Micrc5films International 300 N

OXYGEN ATOM TRANSFER REACTIONS OF NICKEL AND PALLADIUM NITRO COMPLEXES.

Item Type text; Dissertation-Reproduction (electronic)

Authors SIMONDSEN, JEANNE CLARE.

Publisher The University of Arizona.

Rights Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.

Download date 29/09/2021 02:32:32

Link to Item http://hdl.handle.net/10150/184281 INFORMATION TO USERS

This reproduction was made from a copy of a document sent to us for microfilming. While the most advanced technology has been used to photograph and reproduce this document, the quality 01'1 the reproduction is heavily dependent upon the quality of the material submitted.

The following expl~nation of Itechniques is provided to help clarify markings or notations which mar appear on this reproduction.

1. The sign or ~'target" for pages apparently lacking from the document photographed is "Missing Page(s)". If it was possible to obtain the missing page(s) or section, they arel spliced into the film along with adjacent pages. This may have nece~sitated cutting through an image and duplicating adjacent pages to assure complete continu~ty.

2. When an imag~ on the filp1 is obliterated with a round black marko, it is an indication of either bluned copy because of movement during exposure, duplicate copy, or copyrigttted materials that should not have been filmed. For blurred pages, .a good image of the page can be found in the adjacent frame. If copyrighted mflterials werel deleted, a target note will appear listing the pages in

the adjacent fn.lme. I

3. When a map, qrawing or chart, etc., is part of the material being photographed, a definite me~hod of "sectioning" the material has been followed. It is customary to t~egin filming at the upper left hand corner of a large sheet and to continue from left to right in equal sections with small overlaps. If necessary, sectioning is c~mtinued again-beginning below the first row and continuing on

until complete. I

4. For illustrations that cam not be satisfactorily reproduced by xerographic means, photographic prin~s can be purchased at additional cost and inserted into your xerqgraphic copy. These prints are available upon request from the Dissertations Customer Ser:vices Department.

5. Some pages in any document may have indistinct print. In all cases the best available copy p.as been filmed.

University Micrc5films International 300 N. Zeeb Road Ann Arbor, MI48106

8223015

Simondsen, Jeanne Clare

OXYGEN ATOM TRANSFER REACfIONS OF NICKEL AND PALLADIUM NITRO COMPLEXES

The University of Arizona PH.D. 1982

University Microfilms Intern ati 0 nal 300 N. Zeeb Road, Ann Arbor, MI 48106

OXYGEN ATOM TRANSFER REACTIONS OF NICKEL

AND PALLADI~1 NITRO COMPLEXES

Jeanne Clare Simondsen

A Dissertation Submitted to the Faculty of .the

DEPARTMENT OF CHEllISTRY

In Partial Fulfillment of the Requirements For the Degree of

DOCTOR OF PHILOSOPHY

In the Graduate College

THE UNIVERSITY OF ARIZONA

1 982 THE UNIVERSITY OF ARIZONA GRADUATE COLLEGE

As members of the Final Examination Committee, we certify that we have read the dissertation prepared by ------Jeanne Clare Simondsen entitled Oxygen Atom Transfer Reactions of Nickel and Palladium Nitro Complexes

and recomnlend that it be accepted as fulfilling the dissertation requirement for the Degree of Doctor of Philosophy

Date

IJfo, Jj /~i~ Date ) IJ 1£1<. Date2l24j

Final approval and acceptance of this dissertation is contingent upon the candidate's submission of the final copy of the dissertation to the Graduate College.

I hereby certify that I have read this dissertation prepared under my direction and recommend that it be accepted as fulfilling the dissertation requirement.

lssertatlon Dlrector STATEHENT BY AUTHOR

This dissertation has been submitted in partial fulfillment of requirements for an advanced degree at the University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library.

Brief quotations from this dissertation are allowable without special permission, provided that accurate acknowledgment of source is made. Requests for permission for extended quotation from or reprod~c tion of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the Graduate College when in his judgement the proposed use of the material is in the interests of scholarship. In all other instances, however, permission must be obtained from the author. _.

SIGNED: (7 . For my parents

and

For Royce

iii ACKNOWLEDGMENTS

I wish to express my gratitude to Dr. Robert D. Feltham, my research director, for his encouragement and advice during the course of this research.

I also wish to thank Drs. J. H. Enemark and J. V. Rund for their helpful suggestions. Thanks are also due to my colleagues,

Drs. Jules Dubrawski and Trevor Bailey for many helpful discussions both in and out of the laboratory.

iv TABLE OF CONTENTS

Page

LIST OF ILLUSTRATIONS vii

LIST OF TABLES ix

ABSTRACT xi

1. BACKGROUND 1

2. PREPARATION AND CHARACTERIZATION OF COMPLEXES: THE CRYSTAL STRUCTURE OF Ni(N0 ) (NO) (PMe ) 2 ...•• 8 2 3 Introduction 8 Experimental • • • • • . •• .•••. • • • • 9 Preparation of Complexes . • • • . ••.•• 19 Collection and Reduction of X-Ray Intensity Data 20 Solution and Refinement of the Structure 20 Results and Discussion •• • • • • . • • . . . • • • • 23 Complexes of the Formula Ni(N02)2L2 • .• .. • • • 23 Complexes of the Formula Ni(N02) (NO)L2 ••••• 25 Structure of Ni(N02) (NO) (PMe3) 2 . 29 Description of Structure .~ 29 Structurar Interpretation • . 40 Conclusions . • • • • • • • • • 47

3. CRYSTAL STRUCTURES OF Ni(N02)2dppe AND [Ni(ONO)(NO)dppe]2 . 49

Introduction 49 Experimental • • • . • . . • 50 Preparation of Complexes . . • • • • • • • • • . 51 Collection and Reduction of X-Ray Intensity Data 51 Solution and Refinement of Structures . • • . • . 52 Results and Discussion • . • • 57 Description of Structures . . • • • 57 Structural Interpretation • 73

4. NECHANISTIC STUDIES OF THE REACTION OF Ni(N02)2L2 AND CO 75 Introduction • • . . . • . . 75 Experimental 76 Kinetic Studies • . 77 Isotope Scrambling Experiments 79 31p{lH} NMR Studies . . 80

v vi

TABLE OF CONTENTS--Continued

Page

Reversibility and Turnover Experiments 81 Detection of Intermediates • • • . 81 Results and Discussion . • • • • . • • • 82

5. REACTION OF PALLADIUM NITRO COMPLEXES WITH CO • 98 Introduction ...... 98 Experimental • • • • • • • • • 99 Collection and Reduction of X-Ray Intensity Data 99 Solution and Refinement of Structure • • • • • 100 Results and Discussion 102

APPENDIX A: LIST OF ABBREVIATIONS • • • • • 1· 117

REFERENCES ...... -. . . 118 LIST OF ILLUSTRATIONS

Figure Page

1.1. Proposed pathway for the oxygen exchange reaction in CNMe 2 cis-Fe(N02)(NO)(S2 2)2 • .. ••• • . •• • ••• 1.2. Proposed mechanism for the reaction of NiC1 L with 2 2 NaN0 . . . • • • • • • • • . • • • . • • • • . • • 4 2 + 4 1.3. Proposed mechanism for dissociation of [Ni(NO)(PPh3)3] •

2.1. Mechanism proposed for oxygen atom transfer between N0 2 and CO • • • • • • • • • • • • • . • • • • • • 8

2.2. Correlation diagram relating the molecular orbitals of pseudotetrahedral four_coordinate NiNO 10 complexes of C and C symmetry showing allowed electronic 3 tranSl.tl.onsv .. s ...... " . . • . 28

2.3. Perspective view and numbering scheme for Ni(N0 ) (NO) (PMe )2 2 3 at 135 K ...... 34

2.4. Perspective view and numbering scheme for Ni(N02) (NO) (PMe3) 2 at 293 K ...... 35

2.5. Packing diagram for Ni(N0 ) (NO) (PMe ) 2 at 293 K (A) and 2 3 135 K.(B) • • •••• '0' • • •• ••• 38

o 2.6. Ni-N (A) vs Ni-N-O angle (deg) for structurally characterized {NiNO}lO complexes . • • • • . • • • • • • . • 42

2.7. Correlation diagram relating the molecular orbitals of pseudo tetrahedral four-coordinate NiNO 10 complexes of C3v and Cs symmetry to those of four-coordinate square planar {NiNO}lO complexes of Cs symmetry 44 2.8. N-Ni-X angles for Ni(NO) (X) (PR )2 complexes. . 46 3 . . 3.1. Perspective view and numbering scheme for Ni(N02)2dppe 58

3.2. Perspective view and numbering scheme for [Ni(ONO)(NO)- dppe]2 . • • • • . • • . . • . . . • • • 59

4.1. Modified spectrophotometer used for kinetic investigations • • • • . • . • . • . • . . • . •• 78

v;ii viii

LIST OF ILLUSTRATIONS--Continued

Figure Page

Beer's law plot for Ni(N0 ) (NO)dppe in CH C1 •...• 83 4.2. 2 2 2

] 4.3. Plot of -In[(At -Am )/-Am vs t at 570 nm for the reaction of Ni(N02)2dppe with CO in CH C1 at 20°C •.••• 85 2 2 4.4. Dependence of kobsd on PCO for the reaction of Ni(N02)2dppe and CO at 20°C in CH C1 . • • . • • • . • • • 86 2 2 4.5. Scheme I: A mechanism proposed for the reaction of Ni(N02)2L2 anJ CO • . • • • . • • . . . •. • ••• 87

4.6. 31p{lH} nmr spectra of NiI2dppe (CDC1 3) under CO 89 4.7. 3lp{lH} nmr spectra of Ni(15N02)2dppe in CD C1 93 2 2 4.8. Experimental and simulated 31p{lH} nmr spectra of Ni(15NO Ni(15N02)2dppe in CD C1 at 220 K . . • . . • 94 2 2 4.9. 3lp{lH} nmr spectra of Ni(15 ) (15 )dppe in CD C1 • 95 N02 NO 2 2 4.10. Scheme II: An alternative mechanism proposed for the reaction of Ni(N02)2L2 and CO . . • . • • • . • 96

5.1. Perspective view and numbering scheme for Pd (C0 )- 4 5 104 (PMePh2) 4 ...... • 5.2. Mechanistic implications of oxygen atom transfer via a nitrito isomer ...... • . • •• 115

5.3. Possible reaction routes for the formation of Pd4(CO)5L4 ...... • • . • •. • • • • • • • 116 LIST OF TABLES

Table Page

2.1. Elemental analyses for Ni(N02)2L2 complexes •• 10

Eleffiental analyses for Ni(N0 ) (NO)L complexes 11 2.2. 2 2 Crystallographic data for Ni(N0 ) (NO) (PMe ) 2 ...•.••• 12 2.~. 2 3 2.4. N02 vibrational frequencies of Ni(N02)2L2 complexes ••••• 14

2.S. Additional vibrational frequencies not due to phosphine ligands i~'l Ni (N0 2) 2L2 complexes . • . • • . • • • . • • 15

2.6.- Room temperature 3lp{lH} NMR spectra of Ni(N02)2L2 and 16 Ni(N02) (NO)L2 complexes .••.....••.•.•• 2.7. Electronic spectra for Ni(N0 ) (NO)L complexes 17 2 2 2.8. Infrared data for Ni(N0 ) (NO)L complexes ••• 18 2 2 . . 2.9. Selected bond distances and angles of structurally characterized {NiNO}lO complexes . • • • • . • 27

2.10. Atomic positional parameters for the nonhydrogen atoms of Ni(N0 ) (NO) (P(CH )3)2 •.•.. • • 30 2 3 2.11. Thermal parameters for the non-hydrogen atoms of Ni(N0 ) (NO) (P(CH )3)2 •.•.•. • 31 2 3 2.12. Interatomic distances and angles for Ni(N0 ) (NO) (PMe ) 2 • 32 2 3 CI 2.13. Root-mean-square amplitudes of vibration (in A) for Ni(N0 )(NO)(P(CH )3)2 • • • 36 2 3 2.14. Thermal and positional parameters for the hydrogen atoms of Ni(N0 ) (NO) (P(CH )3)2 . • . • • . • • • • . • 37 2 3 3.1. Cryst~llographic data for [Ni(ONO)(NO)«C6HS)2PCH2CH2P(C6HS)]2 and N1(N02)2«C6HS)PCH2CH2P(C6HS)2)-CH2C12 . . • . • • • •• 53

3.2. Final parameters for the nonhydrogen atoms of Ni(N0 2)2- «C6HS)2PCH2CH2P(C6HS)2 • CH2C12 ....••••••. 60

ix x

LIST OF TABLES--Continued

Table Page

3.3. Final parameters for the nonhydrogen atoms of [Ni(NO)(ONO) «C6HS)ZPCHZCHZP(C6HS)Z)]Z .••••••••••••••.• 61

3.4. Final group parameters for Ni(NOZ)Z«C6HS)ZPCHZCHZP(C6HS)Z) . CH C1 • • . • • • . • . • . . • . • • . • • • • • • • 6Z 2 2

3.S. Final group parameters for [Ni(NO)(O~O)«C6HS)ZPCHZCHZP (C H )Z)]Z ..•...... ••.. 63 6 S 3.6. Derived parameters for the group atoms for Ni(NOZ)Z 64 «C6HS)ZPCHZCHZP(C6HS)Z) • CHZC1 Z ...... •• 3.7. Derived parameters for the group atoms for [Ni(NO)(ONO) «C6HS)ZPCHZCHZP(C6HS)Z)]Z ...•...... 6S 3.8. Parameters for the fixed hydrogen atom positions for 67 Ni(NOZ)Z«C6HS)ZPCHZCHZP(C6HS)Z) • CHZC1 Z ...... •.•• 3.9. Parameters for the fixed hydrogan atom positions for [Ni (NO) (ONO) «C H ) 2PCH2CH2P (C H ) 2] 2 • • . • • . • • • 68 6 S 6 S 3.10. Selected interatomic distances and angles for Ni(N0 )2- 2 «C6HS)2PCHZCHZP(C6HS)2)" CH2C1 2 . • •• •.••. 69 3.11. Selected interatomic distances and angles for [Ni(NO)(ONO)- «C6HS)2PCH2CH2P(C6HS)2)]2 • . . . • • . •• 70 S.l. Crystallographic data at 2SoC for Pd «CO)S[P(CH3)- (C H )2]4 • • • • . • • • •. •.•• • . • • • • • • • • 101 6 S S.2. Atomic positional parameters for the non-hydrogen, non-group atoms of Pd (CO) S [P (CH ) (C H ) 2) 4 • . • . • . • • . . • • • • lOS 4 3 6 S S.3. Thermal parameters for the non-hydrogen, non-group atoms of Pd (CO)S[P(CH ) (C H )2]4 •••..•••....•••.•• 106 4 3 6 S S.4. Parameters for the fixed hydrogen atom positions for Pd (CO)S[P(CH ) (C H )2]4 . • . • . • • . • • • • . • 107 4 3 6 S S.S. Final group parameters for Pd (CO)S[P(CH ) (C H )2]4 • • 109 4 3 6 S S.6. Derived parameters for the group atoms for Pd (CO)S- 4 [P(CH ) (C H )2]4 •• . • . . • . . . •• ..•.• 110 3 6 S S.7. Selected interatomic distances and angles for Pd (CO)S- 4 [P(CH ) (C H )2]4 ...... • . •. . • . • • 112 3 6 S ABSTRACT

The reactions of nitro complexes of nickel and palladium 'tvith

CO have been examined to determine the mechanism(s) by which CO is 2 produced.

The solution and solid state structures of square planar

Ni(N0 )2(L)2 reactants and pseudotetrahedra1 Ni(N0 ) (NO) (L)2 products 2 2 have been determined and related to their reactivity. Infrared,

31p{lH}, and crystallographic data indicate rapid isomerization between nitro and nitrito bonding modes of the NO; ligands. The crystal structures of Ni(N02)2(PPh2(CH2)2PPh2) (I), Ni(N02)(NO)(PMe3)2 (II), and [Ni(ONO)(NO)(PPh2(CH2)2PPh2]2 (III), show the NO; groups to be

N-bonded in I and II and Q-bonded in III. The nitrosyl ligands in II and III are non-linear (Ni-N-O = 165.5(8)° and 153.4(8)°, respectively).

Furthermore, III crystallizes as a dimer bridged by two phosphine ligands even though molecular weights show this complex to be monomeric in solution.

Each Ni(N0 ) (NO) (L)2 complex reacts with CO to produce 2 stoichiometric amounts of Ni(N0 ) (NO) (L)2 and CO , Rate data indicate 2 2 the reaction proceeds associatively through formation of a carbonyl intermediate which has been directly observed in the reaction of 18 The reaction of C ° with Ni(N02)2- (PMe )2 results in no incorporation of 180 into the nickel product 3 while 180 is incorporated into CO to form 18 16 • 2 0C 0

xi xii

The mechanism consistent with all of the data involves a rapid equilibrium between both forms of NO; coordination followed by the reaction of CO with either isomer in the rate determining step to form a monocarbonyl complex. Irreversible oxygen atom transfer to CO and loss of CO2 terminate the reaction. The corresponding square planar palladium complexes,

Pd(N02)2L2' react with CO to form N 0, CO and novel tetranuclear 2 2 palladium clusters (Pd (CO)SL ). A crystal structure of Pd (CO)S 4 4 4 (PMePh )4 shows the cluster to be a distorted tetrahedron of metal atoms 2 with one open edge and the five remaining edges each bridged by a carbonyl group. CHAPTER 1

BACKGROUND

The reactions of nitrogen oxides and their salts are important since these species are generated by combustion processes as well as soil fertilization. Although the chemistry of such substances has received considerable attention, there is little known about the chemical pathways by which they are interconverted. Transition metals often 1 catalyze these reactions. Except however, for the catalyzed reactions 2-9 of CO with NO to yield CO and N 0, there are almost no mechanistic 2 2 2-9 studies of the reactions of NO complexes. x Many reactions of transition metal NO complexes are simple atom x transfers. Since atoms are the species being transferred, the oxidation state of the metal is not necessarily affected even though attached ligands may be oxidized or reduced. Intramolecular atom transfer reactions which have been reported include the transfer of oxygen between adjacent NO ligands in cis-nitronitrosy1bisdithiocarbamato iron. x Isotropic labelling studies have shown that when the above complex is prepared at O°C from Fe (lSNO) (S2CNMe2)2' the label is specifically retained at the nitrosyl nitrogen (Reaction 1.1).

(1.1)

- ~·---".-__~.~~a ______~ ______2

On warming tOI ambient temperature, however, the cis is~)mer is fC/rmed and

the label becomes statistically scrambled between the ~litro and nitrosyl

groups Oleaction 1.2).

(1.2)

The kinetic features of this reaction indicated an int:ramo1ecu1~r process 1whereby the atom transfer takes place via an o:~o-bridged inter mediate in wllich both nitrogens become equivalent (Fig. 1.1) .10 I

6 Q 0 o O'{5 1~' 15 / '14- 15 / 14N/ O-N 'N-O O-N N NI-O , / 'IFe/ Fe / Fe -~ " --

Figure 1.1. IProposed pathway for the oxygen exchange reaction in cis-Fe(N0 ) (NO) (S2CNMe2)2. 2

Although this reaction is mechanistically interesting, it is of no synthetic value. It does, however, suggest that under' appropri,ate conditions, NO groups could react with other ligands in the coordina x tion sphere C/f a metal.

Because CO is isoe1ectronic with NO+, it was tihe purpose of this

investigatioT,l to study the reactions of carbon monoxiqe with tr.ansition metal ni.tro ~omp1exes. 3

The oxidation of CO by nitrite is a thermodynamically favored but a kinetically slow process (Reaction 1.3).

2CO(g) 2NO;(aq) H 0(1) 2HCO;(aq) N 0(g) 585.8KJ (1.3) + + 2 ~ + 2 +

No reaction was observed between NaN0 (0.11 M) and CO (1 atm) in water 2 or between [Et N][N0 ]2 (.22 M) and CO in dich1oromethane over a 2 hour 4 2 period. A modest search of the literature revealed no reference to this reaction in the absence of transition metals. Carbon monoxide, however, reacts readily with nitrite when Ni, Fe, Co, Ru, Rh, or Ir are present to yield CO and the corresponding transition metal nitrosyl 2 11-26 complexes. In their report of the reaction between CO and trans- 11 Ni(N0 )2(PEt )2' Booth and Chatt found that production of the 2 3 nitrosyl complex, Ni(N0 ) (NO) (PEt )2 was rapid at ambient conditions. 2 3 The authors also found the {NiNO}10 (notation introduced by Enemark and

Fe1tham)27 product to be air sensitive and difficult to purify.

Although CO 2 was presumed to be the oxidized product, it was not identified in the reaction mixture.

CO is also claimed to be the reducing agent in Reactions 1.4 12 and 1.5.

NiC1 (PMePh )2 + NaN0 + NaPF + CO + PMePh ') 2 2 2 6 2 [Ni(NO)(PMePh )3][PF ] NaC1 CO (1.4) 2 6 + + 2

NaPF 6 >

Ni(C1) (NO) (PPh )2 + NaC1 + CO (1.5) 3 2 4

,~ere the following mechanism is proposed (Fig. 1.2):

NiC1 L + NaN0 NiCl(N0 )L + NaCl (1.6) 2 2 2 > 2 2 NiCl(N0 )L + CO ) NiCl(NO)L + CO (1. 7) 2 2 2 2 NiCl(NO)L + NaPF ) [Ni(NO)L ] [PF ]2 + NaCl (1.8) 2 6 2 6 [Ni(NO)L ][PF ] + L ) [Ni (NO)L ][PF6] (1. 9) 2 6 3 L = PMePh , PPh 2 3

Figure 1.2. Proposed mechanism for the reaction of NiC1 L with NaN0 • 2 2 2

That Ni(Cl) (NO)(PPh ) 2 is the sole product when L = PPh , is attributed 3 3 to the ease with which [Ni(NO) (PPh )3]+ undergoes dissociation 3 according to Fig. 1.3.

[Ni(NO) (PPh )3]+ -~) [Ni(NO)(PPh )2]+ + PPh (1.10) 3 3 3 [Ni(NO) (PPh )2] + X- -~) [Ni(X)(NO)PPh ]2 (1.11) 3 3

Figure 1.3. Proposed mechanism for dissociation of [Ni(NO) (PPh )3]+. 3

In support of this scheme, the authors point out that efforts to re- crystallize the tris[triphenylphosphine]nitrosyl complex in the absence 2 of excess free phosphine invariably resulted in decomposition.1

Although the reaction to generate Ni(Cl) (NO) (PPh )2 occurs in the 3 absence of free ligand, yields are significantly improved when a large excess of free PPh is present. Since excess phosphine acts to 3 suppress reaction 1.10, the significance of Fig. 1.3 is doubtful.

Furthermore, the necessity of carbon monoxide for the production of chloronitrosylbis(triphenylphosphine)nickel has not been demonstrated. '5 28 Fe1tham has reported the preparation of this complex in the absence

of CO (Reaction 1.12).

