Organometallic Chemistry of Transition Metal

Home , Salt metathesis reaction

Organometallic Chemistry of Transition

Metal – Group 13 Complexes and

Metal-Organic Precursor Synthesis for

ALD Cobalt Oxide Thin Films

Cumulative Dissertation

Jiyeon Kim

Part I Organometallic Synthesis of Transition Metal – Group 13 Complexes

Part II Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

Dissertation

Faculty of Chemistry and Biochemistry Ruhr-University Bochum

Submitted by Jiyeon Kim, M.Sc.

February 2017

This work has been performed in the time between May 2013 and January 2017 under supervision of Prof. Roland A. Fischer and Prof. Anjana Devi at the Chair of Inorganic

Chemistry II – Organomatallics & Materials Chemistry of Ruhr-University Bochum, Germany.

1st referee: Prof. Dr. Roland A. Fischer

2nd referee: Prof. Dr. Anjana Devi

Herein I declare that I have written the corresponding cumulative thesis myself independently without any other help or source which are not specifically and explicitly marked or authorized in this dissertation. Furthermore, I declare that I have not applied to any other review procedures or submitted this cumulative thesis in the same, similar or different form to another faculty or university as a dissertation.

Jiyeon Kim, February 2017

Acknowledgement

I am deeply grateful to many people who supported me with every meaningful way to manage this work. Amongst, I sincerely appreciate my supervisor Prof. Roland A. Fischer for his support and generosity for giving the great opportunity to be at AC II. With no doubt, it was a dream to work in organometallics regardless of the frustrating outcome.

I wish to express my gratitude to Prof. Anjana Devi who is the supervisor of the ALD part of my thesis. I appreciate your efforts for giving me guidance from totally different field to another.

I was given the inspiration as well as the motivation of learning about materials science in very productive environment.

I wish to thank to Christian Gemel for his supervision also scientific support.

I am thankful to Prof. Markku Leskelä from University of Helsinki in Finland for providing me the opportunity to learn atomic layer deposition from one of the pioneering group in the ALD field.

I would like to thank Dr, Mariusz Molon, Dr. Arik Puls, Hung Banh who provided me a guidance to establish and settle down at OM group at the beginning.

I am thankful to Dr. Andreas Schneemann, Suttipong Wannapaiboon, Dr. Clarissa Kroll, Dr.

Sunja Kim for welcoming me and giving the introduction of Ruhrgebiet at the beginning.

I would like to thank Dr. Markus Halbherr, Dr. Arik Puls, Dr. Mariusz Molon, Dr. Kerstin

Freitag, Hung Banh, Jana Weßing and Manuela Winter for the introduction and guidance of single crystal X-ray diffraction and synthetic skills in the lab.

I am very thankful to Tomi Iivonen from University of Helsinki in Finland who offered me an excellent supervision and fun while learning the ALD process including the maintenance of F-

120.

I would like to thank the OM group and the CVD group for the scientific support and advise as well as to the whole ACII including the former members for providing me the great and comfortable working environment. Especially, thanks to all the office member at NC 2/27 who always brought interesting conversation topic to drag me out of work.

I would like to thank Prof. Radim Beranek for successfully instructing the 4G Photocat EU project as a leader and organizing the interesting meetings in different countries which allowed me to broaden my knowledge on different field. Also, his contribution of photoelectrochemical catalysis expertise on my work.

I would like to thank the New York and Boston crew, Dr. Christian Wiktor, Dr. Andreas

Schneemann, Dr. Wenhua Zhang, Jana Weßing, Inke Hante who joined the ACS fall meeting

on 2015, especially Wenhua Zhang and Inke Hante for having girls’ night out in Boston and

Jana Weßing and Andreas Schneemann for having together a wonderful time during the round trip in the US and Canada afterwards.

I thank Dr. Clarissa Kroll who joined me to the ICOMC 2014 in Sapporo, Japan and during the travel to Tokyo and Kyoto afterwards.

I would like to thank Dr. Mariusz Molon, Dr. Arik Benjamin Puls and Hung Banh for their guidance at the beginning to help me settle down in the group and learn quickly to handle the air sensitive reactions.

I would like to thank Inke Hante or Schwedler for her friendship and having an awesome chat in the chair as well as outside of the chair. Thanks to Jana Weßing for hanging out together at scientific and non-scientific meetings accompanied with beer. I am thankful to Sarah Karle having a short break together every day at 02 in front of NC. Thanks to Stefan Cwik for his guidance in CVD as well as common humor. Thanks to Alexander Sadlo for introducing

Bouldering which I still cannot manage to do well but it was fun. I would like to thank all the

CVD group for their help and welcoming me as a CVD group.

I would like to thank Dr. Andreas Schneemann and Richard O’Donoghue, Dr. Steven Hughes,

Kathryn Poole for generously proof reading my English during my study.

I would like to thank Sabine Pankau who is a wonderful colleague always being supportive with non-scientific issues even though it was not her task.

I would like to thank Niklas Stegmann for his assistance on the synthesis in the lab during the in-depth practical.

I express my special thanks to my friends Meike Stiepel and Christian Groß and Andreas

Schneemann for their friendship and spending plenty of time together that I could avoid feeling home sick. The chemistry what we did together was the most fun during my PhD while drinking and brewing Panda Gom Pale Ale.

Finally, I wish to thank from the bottom of my heart to Dr. Andreas Schneemann for his unconditional love and support and made it possible for in many ways.

“I am one of those who think, like Nobel, that humanity will draw more good than evil from new discoveries.”

Marie Curie

Contents

Part 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

1 Motivation ...... 1 List of Abbreviation ...... 2 2 Introduction ...... 3 2.1 Organometallic Chemistry ...... 3 2.2 Historic Development and Significance of Organometallic Chemistry...... 3 2.3 Discovery of the Low-Valent Group 13 Organyls and Chronological Development of Their Synthesis ...... 10 2.4 Coordination chemistry of ECp* toward transition metals ...... 12 2.5 Redox Chemistry of Transition Metal Group 13 Complexes...... 16 3. References ...... 19

I 3 Oxidation of Ni-ECp* Complexes: Stable Open-Shell Ni Cations [Ni(ECp*)n(PPh3)4−n]+ (n = 2, 4; E = Al, Ga)...... 24 3.1 Abstract ...... 24 3.2 Introduction ...... 24 3.3 Result and Discussion ...... 26 3.4 Conclusion ...... 29 3.6 Experimental Section ...... 33 3.6.1 General Comments...... 33 3.6.2 Synthetic Procedures ...... 34 3.6.3 Spectroscopic and crystallographic data ...... 36 4 References ...... 47 4. Organoaluminum Palladium Complex via Reductive Elimination ...... 49 4.1 Abstract ...... 49 4.2 Introduction ...... 49 4.3 Result and Discussion ...... 50 4.4 Conclusion ...... 53 4.5 Experimental Section...... 55 4.5.1 General Comments ...... 55 4.5.2 Synthetic procedure ...... 55 4.5.3 Spectroscopic, Spectrometric, Crystallographic Data...... 56 5. References ...... 62

List of Abbreviation ...... 64 5 General Introduction ...... 67 5.1 General Introduction ...... 67 Physical Vapor Deposition ...... 68 Chemical Vapor Deposition ...... 68 5.2 Atomic Layer Deposition ...... 69 5.3 ALD of metal oxides thin films ...... 72 5.3.1 Precursors ...... 72 5.3.2 Transition metal oxides and cobalt ...... 80 6. References ...... 82 6. Low temperature atomic layer deposition of cobalt oxide as an effective catalyst layer for photoelectrochemical water splitting devices ...... 88 6.1 Abstract ...... 88 6.2 Introduction ...... 89 6.3 Results and Discussion...... 91 6.3.3 Film Characterization...... 95 6.4 Conclusion and Outlook ...... 102 6.5 Appindex ...... 102 6.5 Experimental Section...... 107 6.5.1 Precursor synthesis...... 108 6.5.2 Film deposition ...... 109 6.5.3 Film characterization ...... 109 6.5.4 Photoelectrochemistry ...... 111 6.6 References ...... 112 7. Homoleptic all nitrogen coordinated bis-guanidinato cobalt (II) complex: Volatile and thermally stable precursors suitable for ALD/CVD ...... 136 7.1 Abstract ...... 136 7.2 Introduction ...... 136 7.3 Result and Discussion ...... 139 7.4 Conclusion ...... 148 7.5 Experimental Section...... 150 7.5.1 General Comments ...... 150 7.5.2 Synthetic Procedure ...... 150 7.5.3 Crystallographic data ...... 152 8. References ...... 153 Curriculum Vitae ...... 155

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

1 Motivation

This cumulative thesis consists of two parts whereas Part 1 deals with the synthesis and characterization of novel organometallic complexes. On the other hand, Part 2 describes the synthesis of new cobalt precursors for the use in thin film deposition techniques (Part 2 chapter XY) as well as the development of a ALD process for the preparation of efficient cobalt-oxide thin film photocatalysts.

The driving force for pursuing the first project based on pure organometallic chemistry, was the preparation of new organometallic complexes with unusual bonding situation and structural features. In

Part 1 chapter 4 the use of reductive elimination as a tool for the preparation of new building blocks and precursors for metal-rich clusters is explored. These clusters bear potential for a wide range of applications including catalysis and medical application. In chapter 3 of Part 1 the focus lays on the preparation of stable open shell NiI species, which are extremely rare and the exploration of these compounds and in particular the documentation of reliable synthesis conditions can be of extreme usefulness for future studies.

The second part of the thesis is much closer to application and was embedded in the Photocat-4G

Network. The overall goal of this project was the preparation of thin film devices useful for the photocatalytic degradation of pollutants in water. As part of my thesis, new precursors were developed for this project and in close collaboration with the University of Helsinki, we worked on new atomic layer deposition processes for the preparation of spinel-cobaltoxide thin films, which is are promising photocatalytic materials.

My role within this project was on the one hand the preparation of known (commercially unavailable) and novel precursor molecules for the thin film depositions, and additionally, I was during two academic stays at the university of Helsinki deeply involved in the film depositions as well as characterizations.

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

List of Abbreviation F BAr4 B[C6H3(CF3)2]4

DDP 2-Diisopropylphenylamino-4-diisopropylphenylimino-2-pentene

dvds 1,3-Divinyl-1,1,3,3-tetramethyldisiloxane

COD 1,5-Cyclooctadiene

COT Cyclooctatetraene

Cp Cyclopentadienylanion

CpH 1,3-Cyclopentadiene

Cp* Pentamethylcyclopentadienylanion

Cp*H 1,3-Pentamethylcyclopentadiene

E Group 13 metals

EA Elemental Analysis

Et Ethyl

EIR Monovalent group 13 organyls

Et2O Diethylether

iPr iso-propyl

LIFDI Liquid Injection Field Desorption Ionization

Me Methyl

NHC N-heteocyclic-carbene

Ph Phenyl

Rt Room temperature

TM Transition metals

tBu tert-butyl

tmeda N,N,N’,N’-Tetramethylethylen-1,2-diamine

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

2 Introduction

2.1 Organometallic Chemistry Organometallic Chemistry describes the chemistry of compounds, which contain at least one direct carbon to metal bond. This includes all transition metals, earth alkaline and alkaline as well as lanthanides, actinides, post transition metals and metalloids. The field of organometallic chemistry has a huge impact on the day-to-day life of millions of people, since organometallic complexes are used for a wide range of large-scale homogeneously catalyzed reactions as well as precursors for the preparation of thin films in the semiconductor industry. The influence of this field was honoured with several Nobel Prizes in chemistry. For instance, for the discovery of the Grignard Reagent and Metal Catalyzed Hydrogenation for Victor Grignard and Paul

Sabatier in 1912 and the discovery of the Metallocenes by E.O. Fischer and G. Wilkinson in

1973. Also, several Nobel Prizes were awarded for the development of groundbreaking catalysts based on organometallic compounds (e.g. 1963 for Ziegler-Natta Catalysis, 2001 for assymetric hydrogenation, 2005 for metal-catalyzed metathesis and 2010 for palladium catalyzed cross-couplings).

2.2 Historic Development and Significance of Organometallic Chemistry

Figure 2.1: Depiction of the synthesis of one of the first organometallic compounds cacodyl by Louis Claude Cadet de Gassicourt in 1760.

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

As in many cases in the history of new fields in chemistry, the discovery of organometallic compounds was a mere serendipity. The first reported man-made organometallic compound is most likely (As2(CH3)4), or Cacodyl, which was a component of “Cadet’s Fuming Liquid”. This liquid was prepared in 1760 by Louis Claude Cadet de Gassicourt by the reaction of AsO3 and potassium acetate. A side product of this reaction is cacodyl oxide. The presence of methyl radicals, who are bound to the Arsen was later postulated by Robert Bunsen.

Figure 2.2: Schematic of the preparation of Zeise Salt by the reaction of K2PtCl4 in boiling ethanol.

In 1827 William Zeise discovered the first Olefin Transition Metal-Complex, the so called Zeise

Salt which has the formuly [K[PtCl3(C2H4)]·H2O. This compound was prepared by reacting

PtCl4 in boiling ethanol. In this comlex the platinum is fourfold coordinated, with each chlorine

2 taking up one coordination site and the C2H4 coordinating to the fourth site with a η bond.

However, the real composition and structure remained for a long time under controversial discussion. The chemist at that time very influential chemist Justus Liebig highly criticized the assumption by Zeisse that Ethylene is bound within the structure. The ultimate proof for the presence of Ethylene in this structure was finally given by the chemist Birnbaum in 1868, who prepared the same compound using Ethylene gas. During the second half of the 19th century the

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes constitution of this compound remained uncertain and was finally resolved in the 1920 with the development of X-ray diffraction techniques.

Figure 2.3: Depiction of the synthetic route of Edward Frankland towards diethylzinc.

Edward Frankland discovered another milestone compound of historical significance, the very pyrophoric compound diethyl zinc (Zn(Et)2). It was prepared by the reaction of zinc with ethyliodide. Due to its reactive nature it has become a very important reactant in organic chemistry.

Ludwig Mond discovered the first carbonyl compounds in 1890. Ni(CO)4 is an intermediate in the so called Mond process, which is used for the preparation of highly pure Nickel.

Nickeloxide, NiO reacts with Syngas (H2 and CO), forming impure Ni and H2O. At temperatures of roughly 50-60 °C the NiCO4 is formed from the prepared impure Nickel.

NiCO4 is a highly volatile gaseous compound and can be deposited as ultrathin pure Nickel films on substrates or pallets at around 220 °C. Among the carbonyl preparation reactions, this is the one occurring under the mildest conditions.

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

Figure 2.4: Illustration of the three steps occuring during the Mond process for the preparation of ultra-pure Nickel

The chemistry of carbonyls was promoted later on by the likes of Walter Hieber and Walter

Reppe. Walter Hieber did a lot of groundbreaking research at the Technical University Munich since the 1920s and developed among other things the first routes towards Metal Carbonyl

Hydrides, for instance Fe(CO)4H2 or Mn(CO)H5. These hydrides, particularly Co(CO)4H and

Rh(CO)4H are still today important intermediates of the catalysts for the hydroformulation, which is used for the preparation of Aldehydes from Olefins (Figure 5). Walter Repp on the other hand worked for the BASF (Badische Soda und Anillin Fabrik) and developed a manifold of catalytic processes using Ni(CO)4 or Co(CO)4 as catalysts for the hydrocarboxylation of

Olefins and Alkynes using CO as a reagent. The carbonyl chemistry however is not only limited to large scale industrial reactions, also in nature, so far three active centers of enzymes have been identified to be metal-carbonyls.

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

Figure 2.5: Scheme of the reaction process of hydroformulation using one of the metal carbonyl hydrides discovered by Walter Hieber.

A further class of important organometallic compounds are the so called sandwhich complexes.

This class of molecules features usually two arenes bound to a metal center on opposite sites by a haptic covalent bond. The most famous subclass of these materials are the metallocenes, which feature two cyclopentadienyl units bound to a metal center. The first metallocene reported was Ferrocene, which was independently discovered by Kealy and Pauson, as well as

Miller et al. in 1951. Kealy and Pauson were actually trying to prepare Fulvalene from a cylopentadienyl salt using anhydrous ironchloride as an electron acceptor. The reaction of

Miller et al. featured Cyclopentadiene, iron as well as the oxides of either aluminum, potassium or molybdenum. Both reactions yielded a product with the sum formula C10H10Fe.

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

Figure 2.6: Depiction of the two different synthetic pathways of Pauson and Kealey (a) and Miller (b) towards Ferrocene.

In the following year Fischer and Wilkinson determined the structure of the compound and concentrated on the further development of the metallocenes. Their works demonstrated that the electrons located on the cyclopentadienyl ring equally contributed to the bonding of the metal-center between the two aromatic rings, leading to a η5 bond. Fischer and coworkers furthermore developed analogues compounds for Nickel and Cobalt. Metallocenes and their derivatives are of significance for polymerization reactions, and are used as Ziegler Natta

Catalysts.

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

Figure 2.7: Depiction of the Singlet-Fischer-Carbenes as well as the Triplet-Schrock-Carbenes including an example for both classes.

Apart from his research on metallocenes, E.O. Fischer also helped establishing the class of metal-carbenes, with R. Schrock being another important figure in this field of organometallic chemistry. In the honor of these two scientists two classes of metallcarbenes are named Fischer and Schrock Carbenes. The Fischer Carbenes feature high oxidation state metal-centers, usually use middle and late transition metals and are π-acceptor ligands, while the Schrock Carbenes consist of low oxidation state metal centers of early transition metals and are usually π-donor ligands. Apart from these two classes, N-heterocyclic carbenes (NHC) are much more commonly used, since they are purely spectator ligands, that do not influence the electronic structure of the metal as heavily as the afore mentioned Fischer and Schrock Carbenes. These

NHC Ligands are of high relevance, in the research towards new catalysts and have also been established in the chemical industry, e.g. as a part of Grubbs catalysts used for the olefin methathesis (Figure 8).

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

Figure 2.8: Depiction of the first and second generation Grubbs Catalysts, which both employ carbenes in their structure.

2.3 Discovery of the Low-Valent Group 13 Organyls and Chronological Development of Their Synthesis

The exotic chemistry of low valent group 13 organyls EIR (E = Al, Ga, In, Tl) has caught enormous attention by the organometallic community due to their stability and availability as ligands in coordination chemistry. The first synthesized low valent group 13 compounds were

InCp and TlCp reported already in 1957 by E. O. Fischer and H. Meister.1-2 Since the synthetic route of those heavier element containing EIR group 13 organyls is not feasible for the lighter elements such as aluminum and gallium, the syntheses of AlIR and GaIR were published 34 years later. Schnöckel et al. successfully synthesized [AlCp*]4 in 1991. Reaction of AlCl and

3 [MgCp2] led to tetrameric [AlCp*]4. Two years later in 1993, Schnöckel et al. achieved to synthesized the related gallium (I) organyls including [GaCp*]6 and its Cp derivatives using the same synthetic pathway.4 (Scheme 1)

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

Scheme 2.1 Synthesis of ECp*

Thereafter, Jutzi et al. reported an alternative synthetic method which provided more convenient access to EIR.5-6 The former methods required metastable AlCl and GaCl generated at harsh conditions e.g. heating to 1000 °C, while the routes provided by Jutzi comprises reductive elimination of salt in Cp*AlCl2 and Cp*GaI2 employing elemental potassium.

(Scheme 2)

Scheme 2.2 Synthesis of ECp*

Nowadays the most commonly used synthetic method to prepare GaCp* is using metal stable

GaI generated by ultra-sonication in benzene which is a modified procedure of a report by Green et al. who synthesized in 1990 a different, unrelated GaI species.7 (Scheme 3)

Scheme 2.3 Synthesis of GaCp* (most commomly adopted procedure)

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

In 2012, Ganesamoorthy et al. developed a new large scale synthetic route towards AlCp* via thermally induced reductive elimination of Cp*H from the organo-aluminum hydride species

8 Cp*2AlH in an exceptionaly high yield (93 %). The reaction of AlCl3 with LiAlH4 in 3:1 molar ratio leads to Cl2AlH which reacts further with KCp* in Et2O to afford Cp*2AlH. Cp*2AlH reductively eliminates Cp*H in toluene at 110 °C. (Scheme 4)

Scheme 2.4 Synthesis of AlCp* (improved procedure for large scale synthesis)

Due to the stability as well as the isolobal structure to CO or phosphines, EIR species are interesting ligands for transition metals in coordination chemistry. This topic will be discussed more in detail in the next section. (See 2.4)

2.4 Coordination chemistry of ECp* toward transition metals

EIR species are an isolobal analogs of CO or phosphine, they possess two vacant p orbitals perpendicular E−R axis and one lone pair electron in the  orbital. (Figure 9)

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

Figure 2.9 Isolobality of EIR species and CO

Therefore, the interaction between the EIR species and transition metals follows as shown in

Figure 2.

Figure 2.10 Bonding situation in transiton metal-EIR complexes.

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

The synthetic strategy of ECp* transition metal compounds is frequently found in the form of ligand substitution reactions with labile ligands. The first attempt for synthesizing TM-EIR complexes was performed using homoleptic TM−carbonyls. However, the strong -donating properties of ECp* increase the -backdonating properties of carbonyls which prohibits the further substitution of ligands. Accordingly, labile ligands with weaker -accepting properties such as olefins were adopted to access homoleptic ECp*−TM complexes. (Scheme 5) In the case of palladium, the redox reaction takes place at the palladium center transforming one

I III equivalent of Ga Cp* to Me2Ga Cp* as a byproduct.

Scheme 2.5 Synthesis of homoleptic TM−ECp* complexes

The homoleptic [M(ECp*)4] complexes are used as molecular building blocks for multi-nuclear clusters. T. Steinke et al. synthesized homoleptic dimeric clusters including [Pt2(2-

AlCp*)(AlCp*)2] and [Pd2(2-GaCp*)(PPh3)2] by treatment of [M(GaCp*)4] with [M(cod)2] or [Pd(dvds)3] (cod = 1,5-Cyclooctadiene, dvds = 1,3-Divinyl-1,1,3,3-tetramethyldisiloxane) and subsequent addition of ECp*.9 (Scheme 6) Heteroleptic dimeric clusters are accessed by the treatment of [Pd2(dvds)3] or [Pt(cod)2] by 3.0 equiv. of GaCp* and 2.0 equiv. of PMe3.

(Scheme 7)

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

Scheme 2.6 [M(GaCp*)4] as building block for dimeric clusters of the type [MPt(GaCp*)2(-

9 GaCp*)3]. Reprinted with permission from Organometallics, 2003, 22 (13), pp 2705–2710.

10 Scheme 2.7 Synthesis of heteroleptic dimeric clusters of the type [M2(GaCp*)3(PMe3)2].

Reprinted with permission from Dalton Trans., 2014, 43, 3114–3120. Copyright (2014) Royal

Society of Chemistry.

In a similar manner, trimeric homoleptic clusters can be synthesized by controlling the stoichiometry of ECp*.10-11

Recently, it has been discovered a novel pathway to approach heteroloptic dimeric cluster via reductive elimination unlike the conventional method by substitution of labile ligand with ECp*.

This will be discussed in chapter 4 as a major discussion of this thesis.

