YTTERBIUM(II) – GROUP 6, 7 TRANSITION METAL CARBONYL COMPLEXES:
SYSTEMATIC SYNTHESES AND STRUCTURAL CHARACTERIZATION
DISSERTATION
Presented in Partial Fulfillment of the Requirements for the Degree of Doctor of
Philosophy in the Graduate School of The Ohio State University
By
Pavel V. Poplaukhin, B.S.
************
The Ohio State University, 2006
Dissertation Committee: Approved by
Professor Sheldon G. Shore, Adviser ______
Professor Claudia Turro Advisor
Professor Yiying Wu
Professor Richard A. Miller Graduate Program in Chemistry
ABSTRACT
New carbonyl complexes of divalent ytterbium and transition metals of groups 6,
7 of the periodic table have been prepared. The syntheses were carried out in systematic
fashion with the aim of establishing general procedures suitable for preparation of a range
of compounds of this type. The products obtained were characterized by means of IR spectroscopy and X-ray single crystal diffraction. Nineteen X-ray structures are reported herein, of which only one has been published before. The compounds studied can be divided into two major groups: the solvent-separated ion pairs, where the YbII cation is surrounded with solvent molecules acting as ligands, preventing interaction with the metal carbonylate anion; and complexes with the bridging carbonyl ligands (isocarbonyl ligands), where the cation and the anion are bound together through a –CO- link. New instances of condensation of the solvent-separated ion pairs into the isocarbonyl complexes have been discovered, and the mechanism for such transformation was
2- proposed. The novel [Hg(W(CO)5)2] anion was discovered and characterized by X-ray
single crystal diffraction. Its reactivity was briefly investigated.
ii
Посвящаю моим родителям
Dedicated to my parents
iii ACKNOWLEDGMENTS
I would like to thank all the wonderful friends that I made during my stay here at
OSU. This thesis would not be possible without their help and support. I wish to express
my gratitude to my parents also, whose love and support I could sense all this time, no
matter the distance. I am obviously indebted to Dr. Sheldon Shore for his advice,
financial support, and remarkable freedom that I and my fellow coworkers enjoyed in his research group. I also cannot fail to mention Dr. Shore’s jokes and his easy-going personality, qualities that helped everyone working under his guidance.
My special thanks go to Dr. Xuenian Chen, who was a true friend and teacher all
these years. His knowledge, experience and almost unbelievable patience made this thesis
possible. I am grateful to Dr. Edward A. Meyers for his inestimable help with X-ray structure determinations and many interesting conversations we held along the way. I thank Dr. Shengming Liu for teaching me many laboratory techniques and helping with
X-ray, as well as being a great person to talk to. And I also want to thank all Shore group
members that I worked with. It was fun to get to know you guys!
iv VITA
December 5, 1976……………...…Born – Biysk, Russia
1999…………………………….…B.S. Chemistry, Novosibirsk State University
1999 – 2001……………………….Graduate Research Assistant, Institute of Solid State
Chemistry, Siberian Branch of Russian Academy of
Sciences, Novosibirsk, Russia
2001 – present……………….…....Graduate Teaching and Research Associate,
The Ohio State University
PUBLICATIONS
1. Liu, S.; Poplaukhin, P.V.; Ding, E.; Plecnik, C.; Chen, X.; Keane, M.A.; Shore, S.G.
“Extended Lanthanide-Transition Metal Arrays with Cyanide Bridges: Syntheses,
Structures and Catalytic Applications”, J. of Alloys and Comp. 2006, 418, 21.
v 2. Modestov A.N., Poplaukhin P.V., Lyakhov N.Z. “Dehydration kinetics of lithium
sulfate monohydrate single crystals”, Journal of Thermal Analysis and Calorimetry,
2001, 65(1), 121-130.
3. Kundo N.N., Ivanchenko V.A., Mishakov I.V., Poplaukhin P.V. “Effects of an electron
beam on aqueous solutions of Cr(VI) salts”, Khimiya v Interesakh Ustoichivogo Razvitiya
(Chemistry for Sustainable Development), 1999, 5, 491-495.
4. Kundo N.N., Ivanchenko V.A., Mishakov I.V., Poplaukhin P.V. “Study of the
influence of sulfurous reducing agents on Cr(VI) elimination from aqueous solutions”,
Khimiya v Interesakh Ustoichivogo Razvitiya (Chemistry for Sustainable Development),
1999, 5, 485-490.
5. Klyuchnikov O.R., Brylyakov K.P., Poplaukhin P.V. “Reaction of 2,4,6-t-tret-
butilphenol with 2-methyl-5-isopropyl-1,4-dinitrozobenzene”, Izvestiya Vysshih
Uchebnyh Zavedeniy (in Russian), Part ‘Chemistry and Chemical Technology’, 1996,
39(1), 98-99.
vi FIELDS OF STUDY
Major Field: Chemistry
vii TABLE OF CONTENTS
Abstract……………………………………………………………………………………ii
Dedication………………………………………………………………………………...iii
Acknowledgments…………………………………….………………………………….iv
Vita………………………………………………………………………………………...v
List of Tables……………………………………………………………………………xiv
List of Figures………………………………………………………………………..….xix
List of Schemes…………………………………………………………………...….xxxiii
List of Charts…………………………………………………………………………xxxv
List of Abbreviations…………………………………………………………………xxxvi
List of Compound Numbers………………………………………………………...xxxviii
CHAPTER 1. Introduction
1.1 General Review of Ln - TM Complexes…………………………………………..1
1.2 Ln-TM Cyanides…………………………………………………………………..4
1.3 Ln-TM Carbonyls
1.3.1 General Remarks…………………………………………………..…………...8
viii 1.3.2 Systems with direct Ln–M bond…………………………………………...10
1.3.2 Solvent-Separated Ion Pairs………………………………………………..14
1.3.3 Systems with Bridging
Ligands (Ln-TM Isocarbonyls)……………………………………………….…17
1.3.4.1 Transmetalation reaction…………………………………………………22
1.3.4.2 Reduction of Transition Metal Carbonyls over Amalgam……………….24
1.3.4.3 Condensation of Solvent-Separated Ion
Pairs into Extended Arrays…………………………………………………...….26
1.4 Applications……………………………………………………………………..28
1.5 Statement of the Problem……………………………………………………….29
CHAPTER 2. Yb(II) – Group 7 Transition Metal Carbonyl Complexes
2.1 Starting Materials…………………………………………………………….31
2.2 Transmetalation reactions between Yb and
Hg[M(CO)5]2 (M=Mn, Re). Formation of
the Solvent-Separated Ion Pairs…………………………………..………….34
2.3 Solution IR Spectra of [Yb(THF)6][Mn(CO)5] 1,
[Yb(THF)6][Re(CO)5] 2, [Yb(DME)n][Mn(CO)5]2 (1a),
[Yb(DMF)n][Mn(CO)5]2 (1b), [Yb(pyr)6][Mn(CO)5]2 (1c)………………….37
2.4 X-ray structures of 1 and 2………………………………………………..….41
ix 2.5 Condensation of 1 with Et2O into extended
structures {Yb(THF)2(Et2O)2[(μ-CO)2Mn(CO)3]2}∞ (3)
and {Yb(THF)4[(μ-CO)2Mn(CO)3]2}∞ (4)…………………………………..53
2.6 Molecular Structure of {Yb(THF)2(Et2O)2[(μ-CO)2Mn(CO)3]2}∞ (3)………62
2.7 Molecular Structure of {Yb(THF)4[(μ-CO)2Mn(CO)3]2}∞ (4)………………64
2.8 Minor Products of Condensation of 1 with Et2O.
Molecular Structures of (THF)2Mn3(CO)10 (5)
and [(THF)5Yb(μ-CO)Mn3(CO)13][Mn3(CO)14] (6)…………………………67
CHAPTER 3. Yb(II) – Group 6 Transition Metal Carbonyl Complexes.
- Complexes of the [CpM(CO)3] anions
3.1 Starting Materials…………………………………………………………….76
3.2 Notes on Syntheses…………………………………………………………..78
3.3 The Structure of
{(THF)4Yb[Cp(μ-CO)Cr(CO)2][Cp(μ-CO)2Cr(CO)]}∞ (7a)………………..86
3.4 Molecular Structures of the 1-D Polymeric
Chain Compounds {(THF)4Yb[Cp(μ-CO)2Mo(CO)]2}∞ (8a),
{(THF)3Yb[Cp(μ-CO)2W(CO)]2}∞ (9a),
{(CH3CN)4Yb[Cp(μ-CO)2Cr(CO)]2}∞ (7b),
{(CH3CN)4Yb[Cp(μ-CO)2Mo(CO)]2}∞ (8b)………………………………...89
x 3.4.1 One-Dimensional Chain Structures 8a and 9a…………………………90
3.4.2 One-Dimensional Chain Structures 7b and 8b………………………...96
3.5 Molecular Structures of the Discrete
Molecular Compounds (CH3CN)6Yb[Cp(μ-CO)W(CO)2]2 (9b)
and (DME)3Yb[Cp(μ-CO)M(CO)2]2
(M = Cr, 7c; M = Mo, 8c; M = W, 9c)……………………………………...101
3.6 Discussion of the Solution Infrared Spectra………………………………..105
3.7 Discussion of the Molecular Structures…………………………………….109
CHAPTER 4. Yb(II) – Tungsten Carbonyl Complexes Derived from W(CO)6
4.1 Starting Materials…………………………………………………………...114
4.2 Reactions of the Ytterbium Amalgam with the
Tungsten Hexacarbonyl in Various Solvents.