Ni(C1) (NO) (PPh )2 NaC1 (1.12) > 3 +

The presence of CO does not appear to accelerate the reaction or to

increase the yield of product. In addition, CO was not identified as 2 a product in reactions 1.4 or 1.5. For these reasons, the role, if any,

of CO in reactions 1.4 and 1.5 is not clear.

Transition metals are also known to catalyze the oxidation of

CO by NO according to reaction 1.13.

cat> CO + N 0 2NO + CO 2 2 (1.13)

Although reaction 1.13 has been studied extensively, it is complicated

and consists of several steps, including at least one in which the 1-9 transfer of an oxygen atom is required. One of several mechanisms

proposed for this reaction involves the transfer of an oxygen atom from

a co or d1nate· d· n1tro group 1n. one 0 f the k ey react10ns. . 9

The present study was undertaken to explore the specific role

played by transition metal complexes in promoting the oxidation of CO

by N0 • The metals selected for study were nickel and palladium. The 2 investigation was developed by preparing a variety of four coordinate

Ni(N02)2L2 (L PNe , PEt , PMe Ph, PHePh , PCY3' dppe, cis-vdpp, dppp) = 3 3 2 2 complexes and exploring their reactions with CO to yield Ni(N0 ) (NO)L 2 2 (Reaction 1.14). 6

(1.14 )

Reactants and products were fully characterized by elemental anaiysis, electronic and infrared spectroscopy, and 31p{lH} nmr as discussed in

were then examined in detail as representative examples of this class of reactions. The crystal structure of Ni(N0 )2(PMe )2 was determined 2 3 at 298 K and 135 K (Chapter 2). The relationships among Ni-N-Q angle,

Ni-P distance, and stereochemistry for this and related complexes were interpreted in terms of a molecular orbital description (Chapter 2).

Crystal structures of Ni(N02)2dppe and Ni(N0 ) (NO)dppe are presented 2 in Chapter 3.

These studies form the basis for the mechanistic investigation of reaction 1.14 presented in Chapter 4. This investigation required kinetic studies of the reaction of Ni(NOi)2dppe and Ni(N0 )2(PMe )2 with 2 3 18 CO. The reaction of C 0 with Ni(N0 )2(PMe )2 was also studied to 2 3 determine the distribution of isotopes in the products. In addition, variable temperature 31p{lH} nmr spectra of Ni(N02)2L2 and

Ni(N02) (NO)L 2 under CO and N2 were obtained to elucidate the solution structures of these complexes as well as to detect possible reaction intermediates. Finally, a carbonyl intermediate was detected in the reaction of Ni(N02)2(PCY3)2 with co.

The reaction of Pd(N02)2L2 (L = PMe2Ph, PMePh2) with co was found to yield the tetranuc1ear palladium cluster Pd4(CO)5L4 which was characterized by x-ray crystallography (L PMePh ) as discussed in = 2 7

Chapter 5. The mechanism elucidated for the reaction of Ni(N02)2L2 and

CO forms a basis for future understanding of the more complicated palladium reactions. CHAPTER 2

PREPARATION AND CHARACTERIZATION OF COMPLEXES: THE CRYSTAL STRUCTURE OF Ni(N0 ) (NO) (PMe )2 2 3

Introduction

The facile conversion of trans-Ni(N0 )2(PEt )2 to Ni(N0 ) (NO) 2 3 2 (PEt )2 by CO is one of the earliest reported examples of the now well 3 documented reduction of coordinated N0 to NO by CO. 3,11,12,16-22,29-3l 2

(2.1)

Although CO is presumed to be the oxidized product of this reaction, 2 it was not identified in the reaction mixture. Moreover, reports indicated the nickel nitrosyl product to be oily and difficult to purl."f y. 11

The mechanism depicted in Fig. 2.2 has been previously invoked for this class of reactions. 3 ,12,18-2l,29 o / Ln M- N t CO \ o

0 ...... 0 C / I Ln M- N \ o

Figure 2.1. Mechanism proposed for oxygen atom transfer between N0 2 and CO. 8 9

This scheme, however, has _not been experimentally substantiated to date. Consequently, a mechanistic investigation of this oxygen atom transfer reaction was undertaken. This study required the synthesis and characterization of a series of dinitrobis(phosphine) complexes of nickel(II). The reduction of each of these complexes by CO was studied and the resulting {NiNO}lO complexes were fully characterized. In addition, the crystal and molecular structure of Ni(N0 ) (NO) (PMe ) 2 was 2 3 determined at 298 K a?d 135 K. The structure of this complex will be 32 compared to structures of similar low symmetry Ni(X) (NO) (PR )2 ,33 3 complexes to probe the relationships among stereochemistry, ligand environment, Ni-N-O angle, and Ni-P distance.

Experimental

All samples were manipulated under an atmosphere of dry nitrogen using standard Schlenk tube techniques. Solvents were reagent grade and purified by distillation under dry nitrogen according to 15 publ 1S· h e d proce d ures 34.1mme d·1ate 1 y pr10r . to use. Na N0 (95% enriched) 2 15 supplied by Prochem was used to prepare N enriched samples.

Microanalyses were carried out by Huffman Laboratories,

Wheatridge, Colorado, and Atlantic }1icrolabs, Atlanta, Georgia. Solid state infrared spectra were obtained from KBr pellets using a Perkin

Elmer model 387 spectrometer. Electronic spectra were recorded on a

Cary 14 uv-visible spectrophotometer while 3lp{lH} nmr spectra were recorded on a Brucker model WM 250 multinuclear spectrometer and referenced to external T}iP. Physical properties of Ni(N02)2L2 and

Ni(N0 )(NO)L complexes are summarized in Tables 2.1-2.8. 2 2 Table 2.1. Elemental analyses for Ni(N02)2L2 complexes.

Complex Color i.C i.H i.N i.0 Calcd. Found Calcd. Found Calcd. Found Ca1cd. Found

23.80 Ni(N0 2)2(PHe3)2 dark 23.79 5.95 6.03 9.25 9.28 yellow

Ni(N02)2(PEt3)2 ye11ow- 37.24 37.24 7.76 7.75 7.24 7.82 orange

Ni(N02)2(PHe2Ph)2 bright 45.00 44.44 5.16 5.24 6.56 6.62 yellow bright 56.66 56.78 4.72 4.85 5.08 5.19 Ni(N02)2(PMePh2)2 yellow yellow 60.79 9.29 9.48 3.94 3.70 Ni(N02)2(P(C6H1l)3)2 60.31

Ni(N02)2dppe orange 56.86 56.70 4.37 4.43 5.10 4.99 11.66 11.41

4.22 5.12 4.96 Ni(N02)2(cis-vpp) orange 57.07 56.90 4.02

Ni(N02)2dppp orange 57.58 57.36 4.62 4.68 4.97 4.94

I-'o

/ Table 2.2. Elemental analyses for Ni(N0 ) (NO)L complexes. 2 2

Complex Color %C %H %H %0 Calcd. Found Calcd. Found Calcd. Found Calcd. Found

Ni(N0 ) (NO) (PMe )2 blue- 25.12 25.05 6.32 6.66 9.76 9.91 2 3 black Ni(N0 ) (NO) (PEt )2 blue- 38.84 38.70 8.15 8.15 7.55 7.42 2 3 black

Ni(N02) (NO) (PMe 2Ph)2 blue- 46.76 46.75 5.40 4.84 6.80 6.68 black Ni(N0 ) (NO)dppe purple 58.37 4.58 5.26 5.06 9.01 9.59 2 58.57 4.50

Ni(N0 ) (NO) (cis-vpp) purple 58.79 57.25 4.14 4.26 5.28 4.52 2

Ni(N0 ) (NO)dppp purple 4.78 5.12 5.06 2 59.27 59.10 4.76

...... Table 2.3. Crystallographic data for Ni(N0 ) (NO) (PMe )2' 2 3

a a 293 K 135 K molecular formula [Ni(NO) (N0 2) (P(CH3)3)2] [Ni(NO) (N02) (P(CH3)3)2) mol wt 283.36 283.36 cryst shape cubic cubic cryst dimens, mm3 0.35 x 0.38 x 0.47 cryst system monoclinic monoclinic cell dimens b a,A 7.751 (2) 7.648 (5) b,A 12.611 (3) 12.468 (11) c,A 14.328 (4) 13.956 (11) B,deg 96.93 (6) 95.88 (3) V, A3 1390.3 (6) 1324 (2) Z 4 4 dObsd,Cg cm-~ 1. 33 (3) d ca1cd, g cm -3 1.37 1.30 radiation,A (MdKa) 0.7103 (MoKa) 0.7103 monochromator graphite crystal graphite crystal supplied power 50kV, 30 rnA 50kV, 30 rnA data collection method 9-29 scan 9-29 scan scan speed, deg min -1 variable, 2.00-29.30, determined variable, 3.00-29.30, determined as a function of peak intensity as a function of peak intensity

I-' N Table 2.3. Crystallographic data--Continued.

a a 293 K 135 K ratio of total bkgd time to peak scan time 0.5 0.5 scan range (28), deg MoK -0.8 to MoK +0.8 MoK -0.8 to MoK +0.8 l 2 1 2 std reflctns (200),(020),(004) after every (200),(020),(004) after every 97 readings 97 readings decompn of standards 5% 3% 28 limit, deg 0.0-50.0 0.0-50.0 no. of unique data 2584 2481 no. of data used in the calcn 1811 1515, I ~ 30 (I) -1 abs coeff(~), cm 15.79 (MoKa) 15.79 (MoKa) a The standard deviation of the least significant figure is given in parentheses in this and the following tables. b Cell dimensions were obtained from a least-squares refinement of setting angles of 25 reflec tions in the 28 range 5-25° c Density was determined by flotation using a solution of diethyl ether and carbon tetrachloride.

I-' UJ a Table 2.4. N02 vibrational frequencies of Ni(N02)2L2 complexes.

-1 cm v v ass 15 ass v v 15 sym 0 0 15 15 Complex NO NO sym NO NO N02 rock N02 2 2 2 N02 2 N02 rock

1377 1353 1325 802 812 619 604 Ni(N02)2(PMe3)2 1305 1374 1323 818 619 Ni(N02)2(PEt3)2 1320 820 812 617 601 Ni(N02)2(PMe2Ph)2 1380 1355 1300 1385 820 815 620 605 Ni(N0 2)2(PMePh2)2 1362 1327 1306 1390 1370 1318 1303 820 813 Ni(N0 2)2(P(C6H11)3)2

Ni(N02) 2dppe 1410 1380 1340 1315 819 811 1392 1365 1320 1297 1375 1350 1310 1288

810 Ni(N02)2(cis-vpp) 1410 1370 1335 1319 817 1400 1355 1322 1281 1381 1310

Ni(N02)2dppp 1400 1340 815 1390 1310

-1 a All frequencies are calibrated using the polystyrene bands at 1601 and 1028 cm as standards. Spectra were determined in the solid state from KBr pellets.,

f-I ~ 15

Table 2.5. Additional vibrational frequencies not due to phosphine ligands in Ni(N02)2L2 complexes.a

Complex N N

PHe 1155 1126 Ni(N02)2( 3)2 Ni(N0 )2(PEt )2 1149 2 3 Ni(N02)2dppe 1270 1240 1105 1075

Ni(N02)2(cis-vpp) 1275 1245 1090 1065 Ni(N02)2dppp l27Q

a All frequencies are calibrated using the polystyrene bands at 1601 and 1028 cm-l as standards. Spectra were determined in the solid state from KBr pellets. 16

Table 2.6. Room temperature 31p{lH} NMR spectra of Ni(N02)2L2 and a Ni(N02 (NO)L2 comp1exes.

Complex o (ppm) b Peak Width (Hz). 6.8 Ni(N02)2(PMe3)2 -151. 7 -126.7 1.0 Ni(N02)2(PEt3)2 Ni(N02)2(PMe2Ph)2 -146.3 6.8 Ni(N0 )2(PMePh )2c -137.0 11 2 . 2 d Ni(N02)2(PMePh2)2 -117 -111 12 12 Ni(N02)2(PCY3)2 -125 2 Ni(N02)2dppe -87 120

Ni(N02)2(ci~-vpp) -87 120 Ni(N02)2dppp -137 83

Ni(N0 ) (NO)L Complexes 2 2 Ni(N0 ) (NO) (PMe ) 2 -138 2 3 Ni(N0 ) (NO) (PEt )2 -111 7 2 3 Ni(N02) (NO) (PMe 2Ph) 2 -131.3 6 Ni(N0 ) (NO) (PCY )2 4 2 3 -91 Ni(N0 ) (NO)dppe 2 -88 7 Ni(N0 ) (NO) (cis-vpp) 7 2 -86 Ni(N0 ) (NO)dppp 2.9 2 -112.5

a All spectra were obtained in CDC1 • 3 b Chemical shifts are relative to external TMP.

c Spectra recorded after complex was in solution 10 minutes.

d Spectra recorded after complex was in solution 8 hours. 17

Table 2.7. Electronic spectra for Ni(N0 ) (NO)L complexes. 2 2

Absorption Transition ComElex Band Assi~ment 4lA'~5lA' Ni(N02) (NO) (PMe3) 2 571(860) 444(835) 4lA~4lA" Ni(N0 ) (NO)(PEt ) 2 4lA'~5lA' 2 3 584(970) 458(1030) 4lA~ 4lA" 4IA~5lA' Ni(N02) (NO) (PMe2Ph) 2 576(1020) 450(940) 4lA'~ 4lA" 4lA45l A' Ni(N02) (NO)dppe 570(690) 460(480) 4lA441A" 4lA'-....:,.5l A' Ni(N02) (NO) (cis-vpp) 590(615)

a Positions of absorption bands are in nm.· The extinction coefficients in parentheses have units of 1 mol-lcm-l • All spectra were determined in CDC1 using 1 em glass cells. 3 18 a Infrared data for Ni(N0 ) (NO)L complexes. Table 2.8. 2 2

15 Complex NO NO Additional nonphosphine bands

Ni(N0 ) (NO) (PMe ) 2 1720 1270 800 2 3 Ni(N0 ) (NO) (PEt )2 1705 1365 1330 1310 805 2 3 Ni(N0 ) (NO) (PMe Ph) 2 1725 1380 1185 805 2 2

Ni(N02) (NO) (PCY3)2 1725 Ni(N0 ) (NO)dppeb 2 1750 1715 1380/1360 1340/1290 1260/1230 815/812· Ni(N0 ) (NO) (cis-vpp) 2 1745 Ni(N0 ) (NO)dppp 2 1730 a . -1 All frequencies are reported in cm using the polystyrene bands at 1601 and 1028 cm-1 as standards. Spectra determined in the solid state using KBr pellets. bFrequencies. arising from 15N complex are separated by a slash. 19 Preparation of Complexes

Ni(N0212~2 Complexes. (L=PMe , PEt , PMe Ph, PMePh , PCY3' or 3 3 2 2 1/2 dppe, cis-vdpp, and dppp). A solution of nickel nitrite was prepared by dissolving Ni(N03)2·6H20 (1 romol) in a minimum amount of dry methanol and adding NaN0 (2 romol). After vigorous stirring, the 2 dark green solution was cooled to -78DC. Filtering resulted in the removal of NaN0 • Addition of the appropriate phosphine ligan (2 romol) 3 resulted in precipitation nickel products of the formula Ni(N02)2L2 in nearly quantitative yields. All complexes were recrystallized from dichloromethane-hexane and dried in vacuuo.

Ni(N0 ) (NO)L Complexes. Method 1. (L = PMe Ph, PCY3' or 2 2 2 1/2 of dppe, cis-vdpp, or dppp). Ni(N0 )(NO)L complexes were prepared 2 2 by the reaction of Ni(N02)2L2 with CO in 20 ml of carefully degassed dichloromethane for 7 hours (20DC, I atm). The solution volume was then reduced to 5 ml in vacuuo and 20 ml of CO saturated hexane were .added.

After 10 hours, small blue-black crystals were collected by filtration and dried in vacuuo. Yields of nitrosyl product were 85-95% based on nickel. 36 }1ethod 2. (L = PMe , PEt , PMe Ph). Ni(N0 )(NO)L complexes 3 3 2 2 2 were prepared by the reaction of 2 romol of Ni(N02)2L2 with CO in 20 m1 of carefully degassed acetone for 2 hours (20DC, 1 atm). After evaporation of the solvent in vacuuo, the dark residue was dissolved in diethyl ether (30 ml), filtered and cooled in -30DC under an atmosphere of CO. After 3 hours, well formed crystals were collected by 20

filtration and dried ina stream of CO. Yields of nitrosyl product

were 65-70% based on nickel starting materia~.

Collection and Reduction of X-Ray Intensity Data

Several well formed crystals of Ni(N0 ) (NO) (PMe ) 2 obtained by 2 3 method 2, were coated with stopcock grease, mounted in quartz

capillaries, and sealed under nitrogen. One of these crystals was

mounted on a Syntex P21~autodiffractometer equipped with a scintillation

counter, a graphite monochromator, and a Syntex LT1 low-temperature

device. The longest dimension of the crystal (S axis) 'was

approximately parallel to the ~ axis. Automatic centering, indexing, 7 an d 1 east-squares rout1nes. 3 were carr1e. d out to Y1e . 1d t h e ce11

dimensions that are listed in Table 2.3 along with other crysta110-

graphic data. Intensity data were collected at room temperature under 2 the conditions listed in this table. The data were reduced to FO and

O(F~) by published procedures. Lorentz polarization factors were

calculated on the ass~mption of 50% mosaicity and 50% perfection of the monochromator crystal. During the data collection, all three standards

showed a 5% or less decline in intensity. Because of these small

intensity changes and the equidimensiona1 shape of the crystal,

corrections for decomposition and absorption were not significant.

Solution and Refinement of the Structure

Neutral-atom scattering factors for the non-hydrogen atoms and

corrections for the anomalous dispersion made for the nickel and

phosphorus atoms were taken from the tabulations of Cromer and 21 38,39 Wa b er. The hydrogen atom scattering factors were taken from the 40 calculations of Stewart, Davidson, and Simpson. The structures were

refined by full-matrix least-squares techniques minimizing the function 2 2 Lw(IFI - IFI )2 where w = 4F /[o2(F ) + (pF )]2. "P", the factor to o coo1 prevent overweighting of strong reflections, was set equal to 0.03. The

discrepancy indices, R1 and R , are defined in the usual way. All 2 computations were carried out on the CDC Cy-175 computer at the

University of Arizona Computer Center. The major programs used for the

structure determination were FORDAP (Fourier summation program by

A. Zalkin) Ibers' NUCLS (Structure factor calculations and full-matrix

least-squares refinement a modification of ORFLS by W. R .. Busing,

K. O. ~lartin, and H. A. Levy) and Ortep (Thermal ellipsoid drawing

program by C. K. Johnson).

The room-temperature structure was solved by the heavy-atom method in which the position of the nickel atom was readily determined from a three-dimensional Patterson map. The remaining nonhydrogen atoms were located by successive refinements and difference electron density maps. Refinement of this structure with anisotropic thermal parameters for all atoms converged with R1 = 0.0608 and R2 = 0.0952.

At this point, the positions of six hydrogen atoms labeled (1) in

Table 2. ll. (page 37) were located from a series of difference electron density maps. The remaining 12 hydrogen atoms were placed at geo- o metrically idealized positions (C-H = 0.95 A). All hydrogen atoms were 02 assigned an isotropic temperature factor, a = 5.0 A , and were held fixed in subsequent refinements. Final refinements of this model by 22 2 using the 1811 independent reflections with F2 ~ 30(F ) and 28 ~ 50° o 0 converged with R1 = 0.0456 and R2 = 0.0713. All parameter shifts

during the final cycle of refinement were less than 0.050. The

"goodness of fit," defined by [Ew(!F! - !F! )2/(n - m)]1/2 where n is o c the number of reflections used in the refinement and m is the number of

refined parameters was 2.917. The overdetermination ratio (n/m) was

14.3. The largest peak in the final electron density map was °_3 0.497 eA.

The temperature factors for the 0 atom of the nitrosyl group

were large, suggestive of possible dis~rder common in non-linear metal 41 nitrosyl complexes. Consequently, intensity data were collected from

the same crystal at 135 K, with the conditions listed in Table 2.3.

Lower supplied power and faster scanning speed resulted in slightly

fewer data being collected at 135 K than at 293 K. The low-temperature

structure was solved by using the coordinates of the non-hydrogen atoms

from the room-temperature model followed by least-squares refinements

and difference electron density maps, which revealed the positions of

all 18 hydrogen atoms. All atoms were then refined to isotropic

convergence. The positions of the hydrogen atoms were then held fixed while the remain~ng atoms were refined to anisotropic convergence. The

final cycle of refinement, based on the 1512 unique reflections having

F~ ~ 30(F~), converged with R1 = 0.0526 and R2 = 0.0544. All parameter

shifts during the final cycle of refinement were less than 0.040. The

"goodness of fit" was 1.808, and the largest peak in the final electron 23 3 density map (1.55 e A- ) was 1.07 Afrom Ni. The overdetermination ratio was 11.9.

Results and Discussion

Complexes of the Formula Ni(N02)2L2

Several square planar dinitro complexes of Ni(II) have been prepared by the reaction of methanolic solutions of nickel nitrite with stoichiometric amounts of the appropriate phosphine ligand.

Electronic spectra are characterized by rapidly rising charge transfer bands extending into the uv with shoulders between 380 and 420 nm.

Although many similar square planar complexes of divalent nickel are known, they primarily exhibit charge transfer spectra and ligand field bands have not been established with certainty.42

The infrared spectra show frequencies characteristic of nitro ligands. Bands due to antisymmetric and symmetric stretching of the l N-bonded nitro ligands are observed between 1374 and 1410 cm- and 1310 -1 and 1340 cm ,respectively (Table 2.4). Free NO; exhibits these modes l 43 at 1250 and 1335 cm- • Therefore, while V (N0 ) changes only slightly, s 2 V (N0 ) shifts to markedly higher frequencies on coordination. These a 2 -1 band assignments were verified by 23-30 cm shifts to lower frequencies observed in the l5N enriched analogs. Other nitro bands 1 occur in the regions 810-820 and 615-620 cm- These bands, which s h1Of t 5 - 16 cm -1 wi t h 15N subOO st1tut10n, ar1se0 f rom teen h b dO1ng and rocking modes of the nitro group, respectively. It is possible to distinguish cis and trans isomers since the former exhibits more banos 24 43 for v and v than the latter. Accordingly, complexes of the mono- a s dentate phosphines are trans while those of the chelating phosphines are cis. It is noteworthy that, while the rocking mode is observed as -1 a moderately strong band between 615 and 620 cm in the trans dinitro complexes, this mode is not observable in the cis analogs. This maybe the result of changing the symmetry from D2h to C in going from trans 2v to cis isomers thus reducing by one the number of infrared active vibrations of the metal coordination sphere in the cis isomers.

In addition to the nitro bands just assigned, several of the dinitro complexes exhibit additional weak absorptions which shift with l l5N substitution (Table 2.5). Bands at 1090-1155 cm- are observed when L = PMe , PEt , 1/2 dppe, or 1/2 cis-vpp while bands at 3 3 l 1240-1275 cm- are observed when L is one of the chelating phosphines.