The ECp* containing transition metal complexes are used as well in synthesis of metal-rich complexes. M. Molon et al synthesized organozinc-rich palladium and platinum clusters.10

(Scheme 8)

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

Treatment of 6.0 equivalents of ZnMe2 and excess of PMe3 with [M(2-GaCp*)3(PMe3)] (M =

Pd or Pt) leads to mononuclear compounds [M(ZnCp*)2(ZnMe2)(PMe3)2].

Scheme 2.8 Synthesis of zinc rich transition metal clusters.10 Reprinted with permission from

Dalton Trans., 2014, 43, 3114–3120. Copyright (2014) Royal Society of Chemistry.

2.5 Redox Chemistry of Transition Metal Group 13 Complexes

F F The reaction of cationic complexes such as [Ga2Cp*][BAr4 ] and [FeCp2][BAr4 ] with ECp* containing transition metal complexes gives further insight into the redox chemistry of ECp* species. Treatment of homoleptic complex [M(GaCp*)4] with one equivalent of

F F F [Ga2Cp*][BAr4 ] or [FeCp2][BAr4 ] led to the product [MGa(GaCp*)4][BAr4 ] (M = Ni or Pt)

F F 12-13 whereas 1/3 equiv. of [FeCp2][BAr4 ] yielded [(2-Ga)Pt3(GaCp*)6][BAr4 ].

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

F Scheme 2.9 Reaction of [M(GaCp*)4] (M = Ni or Pt)with [FeCp2][BAr4 ] leading to an oxidative cleavage of Cp* to form “naked” Ga+ containing complexes.[Buchin 2006, Halbherr

2010] Reprinted with permission from Angew. Chem. Int. Ed. 2006, 45, 5207 –5210 and Angew.

Chem. Int. Ed. 2010, 49, 1878 –1881 Copyright (2006, 2010) Wiley.

F Reaction behavior of heteroleptic complex of the type [Ni(ECp*)2(PPh3)2] with [FeCp2][BAr4 ] will be discussed as a major topic of this thesis in chapter 3 in detail.

Aforementioned new synthetic route from aluminum (III) hydride species to AlICp* developed by C. Ganesamoorthy et. al.8 is an inspiration of redox chemistry of group 13 complexes. Indeed, hydrides of main group metals are playing a key role in organic synthesis as Lewis acid catalysis14-21 and hydrogen storage materials22-23 as well as in material chemistry due to its volatility in chemical vapor deposition techniques 24-27 Moreover, the comprehensive elucidation of chemical reactivity of hydrides in terms of addition, elimination and redox reactions has been a goal for inorganic chemical society.28-31 Previous studies of main group hydride compounds show the unusual stabilization of aluminum with +1 oxidation state via

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes reductive dehalogenation from the aluminum halide species5, 28, 30, 32 and ECp* is remains arguably the excellent example of those studies due to their exotic chemistry and strong  donating and  accepting properties towards transition metal complexes.6

C. Ganesamoorthy et al reported a novel pathway to AlICp* via reductive elimination of Cp*H

0 from Cp*2AlH while Cp*AlH2 reductively eliminates Cp*H and forms metallic Al . Both results provided very important information in terms of not only the final product but also the type of chemical bonding. Cp*2AlH and Cp*AlH2 lead to the formation of metal-metal bonds where the aluminum metal is clearly containg metal-metal bonds as well as AlCp* known as

8 tetrameric complex [AlCp*]4. This work was an inspiration of preparing intermetallic complexes including bonds between transition metals and group 13 metals via intramolecular reductive elimination. The reaction with aluminum hydride towards transition metal complexes leading to the formation of TM-Al will be discussed in detail in chapter 4.

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

3. References

1. Fischer, E. O.; Hofmann, H. P., Metall-cyclopentadienyle des Indiums. Angew. Chem.

1957, 69, 639-640.

2. Meister, H., Cyclopentadienyl-thallium. Angew. Chem. 1957, 69, 533-534.

3. Dohmeier, C.; Robl, C.; Tacke, M.; Schnöckel, H., The Tetrameric Aluminum(I)

Compound [{Al(η5-C5Me5)}4]. Angew. Chem., Int. Ed. 1991, 30, 564-565.

4. Loos, D.; Schnöckel, H., (Cyclopentadienyl)-Gallium(I)-Verbindungen. J. Organomet.

Chem. 1993, 463, 37-40.

5. Schulz, S.; Roesky, H. W.; Koch, H. J.; Sheldrick, G. M.; Stalke, D.; Kuhn, A., A Simple

Synthesis of [(Cp*Al)4] and Its Conversion to the Heterocubanes [(Cp*AlSe)4] and

[(Cp*AlTe)4] (Cp* = η5-C5(CH3)5). Angew. Chem., Int. Ed. 1993, 32, 1729-1731.

6. Jutzi, P.; Neumann, B.; Reumann, G.; Schebaum, L. O.; Stammler, H.-G.,

Pentamethylcyclopentadienylindium (Cp*In) as Terminal Ligand in the Chemistry of

Chromium. Organometallics 1999, 18, 2550-2552.

7. Green, M. L. H.; Mountford, P.; Smout, G. J.; Speel, S. R., New synthetic pathways into the organometallic chemistry of gallium. Polyhedron 1990, 9, 2763-2765.

8. Ganesamoorthy, C.; Loerke, S.; Gemel, C.; Jerabek, P.; Winter, M.; Frenking, G.;

Fischer, R. A., Reductive elimination: a pathway to low-valent aluminium species. Chem.

Commun. 2013, 49, 2858-2860.

9. Gemel, C.; Steinke, T.; Weiss, D.; Cokoja, M.; Winter, M.; Fischer, R. A., [M(GaCp*)4]

(M = Pd, Pt) as Building Blocks for Dimeric Homoleptic Cluster Compounds of the Type

[MPt(GaCp*)5]. Organometallics 2003, 22, 2705-2710.

10. Molon, M.; Gemel, C.; Fischer, R. A., Organogallium- and organozinc-rich palladium and platinum clusters. Dalton Trans. 2014, 43, 3114-3120.

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

11. Steinke, T.; Gemel, C.; Winter, M.; Fischer, R. A., The Clusters [Ma(ECp*)b] (M=Pd,

Pt; E=Al, Ga, In): Structures, Fluxionality, and Ligand Exchange Reactions. Chem. Eur. J.

2005, 11, 1636-1646.

12. Buchin, B.; Gemel, C.; Cadenbach, T.; Fernández, I.; Frenking, G.; Fischer, R. A.,

“Naked” Ga+ and In+ as Pure Acceptor Ligands: Structure and Bonding of

[GaPt(GaCp*)4][BArF]. Angew. Chem., Int. Ed. 2006, 45, 5207-5210.

13. Halbherr, M.; Bollermann, T.; Gemel, C.; Fischer, R. A., Selective Oxidative Cleavage of Cp* from Coordinated GaCp*: Naked Ga+ in [GaNi(GaCp*)4]+ and [(μ2-

Ga)nM3(GaCp*)6]n+ Angew. Chem., Int. Ed. 2010, 49, 1878-1881.

14. Bulychev, B. M., A new stage in the development of transition metal alumohydrides.

Polyhedron 1990, 9, 387-408.

15. Duchateau, R.; Meetsma, A.; Teuben, J. H., Sterically crowded monomeric neutral bis(benzamidinato) compounds of aluminium, [PhC(NSiMe3)2]2AlX (X = Cl, H); X-ray crystal structure of [PhC(NSiMe3)2]2AlH. Chem. Commun. 1996, 223-224.

16. Natta, G.; Pino, P.; Mazzanti, G.; Longi, P.; Bernardini, F., The Reaction between

Styrene and Triisobutylaluminum. J. Am. Chem. Soc. 1959, 81, 2561-2563.

17. Spielmann, J.; Buch, F.; Harder, S., Early Main-Group Metal Catalysts for the

Hydrogenation of Alkenes with H2. Angew. Chem., Int. Ed. 2008, 47, 9434-9438.

18. Harder, S., Molecular early main group metal hydrides: synthetic challenge, structures and applications. Chem. Commun. 2012, 48, 11165-11177.

19. Mandal, S. K.; Roesky, H. W., Group 14 Hydrides with Low Valent Elements for

Activation of Small Molecules. Acc. Chem. Res. 2012, 45, 298-307.

20. Harder, S., Molecular early main group metal hydrides: synthetic challenge, structures and applications. Chem. Commun. (Cambridge, U. K.) 2012, 48, 11165-11177.

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

21. Spielmann, J.; Buch, F.; Harder, S., Early main-group metal catalysts for the hydrogenation of alkenes with H2. Angew. Chem., Int. Ed. 2008, 47, 9434-9438.

22. Zidan, R.; Garcia-Diaz, B. L.; Fewox, C. S.; Stowe, A. C.; Gray, J. R.; Harter, A. G.,

Aluminium hydride: a reversible material for hydrogen storage. Chem. Commun. (Cambridge,

U. K.) 2009, 3717-3719.

23. Orimo, S.-i.; Nakamori, Y.; Eliseo, J. R.; Zuettel, A.; Jensen, C. M., Complex Hydrides for Hydrogen Storage. Chem. Rev. (Washington, DC, U. S.) 2007, 107, 4111-4132.

24. Choi, H.; Hwang, S., Volatile Amidoalane Compounds for Chemical Vapor Deposition of Aluminum. Chem. Mater. 1998, 10, 2323-2325.

25. Xenidou, T. C.; Prud'homme, N.; Vahlas, C.; Markatos, N. C.; Boudouvis, A. G.,

Reaction and Transport Interplay in Al MOCVD Investigated Through Experiments and

Computational Fluid Dynamic Analysis. J. Electrochem. Soc. 2010, 157, D633-D641.

26. Hanabusa, M.; Ikeda, M., Wavelength dependence in photochemical vapor deposition of aluminum film using dimethylaluminum hydride. Appl. Organomet. Chem. 1991, 5, 289-93.

27. Baum, T. H.; Larson, C. E.; Jackson, R. L., Laser-induced chemical vapor deposition of aluminum. Appl. Phys. Lett. 1989, 55, 1264-6.

28. Aldridge, S.; Downs, A. J., Hydrides of the main-group metals. New variations on an old theme. Chem. Rev. (Washington, D. C.) 2001, 101, 3305-3365.

29. Uhl, W.; Matar, M., Hydroalumination of nitriles and isonitriles. Z. Naturforsch., B:

Chem. Sci. 2004, 59, 1214-1222.

30. Jones, C.; Rose, R. P., Synthesis and characterisation of tetramethylpiperidinyloxide

(TEMPO) complexes of group 13 metal hydrides. New J. Chem. 2007, 31, 1484-1487.

31. Dagorne, S.; Atwood, D. A., Synthesis, Characterization, and Applications of Group 13

Cationic Compounds. Chem. Rev. (Washington, DC, U. S.) 2008, 108, 4037-4071.

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

32. Roesky, H. W., Hydroalumination reactions in organic chemistry. Aldrichimica Acta

2004, 37, 103-108.

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

Oxidation of Ni-ECp* Complexes: Stable Open-Shell NiI Cations

[Ni(ECp*)n(PPh3)4−n]+ (n = 2, 4; E = Al, Ga)

Jiyeon Kim, Markus Halbherr, Christian Gemel, and Roland A. Fischer

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

3 Oxidation of Ni-ECp* Complexes: Stable Open-Shell NiI Cations 1 [Ni(ECp*)n(PPh3)4−n]+ (n = 2, 4; E = Al, Ga)

3.1 Abstract

I Uncommon Ni cationic complexes were synthesized by treating [Ni(ECp*)2(PPh3)2] (E = Al

F or Ga, Cp* = pentamethylcyclopentadienyl) with one equivalent of [FeCp2][BAr4 ]. All compounds have been prepared readily in a high yield. The paramagnetic compounds were characterized by single crystal X-ray crystallography, Mass spectroscopy, elemental analysis, magnetic susceptibility and EPR spectroscopy.

3.2 Introduction During the past two decades the coordination chemistry of low-valent EIR (E = Al, Ga) ligands to transition metals has been thoroughly investigated. In the case of Cp*, so far every reported monomeric homoleptic [M(Ecp*)] complex is a saturated species containing 18 valence electrons (VE). This feature is independent of the used transition metal M. However, for GaR species containing bulkier R groups, i.e. β-diketiminates (e.g. Ga(ddp) with ddp =

i i CH(CMeNC6H3(C6H3Pr 2-2,6))2) or guanidinates (e.g. Ga(guan) with guan = (C6H3Pr 2-

i 4 2,6)NC(NCy2)N(C6H3Pr 2-2,6) unsaturated species including [(1,3-cod)Pt(Ga(ddp))2] (cod = cyclooctadiene) or [Pt(Ga(guan))3] have been reported. The very strong and polar M-E bond is responsible for the reactivity of these class of mixed-metal complexes. In these compounds

1 The results presented in this chapter have been published in a peer reviewed journal. Reprinted with permission from Inorg. Chem., 2015, 54 (20), pp 9675–9677. Copyright (2015) American Chemical Society. 24

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes the ligand dissociation is prevented by the close vicinity of the nucleophilic M and electrophilic

E. Strong and weakly polar bond such as C-H, Si-H or C-C bonds were activated by compounds with this interesting electronic settings. To further elucidate this correlation, recent studies deployed by our group focused on investigating the behavior of [M(ECp*)n] complexes towards strong one-electron oxidizing agents, for example ferrocenium salts. Suprisingly, neither for

[M(GaCp*)4] (M = Ni, Pt) nor for [Pd3(GaCp*)8] a reaction on the metal-site was observed but on the organic ligand sphere, leading to the removal of decamethylfulvalene as a dimerization

+ product of the cp* radical. The reaction products were [Ni(GaCp*)4(Ga)] as well as the

n+ + triangular clusters [M3(GaCp*)6(Ga)n] (M = Pd, Pt; n = 1, 2), which contained “naked” Ga ions as terminal or bridging ligands, respectively. Whithin this chapter, the oxidation of

F [Ni(ECp*)n(PPh3)(4−n)] complexes (E = Al, Ga; n = 2, 4) by [FeCp2][OTf] and [FeCp2][BAr4 ]

F (OTf = triﬂuoromethanesulfonate; Cp = cyclpentadienyl; BAr4 = B[3,5-(CF3)2C6H3]4) will be

+ discussed. Interestingly, [Ni(ECp*)n(PPh3)(4−n)] cations are formed by oxidation of the transition-metal center, which reveal the oxidation state NiI. Outstaningly, these products are the ﬁrst mixed d-block metal/main-group metal open-shell complexes ever reported and represent new examples for only a handful known organometallic monomeric open-shell d9 complexes.8

In general the oxidation state 1+ for nickel is uncommon, however, in fact, a few stable compounds have been reported in the literature. For example the tetraazamacrocycle [Ni(tmc)]+

(tmc = 1,4,8,11-tetramethyl-1,4,8,11-tetraazacyclotetradecane) is an important model for enzymatic processes e.g. Acetyl-CoA synthesis. This structure is also related to the methyl coenzyme M Reductase F430 which generates in metabolic reactions in the Achea methane from thiomethyl compounds. Other stable NiI complexes include [NiCl(NHC)L] [NHC = 1,3- bis(2,6-diisopropylphenyl)imidazol-2-ylidene; L = NHC, PPh3], which is used for the catalysis of cross-coupling and amination reactions.10 A number of low coordinate and neutral NiI 25

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes complexes were investigated by Laskowski and Hillhouse In the recent literature NiI complexes which were supported by NHC and Cp ligands have been explored. Furthermore, a small number of cationic compounds have been discovered, such as the tricoordinated and

I 13 I 14 tetracordinated complexes [Ni (PPh3)3][BF4] and [Ni (PMe3)4][BPh4], respectively. For the latter compound a disproportion into Ni0 and Ni was observed, as it is not stable in solution.

3.3 Result and Discussion The oxidation the Ni(0) centers to Ni(1) is achieved by the addition of one equivalent of

F [FeCp2][BAr4 ] in fluorobenzene at room temperature to the complexes [Ni(ECp*)2(L)2] (L =

AlCp*, PPh3; E = Al, Ga). The E(I) ligands and the Cp* remain unchanged during this reaction.

I I F Isolation of the complexes [Ni (AlCp*)4][Cp*Al(OTf)3] (1), [Ni (AlCp*)2(PPh3)2]-[BAr4 ] (2)

I F and [Ni (GaCp*)2(PPh3)2][BAr4 ] (3) was possible. Noticeably 1 is thermally highly unstable and decomposes during workup, hence only few crystals were isolated and were used for

Scheme 3.3.1 Synthesis of NiI Compounds 1-3

characterization. On the other hand, compounds 2 and 3 were stable at inert conditions at room temperature either in the solid state as well as in solution.

A quite suprising result is the finding that the trasition metal center is oxidized instead of the

ECp* ligand. In particular, because for the analogous compound [Ni(GaCp*)4] the oxidation 26

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

F + by [FeCp2][BAr4 ] resulted in the formation of [NiGa(GaCp*)4] containing Ga+ moieties.

Additionally, in the case of analogous palladium and platinum complexes a similar oxidation accompanied by the formation of decamethylfulvalene.

The compounds 1-3 are crystallized from a mixture of fluorobenzene and hexane. In Figure

3.3.1 depiction of the molecular structures of all three compounds is displayed and in Table

3.6.1 all important structural paramters are summarized. An overview of the important interatomic distances is given in Table 3.6.2. No complete single crystal diffraction datasets of compound 1 could be obtained, because of the high instability of this material. During the experiment the compound slowly decomposed. However, an overall completeness of 80 % with a final data to parameter ratio of 9:1 could be achieved and also a sufficiently good R value leads to high confidence in the interpretation of the diffraction data. More or less strong distortions of the tetrahedral coordination environment of the Ni center was observed in all three compounds. Particularly obvious was this finding for the homoleptic complex 1, where the Al-Ni-Al angles amount to 103.23(15) ° and 132.00(17) °. In the parent, neutral [Ni(AlCp*)4] species the ideal tetrahedral angles of 109.5° are present. The Jahn-Teller distortion is held accountable for this flattening of the structure with the prototypical [CuCl4]2- as an example.

+ For the similar homoleptic cation [Ni(PMe3)4] similar distortions were observed (104.6(1)° to

119.9(1)°). Even larger distortion have been found in the phosphine substituted complex 2 and

3 with angles between 98.07(4)° and 131.24(4)°) and 3 (L-Ni-L angles between 98.75(2)° and

134.35(3)°, respectively. This is caused most probably by the steric repulsion of the very bulky

PPh3 ligands. Notably, no indication for hydride species were found using spectroscopic methods (NMR). Additionally, the compounds are all paramagnetic, which was confirmed by

EPR spectroscopy.

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

Figure 3.3.1 Molecular structure (POV-Ray Plot) of the cationic parts of the salts 1-3, namely + + + [Ni(AlCp*)4] (1), [Ni(AlCp*)2(PPh3)2] (2) and [Ni(GaCp*)2(PPh3)2] (3) in the solid state as determined by single crystal X-ray diffraction; displacement ellipsoids are shown at the 50% probability level, hydrogen atoms are omitted for clarity. Selected interatomic distances and angles are summarized in Table 3.6.2.

The Ni-E bond distances are only slightly affected by the oxidation. The mean Ni-Al distance

+ amounts to 2.17 Å in [Ni(AlCp*)4] and is the same in the parent neutral compound

[Ni(AlCp*)4]. In the phosphine substituted species a slight increase is observed in the Ni-Al bond upon oxidation, from 2.21 Å to 2-27 Å and from 2.25 Å to 2.32 Å in compounds 2 and 3, respectively. Moreover, a decrease of the E-Cp*centroid distance was found, most likely because of the increase of the E-Cp* bonds polarity. This is induced by the over-all positive charge of the complex.

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

Broad signals in the 1H-NMR spectra of compounds 2 and 3 indicate their paramagnetic nature.

Nonetheless, in the 31P-NMR unusually sharp signals were observed. Via Evans NMR method at room temperature the magnetic susceptibility (eff) of these complexes were determined.

Values for eff of 1.09 BM and 1.15 BM were calculated in fluorobenzene (BM = Bohr

Magnetron). These values are slightly lower than the theoretically calculated values for NiI in tetrahedral geometry (1.73 BM). These two compounds showed also EPR activity, with average g-factors of 2.11 and 2.15 for 2 and 3 respectively. (Figure 3.6.3) These findings are in accordance with other NiI complexes. The recorded spectra are highly complex and thus the hyperfine couplings of the EPR signals could not be simulated. The cation peaks of these two complexes were also revealed using liquid injection field desorption ionization mass spectrometry (LIFDI-MS), with the cation peaks at m/z [a.u] of 906 and 993 for compounds 2 and 3, respectively.

3.4 Conclusion

Using the more electrophilic metal ligator Al, both the homoleptic [Ni(AlCp*)4] and the phosphine containing analogue [Ni(AlCp*)2(PPh3)2] yield cationic 17 VE complexes (1 and 2).

For the metal Ligator Ga, only the heteroleptic complex [Ni(GaCp*)2(PPh3)2] is reacting in the same fashion, yielding 3. The homoleptic [Ni(GaCp*)4] complexe yields after oxidation the

+ cation [Ni(GaCp*)4(Ga)] featuring naked Ga. This is because the cleavage of a C5Me5 radical occurs easier on the less polar E-Cp* bond, showcased by the homoleptic GaCp* substituted

Nickel complex. Furthermore, the substitution of ECp* by PPh3 forms more electron deficient metal centers increasing the polarity of the E-Cp* bond.

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

3.5 References

(1) (a) Asay, M.; Jones, C.; Driess, M. Chem. Rev. 2011, 111, 354-396; (b) Molon, M.;

Bollermann, T.; Gemel, C.; Schaumann, J.; Fischer, R. A. Dalton Trans. 2011, 40, 10769-10774;

Fischer, R. A. Angew. Chem., Int. Ed. 2005, 44, 2943-2946; (d) Steinke, T.; Gemel, C.; Cokoja,

M.; Winter, M.; Fischer, R. A. Angew. Chem., Int. Ed. 2004, 43, 2299-2302; (e) Steinke, T.;

Gemel, C.; Winter, M.; Fischer, R. A. Chem. Eur. J. 2005, 11, 1636-1646; (f) Bollermann, T.;

Cadenbach, T.; Gemel, C.; von Hopffgarten, M.; Frenking, G.; Fischer, R. A. Chem. Eur. J.

2010, 16, 13372-13384.

(2) Hardman, N. J.; Eichler, B. E.; Power, P. P. Chem. Commun. 2000, 1991-1992.

(3) Jones, C.; Junk, P. C.; Platts, J. A.; Stasch, A. J. Am. Chem. Soc. 2006, 128, 2206-2207.

(4) Kempter, A.; Gemel, C.; Fischer, R. A. Chem. Eur. J. 2007, 13, 2990-3000.

(5) Green, S. P.; Jones, C.; Stasch, A. Inorg. Chem. 2007, 46, 11-13.

(6) Cadenbach, T.; Gemel, C.; Schmid, R.; Block, S.; Fischer, R. A. Dalton Trans. 2004, 3171-

3172.

(7) Halbherr, M.; Bollermann, T.; Gemel, C.; Fischer, R. A. Angew. Chem., Int. Ed. 2010, 49,

1878-1881.