Overview of Preparation of [Yb(DMF)7][W2(CO)10] (10),
{(CH3CN)6Yb(W2(CO)10)·CH3CN}∞ (11) and
{(pyr)5Yb[Hg(W(CO)5)2]}∞ (12)…………………………………………...115
4.3 Preparation, Molecular Structure and the Solution IR
Spectrum of the Solvent-Separated Ion Pairs
Complex [Yb(DMF)7][W2(CO)10] (10)………………………………..…...117
xi 4.4 Preparation, Molecular Structure and the
Solution IR Spectrum of the 1-D Polymer
{(CH3CN)6Yb(W2(CO)10)·CH3CN}∞ (11)…………………………………127
4.5 Preparation, Molecular Structure and
the Solution IR Spectrum of the 1-D Polymer
{(pyr)5Yb[Hg(W(CO)5)2]}∞ (12)…………………………………………...131
4.6 Reduction of {(pyr)5Yb[Hg(W(CO)5)2]}∞ (12)
with Sodium Amalgam………………………………………….……….…138
4.7 Attempted Reduction of the Chromium Hexacarbonyl
with the Ytterbium Amalgam………………………………………………142
CHAPTER 5. Conclusion………………………………………………………………146
CHAPTER 6. Experimental Section
6.1 General Procedures…………………………………………………………148
6.2 X-ray Structure Determination……………………………………………..149
6.3 Preparation of [Yb(L)n][M(CO)5]2 1a, 1b, 1c, 2…………………………....151
6.3.1 Route (a)………………………………………………………………...151
6.3.2 Route (b)………………………………………………………………..152
6.4 Preparation of {(THF)2(Et2O)2Yb[(μ-(CO)2Mn(CO)3)2]}∞, 3……………..152
6.5 Preparation of {(THF)4Yb[(μ-CO)2Mn(CO)3)2]}∞, 4………………………153
6.6 Preparation of (THF)2Mn3(CO)10, 5………………………………………..153
xii 6.7 Preparation of [(THF)5Yb(μ-CO)Mn3(CO)13][Mn3(CO)14], 6……………..154
6.8 Preparation of compounds 7a, 7b, 7c, 8a, 8b, 8c, 9a, 9b, 9c…………...….154
6.8.1 Preparation via the ytterbium amalgam route (Scheme 3.1)……………154
6.8.2 Preparation via the mercury
bis-di(chromiumcyclopentadienyltricarbonyl) route (Scheme 3.2)…….155
6.8.3 Elemental analyses for compounds 7a, 7b, 7c, 8a, 8b, 8c,
9a, 9b, and 9c……………………………………………………………….155
6.9 Preparation of compounds 10, 11 and 12………………………………155
6.9.1 Preparation of [Yb(DMF)7][W2(CO)10] (10)……………………….156
6.9.2 Preparation of {(CH3CN)6Yb(W2(CO)10)·CH3CN}∞ (11)………….157
6.9.3 Preparation of {(pyr)5Yb[Hg(W(CO)5)2]}∞ (12)…………………...157
6.9.4 Preparation of (pyr)3Cr(CO)3 (13)………………………………….158
APPENDIX A. Disorder in Compounds Studied…………..…………………………..159
APPENDIX B. Infrared Spectra of Compounds Studied………………………………171
LIST OF REFERENCES……………………………………………………………….194
xiii LIST OF TABLES
Table 2.1. Infrared Data for Compounds 1 - 4 in the
Carbonyl Stretching Frequency Region………………………………………………….38
Table 2.2 Crystallographic Data for [Yb(THF)6][M(CO)5]2
(1, M=Mn; 2, M=Re), {Yb(THF)2(Et2O)2(Mn(CO)5)2}∞
(3) and {Yb(THF)4(Mn(CO)5)2}∞ (4)……………………………………………………44
Table 2.3 Selected Bond Distances (Å) and Bond Angles (deg) for [Yb(THF)6][M(CO)5]2
(1, M=Mn; 2, M=Re; data for the non-disordered
- [M(CO)5] anions are given)……………………………………………………………..46
Table 2.4 Bond Distances (Å) and Bond Angles
- (deg) for the Disordered [Mn(CO)5] Anion in
[Yb(THF)6][Mn(CO)5]2 (1)………………………………………………………………48
xiv Table 2.5. Bond Distances (Å) and
Bond Angles (deg) for the Major Portion of the Disordered Pentacarbonylrhenium
Anion in [Yb(THF)6][Re(CO)5]2 (2)……………………………………………………..51
Table 2.6. Bond Distances (Å) and
Bond Angles (deg) for the Minor Portion of the Disordered Pentacarbonylrhenium
Anion in [Yb(THF)6][Re(CO)5]2 (2)……………………………………………………..52
Table 2.7 Selected Bond Lengths (Å) and
Bond Angles (deg) for
{Yb(THF)2(Et2O)2[(μ-(CO)2Mn(CO)3)2]}∞ (3) and {Yb(THF)4[(μ-(CO)2Mn(CO)3)2]}∞ (4)……………………………………………..63
Table 2.8 Crystallographic Data for
(THF)2Mn3(CO)10 (5) and
[(THF)5Yb(μ-CO)Mn3(CO)13][Mn3(CO)14] (6)………………………………………….69
xv
Table 2.9 Selected Bond Distances (Å) and
Bond Angles (deg) for (THF)2Mn3(CO)10 (5)…………………………………………...70
Table 2.10 Selected Bond Distances (Å) and
Bond Angles (deg) for (THF)2Mn3(CO)10 (5)…………………………………………...74
Table 3.1 Compositions of complexes 1a, 1b, 1c,
2a, 2b, 2c, 3a, 3b, 3c……………………………………………………………………..81
Table 3.2 Crystallographic Information for
Compounds 1a, 1b, 1c, 2a, 2b, 2c, 3a, 3b, 3c…………………………………………...83
Table 3.3 Selected Bond Distances (Å) and
Bond Angles (deg) for 7a
{(THF)4Yb[Cp(μ-CO)Cr(CO)2] [Cp(μ-CO)2Cr(CO)]}∞………………………………...88
Table 3.4 Selected Bond Distances (Å) and
Bond Angles (deg) for {(THF)nYb[Cp(μ-CO)2M(CO)]2}∞
(8a, M = Mo, n = 4; 9a, M = W, n = 3)………………………………………………….93
xvi
Table 3.5 Selected Bond Distances (Å) and
Bond Angles (deg) for {(CH3CN)4Yb[Cp(μ-CO)2M(CO)]2}∞
(7b, M = Cr; 8b, M = Mo)……………………………………………………………...100
Table 3.6 Selected Bond Distances (Å) and
Bond Angles (deg) for 9b
(CH3CN)6Yb[Cp(μ-CO)W(CO)2]2 and
(DME)3Yb[Cp(μ-CO)M(CO)2]2
(M = Cr, 7c; M = Mo, 8c; M = W, 9c)………………………………………………….104
Table 3.7 Solution Infrared Data for Compounds
7a, 7b, 7c, 8a, 8b, 8c, 9a, 9b, 9c in the Carbonyl
Stretching Frequency Region…………………………………………………………...106
Table 3.8 Bond distances (Å) for the bridging carbonyl ligands………………………113
Table 4.1. Crystallographic Data for the
[Yb(DMF)7][W2(CO)10] (10),
{(CH3CN)6Yb(W2(CO)10)·CH3CN}∞ (11),
{(pyr)5Yb[Hg(W(CO)5)2]}∞ (12) and (pyr)3Cr(CO)3 (13)……………………………...119
xvii
Table 4.2. Bond Distances (Å) and Bond Angles (deg) for the [Yb(DMF)7][W2(CO)10] (10) and the {(CH3CN)6Yb(W2(CO)10)·CH3CN}∞ (11)………………………………………….121
Table 4.3. Infrared Data in the Carbonyl
Stretching Frequency Region for Complexes
[Yb(DMF)7][W2(CO)10] (10),
{(CH3CN)6Yb(W2(CO)10)·CH3CN}∞ (11),
{(pyr)5Yb[Hg(W(CO)5)2]}∞ (12) and (pyr)3Cr(CO)3 (13)……………………………...126
Table 4.4. Bond Distances (Å) and Bond Angles (deg) for the 1-D Polymeric Chain {(pyr)5Yb[Hg(W(CO)5)2]}∞ (12)………………………..136
Table 4.5. Bond Distances (Å) and
Bond Angles (deg) for the (pyr)3Cr(CO)3 (13)…………………………………………145
xviii
LIST OF FIGURES
Figure 1.1. The {(DMF)10Yb[Pd(CN)4]3}, the precursor for the catalyst from Scheme 1.1.
DMF ligands are omitted for clarity………………………………………………………7
Figure 1.2. The (THF)Cp2LuRu(CO)2Cp molecule containing direct Ru-Lu bond…………………………………………………11
Figure 1.3. Molecular structures of the
{[(CH3CN)3YbFe(CO)4]2•CH3CN}∞ (a) and the {(CH3CN)3YbFe(CO)4}∞ (b).
CH3CN ligands and the solvent of crystallization are omitted for clarity…………………………………………………….13
2+ Figure 1.4. (a) The [Yb(pyr)6] cation,
2+ (b) the [Sm(DIME)3] cation……………………………………………………………15
xix
Figure 1.5. Solution IR spectra of (a) [Yb(Pyr)6] [Co(CO)4]2 ,
(b) [Yb(THF)6][Co(CO)4]2 (0.0013M),
(c) [Yb(THF)6] [Co(CO)4]2 (0.021M),
(d) [Yb(DME)4] [Co(CO)4]2, and (e) [Eu(THF)x][Co(CO)4]2…………………………...16
* Figure 1.6. The Cp 2Yb(THF){(μ-CO)Co(CO)3} complex……………………………..18
Figure 1.7. Complexes {(pyr)4Yb(μ-CO)2Co(CO)2}∞ (C) and {(Et2O)2(THF)Yb[Co4(CO)11]}∞ (D) obtained via condensation of the [Yb(L)n][Co(CO)4]2 parent complex in toluene and diethyl ether, respectively.
Ligands coordinated to the Yb(II) cation are omitted for clarity………………………...21
Figure 1.6. Molecular structure of Hg[Co(CO)4]2………………………………………24
Figure 2.1. Molecular structure of Hg[Mn(CO)5]2……………………………………...33
xx
Figure 2.2. Molecular structure of 1
2+ (a) the [Yb(THF)6] cation (25% thermal ellipsoids),
- (b) the [Mn(CO)5] anion (non-disordered)
(15% thermal ellipsoids). Yb1 lies on a center of symmetry…………………………….42
- Figure 2.3. The [Re(CO)5] anions in 2
(25% thermal ellipsoids) (a) anion without disorder,
(b) disordered anion……………………………………………………………………...43
Figure 2.4. Molecular structure (10% probability thermal ellipsoids)
- showing the disordered [Mn(CO)5] anion in 1. The site occupancy factors are 0.53 for Mn2, C8, O8, C9, O9, C10,
O10; 0.47 for Mn2’, C8’, O8’, C9’, O9’, C10’, O10’…………………………………...49
Figure 2.5. Molecular structure of 3. (a) asymmetric unit (25% thermal ellipsoids), (b) ring containing eight metal atoms, Yb1 lies on a 2-fold rotation axis……………………………………55
xxi
Figure 2.6. Views of a portion of the 2-D polymeric sheet of 3.
(a) top view (along c axic), (b) cross-section (view along b axis).
Ytterbium atoms shown in red, manganese in green, carbon in grey and oxygen in blue……………………………………………………….56
Figure 2.7. Molecular structure of 4. (a) asymmetric unit
(25% thermal ellipsoids), (b) a ‘diamond’
(15% thermal ellipsoids), (c) portion of a 1-D linear chain of 4 (only oxygen atoms of THF ligands are shown for clarity). Yb1 lies on a 2-fold rotation axis……………………………………57
Figure 2.8. Coordination geometry around YbII center in 4.
Yb1 lies on a 2-fold rotation axis………………………………………………………...66
Figure 2.9 Molecular structure of 5. 35% probability ellipsoids shown………………..68
xxii
Figure 2.10 Molecular structure of 6
(35% probability ellipsoids shown): (a) the
+ [(THF)5Yb(μ-CO)Mn3(CO)13] cation (only oxygen
- atoms of THF shown); (b) the [Mn3(CO)14] anion.
Yb1 and Mn2 lie on a common mirror plane and Mn3 lies on a center of symmetry…………………………………………………...73
Figure 3.1 Molecular structure of
{Yb(THF)4[Cp(μ-CO)Cr(CO)2][Cp(μ-CO)2Cr(CO)]}∞ 7a.
(a) asymmetric unit (35% probability ellipsoids);
(b) view down crystallographic b axis;
(c) view down crystallographic a axis. Only oxygen atoms of
THF ligands are shown in (b) and (c)……………………………………………………87
xxiii
Figure 3.2 Molecular structure of {Yb(THF)4(Cp(μ-CO)2Mo(CO)}∞ 8a.
(a) asymmetric unit (35% probability ellipsoids);
(b) view of two ‘eight-member rings’ down crystallographic a axis;
(c) view down crystallographic b axis. Only oxygen atoms of THF ligands are shown in (b) and (c). One of THF ligands is disordered, see Appendix A…………………………………………………...91
Figure 3.3 Molecular structure of {Yb(THF)3(Cp(μ-CO)2W(CO)}∞ 9a.
(a) asymmetric unit (35% probability ellipsoids);
(b) view down crystallographic a axis; (c) view down crystallographic b axis. Only oxygen atoms of
THF ligands are shown in (b) and (c)……………………………………………………92
Figure 3.4 Molecular structure of the discrete molecule
21a (THF)5La[(μ-CO)Mo(CO)2Cp]3 obtained by Beletskaya et al .
Only oxygen atoms of THF ligands are shown for clarity……………………………….95
xxiv
Figure 3.5 Molecular structure of {Yb(CH3CN)4[Cp(μ-CO)2Cr(CO)]2}∞ 7b.
(a) asymmetric unit (35% probability ellipsoids);
(b) view down crystallographic a axis; (c) view down crystallographic b axis. Only nitrogen atoms of acetonitrile ligands are shown in (b) and (c)……………………………………………..98
Figure 3.6 Molecular structure of {Yb(CH3CN)4(Cp(μ-CO)2Mo(CO))2}∞ 8b.
