and Biotechnology, 30 Flemington Road, Parkville, Victoria 3010, Australia. 3 Center of Excellence for Science and Technology – Integration of Mediterranean region (STIM) at Interdisciplinary Center for Advanced Science and Technology (ICAST), University of Split, Mesˇtrovic´evoˇeta s lisˇte 45, 21000 Split, Croatia. 4 Institut des Sciences Analytiques, Universite´ de Lyon, Universite´ Lyon 1-CNRS- ENS Lyon, 69100 Villeurbanne, France. 5 Institut Lumie`re Matie`re, Universite´ Lyon 1-CNRS, Universite´ de Lyon 69622 Villeurbanne Cedex, France. 6 Laboratoire de Chimie Physique, Baˆtiment 349, Universite´ Paris-Sud, CNRS, Universite´ Paris-Saclay, F-91405 Orsay, France. 7 CSIRO Manufacturing, Bayview Avenue, Clayton, Victoria 3168, Australia. 8 Humboldt-Universita¨t Berlin, Institut fu¨r Chemie, 12489 Berlin, Germany. Correspondence and requests for materials should be addressed to V.B.-K. (email: [email protected]) or to P.D. (email: [email protected]) or to R.A.J.O. (email: [email protected]).

NATURE COMMUNICATIONS | 7:11746 | DOI: 10.1038/ncomms11746 | www.nature.com/naturecommunications 1 109 ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms11746

ature uses a number of design principles to create characterized7 and ligated variants can readily be formed8–10. different classes of enzyme catalysts capable of a wide Formic acid was chosen as a substrate since its decomposition is Nrange of chemical transformations of substrates1. A metal one of the most widely studied topics in chemistry, with a rich ion or metal cluster often has a critical role as a co-factor2. history spanning more than a century11–14. Apart from the A key concept in enzyme catalysis is the preorganization of the academic interest in establishing the mechanism(s) of reaction environment by the enzyme, directing the substrate to decomposition, the selective, catalysed decomposition of formic the reaction site, which provides a favourable geometry for acid has potentially important applications in areas ranging from the transition state required for bond activation. In essence, the hydrogen storage15–17 through to the generation of in situ enzyme steers the substrate along the required reaction hydrogenation sources for reduction of organic substrates18,19. coordinate to allow the desired transformation to product(s)3. In the absence of a catalyst, pyrolysis of formic acid proceeds The concept of changing the environment at a metal centre to via two primary pathways: decarboxylation (equation (1)) and switch on reactivity has also been recently exploited in gold dehydration (equation (2)). These reactions are coupled by the chemistry. Au(I) complexes prefer to be linear, which is why they water–gas shift reaction (Equation (3))20,21 and have been widely are unreactive toward oxidative addition of iodobenzene (Fig. 1a). studied experimentally22 and theoretically23. In the gas-phase, the To promote reactivity, ligand-induced preorganization of the dehydration channel (Equation (2)) is the dominant reaction22, metal centre has been shown to accommodate the geometry consistent with a lower activation energy, as predicted by DFT of the ensuing oxidative addition of aryl halides (Fig. 1b)4. calculations23. Embedding the metal centre within a ligated nanocluster also HO CH ! H þ CO ð1Þ facilitates reactivity, which can be further tuned by the choice of 2 2 2 ligand (Fig. 1c)5. HO CH ! H O þ CO ð2Þ Here, we use gas-phase experiments and density functional 2 2 theory (DFT) calculations to examine how the binuclear silver þ H2O þ CO ! H2 þ CO2 ð3Þ hydride cation, [Ag2(H)] (Fig. 1d,f), can be structurally manipulated by the appropriate choice of phosphine ligands6 to The concept of using metal catalysts to selectively decompose switch on the protonation of the hydride by formic acid to formic acid dates back over 100 years to Sabatier’s work on the liberate hydrogen, which is a key step in the selective, catalysed role of metal and metal oxide catalysts11, and the substantial early decomposition of formic acid that does not occur in absence of literature has been reviewed12–14. Over the past century, a wide þ ligands. We chose [Ag2(H)] since it has been spectroscopically range of metal catalysts have been surveyed for their potential to

ab+ R + + R R R R R P R P P 180° Au No reaction 90° Au Au P P P I R R R R R R R Unreactive - linear metal center Reactive - ligand-induced preorganization metal center o = BH c d H + + + Ag Ag R R R R P Au P Au O O O O (CH2)n Au (CH2)n Au H H P Au P Au H H R R + R R H Ag Ag Reactivity is tuned by ‘n’, with the following reaction efficiencies: Ph n = 6 (42%) > 5 (9.9%) > 4 (0.75%) > 3 (0.04%) P O O Ph Ph H H Reactive - metal center embedded within a ligated nanocluster Substrate Undesired substrate coordination - Lewis adduct steering + efH + O O H H O H H O 3.07 Å ~220° H Ligand-induced AgAg reshaping of ~180° Ph Ph Ag H activates AgAg 2 Ph X PP Ph catalytic activity Ph PPPh Ph Ph Ph Ph X = CH2, (CH2)2, C6H4 Desired substrate coordination - weakly bound Desired substrate coordination - reactive ion–molecule complex ion–molecule complex

Figure 1 | Key concepts for switching on reactivity at coinage metal centres. (a) Linear diphosphine Au(I) complexes do not undergo oxidative addition of iodobenzene. Oxidative addition of iodobenzene does occur for: (b) bisphosphine Au(I) complexes with P–Au–P bond angles of E90°; þ and (c) bisphosphine ligated gold cluster. Switching on desired protonation of binuclear silver hydride cations, [LAg2(H)] by formic acid: (d) undesired Lewis adduct formation occurs when silver centres have vacant coordination sites; (e) formic acid is steered to active site by phosphine ligands; (f) bisphosphine ligands reshape geometry of active site to switch on desired protonation reaction.

2 NATURE COMMUNICATIONS | 7:11746 | DOI: 10.1038/ncomms11746 | www.nature.com/naturecommunications 110 NATURE COMMUNICATIONS | DOI: 10.1038/ncomms11746 ARTICLE selectively decarboxylate formic acid (Equation (1)). The types of 1a m/z 215, reacts via sequential addition of formic acid metal catalysts examined include metal and metal oxide (Supplementary Fig.7a,b; equations (5) and (6); Fig. 2a (iii) and 14 24 25 þ surfaces , mononuclear metal complexes , metal clusters (v)), as confirmed via mass selection of [Ag2(H)(HCOH)] m/z and metal nanoparticles26. 261, and subsequent reaction with formic acid, which yields þ þ The powerful combination of gas-phase ion–molecule reac- [Ag2(H)(HO2CH)2] m/z 307. CID of [Ag2(H)(HCOH)] tions (IMRs), collision-induced dissociation (CID), infrared and regenerates 1a via loss of formic acid (Fig. 2b and Supplementary ultraviolet action spectroscopy and computational chemistry Fig. 7c). These results confirm the concept that formic acid is allows us to examine the role of the ligand (L) in promoting trapped down the wrong reaction pathway (Fig. 2a (iii)). decomposition of formic acid catalysed by the binuclear silver ÂÃþ ÂÃþ þ LAg ðÞH þ HO2CH ! LAg ðÞH ðÞHO2CH ð5Þ hydride cations, [LAg2(H)] . Guided by the right choice of 2 2 ligand, we have translated our gas-phase results to achieve the ÂÃ ðÞðÞþ þ selective condensed-phase decarboxylation of formic acid. LAg2 H HO2CH HO2CH ÂÃ ! ðÞðÞþ ð Þ Results LAg2 H HO2CH 2 6 Catalyst systems. The six systems that we have studied to examine the catalytic cycle (Fig. 2a) for decomposition of formic We next tested whether ligation could steer the substrate away acid are designated by the letters, which identify the ligands from coordination to form a Lewis adduct and toward the as: 1a ¼ no ligand; 1b L ¼ PPh ; 1c L ¼ 2 PPh ; 1d hydride site. Blocking one Ag site in 1b results in the formation of 3 3 a mono adduct (equation (5), L ¼ Ph P, Supplementary Fig. 8b). L ¼ bis(diphenylphosphino)methane (dppm); 1e L ¼ 1,2-bis 3 þ Kinetic modelling of the temporal profiles of [(Ph P) Ag (H)] (diphenylphosphino)benzene (dppbz); and 1f L ¼ bis þ 3 n 2 (diphenylphosphino)ethane (dppe)). DFT calculations reveal that and [(Ph3P)nAg2(H)(HO2CH)2 n] (Supplementary Table 1) þ reveals that: (i) the addition of formic acid for n ¼ 0 and 1 is at the ligand(s) can induce changes to the geometry of the Ag2(H) E ¼ scaffold (Fig. 2b). 1% of the collision rate; (ii) in the case of n 0, addition of formic acid is reversible. Blocking both Ag sites makes 1c þ Reactions of 1a–1f with formic acid. [Ag (H)] and its ligated unreactive toward formic acid (Supplementary Fig. 8a). By þ 2 þ variants, [LAg2(H)] , were prepared in the gas-phase via replacing both Ph3P ligands with the dppm ligand, the Ag2(H) well-established ligand fragmentation reactions (Supplementary scaffold of 1e is compressed, with the P–Ag–H angle deviating substantially away from linearity. As a consequence Figs 1–6; Supplementary equations 1 and 2), including þ [dppmAg (H)] 1d (m/z 601) reacts with formic acid to form decarboxylation of coordinated formates, equation (4) (refs 27,28). 2 þ The precursor ions were formed via electrospray ionization (ESI). [dppmAg2(O2CH)] and H2 (equation (7), Fig. 3a), a reaction ÂÃþ ÂÃþ that proceeds at E 1% of the collision rate (Supplementary þ LAg ðÞO2CH ! LAg ðÞH þ CO2 ð4Þ 2 2 Table 1). When [dppmAg2(D)] m/z 602, formed via

a CO2 (i) (iii) (v)

HCO2H HCO2H + + [LAg (H)(HCO H)]+ [LAg (H)(HCO H) ]+ [LAg2(O2CH)] [LAg2(H)] 2 2 2 2 2 –HCO2H –HCO2H (iv) (vi) (ii)

H2 HCO2H b

1.77 173.2° 1.74 1.78 172.2° 1.71 2.95 3.1 2.9 1a 1b 7.36

1.84 1.83 1.82 210.8° 219.8° 226.2° 2.68 2.62 2.98

3.25 3.7 4.55

1d 1e 1f þ Figure 2 | Role of the ligand in the selective decarboxylation of formic acid catalysed by [LAg2(H)] . (a) catalytic cycle, with following steps: (i), þ þ þ þ þ decarboxylation of [LAg2(O2CH)] via CID to generate [LAg2(H)] . (ii), IMR of [LAg2(H)] to regenerate [LAg2(O2CH)] . (iii), IMR of [LAg2(H)] . þ þ þ þ (iv), CID of [LAg2(H)(HO2CH)] to regenerate [LAg2(H)] . (v), IMR of [LAg2(H)(HO2CH)] with HO2CH to yield [LAg2(H)(HO2CH)2] . (vi), CID of þ þ [LAg2(H)(HO2CH)2] to regenerate [LAg2(H)(HO2CH)] . Most stable DFT-calculated structures of systems examined: (b) 1a, 1b, 1c, 1d, 1e and 1f. DFT calculations used the hybrid functional B3LYP53 with def2-TZVP AO basis set54 for all atoms and corresponding relativistic effective core potential for Ag atoms55. Bond distances are given in Å (black) and P–Ag–H bond angles in degrees (red).

NATURE COMMUNICATIONS | 7:11746 | DOI: 10.1038/ncomms11746 | www.nature.com/naturecommunications 3 111 ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms11746

2 þ 8 fragmentation of the cluster cation [Ag3(m3-D)dppm3] , was fragmentation was more easily achieved using the infrared FEL allowed to react with formic acid, the unlabelled formate only, and the auxiliary CO laser was thus not used. þ 2 þ þ [dppmAg2(O2CH)] and HD are formed (equation (8) and For both [dppmAg2(H)] and [dppmAg2(O2CH)] , a good Supplementary Fig. 9b), which is consistent with a mechanism in match was observed between the experimental and theoretically which the hydride is protonated by formic acid to release H2. predicted infrared spectrum for the lowest energy structure. The reactions given by equations (4) and (7) represent A detailed assignment of the main observed infrared features is those associated with the catalytic cycle (Fig. 2a) for the provided (Supplementary Table 2). As expected, infrared bands selective decarboxylation of formic acid (equation (1)). Indeed, associated with the auxiliary dppm ligand are observed. This is þ þ sequential reactions of CID of [dppmAg (O CH)] to form particularly true in the case of the [dppmAg (H)] spectrum þ 2 2 2 [dppmAg (H)] followed by ion-molecule reaction (IMR) with where four bands can be assigned to phenyl in plane ring 2 formic acid, allows completion of the cycle multiple times with no deformation (998 cm 1), CH twist and P–C H stretch 2 6 5 significant loss of signal (Supplementary Fig. 9). The other (1,097 cm 1), phenyl in plane CH bending (1,438 cm 1), and þ 1 binuclear silver hydride cations, [LAg2(H)] , containing bispho- phenyl CH bending and ring deformation (1,478 cm ). sphine ligands also reacted with formic acid to reform the More importantly, diagnostic bands of the coordination mode þ [LAg2(O2CH)] and H2 (equation (7) and Fig. 3b,c), although of the hydride and formate ligands are also observed. In the þ 2 the nature of the ligand influences the reaction efficiencies, which case of [dppmAg2(H)] (Fig. 4a), the m bridging coordination follow the order 1dE1e441f (Supplementary Table 1). mode of the hydride is well characterized by two bands associated ÂÃþ ÂÃþ with the asymmetric and symmetric Ag–H stretching bands ðÞ þ ! ðÞþ ð Þ LAg2 H HO2CH LAg2 O2CH H2 7 observed at 900 and 1,250 cm 1, respectively, in excellent ÂÃ ÂÃ agreement with the theoretical prediction (916 and 1,236 cm 1, ðÞþ þ ! ðÞþ þ ð Þ þ LAg2 D HO2CH LAg2 O2CH HD 8 respectively). In the case of [dppmAg2(O2CH)] (Fig. 4b), it was expected that the positions of the asymmetric and symmetric formate CO stretching bands were sensitive to the formate IR and UV spectroscopy of reactive intermediates.Inorderto coordination modes. These two bands are strongly infrared active, relate the structure of the proposed reactive intermediates 1d and and could thus be revealed through IRMPD without the use of the þ [dppmAg (O CH)] to their observed reactivity (Fig. 2a), we next auxiliary CO laser. As can be seen in Supplementary Table 2, the 2 2 2 turned our attention to their gas-phase characterization29 using observed position of these two bands (1,360 and 1,547 cm 1, infrared multiple photon dissociation (IRMPD) spectroscopy respectively) is in excellent agreement with their predicted (Fig. 4a,b)30 and ultraviolet action-spectroscopy (Fig. 4c,d)7,31.For positions (1,345 and 1,564 cm 1, respectively) for the lowest þ each complex, the experimental IRMPD spectrum is compared energy structure of the [dppmAg (O CH)] complex. It can thus 2 2 þ against the theoretically predicted infrared absorption spectrum of be concluded that IRMPD spectroscopy of [dppmAg (H)] and þ þ 2 thelowestenergyisomer.Inthecaseof[dppmAg2(H)] (m/z 601), [dppmAg2(O2CH)] , in conjunction with electronic structure infrared features could only be observed if mass-selected ions were calculations, provide clear structural diagnostic of the coordina- þ þ irradiated on resonance with the infrared free electron laser (FEL) tion mode within the Ag2H or Ag2(O2CH) scaffolds. 32 and in conjunction with an auxiliary CO2 laser . The enhancement Comparison of the ultravoilet action spectra and calculated of the spectroscopic resolution was such that relatively weak time dependent density functional method (TDDFT) spectra IRMPD features could be observed (Fig. 4a). In the case of using dispersion correction D3 (ref. 33) for the lowest energy þ þ þ [dppmAg2(O2CH)] (m/z 645), however, infrared-induced structures for [dppmAg2(H)] and [dppmAg2(O2CH)] are shown in Fig. 4c,d. Introduction of dispersion correction into

a HO CH TDDFT reduces the distance between two parallel aromatic rings, [dppmAg (H)]+ 2 645 þ 100 * 2 thus preventing their mobility. In the case of [dppmAg2(H)] , 601 H–H the experimental spectrum shows an increase in fragmentation 601 50 + yield as the wavelength decreases, with two superimposed bands [dppmAg (O CH)] HO CH 2 2 2 at 270 and 235 nm (Fig. 4c). The corresponding TDDFT 645 5,000 ms 0 transitions with dominant oscillator strengths are due to leading excitations from HOMO-1 and HOMO to LUMO þ 2 b HO2CH 100 * [dppbzAg (H)]+ 707 2 (Supplementary Fig. 10), respectively. They involve the Ag2H 663 subunit as well as the ligand. In contrast, the S and S states H–H 663 1 2 50 + located close to 300 nm are characterized by HOMO to LUMO [dppbzAg2(O2CH)] HO2CH 707 5,000 ms and HOMO-1 to LUMO excitations, respectively, in which Ag2 or 0 Ag2HP2 are more involved than ring subunits of the ligand.

Relative intensity (%) þ c * HO CH 659 In the case of [dppmAg2(O2CH)] , a similar action spectrum 100 [dppeAg (H)]+ 2 2 was obtained. The main difference is a pronounced shoulder at 615 50 H–H 615 250 nm resulting from an intense S1 transition involving mainly

+ HO2CH the Ag2HP2 subunit (Supplementary Fig. 10). The formate [dppeAg2(O2CH)] 0.4 659 5,000 ms has little role in the excitations leading to absorption in this 0.0 spectral region, which explains the similarity between the 600 800 1,000 two optical spectra. Absorption spectra obtained with TDDFT- m/z D3 are in good agreement with the experimental ultraviolet Figure 3 | Ion–molecule reaction of formic acid with mass-selected photodissociation (UVPD) spectra, thus conﬁrming the calcu- þ 9 hydrides, [LAg2(H)] . (a)L¼ dppm, 1d, [HO2CH]ion trap ¼ 7.19 10 lated structural properties. Altogether, on the basis of UVPD and 3 ¼ ¼ 9 þ molecules cm .(b)L dppbz, 1e, [HO2CH]ion trap 7.30 10 IRMPD spectra, the structural assignments of [dppmAg2H] 3 ¼ ¼ 9 þ molecules cm .(c)L dppe, 1f, [HO2CH]ion trap 7.09 10 and [dppmAg2(O2CH)] are unambiguous. molecules cm 3. The most intense peak of the isotope cluster is represented by the m/z value. * Represents the mass selected precursor DFT-calculated mechanism of steps associated with the ion. catalytic cycle. The catalytic cycle for the selective decomposition

4 NATURE COMMUNICATIONS | 7:11746 | DOI: 10.1038/ncomms11746 | www.nature.com/naturecommunications 112 NATURE COMMUNICATIONS | DOI: 10.1038/ncomms11746 ARTICLE

ac + + [dppmAg2(H)] [dppmAg2(H)] 300 H-Ag2 0.5 bending 2.0 250 0.4 1.5 200 S13 H-Ag2 0.3 stretching 150 1.0 S10 0.2 Intensity

theor, Oscillator strength 100 S Fragmentation yield 0.5 9 exp. S5 0.1 S 50 theor. 2 S1 exp. 0.0 0.0 0 210 220 230 240 250 260 270 280 290 300 1,000 1,500 2,000 Wavelength (nm) (cm–1) bd + + [dppmAg2(O2CH)] [dppmAg2(O2CH)] 600 C-O stretching 1.5 0.6 500

O-C-O 400 stretching 1.0 0.4 + S1 300 H-CO2 bending Intensity 0.5 S8 theor, 0.2 Oscillator stength 200 Fragmentation yield exp. S2 100 theor. 0.0 0.0 exp. 0 210 220 230 240 250 260 270 280 290 300 1,000 1,500 2,000 Wavelength (nm) (cm–1)

Figure 4. | Experimental (red line) and DFT-calculated (blue line) IRMPD and UV spectra (hybrid functional B3LYP with Stuttgart relativistic þ effective core potential for Ag atoms with corresponding def2-TZVP AO basis set, same AO basis set for all other atoms) of (a) [dppmAg2(H)] , 1d. þ (b) [dppmAg2(O2CH)] . A scaling factor of 0.98 was applied to the calculated harmonic frequencies. UV action spectrum (red line) and calculated TDDFT spectrum with dispersion correction D3 (using CAM-B3LYP functional with Stuttgart relativistic effective core potential for Ag atoms with corresponding def2-TZVP AO basis set for all atoms, the black vertical lines correspond to values of oscillator strength frequency fe) for the lowest energy þ þ structure of: (c) [dppmAg2(H)] .(d) [dppmAg2(O2CH)] . The analysis of leading excitations is in Supplementary Fig. 10. of formic acid involves two distinct types of reactions (Fig. 2a for intact as can be seen from Fig. 1. Analysis of the charge þ 1d, Supplementary Fig. 11 for 1e and 1f, Supplementary Table 3 distributions in these [LAg2(H)] complexes reveals delocali- for all systems). The ion–molecule reaction of formic acid with zation of positive charge in this subunit (Supplementary Fig. 12). þ þ þ [LAg2(H)] to produce [LAg2(O2CH)] , and H2 must be an In contrast, decarboxylation of [LAg2(O2CH)] is endother- exothermic process with barriers that lie below the separated mic as it requires energization through multiple collisions with reactants in order for it to occur under the near thermal the helium bath gas during the CID process in order to occur. 34 conditions of the ion-trap . Indeed, DFT calculations reveal this The mechanism for CO2 release involves two steps. First, the to be the case, for 1d (Fig. 5 and Supplementary Table 3). The formate needs to change from an O,O-bridging ligand to an þ binding energy of cis-formic acid to [dppmAg2(H)] is 0.24 eV, O-bound ligand. This involves breaking one of the Ag-O bonds 28 and subsequent reaction via H2 formation proceeds via a via a barrier of 1.7 eV. The next step involves decarboxylation , transition state with barrier of 0.18 eV leading to the formation which proceeds over a barrier of 1.86 eV to release of CO . þ 2 of [dppmAg2(O2CH)] , which is exothermic by 0.95 eV. Altogether, a catalytic cycle involving the selective decomposition Throughout this reaction, the dppm ligand keeps the Ag of formic acid via the release of H and CO (Equation (1)) can 2 2 2 þ subunit intact, while allowing the Ag–Ag bond length to relax. occur according to calculated energy profile for [dppmAg2(H)] The overall exothermicities for this first reaction (step 1) for the (Fig. 5), under experimental conditions involving IMR for the þ other [LAg2(H)] complexes examined follow the order L ¼ dppm, spontaneous release of H2 and activation via CID for the release (0.95 eV)4L ¼ dppe (0.76 eV)4L ¼ dppbz (0.71 eV)4L ¼ 2 Ph P of CO . It is to be expected that this is also the case for 3 2 þ þ (0.68 eV)4L ¼ Ph3P (0.51 eV), which is in qualitative agreement [dppbzAg2(H)] and [dppeAg2(H)] since that Ag2H(X)P2 with experimental findings that dihydrogen release (equation (7)) subunit remains intact, although the energetics are slightly less ¼ þ occurs for L dppm, dppbz and dppe, but not for the favourable than in the case of [dppmAg2(H)] (Supplementary other ligands. The corresponding heights of barriers are 1d Table 3). (0.18 eV)o1f (0.25 eV)o1e (0.36 eV) indicate that all three barriers can overcome under experimental conditions, although Solution-phase selective decarboxylation of formic acid. The þ the efficiency of the reaction will depend on the height of the fact that [dppmAg (O CH)] is both readily decarboxylated in 2 2 þ barrier. Only in these three cases do the Ag2HP2 subunits remain the gas-phase and reformed via the reaction of [dppmAg2(H)]

NATURE COMMUNICATIONS | 7:11746 | DOI: 10.1038/ncomms11746 | www.nature.com/naturecommunications 5 113 ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms11746

2.86

2.96 2.5 3.16 2

1.5 TS for CO2 loss 0.91 1 0.75 TS for H loss 0.5 2 0.00 0.48 0.43 E (eV) –0.06 0 1.7 0.00 –0.5 –0.24 0.18 –0.14 2.68 –1 2.71 2.68 –0.95 2.98 2.69 –1.5 2.74 2.88 + [Ag2HL] +HCOOH + [Ag2HL] +H2+CO2

Step 1 Step 2

Figure 5 | DFT-calculated energy profile for the two reaction steps in the catalytic cycle of Fig. 1a. (Step 1) ion-molecule reaction of formic þ þ acid with [dppmAg2(H)] ; (step 2) CID decarboxylation of [dppmAg2(O2CH)] . Relative energies are in eV. with formic acid prompted us to prepare stoichiometrically well- the tight-bite angles of the bidentate bridging ligands dppm, defined solutions in order to use variable temperature 1H and 13C dppbz and dppe switch on the protonation of the silver hydride in NMR spectroscopy to examine the evolution of the two gaseous step 1 (Fig. 5) by providing an appropriate geometry to weaken products formed in the selective decarboxylation of formic acid the Ag–H bonds and bend the P–Ag–H away from linearity, 1 (equation (1)). No H2 evolution was observed by H NMR when thereby allowing coordination of formic acid and subsequent 13 a solution of AgBF4, C-labelled formic acid and dppm was reaction between the coordinated moieties Ag(OCH(OH) and heated from 25 to 70 °C (Supplementary Fig. 13). When the Ag(H). The ligand further tunes the reactivity as highlighted by experiment was repeated with the addition of sodium formate, no both the experimentally determined reaction efficiencies, which E evolution of H2 was observed from 25 to 55 °C (Supplementary follow the order 1d 1e441f and the DFT-calculated barrier Fig. 14). When the temperature was raised to 70 °C, H2 evolution heights for reaction with cis-formic acid (Supplementary Fig. 18) was observed almost instantly (Supplementary Fig. 15). The to release H2, which follow the order 1d (0.18 eV)o1f evolution of both H2 and CO2 were observed to increase over (0. 25 eV)o1e (0.36 eV). The ligand also exerts an effect in step time at 70 °C, and both H2 and CO2 reached a steady-state 2, with energy resolved CID experiments (Supplementary Fig. 5) concentration after B11.5 min (Supplementary Figs 15 and 16). providing reactivity orders for the ease of decarboxylation that are in agreement with the DFT-calculated barrier þ þ Discussion heights: [dppeAg (O CH)] (1.64 eV)E[dppbzAg (O CH)] 2 2 þ 2 2 While concepts of steric and electronic effects are well established (1.65 eV)o[dppmAg2(O2CH)] (1.86 eV). in guiding the choice of ligand to modulate the reactivity of Finally, the gas-phase results encouraged us to examine related mononuclear catalysts in homogenous catalysis35,36, related selective decarboxylation reactions in solution44,45. We found that 13 concepts for choosing ligands to modulate the reactivity of both H2 and CO2 are evolved when a stoichiometrically well- 5 13 binuclear and cluster catalysts are yet to be fully developed . The defined solution containing dppm, AgBF4, C-labelled formic value of gas-phase studies employing mass spectrometry (MS)- acid and sodium formate was warmed to 70 °C. While the precise based methods37,38 is that they allow a systematic exploration nature of the reactive species in solution is unknown, previous of the factors that control reactivity for all steps in a catalytic studies have shown that: (i) related dppm complexes of silver cycle39,40. When used in conjunction with DFT calculations, carboxylates exist as dimers in solution46,47; (ii) the related þ ¼ where mechanistic pathways can be explored, different types of silver hydride [(NHC)2Ag2(H)] (where, NHC 1,3-bis(2,6- catalysts that vary in the metal, ligand and/or nuclearity that diisopropylphenyl)imidazolin-2-ylidene) exists in solution and 41 þ catalyse the same transformation or which transform the same reacts with CO2 to form [(NHC)2Ag2(O2CH)] , a reaction that substrate in different ways42 can be directly compared43. is the reverse of decarboxylation of a coordinated formate ligand Here, we have shown that ligand choice is a crucial factor in (equation (4)) studied here48. designing a binuclear silver hydride cluster that catalyses the Two key concepts have emerged from this work: (i) that þ selective decarboxylation of formic acid. Ligation of the Ag2(H) ligands can have a vital role in reshaping the scaffold of a metal scaffold clearly has an influence on its geometry by shortening the cluster to activate its reactivity towards a substrate; and (ii) that Ag–Ag distance and increasing the Ag–H distance (Fig. 1). In the fundamental gas-phase studies can be used to direct the search for absence of bidentate bridging ligands, the reactivity pattern of 1a, new types of metal complexes that promote related reactivity in 1b and 1c towards formic acid is consistent with simple Lewis solution49. Together these concepts have allowed us to achieve the acid/base interactions in which the number of vacant coordina- selective extrusion of carbon dioxide from formic acid, an þ 15–17 tion sites in [(Ph3P)nAg2(H)] dictates how many formic acid important process for applications in hydrogen storage . molecules can coordinate via the O atom of the C ¼ O to form the þ adducts [(Ph3P)nAg2(H)(HO2CH)2 n] . Indeed DFT calcula- Methods tions reveal that this coordination mode yields the most stable Materials. Chemicals listed in the Supplementary information were used as adducts for n ¼ 0 and 1, while in the case of n ¼ 2 only a weakly received. bound ion–molecule complex is formed (Supplementary Fig. 17) and this is likely to simply dissociate back to separated reactants, Preparation of silver complexes for MS analysis. In situ silver precursor which is why no adduct is observed experimentally. In contrast, complexes for ESI/MS were typically generated by adding 20 mmol AgX

6 NATURE COMMUNICATIONS | 7:11746 | DOI: 10.1038/ncomms11746 | www.nature.com/naturecommunications 114 NATURE COMMUNICATIONS | DOI: 10.1038/ncomms11746 ARTICLE

¼ ¼ (X NO3 or BF4 ) followed by 10 mmol of phosphine ligand, L (L PPh3, dppm, experimental data. The IR spectrum for neutral monomeric cis-formic acid dppbz or dppe), to 20 ml of freshly prepared solvent mixtures in a 50 ml Quickfit calculated at this level of DFT theory is also in good agreement with experimental round-bottom flask covered in foil and equipped with a glass stopper and magnetic data (Supplementary Fig. 19). For calculations of the absorption spectra TDDFT stir bar. The solution was stirred for at least 5 min and 10 mmol of sodium formate with the long-range corrected version of the hybrid B3LYP functional, the was added. MS experiments were conducted immediately after the addition of Coulomb-attenuated CAM-B3LYP functional and TZVP AO basis set has been sodium formate. employed. Potential interactions between the aromatic rings of dppm, raises the question of whether the dispersion correction within DFT are required. We have tested the Gas-phase studies using CID and IMR . Gas-phase experiments on phosphine influence of dispersion correction on the structural and spectroscopic properties of ligated silver formate clusters were carried out using a Finnigan hybrid linear þ þ [dppmAg H] and [dppmAg (O CH)] complexes by introducing D3 into DFT quadrupole Fourier transform ion-cyclotron resonance mass spectrometer. The 2 2 2 and TDDFT33. Comparison of the measured and calculated infrared spectra silver complexes prepared above were typically diluted in methanol or acetonitrile suggest that dispersion corrections has no influence. In contrast, the absorption m to a final silver(I) concentration of 50 M and at least 1.0 ml. The diluted solution transitions calculated using the D3 correction in which aromatic rings are involved m was drawn into a 500 l gas tight borosilicate glass syringe with polytetra- are only slightly blue shifted, thus improving agreement with the experimental fluoroethylene (PTFE) plunger tips and injected into the Finnigan ESI source at a UVPD spectra. Finally, the energy profile shown in Fig. 5 is almost unchanged m 1. ESI source conditions to yield a stable current of 0.5 mA flow rate of 3–5 lmin when single point B3LYP-D3 energy calculations are carried out for each reaction were: needle potential (3.5–5.0 kV); nitrogen sheath gas pressure (5–10 a.u.). The step of the catalytic cycle (Supplementary Table 3). ion transfer capillary temperature was set to 250 °C. Voltages were: tube lens (E20.0 V) and capillary voltage (10.0 V). The unimolecular fragmentation/dissociation of mass-selected precursor silver complexes occurred via CID using a NMR spectroscopy experiments. The NMR experiments were performed on a normalized collision energy typically between 20 and 25%, and an activation time Bruker Avance Av500 NMR spectrometer (500.13 MHz 1H frequency) equipped of 30 ms. IMR were carried by injecting formic acid into the helium bath gas5.The with a 5 mm triple resonance CryoProbe Prodigy probe (1H/19F–2H/13C/15N). stoichiometry of all ions was confirmed by high-resolution MS experiments Solutions for analysis by NMR were prepared by dissolving: (1) solution A: AgBF4 (Supplementary Table 4). (195 mg, 1 mmol), dppm (192 mg, 0.5 mmol), 13C formic acid (24 mg, 0.5 mmol) and sodium formate (34 mg, 0.5 mmol) in 1 ml of deuteroacetonitrile; (2) solution B: same as solution A, but without sodium formate added. NMR experiments were Energy-resolved CID experiments. Energy-resolved CID experiments were performed with the sample held at temperatures between þ 25 °Cand þ 70 °C carried out using a Finnigan 3D ion trap (LCQ) mass spectrometer. The method of (±0.1 °C). Chemical shifts for 1H experiments are referenced to the residual Brodbelt was adapted and details are given in the Supplementary methods (see text protonated solvent signal (CD HCN, d 1.94 ppm); 13C referenced to the solvent associated with Supplementary Fig. 4)50. The activation voltage was determined by 2 signal (CD CN, d 1.39 ppm). One-dimensional NMR experiments were acquired Supplementary equation 3. 3 using standard Bruker library pulse sequences.

MS for IR action spectroscopy. Infrared spectroscopy of mass-selected ions in Data availability the 800–1,600 cm 1 range was performed using a 7 Tesla Fourier transform ion . The data that support the findings of this study are available cyclotron resonance tandem mass spectrometer (Bruker Apex IV Qe)51 equipped from the corresponding author upon request. with an ESI source and coupled to the infrared FEL beam line of CLIO52. This IR FEL delivers B10 ms long trains of picosecond pulses at 25 Hz. Ions of interest were References accumulated and trapped in a B5 cm long hexapole ion-trap pressurized with B 1. Menger, F. M. An alternative view of enzyme catalysis. Pure Appl. Chem. 77, argon. The trapping delay ( 500 ms) allows for an efficient collisional cooling of 1873–1886 (2005). the ions. Ions are then pulse extracted to the ICR cell where they are mass-selected, 2. Valdez, C. E., Smith, Q. A., Nechay, M. R. & Alexandrova, A. N. Mysteries of and then irradiated for 1 s. Upon resonant vibrational excitation, dissociation of the metals in metalloenzymes. Acc. Chem. Res. 47, 3110–3117 (2014). selected ion can be monitored via its fragment peaks. 3. Cannon, W. R., Singleton, S. F. & Benkovic, S. J. A perspective on biological A significant enhancement of the photofragmentation yield can be observed catalysis. Nat. Struct. Biol. 3, 821–833 (1996). using an auxiliary CO laser (10 W continuous wave, BFi OPTiLAS, France)32. 2 4. Joost, M. et al. Facile Oxidative addition of aryl iodides to gold(I) by ligand For this purpose, a train of CO2 pulses at 25 Hz is generated and synchronized with the IR FEL laser with a retarding delay being on the order of B1 ms. This auxiliary design: bending turns on reactivity. J. Am. Chem. Soc. 136, 14654–14657 þ CO laser was used in the case of [dppmAg (H)] (m/z 601), and the CO laser (2014). 2 2 2 5. Robinson, P. S. D., Khairallah, G. N., da Silva, G., Lioe, H. & O’Hair, R. A. J. pulse length (25 ms) was adjusted to avoid CO2 induced dissociation. þ Gold mediated C–I bond activation of iodobenzene. Angew Chem. Int. Ed. 51, The experimental IRMPD bandwidth (fwhm) for both [dppmAg2(H)] and þ 1 3812–3817 (2012). [dppmAg2(O2CH)] is on the order of 20 cm , as generally observed for IRMPD spectra of other systems obtained using the CLIO FEL. 6. Meijboom, R., Bowen, R. J. & Berners-Price, S. J. Coordination complexes of silver(I) with tertiary phosphine and related ligands. Coord. Chem. Rev. 253, 325–342 (2009). MS for ultravoilet action spectroscopy. For ultravoilet action spectroscopy, a 7. Mitric, R. et al. Gas-phase synthesis and vibronic action spectroscopy of dual linear ion trap (LTQ-VELOS, ThermoScientific) was used to generate, mass þ Ag2H . J. Phys. Chem. Lett. 2, 548–552 (2011). select and trap ions in a first, high-pressure ion trap, for a controlled duration. 8. Zavras, A. et al. Synthesis, Structural characterisation and gas-phase During ion trapping, ions can be activated and fragmented by photons or CID. unimolecular reactivity of the silver hydride nanocluster Fragment ions are transmitted to a second ion trap, with low pressure, where they [Ag3((PPh2)2CH2)3(m3-H)](BF4)2. Inorg. Chem. 53, 7429–7437 (2014). are mass analyzed. A fused-silica window is positioned at the back end of the 9. Girod, M. et al. Formation and characterisation of the silver hydride instrument allowing for the introduction of laser beams in the ultravoilet-visible þ nanocluster cation [Ag H ((Ph P) CH )] and its release of hydrogen. Chem. range along the ion trap axis. Ultravoilet light was generated by doubling the 3 2 2 2 2 Eur. J. 20, 16626–16633 (2014). output of an optical parametric oscillator (Horizon optical parametric oscillator 10. Khairallah, G. N., O’Hair, R. A. J. & Bruce, M. I. Gas-phase synthesis and pumped by the third harmonic of a Surelite II Nd:YAG laser, Continuum). þ reactivity studies of binuclear gold hydride cations, (R PAu) H (R ¼ Me A mechanical shutter, synchronized with the mass spectrometer, is used to stop the 3 2 and Ph). Dalton Trans. 30, 3699–3707 (2006). beam at all times except the ‘ion activation window’—that is the time after ion accumulation and before the mass analysis. A single laser pulse was used for the 11. Sabatier, P. & Mailhe, A. Catalytic decomposition of formic acid. Compt. Rend. irradiation of the trapped ions and when irradiating ions the normalized collision 152, 1212–1215 (1912). energy is kept at zero. The fragmentation yield (FY) is given by equation (9). 12. Sabatier, P. Catalysis in Organic Chemistry 538–539 (Library Press, 1923). 13. Mars, P., Scholten, J. J. F. & Zwietering, P. The catalytic decomposition of P formic acid. Adv. Catal. 14, 35–113 (1963). FY ¼log =l:Pw ð9Þ P þ F 14. Trillo, J. M., Munuera, G. & Criado, J. M. Catalytic decomposition of formic acid on metal oxides. Catal. Rev. 7, 51 (1972). P and F are the intensities on the mass spectrum for respectively the parent ion and 15. Grasemann, M. & Laurenczy, G. Formic acid as a hydrogen source—recent the ensemble of photo-fragment ions. l and Pw are respectively the wavelength and developments and future trends. Energy Environ. Sci. 5, 8171–8181 (2012). measured average power of the incoming ultravoilet laser beam. 16. Boddien, A. et al. CO2-‘‘Neutral’’ hydrogen storage based on bicarbonates and formates. Angew Chem. Int. Ed. 50, 6411–6414 (2011). DFT calculations. The extensive search for lowest energy structures and transi- 17. Enthaler, S., von Langermann, J. & Schmidt, T. Carbon dioxide and formic tions states were performed by the hybrid B3LYP53 functional with def2-TZVP acid–the couple for environmental-friendly hydrogen storage? Energy Environ. atomic basis set54, which has been used for all atoms. Silver atoms have been Sci. 3, 1207–1217 (2010). treated by Stuttgart relativistic effective core potential with corresponding atomic 18. Wienho¨fer, G. et al. General and selective iron-catalyzed transfer orbital (AO) basis set55. The same combination of functional and basis set was used hydrogenation of nitroarenes without base. J. Am. Chem. Soc. 133, for calculation of the infrared spectra, which were scaled by 0.98 to match 12875–12879 (2011).

NATURE COMMUNICATIONS | 7:11746 | DOI: 10.1038/ncomms11746 | www.nature.com/naturecommunications 7 115 ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms11746

¼ 19. Braden, D. J., Henao, C. A., Heltzel, J., Maravelias, C. T. & Dumesic, J. A. addition product [Ag(OAc)(dppm)]2 (dppm Ph2PCH2PPh2). Inorg. Chem. Production of liquid hydrocarbon fuels by catalytic conversion of 34, 520–523 (1995). biomass-derived levulinic acid. Green Chem. 13, 1755–1765 (2011). 47. Szłyk, E. et al. X-ray crystal structure of [Ag4(m-dppm)2(m-C2F5COO)4]. 20. Herrmann, W. A. & Muehlhofer, M. in Applied Homogeneous Catalysis Synthesis and spectroscopy of silver(I) perfluorinated carboxylate with Organometallic Compounds, 2nd edn, Vol. 3 (eds Cornils, B. and complexes with bis(diphenylphosphino)methane. Dalton Trans. 17, 3404–3410 Herrmann, W. A.), 1086 (Wiley-VCH, 2002). (2003). 21. Odabasi, C., Gunay, M. E. & Yildirim, R. Knowledge extraction for water gas 48. Tate, B. K. et al. A dinuclear silver hydride and an umpolung reaction of CO2. shift reaction over noble metal catalysts from publications in the literature Chem. Sci. 4, 3068–3074 (2013). between 2002 and 2012. Int. J. Hydrogen Energy 39, 5733–5746 (2014). 49. Chen, P. Electrospray ionization tandem mass spectrometry in high- 22. Saito, K. et al. Unimolecular decomposition of formic acid in the gas phases on throughput screening of homogeneous catalysts. Angew Chem. Int. Ed. 42, the ratio of the competing reaction channels. J. Phys. Chem. A 109, 5352–5357 2832–2847 (2003). (2005). 50. Colorado, A. & Brodbelt, J. An empirical approach to estimation of critical 23. Chang, J.-G., Chen, H.-T., Xu, S. & Lin, M. C. Computational study on the energies by using a quadrupole ion trap. J. Am. Soc. Mass Spectrom. 7, kinetics and mechanisms for the unimolecular decomposition of formic and 1116–1125 (1996). oxalic acids. J. Phys. Chem. A 111, 6789–6797 (2007). 51. Bakker, J. M., Besson, T., Lemaire, J., Scuderi, D. & Maitre, P. Gas-phase 24. Loges, B., Boddien, A., Ga¨rtner, F., Junge, H. & Beller, M. Catalytic generation structure of a p-allyl-palladium complex: efficient infrared spectroscopy in a of hydrogen from formic acid and its derivatives: useful hydrogen storage 7 T Fourier transform mass spectrometer. J. Phys. Chem. A 111, 13415–13424 materials. Topics Catal. 53, 902–914 (2010). (2007). 25. Bi, Q.-Y. et al. Efficient subnanometric gold-catalyzed hydrogen generation via 52. Prazeres, R., Glotin, F., Insa, C., Jaroszynski, D. A. & Ortega, J. M. Two-colour formic acid decomposition under ambient conditions. J. Am. Chem. Soc. 134, operation of a free-electron laser and applications in the mid-infrared. Eur. 8926–8933 (2012). Phys. J. D 3, 87–93 (1998). 26. Tedsree, K. et al. Hydrogen production from formic acid decomposition at 53. Becke, A. D. Density functional thermochemistry. III. The role of exact room temperature using a Ag–Pd core–shell nanocatalyst. Nat. Nanotechnol. 6, exchange. J. Chem. Phys. 98, 5648–5652 (1993). 302–307 (2011). 54. Scha¨fer, A., Huber, H. & Ahlrichs, R. Fully optimized contracted Gaussian 27. O’Hair, R. A. J. in Reactive Intermediates. MS Investigations in Solution basis sets of triple zeta valence quality for atoms Li to Kr. J. Chem. Phys. 100, (ed Santos, L.S.) Ch. 6, 199–227 (Wiley-VCH, 2010). 5829–5835 (1994). 28. O’Hair, R. A. J. & Rijs, N. J. Gas phase studies of the Pesci decarboxylation 55. Andrae, D., Haeussermann, U., Dolg, M., Stoll, H. & Preuss, H. Energy- reaction: synthesis, structure, and unimolecular and bimolecular reactivity of adjusted ab initio pseudopotentials for the second and third row transition organometallic ions. Acc. Chem. Res. 48, 329–340 (2015). elements. Theor. Chim. Acta 77, 123–141 (1990). 29. Duncan, M. A. Spectroscopy of metal ion complexes: gas-phase models for solvation. Ann. Rev. Phys. Chem. 48, 69–93 (1997). Acknowledgements 30. MacAleese, L. & Maıˆtre, P. Infrared spectroscopy of organometallic ions in the gas phase: from model to real world complexes. Mass Spectrom. Rev. 26, We thank the ARC for financial support via grant DP150101388 (to R.A.J.O. and P.D.). 583–605 (2007). A.Z. acknowledges the award of an Australian Postgraduate PhD Scholarship. The 31. Antoine, R. & Dugourd, P. Visible and ultraviolet spectroscopy of gas phase research leading to these results has received funding from the European Research protein ions. Phys. Chem. Chem. Phys. 13, 16494–16509 (2011). Council under the European Union’s Seventh Framework Programme (FP7/2007-2013 32. Lanucara, F. et al. Naked five-coordinate Fe(III)(NO) porphyrin complexes: Grant Agreement No. 320659). Financial support from the French FT-ICR network (FR vibrational and reactivity features. Inorg. Chem. 50, 4445–4452 (2011). 3624 CNRS) for conducting the research is gratefully acknowledged. V.B.-K. and M.K. 33. Grimme, S., Antony, J., Ehrlich, S. & Krieg, H. A consistent and accurate ab acknowledge Prof. Miroslav Radman at MedILS and Split-Dalmatia County for kind support. initio parameterization of density functional dispersion correction (DFT-D) for the 94 elements H-Pu. J. Chem. Phys. 132, 154104 (2010). 34. Donald, W. A., Khairallah, G. N. & O’Hair, R. A. J. The effective temperature of Author contributions ions stored in a linear quadrupole ion trap mass spectrometer. J. Am. Soc. Mass A.Z. identified and optimized routes to the gas-phase formation of the silver hydride Spectrom. 24, 811–815 (2013). and silver formate complexes, carried out the CID, IMR and UVPD experiments 35. Kamer, P. & van Leeuwen, P. W. N. M. Phosphorus(III) Ligands in and contributed to manuscript preparation. G.N.K carried out the IRMPD Homogeneous Catalysis: Design and Synthesis (Wiley, 2012). experiments and contributed to manuscript preparation. M.K. carried out all the 36. Gorin, D. J., Sherry, B. D. & Toste, F. D. Ligand effects in homogeneous au DFT calculations and contributed to manuscript preparation; M.G., S.D., R.A., P.M., catalysis. Chem. Rev. 108, 3351–3378 (2008). R.J.M., S.-A.A. performed experiments and/or analyzed data and/or provided intellectual 37. Eller, K. & Schwarz, H. Organometallic chemistry in the gas phase. Chem. Rev. input; V.B.-K. and P.D. contributed to the project design, interpretation of data and 91, 1121–1177 (1991). writing of the manuscript; R.A.J.O. devised the project, contributed to the design of 38. Jena, P. & Castleman, Jr A. W. Cluster chemistry and dynamics special feature experiments and interpretation of data, project management and writing of the introductory perspective clusters: a bridge across the disciplines of physics and manuscript. chemistry. Proc. Natl Acad. Sci. USA 103, 10560–10569 (2006). 39. Waters, T., O’Hair, R. A. J. & Wedd, A. G. Catalytic gas phase oxidation of methanol to formaldehyde. J. Am. Chem. Soc. 125, 3384–3396 (2003). Additional information 40. O’Hair, R. A. J. The 3D quadrupole ion trap mass spectrometer as a complete Supplementary Information accompanies this paper at http://www.nature.com/ chemical laboratory for fundamental gas phase studies of metal mediated naturecommunications chemistry. Chem. Commun. 14, 1469–1481 (2006). Competing financial interests: The authors declare no competing financial interests. 41. O’Hair, R. A. J. Gas-phase studies of metal catalyzed decarboxylative cross-coupling reactions of esters. Pure. Appl. Chem. 87, 391–404 (2015). Reprints and permission information is available online at http://npg.nature.com/ 42. Schwarz, H. Chemistry with methane: concepts rather than recipes. Angew reprintsandpermissions/ Chem. Int. Ed. 50, 10096–10115 (2011). How to cite this article: Zavras, A. et al. Ligand-induced substrate steering and 43. Schwarz, H. How and why do cluster size, charge state, and ligands affect the þ course of metal-mediated gas-phase activation of methane? Isr. J. Chem. 54, reshaping of [Ag2(H)] scaffold for selective CO2 extrusion from formic acid. 1413–1431 (2014). Nat. Commun. 7:11746 doi: 10.1038/ncomms11746 (2016). 44. Agrawal, D. & Schro¨der, D. Insight into solution chemistry from gas-phase experiments. Organometallics 30, 32–35 (2011). This work is licensed under a Creative Commons Attribution 4.0 45. Schro¨der, D. Applications of electrospray ionization mass spectrometry in International License. The images or other third party material in this mechanistic studies and catalysis research. Acc. Chem. Res. 45, 1521–1532 article are included in the article’s Creative Commons license, unless indicated otherwise (2012). in the credit line; if the material is not included under the Creative Commons license, 46. Neo, S. P., Zhou, Z. Y., Mak, T. C. W. & Hor, T. S. A. Solid-state tetramer vs. users will need to obtain permission from the license holder to reproduce the material. solution-state dimer reinvestigation of [Ag2(OAc)2(dppm)]2 and its dppm To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/

8 NATURE COMMUNICATIONS | 7:11746 | DOI: 10.1038/ncomms11746 | www.nature.com/naturecommunications 116 4 Selectivity Effects in Bimetallic Catalysis: Role of the Metal Sites in the Decomposition of

Formic Acid into H2 and CO2 by the Coinage Metal Binuclear Complexes [dppmMM’(H)]+

117 118 DOI:10.1002/cctc.201601675 Full Papers

Selectivity Effects in Bimetallic Catalysis:Role of the Metal Sites in the Decomposition of Formic Acid into H2 and CO2 by the Coinage Metal Binuclear Complexes [dppmMM’(H)]+ Athanasios Zavras,[a] Marjan Krstic´,[b] Philippe Dugourd,[c] Vlasta Bonacˇic´-Koutecky´,*[b, d] and Richard A. J. O’Hair*[a]

Design of new bimetallic catalysts requires an understanding magnitude. This is akey step in the selective, catalyzedextru- of how cooperative effects of the metal sites influences reactiv- sion of carbon dioxide from formic acid, HO2CH, with impor- ity.Hereweshow how switching one or both of the silver tant applicationsinhydrogen storage andinsitu generation of + + atoms in binuclear silver hydride cations,[dppmAg2(H)] H2.Decarboxylationof[dppmMM’(O2CH)] throughcollision (dppm =1,1-Bis(diphenylphosphino)-methane), with all combi- induced dissociation regenerates [dppmMM’(H)] +.DFT calcula- nations of copperand/or gold maintains selective dehydrogen- tions provide insights into these cooperative effects. The ation of formic acid, enhancing reactivity by up to 2orders of copperhomobinuclear catalyst performs best overall.

Introduction

Catalytic cooperative effects are ubiquitous in nature and in candidates for developing adeeper understanding of bimetal- synthetic systems.[1] The requirement for two metals,either of lic catalysis sincethey allow the role of the ligand and each of the same element or of two different elements, to complete the metal centres to be examined.[5] By using multistage mass acatalytic cycle represents an important synergistic effect[2] in spectrometry techniques[6] to study these systems in the gas- both heterogeneous[3] and homogenous catalysis[4,5] and has phase[7] it is possible to examine the role of each metal centre been termed bimetallic catalysis.[3–5] Although, single site ho- on the elementary steps of acatalytic cycle involving homo- mogenous catalysis represents an attractiveway of “bottom- and heteronuclear clusters.[8] up” design of catalysts, the design principles for bimetallic cat- As one of the few organic liquids with potential for hydro- alysis are still not well understood.Incases where reactivity is gen storage applications, formicacid has attracted the atten- catalytic in one metal and stoichiometric in the othermetal, tion of the catalysis community.[9] In the absence of acatalyst, transmetallation can be the crucial step for asuccessful catalyt- high temperatures are required forthe gas-phase decarboxyla- ic cycle.[4] Catalyststhat contain two metal sites are attractive tion of formic acid [Eq. (1)],which is in competitionwith dehydration [Eq. (2)].[10] Thus there has been considerable interest in developing metal catalysts to selectively decarboxylate [a] A. Zavras, Prof. R. A. J. O’Hair [11] SchoolofChemistry and Bio21Molecular Science and formic acidatlow temperatures. Biotechnology Institute University of Melbourne HCO2H H2 CO2 1 30 Flemington Rd Parkville, Victoria 3010 (Australia) ! þ ð Þ HCO H H O CO 2 E-mail:[email protected] 2 ! 2 þ ð Þ [b] M. Krstic´,Prof. V. Bonacˇic´-Koutecky´ Center of Excellencefor Science and Technology We recently showed that the choice of ligand is crucial to Integration of MediterraneanRegion (STIM) at Interdisciplinary developing atwo-step catalytic cycle for the selective extru- Center for Advanced Sciences and Technology(ICAST) + sion of carbon dioxide from formic acid by [dppmAg2(H)] University of Split (dppm =bis(diphenylphosphino)-methane) in the gas phase Mesˇtrovic´evo Sˇetalisˇte 45 21000Split (Croatia) (Figure 1).[12] The dppm ligand was found to reshapethe ge- [c] Prof. P. Dugourd + Institut Lumire Matire, CNRS ometry of the binuclear Ag2(H) scaffold, thereby switching on + Univ Lyon, UniversitØ Claude Bernard Lyon 1 dehydrogenation to produce [dppmAg2(O2CH)] and H2 (Fig- F-69622,Lyon (France) + ure 1a,Step 1). Decarboxylation of [dppmAg2(O2CH)] through [d] Prof. V. Bonacˇic´-Koutecky´ + collisioninduced dissociation (CID) regenerates [dppmAg2(H)] ChemistryDepartment (Figure 1a,Step 2). Both silver sites are involvedinthe crucial Humboldt University of Berlin Brook-Taylor-Strasse 212489 Berlin (Germany) transition states for dehydrogenation and decarboxylation, Supporting informationfor this article can be found under: with one acting as an “anchor” for the oxygen of formic acid http://dx.doi.org/10.1002/cctc.201601675. (step 1, Figure 1b)orthe coordinated formate (step 2, Fig-

ChemCatChem 2017, 9,1298 –1302 1298 2017 Wiley-VCH Verlag GmbH &Co. KGaA, Weinheim

119 Full Papers

dride complexes devoid of any isobaricimpurities for ion-molecule reactions with formic acid in aseries of MS3 experiments.

Step 1ofthe catalytic cycle:Ion-moleculereactions between [dppmMM’(H)] + and formicacid All hetero-and homobimetallic hydride complexes react with formicacid to regenerate the formate complex + [dppmMM’(O2CH)] (Figure 1a,Step 1and Figure 2). To evalu-

Figure 1. Selective decarboxylation of formic acid. a) Twostepcatalytic cycle involving bimetallic catalysis. DFT calculations highlight the role of both metal centers for:b)Transition state (TS) for dehydrogenation of formic acid + 12 + [12] for [dppmAg2H] . (c) TS for decarboxylation for [dppmAg2(O2CH)] .

ure 1c)while the other triggersdehydrogenation or hydride + + + transfer.[L2Ag2(H)] ,[L2Ag4(H)3] ,[L2Ag4(O2CH)(H)2] and + [L2Ag4(O2CH)2(H)] are unreactivetowards HCO2Hand as such the stoichiometry of the metal complex is crucial for this catalytic cycle.[13] Here we use MS experiments and DFT calculations to examine the chemistry of all of the related homo-and heterobinu- + 10 Figure 2. MS3 LTQspectra obtained in a2Dlinear-ion trap at 298 Kshow- clear complexes [dppmMM’(H)] of the d coinage metals in ing the ion-moleculereaction of formic acid with mass-selected hydrides, which MorM’= Cu, Ag and Au.[14] This allows an evaluation of [LMM’H]+ (L=dppm) for: how cooperative effects between the metal centers influence 9 3 a) M=M’ =Cu, tAct. =60 ms, [HO2CH]ion trap =2.2410 molecules cmÀ , 9 3 both steps of the catalytic cycle. b) M=M’=Ag, tAct. =1000 ms,[HO2CH]ion trap =4.7510 molecules cmÀ , 9 3 c) M= M’ =Au, tAct. =20 ms, [HO2CH]ion trap =2.2410 molecules cmÀ , 9 3 d) M =Cu;M’ =Ag, tAct. = 30 ms, [HO2CH]ion trap =3.7210 moleculescmÀ , 9 3 e) M=Cu;M’=Au tAct. =50 ms, [HO2CH]ion trap = 8.1910 moleculescmÀ , Results and Discussion 8 3 f) M =Ag;M’=Au tAct. =1000 ms, [HO2CH]ion trap =8.9610 molecules cmÀ . A*represents the mass-selected precursor ion. The most intense peakin Formation of [dppmMM’(H)] + the cluster is representedbythe m/z value. tAct. =Activation time Our entry into the catalytic cycle was via the coordinated for- + mates, [dppmMM’(O2CH)] ,which weretransferred to the gas- phase using electrospray ionization (ESI) of a50mm acetonitrile ate the roles of the metal centers on reactivity,the temporal + solution containing amixture of Cu2O:Ag2O:AuClPPh3 (1:1:2) to decay of the reactantion, [dppmMM’(H)] ,was monitored which 10 equivalents of formic acid was added (Supporting in- over arange of activation times and concentrationsofformic formation Figure S1aand relateddiscussion). All homo and acid (Figure S5) to yield rate constants, which if compared to hetero binuclear formates were formed from this solution, as the predictedcollisionrates gave the reactionefficiencies confirmed through their isotope patterns(Figure S1b) and listed Table 1.[17] The experimentally observedreactivityorder + + + high-resolution mass spectrometry (HRMS)experiments (Fig- follows:[dppmAu2(H)] [dppmCu2(H)] > [dppmCuAu(H)] + + + ure S2). The HRMS experiments also identified the presence of [dppmCuAg(H)] @ [dppmAgAu(H)] [dppmAg2(H)] with + isobaric impurities for [dppmAu2(O2CH)] (m/z 823, Fig- aca. 2orders of magnitude differenceinreactivity between + ure S2b), [dppmCuAu(O2CH)] (m/z 689, Figure S2d), and the most and least reactive complexes. + 2 [dppmAgAu(O2CH)] (m/z 733, Figure S2e). CID (MS )of the mass-selected formate complexes, [dppmMM’(O CH)] + 2 Step 2ofthe catalytic cycle:CID of [dppmMM’(O CH)]+ to (Figure S3), generatedabundant hydride complexes 2 regenerate[dppmMM’(H)] + [dppmMM’(H)]+ through decarboxylation with the exception + + of [dppmAu2(H)] (m/z 779, Figure S3b) and [dppmAgAu(H)] We next examinedthe ease of decarboxylation (Step 2, (m/z 689, Figure S3e). The elemental composition of each of Figure 1). Owing to the presence of isobaric impurities from the hydride complexes generated throughthe decarboxylation ESI/MSasdiscussed previously,the formate complexes formed of the coordinated formates (Figure 1a,Step 2) was confirmed through ion-moleculereactions betweenthe hydridesand by HRMS (Figure S4). Subsequent mass-selection provided hy- formicacid (Step 1), were mass-selected and allowed to under-

ChemCatChem 2017, 9,1298 –1302 www.chemcatchem.org 1299 2017 Wiley-VCH Verlag GmbH &Co. KGaA, Weinheim

120 Full Papers

DFT calculations on the mechanisms and energetics of both Table 1. Rates of ion-molecule reactionsbetween [LMM’(H)] + and formic acid (L=dppm). steps associated with the catalytic cycle

[a,b,c] Reactant ion kexpt ReactionDFT Eact [eV] The DFT calculated energy diagrams are largely consistent with f [d] [f] efficiency ( ) calculated the experimentsand provideinsights into how reactivity is + 9 [LCu2(H)] 1.20 0.0110À 113.6 6.1 0.66 (0.07) modulated by the nature of the metal centers (Figure 3and + Æ 11 Æ À [LAg2(H)] 1.53 0.0310À 1.4 0.1 0.06 (0.18) +[e] Æ 9 Æ À Supporting Information Figures S8–S11). [LAu2(H)] 1.47 0.0910À 141.0 9.2 1.02 (0.06) + Æ 10 Æ À [LCuAg(H)] 3.31 0.210À 31.5 2.1 0.25 (0.47) Step 1ofthe catalytic cycle must be an exothermicprocess + Æ 10 Æ À [LCuAu(H)] 4.51 0.410À 43.1 4.1 0.30 (0.74) with barriers that lie below the separated reactants for it to + Æ 11 Æ À [LAgAu(H)] 3.33 0.310À 3.2 0.3 0.08 (0.18) [18] Æ Æ À occur under the near thermalconditions of the ion-trap. 3 1 1 [a] Mean standard deviation (n=3). [b] In units of cm moleculesÀ sÀ . Indeedthis is the case, with the most exothermic reaction oc- Æ + + [c] Rates for the reaction with formicacid with [LMM’H] to regenerate curring for [dppmCu2(H)] (Table S2). In all cases an initial com- + [LMM’(O2CH)] as the product. Rates were determined by monitoringthe plex between the hydride, [dppmMM’(H)]+,and formic acid is decayofthe reactant ion with aknown concentrationofformic acid over formed, which then proceeds via asingle transition state to time.[d] Reactionefficiency (f)=(kexpt/kADO)100. The kADO is the theoretical ion-molecule collision rate constant obtained from the average-dipole produce H2 and the thermodynamically favoured O,O-bridged [15] + orientation (ADO) theory, which was calculatedusing the Colratepro- formate complex, [dppmMM’(O2CH)] .Inall cases this critical [16] gram. [e] data from ref.[12].[f] Eact relative to separated reactants (or TS has astructure in which one metal site acts as an “anchor” relative to initial complex). for the oxygen of the formic acid whilethe other metal site containsthe hydride that reacts by dehydrogenation.Wehave performed aMulliken chargeanalysisofthe precursor ions and go CID in aMS4 experiment (Figure S6). This precluded energy their complexes with formic acid (Figure S12). With the excep- + + resolved CID measurements of thresholds for decarboxylation. tion of [dppmCu2(H)] and [dppmCuAg(H)] ,all precursor ions In all cases decarboxylation was the major fragmentation path- have partial negative charges on the coordinated hydride and way (Scheme S1), with competing formation of [dppmM]+ apartial positive charge at the metal center(s). For the initial and/or [dppmM’]+ being only minor channels.Atanormalized complexes between the hydride, [dppmMM’(H)]+,and formic collisionenergy of 15 %and an activation time of 10 ms, the acid, the oxygen atom of the coordinated formic acidhas amount of hydride formed (relative to all ions present in the anegative charge, whereas the O-H has the expected ( )(+) À CID spectrum) through decarboxylation follows the order of: dipole.Taken together,this suggests that these can be regard- + + [dppmAg2(O2CH)] (88.6%) [dppmCu2(O2CH)] (85%) > ed as dehydrogenation reactions in which the coordinated hy- + + [dppmCuAu(O2CH)] (76.6 %) [dppmCuAg(O2CH)] (76.5%) dride reacts with the acidic O-H protonofthe formic acid. In + + > [dppmAgAu(O2CH)] (45%) > [dppmAu2(O2CH)] (28.1 %). the case of the homobinuclearcomplexes, the potentialwell

Figure 3. DFT-calculatedenergyprofile for the preferred site of reactivity of the heterobinuclear complex for the two reaction steps in the catalytic cycle of + + Figure 1a.Step 1: ion-molecule reaction of formic acid with [dppmCuAg(H)] at the copper site;step 2: CIDdecarboxylation of [dppmCuAg(O2CH)] with hy- dridetransfer to the silver site. Relative energiesare in eV.All structures were fully optimized using DFT methodwith the hybrid B3LYP functional and def2- TZVPatomic basisset which has been usedfor all atoms. Silveratoms have been treated by Stuttgart relativistic effective core potential(RECP)with correspondingAObasis set. Silver=Ag,Yellow= Cu.