(8) Margulieux, G. W.; Weidemann, N.; Lacy, D. C.; Moore, C. E.; Rheingold, A. L.; Figueroa,

J. S. J. Am. Chem. Soc. 2010, 132, 5033-5035.

(9) (a) Ram, M. S.; Riordan, C. G.; Ostrander, R.; Rheingold, A. L. Inorg. Chem. 1995, 34,

5884-5892; (b) Ram, M. S.; Yap, G. P. A.; Liable-Sands, L.; Rheingold, A. L.; Marchaj, A.; 30

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

Norton, J. R. J. Am. Chem. Soc. 1997, 119, 1648-1655; (c) Lafrance, M.; Gorelsky, S. I.; Fagnou,

K. J. Am. Chem. Soc. 2007, 129, 14570-14571; (d) Telser, J. J. Braz. Chem. Soc. 2010, 21,

1139-1157.

(10) Nagao, S.; Matsumoto, T.; Koga, Y.; Matsubara, K. Chem. Lett. 2011, 40, 1036-1038.

(11) (a) Laskowski, C. A.; Hillhouse, G. L. J. Am. Chem. Soc. 2008, 130, 13846-13847; (b)

Anderson, J. S.; Iluc, V. M.; Hillhouse, G. L. Inorg. Chem. 2010, 49, 10203-10207; (c) Iluc, V.

M.; Hillhouse, G. L. J. Am. Chem. Soc. 2010, 132, 15148-15150; (d) Lipschutz, M. I.; Yang,

X.; Chatterjee, R.; Tilley, T. D. J. Am. Chem. Soc. 2013, 135, 15298-15301.

(12) (a) Wu, J.; Nova, A.; Balcells, D.; Brudvig, G. W.; Dai, W.; Guard, L. M.; Hazari, N.; Lin,

P.-H.; Pokhrel, R.; Takase, M. K. Chem. Eur. J. 2014, 20, 5327-5337; (b) Pelties, S.; Herrmann,

D.; de Bruin, B.; Hartl, F.; Wolf, R. Chem. Commun. 2014, 50, 7014-7016.

(13) Saraev, V. V.; Kraikivskii, P. B.; Svoboda, I.; Kuzakov, A. S.; Jordan, R. F. J. Phys. Chem.

A 2008, 112, 12449-12455.

(14) Gleizes, A.; Dartiguenave, M.; Dartiguenave, Y.; Galy, J.; Klein, H. F. J. Am. Chem. Soc.

1977, 99, 5187-5189.

(15) (a) Helmholz, L.; Kruh, R. F. J. Am. Chem. Soc. 1952, 74, 1176-1181; (b) Willett, R. D.;

Larsen, M. L. Inorg. Chim. Acta 1971, 5, 175-179.

(16) (a) Hüttner, W.; Flygare, W. H. J. Chem. Phys. 1969, 50, 2863-2868; (b) de Luca, G.;

Russo, N.; Sicilia, E.; Toscano, M. J. Chem. Phys. 1996, 105, 3206-3210; (c) Bloor, J. E.;

Maksić, Z. B. Mol. Phys. 1973, 26, 397-408; (d) Hoppe, J. I. J. Chem. Educ. 1972, 49, 505; (e)

Bain, G. A.; Berry, J. F. J. Chem. Educ. 2008, 85, 532; (f) Flygare, W. H.; Benson, R. C. Mol.

Phys. 1971, 20, 225-250; (g) Piguet, C. J. Chem. Educ. 1997, 74, 815; (h) Sur, S. K. J. Magn.

Reson. 1989, 82, 169-173.

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

(17) (a) Kotani, M. Progr. Theor. Exp. Phys. Supp. 1961, 17, 4-13; (b) Petuker, A.; Merten, C.;

Apfel, U.-P. Eur. J. Inorg. Chem. 2015, 2015, 2139-2144.

(18) Eckert, N. A.; Dinescu, A.; Cundari, T. R.; Holland, P. L. Inorg. Chem. 2005, 44, 7702-

7704

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

3.6 Experimental Section

3.6.1 General Comments

Material and Methods. All manipulations were carried out in an atmosphere of purified argon using standard Schlenk and glove-box techniques. n-Hexane was dried using an MBraun

Solvent Purification System and fluorobenzene and d6-benzene were dried by passing through a column of dried activated neutral Al2O3. The final H2O content in all solvents was checked

1 2 3 by Karl Fischer titration and did not exceed 5 ppm. [AlCp*] , [GaCp*] , [Ni(AlCp*)4] ,

4 4 F 5 6 [Ni(AlCp*)2(PPh3)2] , [Ni(GaCp*)2(PPh3)2] , [FeCp2][BAr4 ] and [FeCp2][OTf] were prepared as previously described in the literature. Elemental analyses were performed by the

Microanalytical Laboratory (Rubiospek) of the Ruhr-University Bochum. The LIFDI mass spectra were measured with a Jeol AccuTOF GCv spectrometer with a Linden-CMS.de LIFDI source. NMR spectra of paramagnetic substances were recorded on a Bruker Avance DPX-250 spectrometer in fluorobenzene/benzene (10:1 v/v ratio) solution at 298 K. Effective magnetic moment calculated by Evan’s method was conducted in fluorobenzene and dichloromethane with chemical shifts given relative to TMS and were referenced to the residual solvent. The chemical shifts are described in parts per million (ppm), downﬁeld shifted from TMS and are consecutively reported as position (δ H, δ P, δ B), relative integral, multiplicity (s = singlet, m

= multiplet), coupling constant (J in Hz) and assignment. EPR spectra were measured with an

X-band spectrometer in fluorobezene at 16 K (Bruker ELEXSYS500 X-band spectrometer).

FT-IR spectra were measured on a Bruker Alpha FT-IR spectrometer. The molecular structures were measured on an Agilent Technologies SuperNova diffractometer with an Atlas CCD

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

detector, using CuK radiation (λ = 1.541 84 Å) from multilayer X-ray optics. The crystals were coated with a perflusotopolyether, picked up with a cryo loop, and immediately mounted in the nitrogen cold gas stream of the diffractometer. The data were processes with CrysAlisPro.7 The crystal structure was solved by direct methods using SHELXS-97 and reﬁned with SHELXL-

2013.19 Two of the CF3 groups of the BAr4F anion show rotational disorder.8

3.6.2 Synthetic Procedures

[Ni(AlCp*)4][Cp*Al(OTf)3] (1) [Ni(AlCp*)4] (100 mg, 0.141 mmol) was dissolved in fluorobenzene (5 ml) and [FeCp2][OTf] (47 mg, 0.141 mmol) was suspended in fluorobenzene

(5 ml) in a different schlenk tube and both cooled with an iso-propanol/dry-ice bath to -35°C.

The latter suspension of [FeCp2][OTf] was transferred slowly via cannula to [Ni(AlCp*)4] giving a deep red solution. Storage at -30°C for 3 months gave few crystals suitable for single crystal x-ray diffraction. Because of very low stability of the compound 1, further characterization could not be done.

F [Ni(AlCp*)2(PPh3)2][BAr4 ] (2) [Ni(AlCp*)2(PPh3)2] (100 mg, 0.110 mmol) and

F [FeCp2][BAr4 ] (116 mg, 0.110 mmol) were dissolved in fluorobenzene (5 ml) and stirred for

1 h during which the color changed from deep red to dark green. Crystals suitable for single crystal x-ray diffraction were obtained by layering the fluorobenzene solution with n-hexane.

Anal. Calcd. (found) for C88H72BF24Al2NiP2: C 59.68% (59.04%); H 4.1% (4.11%). LIFDI-MS:

+ + 1 m/z = 906.4 [M] , for the cation [Ni(AlCp*)2(PPh3)2] . H-NMR (C6H5F/10% C6D6): δ 10.10 ppm, 1.10 ppm. 31P-NMR : 53.16 ppm.

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

F [Ni(GaCp*)2(PPh3)2][BAr4 ] (3) [Ni(GaCp*)2(PPh3)2] (100 mg, 0.101 mmol) and

F [FeCp2][BAr4 ] (106 mg, 0.101 mmol) were dissolved in fluorobenzene (5 ml) and stirred for

1 h during which the color changed from deep red to light red. Crystals suitable for single crystal x-ray diffraction were obtained by layering the fluorebenzene solution with n-hexane. Anal.

Calcd. (found) for C88H72BF24Ga2NiP2: C 56.98 % (56.53 %); H 3.92 % (4.01 %). LIFDI-MS:

+ + 1 m/z = 993.0 [M] for the cation [Ni(GaCp*)2(PPh3)2] . H-NMR (C6H5F/10% C6D6): δ 11.32 ppm, 3.77 ppm. 31P-NMR : 48.84 ppm.

Comment to the x-ray diffraction experiment of 1:

The high chemical instability and sensitivity of 1 also at low temperatures leads to slow decomposition of the single crystal during the x-ray diffraction experiment and no complete dataset was available. However, the overall completeness of 80%, a final data:parameter ratio of 9:1 and sufficiently good R values (R1=0.0485) allow a consistent interpretation of the data, and therefore a structural discussion of this compound has been included in the manuscript.

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

3.6.3 Spectroscopic and crystallographic data

31 1 11 F Figure 3.6.1 P, H and B NMR spectra of compound 2 [Ni(AlCp*)2(PPh3)2][BAr4 ] in 10 % d6-benzene in fluorobenzene.

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

F 6 Figure 3.6.2 NMR spectra of compound 3 [Ni(GaCp*)2(PPh3)2][BAr4 ] in 10 % d -benzene in fluorobenzene.

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

1000 2000 3000 4000 5000 6000 1000 2000 3000 4000 5000 B (G) B (G)

F Figure 3.6.3 EPR spectra of compound 2 [Ni(AlCp*)2(PPh3)2][BAr4 ] (left) and 3

F [Ni(GaCp*)2(PPh3)2][BAr4 ] (right). The average g-factors for 2 (g = 2.11) and 3 (g = 2.15) were obtained. The measurements were reproduced at 6 K on a different instrument (Varian

E112 EPR- Spectrometer (X-Band)) with identical results. The interpretation of the hyperfine couplings of the highly complicated EPR signals is not possible.

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

Figure 3.6.4 UV-Vis (black) and Photoluminescence (PL, blue) spectra of compound 2 (left) and 3 (right).The UV-Vis and PL were measured in fluorobenzene solution. The spectra of UV-

Vis/PL of compound 2 and 3 show absorbance band and emission peak at abs = 350 nm, em =

404, 428 nm and abs = 355, 444 nm, em = 404, 428 nm.

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

Crystallographic Data

Table 3.6.1 Important crystallographic data of compounds 1, 2 and 3

1 2 3

Empirical formula C59H75Al5F10NiO9S3 C88H72Al2BF24NiP2 C88H72BF24Ga2NiP2 Formula weight 1407.98 1770.88 1856.36 Temperature (K) 108.15 104.15 110.15 Crystal system monoclinic monoclinic monoclinic

Space group Cc P21/n P21/n a/Å 16.7480(8) 20.4713(5) 20.5202(3) b/Å 15.5917(9) 20.1109(4) 20.1408(2) c/Å 26.5642(15) 20.5530(5) 20.5787(3) α/° 90 90 90 β/° 95.151(5) 101.560(2) 101.965(1) γ/° 90 90 90 Volume/Å3 6908.7(6) 8290.0(3) 8320.26(19) Z 4 4 4 3 ρcalcg/cm 1.354 1.419 1.482 μ/mm-1 0.512 0.393 1.004 F(000) 2928 3620 3764 Crystal size/mm3 0.15 × 0.12 × 0.09 0.14 × 0.12 × 0.10 0.17 × 0.14 × 0.09 Radiation MoKα (λ = 0.71073) MoKα (λ = 0.71073) MoKα (λ = 0.71073) 2Θ range for data collection/° 5.666 to 48.498 5.72 to 58 5.72 to 55.12 Index ranges -17 ≤ h ≤ 17 -26 ≤ h ≤ 27 -26 ≤ h ≤ 26 -11 ≤ k ≤ 17 -26 ≤ k ≤ 26 -26 ≤ k ≤ 26 -28 ≤ l ≤ 27 -26 ≤ l ≤ 27 -25 ≤ l ≤ 26 Reflections collected 11368 157220 152416 Data/restraints/parameters 6835/56/784 20038/0/1063 19082/0/1073 Goodness-of-fit on F2 0.789 1.025 1.02

Final R indexes [I>=2σ (I)] R1 = 0.0485 R1 = 0.0704 R1 = 0.0460 wR2 = 0.0668 wR2 = 0.1517 wR2 = 0.1016

Final R indexes [all data] R1 = 0.1000 R1 = 0.1343 R1 = 0.0736 wR2 = 0.0746 wR2 = 0.1792 wR2 = 0.1136 Largest diff. peak/hole / e Å-3 0.48/-0.30 1.39/-0.88 1.29/-0.85

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

Table 3.6.2 Selected interatomic distance [Å] and angle [°]

1 2 3

Ni-Al1 2.181(4) Ni-Al1 2.260(1) Ni-Ga1 2.310(1)

Ni-Al2 2.147(4) Ni-Al2 2.287(1) Ni-Ga2 2.329(1)

Ni-Al3 2.195(3) Ni-P1 2.178(1) Ni-P1 2.213(1)

Ni-Al4 2.142(4) Ni-P2 2.189(1) Ni-P2 2.204(1)

Al1-Cp*centroid 1.876 Al1-Cp*centroid 1.857 Ga1-Cp*centroid 1.933

Al2-Cp*centroid 1.854 Al2-Cp*centroid 1.873 Ga2-Cp*centroid 1.942

Al3-Cp*centroid 1.851

Al4-Cp*centroid 1.844

Al1-Ni-Al2 103.26(14) Al1-Ni-Al2 108.54(5) Ga1-Ni-Ga2 117.30(2)

Al1-Ni-Al3 112.83(16) P1-Ni-P2 131.24(4) P1-Ni-P2 134.35(3)

Al1-Ni-Al4 100.85(15) P1-Ni-Al1 98.07(4) P1-Ni-Ga1 100.25(2)

Al2-Ni-Al3 103.23(15) P2-Ni-Al1 99.47(4) P2-Ni-Ga1 98.75(2)

Al2-Ni-Al4 132.00(17) P1-Ni-Al2 108.22(4) P1-Ni-Ga2 103.99(2)

Al3-Ni-Al4 104.63(13) P2-Ni-Al2 108.53(4) P2-Ni-Ga2 103.34(2)

Ni-Al1-Cp*centroid 173.13(166) Ni-Al1-Cp*centroid 174.12(7) Ni-Ga1-Cp*centroid 168.74 (2)

Ni-Al2-Cp*centroid 167.44(17) Ni-Al2-Cp*centroid 172.91(6) Ni-Ga2-Cp*centroid 175.45 (2)

Ni-Al3-Cp*centroid 177.88(14)

Ni-Al4-Cp*centroid 170.81(16)

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

4 References

(1) S. Schulz, H. W. Roesky, H. J. Koch, G. M. Sheldrick, D. Stalke and A. Kuhn, Angew.

Chem. Int. Ed., 1993, 32, 1729–1731.

(2) P. Jutzi, B. Neumann, G. Reumann and H.-G. Stammler, Organometallics, 1998, 17, 1305–

1314.

(3) Steinke, T.; Gemel, C.; Cokoja, M.; Winter, M.; Fischer, R. A Angew. Chem. Int. Ed, 2004,

43, 2299–2302.

(4) M. Molon, T. Bollermann, C. Gemel, J. Schaumann, R. A. Fischer, Dalton Trans, 2011.

(5) J. Le Bras, H. Jiao, W. E. Meyer, F. Hampel and J. A. Gladysz, J. Organomet. Chem., 2000,

616, 54–66.

(6) Z.-W. Li, A. Yeh and H. Taube, Inorg. Chem., 1994, 33, 2874–2881.

(7) CrysAlisPro Software System, version 1.171.36.20; AgilentTechnologies UK Ltd.: Oxford,

U.K., 2012.

(8) Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112−122.

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

4. Organoaluminum Palladium Complex via Reductive Elimination

Jiyeon Kim, Hung Banh, Mathies Evers, Christian Gemel, Roland A. Fischer

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

4. Organoaluminum Palladium Complex via Reductive Elimination

4.1 Abstract The reaction of AlIII hydride species and palladium phosphine complexes is described in this chapter. The reaction product [(PCy3)Pd-2-AlCp*]2 was obtained via reductive elimination of hydrogen and characterized by NMR spectroscopy, LIFDI-MS, single crystal X-ray diffraction.

4.2 Introduction

There has been a significant amount of efforts from fundamental organometallic chemists to access heterometallic complexes containing direct metal-metal bonds. Although, several classical methods are known, such as oxidative addition etc., metal-metal coupling still remains a challenge. The coordination chemistry of low valent EIR species towards transition metals opened new doors in intermetallic chemistry. Since EIR is an isolobal analog of CO and phosphine, many group 6-12 transition metal complexes containing EIR with the fomular

I [M(E R)n] have been investigated.

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

I In 2012, Ganesamoorthy et al. reported a new synthetic pathway for synthesizing [Al Cp*]4

(Cp* = pentamethylcyclopentadienyl) via reductive elimination.1 This procedure suggested a route with a mechanism distinct from the literature known procedure (Schnöckel)2. Thermally

III induced reaction of Al Cp*2H leads to the formation of [AlCp*]4 which includes the formation of a direct Al-Al bond via reductive elimination of Cp*H. This work was an inspiration of preparing intermetallic complexes including bonds between transition metals and group 13 metals via intramolecular reductive elimination. Even though, reductive elimination in transition metal-complexes is commonly observed particularly in small molecule activations

(C-H or C-C coupling), reports on intramolecular reductive elimination resulting in direct metal-metal bonds remain scarce. Herein, we report the formation of Pd-Al bonds via reductive elimination of H2.

4.3 Result and Discussion

The stoichiometric reaction of Pd(PCy3)2 (Cy = cyclohexyl) and AlCp*H2 in benzene reductively eliminates H2 thermally at 70 °C. (Scheme 4.1) Additionally, the loss of one equivalent of free PCy3 affords [(PCy3)Pd-2-AlCp*]2 (4). Reaction completeness was monitored by in-situ 1H-NMR spectroscopy. A strong gas evolution was already observed as soon as benzene was added into a J-Young NMR tube before sealing. Although, in-situ 1H

NMR measurements show a weak but clear evidence of H2 evolution at  4.46 ppm from the

H2 cleavage in AlCp*H2. (Figure 4.4) Compound 4 decomposes in air and is moisture sensitive and readily dissolve in non-polar solvents such as hexane or toluene. The chemical shift of bridging AlCp* from compound 4 was observed at  2.14 ppm whereas AlCp* and AlCp*H2 protons appear at  1.89 and 1.91 ppm. Protons in the phosphine ligand are displayed as broad

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes peaks at 1.22 and 1.72 ppm, whereas the protons of free phosphine appear in the area of 1.26

– 2.24. The 31P NMR spectrum of compound 4 shows a resonance at  57.09 ppm with trace amounts of free PCy3 at  9.97 ppm. the reactant Pd(PCy3)2 has the phosphine resonance peak at  39.18 ppm and cannot be identified in the final product. Compound 4 could not be completely isolated from free PCy3 by recrystallization due to the indistinct solubility differences. Liquid injection field desorption ionization mass spectrometry (LIFDI-MS) of 4 clearly identifies molecular ion peak at m/z 1098.17. This observation indicates there is no hydride takes place in the molecule particularly at the two palladium atoms. Moreover, the transition metal hydrides exhibit the IR vibration near 2000 cm-1 where no peak has been observed in this area. (Figure 4.7)

Scheme 4.1 Reaction of Pd(PCy3)2 with AlCp*H2

Crystals suitable for single crystal X-ray diffraction are obtained in pentane at -30 °C. Figure

4 shows a depiction of the molecular structure of 4. Crystallographic data and refinement details are shown in Table 4.1 whereas important bond distance and angles are shown in Table 4.2.

The solvate free compound 4 crystallizes in the monoclinic space group C2/c with two fragments in the asymmetric unit. The molecular structure of the type [Pd2Al2P2] has two bridging AlCp* and two cis-terminal phosphine ligands with a geometry described best as two edge-sharing trigonal planar units. (Figure 4.2) The two aluminum atoms are shared within both units. The two bridging AlCp*s are bent to lopsided with an angle of Pd−Pd axis and two aluminum atoms found 130.7°. Interestingly, the two terminal phosphines are bent to opposite

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes direction of AlCp* ligands resulting in a P−Pd−Pd angle of 152.7 °. (Figure 4.1) These effects explain the biased structure of bridging AlCp* of compound 4. In comparison to previous

3-4 observations by Fischer et al. where the compounds [Pd2(GaCp*)3(PMe3)2] [Mariusz Dalton

2014] and [Pd2(GaCp*)3(PPh3)2] [Steinke 2005 chemeurj] with a similar core building unit but with three bridging ligands of the type [Pd2Ga3P2] was reported, quite different results regarding the bonding situation were found. The angle of P-Pd-Pd in those structures, the tree atoms are in linear positions in contrast to compound 4. Also, the steric bulk cyclohexyl groups in phosphine are inclined to the opened site of Pd−Pd bond which consequently stabilize the structure (Figure 2). The Pd1-Pd2 bond length of 2.592 Å in compound 4 is slightly shortened

0 0 4 than as found for the Pd -Pd complexes [Pd2(AlCp*)5] [Steinke 2005 chemeurj] and

3 [Pd2(GaCp*)3(PMe3)2] [Mariusz Dalton 2014] (2.633(1); and 2.623(1)). The bridging Pd−Al

a a bond length of 4 ( 2.424 Å) is in the same range as found in [Pd2AlCp*5] ( 2.448 Å). [Steinke chemeurj 2005]

Figure 4.1 Povray plot of the molecular structure of 1 in the solid state as determined by single crystal X-ray diffraction (the asymmetric unit contains two molecules; thermal ellipsoids are shown 50 % probability; hydrogen atoms are omitted for clarity). Selected atomic distances (Å) as well as angles (°): Pd1−Pd2: 2.5914(6), Al1−Pd1: 2.4284(15), Pd1−Al2: 2.4102(14),

Pd2−Al2: 2.4201(15), Pd2−Al1: 2.4377(15), Pd1−P1: 2.2807 (13), Pd2−P2: 2.2936(13),

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

Al−Cp*: a1.951, Pd1−Al1−Pd2: 100.25(5), Pd1−Al2−Pd2: 99.71(5), Cp* values are calculated from the centroid of the Cp* moiety.

Figure 4.2 Illustration of coordination geometry of compound 1 (M2Al2P2) and the comparison of molecular type M2Ga3P2.

4.4 Conclusion

Within this chapter a facile route to a dinuclear Pd core structure with heteroleptic ligand coordination sphere was presented. Reductive elimination was used as a tool to prepare this new structure. Regarding the state of the art, this is the first described synthesis of a transition metal−group 13 complex via this mechanistic route. It is highly expected that this synthetic approach can help to extend the possibilities in constructing compounds containing metal-metal

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes bonds and will help open up new doors in the field of organomettalic chemistry. Furthermore, we showed a new Pd2Al2P2 core with aluminum ligands which have been directed by the steric bulk of the R groups on the P ligands. The found geometry is very uncommon and the assymetric palladium dimer with relatively unshielded metal site might be of interest as a building unit to access intermetallic clusters.