(a) asymmetric unit (35% probability ellipsoids);
(b) view down crystallographic a axis; (c) view down crystallographic b axis. Only nitrogen atoms of acetonitrile ligands are shown in (b) and (c)……………………………………………..99
Figure 3.7 Molecular structure of Yb(CH3CN)6(Cp(μ-CO)W(CO)2)2 9b
(35% probability ellipsoids)…………………………………………………………….102
xxv
Figure 3.8. Molecular structure of compounds
Yb(DME)3(Cp(μ-CO)M(CO)2)2 7c (M = Cr), 8c (M = Mo),
9c (M = W); 35% probability ellipsoids. (a) Entire molecule, only oxygen atoms of DME ligands are shown for clarity;
(b) The Yb(II) center with the three DME ligands.
In compounds 7c, 8c, 9c one of the three DME ligands is disordered, only major portion is shown (see Appendix B)………………………….103
Figure 4.1 Molecular structure of the [Yb(DMF)7][W2(CO)10] (10)
2+ (35% probability ellipsoids). (a) the [Yb(DMF)7] cation, only oxygen atoms of the DMF ligands are shown for clarity;
2- (b) the [W2(CO)10] anion……………………………………………………………...118
Figure 4.2 Molecular structure of the
{(CH3CN)6Yb[W2(CO)10]•CH3CN}∞ (11)
(a) the asymmetric unit (35% probability ellipsoids), only oxygen atoms of the CH3CN ligands are shown for clarity, the co-crystallized molecule of CH3CN is omitted; (b) view of the 1-D chain of 11…………………………………129
xxvi
Figure 4.3 Molecular structure of the {(pyr)5Yb(Hg(W(CO)5)2)}∞ (12).
(a) asymmetric unit (35% probability ellipsoids), only nitrogen atoms of pyridine ligands are shown for clarity;
(b) view of the 1-D chain of 12…………………………………………………………135
Figure 4.4 Molecular structure on the (pyr)3Cr(CO)3 (13)
(35% probability ellipsoids)…………………………………………………………….144
Figure A.1 Molecular structure (25% probability thermal ellipsoids) showing the disordered THF ligand in 1. The site occupancy factors are
0.47 for C21, C22, C23, C24; 0.53 for C21’, C22’, C23’, C24’……………………….151
Figure A.2 Molecular structure (25% probability thermal ellipsoids) showing the disordered THF ligands in 2 (asymmetric unit shown).
The site occupancies are 0.29 for C21, C22, C23, C24;
0.71 for C21’, C22’, C23’, C24’; 0.42 for C51, C52, C53, C54;
0.58 for C51’, C52’, C53’, C54’;
0.46 for C61, C62, C63, C64; 0.54 for C61’, C62’, C63’, C64’……………………….152
xxvii
Figure A.3 Molecular structure (15% probability thermal ellipsoids) showing the disordered THF and Et2O ligands in 3.
The occupancy factors are 0.53 for C11, C12, C13, C14;
0.47 for C11’, C12’, C13’, C14’; 0.27 for C21, C22,
C23, C24; 0.73 for C21’, C22’, C23’, C24’……………………………………………153
Figure A.4 Molecular structure (15% probability thermal ellipsoids) showing the disordered THF ligand and disordered oxygen atoms of the carbonyl groups in the asymmetric unit of 4. The site occupancy factors are 0.56 for O1; 0.44 for O1’; 0.74 for O2; 0.26 for O2’;
0.54 for C16, C26, C36, C46; 0.46 for C16’, C26’, C36’, C46’;
0.39 for C17, C27, C37, C47; 0.61 for C17’, C27’, C37’, C47’……………………….154
Figure A.5 Molecular structure (25% probability thermal ellipsoids) showing the disordered THF ligand in 5. The site occupancy factors are 0.36 for C31, C32, C33, C34 and 0.64 for C31’, C32’, C33’, C34’………..155
Figure A.6 Molecular structure (15% probability thermal ellipsoids for two top ligands, 35% for the bottom ligand) showing the disordered THF ligands in 6. The site occupancy factors are 0.59 for C61, C62, C63, C64; 0.41 for C61’, C62’, C63’, C64’……………156
xxviii
Figure A.7 Molecular structure (15% probability thermal ellipsoids
for two top ligands, 35% for the bottom ligand) showing the
disordered THF ligands in 6. The site occupancy factors are
0.50 for C71 C72, C73, C74, C71A, C72A,
C73A, C74A, C81, C82, C81A, C82A…………………………………………………157
Figure A.8 Disordered THF ligand in {(THF)4Yb[Cp(μ-CO)2Mo(CO)]2}∞ (8a),
35% probability ellipsoids. This is the only disordered ligand
in the structure; all other THF molecules do not show disorder.
Atoms C21, C22, C23, C24 have occupancy factor 0.55.
Atoms C21’, C22’, C23’, C24’ have occupancy factor 0.45…………………………...158
Figure A.9 Disordered DME ligand in structures (DME)3Yb[Cp(μ-CO)M(CO)2]2
(M = Cr, 7c; M = Mo, 8c; M = W, 9c), 35% probability ellipsoids.
- The [Cp(μ-CO)M(CO)2] anions are omitted for clarity.
Atoms O6, O7, O6A, O7A, C13, C14, C15, C16,
C13A, C14A, C15A, C16A have occupancy factor 0.50
imposed by a 2-fold rotation axis……………………………………………………….159
Figure B.1 IR spectrum of [Yb(THF)6][Re(CO)5]2 (2) in THF………………………..161
xxix
Figure B.2 IR spectrum of [Yb(THF)6][Mn(CO)5]2 (1) in THF prepared via the Hg[Mn(CO)5]2 route………………………………………….162
Figure B.3 IR spectrum of NaRe(CO)5 in pyridine……………………………………163
Figure B.4 IR spectrum of [Yb(THF)6][Mn(CO)5]2 (1) in
THF prepared via ytterbium amalgam route……………………………………………164
Figure B.5 IR spectrum of [Yb(DME)n][Mn(CO)5]2 (1a) in DME……………………165
Figure B.6 IR spectrum of [Yb(DMF)n][Mn(CO)5]2 (1b) in DMF…………………….166
Figure B.7 IR spectrum of [Yb(pyr)6][Mn(CO)5]2 (1c) after stirring in Et2O overnight; the spectrum is recorded in diethyl ether……………..167
Figure B.8 IR spectrum of [Yb(THF)6][Mn(CO)5]2 (1) freshly dissolved in Et2O; initial stages of formation of 3……………………………...168
Figure B.9 IR spectrum of the solution of [Yb(THF)6][Mn(CO)5]2 (1) in Et2O in ca. 18 hours………………………………………………………………….169
xxx
Figure B.10 IR spectrum of
{(THF)3(Et2O)[(μ-CO)2Mn(CO)3]2}∞ (3) in KBr pellet………………………………..170
Figure B.11 IR spectrum of {(THF)4Yb[(μ-CO)2Mn(CO)3]2}∞ in KBr pellet………...171
Figure B.12 IR spectra of NaMn(CO)5 (dotted line) and
[Yb(pyr)n][Mn(CO)5]2 (1) (solid line) in pyridine……………………………………...172
Figure B.13 IR spectrum of
{(THF)4Yb[Cp(μ-CO)Cr(CO)2][Cp(μ-CO)2Cr(CO)]}∞ (7a) in THF………………….173
Figure B.14 IR spectrum of {(THF)4Yb[Cp(μ-CO)2Mo(CO)]2}∞ (8a) in THF……….174
Figure B.15 IR spectra of Na[CpW(CO)3] (dotted line) and {(THF)3Yb[Cp(μ-CO)2W(CO)]2}∞ (9a) (solid line) in THF………………………175
Figure B.16 IR spectrum of {(CH3CN)4Yb[Cp(μ-CO)2Cr(CO)]2}∞ (7b) in CH3CN (also showing spectrum of a very dilute solution of [CpCr(CO)3]2 in CH3CN, dotted line)……………………………………………….176
xxxi
Figure B.17 IR spectrum of
{(CH3CN)4Yb[Cp(μ-CO)2Mo(CO)]2}∞ (7b) in CH3CN...... 177
Figure B.18 IR spectrum of (CH3CN)6Yb[Cp(μ-CO)W(CO)2]2 (9b) in CH3CN……...178
Figure B.19 IR spectra of Hg[CpCr(CO)3]2 in pyridine
(dotted line) and (DME)3Yb[Cp(μ-CO)Cr(CO)2]2 (7c) in DME……………………….179
Figure B.20 IR spectrum of (DME)3Yb[Cp(μ-CO)Mo(CO)2]2 (8c) in DME………….180
Figure B.21 IR spectrum of (DME)3Yb[Cp(μ-CO)W(CO)2]2 (9c) in DME………….181
- Figure B.22 IR spectra of sodium salts of the [CpCr(CO)3]
- (solid line), the [CpMo(CO)3] (dashed line) and
- the [CpW(CO)3] (dotted line) recorded in pyridine……………………………………182
xxxii
LIST OF SCHEMES
Scheme 1.1. (a) Hydrogenation of phenol;
(b) hydrodechlorination of dichlorobenzene………………………………………………5
Scheme 1.2. Preparation of the {[(CH3CN)3YbFe(CO)4]2•CH3CN}∞
(A) and the {(CH3CN)3YbFe(CO)4}∞ (B), complexes that contain direct Yb – Fe bond……………………………………………..13
- Scheme 1.3. Summary of condensation reactions for the [Co(CO)4] anion…………….20
- Scheme 1.4. Condensation of the [Co(CO)4] anion into
- the [Co4(CO)11] cluster anion accompanied by reduction of the Yb2+ cation to ytterbium metal……………………………………………………27
Scheme 2.1 Proposed mechanism of formation of complexes 3 and 4 out of parent compound 1…………………………………………...60
xxxiii
Scheme 3.1 Preparation of molybdenum- and tungsten-containing complexes 8a, 8b, 8c, 9a, 9b, 9c…………………………………...79
Scheme 3.2 Preparation of chromium-containing complexes 7a, 7b, 7c………………..80
Scheme 3.3 Reduction of the [CpMo(CO)3]2 with ytterbium and with lanthanum……...94
Scheme 4.1 Preparation of compounds [Yb(DMF)7][W2(CO)10] (10),
{(CH3CN)6Yb(W2(CO)10)·CH3CN}∞ (11) and {(pyr)5Yb[Hg(W(CO)5)2]}∞ (12)…………………………………………………..116
Scheme 4.1 Proposed reaction pathway for transformation of
2- 69 the [MFe(CO)4]x into the [Fe2(CO)8] anion …………………………………………141
xxxiv
LIST OF CHARTS
Chart 1.1 Three types of Ln – TM carbonyl compounds………………………………...3
Chart 1.2. (a) Scheme of binding of the carbonyl ligand to the transition metal; (b) Scheme of coordination of the lanthanide cation to the carbonylate anion
(a portion of the anion is shown)………………………………………………………….9
Chart 1.3. IUPAC nomenclature for the three known types of isocarbonyl bridges……………………………………………………..18
xxxv
LIST OF ABBREVIATIONS
CO carbonyl ligand
- Cp Cyclopentadienylide ion, [C5H5]
* - Cp Pentamethylcyclopentadienylide ion, [C5Me5]
DME 1,2-Dimethoxydiethane, glyme
DMF N,N-Dimethylformamide e electron
Et ethyl
Et2O Diethyl ether
IR infrared
IUPAC International Union of Pure and Applied Chemistry
L A 2e ligand, Lewis base
Ln Lanthanide element
M Transition metal pyr pyridine
THF Tetrahydrofuran, C4H8O
xxxvi
Tol Toluene
ηn Descriptor of hapticity, superscript designates number of bridging sites
μ Descriptor for bridging
Tol Toluene
xxxvii
LIST OF COMPOUND NUMBERS
COMPOUND NUMBER
{[(CH3CN)3YbFe(CO)4]2•CH3CN}∞ A
{(CH3CN)3YbFe(CO)4}∞ B
{(pyr)4Yb(μ-CO)2Co(CO)2}∞ C
{(Et2O)2(THF)Yb[Co4(CO)11]}∞ D
(THF)5La[(μ-CO)Mo(CO)2Cp]3 E
[Yb(THF)6][Mn(CO)5]2 1
[Yb(DME)n][Mn(CO)5]2 1a
[Yb(DMF)n][Mn(CO)5]2 1b
[Yb(pyr)n][Mn(CO)5]2 1c
[Yb(THF)6][Re(CO)5]2 2
{Yb(THF)2(Et2O)2[(μ-CO)2Mn(CO)3]2}∞ 3
{Yb(THF)4[(μ-CO)2Mn(CO)3]2}∞ 4
(THF)2Mn3(CO)10 5
[(THF)5Yb(μ-CO)Mn3(CO)13][Mn3(CO)14] 6
xxxviii
{(THF)4Yb[Cp(μ-CO)Cr(CO)2][Cp(μ-CO)2Cr(CO)]}∞ 7a
{(CH3CN)4Yb[Cp(μ-CO)2Cr(CO)]2}∞ 7b
(DME)3Yb[Cp(μ-CO)Cr(CO)2]2 7c
{(THF)4Yb[Cp(μ-CO)2Mo(CO)]2}∞ 8a
{(CH3CN)4Yb[Cp(μ-CO)2Mo(CO)]2}∞ 8b
(DME)3Yb[Cp(μ-CO)Mo(CO)2]2 8c
{(THF)3Yb[Cp(μ-CO)2W(CO)]2}∞ 9a
(CH3CN)6Yb[Cp(μ-CO)W(CO)2]2 9b
(DME)3Yb[Cp(μ-CO)W(CO)2]2 9c
[Yb(DMF)7][W2(CO)10] 10
{(CH3CN)6Yb(W2(CO)10)·CH3CN}∞ 11
{(pyr)5Yb[Hg(W(CO)5)2]}∞ 12
(pyr)3Cr(CO)3 13
xxxix
CHAPTER 1
INTRODUCTION
1.1 General Review of Ln - TM Complexes
The past three decades witnessed a rise of interest in lanthanide – transition-metal
(Ln-TM) complexes. Although first members of this (now) extensive family of
compounds were prepared in the 1910s1, they had to wait until the 1970s2 for a new wave
of interest that was spurred by increased availability of the X-ray diffraction techniques to a wide circle of researchers. Since then, a number of structures that contain a transition metal atom and that of a lanthanide have been prepared and characterized. Some of these
compounds are extended ionic solids, while others have discrete molecular structure. In
some of them, the lanthanide and the transition metal are situated in close proximity to
each other, and in others they are quite apart. Generally speaking, the two different metal atoms have to have an auxiliary ligand of some sort between them, although a few important exceptions from this rule are known3. The closer we want to bring a lanthanide
and a transition metal together, the more compact that ligand should be. Apart from
1
fundamental research interest, such bimetallic arrangements promise new and exciting
opportunities for practical applications in the fields of materials science4 and catalysis5.