ChemCatChem 2017, 9,1298 –1302 www.chemcatchem.org 1300 2017 Wiley-VCH Verlag GmbH &Co. KGaA, Weinheim

121 Full Papers depth associated with the formationofthe initial complex fol- Experimental Section + + lows the order [dppmAu2(H)] ( 1.08 eV) > [dppmCu2(H)] + À Mass spectrometry experiments:Gas-phase experiments on phos- ( 0.73 eV) > [dppmAg2(H)] ( 0.24 eV) and thus dictates the À [19] À phine ligated bimetallic formate clusters, formed as discussed in observedreactivity. For the heterobinuclear complexes, the Supporting Information, were performed using aFinnigan formic acid can approacheither of the two different metal cen- hybrid linear quadrupole Fourier transform ion-cyclotron resonance tres. Adetailed examination of all possible reactant complexes (LTQ FTICR) mass spectrometer modified to allow the study of and transitionstates for attack at both copper and silver was IMR.[20] The unimolecular fragmentation/dissociation of mass-select- performed for [dppmCuAg(H)]+ (Supporting Information ed phosphine ligated bimetallic clusters occurred through CID Table S2). The most favoured site of attack is at copper using anormalized collision energy between 20–25%and an acti- (Figure 3), consistentwith the relative reactivity order for the vation time of 30 ms. The CID isolation width was 5–8 m/z from the centre of the ion cluster distribution. IMRs were carried by de- homobinuclear complexes.For [dppmCuAu(H)] + the preferred livering ameasured concentration of formic acid into the helium site of reactivity is Cu (Figures S10), whereas for [dppmA- bath gas. gAu(H)] + it is Au (Figures S11). Overall, there is good agreement between theory and experiment. The DFT predicted acti- DFT calculations:The extensive search for lowest energy structures and transitions states was performed using the hybrid B3LYP[21] vation energies forstep 1are inversely related to the mea- functional with def2-TZVP atomic basis set for all atoms.[22] Silver sured reaction efficiencies (compare columns 4and 3of and/or gold atoms were treated by the Stuttgart relativistic effec- Table 1). tive core potential (RECP) with the corresponding AO basis set.[23] In contrast, decarboxylation is endothermic as it requires en- Potential interactions between the aromatic rings of the dppm ergization of [dppmMM’(O CH)]+ through multiple collisions 2 ligand, raises the question of whether dispersion corrections within with the helium bath gas during the CID process. The DFT cal- DFT are required. We have tested the influence of dispersion cor- culationsreveal that both metal centers play arole in the + rection on the structural properties of [dppmAg2H] (dppm=1,1- mechanism for CO release, which involves two steps (Figure 3 + 2 Bis(diphenylphosphino)-methane) and [dppmAg2(O2CH)] com- + [24] and Figures S8–10), except for [dppmAgAu(O2CH)] ,which plexes by introducing D3 into the DFT. Since the energy profile only requires asingletransition state (Figures S11). The first remained almost unchanged when single point B3LYP-D3 energy step involves breaking one of the M Obonds to isomerize the calculations were performed for each reaction step of the catalytic À O,O- bridged formate to its O-bound form, and is the rate de- cycle we have not used it for all profiles. The use of other function- [25] [26] + als with high averaged accuracy such as TPSSh and M062X termining step except for [dppmAg2(O2CH)] and + does not change the overall energy profiles. [dppmCuAu(O2CH)] .The next step involves decarboxylation, to give the Obound [dppmMM’(H)(OCO)]+ complex, which then loses CO2.For the heterobinuclear complexes, hydride Acknowledgements transfer from the coordinated formate can occur to either metal centre. Adetailed examination of all possible transition We thank the ARC for financial supportthrough grant states and intermediates associated with hydride transfer to DP150101388 (to RAJO and PD). AZ acknowledges the award of either copper or silver sites was performed for an Australian Postgraduate PhD Scholarship. The research lead- + [dppmCuAg(H)] (Supporting Information Table S2). In step 2, ing to these results has received fundingfrom the European Re- the most favoured site of attack is at silver (Figure3), consis- search Council under the European Union’s Seventh Framework tent with both the experimentally determined relative reactivi- Programme (FP7/2007–2013 Grant agreement No. 320659). VBK ty order and DFT calculated energetics for the homobinuclear and MK acknowledgecomputationalfacilities of the supercom- complexes. puter “Bura” at the Universtiy of Rijeka and SRCE at University of Zagreb as well as Prof. Miroslav Radman at MedILS and Split- DalmatiaCounty for kind support. Conclusions Keywords: bimetallic catalysis · DFT calculations · formic acid · In conclusion, the DFT calculations show that both metal cen- mass spectrometry · selective decarboxylation ters play arole in both steps of the catalytic cycle. One metal site acts as an “anchor” for the oxygen of formic acid (step 1) or formate (step 2) while the other site facilitates dehydrogena- [1] J. I. van derVlugt, Eur.J.Inorg. Chem. 2012,363 –375. [2] For arecent review on synergistic effects in isolatedtransition metal [19] tion (step 1) or hydride transfer during decarboxylation of systems in the gas-phase, see:G.Niedner-Schatteburg, Cooperative Ef- the coordinated formate (step 2). Since each metal centermay fects in Clustersand Oligonuclear Complexes of Transition Metals in Iso- influenceeach step of acatalytic cycle in adifferent way,the lation, Struct. Bond.,inpress DOI:10.1007/430 2016 11. overall preferred bimetallic catalyst is that whichrepresents [3] W. Yu,M.D.Porosoff, J. G. Chen, Chem.Rev. 2012, 112,5780 –5817. [4] M. H. PØrez-Temprano, J. A. Casares, P. Espinet, Chem. Eur.J.2012, 18, acompromise in reactivity for all steps as well as the cost of 1864 –1884. the metal. This is the case here, in which the cheaper,earth [5] N. P. Mankad, Chem. Eur.J.2016, 22,5822 –5829. + [6] a) R. A. J. O’Hair, Chem.Commun. 2006,1469 –1481;b)R.A.J.O’Hair, abundant coppercatalyst[dppmCu2(H)] (dppm= 1,1-Bis(diphenylphosphino)-methane) is the second most reactive com- N. J. Rijs, Acc. Chem. Res. 2015, 48,329 –340;c)R.A.J.O’Hair, Pure. Appl. Chem. 2015, 87,391–404. + plex and is regenerated from [dppmCu2(O2CH)] slightly less [7] For reviews on the use of MS to examine catalysis in the gas phase see: + efficiently that the hydride from CID of [dppmAg2(O2CH)] . a) D. K. Bçhme, H. Schwarz, Angew.Chem. Int. Ed. 2005, 44,2336–2354;

ChemCatChem 2017, 9,1298 –1302 www.chemcatchem.org 1301 2017 Wiley-VCH Verlag GmbH &Co. KGaA, Weinheim

122 Full Papers

Angew.Chem. 2005, 117,2388 –2406;b)D.Schrçder, Acc. Chem. Res. Deakin University,Geelong, Victoria, Australia:http://www.deakin. 2012, 45,1521 –1532;c)H.Schwarz, Isr.J.Chem. 2014, 54,1413 –1431; edu.au/ ~lim/programs/COLRATE.html. d) R. A. J. O’Hair, Int. J. Mass Spectrom. 2015, 377,121–129. [17] Several studies have highlighted that the measured rate constantsof [8] For examples of bimetallic reactivity in the gas-phase, see:a)K.Koszi- metal clusters can exceed the theoreticalADO rate constants, which nowski, D. Schrçder,H.Schwarz, J. Am. Chem. Soc. 2003, 125,3676 – use apointcharge model, by up to afactor of 3–4:a)O.P.Balaj, I. Bal- 3677;b)K.Koszinowski, D. Schrçder,H.Schwarz, ChemPhysChem 2003, teanu, T. T. J. Rossteuscher,M.K.Beyer,V.E.Bondybey, Angew.Chem. 4,1233 –1237;c)K.Koszinowski, D. Schrçder,H.Schwarz, Organometal- Int. Ed. 2004, 43,6519 –6522; Angew.Chem. 2004, 116,6681 –6684; lics 2004, 23,1132–1139;d)H.AlSharif, K. L. Vikse, G. N. Khairallah, b) M. L. Anderson, M. S. Ford, P. J. Derrick, T. Drewello, D. P. Woodruff, R. A. J. O’Hair, Organometallics 2013, 32,5416–5427;e)G.N.Khairallah, S. R. Mackenzie, J. Phys. Chem. A 2006, 110,10992–11000. G. R. da Silva, R. A. J. O’Hair, Angew. Chem. Int. Ed. 2014, 53,10979 – [18] a) S. Gronert, J. Am. Soc. MassSpectrom. 1998, 9,845 –848;b)W.A. 10983; Angew.Chem. 2014, 126,11159–11163;f)K.L.Vikse, A. Zavras, Donald, G. N. Khairallah, R. A. J. O’Hair, J. Am. Soc. MassSpectrom. 2013, T. H. Thomas, A. Ariafard,G.N.Khairallah, A. J. Canty,B.F.Yates, R. A. J. 24,811 –815. O’Hair, Organometallics 2015, 34,3255 –3263. [19] Althoughthere is adearth of experimental thermochemical data for [9] a) M. Grasemann, G. Laurenczy, Energy Environ. Sci. 2012, 5,8171 –8181; these systems,the DFT calculated binding energy orders are consistent b) S. Enthaler,J.von Langermann, T. Schmidt, Energy Environ. Sci. 2010, with the relative atomic metal cation water binding energyorders: + + + 3,1207 –1217. D(H2O-Au ) > D(H2O-Cu ) > D(H2O-Ag )(J. Roithovµ,D.Schrçder, [10] a) K. Saito, T. Shiose, O. Takahashi, Y. Hidaka, F. Aiba, K. Tabayashi, J. Coord.Chem. Rev. 2009, 253,666). Phys. Chem. A 2005, 109,5352 –5357;b)J.-G. Chang, H.-T.Chen,S.Xu, [20] W. A. Donald,C.J.McKenzie, R. A. J. O’Hair, Angew.Chem.Int. Ed. 2011, M. C. Lin, J. Phys. Chem. A 2007, 111,6789 –6797. 50,8379 –8383; Angew.Chem. 2011, 123,8529–8533. [11] B. Loges,A.Boddien,F.Gärtner, H. Junge, M. Beller, Top. Catal. 2010, 53, [21] A. D. Becke, J. Chem.Phys. 1993, 98, 5648 –5652. 902 –914. [22] A. Schäfer,H.Huber,R.Ahlrichs, J. Chem. Phys. 1994, 100,5829–5835. [12] A. Zavras,G.N.Khairallah, M. Krstic´,M.Girod, S. Daly,R.Antoine, P. [23] D. Andrae, U. Haeussermann,M.Dolg, H. Stoll, H. Preuss, Theor.Chim. Maitre, R. J. Mulder, S.-A. Alexander,V.Bonacˇic´-Koutecky´,P.Dugourd, Acta. 1990, 77,123–141. R. A. J. O’Hair, Nat. Commun. 2016, 7,11746. [24] S. Grimme, J. Antony, S. Ehrlich, H. Krieg, J. Chem.Phys. 2010, 132, [13] A. Zavras, J. M. White, R. A. J. O’Hair, Dalton Trans. 2016, 45,19408 – 154104. 19415. [25] J. Tao, J. P. Perdew,V.N.Stavorevov,G.E.Scuseria, Phys. Rev.Lett. 2003, [14] For acomprehensivereview on the formation and reactions of coinage 91,146401. metal hydridessee:A.J.Jordan, G. Lalic, J. P. Sadighi, Chem. Rev. 2016, [26] Y. Zhao, D. G. Truhlar, Theor.Chem. Acc. 2008, 120,215–241. 116,8318 –8372. [15] T. Su, M. T. Bowers, Classical Ion-Molecule CollisionTheory in Gas-Phase Ion Chemistry,Vol1, (Ed.:M.T.Bowers), Academic Press, New York, Manuscript received:December 21, 2016 1979, pp. 84–119 Revised:January 4, 2017 [16] K. F. Lim, Quantum ChemistryProgramExchange Bulletin 1994, 14,3.The Accepted Article published:February 2, 2017 programColrate is availablefor download from the author’swebsite at Final Articlepublished:March 6, 2017

ChemCatChem 2017, 9,1298 –1302 www.chemcatchem.org 1302 2017 Wiley-VCH Verlag GmbH &Co. KGaA, Weinheim

123

124 5 An unusual co-crystal [(μ2- 2 dcpm)Ag2(μ2-O2CH)(η -

NO3)]2·[(μ2-dcpm)2Ag4(μ2-

NO3)4] and its connection to the selective decarboxylation of formic acid in the gas phase

125

126 Dalton Transactions

View Article Online PAPER View Journal | View Issue

An unusual co-crystal [(µ2-dcpm)Ag2(µ2-O2CH) 2 (η -NO3)]2·[(µ2-dcpm)2Ag4(µ2-NO3)4] and its Cite this: Dalton Trans., 2016, 45, 19408 connection to the selective decarboxylation of formic acid in the gas phase†

Athanasios Zavras, Jonathan M. White and Richard A. J. O’Hair*

ESI/MS of an acetonitrile solution containing a mixture of AgNO3 : bis(dicyclohexylphosphino)methane + (dcpm, L) : NaO2CH in a molar ratio of 2 : 1 : 1 gave an abundant peak due to [LAg2(O2CH)] and a minor + peak assigned as [L2Ag4(O2CH)3] . When this acetonitrile solution was frozen and layered with diethyl ether and left undisturbed for six days, crystalline material suitable for X-ray crystallography was identiﬁed and separated from amorphous solids. Single crystal X-ray diﬀraction revealed an unusual co-crystal con- 2 sisting of two discrete tetranuclear silver clusters [(µ2-dcpm)Ag2(µ2-O2CH)(η -NO3)]2·[(µ2-dcpm)2Ag4(µ2-NO3)4]. + + Received 23rd September 2016, While all of the coordinated formates in [LAg2(O2CH)] and [L2Ag4(O2CH)3] can be decarboxylated in the Accepted 14th November 2016 + gas-phase under conditions of collision induced dissociation, only the hydride [LAg2(H)] thus formed DOI: 10.1039/c6dt03700c reacts with formic acid via protonation and liberation of H2 to regenerate to formate, thereby closing a www.rsc.org/dalton catalytic cycle for the selective decomposition of formic acid.

Introduction

Formic acid has emerged as one of the few organic liquids with potential for hydrogen storage applications.1 In the absence of a catalyst, high temperatures are required for decarboxylation to release hydrogen from formic acid. Furthermore, this process is non-selective, being in competition with decomposition into water and carbon monoxide.2 Thus there Published on 24 November 2016. Downloaded by The University of Melbourne Libraries 03/03/2017 10:05:22. has been considerable interest in developing metal catalysts to selectively decarboxylate formic acid at low temperatures.3 We recently described a two-step catalytic cycle for the selec-

tive decomposition of formic acid into CO2 and hydrogen in the gas phase (Scheme 1a, eqn (1)).4 The crucial steps involve: Scheme 1 (a) Catalytic cycle for the selective decarboxylation of (i) collision-induced dissociation (CID) of the binuclear bis- formic acid; previous crystal structures of phosphine ligated silver for- 2 5 6 + mates: (b) [(Ph3P)2Ag(η -O2CH)] and (c) [(Ph3P)3Ag(O2CH)]. R = Ph and phosphine silver formate, [(L)Ag2(O2CH)] 1,whichliberates 1a and 2a (ref. 4); R = Cy and 1b and 2b (this work). CO2 and results in the formation of the coordinated hydride, + [(L)Ag2(H)] 2; (ii) an ion-molecule reaction (IMR) between 2 and formic acid to regenerate 1 via protonation of the coordinated cycle were characterised by gas-phase IR and UV-Vis spectroscopy hydride with concomitant liberation of H2 (eqn (2)), thereby for the case of bis(diphenylphosphino)methane (dppm) ligand closing the catalytic cycle. Both resting states of the catalytic and were found to adopt the bidentate structures [(µ2-dppm) + + Ag2(µ2-O2CH)] 1a,and[(µ2-dppm)Ag2(µ2-H)] 2a.Thesegas phase results prompted variable temperature NMR experiments School of Chemistry and Bio21 Institute of Molecular Science and Biotechnology, on a mixture of AgBF4, dppm, sodium formate and formic acid, The University of Melbourne, Melbourne, Victoria 3010, Australia. which revealed the formation of CO and H at 65 °C. E-mail: [email protected] 2 2 †Electronic supplementary information (ESI) available: Asymmetric unit of 3; þ þ + ½LAg2ðO2CHÞ !½LAg2ðHÞ þ CO2 ð1Þ ion-molecule kinetics for reaction of [LAg2H] , 2b, with formic acid; crystallographic information for 3. CCDC 1505745. For ESI and crystallographic data in þ þ CIF or other electronic format see DOI: 10.1039/c6dt03700c ½LAg2ðHÞ þ HO2CH !½LAg2ðHÞ þ H2 ð2Þ

19408 | Dalton Trans.,2016,45, 19408–19415 This journal is © The Royal Society of Chemistry 2016 127 View Article Online

Dalton Transactions Paper

Since the formate complex, 1, represents the key entry point Mass spectrometry into the catalytic cycle, we were interested in determining its Mass spectra were recorded using a Thermo Finnigan linear structure via X-ray crystallography. While X-ray crystallographic ion trap (LTQ) mass spectrometer modified to allow ion- studies have characterised the structures of mononuclear silver molecule reactions to be carried out. The silver clusters 5,6 phosphine formate complexes (Scheme 1b and c) and several prepared in the condensed phase were introduced into the 7 silver phosphine carboxylate clusters, no cationic structures mass spectrometer via a syringe pump set at a flow rate of − with the required 2 : 1 : 1 stoichiometry for Ag : bisphosphine : 5 µL min 1 to the ESI capillary. The ESI conditions used, for formate have been published. Indeed, it appears that optimum intensity of the target ions, typically were: spray [(µ2-dppm)2Cu2(µ2-O2CCH3)]BF4 and [(µ2-dppm)2(4-vinyl-pyri- voltage, 4.2–5.0 kV, capillary temperature, 250 °C, nitrogen dine)Cu2(µ2-O2CH)]NO3 are the only cationic binuclear bisphos- sheath gas pressure, 5 (arbitrary units), capillary voltage. CID phine coinage metal carboxylates to have been characterised by experiments were carried out by mass selecting the entire 8 X-ray crystallography. Here we report the results of an attempt isotope cluster with a window of 8 m/z and applying an acti- to form crystals with the required stoichiometry and ligand vation energy between 15% and 25% NCE to allow collisions binding modes related to structure 1.Wealsodescribethe with the helium bath gas. An activation (Q) of 0.25 and acti- results of ESI multistage mass spectrometry experiments on vation time of 30 ms were used. IMR were carried by injecting solutions of this stoichiometry aimed at examining what cations formic acid into the helium bath gas, and rates were measured are formed and to provide a link to related gas-phase catalytic by varying the reaction time, as described previously.12 Under cycles for the selective decomposition of formic acid. IMR conditions, collisions with the helium bath gas quasi- thermalizes the ions to room temperature.12c The ADO average-dipole orientation theory rate coeﬃcient was calcu- Experimental section lated using the COLRATE program.13 Chemicals Chemicals from the following suppliers were used without Results and discussion further purification: Aldrich: (i) bis(dicyclohexylphosphino)

methane (dcpm, L) (97%), (ii) silver(I) tetrafluoroborate (AgBF4) ESI/MS to investigate clusters formed in solution from silver (98%), (iii) sodium formate (NaO2CH, 99%). Ajax Finechem: salts with dcpm HCOOH and NaO2CH (iv) formic acid (HCO2H, 99%), (v) silver(I) nitrate (AgNO3). ESI/MS was used to monitor the reactions of silver salts with Merck: (vi) acetronitrile (HPLC grade). bisphosphine ligand and sodium formate in an eﬀort to ident-

2 ify suitable conditions for the synthesis of a ligand protected Preparation of [(µ2-dcpm)Ag2(µ2-O2CH)(η -NO3)]2· silver formate cluster. The reaction of AgBF4 : dcpm : NaO2CH [(µ2-dcpm)2Ag4(µ2-NO3)4]3 + in a molar ratio of 2 : 1 : 1 yielded abundant [LAg2(O2CH)] AgNO3 (0.017 g, 0.10 mmol) and bis(dicyclohexylphosphino) (m/z 669, Fig. 1a) in agreement with the stoichiometric con- methane (0.020 g, 0.05 mmol) were added to a 10 mL glass vial ditions applied in the synthesis. The related dinuclear complex 3 + with a screw-cap. Acetonitrile (5 cm ) was added to the vial [LAg2(O2CH)] was observed for ESI/MS of solutions containing and the solution was sonicated for 1 minute. Formic acid was dppm instead of dcpm (data not shown). Although ESI/MS Published on 24 November 2016. Downloaded by The University of Melbourne Libraries 03/03/2017 10:05:22. added in excess and the solution was sonicated for a further has been previously used to direct the synthesis of silver 14 + 1 minute. NaO2CH (0.004 g, 0.05 mmol) was added to the solu- clusters and the observation of abundant [LAg2(O2CH)] tion under stirring until dissolved, ca. 5 minutes. The solution showed promise for the condensed phase synthesis of a di- was frozen with liquid nitrogen and layered with diethylether. nuclear silver formate, all attempts failed to produce crystalline The screw-cap was immediately fitted and sealed with Teflon material associated with a discrete cluster of the form of

and parafilm tape. The vial was wrapped in aluminium foil [LAg2(O2CH)]X, where L = dcpm or dppm, X = a non-coordinat- ﬀ − − and left undisturbed for 6 days to a ord crystals suitable for ing anion BPh4 and BF4 . Instead, slowly diﬀusing solutions X-ray crystallography. Although some decomposition and silver resulted in the formation of silver mirrors on the glass vial mirrors were observed, crystals suitable for X-ray crystallo- and in amorphous products of decomposition.

graphy were manually collected by sorting through the solid ESI/MS of an acetonitrile solution of AgNO3 : dcpm : NaO2CH material suspended in paraﬃn under a microscope. in a molar ratio of 2 : 1 : 1 also gave an abundant peak due to + [LAg2(O2CH)] (m/z 669, Fig. 1b). A minor peak of ca. 1.25% X-ray crystallography + observed at m/z 1383 was assigned as [L2Ag4(O2CH)3] .

Intensity data for compound 3 was collected on an Oxford When an acetonitrile solution of AgNO3 : dcpm : NaO2CH in Diffraction SuperNova CCD diffractometer using Cu-Kα radiation, a molar ratio of 2 : 1 : 1 was frozen and layered with diethyl the temperature during data collection was maintained at ether and left undisturbed for six days, a mixture of amor- 130.0(1) using an Oxford Cryostream cooling device. The structure phous and crystalline solids were observed at the solvent inter- wassolvedbydirectmethodsanddifference Fourier synthesis.9 face. There were two types of amorphous material. One was a The thermal ellipsoid plot was generated using the program grey flocculant material and the other was a blackened solid. ORTEP-3 10 integrated within the WINGX11 suite of programs. The solid material was suspended in paraffin and examined

This journal is © The Royal Society of Chemistry 2016 Dalton Trans.,2016,45, 19408–19415 | 19409 128 View Article Online

Paper Dalton Transactions

2 of [(µ2-dcpm)Ag2(µ2-O2CH)(η -NO3)]2 3.2294(4) are signifi- cantly longer, however remain below the sum of the van der Waals radii (3.4 Å)17 suggesting the possibility of argentophilic interactions.18 The Ag–P contacts which range from 2.333 to 2.564 Å are typical, while the P–C–P angle of, 112.2(2)°, is more acute than previously reported, this is a reflection of the

greater puckering of the (Ag2 dcpm chelate ring). Opposing the – dinuclear Ag Ag edge lies a formato ligand with a µ2-bridging coordination mode. The Ag–O contacts from the bridging formato ligand are asymmetrical with bond lengths of 2.161(3)

and 2.288(4) Å. The collective µ2-bridging dcpm and the formato ligands give a quasi-eight membered-ring (Ag2P2C2O2) with the silver atoms on opposing sides. The Ag(2) atom is che- lated by an η2-nitrato ligand, related binding has been observed by Mak and coworkers in a series of polymeric silver complexes and clusters.19 The overall [(µ -dcpm)Ag (µ -O CH) Fig. 1 Positive-ion LTQ ESI/MS of acetonitrile solutions of silver 2 2 2 2 (η2-NO )] cluster is a centrosymmetric dimer of [(µ -dcpm) formate clusters prepared from: (a) AgBF4 : dcpm : HCOOH : NaO2CH 3 2 2 η2 (2 : 1 : 1 : 1) and (b) AgNO3 : dcpm : HCOOH : NaO2CH (2 : 1 : 1 : 1). Ag2(µ2-O2CH)( -NO3)] moieties which form a 4-membered Solutions were diluted to 50 µm prior to ESI. L = dcpm = bis(dicyclo- (Ag–O–Ag–O) ring (Scheme 2a). A related four-membered hexylphosphino)methane. The m/z values shown are of the most Ag2O2 ring has previously been observed in a triphenyl- intense peak arising from isotopic abundances of the atoms. Other ions phosphine silver acetate complex [(Ph P) Ag (µ -O CCH ) ] formed in minor abundance are due to other ligated silver complexes 3 2 2 2 2 3 2 2 – formed by various combinations of Ag+, solvent, dppm ligand (or in its (Scheme 2b). While the Ag O distances linking the two halves oxidised formed) and anions. Their assignments are listed in ESI of this cluster are similar, the the formate ligand displays Table S1.† shorter contacts to the silver atom, Ag(1)–O(1) 2.161(3) Å, than the acetate carboxylate ligand (Scheme 2b) (Table 1).

under a light microscope to identify and manually separate Structure of [(µ2-dcpm)2Ag4(µ2-NO3)4] crystalline material, from the amorphous solids, suitable for An examination of the structure of the tetranuclear cluster

X-ray crystallography. [(µ2-dcpm)2Ag4(µ2-NO3)4] reveals that it can be considered as two L Ag fragments held together by four nitrato ligands, Structural characterization of [(µ -dcpm)Ag (µ -O CH)(η2-NO )] · 2 2 2 2 2 2 3 2 giving rise to four fused eight membered rings (Scheme 3). [(µ2-dcpm)2Ag4(µ2-NO3)4] (3) by X-ray crystallography Each L2Ag2 fragment adopts a quasi-staggered conformation. 15 Single crystal X-ray diﬀraction revealed an unusual co-crystal All four Ag–P bonds have comparable lengths, ranging consisting of two discrete tetranuclear silver clusters 2.3473(10)–2.3255(9) Å. The Ag–O distances range between η2 – [(µ2-dcpm)Ag2(µ2-O2CH)( -NO3)]2·[(µ2-dcpm)2Ag4(µ2-NO3)4] 2.217(3) 2.462(3) Å. The nitrato ligands bind in µ2-bridging (Fig. 2).‡ The former cluster has crystallographic inversion mode with the Ag–O bond length of ranging from 2.217(3)– Published on 24 November 2016. Downloaded by The University of Melbourne Libraries 03/03/2017 10:05:22. symmetry. When the asymmetric unit (Fig. S1†) is expanded, 2.462(3) Å. In comparison the Ag–O distance of a similar to show the crystal packing, the two clusters exhibit close inter- binding motif lies within the range 2.304–2.755 Å (Table 2).19a,d,f molecular contacts between axial and equatorial hydrogen Table S1† displays the intermolecular contacts at that exist

atoms from the cyclohexyl rings, from dcpm, on adjacent clus- between the two discrete clusters [(µ2-dcpm)Ag2(µ2-O2CH) † η2 ters (Table S1 ). These van der Waals interactions likely ( -NO3)]2 and [(µ2-dcpm)2Ag4(µ2-NO3)4]. There are short provide a stabilising force as a template for the crystal packing distances between axial and equatorial hydrogen atoms on of the clusters. cyclohexyl rings of adjacent clusters. The nitrato ligand oxygen atoms also form short contacts with the methylene hydrogens Structure of [(µ -dcpm)Ag (µ -O CH)(η2-NO )] 2 2 2 2 3 2 and axial/equatorial cyclohexyl hydrogens. η2 The structure of [(µ2-dcpm)Ag2(µ3-O2CH)( -NO3)]2 displays the Gas-phase reactivity of silver formate clusters typical µ2-bridging of the short bite-angle ligand dcpm. Two “ ” + + structures in the CSD containing the (µ2-dcpm)Ag2 structural The observation of both [LAg2(O2CH)] and [L2Ag4(O2CH)3] motif exhibit short Ag–Ag contacts ranging from clusters in the ESI/MS (Fig. 1b) provided an opportunity to 2.889–2.959 Å, Ag–P bond distances are typically asymmetrical examine whether these clusters take part in catalytic cycles for and range between 2.354–2.409 Å and the P–C–P angle ranges the selective decomposition of formic acid (cf. Scheme 1a). between 113.242–116.846°.16 In comparison the Ag–Ag contacts Since the first crucial step involves decarboxylation, + [L2Ag4(O2CH)3] (m/z 1383) was mass-selected and subjected to

2 CID. A range of ionic products (Fig. 3) were observed, arising ‡The crystallographic information file for [(µ2-dcpm)Ag2(µ2-O2CH)(η -NO3)]2·

[(µ2-dcpm)2Ag4(µ2-NO3)4] has been deposited at the Cambridge Crystallographic from three main reaction pathways: (i) decarboxylation, Data Centre and assigned the code: CCDC 1505745. (ii) cluster fission and (iii) cluster fission coupled to ligand

19410 | Dalton Trans.,2016,45,19408–19415 This journal is © The Royal Society of Chemistry 2016 129 View Article Online

Dalton Transactions Paper

2 Fig. 2 ORTEP-3 representation of [(µ2-dcpm)Ag2(µ2-O2CH)(η -NO3)]2·[(µ2-dcpm)2Ag4(µ2-NO3)], 3. Displacement ellipsoids set at 20% probability level. Ag(1)–Ag(2): 3.2294(4)Å, Ag(1)–O(1): 2.485(3)Å, Ag(2)–O(2): 2.288(4)Å. Ag(3)–Ag(4): 3.3148(4)Å, Ag(5)–Ag(6): 3.1892(4)Å. Hydrogen atoms of the cyclohexyl rings have been omitted for clarity.

2 Scheme 2 A representation highlighting (a) the interaction between the two inversion-related of [(µ2-dcpm)Ag2(µ2-O2CH)(η -NO3)]2, and (b) the 2 20 similarity of [(µ2-dcpm)Ag2(µ2-O2CH)(η -NO3)]2 to [(Ph3P)2Ag2(µ2-O2CCH3)2]2. Published on 24 November 2016. Downloaded by The University of Melbourne Libraries 03/03/2017 10:05:22.

loss. Sequential decarboxylation reactions give rise to Table 1 Selected bond distances (Å) and angles (°) with esd’sin + + [L2Ag4(O2CH)2(H)] (m/z 1339, eqn (3)), [L2Ag4(O2CH)(H)2] 2 parentheses for [(µ2-dcpm)Ag2(µ2-O2CH)(η -NO3)]2 + (m/z 1295, eqn (4)) and [L2Ag4(H)3] (m/z 1251, eqn (5)). The + Ag(1)–Ag(2) 3.2294(4) Ag(1)–O(1)–Ag(1a) 102.89(13) main fragmentation channel for the CID of [L2Ag4(O2CH)3] Ag(1)–O(1) 2.161(3) O(1)–Ag(1)–O(1a) 77.11(13) (m/z 1383) is due to cluster fission via the loss of neutral – a – – Ag(1) O(1 ) 2.485(3) O(1) Ag(1) P(1) 162.80(9) Ag (O CH) 21 to give [L Ag (O CH)]+ (m/z 1077 eqn (6)). The – a – – 2 2 2 2 2 2 Ag(1) O(1 ) 2.485(3) O(2) Ag(2) P(2) 139.22(10) + Ag(1)–P(1) 2.3364(9) O(1)–C(26)–O(2) 128.8(5) next most abundant cluster is the hydride, [L2Ag2(H)] (m/z Ag(2)–O(2) 2.288(4) P(2)–C(13)–P(1) 112.2(2) 1033), which likely arises from combined cluster fission and – – – Ag(2) O(3) 2.423(3) O(3) N(1) O(4) 117.8(3) decarboxylation (eqn (7)). Ag(2)–O(4) 2.536(6) O(3)–N(1)–O(5) 121.1(3) – – – Ag(2) P(2) 2.3591(10) O(4) N(1) O(5) 121.0(3) ½ ð Þ þ !½ ð Þ ð Þþ þ ð Þ N(1)–O(3) 1.257(5) P(2)–Ag(2)–O(3) 133.31(10) L2Ag4 O2CH 3 L2Ag4 O2CH 2 H CO2 3 N(1)–O(4) 1.256(5) P(2)–Ag(2)–O(4) 131.15(8) N(1)–O(5) 1.233(5) Ag(1)–Ag(2)–O(3) 153.45(9) ½ ð Þ þ !½ ð Þð Þ þ þ ð Þ L2Ag4 O2CH 3 L2Ag4 O2CH H 2 2CO2 4 C(26)–O(1) 1.269(6) Ag(1)–Ag(2)–O(4) 115.42(7) – – – C(26) O(2) 1.212(7) N(1) O(3) Ag(2) 98.1(2) ½ ð Þ þ !½ ð Þ þ þ ð Þ C(13)–P(1) 1.847(4) N(1)–O(4)–Ag(2) 92.6(2) L2Ag4 O2CH 3 L2Ag4 H 3 3CO2 5 C(13)–P(2) 1.845(4) O(1a)–Ag(1)–Ag(2) 135.74(8) Ag(1)–O(1a) 2.485(3) (O1a)–Ag(1)–O(1) 77.11(13) þ þ ½L2Ag4ðO2CHÞ !½L2Ag2ðO2CHÞ þ Ag2ðO2CHÞ ð6Þ Ag(1)–P(1) 2.3364(9) (O1a)–Ag(1)–P(1) 120.07(8) 3 2 a ½ ð Þ þ !½ ð Þþ þ ð Þ þ ð Þ Equivalent atom from adjacent unit cell. L2Ag4 O2CH 3 L2Ag2 H Ag2 O2CH 2 CO2 7

This journal is © The Royal Society of Chemistry 2016 Dalton Trans.,2016,45, 19408–19415 | 19411 130 View Article Online

Paper Dalton Transactions

þ þ Table 2 Selected bond distances (Å) and angles (°) with esd’sin ½ ð Þ !½ ð Þ þ ð Þ ð Þ L2Ag4 O2CH 3 LAg2 O2CH LAg2 O2CH 2 8 parentheses for [(µ2-dcpm)2Ag4(µ2-NO3)4] ½ ð Þ þ !½ ð Þþ þ ð Þ þ ð Þ Ag(3)–Ag(4) 3.3148(4) Ag(3)–Ag(4)–O(15) 89.64(10) L2Ag4 O2CH 3 LAg2 H LAg2 O2CH 2 CO2 9 Ag(5)–Ag(6) 3.1892(4) Ag(3)–O(9)–Ag(5) 138.21(15) Ag(3)–O(6) 2.217(3) Ag(4)–O(15)–Ag(5) 118.71(19) – – – ½ ð Þ þ !½ þ þ ð Þ ð Þ Ag(3) O(9) 2.462(3) Ag(4) Ag(3) O(9) 84.17(2) L2Ag4 O2CH 3 LAg Ag3 O2CH 3 10 Ag(3)–P(3) 2.3473(10) O(9)–Ag(5)–O(15) 89.64(10) Ag(4)–O(15) 2.333(4) Ag(3)–Ag(4)–O(12) 79.24(9) Ag(4)–O(12) 2.367(3) Ag(3)–O(6)–Ag(6) 127.42(15) Mass selection followed by CID on [L Ag (O CH) (H)]+ – – – 2 4 2 2 Ag(4) P(4) 2.3380(9) Ag(4) O(12) Ag(6) 141.16(16) (m/z 1339, Fig. 3b) or [L Ag (O CH)(H) ]+ (m/z 1295, Ag(5)–O(15) 2.379(4) Ag(4)–Ag(3)–O(6) 94.03(10) 2 4 2 2 Ag(5)–O(9) 2.236(3) O(12)–Ag(6)–O(6) 81.02(13) Fig. 3c) only gives rise to decarboxylation reactions Ag(5)–P(5) 2.3368(9) Ag(5)–Ag(6)–O(6) 95.24(8) (eqn (11)–(13)), with no cluster fission reactions being Ag(6)–O(6) 2.362(3) Ag(5)–O(9)–Ag(3) 138.21(15) + – – – observed. Finally, the binuclear clusters [L2Ag2(O2CH)] Ag(6) O(12) 2.324(4) Ag(6) O(6) Ag(3) 127.42(15) + Ag(6)–P(6) 2.3255(9) Ag(6)–Ag(5)–O(9) 77.29(13) (Fig. 3d) and [LAg2(O2CH)] (Fig. 4a) also mainly undergo N(2)–O(6) 1.277(6) O(6)–Ag(3)–O(9) 75.35(12) decarboxylation (eqn (14), eqn (1) and step 1 of Scheme 1). – – – N(2) O(7) 1.281(7) Ag(5) Ag(6) O(12) 71.71(11) In the case of [L Ag(O CH)]+ the branching ratio (BR) for N(2)–O(8) 1.204(6) Ag(5)–O(15)–Ag(4) 118.71(19) 2 2 N(3)–O(9) 1.280(5) Ag(4)–O(12)–Ag(6) 141.16(16) decarboxylation is 95%, while the other minor channels do N(3)–O(10) 1.216(6) Ag(6)–Ag(5)–O(15) 104.16(11) not involve fragmentation of the coordinated formate, but – – – N(3) O(11) 1.232(5) O(12) Ag(4) O(15) 82.26(15) rather involve silver formate loss to give [LAg]+ (m/z 517, N(4)–O(12) 1.275(5) P(3)–C(39)–P(4) 116.3(2) N(4)–O(13) 1.231(6) P(5)–C(64)–P(6) 113.6(2) BR = 1.8%) or loss of the protonated dppm ligand (m/z N(4)–O(14) 1.218(6) Ag(3)–P(3)–C(39) 116.33(12) 409, BR = 3.2%). N(5)–O(15) 1.257(6) Ag(4)–P(4)–C(39) 119.04(12) – – – þ þ N(5) O(16) 1.217(6) Ag(5) P(5) C(64) 113.59 ½L Ag ðO CHÞ ðHÞ !½L Ag ðO CHÞðHÞ þ CO ð11Þ N(5)–O(17) 1.235(7) Ag(6)–P(6)–C(64) 115.36(13) 2 4 2 2 2 4 2 2 2 ½ ð Þ ð Þþ !½ ð Þ þ þ ð Þ L2Ag4 O2CH 2 H L2Ag4 H 3 2CO2 12 ½ ð Þð Þ þ !½ ð Þ þ þ ð Þ L2Ag4 O2CH H 2 L2Ag4 H 3 CO2 13 þ þ ½L2Ag2ðO2CHÞ !½L2Ag2ðHÞ þ CO2 ð14Þ

Ion-molecule reactions of silver hydrides with formic acid Since the next step of the catalytic cycle involves protonation of the coordinated hydride to liberate hydrogen and reform the formate complex (step 2 of Scheme 1a and eqn (2)), we next isolated each of the hydride complexes formed via CID and exposed them to formic acid inside the ion-trap. Published on 24 November 2016. Downloaded by The University of Melbourne Libraries 03/03/2017 10:05:22. + + + [L2Ag4(O2CH)2(H)] ,[L2Ag4(O2CH)(H)2] ,[L2Ag4(H)3] and + [L2Ag2(H)] were all unreactive towards formic acid in the gas- phase, even at the longest reaction times studied (10 000 ms, data not shown). In contrast, the only reaction between + [dcpmAg2(H)] and formic acid is the selective protonation of the coordinated hydride to reform the formate, + [dcpmAg2(O2CH)] (Fig. 4b, and step and eqn (2) of Scheme 1a), thereby closing the catalytic cycle (Scheme 1a). Scheme 3 A representation highlighting the four fused eight- Under the pseudo-first order kinetic conditions used, a rate membered rings of [(µ2-dcpm)2Ag4(µ2-NO3)4]. − constant of 2.6 ± 0.03 × 10 10 cm3 per molecules3 per s1 was determined, which when compared to the ADO theory rate − coefficient(2.2×109 cm3 per molecules3 per s1)yieldsa reaction efficiency of 12.0 ± 0.2%. This is approximately Ligand loss coupled to cluster fission drives the fragmenta- an order of magnitude faster than that determined + + tion of [L2Ag4(O2CH)3] to [LAg2(O2CH)] (m/z 669) via the previously for the bis(diphenylphosphino)methane analogue, + 1 neutral loss of LAg2(O2CH)2 (eqn (8)). The related [dppmAg2(H)] . The enhanced reactivity could be due to [dcpmAg2(O2CF3)2] cluster with trifluroacetato ligands in place changes in the steric and electronic contributions from the of the formato ligands has been observed.16b The dinuclear cyclohexyl rings of dcpm compared to the phenyl rings of + 22 silver hydride [LAg2(H)] (m/z 625) is coupled to the loss of dppm and differences in the bite angle of the phosphine 23 neutral LAg2(O2CH)2 and CO2 (eqn (9)). ligands.

19412 | Dalton Trans.,2016,45, 19408–19415 This journal is © The Royal Society of Chemistry 2016 131 View Article Online

Dalton Transactions Paper

n + Fig. 3 LTQ MS experiments obtained for the collision induced dissociation of silver formato clusters: (a) [L2Ag4(O2CH)3] (m/z 1383), + + + (b) [L2Ag4(O2CH)2(H)] (m/z 1339), (c) [L2Ag4(O2CH)(H)2] (m/z 1293), and (d) [L2Ag2(O2CH)] (m/z 1078). The normalised collision energy used for each experiments was 15%. L = dcpm = bis(dicyclohexylphosphino)methane. The m/z values shown are of the most intense peak arising from isotopic abundances of the atoms.

Conclusions

Published on 24 November 2016. Downloaded by The University of Melbourne Libraries 03/03/2017 10:05:22. This study further highlights the labile nature of coinage metal carboxylates.7b,8 Thus while ESI/MS of an acetonitrile

solution containing a mixture of AgNO3 : bis(dicyclohexyl-

phosphino)methane (dcpm, L) : NaO2CH in a molar ratio of + 2 : 1 : 1 gave an abundant peak due to [LAg2(O2CH)] , crystalli- zation of this solution produced a rare example of a co-crystal

consisting of two discrete clusters [(µ2-dcpm)Ag2(µ2-O2CH) η2 ( -NO3)]2 and [(µ2-dcpm)2Ag4(µ2-NO3)4]. Nonetheless, the 2 dimer, [(µ2-dcpm)Ag2(µ2-O2CH)(η -NO3)]2, within the crystal bears resemblance to the previously described gas-phase struc- + ture calculated for [dppmAg2(O2CH)] (dppm = bis(diphenylphosphino)methane) (Scheme 1a). + The observation of the tetranuclear complex [L2Ag4(O2CH)3] . via ESI/MS provided an opportunity to see if it can also partici- Fig. 4 Mass spectra from LTQ MSn experiments supporting a catalytic pate in a related catalytic cycle for the decarboxylation of cycle for the selective decarboxylation of formic acid: (a) CID of formic acid in the gas phase (cf. Scheme 1a). While it + + [LAg2(O2CH)] ; (b) ion-molecule reaction of formic acid with [LAg2(H)] , 9 3 does indeed undergo sequential decarboxylation reactions [HO2CH]ion trap = 4.0 × 10 molecules per cm . Ion-activation times = + 700 ms. L = dcpm = bis(dicyclohexylphosphino)methane. The m/z to provide access to the hydrides [L2Ag4(O2CH)2(H)] , + + – values shown are of the most intense peak arising from isotopic abund- [L2Ag4(O2CH)(H)2] and [L2Ag4(H)3] (eqn (3) (5)), significant ances of the atoms. *Represents the mass selected precursor ion. amounts of cluster fission products are also observed

This journal is © The Royal Society of Chemistry 2016 Dalton Trans.,2016,45, 19408–19415 | 19413 132 View Article Online

Paper Dalton Transactions

(eqn (6)–(10)). More importantly, none of these tetranuclear 4510; (b) S. P. Neo, Z.-Y. Zhou, T. C. W. Mak and hydrides are reactive towards formic acid and cannot thus T. S. A. Hor, Inorg. Chem., 1995, 34, 520; (c)E.Szłyk, participate in the crucial hydride protonation step to liberate I. Szymańska, A. Surdykowski, T. Głowiak, A. Wojtczak and hydrogen and reform the coordinate formate (cf. eqn (2) and A. Goliński, Dalton Trans., 2003, 3404. step 2 of Scheme 1a). 8(a) P. D. Harvey, M. Drouin and T. Zhang, Inorg. Chem., + In contrast, [dcpmAg2(O2CH)] also participates in a gas- 1997, 36, 4998; (b) K. Jiang, D. Zhao, L.-B. Guo, C.-J. Zhang phase catalytic cycle for the selective decomposition of formic and R.-N. Yang, Chin. J. Chem., 2004, 22, 1297. in the gas-phase with essentially complete conversion of reac- 9(a) G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. tant ions to product ions and excellent selectivity for both Crystallogr., 2008, 64, 112; (b) G. M. Sheldrick, Acta steps, highlighting that the nature of the R group on the Crystallogr., Sect. A: Fundam. Crystallogr., 2015, 71,3. bisphosphine ligand is less important than the nuclearity of 10 L. J. Farrugia, J. Appl. Crystallogr., 1997, 30, 565. the cluster. The labile nature of coinage metal carboxylates 11 L. J. Farrugia, J. Appl. Crystallogr., 1999, 32, 837. suggests that there are likely to be complex equilibria and 12 (a) W. A. Donald, C. J. McKenzie and R. A. J. O’Hair, Angew. potentially multiple silver complexes in solution. While Chem., Int. Ed., 2011, 50, 8379; (b) A. K. Y. Lam, C. Li, caution demands that the gas-phase results not be over-inter- G. N. Khairallah, B. B. Kirk, S. J. Blanksby, A. J. Trevitt, preted in terms of the condensed-phase milieu, it is nonethe- U. Wille, R. A. J. O’Hair and G. da Silva, Phys. Chem. Chem. less interesting to speculate that binuclear rather than tetra- Phys., 2012, 14, 2417; (c) W. A. Donald, G. N. Khairallah nuclear intermediates might be associated with the obser- and R. A. J. O’Hair, J. Am. Soc. Mass Spectrom., 2013, 24,

vation of the evolution of CO2 and H2 at 65 °C from a mixture 811.

of AgBF4, dppm, sodium formate and formic acid under vari- 13 (a) T. Su and M. T. Bowers, Int. J. Mass Spectrom. Ion Phys., able temperature NMR conditions.4 Given that a host of silver 1973, 12, 347; (b) K. F. Lim, QCPE Bull., 1994, 14,3. hydrides of diﬀerent nuclearity can be prepared via ESI-MS 14 (a) A. Zavras, G. N. Khairallah, T. U. Connell, J. M. White, and CID,24 we are currently further examining how the nucle- A. J. Edwards, P. S. Donnelly and R. A. J. O’Hair, Angew. arity of ligated silver hydride complexes influences the crucial Chem., Int. Ed., 2013, 52, 8391; (b) A. Zavras, protonation step by formic acid to liberate hydrogen. G. N. Khairallah, T. U. Connell, J. M. White, A. J. Edwards, R. J. Mulder, P. S. Donnelly and R. A. J. O’Hair, Inorg. Chem., 2014, 53, 7429; (c) A. Zavras, A. Ariafard, Acknowledgements G. N. Khairallah, J. M. White, R. J. Mulder, A. J. Canty and R. A. J. O’Hair, Nanoscale, 2015, 7, 18129. RAJO thanks the Australian Research Council for financial 15 (a) A. D. Bond, CrystEngComm, 2007, 9, 833; support DP150101388. AZ acknowledges the award of an APA (b) M. Kriechbaum, D. Otte, M. List and U. Monkowius, scholarship. Z. Naturforsch., B: Chem. Sci., 2014, 69, 1188; (c) M. Tabatabaeea, B.-M. Kukovecb and M. Kazeroonizadeha, Polyhedron, 2011, 30, 1114. Notes and references 16 (a) Y.-Y. Lin, S.-W. Lai, C.-M. Che, W.-F. Fu, Z.-Y. Zhou and N. Zhu, Inorg. Chem., 2005, 44, 1511; (b) C.-M. Che, Published on 24 November 2016. Downloaded by The University of Melbourne Libraries 03/03/2017 10:05:22. 1(a) M. Grasemann and G. Laurenczy, Energy Environ. Sci., M.-C. Tse, M. C. W. Chan, K.-K. Cheung, D. L. Phillips and 2012, 5, 8171; (b) S. Enthaler, J. von Langermann and K.-H. Leung, J. Am. Chem. Soc., 2000, 122, 2464. T. Schmidt, Energy Environ. Sci., 2010, 3, 1207. 17 A. Bondi, J. Phys. Chem., 1964, 68, 441. 2(a) K. Saito, T. Shiose, O. Takahashi, Y. Hidaka, F. Aiba and 18 (a) T. G. Gray and J. P. Sadighi, in Molecular Metal–Metal K. Tabayashi, J. Phys. Chem. A, 2005, 109, 5352; Bonds, ed. S. T. Liddle, Wiley-VCH Verlag GmbH & Co. (b) J.-G. Chang, H.-T. Chen, S. Xu and M. C. Lin, J. Phys. KGaA, Weinheim, Germany, 2015, ch. 11, pp. 397–428; Chem. A, 2007, 111, 6789. (b) H. Schmidbaur, Gold Bull., 2000, 33,3; 3 B. Loges, A. Boddien, F. Gärtner, H. Junge and M. Beller, (c) H. Schmidbaur and A. Schier, Chem. Soc. Rev., 2008, 37, Top. Catal., 2010, 53, 902. 1931; (d) H. Schmidbaur and A. Schier, Angew. Chem., Int. 4 A. Zavras, G. N. Khairallah, M. Krstić, M. Girod, S. Daly, Ed., 2015, 54, 746. R. Antoine, P. Maitre, R. J. Mulder, S.-A. Alexander, 19 (a) P.-S. Cheng, S. C. K. Hau and T. C. W. Mak, Inorg. Chim. V. Bonačić-Koutecký, P. Dugourd and R. A. J. O’Hair, Nat. Acta, 2013, 403, 110; (b) S.-Q. Zang, L. Zhao and Commun., 2016, 7, 11746. T. C. W. Mak, Organometallics, 2008, 27, 2396; (c) Z. Chen, 5 Z. Lan-Sun, Y. Hua-Hui and Z. Qian-Er, Jiegou Huaxue, L. Zhang, F. Liu, R. Wang and D. Sun, CrystEngComm, 2013, 1992, 10, 97. 15, 8877; (d) B. Li, R.-W. Huang, J.-H. Qin, S.-Q. Zang, 6 G. A. Bowmaker, Eﬀendy, J. V. Hanna, P. C. Healy, G.-G. Gao, H.-W. Hou and T. C. W. Mak, Chem. – Eur. J., J. C. Reid, C. E. F. Rickard and A. H. White, J. Chem. Soc., 2014, 20, 12416; (e) S. M. J. Wang and T. C. W. Mak, Dalton Trans., 2000, 753. Polyhedron, 2009, 28, 2684; (f) S. C. K. Hau and 7(a) T. S. A. Hor, S. P. Neo, C. S. Tan, T. C. W. Mak, T. C. W. Mak, Polyhedron, 2013, 64, 63; (g)T.Huand K. W. P. Leung and R. J. Wang, Inorg. Chem., 1992, 31, T. C. W. Mak, Organometallics, 2013, 32, 202.

19414 | Dalton Trans.,2016,45, 19408–19415 This journal is © The Royal Society of Chemistry 2016 133 View Article Online

Dalton Transactions Paper

20 E. T. Blues, M. G. B. Drew and B. Femi-Onadeko, Acta 23 Z. Freixa and P. W. N. M. van Leeuwen, Dalton Trans., 2003, Crystallogr., Sect. B: Struct. Crystallogr. Cryst. Chem., 1977, 1890. 33, 3965. 24 M. Krstić, A. Zavras, G. N. Khairallah, P. Dugourd, 21 N. Y. Kozitsyna, S. E. Nefedov, A. P. Klyagina, A. A. Markov, V. Bonačić-Koutecký and R. A. J. O’Hair, ESI/MS Z. V. Dobrokhotova, Y. A. Velikodny, D. I. Kochubey, Investigation of Routes to the Formation of Silver Hydride 2+ T. S. Zyubina, A. E. Gekhman, M. N. Vargaftik and Nanocluster Dications [AgxHx−2Ly] and Gas-phase 2+ I. I. Moiseev, Inorg. Chim. Acta, 2011, 370, 382. Unimolecular Chemistry of [Ag10H8L6] , Int. J. Mass 22 C. A. Tolman, Chem. Rev., 1977, 77, 313. Spectrom., DOI: 10.1016/j.ijms.2016.05.022, in press. Published on 24 November 2016. Downloaded by The University of Melbourne Libraries 03/03/2017 10:05:22.

This journal is © The Royal Society of Chemistry 2016 Dalton Trans.,2016,45, 19408–19415 | 19415 134 6 Gas-Phase Ion−Molecule Reactions of Copper Hydride − − Anions [CuH2] and [Cu2H3]

135

136 Article

pubs.acs.org/IC

Gas-Phase Ion−Molecule Reactions of Copper Hydride Anions − − [CuH2] and [Cu2H3] Athanasios Zavras,† Hossein Ghari,§ Alireza Ariafard,*,‡,§ Allan J. Canty,‡ and Richard A. J. O’Hair*,†

† School of Chemistry and Bio21 Molecular Science and Biotechnology Institute, University of Melbourne, 30 Flemington Road, Parkville, Victoria 3010, Australia ‡ The School of Physical Sciences, University of Tasmania, Private Bag 75, Hobart, Tasmania 7001, Australia § Department of Chemistry, Faculty of Science, Central Tehran Branch, Islamic Azad University, Shahrak Gharb, Tehran, Iran

*S Supporting Information

ABSTRACT: Gas-phase reactivity of the copper hydride anions − − [CuH2] and [Cu2H3] toward a range of neutral reagents has been examined via multistage mass spectrometry experiments in a linear ion trap mass spectrometer in conjunction with isotope labeling studies and Density Functional Theory (DFT) calculations. − − [CuH2] is more reactive than [Cu2H3] , consistent with DFT calculations, which show ithasahigherenergyHOMO. Experimentally, [CuH ]− was found to react with CS via hydride 2 − 2 transfer to give thioformate (HCS2 ) in competition with the formation of the organometallic [CuCS ]− ion via liberation of 2 − hydrogen; CO2 via insertion to produce [HCuO2CH] ; methyl iodide and allyl iodide to give I− and [CuHI]−; and 2,2,2- trifluoroethanol and 1-butanethiol via protonation to give hydrogen − − and the product anions [CuH(OCH2CF3)] and [CuH(SBu)] . In contrast, the weaker acid methanol was found to be ff π* unreactive. DFT calculations reveal that the di erences in reactivity between CS2 and CO2 are due to the lower lying orbital of the former, which allows it to accept electron density from the Cu center to form the initial three-membered ring complex η2 − intermediate, [H2Cu( -CS2)] . In contrast, CO2 undergoes the barrierless side-on hydride transfer promoted by the high electronegativity of the oxygen atoms. Side-on S 2 mechanisms for reactions of [CuH ]− with methyl iodide and allyl iodide are N 2 − favored on the basis of DFT calculations. Finally, the DFT calculated barriers for protonation of [CuH2] by methanol, 2,2,2- trifluoroethanol, and 1-butanethiol correlate with their gas-phase acidities, suggesting that reactivity is mainly controlled by the acidity of the substrate.

■ INTRODUCTION Given that solvent can influence reactivity of copper hydride 16 1 complexes, and the relationship between cluster size, n, and Copper(I) hydride, first isolated by Wurtz in 1844, is a reactivity of the cuprate complexes, [MCuH ] , has never been pyrophoric red solid that adopts a Wurtzite crystal structure. In 2 n examined previously, we have used the multistage mass the quest to “tame” CuH by using ligands to create soluble 2 spectrometry capabilities of a linear ion trap mass spectrom- copper hydrides as selective reagents in organic synthesis, two eter17 to examine the fundamental gas-phase reactivity of main approaches have been adopted. The first involves the use 3 4 neutral substrates with the mass selected hydrido cuprate of Lewis bases, B, such as pyridine or phosphines to generate −12,13 − complexes, [CuH2] and [Cu2H3] . While a range of 5 18 19 complexes of the type [BCuH]n. This approach has led to the transition metal hydride cations and anions have been ’ 6 successful generation of the widely used Stryker s reagent, and formed in the gas phase, typically as the products of X−H(X= 7 more recently Lipshutz’s “CuH in a bottle”. The second H, C, etc.) bond activation reactions, fewer studies have used approach has involved the generation of hydrido cuprate ion−molecule reactions to examine the reactivity of transition 8,9 20 complexes of the types [MCuH2]n (where M = Li or K) and metal hydride ions with organic substrates. Thus, we chose 10 ff [LiRCuH]n. Related hydrido cuprates have been prepared in neutral substrates representative of di erent classes of the absence of alkali metal cations as part of a series of matrix reactivity: CS2 and CO2 for their ability to undergo insertion isolation studies aimed at examining the formation of transition reactions into M−H bonds; methyl iodide and allyl iodide to metal hydrides.11 In particular, Andrews and co-workers have investigate their reduction to hydrocarbons; and the acids characterized a series of copper hydrides including CF3CH2OH and BuSH to investigate protonation of the −12,13 − 14,15 [CuH2] and [Cu2H] . The former anion was suggested as a potential hydrogen carrier for hydrogen storage Received: September 5, 2016 applications.12 Published: February 10, 2017

137 Inorganic Chemistry Article coordinated hydrides. Where possible, comparisons are made dyotropic rearrangements,24 we found that the copper hydride − − with published reactivity of the same substrates toward the bare anions [CuH2] and [Cu2H3] can be prepared by (i) mass 21 − 22 − hydride anion and dimethyl cuprate, [CuMe2] . Finally, selecting the copper formate anions [Cu(O2CH)2] (Figure DFT calculations were used to examine potential mechanisms S2a and S2b) and [Cu (O CH) ]− (Figure S3a−c), formed − 2 2 3 to account for the observed reactivity of [CuH2] . under gentle electrospray ionization conditions (Figure S1) from an acetonitrile solution containing cuprous oxide and ■ RESULTS AND DISCUSSION formic acid or its deuterium labeled isotopologue, DCO2D − − Gas-Phase Formation of [CuH2] and [Cu2H3] . Electro- (Scheme 1a), and subjecting these ions to multiple stages of spray ionization mass spectrometry (ESI−MS) together with collision-induced dissociation (CID) where sequential decar- various combinations of multistage mass spectrometry (MSn) boxylation reactions of the coordinated formates lead to the experiments were used to prepare hydrido cuprate complexes formation of coordinated hydrides; and (ii) sequential and study their reactivity toward a range of substrates (Scheme fragmentation reactions of precursor copper formate anions under “in source” CID.25 1). While copper hydride anions have been previously prepared − − [CuH2] , 1, and [Cu2H3] , 2, prepared via the former Scheme 1. Multistage Mass Spectrometry (MSn) method, were individually mass selected and allowed to Experiments for the Gas-Phase Preparation of Hydrido undergo ion−molecule reactions (IMRs) with a volatile neutral 4 5 Cuprate Complexes via Collision-Induced Dissociation substrate in a MS or MS experiment, respectively (Scheme 1b a (CID) and c). Of particular interest was to determine which of these cuprates more readily undergoes hydride transfer toward neutral substrates, which is related to their relative “hydricity”.26,27 − − − Ion Molecule Reactions of [CuH2] and [Cu2H3] . The gas-phase ion−molecule reactions of the cuprate anions − − [CuH2] (Figure S4 and Table S1) and [Cu2H3] (Figure S5 and Table S2) were explored with a range of neutral substrates including CS2 and CO2, methyl iodide and allyl iodide, and the acids CF CH OH and CH (CH ) SH. While [CuH ]− was 3 2 3 2 3 − 2 found to react with all of these substrates, [Cu2H3] was found to be generally less reactive (Table 1). Because the nature of the products formed is substrate speciﬁc, in the next sections we discuss the results of experiments by individual class of substrates. − − a − − − Reactions of [CuH ] and [Cu H ] with CS . Two ionic (a) [Cux(O2CH)x+1] (x = 1,2), (b) [CuH2] , 1, and (c) [Cu2H3] , 2 2 3 2 products are observed for the ion−molecule reaction of 2, and their reaction(s) with various neutral substrates to yield anionic − products via ion−molecule reactions (IMRs). [63CuH ] (m/z 65) with CS (Figure 1a). The major reaction 2 2 − pathway proceeds via hydride transfer to produce [HCS2] (m/ z 77) and 63CuH (eq 1). The minor reaction pathway results in − via sequential decarboxylation reactions of aliphatic copper the formation of the organometallic ion [CuCS2] (m/z 139) β 23 carboxylates followed by -hydride fragmentation reactions or via the reductive elimination of H2 (eq 2), a reaction related to

− 63 − m z 63 − m z Table 1. Gas-Phase Kinetics Associated with the Ion Molecule Reactions of [ CuH2] ( / 65) and [ Cu2H3] ( / 129) with Various Neutral Reagents

a b,c,d φe ion neutral reagent reaction channel; branching ratio out of 100%; eq no. kexpt 63 − − − ± × −10 ± [ CuH2] CS2 [HCS2] + CuH; 91%; eq 1; [CuCS2] +H2 ; 9%; eq 2 (2.2 0.4) 10 19 4 − ± × −13 ± × −2 CO2 [CuH(O2CH)] ; 100%; eq 15 (2.8 0.4) 10 (4 0.5) 10 − ± × −10 ± CH3II+ CuH + CH4; 81%; eq 16 (4.7 0.3) 10 34 2 − [CuH(I)] +CH4; 19%; eq 17 − ± × −9 ± C3H5II+ CuH + C3H6; 53%; eq 22 (1.1 0.1) 10 68 7 − [CuH(I)] +C3H6; 47%; eq 23 − ± × −10 ± CF3CH2OH [CuH(OCH2CF3)] +H2; 100; eq 32 (7.1 0.6) 10 45 20 − ± × −9 ± 40 BuSH [CuH(SBu)] +H2; 100%; eq 36 (9.3 1.2) 10 580 76 63 − •− • ± × −11 ± × −1 [ Cu2H3] CS2 [Cu2S2] +CH3 ; 100%; eq 12 (1.7 0.2) 10 (1.7 0.1) 10 − ± × −13 ± × −2 CO2 [Cu2H2(O2CH)] ; 100%; eq 15 (1.3 0.1) 10 (2 0.1) 10 f f f CH3I N.R. N.R. N.R. − ± × −10 ± C3H5I [Cu2H2(I)] +C3H6; 100%; eq 31 (1.3 0.1) 10 10 1 − ± × −12 × −2 CF3CH2OH [Cu2H2(OCH2CF3)] +H2; 100%; eq 35 (5.4 1.0) 10 7 10 − ± × −11 ± BuSH [Cu2H2(SBu)] +H2; 100%; eq 39 (1.4 0.2) 10 1 0.1 aEquation no. is the full equation listed in the text. bMean ± standard deviation (n = 3). cIn units of cm3 molecules−1 s−1. dRates measured for the − − e reaction of neutral reagents with the hydrido cuprate anions [CuH2] or [Cu2H3] were determined from triplicate experiments. Reaction ﬃ φ × − e ciency, =(kexpt/kADO) 100. The kADO is the theoretical ion molecule collision rate constant obtained from the average-dipole orientation (ADO) theory,28 which is calculated using the Colrate program.29 fN.R. = no reaction at the longest reaction time (10 000 ms) examined at a concentration of methyl iodide of 2.9 × 109 molecules cm−3.

2388 DOI: 10.1021/acs.inorgchem.6b02145 Inorg. Chem. 2017, 56, 2387−2399

138 Inorganic Chemistry Article

−− [CuMe22 ]+→ CS [MeCS 2 ] + CuMe (10) − → [MeCS2 CuMe] (11) The observed hydride transfer reaction (eq 1) is related to − the reaction of [CuMe2] with CS2, which proceeds via methyl 22e − anion transfer (eq 10). In contrast, [CuMe2] also forms an adduct with CS (eq 11), but does not form the organometallic − 2 ion, [CuCS2] (cf., eq 2). − 63 − The ion molecule reaction of [ Cu2H3] with CS2 is much slower (1.7 × 10−11 cm3 molecules−1 s−1, reaction efficiency of 0.17%, Table 1) than that of the mononuclear cuprate and proceeds via an entirely different pathway to produce a single product ion at m/z 190, formulated as the mixed valence fi 63 •− dinuclear copper sul de cluster [ Cu2S2] (Figure S6a, eq 63/65 − 12). By mass selecting the [ Cu2H3] (m/z 131) isotopologue, a mass shift of 2 Da to m/z 192 is observed Figure 1. Ion−molecule reaction of various mass selected isotopo- (Figure S6b, eq 13), thus confirming the formulation. We infer − × 9 logues of [CuH2] with CS2 ([CS2] ion trap = 7.2 10 molecules that a neutral methyl radical is also formed, and thus the −3 63 − •− cm ) at an activation time of 200 ms: (a) [ CuH2] m/z 65; (b) reaction between [Cu2S2] and CS2 is likely to be multistep [63CuHD]− m/z 66; (c) [65CuH ]− m/z 67; and (d) [63CuD ]− m/z •− 2 2 and mechanistically complex. Finally, [Cu2S2] has been 67. The most intense peak in the ion is represented by the m/z value. 32 * previously observed in the gas phase and via computational Represents the mass selected precursor ion. chemistry.33

63 −•63 −• [ Cu23 H ]+→ CS 2 [ Cu22 S ] + CH 3 (12) the oxidation of organocuprates.30 These assignments were 63/65 −•+→63/65 − +• conﬁrmed via the following isotope labeling studies: (i) [ Cu23 H ] CS 2 [ Cu22 S ] CH 3 (13) − [63CuHD] (m/z 66, Figure 1b), formed from a solution − − Reactions of [CuH2] and [Cu2H3] with CO2. In contrast containing mixtures of deuterium labeled and unlabeled formic 63 − 63 − − to the reactivity of [ CuH ] toward CS ,[ CuH ] reacts − 2 2 − − 2 acid, reacted with CS2 to give [HCS2] (m/z 77, eq 3), × 13 3 1 1 − 63 − very slowly (2.8 10 cm molecules s , reaction [DCS ] (m/z 78, eq 4), and the minor product [ CuCS ] 63 2 − 2 eﬃciency of 0.04%) with CO to only produce [ CuH- 65 − 2 (m/z 139, eq 5); (ii) [ CuH2] (m/z 67) reacted with CS2 − 65 − (O2CH)] (m/z 109, Figure 2a, eq 14). This reaction proceeds (Figure 1c) to give [HCS2] (m/z 77, eq 6) and [ CuCS2] 63 − via a direct insertion pathway resulting in carbon dioxide (m/z 141, eq 7); and (iii) [ CuD2] (m/z 67) reacts with CS2 − 63 − capture and transformation into a coordinated formate (Figure 1d) to give [DCS2] (m/z 78, eq 8) and [ CuCS2] 34,35 anion. Related insertion reactions between CO2 and copper (m/z 139, eq 9). hydrides play key roles in copper-catalyzed transformations of 36 63 −−63 CO2. Although the bare hydride anion reacts slowly with CO2 [CuH]22+→ CS [HCS] 2 + CuH (1) 63 − →+[CuCS]22 H (2)

63 −−63 [ CuHD]+→ CS22 [HCS ] + CuD (3) − 63 →+[DCS2 ] CuH (4) 63 − →+[CuCS]2 HD (5)

65 −−65 [CuH]22+→ CS [HCS] 2 + CuH (6) 65 − →+[CuCS]22 H (7)

63 −−63 [CuD]22+→ CS [DCS] 2 + CuD (8) 63 − →+[CuCS]22 D (9) − × The rate of reaction between [CuH2] and CS2 is 2.2 10−10 cm3 molecules−1 s−1, corresponding to a reaction ﬃ ﬀ Figure 2. LTQ mass spectra obtained for the ion−molecule reaction of e ciency of 19% (Table 1). The kinetic isotope e ect (KIE) × 12 63 − hydrido cuprate anions with CO2 ([CO2] ion trap = 2.2 10 for the hydride transfer reaction of [ CuHD] (m/z 66) with −3 − molecules cm ) at an activation time of 600 ms: (a) [CuH2] m/z 65; CS2, determined by dividing the integrated ion counts of − − − and (b) [Cu2H3] m/z 129. The most intense peak in the ion is [HCS2] (m/z 77) by [DCS2] (m/z 78), AH/AD, was found to * 31 represented by the m/z value. Represents the mass selected precursor be 0.98. The fact that the KIE is very close to unity suggests ion. For these experiments, the helium bath gas cylinder was replaced that the rate-determining step does not involve the breaking with a helium cylinder seeded with 1.03% carbon dioxide (see and forming of bonds to H/D. Experimental Section for further details).

2389 DOI: 10.1021/acs.inorgchem.6b02145 Inorg. Chem. 2017, 56, 2387−2399

139 Inorganic Chemistry Article to form the formate anion,21 it is noteworthy that the bare [CuMe ]−−+→++ C H I I CuMe C H (24) 63 − 235 48 formate anion is not observed in reactions of [ CuH2] − (experiments were carried out to attempt to detect it at m/z →+[MeCuI] C48 H (25) 63 − 45). Finally, the copper hydride [ CuH2] can be regenerated − 63 − →+[C35 H CuI] C26 H (26) when [ CuH(O2CH)] (m/z 109) is mass selected and subjected to collision-induced dissociation (Figure S2b). [63CuHD]− reacts with allyl iodide to give I− as well as [63CuH(I)]− and [63CuD(I)]− (Figure S10). The observation 63 − 63 − − [CuH]22+→ CO [CuH(OCH)]2 (14) of the latter two isotopologues suggests that I is formed by eqs − 27 and 28. [63Cu H ] (m/z 129) also reacts via CO capture (Figure 2 3 2 − 2b, eq 15), but at approximately one-half the rate (1.3 × 10 13 63 −−63 [ CuHD]+→++ C35 H I I CuH C35 H D (27) cm3 molecules−1 s−1, reaction efficiency of 0.02%) of 63 − − 63 [ CuH2] . →+I CuD + C36 H (28) 63 −−63 63 − [CuH]23+→ CO 2 [CuH(OCH)]22 (15) →+[ CuHI] C35 H D (29) − − 63 − Reactions of [CuH2] and [Cu2H3] with CH3I and →+[ CuDI] C36 H (30)  63 − CH2 CHCH2I. [ CuH2] reacts with methyl iodide (Figure − 63 − Although the isotope effects associated with product S7a) to yield two ionic products, I and [ CuHI] . The − observation of the latter ion suggests that the reaction of methyl channels involving formation of I could not be determined 63 − due to the fact that neutral species containing H or D are not iodide with [ CuH2] proceeds via cross-coupling to yield 63 − 63 − detected (i.e., eqs 18 versus 19 and eqs 27 versus 28), by methane and [ CuH] and I (eq 16) as well as [ CuH(I)] − − comparing the product ion abundances of [63CuH(I)] (eq 20 (eq 17). The ion molecule reaction of methyl iodide with 63 − 63 − × −10 3 −1 −1 or 29) and [ CuD(I)] (eq 21 or 30), kinetic isotope effects of [ CuH2] reacts at a rate of 4.7 10 cm molecules s . 63 − Sequential replacement of methides by hydrides in cuprates 1.51 and 0.9 are calculated for the reactions of [ CuHD] with enhances reactivity toward methyl iodide, with the following methyl iodide and allyl iodide, respectively. Changes in order − − from normal to inverse isotope effects have been reported for reactivity order being observed: [CuH ] > [MeCuH] > − − 2 + 37 22c hydride transfer from tungsten hydrides to Ph3C BF4 . [CuMe2] . 63 − [ Cu2H3] is unreactive toward methyl iodide at ion 63 −−63 activation times of up to 10 000 ms (Figure S7b). However, [ CuH23 ]+→++ CH I I CuH CH4 (16) in contrast with the trends in reactivity of methyl iodide and 63 − 63 − →+[ CuH(I)] CH4 (17) allyl iodide toward [ CuH2] , allyl iodide reacts with [63Cu H ]− (m/z 129) to yield the product ion [63Cu H I]− − − 2 3 2 2 [63CuHD] reacts with methyl iodide to give I as well as (m/z 255, Figure S8b) at a reaction rate of 1.3 × 10−10 and a − − [63CuH(I)] and [63CuD(I)] (Figure S9). The observation of reaction efficiency of 10%. The observation of the latter ion − the latter two isotopologues suggests that I is formed by both suggests the formation of propene via a cross-coupling pathway eqs 18 and 19. (eq 31). 63 −−63 [63 CuH]−−+→ CHI [63 CuH(I)] + CH [ CuHD]+→++ CH3 I I CuH CH3 D (18) 23 35 22 36 (31) − − − →+I− 63 CuD + CH (19) Acid Base Reactions of [CuH2] and [Cu2H3] with 4 Alcohols and 1-Butanethiol. We next examined whether the 63 − 63 − →+[ CuHI] CH D (20) coordinated hydrides of [ CuH2] could be protonated to 3 38 Δ liberate H2. Methanol, with a gas-phase acidity ( Hacid)of 63 − −1 39 →+[ CuDI] CH4 (21) 382 kcal mol , was found to be unreactive (data not shown). In contrast, the bare hydride anion rapidly deprotonates The ion−molecule reaction of [63CuH ]− with allyl iodide 2 − methanol, highlighting that coordination of the hydride to the was slightly faster with a reaction rate of 1.1 × 10 9 cm3 copper center decreases its basicity.21 The stronger acid −1 −1 ffi fl Δ −1 39 molecules s , corresponding to a reaction e ciency of 68%. tri uoroethanol ( Hacid = 361.7 kcal mol ) reacted with − 63 − 63 − − Two ionic products, I and [ CuHI] , are also observed [ CuH2] via an acid base reaction to form the anionic 63 − (Figure S8a), with the latter product ion suggesting formation product [ CuH(OCH2CF3)] (m/z 163, eq 32, Figure S11a). of propene via a cross-coupling pathway to produce I− (m/z × − The rate of this protonation reaction was measured to be 7.1 127) and [63CuH] (eq 22) as well as [63CuHI] (eq 23). In 10−10 cm3 molecules−1 s−1, which corresponds to a reaction contrast, not only is [CuMe ]− an order of magnitude less ffi fl 2 − e ciency of 45%. 2,2,2-Tri uoroethanol reacts with reactive toward allyl iodide, but also produces I (eq 24), [63CuHD]− to protonate both the D and the H sites to form − − 22d 63 − 63 − [MeCuI] (eq 25), as well as [C3H5CuI] (eq 26). The [ CuH(OCH2CF3)] (eq 33) and [ CuD(OCH2CF3)] (eq formation of the latter homocoupling product arises from 34), respectively (Figure S12). The KIE for this reaction can be oxidative addition to give a Cu(III) η3-allyl intermediate. The − directly determined from the ion abundances and was found to lack of such a product in the reaction of [CuH2] suggests that be 0.85. Related reverse isotope effects have been observed for such an intermediate may not be formed in the case of − 38 − the protonation of iron hydride complexes by acids. The [CuH2] . binuclear cuprate undergoes an acid base reaction with trifl uorethanol to form the anionic product [63 CuH ]−−+→++ C H I I63 CuH C H (22) − 235 36 [Cu2H3(OCH2CF3)] (m/z 227, eq 35, Figure S11b), albeit at a slower rate (5.4 × 10−12 cm3 molecules−1 s−1, reaction →+63 − [ CuH(I)] C36 H (23) efficiency of 0.07%).