The former bands can be assigned to nitrito isomers and the latter to 43 chelating nitro isomers by comparison with spectra of known compounds.

Thus the N02 ligand in N~(N02)2L2 complexes exhibits linkage isomerism between the predominant -N-bonded nitro coordination and the nitrito and chelating nitro bonding modes.

The room temperature 3lp{lH} nmr spectra (Table 2.6) of

Ni(N02)2L2 complexes produce a single signal in each case. Except for

Ni(N02)2(PCY3)2' however, the signals are broad indicating that these complexes are fluxional in solution. The 3lp {lH} nmr of Ni(15N02 )2- (PCY3)2 is characterized by a triplet indicating that this complex has trans stereochemistry in solution (J = 7.8 Hz). 15 N-P 25

All of the dinitro complexes are moderately air sensitive in solution. Moreover, Ni(N0 )2(PMePh )2 undergoe~ spontaneous dispropor 2 2 tionation in a variety of solvents to yield two deep blue nitrosyl species over a period of 2-3 hours. When the reaction is monitored by

nmr, a gradual disappearance of the Ni.{N0 2)2(PMePh2)2 signal at -137 ppm is accompanied by the appearance of signals at -117 and

-111 ppm. The signal at -117 ppm initially grows at a somewhat faster rate than that at -111 ppm. On completion of the reaction however, the two species are present in equal concentrations and there is negligible signal due to Ni(N02)2(PMePh2)2~ Infrared spectra of the deep blue products of this reaction are characterized by two nitrosyl bands at l 1700 and 1750 cm- • It was not possible to isolate either species due to similarities in solubility although partial separation was accomplished. Since this reaction was unique to Ni(N0 )2(PMePh )2' it 2 2 was not investigated further.

Each Ni(N02)2L2 complex studied reacted rapidly with CO at ambient temperature and pressure to yield deep blue nitrosyl species.

When L PMePh , characterization of the products is complicated by the = 2 reaction just discussed. In all other cases, however, crystalline complexes of the formula Ni(N0 ) (NO)L were isolated in high yields. 2 2

Complexes of the Formula Ni(N0 ) (NO)L 2 2 31 1 The {H}spectra (Table 2.6) of Ni(N0 ) (NO)L complexes at 2 2 room temperature exhibit single signals which are broad indicating that these complexes are fluxional in solution at room temperatures. 26

Visible spectral data are summarized in Table 2.7. However,

before a detailed description of the electronic spectra is given, a

brief review of the theory concerning the ordering of molecular

orbitals in tetracoordinate {NiNO}lO complexes will be presented. 44 ,45

The orbital ordering is summarized in Fig. 2.2. Molecular

orbitals in this scheme arise from the metal d orbitals and the n*

orbitals of the nitrosyl ligand. Structures of several {NiNO} 10

complexes have been reported. Important bond distances and angles of

these complexes are summarized in Table 2.9. For [Ni(NO)L ]+ 3 complexes of C symmetry (Table 2.9) there are ten electrons in these 3v 2 orbitals. Therefore the HOMO is the filled 4a (d ) level and the LUMO l z is the antibonding 4e[n*(NO),xz,yz] level. Degenerate E orbitals of

(NiNO) result and linear Ni-N-O groups are observed. Replacing one phosphine by an N0 group to give Ni(N0 ) (NO)L complexes, however, 2 2 2 lowers the symmetry to C and lifts the degeneracy of the E levels. l Splitting of the E levels to give AI and A" levels increases the number of allowed electronic transitions. Symmetry allowed electronic transitions are shown in Fig. 2.2 while transition assignments for the observed absorption of Ni(N0 ) (NO)L complexes are summarized in 2 2 Table 2.7.

Infrared spectra (Table 2.8) of tetrahedral Ni(N0 ) (NO)L 2 2 complexes are dominated byv which gives rise to an intense absorp NO l tion band at 1710-1750 cm- • Bands characteristic of nitro and nitrito groups are also observed. In complexes of unidentate phosphine ligands, v decreases with increasing phosphine basicity. An increase NO Table 2.9. Selected bond distances and angles of structurally characterized {NiNO}lO Complexes.

Complex .Syrnrnetry Ni-N(A) N-0(A) Ni-N-O(deg) N-Ni-X(deg) Ni-P(A) Reference

Ni(NO)cp 46 . C5v 1. 58 (1) 1.17(2) 180 [Ni(NO) (TEP)] [BF ] 47,48 4 C3v 1.579(10) 1.199 180 [Ni(NO) (POC)3] [PF ] C 1. 58 (1) 1.12 (1) 176.8(18) P1 2.191(4) 49 4 3v P2 2.176(5) P3 2.191(4) 31 [Ni(NO) (PMe3)3] [PF6] C3v 1. 646 (6) 1.138 (9) 175.4(7) P1 2.239(1) P2 2.229(2) P3 2.239(1)

Ni(N0 2) (NO) (PMe )2 C1 298 K 1.652 (7) 1.158(8) 165.6(8) 126.2(3) P1 2.243(2) 50 3 P2 2.243(3)

Ni(N0 ) (NO) (PMe ) 2 C 135 K 1.648(6) 1.177(7) 166.9(9) 127.8(3) P1 2.228(3) 50 2 3 1 P2 2.244(2)

Ni(NO) (NCS) (PPh3)2 C 1.648(5) 1.159(6) 161.5(5) 116.8(2) P1 2.271(2) 32 l P2 2.238(2)

Ni(N ) (NO) (PPh )2 C 1. 686(7) 1.164(8) 152.7(7) 128.8(3) P1 2.257(2) 33 3 3 l P2 2.306(2)

C 1.666(9) 1.155 (9) 153.4(8) 140.0(4) P1 2.251(3) Table 3.11 [Ni(ONO) (NO)dppe]2 1 P2 2.27l(l~)

N ...... 28

z o N lL:

L~~L * ,,-_ .., ____ T"""T'-4a"(7T (NO), yz ) .... 4e(7T(NO).xz.yz) -=r===--r ----______<...... ''''--'' ...... """"'"hif.r+-ro Sa' (7T °, NO), x z) . 4

~-H------H+T'-'H" 4a'(z~ O""(NO»

...... ---++11 2a'(xz,7To(NO» 2e(xz.yz. 7T *(NO» 1111 ------<_ ...... -.. -..... ---++-11 2a"(yz,7TO(NO»

Figure 2.2. Correlation diagram relating the moleculio orbitals of pseudotetrahedral four-coordinate {NiNO} complexes of C and C symmetry showing allowed electronic transitions. 3v s 29 in phosphine basicity results in greater back donation of electron density to the nickel d orbital which is antibonding with respect to xz NO (Fig. 2.2). Thus phosphines coordinated to nickel indirectly contribute to the population of the antibonding n*(NO) orbital and give rise to the variation in observed v frequencies (Fig. 2.2). In NO complexes of bidentate phosphines, however, the steric constraints imposed by the size of the chelate ring affect orbital overlap and complicate spectral interpretation.

From the preceeding discussion, it is clear that the Ni-N-O angle in Ni(X) (NO) (PR ) 2 complexes is dependent on the geometTY of the 3 complex as well as the nature of the phosphine and X- ligands. In complexes of C or higher symmetry, the NiNO group is rigorously 3v linear while slightly nonlinear NiNO linkages are observed in complexes of lower symmetry. Moreover, inequivalent Ni-P bond lengths have been observed in low symmetry {NiNO}10 complexes (Table 2.9). As a further probe into the relationship among stereochemistry, ligand environment,

Ni-N-O angle, and Ni-P distance, the structure of the low symmetry

Ni(N0 )(NO)(PMe )2 complex has been determined at 298 K and 135 K. 2 3

Description of Structure

Final structural parameters for the nonhydrogen atoms of

Ni(N0 ) (NO) (PMe )2 at both 293 and 135 K are found in Tables 2.10 and 2 3 2.11. The interatomic distances and angles found for the nonhydrogen atoms at both temperature~ are set out in Table 2.12. A perspective Table 2.10. Atomic positional parameters for the nonhydrogen atoms of Ni(N0 ) (NO) (P(CH )3)2' 2 3

x ~ z atom 293 K 135K 293 K 135K 293 K 135 K Ni 0.2329 (1) 0.2394 (1) 0.15156 (6) 0.14685 (7) 0.18590 (5) 0.18337 (6) P(l) 0.3052 (2) 0.3073 (2) 0.1491 (1) ·0.1461 (1) 0.3422 (1) 0.3425 (1) P(2) -0.0535 (2) -0.0497 (2) 0.1842 (1) 0.1821 (1) 0.1623 (1) 0.1618 (1) N(l) 0.3131 (8) 0.3172 (7) 0.2983 (5) 0.2958 (5) 0.1613 (4) 0.1582 (4) N(2) 0.2875 (9) 0.2960 (8) 0.0422 (5) 0.0352 (5) 0.1332 (4) 0.1313 (4) 0(1) 0.2197 (8) 0.2207 (6) 0.3785 (4) 0.3763 (4) 0.1640 (4) 0.1628 (4) 0(2) 0.4576 (7) 0.4677 (6) 0.3099 (5) 0.3103 (4) 0.1350 (5) 0.1329 (4) 0(3) 0.2977 (14) 0.3129 (10) -0.0436 (6) -0.0534 (5) 0.1055 (7) 0.1047 (5) C(l) 0.1739 (11) 0.1719 (10) 0.0638 (7) 0.0606 (6) 0.4066 (5) 0.4099 (5) C(2) 0.2977 (11) 0.2959 (10) 0.2758 (5) 0.2761 (6) 0.3986 (5) 0.3987 (5) C(3) 0.5214 (10) 0.5279 (9) 0.1048 (7) 0.1026 (6) 0.3815 (6) 0.3851 (6) C(4) -0.1363 (9) -0.1329 (9) 0.2733 (7) 0.2719 (7) 0.2463 (5) 0.2484 (5) C(5) -0.1290 (9) -0.1236 (9) 0.2437 (6) 0.2442 (6) 0.0502 (5) 0.0482 (5) C(6) -0.1908 (12) -0.1935 (10) 0.0700 (7) 0.0667 (6) 0.1639 (7) 0.1632 (6)

a x,y, and z are in fractional monoclinic coordinates.

w o 31

Table 2.11. Thermal parameters for the non-hydrogen atoms of a Ni(N02) (NO) (P(CH3)3)2.

Atom T(OK) B B B11 B22 B33 B12 13 23

Ni 293 5.09(4) 4.26(4) 3.33(3) 0.13(3) 0.45(3) -0.11(3) 135 1.35(4) 1.18(4) 1.85(4) 0.09(3) 0.18(3) -0.07(4) P(l) . 293 5.51(8) 3.89(7) 3.50(6) 0.19(6) 0.21(6) 0.12(5) 135 1.49 (7) 1.13 (7) 1. 76(7) 0.05(6) 0.07(6) 0.00(7) P(2) 293 4.88(8) 5.22(9) 4.40(7) -0.69(6) -0.02(6) 0.68(6) 135 1.36 (7) 1.51(8) 1. 98 (8) -0.23(6) -0.02(6) 0.20(6) N(l) 293 5.9(3) 6.5 (3) 4.1(2) -0.6(3) 0.3(2) 0.6(2) 135 1.5 (3) 1. 9 (3) 2.2(3) -0.4(2) -0.0(2) -0.2(2) N(2) 293 10.7(5) 6.6(4) 5.3(3) 0.1(3) 1. 9 (3) -0.9(3) 135 2.3(3) 1. 9 (3) 3.1(3) -0.1(2) 0.8(2) -0.5(2) 0(1) 293 9.2(4) 5.1(3) 11.0(4) 0.7(2) 3.7(3) 0.6(3) 135 2.5(2) 1. 2 (2) 4.0(3) -0.1(2) 0.8(2) -0.0(2) 0(2) 293 5.2(3) 9.2(4) 11.8(5) -1.4(2) 2.6(3) 2.0 (3) 135 1. 2 (2) 3.5(3) 4.9(3) -0.7(2) 0.7(2) 0.4(2) 0(3) 293 26.2(10) 6.6(4) 15.5 (7) 1.1(5) 5.4(6) -5.4(4) 135 7.3 (5) 2.1(3) 7.3(4) -0.1(3) 1.8(3) -2.2(3) C(l) 293 9.4 (5) 7.2(5) 4.8(3) -2.4(4) 1. 0 (3) 0.4(3) 135 3.2(4) 1.8(3) 2.5(3) -0.1(3) 0.9(3) -0.4(3) C(2) 293 10.4(5) 4.4(3) 4.7(3) 0.9(3) -0.3(3) -0.3(3) 135 2.7(4) 1.8(3) 1. 9 (3) 2.9(3) -0.1(3) -0.2(2) C(3) 293 7.5(4) 7.6(5) 6.3(4) 2.3 (4) -0.1(3) 0.7(4) 135 2.1(3) 2.1(3) 2.9(4) 0.5(3) 0.1(3) 0.2(3) C(4) 293 5.4(4) 8.4(5) 6.2(4) 1.5(3) 0.4(3) 0.1(3) 135 1.1(3) 3.3(4) 2.7(3) 0.0(3) 0.1(3) -0.1(3) C(5) 293 6.1(4) 7.8(5) 5.1(3) 0.1(3) -0.6(3) 0.5(3) 135 1. 9 (3) 2.6(4) 2.2(3) -0.2(3) 0.1(3) 0.3(3) C(6) 293 9.1(5) 8.4(6) 9.7(6) -3.3(4) -0.4(4) 2.1 (5) 135 2.3(3) 2.6(4) 3.5(4) -1.1(3) 0.2(3) 0.4 (3)

2 a Anisotropic thermal parameters are in the form exp [-1/4 (B h a, *2+ i1 2 *2' 2 *2 * * * * k1b '* B22k b + B331 c + 2B12hka b + 2B13h1a c + 2B23 c)]. a Table 2.12. Interatomic distances and angles for Ni(N0 2) (NO) (PMe3)2.

Distances atoms 293 K 135 K atoms 293 K 135K Ni-P(l) 2.243(2) 2.228(3) P(l)-C(l) 1.808(7) 1.814(7) Ni-P(2) 2.243(2) 2.244(2) P (1) -C (2) 1. 795 (7) 1. 807 (7) Ni-N(l) 1.997(6) 1.992(6) P(1)-C(3) 1. 791(8) 1.813(7) Ni-N(2) 1.652 (7) 1.648(6) P(2)-C(4) 1.820(8) 1.810(8) N(l)-o(l) 1.247(7) 1. 252 (7) P(2)-C(5) 1. 805 (7) 1.803(7) N(1)-o(2) 1. 233 (7) 1. 251 (7) P(2)-C(6) 1.792(8) 1.812 (7) N(2)-o(3) 1.158 (8) 1.177(7)

Angles atoms 293K 135 K atoms 293 K 135 K N(1)-Ni-N(2) 126.2(3) 127.8 (3) C(1)-P(1)-C(2) 104.4(4) 104.2(3) N(l)-Ni-P(l) 98.3(1) 97.8(2) C(1)-P(1)-C(3) 102.6(4) 102.5(4) N(1)-Ni-P(2) 97.3(2) 95.8(2) C(l)-P(l)-Ni 115.4(2) 115.6(3) N(2)-Ni-P(1) 113.4(2) 113.0(2) C(2)-P(1)-C(3) 102.5(5) 102.1(3) N(2)-Ni-P(2) 112.8(2) 114.0(2) C(2)-P(1)-Ni 114.8(2) 114.3(2) P(1)-Ni-P(2) 106.0(7) 105.1(1) C(3)-P(1)-Ni 115.5(3) 116.2(3) 0(1)-N(1)-0(2) 117.7(6) 117.6(6) C(4)-P(2)-C(5) 103.2(3) 102.8(3) Ni-N(l)-o(l) 123.5(5) 123.4 (4) C(4)-P(2)-C(6) 103.3(4) 103.2(4) Ni-N(1)-0(2) 118.5 (5) 119.0(5) C(4)-P(2)-Ni 115.8(2) 116.1(2) Ni-N(2)-0(3) 165.5(8) 166.9(6) C(5)-P(2)-C(6) 102.5(4) 102.4(4) C(5)-P(2)-Ni 114.8(2) 114.7(2) ______Ci6L-~(~):N! ___ _1!5~4i3L __ _1!5~7i3L _ Dihedral Angles atoms 293 K 135 K atoms 293 K 135 K Ni-N(1)-N(2) vs Ni-P(1)-P(2) 90.4 91.2 Ni-P(1)~P(2) vs Ni-N(2)-o(3) 82.8 88.0 Ni-N(2)-0(3) vs Ni-N(1)-N(2) 11.5 4.8 Ni-P(1)-P(2) vs N(1)-0(1)-0(2) 102.0 99.3 N(1)-O(1)-0(2) vs Ni-N(1)-N(2) ·32.7 32.8 W N a Distances in angstroms and angles in degrees. Standard deviations for the distances and angles were calculated by using a variance-covariance matrix. 33 view and the numbering scheme for the molecules at 135 and 293 K are

shown in Fig. 2.3 and Fig. 2.4. Theroot-mean-square amplitudes of vibration for the nonhydrogen atoms are listed in Table 2.13 while the

thermal and positional parameters for the hydrogen atoms are found in

Table 2.14.

At both temperatures, the crystal structure consists of

discrete molecules of Ni(N0 ) (NO) (PMe ) 2 with the closest inter- 2 3 molecular contacts of 2.761 and 2.437 A° found between 0(1) and lH(l).

The crystal packing of these molecules at 135 and 293 K is shown in

Fig. 2.5. At 135 K, the non-bonded intramolecular distances are of the order of van der Waals distances, and all hydrogen-hydrogen contacts are 2.378 Aor greater. No evidence was found for interaction between the oxygen atoms of adjacent molecules with nickel atoms of adjoining molecules.

Room Temperature Form. The coordination geometry about the nickel atom is significantly distorted from tetrahedral. Although the

P(1)-Ni-P(2) angle of 106.16(7)° is only slightly smaller than that of a regular tetrahedron, the N(1)-Ni-N(2) angle of 126.2(3)° is considerably greater. The dihedral angle of 90.4° between the P-Ni-P and N-Ni-N planes differs only slightly from the idealized angle of

90° for the regular tetrahedron. The two Ni-P bond lengths are the same (2.243(2) A).° The Ni-N(O) bond length of 1.652(2) 0A is among 51 the shortest observed for transition-metal nitrosyl complexes but is . 1 f { }10 1 32,33,49,52-56 rath er typ~ca or NiNO comp exes. These very short distances are indicative of considerable covalent bonding between 34

Perspective view and numbering scheme for Ni(N0 ) (NO) Figure 2.3. 2 (PMe )2 at 135 K. -- The the~a1 ellipsoids are 50% 3 probab~lity ellipsoids. 35

Figure 2.4. Perspective view and numbering scheme for Ni(N02) (NO) (PMe )2 at 293 K. -- The thermal ellipsoids are 50% 3 probab~lity ellipsoids

/ Table 2.13. ° Root-mean-square amplitudes of vibration (in A) for Ni(N02) (NO) (P(CH3)3)2.

Minimum Intermediate Maximum Atom 293°K 135°K 293°K 135°K 293°K 135°K

Ni .205(1) .119(2) .232(1) .133(2) .255(1) .153(2) P(l) .209(2) .119(4) .223(2) .137(3) .267(2) .151 (3) P (2) .225(2) .123(4) .235(2) .140(4) .283(2) .165(3) N(l) .224 (7) .125(13) .263(7) .166(11) .301(8) .170(10) N(2) .241(8) .146(13) .300(8) .163(11) .369(8) .208(10) 0(1) .251 (7) .122(12) .301(6) .171(8) .399 (7) .228(8) 0(2) .231(7) .104(12) .334(7) .214(8) .406 (7) .251(8) 0(3) .214(9) .128(14) .469(10) .287(9) .581(11) .332(9) C(l) .238(9) .140(14) .274(9) .169(12) .373(9) .212(12) C(2) .230(9) .145(14) .244(8) .154(14) .373(9) .194(11) C(3) .241(9) .139 (14) .299(9) .183(12) .356(10) .193(12) C(4) .245 (9) .116(15) .283(9) .186(12) .337 (9) .204(12) C(5) .235(9) .154(13) .300(9) . 161(13) .316(9) .188 (13) . C(6) .257(10) .128(15) .324(10) .199(13) .422(11) .223(11)

W 0\ a Table 2.1/,. Thermal and positional parameters for the hydrogen atoms of Ni(N0 ) (NO) (P(CH )3)2' 2 3

x y. z B,AZ atom 293 K 135 K 293 K 135 K 293 K 135K 295 K 135K lH(l) 0.2322 0.2120 0.05R6 0.0835 0.4804 0.4916 5.00 6.58 lH(2) 0.1632 0.1945 -0.0068 0.0080 0.3812 0.3872 5.00 1.59 lH(3) 0.0560 0.0626 0.0908 0.0711 0.4065 0.3805 5.00 2.10 21I(1) 0.3331 0.3200 0.2777 0.2776 0.4511 0.4643 5.00 2.15 2H(2) 0.1789 0.1711 0.2988 0.3055 0.3942 0.3800 5.00 4.53 2H(3) 0.3618 0.3895 0.3257 0.3174 0.3671 0.3605 5.00 2.92 3H(1) 0.6096 0.6138 0.1365 0.1551 0.3496 0.3638 5.00 3.32 3H(2) 0.5320 0.5344 0.0298 0.0460 0.3739 0.3703 5.00 2.84 3H(3) 0.5523 0.5341 0.1209 0.1110 0.4472 0.4561 5.00 0.67 4H(1) -0.2394 -0.2354 0.2748 0.2851 0.2287 0.23Q3 5.00 0.11 4H(2) -0.0877 -0.0699 0.3451 0.3323 0.2400 0.2500 5.00 1. 70 4H(3) -0.1077 -0.1119 0.2518 0.2395 0.3084 0.3223 5.00 2.31 5H(1) -0.2457 -0.2315 0.2486 0.2529 0.0541 0.0526 5.00 7.42 5H(2) -0.1057 -0.0976 0.2032 0.2044 0.0014 -0.0136 5.00 1. 75 5H(3) -0.0847 -0.0593 0.3141 0.2942 0.0484 0.0405 5.00 2.62 6H(1) -0.2675 -0.3229 0.0818 0.0985 0.1616 0.1431 5.00 5.98 6H(2) -0.1585 -0.1858 0.0291 0.0369 0.2210 0.2301 5.00 2.03 6H(3) -0.1739 -0.1854 0.0229 0.0190 0.1121 0.1099 5.00 4.16 a In the case of the structure determination at 25°C, all hydrogen atoms labeled with a "(1)" were located on a difference electron density map while those labeled with a "(2)" or a "(3)" were placed at geometrically idealized ~ositions. All of these hydrogen atoms were assigned an isotropic temperature factor, B=5.00A. All hydrogen parameters were held fixed during refinement in the case of structure determination at 25°C. For the structure determination at -138°C, all hydrogen atoms were located on a series of difference electron density maps. Parameters for these hydrogen atoms were allowed to vary while the structure was refined to isotropic convergence. Parameters were then held fixed while the structure was converged with anisotropic w thermal parameters on all nonhydrogen atoms. --..J (A) . (8)

Figure 2.5. Packing diagram for Ni(N0 ) (NO) (PMe ) 2 at 293 K (A) and 135 K (B). 2 3 w ():) 39

nickel and the nitrosyl ligand. The nitrosyl ligand is nearly coplanar

with the Ni-N(1)-N(2) plane [0(3)-N(2)-Ni/N(1)-Ni-N(2) = 11.5°,

Table 2.12].