Historically, the first type of Ln-TM compounds prepared was a series of
1 hexacyanocobaltates Ln[Co(CN)6]•nH2O (Ln = La, Ce, Nd, Gd, Yb, Y) . As the later X-
ray structure determination revealed2, the structures of these compounds contained
lanthanide and cobalt atoms linked together by an isocyanide bridge, thus forming the
Ln-N≡C-Co motif. These were the first known Ln-TM cyanides. Much later, in the
1970s, a closely related group of compounds was prepared6, that of Ln-TM carbonyls, where instead of the isocyanide bridge the isocarbonyl –C≡O- is present. Determination of the structure of these compounds and understanding the mode of binding of the lanthanide and the transition metal in them was accomplished soon after7. According to
the present day classification8, one can divide these complexes into three types (Chart
1.1): (I) Systems with direct Ln–M bond, (II) Solvent-separated ion pairs, and (III)
Systems with bridging ligands, cyanide or carbonyl being the most important for this work. It is believed* that the binding mode in the structure – i.e. whether the complex of
type I, II or III forms – is determined by interplay of the Lewis basic/acidic properties of
the three essential components: the lanthanide ion, the carbonyl (or cyanide) group and
the solvent. The Lnn+ cation is the Lewis acid in this ensemble, and it tends to bind to the strongest base available. If this happens to be the transition metal center, then the direct
Ln-TM bond is formed (Type I). When the solvent molecule is the strongest base, it
solvates the cation, and the solvent-separated ion pairs result (Type II).
2
(L)xLn - M(CO)y [Ln(solvent)x]n[M(CO)y]m I II
(L)xLn - OC - M(CO)y III
Chart 1.1 Three types of Ln – TM carbonyl compounds
Finally, if the cyanide or carbonyl ligands are the most basic species available,
they coordinate to the Lnn+ center, and the result is a system with bridging ligands (Type
III). It is necessary to add that this reasoning deals with the compounds’ solid state structures. In solution, two (or possibly three) binding schemes might be present; it is very dependent on the solution’s concentration8, 9.
This dissertation primarily deals with the Ln-TM carbonyl structures, and because
of that, a closer examination of both Ln-TM carbonyls and their close relatives, Ln-TM cyanides would be helpful.
3
1.2 Ln-TM Cyanides
In the solid state, the main mode of bonding between the lanthanide and the
transition metal in compounds of this class is through the isocyanide bridge -N≡C-. The more electron-rich N-end coordinates to the electrophilic Ln(II) or Ln(III) ion, and the C- end is bound to the transition metal atom. These structures are often prepared starting from a simple lanthanide salt and another salt containing the transition metal cyanide
3- 10e anion, such as [Ni(CN)4] (Equation (1.1)) . Varying the lanthanide, the transition
metal cyanide and the solvent, it is possible to synthesize a remarkable range of extended
solid polymers that include chains10, layers11, columns11a, and 3-dimensional ionic
arrays11b.
DMF 2LnCl3 + K2[M(CN)4] {(DMF)10Ln2[M(CN)4]3} (1.1) -6KCl
The structural variety of these systems still wants exploration, despite the vast
amount of research that has already been done on them. One of the reasons that promoted
active research in the field of Ln-TM cyanides is their potential for future applications.
Isocyanide-containing Ln-TM systems proved to be precursors for superior reduced bimetallic heterogeneous catalysts for such important processes as aqueous vapor- 4
Cl Cl Cl OH O H H2 2 catalyst catalyst
2H2 H2
(a)
(b)
Scheme 1.1. (a) Hydrogenation of phenol; (b) hydrodechlorination of
dichlorobenzene.
phase hydrogenation of phenol5e and hydrodechlorination of chlorobenzenes5f (Scheme
1.1, Figure 1.1). These catalysts have demonstrated improved activity and selectivity over the ones containing only the corresponding transition-metal. The reason for such improved characteristics is believed to rest in the close interaction between the reduced lanthanide and the transition-metals produced from the mixed Ln-TM precursors. The polymeric framework of the precursor allows the uniform distribution of the metals over a catalyst’s support surface5d. When the precursor is reduced to bimetallic nanoparticles,
the more electropositive lanthanide can enhance reducing properties of the transition
metal by transferring electron density to the latter. Introduction of such catalysts could be
useful both industrially and environmentally, since chlorobenzenes, as well as phenol, are
5
large-scale industrial toxins. The improved selectivity of the mixed Ln-TM catalysts is
also very advantageous for the process of conversion of phenol to cyclohexanone,
because cyclohexanone is used to prepare caprolactam, a starting material in the
preparation of nylon 6.
Apart from their possible applications in catalysis, the Ln-TM cyanides drew
attention of material scientists. Extended structures, in which a lanthanide and a transition
metal are placed relatively close to each other, are attractive for creating novel
magnetic12, fluorescent13, electroceramic14 and chemical sensor materials15. Researches hunting for materials with new magnetic properties seem to place especially high hopes on Ln-TN combinations, because of the rich variety of spin combinations that could result from interactions between the two metals.
6
Figure 1.1. The {(DMF)10Yb[Pd(CN)4]3}, the precursor for the catalyst from
Scheme 1.1. DMF ligands are omitted for clarity.
7
1.3 Ln-TM Carbonyls
1.3.1 General Remarks
These compounds are closely related to the Ln-TM cyanide family, having a
carbon monoxide CO ligand instead of CN in their structure. Despite such close
relationships, Ln-TM carbonyls are considerably less studied than the cyanide
compounds. This fact naturally invites more investigations to be done on their syntheses
and chemical properties. Ln-TM carbonyls are potentially important for preparing mixed
Ln–TM catalysts5, 16. Such structures allow uniform dispersion of the metals over the
support’s surface, with the thoroughness of mixing the two metals that can hardly be achieved by any other method5d. This and the potentially useful magnetic properties12 make Ln-TM carbonyls an attractive object for research.
Bonding of the CO ligand to the transition metal and the effect of coordination of the Lnn+ cation to the O-end of the CO is represented in Chart 1.2. Carbon monoxide is a
σ-donor, π-acceptor ligand. Its lone pair of electrons donates its density to the vacant d-
orbital of the metal, and the electron density of the filled d-orbitals is donated to the
empty π*-orbitals of the CO. When a positively charged lanthanide cation coordinates to
the oxygen atom, it drains electrons from the ligand. This causes shift of the electron
density from the transition metal to the carbonyl ligand to which the Lnn+ is bound, and
those electrons go to the π*-antibonding orbitals of the CO. As a result, the bonding order
within the CO molecule decreases, and the C-O distance becomes greater. The IR 8
π -acceptance
Ln(L)x σ - donation M C O M CO C O
(b) (a)
Chart 1.2. (a) Scheme of binding of the carbonyl ligand to the transition metal; (b)
Scheme of coordination of the lanthanide cation to the carbonylate anion (a portion
of the anion is shown).
signal of such bridging CO (since it now serves as bridge between the lanthanide cation
and the transition metal) shifts to lower wavenumbers. At the same time, this draining of
the electrons moves the electron density from the rest of the CO ligands to the transition
metal, thus removing it from the π*-orbitals. It leads to higher order of bonding, decreases the C-O distance, and shifts the IR signals of these terminal ligands to the higher
frequency region.
9
As it was mentioned above (Section 1.1), these species can be classified under the
three rubrics. These will be considered in more detail in the following subsections.
1.3.2 Systems with direct Ln–M bond
Systems of this type (Type I, see Chart 1.1) are the rarest among the three types of
Ln-TM carbonyls. To the best of my knowledge, there are only three instances when the
direct Ln-TM bonds were confirmed by X-ray structure determinations3. Other claims
that such compounds had been prepared appeared in the literature6a, 17, based mostly on
evidence obtained from NMR or the solution IR. Beltskaya and coworkers3b, 18 reasoned that a strong Lewis base containing a transition metal atom had a good chance to react with a Ln(III) center, which is a Lewis acid. In order for the direct Ln-TM bond to be formed, the electron density of the base must be localized on the TM atom. Following
- this train of thought, Beletskaya’s group employed the [CpRu(CO)2] anion, that meets
the above requirements, and Ln(III)-containing complex Cp2LuCl(THF) that is soluble in
common polar solvents (THF, DMF, DME). The outcome of the reaction was a discrete
molecule (THF)Cp2LuRu(CO)2Cp (Equation (1.2)).
THF Cp2LuCl(THF) + Na[Ru(CO)2Cp] (THF)Cp2LuRu(CO)2Cp (1.2)
10
This molecule is the first example of the Ln-TM carbonyl compound where the presence of the direct bond between the two metal centers was unequivocally confirmed by means of the X-ray data collection. The molecular structure of this compound is shown in Figure 1.2, and clearly reveals the direct bond between the Ru and Lu atoms.