2390 DOI: 10.1021/acs.inorgchem.6b02145 Inorg. Chem. 2017, 56, 2387−2399

140 Inorganic Chemistry Article

ﬁ − − Figure 3. Energy pro le obtained from DFT calculations, which shows the following: (A) The relationship between [CuH2] and [Cu2H3] and their respective isomers. Key bond lengths are given in angstroms. The HOMO for (B) 1A; and (C) 2A. The relative Gibbs and potential energies (in parentheses) obtained from the M06/BS2//M06/BS1 calculations are given in kcal mol−1.

ﬁ − Figure 4. Energy pro le obtained from DFT calculations for competing mechanisms for the reaction of [CuH2] with CS2: (a) hydride transfer − pathway (red); (b) addition/H2 elimination to form [CuCS2] (black); and (c) adduct formation (blue, not observed experimentally). The relative Gibbs and potential energies (in parentheses) obtained from the M06/BS2//M06/BS1 calculations are given in kcal mol−1.

2391 DOI: 10.1021/acs.inorgchem.6b02145 Inorg. Chem. 2017, 56, 2387−2399

141 Inorganic Chemistry Article

63 −−DFT Calculated Mechanisms for the Reactions of [ CuH]232+→ CFCHOH [CuH(OCHCF)] 232 + H − (32) [CuH2] with CS2 and CO2. DFT calculations were carried 63 − out for the reactions on [ CuH2] (Figure 4) with CS2 to establish the likely mechanisms associated with the hydride transfer reaction (eq 1) and the reaction giving rise to the 63 − organometallic ion, [ CuCS2] (eq 2). The lowest energy 63 − pathway for initial attack by [ CuH2] , 1A,onCS2 involves 63 −− [ CuH]33+→ CFCHOH 3 2 [CuH(OCHCF)] 2 2 3 + H 2 coordination of the carbon atom at the copper center via η2 − (35) TS1A−2 to form the three-membered ring, [H2Cu( -CS2)] , 2. 63 − Δ Related three-membered ring structures are formed in the [ CuH2] reacts with 1-butanethiol ( Hacid = 353.7 kcal 47 48 −1 39 40 63 − reactions of Cu atoms and metal complexes with CS2. Side- mol ) at the collision rate to give [ CuH(S(CH2)3CH3)] 41 on transfer of a hydride via TS1A−5 is not competitive, (m/z 153) and H2 (Figure S13a, eq 36). The reaction of 63 − 63 consistent with previous calculations on the related side-on [ CuHD] with 1-butanethiol (Figure S14) produced [ CuH- − − − 63 − transfer of CH3 from [CuMe2] to CS2, which was found to (S(CH2)3CH3)] (eq 37) and [ CuD(S(CH2)3CH3)] (eq 22e η2 be a high energy process. The initially formed [(H)2Cu( - 38, Figure S14), giving a KIE of 0.86, which is comparable to − CS2)] , 2, can either fragment via H2 loss or via hydride that of 2,2,2-trifluoroethanol. The binuclear cuprate undergoes − a slower (1.4 × 10−11 cm3 molecules−1 s−1, reaction efficiency of transfer to form [HCS2] . The pathway for H2 loss from 2 results in the formation of a σ bond between the hydrides, 1%) acid−base reaction with 1-butanethiol to form the anionic − − − η2 TS2−3, to give the ion molecule complex [H2Cu( -CS2)] , 3, product [Cu2H3(S(CH2)3CH3)] (m/z 217, eq 39, Figure which is then able to lose a molecule of H2, TS3−4, to give the S13b). − observed organometallic ion [CuCS2] (m/z 139), 4. The 63 − hydride transfer pathway from 2 occurs via electrophilic attack [CuH]2323+ CH(CH)SH of the carbon by a hydride, TS2−5, to yield hydridothioformato →+63 − [ CuH(S(CH23 ) CH 3 )] H 2 (36) cuprate, 5, which can isomerize via TS5−6 to 6. Even though H2 − loss is predicted to be thermodynamically favored, [HCS2] formation is kinetically favored due to both TS2−5 and TS5−6 being lower in energy than TS2−3 and TS3−4. Although the − formation of CuH and [HCS2] is predicted to be slightly 63 − endothermic, at the higher CCSDT level of theory, formation [CuH]23+ CH(CH)SH 3 23 of these products is predicted to be exothermic by 4.5 kcal − 63 − mol 1 (Figure S15). Finally, the fact that the isotope effect for →+[ Cu22 H (S(CH 23 ) CH 3 )] H 2 (39) − the formation of [HCS2] is close to unity is consistent with − − DFT Calculated Structures of [CuH2] and [Cu2H3] TS1A−2 being the rate-determining step. and Their HOMOs. The formation of [Cu H ]− via ESI−MS DFT calculations were carried out (Figure 5) to establish the 2 3 − suggests that it is stabilized with respect to its dissociation to mechanism of the CO2 insertion reaction into [CuH2] (eq − form [CuH ] and CuH (eq 40), as indicated in Figure 3A. 14). The electrophilic nature of the carbon atom of CO2 allows 2 − − − The most stable isomer of [Cu2H3] , [(HCu)2H] 2A, [CuH2] to transfer a hydride to the carbon atom as it contains a three-centered (Cu−H−Cu) two-electron bond. approaches. This transforms carbon dioxide to a formate anion, − − The dissociation of [Cu2H3] 2A, to yield CuH and [CuH2] which is coordinated via an oxygen atom in the complex 1A, is endothermic and requires 57.8 kcal mol−1. A related + ligated copper hydride cluster, [(LCu)2H] , has been isolated by Sadighi’s group,42 and related halocuprates43 and organo- − cuprates [(ArCu)2Ar] (Ar = mesityl) have been structurally characterized.44 Other higher energy isomeric structures include linear [HCuHCuH]− 2B, and the T-shaped complex − 45 [H2CuCuH] (2C). −− [Cu23 H ]→+ [CuH2 ] CuH (40) Given that organocuprate reactivity is governed by the orbital interactions between the HOMO of the cuprate and the LUMO of the electrophilic substrate,46 we were interested in establishing whether the enhanced reactivity of the mononuclear hydrido cuprate arises from differences in its HOMO. A comparison of the HOMOs of 1A, Figure 3B, and 2A, Figure 3C, reveals that the HOMO of 1A is higher in energy and is based on the hydride ligands. This contrasts with both the HOMO of dimethyl cuprate, which is based on the Cu center,46 and 2A, which is based on both the copper atoms and the − terminal hydrides. Given the unique HOMO of [CuH2] ,we Figure 5. Energy profile obtained from DFT calculations for the − next examined the potential energy diagrams of various insertion of CO2 into [CuH2] . The relative Gibbs and potential mechanisms associated with its reactions with all of the energies (in parentheses) obtained from the M06/BS2//M06/BS1 substrates examined experimentally. calculations are given in kcal mol−1.

2392 DOI: 10.1021/acs.inorgchem.6b02145 Inorg. Chem. 2017, 56, 2387−2399

142 Inorganic Chemistry Article

− [CuH(O2CH)] , 7, that in turn can isomerize to the thermodynamically favored isomer 8 via TS7−8. It follows ff from the calculations that CS2 is activated through a di erent π* mechanism than CO2. The CS2 molecule, with a lower lying orbital, has the capability to accept electron density from the Cu center and thus binds strongly to Cu. In contrast, due to the  π relatively strong C O bonds in CO2, the coordination of this molecule to Cu is less feasible; for example, attempts to optimize the corresponding three-membered ring complex η2 − [H2Cu( -CO2)] failed. In support of this approach, we found that the analogous transition structure TS2−3 (Figure 4) for CO2 is very unstable with a Gibbs free energy of 44.8 kcal −1 mol . On the other hand, CO2 is very reactive toward side-on hydride transfer and can undergo the relevant reaction without any activation barrier, whereas this is not the case for CS2. Because hydride transfer increases the electron density on the pendant oxygen atoms of CO2, the high electronegativity of the oxygen atoms helps to facilitate this process. The slow experimentally determined rate and efficiency of this insertion reaction can be reconciled with the calculated surface by noting that the initially formed insertion products 7 and 8 are “hot” and must be collisionally cooled with the helium bath gas for them to be trapped and subsequently detected in the mass spectrometry experiments. Thus, it is likely that a large proportion of these “hot” product ions undergo deinsertion − to regenerate the reactant anion, [CuH2] . DFT Calculated Mechanisms for the Reactions of − [CuH2] with Methyl Iodide and Allyl Iodide. Cuprates can react with methyl iodide and allyl iodide via a number of fi mechanisms, including side-on SN2 reactions or via oxidative Figure 6. Energy pro le obtained from DFT calculations for the side- 22a−d − addition followed by reductive elimination. The only on SN2 reaction of [CuH2] with methyl iodide. The relative Gibbs − viable pathway for the reaction of [CuH2] with methyl iodide and potential energies (in parentheses) obtained from the M06/BS2// M06/BS1 calculations are given in kcal mol−1. involves a side-on SN2 reaction as shown in Figure 6. The carbon atom of the methyl group directly approaches the hydride ligand via transition state TS1A−9, with a colinear which of these pathways dominates depends on the nature of − 49 arrangement of H Cu---H---CH3---I, which then produces the substrate, the solvent, as well as the NHC. The SN2 complex 9 between copper hydride, methane, and the iodide − reaction for allyl iodide has a lower barrier than that for methyl anion. Although 9 can dissociate to yield HCuHCH3 10 and I , iodide, consistent with a slightly faster reaction between allyl − it is entropically favored for 9 to dissociate via loss of both iodide and [CuH2] . methane and CuH to form the experimentally observed An oxidative-addition-reductive-elimination pathway was − complex, [CuH(I)] 11 (eq 18), and iodide anion (eq 17). located and involves multiple transition states (Figure S16). All attempts to locate a transition state for oxidative addition via While the initial transition state has an activation energy similar − ′ approach of the copper atom of [CuH2] to the backside of the to those for the SN2 and SN2 pathways, the DFT calculations carbon atom of methyl iodide collapsed to the side-on predict that homocoupling should be the major pathway. Given − transition state TS1A−10. that the product of homocoupling, [C H CuI] ,isnot − 3 5 In the case of [CuH2] reacting with allyl iodide, the observed experimentally, this suggests that oxidative addition following mechanisms were examined via DFT calculations: (i) pathways are unlikely to be important. ′ side-on SN2 and SN2 reactions (Figure 7); and (ii) oxidative DFT Calculated Mechanisms for the Reactions of − addition followed by reductive elimination (Figure S16). The [CuH2] with Acids. The DFT calculated surface for the ′ − transition states for the SN2 and SN2 reactions of allyl iodide reaction of [CuH2] , 1A, with methanol proceeds via the initial (TS1A−12 and TS1A−13) are similar to that of methyl iodide formation of the ion−molecule complex, 16. The transition (TS1A−9) and also produce complexes 12 and 13 between state, TS16−17, barrier for subsequent formation of H2 and the copper hydride, propene, and the iodide anion. These coordinated methoxide, 17, lies above the energy of separated complexes can dissociate via the formation of the organo- reactants (Figure 8), consistent with the fact that no reaction is η2 − − 50 metallic [HCu( -C3H5)] , 14, and I , extrusion of propene, or observed under the near thermal conditions of the ion trap. In loss of both propene and CuH to form the experimentally contrast, the reaction of [CuH ]−, 1A, with 2,2,2-trifluoroetha- − 2 observed complex, [CuH(I)] 15 (eq 23), and iodide anion nol to give 18 and the subsequent formation of the coordinated (eq 22). The SN2 pathway has a slightly lower barrier than that alkoxide, 19, has a transition state barrier for the formation of ′ for the SN2 reaction, but both reactions might occur under the H2, TS18−19, that is below the energy of the separated reactants. experimental conditions. Interestingly, a recent study on the Because the barrier for proton transfer decreases as the acidity reactions of N-heterocyclic carbene (NHC) copper hydride increases from CF3CH2OH (Figure 8)toCH3(CH2)3SH complexes, LCuH, with allyl bromides has found that the (Figure S17), there is a good correlation between the gas-phase ′ products of both SN2 and SN2 pathways can occur and that acidity of the substrates and the DFT calculated activation

2393 DOI: 10.1021/acs.inorgchem.6b02145 Inorg. Chem. 2017, 56, 2387−2399

143 Inorganic Chemistry Article

fi − ′ Figure 7. Energy pro le obtained from DFT calculations for the side-on SN2 reactions of [CuH2] with allyl iodide. Left, SN2 ; right, SN2. The relative Gibbs and potential energies (in parentheses) obtained from the M06/BS2//M06/BS1 calculations are given in kcal mol−1. energies (Figure S18). This supports the view that the reactivity Using the DFT calculated transition states and substituting H is mainly controlled by the acidity of the substrates. The kinetic for D gives KIEs of 0.97 and 0.99 associated with TS1A−9 and 63 − isotope effects (0.85 and 0.86, respectively) are consistent with TS1A−9 for the reactions of [ CuHD] with methyl iodide and rearrangement processes involving hydride transfer in the rate- allyl iodide, respectively (Figure S19). The former value is in poor agreement with the experimentally determined KIE of determining step; for example, see TS18−19 for CF3CH2OH 1.51 (Table 2), highlighting that interpretation of the (Figure 8). ff Experimentally Determined Kinetic Isotope Effects experimentally measured isotope e ects is problematic in the and Their Relationship to the DFT Calculated Mecha- case of reactions such as these, where the formation of iodide dominates the product channels. nisms. We now briefly revisit the kinetic isotope effects Finally, the protonation reactions both give inverse isotope determined from the integrated ion abundances of the product effects. Because there is only one transition state associated ions formed in the reactions of [63CuHD]− with various with these reactions (TS18−19 in Figure 8b and TS26−27 in substrates (Table 2). We note that care must be used in their Figure S17), this is consistent with the DFT calculated interpretation because some reactions approach the collision- transition states, which are found to be “late” (Figure 9). controlled limit, so competitive effects in reaction channels Related inverse isotope effects for protonation reactions of must be considered.51 This is challenging in the cases of the metal hydrides have also been proposed to occur via late 63 − 38e fi reactions of [ CuHD] with CS2, methyl iodide, and allyl transition states. Such late transition states bene t from the iodide, where the isotope effect for the other product channels formation of the stronger H−D bond,52 giving rise to inverse could not be determined due to formation of ions that did not isotope effects. Indeed, using the DFT calculated transition contain H or D (i.e., eqs 5, 18, 19, 27, and 28). The isotope states and substituting H for D gives KIEs of 0.66 and 0.61 ff − associated with TS − and TS − (Figure S19). e ect for formation of [HCS2] , which is the dominant reaction 18 19 26 27 of CS2, is close to unity. This is consistent with DFT calculated potential energy diagram (Figure 4), where the rate- ■ CONCLUSIONS − determining step TS1A−2 does not involve breaking the Cu Mass spectrometry experiments involving isotope labeling H(D) bond. together with DFT calculations highlight that the hydrido

2394 DOI: 10.1021/acs.inorgchem.6b02145 Inorg. Chem. 2017, 56, 2387−2399

144 Inorganic Chemistry Article

fi − Figure 8. Energy pro le obtained from DFT calculations for the reaction of [CuH2] with alcohols, ROH: (a) R = Me; (b) R = CF3CH2. The relative Gibbs and potential energies (in parentheses) obtained from the M06/BS2//M06/BS1 calculations are given in kcal mol−1. ff − Table 2. Kinetic Isotope E ect (KIE) Associated with the cuprate [CuH2] exhibits a rich substrate-dependent bimolec- Formation of Certain Product Ion Channels in the Ion− ular reactivity. [CuH ]− readily reacts with methyl iodide and − 2 Molecule Reactions of [63CuHD] (m/z 66) with Various allyl iodide via substitution, a reaction that has been used in the Neutral Reagents condensed phase to reductively remove halogens from alkyl 8d,9 − a halides. The insertion of CO2 into the Cu H bond leads to ion molecule KIE = AH/AD 63 − b the formation of coordinated formate, while CS2 reacts to give [ CuHD] CS2 AH(eq 3)/AD(eq 4) = 0.98 b the dithioformate anion as the major product. The former CH3I AH(eq 21)/AD(eq 20) = 1.51 36 C H Ib A /A = 0.9 reaction has been observed in solution for copper hydrides. 3 5 H(eq 30) D(eq 29) ff CF CH OH A /A = 0.85 Another di erence between these two systems is that CS2 is 3 2 H(eq 34) D(eq 33) fi CH (CH ) SH A /A = 0.86 capable of releasing H2 but CO2 is not. This nding can be 3 2 3 H(eq 38) D(eq 37) π* a rationalized by the lower energy orbitals on CS2, which AH = ion count of the ionic product where the neutral molecule has promotes coordination to the metal center. reacted with the hydride (light isotope) via given eq no.; and AD = ion − count of the ionic product where the neutral molecule has reacted with The hydrido cuprate [CuH2] has been suggested as a the deuteride (heavy isotope) via given eq no. bThe isotope effects complex for hydrogen storage applications.12 Our studies have associated with other product channels could not be determined due shown that there are two different chemically induced routes to the formation of ions that did not contain H or D. − for the liberation of hydrogen from [CuH2] via the following:

− Figure 9. DFT calculated transition states for the protonation reactions of [CuH2] with (a) CF3CH2OH via TS18−19, and (b) CH3(CH2)3SH via TS26−27.

2395 DOI: 10.1021/acs.inorgchem.6b02145 Inorg. Chem. 2017, 56, 2387−2399

145 Inorganic Chemistry Article

− (i) Homocoupling of both hydride ligands triggered by a extrapolation of plots of ln([hydrido cuprate] intensity/total ions) substrate, to give the substrate coordinated to copper. As versus RD. Rate constants were calculated by dividing the pseudo first- order rate coefficient by the calculated concentration of carbon mentioned above, only CS2 reacts to trigger such a loss via the fi − formation of the organometallic ion [CuCS ]− (eq 2), but this disul de in the ion trap. Theoretical rates for the ion molecule 2 reactions of hydrido cuprates with neutrals were calculated using the is only a minor reaction channel. 28 − Average Dipole Orientation (ADO) theory of Su and Bowers with (ii) Protonation by an acid, AH, to give [CuHA] . This 29 fl the program COLRATE. reaction occurs for both 2,2,2-tri uoroethanol (eq 32) and 1- Determination of Kinetic Isotope Effects. The kinetic isotope butanethiol (eq 36). Related reactions have been proposed to effect was determined for the ion−molecule reactions of the mass occur in the solution for copper hydrides reacting with selected isotopologue [63CuHD]− (m/z 66) with carbon disulfide, isopropanol.38d The observation of facile protonation of methyl iodide, allyl iodide, 2,2,2-trifluoroethanol, and 1-butanethiol − (see Table 2). The ion count for product ion produced from the [CuH2] by these acids suggests the possibility of using activation of the Cu−H bond was divided by the ion count for the hydrido cuprates as catalysts to selectively transform formic − acid into hydrogen and carbon dioxide (eq 41).53 Such studies product ion produced by the activation of the Cu D bond. Depending on the reaction, corrections were made for overlapping isotope are underway and will be reported in due course. contributions, that is, 2H, 13C, 17O, and 33S. Theoretical Methods. 55 HCO222 H→+ H CO (41) Gaussian 09 was used to fully optimize all of the structures reported in this Article at the M06 level of density functional theory (DFT).56 The SDD basis set with Stuttgart ■ EXPERIMENTAL SECTION potentials was used to describe Cu and I.57 The 6-31+G(d,p) basis Materials. Chemicals from the following suppliers were used set was used for other atoms. This basis set combination will be without further purification: Ajax Finechem, (i) copper(I) oxide, referred to as BS1. Frequency calculations were carried out at the same level of theory as those for the structural optimization. Transition Cu2O; (ii) formic acid (98%), HO2CH; MERCK, (iii) acetonitrile, (98%), CH CN; SIGMA, (iv) formic acid d2 (98%), DO CD; (v) 1- structures were located using the Berny algorithm. Intrinsic reaction 3 2 coordinate (IRC)58 calculations were used to confirm the connectivity butanethiol, (99%), CH3(CH2)3SH; (vi) allyl iodide, CH2CHCH2I, between transition structures and minima. To further refine the (98%); (vii) methyl iodide, CH3I (99%); COREGAS, (viii) gaseous energies obtained from the M06/BS1 calculations, we carried out carbon dioxide, CO2 (1.03% in Ultra High Purity helium (99.995%)); fi single-point energy calculations for all of the structures with a larger CHEMSUPPLY, (ix) carbon disul de, CS2 (99.9%); ACROS, (x) 2,2,2-trifluorethanol, CF CH OH (99.8%). basis set (BS2) at the M06 level. BS2 utilizes the def2-TZVP basis set 3 2 ff Mass Spectrometry. Mass spectrometry experiments were on all atoms. An e ective core potential including scalar relativistic ff 59 conducted on a modified Finnigan hybrid linear triple-quadrupole e ect was used for I atom. To estimate the corresponding Gibbs Δ (LTQ) Fourier transform ion cyclotron resonance (FTICR) mass energies, G, the corrections were calculated at the B3LYPD3BJ/BS1 × −3 spectrometer.54 Under ion−molecule reaction conditions, collisions level using the conditions of T = 298.15 K; P =2 10 Torr, which fl ≈ ≈ × with the helium bath gas quasi-thermalize the ions to room re ect the operating conditions of the ion trap (T 298 K; P 2 −3 fi temperature.50 Both CID and IMR experiments were performed on 10 Torr) and nally added to the single-point energies. We have this instrument.54 The solution used to prepare copper hydride anions used the corrected Gibbs free energies and the potential energies in the gas phase was prepared as follows: In situ hydrido formate obtained from the M06/BS2//M06/BS1 calculations throughout this complexes for ESI−MS were typically generated by adding 10 mmol of Article unless otherwise stated. Finally, the DFT calculated KIEs were copper(I) oxide to a 20 mL solution of acetonitrile, generating a determined by calculating the Gibbs free energies of the transition − ’ − reddish colored suspension. To this suspension was immediately states where either of the Cu H s was replaced by Cu D(Figure added 20 mmol of formic acid to give a clear solution. S19). This solution was diluted in acetonitrile to a copper(I) concentration of 50 μM and injected at a flow rate 3−5 μL min−1 ■ ASSOCIATED CONTENT into the Finnigan ESI source. ESI source conditions typically involved *S Supporting Information needle potentials of 3.0−5.0 kV to give a stable source current of ca. The Supporting Information is available free of charge on the 0.5 μA and a nitrogen sheath gas pressure of 8 arbitrary units. The ion ° ACS Publications website at DOI: 10.1021/acs.inorg- transfer capillary temperature was set to 275 C. The tube lens voltage chem.6b02145. was set to −38 V, and the capillary voltage was set to −13 V. Unimolecular fragmentation studies involved the anion of interest Mass spectra, pseudo-first-order kinetics, DFT calculated being mass selected in the linear ion trap (LIT) and then subjected to mechanisms, and Cartesian coordinates and total CID, where the normalized collision energy was set between 10−25 energies for all calculated structures (PDF) arbitrary units to result in the mass selected precursor ion to be − depleted to 10 20% with an activation Q of 0.25 and activation time ■ AUTHOR INFORMATION of 30 ms. Kinetic Measurements. The kinetics for the reaction between the Corresponding Authors hydrido cuprate anions and neutral reagents were examined using the *E-mail: [email protected]. LTQ FT hybrid mass spectrometer. Ion−molecule reaction rates were *E-mail:[email protected]. measured by isolating the reactant ion and allowing it to react with ORCID neutral reagent at various activation times, similar to previously Richard A. J. O’Hair: 0000-0002-8044-0502 reported ion−molecule reactions.54 The neutral substrates were introduced at various concentrations into the ion trap via the helium Notes inlet line. In the case of the reactions with carbon dioxide, the helium The authors declare no competing financial interest. bath gas cylinder was replaced by a helium cylinder seeded with 1.03% CO2, and the entire mass spectrometer was allowed to re-equilibrate ■ ACKNOWLEDGMENTS with this gas mixture for 15 min. Rates were measured by varying the time delay between isolation of the reactant ion and its mass analysis We thank Dr. George Khairallah for useful discussions “ ” regarding the mass spectrometry experiments and the ( reaction delay , RD). The decay of the reactant hydrido cuprate fi anions was monitored over at least eight values of RD. The intensity of Australian Research Council for nancial support the reactant ion was calculated by integration of its ion count within DP150101388 (to R.A.J.O. and A.J.C.). We gratefully acknowl- the mass-selected window. Pseudo first-order rates were estimated by edge the generous allocation of computing time from the

2396 DOI: 10.1021/acs.inorgchem.6b02145 Inorg. Chem. 2017, 56, 2387−2399

146 Inorganic Chemistry Article

University of Tasmania and the National Computing Infra- (f) Semmelhack, M. F.; Stauffer, R. D. Reductions with copper structure. We thank the reviewers for their helpful insights. hydride. New preperative and mechanistic aspects. J. Org. Chem. 1975, 40, 3619−3621. ■ REFERENCES (9) [KCuH2]n complexes: (a) Yoshida, T.; Negishi, E.-I. A novel ̈ ′ copper-containing hydride species and its application to the reduction (1) Wurtz, A. C. R. Sur l hydrure de cuivre. Hebd. Seances Acad. Sci. of organic substances. J. Chem. Soc., Chem. Commun. 1974, 762−763. 1844, 18, 702. (b) Masamune, S.; Rossy, P. A.; Bates, G. S. Reductive removal of halo (2) For reviews on the use of copper hydrides as reagents in organic and mesyloxy groups with a copper(I) complex. J. Am. Chem. Soc. synthesis, see: (a) Lipshutz, B. H. In Modern Organocopper Chemistry; 1973, 95, 6452−6454. Krause, N., Ed.; Wiley-VCH Verlag GmbH Weinheim: Germany, (10) [LiRCuH] complexes: (a) Boeckman, R. K.; Michalak, R. 2002; Chapter 5, pp 167−187. (b) Deutsch, C.; Krause, N. CuH- n Reduction of α,β-unsaturated carbonyl compounds by ″ate″ complexes Catalyzed Reactions. Chem. Rev. 2008, 108, 2916−2927. (c) Lipshutz, of copper(I) hydride. J. Am. Chem. Soc. 1974, 96, 1623−1625. B. H. Rediscovering organocopper chemistry through copper hydride. (b) Masamune, S.; Bates, G. S.; Georghiou, P. E. Reactions of lithium It’s all about the ligand. Synlett 2009, 4, 509−524. (d) Riant, O. In alkyl and alkynyl cuprates. Selective removal of halo and mesyloxy Chemistry of Organocopper Compounds; Rappoport, Z., Marek, I., Eds.; groups and reduction of α,β-unsaturated ketones. J. Am. Chem. Soc. John Wiley & Sons Ltd.: Chichester, UK, 2009; Chapter 15, pp 731− 1974, 96, 3686−3688. (c) House, H. O.; DuBose, J. C. Reduction as a 773. (e) Lipshutz, B. H. In Copper-Catalyzed Asymmetric Synthesis; side reaction arising from the thermal decomposition of lithium Alexakis, A., Krause, N., Woodward, S., Eds.; Wiley-VCH Verlag − organocuprates to form copper hydride derivatives. J. Org. Chem. 1975, GmbH & Co. KGaA: Weinheim, Germany, 2014; Chapter 7, pp 179 − 201. (f) Fujihara, T.; Semba, K.; Terao, J.; Tsuji, Y. Regioselective 40, 788 790. transformation of alkynes catalyzed by a copper hydride or boryl (11) Andrews, L. Matrix infrared spectra and density functional − calculations of transition metal hydrides and dihydrogen complexes. copper species. Catal. Sci. Technol. 2014, 4, 1699 1709. (g) Pirnot, M. − T.;Wang,Y.-M.;Buchwald,S.L.CopperHydride-Catalyzed Chem. Soc. Rev. 2004, 33, 123 132. (12) Andrews, L.; Wang, X. Infrared Spectra and Structures of the Hydroamination of Alkenes and Alkynes. Angew. Chem., Int. Ed. − − − − 2016, 55,48−57. For a review on the synthesis, structure, and Stable CuH2 , AgH2 , AuH2 , and AuH4 Anions and the AuH2 Molecule. J. Am. Chem. Soc. 2003, 125, 11751−11760. reactivity of coinage metal hydrides, see: Jordan, A. J.; Lalic, G.; − Sadighi, J. P. Coinage Metal Hydrides: Synthesis, Characterization, and (13) The gas-phase photo electron spectrum of [CuH2] has also Reactivity. Chem. Rev. 2016, 116, 8318−8372. been reported: Calvi, R. M. D.; Andrews, D. H.; Lineberger, W. C. Negative ion photoelectron spectroscopy of copper hydrides. Chem. (3) Dilts, J. A.; Shriver, D. F. Nature of soluble copper(I) hydride. J. − Am. Chem. Soc. 1968, 90, 5769−5772. Phys. Lett. 2007, 442,12 16. (4) (a) Churchill, M. R.; Bezman, S. A.; Osborn, J. A.; Wormald, J. (14) Wang, X.; Andrews, L.; Manceron, L.; Marsden, C. Infrared Spectra and DFT Calculations for the Coinage Metal Hydrides MH, Preparation and crystallographic characterization of a hexameric · − triphenylphosphinecopper hydride cluster. J. Am. Chem. Soc. 1971, (H2) MH, MH2,M2H, M2H , and (H2) CuHCu in Solid Argon, − Neon, and Hydrogen. J. Phys. Chem. A 2003, 107, 8492−8505. 93, 2063 2065. (b) Churchill, M. R.; Bezman, S. A.; Osborn, J. A.; − Wormald, J. Synthesis and molecular geometry of hexameric (15) The gas-phase photo electron spectrum of [Cu2H] has also triphenylphosphinocopper(I) hydride and the crystal structure of been reported: (a) Xie, H.; Li, X.; Zhao, L.; Liu, Z.; Qin, Z.; Wu, X.; − Tang, Z.; Xing, X. Vibrationally Resolved Photoelectron Imaging of H6Cu6(PPh3)6.HCONMe2. Inorg. Chem. 1972, 11,18181825. − − (c) Bennett, E. L.; Murphy, P. J.; Imberti, S.; Parke, S. F. Cu2H and AgCuH and Theoretical Calculations. J. Phys. Chem. A − Characterization of the Hydrides in Stryker’s Reagent: [HCu{P- 2013, 117, 1706 1711. (b) Vetter, K.; Proch, S.; Gantefoer, G.. F.; − Behera, S.; Jena, P. Hydrogen mimicking the properties of coinage (C6H5)3}]6. Inorg. Chem. 2014, 53, 2963 2967. (d) Whitesides, G. M.; San Filippo, J.; Stredronsky, E. R.; Casey, C. P. Reaction of copper(I) metal atoms in Cu and Ag monohydride clusters. Phys. Chem. Chem. − hydride with organocopper(I) compounds. J. Am. Chem. Soc. 1969, 91, Phys. 2013, 15, 21007 21015. 6542−6544. (16) Sass, D. C.; Heleno, V. C. G.; Cavalcante, S.; da Silva Barbosa, J.; (5) Other ligands have been used to produce a range of copper Soares, A. C. F.; Constantino, M. G. Solvent Effect in Reactions Using ’ − hydride clusters of diﬀerent nuclearity. For a review, see: Dhayal, R. S.; Stryker s Reagent. J. Org. Chem. 2012, 77, 9374 9378. ’ van Zyl, W. E.; Liu, C. W. Polyhydrido Copper Clusters: Synthetic (17) (a) O Hair, R. A. J. The 3D Quadrupole Ion Trap Mass Advances, Structural Diversity, and Nanocluster-to-Nanoparticle Spectrometer as a Complete Chemical Laboratory for Fundamental Conversion. Acc. Chem. Res. 2016, 49,86−95. Gas Phase Studies of Metal Mediated Chemistry. Chem. Commun. (6) (a) Mahoney, W. S.; Brestensky, D. M.; Stryker, J. M. Selective 2006, 1469−1481. (b) O’Hair, R. A. J.; Rijs, N. J. Gas Phase Studies of hydride-mediated conjugate reduction of α,β-unsaturated carbonyl the Pesci Decarboxylation Reaction: Synthesis, Structure, and − compounds using [(Ph3P)CuH]6. J. Am. Chem. Soc. 1988, 110, 291 Unimolecular and Bimolecular Reactivity of Organometallic Ions. 293.(b)Mahoney,W.S.;Stryker,J.M.Hydride-mediated Acc. Chem. Res. 2015, 48, 329−340. homogeneous catalysis. Catalytic reduction of α,β-unsaturated ketones (18) Armentrout, P. B.; Sunderlin, L. S. In Transition Metal Hydrides; − − using [(Ph3P)CuH]6 and H2. J. Am. Chem. Soc. 1989, 111, 8818 8823. Dedieu, A., Ed.; VCH: New York, 1992; pp 1 64. (7) Lipshutz, B. H.; Frieman, B. A. CuH in a Bottle: A Convenient (19) (a) Squires, R. R. Gas-Phase transition-metal negative ion Reagent for Asymmetric Hydrosilylations. Angew. Chem., Int. Ed. 2005, chemistry. Chem. Rev. 1987, 87,623−646. (b) Damrauer, R. 44, 6345−6348. Organometallic Chemistry in the Flowing Afterglow: A Review. Organometallics 2004, 23, 1462−1479. (8) [LiCuH2]n complexes: (a) Ashby, E. C.; Korenowski, T. F.; − Schwartz, R. D. Preparation of the first stable complex metal hydride (20) (a) Lane, K. R.; Squires, R. R. Formation of HCr(CO)3 from − of copper, LiCuH2. J. Chem. Soc., Chem. Commun. 1974, 157 158. the remarkable reaction of hydride ion with benzenechromium (b) Ashby, E. C.; Goel, A. B. Preparation and characterization of tricarbonyl. Gas-phase reactions of a novel 14-electron metal anion − complex metal hydrides of copper, LimCunHm+n. Inorg. Chem. 1977, complex. J. Am. Chem. Soc. 1985, 107, 6403 6404. (b) McDonald, R. 16, 3043−3047. (c) Ashby, E. C.; Goel, A. B.; Lin, J. J. Preparation, N.; Schell, P. L. Gas-phase ligand substitution reactions with the 17- − − properties and applications of some new complex metal hydrides of electron transition-metal complexes (OC)4Fe ,(OC)5Cr ,and − − − copper, LinCumHn+m. Tetrahedron Lett. 1977, 18, 3695 3698. (OC)4MnH . Organometallics 1988, 7, 1820 1827. (c) Khairallah, ’ + (d) Ashby, E. C.; Lin, J.-J.; Goel, A. B. Reactions of complex metal G. N.; O Hair, R. A. J. Gas Phase Synthesis of Ag4H and its Mediation hydrides of copper with alkyl halides, enones, and cyclic ketones. J. of C-C Bond Coupling of Allylbromide. Angew. Chem., Int. Ed. 2005, Org. Chem. 1978, 43, 183−188. (e) Simaan, S.; Marek, I. Copper- 44, 728−731. (d) Schlangen, M.; Schwarz, H. Thermal activation of + catalyzed hydride transfer from LiAlH4 for the formation of methane by group 10 metal hydrides MH : the same and not the alkylidenecyclopropane derivatives. Chem. Commun. 2009, 292−294. same. Angew. Chem., Int. Ed. 2007, 46, 5614−5617. (e) Firouzbakht,

2397 DOI: 10.1021/acs.inorgchem.6b02145 Inorg. Chem. 2017, 56, 2387−2399

147 Inorganic Chemistry Article

M.; Rijs, N. J.; Gonzalez-Navarrete, P.; Schlangen, M.; Kaupp, M.; (33) (a) Vajenine, G. V.; Hoffmann, R. Compounds Containing Schwarz, H. On the Activation of Methane and Carbon Dioxide by Copper-Sulfur Layers: Electronic, Structure, Conductivity and [HTaO]·+ and [TaOH]·+ in the Gas Phase: A Mechanistic Study. Stability. Inorg. Chem. 1996, 35, 451−457. (b) Ni, B.; Kramer, J. R.; Chem. - Eur. J. 2016, 22, 10581−10589. Westiuk, N. H. An ab Initio and AIM Study on the Molecular − (21) Martinez, O., Jr; Yang, Z.; Demarais, N. J.; Snow, T. P.; Structure and Stability of Small CuxSy Clusters. J. Phys. Chem. A 2003, Bierbaum, V. M. Gas-phase reactions of hydride anion, H− Astrophys. 107, 8949−8954. − Astrophys. J. 2010, 720, 173 177. (34) For a review on the gas-phase reactions of CO2 with metal ions − (22) Reaction of [CuMe2] with methyl iodide: (a) James, P. F.; and metal-containing ions using mass spectrometry-based techniques, O’Hair, R. A. J. Dimethyl cuprate undergoes C-C bond coupling with see: Schwarz, H. Metal-mediated activation of carbon dioxide in the methyliodide in the gas phase but dimethyl argenate does not. Org. gas phase: Mechanistic insight derived from a combined experimental/ Lett. 2004, 6, 2761−2764. (b) Rijs, N. J.; Sanvido, G. B.; Khairallah, G. computational approach. Coord. Chem. Rev. 2017, 334, 112−123. N.; O’Hair,R.A.J.Gasphasesynthesisandreactivityof (35) For a review on DFT calculations on the insertion reactions of − dimethylaurate. Dalton Trans. 2010, 39, 8655 8662. (c) Rijs, N. J.; CO2 with transition metal complexes, see: Fan, T.; Chen, X.; Lin, Z. Yoshikai, N.; Nakamura, E.; O’Hair, R. A. J. Unraveling Organocuprate Theoretical studies of reactions of carbon dioxide mediated and Complexity: Fundamental Insights into Intrinsic Group Transfer catalysed by transition metal complexes. Chem. Commun. 2012, 48, Selectivity in Alkylation Reactions. J. Org. Chem. 2014, 79, 1320−1334. 10808−10828. − Reaction of [CuMe2] with allyl iodide: (d) Rijs, N. J.; Yoshikai, N.; (36) (a) Zhang, L.; Cheng, J.; Hou, Z. Highly efficient catalytic Nakamura, E.; O’Hair, R. A. J. Gas-Phase Reactivity of Group 11 hydrosilyation of carbon dioxide by an N-heterocyclic carbine copper Dimethylmetallates with Allyl Iodide. J. Am. Chem. Soc. 2012, 134, catalyst. Chem. Commun. 2013, 49, 4782−4784. (b) Zall, C. M.; − − 2569 2580. Reaction of [CuMe2] with carbon disulfide: (e) Li, J.; Linehan, J. C.; Appel, A. M. Triphosphine-Ligated Copper Hydrides or ’ Khairallah, G. N.; O Hair, R. A. J. Dimethylcuprate-Mediated CO2 Hydrogenation: Structure, Reactivity, and Thermodynamic Transformation of Acetate to Dithioacetate. Organometallics 2015, Studies. J. Am. Chem. Soc. 2016, 138, 9968−9977. (c) Nakamae, K.; 34, 488−493. Kure, B.; Nakajima, T.; Ura, Y.; Tanase, T. Facile Insertion of Carbon ’ μ (23) (a) Rijs, N.; Waters, T.; Khairallah, G. N.; O Hair, R. A. J. Gas- Dioxide into Cu2( -H) Dinuclear Units Supported by Tetraphosphine Phase Synthesis of the Homo and Hetero Organocuprate Anions Ligands. Chem. - Asian J. 2014, 9, 3106−3110. (d) Motokura, K.; [MeCuMe]−, [EtCuEt]−, and [MeCuR]−. J. Am. Chem. Soc. 2008, 130, Kashiwame, D.; Takahashi, N.; Miyaji, A.; Baba, T. Highly Active and 1069−1079. (b) Rijs, N. J.; O’Hair, R. A. J. Unimolecular Reactions of Selective Catalysis of Copper Diphosphine Complexes for the Organocuprates and Organoargenates. Organometallics 2010, 29, Transformation of Carbon Dioxide into Silyl Formate. Chem. - Eur. 2282−2291. J. 2013, 19, 10030−10037. (e) Zall, C. M.; Linehan, J. C.; Appel, A. M. ’ (24) Rijs, N. J.; Yates, B. F.; O Hair, R. A. J. Dimethylcuprate A Molecular Copper Catalyst for Hydrogenation of CO2 to Formate. undergoes a dyotropic rearrangement. Chem. - Eur. J. 2010, 16, 2674− ACS Catal. 2015, 5, 5301−5305. (f) Watari, R.; Kayaki, Y.; Hirano, S.- 2678. i.; Matsumoto, N.; Ikariya, T. Hydrogenation of Carbon Dioxide to (25) Tuytten, R.; Lemiere,̀ F.; Esmans, E. L.; Herrebout, W. A.; van Formate Catalyzed by a Copper/1,8-Diazabicyclo[5.4.0]undec-7-ene der Veken, B. J.; Dudley, E.; Newton, R. P.; Witters, E. In-source CID System. Adv. Synth. Catal. 2015, 357, 1369−1373. of guanosine: gas phase ion−molecule reactions. J. Am. Soc. Mass (37) Cheng, T.-Y.; Bullock, R. M. Isotope Effects on Hydride Spectrom. 2006, 17, 1050−1062. Transfer Reactions from Transition Metal Hydrides to Trityl Cation. (26) For an explanation of hydricity, see: (a) Jacobsen, H.; Berke, H. An Inverse Isotope Effect for a Hydride Transfer. J. Am. Chem. Soc. In Recent Advances in Hydride Chemistry; Peruzzini, M., Poli, R., Eds.; 1999, 121, 3150−3155. Elsevier: Amsterdam, 2001; Chapter 4, pp 89−116. (b) DuBois, D. L.; (38) For reviews on the protonation of transition metal hydrides, see: Berning, D. E. Hydricity of transition-metal hydrides and its role in (a) Bakhmutov, V. I. Proton transfer to hydride ligands with formation − CO2 reduction. Appl. Organomet. Chem. 2000, 14, 860 862. (c) Tsay, of dihydrogen complexes: A physicochemical view. Eur. J. Inorg. Chem. C.; Livesay, B. N.; Ruelas, S.; Yang, J. Y. Solvation Effects on 2005, 2005, 245−255. (b) Besora, M.; Lledos, A.; Maseras, F. Transition Metal Hydricity. J. Am. Chem. Soc. 2015, 137, 14114− Protonation of transition-metal hydrides: a not so simple process. 14121. Chem. Soc. Rev. 2009, 38, 957−966. (c) Belkova, N. V.; Epstein, L. M.; (27) For a discussion of the gas-phase hydride ion affinity scale, see: Shubina, E. S. Dihydrogen Bonding, Proton Transfer and Beyond: (a) Squires, R. R. In Structure/Reactivity and Thermochemistry of Ions; What We Can Learn from Kinetics and Thermodynamics. Eur. J. Inorg. Ausloos, P., Lias, S. G., Eds.; D. Reidel Publsihing Co.: Dordrecht, Chem. 2010, 2010, 3555−3565. (d) Whittaker, A. M.; Lalic, G. 1987; pp 373−375. (b) Goebbert, D. J.; Wenthold, P. G. Gas-phase Monophasic Catalytic System for the Selective Semireduction of hydride affinities of neutral molecules. Int. J. Mass Spectrom. 2006, 257, Alkynes. Org. Lett. 2013, 15, 1112−1115. (e) Basallote, M. G.; Duran,́ 1−11. J.; Fernandez-Trujillo,́ M. J.; Mań̃ez, M. A. Kinetics of protonation of (28) Su, T.; Bowers, M. T. Ion-polar molecule collisions. Effect of cis-[FeH2(dppe)2]: formation of the dihydrogen complex trans- + size on ion-polar molecule rate constants. Paramterization of the [FeH(H2)(dppe)2] (dppe = Ph2PCH2CH2PPh2). J. Chem. Soc., average-dipole-orientation theory. Int. J. Mass Spectrom. Ion Phys. 1973, Dalton Trans. 1998, 2205−2210. 12, 347−356. (39) Lias, S. G.; Bartmess, J. E.; Liebman, J. F.; Holmes, J. L.; Levin, (29) Lim, K. F. COLRATE. QCPE 643: Calculation of gas-kinetic R. D.; Mallard, W. G. In NIST Chemistry WebBook, NIST Standard collision rate coefficients. QCPE Bull. 1994, 14,3. Reference Database Number 69; Linstrom, P. J., Mallard, W. G., Eds.; (30) Surry, D. S.; Spring, D. R. The oxidation of organocupratesan National Institute of Standards and Technology: Gaithersburg, MD; offbeat strategy for synthesis. Chem. Soc. Rev. 2006, 35, 218−225. http://webbook.nist.gov (retrieved January 11, 2016); Chapter “Ion (31) Gomez-Gallego,́ M.; Sierra, M. A. Kinetic Isotope Effects in the Energetics Data”. − Study of Organometallic Reaction Mechanisms. Chem. Rev. 2011, 111, (40) The measured rate constant for the reaction of [CuH2] with − 4857 4963. CH3(CH2)3SH substantially exceeds the theoretical predicted ADO (32) (a) Fisher, K.; Dance, I.; Willet, W.; Yi, M. N. Gas-phase metal rate. The measured rate contant is reproducible under a range of sulphide cluster anions. J. Chem. Soc., Dalton Trans. 1996, 709−718. different experimental conditions (see Table S1), and we routinely (b) Fisher, K.; Dance, I.; Willet, W. Gas phase coordination chemistry: check the performance of our reaction line by measuring the reaction − copper sulphide cluster anions reacting with tertiary phosphine ligands of Br with CH3I, where we obtain rate constants comparable to in the gas phase. Polyhedron 1997, 16, 2731−2735. (c) Fisher, K.; literature values obtained via flowing afterglow techniques at 298 K Dance, I. Gas phase inorganic synthesis: copper sulphide cluster anions (Gronert, S.; De Puy, C. H.; Bierbaum, V. M. J. Am. Chem. Soc. 1991, − react with phosphorus, P4, to generate copper compounds PmSn 113, 4009 4010). Unlike several studies on transition metal cluster ligands. J. Chem. Soc., Dalton Trans. 1997, 2381−2382. ions, which have highlighted that experimental rates can exceed ADO

2398 DOI: 10.1021/acs.inorgchem.6b02145 Inorg. Chem. 2017, 56, 2387−2399

148 Inorganic Chemistry Article rates by up to a factor of 3−4 (see, for example: Balaj, O. P.; Balteanu, (51) For reviews on isotope eﬀects in gas-phase mass spectrometry- I.; Rossteuscher, T. T. J.; Beyer, M. K.; Bondybey, V. E. Catalytic basedexperiments,see:(a)Derrick,P.J.Isotopeeffectsin − oxidation of CO with N2O in gas-phase platinum clusters. Angew., fragmentation. Mass Spectrom. Rev. 1983, 2, 285 298. (b) Lehman, Chem., Int. Ed. 2004, 43, 6519−6522. Anderson, M. L.; Ford, M. S.; T. A. Isotope effects in the bimolecular reactions of gaseous ions. Derrick, P. J.; Drewello, T.; Woodruﬀ, D. P.; Mackenzie, S. R. Nitric Concepts, techniques, and reactions. Mass Spectrom. Rev. 1995, 14, ± − Oxide Decomposition on Small Rhodium Clusters, Rhn J. Phys. Chem. 353 382. A 2006, 110, 10992−11000), the use of a point charge model for the (52) Taking into account zero-point energies, the bond dissociation ADO rates is an entirely reasonable approximation for a small system energy for H−D is 104.0 kcal/mol, while that of H−H is 103.2 kcal/ − ’ − such as [CuH2] . mol. See Figure 1 of: O Neil, J. R. Rev. Mineral. Geochem. 1986, 16,1 (41) The gas-phase photo electron spectrum of the related 40. [CuH(SH)]− has been reported: Qin, Z.; Liu, Z.; Cong, R.; Xie, H.; (53) (a) Zavras, A.; Khairallah, G. N.; Krstic,́ M.; Girod, M.; Daly, S.; Tang, Z.; Fan, H. Photoelectron imaging and theoretical study on the Antoine, R.; Maitre, P.; Mulder, R. J.; Alexander, S.-A.; Bonacič-́ structure and chemical binding of the mixed- ligand M(I) complexes, Koutecky,́ V.; Dugourd, P.; O’Hair, R. A. J. Ligand-induced Substrate + [HMSH] - (M = Cu, Ag, and Au). J. Chem. Phys. 2014, 140, 114307/ Steering and Reshaping of [Ag2(H)] Scaffold for Selective CO2 1−114307/7. Extrusion from Formic Acid. Nat. Commun. 2016, 7, 11746. (b) Zavras, ’ μ (42) Wyss, C. M.; Tate, B. K.; Bacsa, J.; Gray, T. G.; Sadighi, J. P. A.; White, J. M.; O Hair, R. A. J. An unusual co-crystal [( 2- μ η · μ μ Bonding and Reactivity of a μ-Hydrido Dicopper Cation. Angew. dcpm)Ag2( 2-O2CH)( 2-NO3)]2 [( 2-dcpm)2Ag4( 2-NO3)4] and its Chem., Int. Ed. 2013, 52, 12920−12923. connection to the selective decarboxylation of formic acid in the gas (43) (a) Ko, Y. J.; Wang, H.; Pradhan, K.; Koirala, P.; Kandalam, A. phase. Dalton Trans. 2016, 45, 19408−19415. (c) Zavras, A.; Krstic,́ ̌́ ́ ’ K.; Bowen, K. H.; Jena, P. Superhalogen properties of CumCln clusters: M.; Dugourd, P.; Bonacic-Koutecky, V.; O Hair, R. A. J. Selectivity Theory and experiment. J. Chem. Phys. 2011, 135, 244312/1−244312/ Effects in Bimetallic Catalysis: Role of the Metal Sites in the 7. (b) Subramanian, L.; Hoffmann, R. Bonding in Halocuprates. Inorg. Decomposition of Formic Acid into H2 and CO2 by the Coinage Metal + Chem. 1992, 31, 1021−1029. and references cited therein (c) Chen, Binuclear Complexes [dppmMM’(H)] . ChemCatChem., in press, S.; Li, J.; Zhou, C.; Wu, J.; Tempel, D. J.; Henderson, P. B.; DOI: 201710.1002/cctc.201601675. ’ − Brzozowski, J. R.; Cheng, H. Weak Chemical Complexation of PH (54) (a) Donald, W. A.; McKenzie, C. J.; O Hair, R. A. J. C H Bond 3 IV with Ionic Liquids. J. Phys. Chem. B 2010, 114, 904−909. Activation of Methanol and Ethanol by a High-Spin Fe O Biomimetic − (44) (a) Bomparola, R.; Davies, R. P.; Hornauer, S.; White, A. J. P. Complex. Angew. Chem., Int. Ed. 2011, 50, 8379 8383. (b) Lam, A. K. Structural Characterization of Magnesium Organocuprates Derived Y.; Li, C.; Khairallah, G. N.; Kirk, B. B.; Blanksby, S. J.; Trevitt, A. J.; ’ from Grignard Reagents: CuI-Based Inverse Crown Ethers. Angew. Wille, U.; O Hair, R. A. J.; da Silva, G. Gas-phase reactions of aryl Chem., Int. Ed. 2008, 47, 5812−5815. (b) Krieck, S.; Gcrls,̧ H.; radicals with 2-butyne: An experimental and theoretical investigation Westerhausen, M. Structural Diversity of Calcium Organocuprates(I): employing the N-methyl-pyridinium-4-yl radical cation. Phys. Chem. − Synthesis of Mesityl Cuprates via Addition and Transmetalation Chem. Phys. 2012, 14, 2417 2426. Reactions of Mesityl Copper(I). Chem. - Asian J. 2010, 5, 272−277. (55) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; (45) For discussions on metal−metal interactions in coinage metal Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, complexes, see: (a) Gray, T. G.; Sadighi, J. P. In Molecular Metal− B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. Metal Bonds; Liddle, S. T., Ed.; Wiley-VCH Verlag GmbH & Co. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; KGaA: Weinheim, Germany, 2015; Chapter 11, pp 397−428. Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, (b) Schmidbaur, H. The aurophilicity phenomenon: A decade of T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.; experimental findings, theoretical concepts and emerging applications. Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, Gold Bulletin 2000, 33,3−10. (c) Schmidbaur, H.; Schier, A. A briefing K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; on aurophilicity. Chem. Soc. Rev. 2008, 37, 1931−1951. (d) Schmid- Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, baur, H.; Schier, A. Argentophilic Interactions. Angew. Chem., Int. Ed. N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; 2015, 54, 746−784. (e) Cotton, F. A.; Feng, X.; Timmons, D. J. Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Further Study of Very Close Nonbonded CuI− CuI Contacts. Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Molecular Structure of a New Compound and Density Functional Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Theory Calculations. Inorg. Chem. 1998, 37, 4066−4069. Dannenberg,J.J.;Dapprich,S.;Daniels,A.D.;Farkas,O.; (46) Yoshikai, N.; Nakamura, E. Mechanisms of Nucleophilic Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09, Organocopper(I) Reactions. Chem. Rev. 2012, 112, 2339−2372. revision D.01; Gaussian, Inc.: Wallingford, CT, 2009. (47) (a) Zhou, M.; Andrews, L. Reactions of Co, Ni, and Cu Atoms (56) Zhao, Y.; Truhlar, D. G. Density Functionals with Broad Applicability in Chemistry. Acc. Chem. Res. 2008, 41, 157−167. with CS2: Infrared Spectra and Density-Functional Calculations of ̈ η + (57) (a) Bergner, A.; Dolg, M.; Kuchle, W.; Stoll, H.; Preuss, H. Ab SMCS, M-( 2-CS)S, M-CS2, and MCS2 in Solid Argon. J. Phys. Chem. − A 2000, 104, 4394−4401. (b) Dobrogorskaya, Y.; Mascetti, J.; Papai,́ initio energy-adjusted pseudopotentials for elements of groups 13 17. Mol. Phys. 1993, 80, 1431−1441. (b) Dolg, M.; Wedig, U.; Stoll, H.; I.; Nemukhin, A.; Hannachi, Y. Theoretical Investigation of the Preuss, H. Energy-adjusted ab initio pseudopotentials for the first row Reactivity of Copper Atoms with Carbon Disulfide. J. Phys. Chem. A transition elements. J. Chem. Phys. 1987, 86, 866−872. 2003, 107, 2711−2715. (58) (a) Fukui, K. Formulation of the reaction coordinate. J. Phys. (48) For reviews on the reactions of CS with transition metal 2 Chem. 1970, 74, 4161−4163. (b) Fukui, K. The path of chemical complexes, including insertion and formation of η2 complexes, see: reactions − the IRC approach. Acc. Chem. Res. 1981, 14, 363−368. (a) Yaneff, P. V. Thiocarbonyl and related complexes of the transition (59) (a) Andrae, D.; Haeussermann, U.; Dolg, M.; Stoll, H.; Preuss, metals. Coord. Chem. Rev. 1977, 23, 183−220. (b) Werner, H. Novel H. Energy-adjusted ab initio pseudopotentials for the second and third coordination compounds formed from CS and heteroallenes. Coord. 2 row transition elements. Theor. Chim. Acta 1990, 77, 123−141. Chem. Rev. 1982, 44, 165−185. (c) Pandey, K. K. Reactivities of (b) Peterson, K. A.; Figgen, D.; Goll, E.; Stoll, H.; Dolg, M. carbonyl sulphide (COS), carbon disulphide (CS ) and carbon dioxide 2 Systematically convergent basis sets with relativistic pseudopotentials. (CO2) with transition metal complexes. Coord. Chem. Rev. 1995, 140, − II Small-core pseudopotentials and correlation consistent basis sets for 37 114. the post-d group 16−18 element. J. Chem. Phys. 2003, 119, 11113− (49) Nguyen, T. N. T.; Thiel, N. O.; Pape, F.; Teichert, J. F. Org. Lett. 11123. 2016, 18, 2455−2458. (50) Donald, W. A.; Khairallah, G. N.; O’Hair, R. A. J. The Effective Temperature of Ions Stored in a Linear Quadrupole Ion Trap Mass Spectrometer. J. Am. Soc. Mass Spectrom. 2013, 24, 811−815.

2399 DOI: 10.1021/acs.inorgchem.6b02145 Inorg. Chem. 2017, 56, 2387−2399

149

150 7 Structure of the Ligated Ag60 Nanoparticle

[{Cl@Ag12}@Ag48(dppm)12] (where dppm = bis(diphenylphosphino)methan e)

151 152 CHINESE JOURNAL OF CHEMICAL PHYSICS MARCH 18, 2019

ARTICLE

Structure of the Ligated Ag60 Nanoparticle [{Cl@Ag12}@Ag48(dppm)12] (where dppm=bis(diphenylphosphino)methane)†

Athanasios Zavrasa, Antonija Mravakb, Margarita Bu˘zan˘cićb, Jonathan M. Whitea∗, Vlasta Bona˘cić-Kouteckýb,c, Richard A. J. O’Haira∗ a. School of Chemistry and Bio21 Molecular Science and Biotechnology Institute, University of Mel- bourne, 30 Flemington Rd, Parkville, Victoria 3010, Australia b. Center of Excellence for Science and Technology-Integration of Mediterranean Region (STIM) at Interdisciplinary Center for Advanced Sciences and Technology (ICAST), University of Split, Polji˘cka cesta 35, Split 21000, Croatia c. Chemistry Department, Humboldt University of Berlin, Brook-Taylor-Strasse 2, Berlin 12489, Germany

(Dated: Received on December 21, 2018; Accepted on January 26, 2019)

A novel bisphosphine ligated Ag60 nanocluster, [{Cl@Ag12}@Ag48(dppm)12], has been discovered and characterized by X-ray crystallography. It consists of a central chloride located inside an icosahedral silver core layer, which is further encased by a second shell of 48 silver atoms/ions, which are capped with 12 bis(diphenylphosphino)methane (dppm) ligands. Due to lack of suﬃcient material the cluster could not be further characterized by other methods. + DFT calculations were carried out on the cation [{Cl@Ag12}@Ag48(dppm)12] to determine if it corresponds to a superatom with a core count of n=58. The DFT optimized structure is in agreement with X-ray ﬁndings, but the low value of the HOMO-LUMO gap does not support superatom stability. Key words: Nanoparticle, DFT calculations, X-ray crystallography

I. INTRODUCTION [9], [Ag15(Ntriphos)4(Cl4)](NO3)3 [10]; (ii) n=18, 4− Ag44(SR)30 and it selenium variants [11]. We have been using an approach the blends elec- Coinage metal nanoclusters (CMNCs) and nanopar- trospray ionization mass spectrometry to direct the ticles (CMNPs) continue to attract attention for their bulk synthesis of CMNCs, X-ray crystallography, neu- structure and bonding arrangements, for their opti- tron diffraction and NMR spectroscopy for structural cal properties and roles in catalysis [1]. A key re- characterization, and multistage mass spectrometry quirement is that surface bound ligands surround the (MS ) experiments in conjunction with DFT calcula- metal core and prevent further growth into colloids n tions to examine the gas-phase chemistry of CMNCs or metal films [2]. The main classes of ligands are [12]. During our studies we serendipitously discovered thiolate (and related anionic S and Se based ligands) a novel bisphosphine ligated Ag nanocluster, that and phosphine and these give rise to a diverse range 60 may correspond to a superatom with a core count of of structures of ligand protected CMNCs and CMNPs n=58. Thus after publishing our work on [Ag ( - [3]. A key breakthrough in rationalizing their elec- 3 µ3 H)(BH )(dppm) ]BF , 1 [12e], we noticed that an NMR tronic structures has been the concept of superatoms 4 3 4 tube used to characterize this sample contained a crys- [4] that have closed shells with electron counts for the tal with a different colour (orange) and morphology metallic core of n=2, 8, 18, 34, 58, 92, 138 ··· [5]. to 1. X-ray crystallography analysis revealed this to While there are many examples of CMNCs and CM- be a new nanocluster, [{Cl@Ag }@Ag (dppm) ], 2, NPs [6], fewer crystal structures for superatom lig- 12 48 12 whose structure is described in this work. Due to the and protected silver CMNCs and CMNPs have been disorder in the crystal, the presence of potential coun- determined and these include those associated with + teranion(s) could not be established. Thus DFT cal- electron counts of: (i) n=8, [Ag21{S2P(OiPr)2}12] { } { } + culations are carried out on the even electron cation, [7], [Ag20 S2P(OR)2 12] [8], [AuAg20 Se2P(OEt)2 12] + [{Cl@Ag12}@Ag48(dppm)12] , 3.

II. METHODS †Part of the special issue for “the 19th International Symposium A. Compound 2 on Small Particles and Inorganic Clusters”. ∗Authors to whom correspondence should be addressed. E-mail: A 100 mg sample of [Ag3(µ3-H)(BH4)(dppm)3]BF4, [email protected], [email protected] 1 [12e], dissolved in 0.6 mL of deuteroacetonitrile and

DOI:10.1063/1674-0068/cjcp1812285 ⃝c 2019 Chinese Physical Society

153 Chin. J. Chem. Phys. Athanasios Zavras et al.

+ previously characterized via NMR experiments that in- [{Cl@Ag12}@Ag48(dppm)12] nanoparticle using the volved heating the sample from −15 ◦C to +25 ◦C in X-ray structure as the starting geometry. In addition, the NMR probe, was allowed to stand in the capped for the subunit Ag12-Cl, X-ray data have been taken NMR tube wrapped in aluminium foil at room temper- to achieve faster geometry optimization. The Cartesian ature for around 4 months. After this time we noticed coordinates of the optimized structure are given in the that orange crystals of 2 had formed. Crystals of 2 that supplementary materials. were embedded in paraffin oil and immediately mounted on an Oxford Diffraction SuperNova CCD diffractome- III. RESULTS AND DISCUSSION ter were stable for at least 3 h under the paraffin oil. { } In contrast, crystals that were removed from the NMR A. X-ray crystallography analysis of [ Cl@Ag12 @Ag48 tube and left exposed to air changed colour from or- (dppm)12] ange to black over the course of an hour and within a 1. Overview day had decomposed to unknown species. Thus crystals of 2 are stable when surrounded by solvent. Due Results obtained via X-ray crystallography reveal to the large voids in the crystals, there are likely to a Ag60 nanoparticle consisting of a central chloride be two effects that cause crystals of 2 to be unstable located inside an icosahedral silver core layer, which is when removed from solvent and exposed to air. The further encased by a shell of 48 silver atoms/ions, which first involves slow evapouration of the stabilizing sol- are capped with 12 bis(diphenylphosphino)methane { } vent molecules present in the crystals, which causes the ligands [ Cl@Ag12 @Ag48(dppm)12] (FIG. 1), many of crystals to lose their physical integrity. The second the phenyl groups of the dppm ligand exhibited high involves chemical decomposition, whereby diffusion of thermal motion, including torsional motions about the gases such as oxygen into the voids triggers unknown C(Ar)−P bond and ‘wagging’ motions, and required reactions (presumable oxidation) that cause decompo- appropriate restraints on both positional and thermal sition of 2. parameters. The transformation of 1 to 2 results in a change in the ratio of Ag:dppm from 1:1 to 5:1. This B. Crystallography is consistent with the disordered part of the crystal, which contains at least one discernible non-coordinated Intensity data for compound 2 were collected on an dppm ligand in addition to highly disordered acetoni- Oxford Diffraction SuperNova CCD diffractometer us- trile solvent molecules, the number and nature of the ing Mo-Kα radiation, the temperature during data col- counterions (if any) could not be established due to lection was maintained at 130.0(1) K using an Oxford the extensive disorder and high thermal motion of Cryostream cooling device. The structure was solved these species. The Squeeze procedure [22] was used by direct methods and difference Fourier synthesis [13]. to obtain a model for the contents of the disordered The diagrams were produced using the CrystalMaker part of the crystal which occupied 5861 A˚3 being software [14] integrated within the WINGX suite of pro- approximately 30% of the crystal volume and con- grams [15]. tained 3135 electrons. Thus this Ag60 nanoparticle is + Crystal data for compound 2:C300H257P24Ag60Cl: structurally distinct from [Au60Se2(Ph3P)10(SeR)15] , ˚ M=11112.97, T =130.0(2) K, λ=0.71073 A, tri- [Ag60(Mo6O22)2(tBuCC)38](CF3SO3)6 and (H3O)8[S@ clinic, space group P-1, a=23.2056(5), 23.2056(5), Ag60S14(iPrS)24(CF3SO3)14(CH3OH)4(DMF)2·2CH2- ˚ ◦ ◦ 41.9198(8) A, α=83.100(2) , β=79.492(2) , Cl2] previously reported [23]. γ=61.596(2)◦, V =19584.4(8) A˚3, Z=2, D =1.885 mg M−3, µ(Mo-Kα)=3.061 mm−1, c { } F (000)=10508, crystal size 0.37 mm ×0.16 mm×010 2. Shell 1 Cl@Ag12 mm. 130172 reflections measured, 92760 independent The central icosahedral core Cl@Ag12 (FIG. 1(B)) reflections (Rint=0.0598), the final R was 0.1427 (I>3(I) 57520 data) and wR(F ) (all data) was 0.3671. is well defined compared to related Ag12 icosahedral cores in the nanoparticles Na4Ag44(p-SAr)30, 4 [11b,c] The structure has been lodged with the Cambridge − Crystallographic Data Centre (CCDC 1886656). and [Ag216S56Cl7(CCPh)98(H2O)12] , 5 [24]. The Ag−Ag distances within the icosahedral core ranged from 2.811(3) A˚ to 3.194(3) A˚ (all distances are listed C. Density functional theory in the supplementary materials). These are compa- The Turbomole 7.1 [16] program was used with rable with the averages of 2.788−2.881 A,˚ found in the PBE functional including Multipole Accelerated the series of clusters 4 with Ag12 cores missing a Resolution of Identity [17] to accelerate computation. Cl [11b,c] and are slightly shorter than Ag−Ag dis- D3 dispersion [18] has been included and the SVP tances in 5 (which possesses “short edges” in the basis set for Cl and Ag atoms [19] together with range of 2.970−2.991 A˚ and “long edges” in the range RECP for silver atoms [20] and the 3-21G basis set 3.329−3.423 A).˚ Within the core, the Ag−Cl distances for all other atoms (C, H, P) associated with the vary from 2.875(6) A˚ to 3.003(6) A,˚ which is slightly dppm ligands [21] have been used to optimise the more compact compared to 5 which has distances in

DOI:10.1063/1674-0068/cjcp1812285 ⃝c 2019 Chinese Physical Society

154 Chin. J. Chem. Phys. Structure of the Ligated Ag60 Nanoparticle

FIG. 1 X-ray crystallographic results of [{Cl@Ag12}@Ag48(dppm)12]. (A) the complete Ag60 nanoparticle, hydrogens omitted, (B) the icosahedral silver core with a diameter of ca. 5.8 A,˚ and (C) the outer silver shell ca. 11.5 A.˚

FIG. 2 X-ray crystallographic results of [{Cl@Ag12}@Ag48(dppm)12] showing the arrangement of the dppm ligands binding to the silver atoms, (A) complete cluster, (B) cluster with Cl and Ag atoms not bound to a P atom removed. Phenyl groups and H atoms removed for clarity. the range of 3.082−3.152 A.˚ The doping of Cl in the 4. Capping ligands centre of a nanocluster is unusual, but not without precedence. A recent review has highlighted the various The 12 dppm capping ligands effectively shield the roles of halide ions in the growth of colloidal inorganic surface of the 48 silver atoms of shell 2 by provid- nanocrystals, including incorporation in the nanocrys- ing 24 phosphine atoms that coordinate to 24 silver tal core [25]. There are several examples of silver CM- atoms. Three dppm ligands cap the top and another NCs and CMNPs containing Cl in the core, including: three dppm ligands cap the bottom. The remaining 6 5, which has the same central icosahedral core Cl@Ag12 dppm ligands cap as a belt around the middle of the as 2 [24]; and the chloride-ion-templated rhombic do- shell (FIG. 2). decahedron nanoclusters [Ag14(C≡CtBu)12Cl]OH [26] and [Ag8Cu6(C≡CtBu)12Cl]BF4 [27]. Readers inter- + B. DFT calculations on [{Cl@Ag12}@Ag48(dppm)12] ested in anion templating effects [28] of simple inorganic anions on the formation and properties of silver clusters Unfortunately due to the instability of crystals of 2 in are directed to an excellent review [29]. the absence of solvent, we have not been able to further characterize it via mass spectrometry or other methods such as NMR, IR, and UV-Vis spectroscopy. In the ab- 3. Shell 2 {Ag } 48 sence of MS data we cannot exclude the possibility of hydrides binding to the silver atoms. Due to the disor- The second shell consists of 48 silver atoms that can der in the crystal we cannot provide a complete molecu- best be described as an irregular tessellation of trian- lar formula. Nonetheless, we were intrigued by the pos- gles, distorted squares and squares (FIG. 1(C)) dis- sibility that 2 might be a superatom ligand protected playing a range of Ag-Ag distances which can be di- silver nanoparticle. To fulfil this criterion, it would need vided into “short edges” and “long edges” with ranges of to be a monocation. Thus we used DFT calculations to 2.753(3)−3.246(3) A˚ and 3.272(3)−3.792(3) A˚ respec- optimize [{Cl@Ag }@Ag (dppm) ]+ 3 and the resul- tively (all distances are listed in the supplementary ma- 12 48 12 tant structure is shown in FIG. 3. Shell 1 {Cl@Ag } is terials). The Ag−Ag distances between the ‘inner’ and 12 maintained with Ag-Ag distances of 2.76−3.05 A˚ that ‘outer’ shells ranged from 2.812(3) A˚ to 3.357 A.˚

DOI:10.1063/1674-0068/cjcp1812285 ⃝c 2019 Chinese Physical Society

155 Chin. J. Chem. Phys. Athanasios Zavras et al.

FIG. 3 Fully optimised structure of the cation, + [Cl@Ag12@Ag48(dppm)12] .