The Ni-N-O angle of 165.5(8)° is also similar to those found

~for the closely related low-symmetry triphenylphosphine complexes

Ni(NCS)(NO)(PPh )2 (161.5(5)°) and Ni(N ) (NO) (PPh ) 2 (152.7(7)°), 3 3 3 which are significantly less than the 180° observed for {NiNO}lO

comp 1 exes W1t" h C3v or h"19 h er symme t rYe 32,33,49,52-56 Th e N"1- N(O 2 ) b on d

distance of 1.997(6) A° is 0.107 A° longer than the 1.890(1) A° found in

Ni(N02)2dppe (Table 3.6, page 64). The NO group is tipped 30° out of

the plane formed by N(1)-Ni-N(2), removing any semblance of a mirror

plane perpendicular to P(1)-Ni-P(2). The average Ni-P distance in

Ni(N0 )(NO)(PHe )2 is also 0.063 A° longer than that found for Ni(N0 )2- 2 3 2 dppe (Table 3.6, page 64), indicative of a larger radius for the

tetrahedral nickel atom of Ni(N0 ) (NO) (PMe )2 compared with that of 2 3 square-planar nickel complexes. The radius observed for tetrahedral

o Ni(II) complexes is 0.05 A longer than that of isomeric square-planar 57 . complexes. The radius 'found for Ni (0) complexes is even shorter than '

for square-planar Ni(II).58-60 Within experimental error, all of the

C-p-c distances and angles are the same and do not differ significantly

from those of the five-coordinate nickel complexes reported

preV10US" 1 y. 61

Low Temperature Form. The structure found for Ni(N02) (NO)- (PMe3)2 at low temperature (LT) is similar in most essential details

to that found at ambient temperature but the reduced thermal parameters, 40

especially of the nitrosyl ligand, provide a clearer view of its molecular geometry. With the exception of Ni-P(l) , the LT structure has interatomic distances that are indistinguishable from those of the o room temperature (RT) structure. The Ni-P{l) distance of 2.228(3) A

found for the LT structure differs from that found for the RT structure by 5-80, while the Ni-P(2) distance of 2.244(2) A for the LT structure is unchanged. The bond angles and dihedral angles also differ little between the LT and RT structures except for the dihedral angle of

O(3)-N(2)-Ni/N(1)-Ni-N{2), which decreases from 11.5 0 at ambient temperature to 4.8 0 at 135 K (Table 2.12).

Structural Interpretation

The results outlined above establish that the N0 group is 2 bonded to nickel through the nitrogen atom and that the nickel atom has distorted tetrahedral geometry. The nickel nitrosyl group has a slightly bent Ni-N-O linkage well within the range found for other

{NiNO}lO complexes with less than 3-fold symmetry axes (Table 2.9). The

Ni-N-O groups in the other two closely related triphenylphosphine complexes, Ni(NO){NCS)(PPh ) and Ni(N ) (NO) (PPh )2' are also similarly 3 3 3 bent [161.5(5) and l52.7(7)~, respectively] in a plane that is nearly coincident with the N-Ni-X plane.32 ,33 The bending of the Ni-N-D linkage is associated with the loss of n bonding in the direction of " 44 t h e pane 1 0 f b en d1ng. The unequal n bonding results from the disparate donor-acceptor properties of the other ligands attached to the metal, in this case PMe and NO;. 3 41

As was found for the other members of this NiX(NO)L series, 2 there are no inter- or intramolecular contacts with the NiNO group that

are close enough to cause such a large deformation. If the bending of

the NiNO group is indeed due to electronic effects, then some correla-

tion between the Ni-N distance and the NiNO angle would be expected

since the loss of pi bonding should also be accompanied by a

lengthening of the Ni-N bond. There are now a sufficient number of 5l structural data for the {NiNO}lO complexes to provide a test of this hypothesis. A plot of the Ni-N distance vs. NiNO angle for these complexes is shown in Fig. 2.6. Although the Ni-N distance appears to increase as the NiNO angle decreases in these {NiNO}lO complexes, this possible relationship should be viewed with caution since similar plots for nitrosyl complexes of other transition metals fail to evoke such 51 patterns.

The most unusual feature of the structure of these NiX(NO)L 2 complexes, however, is the apparently different Ni-P distances in each

(Table 2.9). Although the two Ni-P distances found for Ni(N0 ) (NO)- 2 o (PMe )2 at room temperature are identical [2.243(2) A], they differ by 3 o 5-80 at low temperature [2.228(3) and 2.244(2) A]. If Ni(N0 ) (NO)- 2 (PMe )2 were the only complex to exhibit inequivalent Ni-P distances, 3 the difference might be attributed to an experimental artifact. How- ever, in view of the fact that each of these complexes has nonequivalent Ni-P distances, it seems likely that these differences are significant. Each of these molecules has only C symmetry, so that 1 nonequivalent Ni-P distances are permissible, but as the discussion in 42

L70 I- -f -f L65 ... l~ +' 15+ -f1 ..z fll z ~ ;11 C en- '.p »...:.t L60 I- Z (') 1'1'1 4-t -» - 1.55 I-

1.501-

..L I , 180 170 160 150 140

Ni-N-O ANGLE (OEG)

o Figure 2.6. Ni-N (A) vs Ni-N-O ay~le (deg) for structurally characterized characterized {NiNO} complexes.

1 = [Ni{NO)TEP][BF ]49, 2 = [Ni(NO)(POC)3][BF ]52, 3,4 = [(C H 0)7Na ] 4 4 4 8 t{Ni(NO)}2(~2-Me2pz)3]55, 5 = [Ni(NO)(Me Pz)]254 , 6 = Ni(NO) (cp)4 ,48, 2 7,8 = [Et4N][{~i(NO)2}2(~3-I)(~2-Ne2pz)2]55, 9 = [Ni(NO)(Me2Pz)2]2Ni54, 10 = [Ni(NO)(NP3)][BPh4]5 , 11 = Ni(NO) (GAP)56, 12 = Ni(NO) (NCS) 32 this 33 (PPh3)2 , 13 = [Ni(ONO) (NO)dPpe]2 work, 14 = Ni(N ) (NO) (PPh )2 50 3 3 15 = Ni(N0 ) (NO) (PMe )2 , 16 = [Ni(NO)(PMe )3][Pf ]36. 2 3 3 6 43 references 32 and 33 points out, there are no obvious sources for the disparity in the Ni-P distances due to inter or intramolecular contacts with other ligands; so it appears the nonequivalence must be electronic in origin. The present complex is particularly amusing since the RT structure has equivalent Ni-P distances but has a less planar

O-N-Ni-N grouping than the LT form, which has a more nearly planar

O-N-Ni-N grouping but inequivalent Ni-P distances.

Each of the low symmetry NiX(NO)L complexes has a slightly 2 bent NiNO linkage while {NiNO}lO complexes of C or higher symmetry 3v have linear NiNO groups (Table 2.9). These results are in accord with the molecular orbital diagram in Fig. 2.7. 44 The molecular orbitals in this diagram arise from the metal d orbitals and the n* orbitals of NO.

In this molecular orbital description, the HOMO consists of the filled

4a (d )2 orbital of the {NiNO}lO group and the LUMO is the antibonding l z 4e(n*(NO),xz,yz). The alternative form for a four-coordinate {NiNO}lO complex is planar, with a strongly bent NiNO group, but this form has not been observed to date. Attainment of square-planar geometry requires 4a (d;) to be hibher in energy than 4e(n*(NO),xz,yz), which l in turn would produce a strongly bent {}INO}lO group. It is clear that although the geometry of these Ni(X) (NO)L complexes is distorted,. the 2 coordination geometry still approximates tetrahedral. Consequently, 2 we conclude that 4a (d ) is the HOMO but that 4e(n*(NO),xz,yz), the l z LUtl0, is rather close in energy, a conclusion supported by the intense absorption bands in the visible region (Table 2.7). 44

o I N L~'L L

501(% 2_y.2, «X(NOl,xz) ~a·(,1{NO).yz) __ ._ .f"(ify(NO),YZ) 48.(d(NO)~,yz) ,,,,-/ / ---( . "'''~a'(rr(NO),XZ)" ---:,...l: __4a~(xz,.... _r----- sp2(N),z2-y2)

40 (z2.c1'(NO» 1 2 ' / ...... !.-----_1 . ____ 40 (z • O-(NO)y

3oll(x2 _.y2) 3e(xy, ,,2 _y2) ,,,., ====-----< 11 "'. 3a' (xy) 30 (xy) '---~--- 2a' (xz,rt(NO» ~=~_-_-=- ?a'(apZ(Nl,xz) ". 2e(xz. yz,fr- (NO» ---<_,," ...... 2~· (JZ,.,r(NO» ------2a ·(yZt1T~. (NO»

Figure 2.7. Correlation diagram relating the molecular ~obitals of pseudotetrahedral four-coordinate {NiNO} comple:{es of C and C symmetro to those of four-coordinate 3 squar~ planaf {NiNO) complexes of C symmetry. s 45

In [Ni(NO)L ]+ complexes of C symmetry, the two n* orbitals 3 3v of NO are degenerate and linear NiNO gr~ups are predicted. Replacing one phosphine by an X- group however, lifts the degeneracy of the e orbitals and results in a bending of the NiNO group away from the region of highest electron density. Fig. 2.8 shows the angles the various sigma donating X- groups form with the Ni-N(O) axis. The axis of the d orbital which is n antibonding to the nitrosyl group is also shown. xz Donation of electron density to this orbital results in bending of the

NiNO linkage associated with loss of n bonding in the direction of the bending. As shown in Fig. 2.8, when the N-Ni-X angle more closely approaches the Ni-N-d angle of 135°, the orbital overlap increases, xz thereby increasing the efficiency of electron donation to the n* orbitals of NO. In all cases, this causes the Ni-N-Q linkage to be bent in a plane which is nearly coincident with the N-Ni-X plane.

In NiX(NO)L complexes, the impending crossing of the filled 2 2 4a (d ) orbital which has a single potential minimum, with the 4e(n*- l z (NO),xz,yz) orbital, which has a double potential minimum, would result in their mixing and in·consequent distortion of the NiNO group and distortion in the coordination geometry at the nickel atom in such a way as to remove the degeneracy of the 4e(n*(NO),xz,yz) orbital. Since in the tetrhaedral geometry, the 4e(n*(NO)xz,yz) is antibonding with respect to the two phosphorus ligands, removal of their degeneracy could then be reflected in unequal Ni-P bond lengths. Clearly, the presence of the X- ligands in these complexes has already removed the 46

z ~x o

• • ::N~RBI~AL ~N~-

/ , , / " dxz·ORBITAL / ANTIBONDING , .; VJS-A-VIS NCS,N02 .~ N3 / Nes of Ni (NO)(NCS)(PPh3~ , N02 of NI(NO)(N02)(PMe3)2 / '\: , , N3 of Ni(NO)(N3)(PPh3)2 / 'd xz AXIS (l35°r N02 of. ~i(NO)(N02)DPP~ 2

Figure 2.8. U-l'1i-X angles for Ni (NO)(X)(P.R ) 2 complexes. 3 47

degeneracy of the 4e orbital, but each component will interact

differently with 4a1"

In Ni(N02)(NO)(~le3)2' the presence of the TI-acceptor orbital

on the N0 ligand should serve to stabilize the bonding with one or the 2 other phosphorus ligands, provided the TI* orbital on the N02 group is

in the same plane as the sigma orbital of the phosphorus ligand. In

both the RT and the LT structures, the N02 group forms a dihedral angle

of 80 0 with the P(l)-Ni-N(l) plane placing the TI* orbital of the N02

group within 100 of the plane containing the sigma orbital of the P(l)

ligand. Thus, the met~ica1 details of the LT structure are in accord w1to h t h e mo 1 ecu 1 ar orbOt 1 a 1 d escr1pt10n.0 0 44 A similar suggestion was 32 33 forwarded for the triphenylphosphine complexes.' The RT

structure has two Ni-P distances that are the same as the longer

Ni-P(2) distance in the LT structure. This fact suggests that Ni(N02) (NO) (PMe ) 2 may have two electronic states of similar energy with 3 slight differences in their geometries. Some confirmation of this point

of view is found in the change of orientation of the NO group as the

temperature is decreased [Ni-N-O/N(1)-Ni-N(2) is 11.50 at RT and 4.8 0

at LT].

Conclusions

A series of Ni(N02)2L2 complexes have been synthesized and

characterized. The N02 groups in these complexes are primarily

N-bonded although there is some infrared evidence for nitrito coordina-

tion. Each complex reacts with CO under ambient conditions of 48 temperature and pressure to yield nitrosyl species of the formula,

Ni(N0 ) (NO)L which have been similarily characterized. The structure 2 2 of one of these complexes, Ni(N0 )(NO)(PMe )2' has been determined by 2 3 x-ray crystallography. The complex has a distorted tetrahedral geometry about the nickel atom, an N-bonded N0 2 group and a slightly non-linear Ni-N-D linkage. Equivalant Ni-P distances are observed at

RT while these distances are different at LT. These and similar results observed for related complexes of the formula Ni(X) (NO)L2 have been interpreted in terms of the molecular orbital scheme shown in

Fig. 2.7. CHAPTER 3

CRYSTAL STRUCWRES Of Ni(N02)2dppe AND [Ni(ONO)(NO)dppe]2

Introduction

The reaction of CO with square planar Ni(N02)2L2 complexes of tertiary phosphines has been shown to yield the corresponding nitro . 11 21 nitrosyl complexes. according to react10n 3.1. '

(3.1)

The reaction of CO with Ni(N02)2dppe was investigated in some detail as a representative example of this class of reactions. A complete understanding of the mechanism of this reaction requires knowledge of the geometries of both reactant and product; Therefore, the present study was initiated to determine the crystal structure of Ni(N02)2dppe and the product from its reaction with CO, Ni(N0 )(NO)dppe. 2 These complexes are of structural interest because of the constraints imposed by the dppe ligand on the coordination geometry of the nickel atom. Although the 89° bite angle of che1ating dppe is compatible with the cis square planar geometry adopted by NiX dppe 2 complexes, it imposes considerable strain on the tetrahedral geometry common for {NiNO}10 and Ni(O) complexes of tertiary phosphines. While 63 a variety of related complexes such as Ni(Co)2dppe62, Ni(dPpe)2 , and 64 NiC1(NO)dppe are claimed to be tetrahedral monomers with che1ating phosphine ligands based on spectroscopic studies of their solutions,

49 50

there are no supporting crystallographic data available for these

complexes.

Additional interest in the structures of the title compounds arises from the structural versatility of the N02 ligand. This ligand most commonly functions ,as a unidentate nitro group although it can also act either as a unidentate or bidentate nitrito group.

Complexes having monodentate nitro groups are quite common and have been well-documented crysta110graphica11y.65 There is less direct evidence for the alternate coordination modes of N0 ligands. In 2 addition, there are almost no crystallographic data on complexes of unidentate nitrito groups although numerous complexes of this type 65 have been claimed on the basis of spectroscopic data.

The crystal structures of Ni(N02)2dppe and {Ni(ONO)(NO)dppe}2 have been investigated in order to determine the coordination geometry about the nickel atom and to determine the bonding mode of the N0 2 ligands in these complexes.

Experimental

All samples were manipulated under an atmosphere of dry nitrogen using standard Sch1enk tube techniques. Solvents were reagent grade and purified by distillation under dry nitrogen immediately prior to use according to published procedures. 34 Solution molecular weights in 1,2 dich1oroethane were determined by vapor phase osmometry using a

Hewlett Packard model 302B instrument calibrated with dppe in

1,2 dich1oroethane. 51

Preparation of Complexes

Ni(N02)2dppe was prepared according to the procedure outlined

in Chapter 2. Mol Wt: Cal (found) 548.65 (560±20). Suitable needle

shaped crystals of this complex for x-ray structure determination were grown by vapor diffusion over a period of four days using dichloro methane/diethyl ether. The complex crystallizes with one molecule of dichloromethane per molecule of nickel complex. To prevent evaporation of this solvent and consequent crystal decomposition, the crystal was coated with epoxy resin and mounted in a hollow glass capillary under a stream of nitrogen prior to data collection.

[Ni(ONO) (NO)dppe]2 was prepared by Method 1, (Chapter 2).

Mol Wt: Cal (found) 532.65 (535±20).

A solution of Ni(N0 ) (NO)dppe in dichloromethane was 2 generated in situ from the reaction of Ni(N02)2dppe with co. Vapor diffusion of diethyl ether into this solution over a period of two weeks yielded dark blue plate-shaped crystals suitable for data collection. Several crystals of appropriate size were coated with cyanoacrylate resin, mounted in hollow glass capillaries and sealed under nitrogen prior to data collection.

Collection and Reduction of X-Ray Intensity Data

Each crystal was mounted on a Syntex P2 autodiffractometer l controlled by a Nova 1200 computer. The longest dimension of each crystal

Ni(N02)2dppe, the intensities of the standard reflections exhibited an average decline of 12% during the first fifth of the data collection and a 3% decline thereafter. The average linear decomposition for the standards of [Ni(ONO) (NO)dppe]2 was 10%.

Solution and Refinement of Structures

Neutral-atom scattering factors for the nonhydrogen atoms and corrections for the anomolous dispersion made for the nickel, phosphorus, and chlorine atoms were taken from the tables of Cromer and 38,39 Wa b ere Hydrogen atom scattering factors were taken from the 40 calculations of Stewart, Davidson, and Simpson. The structures were refined by full-matrix least-squares techniques minimizing the function 2 22 22 Ew( IIF - 11F) where w = 4F /[0 (F ) + (pF )] • The factor to prevent o coo 0 overweighting of strong reflections, p, was set equal to 0.03. The discrepancy indices, RI and R , are defined in the usual manner. All 2 computations were carried out at the University of Arizona Computer

Center using the programs listed in Chapter 2. Table 3.1. Crystallographic data for [Ni(ONO) (NO) «C6Hs ) 2PC1l2CH2P'(C6Hs) 2)] 2 and Ni (N0 2) 2- «C6Hs)PCH2CH2P(C6Hs)2)-CH2C12'

molecular formula [Ni(ONO) (NO) «C6Hs)2PCH2CH2P Ni(N02)2«C6Hs)2PCH2CH2P (C 6Hs)2)]2 (C6Hs)2·CH2C12 molecular weight 1065.3 548.65 crystal shape plate rectangular needle crystal dimensions, mm .42 x .33 x .06 .75 x .30 x .08' crystal system monoclinic monoclinic space group P2/n P2/c cell dimensions b a,A 15.528(10) 11.880(6) b,A 11.897 (7) 15.191(8) c,A 15.023(13) 17.162(6) ,deg 110.71(5) 108.79(4) V,A3 2596. (3) 2932(2) z 2 4 c -3 db' g cm 1.37 1.46 o s -3 1.36 1.44 d cal' c g cm radiation, A (Mo Ka) 0.71073 (Mo Ka) 0.71073 monochromator graphite crystal graphite crystal supplied power 50 kV, 30 rnA 50 kV, 30 rnA data collection method -28 scan -28 scan In W Table 3.1. Crystallographic data--Continued.

-1 scan speed, deg min variable (3.0-29.3) determined variable (2.0-29.3) determined as a function of peak intensity as a function of peak intensity scan range (2 ), deg MoKa -1.0 to MoKa +1.0 MoKa -1.0 to MoKa1 +1.0 1 2 1 ratio of total background 0.5 0.5 time to peak scan time standard reflections (052), (202), (-130) after (006), (-504), (-320) after every 97 readings every 97 readings decomposition of standards 10% 15% 26 limit, deg 4.0-50.0 4.0-50.0 no. of unique data 5033 5612 no. of data used in the calculation 2249 1 ~ 30(1) 2508, 1 ~ 30(1) -1 Absorption Coefficient (~), cm 9.64 (MoKa) 10.03 (MoKa) a The standard deviation of the least significant figure is given in parentheses in this and the following tables. b Cell dimensions were obtained from a least-squares refinement of setting angles of 15 reflections in the 26 range 9 to 21. c Density was. determined by flotation using a solution of hexane and carbon tetrachloride.

\JI ~ 55

Ni(N0 1 dppe·CH C1 • The structure was solved by the heavy atom 2 2 2 2 method in which the position of the nickel atom was readily determined

from a three dimensional Patterson map. The remaining 37 nonhydrogen

atoms were located by successive refinements and difference electron

density maps. This structure was then refined to isotropic convergence. 2 At this point, the unusually high temperature factor (B = 13.6A ) of

the dichloromethane carbon, C(3), suggested a possible disorder

involving this atom. To investigate this possibility, the contribution of this carbon atom was removed from the calculated structure factors and five fourier planes were generated between the two chlorine atoms and perpendicular to the C1(1)-Cl(2) vector. These planes were at the o midpoint of the Cl(1)-C1(2) vector and at intervals of 0.25 A on either side of this point. Although there was considerable electron density in these planes, there was only one clearly defined peak. The position of highest electron density was therefore assigned as C(3).

The carbon atoms of the phenyl rings were then treated as rigid groups o (C-C 1.383 A, D -6/mmm symmetry) with individual isotropic thermal = 6h parameters. Refinement of the structure with an isotropic thermal parameter for C(3) and anisotropic thermal parameters for all remaining nonhydrogen non-group atoms converged with Rl = 0.083 and

R2 ~ 0.089. The hydrogen atoms of the nickel complex were then placed o at geometrically idealized positions (C-H = 1.08 A), assigned an iso- 02 tropic thermal parameter (B = 5.0 A ) and held fixed in subsequent refinements. Final refinements of this model using. 2406 independent reflections with F2 ~ 36(F2) and 4.00 ~ 2e ~ 50.00 converged with o 56

Rl = 0.0795 and R2 = 0.0820. All parameter shifts during the final cycle of refinement were less' than 0.080. The standard deviation of an observation of unit weight (Chapter 2) was 2.37. The number of observations was 2406 and the number of variables was 170. The °_3 largest peak in the final electron density map was 0.99~ eA. All of the prominent peaks on this map were closely associated with phenyl rings.

[Ni(ONO) (NO)dppe]2. This structure was also solved by the heavy atom method in which Ni and P(l) were located from a three dimensional Patterson map. All remaining nonhydrogen atoms were located by successive refinements and difference electron density maps.