This bond’s length is 2.995(2) Å, which is slightly shorter than the sum of the metallic radii, 3.08 Å. The compound is reasonably stable, and the Lu-Ru bond persists in the solution, as evidenced by the IR spectrum that shows carbonyl stretching frequencies at
-1 - 2027, 1965 cm , that are higher in energy than those of the free [CpRu(CO)2] anion.
Figure 1.2. The (THF)Cp2LuRu(CO)2Cp molecule containing direct Ru-Lu bond.
11
The Beletskaya group also carried out reactions analogous to (1.2) using the
- [CpFe(CO)2] anion as a Lewis base, and they claim that the direct bond was formed
between the Lu and Fe atoms based on the IR spectrum3b. However, the product was
unstable and decomposed rapidly, so that no more data could be obtained. This group
* also used the {Cp”2Lu(μ-Cl)}2 and the Cp 2Lu(μ-Cl)2Na(THF)2 in combination with the
- [CpRu(CO)2] anion, with the alleged products being the (THF)Cp”2LuRu(CO)2Cp and
* the Cp 2LuRu(CO)2Cp molecules, respectively. However, no X-ray data were collected
for these structures.
Two more complexes of this type were prepared in the Shore laboratory3a, c. This
research group was focusing on the divalent lanthanides, believing that these are milder
Lewis acids than their trivalent analogs, and because of that have a better chance to form
2- a stable bond to the TM atom. The [Fe(CO)4] anion was chosen as the counterpart base,
and reduction of Fe3(CO)12 with Yb metal in liquid ammonia has lead to an insoluble
solid, for which the formula (NH3)2YbFe(CO)4 was suggested on the basis of the elemental analysis. Dissolving this compound in acetonitrile CH3CN displaced the ammonia ligands, and two slightly different work-up routes produced two structures (A
and B) where there is a direct Yb-Fe bond (Scheme 1.2). These structures are shown in
Figure 1.3. Compound A has a ladder-type 1D chains (Fig. 1.3a), and compound B is a
two-dimensional sheet (Fig. 1.3b). These structures are related, and one can imagine that
the B is created from the A by linking its ‘ladders’ together with the additional CO
bridging ligands.
12
CH3CN
{[(CH3CN)3YbFe(CO)4]2*CH3CN}8 A
NH3(l) 3Yb + Fe3(CO)12 3(NH3)2YbFe(CO)4
{(CH3CN)3YbFe(CO)4}8 1) NH3/CH3CN 2) remove volatiles B 3) CH3CN
Scheme 1.2. Preparation of the {[(CH3CN)3YbFe(CO)4]2•CH3CN}∞ (A) and the
{(CH3CN)3YbFe(CO)4}∞ (B), complexes that contain direct Yb – Fe bond.
(a) (b)
Figure 1.3. Molecular structures of the {[(CH3CN)3YbFe(CO)4]2•CH3CN}∞ (a) and the
{(CH3CN)3YbFe(CO)4}∞ (b). CH3CN ligands and the solvent of crystallization are omitted for clarity. 13
The Yb-Fe bond lengths in A and B are 3.010[2] Å and 3.046(1) Å, respectively.
This is shorter than the sum of their metallic radii (3.2 Å)19, and comparable to the
20 distance in the YbFe2 alloy (3.00 Å) .
1.3.2 Solvent-Separated Ion Pairs
In this group of Ln-TM carbonyl compounds (Type II, Chart 1.1), the cation is
surrounded with molecules of solvent, and thus is well separated from the anion. This
situation usually arises when the metal carbonyl anion is a weak Lewis base, and cannot
compete with the molecules of solvent for the Lnn+ center. A number of complexes of this
type have been prepared, and the [Yb(pyr)6][Co(CO)4]2 and the [Sm(DIME)3][Mn(CO)5]2 can be used as representative examples (Figure 1.4). The former complex was prepared via a transmetallation reaction8a, where the Yb metal reduces mercury in the starting
material (Equation (1.3)). The latter compound was prepared by reduction of the metal
21 carbonyl dimer, Mn2(CO)10, with a samarium amalgam in DIME (Equation (1.4)). In
either case the freshly formed Ln(II) cation coordinates to the maximum number of
solvent molecules. In these examples, a coordination number of six (six nitrogen atoms in
2+ 2+ [Yb(pyr)6] ) or nine (nine oxygen atoms in [Sm(DIME)3] ) is observed, the latter being
somewhat atypical for the lanthanide cations.
Although compounds of this class have ionic salt-like solid structures, some
covalent interaction takes place in solution between the carbonyl ligands of the anion 14
(b)
(a)
2+ 2+ Figure 1.4. (a) The [Yb(pyr)6] cation, (b) the [Sm(DIME)3] cation.
pyr Yb + Hg[Co(CO)4]2 [Yb(pyr)6][Co(CO)4]2 + Hg (1.3)
DIME Sm/Hg + Mn2(CO)10 [Sm(DIME)3][Mn(CO)5]2 (1.4)
15
Figure 1.5. Solution IR spectra of (a) [Yb(Pyr)6] [Co(CO)4]2 , (b) [Yb(THF)6][Co(CO)4]2
(0.0013M), (c) [Yb(THF)6] [Co(CO)4]2 (0.021M), (d) [Yb(DME)4] [Co(CO)4]2, and
(e) [Eu(THF)x][Co(CO)4]2
16
and the lanthanide cation8, 9. Evidence for it is supplied by the solution IR (Figure 1.58e).
m- At low concentrations, only absorptions belonging to the undistorted [M(CO)n] anion are seen.
As the concentration increases, two more sets of peaks emerge, one in the higher and one in the lower frequency region (compared to the original, low-concentration spectrum). This is a sign of formation of weakly coordinated pairs in solution
(S)nLn···OCM(CO)n-1, when the carbonyl ligand penetrates the layer of solvent
surrounding the cation. Thus the formerly equivalent CO ligands are divided into
bridging and terminal ones. According to the well-known explanation8, 9, the bridging set gives rise to the absorption in lower wavenumbers, and the terminal set is seen in the higher energy IR region. The higher the concentration, the greater the intensity of these side bands becomes.
1.3.3 Systems with Bridging Ligands (Ln-TM Isocarbonyls)
Ln-TM isocarbonyls (Type III, Chart 1.1) constitute the third family of structures
depicted in Chart 1.1. The peculiarity of this mode of binding is that the bridging
carbonyl ligand can be attached to more than one transition metal at its C-end. The appropriate IUPAC nomenclature for the three known types of isocarbonyl bridges is illustrated in Chart 1.3.
17
M' M' M' O O O C C C M M M M M M
2 2 2 η ,μ2-CO η ,μ3-CO η ,μ4-CO μ-CO
Chart 1.3. IUPAC nomenclature for the three known types of isocarbonyl bridges.
* Figure 1.6. The Cp 2Yb(THF){(μ-CO)Co(CO)3} complex
18
Andersen and co-workers pioneered the structural investigations of complexes of
7 * this type . Molecular structure of the first Ln-TM carbonyl, Cp 2Yb(THF){(μ-
CO)Co(CO)3}, is shown in Figure 1.6. This is a molecule that contains a CO bridge
between the Yb and Co atoms. Several other Ln-TM isocarbonyl compounds were
reported, most of them possessing the discrete molecular structure21.
Several Ln-TM carbonyl polymers containing cobalt centers linked to Yb or Eu
were prepared recently in the Shore laboratory8. They were obtained via a so-called
condensation reaction. The starting compound [Yb(L)n][Co(CO)4]2 (which is of solvent-
separated ion pairs type) was stirred overnight with toluene or diethyl ether. Upon
- subsequent work-up, the polymeric compounds that contained the [(μ-CO)nCo(CO)4-n]
2- anions (structure C) or the previously unknown [Co4(CO)11] clusters (structure D) were
isolated (Figure 1.7). The starting material, in turn, was prepared via a transmetalation
8e reaction between Ln metal and Hg[Co(CO)4]2 (Scheme 1.3 ). It was discovered later that
the compound D could be prepared more straightforwardly by conducting the
transmetallation reaction in Et2O. Both compounds C and D contain carbonyl bridges that bind cobalt atoms (or clusters) to the lanthanide cation.
19
Tol [(L)xYb{(μ-CO)yCo(CO)4-y}2*zTol]4 C: L = Pyr, x = 4, y = 2, z = 0 Pyr, THF, L = THF, x = 2, y = 3, z = 1 DME [Ln(L)x] [Co(CO)4]2
Ln = Yb, L = Pyr, x = 6 Ln = Yb, L = THF, x = 6 Et2O Ln = Yb, L = DME, x = 4 {(L) Ln[Co (CO) ]} Ln = Eu, L = THF x 4 11 4 D: Ln = Yb, L = Et2O, x = 2, Ln + Hg[Co(CO)4]2 L = THF, x = 1 Ln = Eu, L = THF, x = 5
{(Et2O)3Ln[Co4(CO)11]}4 Et2O Ln = Yb, Eu
- Scheme 1.3. Summary of condensation reactions for the [Co(CO)4] anion.
20
1.3.4 Methods of PreparationD D
Figure 1.7. Complexes {(pyr)4Yb(μ-CO)2Co(CO)2}∞ (C) and
{(Et2O)2(THF)Yb[Co4(CO)11]}∞ (D) obtained via condensation of the
[Yb(L)n][Co(CO)4]2 parent complex in toluene and diethyl ether, respectively. Ligands coordinated to the Yb(II) cation are omitted for clarity.
Most commonly employed preparation techniques for the Ln-TM heterometallics
include transmetallation6a, 23, metathesis6c, 18a-c, 24, M-M bond cleavage (reduction of the
metal carbonyls in liquid ammonia3a,c, 17 or over amalgam8, 22), adduct formation6b,c, 25, M-
X bond cleavage6d, 26, and condensation of solvent-separated ion pairs into extended
arrays8b, c. Transmetalation, reduction of metal carbonyl dimer over amalgam and
condensation of solvent-separated ion pairs were used widely in this work, and will be
described in more detail below. 21
1.3.4.1 Transmetalation reaction
Transmetalation (metal exchange reaction) has been particularly valuable for
preparation of solvent-separated ion pair type of compounds8a, 24a, 22, 27. The starting
material in this reaction is mercury – TM carbonyl compound of general formula m2Hg,
where m is a TM carbonyl moiety. This reaction is also suitable for transferring organic
alkyl or aryl groups from mercury to another metal, which has to be more electropositive
than Hg28. The general reaction scheme is shown in Equation (1.5). Equation (1.6) shows
8a a specific example, the preparation of [Yb(THF)6][Co(CO)4]2 via this route of synthesis.
m2Hg + M m2M + Hg (1.5)
THF Hg[Co(CO) ] + Yb [Yb(THF) ][Co(CO) ] + Hg (1.6) 4 2 6 4 2
The advantages of this procedure is mild reaction conditions (ambient
temperature, 1 atm pressure), a by-product that is easily removed by filtration (metallic
mercury), and the reaction’s remarkable versatility. A great number of m2Hg compounds
can be prepared, and a vast range of m2Hg and M sets can be combined together, leading to the number of possible products that is virtually limited by combinatorics only. At the
same time, in a given reaction there is only one product as a rule (apart from metallic 22
Hg), and the yield is often quantitative. The only significant disadvantage of this
procedure is the fact that mercury compounds are toxic, and the resulting mercury metal
needs to be carefully collected and recycled.
The starting material m2Hg (where m = TM carbonyl moiety) is commonly
prepared from a Hg(II) salt, such as Hg(CN)2 or HgCl2, and an alkali salt of the TM carbonyl anion in a solvent that dissolves both reactants (Equation (1.7)). These compounds are usually moderately soluble in organic solvents and very poorly soluble in water, which is utilized for their isolation29. Most of them are fairly air stable, although
some undergo oxidation by atmospheric oxygen when exposed to light29. Consequently,
isolation of such compounds has to be carried out either on the bench in the dark, or
under an inert atmosphere. The final m2Hg product has no solvent molecules co-
crystallized, which is definitely to the researcher’s advantage. The molecular structure of
Hg[Co(CO)4]2 is shown in Figure 1.6 as a representative example of a complex of this
type.