FIG. 4 HOMO and LUMO of the cation, are somewhat shorter than those found in the crystal [Cl@Ag @Ag (dppm) ]+. structure (2.811(3)−3.194(3) A).˚ The second shell and 12 48 12 arrangements of the bisphosphine ligands are also maintained. termine electronic properties. Nonetheless, this work A key requirement for superatoms is that they have highlights that the use of phosphine based ligands opens well defined large HOMO−LUMO gaps. FIG. 4 shows new routes towards the formation of novel ligated silver that the HOMO is located over 3 of the 12 Ag atoms quantum clusters which have to be stabilized in order of the first shell and 10 of the 48 Ag atoms of the to form new materials for different applications. second shell (red color). No atoms of the first shell are associated with the LUMO, which involves 10 of the 48 Ag atoms of the second shell (blue color). Al- though the calculated HOMO-LUMO gap (0.2 eV) is V. ACKNOWLEDGMENTS underestimated using the PBE functional, it seems to be lower than in the case of ligated gold clusters R. A. J. O’Hair acknowledges funding from the of similar sizes. This suggests that the stability of Australian Research Council (No.DP150101388 and + [{Cl@Ag12}@Ag48(dppm)12] is likely to be low rela- No.DP180101187). This research was partially sup- tive to other superatom CMNCs and CMNPs. ported by the project STIM-REI, Contract Number: KK.01.1.1.01.0003, funded by the European Union through the European Regional Development Fund–the IV. CONCLUSION Operational Programme Competitiveness and Cohesion 2014-2020 (KK.01.1.1.01). Vlasta Bona˘cić-Koutecký, This study has revealed that the trinuclear nanoclus- Margarita Bu˘zan˘cić,and Antonija Mravak acknowledge ter 1 can transform to the Ag60 nanoparticle 2. While computational facilities of the supercomputer “Bura” at the mechanism of this transformation is unknown, we the University of Rijeka and SRCE at University of Za- previously noted that at the highest NMR tempera- greb as well as Prof. Miroslav Radman at MedILS and ture used (+25 ◦C), 1 had undergone decomposition Split-Dalmatia County for support. with liberation of H2. This suggests reduction of Ag(I) to Ag(0) triggers growth of the nanoparticle. Dur- ing this growth stage there is a change in the ratio of Ag:dppm from 1:1 to 5:1. Characterization of [1] R. Jin, C. Zeng, M. Zhou, and Y. Chen, Chem. Rev. 2 by X-ray crystallography and DFT calculations on 116, 10346 (2016). [{Cl@Ag }@Ag (dppm) ]+ reveals its structural and [2] J. Yan, B. K. Teo, and N. Zheng, Acc. Chem. Res. 51, 12 48 12 3084 (2018). electronic properties. In spite of the core count of n=58, [3] (a) T. Higaki, Q. Li, M. Zhou, S. Zhao, Y. Li, S. Li, this ligated silver cluster has a low HOMO-LUMO gap and R. Jin, Acc. Chem. Res. 51, 2764 (2018). and is thus not very stable relative to other superatom (b) S. Sharma, K. K. Chakrahari, J. Y. Saillard, and C. CMNCs and CMNPs. Thus simple electron counting W. Liu, Acc. Chem. Res. 51, 2475 (2018). is not sufficient to establish if a nanoparticle is a su- (c) Q. Tang, G. Hu, V. Fung, and D. Jiang, Acc. Chem. peratom and DFT calculations are mandatory to de- Res. 51, 2793 (2018).

DOI:10.1063/1674-0068/cjcp1812285 ⃝c 2019 Chinese Physical Society

156 Chin. J. Chem. Phys. Structure of the Ligated Ag60 Nanoparticle

[4] A. C. Reber and S. N. Khanna, Acc. Chem. Res. 50, Donnelly, and R. A. J. O’Hair, Inorg. Chem. 55, 9858 255 (2017). (2016). [5] (a) M. Walter, J. Akola, O. Lopez-Acevedo, P. D. (h) J. Li, H. Z. Ma, G. E. Reid, A. J. Edwards, Y. Hong, Jadzinsky, G. Calero, C. J. Ackerson, R. L. Whetten, J. M. White, R. J. Mulder, and R. A. J. O’Hair, Chem. H. Gronbeck, and H. Häkkinen,Proc. Nat. Acad. Sci., Eur. J. 24, 2070 (2018). 105, 9157 (2008). (i) H. Z. Ma, J. Li, A. J. Canty, and R. A. J. O’Hair, (b) H. Häkkinen,Chem. Soc. Rev. 37, 1847 (2008). Dalton Trans. 46, 14995 (2017). (c) S. Kaappa, S. Malola, and H. Häkkine,J. Phys. (j) H. Z. Ma, J. M. White, R. J. Mulder, G. E. Reid, A. Chem. A 122, 8576 (2018). J. Canty, and R. A. J. O’Hair, Dalton Trans. 47, 14713 (d) C. M. Aikens, J. Phys. Chem. Lett. 2, 99 (2011). (2018). (e) W. W. Xu, X. C. Zeng, and Y. Gao, Acc. Chem. [13] G. Sheldrick, Acta Crystallogr. Section C, 71, 3 (2015). Res. 51, 2739 (2018). [14] D. C. Palmer, Crystal Maker, Begbroke, England: [6] D. M. P. Mingos, Gold Clusters, Colloids and Nanopar- CrystalMaker Software Ltd., (2014). ticles, New York, USA: Springer, (2014). [15] (a) L. J. Farrugia, J. Appl. Cryst. 30, 565 (1997). [7] R. S. Dhayal, J. H. Liao, Y. C. Liu, M. H. Chiang, S. (b) L. J. Farrugia, J. Appl. Cryst. 32, 837 (1999). Kahlal, J. Y. Saillard, and C. W. Liu, Angew. Chem. [16] TURBOMOLE V7.1 2016, a development of University Int. Ed. 54, 3702 (2015). of Karlsruhe and Forschungszentrum Karlsruhe GmbH, [8] R. S. Dhayal, Y. R. Lin, J. H. Liao, Y. J. Chen, Y. C. 1989-2007, TURBOMOLE GmbH, since 2007; available Liu, M. H. Chiang, S. Kahlal, J. Y. Saillard, and C. W. from http://www.turbomole.com Liu, Chem. Eur. J. 22, 9943 (2016). [17] (a) J. P. Perdew, K. Burke, and M. Ernzerhof, Phys. [9] W. T. Chang, P. Y. Lee, J. H. Liao, K. K. Chakrahari, Rev. Lett. 77, 3865 (1996). S. Kahlal, Y. C. Liu, M. H. Chiang, J. Y. Saillard, and (b) M. Sierka, A. Hogekamp, and R. Ahlrichs, J. Chem. C. W. Liu, Angew. Chem. Int. Ed. 56, 10178 (2017). Phys. 118, 9136 (2003). [10] X. T. Shen, X. L. Ma, Q. L. Ni, M. X. Ma, L. C. Gui, [18] S. Grimme, J. Antony, S. Ehrlich, and H. Krieg, J. C. Hou, R. B. Hou, and X. J. Wang, Nanoscale 10, 515 Chem. Phys. 132, 154104 (2010). (2018). [19] (a) A. Schäfer,H. Horn, and R. Ahlrichs, J. Chem. [11] (a) K. M. Harkness, Y. Tang, A. Dass, J. Pan, N. Phys. 97, 2571 (1992). Kothalawala, V. J. Reddy, D. E. Cliffel, B. Demeler, (b) F. Weigend and R. Ahlrichs, Phys. Chem. Chem. F. Stellacci, O. M. Bakr, and J. A. McLean, Nanoscale Phys. 7, 3297 (2005). 4, 4269 (2012). [20] D. Andrae, U. Haeussermann, M. Dolg, H. Stoll, and (b) A. Desireddy, B. E. Conn, J. Guo, B. Yoon, R. N. H. Preuss, Theor. Chim. Acta 77, 123 (1990). Barnett, B. M. Monahan, K. Kirschbaum, W. P. Grif- [21] (a) J. S. Binkley, J. A. Pople, and W. J. Hehre, J. Am. fith, R. L. Whetten, U. Landman, and T. P. Bigioni, Chem. Soc. 102, 939 (1980). Nature 501, 399 (2013). (b) M. S. Gordon, J. S. Binkley, J. A. Pople, W. J. (c) H. Yang, Y. Wang, H. Huang, L. Gell, L. Lehto- Pietro, and W. J. Hehre, J. Am. Chem. Soc. 104, 2797 vaara, S. Malola, H. Häkkinen,and N. Zheng, Nature (1983). Comm. 4, 2422 (2013). [22] A. L. Spek, Acta Cryst. C 71, 9 (2015). (d) I. Chakraborty, W. Kurashige, K. Kanehira, L. [23] (a) Y. Song, F. Fu, J. Zhang, J. Chai, X. Kang, P. Li, Gell, H. Häkkine,Y. Negishi, and T. Pradeep, J. Phys. S. Li, H. Zhou, and M. Zhu, Angew. Chem. Int. Ed. 54, Chem. Lett. 4, 3351 (2013). 8430 (2015). [12] (a) A. Zavras, G. N. Khairallah, T.U. Connell, J. M. (b) J. Qiao, K. Shi, and Q. M. Wang, Angew. Chem. White, A. J. Edwards, P. S. Donnelly, and R. A. J. Int. Ed. 49, 1765 (2010). O’Hair, Angew. Chem. Int. Ed. 52, 8391 (2013). (c) Z. Wang, H. F. Su, X. P. Wang, Q. Q. Zhao, C. H. (b) A. Zavras, G. N. Khairallah, T. U. Connell, J. M. Tung, D. Sun, and L. S. Zheng, Chem. Eur. J. 24, 1640 White, A. J. Edwards, R. J. Mulder, P. S. Donnelly, (2018). and R. A. J. O’Hair, Inorg. Chem. 53, 7429 (2014). [24] Z. Y. Chen, D. Y. S. Tam, and T. C. W. Mak, Nanoscale (c) A. J. Clark, A. Zavras, G. N. Khairallah, and R. A. 9, 8930 (2017). J. O’Hair, Int. J. Mass Spectrom. 378, 86 (2015). [25] S. Ghosh and L. Manna, Chem. Rev. 118, 7804 (2018). (d) A. Daly, M. Krsti˘c,A. Giuliani, R. Antoine, L. Na- [26] D. Rais, J. Yau, D. M. P. Mingos, R. Vilar, A. J. P. hon, A. Zavras, G. N. Khairallah, V. Bona˘cić-Koutecký, White, and D. J. Williams, Angew. Chem. Int. Ed. 40, P. Dugourd, and R. A. J. O’Hair, Phys. Chem. Chem. 3464 (2001). Phys. 17, 25772 (2015). [27] T. Connell, S. Sandanayake, G. N. Khairallah, J. M. (e) A. Zavras, A. Ariafard, G. N. Khairallah, J. M. White, R. A. J. O’Hair, P. S. Donnelly, and S. J. White, R. J. Mulder, A. J. Canty, and R. A. J. O’Hair, Williams, Dalton Trans. 42, 4903 (2013). Nanoscale 7, 18129 (2015). [28] R. Vilar, Angew. Chem. Int. Ed., 42, 1460 (2001). (f) M. Krsticć, A. Zavras, G. N. Khairallah, P. [29] Q. M. Wang, Y. M. Lin, and K. G. Liu, Acc. Chem. Dugourd, V. Bona˘cić-Koutecký,R. A. J. O’Hair, Int. Res. 48, 1570 (2015). J. Mass Spectrom. 413, 97 (2017). (g) J. Li, J. M. White, R. J. Mulder, G. E. Reid, P. S.

DOI:10.1063/1674-0068/cjcp1812285 ⃝c 2019 Chinese Physical Society

157

158 8 Conclusions

This thesis investigated coinage metal hydrides as reactive intermediates and their

significance to nanoparticle synthesis. MS provided the foundations to: (i) examine solutions

containing coinage metal hydrides and to monitor experimental conditions such as solvent,

temperature, ratios of ligands and metal salts etc. to successfully direct the condensed phase

synthesis of isolable material, and (ii) use a powerful combination of gas-phase techniques to

examine the structure and reactivity of gas-phase coinage metal hydrides.

Ph The first silver hydride/borohydride nanocluster, [Ag3(μ3-H)(μ3-BH4)L 3]BF4, was isolated

following a mass spectrometry directed approach. Gas-phase studies in combination with DFT

calculations allowed mechanisms of reactivity/degradation to be proposed. Although stable in

solid form, challenges remained in stabilising the nanocluster in solution to harness its

potential as a reagent for developing new organic transformations. This metastable cluster

was also highlighted as a key intermediate in the nanocluster to nanoparticle transformation

to yield the novel Ag60 nanoparticle, [{Cl@Ag12}@Ag48(dppm)12]. Further work is needed to

examine the reaction conditions in reproducing the nanocluster to nanoparticle synthesis to

allow the preparation of stable and monodisperse silver nanoparticles with potentially exciting

optical properties.

For the catalytic reactivity of binuclear silver hydride with formic acid, the reaction environment

+ of [LAg2H] was extensively examined using theory and a suite of gas-phase techniques, which

highlighted that the active site is sensitive to subtly changes in the ligand. Thus reactivity could

be “switched on” and fine-tuned with molecular precision to provide a reactive site with

increased reactivity toward the selective dehydrogenation of formic acid. This work highlighted

the importance of the steric and electronic effects that ligands play in altering the reactivity of

the active site. When larger nuclearity silver hydrides were used no reaction was observed,

further highlighting the importance of the cluster nuclearity and its reactivity. Although the

dehydrogenation of formic acid was observed in the condensed-phase when the same

159 stoichiometric combination of silver salt, ligand and formate were used, attempts to isolate the

+ formate intermediate [LAg2H] of the gas-phase catalytic cycle proved challenging and resulted in larger nuclearity silver formate structure embedded in a cocrystal. The development of a robust and practically applicable system that is structurally well characterized in the condensed-phase remains a challenge. Innovative approaches to harness this selective pathway when used as a heterogenous catalyst have been reported wherein the reactive intermediates are confined within a porous zeolite framework.1

Copper hydride anions were explored for their reactivity in the transformation of organic substrates. Homocoupling and protonation reactions were examined and a difference in

- - reactivity for CuH2 and Cu2H3 was observed. The reaction rates and theoretical calculation

- rationalised the observed reactivities of these species in which CuH2 performed best. These reactions may have the potential to provide isolable intermediates and further work is needed to develop this.

Rational catalyst design may often prove challenging where reactive intermediates are elusive.

MS provides a means to harness otherwise elusive compounds and to thoroughly examine their reactivity and structure in the gas-phase. The knowledge acquired can then be used to design catalysts with atomic and molecular precision around the active site of these catalysts since the steric and electronic effects of ligands on metals can be readily deduced.

1. Krstić, M., Jin, Q., Khairallah, G. N., O'Hair, R. A. J., & Bonačić-Koutecký, V., ChemCatChem 2018, 10(5), 1173-1177.

160 161 162 9 Appendices

163

164 9.1 Appendix A - Supplementary material for Chapter 2

165 166 Electronic Supplementary Material (ESI) for Nanoscale. This journal is © The Royal Society of Chemistry 2015

Supporting Information

Synthesis, Structure and Gas-Phase Reactivity of the Mixed Silver Hydride Ph Ph Borohydride Nanocluster [Ag3(µ3-H)(µ3-BH4)L 3]BF4 (L = bis(diphenylphosphino)methane).

A. Zavras,a A. Ariafard,a-c G. N. Khairallah,a J. M. White,a Roger J. Mulder,d A. J. Canty,b and R. A. J. O’Haira,b

a. School of Chemistry and ‡Bio21 Institute of Molecular Science and Biotechnology, The University of Melbourne, Melbourne, Victoria 3010, Australia. b. School of Physical Sciences, University of Tasmania, Private Bag 75, Hobart, Tasmania 7001, Australia. c. Department of Chemistry, Faculty of Science, Central Tehran Branch, Islamic Azad University, Shahrak Gharb, Tehran, Iran. d. CSIRO Manufacturing, Bayview Avenue, Clayton, VIC 3168, Australia

167 Contents

S3 ...... Complete Citation for ref. 46 and description of conversion of NCE to voltage.

S4 ...... Figure S1

S5 ...... Figure S2

S6 ...... Figure S3

S7 ...... Figure S4

S8 ...... Figure S5

S9 ...... Figure S6

S10 ...... Figure S7

S11 ...... Figure S8

S12 ...... Figure S9

S13 ...... Figure S10

S14 ...... Figure S11

S15 ...... Figure S12

S16 ...... Figure S13

S17 ...... Figure S14

S18 ...... Figure S15

S19 ...... Figure S16

S20 ...... Figure S17

S21 ...... Figure S18

S22 ...... Figure S19

S24 ...... Figure S20

S24 ...... Figure S21

S25 ...... Figure S22

S26 ...... Cartesian coordinates and total energies for all of the calculated structures

168 Complete citation for reference 46

Gaussian 09, revision D.01, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, and D. J. Fox, Gaussian, Inc., Wallingford CT, 2009.

Description of conversion of NCE to voltage for ERCID experiments.

The NCE was converted to an amplitude of the resonance excitation RF voltage (tick amp) (eq. 13). The tick amp slope and tick amp intercept are extracted from the normalised collision energy calibration file of the most recent calibration.

Amplitude = (NCE)% ÷ 30%((parent mass)(tick amp slope) + (eq. 13) (V) tick amp intercept)

169 722.88262 100 1493.12515 1107.00317 (a) (c) (e)

0 1107.00331 100 722.88303 (b) 1493.12427 (d) (f) Relative Intensity (%)

0 1488 1492 1496 1104 1107 1110 722 724 726 m/z

Figure S1: LTQ/FTICR high resolution ESI/MS of hydride/borohydride containing trinuclear silver(I) clusters. Comparison of measured (top panel) vs. simulated isotope ratios (bottom panel) and accurate mass determination for: (a) measured HR FT-ICR MS of + + [Ag3(H)(BH4)L3] , 0.6 ppm error; (b) simulated isotope distribution of [Ag3(H)(BH4)L3] ; (c) + measured HR FT-ICR MS of [Ag3(H)(BH4)L2] , 0.1 ppm error; (d) simulated isotope + + distribution of [Ag3(H)(BH4)L2] ; (e) measured HR FT-ICR MS of [Ag3(H)(BH4)L] , 0.6 + ppm error; (f) simulated isotope distribution of [Ag3(H)(BH4)L] . L = dppm = bis(diphenylphosphino)methane. The most intense isotope peak is represented by the m/z value.

170 100 (a) 746.93356

100 (b) 746.93436 Relative Intensity (%)

0 744 746 748 750 752 m/z

Figure S2: LTQ/FTICR high resolution ESI/MS of hydride/borohydride containing trinuclear silver(I) clusters. Comparison of measured (top panel) vs. simulated isotope ratios (bottom panel) and accurate mass determination for: (a) measured HR FT-ICR MS of + + [Ag3(H)(BH4)L3] , 0.6 ppm error; (b) simulated isotope distribution of [Ag3(H)(BH4)L3] . L = dmpm = bis(dimethylphosphino)methane. The most intense isotope peak is represented by the m/z value.

171 100 11 - [ BF ] 4 80

Relative Intensity (%) 10 - [ BF ] 4 79

0 50 60 70 80 90 100 m/z Figure S3: LTQ ESI/MS in the negative ion polarity mode of an acetonitrile solution of 1. Solutions were diluted to ca. 50 µM in acetonitrile. The most intense isotope peak is represented by the m/z value.

172 100

90 [Ag (H)(BH )L ]B F 3 4 3 4 NaBH 4 NaBF 4 % Transmittance

3500 3000 2500 2000 1500 1000 500

-1 cm

Figure S4: FT-IR spectra of: [Ag3(H)(BH4)L3]BF4 shown by the black line, NaBH4 shown by the red line and NaBF4 shown by the blue line. All spectra were collected from material as fine powders after mechanical grinding in a mortar and pestle and subsequently recorded using a Perkin-Elmer Spectrum one FTIR spectrometer operating in the diffuse reflectance mode. Each spectrum is an average of 32 scans.

173 1 Figure S5: H NMR spectrum of 1 (400.1 MHz, CD3CN) at -15°C. Insets are the µ3-H (5.3 to 3.9 ppm) and µ3-BH4 (1.3 to -0.1 ppm) signals.

174 11 1 Figure S6: B-decoupled H NMR spectrum of 1 (400.1 MHz, CD3CN) at -15°C. Insets are the µ3-H (5.3 to 3.9 ppm) and µ3-BH4 (1.3 to -0.1 ppm) signals.

175 Figure S7: Overlay of the 11B-decoupled 1H (upper) and 1H (lower) NMR spectra of 1 (400.1

MHz, CD3CN) at -15°C of the µ3-BH4 (1.3 to -0.1 ppm) signals.

S10

176 31 1 Figure S8: P-decoupled H NMR spectrum of 1 (400.1 MHz, CD3CN) at -15°C. Insets are the µ3-H (5.3 to 3.9 ppm) and µ3-BH4 (1.3 to -0.1 ppm) signals.

S11

177 Figure S9: Overlay of the 31P-decoupled 1H (upper) and 1H (lower) NMR spectra of 1 (400.1

MHz, CD3CN) at -15°C of the µ3-H (5.3 to 3.9 ppm) signals.

S12

178

1 31 Figure S10: H-decoupled P NMR spectrum of 1 (162.0 MHz, CD3CN) at -15°C.

S13

179 1 11 Figure S11: H-decoupled B NMR spectrum of 1 (128.4 MHz, CD3CN) at -15°C.

S14

180 1 11 1 11 Figure S12: H- B HSQC spectrum of 1 (400.1 MHz ( H), 128.4 MHz ( B), CD3CN) at - 15°C.

S15

181

19 11 19 11 Figure S13: F- B HSQC spectrum of 1 (376.5 MHz ( F), 128.4 MHz ( B), CD3CN) at - 15°C.

S16

182 1 19 Figure S14: H-decoupled F NMR spectrum of 1 (376.5 MHz, CD3CN) at -15°C.

S17

183 1 13 Figure S15: H-decoupled C NMR spectrum of 1 (100.6 MHz, CD3CN) at -15°C.

S18

184 1 Figure S16: H NMR spectrum of 1 (400.1 MHz, CD3CN) recorded at temperatures from -15°C to +25°C; µ3-H (5.3 to 3.9 ppm)and H2 (4.56 ppm) signals.

S19

185 100 (a) (c) 1094.96964 710.84930

100 (b) (d) 710.84965 1094.97003 Relative Intensity (%)

0 1092 1095 1098 706 708 710 712 714

m/z

Figure S17: LTQ/FTICR high resolution ESI/MS of dihydride containing trinuclear silver(I) clusters. Comparison of measured (top panel) vs. simulated isotope ratios (bottom panel) and + accurate mass determination for: (a) measured HR FT-ICR MS of [Ag3(H)2L2] , 0.4 ppm + error; (b) simulated isotope distribution of [Ag3(H)2L2] ; (c) measured HR FT-ICR MS of + + [Ag3(H)2L] , 0.5 ppm error; (d) simulated isotope distribution of [Ag3(H)2L] . L = dppm = bis(diphenylphosphino)methane. The most intense isotope peak is represented by the m/z value.

S20

186

+ Figure S18: LCQ energy resolved CID of [Ag3(H)(BH4)L3] . (a) CID spectrum at 0.92 volts. (b) branching ratio of products as a function of the activation amplitude used.

S21

187 (a) + + [Ag (H)(BH )L ] 100 3 4 3 + [A g (H) L ] [A g (H) L] 3 2 2 3 2 1095 1493 709 + [A g (H)(BH )L] 3 4 + [Ag (H)(BH )L ] 723 3 4 2 50 1109

Relative Intensity (%) 0 500 1000 1500 m/z

+ 1.0 [Ag (H)(BH )L ] (1109) (b) 3 4 2 + [Ag (H) L ] (1095) 3 2 2 0.5 + [Ag (H)(BH )L] (723) 3 4 + [Ag (H) L] (709) 3 2 + [Ag (H)L] (601) 2 + 0.05 [AgL] (491) +

Branching Ratio [HL] (385) 0.00 0.4 0.6 0.8 1.0 1.2 Activation Voltage (V ) p-p

+ Figure S19: LCQ energy resolved CID of [Ag3(H)(BH4)L2] . (a) CID spectrum at 0.90 volts. (b) branching ratio of products as a function of the activation amplitude used.

S22

188 (a) 100 + [A g (H)(BH )L ] 3 4 3 + [A g (H) L] 1493 3 2

709 + * [A g (H)(BH )L] 3 4 50 723

Relative Intensiy (%) 0 600 800 1000 m/z

(b) 1.0

+ [Ag (H)(BH )L] (723) 3 4 0.5 + [Ag (H) L] (709) 3 2 Branching Ratio 0.0 0.3 0.4 0.5 0.6 Activation Voltage (V ) p-p

+ Figure S20: LCQ energy resolved CID of [Ag3(H)(BH4)2] . (a) CID spectrum at 0.46 volts. (b) branching ratio of products as a function of the activation amplitude used.

S23

189 + [Ag (H)(BH )L ] (a) 3 4 3 1493 100 + [A g (H) L] 3 2 + 80 [Ag (H)(BH )L ] 709 3 4 2 1109 60 + [Ag (H) L ] 40 3 2 2 1095 20 *

Relative Intensity (%) 0 400 600 800 1000 1200 1400 m/z

(b) + [Ag (H) L ] (1095) 1.0 3 2 2 + [Ag (H)(BH )L] (723) 3 4 + [Ag (H) L] (709) 3 2 0.5 + [Ag (H)L ] (601) 2 + [AgL] (491) + Branching Ratio [HL] (385) 0.0 0.4 0.6 0.8 Activation Voltage (V ) p-p

+ Figure S21: LCQ energy resolved CID of [Ag3(H)2L2] . (a) CID spectrum at 0.76 volts. (b) branching ratio of products as a function of the activation amplitude used.

S24

190 + (a) [A g (H)(BH )L ] 3 4 3 100 + [A g (H)L] 1493 2 + + [A g L ] [A g (H)(BH )L ] 601 3 4 2 491 1109 50 * + + [H L ] [A g (H) L] 3 2 385 # 709 #

Relative Intensity (%) 0 400 600 800 1000 m/z

(a) 1.0 + [Ag (H) L] (709) 3 2 + [Ag (H)L ] (601) 2 + 0.5 [A g L ] (491) + [H L ] (385) Branching Ratio 0.0 0.4 0.6 0.8 1.0 1.2 Activation Voltage (V ) p-p

+ Figure S22: LCQ energy resolved CID of [Ag3(H)2L] . (a) CID spectrum at 1.04 volts. (b) branching ratio of products as a function of the activation amplitude used.

S25

191 Cartesian coordinates and total energies for all of the calculated structures

Me + [Ag3(H)(BH4)L 3] E(M06/BS1) = -3109.492746 au H(M06/BS1) = -3108.855609 au G(M06/BS1) = -3108.984256 au E(M06/BS2//M06/BS1) = -3113.708582 au C -3.97746700 -0.36313700 0.79349200 H -5.06071400 -0.56258900 0.83956100 H -3.65612600 0.01562300 1.77535900 B -0.01623900 -0.02847300 2.20560000 H -1.05849000 0.53565300 1.81917100 H 0.00027600 -0.01769200 3.42388000 H -0.02278700 -0.01383300 -1.35399800 Ag -1.06204800 1.34923200 -0.32550700 P -3.59226200 0.97080100 -0.42509200 P -3.03609500 -1.93446300 0.49521000 H 0.00400400 -1.21247800 1.81790800 H 0.98892600 0.57697000 1.78816200 C 2.31496900 -3.22047900 0.86812600 H 3.04329300 -4.04521500 0.93438500 H 1.80713900 -3.12939600 1.83996700 Ag -0.63989200 -1.61381200 -0.33211900 P 1.00342900 -3.58570100 -0.38205800 P 3.17978900 -1.60684900 0.56194100 C 1.71114200 3.59430200 0.81033600 H 2.06750800 4.63661500 0.84823700 H 1.94298800 3.11680500 1.77401900 Ag 1.69206200 0.23636500 -0.38177300 P 2.60058300 2.62987200 -0.48998900 P -0.13575200 3.54306400 0.60947500 C 1.98159100 -3.85872700 -1.91465900 H 2.81740600 -4.55161000 -1.74270800 H 1.33280600 -4.27807800 -2.69269500 H 2.36650600 -2.90054600 -2.28658900 C 0.54158600 -5.30382900 0.07988200 H -0.17692100 -5.70397000 -0.64526600 H 1.41865600 -5.96416300 0.10767900 H 0.06600200 -5.30353400 1.06832400 C 4.76832500 -2.13670800 -0.19856900 H 5.42621000 -1.26411700 -0.29412300 H 5.27363200 -2.89219400 0.41806900 H 4.60415400 -2.54465100 -1.20213600 C 3.71554400 -1.14848700 2.25575900 H 4.25166400 -1.97044600 2.74862100 H 4.37728700 -0.27460700 2.19854900 H 2.83382900 -0.87070800 2.84661100 C 4.35035800 3.06317200 -0.12900600 H 5.00712400 2.62199700 -0.88833700 H 4.50421400 4.15061700 -0.11855700 H 4.63532400 2.65553600 0.84879100 C 2.26978900 3.61187300 -2.00724000 H 2.49360100 4.67750500 -1.85716000 H 2.88866900 3.23495700 -2.82991000

S26

192 H 1.21939000 3.49354400 -2.30264300 C -0.49938100 5.20563100 -0.08999700 H -1.58630100 5.34941400 -0.12807100 H -0.06304600 6.00095000 0.52907300 H -0.11262100 5.29475900 -1.11192700 C -0.69503000 3.73635600 2.34602800 H -0.21317200 4.59426500 2.83364800 H -1.78286300 3.88093600 2.36213900 H -0.46612600 2.81500000 2.89677200 C -4.78532400 2.27223300 0.08599200 H -4.73231700 3.11604200 -0.61243600 H -5.81723300 1.89675000 0.10820000 H -4.51885900 2.63903800 1.08516400 C -4.35354400 0.34476700 -1.97557500 H -5.36871300 -0.03970800 -1.80362700 H -4.40720600 1.15988100 -2.70716200 H -3.72784900 -0.44450900 -2.41073200 C -4.27905900 -2.98521500 -0.36256600 H -3.87516600 -3.99876800 -0.47347200 H -5.21592900 -3.04089400 0.20802900 H -4.49269200 -2.59729300 -1.36521800 C -3.02245800 -2.67006500 2.17655200 H -4.03207300 -2.72465100 2.60547200 H -2.60250400 -3.68262600 2.12371500 H -2.37707000 -2.06705000 2.82734800

Me + [Ag3(H)(BH4)L 2] E(M06/BS1) = -2228.000079 au H(M06/BS1) = -2227.556000 au G(M06/BS1) = -2227.659437 au E(M06/BS2//M06/BS1) = -2232.072635 au C 3.52543500 0.63553100 1.06518700 H 4.53836300 1.04914200 1.19727900 H 3.11809100 0.39356000 2.05811900 B -0.03440000 -0.92577500 2.19109600 H 1.10756600 -0.87279900 1.73871300 H -0.00810200 -0.98808400 3.39992700 H -0.23628300 -1.59905400 -1.64906500 Ag 1.37199300 -1.39535200 -0.85704800 P 3.59396400 -0.95238900 0.11909500 P 2.39794400 1.90376100 0.33540800 H -0.72275900 0.10002400 1.92976600 H -0.62276000 -1.94962400 1.77209700 C -3.61950400 1.06935800 0.56658100 H -4.60963700 1.50267600 0.35115200 H -3.36523200 1.31508600 1.60862800 Ag -0.00249100 1.14231700 0.26066400 P -2.32248100 1.85356800 -0.49234800 P -3.68005300 -0.78495500 0.43856000 Ag -1.36867200 -1.56625100 -0.14988600 C -2.85618400 1.46611200 -2.20256400 H -3.91876500 1.70247300 -2.35195600 H -2.26297000 2.06344400 -2.90503900 H -2.67395500 0.40890500 -2.43288900

S27

193 C -2.75123400 3.63165200 -0.34193300 H -2.10442300 4.22608800 -0.99782500 H -3.79805600 3.81338900 -0.62015300 H -2.59435600 3.96815700 0.68953500 C -5.17831500 -1.07728400 -0.57814000 H -5.37228200 -2.15452100 -0.63587700 H -6.05460400 -0.58513300 -0.13585000 H -5.03298000 -0.70737900 -1.59936600 C -4.22829200 -1.26656400 2.11808900 H -5.12783300 -0.71448000 2.42077500 H -4.44808200 -2.34050300 2.13600500 H -3.41988100 -1.07529800 2.83446300 C 4.10970800 -2.15233400 1.39918500 H 4.30936600 -3.12495400 0.93493000 H 5.01775100 -1.80939400 1.91205400 H 3.30147900 -2.27601800 2.12945500 C 5.09316900 -0.78680500 -0.92217400 H 5.96755400 -0.53129700 -0.30929900 H 5.28629000 -1.74056400 -1.42709300 H 4.95686300 -0.01946500 -1.69243900 C 3.22872200 2.35999900 -1.23586700 H 2.76246300 3.26730700 -1.63797900 H 4.29797900 2.55726500 -1.07723700 H 3.10646000 1.56788300 -1.98475500 C 2.75099000 3.35622300 1.39775400 H 3.82351800 3.59226100 1.39967700 H 2.19752500 4.22841500 1.03094800 H 2.42711900 3.15274000 2.42504900

Me + [Ag3(H)2L 3] E(M06/BS1) = -3082.838167 au H(M06/BS1) = -3082.236005 au G(M06/BS1) = -3082.360653 au E(M06/BS2//M06/BS1) = -3087.044329 au C 0.32077800 3.60658200 -1.30541800 H 0.56269200 4.65452300 -1.54764300 H 0.05207600 3.08814100 -2.23749500 H -0.68162100 0.38340600 -1.66917700 H -0.40724600 0.21797300 1.60544500 Ag -1.53893500 0.95984300 0.10344300 P -1.16343500 3.48804500 -0.20362100 P 1.81182700 2.74457500 -0.61277400 C 3.56395700 -1.84375700 -0.37535800 H 4.36368000 -2.58337100 -0.19899100 H 3.84549900 -1.25156400 -1.26019700 Ag 1.23470600 0.65865500 0.87829600 P 3.44651100 -0.65314100 1.04106900 P 1.97309700 -2.68717000 -0.80493500 C -3.70227300 -1.95544700 -0.26258500 H -4.50668000 -2.57878800 0.16464500 H -3.90067200 -1.84439800 -1.33993200 Ag -0.10627500 -1.25341400 -1.08134800 P -2.06705300 -2.81726200 -0.13106700 P -3.76334900 -0.24273300 0.45133000

S28

194 C 2.68673000 2.26834700 -2.16019900 H 3.69101900 1.89613500 -1.91954800 H 2.12792200 1.46313100 -2.65393000 H 2.78609100 3.11628200 -2.85145900 C 2.85147900 4.15410300 -0.03988500 H 2.39861600 4.64022900 0.83230700 H 3.83666600 3.77745200 0.26300800 H 2.98996500 4.90107300 -0.83326300 C -2.30362800 4.65938600 -1.04798700 H -2.58681800 4.26024600 -2.02899000 H -3.21715600 4.77850800 -0.45356100 H -1.84032500 5.64626900 -1.18326600 C -0.63298700 4.49104300 1.24513500 H 0.13055200 3.94651200 1.81518500 H -0.23827800 5.46958300 0.93646300 H -1.48738300 4.65208300 1.91295900 C -5.36774000 0.32896100 -0.24640200 H -5.64062200 1.29134100 0.20289600 H -5.27536200 0.47274900 -1.32911500 H -6.17110300 -0.39323700 -0.04645400 C -4.25044400 -0.54609600 2.19941200 H -3.39418500 -0.89658500 2.78593900 H -4.58391300 0.40234800 2.63803200 H -5.06996500 -1.27444200 2.27482400 C -1.93854700 -3.15510700 1.67561200 H -1.61826100 -2.24871400 2.20537200 H -2.88971900 -3.51524900 2.09302900 H -1.17695800 -3.92806700 1.84139400 C -2.54504100 -4.48676900 -0.75312900 H -2.81975600 -4.42967800 -1.81297400 H -1.69593400 -5.17577400 -0.66093700 H -3.39280200 -4.89912500 -0.18907100 C 1.72937400 -3.83788800 0.60854000 H 0.97886300 -4.58647300 0.32102400 H 2.65487800 -4.35999500 0.88914200 H 1.33599000 -3.28716900 1.47347800 C 2.53571600 -3.83950400 -2.12082300 H 3.37561600 -4.45913800 -1.77925400 H 1.70579600 -4.49507700 -2.41137600 H 2.84457300 -3.27421200 -3.00763700 C 5.05220900 0.23000000 0.86894200 H 5.18706500 0.91038400 1.71846400 H 5.89751400 -0.47095500 0.84175200 H 5.05912200 0.82869100 -0.05001400 C 3.80890700 -1.70244300 2.50761600 H 4.68520300 -2.34282900 2.33597400 H 4.01588700 -1.04835700 3.36322400 H 2.94828300 -2.32603100 2.77053600

Me + [Ag3(H)2L 2] E(M06/BS1) = -2201.358866 au H(M06/BS1) = -2200.950546 au G(M06/BS1) = -2201.049756 au E(M06/BS2//M06/BS1) = -2205.421185 au

S29

195 C -3.69818300 0.61344000 -0.67045800 H -4.69083200 1.04474900 -0.45772900 H -3.67123500 0.34775000 -1.73823700 H 0.50430200 -2.33522700 0.87657300 Ag -1.14213600 -1.69924800 0.49701000 P -3.48877300 -0.96307400 0.27340800 P -2.36871900 1.87889600 -0.42529300 H 0.44894700 -0.48196200 -1.94541600 C 3.57250000 1.27458000 0.07352800 H 4.36611600 1.69516400 0.71379200 H 3.71242000 1.68179500 -0.93941300 Ag -0.05519700 0.96620900 -0.86814000 P 1.90474800 1.85529900 0.64126600 P 3.76549500 -0.57118900 -0.06253200 Ag 1.43715800 -1.43947800 -0.57573300 C 1.80558800 1.21758900 2.36365000 H 2.70702700 1.46485500 2.94222400 H 0.93773100 1.67140900 2.85969600 H 1.65382300 0.13062800 2.36179900 C 2.24496300 3.63869400 0.93792000 H 1.35902700 4.11364500 1.37699300 H 3.09254400 3.77410600 1.62330200 H 2.46874200 4.14455600 -0.00833500 C 4.56591100 -1.01039600 1.53133100 H 4.87201100 -2.06265600 1.49972200 H 5.45300500 -0.39004800 1.71801200 H 3.86074000 -0.89515900 2.36176000 C 5.17975500 -0.68769300 -1.22701900 H 6.01896600 -0.06317600 -0.89290800 H 5.51815300 -1.72847100 -1.28964000 H 4.86842800 -0.37064800 -2.22872000 C -4.59534800 -2.09970400 -0.64187200 H -4.67105300 -3.05241000 -0.10502700 H -5.59979400 -1.66810900 -0.74385900 H -4.18603500 -2.30155500 -1.63810800 C -4.42248400 -0.67560000 1.82480000 H -5.44016400 -0.32390800 1.60860900 H -4.48781500 -1.61850400 2.38040800 H -3.91627800 0.05558900 2.46443400 C -2.60358200 2.41457900 1.31673600 H -2.03609300 3.34063300 1.47617900 H -3.65909700 2.60891600 1.55330100 H -2.20178200 1.66125700 2.00696200 C -3.06998100 3.29626000 -1.35545600 H -4.07444900 3.55089600 -0.99209300 H -2.41980900 4.17119200 -1.24070100 H -3.12544200 3.05351900 -2.42273100

Me + [Ag3(H)(BH4)L ] E(M06/BS1) = -1346.503099 au H(M06/BS1) = -1346.252796 au G(M06/BS1) = -1346.328975 au E(M06/BS2//M06/BS1) = -1350.434988 au

S30

196 C 2.77304200 0.44787800 -0.70082800 H 3.84707500 0.61820200 -0.51927800 H 2.59451000 0.58715900 -1.77755000 B -3.00308700 1.72117500 -0.55017900 H -2.49878700 2.03314500 0.55214400 H -3.68305700 2.65193700 -0.88278900 H -1.76354300 -2.07393000 0.60169200 Ag -0.62439200 1.58915600 -0.21635500 P 1.77580900 1.73168200 0.17333300 P 2.33871100 -1.30282300 -0.29713400 H -2.22756900 1.49915500 -1.51097700 H -3.84580500 0.78856700 -0.42172500 Ag -0.05758400 -1.62705800 0.17695000 Ag -2.71306200 -0.63870500 0.11998900 C 2.57892800 3.28527100 -0.35670800 H 2.10385300 4.13446000 0.14803200 H 3.64810500 3.27687400 -0.10771500 H 2.46061400 3.42276100 -1.43748000 C 2.25740400 1.57696700 1.93091900 H 3.34975700 1.53883400 2.04024200 H 1.88257500 2.45016900 2.47826000 H 1.80673900 0.68712100 2.38712900 C 3.47537900 -1.75476600 1.06441200 H 3.37805400 -2.82707300 1.27081400 H 4.51608800 -1.54519300 0.78429300 H 3.23211000 -1.21237200 1.98442100 C 3.01622600 -2.22316400 -1.72417000 H 4.07510000 -1.97656000 -1.87547300 H 2.92754200 -3.30023300 -1.54094700 H 2.45515700 -1.98435300 -2.63438800

Me + [Ag3(H)2 L ] E(M06/BS1) = -1319.874473 au H(M06/BS1) = -1319.659579 au G(M06/BS1) = -1319.730716 au E(M06/BS2//M06/BS1) = -1323.792176 au C 2.57420300 0.00000700 -0.73664800 H 3.67526200 0.00001300 -0.66019700 H 2.32387200 0.00000100 -1.80881400 H -2.36307400 -1.81378300 0.01483500 Ag -0.53275500 1.75070900 -0.01074100 P 1.92552200 1.59422200 -0.05709700 P 1.92553700 -1.59421200 -0.05708800 H -2.36309000 1.81376400 0.01491800 Ag -0.53273900 -1.75070600 -0.01073800 Ag -2.69224700 -0.00001100 0.01999200 C 2.79443600 2.80749200 -1.11634200 H 2.57080200 3.82279000 -0.76902100 H 3.88047900 2.65173500 -1.07689100 H 2.45488800 2.72097600 -2.15454300 C 2.76480000 1.79666000 1.55632200 H 3.83696200 1.57353000 1.47327800 H 2.64987000 2.83740000 1.88245900 H 2.31630400 1.16017600 2.32584800

S31

197 C 2.76484300 -1.79665700 1.55631400 H 2.64993400 -2.83740500 1.88243700 H 3.83700200 -1.57351100 1.47325500 H 2.31635300 -1.16019200 2.32585700 C 2.79442800 -2.80747700 -1.11635500 H 3.88047100 -2.65171300 -1.07694100 H 2.57081200 -3.82277600 -0.76902300 H 2.45484500 -2.72096900 -2.15454600

+ [Ag3(H)(BH4)] E(M06/BS1) = -464.9514358 au H(M06/BS1) = -464.896099 au G(M06/BS1) = -464.945194 au E(M06/BS2//M06/BS1) = -468.728454 au B -1.61211900 0.10699200 -0.00489400 H -2.16990200 -0.32646200 -1.03259000 H -1.19496000 1.22137900 -0.32689900 H 2.43333200 1.11782800 -0.06039000 Ag -3.95820800 -0.19732000 -0.00008400 H -2.24649000 0.29880400 1.05155600 H -0.82921000 -0.80479600 0.29407400 Ag 3.46187400 -0.33977700 -0.00008400 Ag 0.75309700 0.49365700 0.00226800

LMe E(M06/BS1) = -881.4273069 au H(M06/BS1) = -881.235789 au G(M06/BS1) = -881.285286 au E(M06/BS2//M06/BS1) = -881.5738711 au C 0.00312500 -0.73330400 0.21312300 H 0.02496600 -0.55105100 1.30301000 H -0.08841400 -1.82033900 0.06500200 P 1.60005000 -0.22672900 -0.60293900 P -1.51487300 0.04014800 -0.55846900 C 2.76542600 -0.91358000 0.67191400 H 3.79286800 -0.60506400 0.44095300 H 2.51732700 -0.57276300 1.68770600 H 2.73496900 -2.01000900 0.65393000 C 1.67505000 1.54994600 -0.06831800 H 1.42201600 1.67762900 0.99508700 H 2.69123100 1.93075000 -0.23174800 H 0.99595300 2.16149400 -0.67515400 C -1.78265900 1.44741400 0.62506000 H -2.77811600 1.87600600 0.45220500 H -1.72101100 1.12070800 1.67369900 H -1.04700500 2.24374800 0.46185500 C -2.77624600 -1.11506200 0.16918500 H -2.63021100 -1.25381900 1.25037100 H -3.78492300 -0.71741500 0.00085400 H -2.71546500 -2.09365600 -0.32244800

BH3

S32

198 E(M06/BS1) = -26.582141 au H(M06/BS1) = -26.551922 au G(M06/BS1) = -26.574995 au E(M06/BS2//M06/BS1) = -26.594077 au E(B3LYP-D3BJ/BS2//M06/BS1) = -26.625207 au B 0.00015600 0.00001000 -0.00003800 H -0.60766600 -1.02353700 0.00006300 H -0.58345700 1.03747200 0.00006300 H 1.19034500 -0.01398600 0.00006300

Ph + [Ag3(H)(BH4)L 3] E(M06/BS1) = -5408.503573 au H(M06/BS1) = -5407.186337 au G(M06/BS1) = -5407.398341 au E(M06/BS2//M06/BS1) = -5413.533698 au E(B3LYP-D3BJ/BS2//M06/BS1) = -5416.591986 au C 3.68616500 -1.26909300 0.99319100 H 4.73727500 -1.58118000 1.08180400 H 3.22382100 -1.31528500 1.99024600 B -0.01174500 0.05997400 2.42879100 H 0.60064700 -0.92112200 1.96641200 H 0.10944700 0.02953700 3.64241500 H -0.18314400 0.04741100 -1.12840100 Ag 0.26851600 -1.61287500 -0.15355400 P 2.68759200 -2.38501800 -0.09298800 P 3.55578000 0.50306400 0.43824400 H 0.47855000 1.12213100 2.01083200 H -1.22982300 -0.01338300 2.16571600 C -0.70620700 3.91417200 1.13768300 H -0.91230300 4.98564200 1.28210700 H -0.37137500 3.48589300 2.09484500 Ag 1.16606800 1.14814000 -0.19344100 P 0.68874000 3.63340000 -0.04370100 P -2.22933300 2.95611900 0.69661900 C -3.01246900 -2.73664000 1.06209500 H -3.77919200 -3.52442500 1.09966900 H -2.88059000 -2.33314100 2.07670300 Ag -1.74578700 0.53699100 0.08171100 P -3.50855800 -1.31498500 -0.01542400 P -1.36390200 -3.42623400 0.53515200 C 4.10294900 1.42079000 1.92423300 C 5.13202000 2.36496900 1.87870200 C 3.38614500 1.24264900 3.11530300 C 5.46195100 3.09489500 3.01673900 H 5.67917300 2.53485800 0.95209300 C 3.72583700 1.97012600 4.24934400 H 2.53907900 0.55464500 3.15850200 C 4.76894800 2.89193400 4.20506100 H 6.26675200 3.82711400 2.96959800 H 3.16845000 1.81552300 5.17163600 H 5.03493600 3.45936100 5.09567100 C 4.91370300 0.68392400 -0.76707200

S33

199 C 6.23494800 0.33434100 -0.46269200 C 4.61259400 1.17679400 -2.03830900 C 7.23501200 0.47805800 -1.41509900 H 6.48593400 -0.03478500 0.53307700 C 5.61482700 1.31688100 -2.99453500 H 3.58077500 1.44459900 -2.27819300 C 6.92370500 0.96677600 -2.68296200 H 8.26140700 0.21057300 -1.16996400 H 5.37075200 1.69652300 -3.98509800 H 7.70823900 1.07827700 -3.42946700 C 2.02529800 4.68113300 0.63308400 C 3.00677100 5.15915200 -0.24306200 C 2.15791500 4.92253400 2.00299300 C 4.08622300 5.88848400 0.24140300 H 2.91551800 4.97092400 -1.31435100 C 3.24087900 5.65234500 2.48351100 H 1.42233500 4.54001700 2.71032900 C 4.20143900 6.14238400 1.60542100 H 4.83529700 6.26800700 -0.45151300 H 3.33285500 5.83526200 3.55287900 H 5.04404600 6.71862800 1.98391100 C 0.19155600 4.47891200 -1.58253800 C -0.20035900 5.82191600 -1.60310300 C 0.20808800 3.74891700 -2.77314200 C -0.57740900 6.42007300 -2.79796700 H -0.20016700 6.40627600 -0.68154600 C -0.16204000 4.35224500 -3.97271100 H 0.50662400 2.69815600 -2.75247200 C -0.55594900 5.68546700 -3.98280400 H -0.88227800 7.46495200 -2.80979700 H -0.14405400 3.77920000 -4.89813400 H -0.84653600 6.15967900 -4.91873500 C -3.09858000 3.99456300 -0.53062800 C -3.58812500 5.26678400 -0.21721200 C -3.28137300 3.48410900 -1.81734200 C -4.24207600 6.01929500 -1.18447000 H -3.47015000 5.66357600 0.79275800 C -3.93960700 4.23776900 -2.78507100 H -2.90276300 2.48657200 -2.05650700 C -4.41648600 5.50496600 -2.46810500 H -4.62547500 7.00754700 -0.93649800 H -4.07747100 3.83420600 -3.78678500 H -4.93405400 6.09482300 -3.22268200 C -3.26048400 3.10490400 2.19757000 C -4.43262900 2.34027700 2.23163400 C -2.94391900 3.91561500 3.28906000 C -5.27616000 2.39666000 3.33337400 H -4.69208900 1.69636600 1.38727500 C -3.78504000 3.96088400 4.39773500 H -2.03760200 4.52013300 3.28607100 C -4.95037100 3.20350500 4.42099400 H -6.18535800 1.79773500 3.34223300 H -3.52568000 4.59190500 5.24594100 H -5.60519100 3.23854900 5.28972900 C -5.21514800 -0.92215300 0.50530500 C -5.93289000 -0.02737900 -0.29999800

S34

200 C -5.78213700 -1.36735000 1.70045200 C -7.19897800 0.39976600 0.07666000 H -5.49006600 0.34122000 -1.22855600 C -7.05182400 -0.93415200 2.07829300 H -5.24099700 -2.05498900 2.34890700 C -7.76137100 -0.05375000 1.26935600 H -7.74955200 1.09057400 -0.55909800 H -7.48646300 -1.29246400 3.00997400 H -8.75356200 0.28117700 1.56601400 C -3.72825800 -2.06346900 -1.66957200 C -4.58016600 -3.15373200 -1.87967800 C -3.00471800 -1.53931200 -2.74350200 C -4.68569800 -3.72478000 -3.14074900 H -5.16037300 -3.56280000 -1.05070300 C -3.12232200 -2.10430100 -4.01147800 H -2.33551100 -0.69210500 -2.57678400 C -3.95594600 -3.19976600 -4.20676900 H -5.33945000 -4.58128400 -3.29560900 H -2.56018300 -1.68799400 -4.84571300 H -4.04597100 -3.64606500 -5.19589900 C -0.76322500 -4.27052000 2.04131100 C -0.34249400 -5.60283100 2.03515800 C -0.59617200 -3.50266400 3.20174700 C 0.19757900 -6.16969100 3.18592200 H -0.43202800 -6.20325600 1.13025700 C -0.05994500 -4.07609700 4.34754400 H -0.86003000 -2.44351500 3.21016700 C 0.33105900 -5.41297100 4.34436800 H 0.51382700 -7.21170900 3.17305300 H 0.05641500 -3.47126400 5.24499600 H 0.74821300 -5.86129100 5.24454100 C -1.80201900 -4.74379100 -0.64796600 C -2.65540100 -5.80229300 -0.31308500 C -1.28704500 -4.66511400 -1.94377400 C -2.97893200 -6.76738900 -1.25848200 H -3.05671900 -5.88161800 0.69863000 C -1.61766300 -5.62812200 -2.89282600 H -0.64090200 -3.82661200 -2.21551700 C -2.45866700 -6.68080000 -2.54886500 H -3.63570400 -7.59260900 -0.98896100 H -1.21881700 -5.55223200 -3.90304300 H -2.71307700 -7.43801300 -3.28848700 C 3.55992900 -2.37400700 -1.69839200 C 4.90541000 -2.73875800 -1.81168000 C 2.85448200 -1.98592500 -2.84005700 C 5.53802800 -2.69951600 -3.04692800 H 5.45974600 -3.06181100 -0.92887800 C 3.48721700 -1.95733600 -4.08087600 H 1.80485000 -1.69530900 -2.74859800 C 4.82811900 -2.31079300 -4.18202200 H 6.58776400 -2.97608700 -3.12790000 H 2.93068400 -1.65919800 -4.96782800 H 5.32519700 -2.28766200 -5.15035300 C 2.98613100 -4.06566100 0.55590100 C 2.53253500 -5.13564800 -0.22660900 C 3.54936000 -4.32680100 1.80618900

S35

201 C 2.66784900 -6.44239500 0.22156300 H 2.07964600 -4.94369700 -1.20213500 C 3.67111100 -5.63869700 2.25861400 H 3.90124600 -3.51211300 2.43795600 C 3.24168900 -6.69631500 1.46588100 H 2.32660800 -7.26676500 -0.40258000 H 4.11527800 -5.83079800 3.23388600 H 3.34852400 -7.72067000 1.81896900

Ph + [Ag3(H)(BH4)L 2] E(M06/BS1) = -3760.671618 au H(M06/BS1) = -3759.774952 au G(M06/BS1) = -3759.936572 au E(M06/BS2//M06/BS1) = -3765.285290 au E(B3LYP-D3BJ/BS2//M06/BS1) = -3767.347609 au C 3.50633200 -0.16415800 1.16098600 H 4.50558400 0.20280000 1.43647600 H 3.06990300 -0.67696200 2.03106200 B -0.09044200 -1.80449600 2.00925600 H 0.97953200 -1.71979500 1.39348600 H 0.15860500 -2.12968900 3.15045600 H -0.30582100 -1.99452100 -1.90040300 Ag 1.30411100 -1.77829600 -1.11211100 P 3.58069400 -1.41426400 -0.20091700 P 2.35811700 1.23837100 0.78046400 H -0.71625900 -0.71517100 2.09309000 H -0.83078300 -2.66748900 1.49731100 C -3.57384200 0.57284400 0.92671600 H -4.52699500 1.12166300 0.92678800 H -3.23108300 0.47754500 1.96834400 Ag -0.00302300 0.46594100 0.49790900 P -2.24349400 1.51704400 0.04495300 P -3.75168400 -1.15269500 0.25755500 Ag -1.47459800 -1.98158200 -0.42908900 C -2.23596700 3.13154100 0.90147700 C -2.79627800 4.29777000 0.37754200 C -1.56663000 3.18177100 2.13237400 C -2.69705400 5.49474600 1.08157400 H -3.30047100 4.27898300 -0.58794500 C -1.47699200 4.37613600 2.83529200 H -1.09969500 2.28031800 2.54170600 C -2.04232300 5.53510200 2.30784900 H -3.13150300 6.40195600 0.66571400 H -0.95523200 4.40435400 3.79010300 H -1.96540500 6.47434300 2.85226100 C -4.54108400 -2.06863100 1.61817400 C -4.28027300 -3.43981700 1.69806100 C -5.39164800 -1.47635300 2.55800100 C -4.87344200 -4.21147100 2.69025300 H -3.59660600 -3.90226200 0.98422100 C -5.97759400 -2.24906000 3.55409400 H -5.59787600 -0.40615500 2.52364300 C -5.72182200 -3.61592900 3.61827700 H -4.66213500 -5.27732400 2.74613500

S36

202 H -6.63598700 -1.78218200 4.28415800 H -6.18056000 -4.21746600 4.40050400 C -2.89016700 1.78169900 -1.63356300 C -4.20645700 2.19587000 -1.87450700 C -2.04325900 1.52388200 -2.71503000 C -4.66088600 2.35554200 -3.17688500 H -4.88709900 2.38248000 -1.04272300 C -2.49953500 1.68775200 -4.01969800 H -1.02743200 1.16952700 -2.53092200 C -3.80648300 2.10442300 -4.24892000 H -5.68826100 2.66602600 -3.35759900 H -1.83558700 1.48102600 -4.85675000 H -4.16741600 2.22752700 -5.26842900 C -5.00819900 -1.00239300 -1.05913100 C -6.33469700 -0.64979100 -0.78765300 C -4.61316400 -1.23137600 -2.37907800 C -7.24600400 -0.51252300 -1.82701700 H -6.66364900 -0.49913100 0.24162600 C -5.52816300 -1.09532900 -3.41845600 H -3.57677500 -1.50100400 -2.59303400 C -6.84289300 -0.73548000 -3.14264000 H -8.27844700 -0.24413400 -1.61040000 H -5.21175900 -1.27235400 -4.44487700 H -7.56221400 -0.63447400 -3.95340700 C 4.24925600 -2.91123500 0.59065400 C 4.03118200 -4.12884200 -0.06270800 C 4.97711600 -2.89325000 1.78398000 C 4.54255400 -5.30902300 0.46283500 H 3.45283000 -4.15281600 -0.98860800 C 5.48051500 -4.07763700 2.31140800 H 5.15538100 -1.95820200 2.31475800 C 5.26597300 -5.28347800 1.65175400 H 4.36702400 -6.25203900 -0.05083300 H 6.04253100 -4.05744500 3.24303200 H 5.65934400 -6.20867500 2.06818600 C 4.89314500 -0.81537500 -1.31439800 C 6.22791200 -0.76554400 -0.89691300 C 4.55261500 -0.38235600 -2.59685100 C 7.20478300 -0.26859200 -1.74977300 H 6.50591200 -1.13311500 0.09216400 C 5.53247900 0.11948600 -3.44770000 H 3.51346900 -0.43386000 -2.92690700 C 6.85618700 0.17618300 -3.02381000 H 8.24350500 -0.23826600 -1.42626100 H 5.26152300 0.45958200 -4.44546500 H 7.62409300 0.55992500 -3.69311500 C 2.42836800 2.23641000 2.30834200 C 2.94984700 3.53045600 2.35306900 C 1.84986700 1.68247600 3.45927500 C 2.90790500 4.25728800 3.54007200 H 3.38707500 3.97670300 1.46041600 C 1.82243200 2.40830500 4.64349700 H 1.41507900 0.67982000 3.43228600 C 2.35158300 3.69728700 4.68480400 H 3.31750200 5.26524300 3.56876600 H 1.38362900 1.96745200 5.53683200

S37

203 H 2.32765200 4.26588400 5.61259900 C 3.18420200 2.19033800 -0.53043000 C 4.55472400 2.47594100 -0.50118800 C 2.41660700 2.62015000 -1.61631200 C 5.14596800 3.16978300 -1.54850300 H 5.16665800 2.15384100 0.34253900 C 3.00861700 3.32483700 -2.66050600 H 1.34827800 2.39669500 -1.63784800 C 4.37233600 3.59514600 -2.62751100 H 6.21361300 3.37956600 -1.52476800 H 2.40416900 3.65944500 -3.50134100 H 4.83894400 4.13942500 -3.44656700

Ph + [Ag3(H)2L 3] E(M06/BS1) = -5381.859998 au H(M06/BS1) = -5380.577766 au G(M06/BS1) = -5380.779571 au E(M06/BS2//M06/BS1) = -5386.877760 au E(B3LYP-D3BJ/BS2//M06/BS1) = -5389.905173 au C 3.22338900 1.25157600 1.49458000 H 4.09480700 1.76418500 1.92775500 H 2.61113300 0.83753500 2.30755900 H 0.72527500 -0.71182700 1.53989600 H 0.58782500 -0.34978800 -1.86190900 Ag 1.46324900 -1.24625100 -0.33328900 P 3.68397700 -0.18478000 0.41375800 P 2.17043000 2.46226900 0.54625900 C -2.51825100 2.96031300 0.49444200 H -3.44946900 3.53894500 0.39497100 H -1.89791700 3.41391700 1.28424100 Ag 0.44364400 1.32152800 -1.08188100 P -1.50559300 3.00516500 -1.05531800 P -2.88105300 1.22905900 1.04202500 C -0.64743500 -4.02630400 0.71630100 H -1.01424900 -5.05930600 0.62173300 H -0.36366900 -3.86018300 1.76538400 Ag -0.97213000 -0.39410700 0.96449000 P -1.95744000 -2.77349600 0.31518000 P 0.92722800 -3.73595800 -0.24102700 C 1.19310500 3.30836700 1.84852000 C 0.96641800 4.68823400 1.79919700 C 0.49182800 2.54047800 2.79002300 C 0.07462500 5.28705100 2.68694100 H 1.48145700 5.30516600 1.06147100 C -0.39775200 3.14175200 3.67376300 H 0.61168400 1.45555600 2.81635100 C -0.60815000 4.51922200 3.62490700 H -0.08407900 6.36352400 2.63890400 H -0.93627500 2.52960400 4.39769000 H -1.30574900 4.98877000 4.31744700 C 3.38287500 3.70029800 -0.03032700 C 4.19853200 4.42739700 0.84481900 C 3.51429000 3.89403100 -1.40669700 C 5.13018200 5.32927300 0.34587300

S38

204 H 4.09367900 4.29547300 1.92299400 C 4.44967900 4.79590300 -1.90641100 H 2.89337700 3.31479900 -2.09291500 C 5.25597200 5.51340200 -1.03008300 H 5.75930400 5.89407900 1.03160700 H 4.55281100 4.93048300 -2.98188800 H 5.98809900 6.21936200 -1.41820900 C -1.01434000 4.76167800 -1.18964000 C 0.04746300 5.04615800 -2.05533100 C -1.59192300 5.80217400 -0.45817700 C 0.51871000 6.34618700 -2.19389300 H 0.51569200 4.23620800 -2.62037500 C -1.11173500 7.10176400 -0.58793500 H -2.41285700 5.60533000 0.23122000 C -0.05761800 7.37530600 -1.45406200 H 1.34591600 6.55565700 -2.86969200 H -1.56635600 7.90488600 -0.01015500 H 0.31670000 8.39259900 -1.55150600 C -2.72400200 2.73863000 -2.39384200 C -3.93192800 3.44034700 -2.45986300 C -2.43221000 1.78230800 -3.36949400 C -4.83978100 3.17516200 -3.47660400 H -4.16613200 4.20407500 -1.71643800 C -3.33896300 1.52338000 -4.39339500 H -1.49400300 1.22450300 -3.31165000 C -4.54338500 2.21639000 -4.44411300 H -5.78006400 3.72161500 -3.52092700 H -3.10368200 0.77614700 -5.14967800 H -5.25544200 2.01402900 -5.24226000 C -4.32399400 0.73279900 0.03700300 C -5.63665500 1.07157700 0.37437700 C -4.08312200 -0.03445700 -1.10579100 C -6.69131500 0.64213700 -0.42485600 H -5.83596300 1.65703300 1.27318200 C -5.13607000 -0.45307700 -1.91201100 H -3.06014300 -0.32187100 -1.35758400 C -6.44276000 -0.11899300 -1.56576300 H -7.71394900 0.90002100 -0.15477000 H -4.92825900 -1.05586500 -2.79637800 H -7.27370400 -0.45507200 -2.18357400 C -3.52545700 1.42301800 2.73620300 C -3.61835600 0.25780100 3.50702000 C -3.87496900 2.64993800 3.30546200 C -4.05522700 0.32108800 4.82447900 H -3.34427200 -0.70858500 3.07422400 C -4.30030400 2.71181000 4.62985700 H -3.80298500 3.56981900 2.72487000 C -4.38867300 1.54963600 5.39002200 H -4.12581800 -0.59376600 5.41066600 H -4.56497200 3.67275400 5.06806300 H -4.71817700 1.60182600 6.42607400 C -3.31219200 -3.13955600 1.49167800 C -4.61524600 -2.72005700 1.18438200 C -3.04749100 -3.61497500 2.78100600 C -5.62359300 -2.78926500 2.13822800 H -4.84453300 -2.32892000 0.19166800

S39

205 C -4.06025100 -3.68197500 3.73460700 H -2.04563800 -3.94408100 3.05554800 C -5.35013200 -3.26893400 3.41693600 H -6.62887400 -2.46177200 1.87784500 H -3.83839500 -4.06706100 4.72872200 H -6.14143300 -3.32230200 4.16249200 C -2.57229400 -3.31642700 -1.31964600 C -3.47200100 -4.37493500 -1.47965200 C -2.08618800 -2.65008600 -2.45031200 C -3.88859400 -4.75238600 -2.75113700 H -3.85369400 -4.89991400 -0.60328000 C -2.49322500 -3.04135100 -3.72266100 H -1.36728600 -1.83355900 -2.33408200 C -3.40098200 -4.08568000 -3.87285400 H -4.59347800 -5.57369200 -2.86887000 H -2.09506900 -2.52999000 -4.59791100 H -3.72462400 -4.38820500 -4.86729500 C 2.13019500 -4.67691300 0.77213600 C 2.82416500 -5.79765900 0.31476000 C 2.41512900 -4.16948400 2.04780700 C 3.76914600 -6.41628700 1.13136100 H 2.62729700 -6.19514000 -0.68053800 C 3.34536900 -4.79752600 2.86373700 H 1.91866500 -3.26136000 2.39874800 C 4.02444900 -5.92502100 2.40637300 H 4.30189200 -7.29288900 0.76628100 H 3.56456000 -4.38757400 3.84849900 H 4.76297100 -6.41196600 3.04134100 C 0.71846500 -4.67979900 -1.78776900 C 0.17231700 -5.96730100 -1.83686300 C 1.12999000 -4.06687500 -2.97441800 C 0.04794100 -6.62998100 -3.05171500 H -0.14240300 -6.46702700 -0.91935300 C 1.00453100 -4.73085500 -4.19085400 H 1.53584300 -3.05364300 -2.93885300 C 0.46292100 -6.01096500 -4.22916000 H -0.37579700 -7.63219700 -3.08146100 H 1.32665400 -4.24456300 -5.11001000 H 0.36272800 -6.53257900 -5.17943500 C 4.83351800 0.55945700 -0.80260600 C 6.01600800 1.19574600 -0.40995200 C 4.51761300 0.47094800 -2.16004200 C 6.86752400 1.73836500 -1.36300400 H 6.27990400 1.25424900 0.64754600 C 5.37701200 1.00870600 -3.11638600 H 3.59015500 -0.02179600 -2.45995100 C 6.55006200 1.64022700 -2.71707100 H 7.78494700 2.23494900 -1.05187400 H 5.13153000 0.92780700 -4.17410000 H 7.22491000 2.05603000 -3.46346400 C 4.78662900 -1.20730200 1.45325300 C 5.29749000 -2.36377600 0.84969800 C 5.10600900 -0.93761400 2.78507700 C 6.14158300 -3.21176900 1.55151800 H 5.02658500 -2.60020400 -0.18201600 C 5.93838000 -1.80187300 3.49493700