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

4.5 Experimental Section

4.5.1 General Comments

All manipulations were carried out in an atmosphere of purified argon using standard Schlenk and glove-box techniques. n-Hexane and n-pentane were dried using an MBraun Solvent

Purification System. Benzene was dried by passing through a column of dried activated neutral

6 Al2O3. d -Benzene was degassed and stored under the molecular sieves (4 Å). All the instruments for characterization were identically used described in section 3.6. [AlCp*H2] and

[Pd(PCy3)2] were prepared as previously described in the literature. [Ganesh + a]

4.5.2 Synthetic procedure

[(Pd(AlCp*)(PCy3))2] Pd(PCy3)2 (200 mg, 0.300 mmol) and AlCp*H2 (50 mg, 0.305 mmol) were dissolved in benzene (5 ml), immediate colour change to dark red and gas evolution were observed. The reaction mixture was stirred for 1 h at 80 °C. All the volatiles were removed under vacuum and the residue was extracted with cold (- 60 °C) pentane. Dark-red crystals suitable for single crystal x-ray diffraction were obtained at -30 °C in the concentrated pentane

1 31 solution. H-NMR (C6D6):  2.14 ppm (Cp*Al),  1.77-1.72 ppm (Cy3P). P-NMR (C6D6): 

+ + 57.09 ppm (P-Pd). LIFDI-MS: m/z = 1098.2 [M] , for the cation [(Pd(AlCp*)(PCy3))2] .

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

4.5.3 Spectroscopic, Spectrometric, Crystallographic Data

Figure 4.3 LIFDI-MS data of 1 in toluene solution

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

1 6 Figure 4.4 H-NMR from the NMR reaction of Pd(PCy3)3 and AlCp*H2 in 1:1 ratio in d - benzene. Hydrogen (H2) is observed at  4.46 ppm.

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

Figure 4.5 1H-NMR of compound 4 in d6-benzene.  2.14 ppm (Cp*Al);  1.77-1.72 ppm (Cy3P).

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

Figure 4.6 31P-NMR of compound 4 in d6-benzene.  57.09 ppm, (P-Pd);  9.97 ppm, free phosphine.

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

Figure 4.7 FT-IR spectrum of compound 4.

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

Table 4.1 Crystallographic data and refinement details of compound 4.

Chemical formula C56H96Al2P2Pd2

Formula weight 1098.02

Crystal system Monoclinic

Space group C2/c

a (Å) 71.238(5)

b (Å) 10.2931(3)

c (Å) 45.665(3)

 (°) 136.980(12)

Volume (Å3) 22845(4)

Z 16

-3 Dcalc (g × cm ) 1.277

/mm-1 6.142

F(000) 9280.0

2θ range for data collection (°) 7.276 to 139.998

Reﬂections collected/unique 42875/20848 [Rint = 0.0564]

Data/restraints/parameters 20848/0/1137 2 goodness-of-ﬁt on F 1.006

Final R indices [I > 2(I)] R1 = 0.0469, wR2 = 0.1028 R indices (all data) R1 = 0.0717, wR2 = 0.1503 Table S1 Selected Bond Length (Å) and Angles (°) for compound 4.

Pd(1)−Pd(2) 2.5914(6) Pd(1)−Al(1)−Pd(2) 100.25(5) Pd(1)−Al(1) 2.4284(15) Pd(1)−Al(2)−Pd(2) 99.71(5) Pd(1)−Al(2) 2.4102(14) P(1)−Pd(1)−Pd(2) 153.70(4) Pd(2)−Al(2) 2.4201(15) P(1)−Pd(2)−Pd(1) 151.74(4)

Pd(2)−Al(1) 2.4377(15) Al(1)−Pdaxis−Al(2) 130.7 Pd(1)−P(1) 2.2807 (13) Pd(2)−P(2) 2.2936(13)

a Al−Cp*centroid 1.951 a means value 61

PART 1 Organometallic Synthesis of Transition Metal – Group 13 Complexes

5. References

1. Ganesamoorthy, C.; Loerke, S.; Gemel, C.; Jerabek, P.; Winter, M.; Frenking, G.;

Fischer, R. A., Reductive elimination: a pathway to low-valent aluminium species. Chem.

Commun. 2013, 49, 2858-2860.

2. Dohmeier, C.; Robl, C.; Tacke, M.; Schnöckel, H., The Tetrameric Aluminum(I)

Compound [{Al(η5-C5Me5)}4]. Angew. Chem., Int. Ed. 1991, 30, 564-565.

3. Molon, M.; Gemel, C.; Fischer, R. A., Organogallium- and organozinc-rich palladium and platinum clusters. Dalton Trans. 2014, 43, 3114-3120.

4. Steinke, T.; Gemel, C.; Winter, M.; Fischer, R. A., The Clusters [Ma(ECp*)b] (M=Pd,

Pt; E=Al, Ga, In): Structures, Fluxionality, and Ligand Exchange Reactions. Chem. Eur. J.

2005, 11, 1636-1646.

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

List of Abbreviation

AFM Atomic Force Microscopy

ALD Atomic Layer Deposition

AMD Amidinate

CDI Carbodiimide

Cp Cyclopentadiene

CVD Chemical Vapor Deposition

Cy Cyclohexyl

DTA Differential Thermal Analysis

DSC Differential Scanning Calorimetry

EA Elemental Analysis (CHN)

EDX Energy Dispersive X-ray

EI-MS Electron Impact Mass Spectrometry

Et2O Diethylether

tBu2DAD 1,4-di-tert-butyl-1,3-diazadiene

GI Grazing incidence

GPC Growth per cycle

Guan Guanidinate

Hacac Acetylacetone

HTXRD High Temperature X-ray Diffraction

IR Infra-Red

Me Methyl

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

MOCVD Metalorganic Chemical Vapor Deposition

NMR Nuclear Magnetic Resonance

iPr Isopropyl

PEALD Plasma Enhanced ALD

PVD Physical Vapor Depsoition

RBS Rutherford Backscattering Spectrometry

Rms Root mean square

RT Room Temperature

SEM Scanning Electron Microscopy

tBu tert-butyl

TEM Transmission Electron Microscopy

TGA Thermogravimetric Analysis

Thd 2,2,6,6,-tetramethyl-3,5-heptadione, C11H20O2

THF Tetrahydrofuran

ToF-ERDA Time of Flight Elastic Recoil Detection Analysis

UV Ultraviolet

XPS X-ray Photoelectron Spectroscopy

XRD X-ray Diffraction

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

5. General Introduction

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

5 General Introduction

5.1 General Introduction

Already many examples for thin films possessing functional properties exist in nature, for instance the brilliant colors of the feathers of a peacock or the wings of butterflies caused by the interference of light with extremely thin films covering these. Another example for thin films in daily life are simple household mirrors. Mirrors are nowadays produced by covering a suitable substrate with a reflective coating. The first attempts on making very crude mirrors in this fashion were already reported 2000 years ago in Pliny the Elders book on Natural History.

In Sidon, which is located in nowadays Lebanon, circular cuts of brown glass were coated with molten lead. Since these early reports on the preparation of functional films a lot happened and many methods have been developed to coat substrate materials in order to improve their properties for certain useful applications and alter their mechanic, optical, chemical or electronic functions.

Rusting of steel and iron parts was a big issue in former times, which reduced the stability and longevity of tools and parts in i.e. machinery and buildings drastically. By applying a thin film of zinc to the used parts the iron was protected from outer influences and the reaction to iron oxide (rust) is inhibited. Even in the case of scratching or partly destruction of the zinc layer, still a protective effect is occurring, as the zinc acts as a sacrificial anode, which reacts with corrosive reagents from the surroundings. A method to apply these films was developed by Paul

Jacques Malouin, who coated iron with zinc by dipping it into molten zinc in 1742.

For the preparation of thin films nowadays, two general concepts are of outstanding importance and will be introduced in the following, (1) physical vapor deposition and (2) chemical vapor 67

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

deposition. These methods have a plethora of sub-methods and variations and the following description of methods is by no means complete. Both methods are very powerful to prepare thin films and are conducted at vacuum conditions, and the deposited thin film is transported through the gas phase and deposited on a substrate.

Physical Vapor Deposition The most common methods of physical vapor deposition are sputtering and evaporation techniques. In sputtering techniques, atoms from a metal target are ejected by collisions of high energetic particles, most commonly gas atoms. The ejected metal atoms are travelling from the target to the substrate where they are condensing, forming a thin film. In physical evaporation techniques, the target is heated to very high temperatures at very low pressures, leading to the evaporation of atoms from the target. The evaporated atoms can then react with the substrate and deposit on its surface forming a thin film. The advantages of these physical deposition techniques are the ability to use nearly any kind of target and substrate. The biggest disadvantage is the so called “line-of-sight-coverage” which makes the coverage of complex samples with advanced geometrical features, e.g. protrusions unfeasible

Chemical Vapor Deposition Chemical Vapor deposition summarizes all methods in which a chemical precursor is decomposed and deposited on a surface. Many variations of this method exist, but the general advantages of this method is the uniform coverage of complex shaped substrates. A big drawback of this method is the limitation of precursors, not for every kind of coating suitable precursor molecules have been developed. A very useful variation of this method, which will be the focal point in the following is atomic layer deposition, or short ALD.

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

5.2 Atomic Layer Deposition The fabrication of highly conformal thin films of metal or metal oxides are of eminent importance for the production of modern devices in electronics and nanotechnology. There are various methods to prepare thin films, the most common used techniques are either physical or chemical methods. For instance, The most common methods to prepare cobalt oxide thin films are for example sol-gel processes,1-3 spray pyrolysis,4-7 electrodeposition and physical vapor deposition (PVD).8-12 The precision of these methods is still limited and hence the development of new deposition techniques for these films is necessary. Up to date, the most promising methods, which outperform the afore mentioned by means of film thickness and phase purity are chemical vapor deposition (CVD) and atomic layer deposition (ALD).13 Among these techniques, specifically ALD is a very powerful method to deposit films which possess exceptional conformality and distinct morphology and phase control on the nanometer scale is possible.9-10 Furthermore, the deposition of thin films on complex-shaped substrates is possible by this method.

ALD also called ALE (atomic layer epitaxy) was developed in the 1970s by T. Suntola and J.

Antson to deposit thin films from the chemical vapor phase. The ALD process allowed producing high quality, uniform and conformal thin films over large and complex-shaped areas with precise control of thickness. These characteristic function of ALD has caught an enormous attention from the microdevice and semiconductor industries. Fundamental studies as well as todays application fields of ALD have been excellently reviewed in several reports by T.

Suntola, L. Niinistö, M. Leskelä, M. Ritala.14-20

ALD is considered as an advanced version CVD, because ALD shows two unique features compared to CVD processes. ALD is a thin film deposition technique, which relies on

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

sequential and self-limiting surface reactions in the gas phase. The precursor reacts with the surface of the substrate as well as the surface created by the complementary reactant in a self- limiting and alternating fashion which leads to a discontinuous process while CVD provides a simultaneous process via thermally induced decomposition of precursors on the surface.

Figure 5.1 Schematic illustration of the ALD reaction cycle

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

Unlike precursors used for CVD growth, an ALD precursor should be thermally stable at deposition temperature to guarantee the self-limiting saturation of the surface. (Criteria of ALD precursors will be discussed in detail in chapter 5.3) Full ALD cycles consist of metal containing precursor and the complementary reactant. In the first step, the surface (most likely

OH groups on the surface) is exposed to a pulse of gaseous metal-containing precursor and saturated by the precursor via chemisorption which inhibits the further growth (Figure 5.1a).

Excess precursor and by-products are purged out with an inert gas (e.g. nitrogen) (Figure 5.1b).

The resulting chemisorbed layer from the first step, is then exposed to the second complementary reactant (e.g. H2O, O2, O3 H2, NH3 etc) (Figure 5.1c). This leads to the formation of the desired material (for instance metals, metal oxides or metal nitrides depending on the second reactant) and the excess of reactants and reaction by-products are subsequently purged by inert gas. (Figure 5.1d) Repetition of these steps initiates the film growth restrained by saturative and self-limiting surface reactions leading to uniform, conformal and precisely controlled thickness (simply tuning the applied number of cycles controls film thickness).

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

Figure 5.2 Possible film growth behavior of ALD growth per cycles versus deposition temperature.

To obtain an ideal ALD grown thin film, an optimized deposition temperature (Tdep) is required.

The temperature range where a constant growth rate is achieved is the so-called ALD window

(Figure 5.2).21 At temperatures below the ALD window the precursor either condenses on the surface, or the surface reaction remains incomplete, due to the low temperature. Above the ALD temperatures window, the precursors can decompose on the surface, which will lead to a CVD like film growth or the surface species desorbs again.

The presence of an ALD window and the linearity of the thickness vs number of applied cycles are typical features of the self-limiting character, however, it does not necessarily guarantee the self-limiting growth process will actually take place. Only saturation studies with the growth rate as a function of the precursor pulses finally demonstrate the self-limiting process.

5.3 ALD of metal oxides thin films

5.3.1 Precursors Although atomic layer deposition techniques are widely used to provide very uniform and conformal thin films, suitable and effective precursors remain scarce. Precursors employed for

ALD require several conditions. The criteria of ALD precursors are as follow. 1) Highly reactive towards the functionalities on the surface as well as the surface prepared by the complementary reactant. Furthermore, a complete reaction is desired. 2) Volatile (vapor pressure > 0.1 Torr;)22-23 and thermally stable at growth (> seconds) and sublimation temperatures (> months). Evaporation should occur in a single step without thermal

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

decomposition. 3) Non-reactive and non-corrosive byproduct formation is strongly preferred.

As is often the case, the reactive by-products can damage the film materials by etching or reabsorption onto the reactive sites of the surface, thereby inducing non-uniformity and low growth rates.23

ALD precursors can occur in the solid, liquid or gaseous state, but solid precursors own certain drawbacks as unwanted particles might be transported by the carrier gas and incorporated into the films. This in turn might cause defects in the films. Thus, the liquid or gaseous precursors are preferred particularly in industrial scale application, if those are accessible.

Chronological development of precursors including metal halides, cyclopentadienyls (Cps), alkoxides, -ketonates, alkylamides, trimethylsilanes (TMS), heteroleptic compounds of mixture of theses ligands.

Figure 5.3 shows a list of the common ligands used for ALD precursors. In the following chapters, the ALD precursors properties in detail will be discussed.

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

Figure 5.3 Outlook of ligand system used as precursors for metal oxides ALD. (dmap: dimethylaminopropoxide, DAD: diazadienyl)

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

A. Metal halides

The metal halides with the general formula MXn (X = F, Al, Br or I; n = depends on the oxidation state of metals) have the general advantage of stability at a broad range of temperatures and a high reactivity. However, metal halides produce HX which is a strong acid can damage or etch the surface of the metal/metal oxide thin films and this cause the inhomogeneous film thickness. Moreover, metal halides precursors are not preferred in many occasions in particular in our group due to the resulting halogen contamination.

B. ß-diketonates

Metal ß-diketonates with the general formula of M(thd)n (thd = 2,2,6,6-tetramethyl-3,5- heptanedionyl) are one of the most common precursors used for gas phase depositions and nearly every corresponding transition metal complexes across the periodic table as well as rare earth metals compounds have been explored.24-25 The synthesis is relatively easy and straight forward and very often they are commercially available. Owing to four strong M−O bonds in the structure (low basicity of the ß-diketonate ligand), metal ß-diketonates possess low reactivity towards water. Thus, metal ß-diketonates are rather developed for CVD precursor or

ALD with powerful oxidant such as ozone to combust the ligands. Metal ß-diketonates have relatively low volatility and evaporation late but they have very high thermal stability (>

300 °C). Although metal ß-diketonates are the most intensively used precursor, they cause the major problems concerning the low growth rates (0.2 – 0.4 Å/cycle), high quantity of residue related to the high evaporation temperature due to the high thermal stability.26 [Andrian 167,168]

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

C. Amidinates

The bidendate chelating effects of amidinates improve the thermal stability which is the major drawback of many monodendate alkylamides. Moreover, easily tunable steric crowding in the ligand sphere of the system prevents the oligomerization, which has a huge impact on the volatility of the compounds. In fact, the presence of two substituents (NR2) at the nitrogen atom readily leads to a distinct increase of steric hindrance compared to oxygen based systems

(OR).27-28 Also amidinates have good reactivity towards oxidizing reagent in the ALD process such as water and ozone. Amidinates were intensively investigated which diminish the drawbacks of conventional precursors. Although numerous amidinates have been reported, only a few ALD process are known for the preparation of metal oxides.29 Details of amnidinates synthetic route and their electronic structures will be dicussed in the following chapter in comparison to the guanidinates. (See 1.2.1.7 Guanidinates)

D. Guanidinates

Guanidinates are closely related to the amidinates regarding the tunability of their steric hindrance, sturcture and bidendate chelating effect, however guanidinates possess an extended electronic flexibility since the lone pair electrons of the amido-nitrogen is delocalized in the ligand -system.30-32 Additionally, guanidinate ligands are stronger bases in comparison to the

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

amidinates, which increases the reactivity of the corresponding guanidinato metal complexes.

As an ALD precursor, these features give rise to advantages, in particular from the increase of reactivity towards hydroxyl groups on the surface leading to a high surface coverage and high growth rates.

Amidinate and guanidinate compounds, particularly Group 13 and transition metals are investigated intensively. Synthetic strategies to prepare metal guanidinate complexes include several different routes, most common approaches are summarized in the following.30, 33

(a) Salt metathesis reaction

The principle of salt metathesis reaction is the chemical process of the exchange of cations between a metal halide and an alkali metal (mostly Li or Na) guanidinates. This approach includes a two-step reaction, where an alkali metal guanidinate [Li(NR’}2CNR2]x is created by insertion of carbodiimide into the alkyl lithium amide. Normally this step is a quantitative reaction, [Li(NR’)2CNR2]n is not separated but its solution (most of the case Et2O) is used directly for the next step. In the second step, reaction with metal halides give the metal

n guanidinates with the general formula of [M (NR’)2CNR2]n (n is related to the oxidation state of the metals) and LiCl as a byproduct.30, 34-35 (Scheme 1) This synthetic route is most common way to synthesize the metal guanidinates. Reaction of dialkylamide and N,N’-dialkyl- carbodiimide gives almost quantitative yield of lithiated guanidinato complex, normally metal halides are added in-situ without isolation of lithiated guanidinate.

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

Scheme 5.1 Synthetic scheme of metal guanidinates via salt metathesis reaction. R, R’ = alkyl or aryl group, MXn = metal halides.

(b) Insertion of carbodiimide into a metal−amide bond

It is known that, N,N’-dialkylcarbodiimides cleave the bond between metal and nitrogen in alkyl

32, 36-37 amides (M−NR2) and are inserted into the M−NR2 bonds. A lone pair electron of the nitrogen atom in the dialkyl-carbodiimide attacks the metal center, then the dialkylamido (NR2) ligands from the metal migrates into the carbon of the dialkyl-carbodiime forms an intermediate state where the guanidinate moiety coordinates to the metal center in a monodendate fashion.

Consequently, the other nitrogen atom from the dialkyl-carbodiimide coordinates to the metal center to form the final chelating metal guanidinate complex. (Scheme 2)

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

Scheme 5.2 Mechanistic procedure of the insertion and rearrangement of dialkyl-carbodiimide into a metal-amide bond.37

E. Oxygen sources

In metal oxide thin film process, the oxygen source is needed for oxidation of the metal precursor. H2O, O2 and O3 are by far the most commonly adopted oxygen sources for oxide processes.

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

5.3.2 Transition metal oxides and cobalt

Transition metal oxides are known to exhibit distinctive physical/chemical properties including electric/optical and catalytic/semiconductive properties due to the concurrence of delocalized and localized d-electrons. Notwithstanding, the performance of those properties are highly related to the morphology, structure and impurities of the materials. These properties can be reasoned by the band structure of transition metal oxides. The bulk band structure of transition metal oxides consists of overlapping p-orbitals from the oxygen atoms. These orbitals are highly populated and of comparatively low energy, forming the valence band. Additionally, the d- orbitals of the metals are also overlapping, although they are only sparsely populated thus forming the conduction band. This band is of higher energy than the valence band and the energy difference is called band gap. This band gap prevents the recombination of electron-hole pairs which are separated into the valence and conduction band. Exactly this separation allows the charge transfer onto adsorbed surface species, a requirement for using transition metal oxides in a manifold of catalytic processes.

Owing to its peculiar and versatile physical/chemical properties of cobalt oxides such as electrochemical capacitors, high energy density as Li ion batteries, oxygen evolution reaction

(OER) catalyst, optical gas sensors as well as their abundacy in earth, they have attracted substantial attention from the material scientists. In general, cobalt oxide exists in three different

II III II,III forms, Co O, Co 2O3 and Co 3O4 and the latest stoichiometry is predominantly studied for the applications due to its chemical stability and electrochemical properties.6 The focus in this thesis will lay on the description of cobalt-oxide materials in particular the mixed valent CoII

III Co containing spinel Co3O4. Co3O4 is studied in the field of water splitting system as an

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

effective photoelectrocatalyst. The water splitting system consists two half reactions: namely hydrogen evolution reaction (HER) and oxygen evolution reaction (OER) and the latter reaction is the rate-determining step because it requires the coupling of four electrons and four proton transfers and formation of O−O bond which normally demands high overpotential to initiate

38 the reaction. The Co3O4 unit cell has a spinel structure and direct optical band gap of 1.4-2 eV in the bulk solid state. However, nanostructured Co3O4 shows considerably larger band gap

(ca 2.5 eV).39 A prerequisite for the high activity in OER is the presence of two distinct cobalt

2+ sites in spinel Co3O4. There are tetrahedrally coordinated Co Td sites and octahedrally

3+ 40 coordinated Co Oh sites. It was shown in different studies, that the OER reaction is highly dependent on the nature of the exposed surface sites, which in turn are interacting with adsorbed species. The importance of both sites remains controversially discussed in the literature, even though, both sites are important as they are building up the geometrical environment for the catalytically active site. In a study by Wang et al. the dependence of the OER on both metal sites was probed using an ion substitution technique. They used CoAl2O4 only containing

2+ 3+ Co Td sites as well as ZnCo2O4, which only contains Co Oh sites. From their study using in

2+ operando spectroscopic methods it was revealed that the Co Td site is forming CoOOH species

3+ during the reaction which is an important intermediate in the OER reaction, whilst the Co Oh containing species remained catalytically inactive.40

Photocatalytic reactions often require a harsh alkali environment which result in a corrosion of the layers, limiting the catalysis operation time.

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

6. References

1. Barrera, E.; Viveros, T.; Avila, A.; Quintana, P.; Morales, M.; Batina, N., Cobalt oxide films grown by a dipping sol-gel process. Thin Solid Films 1999, 346, 138-144.

2. Armelao, L.; Barreca, D.; Gross, S.; Martucci, A.; Tieto, M.; Tondello, E., Cobalt oxide- based films: sol–gel synthesis and characterization. J. Non-Cryst. Solids 2001, 293–295, 477-

482.