THF HgCl2 + 2NaCo(CO)4 Hg[Co(CO)4]2 (1.7) -2NaCl
23
Figure 1.6 goes here
Figure 1.6. Molecular structure of Hg[Co(CO)4]2
1.3.4.2 Reduction of Transition Metal Carbonyls over Amalgam
In this reaction, a metal – metal bond within a TM carbonyl is cleaved, and a TM carbonyl anion is produced. The route is often employed to prepare alkali salts of these anions that are used in turn as starting materials in the preparation of m2Hg derivatives
(see Section 1.3.4.1). Preparation of Na[CpCr(CO)3] from [CpCr(CO)3]2 is shown in
Equation (1.8) as an example. The conversion is quantitative, and the resulting solution of a TM carbonyl anion can be used directly in other reactions, or the solvent can be evaporated and the solid salt obtained instead.
24
THF 2Na/Hg + [CpCr(CO) ] 2[(THF) Na][Cr(CO) Cp] (1.8) 3 2 x 3
This route has also been used for producing lanthanide(II) – TM carbonyl
complexes8, 22, most of them having solvent-separated ion pair structures. Sm, Eu and Yb
were traditionally considered the only three elements among the lanthanide family that
demonstrate a stable oxidation state +2 30, but it was reported recently that many other
lanthanides can be in +2 oxidation state under argon atmosphere31. Their redox potentials are not much different from those of Na or K30, 32. These metals can be dissolved in
mercury, and such amalgams will reduce many TM carbonyls (Equation (1.9)). The
yields are apparently quantitative, and the product is pure enough for X-ray quality
crystals to be grown after filtration from the reaction solution.
THF Yb/Hg + Mn2(CO)10 [Yb(THF)6]Mn(CO)5]2 (1.9)
It should be noted though that the reducing ability of a lanthanide amalgam is
inferior to that of an alkali metal amalgam, because of the differences in the standard
reduction potentials of the metals. Thus, in the author’s experience, the ytterbium 25
- amalgam readily reduces the Mn2(CO)10 to the [Mn(CO)5] , but the Re2(CO)10 does not react under the same conditions.
The advantage of this procedure is that it offers a one-step synthesis of the desired
product, which solution can be used directly in subsequent preparations, because the
conversion is clean and nearly quantitative. The disadvantage is the necessity to handle
large amounts of metallic mercury and mercury amalgam, and to dispose of them in a proper, environmentally-friendly manner.
1.3.4.3 Condensation of Solvent-Separated Ion Pairs into Extended Arrays
This is a relatively novel preparative technique. It was introduced and elaborated
by Shore and co-workers8b, c. They used the solvent-separated ion pairs - type compounds
[Ln(L)n][Co(CO)4]2 as starting material. Refluxing these in non-coordinating solvents
toluene or diethyl ether prompted conversion of these salt-like compounds into extended
polymeric structures (see Scheme 1.2 and Figure 1.7 in Section 1.3.3). These reactions
illustrate the important role played by the solvent in formation of solid structure of Ln –
TM carbonyls (see Section 1.1). The original material ([Ln(L)n][Co(CO)4]2) contained
THF or pyridine, strongly coordinating solvents (strong Lewis bases) surrounding the
Yb(II) cation. When these are removed, and much weaker bases Et2O or toluene are
introduced instead, the latter’s coordinating ability is insufficient to prevent the carbonyl
- ligands of the [Co(CO)4] anion from coordinating to the Yb(II) core. Thus 26
bridging isocarbonyl linkages are formed, and the initial solvent – separated ion pairs are condensed into more ‘compact’ structure, with higher metal content per unit of mass.
It is also noteworthy that in certain instances this process of condensation not only involves elimination of extra ligands but a redox reaction takes place also (Scheme 1.4)8b.
- 2- The [Co(CO)4] anion is converted into the [Co4(CO)11] cluster, and the cobalt atom is
oxidized from formal -1 to formal -½ charge. The Ln(II) cation is reduced to Ln(0), and
precipitates as a metallic powder. It is clear that solvent affects the redox chemistry of the
charged species in the solution. The extent of this influence has not yet been fully
investigated.
Et2O
2[Yb(T HF)6][Co(CO)4]2 {(Et2O)2(THF)Yb[Co4(CO)11]}8 + 5CO + Yb + 11THF
- 2- - 4[Co(CO)4] [Co4(CO)11] + 2e + 5CO
2+ - 0 Yb + 2e Yb
- - Scheme 1.4. Condensation of the [Co(CO)4] anion into the [Co4(CO)11] cluster anion
accompanied by reduction of the Yb2+ cation to ytterbium metal.
27
1.4 Applications
Most application studies have been performed on Ln – TM isocyanide
compounds. As it was mentioned in Section 1.2, compounds of this class are proven
precursors for heterobimetallic catalysts, that are superior to those containing a transition
metal only5d, e. Isocyanide compounds have drawn a lot of attention recently for their
unusual magnetic properties12, 27, and appear to be promising for applications in material science and technology13-15.
Since the Ln – TM carbonyls are closely related to the Ln – TM cyanides, it is
plausible to think that the carbonyl family can function in similar capacity, and offer a
comparable range of potential applications. Few application studies have been performed
on this class of compounds so far. Some researchers suggested that these systems may
function as homogeneous catalysts for Fischer-Tropsch processes23c. A report was also
33 published by Beletskaya on catalytic activity of Cp2Yb{(μ-CO)Co(CO)3} in the hydroformylation of 1-octene. It is possible that the Ln – TM carbonyls could be employed as sources for the preparation of perovskite-type oxides LnMO3 that are used as methane oxidation catalysts34.
It is also worth noting that synthesis and characterization of such new types of
heterometallic systems have made contributions to the realm of fundamental science,
improving our knowledge of how lanthanides and transition metals can be arranged
together in the presence of various ligands. New ways to synthesize compounds that 28
hold metals of very different kinds in close proximity were developed. Research of this kind ultimately paves the way to rational preparation of materials with pre-determined, desired properties.
1.5 Statement of the Problem
The goal of this research is synthesis and characterization of new Yb(II) – TM carbonyl compounds. Carbonyls of transition metals belonging to two groups of the periodic table will be surveyed, those of groups 7 and 6 (Mn, Re; Cr. Mo, W). The reasons for our interest in this research are the following:
1. The chemistry of divalent lanthanide – TM carbonyl systems still lacks
exploration. Numerous reports on their trivalent analogues have been
published, but our knowledge about the Ln(II) derivatives still remains
limited. This work aims at systematic synthesis of the Ln(II) – TM
carbonyl compounds and their structural characterization by means of X-
ray diffractometry. The IR spectroscopy will be routinely employed in
order to elucidate interactions of different species present in the solutions
of the compounds studied.
2. We are interested to learn more about condensation of solvent-separated
ion pairs into extended structures. This reaction was discovered only
recently8b, and it is not clear yet how large the scope of 29
this preparation method is. We need to know whether this is a general
phenomenon that can occur to any compound of the solvent-separated ion
pair type, or to a certain range of them only. If the latter case is true, the
question of limitations of this method needs to be answered. Elucidation of
the mechanism of such condensation would be of great value.
3. We also want to probe the reactivity of carbonyls of transition metals of
group 6 of the periodic table with Yb(II) species. The distinguishing
feature of these carbonyls is that they are mononuclear compounds of
general formula M(CO)6. Most research in the field of Ln(II) – TM
carbonyl chemistry has been done using TM carbonyl dimers, trimers or
clusters of higher nuclearity. The M-M bond of these compounds is
m- cleaved upon reduction, thus affording the [M(CO)n] anions. This
opportunity is obviously absent in case of the M(CO)6 structures. It is well
known that such carbonyls lose one of their CO ligands readily upon
35 heating or irradiation with UV radiation . The resulting M(CO)5 fragment
is highly reactive due to its electron-deficient nature. Formation of the
2- 36 [M(CO)5] anion is documented in the literature , and we believe this can
afford new possibilities for creating Yb(II) – TM carbonyl systems.
30
CHAPTER 2
Yb(II) – GROUP 7 TRANSITION METAL CARBONYL COMPLEXES
2.1 Starting Materials
Decacarbonyl dimers of the group 6 transition metals Mn2(CO)10 and Re2(CO)10
- were employed as sources of the [M(CO)5] anion. Both compounds are commercially
available in good purity (98% or higher), and can be further purified easily by
sublimation on boiling water bath. These carbonyls show three strong absorptions in
-1 -1 solution IR at 2045, 2008, 1980 cm (Mn2(CO)10, in THF) and 2070, 2009, 1966 cm
(Re2(CO)10, in THF) (the middle band is the strongest in both cases). Reduction with an
- alkali metal amalgam readily affords the [M(CO)5] anion in a matter of hours (Equation
(2.1)). The IR spectra of those consist of three bands situated at 1896, 1862, 1830 cm-1
THF 2Na/Hg + Mn2(CO)10 2NaMn(CO)5 (2.1)
31
- -1 - 37 ([Mn(CO)5] in THF); 1911, 1864, 1835 cm ([Re(CO)5] in THF) (see also Figures B.3
and B.12 in Appendix B). Both species possess D3h symmetry (trigonal bipyramid) in the
solid state, and group theory predicts only two IR-active bands in the carbonyl region.
Explanation for this discrepancy can be found in the formation of contact ion pairs in
solution. This will be discussed in more detail later (see Section 2.3). For now suffice it
to say that in a stronger coordinating solvent pyridine we do observe just two bands at
-1 -1 1897, 1860 cm ([Na(pyr)x][Mn(CO)5]) and 1913, 1860 cm ([Na(pyr)x][Re(CO)5]). This is because in these solvent-separated ion pair compounds pyridine binds strongly to the
+ - Na cation, and the [M(CO)5] anion do not interact with it, thus preserving its
undisturbed trigonal bipyramidal shape.
THF 2NaM(CO)5 + HgCl2 Hg[M(CO)5]2 (2.2) -2NaCl M = Mn, Re
The alkali salts of these pentacarbonylmetal anions were further converted into
the respective mercury derivatives Hg[M(CO)5]2. The conversion is a simple metathesis reaction (Equation (2.2))29, 38. As it was mentioned before (Section 1.3.4.1), the mercury
– TM carbonyls are convenient starting point for carrying out the transmetalation
reaction. Both Hg[Mn(CO)5]2 and Hg[Re(CO)5]2 can be prepared via reaction (2.2) in
good yields (> 70%), and further purified by sublimation. The IR data: Hg[Mn(CO)5]2,
32
-1 38 2065(s), 1995(s), 1955(vs) cm (in KBr) ; Hg[Re(CO)5]2, 2073(ms), 1974(vs), 1935(m, sh) cm-1(in Nujol mull)29. It should be noted that these compounds are poorly soluble in
almost any solvent at room temperature, thus IR data are reported only for the solids.
39 The X-ray structure of the Hg[Mn(CO)5]2 has been reported in the literature
(Figure 2.1). It has D4h symmetry, consistent with its IR spectrum (three bands expected).
The Hg-Mn bond length is 2.614(1) Å; the axial Mn-C distance is 1.829(6) Å, and the average equatorial Mn-C distance is 1.843[6] Å. The average Hg-Mn-C angles are
84.9[2]º (equatorial CO) and 179.0(2)º (axial CO).
Figure 2.1. Molecular structure of Hg[Mn(CO)5]2
33
We could find no X-ray data for the Hg[Re(CO)5]2 in the literature. Despite the
fact that the synthesis of this complex was published in 197129, no X-ray structure determination has been carried out. We might speculate that the structure is essentially analogous to that of the Hg[Mn(CO)5]2. Atomic radii of the Mn and Re atoms are very
close (1.40 and 1.35 Å, respectively40), so there should be no significant difference in the
Hg-M and the M-C bond lengths. Some differences could arise because of the different
electronegativities of the two elements (Mn, 1.55; Re, 1.9 Pauling scale)41.