S40

206 H 4.71481000 -0.04933400 3.28061400 C 6.46298000 -2.93197600 2.87851300 H 6.53655800 -4.10387300 1.06848200 H 6.18465100 -1.58115700 4.53231500 H 7.12029200 -3.60035200 3.43236600

Lph E(M06/BS1) = -1647.753043 au H(M06/BS1) = -1647.335460 au G(M06/BS1) = -1647.415099 au E(M06/BS2//M06/BS1) = -1648.173935 au E(B3LYP-D3BJ/BS2//M06/BS1) = -1649.152524 au C -0.15652900 -1.16618900 -0.32843300 H -0.01965600 -1.01525600 0.75308800 H -0.48718000 -2.20281600 -0.48187100 P 1.43122900 -0.93731700 -1.28097900 P -1.50014200 -0.04756900 -1.01543400 C -2.98624000 -1.00570700 -0.47638100 C -4.00413700 -0.48090800 0.32459500 C -3.15925000 -2.28838500 -1.01298600 C -5.14361200 -1.22983100 0.60828400 H -3.90657600 0.52216100 0.73803000 C -4.29163200 -3.03932300 -0.72442100 H -2.40075400 -2.70524100 -1.67893700 C -5.28857200 -2.51225400 0.09249000 H -5.92152400 -0.80484000 1.24132800 H -4.40090100 -4.03758600 -1.14615000 H -6.17811700 -3.09761400 0.31961700 C -1.45242200 1.33328800 0.20555700 C -1.45638600 1.14268200 1.59304100 C -1.37363700 2.63548700 -0.29091500 C -1.37620600 2.22620600 2.45821800 H -1.53502800 0.13415500 2.00205000 C -1.29088500 3.72456100 0.57368100 H -1.35673300 2.79521600 -1.36977700 C -1.29037300 3.52070000 1.94809300 H -1.38207900 2.06334000 3.53517100 H -1.21822400 4.73296000 0.16837600 H -1.22481400 4.37010700 2.62661500 C 1.75308500 0.84945900 -0.94751900 C 1.97206100 1.36079600 0.33703300 C 1.72792000 1.73045000 -2.03084300 C 2.14983300 2.72362300 0.53239100 H 1.99669900 0.68192100 1.19117300 C 1.91429400 3.09792600 -1.83844100 H 1.55606700 1.33501700 -3.03289200 C 2.12342100 3.59361500 -0.55697600 H 2.29994500 3.11411700 1.53818800 H 1.89692700 3.77409700 -2.69206600 H 2.26492300 4.66265300 -0.40220800 C 2.64746700 -1.71013800 -0.13113300 C 3.99979500 -1.36830800 -0.26749900 C 2.30296500 -2.69393400 0.79988700 C 4.97222000 -1.97690400 0.51449300

S41

207 H 4.29113500 -0.60439300 -0.99028900 C 3.27865400 -3.30929600 1.58109100 H 1.26168800 -2.98994100 0.92542700 C 4.61432500 -2.95157100 1.44366300 H 6.01645700 -1.68974300 0.39916900 H 2.98827000 -4.07056100 2.30402000 H 5.37623700 -3.43074400 2.05631200

Ph + [Ag3(H)(BH4)L ] E(M06/BS1) = -2112.843086 au H(M06/BS1) = -2112.366417 au G(M06/BS1) = -2112.471802 au E(M06/BS2//M06/BS1) = -2117.045602 au C 1.19804900 0.72864200 -1.23993400 H 2.09599000 1.36379400 -1.27561200 H 0.90555200 0.48283300 -2.27061100 B -4.56184100 -0.45392400 -0.73297900 H -4.23791100 0.23551400 0.25873100 H -5.61405600 -0.01354000 -1.10526500 H -1.84636300 -2.74206500 1.43493400 Ag -2.34489000 0.51184800 -0.68289100 P -0.20280200 1.65297600 -0.46555200 P 1.55043800 -0.87179200 -0.37385400 H -3.79336000 -0.44366400 -1.72539700 H -4.83131400 -1.65514200 -0.44645100 Ag -0.47539800 -1.86546000 0.61458000 Ag -3.28976100 -2.11208100 0.58705300 C 0.28827200 1.89051100 1.26738700 C 1.47473100 2.56566000 1.58107700 C -0.50642600 1.37191400 2.29368500 C 1.87011700 2.69274000 2.90493600 H 2.08899800 2.99382900 0.78741500 C -0.11047400 1.50800700 3.62199200 H -1.44634100 0.86707900 2.05576300 C 1.07978000 2.16076500 3.92380600 H 2.79612700 3.21085300 3.14647100 H -0.73693600 1.11087100 4.41790800 H 1.39053300 2.26807900 4.96131500 C -0.15267500 3.30395500 -1.22482800 C 0.40281100 3.52785100 -2.48725200 C -0.77452500 4.35583600 -0.54204500 C 0.34737200 4.79771900 -3.05331600 H 0.88247400 2.72074600 -3.04008800 C -0.82653500 5.62030100 -1.11331500 H -1.21144400 4.18867000 0.44367800 C -0.26547100 5.84176700 -2.36893000 H 0.78568300 4.96923100 -4.03431100 H -1.30771700 6.43586000 -0.57770200 H -0.30947900 6.83246700 -2.81630500 C 2.82253700 -0.48003700 0.86604700 C 4.04383700 0.09336500 0.49337700 C 2.57789700 -0.77303400 2.20889600 C 4.99762000 0.38522400 1.45912300 H 4.26251300 0.29657300 -0.55597600

S42

208 C 3.53635200 -0.48097400 3.17472200 H 1.62779800 -1.22159100 2.50685300 C 4.74261000 0.09994800 2.79984600 H 5.94828500 0.82599800 1.16597200 H 3.33820800 -0.71019900 4.21999200 H 5.49525100 0.32279600 3.55358500 C 2.34645500 -1.90127900 -1.64159300 C 2.20743200 -3.28853200 -1.53108000 C 3.11602400 -1.36532100 -2.68012000 C 2.83545500 -4.13024100 -2.44193500 H 1.60281900 -3.71385200 -0.72769100 C 3.73639200 -2.20997400 -3.59281400 H 3.23402800 -0.28670200 -2.78721100 C 3.59774300 -3.59036100 -3.47302800 H 2.72187900 -5.20821600 -2.35079300 H 4.33040500 -1.78925100 -4.40145700 H 4.08304800 -4.24827100 -4.19081000

Ph + [Ag3(H)2L 2] E(M06/BS1) = -3734.032094 au H(M06/BS1) = -3733.170615 au G(M06/BS1) = -3733.326060 au E(M06/BS2//M06/BS1) = -3738.636247 au C -3.70899500 0.17538600 -0.68329400 H -4.59355000 0.59071100 -0.17550400 H -3.96686000 0.04274600 -1.74521000 H 0.56395700 -3.26033400 -0.42095000 Ag -1.02965100 -2.41577900 -0.50901000 P -3.29492100 -1.51973400 -0.05145700 P -2.31179300 1.39979600 -0.72520000 H 0.55303800 -0.79072300 -2.64175000 C 3.53231000 0.87882500 -0.47531700 H 4.33186400 1.38015500 0.09111600 H 3.58885700 1.22873000 -1.51816200 Ag -0.09794200 0.38924600 -1.37299300 P 1.84534600 1.39490200 0.11074700 P 3.75083500 -0.96603200 -0.52743900 Ag 1.57695200 -1.91989600 -1.40009100 C -2.92090900 2.62752600 -1.92777400 C -3.95780900 3.51188300 -1.61259300 C -2.34559200 2.65709000 -3.20110200 C -4.41662600 4.41107300 -2.56669900 H -4.39623200 3.50345800 -0.61348100 C -2.80891600 3.55901000 -4.15467500 H -1.53559500 1.96598700 -3.44671700 C -3.84246900 4.43375400 -3.83664400 H -5.22284000 5.09939200 -2.32047300 H -2.35989100 3.57930100 -5.14578800 H -4.20362100 5.14105600 -4.58087100 C -2.32625300 2.29038000 0.86679100 C -3.07817500 1.90800600 1.97809100 C -1.44615900 3.37713100 0.97727900 C -2.93415600 2.58239200 3.18925000 H -3.77880100 1.07718900 1.92138000

S43

209 C -1.32145200 4.05945000 2.17907600 H -0.85234700 3.68993200 0.11446100 C -2.05683900 3.65506700 3.29244100 H -3.51811400 2.26033000 4.05044200 H -0.63816600 4.90418800 2.24738200 H -1.94631200 4.18079500 4.23897000 C -4.63231600 -2.57991000 -0.68958100 C -4.40009900 -3.95973700 -0.66929100 C -5.85837800 -2.09721200 -1.15674300 C -5.38173400 -4.84372600 -1.09879900 H -3.44180300 -4.34445700 -0.31334300 C -6.83567600 -2.98539600 -1.59485900 H -6.06333200 -1.02706400 -1.18409600 C -6.59966000 -4.35592300 -1.56443800 H -5.19138600 -5.91477600 -1.07994100 H -7.78588000 -2.60283400 -1.96241300 H -7.36542900 -5.04708300 -1.91069500 C -3.59345000 -1.43472000 1.74780900 C -4.89003600 -1.30131100 2.25717600 C -2.50565900 -1.44887000 2.62433300 C -5.09102300 -1.15258300 3.62388700 H -5.74707500 -1.31978300 1.58221400 C -2.71132100 -1.31137500 3.99450000 H -1.49139500 -1.56889400 2.23432000 C -4.00052100 -1.15437500 4.49249800 H -6.10052700 -1.04789200 4.01661400 H -1.86107700 -1.34080300 4.67419300 H -4.16063100 -1.04962200 5.56397000 C 1.89590100 3.20592200 -0.16117100 C 2.08756900 4.13749900 0.86065900 C 1.63950300 3.65807200 -1.46254200 C 2.02648600 5.50070700 0.58319400 H 2.26581500 3.79919200 1.88146800 C 1.58498600 5.01907300 -1.73813300 H 1.47146200 2.93420400 -2.26535100 C 1.77350300 5.94248700 -0.71168900 H 2.17635600 6.22254800 1.38456500 H 1.38676800 5.36068000 -2.75237000 H 1.72200700 7.00871300 -0.92335800 C 1.85373400 1.18200300 1.92219500 C 2.98335600 1.37902100 2.72398600 C 0.64237400 0.83605100 2.52884700 C 2.89965800 1.23120700 4.10335300 H 3.94180800 1.64282800 2.27639700 C 0.55370800 0.70956600 3.91131700 H -0.24686500 0.67596300 1.91427700 C 1.68408800 0.90141600 4.69907700 H 3.78658300 1.37942900 4.71654400 H -0.40768500 0.47349000 4.36657400 H 1.62015000 0.79741000 5.78071300 C 5.28921600 -1.20082400 -1.47630100 C 5.48582600 -2.45819600 -2.05764400 C 6.26337100 -0.21125800 -1.63940000 C 6.64376700 -2.72701200 -2.77725900 H 4.72290000 -3.23070700 -1.94440900 C 7.41794300 -0.48037600 -2.36718100

S44

210 H 6.12754700 0.77709700 -1.19986700 C 7.60931300 -1.73674700 -2.93380300 H 6.78812000 -3.70839200 -3.22485100 H 8.17090200 0.29542600 -2.49316500 H 8.51276600 -1.94417100 -3.50393500 C 4.17722500 -1.38641300 1.20016200 C 5.44380900 -1.12580600 1.73242800 C 3.18872100 -1.94683100 2.01380300 C 5.71328700 -1.41342500 3.06538600 H 6.22846300 -0.71053700 1.09806500 C 3.45895500 -2.22570300 3.34966200 H 2.20106000 -2.15874500 1.59723200 C 4.72052900 -1.96294700 3.87406900 H 6.70418800 -1.22017900 3.47259000 H 2.68342900 -2.65742300 3.97994900 H 4.93602200 -2.19428100 4.91589800

Ph + [Ag3(H)2L ] E(M06/BS1) = -2086.21344 au H(M06/BS1) = -2085.773971 au G(M06/BS1) = -2085.871002 au E(M06/BS2//M06/BS1) = -2090.401324 au C -1.12363400 0.17555400 1.31058300 H -2.19262400 0.37700400 1.47610700 H -0.66135700 0.00238500 2.29231300 H 2.94677600 -1.92045100 -1.44646500 Ag 2.15296200 1.44178600 0.51499900 P -0.30849100 1.64206700 0.52958900 P -0.93478900 -1.39337500 0.32988400 H 3.94983000 1.11588900 0.28373000 Ag 1.25822100 -1.65536900 -0.77041100 Ag 3.77694700 -0.47231100 -0.65781100 C -0.96337100 1.66486900 -1.16827500 C -2.33068500 1.85011300 -1.40797400 C -0.09777800 1.43853200 -2.24221100 C -2.82199400 1.79082400 -2.70495100 H -3.01205100 2.04298000 -0.57803700 C -0.59274200 1.38382900 -3.54263100 H 0.97264400 1.31087000 -2.06240700 C -1.95409000 1.55497400 -3.77066100 H -3.88660500 1.92577800 -2.88692800 H 0.08723600 1.21704200 -4.37551000 H -2.34279900 1.51486400 -4.78656900 C -1.03691100 3.08528000 1.36218800 C -0.93352300 4.32706700 0.72326200 C -1.61120900 3.00634800 2.63322800 C -1.41577200 5.47104600 1.34472800 H -0.48509500 4.39846800 -0.26893100 C -2.09093700 4.15757900 3.25126100 H -1.69009300 2.05299300 3.15484200 C -1.99525400 5.38693200 2.60878600 H -1.33873300 6.43254500 0.84138500 H -2.54150900 4.09005300 4.23932500 H -2.37136200 6.28488300 3.09454800

S45

211 C -1.11500700 -2.67665500 1.61130200 C -2.16523000 -3.59553800 1.62751100 C -0.10049800 -2.76437300 2.57343900 C -2.20754300 -4.58011600 2.61146900 H -2.94934600 -3.54919600 0.87310900 C -0.15353800 -3.74153300 3.55786100 H 0.74398900 -2.06967800 2.54884600 C -1.20948300 -4.65093400 3.57651500 H -3.02764600 -5.29506600 2.62101700 H 0.63503800 -3.80310100 4.30492700 H -1.24776600 -5.42212600 4.34291700 C -2.39430800 -1.45878500 -0.74610000 C -3.68835100 -1.25525700 -0.24852200 C -2.21218200 -1.71260700 -2.10683700 C -4.77965700 -1.30377200 -1.10598600 H -3.84922400 -1.07356200 0.81544900 C -3.30735800 -1.75849100 -2.96400800 H -1.20724900 -1.86586700 -2.50359100 C -4.58876600 -1.55499800 -2.46399300 H -5.78327500 -1.15238200 -0.71358100 H -3.15652800 -1.95291900 -4.02403800 H -5.44620900 -1.59407800 -3.13295600

S46

212 9.2 Appendix B - Supplementary material for Chapter 3

213 214 + [(Ph3P)Ag(CH3CN)] + [(Ph3P)2Ag2(O2CH)] 100 (a) 410 785

0 + + 100 [dppm2Ag2(O2CH)] (b) [dppmAg2(O2CH)] 645 1029 0 + 100 (c) [dppmAg2(O2CH)] 645 0 + [dppbzAg] + 100 [dppbzAg (O CH)] (d) 553 2 2

Relative Intensity (%) 707 0 + + (e) [dppeAg2(O2CH)] 100 [dppe2Ag] 659 905 0 200 400 600 800 1000 1200 1400 1600 1800 2000 m/z

Supplementary Figure 1│LTQ ESI-MS of ligated silver formate clusters formed

in a 10 mL solution. (a) methanol:dichloromethane (1:1) of 20 mmol AgNO3 and 10

mmol PPh3 5 minutes after 10 mmol NaO2CH was added; (b) acetonitrile:water (95:5)

of 20 mmol AgBF4, 10 mmol dppm and 10 mmol NaO2CH; (c) methanol:formic acid

(1:1) of 20 mmol AgNO3, 10 mmol dppm and 10 mmol NaO2CH; (d)

dichloromethane:acetonitrile:formic acid (45:45:10) 20 mmol AgNO3, 10 mmol dppm

and 10 mmol NaO2CH; (e) dichloromethane:acetonitrile:formic acid (45:45:10) of 20

mmol AgNO3, 10 mmol dppm and 10 mmol NaO2CH. Solutions containing silver clusters were diluted to 50 μM prior to ESI. The m/z represents the most abundant isotope peak in the cluster.

215 100 + (a) [Ag2(O2C(CH2)2Ph)] + [Ag (O C(CH ) Ph)(MeCN)] 365 2 2 2 2 406 406 50 *

+ [Ag(H2C=CH-Ph)] 100 + 406 (b) 211 [Ag2((CH2)2Ph)] [Ag (H)]+ 2 321 215 365

Relative Intensity (%) 50 *[Ag (O C(CH ) Ph)]+ 2 2 2 2 365

# 0 200 300 400 500 600

m/z

Supplementary Figure 2│LTQ collision-induced dissociation of dinuclear silver + + clusters to generate [Ag2(H)] m/z 215. (a) [Ag2(O2C(CH2)2Ph)(MeCN)] m/z 406, + and (b) [Ag2(O2C(CH2)2Ph)] m/z 365. The most intense peak in the cluster is represented by the m/z value. * Refers to the mass selected precursor cluster. # Refers to reaction with background solvent.

216 Supplementary Figure 3│LTQ collision-induced dissociation to generate + + + [(Ph3P)Ag2(H)] m/z 479. (a) [(Ph3P)2Ag2(O2CH)] m/z 785. (b) [(Ph3P)Ag2(O2CH)] m/z 523. The most intense peak in the cluster is represented by the m/z value. * Refers to the mass selected precursor cluster.

217 [dppmAg (H)]+ 100 (a) 2 601 645 + [dppmAg2(O2CH)] 50 645 * 0

+ + [dppbzAg (H)] 100 (b) [dppbzAg] 2 553 663 707 [dppbzAg (O CH)]+ 50 2 2 # 707 * 0 [dppeAg (H)]+ Relative Intensity (%) 100 (c) 2 615 659 + [dppeAg] [dppeAg (O CH)]+ 50 2 2 505 659 # * 0 400 600 800 1000 m/z

+ Supplementary Figure 4│Collision-induced dissociation of [LAg2(O2CH)] . (a) L = bis(diphenylphosphino)methane (dppm); (b) 1,2-bis(diphenylphosphino)benzene (dppbz) and (b) bis(diphenylphosphino)ethane (dppe). The most intense peak in the ion is represented by the m/z value. * Represents the mass selected precursor ion. # Represents reaction of [LAg]+ (L = dppbz and dppe) with background acetonitrile in the ion trap from the ESI solvent.

218

Supplementary Figure 5│LCQ energy-resolved CID data for the + decarboxylation of [LAg2(O2CH)] . (a) Plot of reaction extent (Σproduct ions/total + ion count) vs activation voltage (Vp-p) for the decarboxylation of [LAg2(O2CH)] . The solid red circle corresponds to L = dppm = bis(diphenylphosphino)methane; the solid blue star corresponds to L = dppbz = 1,2-bis(diphenylphosphino)benzene, and; the solid black square corresponds to L = dppe = 1,2-bis(diphenylphosphino)ethane. Each coordinate is the average of at least three separate experiments from which the standard deviation is shown as error bars. The threshold activation voltages correspond to the dashed olive line at 10% reaction extent. The thresholds for dissociation were also measured for two systems were the critical energies for + dissociation are known from other gas-phase experiments: Ag(CH3OH) (1.43 ± 0.16 + eV) and Fe(C5H5)2 (3.7 ± 0.3 eV).

219

Supplementary Figure 5│LCQ energy-resolved CID data for the + decarboxylation of [LAg2(O2CH)] . (b) Expanded view of Figure S5a showing the + corresponding activation voltage (Vp-p) for [dppeAg2(O2CH)] (0.582) ≈ + + [dppbzAg2(O2CH)] (0.585) < [dppmAg2(O2CH)] (0.621) at 10% reaction extent. (c) + Plot of relative abundance for the formation of [dppeAg2(H)] from the ERCID of 1f

as a function of activation voltage (Vp-p). Linear regression analyses using data points 2 from 0.58 – 0.70 (Vp-p). R = 0.98.

220 Supplementary Figure 5│LCQ energy-resolved CID data for the + decarboxylation of [LAg2(O2CH)] . (d) Plot of relative abundance for the formation + of [dppbzAg2(H)] from the ERCID of 1e as a function of activation voltage (Vp-p). 2 Linear regression analyses using data points from 0.58 – 0.67 (Vp-p). R = 0.99. (e) + Plot of relative abundance for the formation of [dppmAg2(H)] from the ERCID of 1d

as a function of activation voltage (Vp-p). Linear regression analyses using data points 2 from 0.66 – 0.80 (Vp-p). R = 0.98.

221

Supplementary Figure 5│LCQ energy-resolved CID data for the + decarboxylation of [LAg2(O2CH)] . ((f) Comparison of the experimentally

determined activation voltage (Vp-p) by the 10% reaction extent method and the linear regression analyses method (x-axis) vs. the DFT calculated energy (eV) for ease of decarboxylation (y-axis) for the decarboxylation transition states given in Supplementary table 3.

222 + [dppbzAg2(O2CH)] [dppbzAg (H)]+ 1.0 2 [dppbzAg]+ + [dppbzAg(CH3CN)]

0.5 Relative Intensity

0.0 15 20 25 30 Normalized Collision Energy (%)

+ Supplementary Figure 6│LCQ energy-resolved CID of [dppbzAg2(O2CH)] . Branching ratios as a function of CID energy.

223 (a) 107 + 100 * 406 [ Ag2(H)(HO2CH)] 50 261 107 + 363 [ Ag2(H)(HO2CH)2] 215 307 HO CH 5 2 0 406 100 (b) * 107 + [ Ag2(H)(HO2CH)2] 363 307 215 HO CH 50 2 261 HO2CH 0 (c) 406 100 107 + Relative Intensiy (%) 107 + [ Ag2(H)] 363 [ Ag2(H)(HO2CH)2] 215 307 215 50 HO2CH 261 HO CH * 2 0 200 250 300 350 400 m/z

Supplementary Figure 7│LTQ-FT gas-phase reactions of silver hydrides, + 107 + [Ag2(H)] , with formic acid. (a) ion-molecule reaction of [ Ag2(H)] m/z 215 with 107 + formic acid for 1000 ms; (b) ion-molecule reaction of [ Ag2(H)( HO2CH)] m/z 261 107 + with formic acid for 5000 ms; (c) CID of [ Ag2(H)(HO2CH)] m/z 215. The most intense peak in the cluster is represented by the m/z value. * Refers to the mass selected precursor cluster. # Refers to noise peaks.

224 100 (a) + * 785 [(Ph3P)2Ag2(H)] 741 741

HO2CH 50

# 0

100 (b) * 785 + [(Ph3P)Ag2(H)] 479 523

479 Relative Intensity (%) + 50 HO2CH [(Ph3P)Ag2(H)(HO2CH)] 525

# 0 200 300 400 500 600 700 800 900 m/z

Supplementary Figure 8│LTQ-FT ion-molecule reaction of dinuclear silver

hydrides coordinated to triphenylphosphine (PPh3) with formic acid. (a) + [(Ph3P)Ag2(H)] m/z 741 for 10000 ms. (b) LTQ ion-molecule reaction of + [(Ph3P)Ag2(H)] m/z 479 for 100 ms. The most intense peak in the cluster is represented by the m/z value. * Refers to the mass selected precursor cluster. # Refers to noise peaks.

225

+ Supplementary Figure 9│Synthesis of deuterium labelled [dppmAg2(D)] and reaction with formic acid. (a) LTQ collision-induced dissociation of 2+ [dppm3Ag3(D)] m/z 740, prepared from [Ag3(µ3-D)dppm3](BF4)2. (b) LTQ ion- + molecule reaction of [dppmAg2(D)] m/z 602 with HO2CH for 1000 ms. (c) LTQ + collision-induced dissociation of [dppmAg2(O2CH)] m/z 645. (c) LTQ ion-molecule + reaction of [dppmAg2(H)] m/z 601 with HO2CH for 1000 ms. The most intense peak in the cluster is represented by the m/z value. * Refers to the mass selected precursor cluster.

226 Supplementarry Figure 10│The analysis of leading excitations conttributing to the most intense transitions of UV absorption spectra shown in Figures 4c and 4d. + + (a) [dppmAg2(H)] and (b) [dppmAg2(O2CH)] respectively; cut-off for MOs is 0.04, minus and plus are labeled by blue and grey colors, respectively.

227

Supplementary Figure 11│DFT-calculated energy profile for the two reaction steps in the catalytic cycle for: (a) 1e and (b) 1ff.

228

Supplementarry Figure 12│Calculated chargee distribution. (positive charge blue, negative charge red), blue lines close to atoms label positive charge and the corresponding region in which the charge is delocalized is denoted by (+). Bond distances are given in Å (black).

229

Supplementary Figure 13│Variable temperature (+25 – +70°C) 1H NMR of Ag, dppm, formic acid, sodium formate in CD3CN. Note the evolution of H2 gas at 4.65 ppm.

230 Supplementary Figure 14│Variable temperature (+25 – +70°C) 1H NMR of Ag, dppm, formic acid in CD3CN. Note that no H2 gas is evolved at 4.65 ppm.

231

Supplementary Figure 15│1H NMR at 70°C of Ag, dppm, 13C-formic acid and sodium formate in CD3CN. Note the evolution of H2 gas at 4.6 ppm.

232

Supplementary Figure 16│13C NMR at 70°C of Ag, dppm, 13C-formic acid and 13 sodium formate in CD3CN. Note the evolution of CO2 gas at 126 ppm.

233

Supplementarry Figure 17│The most stable DFT calculated structures for the + interaction of formic acid with [(Ph3P)nAg2(H)(HO2CH)2-n] . For n = 0, both formic acid molecules coordinate to a separate Ag site. For n = 1, tthe formic acid coordinates at the vacant coordination site. For n = 2, there is not vacant coordination site and so the formic acid is weakly bound in an ion-molecule complex. Note the formic acid molecules are in trans conformation.

234 Supplementarry Figure 18│ DFT calculated structures and relative energies of cis + versus trans conformers of formic acid and their adducts with [LAg2(H)] : (a) cis formic acid; (b) trans formic acid; (c) cis formic acid adduct of 1d (dppm); (d) trrans formic acid adduct of 1d (dppm); (e) cis formic acid adduct of 1e (dppbz); (f) trans formic acid adduct of 1e (dppbz); (g) cis formicc acid adduct of 1f (dppe); (h) trans formic acid adduct of 1f (dppe).

235

Supplementarry Figure 19│Experimmental (matrix isolation) vs DFT calculated gas-phase IR scaled by 0.98 of monnomeric formic acid (in cis conformation 5).

236 Supplementary Table 1│Kinetic data associated with the ion-molecule dinuclear + + silver hydrides with formic acid. 1a, [Ag2(H)] ; [LAg2(H)] , where L = Ph3P, 1b; 2

x PPh3, 1c; dppm, 1d; dppb, 1e and; dppe, 1f with formic acid.

Reaction Reactant Product k a,b expt Efficiencyc 1.25 x 10-11 ± 1.53 1.12 ± 0.014 1a [Ag (H)]+ [Ag (H)(HO CH)]+ 2 2 2 x 10-13

8.45 x 10-10 ± 8 x 79.52 ± 0.75 + + -12 1b [(Ph3P)Ag2(H)] [(Ph3P)Ag2(H)(HO2CH)] 10

+ d 1c [(Ph3P)2Ag2(H)] N.R. N.A. N.A.

1.53 x 10-11 ± 2.83 1.45 ± 0.027 1d [dppmAg (H)]+ f [dppmAg (O CH)]+ 2 2 2 x 10-13

1.25 x 10-11 ± 6.18 1.19 ± 0.059 1e [dppbzAg (H)]+ f [dppbzAg (O CH)]+ 2 2 2 x 10-13

2.27 x 10-14 ± 6.74 2.15 x 10-3 ± 6.4 1f [dppeAg (H)]+ f [dppeAg (O CH)]+ 2 2 2 x 10-16 x 10-5

a In units of cm3.molecules-1.s-1. b + Rates for the reaction with formic acid with [LAg2(H)] to regenerate + [LAg2(O2CH)] as the product. Rates were determined by monitoring the ion evolution with time. c Reaction efficiency = (kexpt/kADO) x 100. The kADO is the theoretical ion-molecule collision rate constant obtained from the average-dipole orientation (ADO) theory6, which was calculated using the Colrate program7. d No reaction. The data shown for each reaction is the average of three separate experiments using different concentrations of neutral reagent.

237 Supplementary Table 2│Assignment of experimentally observed vibrational + + bands of the [(dppm)Ag2(O2CH)] (m/z 645) and [(dppm)Ag2(H)] (m/z 601) ions. (a) the position of the experimental bands are given in cm-1. (b) Harmonic frequencies were scaled by 0.98, and predicted intensities (in parentheses) are given in km.mol-1. Highlighted in green are the modes associated with the auxiliary ligands, and highlighted in red are the most structurally diagnostic band.

[(dppm)Ag (O CH)]+ [(dppm)Ag (H)]+ Mode Assignment 2 2 2 (m/z 645) (m/z 601) Exp.a) Theoryb) Exp. a) Theoryb) Ag as. st. 900 916 (245) 998 (9) 998 (8) C H in plane ring 998 (10) 998 (9) 6 5 not observed 998 deformation 999 (5) 999 (4) 1000 (6) 1000 (6) 1086 (1) 1085 (6) 1087 (14) 1086 (1) 1087 (11) 1087 (2) CH2 twist; P-C6H5 st. (also 1088 (16) 1088 (24) C6H5 in plane CH bending 1103 1097 and ring deformation) 1090 (17) 1089 (30) 1093 (33) 1091 (17) 1094 (19) 1093 (26) 1099 (19) 1099 (9) AgH s. st. 1250 1236 (150) formate s. CO st. 1360 1345 (121)

CH2 scissoring 1393 (12) 1395 (13) 1444 (11) 1443 (13) not observed 1445 (19) 1438 1444 (22) C H in plane CH bending 6 5 1445 (9) 1445 (9) 1446 (31) 1445 (21) 1490 (7) 1489 (7) C H in plane CH bending 1490 (5) 1489 (6) 6 5 not observed 1478 and ring deformation 1492 (4) 1491 (3) 1492 (7) 1491 (7) formate as. CO st. 1547 1564 (504)

238 Supplementary Table 3│DFT (and DFT-D3) -calculated energies (eV) for steps 1 and 2 relative to the initial reactants.

The energies (E) of two reaction steps in the catalytic cycle + of [LAg2(H)] 1d 1e 1f + + + [dppmAg2(H)] [dppbzAg2(H)] [dppeAg2(H)]

Relative E (eV) Relative E (eV) Relative E (eV)

Reactants 0.00 (0.00) 0.00 (0.00) 0.00 (0.00)

Eb of cis-formic acid -0.24 (-0.31) -0.22 (0.29) -0.16 (-0.23) TS for H loss STEP 1 2 -0.06 (-0.09) 0.14 (0.10) 0.09 (0.06)

H2 loss -0.95 (-1.04) -0.71 (-0.86) -0.76 (-0.89)

H2 loss 0.00 (0.00) 0.00 (0.00) 0.00 (0.00)

TS breaking Ag-O bond 1.70 (1.75) 1.67 (1.69) 1.69 (1.73)

O-bound ligand 1.43 (1.37) 1.22 (1.20) 1.30 (1.27) TS for CO loss STEP 2 2 1.86 (1.80) 1.65 (1.69) 1.64 (1.69)

Formation of CO2 0.81 (0.83) 0.60 (0.63) 0.67 (0.71)

Separated products 0.95 (1.04) 0.71 (0.71) 0.76 (0.76)

239 Supplementary Table 4│High-resolution mass spectrometry experiments comparing the mass-to-charge ratio (m/z) of the most abundant peak in the isotope pattern of a given ion. Experiments were recorded using an FT-ICR mass analyser from a Thermo LTQ-FT hybrid instrument. The theoretical m/z was determined from the molecular formula and charge state of the silver cluster using the Thermo Xcalibur Qual Browser version 2.2 software with the output style set to Profile (Gaussian) and the resolution set to 20000 and the valley set to full-width-half- mass (FWHM).

Experimental Theoretical Error Ion Formula (m/z) (m/z) (ppm)

+ 1a [Ag2(H)] Ag2H 216.81726 216.81713 0.6 + [Ag2(H)(HO2CH)] Ag2H3C1O2 262.82273 262.82262 0.4 + [Ag2(H)(HO2CH)2] Ag2H5C2O4 308.82835 308.82811 0.8 + 1b [(Ph3P)Ag2(H)] Ag2H16C18P 478.90842 478.90833 0.2 + [(Ph3P)Ag2(H)(HO2CH)] Ag2H18C19PO2 524.91391 524.91383 0.1 + 1c [(Ph3P)2Ag2(H)] Ag2H31C36P2 740.99997 740.99968 0.4 + 1d [dppmAg2(H)] Ag2H23C25P2 600.93701 600.93693 0.1 + [dppmAg2(O2CH)] Ag2H23C26P2O2 644.92690 644.92678 0.2 + 1e [dppbzAg2(H)] Ag2H25C30P2 662.95277 662.95264 0.2 + [dppbzAg2(O2CH)] Ag2H25C31P2O2 706.94262 706.94249 0.2 + 1f [dppeAg2(H)] Ag2H25C26P2 614.95271 614.95259 0.2 + [dppeAg2(O2CH)] Ag2H25C27P2O2 658.94256 658.94245 0.1

240 Supplementary Methods

Chemicals used

Chemicals purchased from the following suppliers were used without further

purification. (i) Ajax Finechem: formic acid (HCO2H, 99%), silver(I) nitrate

(AgNO3),. (ii) Sigma-Aldrich: silver(I) tetrafluoroborate (AgBF4, 98%), sodium

formate (NaO2CH, 99%) bis(dphenylphosphino)methane (dppm, 97%), 1,2- bis(diphenylphosphino)ethane (dppe, 99%), 1,2-bis(diphenylphosphino)benzene (dppb, 97%) (iii) Merck: acetonitrile (HPLC grade), methanol (AR grade)

dichloromethane (AR grade). (iv) Riedel-de Haën: triphenylphosphine (PPh3).

Gas-phase preparation of 1a-1f

+ All of the binuclear silver hydride cations, [LAg2(H)] , were synthesized by CID of appropriate precursor complexes formed via electrospray ionization (ESI) of solutions containing silver(I) salts, a carboxylic acid or sodium formate and where required, the phosphine ligand. The solvent was found to play a key role in the formation of + [LAg2(H)] via ESI/MS (Supplementary figure 1). In the case of triphenyl phosphine, + [(Ph3P)2Ag2(O2CH)] , was only formed in a 1:1 methanol:dichloromethane mixture (Supplementary figure 1a). For the bisphosphine ligands, dppm gave the simplest ESI + mass spectrum, with the base peak being [dppmAg2(O2CH)] m/z 645, + (Supplementary figure 1b). Modest signals of [dppbAg2(O2CH)] m/z 707, + (Supplementary figure 1d) and [dppeAg2(O2CH)] m/z 659, (Supplementary figure 1e) were obtained, with other ions dominating the ESI mass spectra.

The hydride ligand is unmasked via fragmentation of a coordinated carboxylate + ligand. Thus sequential CID of [Ag2(O2C(CH2)2Ph)(MeCN)] (m/z 406, + Supplementary figure 2a and 2b) gave 1a (m/z 215), while [(Ph3P)2Ag2(O2CH)] + (m/z 785) underwent decarboxylation to give [(Ph3P)2Ag2(H)] (1c m/z 741, equation + (1a)) in competition with loss of a phosphine ligand to give [(Ph3P)Ag2(O2CH)] (m/z 523, equation (2b), Supplementary figure 3a). The latter can be mass selected and

241 + subjected to CID to yield [(Ph3P)Ag2(H)] via decarboxylation (1b m/z 741, Supplementary equation (2), Supplementary figure 3b).

+ + [(Ph3P)2Ag2(O2CH)] → [(Ph3P)2Ag2(H)] + CO2 (1a)

+ → [(Ph3P)Ag2(O2CH)] + Ph3P (1b)

+ + [(Ph3P)Ag2(O2CH)] → [(Ph3P)Ag2(H)] + CO2 (2)

+ CID of [dppmAg2(O2CH)] m/z 645, results in the loss of CO2 (equation (4), + Supplementary figure 4a) yielding [dppmAg2(H)] m/z 601, 1d. The other binuclear + + silver hydride cations, [dppbzAg2(H)] , 1e and [dppeAg2(H)] , 1f were formed in similar decarboxylation reactions (equation (4), Supplementary figure 4b and 4c), although the nature of the ligand influences both the relative ease of decarboxylation + as determined via resolved CID, following the order [dppeAg2(O2CH)] ≈ + + [dppbzAg2(O2CH)] > [dppmAg2(O2CH)] (Supplementary figure 5). The energy resolved CID experiments show that [dppbzAg]+ arises from secondary fragmentation + of [dppbzAg2(H)] (Supplementary figure 6).

Energy-resolved collision-induced dissociation

Energy-resolved CID experiments were carried out using a Finnigan 3D ion trap (LCQ) mass spectrometer. The experimental method of Brodbelt was adapted1, whereby the silver clusters were diluted to 50 µM and introduced into the mass spectrometer via a syringe pump set at 5 µL.min-1 through a Finnigan ESI source. The source conditions used for optimum intensity of the target ions were: spray voltage 4.6 – 5.2 kV, capillary temperature 200°C, nitrogen sheath gas pressure, 40 (arbitrary units), capillary voltage 42 V, tube lens voltage -15 V. The mass-selected precursor ion was isolated with a mass selection window of 8 Th and an activation time of 10 ms. The normalised collision energy (NCE) was increased incrementally by 1.0 % typically starting from a NCE where no fragmentation is observed, until reaching the NCE required for depleting the precursor ion to < 5 % relative intensity. The NCE was converted to an amplitude of the resonance excitation RF voltage (tick amp),

242 Supplementary equation 3. The ‘tick amp slope” and “tick amp intercept” are taken from the latest calibration file prior to experiments. The Σproduct ions/total ion count (Y-axis) was plotted against the Activation Voltage determined from Supplementary equation 3.

Activation = (NCE)% 30% ((parent mass)(tick amp slope) + (3) Voltage tick amp intercept) (V)

Relative ease of decarboxylation of 1d, 1e and 1f.

Method 1, Colorado and Brodbelt1:

The threshold activation voltage corresponding to the critical energy of decarboxylation was determined at 10% reaction extent, i.e. when Σproduct ions/total ion count = 0.1. The thresholds for dissociation were also measured for two systems were the critical energies for dissociation are known from other gas-phase + 2 + 3 experiments: Ag(CH3OH) (1.43 ± 0.16 eV) and Fe(C5H5)2 (3.7 ± 0.3 eV) . The results of this plot are shown in Supplementary figure 5a and 5b. We find the + following activation voltages (Vp-p): [dppeAg2(O2CH)] (0.582) ≈ + + [dppbzAg2(O2CH)] (0.585) < [dppmAg2(O2CH)] (0.621).

Method 2, Falvo et al4:

The threshold activation voltage corresponding to the critical energy of decarboxylation for 1d, 1e and 1f was determined by linear regression analysis of the relative abundance for the product of decarboxylation. The results of these plots are shown in Supplementary figure 5c – 5e. We find the following activation voltages + + (Vp-p): [dppeAg2(O2CH)] (0.564) ≈ [dppbzAg2(O2CH)] (0.566) < + [dppmAg2(O2CH)] (0.627).

243 Comparisons of experimental methods 1 and 2 to Theory:

The experimentally determined relative ease of decarboxylation via both the 10% reaction extent method of Brodbelt and the linear regression analysis method of Falvo are in agreement with the DFT calculated activation energies (eV) for decarboxylation + + + of [dppmAg2(O2CH)] , [dppbzAg2(O2CH)] and [dppeAg2(O2CH)] .

244 Supplementary References

(1) Colorado, A. & Brodbelt, J. An empirical approach to estimation of critical energies by using a quadrupole ion trap J. Am. Soc. Mass Spectrom., 7, 1116– 1125 (1996). (2) El Aribi, H.; Shoeib, T.; Ling, Y.; Rodriquez, C. F.; Hopkinson, A. C. & Siu, K. W. M. Binding Energies of the Silver Ion to Small Oxygen-Containing Ligands: Determination by Means of Density Functional Theory and Threshold Collision-Induced Dissociation J. Phys. Chem. A 106, 2908–2914 (2002). (3) Faulk, J. D. & Dunbar, R. C. Time-resolved photodissociation of gas-phase ferrocene cation: energetics of fragmentation and radiative relaxation rate at near-thermal energies J. Am. Chem. Soc. 114, 8596–8600 (1992). (4) Falvo, F.; Fiebig, F.; Dreiocker, F.; Wang, R.; Armentrout, P.B. & Schäfer, M. Fragmentation reactions of thiourea- and urea-compounds examined by tandem MS-, energy-resolved CID experiments, and theory, Int. J. Mass Spectrom., 330–332, 124–133 (2012). (5) Halupka, M., Sander, W. A simple method for the matrix isolation of monomeric and dimeric carboxylic acids. Spectrochim. Acta A, 54, 495–500 (1998). (6) Su, T. & Bowers, M. T. Classical ion-molecule collision theory Gas-Phase Ion Chemistry 1, 83–118 (1979). (7) Lim, K. F. Quantum Chemistry Program Exchange 14, 3 (1994). The program Colrate is available for download from the author’s website at Deakin University, Geelong, Victoria, Australia: http://www.deakin.edu.au/~lim/programs/COLRATE.html.

245 246 9.3 Appendix C - Supplementary material for Chapter 4

247 248 Supporting Information

Selectivity Effects in Bimetallic Catalysis: Role of the Metal Sites in the Decomposition of Formic Acid into H2 and CO2 by the Coinage Metal Binuclear Complexes [dppmMM’(H)]+ Athanasios Zavras,[a] Marjan Krstic´,[b] Philippe Dugourd,[c] Vlasta Bonacˇic´-Koutecky´,*[b, d] and Richard A. J. O’Hair*[a]

cctc_201601675_sm_miscellaneous_information.pdf

249 Supporting Information

250 Table of Contents Experimental ...... 2 Figure S1 ...... 4 Figure S2 ...... 5 Figure S3 ...... 6 Figure S4 ...... 7 Figure S5 ...... 8 Figure S6 ...... 9 Table S1 ...... 10 Figure S7 ...... 11 Scheme S1 ...... 12 Figure S8 ...... 13 Figure S9 ...... 14 Figure S10 ...... 15 Figure S11 ...... 16 Figure S12 ...... 17 Table S2 ...... 19 Table S3 ...... 19 Cartesian coordinates ...... 20

251 EXPERIMENTAL

Materials Chemicals purchased from the following suppliers were used without further purification. (i) Ajax

Finechem: formic acid (HCO2H, 99%), copper(I) oxide (Cu2O) silver(I) oxide (Ag2O). (ii) Sigma- Aldrich: (dppm, 97%), 1,2-bis(diphenylphosphino). (iii) Merck: acetonitrile (HPLC grade), methanol (AR grade). (iv) STREM: chloro(triphenylphosphine)gold(I) (99%). Experimental Section

Preparation of in situ phosphine ligated bimetallic formate clusters In situ complexes for ESI/MS were prepared by adding 20 mL of acetonitrile to 10 mmol Cu2O, 10 mmol Ag2O and 20 mmol ClAuPPh3 resulting in a reddish-brown suspension. To this suspension was added 100 mmol of formic acid followed by stirring for 2 minutes resulting in a clear solution. To this clear solution immediately was added 50 mmol of bis(diphenylphosphino)methane and the solution was stirred for a further 2 minutes.

Discussion of the generation of the bimetallic binuclear complexes Electrospray ionization (ESI) was performed on an acetonitrile solution containing a mixture of

Cu2O:Ag2O:AuClPPh3 (1:1:2) to which 10 equivalents of formic acid was added and then mixed for 5 minutes before a further 5 equivalents of dppm was added, and then finally diluted to 50 µM. The diluted solution was drawn into a 500 µL gas tight borosilicate glass syringe, with polytetrafluorethylene plunger tips, and injected into the Finnigan ESI source at a flow rate of 3 – 5 µL. min-1. ESI source conditions to yield a stable current of 0.5 µA were: needle potential (2.5 – 4.0 kV); nitrogen sheath gas pressure (5 arbitrary units). The ion transfer capillary temperature was set to 250°C. Voltages were: tube lens (10 – 30 V) and capillary voltage (20 – 40 V).

An examination of the resultant ESI/MS spectrum, in the positive-ion polarity mode, indicated the presence of a polydisperse solution (Figure S1a). Ligand coordinated bimetallic coinage metal formate cations were observed. The relative abundance (RA) for each coinage metal + combination of [dppmMM’(O2CH)] in the mass range of 400 – 1000 m/z provided sufficient + + signal for mass-selection: [dppmCu2(O2CH)] (m/z 555, RA = 100%), [LCuAg(O2CH)] (m/z 601, + + RA ≈ 50%), [LCuAu(O2CH)] (m/z 689, RA ≈12%) [LAgAu(O2CH)] (m/z 733, < 1%) and + + [LCuAg(O2CH)] (m/z 823, RA < 1%). The apparent RA of [LAgAu(O2CH)] (m/z 733) is ca. + 50% and [LAu2(O2CH)] (m/z 823) is ca. 60%, however the simulated isotope distribution (Figure S1b) does not match well with the experimental spectrum suggesting the presence of isobaric impurities. The stoichiometry of all ligand coordinated bimetallic coinage metal formate cations was confirmed by high-resolution mass spectrometry experiments (HRMS) obtained from FTICR-MS (Figure S2) experiments. This clearly identified the presence of isobaric + + impurities for [LAgAu(O2CH)] (m/z 733, RA ≈ 5%) (Figure S2d) and [LAu2(O2CH)] (m/z 823, RA ≈ 2%) (see Figure S2b) with sufficient resolution to obtain accurate mass measurements of < 1 ppm error. Attempts to remove isobaric impurities via changes in the synthetic procedure were unsuccessful (data not shown).

252 Tandem mass spectrometry experiments were used to generate bimetallic hydrides, [LMM’(H)]+, 2 + via CID (MS ) of the mass selected bimetallic formate complexes [LMM’(O2CH)] (Figure S3). The hydride was generated via the decarboxylation of the coordinated formate (Figure 1a, Step + 2). From Figure S3 we identified the RA of the reactive intermediates: [dppmCu2(H)] (m/z 511, + + RA = 100%), [LAu2(H)] (m/z 779, RA ≈ 5%), [LCuAg(H)] (m/z 557, RA ≈ 100%) + + + [LCuAu(O2CH)] (m/z 645, ≈100%) and [LCuAg(O2CH)] (m/z 689, RA < 1%). All [LMM’H] species could be further mass selected for MS3 experiments. The stoichiometry of all ions was confirmed by high-resolution mass spectrometry experiments (HRMS) obtained by a FTICR-MS (Figure S4).

The stoichiometry of all ions were confirmed by HRMS experiments (Figures S2 and S4). The experimental rates (with the present instrumentation) for the ion-molecule reactions of [dppmMM’H]+ with formic acid were in good agreement with literature values obtained for the - reaction of Br with CH3I (Figure S7).

253

Figure S1 (a) LTQ mass spectrum obtained from a solution of Cu2O:Ag2O:AuClPPh3 (1:1:2), formic acid (10 equivalents) and bis(diphenylphosphino)methane = dppm = L (5 equivalents) in acetonitrile. (b) Simulated mass spectrum obtained using the Thermo Scientific Xcalibur 2.2 + software to show the isotope patterns expected for (i) [LCu2(O2CH)] (m/z 555) = + + C26H23O2P2Cu2 (ii) [LCuAg(O2CH)] (m/z 601) = C26H23O2P2CuAg (iii) [LCuAu(O2CH)] (m/z + 689) = C26H23O2P2CuAu (iv) [LAgAu(O2CH)] (m/z 733) = C26H23O2P2AgAu and (v) + [LAu2(O2CH)] (m/z 823) = C26H23O2P2Au2. L = dppm = bis(diphenylphosphino)methane.

254 + + Figure S2 FTMS spectra obtained for (a) [LCu2(O2CH)] = C26H23O2P2Cu2 (b) [LAu2(O2CH)] = + + C26H23O2P2Au2 (c) [LCuAg(O2CH)] = C26H23O2P2CuAg (d) [LCuAu(O2CH)] = C26H23O2P2CuAu + and (e) [LAgAu(O2CH)] = C26H23O2P2AgAu. The black line is the experimental spectrum. The red line is the simulated spectrum. L = dppm = bis(diphenylphosphino)methane. The theoretical m/z of the cation is shown beneath the experimentally measured values and is displayed within brackets (). L = dppm = bis(diphenylphosphino)methane.

255

+ Figure S3 LTQ spectra obtained for the CID of mass-selected (a) [LCu2(O2CH)] (m/z 555) (b) + + + [LAu2(O2CH)] (m/z 823) (c) [LCuAg(O2CH)] (m/z 601) (d) [LCuAu(O2CH)] (m/z 689), and (e) + [LAgAu(O2CH)] (m/z 733). A * represents the mass-selected precursor ion. A # represents product ions from the CID of isobaric cations that overlap with the precursor ion. L = dppm = bis(diphenylphosphino)methane.

256 + + Figure S4 FTMS spectra obtained for (a) [LCu2(H)] (m/z 511) = C25H22P2Cu2 (b) [LAu2(H)] + + (m/z 779) = C25H22P2Au2 (c) [LCuAg(H)] (m/z 557) = C25H22P2CuAg (d) [LCuAu(H)] (m/z 645) + = C25H22P2CuAu and (e) [LAgAu(H)] (m/z 689) = C25H22P2AgAu. The black line is the experimental spectrum. The red line is the simulated spectrum. The theoretical m/z of the cation is shown beneath the experimentally measured values and is displayed within brackets ().L = dppm = bis(diphenylphosphino)methane.

257

Figure S5 Pseudo-first order kinetics obtained in a LTQ 2D linear-ion trap quasi thermalized to + 298 K for the ion-molecule reaction of formic acid with (a) [dppmCu2(H)] (m/z 511) (b) + + + [dppmAu2(H)] (m/z 779) (c) [dppmCuAg(H)] (m/z 557) (d) [dppmCuAu(H)] (m/z 645) and (e) [dppmAgAu(H)]+ (m/z 689). The black dots are plotted coordinates for the activation time (s) (x- axis) vs. ln(Relative Intensity of [LMM’H]+) (y-axis). The red line represents the linear trendline of best-fit using Excel. dppm = bis(diphenylphosphino)methane. n = 3.

258 4 + Figure S6 LTQ mass spectra obtained for the MS CID experiments of: (a) [dppmCu2(O2CH)] + + (m/z 555), (b) [dppmAg2(O2CH)] (m/z 645), (c) [dppmAu2(O2CH)] (m/z 823), (d) + + + [dppmCuAg(O2CH)] (m/z 601), (e) [dppmCuAu(O2CH)] (m/z 689), and (e) [dppmAgAu(H)] (m/z 733). A normalized collision energy of 15% was used in all cases. Activation time 10 ms. 8 -3 The concentration of formic acid ([HO2CH]ion trap = 3.5 x 10 molecules.cm ) was kept low so as to mimimize ion-molecule reactions during the CID time. A * represents the mass-selected precursor ion. The amount of the formate precursor ion and the hydride product ion are given as percentages of their peak areas divided by the total ion counts for all ions present in the CID spectrum. The reason that these do not sum to 100% is that there are other fragmentation pathways (see Scheme S1 for all fragmentation pathways).

259 Table S1:

Reactant ion kexpt run1 kexpt run2 kexpt run3 kexpt AVE kADO Reaction (R2) (R2) (R2) efficiency + -9 -9 -9 -9 -9 [dppmCu2H] , 1.20 x 10 1.14 x 10 1.27 x 10 1.20±0.01 x 10 1.06 x 10 113.6±6.1 m/z 511 (0.99) (0.98) (0.99) (0.99) + -11 -11 -11 -11 -9 (a) [dppmAg2H] , 2.01 x 10 1.78 x 10 1.89 x 10 1.89±0.1 x 10 1.05 x 10 1.8±0.1 m/z 601 (1.00) (0.99) (0.99) (1.00) + -9 -9 -9 -9 -9 [dppmAu2H] , 1.45 x 10 1.57 x 10 1.38 x 10 1.47±0.09 x 10 1.04 x 10 141.0±9.2 m/z 779 (0.98) (0.98) (0.98) (0.98) + -10 -10 -10 -10 -9 [dppmCuAgH] 3.20 x 10 3.17 x 10 3.56 x 10 3.31±0.2 x 10 1.05 x 10 31.5±2.1 m/z 557 (0.99) (1.00) (0.99) (1.00) + -10 -10 -10 -10 -9 [dppmCuAuH] 4.34 x 10 5.00 x 10 4.19 x 10 4.51±0.4 x 10 1.05 x 10 43.1±4.1 m/z 645 (0.98) (0.98) (0.99) (0.99) + -11 -11 -11 -11 -9 [dppmAgAuH] 3.25 x 10 3.65 x 10 3.09 x 10 3.33±0.3 x 10 1.04 x 10 3.2±0.3 m/z 689 (0.99) (0.99) (0.98) (0.99) (a) Compares favourable with the previously reported reaction efficiency of 1.4 reported in ref 8.

260 Figure S7 (a) LTQ spectrum obtained in a LTQ 2D linear-ion trap at 298 K for the ion-molecule - 9 -3 reaction of CH3I with [Br] (m/z 79) at 298 K. [CH3I]ion trap = 6.32 x 10 molecules.cm . (b) - Pseudo-first order kinetics studies for the ion-molecule reaction of CH3I with [Br] . This reaction rate is comparable to literature values obtained via flowing afterglow techniques at 298 K (S. Gronert, C. H. De Puy, V. M. Bierbaum, J. Am. Chem. Soc., 1991, 113, 4009-4010.).

261

Scheme S1 Summary of fragmentation channels for MS4 CID experiments (Figure S7) of: (a) + + + [dppmCu2(O2CH)] (m/z 555), (b) [dppmAg2(O2CH)] (m/z 645), (c) [dppmAu2(O2CH)] (m/z + + 823), (d) [dppmCuAg(O2CH)] (m/z 601), (e) [dppmCuAu(O2CH)] (m/z 689), and (e) + [dppmAgAu(H)] (m/z 733).

262 Figure S8 DFT-calculated energy profile for the two reaction steps in the catalytic cycle for + [dppmCu2H] . All structures were fully optimized using DFT method with the hybrid B3LYP functional and def2-TZVP atomic basis set which has been used for all atoms.

263

Figure S9 DFT-calculated energy profile for the two reaction steps in the catalytic cycle for + [dppmAu2H] . All structures were fully optimized using DFT method with the hybrid B3LYP functional and def2-TZVP atomic basis set which has been used for all atoms. Gold atoms have been treated by Stuttgart relativistic effective core potential (RECP) with corresponding AO basis set.

264 Figure S10 DFT-calculated energy profile for the two reaction steps in the catalytic cycle for [dppmCuAuH]+ All structures were fully optimized using DFT method with the hybrid B3LYP functional and def2-TZVP atomic basis set which has been used for all atoms. Gold atoms have been treated by Stuttgart relativistic effective core potential (RECP) with corresponding AO basis set. Yellow = Cu, Orange = Au.

265

Figure S11 DFT-calculated energy profile for the two reaction steps in the catalytic cycle for [dppmAgAuH]+ All structures were fully optimized using DFT method with the hybrid B3LYP functional and def2-TZVP atomic basis set which has been used for all atoms. Silver/Gold atoms have been treated by Stuttgart relativistic effective core potential (RECP) with corresponding AO basis set. Silver = Ag, Orange = Au.

266 Figure S12 DFT calculated Mulliken charges for precursor hydrides and their complexes with formic acid.

267 Figure S12 DFT calculated Mulliken charges for precursor hydrides and their complexes with formic acid (continued).

268 Table S2 DFT (and DFT-D3) -calculated energies (eV) for steps 1 and 2 relative to the initial reactants.

filename Site of IMC TS H2 ΔH H2 TS1 CO2 loss TS2 CO2 loss attack loss loss 1. Ag2HL+HCOOH- Ag -0.24 -0.06 -0.95 1.7 1.86 profile (0.18) 3. Cu2HL+HCOOH- Cu -0.73 -0.66 -1.29 1.93 1.88 profile (0.07) 2. Au2HL+HCOOH- Au -1.08 -1.02 -1.23 1.85 1.45 profile (0.06) 5.CuAgHL+HCOOH- Cu -0.717 -0.25 -1.14 2.14 1.65 profile -0.723 (0.47) Ag -0.538 -0.07 -1.14 2.26 1.81 -0.539 (0.47) 4.CuAuHL+HCOOH- Cu -1.04 -0.3 -1.18 1.88 1.98 profile -1.04 (0.74) 6.AgAuHL+HCOOH- Au -0.248 -0.08 -0.92 2.02 (a) profile -0.259 (0.18) (a) TS2 converged to TS1 during optimisation

+ + + Table S3 Bond lengths (Å) and angles (°) for 1a [LCu2H] ,1b [LAu2H] , 1c [LCuAgH] , 1d [LCuAuH]+, and 1e [LAgAuH]+.

Bond lengths (Å) and angles (°) for 1a – 1e 1a 1b 1c 1d 1e + + + + + [LCu2(H)] [LAu2(H)] [LCuAg(H)] [LCuAu(H)] [LAgAu(H)] (M=M’=Cu) (M=M’=Au) (M=Cu,M’=Ag) (M=Cu,M’=Au) (M=Ag,M’=Au)

M-M’ 2.38 2.66 2.53 2.49 2.68

M-H 1.66 1.85 1.66 1.87 2.26

M’-H 1.66 1.86 1.84 1.72 1.67

M-P 2.27 2.34 2.27 2.29 2.49

M’-P 2.27 2.34 2.46 2.35 2.37

M-H-M’ 91.46 91.69 92.35 87.85 84.55

P-C-P 117.86 123.52 119.27 120.03 120.45

269 Cartesian coordinates for H2: H 0.000000 0.000000 0.132038 H 0.000000 0.000000 0.875962

Cartesian coordinates for CO2: C 0.233157 0.000000 0.062474 O -0.066772 0.000000 1.182882 O 0.533615 0.000000 -1.057791

Cartesian coordinates for cis formic acid: C 0.078760 0.000000 0.003152 O -0.828966 0.000000 -0.996766 O -0.233821 0.000000 1.151405 H 1.124477 0.000000 -0.351879 H -0.379482 0.000000 -1.852606

Cartesian coordinates for trans formic acid: C 0.059488 0.000000 0.009860 O -0.219490 0.000000 1.173505 H 1.078732 0.000000 -0.397128 O -0.828365 0.000000 -0.999448 H -1.719217 0.000000 -0.610906

+ Cartesian coordinates for [dppmCu2(O2CH)] : Cu 0.367121 -1.211861 4.403835 Cu 1.595531 -1.425933 2.375923 P 1.569557 -3.682285 2.141479 P -0.171562 -3.378229 4.808966 C 0.130460 -4.253191 3.186179 H -0.744047 -4.035337 2.568102 H 0.163194 -5.335029 3.323073 C -1.908092 -3.761117 5.210531 C -2.622522 -4.821154 4.645809 C -3.940321 -5.056356 5.020842 C -4.554987 -4.242189 5.964305 C -3.851343 -3.185465 6.532286 C -2.538543 -2.942168 6.153679 H -2.002686 -2.110608 6.596233 H -4.326976 -2.545293 7.263628 H -5.581278 -4.427309 6.253230 H -4.485425 -5.877765 4.574271 H -2.166707 -5.470954 3.910871 C 1.994151 -3.609893 6.542274 C 2.791925 -4.241457 7.488855 H -0.403902 -5.992370 6.194974 C 2.444208 -5.501539 7.959390 C 1.297137 -6.130444 7.483791 C 0.499432 -5.503291 6.536722 H 2.267146 -2.622602 6.189134 H 3.677034 -3.744692 7.864011 H 3.058756 -5.989700 8.704657 H 1.017089 -7.105666 7.860600 C 3.071088 -4.484117 2.780934 C 4.289554 -3.835606 2.552165 C 5.483674 -4.407635 2.969595 C 5.473357 -5.631224 3.629398 C 4.267311 -6.280593 3.868490 C 3.071733 -5.713404 3.446001 H 2.147075 -6.237688 3.645192 H 4.255564 -7.231329 4.385176 H 6.402835 -6.076508 3.959562 H 6.418974 -3.896383 2.783256

270 H 4.308056 -2.880038 2.041138 C 1.198084 -4.431371 0.520085 C 0.357687 -3.726713 -0.348793 C 0.021278 -4.258809 -1.585806 C 0.532564 -5.493013 -1.973294 C 1.379262 -6.192471 -1.121265 C 1.712012 -5.667716 0.122033 H 2.381762 -6.217376 0.769595 H 1.786577 -7.148104 -1.424788 H 0.278663 -5.903796 -2.941756 H -0.628440 -3.705379 -2.251133 H -0.030088 -2.754645 -0.065437 C 0.846285 -4.237676 6.053086 H 1.113842 -0.166944 3.347998

+ Cartesian coordinates for [dppmCu2(H)] : Cu 1.466676 0.702912 8.232320 Cu 3.631070 -0.853881 7.572833 P 0.124651 -0.123784 6.674703 C 1.125989 -0.986076 5.372574 P 2.576478 -1.972936 5.978079 H 1.565285 -0.198219 4.756047 H 0.492310 -1.591614 4.723499 C 3.551732 -2.244601 4.460170 C 4.398651 -1.215082 4.033685 C 5.149249 -1.358554 2.875406 C 5.073346 -2.536716 2.139490 C 4.245966 -3.568818 2.565087 H 2.767200 -5.974584 8.806282 C 3.485506 -3.426984 3.720119 C 2.630794 -4.230207 7.569417 C 1.951676 -3.588022 6.529037 C 0.859756 -4.227418 5.934112 C 0.458524 -5.482084 6.372872 C 1.143916 -6.114475 7.404799 C 2.231069 -5.487871 8.002212 C -0.900023 -1.897174 8.563796 C -1.082019 -1.339742 7.296114 C -1.803869 -2.830499 9.056856 C -2.191187 -1.718905 6.533357 C -2.899502 -3.207443 8.290439 C -3.092865 -2.649728 7.029787 C -0.831432 1.134381 5.766494 C -1.050289 1.092962 4.386884 C -1.386370 2.179342 6.513371 C -2.153323 3.156114 5.894202 C -2.366209 3.107486 4.520628 C -1.812683 2.077960 3.769373 O 2.676490 1.387337 9.526665 C 3.869151 1.007457 9.649132 O 4.460383 0.129662 8.969180 H 4.458294 1.494694 10.434852 H -1.220965 2.232990 7.583021 H -2.577139 3.958970 6.482936 H -2.958255 3.873169 4.036675 H -1.973410 2.038274 2.699890 H -0.632130 0.301185 3.779740 H -2.366976 -1.273598 5.562341 H -0.055584 -1.599802 9.173919 H -1.656958 -3.252376 10.042412 H -3.609941 -3.926025 8.678145 H -3.954087 -2.931418 6.437868

271 H 0.313210 -3.757699 5.127309 H -0.389425 -5.966677 5.906903 H 0.829633 -7.093214 7.743232 H 3.477702 -3.748871 8.044003 H 2.855568 -4.243727 4.044859 H 4.193173 -4.490617 2.000573 H 5.666312 -2.652804 1.241717 H 5.802374 -0.557395 2.555154 H 4.484650 -0.301585 4.611451

+ Cartesian coordinates for [dppmAu2(O2CH)] : O 4.519557 0.314197 10.701821 C 5.490436 -0.413825 10.367733 O 5.571609 -1.248411 9.428769 H 6.393209 -0.305882 10.980961 Au 3.998669 -1.707612 8.102897 Au 2.658647 0.361594 9.715811 P 0.746521 0.297046 8.496589 C 1.100700 -0.529302 6.869643 P 2.171133 -2.048909 6.800897 H 0.166370 -0.727862 6.342196 H 1.651350 0.207701 6.280015 C 2.505829 -2.210713 5.018626 C 3.553459 -1.465792 4.465559 C 3.814792 -1.532348 3.104363 C 3.042364 -2.350508 2.286203 C 2.009409 -3.103135 2.831939 C 1.738295 -3.036237 4.193389 C -0.618196 -0.612455 9.281304 C -0.360267 -1.467678 10.355125 C -1.398968 -2.168424 10.955363 C -2.700289 -2.015781 10.493047 C -2.965445 -1.159594 9.428439 C -1.931477 -0.457854 8.824622 C 0.100731 1.936383 8.043236 C 0.106670 2.925781 9.032795 C -0.402491 4.188099 8.764657 C -0.912957 4.481154 7.504471 C -0.916959 3.506593 6.514232 C -0.415847 2.237467 6.779932 C 1.170035 -3.471458 7.313826 C 1.821859 -4.558335 7.905708 C 1.110049 -5.697793 8.256089 C -0.259361 -5.761258 8.026460 C -0.916742 -4.683294 7.443828 C -0.207991 -3.544279 7.085939 H -1.309260 3.731576 5.531123 H 0.515055 2.712532 10.012962 H -0.390985 4.945482 9.537316 H -1.301464 5.468896 7.293697 H -0.431583 1.497771 5.990958 H -2.155196 0.226185 8.015960 H 0.648553 -1.578810 10.731602 H -1.191127 -2.823749 11.790932 H -3.510383 -2.552899 10.969142 H -3.980525 -1.026980 9.077308 H -0.742157 -2.721013 6.631558 H -1.983162 -4.727936 7.266145 H -0.814452 -6.648185 8.303059 H 1.625737 -6.532904 8.711488 H 2.887221 -4.514455 8.093915 H 0.941501 -3.638982 4.606799

272 H 1.415385 -3.750064 2.199725 H 3.253057 -2.408825 1.226306 H 4.628970 -0.955900 2.685199 H 4.176027 -0.844887 5.098783

+ Cartesian coordinates for [dppmAu2(H)] : Au 0.305286 -1.054949 4.543157 Au 1.708224 -1.279562 2.293193 P 1.639244 -3.613734 2.123898 P -0.183176 -3.320026 4.872145 C 0.183495 -4.105510 3.204651 H -0.678622 -3.850699 2.583088 H 0.182520 -5.192209 3.309603 C -1.909663 -3.789177 5.204847 C -2.573168 -4.833275 4.554612 C -3.885678 -5.144364 4.890972 C -4.544150 -4.423706 5.879488 C -3.890229 -3.383248 6.531292 C -2.583783 -3.062436 6.192404 H -2.088002 -2.240680 6.694977 H -4.400720 -2.813974 7.296901 H -5.566405 -4.667452 6.137507 H -4.391986 -5.951463 4.377721 H -2.084405 -5.412202 3.782951 C 1.966808 -3.546978 6.633856 C 2.760797 -4.196695 7.570910 H -0.372344 -5.968269 6.158279 C 2.434094 -5.481070 7.988488 C 1.310139 -6.117296 7.469848 C 0.514521 -5.472962 6.532545 H 2.218835 -2.540158 6.324754 H 3.626976 -3.694292 7.981006 H 3.046150 -5.982243 8.727137 H 1.044481 -7.111039 7.806457 C 3.123296 -4.446692 2.752969 C 4.363917 -3.919316 2.377217 C 5.539789 -4.542892 2.769722 C 5.490703 -5.692966 3.550158 C 4.263068 -6.217921 3.936179 C 3.082968 -5.601467 3.538887 H 2.141248 -6.028430 3.854365 H 4.220813 -7.108969 4.548657 H 6.407958 -6.176027 3.860783 H 6.493118 -4.127239 2.471095 H 4.411097 -3.019399 1.776174 C 1.200092 -4.391087 0.538183 C 0.513912 -3.631126 -0.413907 C 0.128994 -4.202842 -1.619265 C 0.435575 -5.532320 -1.887717 C 1.128818 -6.290023 -0.950037 C 1.511202 -5.725442 0.260007 H 2.064891 -6.321169 0.973634 H 1.379541 -7.321075 -1.162779 H 0.144106 -5.974469 -2.831502 H -0.398598 -3.606919 -2.352372 H 0.290313 -2.588864 -0.219712 C 0.842342 -4.182609 6.102639 H 1.144660 0.116501 3.379015