This model was refined to convergence with isotropic phenyl carbon atoms and all other atoms anisotropic. Hydrogens were then located at geometrically idealized positions (C-H = 1.08 A),° assigned isotropic thermal parameters (S 5.0 A°2 ) and held fixed. After refining this model to convergence, the C-C distances for the phenyl carbons ranged from 1.300(14) to 1.443(14) A. The phenyl rings were therefore treated as rigid groups (C-C = 1.383 °A, D -6/mmm symmetry) with 6h individual isotropic thermal parameters for each group atom. 'Using this model final refinements based on the 2249 independent reflections 2 with F2 ~ 30(F ) and 4.0° ~ 28 ~ 50.0° converged with R1 = .0717 and o 0 R2 = 0.0735. All parameter shifts during the final cycle of refinement were less than 0.070. The standard deviation of an observation of unit weight, defined above was 2.18. The largest residual in the final difference electron density map was 1.037 e A°-3 located near ring P1R2. 57

Several anomalous features of the nitrito ligand remained:

0(2)-N(2) = 1.097(14) A° while N(2)-0(3) = 1.192(13) A.° Thus the' terminal oxygen, 0(3), appears to have a longer bond to N(2) than the oxygen [0(2)] bonded to both Ni and N(2). In addition, several angles about the nickel atom involving the nitrito group especially 0(2)-Ni-

N(l) (140.0(4)°), are severely distorted from tetrahedral. There was no evidence for steric crowding that would account for these distor- tions. Although these facts could be interpreted as resulting from the quality of the data set, they are also suggestive of disorder.

Difference fourier maps suggested the presence of another nitrito unit

(occupancy approximately 20%) coplanar with and 180° from the established nitrito atom positions. The position of the metal-bonded oxygen was the same for each nitrito group. The low occupancy of the disordered nitrito atoms, however, prevented their successful refinement.

Results and Discussion

Description of Structures

Perspective views showing the numbering schemes of Ni(N02)2dppe-

CH2C1 2 and [Ni(ONO) (NO)dppe]2 are shown in Fig. 3.1 and Fig. 3.2, respectively. The final structural parameters for the nonhydrogem atoms are given in Tabae 3.2 for the former complex and in Table 3.3 for the latter. The final group parameters for the phenyl rings of the dinitro complex are in Table 3.4 while Table 3.5 contains the corresponding data for the nitrosyl complex. The derived positions for phenyl carbons of the dinitro complex are in Table 3.6 while the 58

Figure 3.1. Perspective view and numbering scheme for Ni(N02)2dppe. The thermal ellipsoids are 50% probability ellipsoids. 01

Perspective view and numbering scheme for [Ni(ONO) (NO)dppe]2. -- The phenyl rings Figure 3.2~ have been deleted for clarity. Thermal ellipsoids are 50% probability ellipsoids.

VI \0 60

Table 3.2. Final parameters for the nonhydrogen atoms of Ni(N02)2«C6H5)2PCH2CH2P(C6H5)2 • CH2C12•

Atom x ~ z Ni 0.23072(13) 0.19654(9) 0.16347(8) C1(1) 0.7606(6) 0.2434(5) 0.3653(4) C1(2) 0.8493(6) 0.3931(4) 0.3098(5) P(l) 0.30314(26) 0.16883(18) 0.29472(16) P(2) 0;06895(27) 0.23716(20) 0.18770 (18) 0(1) 0:3491(12) 0.0721(7) 0.1140(7) 0(2) 0.4568(10) 0.1855(9) 0.1554(6) 0(3) 0.1096(9) 0.1983 (7) -0.0056(5) 0(4) 0.2106(8) 0.3146 (6) 0.0427(5) N(l) 0.3608(12) 0.1475(10) 0.1410(7) N(2) 0.1755 (9) 0.2400 (7) 0.0528(6) C(l) 0.0990(9) 0.2444 (7) 0.3006(6) C(2) 0.1764 (9) 0.1651(7) 0.3355(6) C(3) 0.8054(23) 0.3496(19) 0.3877(17)

Atom B B B11 B22 B33 B12 23 ------13 -- -- Ni 3.63(7) 3.08(6) 2.36 (5) 0.43 (7) 1. 06 (5) 0.14(6) C1(1) 16.3(5) 16.6(6) 17.3(6) -3.7(4) 7.2(5) 0.7(4) C1(2) 17.7(6) 11.4(5) 20.2(6) 1.5(4) 6.1(5) 0.4(4) P(l) 3.76(15) 3.02(15) 2.35(12) -0.04(12) 0.89(11) 0.21(10) P(2) 3.75(16) 4.22(17) 3.26(14) 0.80(13) 1.15 (12) 0.25(12) 0(1) 16.3(11) 5.8(6) 8.8(7) 4.5(7) 6.2(7) -0.7(5) 0(2) 6.4(6) 15.2(10) 6.7(6) 1. 3(7) 3.0(5) -0.9(7) 0(3) 11. 5 (7) 8.8(7) 3.0(4) -3.1(6) -0.1(4) -1.0(5) 0(4) 9.0(6) 4.3(5) 5.8 (5) -0.8(5) 2.5(4) 1. 8 (4) N(l) 6.0(7) 10.5(11) 3.5(6) 1. 7 (8) 1. 7 (6) 1.0(6) N(2) 4.8(6) 5.3(6) 4.0(5) 1. 0(5) 1.8 (5) -0.7(5) C(l) 3.5(6) 4.9(6) 2.3 (5) 1. 2 (5) 0.3(4) -0.3(4) C (2) 2.6(5) 5.2(7) 3.7(5) -0.1(5) 1. 0(4) 1. 4 (5)

Atom------B ------C(3) 15.7(8)

a x, y, and z are in fractional monoclinic coordinates. Anisotropic thermal parameters are in the form exp[-1/4(B11h2a*2 + B22k2b*2 + B33l2c*2 + 2B12hka*b* + 2B13hla*c* + 2B23klb*c »). B is the isotropic thermal parameter in square angstroms of the given atom. 61

Table 3.3. Final parameters for the nonhydrogen atom~ of [Ni(NO) (ONO) «C6H5)2PCH2CH2P(C6H5)2)]2'

Atom x y z Ni 0.46087(8) 0.05148(10) 0.18019(9) P(l) 0.33312(16) 0.00555(21) 0.05428(18) P(2) 0.58008(15) -0.06392(21) 0.19517(16) 0(1) 0.4315 (7) -0.0341(8) 0.3368(7) 0(2) 0.5039(8) 0.1854(8) 0.1123(8) 0(3) 0.4567 (7) 0.2793(8) 0.2002 (7) N(l) 0.4404 (6) 0.0249(7) 0.2796(7) N(2) 0.4879(8) 0.2656(13) 0.1389(9) C(l) 0.6152(6) -0.0888(8) 0.0919(6) C(2) 0.6507(6) 0.0174(8) 0.0595(6) Atom B22 B12

Ni 2.92(5) 3.31(6) 3.47(6) 0.24 (5) 1. 63 (5) 0.00(6) P(l) 2.92(11) 3.78(13) 3.46(12) 0.10(10) 1. 78 (10) 0.01(10) P(2) 3.12(11) 3.24(12) 3.11(12) 0.26(10) 1. 62 (10) -,0.06 (11) 0(1) 14.9(8) 9.1(7) 12.2(7) 4.6(6) 10.7(7) 5.9(6) 0(2) 9.3(6) 5.9(6) 8.7(7) 1. 7 (5) 1.1 (5) -3.6(5) 0(3) 11.2 (7) 5.4 (5) 9.0(7) 0.3(5) 4.4(6) -1.9(5) N(l) 5.4 (5) 5.3(6) 6.4(5) 1.8(4) 3.5(4) 0.7(4) N(2) 6.2(7) 12.2(12) 7.6(9) -1. 7 (8) 1. 8 (6) -0.1(8) C(l) 4.0(5) 3.9(5) 3.4(5) -0.1(4) 2.0(4) -0.8(4) C (2) 3.1(4) 4.1(5) 3.1(4) -0.3(4) 1.4(4) -0.2(4) a x, y, and z are in fractional monoclinic coordinates. Anisotropic thermal parameters are in the form exp[-1/4(B11h2a*2 + . B22k2b*2 + B3312c*2 + 2B12hka*b* + 2B13h1a*c* + 3B23k1b*c*)]. Table 3.4. a Final group parameters for Ni(N02)2«C6H5)2PCH2CH2P(C6H5)2) • CH2C12•

Group x z p c Yc c ~ e

PIRl 0.4665(5} 0.0000(4) 0.3623(3) 2.115 (6) -2.709(6) -0.178(6) P1R2 0.4537(4) 0.3352(3) 0.3847(3) -2.990(12) -1.987(5) -0.889(11) P2Rl -0.0278(6) 0.4224(4) 0.1083(4) -1. 062 (7) -2.708(6) 0.408(6) P2R2 -0.1343(5) 0.0900(4) 0.1289(3) -2.293(8) 2.432(5) 3.058(7)

Group Bl B2 B3 B4 B5 B6

PIRl 3.96(24) 5.9(3) 6.6(3) 6.7(3) 9.4(5) 7.5(4) P1R2 3.27(22) 4.23(25) 5.4(3) 5.22(28) 5.20(29) 4.82(28) P2Rl 4.49 (26) 7.0(4) 8.6(4) 7.3(4) 8.3(4) 6.5(3) P2R2 4.72(27) 6.7(4) 7.6(4) 8.1(4) 7.6(4) 5.8(3)

a xc' Yc' and Zc are the fractional coordinates of the group or~g~ns. The angles ~,.'e, and p (in radians) are the rotations necessary to bring about alignment (except for translation) of the group internal coordinate system with the fixed crystallographic coordinate system. Bi is the isotropic thermal parameter in square angstroms of the atom i in a given group. The rings are numbered so that atom Cl is attached to P and atom C4 is para to Cl. The standard deviations of the least significant digits are given in parentheses.

0' N a Table 3.5. Final group parameters for [Ni(NO) (ONO) «C6H5)2PCH2CH2P(C6H5)2)]2.

Group x z p c Yc c $ e

P1R1 0.1688(3) 0.1831(4) 0.0148(3) 2.254(7) 2.204 (4) -2.910(7) P1R2 0.2351(3) -0.2180(4) 0.0916(3) 0.632(7) -2.294(5) -0.584(7) P2R1 0.7670(3) 0.0106(4) 0.3596(3) 0.522(4) -2.780(4) -2.634(4) P2R2 0.52961(28) -0.3071(4) 0.2536(4) -0.870(10) -2.027(4) -2.098(10)

Group B1 B2 B3 B4 B5 B6 - - -- P1R1 3.28(19) 5.23(25) 7.1(3) 6.26(29) 5.11(25) 4.17(21) P1R2 3.56(20) 5.48(26) 7.1(3) 6.9(3) 6.44(29) 5.64 (26) P2R1 3.13(19) 4.52(23) 5.90(26) 5.74(25) 5.39(26) 4.09(21) P2R2 3.34(19) 5.44(26) 6.7(3) 6.00(28) 6.22(28) 5.26(25)

a xc, Yc' and Zc are the fractional coordinates of the group origins. The angles $, a, and p (in radians) are the rotations necessary to bring about alignment (except for translation) of the group internal coordinate system with the fixed crystallographic coordinate system. Bi is the isotropic thermal parameter in square angstroms of the atom i in a given group. The rings are numbered so that atom C1 is attached to P and C4 is para to C1' The standard deviations of the least significant digits are given in parentheses.

0\ W 64

Table 3.6. Derived parameters for the group atomsa for Ni(N02)2- «C6H5)2PCH2CH2P(C6H5)2) • CH2C1 2•

Group Atom .x y z

P1R1 C1 0.3960(7) 0.0731(5) 0.3336(5) C2 0.5140(8) 0.0736(5) 0.3371(5) C3 0.5845(6) 0.0005 (7) 0.3658(5) C4 0.5371(9) -0.0732(5) 0.3910(5) C5 0.4191(9) -0.0737(5) 0.3875(6) C6 0.3486(6) -0.0006 (7) 0.3588(6)

P1R2 C1 0.3903(6) 0.2627(4) 0.3452(4) C2 0.4155 (7) 0.3305(5) 0.2997(3) C3 0.4789(7) 0.4031(4) 0.3392(5) c4 0.5171 (7) 0.4078(4) 0.4242(5) C5 0.4919(7) 0.3400(5) 0.4697(3) c6 . 0.4285(7) 0.2674(4) 0.4303(4)

P2R1 C1 0.0092(8) 0.3420(5) 0.1456(5) C2 -0.0937(8) 0.3473(6) 0.0788(6) C3 -0.1307 (7) 0.4277(7) 0.0415(5) C4 -0.0648(10) 0.5028(5) 0.0710(6) C5 0.0381(9) 0.4974(5) 0.1378 (6) C6 0.0752(7) 0.4170(7) 0.1751(5)

P2R2 C1 -0.0483 (7) 0.1548(5) 0.1532(5) : C2 -0.1519(9) 0.1619(5) 0.1730(5) C3 -0.2379(7) 0.0971(7) 0.1487(6) C4 -0.2203(8) 0.0252(6) 0.1046(6) C5 -0.1166(9) 0.0180(5) 0.0848(5) C6 -0.0306 (7) 0.0829(6) 0.1092(5)

a x, y, and z are in fractional monoclinic coordinates. Estimated standard deviations (given in parentheses) are derived from those of the group parameters by NUCLS.

------~- _._- -- 65

Table 3.7. Derived .parameters for the group atomsa for [Ni (NO) (ONO) «C6H5)2PCH2CH2P(C6H5)2)]2·

Group Atom x y z

P1R1 C1 0.2409(4) 0.1089(5) 0.0287(5) C2 0.2427(4) 0.1836(6) 0.0998(4) C3 0.1706(5) 0.2578(6) 0.0860(5) C4 0.0966(4) 0.2573(6) 0.0009(6) C5 0.0948(4) 0.1826(6) -0.0702(4) C6 0.1670(5) 0.1084(5) -0.0564(4)

P1R? C1 0.2759(4) -0.1219(5) 0.0729(5) C2 0.2877 (4) -0.2238(6) 0.0343(4) C3 0.2469(5) -0.3198(5) 0.0530(5) C4 0.1942(5) -0.3141(5) 0.1103(6) C5 0.1824(5) -0.2123 (7) 0.1490(5) C6 0.2232(5) -0.1162(5) 0.1303(5)

P2R1 C1 0.6867(3) -0.0227(5) 0.2884(4) C2 0.7653(5) -0.0877 (4) 0.3099(5) C3 0.8456(4) -0.0544(6) 0.3811(5) C4 0.8473(4) 0.0439(6) 0.4308(4) C5 0.7687(5) 0.1089(5) 0.4094(5) C6 0.6884(4) 0.0756(5) 0.3382(5)

P2R2 Cl 0.555.3 (4) -0.2043(5) 0.2279(5) C2 0.5147(5) -0.2843(6) 0.1590(4) C3 0.4890(5) -0.3871(6) 0.1847(5) C4 0.5040(5) -0.4099(5) 0.2793(6) C5 0.5446(5) -0.3299(7) 0.3482(4) C6 0.5702(4) -0.2271 (5) 0.3225(4)

a x, y, and z are in fractional monoclinic coordinates. Estimated standard deviations (given in parentheses) are derived from those of the group parameters by NUCLS. 66 corresponding data for the nitrosyl complex is in Table 3.7. Fixed hydrogen positions are in Tables 3.8 and 3.9, respectively. Table

3.10 contains interatomic distances and angles for Ni(N02)2dppe-CH2C12'

The corresponding information for [Ni(ONO) (NO)dppe]2 is in Table 3.11.

Ni(N0 1 dppe.CH C1 • The crystal structure consists of 2 2 2 2 discrete molecules of the nickel complex well separated from the dichloromethane molecules. The closest contact between the solvent o molecule and the nickel complex is 2.905 A found between Cl(l) and

PlRlH2.

The coordination geometry about the nickel atom is essentially

o square planar with the central atom displaced only 0.17 A from the plane defined by the phosphorus and nitrogen atoms to which it is bonded. The two N-bonded nitro groups are cis to each other in the square plane. The N(l)-Ni-P(l) and N(2)-Ni-P(2) angles are equivalent and are slightly larger than the N(1)-Ni-N(2) and P(1)-Ni-P(2) angles.

This slight distortion from square planarity accomodates the restricted bite angle of dppe and minimizes crowding between the bulky phenyl rings of this ligand and the atoms of the nitro groups. The Ni-P bond lengths are equivalent and are typical of those reported for Ni(II) 66 phosphine complexes. There is no evidence for disorder of the ethylene bridge wh ich forms the backbone of the phosphine ligand. The

o average P-C(CH ) distance of 1.865(10) A is slightly longer than the 2 P-C (C H ) distance of 1.812(8) All P-C distances are typical of 6 5 X. 61 those found in related transition metal phosphine complexes. The geometry about each phosphorus atom is significantly distorted from 67 a Table 3.8. Parameters for the f.ixed hydrogen atom positions for Ni(N02)2«C6H5)2PCH2CH2P(C6H5)2) • CH2C12 •

-= Group -Atom x y z

P1R1 H2 0.5514 0.1312 0.3186 H3 0.6775 0.0011 0.3692 H4 0.5934 -0.1308 0.4132 H5 0.3832 -0.1325 0.4066 H6 0.2571 -0.0023 0.3561

P1R2 H2 0.3860 0.3271 0.2336 H3 0.4998 0.4561 0.3038 H4 0.5676 0.4648 0.4551 H5 0.5217 0.3445 0.5363 H6 0.4079 0.2156 0.4661

P2R1 H2 -0.1447 0.2902 0.0539 H3 -0.2094 0.4341 -0.0112 H4 -0.0917 0.5671 0.0432 H5 0.0907 0.5563 0.1628 H6 0.1554 0.4123 0.2279

P2R2 H2 -0.1644 0.2181 0.2085 H3 -0.3185 0.1033 0.1652 H4 -0.2888 -0.0245 0.0857 H5 -0.1051 -0.0375 0.0493 H6 0.0489 0.0773 0.0926

C1 H1 0.0981 0.3043 0.3359 H2 0.0155 0.2404 0.2508

C2 H1 0.1441 0.1068 0.3583 H2 0.2725 0.1698 0.3675

a x, y, and z are in fractional monoclinic coordinates. 68 a Table 3.9. Parameters for the fixed hydrogen ..tom Positions for [Ni (NO) (ONO) «C6H5) 2PCH2CH2P (C 6H5) 2 ]2·

Group Atom x y z

------~ P1R1 H2 0.3021 0.1874 0.1667 H3 0.1688 0.3170 0.1426 H4 0.0404 0.3179 -0.0124 H5 0.0350 0.1839 -0.1353 H6 0.1639 0.0492 -0.1133

P1R2 H2 0.3302 -0.2265 -0.0094 H3 0.2538 -0.4059 0.0231 H4 0.1604 -0.3871 0.1274 H5 0.1396 -0.2140 0.1922 H6 0.2138 -0.0344 0.1619

P2R2 H2 0.7627 -0.1671 0.2707 H3 0.9102 -0.1023 0.3988 H4 0.9107 0.0724 0.4880 H5 0.7691 0.1848 0.4523 H6 0.6272 0.1284 0.3194

P2R2 H2 0.4986 -0.2652 0.0834 H3 0.4586 -0.4547 0.1329 H4 0.4848 -0.4898 0.3003 H5 0.5531 -0.3499 0.4228 H6 0.6046 -0.1641 0.3774

C1 H1 0.3306 0.1526 -0.1110 H2 0.4431 0.1210 -0.0335

C2 H1 0.2835 -0.0424 -0.1138 H2 0.3978 -0.0859 -0.0526

a x, y, and z are in fractional monoclinic coordinates. 69

Table 3.10. Selected ~nteratomic distances and a.ng1es for Ni(N0 )2- a 2 «C6H5)2PCH2CH2P(C6H5)2) • CH2C12 ·

Distances

Ni-N{l) 1. 864 (12) P(1)-P1R1C1 1.817(8) Ni":N(2) 1. 916 (10) P(1)-P1R2C1 1.811(6) Ni-P(l) 2.178(3) P(2)-C(1) 1.857-(10) . Ni-P(2) 2.182(4) P(2)-P2R1C1 1. 798 (8) N(l)-D(l) 1.227 (14) P(2)-P2R2C1 1.823(8) N(1)-0(2) 1. 229 (14) C(1)-C(2) 1.517 (14) N(2)-D(3) 1.229(11) C1(1)-C(3) 1. 704(27) N(2)-0(4) 1. 240 (11) C1(2)-C(3) 1.716(28) P(1)-C(2) 1.854(10) C1(1)-C1(2) 2.799(10)

Angles

N(1)-Ni-N(2) 88.5(4) C(2)-P(1)-Ni 107.3(3) N(l)-Ni-P(l) 93.4(3) C(2)-P(1)-P1R1C1 107.9(5) N(1)-Ni-P(2) 172.8(5) C(2)-P(1)-P1R2C1 104.6(5) N(2)-Ni-P(1) 170.7(4) P1R1C1-P(1)-P1R2C1 105.6(4) N(2)-Ni-P(2) 92.9(3) P1R1C1-P(1)-Ni 121.7(4) P(1)-Ni-P(2) 86.3(1) P1R2C1-P(1)-Ni 108.5(3) 0(1)-N(1)-D(2) 119.9(15) C(1)-P(2)-Ni 109.3(4) O(l)-N(l)-Ni 116.7(12) C(1)-P(2)-P2R1C1 106.6(5) O(2)-N(1)-Ni 123.3(13) C(1)-P(2)-PZR2C1 104.4(5) O(3)-N(2)-O(4) 120.7(10) P2R1C1-P(2)-P2R2C1 108.4(5) O(3)-N(2)-Ni 123.7(9) P2R1C1-P(2)-Ni 115.6(4) O(4)-N(2)-Ni 115.5(8) P2R2C1-P(2)-Ni 111.8 (4) C1(1)-C(3)-C1(2) 109.8(15) aDistances in angstroms and angles in degrees. Standard deviations (given in parentheses) for the distances and angles involving the non-group atoms were calculated using a variance covariance matrix. Standard deviations for distances and angles involving the 8rouP atoms were calculated using only variances. The estimated variances for the carbon atoms of the rigid groups are those given in Table 3.6. 70

Table 3.11. Selected interatomic distances and angles fo~ [Ni(NO) (ONO) «C6H5)2PCH2CH2P(C6H5)2)]2.a

Distances

Ni-D(2) 2.123(12) P(1)-C(2) 1.842 (9) Ni-N(l) 1.664(9) P(1)-P1RlC1 1.822(6) Ni-P(l) 2.271(4) P(1)-P1R2Cl 1.828(7) Ni-P(2) 2.251(3) P(2)-C(1) 1.834(9) N(l)-O(l) 1.155 (9) P(2)-P2R1Cl 1.817 (5) O(2)-N(2) 1. 097 (14) P(2)-P2R2C1 1.821(7) N(2)-O(3) 1.192 (13) C(1)-C(2) 1.524(11)

Angles

O(2)-Ni-N(1) 140.0(4) C(2)-P(1)-P1R2Cl 104.6(4) O(2)-Ni-P(1) 96.4(3) P1RlC1-P(1)-P1R2C2 101.8(3) O(2)-Ni-P(2) 96.8(3) PlR1Cl-P(1}-N:i ll.~.8 (3) N(l)-Ni-P(l) 108.3(3) P1R2C1'::'P(1)-Ni 113.6(3) N(1)-Ni-P(2) 101.9(3) C(1)-P(2)-Ni 119.4(3) P(1)-Ni-P(2) 112.4(1) C(1)-P(2)-P2R2C1 103.4(5) O(l)-N(l)-Ni 153.4(8) C(1)~P(2)-P2R2C1 103.5(5) N(2)-D(2)-Ni 109.0(12) P2RlC1-P(2)-P2R2C1 105.2(4) O(2)-N(2)-O(3) 127.4 (17) P2RlC1-P(2)-Ni 113.9(4) C(2)-P(1)-Ni 119.4(3) P2R2Cl-P(2)-Ni 110.0(4) C(2)-P(1)-PlRlC1 105.1(4) C(2)-C(1)-P(2) 112.6(6) C(1)-C(2)-P(1) 111. 0 (6) a Distances in angstroms and angles in degrees. Standard deviations (given in parentheses) for the distances and angles involving the non-group atoms were calculated using a variance covariance matrix. Standard deviations for distances and angles involving the group atoms were calculated using only variances. The estimated variances for the carbon atoms of the rigid groups are those given in Table 3.7.