3. Quinlan, F. T.; Vidu, R.; Predoana, L.; Zaharescu, M.; Gartrner, M.; Groza, J.; Stroeve,

P., Lithium Cobalt Oxide (LiCoO2) Nanocoatings by Sol−Gel Methods. Ind. Eng. Chem. Res.

2004, 43, 2468-2477.

4. Athey, P. R.; Urban, F. K.; Tabet, M. F.; McGahan, W. A., Optical properties of cobalt oxide films deposited by spray pyrolysis. J. Vac. Sci. Technol., A 1996, 14, 685-692.

5. Shinde, V. R.; Mahadik, S. B.; Gujar, T. P.; Lokhande, C. D., Supercapacitive cobalt oxide (Co3O4) thin films by spray pyrolysis. Appl. Surf. Sci. 2006, 252, 7487-7492.

6. Louardi, A.; Rmili, A.; Ouachtari, F.; Bouaoud, A.; Elidrissi, B.; Erguig, H.,

Characterization of cobalt oxide thin films prepared by a facile spray pyrolysis technique using perfume atomizer. J. Alloy. Compd. 2011, 509, 9183-9189.

7. Abbas, T. A.-H.; Slewa, L. H.; Khizir, H. A.; Kakil, S. A., Synthesis of cobalt oxide

(Co3O4) thin films by electrostatic spray pyrolysis technique (ESP). J. Mater. Sci. Mater.

Electron. 2016, 1-7.

8. Meyer, W.; Biedermann, K.; Gubo, M.; Hammer, L.; Heinz, K., Surface structure of polar Co3O4 (111) films grown epitaxially on Ir(100)-(1 × 1). J. Phys. Condens. Matter 2008,

20, 265011.

9. Ngo, T. Q.; Posadas, A.; Seo, H.; Hoang, S.; McDaniel, M. D.; Utess, D.; Triyoso, D.

H.; Buddie Mullins, C.; Demkov, A. A.; Ekerdt, J. G., Atomic layer deposition of photoactive

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

CoO/SrTiO3 and CoO/TiO2 on Si(001) for visible light driven photoelectrochemical water oxidation. J. Appl. Phys. 2013, 114, 084901.

10. Johnson, R. W.; Hultqvist, A.; Bent, S. F., A brief review of atomic layer deposition: from fundamentals to applications. Mater. Today 2014, 17, 236-246.

11. Mazzi, A.; Bazzanella, N.; Orlandi, M.; Edla, R.; Patel, N.; Fernandes, R.; Miotello, A.,

Physical vapor deposition of mixed-metal oxides based on Fe, Co and Ni as water oxidation catalysts. Mater. Sci. Semicond. Process. 2016, 42, Part 1, 155-158.

12. Xu, T.; Schwarz, M.; Werner, K.; Mohr, S.; Amende, M.; Libuda, J., Structure-

Dependent Anchoring of Organic Molecules to Atomically Defined Oxide Surfaces: Phthalic

Acid on Co3O4(111), CoO(100), and CoO(111). Chem. Eur. J. 2016, 22, 5384-5396.

13. Klepper, K. B.; Nilsen, O.; Fjellvåg, H., Growth of thin films of Co3O4 by atomic layer deposition. Thin Solid Films 2007, 515, 7772-7781.

14. Leskelä, M.; Ritala, M., Atomic Layer Deposition Chemistry: Recent Developments and Future Challenges. Angew. Chem. Int. Ed. 2003, 42, 5548-5554.

15. Leskelä, M.; Ritala, M., Rare-earth oxide thin films as gate oxides in MOSFET transistors. J. Solid State Chem. 171, 170-174.

16. Tuomo Suntola, J. H., Atomic Layer Epitaxy. Annual Review of Materials Science 1985,

15, 177-195.

17. Suntola, T., Atomic layer epitaxy. Materials Science Reports 1989, 4, 261-312.

18. Suntola, T., Surface chemistry of materials deposition at atomic layer level. Appl. Surf.

Sci. 1996, 100, 391-398.

19. Niinistö, L., Atomic layer epitaxy. Current Opinion in Solid State and Materials Science

1998, 3, 147-152.

20. Niinistö, L.; Päiväsaari, J.; Niinistö, J.; Putkonen, M.; Nieminen, M., Feature Article:

Advanced electronic and optoelectronic materials by Atomic Layer Deposition: An overview 83

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

with special emphasis on recent progress in processing of high-k dielectrics and other oxide materials. physica status solidi (a) 2004, 201, 1375-1375.

21. Adelmann, C.; Delabie, A.; Schepers, B.; Rodriguez, L. N. J.; Franquet, A.; Conard, T.;

Opsomer, K.; Vaesen, I.; Moussa, A.; Pourtois, G.; Pierloot, K.; Caymax, M.; Van Elshocht,

S., Atomic Layer Deposition of Tantalum Oxide and Tantalum Silicate from Chloride

Precursors. Chem. Vap. Deposition 2012, 18, 225-238.

22. Mikko Ritala, M. L., Handbook of Thin Films Materials: Deposition and Processing of

Thin Films. In Handbook of Thin Films, Nalwa, H. S., Ed. Academic Press: Burlington, 2002;

Vol. 1.

23. Ritala, M.; Niinisto, J., Chapter 4 Atomic Layer Deposition. In Chemical Vapour

Deposition: Precursors, Processes and Applications, The Royal Society of Chemistry: 2009; pp 158-206.

24. Tiitta, M.; Niinistou, L., Volatile Metal β-Diketonates: ALE and CVD precursors for electroluminescent device thin films. Chem. Vap. Deposition 1997, 3, 167-182.

25. Binnemans, K., Chapter 225 - Rare-earth beta-diketonates. In Handbook on the Physics and Chemistry of Rare Earths, Karl A. Gschneidner, J.-C. G. B.; Vitalij, K. P., Eds. Elsevier:

2005; Vol. Volume 35, pp 107-272.

26. Luten, H. A.; Rees, W. S.; Goedken, V. L., Preparation and structural characterization of, and chemical vapor deposition studies with, certain yttrium tris(β-diketonate) compounds.

Chem. Vap. Deposition 1996, 2, 149-161.

27. Lim, B. S.; Rahtu, A.; Park, J.-S.; Gordon, R. G., Synthesis and Characterization of

Volatile, Thermally Stable, Reactive Transition Metal Amidinates. Inorg. Chem. 2003, 42,

7951-7958.

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

28. Milanov, A. P.; Thiede, T. B.; Devi, A.; Fischer, R. A., Homoleptic Gadolinium

Guanidinate: A Single Source Precursor for Metal−Organic Chemical Vapor Deposition of

Gadolinium Nitride Thin Films. J. Am. Chem. Soc. 2009, 131, 17062-17063.

29. Lim, B. S.; Rahtu, A.; Gordon, R. G., Atomic layer deposition of transition metals. Nat.

Mater. 2003, 2, 749-754.

30. Bailey, P. J.; Pace, S., The coordination chemistry of guanidines and guanidinates.

Coord. Chem. Rev. 2001, 214, 91-141.

31. Coyle, J. P.; Johnson, P. A.; DiLabio, G. A.; Barry, S. T.; Müller, J., Gas-Phase

Thermolysis of a Guanidinate Precursor of Copper Studied by Matrix Isolation, Time-of-Flight

Mass Spectrometry, and Computational Chemistry. Inorg. Chem. 2010, 49, 2844-2850.

32. Barry, S. T., Amidinates, guanidinates and iminopyrrolidinates: Understanding precursor thermolysis to design a better ligand. Coord. Chem. Rev. 2013, 257, 3192-3201.

33. Musashi, Y.; Sakaki, S., Insertion of carbon dioxide into a rhodium(III)-hydride bond: a theoretical study [dagger]. Journal of the Chemical Society, Dalton Transactions 1998, 577-

584.

34. Edelmann, F. T., Chapter 3 - Advances in the Coordination Chemistry of Amidinate and

Guanidinate Ligands. In Adv. Organomet. Chem., Anthony, F. H.; Mark, J. F., Eds. Academic

Press: 2008; Vol. Volume 57, pp 183-352.

35. Edelmann, F. T., Lanthanide amidinates and guanidinates: from laboratory curiosities to efficient homogeneous catalysts and precursors for rare-earth oxide thin films. Chem. Soc.

Rev. 2009, 38, 2253-2268.

36. Chang, C.-C.; Hsiung, C.-S.; Su, H.-L.; Srinivas, B.; Chiang, M. Y.; Lee, G.-H.; Wang,

Y., Carbodiimide Insertion into Organoaluminum Compounds and Thermal Rearrangement of the Products. Organometallics 1998, 17, 1595-1601.

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

37. Rowley, C. N.; DiLabio, G. A.; Barry, S. T., Theoretical and Synthetic Investigations of Carbodiimide Insertions into Al−CH3 and Al−N(CH3)2 Bonds. Inorg. Chem. 2005, 44,

1983-1991.

38. Koper, M. T. M., Thermodynamic theory of multi-electron transfer reactions:

Implications for electrocatalysis. J. Electroanal. Chem. 2011, 660, 254-260.

39. Ramakrishnan, V.; Kim, H.; Park, J.; Yang, B., Cobalt oxide nanoparticles on TiO2 nanorod/FTO as a photoanode with enhanced visible light sensitization. RSC Adv. 2016, 6,

9789-9795.

40. Wang, H.-Y.; Hung, S.-F.; Chen, H.-Y.; Chan, T.-S.; Chen, H. M.; Liu, B., In Operando

Identification of Geometrical-Site-Dependent Water Oxidation Activity of Spinel Co3O4. J.

Am. Chem. Soc. 2016, 138, 36-39.

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

6. Low temperature atomic layer deposition of cobalt oxide as an effective catalyst layer for photoelectrochemical water splitting devices

Jiyeon Kim, Tomi Iivonen, Jani Hämäläinen, Marianna Kemell, Kristoffer Meinander, Kenichiro Mizohata, Lidong Wang, Radim Beranek, Markku Leskelä, Anjana Devi

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

6. Low temperature atomic layer deposition of cobalt oxide as an effective catalyst layer for photoelectrochemical water splitting devices2

6.1 Abstract We have developed a low-temperature atomic layer deposition (ALD) process for depositing

tBu2 crystalline and phase pure spinel cobalt oxide (Co3O4) films at 120 °C using Co( DAD)2 and ozone as co-reagent. X-ray diffraction, UV-Vis spectroscopy, atomic force microscopy, field emission scanning electron microscopy, X-ray photoelectron spectroscopy and time-of-flight elastic recoil detection analysis were performed to characterize the structure and properties of the films. Full characterization of the film demonstrates that as-deposited Co3O4 films are crystalline with low amount of impurities (<5 % hydrogen) despite low deposition temperatures.

Deposition of ultra-thin Co3O4 (< 1.1 nm) layers onto thin TiO2 films (100 nm) resulted in 30 % improvement of photocurrent compared to pristine TiO2 films, demonstrating the applicability of the developed ALD process for deposition of effective, conformal electrocatalytic layers in photoelectrochemical water-splitting devices.

2 This chapter has been in part submitted for publication and is currently under revision. 88

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

6.2 Introduction

The interest in nanostructured cobal(II,III) oxide has been growing enormously over the past few years, due to its gigantic potential for a variety of applications, in such areas as

1-5 electrochemistry, catalysis, sensing, batteries and energy storage. Co3O4 proved to be an excellent electrocatalyst for water oxidation and is hence of outstanding importance, since the water oxidation reaction remains the kinetic bottleneck in water splitting systems. Because of that, CO3O4 catalysts are among the most favorable materials in this field, which is directly connected to a sustainable and low-cost energy supply. 6-13 Electrocatalyst activity and their use in thin-film devices (e.g. for chemical water splitting) generally depends on a variety of factors, such as composition, morphology, thickness and optical properties.8, 14-17 The most common methods to prepare cobalt oxide thin films are for example sol-gel processes,18-20 spray pyrolysis,21-24 electrodeposition14 and physical vapor deposition (PVD).25-28 The precision of these methods is still limited and hence the development of new deposition techniques for these films is necessary. Up to date, the most promising methods, which outperform the afore mentioned by means of film thickness and phase purity are chemical vapor deposition (CVD) and atomic layer deposition (ALD).29 Among these techniques, specifically ALD is a very powerful method to deposit films which possess exceptional conformality and distinct morphology and phase control on the nanometer scale is possible. 26, 30 Furthermore, the deposition of thin films on complex-shaped substrates is possible by this method. The before mentioned benefits of ALD are in particular in the light of applicability of Co3O4 in devices of marked importance. 31-33 These applications include for example the deposition of light adsorbers, protective layers or catalysts film for photoelectrochemical water splitting devices.

34-39 Noticeably, only very few reports on the preparation of cobalt oxide exist, eventhough for the deposition of other metal oxides ALD processes are quite common. Among the few 89

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

described ALD based routes towards cobalt oxides are for instance the deposition of cobalt

i i oxide from the cobalt amidinates precursor Co( PrNCMeN Pr)2 in a water assisted ALD

30, 40 11, 41 process. Other examples are the use of CoCp2 (Cp = cyclopentadienyl), CCTBA

42 (dicobalthexacarbonyl-tert-butylacetylene), Co(thd)2 (thd = 2,2,6,6-tetramethylheptan-3,5-

29 32, 43 dionate) and Co(acac)2 (acac = acetylacetonate) combined with either O2 plasma or O3 as the oxygen source. Drawbacks of these processes are however the high deposition temperatures in combination with low film growth rates and impure thin films. 29, 32 Therefore, the development of new cobalt precursors made from bidentate chelating ligands, for example amidinates was an interesting alternative.44 By the bidentate binding of the amidinate counteracts effects observed in monodentate compounds (i.e. amido or imido compounds), particularly low thermal stability. At the same time, the all nitrogen coordinated amidinates possess low deposition temperatures caused by the lower bond strength of M-N compared to

M-O.45 Remarkably, in the group of Winter, a amidinate build up from redox non-innocent

tBu2 46 linkers was prepared, namely bis(1,4-di-tert-butyl-1,3-diazadiene) cobalt (II) Co( DAD)2.

This kind of compound bear much higher redox activity under mild conditions. In a recent report, a low temperature ALD process for the preparation of metallic Co0 films using formic

tBu2 5 acid as a reducing agent and the before mentioned Co( DAD)2 was described.

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

6.3 Results and Discussion

tBu2 2.3.1 Thermal properties of precursor [Co( DAD)2]

This chapter describes the development of a low temperature ALD process used for the

tBu2 fabrication of phase pure Co3O4 thin films. The process employs [Co( DAD)2] as cobalt and ozone as oxygen sources respectively. Additionally, it will be demonstrated that the prepared films can be effectively used for the preparation of Co3O4/TiO2 electrocatalysts. TiO2 thin film have gained massive attention and are utilized as protective layers in novel water splitting devices.47-50

tBu2 Scheme 6.1: Synthetic route towards [Co( DAD)2]

tBu2 Scheme 6.1 depicts the synthesis route towards [Co( DAD)2], which was prepared in analogy to the known procedure from the literature (Details Supporting Information). The precursor shows very promising thermal properties, TGA/DTA and isothermal TG studies are presented

tBu2 in Figure 6.1. Thermogravimetric studies of [Co( DAD)2] show a single step weight loss with an onset at 160 °C. The residual mass is very low and accounts to 4 % of the initial mass.

Differential Thermal Analysis (DTA) reveals a melting point of 175 °C, which is in line with the previously reported value (174-175 °C).46 The conducted isothermal studies show a linear weight loss, and constant evaporation rate over a time range of 180 minutes. This makes them

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

suitable candidates as ALD precursors and guarantees sustainable precursor transport into the

ALD reactor (Figure 1b).

tBu2 Figure 6.1 Thermal properties of Co( DAD)2, TGA/DTA (a) and isothermal TG studies (b) measured at 100 – 120 °C.

The average of the slope of the isothermal studies was used to determine the evaporation rates of the precursors at different temperatures. The obtained values are 2.52 µg/min (100 °C), 5.17

µg/min (110 °C) and 11.14 µg/min (120 °C). 100 °C was choosen as the evaporation temperature for the deposition experiments. In order to maintain a thermal gradient between precursor source and substrate chamber, the evaporation temperature was set to 95 °C.

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

6.3.2 Film Deposition

In order to verify the ALD growth of Co3O4, the deposition temperature, the precursor pulse

tBu2 lengths of Co( DAD)2/O3 and the number of deposition cycles were varied (Figure 6.1). In

Figure 6.1a the growth rate is depicted as a function of temperature in the range 100 – 150 °C.

The temperature window with a constant growth rate is very narrow (110 − 120°C), which has been observed for other ozone assisted metal oxide ALD proceses at high temperatures, due to decomposition of ozone on the metal oxide surface.51-52 At deposition temperatures higher than

120 °C full coverage was not feasible over the whole substrate, due to the configuration of the used F-120 reactor. Using 40 mm Si trenches, fully conformal Co3O4 deposition was also not achieved, although 6 s pulses and 10 s purge times were used for both precursors. An explanation for not reaching conformal depositions in the trenches are the ozone decomposition of Co3O4 and the large surface area of the trench substrate. Hence, 120 °C were chosen as the ideal deposition temperature for all further experiments (See Appendix, Figure 6.9).

After 2 second pulses of Co precursors followed by 2 sec nitrogen purges saturation of film growth was observed (Figure 6.2b). For O3 saturation was also found to occur after 2 second ozone purges. These findings are proof for the self-limiting character of the deposition at a temperature of 120 °C. Noticeably, the linear dependence of the film thickness on the number of deposition cycles further underlines the occurrence of a typical ALD process. Astonishingly, the determined growth rate of 1.1 Å/cycle is five times higher than the previously reported

29 Co3O4 ALD deposition rate (0.25 Å/cycle, CoCp2/O3 ALD Window 110-300 °C). This rate is also slightly higher as recently found results, which used a different cobalt precursor

53 (diazadienyl Co precursor) and O3 as well as O2 as oxygen sources.

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

Figure 6.2 Growth per cycle (GPC) as a function of (a) deposition temperature, (b) tBu2 [Co( DAD)2] pulse length, (c) ozone pulse length. (d, black line) Film thickness as a function of number of applied cycles. The R2 value for a goodness of fit is 0.99967. (d, blue line) Average rms roughness as a function of number of applied cycles. Films were grown with 300 cycles tBu2 using 2 s pulses and purges for both [Co( DAD)2] and ozone and deposition temperature 120 °C unless otherwise noted.

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

6.3.3 Film Characterization A. Structure and Morphology

In order to determine the structure and the composition of the prepared Cobalt oxide thin films,

X-ray diffraction measurements, FESEM and atomic force microscopy measurements were performed.

The X-ray diffraction measurements were performed in grazing incidence mode and are shown in Figure 3. The films deposited at 120 °C are of polycrystalline nature and the reflections found can be attributed to spinel cobalt lattice planes, namely the (111), (220), (311), (222),

(400), (511) and (440) planes.54 Already for the as-deposited films high intensities indicating high crystallinity can be observed, which is in strong contrast to previously reported, much less

29, 53, 55 crystalline Co3O4 thin films prepared by ALD. High Temperature XRD showed in air and nitrogen atmosphere an increase in crystallinity in the range from room temperature to 700

°C. However, in oxygen free environment a reduction of the prepared Co3O4 thin films to CoO was observed. This was indicated by new Bragg reflections fitting to the (111), (200) and (220) lattice planes of periclase CoO. 56

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

Figure 6.3 HTXRD patterns of Co3O4 films deposited over 1000 cycles at 120 °C in a) air b) nitrogen. The measurement temperature is shown under each diffractogram.

Via FSEM and AFM the surface morphology of the as-deposited films was analyzed. In Figure

6.4a and b the SEM and AFM images of the Co3O4 films deposited at 120 °C over 1000 cycles are displayed. It is visible in both figures that homogeneous, well defined nano crystallites were deposited without the necessity of any post annealing processes. The grain size of the nano crystallites were in a range of 75 – 90 nm. In Figure 6.4c roughness studies with increasing number of deposition cycles are depicted. Noticeably, the Surface average Rq values rise linearly with the number of applied deposition cycles, with values of 0.81 nm, 1.1nm, 2.2 nm and 3.3 nm for 150, 300, 600 and 1000 deposition cycles respectively (dependency of roughness on number of cycles depicted in Figure 6.4c).

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

Figure 6.4 (a) Plane-view SEM and (b) representative AFM images of Co3O4 thin film deposited over 1000 cycles at 120 °C. (c) AFM topographs of Co3O4 thin films deposited over different cycles at 120 °C and the corresponding Rq values.

B. Film Composition

To confirm the chemical composition and the surface properties of the Co3O4 films prepared at 120 °C over 1000 cycles, XPS and ToF-ERDA measurements were conducted. The XPS measurements were performed in order to analyze the chemical states of surface cobalt and oxygen in the prepared samples. In Figure 6.5 the survey spectrum of CO3O4 deposited on

Si(100) is depicted. Figure 6.5a shows the core level spectrum of Co 2p, which indicates spin orbit splitting into 2p1/2 and 2p3/2. The plateau shake-up peaks confirm the presence of a mixed Co2+/3+ oxidation state. From the satellite line tit is possible to make a distinction between Co2+ and Co3+ states. Pure Co2+ shows a projecting shake-up peak within the areas of

3+ 57 786 and 790 eV. For Co this shake-up peak can be found at 790 eV. Because the 2p1/2 and

2p3/2 components contain identical chemical information, the 2p3/2 peak was chosen for curve fitting and qualitative analysis, due to higher intensity. The binding energy peaks at 779.6 and

781.1 eV strongly suggest the presence of Co3+ and Co2+ species within the sample and the shake up peaks belonging to these peaks can be found at 786.0 and 789.7 eV. 57-59 At 782.7 eV a peak that can be ascribed to Co(OH)2 is identified, which suggests the presence of this species at the surface. 57 The O1s area also shows two peaks, one located at 531.3 eV and

529.6 eV. The first one is attributed to the lattice oxygen present in spinel Co3O4 and the latter arises from hydroxyl groups present in Co(OH)2.

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

Figure 6.5 XPS Co 2p3/2 and O 1s photoelectron spectra of Co3O4 thin film (thickness 110 nm) deposited at 120 °C.

Depths profiles and Co:O ratios were provided by ToF-ERDA. One of the marked benefits of this methods is its ability to quantitavively detect light elements including hydrogen which cannot be detected using Rutherford backscattering Spectrometery (RBS), making it a powerful method in combination with XPS measurements. The measurements proved that the films contain a substantial hydrogen content, although it is below 5%. Interestingly, the amount of hydrogen is highest close to the surface, which can be explained by the presence of surface hydroxyls which were also found via XPS (Figure 6.6 and Table 6.1).

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

Table 6.1 (left) and Figure 6.6 (right) ToF-ERDA depth profile of a Co3O4 film deposited over 1000 cycles at 120 °C

The ToF-ERDA measurement revealed only low impurities, with 1.6 atomic percent of carbon in the sample. Furthermore, also the nitrogen impurity is neglectable with only 0.7 atomic percent. The found Co:O ratio amounts to 0.7 and is in line with the expected ratio for spinel

Co3O4.