+ Cartesian coordinates for [dppmCuAg(O2CH)] : C 0.851634 -4.225722 6.054799

273 C 1.994369 -3.606577 6.565246 C 2.792828 -4.262941 7.494289 C 2.451329 -5.538739 7.924971 C 1.309019 -6.159424 7.427417 C 0.509915 -5.507685 6.498823 P -0.179405 -3.349856 4.835402 C -1.910236 -3.734325 5.255856 C -2.641148 -4.760642 4.651301 C -3.952484 -5.009159 5.040371 C -4.543328 -4.242667 6.037032 C -3.822286 -3.220204 6.645062 C -2.516174 -2.963110 6.253685 Ag 0.283401 -1.032050 4.608200 Cu 1.689610 -1.388763 2.171048 P 1.545870 -3.602163 2.183680 C 1.198415 -4.336209 0.549156 C 0.359874 -3.629917 -0.320479 C 0.042349 -4.151430 -1.566904 C 0.569810 -5.376165 -1.962462 C 1.414492 -6.076762 -1.109412 C 1.729247 -5.562613 0.142923 C 0.101202 -4.180151 3.199575 C 3.040936 -4.403106 2.837563 C 4.255615 -3.726847 2.680939 C 5.444325 -4.302380 3.109471 C 5.431805 -5.557036 3.707333 C 4.228900 -6.234827 3.874171 C 3.039227 -5.664585 3.440433 H -0.771102 -3.939784 2.587017 H 0.127786 -5.264785 3.312155 H -1.968035 -2.157374 6.727648 H -4.279442 -2.616662 7.418072 H -5.564727 -4.437608 6.336555 H -4.510724 -5.803630 4.562475 H -2.204586 -5.374138 3.874868 H -0.390040 -5.991193 6.140478 H 2.261604 -2.606865 6.244946 H 3.673688 -3.772508 7.887197 H 3.066678 -6.045919 8.656733 H 1.033391 -7.147261 7.773251 H 2.117107 -6.212174 3.581132 H 4.215616 -7.210453 4.342078 H 6.356981 -6.005350 4.045403 H 6.376870 -3.769240 2.979344 H 4.276041 -2.746034 2.220604 H 2.398980 -6.112281 0.790244 H 1.835176 -7.024456 -1.419402 H 0.330725 -5.778213 -2.938325 H -0.604827 -3.596243 -2.233171 H -0.038397 -2.663615 -0.032210 O 0.776195 0.987775 4.146370 C 1.396461 1.282123 3.098358 O 1.820960 0.498060 2.205785 H 1.596711 2.348249 2.932114

Cartesian coordinates for [dppmCuAg(H)]+: Ag 0.376003 -0.986389 4.405307 Cu 1.670831 -1.385190 2.270594 P 1.582229 -3.654186 2.137846 P -0.181960 -3.348663 4.819743 C 0.131346 -4.195650 3.184293 H -0.736375 -3.959127 2.563363

274 H 0.146191 -5.279553 3.309672 C -1.911113 -3.772811 5.215613 C -2.630327 -4.790162 4.583037 C -3.942840 -5.057547 4.956372 C -4.547431 -4.319160 5.965665 C -3.838512 -3.305288 6.601990 C -2.532065 -3.029080 6.225278 H -1.993144 -2.230426 6.721565 H -4.305743 -2.723411 7.385620 H -5.569344 -4.528936 6.253201 H -4.490933 -5.845517 4.456363 H -2.183235 -5.383514 3.797151 C 1.993568 -3.593863 6.544649 C 2.794723 -4.235782 7.481333 H -0.402546 -5.974383 6.175436 C 2.449422 -5.501437 7.938620 C 1.301585 -6.125646 7.458988 C 0.500805 -5.488265 6.521340 H 2.264456 -2.601902 6.203214 H 3.680481 -3.742544 7.859597 H 3.066326 -5.997424 8.676740 H 1.023076 -7.105340 7.825244 C 3.069814 -4.471741 2.790699 C 4.297401 -3.839173 2.566460 C 5.482322 -4.424107 2.992460 C 5.453624 -5.644969 3.656511 C 4.238383 -6.278722 3.891352 C 3.052240 -5.698555 3.460286 H 2.120247 -6.210974 3.655969 H 4.212182 -7.227385 4.411309 H 6.375824 -6.100430 3.993160 H 6.424735 -3.924911 2.809280 H 4.330587 -2.885660 2.052341 C 1.196038 -4.428837 0.530650 C 0.375153 -3.721066 -0.354272 C 0.028009 -4.268876 -1.581513 C 0.509041 -5.522775 -1.943402 C 1.336496 -6.225932 -1.075633 C 1.679730 -5.685073 0.157900 H 2.335154 -6.238128 0.817108 H 1.720860 -7.197017 -1.359431 H 0.247021 -5.946123 -2.904265 H -0.606216 -3.712499 -2.259229 H 0.011123 -2.734214 -0.091151 C 0.845251 -4.216693 6.051073 H 1.262571 -0.017377 3.113588

+ Cartesian coordinates for [dppmAgAu(O2CH)] : C -0.204867 -3.551064 7.080611 C 1.167722 -3.452704 7.331766 C 1.824374 -4.517389 7.957537 C 1.123083 -5.660106 8.318854 C -0.240697 -5.748843 8.065838 C -0.903121 -4.693013 7.449111 P 2.152247 -2.022781 6.806416 C 2.497720 -2.201782 5.026930 C 3.511463 -1.419007 4.462553 C 3.783917 -1.502151 3.104615 C 3.057190 -2.374939 2.300897 C 2.058689 -3.164483 2.857939 C 1.776197 -3.081108 4.216357 Au 4.002838 -1.680106 8.081496

275 Ag 2.732721 0.344950 9.785863 P 0.725641 0.293698 8.503645 C 0.082755 1.939591 8.054626 C 0.085902 2.918933 9.054072 C -0.410992 4.188649 8.797292 C -0.904906 4.501148 7.535253 C -0.905232 3.538003 6.533894 C -0.417881 2.261369 6.790078 O 5.647740 -1.293365 9.316624 C 5.623430 -0.477926 10.285733 O 4.679928 0.239376 10.679544 C 1.068450 -0.518435 6.864613 C -0.663409 -0.613134 9.256818 C -0.422258 -1.464185 10.337483 C -1.468355 -2.163438 10.927295 C -2.763169 -2.011304 10.447324 C -3.013112 -1.158556 9.375994 C -1.970673 -0.460198 8.782923 H 6.566596 -0.404659 10.842527 H 0.136084 -0.727907 6.338007 H 1.611780 0.222818 6.273522 H -1.285236 3.777562 5.549328 H 0.479637 2.691692 10.037737 H -0.402653 4.936223 9.579536 H -1.283273 5.494359 7.331988 H -0.432234 1.530914 5.992372 H -2.182771 0.220138 7.967961 H 0.581880 -1.578369 10.727422 H -1.271789 -2.816651 11.767248 H -3.580018 -2.546274 10.914188 H -4.023396 -1.027141 9.010649 H -0.742102 -2.745200 6.599395 H -1.965263 -4.757570 7.252897 H -0.787627 -6.638337 8.350359 H 1.642629 -6.477966 8.800440 H 2.885408 -4.454272 8.163756 H 1.006025 -3.711231 4.639002 H 1.500257 -3.852826 2.237155 H 3.277107 -2.446167 1.243671 H 4.571688 -0.896049 2.676894 H 4.099854 -0.754657 5.084433

Cartesian coordinates for [dppmAgAu(H)]+: Ag 0.160112 -0.878875 4.526463 Au 1.678335 -1.309401 2.359650 P 1.605948 -3.666189 2.160684 P -0.237311 -3.321179 4.807423 C 0.118314 -4.161518 3.176526 H -0.726342 -3.909636 2.530270 H 0.105080 -5.244711 3.309473 C -1.937422 -3.842107 5.213440 C -2.632819 -4.846108 4.535395 C -3.928554 -5.180449 4.914935 C -4.539584 -4.522658 5.974353 C -3.854037 -3.521640 6.655819 C -2.565715 -3.178562 6.273780 H -2.046148 -2.390174 6.806129 H -4.326305 -3.001707 7.478995 H -5.548358 -4.783974 6.265923 H -4.458219 -5.957324 4.379064 H -2.181512 -5.377147 3.708734 C 1.996568 -3.449284 6.468630

276 C 2.843351 -4.036550 7.400420 H -0.359411 -5.890568 6.277559 C 2.540214 -5.288732 7.920922 C 1.389754 -5.954217 7.508560 C 0.543719 -5.372534 6.573943 H 2.235979 -2.467915 6.076639 H 3.731927 -3.511737 7.725802 H 3.192765 -5.741847 8.655923 H 1.144622 -6.923221 7.923811 C 3.053539 -4.545475 2.816704 C 4.309528 -3.975655 2.581313 C 5.464932 -4.621313 3.000235 C 5.378961 -5.837943 3.667401 C 4.134936 -6.407707 3.913832 C 2.977132 -5.768089 3.489713 H 2.021993 -6.231932 3.694544 H 4.063678 -7.352375 4.436858 H 6.279223 -6.339584 3.997777 H 6.430412 -4.171565 2.809263 H 4.385967 -3.024431 2.068823 C 1.216073 -4.389919 0.534054 C 0.551814 -3.591535 -0.402775 C 0.203749 -4.109392 -1.643205 C 0.525967 -5.423614 -1.963178 C 1.198238 -6.218893 -1.041627 C 1.543334 -5.707808 0.203119 H 2.081084 -6.332931 0.903457 H 1.461256 -7.237990 -1.293664 H 0.263247 -5.824161 -2.933620 H -0.306753 -3.483019 -2.362971 H 0.316116 -2.560081 -0.168842 C 0.845141 -4.114596 6.041714 H 1.368541 0.217542 2.960894

+ Cartesian coordinates for [dppmCuAu(O2CH)] : C 0.864021 -4.197047 6.051487 C 2.017640 -3.581233 6.541731 C 2.813667 -4.232109 7.476312 C 2.461155 -5.496894 7.930550 C 1.308859 -6.113264 7.451637 C 0.510202 -5.467703 6.518431 P -0.171236 -3.351420 4.820137 C -1.901315 -3.720657 5.242016 C -2.630160 -4.750209 4.640587 C -3.941499 -4.997514 5.030011 C -4.532989 -4.225743 6.022169 C -3.812926 -3.200163 6.626347 C -2.506118 -2.944717 6.236981 Au 0.254752 -1.132766 4.585158 Cu 1.687424 -1.385791 2.202780 P 1.567092 -3.603373 2.168076 C 1.207958 -4.342850 0.538955 C 0.394365 -3.620805 -0.341161 C 0.065486 -4.145725 -1.583305 C 0.556832 -5.389986 -1.963830 C 1.377254 -6.106772 -1.100375 C 1.703070 -5.589087 0.147507 C 0.118137 -4.177335 3.187948 C 3.056243 -4.415626 2.822092 C 4.275987 -3.750748 2.656658 C 5.461357 -4.333867 3.084188 C 5.440612 -5.584538 3.690155

277 C 4.232651 -6.250875 3.866049 C 3.046264 -5.673071 3.433229 H -0.753288 -3.938147 2.573822 H 0.141985 -5.261208 3.307870 H -1.958300 -2.136241 6.705470 H -4.271469 -2.592663 7.395378 H -5.554818 -4.419053 6.321344 H -4.499368 -5.793967 4.555141 H -2.192856 -5.366219 3.866614 H -0.398121 -5.946518 6.175356 H 2.291152 -2.590134 6.203067 H 3.702487 -3.745218 7.855410 H 3.076105 -5.998584 8.666433 H 1.024489 -7.091945 7.815732 H 2.120248 -6.212149 3.581206 H 4.212791 -7.223306 4.340365 H 6.363252 -6.038586 4.027504 H 6.397931 -3.809579 2.947169 H 4.302389 -2.773072 2.190143 H 2.354423 -6.151515 0.802635 H 1.770322 -7.069826 -1.399097 H 0.309020 -5.795059 -2.936272 H -0.562183 -3.578340 -2.257908 H 0.024804 -2.640086 -0.063782 O 0.665121 0.889832 4.224037 C 1.308843 1.247976 3.198103 O 1.782831 0.521491 2.292902 H 1.465628 2.328371 3.098046

Cartesian coordinates for [dppmCuAu(H)]+: Au 0.470109 -1.164007 4.397488 Cu 1.679997 -1.351009 2.225751 P 1.577763 -3.634913 2.113079 P -0.151084 -3.382074 4.847447 C 0.147043 -4.194484 3.188425 H -0.735459 -3.956039 2.589761 H 0.174604 -5.280410 3.291953 C -1.895756 -3.738249 5.220257 C -2.604109 -4.802782 4.655329 C -3.928166 -5.028819 5.011996 C -4.554088 -4.200948 5.936337 C -3.856088 -3.139855 6.502777 C -2.536263 -2.905060 6.143399 H -2.004344 -2.068355 6.579821 H -4.341486 -2.488669 7.217739 H -5.585635 -4.378808 6.210812 H -4.469685 -5.852915 4.566095 H -2.138657 -5.461889 3.934693 C 1.995391 -3.688683 6.597873 C 2.771434 -4.365204 7.531270 H -0.422436 -6.032016 6.128160 C 2.404736 -5.638777 7.948203 C 1.258732 -6.237316 7.432641 C 0.481075 -5.565626 6.499766 H 2.279582 -2.691088 6.287004 H 3.655201 -3.891962 7.938612 H 3.003271 -6.161228 8.683204 H 0.962991 -7.223029 7.767771 C 3.086006 -4.417239 2.761805 C 4.300764 -3.763628 2.529633 C 5.497913 -4.323636 2.954894 C 5.494030 -5.540803 3.626283

278 C 4.291602 -6.195678 3.869005 C 3.093397 -5.640257 3.438951 H 2.171919 -6.169006 3.640771 H 4.284715 -7.141426 4.394805 H 6.425637 -5.976850 3.962697 H 6.430197 -3.807979 2.765758 H 4.314500 -2.813005 2.009050 C 1.183157 -4.444575 0.525624 C 0.289547 -3.796776 -0.334731 C -0.060426 -4.373244 -1.547690 C 0.490241 -5.595044 -1.920999 C 1.389309 -6.237519 -1.078209 C 1.735622 -5.668711 0.142065 H 2.445165 -6.174902 0.782461 H 1.827292 -7.182978 -1.370689 H 0.225618 -6.039889 -2.871357 H -0.751860 -3.863714 -2.206006 H -0.129814 -2.834063 -0.064284 C 0.848626 -4.286084 6.070136 H 1.191231 0.028787 3.391448

+ Cartesian coordinates for the [dppmCu2(H)+HCO2H] transition state C -2.579111 -3.162666 6.334113 C -1.938018 -3.917690 5.345177 C -2.628253 -4.971193 4.739613 C -3.933598 -5.262741 5.118452 C -4.559734 -4.511123 6.105251 C -3.879640 -3.461695 6.714386 P -0.223793 -3.462334 4.924952 C 0.851825 -4.296383 6.134693 C 1.998241 -3.642017 6.592083 C 2.838810 -4.258399 7.510972 C 2.535177 -5.528271 7.985795 C 1.389224 -6.182697 7.542843 C 0.548533 -5.571568 6.622991 Cu 0.021799 -1.266920 4.877657 O -0.037425 0.585660 5.079822 C 0.186938 1.569627 4.311541 O 0.726837 1.547711 3.189753 Cu 1.793062 -1.423937 2.559596 P 1.546487 -3.646955 2.340862 C 3.023134 -4.536752 2.917609 C 4.262911 -3.924405 2.700540 C 5.439733 -4.570064 3.053462 C 5.391722 -5.831303 3.636655 C 4.164892 -6.444704 3.863465 C 2.985359 -5.804275 3.504648 C 0.082408 -4.267853 3.286358 C 1.210617 -4.209219 0.638850 C 1.635006 -5.454855 0.169195 C 1.316987 -5.855380 -1.122515 C 0.577422 -5.020385 -1.952883 C 0.159733 -3.775609 -1.494814 C 0.479200 -3.368207 -0.206370 H -0.775572 -4.014054 2.659018 H 0.109462 -5.354404 3.374773 H 0.163240 -2.388922 0.135031 H -0.404174 -3.117567 -2.142790 H 0.335297 -5.335430 -2.959548 H 1.651760 -6.819933 -1.481521 H 2.223120 -6.108100 0.799763 H 4.312792 -2.940529 2.248746

279 H 6.391575 -4.086820 2.876319 H 6.308396 -6.334136 3.916009 H 4.123817 -7.425129 4.319652 H 2.044836 -6.304141 3.691188 H -0.353589 -6.079653 6.306647 H 1.143314 -7.165139 7.924645 H 3.182876 -6.004065 8.710684 H 3.721954 -3.741482 7.862717 H 2.234248 -2.645142 6.239486 H -2.064816 -2.335232 6.809040 H -4.364428 -2.869008 7.479073 H -5.576986 -4.738585 6.395847 H -4.460005 -6.078280 4.639789 H -2.164734 -5.572272 3.968894 H 2.359303 0.080021 2.549574 H -0.139271 2.545756 4.698866 H 1.613216 0.500194 2.832705

+ Cartesian coordinates for the [dppmAu2(H)+HCO2H] transition state Au 0.984861 -1.106332 4.426073 Au 1.746174 -1.701682 1.322273 P 0.086317 -3.281170 4.630267 C 0.086074 -4.173904 2.998528 P 1.533209 -3.890003 1.889284 H -0.789725 -3.808846 2.459984 H -0.031312 -5.248697 3.146465 C 1.225700 -4.943748 0.437291 C -0.065090 -5.184143 -0.037668 C -0.256088 -5.930226 -1.195363 C 0.835651 -6.431336 -1.892184 C 2.122176 -6.184866 -1.429684 C 2.320482 -5.444248 -0.272071 H 3.325155 -5.262892 0.079021 H 2.973945 -6.571060 -1.969461 H 0.685164 -7.010743 -2.793862 H -1.257609 -6.116598 -1.552143 H -0.927165 -4.795054 0.480755 C 3.006105 -4.563783 2.707424 C 4.091891 -3.740974 3.013704 C 5.216865 -4.269121 3.637426 C 5.262748 -5.617430 3.960500 C 4.184219 -6.443243 3.660430 C 3.061590 -5.923740 3.035174 H 2.238691 -6.583506 2.791912 H 4.221070 -7.494935 3.909662 H 6.139669 -6.029248 4.442968 H 6.056315 -3.625975 3.865919 H 4.061417 -2.689051 2.764398 C 0.944154 -4.386766 5.802379 C 2.062924 -3.907443 6.487253 C 2.715262 -4.709434 7.415578 C 1.138568 -6.478572 6.995023 C 2.255359 -5.993571 7.669145 C 0.483923 -5.680644 6.069085 H -0.397249 -6.066973 5.573800 H 2.421990 -2.903956 6.299487 H 3.578002 -4.325538 7.943711 H 2.759068 -6.615776 8.397320 H 0.771599 -7.475645 7.200287 C -1.670515 -3.290490 5.115791 C -2.113910 -2.254376 5.941627 C -3.427857 -2.225879 6.390368

280 C -4.316073 -3.226426 6.011806 C -3.885273 -4.257983 5.182698 C -2.569470 -4.292273 4.738709 H -2.259240 -5.106248 4.096819 H -4.572962 -5.037812 4.881859 H -5.341574 -3.200997 6.357320 H -3.757274 -1.417362 7.029464 H -1.432210 -1.464749 6.231974 H 1.559556 0.413751 4.303225 O 2.048152 0.267124 0.532366 C 2.316851 1.351543 1.056548 O 2.294883 1.644605 2.313847 H 2.618664 2.190869 0.425829 H 1.940368 0.900153 2.885452

+ Cartesian coordinates for the [dppmCuAg(H)+HCO2H] transition state Cu 0.643972 -0.622775 -0.061456 Ag 1.535064 -1.334164 2.767647 P 0.849213 -2.701095 -0.790512 C 1.867815 -3.803690 0.296244 P 1.610615 -3.648737 2.121898 H 2.909019 -3.514260 0.135558 H 1.770209 -4.846523 -0.007866 H 1.665211 0.314240 3.425491 O 0.357270 1.197261 0.205956 C 0.095904 1.976960 1.174295 O 0.158529 1.728754 2.390072 H -0.222487 2.987186 0.877240 H 1.048783 0.719849 2.931522 C 3.045204 -4.538794 2.814556 C 2.938691 -5.812538 3.376185 C 4.067513 -6.444675 3.886111 C 5.304989 -5.815489 3.838233 C 5.416698 -4.542680 3.286614 C 4.292446 -3.903754 2.784542 C 0.116058 -4.582483 2.558338 C -0.244243 -5.768293 1.909720 C -1.371995 -6.470625 2.311331 C -2.148599 -6.002552 3.366260 C -1.798255 -4.826470 4.018149 C -0.674006 -4.117669 3.614292 C 1.691412 -2.769937 -2.406809 C 2.563750 -3.796422 -2.780226 C 3.146152 -3.796312 -4.042232 C 2.861507 -2.777529 -4.943424 C 1.995627 -1.751839 -4.580142 C 1.417559 -1.744224 -3.318341 C -0.757249 -3.534375 -1.010523 C -1.887831 -3.012626 -0.377642 C -3.124616 -3.628106 -0.529509 C -3.241998 -4.763851 -1.320593 C -2.121852 -5.284644 -1.962352 C -0.885015 -4.673644 -1.811755 H 0.751610 -0.934179 -3.044986 H 1.775460 -0.952692 -5.275597 H 3.317087 -2.779810 -5.924975 H 3.822435 -4.594329 -4.319718 H 2.801088 -4.602070 -2.098457 H -0.027952 -5.073526 -2.338552 H -1.808412 -2.118377 0.228756 H -3.995318 -3.212669 -0.039419 H -4.206909 -5.237141 -1.448458

281 H -2.214888 -6.160400 -2.591476 H 0.347325 -6.155849 1.091290 H -1.643440 -7.385295 1.800898 H -3.026258 -6.553590 3.678162 H -2.399491 -4.457315 4.838487 H -0.411902 -3.200147 4.127466 H 1.981047 -6.311459 3.427210 H 3.975601 -7.430235 4.323517 H 6.180466 -6.310111 4.238119 H 6.376312 -4.043381 3.258740 H 4.390238 -2.902795 2.378849

+ Cartesian coordinates for the [dppmAgAu(H)+HCO2H] transition state Au 0.689805 -0.526065 -0.113637 Ag 1.455692 -1.338806 2.828010 P 0.857862 -2.687297 -0.793023 C 1.865676 -3.757289 0.324042 P 1.578662 -3.645970 2.152995 H 2.903484 -3.443323 0.190148 H 1.795298 -4.795890 -0.002227 H 1.493066 0.295857 3.461070 O 0.460220 1.507563 0.181574 C 0.255804 2.219632 1.211502 O 0.283490 1.884777 2.410465 H 0.019606 3.270012 0.990941 H 0.960511 0.814580 2.907176 C 3.024391 -4.538637 2.821573 C 2.941734 -5.841028 3.317844 C 4.078789 -6.471515 3.811097 C 5.301438 -5.812173 3.811495 C 5.389226 -4.510911 3.326142 C 4.256140 -3.873865 2.841358 C 0.099924 -4.624833 2.546120 C -0.236138 -5.798023 1.863002 C -1.352981 -6.531536 2.238788 C -2.142135 -6.107794 3.303117 C -1.815777 -4.944537 3.989507 C -0.703237 -4.204111 3.610628 C 1.688417 -2.799890 -2.410548 C 2.665502 -3.753650 -2.707566 C 3.238660 -3.796743 -3.973800 C 2.840561 -2.895075 -4.952285 C 1.867696 -1.943140 -4.664323 C 1.297091 -1.891647 -3.401209 C -0.761022 -3.493561 -0.988799 C -1.884363 -2.980511 -0.336477 C -3.119453 -3.600839 -0.478716 C -3.242542 -4.731248 -1.276946 C -2.128914 -5.243357 -1.936051 C -0.892826 -4.628323 -1.796470 H 0.547723 -1.139641 -3.186620 H 1.558205 -1.234538 -5.421174 H 3.289707 -2.930036 -5.936181 H 3.997152 -4.537027 -4.192180 H 2.993233 -4.469249 -1.966086 H -0.040254 -5.021328 -2.335719 H -1.799288 -2.090262 0.273394 H -3.985702 -3.192644 0.025095 H -4.207337 -5.206956 -1.396423 H -2.225929 -6.114811 -2.570410 H 0.366890 -6.152227 1.037780 H -1.605485 -7.436420 1.701814

282 H -3.010683 -6.683477 3.595477 H -2.426891 -4.609776 4.817300 H -0.459980 -3.295997 4.149101 H 1.995574 -6.363731 3.332504 H 4.004881 -7.479455 4.198085 H 6.183370 -6.305767 4.198295 H 6.336627 -3.988159 3.337207 H 4.333991 -2.851232 2.488944

+ Cartesian coordinates for the [dppmCuAu(H)+HCO2H] transition state Cu 0.186696 -0.894719 0.122666 Au 2.048055 -1.335670 2.468679 P 0.717936 -2.830242 -0.810442 C 1.855996 -3.871771 0.231631 P 1.752580 -3.577780 2.047186 H 2.878500 -3.612184 -0.047123 H 1.718064 -4.934021 0.026270 H 2.743546 0.307812 2.901578 O -0.503504 0.762136 0.632802 C -0.358486 1.431524 1.709031 O 0.349460 1.150624 2.683506 H -0.951631 2.359205 1.740835 H 1.952496 0.535050 2.772099 C 3.080283 -4.599828 2.754245 C 2.813826 -5.670179 3.610251 C 3.862794 -6.418692 4.133090 C 5.176263 -6.106530 3.807928 C 5.447902 -5.034366 2.962451 C 4.407635 -4.279515 2.442911 C 0.157950 -4.203232 2.630726 C -0.329909 -5.442373 2.198463 C -1.538249 -5.921463 2.682567 C -2.269079 -5.173150 3.600688 C -1.791942 -3.942346 4.033860 C -0.583221 -3.456360 3.551090 C 1.609549 -2.647108 -2.390576 C 2.541416 -3.575291 -2.865499 C 3.154116 -3.384983 -4.097700 C 2.840925 -2.271699 -4.869638 C 1.915916 -1.343932 -4.405586 C 1.306598 -1.527023 -3.171298 C -0.735070 -3.880261 -1.148506 C -1.991213 -3.457227 -0.709300 C -3.122096 -4.221945 -0.972901 C -3.006333 -5.413276 -1.676899 C -1.757943 -5.840456 -2.122310 C -0.628144 -5.078009 -1.863839 H 0.594071 -0.792137 -2.815966 H 1.673166 -0.472539 -4.999359 H 3.320758 -2.125785 -5.828548 H 3.876105 -4.107946 -4.454525 H 2.801707 -4.451209 -2.285089 H 0.329937 -5.410427 -2.243407 H -2.092182 -2.526170 -0.164970 H -4.091403 -3.880798 -0.633768 H -3.887218 -6.005285 -1.888972 H -1.667573 -6.761575 -2.683335 H 0.225863 -6.040037 1.487840 H -1.910126 -6.878914 2.342521 H -3.211252 -5.550620 3.976398 H -2.358406 -3.356802 4.745718 H -0.221521 -2.494051 3.890807

283 H 1.796914 -5.920771 3.877457 H 3.648715 -7.244964 4.798168 H 5.989538 -6.690488 4.218500 H 6.470645 -4.780326 2.716889 H 4.635739 -3.431773 1.807430

284 9.4 Appendix D - Supplementary material for Chapter 5

285 286 Electronic Supplementary Material (ESI) for Dalton Transactions. This journal is © The Royal Society of Chemistry 2016

An unusual co-crystal [(µ2-dcpm)Ag2(µ2-O2CH)(η2-NO3)]2·[(µ2- dcpm)2Ag4(µ2-NO3)4] and its connection to the selective decarboxylation of formic acid in the gas phase. Athanasios Zavras,a Jonathan M. Whitea and Richard A. J. O’Hair*a Supplementary Information

Table of contents: Figure S1. A ball-and-stick representation of the asymmetric unit obtained using the Mercury

CSD 3.7 program from the crystallographic information file of [(µ2-dcpm)Ag2(µ2-O2CH)(η2- NO3)]2·[(µ2-dcpm)2Ag4(µ2-NO3)4] (1). Atom colour designation: light grey = silver; dark grey = carbon; orange = phosphorus; red = oxygen; blue = nitrogen. Hydrogen atoms omitted for clarity...... 2 Figure S2. Triplicate experiments for pseudo-first order kinetics obtained in a LTQ 2D + linear-ion trap quasi thermalized to 298 K for the ion-molecule reaction of [LAg2H] , 2b, m/z 625, with formic acid. The black dots are the plotted coordinates for the activation time (s) + (x-axis) vs. ln(Relative Intensity of [LAg2H] ) (y-axis). The red line represents the linear trendline of best-fit using Excel...... 3 Table S1. Peak assignment for ions of low abundance found in the ESI mass spectra shown in Figure 1...... 4

2 Table S2. Intermolecular contacts between [(µ2-dcpm)Ag2(µ2-O2CH)(η -NO3)]2 and [(µ2-

dcpm)2Ag4(µ2-NO3)4]...... 5 Table S3. Crystal data and structure refinement for 3...... 6

Table S4. Atomic coordinates ( x 104) and equivalent isotropic displacement parameters (Å2x 103) for 3. U(eq) is defined as one third of the trace of the orthogonalized Uij tensor. .... 7 Table S5. Bond lengths [Å] and angles [°] for 3...... 11

Table S6. Anisotropic displacement parameters (Å2x 103) for 3. The anisotropic displacement factor exponent takes the form: -2π2[ h2a*2U11 + ... + 2 h k a* b* U12 ] ...... 23

Table S7. Hydrogen coordinates (x 104) and isotropic displacement parameters (Å2x 103) for 3...... 27 Table S8. Torsion angles [°] for 3...... 31

287

Figure S1. A ball-and-stick representation of the asymmetric unit obtained using the Mercury

CSD 3.7 program from the crystallographic information file of [(µ2-dcpm)Ag2(µ2-O2CH)(η2-

NO3)]2·[(µ2-dcpm)2Ag4(µ2-NO3)4] (3). Atom colour designation: light grey = silver; dark grey = carbon; orange = phosphorus; red = oxygen; blue = nitrogen. Hydrogen atoms omitted for clarity.

288 Figure S2. Triplicate experiments for pseudo-first order kinetics obtained in a LTQ 2D + linear-ion trap quasi thermalized to 298 K for the ion-molecule reaction of [LAg2H] , 2b, m/z 625, with formic acid. The black dots are the plotted coordinates for the activation time (s) + (x-axis) vs. ln(Relative Intensity of [LAg2H] ) (y-axis). The red line represents the linear trendline of best-fit using Excel.

289 Table S1. Peak assignment for ions of low abundance found in the ESI mass spectra shown in Figure 1.

m/z a Ion Fig. 1a and/or 1b

515 [LAg]+ 1b

533 [L(=O)Ag]+ b 1a, 1b

+ b 549 [L(=O)2Ag] 1a, 1b

+ 556 [LAgCH3CN] 1b

+ b 572 [L(=O)AgCH3CN] 1a, 1b

+ b 588 [L(=O)2AgCH3CN] 1a, 1b

+ 625 [LAg2H] 1b

+ b 974 [L2(=O)3Ag] 1a, 1b

+ b c 1126 [L2(=O)2Ag2(NO3)] 1a , 1b a The most abundant peak of the isotope cluster is given. The isotope distributions are also consistent with the assignments. b Peaks assigned with components “Lx(=O)y” are likely due to partial (or complete) oxidation of the bisphosphine ligand(s) to the mono (or bis) oxide. c + The presence of [L(=O)2Ag2(NO3)] in Fig. 1a is likely from residual AgNO3 in the ESI sample transfer capillary from prior experiments.

290 2 Table S2. Intermolecular contacts between [(µ2-dcpm)Ag2(µ2-O2CH)(η -NO3)]2 and [(µ2- dcpm)2Ag4(µ2-NO3)4].

Atom1 Atom2 Symm. op. 1 Symm. op. 2 Length (esd’s?)

H(73B) H(51B) x,y,z x,-1+y,z 2.383 H(60A) H(44A) x,y,z x,-1+y,z 2.283 H(52) H(50A) x,y,z 1-x,-1/2+y,1/2-z 2.37 H(72A) O(13) x,y,z 1-x,-1/2+y,1/2-z 2.707 H(72B) H(54A) x,y,z 1-x,-1/2+y,1/2-z 2.374 H(65) O(14) x,y,z 1-x,-1/2+y,1/2-z 2.414 H(64A) O(14) x,y,z 1-x,-1/2+y,1/2-z 2.646 H(61A) O(8) x,y,z x,-1/2-y,-1/2+z 2.515 O(11) H(8B) x,y,z x,y,z 2.54 H(35A) H(19B) x,y,z x,y,z 2.287 H(39A) H(3A) x,y,z x,1+y,z 2.303 H(39B) O(5) x,y,z x,1+y,z 2.41 H(33) O(5) x,y,z x,1+y,z 2.632 H(45A) O(4) x,y,z x,1+y,z 2.502 H(35A) O(3) x,y,z -x,1/2+y,1/2-z 2.582 C(57) H(9B) x,y,z x,-1/2-y,-1/2+z 2.886 H(29A) H(24A) x,y,z -x,1/2+y,1/2-z 2.357 O(16) H(6A) x,y,z x,-1/2-y,-1/2+z 2.538 O(16) H(9A) x,y,z x,-1/2-y,-1/2+z 2.685 O(5) H(17B) x,y,z -x,-1/2+y,1/2-z 2.65

291 Table S3. Crystal data and structure refinement for 3.

Identification code shelx Empirical formula C76 H139 Ag6 N5 O17 P6 Formula weight 2227.95 Temperature 130.0(2) K Wavelength 1.5418 Å Crystal system Monoclinic Space group P 21/c Unit cell dimensions a = 24.6390(2) Å α= 90°. b = 16.08250(10) Å β= 94.7980(10)°. c = 23.41430(10) Å γ = 90°. Volume 9245.56(10) Å3 Z 4 Density (calculated) 1.601 Mg/m3 Absorption coefficient 11.484 mm-1 F(000) 4552 Crystal size 0.4939 x 0.0728 x 0.0600 mm3 Theta range for data collection 3.285 to 77.202°. Index ranges -30<=h<=31, -20<=k<=20, -29<=l<=27 Reflections collected 99896 Independent reflections 19442 [R(int) = 0.0694] Completeness to theta = 67.684° 100.0 % Absorption correction Gaussian Max. and min. transmission 0.560 and 0.100 Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 19442 / 0 / 995 Goodness-of-fit on F2 1.039 Final R indices [I>2sigma(I)] R1 = 0.0449, wR2 = 0.1177 R indices (all data) R1 = 0.0506, wR2 = 0.1226 Extinction coefficient n/a Largest diff. peak and hole 1.873 and -1.681 e.Å-3

292 Table S4. Atomic coordinates ( x 104) and equivalent isotropic displacement parameters (Å2x 103) for 3. U(eq) is defined as one third of the trace of the orthogonalized Uij tensor.

______x y z U(eq) ______C(1) 2030(2) -5269(2) 4605(2) 25(1) C(2) 1871(2) -6046(3) 4250(2) 30(1) C(3) 2272(2) -6752(3) 4375(2) 35(1) C(4) 2337(2) -6965(3) 5009(2) 38(1) C(5) 2513(2) -6198(3) 5359(2) 40(1) C(6) 2106(2) -5486(3) 5247(2) 32(1) C(7) 1743(2) -3527(3) 4875(2) 36(1) C(8) 2315(2) -3249(3) 4743(2) 39(1) C(9) 2489(3) -2458(4) 5071(3) 63(2) C(10) 2092(3) -1770(4) 4961(3) 64(2) C(11) 1520(3) -2052(4) 5117(3) 65(2) C(12) 1347(2) -2825(3) 4771(3) 52(1) C(13) 1480(2) -4182(3) 3722(2) 29(1) C(14) 338(2) -3687(3) 3648(2) 29(1) C(15) -255(2) -4002(3) 3574(2) 34(1) C(16) -631(2) -3447(4) 3897(2) 45(1) C(17) -587(2) -2544(4) 3721(2) 48(1) C(18) 0(2) -2236(3) 3793(2) 45(1) C(19) 371(2) -2778(3) 3458(2) 38(1) C(20) 903(2) -4016(3) 2604(2) 32(1) C(21) 1370(2) -4469(3) 2343(2) 41(1) C(22) 1426(2) -4162(4) 1729(2) 50(1) C(23) 897(3) -4240(4) 1353(2) 60(2) C(24) 446(2) -3771(5) 1618(2) 62(2) C(25) 370(2) -4096(4) 2222(2) 49(1) C(26) -79(3) -6483(4) 4447(3) 59(2) C(27) 1543(2) 227(3) 4099(2) 34(1) C(28) 1049(2) 804(3) 4016(2) 42(1) C(29) 733(2) 798(5) 4556(2) 58(2) C(30) 1099(3) 1056(4) 5086(2) 59(2) C(31) 1599(2) 509(4) 5166(2) 55(1) C(32) 1915(2) 493(3) 4627(2) 41(1) C(33) 1456(2) -144(2) 2851(2) 30(1) 7

293 C(34) 1124(2) -921(3) 2967(2) 43(1) C(35) 766(2) -1174(4) 2429(3) 57(2) C(36) 1106(3) -1335(4) 1928(3) 60(2) C(37) 1427(2) -563(3) 1797(2) 46(1) C(38) 1783(2) -277(3) 2321(2) 36(1) C(39) 2128(2) 1205(2) 3320(2) 26(1) C(40) 2636(2) 2256(2) 2512(2) 26(1) C(41) 2244(2) 2024(3) 1997(2) 31(1) C(42) 2156(2) 2759(3) 1581(2) 39(1) C(43) 1960(2) 3524(3) 1886(2) 39(1) C(44) 2355(2) 3756(3) 2399(2) 39(1) C(45) 2438(2) 3029(2) 2815(2) 28(1) C(46) 3245(2) 1726(2) 3586(2) 26(1) C(47) 3408(2) 1010(3) 3996(2) 34(1) C(48) 3812(2) 1288(4) 4483(2) 46(1) C(49) 4318(2) 1685(4) 4258(2) 47(1) C(50) 4156(2) 2403(3) 3860(2) 41(1) C(51) 3757(2) 2118(3) 3361(2) 31(1) C(52) 4275(2) -2408(2) 1159(2) 28(1) C(53) 4441(2) -1520(3) 1350(2) 34(1) C(54) 4623(2) -1000(3) 851(2) 43(1) C(55) 4183(2) -973(3) 356(2) 48(1) C(56) 4028(2) -1851(3) 162(2) 47(1) C(57) 3843(2) -2373(3) 654(2) 32(1) C(58) 3789(2) -4024(3) 1537(2) 30(1) C(59) 3196(2) -4055(3) 1291(2) 36(1) C(60) 3003(2) -4952(4) 1215(2) 49(1) C(61) 3371(2) -5446(3) 847(2) 48(1) C(62) 3974(3) -5392(3) 1071(2) 51(1) C(63) 4160(2) -4487(3) 1149(2) 35(1) C(64) 4626(2) -3165(3) 2253(2) 28(1) C(65) 5180(1) -3338(2) 3383(2) 24(1) C(66) 5586(2) -2670(3) 3222(2) 30(1) C(67) 6145(2) -2783(3) 3545(2) 40(1) C(68) 6102(2) -2804(3) 4192(2) 39(1) C(69) 5712(2) -3483(3) 4349(2) 41(1) C(70) 5149(2) -3372(3) 4035(2) 32(1) C(71) 4040(2) -3978(2) 3113(2) 26(1) 8

294 C(72) 4297(2) -4838(3) 3064(2) 34(1) C(73) 3875(3) -5524(3) 3106(2) 52(1) C(74) 3598(3) -5446(4) 3659(3) 62(2) C(75) 3335(2) -4599(4) 3703(2) 57(2) C(76) 3745(2) -3891(3) 3663(2) 37(1) N(1) 855(1) -7409(2) 2943(1) 30(1) N(2) 3684(2) -1175(3) 4353(2) 50(1) N(3) 2472(2) -2554(2) 2983(2) 41(1) N(4) 4354(2) -190(3) 2789(2) 42(1) N(5) 2994(2) -590(3) 1425(2) 47(1) O(1) -37(1) -5826(2) 4747(2) 43(1) O(2) 160(2) -6670(3) 4034(2) 66(1) O(3) 456(1) -6967(2) 2763(2) 45(1) O(4) 1173(1) -7127(2) 3344(1) 39(1) O(5) 939(1) -8091(2) 2725(1) 41(1) O(6) 3447(1) -1409(2) 3874(1) 45(1) O(7) 4202(2) -1272(4) 4395(3) 94(2) O(8) 3418(2) -859(3) 4704(2) 73(1) O(9) 2785(1) -1947(2) 2878(2) 42(1) O(10) 2508(2) -3192(3) 2710(3) 86(2) O(11) 2160(2) -2452(3) 3362(2) 59(1) O(12) 3918(1) -596(2) 2850(2) 55(1) O(13) 4793(2) -500(3) 2964(3) 77(1) O(14) 4310(2) 487(3) 2557(2) 62(1) O(15) 2976(2) -900(2) 1915(2) 55(1) O(16) 2879(2) -1013(3) 1001(2) 90(2) O(17) 3146(3) 140(3) 1404(2) 82(2) P(1) 1497(1) -4473(1) 4487(1) 24(1) P(2) 815(1) -4410(1) 3332(1) 25(1) P(3) 1937(1) 125(1) 3467(1) 25(1) P(4) 2775(1) 1353(1) 2987(1) 22(1) P(5) 4012(1) -2969(1) 1766(1) 23(1) P(6) 4505(1) -3101(1) 3015(1) 21(1) Ag(1) 658(1) -5017(1) 4701(1) 33(1) Ag(2) 602(1) -5829(1) 3436(1) 34(1) Ag(3) 2653(1) -840(1) 3579(1) 33(1) Ag(4) 3154(1) 226(1) 2531(1) 34(1) Ag(5) 3338(1) -2226(1) 2192(1) 35(1) 9

295 Ag(6) 4139(1) -1851(1) 3303(1) 39(1) ______

296 Table S5. Bond lengths [Å] and angles [°] for 3.

______C(1)-C(2) 1.534(5) C(12)-H(12B) 0.9700 C(1)-C(6) 1.540(5) C(13)-P(2) 1.845(4) C(1)-P(1) 1.840(4) C(13)-P(1) 1.847(4) C(1)-H(1) 0.9800 C(13)-H(13A) 0.9700 C(2)-C(3) 1.517(5) C(13)-H(13B) 0.9700 C(2)-H(2A) 0.9700 C(14)-C(19) 1.532(6) C(2)-H(2B) 0.9700 C(14)-C(15) 1.543(5) C(3)-C(4) 1.519(6) C(14)-P(2) 1.851(4) C(3)-H(3A) 0.9700 C(14)-H(14) 0.9800 C(3)-H(3B) 0.9700 C(15)-C(16) 1.531(6) C(4)-C(5) 1.523(6) C(15)-H(15A) 0.9700 C(4)-H(4A) 0.9700 C(15)-H(15B) 0.9700 C(4)-H(4B) 0.9700 C(16)-C(17) 1.515(8) C(5)-C(6) 1.531(6) C(16)-H(16A) 0.9700 C(5)-H(5A) 0.9700 C(16)-H(16B) 0.9700 C(5)-H(5B) 0.9700 C(17)-C(18) 1.523(8) C(6)-H(6A) 0.9700 C(17)-H(17A) 0.9700 C(6)-H(6B) 0.9700 C(17)-H(17B) 0.9700 C(7)-C(12) 1.498(7) C(18)-C(19) 1.528(6) C(7)-C(8) 1.533(6) C(18)-H(18A) 0.9700 C(7)-P(1) 1.848(4) C(18)-H(18B) 0.9700 C(7)-H(7) 0.9800 C(19)-H(19A) 0.9700 C(8)-C(9) 1.528(7) C(19)-H(19B) 0.9700 C(8)-H(8A) 0.9700 C(20)-C(21) 1.531(6) C(8)-H(8B) 0.9700 C(20)-C(25) 1.531(6) C(9)-C(10) 1.485(9) C(20)-P(2) 1.847(4) C(9)-H(9A) 0.9700 C(20)-H(20) 0.9800 C(9)-H(9B) 0.9700 C(21)-C(22) 1.538(7) C(10)-C(11) 1.553(10) C(21)-H(21A) 0.9700 C(10)-H(10A) 0.9700 C(21)-H(21B) 0.9700 C(10)-H(10B) 0.9700 C(22)-C(23) 1.514(9) C(11)-C(12) 1.526(8) C(22)-H(22A) 0.9700 C(11)-H(11A) 0.9700 C(22)-H(22B) 0.9700 C(11)-H(11B) 0.9700 C(23)-C(24) 1.519(9) C(12)-H(12A) 0.9700 C(23)-H(23A) 0.9700

297 C(23)-H(23B) 0.9700 C(36)-H(36A) 0.9700 C(24)-C(25) 1.532(7) C(36)-H(36B) 0.9700 C(24)-H(24A) 0.9700 C(37)-C(38) 1.519(6) C(24)-H(24B) 0.9700 C(37)-H(37A) 0.9700 C(25)-H(25A) 0.9700 C(37)-H(37B) 0.9700 C(25)-H(25B) 0.9700 C(38)-H(38A) 0.9700 C(26)-O(2) 1.212(7) C(38)-H(38B) 0.9700 C(26)-O(1) 1.269(6) C(39)-P(3) 1.841(4) C(26)-H 1.11(8) C(39)-P(4) 1.846(4) C(27)-C(28) 1.530(6) C(39)-H(39A) 0.9700 C(27)-C(32) 1.539(6) C(39)-H(39B) 0.9700 C(27)-P(3) 1.844(4) C(40)-C(41) 1.528(5) C(27)-H(27) 0.9800 C(40)-C(45) 1.532(5) C(28)-C(29) 1.540(6) C(40)-P(4) 1.843(4) C(28)-H(28A) 0.9700 C(40)-H(40) 0.9800 C(28)-H(28B) 0.9700 C(41)-C(42) 1.536(6) C(29)-C(30) 1.529(9) C(41)-H(41A) 0.9700 C(29)-H(29A) 0.9700 C(41)-H(41B) 0.9700 C(29)-H(29B) 0.9700 C(42)-C(43) 1.523(7) C(30)-C(31) 1.511(8) C(42)-H(42A) 0.9700 C(30)-H(30A) 0.9700 C(42)-H(42B) 0.9700 C(30)-H(30B) 0.9700 C(43)-C(44) 1.527(6) C(31)-C(32) 1.538(7) C(43)-H(43A) 0.9700 C(31)-H(31A) 0.9700 C(43)-H(43B) 0.9700 C(31)-H(31B) 0.9700 C(44)-C(45) 1.524(5) C(32)-H(32A) 0.9700 C(44)-H(44A) 0.9700 C(32)-H(32B) 0.9700 C(44)-H(44B) 0.9700 C(33)-C(34) 1.531(6) C(45)-H(45A) 0.9700 C(33)-C(38) 1.549(6) C(45)-H(45B) 0.9700 C(33)-P(3) 1.839(4) C(46)-C(47) 1.532(5) C(33)-H(33) 0.9800 C(46)-C(51) 1.542(5) C(34)-C(35) 1.531(7) C(46)-P(4) 1.842(4) C(34)-H(34A) 0.9700 C(46)-H(46) 0.9800 C(34)-H(34B) 0.9700 C(47)-C(48) 1.517(6) C(35)-C(36) 1.521(9) C(47)-H(47A) 0.9700 C(35)-H(35A) 0.9700 C(47)-H(47B) 0.9700 C(35)-H(35B) 0.9700 C(48)-C(49) 1.532(7) C(36)-C(37) 1.517(8) C(48)-H(48A) 0.9700 12

298 C(48)-H(48B) 0.9700 C(61)-H(61A) 0.9700 C(49)-C(50) 1.517(7) C(61)-H(61B) 0.9700 C(49)-H(49A) 0.9700 C(62)-C(63) 1.532(6) C(49)-H(49B) 0.9700 C(62)-H(62A) 0.9700 C(50)-C(51) 1.533(6) C(62)-H(62B) 0.9700 C(50)-H(50A) 0.9700 C(63)-H(63A) 0.9700 C(50)-H(50B) 0.9700 C(63)-H(63B) 0.9700 C(51)-H(51A) 0.9700 C(64)-P(6) 1.837(4) C(51)-H(51B) 0.9700 C(64)-P(5) 1.843(4) C(52)-C(57) 1.525(5) C(64)-H(64A) 0.9700 C(52)-C(53) 1.542(5) C(64)-H(64B) 0.9700 C(52)-P(5) 1.846(4) C(65)-C(70) 1.535(5) C(52)-H(52) 0.9800 C(65)-C(66) 1.537(5) C(53)-C(54) 1.534(6) C(65)-P(6) 1.848(4) C(53)-H(53A) 0.9700 C(65)-H(65) 0.9800 C(53)-H(53B) 0.9700 C(66)-C(67) 1.525(6) C(54)-C(55) 1.521(7) C(66)-H(66A) 0.9700 C(54)-H(54A) 0.9700 C(66)-H(66B) 0.9700 C(54)-H(54B) 0.9700 C(67)-C(68) 1.529(7) C(55)-C(56) 1.522(7) C(67)-H(67A) 0.9700 C(55)-H(55A) 0.9700 C(67)-H(67B) 0.9700 C(55)-H(55B) 0.9700 C(68)-C(69) 1.519(7) C(56)-C(57) 1.525(6) C(68)-H(68A) 0.9700 C(56)-H(56A) 0.9700 C(68)-H(68B) 0.9700 C(56)-H(56B) 0.9700 C(69)-C(70) 1.526(6) C(57)-H(57A) 0.9700 C(69)-H(69A) 0.9700 C(57)-H(57B) 0.9700 C(69)-H(69B) 0.9700 C(58)-C(59) 1.526(6) C(70)-H(70A) 0.9700 C(58)-C(63) 1.534(6) C(70)-H(70B) 0.9700 C(58)-P(5) 1.849(4) C(71)-C(72) 1.530(6) C(58)-H(58) 0.9800 C(71)-C(76) 1.538(5) C(59)-C(60) 1.526(7) C(71)-P(6) 1.843(4) C(59)-H(59A) 0.9700 C(71)-H(71) 0.9800 C(59)-H(59B) 0.9700 C(72)-C(73) 1.525(6) C(60)-C(61) 1.526(8) C(72)-H(72A) 0.9700 C(60)-H(60A) 0.9700 C(72)-H(72B) 0.9700 C(60)-H(60B) 0.9700 C(73)-C(74) 1.518(8) C(61)-C(62) 1.535(8) C(73)-H(73A) 0.9700 13

299 C(73)-H(73B) 0.9700 P(2)-Ag(2) 2.3591(10) C(74)-C(75) 1.515(10) P(3)-Ag(3) 2.3473(10) C(74)-H(74A) 0.9700 P(4)-Ag(4) 2.3380(9) C(74)-H(74B) 0.9700 P(5)-Ag(5) 2.3368(9) C(75)-C(76) 1.529(6) P(6)-Ag(6) 2.3255(9) C(75)-H(75A) 0.9700 Ag(1)-O(1)#1 2.485(3) C(75)-H(75B) 0.9700 Ag(1)-Ag(2) 3.2294(4) C(76)-H(76A) 0.9700 Ag(3)-Ag(4) 3.3148(4) C(76)-H(76B) 0.9700 Ag(5)-Ag(6) 3.1892(4) N(1)-O(5) 1.233(5) N(1)-O(4) 1.256(5) C(2)-C(1)-C(6) 110.4(3) N(1)-O(3) 1.257(5) C(2)-C(1)-P(1) 109.8(3) N(2)-O(8) 1.204(6) C(6)-C(1)-P(1) 109.3(3) N(2)-O(6) 1.277(6) C(2)-C(1)-H(1) 109.1 N(2)-O(7) 1.281(7) C(6)-C(1)-H(1) 109.1 N(3)-O(10) 1.216(6) P(1)-C(1)-H(1) 109.1 N(3)-O(11) 1.232(5) C(3)-C(2)-C(1) 111.8(3) N(3)-O(9) 1.280(5) C(3)-C(2)-H(2A) 109.3 N(4)-O(14) 1.218(6) C(1)-C(2)-H(2A) 109.3 N(4)-O(13) 1.231(6) C(3)-C(2)-H(2B) 109.3 N(4)-O(12) 1.275(5) C(1)-C(2)-H(2B) 109.3 N(5)-O(16) 1.217(6) H(2A)-C(2)-H(2B) 107.9 N(5)-O(17) 1.235(7) C(2)-C(3)-C(4) 111.8(3) N(5)-O(15) 1.257(6) C(2)-C(3)-H(3A) 109.3 O(1)-Ag(1) 2.161(3) C(4)-C(3)-H(3A) 109.3 O(1)-Ag(1)#1 2.485(3) C(2)-C(3)-H(3B) 109.3 O(2)-Ag(2) 2.288(4) C(4)-C(3)-H(3B) 109.3 O(3)-Ag(2) 2.423(3) H(3A)-C(3)-H(3B) 107.9 O(4)-Ag(2) 2.536(3) C(3)-C(4)-C(5) 110.2(4) O(6)-Ag(3) 2.217(3) C(3)-C(4)-H(4A) 109.6 O(6)-Ag(6) 2.362(3) C(5)-C(4)-H(4A) 109.6 O(9)-Ag(5) 2.236(3) C(3)-C(4)-H(4B) 109.6 O(9)-Ag(3) 2.462(3) C(5)-C(4)-H(4B) 109.6 O(12)-Ag(6) 2.324(4) H(4A)-C(4)-H(4B) 108.1 O(12)-Ag(4) 2.367(3) C(4)-C(5)-C(6) 111.1(4) O(15)-Ag(4) 2.333(4) C(4)-C(5)-H(5A) 109.4 O(15)-Ag(5) 2.379(4) C(6)-C(5)-H(5A) 109.4 P(1)-Ag(1) 2.3364(9) C(4)-C(5)-H(5B) 109.4 14

300 C(6)-C(5)-H(5B) 109.4 C(7)-C(12)-C(11) 112.2(5) H(5A)-C(5)-H(5B) 108.0 C(7)-C(12)-H(12A) 109.2 C(5)-C(6)-C(1) 111.1(3) C(11)-C(12)-H(12A) 109.2 C(5)-C(6)-H(6A) 109.4 C(7)-C(12)-H(12B) 109.2 C(1)-C(6)-H(6A) 109.4 C(11)-C(12)-H(12B) 109.2 C(5)-C(6)-H(6B) 109.4 H(12A)-C(12)-H(12B) 107.9 C(1)-C(6)-H(6B) 109.4 P(2)-C(13)-P(1) 112.2(2) H(6A)-C(6)-H(6B) 108.0 P(2)-C(13)-H(13A) 109.2 C(12)-C(7)-C(8) 110.1(4) P(1)-C(13)-H(13A) 109.2 C(12)-C(7)-P(1) 110.9(3) P(2)-C(13)-H(13B) 109.2 C(8)-C(7)-P(1) 114.3(3) P(1)-C(13)-H(13B) 109.2 C(12)-C(7)-H(7) 107.1 H(13A)-C(13)-H(13B) 107.9 C(8)-C(7)-H(7) 107.1 C(19)-C(14)-C(15) 110.7(3) P(1)-C(7)-H(7) 107.1 C(19)-C(14)-P(2) 115.6(3) C(9)-C(8)-C(7) 111.5(5) C(15)-C(14)-P(2) 112.1(3) C(9)-C(8)-H(8A) 109.3 C(19)-C(14)-H(14) 105.9 C(7)-C(8)-H(8A) 109.3 C(15)-C(14)-H(14) 105.9 C(9)-C(8)-H(8B) 109.3 P(2)-C(14)-H(14) 105.9 C(7)-C(8)-H(8B) 109.3 C(16)-C(15)-C(14) 111.0(4) H(8A)-C(8)-H(8B) 108.0 C(16)-C(15)-H(15A) 109.4 C(10)-C(9)-C(8) 112.3(5) C(14)-C(15)-H(15A) 109.4 C(10)-C(9)-H(9A) 109.2 C(16)-C(15)-H(15B) 109.4 C(8)-C(9)-H(9A) 109.2 C(14)-C(15)-H(15B) 109.4 C(10)-C(9)-H(9B) 109.2 H(15A)-C(15)-H(15B) 108.0 C(8)-C(9)-H(9B) 109.2 C(17)-C(16)-C(15) 111.4(4) H(9A)-C(9)-H(9B) 107.9 C(17)-C(16)-H(16A) 109.3 C(9)-C(10)-C(11) 109.8(5) C(15)-C(16)-H(16A) 109.3 C(9)-C(10)-H(10A) 109.7 C(17)-C(16)-H(16B) 109.3 C(11)-C(10)-H(10A) 109.7 C(15)-C(16)-H(16B) 109.3 C(9)-C(10)-H(10B) 109.7 H(16A)-C(16)-H(16B) 108.0 C(11)-C(10)-H(10B) 109.7 C(16)-C(17)-C(18) 111.7(4) H(10A)-C(10)-H(10B) 108.2 C(16)-C(17)-H(17A) 109.3 C(12)-C(11)-C(10) 109.3(5) C(18)-C(17)-H(17A) 109.3 C(12)-C(11)-H(11A) 109.8 C(16)-C(17)-H(17B) 109.3 C(10)-C(11)-H(11A) 109.8 C(18)-C(17)-H(17B) 109.3 C(12)-C(11)-H(11B) 109.8 H(17A)-C(17)-H(17B) 107.9 C(10)-C(11)-H(11B) 109.8 C(17)-C(18)-C(19) 111.1(4) H(11A)-C(11)-H(11B) 108.3 C(17)-C(18)-H(18A) 109.4 15

301 C(19)-C(18)-H(18A) 109.4 C(25)-C(24)-H(24B) 109.4 C(17)-C(18)-H(18B) 109.4 H(24A)-C(24)-H(24B) 108.0 C(19)-C(18)-H(18B) 109.4 C(20)-C(25)-C(24) 110.2(4) H(18A)-C(18)-H(18B) 108.0 C(20)-C(25)-H(25A) 109.6 C(18)-C(19)-C(14) 110.2(4) C(24)-C(25)-H(25A) 109.6 C(18)-C(19)-H(19A) 109.6 C(20)-C(25)-H(25B) 109.6 C(14)-C(19)-H(19A) 109.6 C(24)-C(25)-H(25B) 109.6 C(18)-C(19)-H(19B) 109.6 H(25A)-C(25)-H(25B) 108.1 C(14)-C(19)-H(19B) 109.6 O(2)-C(26)-O(1) 128.8(5) H(19A)-C(19)-H(19B) 108.1 O(2)-C(26)-H 117(4) C(21)-C(20)-C(25) 111.4(4) O(1)-C(26)-H 114(4) C(21)-C(20)-P(2) 110.8(3) C(28)-C(27)-C(32) 110.4(4) C(25)-C(20)-P(2) 110.4(3) C(28)-C(27)-P(3) 114.9(3) C(21)-C(20)-H(20) 108.0 C(32)-C(27)-P(3) 110.6(3) C(25)-C(20)-H(20) 108.0 C(28)-C(27)-H(27) 106.9 P(2)-C(20)-H(20) 108.0 C(32)-C(27)-H(27) 106.9 C(20)-C(21)-C(22) 110.4(4) P(3)-C(27)-H(27) 106.9 C(20)-C(21)-H(21A) 109.6 C(27)-C(28)-C(29) 110.1(4) C(22)-C(21)-H(21A) 109.6 C(27)-C(28)-H(28A) 109.6 C(20)-C(21)-H(21B) 109.6 C(29)-C(28)-H(28A) 109.6 C(22)-C(21)-H(21B) 109.6 C(27)-C(28)-H(28B) 109.6 H(21A)-C(21)-H(21B) 108.1 C(29)-C(28)-H(28B) 109.6 C(23)-C(22)-C(21) 112.1(4) H(28A)-C(28)-H(28B) 108.2 C(23)-C(22)-H(22A) 109.2 C(30)-C(29)-C(28) 111.1(5) C(21)-C(22)-H(22A) 109.2 C(30)-C(29)-H(29A) 109.4 C(23)-C(22)-H(22B) 109.2 C(28)-C(29)-H(29A) 109.4 C(21)-C(22)-H(22B) 109.2 C(30)-C(29)-H(29B) 109.4 H(22A)-C(22)-H(22B) 107.9 C(28)-C(29)-H(29B) 109.4 C(22)-C(23)-C(24) 110.2(4) H(29A)-C(29)-H(29B) 108.0 C(22)-C(23)-H(23A) 109.6 C(31)-C(30)-C(29) 111.3(5) C(24)-C(23)-H(23A) 109.6 C(31)-C(30)-H(30A) 109.4 C(22)-C(23)-H(23B) 109.6 C(29)-C(30)-H(30A) 109.4 C(24)-C(23)-H(23B) 109.6 C(31)-C(30)-H(30B) 109.4 H(23A)-C(23)-H(23B) 108.1 C(29)-C(30)-H(30B) 109.4 C(23)-C(24)-C(25) 111.0(5) H(30A)-C(30)-H(30B) 108.0 C(23)-C(24)-H(24A) 109.4 C(30)-C(31)-C(32) 111.8(4) C(25)-C(24)-H(24A) 109.4 C(30)-C(31)-H(31A) 109.3 C(23)-C(24)-H(24B) 109.4 C(32)-C(31)-H(31A) 109.3 16

302 C(30)-C(31)-H(31B) 109.3 H(37A)-C(37)-H(37B) 108.0 C(32)-C(31)-H(31B) 109.3 C(37)-C(38)-C(33) 112.7(4) H(31A)-C(31)-H(31B) 107.9 C(37)-C(38)-H(38A) 109.0 C(31)-C(32)-C(27) 110.8(4) C(33)-C(38)-H(38A) 109.0 C(31)-C(32)-H(32A) 109.5 C(37)-C(38)-H(38B) 109.0 C(27)-C(32)-H(32A) 109.5 C(33)-C(38)-H(38B) 109.0 C(31)-C(32)-H(32B) 109.5 H(38A)-C(38)-H(38B) 107.8 C(27)-C(32)-H(32B) 109.5 P(3)-C(39)-P(4) 116.3(2) H(32A)-C(32)-H(32B) 108.1 P(3)-C(39)-H(39A) 108.2 C(34)-C(33)-C(38) 110.5(4) P(4)-C(39)-H(39A) 108.2 C(34)-C(33)-P(3) 111.8(3) P(3)-C(39)-H(39B) 108.2 C(38)-C(33)-P(3) 108.4(3) P(4)-C(39)-H(39B) 108.2 C(34)-C(33)-H(33) 108.7 H(39A)-C(39)-H(39B) 107.4 C(38)-C(33)-H(33) 108.7 C(41)-C(40)-C(45) 111.0(3) P(3)-C(33)-H(33) 108.7 C(41)-C(40)-P(4) 110.9(3) C(33)-C(34)-C(35) 110.6(4) C(45)-C(40)-P(4) 114.3(3) C(33)-C(34)-H(34A) 109.5 C(41)-C(40)-H(40) 106.7 C(35)-C(34)-H(34A) 109.5 C(45)-C(40)-H(40) 106.7 C(33)-C(34)-H(34B) 109.5 P(4)-C(40)-H(40) 106.7 C(35)-C(34)-H(34B) 109.5 C(40)-C(41)-C(42) 111.1(4) H(34A)-C(34)-H(34B) 108.1 C(40)-C(41)-H(41A) 109.4 C(36)-C(35)-C(34) 111.3(4) C(42)-C(41)-H(41A) 109.4 C(36)-C(35)-H(35A) 109.4 C(40)-C(41)-H(41B) 109.4 C(34)-C(35)-H(35A) 109.4 C(42)-C(41)-H(41B) 109.4 C(36)-C(35)-H(35B) 109.4 H(41A)-C(41)-H(41B) 108.0 C(34)-C(35)-H(35B) 109.4 C(43)-C(42)-C(41) 110.9(3) H(35A)-C(35)-H(35B) 108.0 C(43)-C(42)-H(42A) 109.5 C(37)-C(36)-C(35) 110.3(5) C(41)-C(42)-H(42A) 109.5 C(37)-C(36)-H(36A) 109.6 C(43)-C(42)-H(42B) 109.5 C(35)-C(36)-H(36A) 109.6 C(41)-C(42)-H(42B) 109.5 C(37)-C(36)-H(36B) 109.6 H(42A)-C(42)-H(42B) 108.0 C(35)-C(36)-H(36B) 109.6 C(42)-C(43)-C(44) 111.1(4) H(36A)-C(36)-H(36B) 108.1 C(42)-C(43)-H(43A) 109.4 C(36)-C(37)-C(38) 111.0(4) C(44)-C(43)-H(43A) 109.4 C(36)-C(37)-H(37A) 109.4 C(42)-C(43)-H(43B) 109.4 C(38)-C(37)-H(37A) 109.4 C(44)-C(43)-H(43B) 109.4 C(36)-C(37)-H(37B) 109.4 H(43A)-C(43)-H(43B) 108.0 C(38)-C(37)-H(37B) 109.4 C(45)-C(44)-C(43) 111.0(3) 17

303 C(45)-C(44)-H(44A) 109.4 C(49)-C(50)-H(50B) 109.4 C(43)-C(44)-H(44A) 109.4 C(51)-C(50)-H(50B) 109.4 C(45)-C(44)-H(44B) 109.4 H(50A)-C(50)-H(50B) 108.0 C(43)-C(44)-H(44B) 109.4 C(50)-C(51)-C(46) 110.7(3) H(44A)-C(44)-H(44B) 108.0 C(50)-C(51)-H(51A) 109.5 C(44)-C(45)-C(40) 110.9(3) C(46)-C(51)-H(51A) 109.5 C(44)-C(45)-H(45A) 109.5 C(50)-C(51)-H(51B) 109.5 C(40)-C(45)-H(45A) 109.5 C(46)-C(51)-H(51B) 109.5 C(44)-C(45)-H(45B) 109.5 H(51A)-C(51)-H(51B) 108.1 C(40)-C(45)-H(45B) 109.5 C(57)-C(52)-C(53) 110.0(3) H(45A)-C(45)-H(45B) 108.0 C(57)-C(52)-P(5) 110.4(3) C(47)-C(46)-C(51) 110.0(3) C(53)-C(52)-P(5) 109.3(3) C(47)-C(46)-P(4) 110.4(3) C(57)-C(52)-H(52) 109.0 C(51)-C(46)-P(4) 110.7(3) C(53)-C(52)-H(52) 109.0 C(47)-C(46)-H(46) 108.6 P(5)-C(52)-H(52) 109.0 C(51)-C(46)-H(46) 108.6 C(54)-C(53)-C(52) 111.8(4) P(4)-C(46)-H(46) 108.6 C(54)-C(53)-H(53A) 109.3 C(48)-C(47)-C(46) 111.8(4) C(52)-C(53)-H(53A) 109.3 C(48)-C(47)-H(47A) 109.3 C(54)-C(53)-H(53B) 109.3 C(46)-C(47)-H(47A) 109.3 C(52)-C(53)-H(53B) 109.3 C(48)-C(47)-H(47B) 109.3 H(53A)-C(53)-H(53B) 107.9 C(46)-C(47)-H(47B) 109.3 C(55)-C(54)-C(53) 111.1(4) H(47A)-C(47)-H(47B) 107.9 C(55)-C(54)-H(54A) 109.4 C(47)-C(48)-C(49) 111.4(4) C(53)-C(54)-H(54A) 109.4 C(47)-C(48)-H(48A) 109.3 C(55)-C(54)-H(54B) 109.4 C(49)-C(48)-H(48A) 109.3 C(53)-C(54)-H(54B) 109.4 C(47)-C(48)-H(48B) 109.3 H(54A)-C(54)-H(54B) 108.0 C(49)-C(48)-H(48B) 109.3 C(54)-C(55)-C(56) 110.2(4) H(48A)-C(48)-H(48B) 108.0 C(54)-C(55)-H(55A) 109.6 C(50)-C(49)-C(48) 110.5(4) C(56)-C(55)-H(55A) 109.6 C(50)-C(49)-H(49A) 109.6 C(54)-C(55)-H(55B) 109.6 C(48)-C(49)-H(49A) 109.6 C(56)-C(55)-H(55B) 109.6 C(50)-C(49)-H(49B) 109.6 H(55A)-C(55)-H(55B) 108.1 C(48)-C(49)-H(49B) 109.6 C(55)-C(56)-C(57) 111.7(4) H(49A)-C(49)-H(49B) 108.1 C(55)-C(56)-H(56A) 109.3 C(49)-C(50)-C(51) 111.1(4) C(57)-C(56)-H(56A) 109.3 C(49)-C(50)-H(50A) 109.4 C(55)-C(56)-H(56B) 109.3 C(51)-C(50)-H(50A) 109.4 C(57)-C(56)-H(56B) 109.3 18