------~----- ~ ~.-- -- 71 tetrahedral. lYhile the PlRlCl-P(l)-Ni angle of 108.5(3)° is equal within experimental error to that found in a regular tetrahedron, the

PlR2Cl-P(1)-Ni angle of 121.7(4)° is significantly larger. In the case of P(2), the corresponding angles are 115.6(4) and 111.8(4)°, respectively.

The N(1)-O(1)-o(2) and N(2)-o(3)-o(4) planes form dihedral angles of 96.5° and 86.4° respectively with the P(1)-P(2)-N(1)-N(2) plane. The Ni-N bond lengths of 1.846(10) and 1.916(10) A° differ by

50 while the Ni-N-O angles span-a range of 70. There are no obvious sources for the disparity in the Ni-N distances. This difference is likely an experimental artifact arising from the unsymmetric shape of the data crystal and the nonlinear decomposition of the crystal which occurred during data collection. The O-N-O angles are equivalent.

The average Ni-N bond length of 1.890(11) A is 0.107 A° shorter than that found in Ni(N0 ) (NO) (PMe )2. The average Ni-P distance of 2 3 2.180(4) A is also shorter than that found in either [Ni(ONO)(NO)dppe]2

(Table 3.11) or Ni(N0 ) (NO) (PMe )2. 50 Within experimental error the 2 3 remaining distances and angles do not differ significantly from those reported for related complexes.

[Ni(NO) (ONO)dppe]2. This complex crystallizes in the mono clinic space group P2 /n with two discrete dimers per unit cell. The l dimer spans two asymmetric units with each half related to the other by a crystallographic inversion center. Halves of the dimer are linked by two dppe ligands forming a ten membered ring including the two nickel 72 atoms. The dppe bridges adopt a chair configuration with no evidence for disorder of the ethylene backbones of these ligands.

The bridging geometry of the dppe ligands allows each Ni atom to attain a distorted tetrahedral geometry. The dihedral angle of

90.4° between the P(1)-Ni-P(2) and 0(2)-Ni-N(1) planes is similar to the idealized angle of 90° for a regular tetrahedron. While the

P(1)-Ni-P(2) angle of l12.4(lj is only slightly larger than that of a regular tetrahedron, the 0(2)-Ni-N(1) angle of 140.0(4)° is considerably larger. Although this structure is distorted, the geometry at the nickel atoms more closely resembles tetrahedral geometry than it does square planar.

The Ni-P(l) distance of 2.271(4) A° differs by 5-70 from the

Ni-P(2) distance of 2.251(3) A. The Ni-N(Ol) bond length of 1.664(9) Ais typical of that found in other low symmetry {NiNO}lO complexes and indicates considerable covalent bonding between nickel and the nitrosyl ligand. The intermediate Ni-N(l)-O(l) angle of l53.4(8)0 is similar to that found in closely related low symmetry

{NiNO}lO complexes of tertiary phosphines (Fig. 2~6).

The nitrito group is bonded to nickel through oxygen as an unidentate ligand where it adopts a cis configuration similar to that 67 observed K CU{N0 )5. There is no bonding interaction between the ~n 3 2 exo oxygen (0(3» of the nitrito ligand and the nickel atom to which this group is bonded. The N(2)-0{3) distance of 1.192(13) is 0.095 A° longer than the N(2)-0{2) distance of 1.097(13) A. This distortion arises from a slight unresolvable disorder involving this ligand. 73

Despite this disorder, however, it is seen that the -N0 ligand is 2 coordinated to nickel as unidentate nitrito groups. Within experimental error, all of the C-P-C distances and angles differ 61 insignificantly from those reported for related complexes.

Structural Interpretation

The results outlined above establish that Ni(N02)2dppe is

square planar with a che1ating dppe ligand. Many d 8 transition metal

complexes of dppe adopt the former configuration and this type of

structure has been well documented crysta110graphical1y. A 68 representative example is the structure of PdC1 dPpe. The metrical 2 details of Ni(N02)2dppe are consistent with these reported structures.

[Ni(ONO) (NO)dppe]2 is binuclear with distorted tetrahedral nickel atoms bridged by two dppe ligands. Complexes bridged by two dppe ligands however are not common. To our knowledge, [Ni(~NO)(NO)- dppe]2 is the first example of a structurally characterized complex in which two bridging dppe ligands form a monocyc1ic system with two metal atoms. (The crystal str"ucture of Re C1 dppe in which two Re atoms are 2 4 2 bridged by two dppe ligands has been reported. However, the metal atoms are also joined by a Re-Re triple bond so that two fused six membered rings are formed.)69 The bridging by dppe ligands in the present complex permits "the {NiNO}lO group to adopt the appropriate tetrahedral geometry by relieving the steric constraint imposed by the

89° bite angle of dppe.

The bending of Ni-NO l~nkage and apparently different Ni-P bond lengths in this complex have also been observed in a number of closely 74 related Ni(X) (NO)L2 complexes (Table 2.9). The low symmetry adopted by these complexes and the bending of the nitrosyl groups are in accord with the molecular orbital diagram for' {NiNO}lO complexes given in

Figure 2.7. A discussion of these structure-bonding relationships has also been presented in Chapter 2.

Another interesting feature of these two structures is the linkage isomerism displayed by the N0 ligand. Each N0 group in 2 2 Ni(N02)2dppe is bonded to the metal atom through the nitrogen atom.

This mode of coordination has been well documented crystallographically and the metrical details of the present structure are consistent with t h ese ear1 1er· reports. 65 In [Ni(ONO)(NO)dppe]2' however, the N0 2 group is bonded to nickel as a unidentate nitrito ligand. Together with the structures reported for Ni(Me NCH CH NH )2(ONO)2,70 Cr(ONO) (NO) 2 2 2 2

CSHSN)3'CSHSN,6S and K3CU(N02)S,67 this complex is one of the few structurally characterized examples of unidentate nitrito coordination.

Infrared spectra (Table 2.8) of [Ni(ONO)(NO)dppe]2 immediately after formation of the nitrosyl complex show predominantly N-bonded nitro groups. In addition, molecular weight determinations indicate that [Ni(ONO)(NO)dppe]2 is monomeric in solution where it likely adopts a distorted tetrahedral structure similar to that reported for bis[bis-(dicyclohexylphosphino)methane]nickel. 7l Thus, while the bis dppe bridged dimer with an -0 bonded N0 group is the crystallographic 2 form of this complex, it is not the predominant form of this complex in solution. Further structural studies may demonstrate the relationship between chelate bite angle and the geometry of Ni(O) complexes. CHAPTER 4

MECHANISTIC STUDIES OF THE REACTION OF Ni(N02)2L2 AND CO

Introduction

The reaction of carbon monoxide with Ni(N02)2L2 complexes has

now been studied in some detail. A series of Ni(N02)2L2 complexes was

prepared, fully characterized and each was allowed to react with CO to

yield Ni(N0 ) (NO)L complexes which were also well characterized 2 2 (Chapter 2). However, the mechanism of reaction 4.1 remains to be

elucidated.

(4.1)

Therefore the reactions of Ni(N0 )2(PMe )2 and Ni(N02)2dppe with CO 2 3 were examined in detail as representative examples of reaction 1. The

structure of Ni(N0 ) (NO) (PMe ) 2 was determined crystallographically 2 3 (Chapter 2) at 298 K and 135 K. Relationships among stereochemistry,

ligand environment, Ni-N-O angle, and Ni-P distance for this and

related complexes were interpreted in terms of the molecular orbital

scheme in Fig. 2.6. The crystal stru~tures of Ni(N02)2dppe and

[Ni(ONO) (NO)dppe]2 were also discussed in Chap~er 3. The latter complex

is particularly interesting since it crystallizes as a dimer with the

N0 group coordinated as a unidentate nitrito -ligand. However, 2 solution studies of its molecular weight show this complex to be monomeric in solution. 75 76

These studies form the basis for the mechanistic studies of reaction 4.1 presented in this chapter. This investigation required 36 kinetic studies of the reactions of Ni(N02)2dppe and Ni(N0 )2(PMe )2 2 3 18 with CO. The reaction of Ni(N0 )2(PMe )2 with c 0 was also studied 2 3 to determine the distribution of ~80 in the products. In addition, variable temperature 3Ip{lH} nmr spectra of Ni(15N02)2dppe and

Ni(15N02 ) (15NO )dppe were obtained to elucidate the solution structures of these complexes.

Experimental

Dichloromethane was purified immediately prior to use by triple distillation in the dark under nitrogen from P40l0?Na2(C03)2' and finally Na S 0 to remove all traces of water, acids, and free radicals. 2 2 4 All other solvents were purified by distillation under dry nitrogen 34 according to published procedures immediately prior to use. The

_solubility of CO in CH C1 was determined manometrically.72 2 2 -1 eCO = PCO/(CO)sol'n = 114. atm 1 mol

The molar absorbtivity at 570 nm of Ni(N0 ) (NO)dppe in CH C1 was 2 2 2 determined at several concentrations using a Cary 14 spectrophotometer and inert atmosphere 1 cm quartz cells. The gasses evolved from the reaction of Ni(N02)2dppe with CO were collected at -196°c and fractionated, and the amount of CO 2 was determined by PV measurements at room temperature. The mole ratio found for CO and Ni was 2 0.94:1.00. 77

Kinetic Studies

The kinetics of reaction 4.1 were followed by visible spectroscopy using a Zeiss PMQ II single beam spectrophotometer modified to accommodate the reaction vessel and temperature control unit shown in Fig. 4.1. Constant temperatures were maintained using a Barnes

Engineering variable temperature chamber and were monitored using a calibrated iron-constantin thermocouple with the reference junction at

O°C. The maximum temperature variation was a 2 DC increase observed over the course of the slowest reaction. All reaction rates were determined at 20°C (±0.5). The reaction vessel consisted of a Pyrex cylinder

(diameter = 1.5 cm) sealed at both ends a~d equipped with a side arm and a stopcock. Solution gas mixing was accomplished by vigorous stirring using a Spinfin (Bel-Art Products) stirring bar sealed in the reaction vessel. This procedure produced a wide vortex throughout the entire column of liquid. Reaction rates were found to be independent of stirring rate. Zero and 100% transmittance were set using the reaction vessel filled with CH C1 before every experiment. 2 2 Monitoring of these settings at the conclusion of each experiment revealed negligible baseline drift.

Solutions of known Ni(N02)2dppe concentration were prepared under nitrogen in 25 ml volumetric flasks. Aliquots were then syringed into the reaction vessel through the stopcock. Three freeze-thaw cycles were performed to insure complete oxygen removal, the vessel was cooled to O°C and the pressure of CO over the solution was adjusted.

After rapid warming to 20°C, the reaction vessel was transferred to E.f=C F( A D J H J I ~ rJ K • ~ \ -f1 nLJ-I ~B~

Figure 4.1. Modified spectrophotometer used for kinetic investigations.

A = photomultiplier detector, B = leveling screws, C = magnetic stirrer, D = reaction vessel with stirring bar, E = constant temperature chamber, F = shutter, G = variable slit, H = monochromator, I = source, J = collimators, K = optical bench. The system also incorporates a temperature controller for the constant temperature cell, a copper-constantin thermocouple for temperature monitoring, a potentiometer for thermocouple output, and a galvanometer bridge for absorbance/transmittance readings.

-.,J 00 79 the constant temperature apparatus and vigorous stirring was initiated

(t ). o The rate data was obtained by following the appearance of the absorption band of the" {NiNO}lO product at 570 nm since Ni(N02)2dppe has negligible absorbance at thiw wavelength. All reactions were run in the presence of excess CO by measuring absorption of the nickel solution in the absence of stirring while vigorous stirring was maintained between absorption measurements. Under these conditions, there was no evidence for the presence of intermediate species. Small positive values for the Y-intercepts observed in the resulting pseudo first order rate plots result from and are dependent on the time required for sample manipulation.

Isotope Scrambling Experiments 18 C 0 (99.4% enriched) was purchased from Prochem. All opera- tions were carried out in an all-glass vacuum line equipped with a mercury diffusion pump and a Toeppler pump for the transfer of carbon 18 monoxide. C 0 (.557 romol). With the connecting stopcock between bulb and flask closed, 20 ml of nitrogen-saturated benzene were added to the flask and three cycles of freeze-thaw were carried out to insure complete oxygen removal. The solution was then cooled to -196°C and the connecting stopcock opened. Four cycles of cooling to -196°C and rapid warming to 500 C accompanied by vigorous stirring were performed to insure solution-gas mixing. The reaction was then allowed to 36 proceed to approximately 93% completion (based on the rate constant , the Ni(N0 )2(PMe )2 concentration, and the estimated pressure) 2 3 ~l80 80 befor~ it was quenched by freezing with a salt-ice slush at -10°C. The gaseous products collected from the frozen solution were passed through three traps at -45°C (chlorobenzene slush) to insure removal of benzene and trapped at -196°C. Any traces of unreacted carbon monoxide were then removed by pumping the collected gas (at -196°C) for 10 minutes. The isolated carbon dioxide was analyzed by mass spectroscopy at the University of Arizona Analytical Center. The nickel containing product was isolated by pumping the frozen solution to dryness and analyzed by infrared spectroscopy (KBr pellet) using a

Perkin Elmer model 387 spectrometer.

A,blank experiment using CO of natural abundance was also carried out under the same conditions.

31p{lH} NMR Studies 15 . 15 15 Ni( N02)2dppe and N~( N0 2) ( NO)dppe were prepared by the procedures outlined in Chapter 2. NiI dppe was synthesized according 2 to published procedures. 73 All 3lp{lH} nmr spectra were obtained using a Brucker WM 250 multinuclear spectrometer equipped with a variable temperature apparatus and are referenced to external TMP. Saturated solutions of each sample in four m1 of CH C1 :CH C1 (9:1) were 2 2 2 2 prepared in 10 rom inert atmosphere tubes. These large volumes were required to prevent vortexing in the probe region. spectrum was simulated using the Panic program supplied with the 74 spectrometer. 81

Reversibility I'lnd Turnovrer Experiments

Identi~al solid Istate infrared spectra (KBr pellets) were obtained befor~ and after Ni(N02)2dppe was stirred in contact with CO 2 for two days.

Ni(N02)2dppe in degassE7d DMSO was stirred under a CO:0 (2:1) l 2 atmosphere whi;Le being irradiated with a 450 watt medium pressure mercury lamp. Once decomposition of the nickel complex was observed, the CO2 produc~d was fractionated and collected as described on the previous page. Moles CO :mo1es Ni(N02)2dppe = 9.6. Similar studies 2 using Ni(N0 )2(PMe )2 in benzene gave the following: Moles CO :moles 2 3 . I 2 36 Ni(N02)2(PMe3)~ = 47. I

Detection of lptermediattes

Saturated solutions of Ni(N02)2(PCY3)2 in CH2C1 2 were stirred under CO (1 aqn, 22°C) until a deep brown color developed (.75 hour).

Solvent was evaporated with a stream of CO and the infrared spectrum of the resulting brown powde: was determined (KBr pellet) (v CO = 1985,

-1 31 1 I 2065 cm ). p{ H} nm:t spectra of Ni (N0 ) 2 (PCy 3) 2 in CO saturated 2 D8 toluene were obtainem at 240 K. In addition to signals arising from reactant and product, slPectra were characterized by a weak signal present only in reacting solutions which was assigned to the carbonyl intermediate. 82

Results and Discussion

The reaction of CO with coordinatively unsaturated dinitro complexes of nickel is quite rapid with reaction rates depending upon the nature of the phosphorus ligands which are attached to the metal.

Reaction times range from 15 minutes for Ni(N0 )2(PMe )2 to one hour 2 3 for Ni(N02)2dppe under ambient conditions of temperature and pressure.

Though reaction 4.1 occurs readily, the reverse reaction of CO2 with the nitrosyl product does not occur over time periods up to 4 hours.

However oxygen, in the presence of light, will reoxidize the nitrosyl complex to the dinitro species. Orjgen has also been observed to 20 oxidize Ni(NO) (C1)dppe. Many cycles of CO reduction followed by 02 oxidation can be carried out before significant decomposition of the metal complex occurs. Thus these complexes are potential catalysts for the oxidation of CO to CO2 at low partial pressures of CO. In studying reaction 4.1, Ni(N02)2dppe had the combination of solubility, reaction rates, and ease of product isolation most suited to a mechanistic investigation. The reaction of Ni(N02)2dppe with carbon monoxide in dry, oxygen-free dichloromethane produced a deep purple solution from which dark blue microcrystalline Ni(N02) (NO)dppe was isolated in nearly quantitative yield. Ni(N02) (NO)dppe is monomeric in solution (Chapter 3) and solutions of this complex in -1 -1 dichloromethane follow Beers law (£ = 690 ± 40 1 mol cm ) over a ten- fold change in concentration (Figure 4.2). The CO evolv~d in this 2 reaction was identified by IR spectroscopy and gas chromatography and 83

1.5

CD (,) c C .Q a 1.0 en .Q

0.5 1.0 3 Ni(N0 HNO)dppe concentration (M) x 10 2

Figure 4.2. Beer's law plot for Ni(N0 ) (NO)dppe in CH Cl • 2 2 2 84 and analyzed by standard vacuum line techniques. The stoichiometry corresponds to reaction 4.2.

(4.2)

The kinetic features of this reaction in dichloromethane were observed at 20°C by following the appearance of the absorption band of the {NiNO}lO product at 570 nm. The dinitro complex has no absorption in this spectral region. The reaction rate was determined at several pressures with CO present in a three- to ten-fold excess. Under these conditions, the reaction is first order in Ni(~02)2dppe, kobsd is linearly dependent on P (Figures 4.3 and 4.4). The overall reaction CO is second order, and there is no evidence for a first-order term in the rate law, since, within experimental error, the plot of k vs P obsd CO has a zero intercept (Figure 4.4). Addition of the free radical inhibitor, 3-tert-butyl-4-hydroxy-5-methylphenyl sulfide, had no effect on the rate of the reaction. Thus the rate law, -d[Ni(N02)2- dppe]/dt = k2[(Ni(N02)2dppe][CO] is applicable with a value of 2.1 x -1 3 -1-1 10 dm mol s for k2 at 20°C. Analogous rate studies of the reaction of Ni(N0 )2(PMe )2 with CO show that this reaction also 2 3 3 -1 -1 36 follows a second order rate law with k2 = 0.6 dm mol s at 20°C.

It was concluded that this reaction is associative, which is 75-79 typical for square planar complexes of nickel (II). Based on these results and other chemical properties of square planar NiX2L2 complexes discussed below, the proposed mechanism for this reaction is outlined in Fig. 4.5. In this mechanism, the rate-determining step

-~ - ____L_._~ ___ ~ ______~ ____ ._~. ______85

4.0

,--, 3.0 «8 ~ -8

1.0

1.0 2.0 3.0 4.0

t (sec) x 10-3

Figure 4.3. Plot of -In[(At-A )/-A ] vs t at 570. nrn for .the reaction co co " of Ni(N02)2dppe with CO in CH2C12 at 20 C. 86

100 200 300 400 Pco(mmHg)

Figure 4.4. 87

+ CO • k-2

.. " +

Figure 4.5. Scheme I: A mechanism proposed for the reaction of . Ni(N02)2L2 and CO. is the formation of the mono carbonyl complex, Ni(CO)(N02)2L2 (k2) , followed by the transfer of an oxygen atom to CO (k ), and terminated 3 by the loss of CO (k ). The trimethylphosphine complex reacts at a 2 4 faster rate than the dppe complex due to the greater ease of formation of the carbony intermediate. Based on the results of Saint-Joly 80 et al., the five coordinate trimethylphosphine complex is expected to be more stable than the corresponding dppe complex due to the greater electron donating power of the trialkylphosphine ligand.

The isolation and characterization of Ni(CO)I (PMe )2 by 2 3 BO Saint-Joly et al. shows that five coordinate nickel complexes of the type Ni(CO)X L do indeed exist. Moreover, 31p nmr studies of Nildppe 2 2 in solution under carbon monoxide demonstrate that this complex adds 88 31 CO reversibly (Fig. 4.6). P nmr spectra of solutions of NiI2dppe under CO (1 atm) are characterized by a single sharp signal unshi£teid from the resonance of Nilidppe indicating a rapid equilibrium which" lies far to the reactant side in reaction 4.3.

(~f .3)

As the temperature is lowered however, signal broadening occurs followed by the appearance of two doublets (Jp_p = 30.5 Hz) consistEmt with the formation of Ni(CO)I dppe. Based on 31p nmr signal 2 intensities, K 1.3 for reaction .4~3 at' 185 K•. " Therefo):"e, 1;ihi1e -low eq = temperatures favor the five coordinate monocarbony1 adduct, increasing the temperature causes evolution of CO to give the four coordinate nickel complex. Similar temperature dependence has also been obser'ved si in the equilibrium between NiX L and NiX L . 2 3 2 2 In the absence of an oxidizing N0 ligand, no further reaction 2 of CO takes place and the five-coordinate monocaroony1 complex can 'be isolated (L = PMe ) or detected (L = 1/2dppe). When the N0 ligand is 3 2 present, an intramolecular attack on the CO ligand analogous to that found for the reaction between the isoe1ectronic NO + ligand and the: . 101 nitro group of cis-[Fe(N02)(S2CNMe2)2] which was reported ear1~er can take place. The oxygen atom transfer from N0 to CO would thenl 2 produce an unstable intermediate similar to the C-bonded CO comp1eix, 2 82 (K)(THF)n(Co(pr-sa1en)(C0 ) whose structure has been reported. ~~n 2 alternative mechanism in which CO attacks the oxygen atom of the 83 coordinated N0 group is also consistent with the observed rate 1a~ •• 2 89

Q) Po Po "t:l N H oM Z 4-1 o t1! J.I +J tJ Q) Po IJl r=J.I ~

l 90

However, this mechanism provides a less than satisfactory explanation -1 for the lack of reaction between N0 and CO in the absence of 2 transition metals (Chapter 1). Moreover a carbonyl intermediate has been observed in the reaction of Ni(N0 )2(PCY3)2 with CO (v = 2 CO -1 -1 31 L_ 1985 cm 2065 em ,15 P {II} = -100.9 ppm). In addition, Scheme I

(Fig. 4.5) requires the rate of formation of Ni(CO) (N02)2L2 to be dependent upon the electronic and steric requirements imposed by 84 L. ,85 A mechanism similar to Scheme I was independently proposed by 2l Doughty et al. whose rece~t study using 180-labelled trans-

[Ni(N02)2(PEt3)~ has shown that N02 is the oxygen source for CO 2 production. 18 The reaction of Ni(N0 )2(PMe )2 with c 0 (99.36% enriched) 2 3 2 yielded additional mechanistic information. Stoichiometric amounts of these species were allowed to react for 5.S hours. The nickel complex isolated from this reaction was analyzed by infrared spectroscopy. No evidence for incorporation of 180 into the nitrosyl group was found.