2.3.4 Film Photoelectrochemical Properties

Cobalt oxide has proven to be an excellent electrocatalyst for the oxygen evolution reaction, which is a important process in all photocatalytic water splitting devices. 60 Remarkably, the disordered heterogenite structure of CoO(OH) has shown the highest activity among cobalt- based electrocatalysts. 14, 16, 61-62 The simultaneous presence of hydroxylated Co2+ and Co3+ at the surface of the thin films as previously discussed, strongly suggests that the highly active

CoO(OH) species is formed under electrocatalytic conditions.16 To confirm the applicability of 99

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

the prepared cobalt oxide thin films for water splitting, ultrathin films (1.1 nm) of the material were deposited on 100 nm thick TiO2 layers on FTO glass substrate. For future photoelectrochemical water splitting systems these composite TiO2/Co3O4 composites are of very high interest. This is because currently TiO2 film are in the focus of research as highly effective protection layers for low-gap semiconductors including Si, GaAs, Gap and CdTe.

These semiconductors usually undergo severe photocorrosion.47-50

The photoaction spectra (Figure 6.7) as well as the spectra under simulated solar light (Figure

6.8) show a striking increase of the photocurrent response of 30% in comparison to pristine

TiO2. Notably, the improvement excludes the presence of any harmful surface defect states in

TiO2 caused by the ALD-grown Co3O4. These effects could lead to enhanced surface recombination. It is noticeable that the charge separation is enhanced, because the Co3O4 effectively extracts the holes from TiO2. This finally leads to the observed dramatic increase of

63 photocurrent. Additionally, another advantage of Co3O4 is its role in the passivation of surface states in TiO2, which has been recently reported for some photoanodes based on Fe2O3 or

64-66 BiVO4. The photoholes in TiO2 have a highly positive nature and the presence of a co- catalyst to induce full water oxidation under photoelectrochemical conditions is not necessary.

Therefore, the increased oxygen evolution rate can be compared to an enhancement of photocurrent without a severe change in the faradaic oxygen evolution efficiency. Anyhow, these results showcase that the developed low temperature ALD process is useful for the preparation of electrocatalytic cobalt oxide layers on top of TiO2 layers. These TiO2 layers are of crucial interest for the preparation of photoanodes for tandem water-splitting devices, which are currently very susceptible towards photocorossion.47-50

100

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

Figure 6.7 Photocurrent transients under intermittent irradiation at different wavelengths (a) and the corresponding photoaction spectrum (b) recorded in KOH (0.1 M) at 1.48 V vs. RHE at TiO2 (100 nm) and TiO2/Co3O4 (10 cycles) photoanodes.

Figure 6.8 Photoelectrocatalytic properties of TiO2 (100 nm) and TiO2/Co3O4 (10 cycles) photoanodes: (a) photocurrents (a) and oxygen evolution (b) during 1 hour irradiation by simulated sunlight (Xenon lamp, AM1.5 filter, 100 mW/cm2) in borate buffer solution (0.1 M; pH 9.2) at 1.25 V vs. RHE

101

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

6.4 Conclusion and Outlook

This chapter nicely demonstrates the possibility to deposit spinel cobalt oxide via ALD by using

tBu2 an ozone assisted process with [Co( DAD)2]as a precursor. The process was conducted at low temperatures (120 °C) and the obtained films are conformal, crystalline, phase pure and only contain extremely low carbon and hydrogen contamination. A Post annealing treatment to increase these films properties is not necessary. Furthermore, by the ALD process ultrathin

Co3O4 electrocatalyst films were prepared on top of TiO2 films, which lead to a striking improvement of photoconversion efficiency. Due to the conformal nature of the ALD grown films and the importance of TiO2 protective layers for photoelectrodes these finding will be of very high relevance for the fabrication of a variety of solar water splitting devices.

6.5 Appindex

Thickness(nm) Top Middle Bottom

114 101 84

122 93 85

111 93 77

119 93 71

102

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

130 89 77

138 92 71

Average

thickness (nm) 122 94 78

Figure 6.9 SEM image of Si trench with an aspect ratio of 20:1 (depth 40 m, width 2 m).

Co3O4 film has been deposited over 1000 cycles at 120 °C using pulse lengths of 6 seconds

tBu2 for both Co( DAD)2 and ozone. The purge length for both precursors was 10 seconds.

Fully conformal Co3O4 films could not be deposited on 40 m Si trenches (aspect ratio of

1:20) despite using 6 s pulse and 10 s purge times of both precursors. A conceivable explanation of for not achieving full conformality in the trench structure is the combined effect of ozone decomposing effect of Co3O4 and high surface area of the substrate.

103

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

Figure 6.10 XPS survey spectrum of Co3O4 thin film deposited over 1000 cycles at 120 °C.

Figure 6.10 displays the XPS survey scan of Co3O4 that is dominated by cobalt and oxygen photopeaks. No Ar+ sputtering experiment was performed.

104

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

Figure 6.10 UV-Vis transmission spectrum and corresponding Tauc plot of Co3O4 thin film deposited over 1000 cycles at 120 °C.

UV-Vis spectroscopy was performed on the as-deposited Co3O4 film grown on soda-lime glass to determine the optical bandgap of Co3O4 thin films via Tauc plot. Obtained data indicated a direct band gap value of 2.02 eV.

105

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

Figure 6.11 UV-Vis transmission spectra of TiO2 and Co3O4

UV-Vis spectroscopy was performed on pristine TiO2 (black line) and TiO2/Co3O4 (red line) grown on FTO glass. As can be seen in Figure S5, the shift between pristine TiO2 and

TiO2/Co3O4 was not observed.

106

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

Figure 6.12 XRD patterns of TiO2 deposited at 275 °C over 2000 cycles on FTO glass

XRD reflections of TiO2 films deposited at 275 °C on FTO show mixed phases of anatase/rutile dominanted by the anatase phase.72 (Figure S4)

6.5 Experimental Section

107

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

6.5.1 Precursor synthesis

tBu2 46 [Co( DAD)2] was synthesized modifying the literature reported procedure. All manipulations were carried out in an atmosphere of argon purified by Cu catalysts and molecular sieves using standard Schlenk and glove-box techniques. All solvents were dried using an MBraun Solvent Purification System. The crude mixture after the synthesis was sublimed at 60 °C in vacuo (10-3 mbar), then the temperature was gradually increased up to 90

°C to remove excess amount of free ligands. Deep dark- green crystals were collected on a cooling finger by slow sublimation (100 °C, 10-3 mbar). The purity and the thermal properties were analyzed using elemental analysis, mass spectrometry and thermogravimetric analysis

(TGA, EXSTAR 6000 TG/DTA 6200, Seiko Instruments Inc.). The measurements were performed under inert gas atmosphere using nitrogen (99.999 %) at ambient pressure.

Approximately 10 mg of samples were filled in aluminum crucibles with a circular opening.

The heating rate was 5 °C/min and a nitrogen flow of 300 ml/min was used. For isothermal TG studies, the sample was heated with a rate of 10 °C/min until the desired temperature was reached, then mass loss was measured for 180 min at constant temperature.

+ + + + + EI-MS [m/z]: 395.3 [M] , 380.2 [M] -CH3, 338.2 [M] -C(CH3)3, 227.1 [M] -L, 212.1 [M] -L-

CH3

108

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

6.5.2 Film deposition

Cobalt oxide thin ﬁlms were deposited using a hot-wall ﬂow-type ALD reactor (ASM

Microchemistry F-120) operated under a nitrogen atmosphere of approximately 5 mbar.

Nitrogen (99.999 %) was used as both carrier and purging gas. Si (100) and soda lime glass cut

2 tBu to 5 × 5 cm pieces were used as substrates. Co( DAD)2 was sublimed from an open glass boat held inside the reactor at 95 − 100 °C. Ozone was produced from oxygen (99.999 %) using a Weeco Ozomatic Modulator 4 HC Lab Ozone generator (ozone concentration of approximately 100 g/Nm3) and introduced to the reactor using needle and solenoid valves. The deposition temperature was varied from 100 – 150 °C. Thin films for photoelectrochemical

2 studies were prepared as follows: 100 nm TiO2 thin films were deposited on 1.5 x 2.0 cm FTO

67 substrates (Solems TEC7) by applying the TiOMe4 / H2O ALD process at 275 °C. The

TiO2/Co3O4 structure was obtained by applying 10 deposition cycles of the Co3O4 process at a temperature of 120 °C.

6.5.3 Film characterization

All characterizations were conducted using films deposited over 300 cycles (film thickness of approximately 30 nm) on Si (100) substrates unless otherwise noted. All thickness measurements related to saturation studies and the effect of deposition temperature were done on Si (100) with an in situ aluminum oxide layer deposited over 100 cycles using trimethylaluminum (TMA) and ozone (Al2O3 film thickness of approximately 10 nm). The purpose of the Al2O3 layer was to twofold: I) To produce a reproducible, hydroxylated starting surface68 for each cobalt oxide film deposition and II) to cover and thus passivate cobalt oxide that would grow on the ALD reactor walls and eventually cause ozone decomposition.52

109

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

Film thickness was measured with energy dispersive X-ray spectrometry (EDX). The EDX spectra were collected using an Oxford INCA 350 microanalysis system connected to a Hitachi

S-4800 field emission scanning electron microscope (FESEM). Film thicknesses were calculated from the EDX spectra using the GMRfilm software and assuming bulk density values

-3 69 of 6.1 g cm for the Co3O4 films.

All X-ray diffraction measurements were performed in the grazing incidence (GI) geometry using an incident angle of 1 °. Film crystal structure was identified GI-(XRD) patterns obtained with a PANalytical X’Pert Pro MPD diffractometer. The high temperature XRD (HTXRD) measurements were done using an Anton-Paar HTK1200N oven in both air and nitrogen atmospheres. The nitrogen gas (99.999%) was purified in situ prior to the experiment in an

Entegris 35KF-I-4R gas purification system.

Film morphology was studied using both FESEM and atomic force microscopy (AFM). AFM images were captured in tapping mode using a Veeco Multimode V tool equipped with a

Nanoscope V controller. Film roughness was calculated as average root mean square (Rq) values from 2 x 2 µm2 images obtained using Si probes (RTESP, Bruker).

X-ray photoelectron spectra (XPS) were obtained using an Omicron ARGUS spectrometer operated at a pass energy of 20 eV. Samples were illuminated with X-rays emitted from a standard Mg source (K alpha line) at a photon energy of 1253.6 eV. No sputtering was done on the films studied. Binding energies were calibrated using the C 1s peak of ambient hydrocarbons

(284.8 eV). Peak fitting was done using the CasaXPS software package.

The time-of-flight elastic recoil detection analysis (ToF-ERDA) measurements were done on films deposited on Si (100) over 1000 cycles using a 50 MeV 127I7+ ion beam in a setup described in full elsewhere.70

110

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

UV-Vis transmittance measurements were performed on films deposited on soda lime glass substrates using a Hitachi U2000 spectrophotometer.

6.5.4 Photoelectrochemistry

Photocurrent measurements were carried out using a three electrode setup with a platinum counter electrode and a Ag/AgCl (3M KCl) reference electrode using a SP-300 BioLogic potentiostat. The photoelectrodes were pressed against an O-ring of the cell leaving an irradiated area of 0.5 cm2. The electrodes were irradiated from the back side (through the FTO glass). The monochromatic wavelength- resolved photocurrent measurements were performed using a tunable monochromatic light source (Instytut Fotonowy) provided with a 150 W Xenon lamp and a grating monochromator with a bandwidth of 10 nm and SP-300 BioLogic potentiostat. The monochromatic intensities between 300 nm and 800 nm were in the range of 0.16–2.48 mW/cm2. Appropriate cut-off filters were used in order to eliminate the second-order diffraction radiation. The oxygen evolution was monitored by an OxySense 325i oxygen analyzer in a two-compartment cell, and the electrodes were irradiated by a 150 W Xenon lamp (LOT Oriel) equipped with a KG-3 (Schott) heat-absorbing filter and an AM 1.5 filter. Prior to experiments the electrolyte solutions were purged with argon for 30 minutes.

111

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

6.6 References

1. Švegl, F.; Orel, B.; Hutchins, M. G.; Kalcher, K., Structural and Spectroelectrochemical

Investigations of Sol‐Gel Derived Electrochromic Spinel Co3O4 Films. J. Electrochem. Soc.

1996, 143, 1532-1539.

2. Ando, M.; Kobayashi, T.; Iijima, S.; Haruta, M., Optical recognition of CO and H2 by use of gas-sensitive Au-Co3O4 composite films. J. Mater. Chem. 1997, 7, 1779-1783.

3. Li, W. Y.; Xu, L. N.; Chen, J., Co3O4 Nanomaterials in Lithium-Ion Batteries and Gas

Sensors. Adv. Funct. Mater. 2005, 15, 851-857.

4. Bekermann, D.; Gasparotto, A.; Barreca, D.; Maccato, C.; Comini, E.; Sada, C.;

Sberveglieri, G.; Devi, A.; Fischer, R. A., Co3O4/ZnO Nanocomposites: From Plasma Synthesis to Gas Sensing Applications. ACS Appl. Mater. Interfaces 2012, 4, 928-934.

5. Klesko, J. P.; Kerrigan, M. M.; Winter, C. H., Low Temperature Thermal Atomic Layer

Deposition of Cobalt Metal Films. Chem. Mater. 2016, 28, 700-703.

6. Lewis, N. S.; Nocera, D. G., Powering the planet: Chemical challenges in solar energy utilization. Proc. Natl. Acad. Sci. 2006, 103, 15729-15735.

7. Dau, H.; Limberg, C.; Reier, T.; Risch, M.; Roggan, S.; Strasser, P., The Mechanism of

Water Oxidation: From Electrolysis via Homogeneous to Biological Catalysis. ChemCatChem

2010, 2, 724-761.

8. Trotochaud, L.; Mills, T. J.; Boettcher, S. W., An Optocatalytic Model for

Semiconductor–Catalyst Water-Splitting Photoelectrodes Based on In Situ Optical

Measurements on Operational Catalysts. J. Phys. Chem. Lett. 2013, 4, 931-935.

9. Yang, J.; Wang, D.; Han, H.; Li, C., Roles of Cocatalysts in Photocatalysis and

Photoelectrocatalysis. Acc. Chem. Res. 2013, 46, 1900-1909. 112

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

10. Peter, L. M., Photoelectrochemical Water Splitting. A Status Assessment.

Electroanalysis 2015, 27, 864-871.

11. Ager, J. W.; Shaner, M. R.; Walczak, K. A.; Sharp, I. D.; Ardo, S., Experimental demonstrations of spontaneous, solar-driven photoelectrochemical water splitting. Energy

Environ. Sci. 2015, 8, 2811-2824.

12. Zhou, X.; Shen, X.; Xia, Z.; Zhang, Z.; Li, J.; Ma, Y.; Qu, Y., Hollow Fluffy Co3O4

Cages as Efficient Electroactive Materials for Supercapacitors and Oxygen Evolution Reaction.

ACS Appl. Mater. Interfaces 2015, 7, 20322-20331.

13. Chua, C. S.; Ansovini, D.; Lee, C. J. J.; Teng, Y. T.; Ong, L. T.; Chi, D.; Hor, T. S. A.;

Raja, R.; Lim, Y.-F., The effect of crystallinity on photocatalytic performance of Co3O4 water- splitting cocatalysts. Phys. Chem. Chem. Phys. 2016, 18, 5172-5178.

14. Kanan, M. W.; Nocera, D. G., In Situ Formation of an Oxygen-Evolving Catalyst in

Neutral Water Containing Phosphate and Co2+. Science 2008, 321, 1072-1075.

15. Lin, F.; Boettcher, S. W., Adaptive semiconductor/electrocatalyst junctions in water- splitting photoanodes. Nat. Mater. 2014, 13, 81-86.

16. Bergmann, A.; Martinez-Moreno, E.; Teschner, D.; Chernev, P.; Gliech, M.; de Araújo,

J. F.; Reier, T.; Dau, H.; Strasser, P., Reversible amorphization and the catalytically active state of crystalline Co3O4 during oxygen evolution. Nat. Commun. 2015, 6, 8625.

17. Nellist, M. R.; Laskowski, F. A. L.; Lin, F.; Mills, T. J.; Boettcher, S. W.,

Semiconductor–Electrocatalyst Interfaces: Theory, Experiment, and Applications in

Photoelectrochemical Water Splitting. Acc. Chem. Res. 2016, 49, 733-740.

18. Barrera, E.; Viveros, T.; Avila, A.; Quintana, P.; Morales, M.; Batina, N., Cobalt oxide films grown by a dipping sol-gel process. Thin Solid Films 1999, 346, 138-144.

113

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

19. Armelao, L.; Barreca, D.; Gross, S.; Martucci, A.; Tieto, M.; Tondello, E., Cobalt oxide- based films: sol–gel synthesis and characterization. J. Non-Cryst. Solids 2001, 293–295, 477-

482.

20. Quinlan, F. T.; Vidu, R.; Predoana, L.; Zaharescu, M.; Gartrner, M.; Groza, J.; Stroeve,

P., Lithium Cobalt Oxide (LiCoO2) Nanocoatings by Sol−Gel Methods. Ind. Eng. Chem. Res.

2004, 43, 2468-2477.

21. Athey, P. R.; Urban, F. K.; Tabet, M. F.; McGahan, W. A., Optical properties of cobalt oxide films deposited by spray pyrolysis. J. Vac. Sci. Technol., A 1996, 14, 685-692.

22. Shinde, V. R.; Mahadik, S. B.; Gujar, T. P.; Lokhande, C. D., Supercapacitive cobalt oxide (Co3O4) thin films by spray pyrolysis. Appl. Surf. Sci. 2006, 252, 7487-7492.

23. Louardi, A.; Rmili, A.; Ouachtari, F.; Bouaoud, A.; Elidrissi, B.; Erguig, H.,

Characterization of cobalt oxide thin films prepared by a facile spray pyrolysis technique using perfume atomizer. J. Alloy. Compd. 2011, 509, 9183-9189.

24. Abbas, T. A.-H.; Slewa, L. H.; Khizir, H. A.; Kakil, S. A., Synthesis of cobalt oxide

(Co3O4) thin films by electrostatic spray pyrolysis technique (ESP). J. Mater. Sci. Mater.

Electron. 2016, 1-7.

25. Meyer, W.; Biedermann, K.; Gubo, M.; Hammer, L.; Heinz, K., Surface structure of polar Co3O4 (111) films grown epitaxially on Ir(100)-(1 × 1). J. Phys. Condens. Matter 2008,

20, 265011.

26. Johnson, R. W.; Hultqvist, A.; Bent, S. F., A brief review of atomic layer deposition: from fundamentals to applications. Mater. Today 2014, 17, 236-246.

27. Mazzi, A.; Bazzanella, N.; Orlandi, M.; Edla, R.; Patel, N.; Fernandes, R.; Miotello, A.,

Physical vapor deposition of mixed-metal oxides based on Fe, Co and Ni as water oxidation catalysts. Mater. Sci. Semicond. Process. 2016, 42, Part 1, 155-158.

114

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

28. Xu, T.; Schwarz, M.; Werner, K.; Mohr, S.; Amende, M.; Libuda, J., Structure-

Dependent Anchoring of Organic Molecules to Atomically Defined Oxide Surfaces: Phthalic

Acid on Co3O4(111), CoO(100), and CoO(111). Chem. Eur. J. 2016, 22, 5384-5396.

29. Klepper, K. B.; Nilsen, O.; Fjellvåg, H., Growth of thin films of Co3O4 by atomic layer deposition. Thin Solid Films 2007, 515, 7772-7781.

30. Ngo, T. Q.; Posadas, A.; Seo, H.; Hoang, S.; McDaniel, M. D.; Utess, D.; Triyoso, D.

H.; Buddie Mullins, C.; Demkov, A. A.; Ekerdt, J. G., Atomic layer deposition of photoactive

CoO/SrTiO3 and CoO/TiO2 on Si(001) for visible light driven photoelectrochemical water oxidation. J. Appl. Phys. 2013, 114, 084901.

31. Leskelä, M.; Ritala, M., Atomic Layer Deposition Chemistry: Recent Developments and Future Challenges. Angew. Chem. Int. Ed. 2003, 42, 5548-5554.

32. Backman, L. B.; Rautiainen, A.; Lindblad, M.; Krause, A. O. I., The interaction of cobalt species with alumina on Co/Al2O3 catalysts prepared by atomic layer deposition. Appl. Catal.,

A 2009, 360, 183-191.

33. Barreca, D.; Devi, A.; Fischer, R. A.; Bekermann, D.; Gasparotto, A.; Gavagnin, M.;

Maccato, C.; Tondello, E.; Bontempi, E.; Depero, L. E.; Sada, C., Strongly oriented Co3O4 thin films on MgO(100) and MgAl2O4(100) substrates by PE-CVD. CrystEngComm 2011, 13,

3670-3673.

34. Liu, R.; Lin, Y.; Chou, L.-Y.; Sheehan, S. W.; He, W.; Zhang, F.; Hou, H. J. M.; Wang,

D., Water Splitting by Tungsten Oxide Prepared by Atomic Layer Deposition and Decorated with an Oxygen-Evolving Catalyst. Angew. Chem. Int. Ed. 2011, 50, 499-502.

35. Paracchino, A.; Laporte, V.; Sivula, K.; Grätzel, M.; Thimsen, E., Highly active oxide photocathode for photoelectrochemical water reduction. Nat. Mater. 2011, 10, 456-461.

115

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

36. Morales-Guio, C. G.; Tilley, S. D.; Vrubel, H.; Grätzel, M.; Hu, X., Hydrogen evolution from a copper(I) oxide photocathode coated with an amorphous molybdenum sulphide catalyst.

Nat. Commun. 2014, 5, 3059.

37. Young, K. M. H.; Hamann, T. W., Enhanced photocatalytic water oxidation efficiency with Ni(OH)2 catalysts deposited on [small alpha]-Fe2O3 via ALD. Chem. Commun. 2014, 50,

8727-8730.

38. Steier, L.; Luo, J.; Schreier, M.; Mayer, M. T.; Sajavaara, T.; Grätzel, M., Low-

Temperature Atomic Layer Deposition of Crystalline and Photoactive Ultrathin Hematite Films for Solar Water Splitting. ACS Nano 2015, 9, 11775-11783.

39. Hajibabaei, H.; Zandi, O.; Hamann, T. W., Tantalum nitride films integrated with transparent conductive oxide substrates via atomic layer deposition for photoelectrochemical water splitting. Chem. Sci. 2016, 7, 6760-6767.

40. Lim, B. S.; Rahtu, A.; Gordon, R. G., Atomic layer deposition of transition metals. Nat.

Mater. 2003, 2, 749-754.

41. Lichterman, M. F.; Shaner, M. R.; Handler, S. G.; Brunschwig, B. S.; Gray, H. B.;

Lewis, N. S.; Spurgeon, J. M., Enhanced Stability and Activity for Water Oxidation in Alkaline

Media with Bismuth Vanadate Photoelectrodes Modified with a Cobalt Oxide Catalytic Layer

Produced by Atomic Layer Deposition. J. Phys. Chem. Lett. 2013, 4, 4188-4191.