2.2 Transmetalation reactions between Yb and Hg[M(CO)5]2 (M=Mn, Re).
Formation of the Solvent-Separated Ion Pairs.
Direct reactions between Yb metal and Hg[M(CO)5]2 in various polar solvents
produced solvent-separated ion pair species [Yb(L)n][M(CO)5]2 (M = Mn, 1; Re, 2;
Equation (2.3)). Using this approach, similar results have been obtained for divalent
- 8, 18d - lanthanides with [Co(CO)4] . Despite a vast difference in Lewis basicity ([Mn(CO)5]
- 4 - , 77; [Re(CO)5] , 2.5⋅10 ; compared to the nucleophilicity of [Co(CO)4] , which is set
arbitrarily as 1)42 and the softer Lewis acidic nature of Yb2+ compared to Yb3+, the products are still ion paired compounds with no isocarbonyl linkages formed. Anions
- [M(CO)5] (M = Mn, Re), were unable to penetrate the coordination sphere of solvent
around the Yb2+ cation to replace THF molecules in their role as Lewis base. It has been
9 - shown that the Lewis basicities of [Co(CO)4] and THF are similar; i.e. the basicity of
THF should only slightly exceed 1 and according to the results of Dessy et al42 it 34
L (2.3) Hg[M(CO)5]2 + Yb [Yb(L)n][M(CO)5]2 + Hg
1, M = Mn, L = THF, n = 6
2, M = Re, L = THF, n = 6
1a, M = Mn, L = DME, n not established, presumably 48
1b, M = Mn, L = DMF, n not established
1c, M = Mn, L = pyridine, n not established, presumably 68
is inferior to values for anions studied herein. Our results suggest that an approach in
which the electrophilic center (Yb2+ in this case) binds to the strongest Lewis base available8, 9, is not sufficient to predict what ligands will coordinate to the Yb2+ cation
first.
Compounds with similar structure were synthesized by reduction of metal
carbonyl dimers with lanthanide amalgam22. Using methods described herein, we carried
out reduction of Mn2(CO)10 (but not that of Re2(CO)10) with ytterbium amalgam in THF,
and found that product from this reaction is identical to 1 [Yb(THF)6][Mn(CO)5]2
(Equation (2.4)). Having established that, we routinely employed this procedure to synthesize 1.
35
THF Yb/Hg + Mn2(CO)10 [Yb(THF)6][Mn(CO)5]2 (2.4)
Syntheses analogous to (2.3) were also carried out in diethyl ether and toluene.
Earlier, a similar approach employing Hg[Co(CO)4]2 yielded the previously unknown
8b 2- cluster [Co4(CO)11] . In the present study, in both solvents some reaction occurred as evidenced by a solution color change, and precipitation of a homogeneous mixture of product and a finely divided black powder. Unfortunately, the precipitates were only sparingly soluble, precluding the possibility of separating the products from the black material. Metallic mercury also precipitated along with the products, to produce a homogeneous mixture. For these reasons, no solid state IR or elemental analyses were attempted. Upon removing the original solvent and dissolution of the precipitates in THF,
IR spectra were like those of 1 and 2, [Yb(THF)6][Re(CO)5]2. It is reasonable to assume
some polymeric structures were formed in toluene and ether, and cleaved later when THF
was added to them.
36
2.3 Solution IR Spectra of [Yb(THF)6][Mn(CO)5] 1, [Yb(THF)6][Re(CO)5] 2,
[Yb(DME)n][Mn(CO)5]2 (1a), [Yb(DMF)n][Mn(CO)5]2 (1b), [Yb(pyr)6][Mn(CO)5]2
(1c)
IR spectra of compounds 1 and 1a (Table 2.1, see Figures B.4 and B.5 in
Appendix B) reveal features typical for solvent separated ion pair species. Thorough discussion of how such characteristic patterns arise can be found in the literature8, 9. In
each spectrum, three groups of peaks can be clearly seen. The central group (in square
parentheses in Table 2.1) strongly resembles the spectrum of the corresponding pure43
- [M(CO)5] anion in pyridine. These absorbances are assigned to the isolated transition metal carbonylate anion as it exists in solution of ligand. For 1 this group is shifted to
higher frequencies by about 5 cm-1 compared to the ‘pure’ anion, suggesting that some
extra electron density is drained from the anion, probably through weak
7a, 8 (S)nYb···OCM(CO)4 interaction . For 1a, position of this group coincides with the
- [Mn(CO)5] spectrum, but an extra band appears on the left, which is discussed later.
Spectra of 1b and 1c coincide entirely with the ‘pure’ anion spectrum, with no extra
peaks emerging. This is ascribed to the very high Lewis basicities of pyridine and DMF,
which enables them to bind to the Yb2+ cation strongly to completely preclude electron
density transfer from the anion to the
37
Table 2.1. Infrared Data for Compounds 1 - 4 in the Carbonyl Stretching Frequency
Region.
-1 compound medium υCO, cm
Mn2(CO)10 THF or Et2O 2045(m, sh), 2008(s, sh), 1980(m, sh)
Re2(CO)10 THF or Et2O 2070(m, sh), 2009(s, sh), 1966(m, sh)
KMn(CO)5 THF 2012(vw), 1899(s), 1865(s), 1835(m);
1896(s), 1862(s), 1830(m)a
Pyr 1978(vw), 1897(s), 1861(s)
NaRe(CO)5 THF 2010(vw), 1968(w), 1910(s), 1863(s), 1830(m)
1911(s), 1864(s), 1835(sh)a
Pyr 2009(w), 1968(mw), 1913(s), 1860(vs)
b Hg[Mn(CO)5]2 KBr 2065(s), 1995(s, sh), 1955(vs), 645(m)
2067(s), 2008(sh), 1975(vs)c
d Hg[Re(CO)5]2 Nujol 2073(ms), 2014(w, sh), 1974(vs), 1935(m, sh)
d CH2Cl2 2075(ms), 2024(ms), 1995(vs)
1 THF 2011(m to s), [1918(s), 1892(s), 1865(m)], 1760(ms)
2 THF 2043(w), 2011(s), 1972(m), 1935(s), 1890(m), 1754(m)
1a DME 2012(m), [1897(s), 1864(s)]
1b DMF 2014(vw), [1898(s), 1865(s)]
1c Pyr 1978(vw), [1897(s), 1861(s)]
Et2O 2015(m), [1929(s), 1903(s)], 1724(ms)
Continued
38
Table 2.1 continued
3 Et2O 2015(s, sh), 1931(mw), 1908(mw), 1707(mw)
KBr 1893(s)
4 Et2O 2046(vw), 2015(s, sh)
KBr 1997, 1819(s, br)
aJ. E. Ellis, E. A. Flom, J. Organomet. Chem., 1975, 99, 263. bJ.M. Burlitch, Chem.
Commun., 1968, 887. cW. Hieber, W. Schropp, jun., Chem. Ber., 1960, 93, 455.
dA.T.T. Hsieh, M.J. Mays, J. Chem. Soc. (A), 1971, 2648
cation. Another two groups of peaks arise on the right and left sides of the central group.
Their appearance can be accounted for in terms of formation of contact ion pairs8, 9. In
solution, the Yb2+ cation is coated with a layer of ligand molecules (THF in this case).
THF is coordinated to the cation through its oxygen atom due to the oxophilic nature of
- the lanthanide species. However, the [M(CO)5] anion contains oxygen atoms also, and
those can compete with the THF molecules for coordination sites around the Yb2+ cation.
When this happens, the resulting weakly held together complex (S)nYb···OCM(CO)4 is
called a contact ion pair. In such a complex there is at least one bridging carbonyl
(depending on how many cations the anion is coordinated to), and a few terminal ones.
As it was explained earlier (Section 1.3.1), the bridging carbonyl ligands absorb in 39
the lower frequency region (compared to the unperturbed carbonylate anion), and the terminal ones absorb in higher wavenumbers. Thus the presence of the contact ion pairs,
- along with some amount of the [M(CO)5] anions with undisturbed D3h geometry, gives rise to the characteristic three-band IR pattern in solution.
The IR spectrum of 2 (Table 2.1) is more complicated than that of 1, but can be
analyzed along the same lines. It seems that the symmetry of the species present in the
- solution is lower than the D3h of the [Mn(CO)5] , or that there is a mixture of various
species with different geometries. The latter suggestion is plausible in the light that in the
- solid structure, there seem to be two different types of the [Re(CO)5] anion: one
possessing the D3h geometry, and another one of the C4v symmetry (Section 2.4). Three
active IR bands are predicted for compounds of the latter symmetry. Obviously, the
presence of such mixture in the solution makes spectral analysis rather difficult. Both the
- D3h and the C4v type [Re(CO)5] anions can form contact ion pairs, adding another level
- of complication. Since the [Re(CO)5] anion is a much stronger Lewis base than the
- - - 4 42 2+ [Mn(CO)5] anion ([Mn(CO)5] , 77; [Re(CO)5] , 2.5⋅10 ) , the Yb cation should
coordinate more tightly to its carbonyl ligands, causing greater splitting between the
terminal and the bridging sets of ligands in terms of their IR peak positions. Finally, it is
unclear whether the two different geometries are present in solution, or is it just an effect
of packing, and thus only takes place in solid structure.
The difference in spectra of 1c in pyridine and in Et2O (Table 2.1) is of interest.
The compound was synthesized in pyridine, and its spectrum coincides completely 40
with that of Na[Mn(CO)5] in the same solvent. However, when pyridine is removed and
Et2O added, the IR spectrum is very much like the spectra of 1, 1a and 2, with three
distinct groups of peaks, indicating that contact ion pairs are present in solution. Both the pyridine solution and the Et2O solution are dark brown, from which we conclude that
even in ether solution some pyridine molecules are retained around the Yb2+ cation, but
as some exchange occurs the solvation layer is no longer impermeable, and interaction
between ion pairs takes place.
2.4 X-ray structures of 1 and 2
X-ray data were collected at –73°C. Structures of the anions of these salts are
2+ represented in Figures 2.2 and 2.3. As the [Yb(THF)6] cation is essentially the same for
both complexes, its structure is shown only once (Fig. 2.5a). Crystallographic data and selected bond distances and bond angles are given in Tables 2.2 and 2.3. The
2+ 8a [Yb(THF)6] cation closely resembles its counterpart in the tetracarbonylcobaltate salt .
In both structures, two independent ytterbium atoms reside on special positions, each with three independent THF ligands; additional THF ligands are generated by symmetry transformations, resulting in nearly octahedral coordination of the Yb(II) centers. In 1, the average Yb-O bond lengths are 2.389(3) and 2.391(6) Å, in 2 these are 2.387(6) and
2.391(9) Å. Lengths of Yb-O bonds are 2.393(7), 2.394(7), and 2.370(7) Å. In both structures there is disorder indicated in the carbon atoms of the THF ligands. This cation
41
2+ Figure 2.2. Molecular structure of 1 (a) the [Yb(THF)6] cation (25% thermal
- ellipsoids), (b) the [Mn(CO)5] anion (non-disordered) (15% thermal ellipsoids). Yb1 lies on a center of symmetry.
42
- Figure 2.3. The [Re(CO)5] anions in 2 (25% thermal ellipsoids) (a) anion without disorder, (b) disordered anion.