Additionally, no other IR bands were shifted from those of the nickel 16 complex isolated from the reaction of Ni(N0 )2(PMe )2 with C 02under 2 3 18 identical conditions. The carbon dioxide evolved in the c 0 reaction 2 was collected at -196°C, fractionated, and identified by mass 16 l8 spectroscopy. The predominant species was C 0 0 (mass 46) with the distribution of isotopes as follows: C160lB:c160l80:c180l80 =

1.4:6.8:1.0. The observed isotope scrambling in the carbon dioxide is the result of a secondary process unrelated to reaction 4.4. 91

16 Ni(N 0 )2(PMe )2 + CIBO' ) 2 3 16 16 Ni(N 0 ) (N 0) (PMe ) 2 + c16/lB02 (4.4) 2 3

From these experiments, it is clear that, within the -

detectability limits of the IR experiment (5%), there is no incorpora

tion of lBO into the nickel product of reaction 4.4. The observed

enrichment of evolved CO confirms this species as the oxidized product 2 of this reaction. These results demonstrate that once the CO adduct 2 is formed (Fig. 4.5), there is negligible back reaction to either the

mono carbonyl intermediate or the dinitro starting material. Further-

more, the reaction is terminated by the irreversible loss of CO as was 2 2l suggested by Doughty et al.

All of the data presented thus far is consistent with the mechanism proposed in Scheme I (Fig. 4.5) and conforms to it in some

detail. Experiments to directly observe the five coordinate mono-

carbonyl intermediate, Ni(CO)(N02)2dppe, were therefore initiated.

3lp {lH} spectra of N°I~ 2 d ppe un d er CO sugges t th a tItow empera t ures

favor formation of five coordinate nickel carbonyl complexes. Addi-

tionally, less energy is available and therefore, the oxygen atom

transfer between N02 and CO (Fig. 4.5) is inhibited as the temperature 31 is lowered. Therefore a P nmr study of reaction 4.5 was undertaken

at low temperatures. Although observation of the intermediate was not

(4.5) 92 possible, using this nickel complex, these nmr studies revealed a wealth of important data. Both Ni(N02)2dppe and Ni(N0 ) (NO)dppe were 2 15 labelled with N (95 atom % enrichment) in order to obtain the maximum amount of structural information. Variable temperature 3~ nmr spectra of Ni(15N02)2dPpe are shown in Fig. 4.7. At ambient temperature, the observed broad signal is indicative rapid isomerization of the N02 ligands. As the temperature is lowered however, the exchange reaction is frozen out and a doublet of doublets arising from cis~ Ni(15N02 )2- dppe~ is observed. The spectrum is second order and therefore additional small peaks are also observed (J _ = 47 Hz, J = p p 15N-15N 3 Hz, J. 15 + -7 Hz, J 5 = 38 Hz). The spectrum is c~s N-P trans 1 N..:p also characterized by a doublet of quartets (intensity 5% of main signal). These signals are not due to a removable impurity since independent preparations of Ni(15N02)2dPpe resulted in identical spectra. Moreover the nmr spectrum of these weak signals is consistent with an ABX pattern expected for a nitro nitrito isomer (Jp~p = 64 Hz,

J cis 15N-P = -7 Hz, Jtrans 15N-P = 51 Hz). The observed and simulated 15 spectra for Ni( N02)2dppe are shown in Fig. 4.8.

This result is consistent with the observation of weak nitrito hands in the infrared sp,ectra of Ni (N0 ) 2dppe (Table 2.5). Similarly, 2 two forms of Ni(N0 ) (NO)dppe are present in solution (Fig. 4.9). The 2 major species present is the nitronitrosy1 and the minor one is the 31 nitrito nitrosyl isomer. The temperature dependence of the P spectra demonstrates that the nitro and nitrito species are in equilibrium. 93

Reference =TMP 300 K

280 K

240 K

-84 -85 -86 -87 -88 -89 ppm

Figure 4.7. 94

NXP N P

B °XP1 P1 N P2 P2

C____ ~ ______~ __ __

-84 -85 -86 -ds -89 ppm

Figure 4.8. EXPisimental and simulated 3lp{lHJ nmr spect~~of Ni( N02)2dppe in CD2C1 2 at 220 K.

A = Experimental spectrum: 90!SNi(lSN02)2dPpe, 10% Ni(lSNO )(OlSNO)dppe, B =15imulatIg spectrum for Ni( N02)2dppe, C = Simulated spectrum for Ni( N0 )(O NO)dppe. 2 95

Reference = TMP

260K

240K

220K J ~ \

200 K ~ J'o-A-.. 1 I I , , -83 -84 -S5 -S6 -g7 -S8 ppm

31 1 15 15 Figure 4.9. p{ H} nmr spectra of Ni( N0 ) ( NO)dppe in CD C1 2 2 2 96

Although the presence of the nitrito isomer does not affect the overall conclusion that the reaction of Ni(NOZ)ZL with CO is an Z associative oxygen atom transfer reaction~ it does raise the question as to which of the several isomers are responsible for the reactions with CO. The mechanistic implications are summarized in Fig. 4.10.

Pre equilibrium ..

.. k-2

Figure 4.10. Scheme II: An alternative mechanism proposed for the reaction of Ni(NOZ)ZL and CO. Z

In this scheme, a rapid equilibrium between both forms of NO coordina Z tion preceeds reaction of the nickel complex with CO. The attack of 97 carbon monoxide and subsequent oxygen atom transfer may then occur via either a nitrito (Fig. 4.10) or a nitro (Fig. 4.5) isomer. While some arguments could be made as to which isomer would give the more favorable equilibrium with the carbonyl complex, both are reasonable.

The ring size for direct transfer of an oxygen atom from the N-bonded

N02 group to the CO is four, while the ring size for the nitrito isomer is five. This ring siz~ factor would favor oxygen atom transfer from the nitrito isomer. This however, leaves the nitrosyl group oxygen bonded rather than nitrogen-bonded as is found in the final product.

Despite the uncertainty as to which isomeric form of Ni(N02)2L2 is the most reactive species, it is clear that the reaction of CO with these complexes produces CO and Ni(N0 ) (NO)L in stoichiometric amounts 2 2 2 from a reaction whose rate is dependent on the nature of L. Further more, the reaction proceeds associatively through a mono carbonyl adduct.

Finally, oxygen atom transfer to form the CO intermediate and sub 2 sequent loss of CO to terminate this reaction are essentially 2 irreversible. CHAPTER 5

REACTION OF PALLADIUM NITRO COMPLEXES WITH CO

Introduction

Oxygen atom transfer from coordinated N0 ligands has now been 2 studied in some detail. The reactions are favorable when .the following conditions are met:

1. An empty coordination site at the metal is available. (The reaction of CO with the following nitro complexes (for which the corresponding metal nitrosyl complexes are known) does not occur: 2+ CNMe [Co(N02) (NH3)5] ,Fe(N02)Cl(das)2' Ru(N02)(NO)(S2 2)2·) 2. The nitrogen oxide and oxygen acceptor are bound to the metal in adjacent coordination sites. (While oxygen exchange between the nitro and nitrosyl groups of cis-Fe-(N0 ) (NO) (S2CNMe2) 2 occurs readily, 2 lO this reaction does not occur in the trans complex .)

3. Electronic factors including the thermodynamic stability of the products are favorable.

The above requirements may lead to SOme common pathways for oxygen atom transfer reactions of NO; ligands and to a better under standing of reactions under less defined conditions. Therefore, in order to gain additional insight into these reactions and to assess the role of the metal in the oxidation of CO by NO ' the reaction of CO with X -NO- complexes of Pd(II) was examined. 2

98 99 In contrast to reaction 1, carbon monoxide reacts with

Pd(N02)2L2 to form the previously unknown tetranuclear palladium (0) cluster, Pd (CO)SL (Reaction S.2). 4 4

(S.l)

(S.2)

The molecular structure of one such palladium cluster has been determined and is the subject of this chapter.

Experimental

The reaction of trans Pd(N0 )Z(PMePh )2 with excess CO (1 atm) 2 2 in dry, oxygen-free dichloromethane produced deep red solutions from which Pd (CO)S(PMePh )4 was isolated by the addition of N -saturated 4 2 2 n-hexane. Anal: Cal cd (found) for Pd4CS7HSZOSP4 c, SO.09(SO.36);

H, 3.84(3.92); 0, S.8S(S.67) (Huffman Laboratories, Wheatridge, CO).

V = 1840 em-I, 1820 cm-l (sh) CO The gasses evolved in this reaction were identified by IR spectroscopy and gas chromatography and analyzed by standard vacuum line techniques.

Collection and Reduction of X-Ray Intensity Data

Deep red crystals of Pd (CO)S(PMePh )4 were kindly provided by 4 2 3l Dr. Jules Dubrawski. A well formed crystal of structural quality was sealed in a hollow glass capillary under nitrogen in an inert atmosphere glove box. The crystal was then mounted on a Syntex PZ l 100

autodiffractiometer with the longest dimension of the crystal (8 axis)

approximately parallel to the ~ axis. Automatic centering, indexing,

and 1 eas t squares rout1nes. 37 Y1e. 1de d t h e ce1 1 d·1menS10ns . 1n . Ta b1 e 5 • 1 •

Intensity data were collected under the conditions listed in this table 2 and reduced to F2 and cr(F ) by published procedures. Lorentz po1ariza- o 0 tion factors were calculated on the assumption of 50% perfection and

50% mosaicity of the monochromator crystal.

Solution and Refinement of Structure

Sources for scattering factors and corrections for anomalous

dispersion of pal1a~ium and phosphorus as well as a listing of major

programs used in solving this structure have been presented in Chapter

2. The structure was solved using the MULTAN package based on 400

reflections. An E-map based on these reflections revealed the positions

of the four palladium atoms. The remaining nonhydrogen atoms were

located by successive refinements and difference electron density maps.

The eight phenyl rings were treated as rigid groups (D -6/mmm 6h symmetry) with a single isotropic thermal parameter assigned to each

ring. This model was refined to convergence with anisotropic palladium

and phosphorus atoms and all remaining non-group atoms isotropic. 2 Refinements were based on reflections with F2 > 3cr(F ) and o 0

At this point, the data were expanded to include all reflections 2 with F2 ~ 36(F ) and 4.0° ~ 26 ~ 50.0°, a linear decomposition o 0 correction was applied and the structure refined to convergence. The

positions of the 40 phenyl hydrogen atoms were then calculated 101

Molecular formula Pd4(CO)5(P(CH3)(C6H5)~)4 Molecular weight 1365.48 Crystal shape Rectangu10id Crystal dimensions (rom) .5 x .25 x .25 Crystal system Tric1inic -Space group P'l!. PI Cell dimensions o

- a , A0 11. 966 (1) -b,~ 12.228(3) -c,A 21.001(3) a,deg 103.127(9) B,deg 101. 794(1) y,~eg 96.506 (1) " -V ,A3 2888. (1) Z 2 -3 d b g cm 1.56 o s -3 dIg cm 1.57 ca c D - Radiation, A (MoK) 0.71073 a Monochromator Graphite crystal Supplied power 50 kV 30 rnA Data collection method 9-29 scan -1 Scan speed, deg min variable (3.0-29.3) determined as a function of peak intensity Scan range (29), deg 1-1oKal -1:. 0' to HoKa2 + 1.0 Ratio of total background time to peak scan time 0.5

Standard reflections (200), (040), (004) aft~r every 97 readings Decomposition of standards 30% 29 limit, deg. 4.0-50.0 No. of unique data 10893

--No. of data used in the calculation 5514 (1 ~ 30(1) Absorption coefficient (~), cm-l 13.55 10Z o (C-H = 1.08 A). These atoms were assigned an isotropic thermal Z parameter (B = 5.0 )and included as fixed contributors to F. Final l c refinements of this model converged with Rl = 0.0653 and RZ = 0.1179. The "goodness of fit" value (Chapter Z) was 7.96 •. The discrepancy

and R as well as likely between Rl Z the large "goodness of fit" value arise as a result of the 30% decomposition of the data crystal and the treatment of the phenyl rings (group thermal parameters). All parameter shifts during the final cycle of refinement were less than

0.12.Q' .. The overdetermination ratio (n/m) was Z9.8.

Results and Discussion

The reaction of trans-Pd(NOZ)Z(PMePhZ)Z with carbon produces

Pd (CO)5(PMePh Examination of the gases evolved from this 4 Z)4' reaction gave a ratio of CO :N 0:Pd(N0 ) (PMePh ) 4 of Z.O:l.O:l.O. Z 2 2 2 Infrared spectra of the solid product show bands characteristic of bridging carbonyl gro'ups while no bands attributable to .Pd-H were observed. The compound was also analyzed for C, H, and 0. The elemental analyses were in excellent agreement with the formulation of the compound as the Pd (CO)5(PMePh )4 tetramer. Howe~er, the 4 Z calculated values of C, H, and ° (C, 50.14; H, 3.91; 0, 4.77) for another possible cluster Pd (CO)3(PMePh )3 are barely distinguishable 3 Z from those of the tetramer. Moreover, the present complex was too air sensitive and nonvolatile to obtain a reliable molecular weight.

Consequently, the single-crystal x-ray structure study was carried out to establish the nuclearity and 'molecular geometry of this novel compound. 103

Pd4 (CO)5(PMePh2)4 crystallizes in the triclinic space group Pi with two molecules per unit cell. A perspective view and the numbering scheme for this complex are shown in Figure 5.1. Thermal and positional parameters for the nonhydrogen, non-group atoms are in

Tables 5.2 and 5.3, respectively, while tpe positional parameters for hydrogen atoms are in Table 5.4. Final parameters for the rigid phenyl rings are in Table 5.5 and derived pOSitional parameters for the phenyl carbon atoms are in Table 5.6. Table 5.7 contains selected interatomic distances and angles.

The molecule consists of a distorted tetrahedron of palladium atoms. One edge of the tetrahedron is open with a non-bonded

Pd(3)-Pd(4) distance of 3.365(2) X giving the molecule a nido structure. The four phosphine ligands are each bonded to different metal atoms. The bridging carbonyl groups span the five bonding edges of this distorted tetrahedron. The resulting molecular geometry therefore contains two different kinds of palladium atoms: one set of

Pd atoms is bonded to three other Pd atoms, one phosphorus atom and three carbonyl groups, and the other set of Pd atoms is bonded to two other Pd atoms, one phosphorus atom and two carbonyl groups.

The average Pd-Pd bonding distance, 2.7506(16) A,° compares . ° favorably with the 2.742 A reported for the other Pd(O) cluster which t 86 has been structurally characterized, Pd (Bt -NC)5(S02)~' 2C H " The 3 6 6 average Pd-P distance, 2.314(4) A, lies between the Pd-P distances t 'iI 87 88 reported for Pd(PBu Ph)2 (2.285 A) and Pd(PPh )2(C (C0 Me)2) 3 2 2 o (2.326 It). The average Pd-C(O) distance is 2.063 A while the average

Pd-C-Pd angle is 82.5(6)°. 104

Figure 5.1. Pe·rspective view and numbering scheme for Pda(CO)s(PMePh2)4. -- The thermal ellipsoids are 507. probability ellipsoids. 105

__ .__ .Ta b1e 5. 2. Atomic positional parametersa for the nonhydrogen , non-group atoms of Pd (CO)5'(~(CH3) (C H )2) 4. 4 6 5

Atom x y z

pa(l) .14821(10) .36729 (10) .25913(6) Pd(2) .24033(10) .17420(10) .22133(6) Pd(3) .03256(10) .16076(10) .26090(6) Pd(4) .29672(10) .29480(10) .35324(6) P1 .1280(4) .5448(3) .24143(22) P2 .3235(3) .0716(4) .14216(20) P3 -.1204(4) .0825(4) .29690(22) P4 .3925(3) .2878(3) .45972 (19) 01 -.0944(14) .3663(14) .2802(8) 02 .1242(15) -.0624(15) .2248(9) 03 .4593(13) .1319(13) .3126(8) 04 .2222(13) .5216(13) .4056(8) 05 .1883(11) .3139(12) .1165 (7) C1 -.0166(13) .3161(13) .2711(8) C2 .1206(15) .0368(15) .2344(9) C3 .3755(15) .1728 (15) .3049(9) C4 .2242(13) .4351(13) .3660(8) C5 .1928(12) .2957(12) .1685 (7) C6 .0547(16) -.3687(16) .3020(9) C7 .3708(18) -.0605(18) .1586(11) C8 -.2669(18) .1084(19) .2615(11) C9 -.4461(13) .3042(14) .4737(8)

a x, y, and z are in fractional tric1inic coordinates.

------~ ------.-~-- 106

Table 5.3. Thermal parameters for the 'non-hydrogen,non-group atoms of Pd (CO)5 .P(CH )(C H )2 ,4. a 4 3 6 5

Atom B B B11 B22 B33 B12 13 23 - Pd(l) 35.7(8) 21. 7 (5) 10.83(24) 7.7 (5) 6.4(3) 4.96 (27) Pd(2) "36.7(8) 23.6(5) 9.18(24) 7.9 (5) 7.3 (3) 2.51(27) Pd(3) 32.9(8) 24.9(5) 11.92(26) 4.2 (5) 8.1(4) 5.46(29) Pd(4) 36.9(8) 24.9(5) 8.04 (23) 8.1 (5) 2.7(3) 2.81(27) P(l) 51. (3) 22.3(17) 15.1(9) 13.8(18) 6.5(14) 3.5(10) P(2) 40.8(27) 30.7(19) 9.9 (8) 15.9(18) 6.2(12) 3.7(10) P(3) 39.6(27) 31.5 (19) 14.5(9) 6.4(18) 10.9(13) 9.0(11) P(4) 37.9(26) 25.0(17) 8.5(8) 7.1(17) 1. 7 (12) 2.5(9) ------Atom B

0(1) 5.7(3) 0(2) 6.6 (4) 0(3) 5.3(3) 0(4) 5.2(3) 0(5) 4.46(27) C(l) 2.50(27) C(2) 3.3(3) C(3) 3.1(3) C(4) 2.29(26) C(5) 2.02(24) C(6) 3.6(3) C(7) 4.7(4) C(8) 4.8(4) C(9) 2.74(29)

2 a Anisotropic thermal p'arameters are in the form exp [-1/4(b h a*2 + ll B22k2b*2 + B3312c*2 + 2B12hka*b* + 2B13h1a*c* + 2B23k1b*c*)J. The scale factor for anisotropic parameters is 10. B is the isotropic thermal paremeter in square angstroms of the given atom. 107 a Table 5.4. Parameters for the "fixed hydrogen atom positions for Pd4 (CO)5(P(CH3) (C 6H5)2)4'

Group Atom x y z

P1R1 H2 -0.0477 -0.6091 0.1353 H3 -0.1669 -0.6008 0.0343 H4 -0.1499 -0.4340 0.0018 H5 -0.0138 -0.2757 0.0632 H6 0.1054 -0.2840 0~1642

P1R2 H2 0.3303 -0.4766 0.1916 H3 0.5077 -0.3606 0.2074 H4 0.5447 -0.1753 0.2762 H5 0.4043 -0.1060 0.3293 H6 0.2269 -0.2220 0.3135

P2R1 H2 0.1687 -0.1343 0.0675 H3 0.0441 -0.1995 -0.0393 H4 0.0341 -0.0849 -0.1145 H5 0.1487 0.0951 -0.0829 H6 0.2733 0.1604 0.0239

P2R2 H2 0.4953 0.0134 0.0669 H3 0.6681 0.1096 0.0569 H4 0.7330 0.2999 0.1159 H5 0.6250 0.3942 0.1850 H6 0.4521 0.2980 0.1949

P3R1 H2 -0.1904 0.2709 0.3845 H3 -0.1310 0.3612 0.4992 H4 0.0055 0.2926 0.5695 H5 0.0828 0.1335 0.5251 H6 0.2340 0.0431 0.4104

P3R2 H2 _0.1188 -0.1196 0.1961 H3 -0.1771 -0.3148 0.1792 H4 -0.2479 -0.3831 0.2609 H5 -0.2605 -0.2561 0.3595 H6 -0.2022 -0.0609 0.3765 108

Table 5.4. Parameters--Continued.

Group Atom x y x

P4Rl H2 0.4654 0.1786 0.5666 H3 0.4169 -0.0018 0.5847 H4 0.2760 -0.1379 0.5040 H5 0.1837 -0.0968 0.4051 H6 0.2323 0.0820 0.3870

P4R2 H2 0.5336 0.4226 -0.4080 H3 0.4959 0.5448 -0.3150 H4 0.3079 0.5800 -0.3164 H5 0.1577 0.4929 -0.4107 H6 0.1954 0.3707 -0.5038

a x, y, and z are in fractional monoclinic coordinates. Table 5.5. Final group parameters for Pd (CO)5(P(CH ) (C H )2)4' a 4 3 6 5

Group x z p c yc c

P1R1 -.0321 (7) -.4429(7) .0992(4) 1.802(9) 2.437(8) -3.600(10) 4.19(15) P1R2 .3692 (10) -.2900(10) .2610(5) . -.680(10) -2.857(10) -.685(11) 6.20(23) P3R1 -.0542(8) .2005(8) .4552(5) .662(8) -3.118(8) 1.193(9) 4.91 (17) P3R2 -.1887(6) -.1877 (7) .2781(4) 3.000(9) 2.555(6) -1.847(9) 3.63(14) P2R1 .1598(8) -.0179(8) -.0070(5) 2.840(9) -2.844(8) -.950(9) 5.16(19) P2R2 .5613 (8) .2017 (8) .1241(5) -2.994(11) 2.402(9) -2.706(12) 5.07(19) P4R1 .3249 (7) .0408 (7) .4861(4) 3.286(8) -2.539 (7) 2.444(9) 3.75(14) P4R2 .3471(6) .4580(6) ,:".4088(4) -1.642 (7) -2.875 (6) 2.053(8) 3.44(13)

a xc' y c' and Zc ··are the frac tiona! coord ina tes of the group origins. The angles iP, e , and po (in radians) are the rotations necessary to bring about alignment (except for translation) , of the group internal coordinate system with the fixed crystallographic coordinate system. B is the isotropic thermal parameter in square angstroms for a given group.