42. Han, B.; Choi, K. H.; Park, K.; Han, W. S.; Lee, W.-J., Low-Temperature Atomic Layer

Deposition of Cobalt Oxide Thin Films Using Dicobalt Hexacarbonyl tert-Butylacetylene and

Ozone. Electrochem. Solid State Lett. 2011, 15, D14-D17.

43. Taheri Najafabadi, A.; Khodadadi, A. A.; Parnian, M. J.; Mortazavi, Y., Atomic layer deposited Co/γ-Al2O3 catalyst with enhanced cobalt dispersion and Fischer–Tropsch synthesis activity and selectivity. Appl. Catal., A 2016, 511, 31-46.

116

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

44. Lim, B. S.; Rahtu, A.; Park, J.-S.; Gordon, R. G., Synthesis and Characterization of

Volatile, Thermally Stable, Reactive Transition Metal Amidinates. Inorg. Chem. 2003, 42,

7951-7958.

45. Barry, S. T., Amidinates, guanidinates and iminopyrrolidinates: Understanding precursor thermolysis to design a better ligand. Coord. Chem. Rev. 2013, 257, 3192-3201.

46. Knisley, T. J.; Saly, M. J.; Heeg, M. J.; Roberts, J. L.; Winter, C. H., Volatility and High

Thermal Stability in Mid- to Late-First-Row Transition-Metal Diazadienyl Complexes.

Organometallics 2011, 30, 5010-5017.

47. Chen, Y. W.; Prange, J. D.; Duehnen, S.; Park, Y.; Gunji, M.; Chidsey, C. E. D.;

McIntyre, P. C., Atomic layer-deposited tunnel oxide stabilizes silicon photoanodes for water oxidation. Nat. Mater. 2011, 10, 539-544.

48. Scheuermann, A. G.; Prange, J. D.; Gunji, M.; Chidsey, C. E. D.; McIntyre, P. C.,

Effects of catalyst material and atomic layer deposited TiO2 oxide thickness on the water oxidation performance of metal-insulator-silicon anodes. Energy Environ. Sci. 2013, 6, 2487-

2496.

49. Hu, S.; Shaner, M. R.; Beardslee, J. A.; Lichterman, M.; Brunschwig, B. S.; Lewis, N.

S., Amorphous TiO2 coatings stabilize Si, GaAs, and GaP photoanodes for efficient water oxidation. Science 2014, 344, 1005-1009.

50. Lichterman, M. F.; Carim, A. I.; McDowell, M. T.; Hu, S.; Gray, H. B.; Brunschwig, B.

S.; Lewis, N. S., Stabilization of n-cadmium telluride photoanodes for water oxidation to O2(g) in aqueous alkaline electrolytes using amorphous TiO2 films formed by atomic-layer deposition. Energy Environ. Sci. 2014, 7, 3334-3337.

51. Li, W.; Gibbs, G. V.; Oyama, S. T., Mechanism of Ozone Decomposition on a

Manganese Oxide Catalyst. 1. In Situ Raman Spectroscopy and Ab Initio Molecular Orbital

Calculations. J. Am. Chem. Soc. 1998, 120, 9041-9046. 117

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

52. Knoops, H. C. M.; Elam, J. W.; Libera, J. A.; Kessels, W. M. M., Surface Loss in Ozone-

Based Atomic Layer Deposition Processes. Chem. Mater. 2011, 23, 2381-2387.

53. Han, B.; Park, J.-M.; Choi, K. H.; Lim, W.-K.; Mayangsari, T. R.; Koh, W.; Lee, W.-J.,

Atomic layer deposition of stoichiometric Co3O4 films using bis(1,4-di-iso-propyl-1,4- diazabutadiene) cobalt. Thin Solid Films 2015, 589, 718-722.

54. Will, G.; Masciocchi, N.; Parrish, W.; Hart, M., Refinement of simple crystal structures from synchrotron radiation powder diffraction data. J. Appl. Crystallogr. 1987, 20, 394-401.

55. Nandi, D. K.; Manna, J.; Dhara, A.; Sharma, P.; Sarkar, S. K., Atomic layer deposited cobalt oxide: An efficient catalyst for NaBH4 hydrolysis. J. Vac. Sci. Technol., A 2016, 34,

01A115.

56. Liu, J. F.; Yin, S.; Wu, H. P.; Zeng, Y. W.; Hu, X. R.; Wang, Y. W.; Lv, G. L.; Jiang, J.

Z., Wurtzite-to-Rocksalt Structural Transformation in Nanocrystalline CoO. J. Phys. Chem. B

2006, 110, 21588-21592.

57. Yang, J.; Liu, H.; Martens, W. N.; Frost, R. L., Synthesis and Characterization of Cobalt

Hydroxide, Cobalt Oxyhydroxide, and Cobalt Oxide Nanodiscs. J. Phys. Chem. C 2010, 114,

111-119.

58. Biesinger, M. C.; Payne, B. P.; Grosvenor, A. P.; Lau, L. W. M.; Gerson, A. R.; Smart,

R. S. C., Resolving surface chemical states in XPS analysis of first row transition metals, oxides and hydroxides: Cr, Mn, Fe, Co and Ni. Appl. Surf. Sci. 2011, 257, 2717-2730.

59. Li, J.; Lu, G.; Wu, G.; Mao, D.; Guo, Y.; Wang, Y.; Guo, Y., Effect of TiO2 crystal structure on the catalytic performance of Co3O4/TiO2 catalyst for low-temperature CO oxidation. Catal. Sci. Tech. 2014, 4, 1268-1275.

60. Harriman, A.; Pickering, I. J.; Thomas, J. M.; Christensen, P. A., Metal oxides as heterogeneous catalysts for oxygen evolution under photochemical conditions. J. Chem. Soc.

Faraday Trans. 1988, 84, 2795-2806. 118

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

61. Kanan, M. W.; Yano, J.; Surendranath, Y.; Dincă, M.; Yachandra, V. K.; Nocera, D. G.,

Structure and Valency of a Cobalt−Phosphate Water Oxidation Catalyst Determined by in Situ

X-ray Spectroscopy. J. Am. Chem. Soc. 2010, 132, 13692-13701.

62. Risch, M.; Ringleb, F.; Kohlhoff, M.; Bogdanoff, P.; Chernev, P.; Zaharieva, I.; Dau,

H., Water oxidation by amorphous cobalt-based oxides: in situ tracking of redox transitions and mode of catalysis. Energy Environ. Sci. 2015, 8, 661-674.

63. Ramakrishnan, V.; Kim, H.; Park, J.; Yang, B., Cobalt oxide nanoparticles on TiO2 nanorod/FTO as a photoanode with enhanced visible light sensitization. RSC Adv. 2016, 6,

9789-9795.

64. Barroso, M.; Cowan, A. J.; Pendlebury, S. R.; Gratzel, M.; Klug, D. R.; Durrant, J. R.,

The Role of Cobalt Phosphate in Enhancing the Photocatalytic Activity of α-Fe2O3 toward

Water Oxidation. J. Am. Chem. Soc. 2011, 133, 14868-14871.

65. Cummings, C. Y.; Marken, F.; Peter, L. M.; Tahir, A. A.; Wijayantha, K. G. U., Kinetics and mechanism of light-driven oxygen evolution at thin film α-Fe2O3 electrodes. Chem.

Commun. 2012, 48, 2027-2029.

66. Ma, Y.; Le Formal, F.; Kafizas, A.; Pendlebury, S. R.; Durrant, J. R., Efficient suppression of back electron/hole recombination in cobalt phosphate surface-modified undoped bismuth vanadate photoanodes. J. Mater. Chem. A 2015, 3, 20649-20657.

67. Pore, V.; Rahtu, A.; Leskelä, M.; Ritala, M.; Sajavaara, T.; Keinonen, J., Atomic Layer

Deposition of Photocatalytic TiO2 Thin Films from Titanium Tetramethoxide and Water. Chem.

Vap. Deposition 2004, 10, 143-148.

68. Goldstein, D. N.; McCormick, J. A.; George, S. M., Al2O3 Atomic Layer Deposition with Trimethylaluminum and Ozone Studied by in Situ Transmission FTIR Spectroscopy and

Quadrupole Mass Spectrometry. J. Phys. Chem. C 2008, 112, 19530-19539.

119

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

69. Waldo, R. A., An iteration procedure to calculate film compositions and thicknesses in electron-probe microanalysis. Microbeam Anal. 1988, 23rd, 310-14.

70. Jokinen, J.; Keinonen, J.; Tikkanen, P.; Kuronen, A.; Ahlgren, T.; Nordlund, K.,

Comparison of TOF-ERDA and nuclear resonance reaction techniques for range profile measurements of keV energy implants. Nucl. Instr. Meth. Phys. Res. 1996, 119, 533-542.

120

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

Supporting Information

Jiyeon Kim,a Tomi Iivonen,b Jani Hämäläinen,b Marianna Kemell,b Kristoffer Meinander,c Kenichiro Mizohata,c Lidong Wang,a Jyrki Räisänen,c Radim Beranek,a,d Markku Leskelä,b Anjana Devia*

a) Inorganic Materials Chemistry, Chair of Inorganic Chemistry II, Ruhr-University Bochum, Universitätstr. 150, 44801 Bochum, Germany b) Laboratory of Inorganic Chemistry, Department of Chemistry, University of Helsinki, P.O. Box 55, FI-00014 Helsinki, Finland c) Division of Materials Physics, Department of Physics, University of Helsinki, P.O. Box 43, FI-00014 Helsinki, Finland d) Institute of Electrochemistry, Ulm University, Albert-Einstein-Allee 47, 89081 Ulm, Germany

Table of Contents

tBu2 Figure S1 EI-MS fragmentation pattern of [Co( DAD)2]. Figure S2 Schematic depiction of the reaction chamber of F-120 ALD reactor

2 Figure S3 Co3O4 film thickness vs distance from leading edge of the 5 x 5 cm silicon substrate depending on the deposition temperature.

Figure S4 The photograph of Co3O4 thin film on Si and soda lime glass deposited at 120 °C over 1000 cycles.

Figure S5 Plane view SEM of Co3O4 thin film deposited over 1000 cycles at 120 °C.

Figure S6 SEM image of Si trench with an aspect ratio of 20:1 (depth 40 m, width 2 m).

Figure S7X-ray Photoelectron Spectroscopy survey spectrum of Co3O4 thin films

Figure S8 UV-Vis transmission spectrum and corresponding Tauc plot of Co3O4 thin film deposited over 1000 cycles at 120 °C.

Figure S9 UV-Vis transmittance plots of a 100 nm TiO2 film (pristine TiO2) modified by the depositing series of Co3O4 (10, 50, 100, 300, 500, and 10000 cycles) on the TiO2 films.

Figure S10 Photocurrent transients of TiO2/Co3O4 structures with depositing series of Co3O4 catalyst (10 – 1000 ALD cycles) on FTO glass under irradiation at different wavelengths (a),

121

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

selected area (b, c) and the corresponding photoaction spectra (d) measured in KOH (0.1 M) at 1.48 V vs RHE at TiO2 and TiO2/Co3O4 photoanodes.

Figure S11 AFM image of Co3O4 deposited over 10 cycles at 120 °C on Si (100) substrate.

Figure S12 XPS spectra of Co3O4 thin films deposited over 10 cycles on Si and TiO2 at 120 °C.

Figure S13 XRD patterns of TiO2 deposited at 275 °C over 2000 cycles on FTO glass

Table S1 XPS surface distribution of Co, O, C, Ti and Si on samples deposited over 10 and 50 cycles on a 100 nm thick TiO2 film and Si (100).

122

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

tBu2 Figure S1 EI-MS fragmentation pattern of [Co( DAD)2].

tBu2 The electron ionization mass spectrometry clearly identifies [Co( DAD)2] molecular ion peak at m/z 395.3.

Figure S2 Schematic depiction of the reaction chamber of F-120 ALD reactor (cassette configuration).

123

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

Figure S2. A schematic representation of the reaction chamber of the F-120 ALD reactor in the “cassette configuration”. The dimensions of the schematic are not to scale.

2 Figure S3 Co3O4 film thickness vs distance from leading edge of the 5 x 5 cm silicon substrate depending on the deposition temperature. All the samples were applied 300 deposition cycles.

124

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

Figure S4 The photograph of Co3O4 thin film on Si and soda lime glass deposited at 120 °C over 300 cycles.

Co3O4 on Si (100) Co3O4 on soda lime glass Soda lime glass

Figure S5 Plane view SEM of Co3O4 thin film deposited over 1000 cycles at 120 °C.

500 nm 1.00 µm

125

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

Figure S6 SEM image of Si trench with an aspect ratio of 20:1 (depth 40 m, width 2 m).

Co3O4 film has been deposited over 1000 cycles at 120 °C using pulse lengths of 6 seconds

tBu2 for both Co( DAD)2 and ozone. The purge length for both precursors was 10 seconds.

Thickness(nm) Top Middle Bottom

114 101 84

122 93 85

111 93 77

119 93 71

130 89 77

138 92 71

Average

thickness (nm) 122 94 78

Fully conformal Co3O4 films could not be deposited on 40 m Si trenches (aspect ratio of

126

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

Figure S7 XPS survey spectrum of Co3O4 thin film deposited over 1000 cycles at 120 °C.

Figure S6 displays the XPS survey scan of Co3O4 that is dominated by cobalt and oxygen photopeaks. No Ar+ sputtering experiment was performed.

Figure S8 UV-Vis transmission spectrum and corresponding Tauc plot of Co3O4 thin film deposited over 1000 cycles at 120 °C.

127

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

Figure S9 UV-Vis transmittance plots of a 100 nm TiO2 film (pristine TiO2) modified by the depositing series of Co3O4 (10, 50, 100, 300, 500, and 10000 cycles) on the TiO2 films.

128

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

Figure S10 Photocurrent transients of TiO2/Co3O4 structures on FTO glass with varying number of ALD cycles (10 – 1000) under irradiation at different wavelengths (a), selected transients (b, c), and the corresponding photoaction spectra (d) measured in KOH (0.1 M) at

1.48 V vs RHE at TiO2 and TiO2/Co3O4 photoanodes.

129

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

130

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

Figure S11 AFM images of Co3O4 deposited over 10 cycles (a,b) and 50 cycles (c,d) at 120

°C on Si (100) substrate. After 10 deposition cycles, island-like nanoparticles are formed instead of a closed film. After 50 deposition cycles, the island-like features are no longer present and a homogeneous and smooth surface is observed, indicating that a closed film has been formed. Images (a) and (b) were captured from different positions of a sample deposited over 10 cycles. Images (c) and (d) were captured from different positions of a sample deposited over 50 cycles.

Figure S12 XPS spectra of cobalt oxide deposited over 10 and 50 cycles on a 100 nm thick

TiO2 layer and on Si (100) substrate at 120 °C.

131

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

132

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

Table S1. XPS surface distribution of Co, O, C, Ti and Si on samples deposited over 10 and 50 cycles on a 100 nm thick TiO2 film and Si (100).

In samples deposited over 10 cycles on Si (100) and TiO2, the signal from the substrate is observed, indicating that the formed cobalt oxide film is not closed, but island-like nanoparticles are formed instead. After 50 deposition cycles on both Si (100) and TiO2, the signal from the substrate is no longer observed, indicating that a closed layer has been formed.

Co % O % 2- - Co O C Ti Si Total CoO Co(OH)2 Co3O4 O OH 10 cycles on TiO2 4.82 44.38 41.38 9.43 0.00 100.01 24.90 52.28 22.82 51.17 48.83 50 cycles on TiO2 14.41 47.92 37.37 0.00 0.00 100.00 50.94 0.00 49.06 41.39 58.31 10 cycles on Si (100) 3.80 33.01 25.34 0.00 37.86 100.01 35.53 57.37 7.11 16.69 83.31 50 cycles on Si (100) 13.15 47.42 39.43 0.00 0.00 100.0 39.62 0.23 60.15 37.03 62.97

133

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

Figure S13 XRD patterns of TiO2 deposited at 275 °C over 2000 cycles on FTO glass

GI-XRD reflections of TiO2 films deposited at 275 °C on FTO show mixed phases of anatase/rutile dominated by the anatase phase.1

References:

134

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

Homoleptic all nitrogen coordinated bis-guanidinato cobalt (II) complex: Volatile and thermally stable precursors suitable for ALD/CVD

135

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

7. Homoleptic all nitrogen coordinated bis-guanidinato cobalt (II) complex: Volatile and thermally stable precursors suitable for ALD/CVD

7.1 Abstract This work describes the synthesis of novel homoleptic all nitrogen coordinated bis-guanidinato cobalt (II) complexes. The used guanidinato ligands feature different side groups attached to the guanidinates, namely isopropyl and cycohexyl. Interestingly, after reaction of the ligand precursors with the cobalt (II) source a carbodiimide insertion occurs in the case of the isopropyl containing guanidinato ligands, which is not observed in the case of cyclohexyl. In this work the chemistry of steric hindrance effect of cobalt guanidinate complexes will be discussed. The cobalt (II) complexes were fully characterized using NMR spectroscopy, single crystal X-ray diffraction and TGA-DSC.

7.2 Introduction

Highly conformal thin films of metal or metal oxides are highly desired for their applications in worldwide field including microelectronics, optics, protection against corrosion and oxidation, diffusion barriers, sensor for gas or liquid chemicals, photo sensitive coatings and so on. Although chemical vapor deposition techniques including atomic layer depositions are widely used to provide very uniform and conformal thin films, suitable and effective precursors remained scarce. The criteria of ALD precursors are as follow. 1) Highly reactive towards the functionalities on the surface as well as surface prepared by complementary reactant. 2) Volatile and thermally stable at growth temperatures (> seconds) and sublimation temperatures (> months). 3) Non-reactive and non-corrosive byproduct formation is preferred. Chronological 136

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

development of precursors including metal halides, Cps, alkoxides, b-ketonates, alkylamides,

TMS, heteroleptic compounds of mixture of theses ligands meet several limits. Metal halides produce highly corrosive byproducts and precursors involving M¬C bonds lead to carbon incorporation into the resulting films.2-3

Precursors containing metals for chemical vapor depositions were investigated across the across the periodic table for the last century. While oxygen based precursors has reached the limit, nitrogen based ligand system has caught an attention due to its ease of tunable steric bulk and electronic saturation. In fact, the presence of two substituents (NR2) at the nitrogen atom readily leads to increase of steric hinderance compared to oxygen based system (OR). Consequently, a higher steric crowding surround the electro positive metal atom prevents the oligomerization which influence the volatility of the compounds. Amidinates were intensively investigated which diminish the drawbacks of conventional precursors.3 Our interest to develop reactive, volatile and thermally stable all nitrogen coordinated cobalt complexes as CVD/ALD precursors, we were inspired by the guanidinates as versatile ligand system closely related to amidinates. Guanidinates have caught attention as ligand for transition metals, main group3-4 and lanthanides. Rare earth guanidinate are not so common still can be found in several literatures.5

Guanidinates are in similarity of amidinates in terms of steric hinderance, however guanidinates possess extended electronic flexibility where the lone pair electrons of the amido-nitrogen are delocalized in the ligand -system.2 Additionally, guanidinate ligand is stronger base than the amidinate which will prompt the reactivity of corresponding guanidinato metal complexes.

Synthetic strategy of guanidinato complex include several routes,2, 6 however we will focus on salt metathesis approach in the paper.2, 7-8

137

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

We have synthesized homoleptic all nitrogen coordinated bis-guanidinato cobalt (II) complexes with the general formula of [M((NR’)2CNR2)2]and the steric bulk effect to stabilize the bis- guanidinato complexes depends on R’ substituent.

138

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

7.3 Result and Discussion

Synthetic strategy of guanidinato complex include several routes,6 however we will focus on salt metathesis approach in this work.7-8 The salt metathesis contains two step reaction.

Stoichiometric reaction of a carbodiimide (CDI) with lithium amide generates a lithium guanidinate [Li((NR’)2CNR2)]. An in-situ reaction with metal halide affords the desired guanidinate complex along with LiCl as a byproduct.

iPr iPr2 iPr2 Stoichiometric reaction of CDI and LiNMe2 results in Li( Guan) ( Guan =

iPr2 (NiPr)2CNMe2). The reaction of two equivalents of Li( Guan) with CoCl2 was expected to

iPr2 II result in Co( Guan)2. Unexpectedly, heteroleptic compound [Co ((NiPr)2CNMe2)8

iPr (NiPr)2CNMe2(iPrNHCNiPr))] (5) was obtained whereas CDI insertion occurred on the guanidinate ligand (Scheme 7.1 and Figure 7.1). In fact, homoleptic six-fold coordinated guanidinate compounds (tris-guanidinates) are already known for second/third row transition metals, metalloid such as antimony and rare earth metals. Meanwhile guanidinates (e.g. 2,6- diisopropylphenyl, 2,4,6-mesityl) with high streric bulk ligands were used to stabilize the low coordination first row transition metal (I) complexes9 but two guanidinates coordination complexes were not investigated but still bis-guanidinate compounds are not investigated.

One molecule of iPrCDI cleaves the cobalt-N bond and lead to a wider N−M−N angel to stabilize the compound 5. Interestingly, the structurally similar ligand system iPrAMD is known that it leads to a homoleptic cobalt amidinate compound.10 This can be explained by the difference of the electronic situation between the amidinates and guanidinates. Richer electron density and electronical flexibility on the iPrguan ligand discourage the iPr substituent to be not bulky enough for stabilization of the Co(II) metal center. On the other hand, our attempt of

iPr 10 reproducing Co( AMD)2 afforded contrasting results. According to our observation,

139

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

dimerization occurred on Co(iPrAMD)2 to yield [Co(iPrAMD)2]2 (7) which has been observed for the Fe and Mn amidinates with iPr-MeAMD ligand.10 (Scheme 7.1 and Figure 7.3) The resulting compounds 5 and 7 suggest that it is challenging for the iPr group to stabilize the cobalt bis- amidinates/guanidinates complexes.

Reaction with employing cyclohexyl group instead of iPr in a similar fashion as compound 5 led to homoleptic bis-guanidinate complex, [Co((CyN)2CN(CH3)2)2] that the steric hindrance from dialkylcarbodiimide has increased. (Scheme 7.1)

Scheme 7.1 Synthetic scheme of guanidinates according to the steric hinderance.

II The solid state molecular structures of [Co ((NiPr)2CNMe2)((NiPr)2CNMe2(iPrNHCNiPr))] (5)

Cy and [Co( guan)2] (6) determined by single crystal X-ray diffraction analysis. The crystal system of compounds 5 and 6 were monoclinic (space group P2/1c) and triclinic (space group

140

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

P-1). Crystallographic refinement data of the compounds 5-7 can be found section in 7.5.3. All the Co−N bond lengths of heteroleptic compound 5 show in the same range (2.022 Å). The bond angle of N4−Co−N5 from the ligand where the CDI insertion took place displays much wider angle (96.16(12) °) than N1−Co−N2 in the guanidinate ligand (65.97(12) °). (Figure 7.1 and Table 7.1) Predictably, in compound 6 the bond lengths of Co−N and the angles of

N−Co−N are in the same range (2.016 Å and 66.62 °) Both structures show more or less strong distortion of tetrahedron geometry, however the distortion observed in 6 was more prominent than in 5. This is ascribed to the cyclohexyl group which is much higher sterically demanding compared to iso-propyl group, enables the nitrogen atoms from each guanidinates ligands are more distanced. (Figure 7.2 and Table 7.2)

iPr The solid state X-ray crystal structure of [Co( AMD)2]2 (7) shows the dimerized structure

iPr which has not been reported where only the monomeric Co( AMD)2 is reported. [Gordon] We postulate the both monomeric and dimeric structures are in the equilibrium in the solution so that the crystallization behavior can differ dependent on the applied crystallization condition.