304 H(56A)-C(56)-H(56B) 107.9 C(62)-C(63)-H(63A) 109.6 C(52)-C(57)-C(56) 111.8(4) C(58)-C(63)-H(63A) 109.6 C(52)-C(57)-H(57A) 109.2 C(62)-C(63)-H(63B) 109.6 C(56)-C(57)-H(57A) 109.2 C(58)-C(63)-H(63B) 109.6 C(52)-C(57)-H(57B) 109.2 H(63A)-C(63)-H(63B) 108.1 C(56)-C(57)-H(57B) 109.2 P(6)-C(64)-P(5) 113.6(2) H(57A)-C(57)-H(57B) 107.9 P(6)-C(64)-H(64A) 108.9 C(59)-C(58)-C(63) 111.1(3) P(5)-C(64)-H(64A) 108.9 C(59)-C(58)-P(5) 113.0(3) P(6)-C(64)-H(64B) 108.9 C(63)-C(58)-P(5) 116.1(3) P(5)-C(64)-H(64B) 108.9 C(59)-C(58)-H(58) 105.2 H(64A)-C(64)-H(64B) 107.7 C(63)-C(58)-H(58) 105.2 C(70)-C(65)-C(66) 110.8(3) P(5)-C(58)-H(58) 105.2 C(70)-C(65)-P(6) 110.7(3) C(60)-C(59)-C(58) 110.8(4) C(66)-C(65)-P(6) 108.7(3) C(60)-C(59)-H(59A) 109.5 C(70)-C(65)-H(65) 108.9 C(58)-C(59)-H(59A) 109.5 C(66)-C(65)-H(65) 108.9 C(60)-C(59)-H(59B) 109.5 P(6)-C(65)-H(65) 108.9 C(58)-C(59)-H(59B) 109.5 C(67)-C(66)-C(65) 112.0(4) H(59A)-C(59)-H(59B) 108.1 C(67)-C(66)-H(66A) 109.2 C(59)-C(60)-C(61) 111.3(4) C(65)-C(66)-H(66A) 109.2 C(59)-C(60)-H(60A) 109.4 C(67)-C(66)-H(66B) 109.2 C(61)-C(60)-H(60A) 109.4 C(65)-C(66)-H(66B) 109.2 C(59)-C(60)-H(60B) 109.4 H(66A)-C(66)-H(66B) 107.9 C(61)-C(60)-H(60B) 109.4 C(66)-C(67)-C(68) 111.1(4) H(60A)-C(60)-H(60B) 108.0 C(66)-C(67)-H(67A) 109.4 C(60)-C(61)-C(62) 112.5(4) C(68)-C(67)-H(67A) 109.4 C(60)-C(61)-H(61A) 109.1 C(66)-C(67)-H(67B) 109.4 C(62)-C(61)-H(61A) 109.1 C(68)-C(67)-H(67B) 109.4 C(60)-C(61)-H(61B) 109.1 H(67A)-C(67)-H(67B) 108.0 C(62)-C(61)-H(61B) 109.1 C(69)-C(68)-C(67) 110.7(4) H(61A)-C(61)-H(61B) 107.8 C(69)-C(68)-H(68A) 109.5 C(63)-C(62)-C(61) 111.5(4) C(67)-C(68)-H(68A) 109.5 C(63)-C(62)-H(62A) 109.3 C(69)-C(68)-H(68B) 109.5 C(61)-C(62)-H(62A) 109.3 C(67)-C(68)-H(68B) 109.5 C(63)-C(62)-H(62B) 109.3 H(68A)-C(68)-H(68B) 108.1 C(61)-C(62)-H(62B) 109.3 C(68)-C(69)-C(70) 111.5(4) H(62A)-C(62)-H(62B) 108.0 C(68)-C(69)-H(69A) 109.3 C(62)-C(63)-C(58) 110.2(4) C(70)-C(69)-H(69A) 109.3 19

305 C(68)-C(69)-H(69B) 109.3 H(75A)-C(75)-H(75B) 107.9 C(70)-C(69)-H(69B) 109.3 C(75)-C(76)-C(71) 110.2(4) H(69A)-C(69)-H(69B) 108.0 C(75)-C(76)-H(76A) 109.6 C(69)-C(70)-C(65) 111.5(4) C(71)-C(76)-H(76A) 109.6 C(69)-C(70)-H(70A) 109.3 C(75)-C(76)-H(76B) 109.6 C(65)-C(70)-H(70A) 109.3 C(71)-C(76)-H(76B) 109.6 C(69)-C(70)-H(70B) 109.3 H(76A)-C(76)-H(76B) 108.1 C(65)-C(70)-H(70B) 109.3 O(5)-N(1)-O(4) 121.0(3) H(70A)-C(70)-H(70B) 108.0 O(5)-N(1)-O(3) 121.1(3) C(72)-C(71)-C(76) 111.9(3) O(4)-N(1)-O(3) 117.8(3) C(72)-C(71)-P(6) 114.7(3) O(8)-N(2)-O(6) 119.3(5) C(76)-C(71)-P(6) 112.2(3) O(8)-N(2)-O(7) 126.4(5) C(72)-C(71)-H(71) 105.7 O(6)-N(2)-O(7) 114.2(5) C(76)-C(71)-H(71) 105.7 O(10)-N(3)-O(11) 124.7(5) P(6)-C(71)-H(71) 105.7 O(10)-N(3)-O(9) 118.0(4) C(73)-C(72)-C(71) 111.2(4) O(11)-N(3)-O(9) 117.3(4) C(73)-C(72)-H(72A) 109.4 O(14)-N(4)-O(13) 123.5(5) C(71)-C(72)-H(72A) 109.4 O(14)-N(4)-O(12) 117.6(4) C(73)-C(72)-H(72B) 109.4 O(13)-N(4)-O(12) 118.8(4) C(71)-C(72)-H(72B) 109.4 O(16)-N(5)-O(17) 123.5(6) H(72A)-C(72)-H(72B) 108.0 O(16)-N(5)-O(15) 120.0(5) C(74)-C(73)-C(72) 110.7(4) O(17)-N(5)-O(15) 116.5(4) C(74)-C(73)-H(73A) 109.5 C(26)-O(1)-Ag(1) 120.2(3) C(72)-C(73)-H(73A) 109.5 C(26)-O(1)-Ag(1)#1 135.8(3) C(74)-C(73)-H(73B) 109.5 Ag(1)-O(1)-Ag(1)#1 102.89(13) C(72)-C(73)-H(73B) 109.5 C(26)-O(2)-Ag(2) 129.0(4) H(73A)-C(73)-H(73B) 108.1 N(1)-O(3)-Ag(2) 98.1(2) C(75)-C(74)-C(73) 110.9(5) N(1)-O(4)-Ag(2) 92.6(2) C(75)-C(74)-H(74A) 109.5 N(2)-O(6)-Ag(3) 118.5(3) C(73)-C(74)-H(74A) 109.5 N(2)-O(6)-Ag(6) 106.8(3) C(75)-C(74)-H(74B) 109.5 Ag(3)-O(6)-Ag(6) 127.42(15) C(73)-C(74)-H(74B) 109.5 N(3)-O(9)-Ag(5) 113.5(3) H(74A)-C(74)-H(74B) 108.0 N(3)-O(9)-Ag(3) 108.0(2) C(74)-C(75)-C(76) 112.2(5) Ag(5)-O(9)-Ag(3) 138.21(15) C(74)-C(75)-H(75A) 109.2 N(4)-O(12)-Ag(6) 109.1(3) C(76)-C(75)-H(75A) 109.2 N(4)-O(12)-Ag(4) 109.5(3) C(74)-C(75)-H(75B) 109.2 Ag(6)-O(12)-Ag(4) 141.16(16) C(76)-C(75)-H(75B) 109.2 N(5)-O(15)-Ag(4) 103.8(3) 20

306 N(5)-O(15)-Ag(5) 124.3(3) O(1)-Ag(1)-P(1) 162.80(9) Ag(4)-O(15)-Ag(5) 118.71(19) O(1)-Ag(1)-O(1)#1 77.11(13) C(1)-P(1)-C(13) 106.20(17) P(1)-Ag(1)-O(1)#1 120.07(8) C(1)-P(1)-C(7) 107.33(19) O(1)-Ag(1)-Ag(2) 80.17(9) C(13)-P(1)-C(7) 104.4(2) P(1)-Ag(1)-Ag(2) 85.72(2) C(1)-P(1)-Ag(1) 109.89(12) O(1)#1-Ag(1)-Ag(2) 135.74(8) C(13)-P(1)-Ag(1) 110.75(13) O(2)-Ag(2)-P(2) 139.22(10) C(7)-P(1)-Ag(1) 117.59(15) O(2)-Ag(2)-O(3) 84.15(14) C(13)-P(2)-C(20) 102.84(18) P(2)-Ag(2)-O(3) 133.31(10) C(13)-P(2)-C(14) 104.15(18) O(2)-Ag(2)-O(4) 82.01(14) C(20)-P(2)-C(14) 106.32(18) P(2)-Ag(2)-O(4) 131.15(8) C(13)-P(2)-Ag(2) 109.65(13) O(3)-Ag(2)-O(4) 51.39(10) C(20)-P(2)-Ag(2) 118.11(15) O(2)-Ag(2)-Ag(1) 70.36(10) C(14)-P(2)-Ag(2) 114.32(14) P(2)-Ag(2)-Ag(1) 73.22(2) C(33)-P(3)-C(39) 103.47(18) O(3)-Ag(2)-Ag(1) 153.45(9) C(33)-P(3)-C(27) 107.51(19) O(4)-Ag(2)-Ag(1) 115.42(7) C(39)-P(3)-C(27) 103.03(19) O(6)-Ag(3)-P(3) 160.55(10) C(33)-P(3)-Ag(3) 111.22(13) O(6)-Ag(3)-O(9) 75.35(12) C(39)-P(3)-Ag(3) 116.33(12) P(3)-Ag(3)-O(9) 122.90(9) C(27)-P(3)-Ag(3) 114.29(14) O(6)-Ag(3)-Ag(4) 94.03(10) C(46)-P(4)-C(40) 105.95(17) P(3)-Ag(3)-Ag(4) 84.17(2) C(46)-P(4)-C(39) 103.31(17) O(9)-Ag(3)-Ag(4) 78.61(8) C(40)-P(4)-C(39) 103.48(17) O(15)-Ag(4)-P(4) 145.28(11) C(46)-P(4)-Ag(4) 110.47(12) O(15)-Ag(4)-O(12) 82.26(15) C(40)-P(4)-Ag(4) 113.37(12) P(4)-Ag(4)-O(12) 128.77(12) C(39)-P(4)-Ag(4) 119.04(12) O(15)-Ag(4)-Ag(3) 89.64(10) C(64)-P(5)-C(52) 103.71(17) P(4)-Ag(4)-Ag(3) 82.81(2) C(64)-P(5)-C(58) 103.34(18) O(12)-Ag(4)-Ag(3) 79.24(9) C(52)-P(5)-C(58) 109.86(18) O(9)-Ag(5)-P(5) 154.96(10) C(64)-P(5)-Ag(5) 113.59(12) O(9)-Ag(5)-O(15) 77.29(13) C(52)-P(5)-Ag(5) 112.72(13) P(5)-Ag(5)-O(15) 127.28(10) C(58)-P(5)-Ag(5) 112.88(14) O(9)-Ag(5)-Ag(6) 75.48(9) C(64)-P(6)-C(71) 103.34(18) P(5)-Ag(5)-Ag(6) 91.74(2) C(64)-P(6)-C(65) 103.19(17) O(15)-Ag(5)-Ag(6) 104.16(11) C(71)-P(6)-C(65) 109.15(17) O(12)-Ag(6)-P(6) 134.50(12) C(64)-P(6)-Ag(6) 115.36(13) O(12)-Ag(6)-O(6) 81.02(13) C(71)-P(6)-Ag(6) 111.49(13) P(6)-Ag(6)-O(6) 137.65(10) C(65)-P(6)-Ag(6) 113.51(12) O(12)-Ag(6)-Ag(5) 71.71(11) 21

307 P(6)-Ag(6)-Ag(5) 80.15(2) O(6)-Ag(6)-Ag(5) 95.24(8) ______Symmetry transformations used to generate equivalent atoms: #1 -x,-y-1,-z+1

308 Table S6. Anisotropic displacement parameters (Å2x 103) for 3. The anisotropic displacement factor exponent takes the form: -2π2[ h2a*2U11 + ... + 2 h k a* b* U12 ]

______U11 U22 U33 U23 U13 U12 ______C(1) 24(2) 24(2) 26(2) 0(1) 2(1) 0(1) C(2) 37(2) 28(2) 25(2) -4(2) 0(2) 4(2) C(3) 42(2) 27(2) 36(2) -6(2) 2(2) 7(2) C(4) 50(3) 25(2) 37(2) 1(2) -3(2) 5(2) C(5) 50(3) 32(2) 34(2) -1(2) -11(2) 4(2) C(6) 40(2) 30(2) 26(2) -4(2) -3(2) 4(2) C(7) 42(2) 26(2) 39(2) -5(2) 7(2) -2(2) C(8) 40(2) 31(2) 45(2) 0(2) -4(2) -9(2) C(9) 69(4) 39(3) 76(4) -6(3) -19(3) -15(3) C(10) 63(4) 42(3) 81(4) -3(3) -23(3) -6(3) C(11) 82(4) 40(3) 70(4) -9(3) -6(3) 14(3) C(12) 47(3) 46(3) 62(3) -3(2) 3(2) 10(2) C(13) 23(2) 32(2) 32(2) 7(2) 5(1) 1(1) C(14) 27(2) 35(2) 24(2) 3(2) 1(1) 5(2) C(15) 26(2) 43(2) 33(2) 2(2) 2(2) 2(2) C(16) 32(2) 64(3) 39(2) -3(2) 6(2) 8(2) C(17) 46(3) 55(3) 43(2) -6(2) 5(2) 23(2) C(18) 52(3) 38(2) 46(3) -3(2) 3(2) 14(2) C(19) 39(2) 32(2) 43(2) 9(2) 8(2) 6(2) C(20) 32(2) 34(2) 31(2) 3(2) 7(2) 0(2) C(21) 43(2) 40(2) 41(2) -1(2) 15(2) 4(2) C(22) 58(3) 52(3) 44(3) 0(2) 24(2) -1(2) C(23) 75(4) 78(4) 30(2) -6(2) 16(2) -23(3) C(24) 47(3) 104(5) 34(2) 10(3) 0(2) -12(3) C(25) 38(2) 78(4) 31(2) 2(2) 2(2) -8(2) C(26) 66(3) 55(3) 60(3) -19(3) 30(3) -28(3) C(27) 31(2) 37(2) 36(2) 4(2) 12(2) 0(2) C(28) 38(2) 57(3) 31(2) 3(2) 6(2) 12(2) C(29) 45(3) 90(5) 41(3) -1(3) 15(2) 17(3) C(30) 65(4) 74(4) 40(3) -4(3) 16(2) 13(3) C(31) 61(3) 71(4) 32(2) 5(2) 8(2) 11(3) C(32) 42(2) 49(3) 32(2) 4(2) 4(2) 7(2)

309 C(33) 25(2) 25(2) 40(2) 1(2) 5(2) -2(1) C(34) 34(2) 39(2) 56(3) 5(2) -2(2) -15(2) C(35) 42(3) 57(3) 70(4) 8(3) -15(2) -23(2) C(36) 64(3) 45(3) 69(4) -14(3) -20(3) -17(3) C(37) 41(2) 48(3) 48(3) -10(2) -4(2) -2(2) C(38) 35(2) 33(2) 41(2) -8(2) 4(2) -7(2) C(39) 26(2) 24(2) 29(2) -2(1) 5(1) 1(1) C(40) 24(2) 27(2) 29(2) 2(1) 7(1) 0(1) C(41) 34(2) 31(2) 27(2) -2(2) 1(2) 3(2) C(42) 40(2) 46(3) 30(2) 9(2) 3(2) 4(2) C(43) 43(2) 32(2) 42(2) 10(2) -2(2) 6(2) C(44) 39(2) 24(2) 51(3) 9(2) -3(2) -2(2) C(45) 33(2) 20(2) 32(2) -1(1) -1(2) 0(1) C(46) 26(2) 25(2) 27(2) -2(1) 5(1) 0(1) C(47) 32(2) 35(2) 34(2) 7(2) 5(2) -5(2) C(48) 43(2) 61(3) 34(2) 8(2) -1(2) -13(2) C(49) 34(2) 59(3) 45(3) 8(2) -7(2) -13(2) C(50) 31(2) 40(2) 51(3) 1(2) -3(2) -11(2) C(51) 29(2) 28(2) 36(2) 4(2) 1(2) -7(2) C(52) 29(2) 27(2) 26(2) 2(1) -2(1) 2(1) C(53) 38(2) 28(2) 33(2) 0(2) -9(2) -2(2) C(54) 50(3) 30(2) 47(3) 10(2) -8(2) -9(2) C(55) 66(3) 34(2) 42(2) 16(2) -8(2) -12(2) C(56) 69(3) 43(3) 29(2) 11(2) -4(2) -13(2) C(57) 40(2) 31(2) 23(2) 3(2) -3(2) -6(2) C(58) 35(2) 29(2) 24(2) 3(1) 0(2) -3(2) C(59) 32(2) 44(2) 32(2) -5(2) 3(2) -6(2) C(60) 56(3) 58(3) 34(2) 2(2) 3(2) -29(3) C(61) 72(3) 32(2) 38(2) -1(2) -9(2) -12(2) C(62) 72(4) 30(2) 48(3) -3(2) -18(2) 6(2) C(63) 38(2) 28(2) 38(2) -6(2) -4(2) 6(2) C(64) 29(2) 34(2) 21(2) 1(1) 2(1) 3(2) C(65) 24(2) 21(2) 26(2) -4(1) 0(1) 1(1) C(66) 27(2) 34(2) 30(2) -3(2) 1(1) -5(2) C(67) 26(2) 41(2) 53(3) -13(2) 1(2) -3(2) C(68) 32(2) 36(2) 45(2) -5(2) -14(2) 1(2) C(69) 47(2) 37(2) 35(2) 3(2) -14(2) 0(2) C(70) 35(2) 33(2) 29(2) 2(2) -3(2) -2(2) 24

310 C(71) 29(2) 27(2) 24(2) 2(1) 2(1) -4(1) C(72) 37(2) 25(2) 37(2) 3(2) -2(2) -5(2) C(73) 67(3) 34(2) 54(3) -1(2) 2(2) -22(2) C(74) 79(4) 49(3) 60(3) 11(3) 14(3) -34(3) C(75) 52(3) 72(4) 50(3) 2(3) 20(2) -29(3) C(76) 36(2) 45(2) 32(2) 0(2) 11(2) -9(2) N(1) 29(2) 30(2) 29(2) -5(1) 2(1) 3(1) N(2) 71(3) 38(2) 40(2) 1(2) -7(2) 14(2) N(3) 47(2) 29(2) 46(2) 1(2) 10(2) 0(2) N(4) 37(2) 34(2) 58(2) 7(2) 14(2) -2(2) N(5) 57(2) 40(2) 42(2) -14(2) -14(2) 20(2) O(1) 37(2) 44(2) 49(2) -11(1) 17(1) -12(1) O(2) 92(3) 45(2) 65(2) -17(2) 43(2) -23(2) O(3) 39(2) 43(2) 49(2) -13(2) -13(1) 15(1) O(4) 33(2) 41(2) 40(2) -11(1) -7(1) 2(1) O(5) 43(2) 35(2) 44(2) -13(1) -6(1) 9(1) O(6) 45(2) 54(2) 37(2) -6(1) 2(1) 21(2) O(7) 75(3) 85(4) 117(4) -23(3) -30(3) 21(3) O(8) 122(4) 59(3) 41(2) 0(2) 17(2) 21(3) O(9) 45(2) 33(2) 51(2) -7(1) 21(2) 0(1) O(10) 81(3) 58(3) 123(4) -46(3) 30(3) -22(2) O(11) 70(3) 46(2) 66(2) 11(2) 33(2) -5(2) O(12) 30(2) 32(2) 105(3) 16(2) 15(2) 2(1) O(13) 38(2) 69(3) 123(4) 16(3) -8(2) -3(2) O(14) 66(3) 42(2) 84(3) 19(2) 35(2) 0(2) O(15) 83(3) 30(2) 56(2) -4(2) 26(2) 13(2) O(16) 103(4) 79(3) 81(3) -43(3) -38(3) 43(3) O(17) 160(6) 39(2) 48(2) 3(2) 12(3) 8(3) P(1) 21(1) 23(1) 27(1) 1(1) 2(1) 1(1) P(2) 23(1) 28(1) 25(1) 2(1) 2(1) 1(1) P(3) 24(1) 22(1) 31(1) 1(1) 5(1) -2(1) P(4) 23(1) 19(1) 25(1) -2(1) 5(1) 1(1) P(5) 24(1) 24(1) 20(1) 0(1) 0(1) 3(1) P(6) 23(1) 20(1) 21(1) 0(1) 2(1) 1(1) Ag(1) 24(1) 41(1) 35(1) 8(1) 7(1) -1(1) Ag(2) 33(1) 26(1) 43(1) -4(1) 3(1) -1(1) Ag(3) 34(1) 26(1) 40(1) 2(1) 7(1) 6(1) Ag(4) 35(1) 26(1) 42(1) -5(1) 11(1) 6(1) 25

311 Ag(5) 30(1) 40(1) 35(1) -8(1) 2(1) 11(1) Ag(6) 43(1) 29(1) 44(1) -5(1) 3(1) 12(1) ______

312 Table S7. Hydrogen coordinates (x 104) and isotropic displacement parameters (Å2x 103) for 3.

______x y z U(eq) ______

H(1) 2373 -5048 4485 30 H(2A) 1511 -6227 4334 36 H(2B) 1857 -5909 3846 36 H(3A) 2623 -6595 4250 42 H(3B) 2146 -7240 4160 42 H(4A) 2608 -7400 5077 46 H(4B) 1994 -7169 5129 46 H(5A) 2870 -6020 5259 48 H(5B) 2542 -6337 5763 48 H(6A) 2234 -5000 5463 39 H(6B) 1757 -5647 5379 39 H(7) 1760 -3656 5285 43 H(8A) 2573 -3689 4847 47 H(8B) 2318 -3148 4335 47 H(9A) 2528 -2578 5478 75 H(9B) 2842 -2282 4959 75 H(10A) 2075 -1613 4559 77 H(10B) 2209 -1289 5189 77 H(11A) 1532 -2176 5524 78 H(11B) 1258 -1610 5033 78 H(12A) 991 -3003 4871 62 H(12B) 1317 -2687 4366 62 H(13A) 1557 -3593 3692 35 H(13B) 1763 -4485 3546 35 H(14) 441 -3688 4061 35 H(15A) -380 -4008 3170 41 H(15B) -270 -4567 3717 41 H(16A) -1004 -3632 3819 54 H(16B) -535 -3497 4305 54 H(17A) -809 -2205 3952 57 H(17B) -726 -2483 3323 57 27

313 H(18A) 128 -2245 4197 54 H(18B) 15 -1666 3660 54 H(19A) 744 -2583 3522 45 H(19B) 262 -2735 3051 45 H(20) 997 -3425 2634 39 H(21A) 1301 -5063 2339 49 H(21B) 1708 -4369 2577 49 H(22A) 1540 -3584 1741 60 H(22B) 1706 -4483 1562 60 H(23A) 799 -4822 1310 72 H(23B) 944 -4016 976 72 H(24A) 108 -3836 1379 74 H(24B) 534 -3184 1638 74 H(25A) 86 -3780 2387 59 H(25B) 258 -4674 2201 59 H(27) 1404 -327 4180 41 H(28A) 812 622 3688 50 H(28B) 1169 1366 3941 50 H(29A) 427 1178 4504 70 H(29B) 591 245 4613 70 H(30A) 1211 1630 5046 71 H(30B) 896 1017 5422 71 H(31A) 1489 -53 5255 65 H(31B) 1835 712 5488 65 H(32A) 2217 107 4685 49 H(32B) 2062 1041 4563 49 H(33) 1205 324 2774 36 H(34A) 896 -810 3277 52 H(34B) 1368 -1374 3084 52 H(35A) 564 -1673 2508 69 H(35B) 505 -735 2328 69 H(36A) 1355 -1792 2021 72 H(36B) 870 -1491 1593 72 H(37A) 1176 -121 1671 55 H(37B) 1654 -681 1488 55 H(38A) 1960 240 2233 44 H(38B) 2064 -689 2413 44 H(39A) 2150 1510 3678 31 28

314 H(39B) 1838 1453 3071 31 H(40) 2983 2406 2362 32 H(41A) 1897 1858 2129 37 H(41B) 2390 1555 1799 37 H(42A) 2496 2886 1417 47 H(42B) 1889 2607 1270 47 H(43A) 1603 3415 2016 47 H(43B) 1926 3987 1620 47 H(44A) 2702 3915 2266 46 H(44B) 2213 4229 2596 46 H(45A) 2098 2905 2977 34 H(45B) 2704 3182 3127 34 H(46) 3058 2151 3796 31 H(47A) 3085 790 4153 41 H(47B) 3569 567 3785 41 H(48A) 3922 812 4720 56 H(48B) 3639 1686 4720 56 H(49A) 4558 1885 4578 56 H(49B) 4513 1272 4054 56 H(50A) 4478 2635 3710 50 H(50B) 3987 2836 4073 50 H(51A) 3934 1713 3131 37 H(51B) 3652 2590 3118 37 H(52) 4596 -2701 1041 33 H(53A) 4135 -1249 1507 40 H(53B) 4736 -1549 1651 40 H(54A) 4706 -439 983 52 H(54B) 4953 -1238 720 52 H(55A) 4315 -666 38 57 H(55B) 3865 -688 475 57 H(56A) 4340 -2116 11 57 H(56B) 3737 -1825 -143 57 H(57A) 3511 -2137 781 38 H(57B) 3762 -2933 519 38 H(58) 3801 -4352 1891 36 H(59A) 2969 -3767 1547 43 H(59B) 3160 -3773 923 43 H(60A) 2633 -4958 1035 59 29

315 H(60B) 3000 -5215 1588 59 H(61A) 3329 -5236 457 58 H(61B) 3259 -6024 838 58 H(62A) 4028 -5680 1435 61 H(62B) 4194 -5666 803 61 H(63A) 4149 -4215 779 42 H(63B) 4533 -4473 1319 42 H(64A) 4764 -3715 2175 34 H(64B) 4903 -2764 2174 34 H(65) 5301 -3880 3251 28 H(66A) 5623 -2695 2813 36 H(66B) 5445 -2125 3308 36 H(67A) 6381 -2329 3452 48 H(67B) 6306 -3298 3425 48 H(68A) 6459 -2904 4387 46 H(68B) 5974 -2270 4319 46 H(69A) 5681 -3471 4760 49 H(69B) 5858 -4020 4253 49 H(70A) 4987 -2862 4162 39 H(70B) 4917 -3831 4129 39 H(71) 3756 -3941 2796 32 H(72A) 4456 -4882 2700 40 H(72B) 4586 -4908 3367 40 H(73A) 3604 -5488 2782 62 H(73B) 4051 -6062 3094 62 H(74A) 3322 -5875 3672 75 H(74B) 3864 -5528 3983 75 H(75A) 3173 -4560 4065 68 H(75B) 3047 -4540 3398 68 H(76A) 4008 -3904 3995 45 H(76B) 3556 -3362 3661 45 H -370(30) -6950(50) 4590(30) 90(30) ______

316 Table S8. Torsion angles [°] for 3.

______C(6)-C(1)-C(2)-C(3) -54.0(4) C(28)-C(29)-C(30)-C(31) -55.9(7) P(1)-C(1)-C(2)-C(3) -174.6(3) C(29)-C(30)-C(31)-C(32) 54.8(7) C(1)-C(2)-C(3)-C(4) 55.7(5) C(30)-C(31)-C(32)-C(27) -55.2(7) C(2)-C(3)-C(4)-C(5) -56.7(5) C(28)-C(27)-C(32)-C(31) 56.5(5) C(3)-C(4)-C(5)-C(6) 57.3(5) P(3)-C(27)-C(32)-C(31) -175.3(4) C(4)-C(5)-C(6)-C(1) -56.8(5) C(38)-C(33)-C(34)-C(35) -53.9(5) C(2)-C(1)-C(6)-C(5) 54.5(4) P(3)-C(33)-C(34)-C(35) -174.8(4) P(1)-C(1)-C(6)-C(5) 175.4(3) C(33)-C(34)-C(35)-C(36) 58.0(6) C(12)-C(7)-C(8)-C(9) 53.6(6) C(34)-C(35)-C(36)-C(37) -59.0(6) P(1)-C(7)-C(8)-C(9) 179.1(4) C(35)-C(36)-C(37)-C(38) 56.7(6) C(7)-C(8)-C(9)-C(10) -55.3(7) C(36)-C(37)-C(38)-C(33) -54.4(6) C(8)-C(9)-C(10)-C(11) 56.8(8) C(34)-C(33)-C(38)-C(37) 53.0(5) C(9)-C(10)-C(11)-C(12) -57.6(7) P(3)-C(33)-C(38)-C(37) 175.9(3) C(8)-C(7)-C(12)-C(11) -56.4(6) C(45)-C(40)-C(41)-C(42) 55.5(4) P(1)-C(7)-C(12)-C(11) 176.1(4) P(4)-C(40)-C(41)-C(42) -176.3(3) C(10)-C(11)-C(12)-C(7) 58.4(7) C(40)-C(41)-C(42)-C(43) -55.6(5) C(19)-C(14)-C(15)-C(16) -55.8(5) C(41)-C(42)-C(43)-C(44) 56.1(5) P(2)-C(14)-C(15)-C(16) 173.5(3) C(42)-C(43)-C(44)-C(45) -56.6(5) C(14)-C(15)-C(16)-C(17) 54.6(5) C(43)-C(44)-C(45)-C(40) 56.3(5) C(15)-C(16)-C(17)-C(18) -54.9(5) C(41)-C(40)-C(45)-C(44) -55.9(4) C(16)-C(17)-C(18)-C(19) 56.3(5) P(4)-C(40)-C(45)-C(44) 177.8(3) C(17)-C(18)-C(19)-C(14) -57.1(5) C(51)-C(46)-C(47)-C(48) 55.5(5) C(15)-C(14)-C(19)-C(18) 56.8(5) P(4)-C(46)-C(47)-C(48) 177.9(3) P(2)-C(14)-C(19)-C(18) -174.3(3) C(46)-C(47)-C(48)-C(49) -55.7(6) C(25)-C(20)-C(21)-C(22) 54.5(6) C(47)-C(48)-C(49)-C(50) 56.0(6) P(2)-C(20)-C(21)-C(22) 177.8(4) C(48)-C(49)-C(50)-C(51) -57.0(6) C(20)-C(21)-C(22)-C(23) -54.9(6) C(49)-C(50)-C(51)-C(46) 57.7(5) C(21)-C(22)-C(23)-C(24) 56.5(7) C(47)-C(46)-C(51)-C(50) -56.0(5) C(22)-C(23)-C(24)-C(25) -57.9(7) P(4)-C(46)-C(51)-C(50) -178.3(3) C(21)-C(20)-C(25)-C(24) -56.3(6) C(57)-C(52)-C(53)-C(54) -54.4(5) P(2)-C(20)-C(25)-C(24) -179.8(4) P(5)-C(52)-C(53)-C(54) -175.7(3) C(23)-C(24)-C(25)-C(20) 57.9(7) C(52)-C(53)-C(54)-C(55) 56.0(5) C(32)-C(27)-C(28)-C(29) -57.6(6) C(53)-C(54)-C(55)-C(56) -56.3(6) P(3)-C(27)-C(28)-C(29) 176.5(4) C(54)-C(55)-C(56)-C(57) 56.7(6) C(27)-C(28)-C(29)-C(30) 57.4(7) C(53)-C(52)-C(57)-C(56) 54.5(5)

317 P(5)-C(52)-C(57)-C(56) 175.2(3) O(11)-N(3)-O(9)-Ag(5) 179.0(4) C(55)-C(56)-C(57)-C(52) -56.6(6) O(10)-N(3)-O(9)-Ag(3) -174.3(5) C(63)-C(58)-C(59)-C(60) 58.1(5) O(11)-N(3)-O(9)-Ag(3) 4.5(5) P(5)-C(58)-C(59)-C(60) -169.3(3) O(14)-N(4)-O(12)-Ag(6) 180.0(4) C(58)-C(59)-C(60)-C(61) -55.4(5) O(13)-N(4)-O(12)-Ag(6) 0.2(6) C(59)-C(60)-C(61)-C(62) 53.2(6) O(14)-N(4)-O(12)-Ag(4) -4.5(6) C(60)-C(61)-C(62)-C(63) -53.2(6) O(13)-N(4)-O(12)-Ag(4) 175.7(4) C(61)-C(62)-C(63)-C(58) 54.8(5) O(16)-N(5)-O(15)-Ag(4) -174.8(4) C(59)-C(58)-C(63)-C(62) -57.7(5) O(17)-N(5)-O(15)-Ag(4) 7.0(6) P(5)-C(58)-C(63)-C(62) 171.3(3) O(16)-N(5)-O(15)-Ag(5) 45.2(7) C(70)-C(65)-C(66)-C(67) 53.7(4) O(17)-N(5)-O(15)-Ag(5) -132.9(5) P(6)-C(65)-C(66)-C(67) 175.6(3) C(2)-C(1)-P(1)-C(13) -63.9(3) C(65)-C(66)-C(67)-C(68) -55.2(5) C(6)-C(1)-P(1)-C(13) 174.8(3) C(66)-C(67)-C(68)-C(69) 56.3(5) C(2)-C(1)-P(1)-C(7) -175.0(3) C(67)-C(68)-C(69)-C(70) -56.9(5) C(6)-C(1)-P(1)-C(7) 63.6(3) C(68)-C(69)-C(70)-C(65) 56.0(5) C(2)-C(1)-P(1)-Ag(1) 56.0(3) C(66)-C(65)-C(70)-C(69) -53.8(4) C(6)-C(1)-P(1)-Ag(1) -65.3(3) P(6)-C(65)-C(70)-C(69) -174.4(3) P(2)-C(13)-P(1)-C(1) 118.5(2) C(76)-C(71)-C(72)-C(73) 54.9(5) P(2)-C(13)-P(1)-C(7) -128.2(2) P(6)-C(71)-C(72)-C(73) -175.8(3) P(2)-C(13)-P(1)-Ag(1) -0.8(2) C(71)-C(72)-C(73)-C(74) -56.2(6) C(12)-C(7)-P(1)-C(1) 176.3(3) C(72)-C(73)-C(74)-C(75) 57.1(7) C(8)-C(7)-P(1)-C(1) 51.2(4) C(73)-C(74)-C(75)-C(76) -56.9(6) C(12)-C(7)-P(1)-C(13) 63.9(4) C(74)-C(75)-C(76)-C(71) 54.6(6) C(8)-C(7)-P(1)-C(13) -61.3(4) C(72)-C(71)-C(76)-C(75) -53.4(5) C(12)-C(7)-P(1)-Ag(1) -59.3(4) P(6)-C(71)-C(76)-C(75) 176.0(4) C(8)-C(7)-P(1)-Ag(1) 175.6(3) O(2)-C(26)-O(1)-Ag(1) -16.9(11) P(1)-C(13)-P(2)-C(20) 177.3(2) O(2)-C(26)-O(1)-Ag(1)#1 148.3(6) P(1)-C(13)-P(2)-C(14) 66.5(2) O(1)-C(26)-O(2)-Ag(2) -19.8(12) P(1)-C(13)-P(2)-Ag(2) -56.2(2) O(5)-N(1)-O(3)-Ag(2) 178.4(3) C(21)-C(20)-P(2)-C(13) 60.2(4) O(4)-N(1)-O(3)-Ag(2) -3.1(4) C(25)-C(20)-P(2)-C(13) -175.9(4) O(5)-N(1)-O(4)-Ag(2) -178.5(3) C(21)-C(20)-P(2)-C(14) 169.4(3) O(3)-N(1)-O(4)-Ag(2) 2.9(4) C(25)-C(20)-P(2)-C(14) -66.8(4) O(8)-N(2)-O(6)-Ag(3) 21.8(6) C(21)-C(20)-P(2)-Ag(2) -60.6(3) O(7)-N(2)-O(6)-Ag(3) -154.9(4) C(25)-C(20)-P(2)-Ag(2) 63.2(4) O(8)-N(2)-O(6)-Ag(6) 174.2(4) C(19)-C(14)-P(2)-C(13) 75.6(3) O(7)-N(2)-O(6)-Ag(6) -2.5(6) C(15)-C(14)-P(2)-C(13) -156.2(3) O(10)-N(3)-O(9)-Ag(5) 0.2(6) C(19)-C(14)-P(2)-C(20) -32.6(4) 32

318 C(15)-C(14)-P(2)-C(20) 95.5(3) P(3)-C(39)-P(4)-Ag(4) 18.4(3) C(19)-C(14)-P(2)-Ag(2) -164.8(3) P(6)-C(64)-P(5)-C(52) 144.1(2) C(15)-C(14)-P(2)-Ag(2) -36.6(3) P(6)-C(64)-P(5)-C(58) -101.2(2) C(34)-C(33)-P(3)-C(39) -163.7(3) P(6)-C(64)-P(5)-Ag(5) 21.4(3) C(38)-C(33)-P(3)-C(39) 74.3(3) C(57)-C(52)-P(5)-C(64) 166.9(3) C(34)-C(33)-P(3)-C(27) -55.1(4) C(53)-C(52)-P(5)-C(64) -72.0(3) C(38)-C(33)-P(3)-C(27) -177.2(3) C(57)-C(52)-P(5)-C(58) 57.0(3) C(34)-C(33)-P(3)-Ag(3) 70.7(3) C(53)-C(52)-P(5)-C(58) 178.1(3) C(38)-C(33)-P(3)-Ag(3) -51.4(3) C(57)-C(52)-P(5)-Ag(5) -69.8(3) P(4)-C(39)-P(3)-C(33) -97.8(2) C(53)-C(52)-P(5)-Ag(5) 51.3(3) P(4)-C(39)-P(3)-C(27) 150.3(2) C(59)-C(58)-P(5)-C(64) 161.2(3) P(4)-C(39)-P(3)-Ag(3) 24.4(3) C(63)-C(58)-P(5)-C(64) -68.7(3) C(28)-C(27)-P(3)-C(33) -51.6(4) C(59)-C(58)-P(5)-C(52) -88.6(3) C(32)-C(27)-P(3)-C(33) -177.3(3) C(63)-C(58)-P(5)-C(52) 41.5(3) C(28)-C(27)-P(3)-C(39) 57.3(4) C(59)-C(58)-P(5)-Ag(5) 38.1(3) C(32)-C(27)-P(3)-C(39) -68.5(3) C(63)-C(58)-P(5)-Ag(5) 168.2(3) C(28)-C(27)-P(3)-Ag(3) -175.6(3) P(5)-C(64)-P(6)-C(71) 65.5(3) C(32)-C(27)-P(3)-Ag(3) 58.7(3) P(5)-C(64)-P(6)-C(65) 179.2(2) C(47)-C(46)-P(4)-C(40) -177.5(3) P(5)-C(64)-P(6)-Ag(6) -56.4(2) C(51)-C(46)-P(4)-C(40) -55.5(3) C(72)-C(71)-P(6)-C(64) 68.6(3) C(47)-C(46)-P(4)-C(39) 74.1(3) C(76)-C(71)-P(6)-C(64) -162.3(3) C(51)-C(46)-P(4)-C(39) -163.9(3) C(72)-C(71)-P(6)-C(65) -40.7(3) C(47)-C(46)-P(4)-Ag(4) -54.3(3) C(76)-C(71)-P(6)-C(65) 88.4(3) C(51)-C(46)-P(4)-Ag(4) 67.7(3) C(72)-C(71)-P(6)-Ag(6) -166.9(2) C(41)-C(40)-P(4)-C(46) -179.6(3) C(76)-C(71)-P(6)-Ag(6) -37.8(3) C(45)-C(40)-P(4)-C(46) -53.2(3) C(70)-C(65)-P(6)-C(64) -176.1(3) C(41)-C(40)-P(4)-C(39) -71.2(3) C(66)-C(65)-P(6)-C(64) 61.9(3) C(45)-C(40)-P(4)-C(39) 55.1(3) C(70)-C(65)-P(6)-C(71) -66.7(3) C(41)-C(40)-P(4)-Ag(4) 59.1(3) C(66)-C(65)-P(6)-C(71) 171.3(2) C(45)-C(40)-P(4)-Ag(4) -174.5(2) C(70)-C(65)-P(6)-Ag(6) 58.3(3) P(3)-C(39)-P(4)-C(46) -104.4(2) C(66)-C(65)-P(6)-Ag(6) -63.6(3) P(3)-C(39)-P(4)-C(40) 145.3(2) ______Symmetry transformations used to generate equivalent atoms: #1 -x,-y-1,-z+1

319 320 9.5 Appendix E - Supplementary material for Chapter 6

321 322 Supplementary Information

Gas-Phase Ion-Molecule Reactions of Copper

– – Hydride Anions [CuH2] and [Cu2H3]

Athanasios Zavras,a Hossein Ghari, c Alireza Ariafard*,b,c Allan J. Cantyb and Richard A. J.

O’Hair*a

a School of Chemistry and Bio21 Molecular Science and Biotechnology Institute, University of Melbourne, 30 Flemington Rd, Parkville, Victoria 3010 (Australia). Fax: (+) 61 3 9347

8124; E-mail: [email protected]

b The School of Physical Sciences, University of Tasmania, Private Bag 75, Hobart,

Tasmania 7001, Australia.

c Department of Chemistry, Faculty of Science, Central Tehran Branch, Islamic Azad

University, Shahrak Gharb, Tehran, Iran.

323 Table of Contents Figure S1. LTQ full negative-ion mode ESI mass spectrum of a solution containing, copper(I) oxide (1 eq.) and formic acid (2 eq.) in an acetonitrile solution obtained immediately after the addition of formic acid. The m/z value represents the most intense isotope contribution of a given ion...... S4 Figure S2. LTQ mass spectra in the negative-ion mode obtained for the MSn CID of: (a) - 2 63 - 3 [Cu(O2CH)2] , m/z 153, N.C.E = 15%, MS , (b) [ CuH(O2CH)] , m/z 109, N.C.E = 18%, MS . A * represents the mass selected precursor ion...... S5 Figure S3. LTQ mass spectra in the negative-ion mode obtained for the MSn CID of: (a) 63 - 2 63 - [ Cu2(O2CH)3] , m/z 261, N.C.E = 14%, MS , (b) [ Cu2H(O2CH)2] , m/z 217, N.C.E = 12%, 3 63 - 4 MS , (c) [ Cu2H2(O2CH)] , m/z 172, N.C.E = 12%, MS . A * represents the mass selected precursor ion...... S6 Figure S4. Triplicate experiments for pseudo-first order kinetics obtained in a LTQ 2D 63 - linear-ion trap quasi thermalized to 298 K for the ion-molecule reaction of [ CuH2] , m/z 65, with the neutral reagents: (a) 1-butanethiol, (b) allyl iodide, (c) methyl iodide, (d) 2,2,2- trifluoroethanol (e) carbon disulfide, and (f) carbon dioxide. The black dots are the plotted 63 + coordinates for the activation time (s) (x-axis) vs. ln(Relative Intensity of [ CuH2] ) (y-axis). The red line represents the linear trendline of best-fit using Excel...... S7 Figure S5. Triplicate experiments for pseudo-first order kinetics obtained in a LTQ 2D 63 - linear-ion trap quasi thermalized to 298 K for the ion-molecule reaction of [ Cu2H3] , m/z 129, with the neutral reagents: (a) 1-butanethiol, (b) allyl iodide, (c) methyl iodide, (d) 2,2,2- trifluoroethanol (e) carbon disulfide, and (f) carbon dioxide. The black dots are the plotted 63 + coordinates for the activation time (s) (x-axis) vs. ln(Relative Intensity of [ CuH2] ) (y-axis). The red line represents the linear trendline of best-fit using Excel...... S8 Figure S6. LTQ mass spectra obtained in the negative-ion mode for the ion-molecule - 9 reaction of different isotopologues of [Cu2H3] at 3000 ms with CS2 ([CS2]ion trap = 6.4 x 10 -3 63 - 63/65 - molecules cm ) for: (a) [ Cu2H3] , m/z 129, and (b) [ Cu2H3] . A * represents the mass selected precursor ion...... S9 Figure S7. LTQ mass spectra obtained in the negative-ion mode for the ion-molecule 63 – reactions of hydrido cuprates with methyl iodide for: (a) [ CuH2] m/z 65, activation time = 9 -3 63 – 10 ms [methyl iodide]ion trap = 2.9 x 10 molecules cm ; (b) [ Cu2H3] m/z 129, activation time 9 -3 = 10000 ms, [methyl iodide]ion trap = 8.4 x 10 molecules cm . The most intense peak in the ion is represented by the m/z value. * Represents the mass selected precursor ion...... S10 Figure S8. LTQ mass spectra obtained in the negative-ion mode for the ion-molecule 63 – reactions of hydrido cuprates with allyliodide for: (a) [ CuH2] m/z 65, activation time = 60 9 -3 – ms, [allyliodide]ion trap = 2.8 x 10 molecules cm ; (b) [Cu2H3] m/z 129, activation time = 100 9 -3 ms, [allyliodide]ion trap = 4.3 x 10 molecules cm . The most intense peak in the ion is represented by the m/z value. * Represents the mass selected precursor ion...... S11

63 - Figure S9. Ion-molecule reaction of methyl iodide (CH3I) with mass selected [ CuHD] . 9 -3 [CH3I]ion trap = 6.8 x 10 molecules cm , activation time = 2000 ms. The most intense peak in the ion is represented by the m/z value. * Represents the mass selected precursor ion...... S12

Figure S10. Ion-molecule reaction of allyl iodide (CH2=CHCH2I) with mass selected 63 - 9 -3 [ CuHD] . [CH2=CHCH2I]ion trap = 4.7 x 10 molecules cm , activation time = 2000 ms. The S2

324 most intense peak in the ion is represented by the m/z value. * Represents the mass selected precursor ion...... S13 Figure S11. LTQ mass spectra obtained in the negative-ion mode for the ion-molecule 63 – reactions of hydrido cuprates with 2,2,2-trifluoroethanol for: (a) [ CuH2] m/z 65, activation 9 -3 63 – time = 10 ms, [2,2,2-trifluorethanol]ion trap = 2.2 x 10 molecules cm ; (b) [ Cu2H3] m/z 129, 9 -3 activation time = 3000 ms, [2,2,2-trifluorethanol]ion trap = 6.1 x 10 molecules cm . The most intense peak in the ion is represented by the m/z value. * Represents the mass selected precursor ion...... S14

Figure S12. Ion-molecule reaction of 2,2,2-trifluorethanol (CF3CH2OH) with mass selected 63 - 9 -3 [ CuHD] . [CF3CH2OH]ion trap = 3.1 x 10 molecules cm , activation time = 3200 ms. The most intense peak in the ion is represented by the m/z value. * Represents the mass selected precursor ion...... S15 Figure S13. LTQ mass spectra obtained for the ion-molecule reactions of hydrido cuprates – with 1-butanethiol for: (a) [CuH2] m/z 65, activation time = 10 ms, [1-butanethiol]ion trap = 9 -3 – 2.20 x 10 molecules cm ; (b) [Cu2H3] m/z 129, activation time = 3000 ms, [1-butanethiol]ion 9 -3 trap = 3.9 x 10 molecules cm . The most intense peak in the ion is represented by the m/z value. * Represents the mass selected precursor ion...... S16

Figure S14. Ion-molecule reaction of 1-butanethiol (CH3(CH2)3SH) with mass selected 63 - 9 -3 [ CuHD] . [CH3(CH2)3SH]ion trap = 2.2 x 10 molecules cm , activation time = 300 ms. The most intense peak in the ion is represented by the m/z value. * Represents the mass selected precursor ion...... S17 Figure S15. Energy profile obtained from DFT calculations of the competing mechanisms – for the reaction of [CuH2] with CS2: (a) hydride transfer pathway; (b) addition/H2 elimination – to form [CuCS2] . The relative Gibbs and potential energies (in parentheses) obtained from the CCSD(T)/BS2//M06/BS1 calculations are given in kcal mol-1...... S18

– Figure S16. Energy profile obtained from DFT calculations for the reaction of [CuH2] with allyl iodide via oxidative addition. The relative Gibbs and potential energies (in parentheses) obtained from the M06/BS2//M06/BS1 calculations are given in kcal mol-1...... S19

– Figure S17. Energy profile obtained from DFT calculations for the reaction of [CuH2] with 1-butanethiol. The relative Gibbs and potential energies (in parentheses) obtained from the M06/BS2//M06/BS1 calculations are given in kcal mol-1...... S20

Figure S18. Correlation between the gas-phase acidity, ΔHacid(kcal/mol) (y-axis) and activation potentials (kcal/mol) (x-axis) of 1-buanethiol, 2,2,2-trifluoroethanol and methanol.S21 Figure S19. DFT calculated KIEs based on the Gibbs free energies obtained at the M06 level using the conditions of T = 298.15 K; P = 2 × 10−3 Torr...... S22

63 - Table S1. Gas-phase kinetics associated with the ion-molecule reactions of [ CuH2] , m/z 64, with various neutral reagents...... S23

63 - Table S2. Gas-phase kinetics associated with the ion-molecule reactions of [ Cu2H3] , m/z 129, with various neutral reagents...... S24 Cartesian coordinates and total energies for all calculated structures...... S25

325

Figure S1. LTQ full negative-ion mode ESI mass spectrum of a solution containing, copper(I) oxide (1 eq.) and formic acid (2 eq.) in an acetonitrile solution obtained immediately after the addition of formic acid. The m/z value represents the most intense isotope contribution of a given ion.

326

n - Figure S2. LTQ MS CID spectra showing decarboxylation: (a) [Cu(O2CH)2] , eq. S1a, m/z 2 63 - 3 153, N.C.E = 15%, MS , (b) [ CuH(O2CH)] , eq. S1b, m/z 109, N.C.E = 18%, MS . A * represents the mass selected precursor ion.

CID 63 - 63 - [ Cu(O2CH)2] → [ CuH(O2CH)] + CO2 (S1a) 153 109 CID 63 - 63 - [ CuH(O2CH)] → [ CuH2] + CO2 (S1b) 109 65

327

n 63 - Figure S3. LTQ MS CID spectra showing decarboxylation: (a) [ Cu2(O2CH)3] , eq. S2a, 2 63 - 3 m/z 261, N.C.E = 14%, MS , (b) [ Cu2H(O2CH)2] , eq. S2b, m/z 217, N.C.E = 12%, MS , (c) 63 - 4 [ Cu2H2(O2CH)] , eq. S2c, m/z 172, N.C.E = 12%, MS . A * represents the mass selected precursor ion.

CID 63 - 63 - [ Cu2(O2CH)3] → [ Cu2H(O2CH)2] + CO2 (S2a) 261 217 CID 63 - 63 - [ CuH(O2CH)2] → [ Cu2H(O2CH)2] + CO2 (S2b) 217 172 CID 63 - 63 - [ Cu2H2(O2CH)] → [ Cu2H3] + CO2 (S2c) 172 129

328 Figure S4. Results of triplicate experiments for pseudo-first order kinetics obtained in a LTQ 63 - 2D linear-ion trap quasi thermalized to 298 K for the ion-molecule reaction of [ CuH2] , m/z 65, with the neutral reagents: (a) 1-butanethiol, (b) allyl iodide, (c) methyl iodide, (d) 2,2,2- trifluoroethanol (e) carbon disulfide, and (f) carbon dioxide. The black dots are the plotted 63 + coordinates for the activation time (s) (x-axis) vs. ln(Relative Intensity of [ CuH2] ) (y-axis). The red line represents the linear trendline of best-fit using Excel.

329 Figure S5. Results of triplicate experiments for pseudo-first order kinetics obtained in a LTQ 63 - 2D linear-ion trap quasi thermalized to 298 K for the ion-molecule reaction of [ Cu2H3] , m/z 129, with the neutral reagents: (a) 1-butanethiol, (b) allyl iodide, (c) 2,2,2-trifluoroethanol (d) carbon disulfide, and (e) carbon dioxide. The black dots are the plotted coordinates for the 63 + activation time (s) (x-axis) vs. ln(Relative Intensity of [ CuH2] ) (y-axis). The red line represents the linear trendline of best-fit using Excel.

330

Figure S6. LTQ mass spectra showing the ion-molecule reaction of different isotopologues - 9 -3 63 - of [Cu2H3] at 3000 ms with CS2 ([CS2]ion trap = 6.4 x 10 molecules cm ) for: (a) [ Cu2H3] , 63/65 - m/z 129, and (b) [ Cu2H3] . A * represents the mass selected precursor ion.

331 Figure S7. LTQ mass spectra showing the ion-molecule reactions of hydrido cuprates with 63 – methyl iodide for: (a) [ CuH2] m/z 65, activation time = 10 ms [methyl iodide]ion trap = 2.9 x 9 -3 63 – 10 molecules cm ; (b) [ Cu2H3] m/z 129, activation time = 10000 ms, [methyl iodide]ion trap = 8.4 x 109 molecules cm-3. The most intense peak in the ion is represented by the m/z value. * Represents the mass selected precursor ion.

S10

332 Figure S8. LTQ mass spectra showing the ion-molecule reactions of hydrido cuprates with 63 – 9 allyliodide for: (a) [ CuH2] m/z 65, activation time = 60 ms, [allyliodide]ion trap = 2.8 x 10 -3 – 9 molecules cm ; (b) [Cu2H3] m/z 129, activation time = 100 ms, [allyliodide]ion trap = 4.3 x 10 molecules cm-3. The most intense peak in the ion is represented by the m/z value. * Represents the mass selected precursor ion.

S11

333

Figure S9. LTQ mass spectrum showing the ion-molecule reaction of methyl iodide (CH3I) 63 - 9 -3 with mass selected [ CuHD] . [CH3I]ion trap = 6.8 x 10 molecules cm , activation time = 2000 ms. The most intense peak in the ion is represented by the m/z value. * Represents the mass selected precursor ion. K.I.E = 1.51, as derived from a comparison of the integrated ion abundances of the peaks at m/z 191 and 192.

S12

334 Figure S10. LTQ mass spectrum showing the ion-molecule reaction of allyl iodide 63 - 9 - (CH2=CHCH2I) with mass selected [ CuHD] . [CH2=CHCH2I]ion trap = 4.7 x 10 molecules cm 3, activation time = 2000 ms. The most intense peak in the ion is represented by the m/z value. * Represents the mass selected precursor ion. K.I.E = 0.9, as derived from a comparison of the integrated ion abundances of the peaks at m/z 191 and 192.

S13

335 Figure S11. LTQ mass spectra showing the ion-molecule reactions of hydrido cuprates with 63 – 2,2,2-trifluoroethanol for: (a) [ CuH2] m/z 65, activation time = 10 ms, [2,2,2- 9 -3 63 – trifluorethanol]ion trap = 2.2 x 10 molecules cm ; (b) [ Cu2H3] m/z 129, activation time = 9 -3 3000 ms, [2,2,2-trifluorethanol]ion trap = 6.1 x 10 molecules cm . The most intense peak in the ion is represented by the m/z value. * Represents the mass selected precursor ion.

S14

336

Figure S12. LTQ mass spectrum showing the ion-molecule reaction of 2,2,2-trifluorethanol 63 - 9 -3 (CF3CH2OH) with mass selected [ CuHD] . [CF3CH2OH]ion trap = 3.1 x 10 molecules cm , activation time = 3200 ms. The most intense peak in the ion is represented by the m/z value. * Represents the mass selected precursor ion. K.I.E = 0.85, as derived from a comparison of the integrated ion abundances of the peaks at m/z 163 and 164.

S15

337

Figure S13. LTQ mass spectra showing the ion-molecule reactions of hydrido cuprates with – 1-butanethiol for: (a) [CuH2] m/z 65, activation time = 10 ms, [1-butanethiol]ion trap = 2.20 x 9 -3 – 10 molecules cm ; (b) [Cu2H3] m/z 129, activation time = 3000 ms, [1-butanethiol]ion trap = 3.9 x 109 molecules cm-3. The most intense peak in the ion is represented by the m/z value. * Represents the mass selected precursor ion.

S16

338

Figure S14. LTQ mass spectrum showing the ion-molecule reaction of 1-butanethiol 63 - 9 - (CH3(CH2)3SH) with mass selected [ CuHD] . [CH3(CH2)3SH]ion trap = 2.2 x 10 molecules cm 3, activation time = 300 ms. The most intense peak in the ion is represented by the m/z value. * Represents the mass selected precursor ion. K.I.E = 0.86, as derived from a comparison of the integrated ion abundances of the peaks at m/z 153 and 154.

S17

339 CCSD(T) (b) S S S C S S C Cu C H Cu H Cu H S TS 1A-5 H H 17.8 (3.1) H TS1A-2 S TS3-4 C S 13.5 (-1.1) 13.6 (-3.5) Cu

CS2 TS2-3 H2 4.2 (-11.7)

3.7 (-14.2) S 0.0 (0.0) -0.7 (-16.4) C S

S Cu Cu HH -3.0 (-19.4) -4.5 (-4.8) 1A C S H S H Cu H C S H Cu -7.9 (-8.0) TS5-6 3 H S -29.8 (-48.0) Cu S TS2-5 C S C H HH TS5-6 2 Cu S -35.4 (-53.8) -29.8 (-48.0) 4 H -36.4 (-54.9) S C S S CS -35.4 (-53.8) Cu H -36.4 (-54.9) Cu S H H CS H C S H S 6 5 Cu Cu H H 5 6

(c) (a) Figure S15. Energy profile obtained from DFT calculations of the competing mechanisms – for the reaction of [CuH2] with CS2: (a) hydride transfer pathway (red); (b) addition/H2 – elimination to form [CuCS2] (black); (c) adduct formation (blue, not observed experimentally). The relative Gibbs and potential energies (in parentheses) obtained from the CCSD(T)/BS2//M06/BS1 calculations are given in kcal mol-1.

S18

340

S19

341

S20

342 385

380 y = 1.8532x + 373.03 methanol R² = 0.97 375

370 (kcal/mol)

acid 365 triﬂuorethanol

Δ H 360

355 1-butanthiol

350 -10 -8 -6 -4 -2 0 2 4 6 activation potential energy (kcal/mol)

Figure S18. Correlation between the gas-phase acidity, ΔHacid(kcal/mol) (y-axis) and activation potentials (kcal/mol) (x-axis) of 1-buanethiol, 2,2,2-trifluoroethanol and methanol.

S21

343 1.567 Cu HH

1.588 H Cu D C I TS1A-9 D Cu H C I H3 H3

G1 = -249.862522 au G2 = -249.862554 au G1 - G2 = 0.000032 au KIE = 0.97

1.565 H2 H2 D H I H D I TS1A-12 Cu C Cu C

G1 = -327.167253 au G2 = -327.167260 au G1 - G2 = 0.000007 au KIE = 0.99

H 1.750 D TS18-19 D Cu H H Cu H CF O 3 CF C O 3 C H2 H2 G1 = -651.197091 au G2 = -651.197496 au G1 - G2 = 0.000405 au KIE = 0.66

1.638 H D TS26-27 D Cu H H Cu H

S C3H7 S C3H7 C C H2 H2

G1 = -754.985543 au G2 = -754.986013 au G1 - G2 = 0.000470 au KIE = 0.61

Figure S19. DFT calculated KIEs based on the Gibbs free energies obtained at the M06 level using the conditions of T = 298.15 K; P = 2 × 10−3 Torr.

S22

344 63 - Table S1: Gas-phase kinetics associated with the ion-molecule reactions of [ CuH2] , m/z 65, with various neutral reagents.

a Reactant ion kexpt kexpt kexpt kexpt kexpt kexpt run1 run2 run3 run4 run5 run6 (R2) (R2) (R2) (R2) (R2) (R2)

1-butanethiol 7.9 x 1.0 x 9.8 x 8.9 x 8.4 x 7.6 x 10-09 10-08 10-09 10-09 10-09 10-09

(0.99) (0.99) (0.99) (0.99) b (0.99) c (0.99) c

Allyl iodide 9.9 x 1.1 x 1.2 x 10-10 10-09 10-09

(0.99) (0.99) (0.99)

Methyl iodide 4.5 x 4.5 x 5.1 x 10-10 10-10 10-10

(0.98) (0.98) (0.98)

2,2,2- 1.1 x 5.8 x 4.9 x trifluoroethanol 10-09 10-10 10-10

(0.99) (0.99) (0.99)

Carbon disulphide 2.1 x 2.6 x 1.8 x 10-10 10-10 10-10

(0.97) (0.99) (0.99)

Carbon dioxide 2.6 x 2.6 x 3.2 x 10-15 10-15 10-15

(0.99) (0.99) (0.99)

a In units of: cm3 molecules-1 s-1. b a much faster flow rate of helium and a lower flow rate of neutral was used to lower the concentration of 1-butanethiol in the ion trap. c 1-butanethiol was diluted with methanol to half its concentration, 1-butanethiol:methanol (1:1 v/v), and this was injected into the ion trap to achieve a lower the concentration of 1- butanethiol in the ion trap.

S23

345 63 - Table S2: Gas-phase kinetics associated with the ion-molecule reactions of [ Cu2H3] , m/z 129, with various neutral reagents.

Reactant ion kexpt run1 kexpt run2 kexpt run3 (R2) (R2) (R2)

1-butanethiol 1.3 x 10-11 1.3 x 10-11 1.6 x 10-11

(0.91) (0.96) (0.95)

Allyl iodide 1.2 x 10-10 1.2 x 10-10 1.4 x 10-10

(0.99) (0.98) (0.98)

Methyl iodide N.R. N.R. N.R.