I-'o \0 110

Table 5.6. Derived parameters for the group atoms a for Pd (CO)5- 4 (P(CH3) (C 6H5)2)'4

Group Atom x y z

P1R1 C1 .0404(10) -.4477(11) .1586(5) C2 -.0410(11) -.5416(9) .1208 (7) C3 -.1135(10) -.5368(9) .0614(6) C4 -.1047(10) -.4381(11) .0399(5) C5 -.0233(11) -.3442(9) .0776(6) C6 .0492(10) -.3489(9) .1370(6)

P1R2 C1 .2639(11) -.3592(13) .2505(9) C2 .3494(15) -.4005(11) .2212(8) C3 .4547(13) -.3313(14) .2317 (8) C4 .4745(11) -:-.2207(13) .2714(9) C5 .3890(15) -.1794(11) .3007(8) C6 .2837(13) -.2486(14) .2902(8)

P2R1 C1 .0847(12) -.0570(12) -.0699(6) C2 .1517(13) .0502(11) -.0516(7) C3 2267(12) .0893(9) .0114 (7) C4 .2348(12) .0213(12) .0560(6) C5 .1679 (13) -.0859(11) .0377 (7) C6 .0928(12) -.1250(9) -.0253 (7)

P2R2 C1 .6637 (10) .2573(12) .1172(7) C2 .6012(12) .3147(9) .1586 (7) C3 .4988 (12) .2591(11) .1655 (7) C4 .4589(10) .1462(11) .1311(8) C5 .5214(12) .0888(9) .0897 (7) C6 .6238(12) .1444(12) .0828 (7)

P3R1 C1 -.0178(12) .2547(11) .5231(5) C2 .0277 (11) .1603(12) .4966(6) C3 -.0087 (12) .1061(10) .4287 (7) C4 -.0906(12) .1464(11) .3873(5) C5 -.1362(11) .2408(12) .4138(6) C6 -.0997(12) .2950(10) .4817 (7)

P3R2 C1 -.1559(10) -.0718 (7) .2880(6) C2 -.1474(10) . -.1467 (9) .2297(5) C3 -.1802(11) -.2627(9) .2198(5) C4 -.2215(10) -.3036 (7) .2683(6) C5 -.2300(10) -.2287(9) .3266(5) C6 -.1972 (11) -.1127(9) .3365(5) 111

Table 5.0. Derived parameters-Continued.

Group Atom x y z

P4R1 C1 .3549(10) .1462 (8) .4749(6) C2 .4095(9) .1219(9) .5336(5) C3 .3794(10) .0164(10) .5447(5) C4 .2949(10) -.0647(8) .4972 (6) C5 .2403(9) -.0403(9) .4386(5) C6 .2703(10) .0652(10) .4275(5)

P4R2 C1 .3710(9) .3852(9) -.4633(5) C2 . .4589 (7) .4373(10) -.4068(6) C3 .4350(8) .5102(10) -.3523(5) C4 .3233(9) .5309'(9) -.3543(5) C5 .2354 (7) .4787(10) -.4108(6) C6 .2592(8) .4059(10) -.4653(5)

a x, y, and z are in fractional tric1inic coordinates. Estimated standard deviations (given in parentheses) alre derived from those of the group parameters by NUCLS. The rings are numbered so that C(l) is attached to P and atom C(4) is para to atom C(l). 112 Table 5.7. Selected interatomic dist?nces and angles for a Pd4 (CO)5(P(CH3) (C6H5)2)4'

Atoms Distance Atoms Angles

Pd(1)-Pd(2) 2.7416(16) Pd(2)-Pd(1)-Pd(3) 60.75(4) Pd(1)-Pd(3) 2.7507(17) Pd(2)-Pd(1)-Pd(4) 59.65(4) Pd(1)-Pd(4) 2.7512(17) Pd(3)-Pd(1)-Pd(4) 75.38 (5) Pd(2)-Pd(3) 2.7773 (15) Pd(1)-Pd(2)-Pd(3) 59.79(4) Pd(2)-Pd(4) 2.7320(16) Pd(1)-Pd(2)-Pd(4) 60.35(4) Pd(3)-Pd(2)-Pd(4) 75.26(4) Pd(l)-P(l) 2.314(4) Pd(1)-Pd(3)-Pd(2) 59.46(4) Pd(2)-P(2) 2.311(4) Pd(1)-Pd(4)-Pd(2) 60.00(4) Pd(3)-P(3) 2.310(4) Pd(4)-P(4) 2.321(4) P(1)-Pd(1)-Pd(2) 144.9(1) P(1)-Pd(1)-Pd(3) 145.1(1) Pd(l)-C(l) 2.083(15) P(1)-Pd(1)-Pd(4) 133.3(1) Pd(1)-C(4) 2.168(15) P(l)-Pd(l)-C(l) 99.8(4) Pd(1)-C(5) 2.107(14) P(1)-Pd(1)-C(4) 93.6(4) Pd(2)-C(2) 2.186(17) P(1)-Pd(1)-C(5) 97.9,(4) Pd(2)-C(3) 2.134(17) P(1)-Pd(2)-Pd(1) 145.5(1) Pd(2)-C(5) 2.107(14) P(2)-Pd(2)-Pd(3) 138.7(1) Pd(3)-C(1) 2.030(16) P(2)-Pd(2)-Pd(4) 140.1(1) Pd(3)-C(2) 1. 984 (18) P(2)-Pd(2)-C(2) 95.5(5) Pd(4)-C(3) 2.041(17) P(2)-Pd(2)-C(3) 94.3(5) Pd(4)-C(4) 1. 998 (15) P(2)-Pd(2)-C(5) 96.5(4) P(3)-Pd(3)-Pd(1) 141. 5 (1) P(1)-C(6) 1. 887 (19) P(3)-Pd(3)-Pd(2) 157.7(1) P(2)-C(7) 1. 853 (22) P(3)-Pd(3)-C(1) 94.4(4) P(3)-C(8) 1.852(22) P(3)-Pd(3)-C(2) 106.8(5) P(4)-C(9) 1. 873 (16) P(4)-Pd(4)-Pd(1) 157.1(1) - P(4)-Pd(4)-Pd(2) 142.2(1) C(l)-O(l) 1.195(20) P(4)-Pd(4)-C(3) 93.8(5) C(2)-0 (2) 1.191(22) P(4)-Pd(4)-C(4) 106.4(4) C(3)-0(3) 1.170(21) C(4)-0(4) - 1.193 (20) Pd(1)-C(1)-Pd(3) 83.9(6) C(5)-0(5) 1.156 (18) Pd(2)-C(2)-Pd(3) 83.4(6) Pd(2)-C(3)-Pd(4) 81.7(6) Pd(1)-C(4)-Pd(4) 82.6(6) Pd(1)-C(5)-Pd(2) 81.2(5)

:.a Distances in angstroms and angles in degrees. Standard deviations (given in parentheses) for the distances and angles were calculated as described in Table 3.10. 113

Although there are no other Pd(O) complexes with bridging

carbonyl ligands with which the present structure can be compared, a o . similar Pd-C distance (2.063 A) was reported for Pd(PPh3)2~ 88 (C2(C02Me)2~ Moreover, the average Pd-C-Pd angle in Pd4 (CO)S- (PMePh )4 is typical of those normally found for complexes with 2 bridging carbonyl ligands and metal-metal bonds. 89 ,90

The metal carbonyl framework of Pd (CO)S(PMePh )4 is 4 2 essentially the same as that reported for Pt (CO)S(PHePh )4 by 4 2 Vranka et al. 91.. Since individual bond distances and angles were not

reported for the tetranuclear platinum cluster, a detailed comparison of the palladium and platinum compounds cannot be made. However, the average of the five M-M bonding distances and the one non-bonded distance compare favorably as do the dihedral angles. The angles defined by the two planes formed by three metal atoms are: Pd, 93°; ° -., Pt, 97°. The non-bonding M-M distances are Pd, 13.365 A; Pt, 3.S48 A. - The nearly indistinguishable bond distances reported for the tetranuclear palladium and platinum clusters are consistent with other

structural findings which indicate that the Pd-X and Pt-X distances are 92 nearly the same for isoelectronic and isostructural species.

A second palladium complex is also formed furing the reaction between CO and Pd(N0 )Z(PMePh )2' This orange compound has a 2 Z prominent band at 1730 cm-l Additionally 3lp NMR studies of reacting solut'ions show the presence of at least two intermediate species which are formed prior to formation of the tetranuclear palladium cluster. 114

Reaction ).2 is clearly more complicated than the reaction of

Ni(N02)2L2 complexes. The differences between the nickel and palladium case may be evaluated on the basis of Fig. 4.10. Oxygen atom transfer from a coordinated nitrito group to CO leaves an O-bound nitrosyl ligand which can rearrange to form the commonly observed N-bound nitrosyl group. Alternatively, hyponitrito bridged dimers may form.

While the former likely occurs in the nickel case, the latter may be 13 93 operative for palladium complexes.' The results of these possible reaction routes are summarized in Figs. 5.2 and 5.3.

In conclusion, this investigation has led to a better understanding of the oxygen atom transfer reactions of coordinated N0 2 groups. However, there still remain some unanswered questions. It is not known whether the nit~o or the nitrito isomer is the active form jn the oxygen atom transfer from Ni(N02)2L2 to CO. Moreover, the difference in the reactivity between Ni(N02)2L2 and Pd(N02)2L2 and the mechanism of the palladium reactions warrant further investigation. o 0 'N 0'):/0 o 'y/ N L XM O/N LiCM-o-N/ + co LiCM- O/ ~ LiCM-o/ 2 + C~

N L2,XM-O ...... ---- L2XM-N-0

N /N=N, L2XM- O/ ~ ~XM-O O-MX~ L2XM-0-MX12 + N20

M=Pd. Ni

Figure 5.2. Mechanistic implications~f oxygen atom transfer via an M-O-N=O isomer.

...... Vt /0 o co 9 co /0, '-2 (N02)Pd-N L2(N02)Pd-N L2Pd N \0 1'o/~

, _ /N=N, / o L2(N02)Pd-0 /1 0-Pd(N02)~ L2Pd", "~O

-1 1/=1730 em

L2(N0 )Pd-O- Pd(N0 )L H 0 2 2 2 + 2 ~ Pd4 (CO>SL4

Figure S.3. Possible reaction routes for the formation of Pd (CO)SL • 4 4

I-" I-" (J'\ APPENDIX A

LIST OF ABBREVIATIONS

Me = methyl

Et = ethyl

Ph = phenyl

Cy = cyclohexyl

dppe = l,Z-diphenylphosphinoethane

cis-vdpp = cis l,Z-diphenylphosphinoethylene

dppp = 1,3-diphenylphosphinopropane

X = a halogen or a pseudohalogen

R = aryl or alkyl

TMP = trimethylphosphite

Cp = cyclopentadienyl

TEP C(CH ) (CH P(C H ))3 = 3 Z Z S POC = P(OCHZ)3C(CH3) DMSO = dimethylsulfoxide

117 REFERENCES

1. Cotton, F. A.; Wilkinson, G. "Advanced Inorganic Chemistry", 3rd Ed. John Wiley and Sons: New York, p. 354-362, and references therein.

2. Johnson, B. F. G.; Bhaduri, S. J. Chern. Soc., Chern. ·Commun. 1973~ 650.

3. Haymore, B. L. ; Ibers, J. A. J. Am. Chern. Soc. 1974, 96, 325.

4. Hayrnore, B. L. ; Ibers, J. A. J. Am. Chern. Soc. 1974, 98, 1364.

5. Hendricksen, D. E. ; Eisenberg, R. J. Am. Chern. Soc. 1976, 98, 4662.

6. Hendricksen, D. E. ; Meyer, C. D. ; Eisenberg, R. Inorg. Chern. 1977, 16, 970.

7. Bhaduri, S.; Johnson, B. F. G. Trans. Met. Chern. (Weinheirn Ger.), 1978, 1, 156.

8. Kubota, M.; Evans, K. J.; Koerntgen, C. A.; Marsters, J. Am. Chern. Soc. 1978, 100, 342.

9. Eisenberg, R.; Meyer, C. D. Ace. Chern. Resch. 1975, ~, 26.

10. I1eperuma, O. A.; Fe1tham, R. D. J. Am. Chern. Soc. 1976, 98, 6039.

11. Booth, G. ; Chatt, J. J. Chern. Soc. 1962~ 2099.

12. Bhaduri, S. ; Johnson, B. F. G.; Matheson, T. W. J. Chern. Soc. , Dalton Trans. 1977, 561.

13. Manchot, W. ; Wa1dmu11er, A. Chern. Ber. 1926, ~, 2363. 14. Hieber, W. ; Anderson, J. S. z. Anorg. A11gern. Chern. 1932, 208, 238. 15. Hieber, W. ; Anderson, J. S. z. Anorg. A11gem. Chern. 1933, 221, 132. 16. Hieber, W.; Beutner, H. z. Anorg. A11g. Chern. 1963, 320, 101. 17. Hughes, W. B. J. Chern. Soc., Chern. Commun. 1969,1126.

18. Kiji, J.; Yoshikawa, S.; Furukawa, J. Bull. Chern, Soc. Japan, 1970, !!1, 3614.

118 119

19. Bhaduri, S.; Johnson, B. F. G.; Savory, C. J.; Segal, J. A.; Walter, R. H. J. Chern. Soc., Chern. Commun. i974, 809.

20. Ugo, R.; Bhaduri,S.; Johnson, B. F. G.; Khair, A.; Pickard, A.; Benn-Taarit, Y. J. Chern. Soc., Chern. Commun. 1976, 694.

21. Doughty, D. T.; Gordon, G.; Stewart, R. P., Jr. J. Am. Chern. Soc., 1979, 101, 2645.

22. Kirtley, S. W.; Chanton, J. P.; Love, R. A.; Topton, D. L.; Sorrell, T. N.; Bau, R. J. Am. Chern. Soc~ 1980, 102, 3451.

23. Johnson, B. F. G.; McCleverty, J. A. Prog. lnorg. Chern., 1966,1, 277.

24. Griffith, W. P. Adv. Organornet. Chern. 1968, I, 211.

25. Connelly, N. G.lnorg.Chirn. Acta Rev. 1972, 6, 48.

26. Caulton, K. G. Coord. Chern. Rev. 1975, 14, 317.

27. Enemark, J. H.; Feltharn, R. D. J. Am.Chem~ ·Soc. 1974, ~, 5002.

28. Feltharn, R. D. Inorg. Chern. 1964, 1, 116.

29. Grundy, K. R.; Laing, K. R.; Roper, W. R. J. Chern. Soc., Chern. Commun. 1970, 1500.

30. Feltharn, R. D.; Kriege, J. C. J. Am. Chern. Soc. 1979, 101, 5064.

31. Dubrawski, J.; Kriege-Sirnondsen, J. C.; Fel tl'tarn, R. D. J. Am. Chern. Soc. 1980, 102, 2089.

32. Haller, K. J.; Enernark, J. H. Inorg. Chern. 1978, 17, 3552.

33. Enemark, J. H. Inorg. Chern. 1971, 10, 1952.

34. Gordon, A. J.; Ford, R. A. "The Chernist's Cornpanion A Handbook of Practical Data, Techniques, and References," John Wiley and Sons: New York, 1976, p. 429.

35. Basolo, F., Ed: "Inorganic Synthesis," McGraw Hill: New York, 1976: Vol. XVI, p. 157.

36. Elbaze, G.; Ph. D. Dissertation, 1981, Universite Paul Sabatier, Toulouse. 120

37. Programs used for centering of reflections, autoindexing, least squares refinement of cell parameters, and data collection are in "Syntex P2l Fortran Ope'ration Manual" Syntex Analytical Instruments: Cupertino, CA 1975.

38. Cromer, D. T.; Waber, J. T. "In terna tional Tables for X-Ray Crystallography;" Ibers, ~J. A.I; Hamilton, W. C., Eds., Kynoch Press: Birmingham, England, 1974; Vol. IV, Table. 2.2A, p. 149.

39. Cromer, D. T. Ibid., Table 2,1, p. 149.

40. Davidson, E. R.; Simpson, W. 1-. Ji. Chern. Phys. 196~, 42, 3175.

41. Johnson, P. L.; Enemark, J. H.; Fe;ltharn, R. D.; Swedo, K. B.

Inorg. Chern. 1976, 15, 2989. !

42. Figgis, B. N. "Introduction '~o Ligand Fields," John Wiley and Sons: New York, 1966, p. ,~15. i

43. Nakamoto, K. "Infrared and Rflman Spectra of Inorganic and Coordination Compounds," ;3rd Edition, John Wiley and Sons: New York, 1978, Chpt. III-4, p., 220.

44. Enernark, J. H.; Feltharn, R. D. Cqord. Chem. Rev. 1974,11, 339.

45. Mingos, D. M. P. Coord. Chern. Rev. 1973, 12, 1205. ,----.- - 46. Huheey, J. E. "Inorganic ChejIlistx'y: Principles of Structure and Reactivity," Harper Interpational Edition, 1972.

47. Ronova, I. A.; Alekseeva, N. V.; Veniarninov, N. N.; Kravers, M. A. Xh. Strukt. Khim. 1975, 16, 476.

48. Cox, A. P.; Brittain, A. Traps. F'araday Soc. 1970 , 66, 557.

49. Berglund, D.; Meek, D. W. IngE..S..:JChern. 1972, 11, 1493.

50. Krieg~Simondsen, J.; Elbaze, G.; IDartiguenave, M.: Feltham, R.D.: Dartiguenave, Y. Inorg. C'~em. 11982, 21, 230.

51. Feltham, R. D.; Enemark, J. H. Tqp. Stereochern. 1981, 12, 155.

52. Meineres, J. H.; Rix, C. J.; Clardy, J. C.; Verkade, J. G.

Inorg. Chern. 1975, 14, 705. !

53. DiVaira, M.; Ghi1ardi, C. A.; Sacconi, L. Inorg. Chern. 1976, 15, 1555. 121

54. Chong, K. S.; Retting, S. J.; Sto'rr, A.; Trotter, J. Can. J. ~. 1979, 57, 3090.

55. Ibid. 3099.

56. Ibid. 3107.

57. Kilbourn, R. T.; Powell, H. M. J. Chem. Soc.~ Dalton Trans. 1970, 1688.

58. Kruger, C.; Tsay, Y. H. Cryst. Struct. Connnun. 1974, l., [:55.

59. Mais, R. H. B.; Owston, P. G.; Thompson, D. T.; Wood, A. M. J. Chem. Soc. A. 1967, 1744.

60. Mais, R. H. B.; Owston, P. G.; Thompson, D. T.; Wood, A. M. Inorg. Chem. 1976, 15~ 2763.

61. Dawson, J. K.; McLennan, T. J.; Robon'son, W.; Merle, A.; Dartiguenave, M.; Dartiguenave, Y.; Gray, H. B. J. Am. Chem. Soc. 1974, 96, 4428.

62. Corain, B.; Bressan, M.; Favero, G.; Inorg. Nuc1. Chem. Lett. 1971, 1., 197. 63. Cundy, C. S. J. Organomet. Chem., 1974, 69, 305.

64. Hidai, M.; Kokura, M.; Uchida, Y. -Bull. Chem. Soc. Japan, 1973, 46, 686.

65. Lukehart, C. M.; Troup, J. M. Inorg. Chim. Acta, 1977, 22, 81 and references therein.

66. Macdougall, J. J.; Nelson, J. H.; Babich, M. W.; Fuller, C. C.; Jacobson, R. A. Inorg. Chim. Acta,1978, 11, 201 and references contained therein.

67. Klanderman, K. A.; Hamilton, W. C.; Bernal, I. Inorg. Chim. Acta, 1977, 23, 117.

68. Steffen, W. L.; Pa1enik, G. J. Inorg. Chem. 1976, 15, 2432.

69. Cotton, F. A.; Stanley, G. G.; Walton, R. A. Inorg. Chem. 1978, 17, 2099.

70. Drew, M. G. B.; Goodgame, D. M. L.; Hitchman, M. A.; Rogers, D. Proc. Chem. Soc. 1964, 363.

71. Kruger, C.; Tsay, Y. Acta. Cryst. 1972, 1941, B28. 122

72. A procedure similar to that described in reference 36 was used.

73. Hudson, M. J.; Nyholm, R. S.; Stiddard, M. H. B. J. Chem. Soc. A. 1968 , 40.

74. Details of this program are summarized in "Aspect 2000 NMR Software Manual 1, Bruker Instruments ,Inc., 1981.

75. Cusmano, M.; Ricevuto, V. J. Chern. Soc., Dalton Trnns. 1978,1682.

76. Murmann, R. K. Inorg. Chern. 1963, ~ 116.

77. Billo, E. J. Inorg. Chern. 1973, 12, 2783.

78. Basolo, F.; Chatt, J.; Gray, H. B.; Pearson, R. G.; Shaw, B. L. J. Chern. Soc. 1961, 2207.

79. Cattaline, L.; Martelli, M.; Rigo, P. Inorg. Chirn. Acta, 1967, ...!.' 149. 80. Saint-Joly, S.; Mari, A.; Gleizes, A.; Dartiguenave, M.; Dartiguenave, Y.; Galy, J. Inorg. Chern. 1980, 19, 2403.

81. Saint-Joly, S.; Galy, J.; Meier, P.; Merbach, A. Inorg. Chern. 1978, 17, 3503.

82. Fachinetti, G.; Floriani, C.; Zanazzi, P. F. J. Am. Chern. Soc. 1978, 100, 7405.

83. Tovrog, B. S.; Diamond, S. E.; Mares, F. J. Am. Chern. Soc. 1979, 101, 270.

84. Tolman, C. A.; Seidel, W. C.; Gosser, L. W. J. Am. Chern. Soc. 1974, 96, 53.

85. Nakamura, Y.; Maruya, K. I.; Mizoroki, T. J. Organornet. Chern. 1976, 104, C5.

86. Otsuka, S.; Tatsurno, M.; Aoki, T.; Matsurnoto, M.; Yoshika, H.; Nakatsu, K. J. Chern. Soc., Chem.Conunun. 1973, 445.

87. Otsuka, S.; Yoshida, T.; Matsumoto, M.; Nakatsu, K. J. Am. Chern. Soc. 1976, 98, 5850.

88. McGinnety, J. A. J. Chern. Soc., Dalton Trans., 1974) 1038.

89. Colton, R.; McCormick, M. M.; Pannan, C. D. Aust. J. Chern. 1978, 31. 1425. 123

90. Braterman, P. S. Struct. Bonding (Berlin), 1972, 10, 57.

91. Vranka, R. G.; Dahl, L. F.; Chini, P.; Chatt, J. J. Am. Chem. Soc. 1969, 2!,., 1574. 92. Hartley, F. R. "The Chemistry of Platinum and Palladium," Halstead Press: New York, 1973.

93. Andrews, M. A.; Kelly, D. P. J. Am. Chem. Soc. 1981, 103, 2894.