The structure of dimeric compound 7 is shown in Figure 7.3, and selected bond distances and angles are contained in Table 7.3. Crystallographic data can be found in Table 7.4 in section

7.5.3. Each cobalt atoms in dimeric compound 7 are bridged by two 1-amidinate ligands and one chelated by one 2-amidinate ligand. The Co−N bond lengths of (2.0880(12)-2.0731(12)

Å) in the terminal chelating ligands are slightly enlongated than the Co−N distances in the bridging ligand (2.0450(12)-2.430(12) Å), but are in the same range. The compound 7 is not of direct goal and interest in this work, further characterization of 7 was not performed.

141

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

Figure 7.1 Povray plot of the molecular structure of compound 5 in the solid state as determined by single crystal X-ray diffraction. Thermal ellipsoids are shown at the 50 % probability level and the hydrogen atoms are omitted for clarity.

Table 7.1 Selected Bond Length (Å) and Angles (°) for

II [Co ((NiPr)2CNMe2)((NiPr)2CNMe2(iPrNHCNiPr))] (5)

Co−N(1) 2.074(3) N(1)−Co−N(2) 65.97(12)

Co−N(2) 2.017(3) N(4)−Co−N(5) 96.16(12)

Co−N(4) 1.969(3) N(1)−Co−N(5) 115.87(12)

Co−N(5) 2.029(3) N(2)−Co−N(4) 128.34(13)

N(1)−C(1) 1.328(5) Co−N(4)−C(10) 110.6(2)

N(2)−C(1) 1.340(5) Co−N(5)−C(20) 116.3(2)

N(4)−C(10) 1.360(5) N(4)−C(10)−N(6) 112.2(3)

N(5)−C(20) 1.316(5) N(5)−C(20)−N(6) 119.5(3)

N(6)−C(10) 1.478(5) C(10)−N(6)−C(20) 119.3(3)

N(6)−C(20) 1.376(5)

142

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

N(7)−C(10) 1.267(5)

Cy Figure 7.2 Povray plot of the molecular structure of [Co( guan)2] (6) (one of the two whole fragments in the asymmetric unit are shown) in the solid state as determined by single crystal

X-ray diffraction. Thermal ellipsoids are shown at the 50 % probability level and the hydrogen atoms are omitted for clarity. Structural disorder is confined to the cyclohexyl groups in the other half of fragments.

Cy Table 7.2 Selected Bond Length (Å) and Angles (°) for [Co( Guan)2] (6)

Co−N(1) 2.023(3) N(1)−Co−N(2) 66.53(13)

Co−N(2) 2.012(3) N(4)−Co−N(5) 66.70(13)

Co−N(4) 2.009(3) N(1)−Co−N(4) 129.77(14)

Co−N(5) 2.021(3) N(2)−Co−N(5) 132.40(14)

N(1)−C(1) 1.336(5)

N(2)−C(1) 1.345(5)

N(4)−C(16) 1.348(5)

N(5)−C(16) 1.348(5)

143

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

N(3)−C(1) 1.374(5)

N(6)−C(16) 1.363(5)

iPr Figure 7.3 Molecular structure (POV-ray plot) of [Co( AMD)2]2 as determined by single- crystal X-ray diﬀraction. Displacement ellipsoids are shown at the 50% probability level, and hydrogen atoms are omitted for clarity.

iPr Table 7.3 Selected Bond Length (Å) and Angles (°) for [Co( AMD)2]2 (7)

Co−N(1) 2.0880(12) N(1)−Co−N(2) 64.50(5) Co−N(2) 2.0731(12) N(3)−Co−N(4) 132.45(5) Co−N(3) 2.0450(12) N(1)−Co−N(3) 113.42(5) Co−N(4) 2.0430(12) N(1)−Co−N(4) 106.97(5) N(1)−C(1) 1.333(2) N(2)−Co−N(3) 105.80(5) N(2)−C(1) 1.325(2) N(2)−Co−N(4) 113.70(5) N(3)−C(9) 1.329(2) N−C−Nchelating 113.28(13) N(4)−C(9) 1.332 N−C−Nbridging 116.01(13)

144

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

A crude oil of compound 5 as synthesized without crystallization was measured TGA to study thermal behavior. (Figure 7.4) TGA curve shows onset of volatilization (1 % weight loss) at

73 °C and a shouldered curve can be seen at ca. 220 °C that is a phenomenon named CDI deinsertion of guanidinate demonstrated in previous report.11 The final residual mass show ca.

10 %. Isothermal studies at ambient pressure exhibit non-linear weight loss which is attributed to a partial decomposition of ligand (CDI deinsertion from ((NiPr)2CNMe2(iPrNHCNiPr))).

TGA studies were repeated on crystallized compound 5 in combination with DSC (Differential

Scanning Calorimetry) to screen the solid-liquid phase transition. (Figure 7.5) In the solid state

TGA, a single step weight loss was observed due to the higher purity and less contact to air during the sample loading. The compound 5 is in the liquid state at room temperature and crystallizes at ca. -30 °C. Notably, it is observed that after warming up the crystals to room temperature, the compound remains in the solid state. DSC studies confirm this behavior. After cooling, a phase transition is observed in the DSC trace, however, on the heating branch the phase transition to solid occurs at substantially higher temperature (75 °C). This result indicates a good agreement with the melting point found in DTA. Both measurements TGA/DTA and

DSC were using a heating rate of 5 K/min which was apparently insufficient to trigger the phase transition in the used temperature range.

145

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

Figure 7.4 Thermal properties of compound 5 as crude oil, TGA/DTA (left) and isothermal

TG studies (right).

Figure 7.5 Thermal properties of compound 5 as crystal pellets (left) and DSC studies (right).

146

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

TGA studies of 6 at ambient pressure displays onset of volatilization at 104 °C and a shouldered curve can be seen at 180 °C which is caused by aforementioned CDI deinsertion.11 The residual weight was found approximately 10 %. (Figure 7.6) Isothermal studies at ambient pressure at

90 °C showed linear weight loss whereas at 100 °C exhibited relatively non-linear behavior.

The evaporation rates as calculated from the slope of the isothermal TGA curves at 90 and 100

C were 3.67 mg/min and 6.95 mg/min. The isothermal studies at 90 °C indicate capability of sustainable transport of precursor during the deposition time (180 min).

Figure 7.6 Thermal properties of compound 6 as crude oil, TGA/DTA (left) and isothermal

TG studies (right).

147

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

Figure 7.7 Thermal properties of compound 5 as crystal pellets (left) and DSC studies (right).

7.4 Conclusion

Cy Within this chapter the synthesis of a new cobalt-guanidinate precursor, namely [Co( Guan)2] is discussed. We found an astonishing dependence of the bonding situation at the final precursor and the R-groups on the used guanidinates. In the case of small iPr groups, a CDI-insertion occurs, leading to the formation of a heteroleptic complex. When a much bulkier R-group is used in the preparation of the precursor, for instance cyclohexyl, a homoleptic guanidinate complex is achieved. This insertion has a dramatic effect on the thermal properties and the volatilization of the compound, increasing the temperature when the compound enters the gas phase. We showed that the use of different R-groups is controlling the insertion, which weakens the usability of the compounds as precursors. By employing bulky side groups, it was possible to prevent the insertion effect and to obtain the homoleptic cobalt guanidinate compounds.

It is planned in future studies to analyze the preparation of cobalt oxide thin films using ALD and CVD methods. We expect to be able to utilize this precursor in the development of a systematic deposition study towards the before mentioned materials. Further studies include the alteration of the R-groups on the guanidinates in order to understand their influence on the

148

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

inclusion reaction and to demonstrate the tunability of this class of ligands by slight alterations on the ligand backbone.

149

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

7.5 Experimental Section

7.5.1 General Comments

Material and Methods. All manipulations were carried out in an atmosphere of purified argon using standard Schlenk and glove-box techniques. n-Hexane, n-pentane, toluene, diethylether were dried using an MBraun Solvent Purification System. Tetrahydrofuran and d6-benzene were dried by passing through a column of dried activated neutral Al2O3. The final H2O content in all solvents was checked by Karl Fischer titration and did not exceed 5 ppm. All characterization and instrumental details are identical described in chapter 3.6 and 6.5. Lithium guanidinate species, [Li(NR)2CNMe2] were prepared by stoichiometric reaction of LiNMe2 and corresponding dialkylcarbodiimide in diethylether in prior to the salt metathesis reaction and used without any further isolation or purification.

7.5.2 Synthetic Procedure

II i i i i i [Co ((N Pr)2CNMe2)((N Pr)2CNMe2( PrNHCN Pr))] (5) Freshly prepared [Li( Pr)2CNMe2] etherate solution was added into a CoCl2 (0.5 equivalent) etherate solution at 0 °C. The reaction mixture was stirred overnight at room temperature. All the volatiles were removed under vacuum and the residue was extracted with n-hexane. The residual solution was filtered through the celite pad in a shlenk frit to remove LiCl salts. The filterate was concentrated to a quarter of initial volume, then recrystallized at -30 °C. Blue crystals suitable for single crystal X-ray diffraction were obtained.

150

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

Cy [Co( Guan)2] (6) Freshly prepared [Li(Cy)2CNMe2] etherate solution was added into a CoCl2

(0.5 equivalent) etherate solution at 0 °C. The reaction mixture was stirred overnight at room temperature. All the volatiles were removed under vacuum and the residue was extracted with n-hexane. The residual solution was filtered through the celite pad in a shlenk frit to remove

LiCl salts. The filterate was concentrated to a quarter of initial volume, then recrystallized at -

30 °C. Blue-violate crystals suitable for single crystal X-ray diffraction were obtained.

iPr 10 [Co( AMD)2]2 (7) was prepared by a modified procedure described in the literature. A solution of methyllithium (1.6 M in Et2O) was added dropwise to a etherate solution of 1,3- diisopropylcarbodiimide in 1:1 molar ratio at -30 °C. The reaction mixture was stirred for

iPr overnight at room temperature. The formed Li( AMD) solution was added into an Et2O solution of CoCl2 (0.5 equivalent) by means of cannula at 0 °C. The reaction mixture was stirred overnight at room temperature. All the volatiles were removed under vacuum and the residue was extracted with n-hexane. The extract was filtered through a celite pad on a schlenk frit to afford dark blue solution. The filtrate was concentrated to 10-15 mL and recrystallized at -30 °C.

151

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

7.5.3 Crystallographic data

Table 7.4 Crystallographic Data Crystallographic data and refinement details of compounds 5-7.

5 6 7

Chemical formula C25H55CoN8 C30H68Co1N6 C32H68Co2N8 (C60H116Co2N12) Formula weight 526.70 561.75 (1123.50) 682.80

Crystal system monoclinic triclinic monoclinic

Space group P21/c P-1 C2/c

a (Å) 20.1386(3) 10.5285(3) 23.1963(4)

b (Å) 9.42059(14) 16.0905(6) 9.5810(2)

c (Å) 16.3758(3) 19.0321(6) 18.0658(3)

 (°) 100.9386(19) 96.101(2) 110.788(2)

Volume (Å3) 3050.33(9) 3159.69(17) 3753.63(13)

Z 4 2 4

-3 Dcalc (g × cm ) 1.147 1.181 1.208

Absorption coeﬃcient (mm-1) 4.599 4.453 7.151

F(000) 1152.0 1228.0 1480.0

2θ range for data collection (°) 8.944 to 152.716 6.864 to 153.088 8.154 to 152.774

Reﬂections collected/unique 25068/6271 [Rint = 57520/13133 [Rint = 19281/3903 [Rint = 0.0548] 0.1086] 0.0315] Data/restraints/parameters 6271/0/324 13133/0/675 3903/0/200

goodness-of-ﬁt on F2 1.213 1.029 1.050

Final R indices [I > 2(I)] R1 = 0.0566, wR2 = R1 = 0.0670, wR2 = R1 = 0.0281, wR2 = 0.1447 0.1627 0.0703 R indices (all data) R1 = 0.0726, wR2 = R1 = 0.0897, wR2 = R1 = 0.0331, wR2 = 0.1503 0.1752 0.0727

152

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

8. References

1. Ballirano, P.; Caminiti, R., Rietveld refinements on laboratory energy dispersive X-ray diffraction (EDXD) data. J. Appl. Crystallogr. 2001, 34, 757-762.

2. Bailey, P. J.; Pace, S., The coordination chemistry of guanidines and guanidinates.

Coord. Chem. Rev. 2001, 214, 91-141.

3. Barry, S. T., Amidinates, guanidinates and iminopyrrolidinates: Understanding precursor thermolysis to design a better ligand. Coord. Chem. Rev. 2013, 257, 3192-3201.

4. Coyle, J. P.; Johnson, P. A.; DiLabio, G. A.; Barry, S. T.; Müller, J., Gas-Phase

Thermolysis of a Guanidinate Precursor of Copper Studied by Matrix Isolation, Time-of-Flight

Mass Spectrometry, and Computational Chemistry. Inorg. Chem. 2010, 49, 2844-2850.

5. Milanov, A. P. MOCVD and ALD of Rare Earth Containing Multifunctional Materials:

From Precursor Chemistry to Thin Film Deposition and Application. Dissertation, Ruhr-

University Bochum, Bochum, Germany, 2010.

6. Musashi, Y.; Sakaki, S., Insertion of carbon dioxide into a rhodium(III)-hydride bond: a theoretical study [dagger]. Journal of the Chemical Society, Dalton Transactions 1998, 577-

584.

7. Edelmann, F. T., Chapter 3 - Advances in the Coordination Chemistry of Amidinate and

Guanidinate Ligands. In Adv. Organomet. Chem., Anthony, F. H.; Mark, J. F., Eds. Academic

Press: 2008; Vol. Volume 57, pp 183-352.

8. Edelmann, F. T., Lanthanide amidinates and guanidinates: from laboratory curiosities to efficient homogeneous catalysts and precursors for rare-earth oxide thin films. Chem. Soc.

Rev. 2009, 38, 2253-2268.

9. Jones, C.; Schulten, C.; Rose, R. P.; Stasch, A.; Aldridge, S.; Woodul, W. D.; Murray,

K. S.; Moubaraki, B.; Brynda, M.; La Macchia, G.; Gagliardi, L., Amidinato– and Guanidinato–

153

PART 2 Precursor Synthesis and Atomic Layer Deposition of Cobalt Oxide

Cobalt(I) Complexes: Characterization of Exceptionally Short Co–Co Interactions. Angew.

Chem., Int. Ed. 2009, 48, 7406-7410.

10. Lim, B. S.; Rahtu, A.; Park, J.-S.; Gordon, R. G., Synthesis and Characterization of

Volatile, Thermally Stable, Reactive Transition Metal Amidinates. Inorg. Chem. 2003, 42,

7951-7958.

11. Kenney, A. P.; Yap, G. P. A.; Richeson, D. S.; Barry, S. T., The Insertion of

Carbodiimides into Al and Ga Amido Linkages. Guanidinates and Mixed Amido Guanidinates of Aluminum and Gallium. Inorg. Chem. 2005, 44, 2926-2933.

154

Curriculum Vitae

Curriculum Vitae Jiyeon Kim

E-mail: [email protected]

PROFESSIONAL EXPERIENCE

1/2017 6/2017 Scientific Researcher Technical University Munich, Chair of Inorganic and Metal Organic Chemistry

 Synthesis of Rhenium Sulfide nanoparticles for anti-skin cancer application

12/2012 – 12/2016 Scientific Researcher Ruhr-University Bochum, Chair of Inorganic Chemistry II  Atomic Layer Deposition (ALD) of Transition Metal Oxides Thin Film and Precursor Chemistry  Synthesis of Organometallic Complexes Expertized in Handling Extremely Air Sensitive Materials  Analytic Techniques; Single Crystal X-ray Diffraction, TG, XRD, XPS, AFM, SEM, NMR, IR, EI-MS etc.  ALD Process Development  Teaching Assistant in Advanced Inorganic Chemistry (Lab Course, 4 semesters in total) and Supervision of Bachelor and Master Students

09/2012 – 11/2012 Research Assistant Kookmin University, Laboratory of Inorganic Chemistry  Synthesis of Metal/Metal Oxide Nanoparticles and Their Surface Modification via Ligand Exchange

03/2012 – 05/2012 Visiting Researcher Ruhr-University Bochum, Chair of Inorganic Chemistry II  Metal Organic Framework coated Silver Nanowires

03/2010 – 02/2012 Research Assistant Kookmin University, Laboratory of Inorganic Chemistry

 Photo-luminescent Ceramic Nano Materials  Analysis of Fluorescent Materials (NMR, UV-Vis, PL)

01/2009 – 02/2010 Undergraduate Research Opportunities Program (UROP) Kookmin University, LABORATORY of Inorganic Chemistry

155

Curriculum Vitae

EDUCATION

12/2012 – 4/2017 Doctor of Philosophy (Ph.D.) Inorganic Chemistry Bochum, Germany Chair of Inorganic Chemistry II, Ruhr-University Bochum Prof. Dr. Anjana Devi and Prof. Dr. Roland A. Fischer

“Precursor Chemistry for Atomic Layer Deposition of Metal Oxides” Development of novel cobalt precursor for ALD and process aiming for photoelectrocatalytic water splitting application.

“Synthesis of Heteronuclear Metal Complexes Containing Direct Metal-Metal Bonds” Metal-Metal bond formation via Reductive Elimination

03/2010 – 02/2012 Mater of Science (M.Sc.) Chemistry Seoul, Republic of Korea Inorganic Chemistry, Kookmin University, Prof. Dr. Sungho Yoon GPA 4.36 / 4.5 “Surface Modification of Silica Nanoparticles with Zn(II) Complexes Mimicking the Active Site of Carbonic Anhydrase”

03/2006 – 02/2010 Bachelor of Science (B.Sc.) Chemistry Seoul, Republic of Korea Inorganic Chemistry, Kookmin University, Prof. Dr. Sungho Yoon GPA 3.26 / 4.5 “Investigating the Gelation Properties of Mixed Ag(I) Carboxylates”

INTERNATIONAL EXPERIENCE

03/2016 and 06/2015 Research Stay Helsinki, Finland University of Helsinki, Helsinki, Finland Prof. Dr. Markku Leskelä “Atomic Layer Deposition of Cobalt Oxide”

Development of Ozone Assisted ALD Process Resulting in Co3O4 Thin Films for Photoelectrocatalysis

01/2013 – 12/2015 Research Contribution 4G-Photocat EU Project (FP7-309636) Objectives: Development of Highly Active Composite Photocatalysts Based on Cheap and Abundant Elements, and Their Deposition onto Various Surfaces. Lab Visits for Scientific Presentation and Discussion during the Project in Czech Republic, Finland, Vietnam, Poland, Malaysia

07/2014 Research Visit Sapporo, Japan Hokkaido University, Sapporo, Japan Prof. Dr. Hajime Ito Organometallic Chemistry and Catalysis (C-H Activation)

156

Curriculum Vitae

FELLOWSHIPS & HONOURS

05/2013 – 12/2015 4G-Photocat EU Project Ph.D. Fellowship 07/2014 – 07/2016 Two Ruhr-University Bochum Research School Plus Grants for International Projects 02/2012 Excellent Researcher Scholarship from Kookmin University

LANGUAGE SKILLS

Korean Native

English Fluent in Spoken and Written Form 2.5 Years Elementary School in USA (Los Angeles County, CA) TOFEL Score of 91 in 2012 Official Language during Ph.D. Project German Basic in Spoken and Written Form

SOFTWARE SKILLS

Microsoft Office (Word, Power Point, Excel), Origin Photoshop, CorelDraw, ChemDraw, POV-Ray Shelxl, Olex

Jiyeon Kim July, 20th, 2017

157

Curriculum Vitae

PUBLICATIONS

Homoleptic all nitrogen coordinated bis-guanidinato cobalt (II) complex: Volatile and thermally stable precursor suitable for ALD/CVD Jiyeon Kim, Niklas Stegmann, Stefan Cwik and Anjana Devi* Manuscript in progress 2017

Low temperature atomic layer deposition of cobalt oxide as an effective catalyst layer for photoelectrochemical water splitting devices Jiyeon Kim, Tomi Iivonen, Jani Hämäläinen, Marianna Kemell, Kristoffer Meinander, Kenichiro Mizohata, Radim Beranek, Markku Leskelä and Anjana Devi* Accepted in Chemistry of Materials (ACS) 2017 (DOI: 10.1021/acs.chemmater.6b05346)

I + Oxidation of Nickel-ECp* Complexes: Stable Open-Shell Ni Cations [Ni(ECp*)n(PPh3)4-n] (n = 2, 4; E = Al, Ga) Jiyeon Kim, Markus Halbherr, Christian Gemel and Roland A. Fischer* Inorganic Chemistry, 2015, 54, 9675

Comparative Synthesis of Cu and Cu2O Nanoparticles from Different Copper Precursors in an Ionic Liquid or Propylene Carbonate Raquel Marcos Esteban, Hajo Meyer, Jiyeon Kim, Christian Gemel, Roland A. Fischer* and Christoph Janiak* Eur. J. Inorg. Chem. 2016, 2016, 2106

Low-temperature atomic layer deposition of copper(II) oxide thin films Tomi Iivonen*, Jani Hämäläinen, Benoît Marchand, Kenichiro Mizohata, Miika Mattinen, Georgi Popov, Jiyeon Kim, Roland A. Fischer, and Markku Leskelä Journal of Vacuum Science & Technology A, 2016, 34, 01A109

A fluorescent ammonia sensor based on a porphyrin cobalt(II)–dansyl complex Jiyeon Kim, Si Hyung Lim, Yeoil Yoon, T. Daniel Thangadurai and Sungho Yoon* Tetrahedron Letters, 2011, 52, 2645

CONFERENCE

7/2016 Low Temperature Ozone Assisted ALD of Co3O4 Dublin, Ireland Jiyeon Kim, Tomi Iivonen, Radim Beranek, Markku Leskelä and Anjana Devi Oral 16th Atomic Layer Deposition Conference, Dublin, Ireland

8/2015 Low Valent Group 13 Organyls at Transition Metals Boston, USA Jiyeon Kim, Christian Gemel and Roland A. Fischer Oral + Poster

Precursor Synthesis and Atomic Layer Deposition of CoOx Thin Films Jiyeon Kim, Tomi Iivonen, Roland A. Fischer, Markku Leskelä and Anjana Devi 250th National Meeting of the American Chemical Society

7/2014 Transition Metal-mediated Elimination of H2 from (Organo-)Al Hydrides Sapporo, Japan Jiyeon Kim, Christian Gemel and Roland A. Fischer Poster XXVI International Conference on Organometallic Chemistry

Oral The Coordination Chemistry of Low Valent Group 13 Species at Transition Metal Complexes Jiyeon Kim, Christian Gemel and Roland A. Fischer International Workshop of the Japanese Society of Coordination Chemistry 158

Curriculum Vitae

159