2,2,2- 5.0 x 10-12 5.0 x 10-12 6.2 x 10-12 trifluoroethanol (0.91) (0.99) (0.99)

Carbon 1.6E x 10-11 1.6 x 10-11 1.8 x 10-12 disulphide (0.99) (0.98) (0.99)

Carbon dioxide 1.3 x 10-15 1.3 x 10-15 1.3 x 10-15

(0.99) (0.99) (0.99) a In units of: cm3 molecules-1 s-1. N.R. = No reaction

S24

346 Cartesian Coordinates for:

1A E(M06/BS1) = -198.582625 au G(M06/BS1) = -198.591504 au H(M06/BS1) = -198.56871 au E(M06/BS2//M06/BS1) = -1641.685465 au

Cu 0.00000000 0.00000000 0.00000000 H 0.00000000 0.00000000 1.56656200 H 0.00000000 0.00000000 -1.56656200

2C E(M06/BS1) = -396.5824868 au H(M06/BS1) = -396.560696 au G(M06/BS1) = -396.593119 au E(M06/BS2//M06/BS1) = -3282.79151 au

Cu 0.00000000 0.00000000 1.17575500 H 0.00000000 1.53987200 1.00649400 H 0.00000000 -1.53987200 1.00649400 Cu 0.00000000 0.00000000 -1.15313000 H 0.00000000 0.00000000 -2.66911300

2B E(M06/BS1) = -396.6053938 au H(M06/BS1) = -396.58291 au G(M06/BS1) = -396.61266 au E(M06/BS2//M06/BS1) = -3282.81705 au

Cu 0.00000000 0.00000000 1.59257500 H 0.00000000 0.00000000 3.11917200 H 0.00000000 0.00000000 0.00000000 Cu 0.00000000 0.00000000 -1.59257500 H 0.00000000 0.00000000 -3.11917200

2A E(M06/BS1) = -396.6124253 au H(M06/BS1) = -396.587827 au G(M06/BS1) = -396.620867 au E(M06/BS2//M06/BS1) = -3282.823753 au

Cu 0.00873200 -1.26865800 0.00000000 H -0.75591400 -2.59799900 0.00000000 H 1.01510900 0.00063500 0.00000000 Cu 0.00873200 1.26881300 0.00000000 H -0.76565900 2.59285700 0.00000000

TS1A-2 E(M06/BS1) = -1032.978881 au

S25

347 H(M06/BS1) = -1032.954689 au G(M06/BS1) = -1032.992653 au E(M06/BS2//M06/BS1) = -2476.158162 au

Cu 1.42637200 -0.48486500 -0.00000300 C -1.07673800 0.29369700 0.00008200 S -2.39370200 -0.59571700 -0.00005700 S -0.01179800 1.46166600 0.00003700 H 2.74309500 0.37964500 -0.00075800 H 0.84054800 -1.93591400 0.00068200

2 E(M06/BS1) = -1032.999907 au H(M06/BS1) = -1032.973786 au G(M06/BS1) = -1033.010586 au E(M06/BS2//M06/BS1) = -2476.176596 au

Cu 1.15662500 -0.50923900 0.00012700 C -0.68808200 0.06601000 0.00019300 S -2.23969200 -0.45592100 -0.00005900 S 0.17566200 1.51117300 -0.00007200 H 2.70182200 -0.52884200 -0.00159800 H 0.90903600 -1.98332300 -0.00114100

TS2-3 E(M06/BS1) = -1032.989927 au H(M06/BS1) = -1032.964986 au G(M06/BS1) = -1033.00147 au E(M06/BS2//M06/BS1) = -2476.164806 au

Cu -1.19388300 -0.44578700 0.00002500 C 0.68433700 -0.07003700 -0.00015800 S 2.27914800 -0.50585800 0.00003100 S -0.08141600 1.49149300 -0.00000200 H -2.69953400 -0.69914500 0.00013000 H -1.94760000 -1.72296300 -0.00036900

3 E(M06/BS1) = -1032.996663 au H(M06/BS1) = -1032.967602 au G(M06/BS1) = -1033.00507 au E(M06/BS2//M06/BS1) = -2476.170447 au Cu 1.20017700 -0.44509600 -0.00006700 C -0.66481300 -0.10454600 0.00000800 S -2.26300700 -0.55322800 0.00024000 S 0.00486000 1.52156100 -0.00009000 H 2.81857800 -0.58244400 -0.00028900 H 2.49551800 -1.37581600 -0.00023400

TS3-4

S26

348 E(M06/BS1) = -1032.989496 au H(M06/BS1) = -1032.96081 au G(M06/BS1) = -1032.999101 au E(M06/BS2//M06/BS1) = -2476.157483 au

Cu 1.27691900 -0.29340000 -0.00125000 C -0.58315900 0.02094000 0.00318400 S -1.77194800 -1.15372700 -0.00209900 S -0.72748800 1.71354100 0.00240800 H 3.27051900 0.05396400 -0.15857500 H 3.18874700 -0.62802300 0.17077600

4 E(M06/BS1) = -1031.824016 au H(M06/BS1) = -1031.811372 au G(M06/BS1) = -1031.8472 au E(M06/BS2//M06/BS1) = -2474.992154 au

Cu 1.38775000 -0.00002400 -0.00006300 C -0.52144500 0.00000500 0.00003000 S -1.15984200 1.53348800 0.00003200 S -1.15991300 -1.53344600 0.00007000

TS2-5 E(M06/BS1) = -1032.995042 au H(M06/BS1) = -1032.97027 au G(M06/BS1) = -1033.006882 au E(M06/BS2//M06/BS1) = -2476.171298 au

Cu 1.21567200 -0.49128500 0.00003000 C -0.67065400 0.02160300 -0.00001800 S -2.23702500 -0.48785900 -0.00006300 S 0.09841700 1.51957000 0.00001000 H 2.73353900 -0.74255300 0.00007500 H 0.25362600 -1.64719000 0.00000000

5 E(M06/BS1) = -1033.063851 au H(M06/BS1) = -1033.033233 au G(M06/BS1) = -1033.071388 au E(M06/BS2//M06/BS1) = -2476.237275 au

Cu 1.84748900 -0.19950300 -0.00003600 C -1.25962600 -0.37685700 0.00023400 S -2.91862000 -0.23770900 -0.00013600 S -0.10470300 0.88487900 0.00007000 H 3.18646700 -0.92045200 -0.00048600 H -0.83272700 -1.38754000 0.00118400

TS5-6

S27

349 E(M06/BS1) = -1033.051847 au H(M06/BS1) = -1033.022391 au G(M06/BS1) = -1033.05972 au E(M06/BS2//M06/BS1) = -2476.225305 au

Cu -1.48747800 -0.40443600 0.01007600 C 1.26612000 0.40502700 0.54282600 H 1.40550300 0.61977200 1.61116800 H -2.55112500 -1.48326300 0.12276700 S -0.04121900 1.28659900 -0.17605300 S 2.33408000 -0.65147500 -0.15414200

6 E(M06/BS1) = -1033.060937 au H(M06/BS1) = -1033.030331 au G(M06/BS1) = -1033.068479 au E(M06/BS2//M06/BS1) = -2476.234827 au

Cu 1.46181400 -0.39097700 -0.00000100 C -1.55830000 0.69013300 -0.00001700 H -2.31576600 1.48671100 -0.00005300 H 2.55996100 -1.44645600 -0.00002900 S 0.02537900 1.32686100 0.00000800 S -2.10581600 -0.87953100 0.00000500

TS1A-5 E(M06/BS1) = -1032.97738 au H(M06/BS1) = -1032.954827 au G(M06/BS1) = -1032.990788 au E(M06/BS2//M06/BS1) = -2476.154922 au

Cu 2.16847500 0.00016900 0.00009800 H 3.72084100 -0.00017700 0.00060100 H 0.58601300 0.00050100 -0.00038200 C -1.53383300 -0.00007900 -0.00010400 S -1.81256900 1.55785100 -0.00007400 S -1.81178200 -1.55814800 -0.00007800

7 E(M06/BS1) = -387.1460826 au H(M06/BS1) = -387.11125 au G(M06/BS1) = -387.14629 au E(M06/BS2//M06/BS1) = -1830.31758 au

Cu 1.22053500 0.02466200 -0.00000900 O -2.79238000 0.00094600 -0.00006800 C -1.59435700 0.28552400 0.00005100 O -0.60615400 -0.52826100 0.00005700 H -1.30902400 1.37244200 0.00014400 H 2.66791800 0.41773400 -0.00009100

S28

350 TS7-8 E(M06/BS1) = -387.1403575 au H(M06/BS1) = -387.106748 au G(M06/BS1) = -387.141244 au E(M06/BS2//M06/BS1) = -1830.312246 au

Cu -1.20964700 -0.06633400 0.02349700 O 2.57354400 -0.55866300 -0.01392100 C 1.74498200 0.33433600 0.16465900 O 0.57715600 0.45313900 -0.33412200 H 2.04132600 1.19513700 0.82718400 H -2.63705100 -0.43328200 0.28778800

8 E(M06/BS1) = -387.1461532 au H(M06/BS1) = -387.111418 au G(M06/BS1) = -387.1464 au E(M06/BS2//M06/BS1) = -1830.317911 au

Cu 1.07512200 -0.07598200 0.00006100 O -1.99740600 -0.88950900 -0.00036800 C -1.72666600 0.31227500 0.00063600 O -0.58759600 0.88859200 -0.00086100 H -2.57602200 1.04802600 0.00310800 H 2.43750700 -0.71085000 0.00113200

TS1A-9 E(M06/BS1) = -249.8633246 au H(M06/BS1) = -249.809773 au G(M06/BS1) = -249.848589 au E(M06/BS2//M06/BS1) = -1979.353332 au

Cu -3.75950600 0.00265300 0.00016900 H -5.30214200 0.06553800 -0.00126500 H -2.17266900 -0.05954700 -0.00038700 C -0.19935200 -0.03095400 -0.00080200 H -0.37709100 -0.61244700 -0.89454700 H -0.40061100 1.02961000 -0.05540800 H -0.37729600 -0.51789000 0.94794400 I 2.24248300 0.00384000 0.00006800

9 E(M06/BS1) = -249.9370261 au H(M06/BS1) = -249.877628 au G(M06/BS1) = -249.919997 au E(M06/BS2//M06/BS1) = -1979.423926 au

Cu -3.62230300 0.05401300 -0.00003000 H -5.09356100 0.27433400 0.00012500

S29

351 H -2.13576500 -1.14675800 -0.00219800 C -1.41154400 -0.30376500 0.00016000 H -0.77675900 -0.38794300 -0.88454500 H -1.83514500 0.72361300 0.00091400 H -0.77886300 -0.39049700 0.88610100 I 2.34219100 0.02232900 -0.00000900

10 E(M06/BS1) = -238.4263501 au H(M06/BS1) = -238.371014 au G(M06/BS1) = -238.402168 au E(M06/BS2//M06/BS1) = -1681.545722 au

Cu 0.00000000 0.00000000 0.70197100 H 0.00000000 0.00000000 2.17693200 H 0.00000000 0.00000000 -1.15822800 C 0.00000000 0.00000000 -2.26243100 H 0.00000000 1.03682200 -2.60042300 H 0.89791400 -0.51841100 -2.60042300 H -0.89791400 -0.51841100 -2.60042300

11 E(M06/BS1) = -209.4958715 au H(M06/BS1) = -209.485173 au G(M06/BS1) = -209.515601 au E(M06/BS2//M06/BS1) = -1938.972131 au

Cu 0.00000000 0.00000000 -1.59632500 H 0.00000000 0.00000000 -3.11328500 I 0.00000000 0.00000000 0.93220200

TS1A-12 E(M06/BS1) = -327.1970706 au H(M06/BS1) = -327.106986 au G(M06/BS1) = -327.153219 au E(M06/BS2//M06/BS1) = -2056.709988 au

Cu -3.06750700 -0.49153100 -0.01048700 H -3.76933900 -0.49277400 -1.39453800 H -2.43865700 -0.53889100 1.42173100 C -0.45316800 -0.05898900 0.40362800 H -0.34205000 0.02215200 1.47701000 H -0.59821600 -1.05341500 0.00283100 I 2.11588100 -0.20317800 -0.02932100 C -0.82254900 1.11403900 -0.37442400 C -0.91015100 2.34678400 0.14529400 H -1.17906200 3.20338000 -0.46642000 H -0.73211900 2.52178400 1.20580500 H -1.00936000 0.94959400 -1.43525900

S30

352 12 E (M06/BS1) = -327.295575 au H(M06/BS1) = -327.199184 au G(M06/BS1) = -327.246247 au E(M06/BS2//M06/BS1) = -2056.805445 au

Cu 3.29617800 -0.19807700 -0.28817600 H 4.59118100 -0.09812200 -1.06736600 H 1.57116400 2.08611200 1.05714500 C 1.01047300 1.61688500 0.23990200 H -0.06474600 1.82149900 0.35478300 H 1.32333500 2.06481300 -0.71237300 I -2.48949500 -0.09509400 -0.07074600 C 1.18441700 0.13853200 0.22090800 C 1.84801100 -0.57857400 1.16992300 H 1.78590100 -1.66550500 1.18983200 H 2.26029400 -0.08994500 2.05510500 H 0.62954800 -0.39569200 -0.55491800

13 E(M06/BS1) = -327.270741 au H(M06/BS1) = -327.174907 au G(M06/BS1) = -327.223386 au E(M06/BS2//M06/BS1) = -2056.782509 au

Cu 3.45230300 -0.63685000 -0.00005800 H 4.85790100 -1.12303000 0.00066600 H 1.84567200 -0.15868900 -0.93174400 C 1.40677100 0.27848200 -0.00092300 C 1.49855500 1.77163000 0.00181800 C 0.42412700 2.56056600 -0.00132200 H 0.52504200 3.64443500 0.00111300 H -0.58316800 2.13563800 -0.00593300 H 2.50094200 2.20971400 0.00685400 H 1.84313400 -0.16188100 0.92953900 H 0.36185700 -0.07909900 -0.00282600 I -2.48009200 -0.29551900 0.00012400

14 E(M06/BS1) = -315.7911609 au H(M06/BS1) = -315.697143 au G(M06/BS1) = -315.733776 au E(M06/BS2//M06/BS1) = -1758.934305 au

Cu -0.93683900 -0.18361500 -0.00349300 H -2.18727300 -1.01648800 -0.09212000 C 1.92772000 -0.74864200 -0.16721000 C 1.10126300 0.32141700 0.46722800 C 0.50264700 1.34239700 -0.20514500 H 0.61390000 1.44609900 -1.28518300

S31

353 H 0.03107100 2.16943800 0.32292900 H 1.06911900 0.32924600 1.55949400 H 1.58740000 -1.74576200 0.13888000 H 2.97508200 -0.65660600 0.14826300 H 1.88926300 -0.69212600 -1.26019700

15 E(M06/BS1) = -209.4958715 au H(M06/BS1) = -209.485173 au G(M06/BS1) = -209.515601 au E(M06/BS2//M06/BS1) = -1938.972131 au

Cu 0.00000000 0.00000000 -1.59632500 H 0.00000000 0.00000000 -3.11328500 I 0.00000000 0.00000000 0.93220200

TS1A-14 E(M06/BS1) = -327.191274 au H(M06/BS1) = -327.102323 au G(M06/BS1) = -327.146592 au E(M06/BS2//M06/BS1) = -2056.706696 au

Cu -4.54998500 -0.18320000 0.04890600 H -6.05678100 -0.51180300 -0.07575700 H -3.00610000 0.15890800 0.18081800 C -1.21147100 0.42795000 -0.75641200 C -0.26637100 1.06162000 -0.00466300 C 0.59257800 0.35655600 0.88851400 H 0.98498500 0.89529500 1.74790800 H 0.30082300 -0.66726600 1.12107000 H -0.17652300 2.14668100 -0.05115100 H -1.83112000 0.95703100 -1.47058600 H -1.24766400 -0.65670300 -0.81164800 I 2.79799100 -0.15256800 -0.05327500

16 E(M06/BS1) = -314.2662771 au H(M06/BS1) = -314.196546 au G(M06/BS1) = -314.23439 au E(M06/BS2//M06/BS1) = -1757.409877 au

Cu 1.25490800 -0.01412400 0.00103600 H 2.22181200 1.19656800 -0.07416000 H 0.29054000 -1.25811700 0.08133500 H -1.29058500 -1.02710400 0.01573900 O -2.23175200 -0.72248700 -0.02199700 C -2.19800400 0.67622500 0.01108100 H -1.16298500 1.05523100 -0.04681700 H -2.76207100 1.09794700 -0.83706300 H -2.64700600 1.06760900 0.94042700

S32

354 TS16-17 E(M06/BS1) = -314.2307774 au H(M06/BS1) = -314.165429 au G(M06/BS1) = -314.201248 au E(M06/BS2//M06/BS1) = -1757.376872 au

Cu -0.93697300 -0.04670700 0.00754300 H -2.07244600 -0.91523800 -0.50122100 H -0.27860900 1.10519600 1.23942500 H 0.36819000 1.02663600 0.64046900 O 1.25156500 0.57428800 -0.42157200 C 2.04290000 -0.39165500 0.13514900 H 1.74645300 -0.65144600 1.18682900 H 2.02103900 -1.35744500 -0.42068300 H 3.11765600 -0.09758000 0.19810200

17 E (M06/BS1) = -313.0879309 au H(M06/BS1) = -313.036811 au G(M06/BS1) = -313.070977 au E(M06/BS2//M06/BS1) = -1756.231938 au

Cu 0.82792000 0.03428400 0.00000000 H 2.27147100 0.48242100 0.00000000 O -0.93073200 -0.60044400 0.00000100 C -1.97607000 0.27764000 0.00000100 H -2.01019000 0.95728600 0.88699700 H -2.00987900 0.95765100 -0.88672900 H -2.95881700 -0.25386600 -0.00028000

18 E(M06/BS1) = -651.2383524 au H(M06/BS1) = -651.158633 au G(M06/BS1) = -651.205408 au E(M06/BS2//M06/BS1) = -2094.506548 au

Cu 2.55617400 0.16499400 0.03120100 H 3.28264000 1.20239000 -0.85410800 H 1.82267800 -0.89920300 0.93803600 H 0.54676000 -1.44977900 0.35416700 O -0.33450900 -1.64047800 -0.08291300 C -0.71171000 -0.47269800 -0.71719200 H 0.11560500 0.26121400 -0.76469500 H -1.06115100 -0.66506000 -1.74408600 C -1.85823900 0.19645100 0.00263200 F -1.55010800 0.56881200 1.24985500 F -2.94590400 -0.59567400 0.08903200 F -2.25285500 1.30985200 -0.65927300

S33

355

TS18-19 E(M06/BS1) = -651.212944 au H(M06/BS1) = -651.137543 au G(M06/BS1) = -651.183009 au E(M06/BS2//M06/BS1) = -2094.48256 au

Cu 2.53201200 0.06732600 0.04287300 H 3.66143900 0.83137900 -0.60842400 H 1.74038400 -0.77057500 1.35898400 H 1.15289700 -0.82181300 0.73301300 O 0.26086500 -0.63253300 -0.51031900 C -0.51431800 0.43588300 -0.25030800 H -0.22046700 1.01863900 0.66589800 H -0.56934900 1.18980900 -1.07174000 C -1.96056000 0.06206100 0.01971500 F -2.10601700 -0.73627100 1.09496600 F -2.54931800 -0.57425800 -1.01004800 F -2.72587600 1.16305200 0.26453300

19 E(M06/BS1) = -650.0703816 au H(M06/BS1) = -650.009851 au G(M06/BS1) = -650.054903 au E(M06/BS2//M06/BS1) = -2093.337986 au

Cu -2.35256500 0.01804300 0.00001300 H -3.82560700 0.32192300 0.00006700 O -0.51749800 -0.40850500 -0.00006600 C 0.39742000 0.58178500 -0.00014300 H 0.37615400 1.25896600 -0.88867500 H 0.37604600 1.25920400 0.88820900 C 1.81094300 0.02941700 0.00001900 F 2.08940000 -0.72676100 -1.07778900 F 2.08929400 -0.72648500 1.07800300 F 2.73104000 1.03519000 -0.00007000

CuH E(M06/BS1) = -197.9415767 au H(M06/BS1) = -197.9338 au G(M06/BS1) = -197.956067 au E(M06/BS2//M06/BS1) = -1641.046206 au

Cu 0.00000000 0.00000000 0.04887800 H 0.00000000 0.00000000 -1.41745500

HCS2 E(M06/BS1) = -835.0450981 au H(M06/BS1) = -835.024301 au G(M06/BS1) = -835.054858 au

S34

356 E(M06/BS2//M06/BS1) = -835.1093429 au

C -0.00000100 0.55031400 -0.00022800 S -1.53288000 -0.15477900 0.00002600 S 1.53288000 -0.15477900 0.00002600 H 0.00000200 1.65105500 0.00053500

HCO2 E(M06/BS1) = -189.1243809 au H(M06/BS1) = -189.100196 au G(M06/BS1) = -189.127916 au E(M06/BS2//M06/BS1) = -189.1859592 au

C 0.00000000 0.31417200 -0.00000400 H 0.00001000 1.45714900 0.00000700 O -1.13668600 -0.20888500 0.00000100 O 1.13668400 -0.20888700 0.00000100

H2 E(M06/BS1) = -1.169301 au H(M06/BS1) = -1.155906 au G(M06/BS1) = -1.170694 au E(M06/BS2//M06/BS1) = -1.170482au

H 0.00000000 0.00000000 0.37045200 H 0.00000000 0.00000000 -0.37045200

CS2 E(M06/BS1) = -834.394854 au H(M06/BS1) = -834.383801 au G(M06/BS1) = -834.410759 au E(M06/BS2//M06/BS1) = -834.470684 au

C 0.00000000 0.00000000 0.00000000 S 0.00000000 0.00000000 1.55997500 S 0.00000000 0.00000000 -1.55997500

CO2 E(M06/BS1) = -188.505962 au H(M06/BS1) = -188.490449 au G(M06/BS1) = -188.514696 au E(M06/BS2//M06/BS1) = -188.584194 au

C 0.00000000 0.00000000 0.00000000 O 0.00000000 0.00000000 1.16473700 O 0.00000000 0.00000000 -1.16473700

CH3I E(M06/BS1) = -51.265134 au H(M06/BS1) = -51.224519 au G(M06/BS1) = -51.253348 au

S35

357 E(M06/BS2//M06/BS1) = -337.658427 au C 0.00000000 0.00000000 -1.83458200 H 0.00000000 1.03690500 -2.16917700 H -0.89798600 -0.51845200 -2.16917700 H 0.89798600 -0.51845200 -2.16917700 I 0.00000000 0.00000000 0.33047200

CH4 E(M06/BS1) = -40.479276 au H(M06/BS1) = -40.430839 au G(M06/BS1) = -40.451977 au E(M06/BS2//M06/BS1) = -40.492827 au

C 0.00000000 0.00000000 0.00000000 H 0.63072600 0.63072600 0.63072600 H -0.63072600 -0.63072600 0.63072600 H -0.63072600 0.63072600 -0.63072600 H 0.63072600 -0.63072600 -0.63072600

I E(M06/BS1) = -11.484070 au H(M06/BS1) = -11.48171 au G(M06/BS1) = -11.500919 au E(M06/BS2//M06/BS1) = -297.849722 au

I 0.00000000 0.00000000 0.00000000

CH2=CHCH2I E(M06/BS1) = -128.596839 au H(M06/BS1) = -128.521161 au G(M06/BS1) = -128.557068 au E(M06/BS2//M06/BS1) = -415.014092 au

C 0.98604200 1.11368600 0.00014600 C 2.26757900 0.35770500 -0.00018000 C 2.46375600 -0.95783600 0.00005600 H 3.46907600 -1.36958200 0.00003400 H 1.63414800 -1.66276700 0.00038300 H 0.89847800 1.75188600 -0.88450500 H 0.89849200 1.75114500 0.88535600 H 3.13928300 1.01931300 -0.00048800 I -0.83667400 -0.08625100 -0.00001700

CH3OH E(M06/BS1) = -115.659462 au H(M06/BS1) = -115.603973 au G(M06/BS1) = -115.631132 au E(M06/BS2//M06/BS1) = -115.699441 au

O 0.74283000 -0.12217900 0.00000200 C -0.65929100 0.02096500 -0.00000100

S36

358 H -1.02778000 0.54615900 0.89345300 H -1.02794600 0.54437800 -0.89443500 H -1.08283100 -0.98682100 0.00102900 H 1.15166600 0.74792400 -0.00005300

CF3CH2OH E(M06/BS1) = -452.61921 au H(M06/BS1) = -452.554685 au G(M06/BS1) = -452.591094 au E(M06/BS2//M06/BS1) = -452.784961 au

O 1.98816500 -0.07884500 0.11291300 C 0.89449300 0.76300000 -0.08614000 H 0.90816600 1.27513200 -1.06006000 H 0.90550700 1.52115900 0.70279900 C -0.41668500 0.01383600 0.00638700 F -1.44527200 0.83785000 -0.23649800 F -0.46147900 -0.97564800 -0.90658200 F -0.60775200 -0.54150200 1.20350400 H 2.04468000 -0.71285100 -0.61133700

TS'1A-20 E(M06/BS1) = -327.191687 au H(M06/BS1) = -327.101822 au G(M06/BS1) = -327.146095 au E(M06/BS2//M06/BS1) = -2056.706090 au Cu -2.33005300 -0.73489900 0.02475500 H -2.45632400 -0.99044600 1.56971700 C 0.44070400 1.81443400 0.01791700 C -0.89866000 1.66511100 0.46764600 C -1.94689800 1.47326100 -0.40055600 H -1.79660700 1.57953300 -1.47155700 H -2.97716500 1.51932800 -0.05067900 H -1.08109600 1.56395800 1.53766000 H 1.13887100 2.32635100 0.67577200 H 0.57522400 2.05785700 -1.03571100 H -2.36542400 -0.67869600 -1.52574000 I 1.71628600 -0.29778600 -0.01749800

TS1A-20 E(M06/BS1) = -327.196909 au H(M06/BS1) = -327.106900 au G(M06/BS1) = -327.153808 au E(M06/BS2//M06/BS1) = -2056.708325 au C 0.64958600 -0.05733000 -0.19999900 H 0.56451000 -0.10668600 -1.27672700 H 0.65553200 -0.98675200 0.35648600 I -2.09563700 -0.19517200 0.00422700 C 0.88961900 1.20885200 0.45964900 C 1.01074300 2.37003100 -0.20369300

S37

359 Cu 2.94741900 -0.53296700 -0.00689000 H 3.14333800 -0.60153200 1.53971300 H 2.82549700 -0.51926500 -1.54987500 H 0.97013900 1.17747000 1.54515600 H 1.18413900 3.30825000 0.31619500 H 0.95074200 2.39937900 -1.29091100

TS1A-21 E(M06/BS1) = -327.198799 au H(M06/BS1) = -327.109129 au G(M06/BS1) = -327.154477 au E(M06/BS2//M06/BS1) = -2056.715048 au Cu -3.20706000 -0.34969000 -0.05093900 H -2.55536500 -1.64304000 0.57072900 C -0.10002600 -0.41777700 0.12882500 C -0.74629300 0.79173800 0.62129200 C -1.40652600 1.64931300 -0.17762900 H -1.45621700 1.49998000 -1.25463300 H -1.92303900 2.52020400 0.21471100 H -0.71726200 0.96116600 1.69762000 H -0.11430800 -1.27357400 0.79802400 H -4.11044000 0.74162600 -0.71463100 I 2.22172700 -0.07754100 -0.04481700 H -0.35308000 -0.69529500 -0.89419300

21 E(M06/BS1) = -327.233178 au H(M06/BS1) = -327.141184 au G(M06/BS1) = -327.186684 au E(M06/BS2//M06/BS1) = -2056.744186 au Cu -2.83124600 -0.41895900 0.26010300 H -3.16152900 -1.85041200 0.06658200 C -1.45315600 -0.24349200 -1.28376600 C -1.27248000 0.78671700 -0.34897700 C -2.35307100 1.58069000 0.07709900 H -3.16682400 1.82183400 -0.61162100 H -2.23808900 2.25032400 0.92423000 H -0.35271000 0.73934100 0.24929700 H -0.63725000 -0.94264700 -1.45089400 H -3.95351400 -0.31388700 1.22614200 I 2.42066900 -0.04034900 0.06524000 H -2.20714700 -0.15975700 -2.07059100

TS21-12 E(M06/BS1) = -327.228153 au H(M06/BS1) = -327.137294 au G(M06/BS1) = -327.182474 au E(M06/BS2//M06/BS1) = -2056.739013 au Cu -2.88895800 -0.39326300 -0.29538900 H -3.67940600 0.62542900 -1.07603900

S38

360 C -2.33374300 1.53195400 0.03565300 C -1.25420500 0.67002100 0.41409900 C -1.43573200 -0.37494300 1.30381900 H -2.23116800 -0.35849400 2.05253900 H -0.63052300 -1.09266200 1.44320000 H -0.33838700 0.64383500 -0.19089000 H -3.07186400 1.82080200 0.78721300 H -2.15683200 2.27910900 -0.73242400 H -3.57366900 -1.72020600 -0.46000500 I 2.44535300 -0.03312000 -0.07129700

TS21-22 E(M06/BS1) = -327.232336 au H(M06/BS1) = -327.141412 au G(M06/BS1) = -327.185206 au E(M06/BS2//M06/BS1) = -2056.743305 au Cu -2.32752900 -0.64058500 -0.03337000 H -2.82870200 -1.55207300 -1.08443800 C -2.16853600 1.11341900 -1.12660500 C -1.40296500 1.19002600 0.04998200 C -1.95520300 0.89490100 1.30844900 H -3.00206900 1.11771600 1.52944100 H -1.29275000 0.76438300 2.15992100 H -0.31036200 1.11501500 -0.04355000 H -3.23497400 1.35286900 -1.11874700 H -1.67237600 1.15292900 -2.09230200 H -2.65697400 -1.72086500 0.91961600 I 2.18220300 -0.05364300 -0.01307900

22 E(M06/BS1) = -327.301549 au H(M06/BS1) = -327.207128 au G(M06/BS1) = -327.254203 au E(M06/BS2//M06/BS1) = -2056.815551 au Cu -0.59819300 0.40285900 -0.14320400 C -2.47933900 0.87008100 -0.34108800 C -3.30127700 0.06363500 0.56310000 C -4.11624500 -0.96681900 0.25383400 H -4.25170100 -1.28898400 -0.78028900 H -4.63249000 -1.53927900 1.02297400 H -3.21151200 0.31677200 1.62648500 H -2.76525800 0.73651700 -1.39556800 H -2.51003300 1.93947200 -0.08377800 I 1.83522600 -0.14278100 0.07632000 H -1.48724300 -1.78130600 -1.31582400 H -1.68000200 -2.30009700 -1.82114200

23 E(M06/BS1) = -326.128171 au H(M06/BS1) = -326.051436 au

S39

361 G(M06/BS1) = -326.094424 au E(M06/BS2//M06/BS1) = -2055.64140231 au Cu 0.63331800 -0.37987400 -0.03626600 C 2.51420200 -0.88202300 -0.08316900 C 3.35566900 0.19041200 0.45097800 C 4.15081500 1.04374400 -0.22785400 H 4.24456000 0.98310700 -1.31373700 H 4.68922400 1.84743400 0.27208300 H 3.30474900 0.32611700 1.53821200 H 2.77774000 -1.13907200 -1.12043200 H 2.55777800 -1.78803400 0.53991800 I -1.81253500 0.16366000 0.00558400

TS22-24 E(M06/BS1) = -327.232091 au H(M06/BS1) = -327.142161 au G(M06/BS1) = -327.185898 au E(M06/BS2//M06/BS1) = -2056.744210 au Cu -1.17844800 0.74099700 -0.08177300 C -2.08656600 -0.51861000 1.31140100 C -1.80373800 -1.21921600 0.09919200 C -2.33767000 -0.84473800 -1.12709600 H -3.25648000 -0.26174900 -1.18768600 H -1.97593800 -1.29782900 -2.04595200 H -0.93015000 -1.87153900 0.09395300 H -3.08767400 -0.11981600 1.47168800 H -1.53465200 -0.79037500 2.20792800 I 1.60544600 -0.09674300 -0.00344200 H -1.24535500 2.01971300 -0.87039500 H -1.51554600 1.45544100 1.18332700

24 E(M06/BS1) = -327.304233 au H(M06/BS1) = -327.208626 au G(M06/BS1) = -327.252880 au E(M06/BS2//M06/BS1) = -2056.817307 au Cu 0.92171300 0.91446300 -0.17067400 C 1.82226100 -1.95086000 -0.45681000 C 1.93199000 -0.85586900 0.56049600 C 2.77952300 0.21989300 0.42270500 H 3.41602100 0.32705200 -0.45593500 H 3.00242200 0.88503500 1.25519000 H 1.43465600 -1.02405200 1.51705400 H 2.42063600 -2.82970600 -0.16808400 H 0.78080700 -2.27471300 -0.57654900 I -1.51454300 -0.12909400 0.04478700 H 1.10146900 2.36817200 -0.71762800 H 2.18244500 -1.60822800 -1.43652800

TS22-25

S40

362 E(M06/BS1) = -327.238574 au H(M06/BS1) = -327.148352 au G(M06/BS1) = -327.192532 au E(M06/BS2//M06/BS1) = -2056.752225 au Cu -0.63016100 1.13324500 -0.15395800 C -2.01741800 0.11066200 0.96994500 C -2.24677600 -1.08162100 0.15402300 C -3.37232900 -1.37595300 -0.52503400 H -4.24383400 -0.71958300 -0.48940100 H -3.44344000 -2.25984900 -1.15565000 H -1.39019200 -1.75447700 0.05647900 H -2.94853200 0.58120900 1.30188300 H -1.34467700 -0.05104100 1.82078100 I 1.50240600 -0.35251400 0.00610100 H -0.24120500 2.35354800 -1.00686600 H -1.92179800 1.75080400 0.02057000

25 E(M06/BS1) = -327.312758 au H(M06/BS1) = -327.216687 au G(M06/BS1) = -327.265791 au E(M06/BS2//M06/BS1) = -2056.827252 au Cu 0.40164100 1.58751100 -0.18052400 C -2.71615900 0.46457900 0.92268000 C -2.92379300 -0.56566000 -0.13123800 C -4.11220600 -1.00090000 -0.55750500 H -5.04272200 -0.61115000 -0.14038300 H -4.20088900 -1.75381100 -1.33844400 H -2.00845500 -0.97282800 -0.57057600 H -3.66890900 0.84154900 1.31759000 H -2.11959600 0.05630700 1.74947200 I 1.24837700 -0.77921000 0.04889500 H -0.13169300 3.00315900 -0.30807900 H -2.12637800 1.30897300 0.53054300

26 E(M06/BS1) = -755.083473 au H(M06/BS1) = -754.929219 au G(M06/BS1) = -754.978741 au E(M06/BS2//M06/BS1) = -2198.265319 au Cu 2.37021600 -0.60887100 -0.08204300 H 2.54742400 -1.86245900 0.81992800 H 2.17261800 0.66254000 -0.99244600 H 0.72335800 1.61996000 -0.79450100 C -0.57975000 0.72974600 0.93050800 H 0.45209200 0.49113600 1.22889600 H -1.11410700 1.11689000 1.80932800 C -1.25393400 -0.51863900 0.38896700 S -0.46178700 2.10607200 -0.28056100 H -0.71295900 -0.84345700 -0.51394000

S41

363 H -1.11856800 -1.32977200 1.12471200 C -2.73328400 -0.34454100 0.08366100 H -2.85284100 0.47737600 -0.63767000 H -3.25950300 -0.02823400 1.00013700 C -3.36781500 -1.61462000 -0.46257300 H -3.26961300 -2.44385800 0.25162600 H -4.43574400 -1.48448400 -0.68081500 H -2.87111000 -1.92722900 -1.39042600

TS26-27 E(M06/BS1) = -755.077078 au H(M06/BS1) = -754.924162 au G(M06/BS1) = -754.971446 au E(M06/BS2//M06/BS1) = -2198.258587 au Cu 2.33491200 -0.79813100 -0.04320300 H 3.21238500 -1.52168600 0.95498300 H 1.44998200 -0.18174300 -1.27611300 H 1.22089900 0.61345500 -0.88705500 C -0.68438500 0.72154200 0.64650100 H -0.06667800 -0.18654400 0.80263900 H -1.00256500 1.04934000 1.64777600 C -1.89893600 0.34224900 -0.18625600 H -1.55121200 -0.02022000 -1.16806100 H -2.49040400 1.24897200 -0.39519500 S 0.35303500 2.01633800 -0.11797200 C -2.77982600 -0.71521900 0.46936800 H -2.17104500 -1.60505600 0.69407100 H -3.13444900 -0.33968100 1.44303000 C -3.96917800 -1.10812300 -0.39435700 H -3.63244500 -1.51827100 -1.35612900 H -4.60517500 -1.86291200 0.08664100 H -4.59634300 -0.23394000 -0.61769800

27 E(M06/BS1) = -753.943495 au H(M06/BS1) = -753.805913 au G(M06/BS1) = -753.849963 au E(M06/BS2//M06/BS1) = -2197.123543 au Cu -2.51354900 -0.28607400 -0.00025000 H -3.76785900 -1.16344500 -0.00067700 C 0.62855600 -0.29765100 0.00092700 H 0.52316300 -0.94432900 0.88503900 H 0.52252400 -0.94588900 -0.88196100 C 2.01083200 0.33365100 -0.00013900 S -0.69664200 0.98000800 0.00027000 H 2.10252900 0.99398700 0.87799000 H 2.10195200 0.99223700 -0.87963900 C 3.14531000 -0.68349600 0.00050400 H 3.04463700 -1.34182200 -0.87724600 H 3.04537100 -1.33989800 0.87977700

S42

364 C 4.52089100 -0.03241600 -0.00078500 H 4.65022800 0.60861800 0.88193000 H 5.33358100 -0.77057200 -0.00027600 H 4.64951100 0.60661200 -0.88505600

C3H7CH2SH E(M06/BS1) = -556.478612 au H(M06/BS1) = -556.339217 au G(M06/BS1) = -556.378368 au E(M06/BS2//M06/BS1) = -556.557843 au C -0.74408500 0.66486900 0.05323100 H -0.72887200 1.26388500 0.97273300 H -0.74292200 1.37198400 -0.78599600 C 0.46180400 -0.25282700 -0.02165600 H -2.16637100 -0.99493500 0.99908900 H 0.40997100 -0.84687100 -0.94645900 H 0.42307900 -0.97832300 0.80716900 S -2.35025800 -0.20320600 -0.07922400 C 1.77868300 0.51125300 0.03525000 H 1.81356100 1.23479500 -0.79348600 H 1.81485600 1.10827100 0.95939600 C 2.98502000 -0.41181000 -0.03241300 H 3.92817600 0.14417800 0.00522700 H 2.98378800 -1.12272800 0.80375400 H 2.98031900 -0.99788500 -0.96032100

S43

365 366 9.6 Appendix F - Supplementary material for Chapter 7

367 To be submitted to special issue of the Chinese Journal of Chemical Physics (CJCP), proceedings of ISSPIC-19. Guest Editors: Profs. Min Han & Lai-Sheng Wang Deadline for submission: 31 December 2018 Version: 21 December 2018

Structure of the Ligated Ag60 Nanoparticle

[{Cl@Ag12}@Ag48(dppm)12] (where dppm = bis(diphenylphosphino)methane).

Athanasios Zavras,(a) Antonija Mravak,(b) Margarita Bužančić,(b) Jonathan M. White,(a)* Vlasta Bonačić-Koutecký,(b,c)* Richard A. J. O’Hair (a)*

(a) School of Chemistry and Bio21 Molecular Science and Biotechnology Institute University of Melbourne 30 Flemington Rd, Parkville, Victoria 3010, Australia. (b) Center of Excellence for Science and Technology – Integration of Mediterranean Region (STIM) at Interdisciplinary Center for Advanced Sciences and Technology (ICAST), University of Split, Poljička cesta 35, 21000 Split, Croatia. (c) Chemistry Department, Humboldt University of Berlin, Brook – Taylor – Strasse 2, 12489 Berlin, Germany.

Supporting Information:

Inner shell Ag-Cl distances Å

Cl-Ag5 3.003(6) Cl-Ag6 2.950(6) Cl-Ag7 2.988(6) Cl-Ag8 2.936(6) Cl-Ag9 2.931(6) Cl-Ag10 2.899(6) Cl-Ag30 2.872(6) Cl-Ag31 2.931(6)

368 Cl-Ag32 2.896(6) Cl-Ag33 2.907(6) Cl-Ag37 2.875(6) Cl-Ag52 2.818(6)

Inner shell Ag-Ag distances Å

Ag5-Ag6 3.091(2) Ag5-Ag7 3.099(2) Ag5-Ag8 3.018(2) Ag5-Ag9 3.134(3) Ag5-Ag10 2.855(3) Ag6-Ag7 3.092(3) Ag6-Ag10 3.017(3) Ag6-Ag33 2.868(3) Ag6-Ag37 3.084(3) Ag7-Ag8 2.860(3) Ag7-Ag33 3.014(3) Ag7-Ag32 3.089(3) Ag8-Ag9 3.073(3) Ag8-Ag32 3.234(3) Ag8-Ag52 2.811(3) Ag9-Ag10 3.154(3) Ag9-Ag52 2.811(3) Ag9-Ag30 3.252(3) Ag10-Ag30 3.284(3) Ag10-Ag37 2.982(3) Ag31-Ag30 3.036(3) Ag31-Ag32 2.841(3) Ag31-Ag33 3.296(3) Ag31-Ag37 3.361(3) Ag31-Ag52 3.086(3) Ag30-Ag37 2.833(3) Ag30-Ag52 3.081(3) Ag37-Ag33 3.271(3) Ag33-Ag32 2.970(3) Ag32-Ag52 3.194(3)

Outer shell Ag-Ag distances Å

Ag57-Ag39 3.247(3) Ag57-Ag38 2.900(3) Ag57-Ag58 3.141(3) Ag57-Ag59 3.687(3) Ag57-Ag51 3.023(3) Ag57-Ag56 2.947(3) Ag38-Ag36 2.753(3) Ag38-Ag29 2.883(3) Ag38-Ag39 2.958(3) Ag38-Ag58 2.938(3)

369 Ag39-Ag29 3.625(3) Ag39-Ag40 3.220(3) Ag39-Ag41 2.955(3) Ag39-Ag56 2.932(2) Ag56-Ag51 2.775(3) Ag56-Ag54 3.792(3) Ag56-Ag41 3.014(3) Ag56-Ag55 3.272(3) Ag51-Ag50 2.853(3) Ag51-Ag55 2.975(3) Ag51 Ag47 2.753(3) Ag50-Ag47 2.932(3) Ag50-Ag48 2.909(3) Ag50 Ag49 2.790(3) Ag50-Ag58 3.263(3) Ag58-Ag49 3.085(3) Ag58-Ag59 2.754(3) Ag47-Ag46 3.240(3) Ag47-Ag48 3.202(3) Ag47-Ag60 2.929(3) Ag48-Ag18 2.993(3) Ag48-Ag19 3.187(3) Ag48-Ag49 3.014(3) Ag48-Ag46 3.626(3) Ag49-Ag59 2.893(3) Ag49-Ag19 3.082(3) Ag49-Ag34 3.413(3) Ag59-Ag34 2.972(3) Ag59-Ag35 2.832(3) Ag59-Ag36 2.946(3) Ag36-Ag35 3.256(3) Ag36-Ag28 3.225(3) Ag36-Ag29 2.897(3) Ag29-Ag28 2.977(3) Ag29-Ag27 2.802(3) Ag29-Ag40 3.129(3) Ag40-Ag27 3.025(3) Ag40-Ag26 2.754(3) Ag40-Ag41 2.991(3) Ag41-Ag42 2.723(3) Ag41-Ag54 3.014(3) Ag54-Ag42 3.036(3) Ag54-Ag43 3.123(3) Ag54-Ag53 2.930(3) Ag54-Ag55 3.164(3) Ag55-Ag53 2.991(3) Ag55-Ag60 2.742(3) Ag60-Ag53 2.905(3) Ag60-Ag46 2.834(3) Ag53-Ag43 2.922(3)

370 Ag53-Ag44 3.115(3) Ag53-Ag45 3.490(3) Ag45-Ag44 3.132(3) Ag45-Ag16 2.889(3) Ag45-Ag17 3.397(3) Ag45-Ag46 2.946(3) Ag46-Ag17 2.925(3) Ag46-Ag18 2.830(3) Ag18-Ag17 2.975(3) Ag18-Ag14 2.923(3) Ag18-Ag19 2.770(3) Ag19-Ag20 2.911(3) Ag19-Ag34 3.231(3) Ag34-Ag20 2.903(3) Ag34-Ag21 3.343(3) Ag34-Ag35 2.940(3) Ag35-Ag21 2.923(3) Ag35-Ag22 2.845(3) Ag35-Ag28 3.602(3) Ag28-Ag22 2.966(3) Ag28-Ag23 3.170(3) Ag28-Ag27 2.971(3) Ag27-Ag23 3.081(3) Ag27-Ag24 3.419(3) Ag27-Ag26 2.907(3) Ag26-Ag24 2.988(3) Ag26-Ag25 2.827(3) Ag26-Ag42 2.988(3) Ag42-Ag25 3.378(3) Ag42-Ag43 3.107(3) Ag43-Ag25 3.587(3) Ag43-Ag4 3.052(3) Ag43-Ag44 3.246(3) Ag43-Ag53 2.922(3) Ag21-Ag20 2.990(3) Ag21-Ag13 3.189(3) Ag21-Ag12 3.189(3) Ag21-Ag22 2.971(3) Ag22-Ag12 2.922(3) Ag22 Ag23 2.768(3) Ag23-Ag11 2.915(3) Ag23-Ag24 3.189(3) Ag24-Ag11 2.899(3) Ag24-Ag3 3.350(3) Ag24-Ag25 2.928(3) Ag25-Ag3 2.925(3) Ag25-Ag4 2.830(3) Ag4-Ag2 2.923(3) Ag4-Ag44 2.756(3) Ag44-Ag16 2.918(3)

371 Ag16-Ag2 2.778(3) Ag16-Ag15 2.835(3) Ag16-Ag17 2.987(3) Ag17-Ag15 3.435(3) Ag17-Ag14 3.209(3) Ag14-Ag15 3.194(3) Ag14-Ag13 2.929(3) Ag14-Ag20 2.775(3) Ag20-Ag13 2.841(3) Ag20-Ag21 2.990(3) Ag12-Ag13 3.167(3) Ag12-Ag1 2.940(3) Ag12-Ag11 2.774(3) Ag11-Ag1 2.841(3) Ag11-Ag3 3.003(3) Ag3-Ag1 3.442(3) Ag3-Ag2 3.204(3) Ag2-Ag1 3.205(3) Ag2-Ag15 2.953(3) Ag15-Ag1 2.861(3) Ag15-Ag13 2.873(3) Ag13-Ag1 2.862(3)

Inter-shell Ag-Ag distances Å

Ag1 Ag5 2.942(3) Ag2 Ag5 3.093(3) Ag3 Ag4 2.970(3) Ag3 Ag5 3.223(2)

Ag4 Ag5 3.250(3 Ag6 Ag13 2.963(3) Ag6 Ag12 3.090(3) Ag6 Ag21 3.236(3) Ag6 Ag22 3.244(2) Ag7 Ag15 2.953(3) Ag7 Ag14 3.078(3) Ag7 Ag17 3.243(2) Ag7 Ag18 3.252(3) Ag8 Ag53 2.955(3) Ag8 Ag45 2.987(2) Ag8 Ag16 3.121(3) Ag8 Ag44 3.320(3) Ag9 Ag25 2.812(3) Ag9 Ag42 3.078(3)

372 Ag9 Ag43 3.102(3) Ag10 Ag27 2.939(3) Ag10 Ag24 3.019(3) Ag10 Ag11 3.133(3) Ag12 Ag21 3.189(3) Ag19 Ag33 3.357(3) Ag20 Ag33 3.121(3) Ag28 Ag37 3.139(3) Ag29 Ag30 3.052(3) Ag29 Ag37 3.134(3) Ag30 Ag41 2.981(3) Ag30 Ag40 3.026(3) Ag30 Ag39 3.054(3) Ag30 Ag38 3.325(3) Ag31 Ag38 2.996(3) Ag31 Ag57 3.032(3) Ag31 Ag50 3.043(3) Ag31 Ag58 3.052(3) Ag31 Ag51 3.292(3) Ag32 Ag46 2.832(3) Ag32 Ag47 3.102(3) Ag32 Ag48 3.153(3) Ag32 Ag50 3.187(3) Ag33 Ag49 2.923(3) Ag33 Ag34 3.030(3) Ag34 Ag35 2.940(3) Ag35 Ag37 2.829(3) Ag36 Ag37 3.110(3) Ag39 Ag40 3.220(3) Ag45 Ag60 2.941(3) Ag49 Ag50 2.790(3) Ag51 Ag52 3.132(3) Ag52 Ag56 2.874(3) Ag52 Ag55 2.967(3) Ag52 Ag53 3.276(3) Ag53 Ag54 2.930(3)

Cartesian coordinates of DFT optimized structure: Ag -1.676030 0.469470 -2.443870 Ag -0.810900 2.661350 -0.827580 Ag 1.818680 2.195680 0.499070 Ag 2.753100 -0.682260 -0.746270 Ag 0.963110 -2.675100 0.537790

373 Ag -0.330390 1.704070 2.274910 Ag 1.327130 1.195600 -2.377640 Ag 0.461030 -1.772110 -2.372760 Ag -1.911770 -2.051390 -0.801220 Ag -1.380260 -1.186810 2.117210 Cl 0.033170 0.006840 -0.010260 Ag 1.641850 -0.600830 2.355690 Ag -2.805080 0.596390 0.477380 C -1.675275 -8.232249 1.630967 C -2.838830 -7.567885 2.067960 C -4.101513 -8.112722 1.788164 C -4.207455 -9.311527 1.061870 C -3.051338 -9.972205 0.619060 C -1.784446 -9.436563 0.908717 Ag -5.657765 0.187411 0.591475 C -8.319422 0.680672 7.787107 C -7.746533 -0.072192 6.746763 Ag -3.481977 0.404778 -4.546540 P -3.828331 1.199650 -6.835151 C -2.939339 2.866523 -7.080294 P -3.142161 3.881512 -5.451389 C -2.218043 5.426092 -5.944028 C -0.872914 5.263282 -6.353309 C -0.085017 6.389403 -6.626132 C -0.625511 7.681203 -6.486186 C -1.954419 7.838139 -6.054790 C -2.754632 6.714558 -5.778038 Ag -3.833275 2.924687 -0.805409 Ag -4.407925 0.422986 -1.941318 Ag -5.616626 -1.979581 -1.375582 P -7.250182 -3.546412 -2.429598 C -6.625857 -5.273374 -2.942480 P -5.211355 -5.860362 -1.802393 C -4.637337 -7.440902 -2.622366

374 C -3.308814 -7.833287 -2.371740 C -2.800203 -8.993087 -2.975492 C -3.608279 -9.754144 -3.836667 C -4.936199 -9.365450 -4.086194 C -5.452762 -8.209591 -3.475562 Ag -1.766774 5.109390 -1.265170 C -6.421045 0.177089 6.343102 P -5.453342 -0.898311 5.121365 C -6.723300 -2.191121 4.619989 Ag -3.951411 -0.140411 3.081846 Ag 2.118964 -3.387270 3.067248 Ag 3.925735 -2.546340 1.100086 Ag 4.539494 -3.870773 -1.240266 Ag 1.915874 -4.092273 -1.987724 Ag -0.526475 -4.722368 -0.761531 Ag -2.425761 -4.247232 1.198860 Ag -3.132755 -2.869444 3.395281 P -3.759060 -3.411080 5.681370 C -4.483198 -1.861149 6.497958 P -2.430964 7.427779 -1.714743 C -2.531295 8.554910 -0.218705 C -1.889876 8.127895 0.956432 C -1.897455 8.953353 2.091896 C -2.546353 10.199095 2.052359 C -3.191295 10.623019 0.875364 C -3.182473 9.800007 -0.263618 C -4.014392 7.706149 -2.676318 C -4.983979 6.688102 -2.610444 C -6.154673 6.786192 -3.378105 C -6.357124 7.892314 -4.220211 C -5.396149 8.917687 -4.278667 C -4.226218 8.826254 -3.505006 C -1.159847 8.386606 -2.757023 P 0.596832 8.114945 -2.075309

375 C 0.804607 9.350099 -0.680207 C 1.743100 9.019675 0.317904 C 2.042068 9.939036 1.336765 C 1.364360 11.168922 1.391501 C 0.388146 11.477834 0.428500 C 0.119315 10.578166 -0.615340 Ag 1.317876 5.786533 -1.503713 Ag 0.313250 4.679919 0.856877 Ag -1.452311 -0.402084 4.726517 Ag -2.544266 2.109251 3.961393 Ag -4.490783 2.379075 1.978602 P -6.475173 3.854839 1.521460 C -7.670910 3.194940 0.181539 P -7.912478 1.322572 0.404691 C -8.864606 0.917219 -1.164606 C -10.267022 0.804459 -1.186110 C -10.915847 0.393188 -2.363854 C -10.168283 0.095351 -3.515700 C -8.767000 0.216985 -3.494440 C -8.113519 0.627972 -2.323655 C 1.555964 8.915101 -3.483614 C 1.866919 10.285781 -3.519941 C 2.558130 10.813911 -4.624455 C 2.932948 9.974017 -5.689220 C 2.637049 8.600782 -5.640003 C 1.956631 8.067535 -4.532648 Ag 2.783748 3.551232 -2.118834 Ag 3.933654 0.794233 -3.067792 Ag 5.406964 -0.890721 -1.000886 P 7.779047 -1.533299 -1.230766 C 8.914597 -0.283032 -2.037750 C 9.975925 -0.642001 -2.891626 C 10.754599 0.356748 -3.501351 C 10.479339 1.711640 -3.248760

376 C 9.431334 2.066726 -2.383206 C 8.642464 1.074652 -1.782903 Ag 0.129777 3.942685 -3.199417 Ag -0.416969 -3.044949 4.099073 Ag 0.085177 -4.939907 2.075106 Ag 2.580391 -4.924040 0.571479 Ag -3.481438 -4.158195 -1.390412 Ag -4.099898 -2.046128 0.898188 P 2.719867 -7.418581 0.486687 C 1.082884 -8.348118 0.760036 P -0.018781 -7.444680 2.039347 Ag 1.327125 -1.007101 5.060611 P 2.551197 -1.451602 7.240248 C 1.851775 -0.822992 8.866203 C 2.625780 -0.681750 10.035069 C 2.034465 -0.219705 11.222719 C 0.669525 0.117587 11.241239 C -0.100569 -0.005173 10.072616 C 0.488789 -0.471945 8.885764 Ag 3.083699 4.638624 0.415807 P 5.089432 6.154592 0.283843 C 6.753698 5.310574 0.681062 P 6.545461 3.974450 2.028067 C 6.937467 4.954536 3.589734 C 8.254211 5.247606 3.989759 C 8.480297 5.925806 5.199877 C 7.393954 6.329608 5.997813 C 6.079205 6.045641 5.589199 C 5.851455 5.340447 4.395289 Ag -2.351573 3.968617 1.304327 Ag -0.702777 4.193268 3.458764 Ag 0.360132 1.757706 4.873796 P 0.440229 3.091327 6.973944 C 0.225384 4.992642 6.820200

377 P -1.087087 5.454406 5.518562 C -0.846992 7.315096 5.485946 C -1.675558 8.208704 6.187476 C -1.446583 9.592476 6.089525 C -0.391189 10.081805 5.299522 C 0.431385 9.185466 4.592737 C 0.199088 7.803249 4.676422 Ag -2.494786 2.976915 -3.213520 Ag -0.489166 1.508971 -4.745774 Ag -1.039008 -1.270154 -4.744178 Ag -3.470065 -1.991390 -3.260870 Ag -1.236274 -3.743209 -3.252807 Ag 1.409348 -3.043295 -4.515929 P 1.339153 -4.368920 -6.521621 C -0.481581 -4.715613 -6.966332 P -1.441040 -5.109612 -5.333545 C -3.189974 -5.061918 -5.982956 C -4.106591 -6.107269 -5.774583 C -5.443661 -5.949783 -6.184754 C -5.865194 -4.756947 -6.795882 C -4.946613 -3.708279 -6.985749 C -3.616433 -3.849393 -6.572381 C -9.233999 1.172537 1.729383 C -9.263331 -0.051496 2.420588 C -10.201889 -0.274833 3.440517 C -11.114560 0.738995 3.778580 C -11.086055 1.966172 3.092593 C -10.153781 2.183341 2.064601 Ag 2.059539 2.594947 -4.621904 Ag 1.722858 -0.297632 -4.733613 Ag 3.571788 -2.000929 -3.111084 Ag 4.437757 1.886402 -0.580142 Ag 4.497482 0.131103 1.550633 Ag 3.950233 -1.453476 3.716324

378 P 5.299773 -1.244948 5.746791 C 4.311979 -0.697404 7.280418 C -4.982870 4.229548 -5.499968 C -5.767116 3.577553 -4.527200 C -7.167570 3.670985 -4.573969 C -7.781973 4.431880 -5.583935 C -6.997781 5.101115 -6.540616 C -5.597994 4.993325 -6.506165 Ag 4.381482 2.789019 2.140103 Ag 3.032448 1.228682 3.993781 Ag 1.929394 3.603908 2.986600 C 8.102824 2.967242 1.766425 C 8.100952 1.642970 2.247054 C 9.247013 0.846655 2.096033 C 10.378860 1.357077 1.437940 C 10.377830 2.672013 0.945168 C 9.243528 3.483577 1.118979 C -2.760536 5.227264 6.322410 C -3.835756 4.945931 5.452795 C -5.142707 4.849638 5.956567 C -5.376508 5.040861 7.329272 C -4.301867 5.284920 8.204405 C -2.991796 5.378919 7.702109 P 3.487679 2.913413 -6.547277 C 4.725604 1.483612 -6.774877 P 5.424471 0.970932 -5.050678 C 6.752159 2.282794 -4.852907 C 6.511405 3.266111 -3.873236 C 7.343571 4.393912 -3.788187 C 8.438723 4.520842 -4.659135 C 8.701803 3.521249 -5.612902 C 7.850338 2.409409 -5.723052 C -6.332838 5.659250 1.049096 C -7.252082 6.295936 0.193715

379 C -7.135601 7.673925 -0.059161 C -6.119007 8.414699 0.560303 C -5.196033 7.779149 1.409830 C -5.288889 6.399121 1.639670 C -7.589423 3.949873 3.035425 C -7.530500 2.880045 3.949110 C -8.346395 2.892093 5.090342 C -9.204911 3.978094 5.327829 C -9.250587 5.055232 4.425383 C -8.442775 5.042348 3.275259 C -0.945535 -6.904767 -5.110952 C -1.325468 -7.923367 -6.003035 C -0.799635 -9.216637 -5.848803 C 0.129314 -9.482769 -4.827098 C 0.511177 -8.464690 -3.936399 C -0.039118 -7.179401 -4.068341 C 6.234445 -0.646493 -5.529883 C 7.592381 -0.925546 -5.301708 C 8.096886 -2.203968 -5.607409 C 7.252529 -3.200274 -6.127736 C 5.886535 -2.925603 -6.329329 C 5.376892 -1.654937 -6.029031 C 6.458593 -5.674638 -4.174000 C 7.256316 -5.768277 -3.018101 C 8.222827 -6.784097 -2.905625 C 8.409763 -7.689006 -3.965731 C 7.637322 -7.575199 -5.135881 C 6.658862 -6.571263 -5.237060 P 6.929143 -4.495946 -1.669554 C 8.097853 -3.108343 -2.252706 C 7.777161 -5.252194 -0.178367 C 6.948224 -5.843439 0.794925 C 7.518826 -6.527110 1.881974 C 8.915225 -6.576306 2.026461

380 C 9.742958 -5.945321 1.081795 C 9.175899 -5.296950 -0.026595 C 4.534080 4.459285 -6.509451 C 5.578116 4.718503 -7.418623 C 6.352557 5.882350 -7.277575 C 6.096206 6.773368 -6.220793 C 5.066751 6.506528 -5.303906 C 4.279173 5.355622 -5.453689 C 2.520367 2.953674 -8.151542 C 2.683965 3.953309 -9.126047 C 1.845546 3.969152 -10.255433 C 0.850479 2.989989 -10.413431 C 0.679654 1.994332 -9.433183 C 1.504607 1.982352 -8.297435 C 2.141475 -6.049068 -6.383560 C 1.870052 -7.119585 -7.256554 C 2.483425 -8.365741 -7.044768 C 3.347356 -8.546106 -5.951584 C 3.600810 -7.484870 -5.068124 C 3.008799 -6.233202 -5.289491 C 2.103430 -3.549852 -8.028511 C 2.741909 -4.264094 -9.058178 C 3.338500 -3.566715 -10.123620 C 3.294195 -2.162262 -10.165615 C 2.667622 -1.446263 -9.129788 C 2.084723 -2.138618 -8.056006 C 4.915973 7.386279 1.707867 C 5.783613 8.483544 1.849786 C 5.648183 9.345114 2.950933 C 4.649403 9.110403 3.912919 C 3.771126 8.024250 3.759551 C 3.896317 7.167868 2.652794 C 5.625102 7.282383 -1.125963 C 4.629917 8.021492 -1.799158

381 C 4.981854 8.907877 -2.829074 C 6.327329 9.040875 -3.217417 C 7.318101 8.278085 -2.579493 C 6.970521 7.408600 -1.530128 C 6.123853 -2.871461 6.183721 C 6.543880 -3.222240 7.478492 C 7.216965 -4.437189 7.690135 C 7.518979 -5.276661 6.602655 C 7.106740 -4.923514 5.305976 C 6.373186 -3.741976 5.103707 C 6.727613 -0.028846 5.785395 C 6.424706 1.338717 5.635844 C 7.461947 2.282629 5.628149 C 8.800379 1.863696 5.717423 C 9.101094 0.497659 5.838510 C 8.065023 -0.451951 5.887298 C -6.196096 -6.496462 -0.333907 C -6.290184 -5.685005 0.808458 C -7.056754 -6.120869 1.901773 C -7.737445 -7.348359 1.850257 C -7.639208 -8.160677 0.705965 C -6.861128 -7.736399 -0.383954 C 2.979008 -3.249612 7.623447 C 3.143693 -4.115048 6.526951 C 3.503778 -5.453529 6.737274 C 3.683482 -5.942361 8.042378 C 3.487807 -5.086529 9.140909 C 3.140581 -3.740150 8.932254 C -3.093342 0.035300 -8.110779 C -1.734011 -0.326810 -7.991146 C -1.194087 -1.312647 -8.834724 C -2.007257 -1.949985 -9.789420 C -3.361284 -1.592942 -9.904487 C -3.905001 -0.601415 -9.069165

382 C -5.562783 1.504734 -7.472928 C -6.626571 1.053922 -6.673266 C -7.952610 1.216706 -7.108892 C -8.214131 1.832606 -8.342148 C -7.151129 2.297710 -9.138472 C -5.826104 2.134943 -8.705451 C 8.693804 -1.963757 0.355739 C 10.099486 -1.995815 0.389801 C 10.763123 -2.343012 1.577709 C 10.019524 -2.645950 2.732518 C 8.616350 -2.599589 2.697279 C 7.947767 -2.263060 1.508008 C 0.448906 -8.336582 3.633055 C 1.250581 -7.654433 4.565446 C 1.602260 -8.285673 5.772084 C 1.148094 -9.584510 6.050033 C 0.340573 -10.260837 5.119564 C -0.010504 -9.637171 3.911962 C 3.495765 -8.457734 -0.878282 C 4.716578 -8.013582 -1.429863 C 5.360205 -8.767579 -2.423558 C 4.775806 -9.958127 -2.894415 C 3.541928 -10.384872 -2.378012 C 2.906703 -9.642138 -1.366450 C 3.737549 -7.956493 1.982398 C 4.090217 -9.302073 2.186524 C 4.854854 -9.665168 3.306650 C 5.280278 -8.681289 4.216362 C 4.941314 -7.334432 4.003023 C 4.162921 -6.969800 2.889923 C -4.947819 -4.811040 6.050828 C -5.390113 -5.103849 7.353399 C -6.228516 -6.209088 7.569796 C -6.601333 -7.031753 6.490558

383 C -6.140958 -6.747062 5.194417 C -5.319655 -5.629446 4.971349 C -2.264388 -3.939457 6.671644 C -1.672600 -3.185398 7.696205 C -0.480400 -3.640600 8.289746 C 0.110497 -4.840580 7.862330 C -0.484616 -5.593338 6.835946 C -1.664856 -5.140724 6.231381 C -8.697878 -3.811356 -1.269827 C -9.159620 -5.063879 -0.832148 C -10.198593 -5.136033 0.113717 C -10.785930 -3.961700 0.612962 C -10.326783 -2.706284 0.171256 C -9.281958 -2.628975 -0.760942 C -8.091157 -3.045506 -4.031786 C -9.467467 -3.234305 -4.244136 C -10.045972 -2.838580 -5.461688 C -9.247984 -2.257550 -6.464461 C -7.871831 -2.069740 -6.245370 C -7.288989 -2.456275 -5.026876 C 2.149004 3.045527 7.744454 C 2.438192 2.469555 8.992802 C 3.774724 2.392360 9.428929 C 4.817129 2.885538 8.628339 C 4.524905 3.452319 7.373664 C 3.199102 3.518186 6.922457 C -0.810499 2.756201 8.332567 C -1.913211 1.943428 8.010587 C -2.914767 1.709914 8.969167 C -2.823227 2.300118 10.242753 C -1.723339 3.117343 10.561902 C -0.718221 3.347729 9.607046 H 1.971997 4.748346 -11.011853 H 0.201861 3.006088 -11.293047

384 H 3.454778 4.718442 -8.995341 H -0.100431 1.235650 -9.540946 H 1.336738 1.251384 -7.493871 H 3.496048 5.121235 -4.719965 H 5.800089 4.006532 -8.219196 H 7.170831 6.078810 -7.975980 H 6.716439 7.663896 -6.089423 H 4.894786 7.178209 -4.462017 H 7.649797 -4.192876 -6.350139 H 9.156863 -2.417025 -5.440167 H 5.210086 -3.692176 -6.721817 H 4.305584 -1.462573 -6.159615 H 8.251345 -0.166913 -4.871586 H 9.074124 5.409380 -4.608769 H 9.553910 3.621721 -6.291082 H 7.109567 5.187166 -3.071786 H 8.032628 1.649036 -6.488664 H 5.649785 3.164031 -3.199254 H -9.245452 1.963590 -8.679624 H -7.355579 2.790633 -10.092741 H -8.773812 0.865975 -6.478937 H -4.998005 2.490645 -9.325796 H -6.405852 0.586846 -5.706331 H -8.871982 4.492495 -5.637660 H -7.476526 5.688050 -7.328713 H -4.984625 5.496575 -7.259325 H -7.771564 3.128915 -3.839865 H -5.279576 2.983534 -3.742487 H -1.583441 -2.721668 -10.437591 H -0.139462 -1.583991 -8.744061 H -3.997710 -2.082112 -10.646225 H -4.956316 -0.315555 -9.163977 H -1.103453 0.139110 -7.220743 H -0.000756 8.556855 -6.675294

385 H -2.377970 8.838346 -5.929003 H 0.952250 6.249238 -6.946067 H -3.780255 6.847451 -5.423148 H -0.425896 4.266357 -6.439040 H -2.553111 10.829060 2.945712 H -3.708271 11.586805 0.848268 H -3.716602 10.105604 -1.168696 H -1.394681 8.625410 3.001150 H -1.411666 7.139789 0.995215 H -7.253315 7.941836 -4.843667 H -5.549342 9.778396 -4.936000 H -3.466979 9.609751 -3.579597 H -6.884101 5.975036 -3.353462 H -4.805318 5.810716 -1.977792 H 3.475199 10.385054 -6.544151 H 2.812454 11.876903 -4.645944 H 1.593089 10.926045 -2.677296 H 1.739707 6.994720 -4.470374 H 2.958037 7.935967 -6.444818 H 1.584969 11.879876 2.192786 H 2.798582 9.683609 2.082312 H -0.165872 12.418106 0.488030 H 2.230688 8.038051 0.309453 H -0.645182 10.827354 -1.355191 H 6.595066 9.725892 -4.026175 H 4.204669 9.488240 -3.327722 H 3.578204 7.900131 -1.525814 H 8.365926 8.369403 -2.878491 H 7.757363 6.846688 -1.022429 H 7.572217 6.856148 6.938668 H 9.503251 6.134631 5.525394 H 5.230371 6.358784 6.202724 H 4.835374 5.059915 4.092052 H 9.096165 4.903186 3.382264

386 H 4.553098 9.774662 4.775754 H 6.324352 10.196435 3.061259 H 2.991286 7.839094 4.503595 H 6.551932 8.670534 1.093567 H 3.210673 6.316271 2.529385 H 7.201880 1.238090 2.726757 H 11.251444 3.062663 0.416255 H 11.251351 0.715847 1.295474 H 9.249963 -0.173066 2.481391 H 9.250154 4.511020 0.744030 H 11.567269 0.076932 -4.177730 H 11.062840 2.491093 -3.743747 H 9.199466 3.117432 -2.205722 H 10.182072 -1.694503 -3.104736 H 7.805736 1.350512 -1.130331 H 10.530260 -2.904183 3.663497 H 11.856003 -2.366750 1.606522 H 8.042688 -2.809178 3.600668 H 10.672585 -1.718798 -0.500198 H 6.851409 -2.203779 1.484806 H 7.781791 -8.281383 -5.957380 H 9.152826 -8.485508 -3.872444 H 8.806252 -6.877450 -1.985826 H 5.677659 -4.908281 -4.230567 H 6.026191 -6.502344 -6.125047 H 9.358932 -7.098173 2.878423 H 6.861844 -7.017188 2.604092 H 10.828766 -5.958400 1.204175 H 9.827522 -4.811419 -0.756710 H 5.857971 -5.752718 0.713167 H -6.907270 -4.629312 -7.097880 H -6.151198 -6.768440 -6.026079 H -5.258469 -2.769941 -7.451277 H -3.792872 -7.031774 -5.283129

387 H -2.929106 -3.004044 -6.690685 H 0.574514 -10.477707 -4.737742 H -1.094923 -10.008620 -6.542585 H 1.263635 -8.655725 -3.166087 H -2.026203 -7.705592 -6.814853 H 0.256146 -6.378927 -3.376702 H 3.754756 -1.625460 -10.998943 H 3.840066 -4.121602 -10.920973 H 2.628664 -0.352863 -9.153649 H 2.792414 -5.356044 -9.014597 H 1.643793 -1.589092 -7.209704 H 3.797484 -9.523905 -5.761357 H 4.228716 -7.642392 -4.190867 H 2.259308 -9.201815 -7.712603 H 3.169308 -5.410901 -4.579684 H 1.157691 -6.997173 -8.077490 H 8.071320 -6.205236 6.767553 H 7.521620 -4.719845 8.701227 H 7.360062 -5.548809 4.445834 H 5.976699 -3.495372 4.109157 H 6.343355 -2.557313 8.322381 H 8.293498 -1.514023 6.007479 H 10.141093 0.166553 5.898999 H 9.600429 2.606084 5.673734 H 7.233222 3.343599 5.524520 H 5.384854 1.664208 5.498062 H 3.962455 -6.987577 8.202951 H 3.601417 -5.466494 10.160013 H 2.960944 -3.085852 9.788946 H 3.630150 -6.113602 5.877193 H 2.968776 -3.749551 5.504727 H 3.695283 -0.905601 10.017157 H 2.640897 -0.110833 12.126420 H 0.209816 0.496720 12.157003

388 H -1.148611 0.299386 10.071479 H -0.097278 -0.548632 7.962074 H 5.855220 2.817257 8.962676 H 3.990536 1.937406 10.398792 H 5.328394 3.838586 6.743671 H 1.639537 2.056126 9.613454 H 2.986225 3.893200 5.911734 H -0.219529 11.158280 5.218546 H -2.094297 10.287367 6.631220 H -2.500109 7.821819 6.793123 H 1.216327 9.557968 3.931816 H 0.784872 7.100253 4.068846 H -6.398102 4.976838 7.712386 H -4.477892 5.398284 9.277952 H -2.160489 5.551203 8.389843 H -5.970307 4.619762 5.281405 H -3.644056 4.789603 4.383924 H -3.606637 2.121088 10.983955 H -1.648402 3.573829 11.552994 H 0.141119 3.978473 9.854355 H -3.771038 1.078865 8.718019 H -1.982177 1.499427 7.006937 H -6.021608 9.485266 0.367997 H -4.396481 8.356449 1.875764 H -7.833809 8.161813 -0.744077 H -4.543073 5.887260 2.260138 H -8.056781 5.726893 -0.278270 H -9.832892 3.984532 6.222838 H -8.458223 5.881826 2.574512 H -8.309005 2.056739 5.787072 H -6.843362 2.044176 3.766638 H -11.839204 0.580704 4.581566 H -11.779733 2.764619 3.367969 H -10.201504 -1.235306 3.964020

389 H -8.539353 -0.828892 2.173706 H -10.127549 3.151464 1.561203 H -10.662522 -0.266688 -4.419724 H -12.004015 0.289997 -2.373682 H -8.178556 -0.063623 -4.369120 H -10.842070 1.016810 -0.280079 H -7.014602 0.666638 -2.293271 H -7.240705 -7.900919 6.665084 H -6.574663 -6.442352 8.580401 H -6.412036 -7.392705 4.355507 H -5.063224 -4.484446 8.194592 H -4.965033 -5.379228 3.963460 H -0.013166 -3.051176 9.082771 H 1.044130 -5.177709 8.315929 H -2.122083 -2.245248 8.030189 H -0.028187 -6.524328 6.493416 H -2.120815 -5.714915 5.416846 H 5.886229 -8.960968 5.082056 H 5.125675 -10.711904 3.464723 H 3.785056 -10.060093 1.458931 H 5.297812 -6.561648 4.688629 H 3.891226 -5.916562 2.723252 H 1.410465 -10.062950 6.996876 H -0.025576 -11.266968 5.340131 H 2.206620 -7.748382 6.505719 H 1.571207 -6.626843 4.354181 H -0.666185 -10.144785 3.199165 H 5.277444 -10.540495 -3.671663 H 6.311381 -8.420005 -2.831536 H 3.077242 -11.304596 -2.744821 H 5.163978 -7.076185 -1.086918 H 1.960261 -10.000280 -0.954574 H -5.195920 -9.711188 0.828616 H -3.132196 -10.896968 0.041879

390 H -5.005090 -7.591166 2.108614 H -0.888289 -9.962341 0.567922 H -2.749110 -6.610698 2.599931 H -3.191559 -10.633216 -4.334194 H -5.565200 -9.952315 -4.761568 H -6.478029 -7.894406 -3.688837 H -1.757125 -9.270451 -2.815339 H -2.671752 -7.215663 -1.727105 H -8.156844 -9.123353 0.666842 H -8.335778 -7.673119 2.706193 H -6.741895 -8.382070 -1.258716 H -7.135667 -5.488891 2.785509 H -5.744086 -4.733434 0.854820 H -9.696157 -1.946001 -7.411826 H -11.117565 -2.981394 -5.624985 H -7.254328 -1.603161 -7.016067 H -10.082468 -3.668792 -3.451432 H -6.220862 -2.283112 -4.839698 H -11.587699 -4.020443 1.354254 H -10.534627 -6.115402 0.463346 H -10.758486 -1.784300 0.568305 H -8.913899 -1.651160 -1.086559 H -8.695510 -5.986910 -1.184165 H 7.511367 6.050994 0.992679 H 7.064222 4.783974 -0.239310 H 1.175035 5.395442 6.427021 H 0.021085 5.446060 7.805720 H -7.156223 3.319708 -0.787640 H -3.325326 3.432710 -7.945801 H -1.861504 2.664123 -7.207666 H -1.172195 7.958231 -3.774110 H -8.637633 3.726538 0.169725 H -1.392290 9.464840 -2.806349 H 4.849279 -0.953190 8.209251

391 H 4.199800 0.398972 7.212723 H 1.258248 -9.392352 1.076133 H 0.535300 -8.305802 -0.198560 H -5.122653 -2.071121 7.372299 H -3.646564 -1.197367 6.783442 H 9.153664 -3.426937 -2.199340 H 7.823555 -2.889078 -3.299585 H -7.460492 -5.995139 -2.987171 H -6.188399 -5.149762 -3.949545 H -0.615345 -5.539215 -7.689303 H -0.891508 -3.778251 -7.377811 H 5.551337 1.738731 -7.461575 H 4.156662 0.624404 -7.166980 C -6.967002 -2.354289 3.244150 C -7.988739 -3.207510 2.801322 C -8.763197 -3.917019 3.737615 C -8.475408 -3.814112 5.109773 C -7.456901 -2.952950 5.551780 H -6.359259 -1.802230 2.513964 H -8.180914 -3.318468 1.731472 H -9.561040 -4.575073 3.384141 H -9.032424 -4.408837 5.837970 H -7.235000 -2.882119 6.619220 H -9.910326 5.905911 4.616788 C -7.572667 1.681775 8.430455 C -6.270541 1.969462 7.987431 C -5.703577 1.237522 6.931788 H -8.012884 2.252002 9.252346 H -8.329669 -0.847033 6.245950 H -5.688283 2.770378 8.445350 H -4.709647 1.502786 6.550533 H -9.349477 0.479746 8.093326

392

Minerva Access is the Institutional Repository of The University of Melbourne

Author/s: Zavras, Athanasios

Title: Coinage metal hydrides: reactive intermediates in catalysis and significance to nanoparticle synthesis

Date: 2019

Persistent Link: http://hdl.handle.net/11343/224142

File Description: thesis

Terms and Conditions: Terms and Conditions: Copyright in works deposited in Minerva Access is retained by the copyright owner. The work may not be altered without permission from the copyright owner. Readers may only download, print and save electronic copies of whole works for their own personal non-commercial use. Any use that exceeds these limits requires permission from the copyright owner. Attribution is essential when quoting or paraphrasing from these works.