43
Table 2.2 Crystallographic Data for [Yb(THF)6][M(CO)5]2 (1, M=Mn; 2, M=Re),
{Yb(THF)2(Et2O)2(Mn(CO)5)2}∞ (3) and {Yb(THF)4(Mn(CO)5)2}∞ (4)
1 2 3 4
empirical C34H48O16Mn2Yb C34H48O16Re2Yb C26H36O14Mn2 C26H32Mn2O1
formula Yb 4Yb
formula 995.64 1258.16 855.47 851.44
weight
space group P-1 P-1 C2/c C2/c
a, Å 10.8799(10) 10.0421(10) 16.7913(10) 11.3065(10)
b, Å 11.1321(10) 10.7667(10) 10.3795(10) 18.9475(10)
c, Å 19.5728(10) 19.7414(10) 19.6009(10) 15.6224(10)
α, deg 78.261(10) 94.585(10) 90 90
β, deg 89.715(10) 92.613(10) 93.357(10) 95.881(10)
γ, deg 68.699(10) 92.898(10) 90 90
V, Å3 2156.3(3) 2122.3(3) 3410.2(4) 3329.2(4)
Z 2 2 4 4
- Dcalcd, g⋅cm 1.533 1.969 1.666 1.699
3
T, °C -73 -73 -73 -73
μ, mm-1 2.790 7.939 3.509 3.594
a R1[I>2σ(I)] 0.0375 0.0517 0.0443 0.0362
44 Continued
Table 2.2 (continued) wR2 (all 0.1053 0.1400 0.1052 0.0870 data)b
GOF on F2 1.029 1.076 1.020 1.058
45
Table 2.3 Selected Bond Distances (Å) and Bond Angles (deg) for
- [Yb(THF)6][M(CO)5]2 (1, M=Mn; 2, M=Re; data for the non-disordered [M(CO)5] anions are given)
Bonds 1 2 Angles 1 2
Yb(1)-O(11) 2.400(4) 2.392(7) O(12)-Yb(1)-O(13) 89.46(13) 88.3(3)
Yb(1)-O(12) 2.385(4) 2.390(7) O(11)-Yb(1)-O(12) 89.76(15) 89.8(3)
Yb(1)-O(13) 2.392(4) 2.384(7) O(11)-Yb(1)-O(13) 89.29(14) 89.7(3)
Yb(2)-O(14) 2.388(4) 2.405(8) C(1)-M(1)-C(3) 178.5(3) 175.6(6)
Yb(2)-O(15) 2.390(4) 2.383(7) C(2)-M(1)-C(4) 116.9(4) 118.1(7)
Yb(2)-O(16) 2.394(4) 2.388(6) C(2)-M(1)-C(5) 121.2(4) 116.8(5)
M(1)-C(1) 1.829(8) 1.971(13) C(4)-M(1)-C(5) 121.9(4) 124.9(6)
M(1)-C(2) 1.775(9) 1.932(14) C(3)-M(1)-C(2) 90.0(4) 94.7(5)
M(1)-C(3) 1.797(8) 1.966(14) C(3)-M(1)-C(4) 90.5(3) 92.1(6)
M(1)-C(4) 1.778(8) 1.935(12) C(3)-M(1)-C(5) 89.8(3) 88.1(5)
M(1)-C(5) 1.796(9) 1.941(12) C(1)-O(1) 1.143(8) 1.141(14)
C(2)-O(2) 1.160(9) 1.154(14) C(3)-O(3) 1.153(8) 1.157(15)
C(4)-O(4) 1.163(8) 1.175(15) C(5)-O(5) 1.154(8) 1.138(13)
46
as reported elsewhere8a has Yb-O distances of 2.282(2), 2.387(2) and 2.396(3) Å. A
44a shorter distance (2.298 Å) was reported for [Yb(THF)6][(nido-7,8-C2B9B H11)2] . Other
Yb(II) compounds with coordinated THF ligands and their Yb-O bond lengths5 include
44b 44c [(MeOCH2CH2C5H4)2Yb(THF)] (2.496(4) Å) , Cp*2Yb(THF) (2.412(5) Å) ,
44d tBu,Me 5 (MeC5H4)2Yb(THF) (2.53(2) Å) , and [{Yb(Tp )(THF)(μ-CO)2Mo(η -
21f C5H4Me)(CO)3}2] (2.510(3) Å) . For comparison, the Yb-O distance in
[Yb(DIME)3][Co(CO)4]2 (Yb 9-coordinated, tricapped trigonal arrangement) are from
2.49(1) to 2.69(2) Å22.
The manganese pentacarbonyl anion of 1 is similar to the same anion in other
salts45. The structure contains two crystallographically independent anions, one of which is ordered and one disordered. In ordered Mn(CO)5, the average Mn-C bond length is
- 1.793(19) Å, average C-O bond length is 1.156(5) Å. The disordered [Mn(CO)5] (Figure
2.4) is divided into part 1 (0.57 occupancy factor) and part 2 (0.43 occupancy factor) contributions to the X-ray scattering with SHELXTL-97. For part 1, the average Mn-C bond length is 1.73(6) Å, average C-O bond length is 1.16(4) Å; for part 2, these distances are 1.80(5) Å and 1.14(2) Å, respectively (Figure 2.4, Table 2.4).
47
- Table 2.4 Bond Distances (Å) and Bond Angles (deg) for the Disordered [Mn(CO)5]
Anion in [Yb(THF)6][Mn(CO)5]2 (1)
Bond Distances, Å Bond Angles, deg
Mn2-C6 1.750(8) C9-Mn2-C10 173.8(11)
Mn2-C7 1.831(8) C6-Mn2-C9 88.2(9)
Mn2-C8 1.793(16) C7-Mn2-C9 94.6(9)
Mn2-C9 1.803(17) C8-Mn2-C9 96.2(11)
Mn2-C10 1.703(16) C6-Mn2-C7 119.7(4)
C6-O6 1.116(9) C6-Mn2-C8 114.2(8)
C7-O7 1.160(8) C7-Mn2-C8 125.2(8)
C8-O8 1.135(15) C9’-Mn2’-C10’ 176.3(9)
C9-O9 1.144(17) C6-Mn2’-C9’ 89.3(7)
C10-O10 1.205(15) C7-Mn2’-C9’ 89.6(7)
Mn2’-C6 1.829(8) C8’-Mn2’-C9’ 87.6(8)
Mn2’-C7 1.717(8) C6-Mn2’-C7 121.7(5)
Mn2’-C8’ 1.825(13) C6-Mn2’-C8’ 120.8(6)
Mn2’-C9’ 1.730(15) C7-Mn2’-C8’ 117.3(5)
Mn2’-C10’ 1.776(14)
C8’-O8’ 1.131(12)
C9’-O9’ 1.171(15)
C10’-O10’ 1.161(14)
48
Figure 2.4. Molecular structure (10% probability thermal ellipsoids) showing the
- disordered [Mn(CO)5] anion in 1. The site occupancy factors are 0.53 for Mn2, C8, O8,
C9, O9, C10, O10; 0.47 for Mn2’, C8’, O8’, C9’, O9’, C10’, O10’.
49
We could find no record in the Cambridge X-ray Structure Database describing
- the [Re(CO)5] anion. It is reasonable to assume that this anion would be analogous to
- - [Mn(CO)5] . In the structure of 2, there are also two [Re(CO)5] anions one of them
- possessing the trigonal bipyramidal geometry analogously to that of [Mn(CO)5] (Figure
2.3a), and the other distorted and disordered (Figure. 2.3b). The trigonal bipyramidal rhenium pentacarbonyl anion (Fig. 2.3a) closely resembles its manganese analog. The average Re-C distance is 1.94[3] Å, C-O 1.16[1] Å. All angles are close to the values for a trigonal bipyramid, and the axial bond distances appear to be longer than the equatorial ones (Table 2.3).
- The disordered site contains two [Re(CO)5] anions of a trigonal bipyramidal
structure, with occupancies 0.56 and 0.44 (Figure 2.3b). The larger occupancy structure
has angles C8-Re2-C6, C6-Re2-C10 of 160.2(12) º and 113.4(9)º, respectively, and the
average Re-C bond length 1.937[19] Å (Tables 2.5a and 2.5b). In the smaller occupancy
structure, the angles C8-Re2-C6’, C6’-Re2-C10 are 142.9(9) º and 113.4(9) º, and the
average Re-C distance is 1.938[19] Å. Normally, transition metal pentacarbonyls possess
2- trigonal bipyramidal coordination, as observed for isoelectronic [M(CO)5] (M=Cr, Mo,
46 - 47 W) , [Mn(CO)5] and Fe(CO)5 . Angles between base ligands vary from 81.0(10) to
89.1(10) degrees, and those between axial and base ligands from 95.3(12) to 113.7(11)
degrees. To the best of our knowledge, there is only one previously known example of a
46e – pentacarbonyl transition metal anion with square pyramidal geometry , [Mn(CO)5] in the complex [Ph4P][Mn(CO)5]. It is unknown what causes this change from trigonal
bipyramidal to square pyramidal geometry as well as why it happens to that anion. 50
Table 2.5. Bond Distances (Å) and Bond Angles (deg) for the Major Portion of the
Disordered Pentacarbonylrhenium Anion in [Yb(THF)6][Re(CO)5]2 (2)
Bond Lengths Bond Angles
Re(2)-C(7) 1.936(16) C(7)-Re(2)-C(10) 114.0(11)
Re(2)-C(6) 1.919(16) C(6)-Re(2)-C(10) 95.5(12)
Re(2)-C(8) 1.971(19) C(8)-Re(2)-C(10) 103.5(7)
Re(2)-C(9) 1.945(15) C(9)-Re(2)-C(10) 109.9(11)
Re(2)-C(10) 1.921(14) C(7)-Re(2)-C(6) 96.4(14)
C(7)-O(7) 1.151(17) C(8)-Re(2)-C(6) 160.2(12)
C(6)-O(6) 1.157(17) C(7)-Re(2)-C(9) 135.3(15)
C(8)-O(8) 1.157(18) C(8)-Re(2)-C(9) 81.3(10)
C(9)-O(9) 1.172(16) C(7)-Re(2)-C(8) 80.8(10)
C(10)-O(10) 1.148(15) C(6)-Re(2)-C(9) 87.3(13)
51
Table 2.6. Bond Distances (Å) and Bond Angles (deg) for the Minor Portion of the
Disordered Pentacarbonylrhenium Anion in [Yb(THF)6][Re(CO)5]2 (2)
Bond Distances Bond Angles
Re2-C7’ 1.934(14) C9’-Re2-C7’ 170.3(11)
Re2-C6’ 1.945(14) C9’-Re2-C10 93.6(10)
Re2-C9’ 1.915(16) C9’-Re2-C8 93.5(10)
C7’-O7’ 1.160(16) C9’-Re2-C6’ 87.9(10)
C6’-O6’ 1.146(15) C6’-Re2-C8 142.9(9)
C9’-O9’ 1.173(16) C6’-Re2-C10 113.4(9)
C7’-Re2-C8 90.4(8)
C7’-Re2-C10 94.1(9)
52
2.5 Condensation of 1 with Et2O into extended structures
{Yb(THF)2(Et2O)2[(μ-CO)2Mn(CO)3]2}∞ (3) and {Yb(THF)4[(μ-CO)2Mn(CO)3]2}∞
(4)
It has been shown that dissolution of [Ln(S)6][Co(CO)4]2 (S = THF, pyridine) in
diethyl ether and toluene facilitates condensation of solvent-separated ion pairs into
extended linear structures or two-dimensional arrays where the cation and the anion are
linked together by isocarbonyl bridges8. In diethyl ether, novel cobalt clusters that are
still bonded to the lanthanide moiety through bridging isocarbonyls were obtained8b. We
carried out similar reactions on salts 1 and 2. Condensation with ether, but not with
toluene proved to be possible for 1 (Equation (2.5)), which is poorly soluble in Et2O;
However, after stirring overnight a green solution was obtained and a vast quantity of undissolved material remained in the reaction flask. Emerald-green crystals grown from this solution have a sheet structure, as elucidated by single crystal X-ray analysis (Figures
2.5 and 2.6). These crystals had to be grown relatively rapidly, taking no more than two days to evaporate the solvent. When the original solution was allowed to spend some time in the flask, or evaporated more slowly (7-10 days), some black decomposition products precipitated, and the solution gradually changed in color from green to yellow and yellow crystals of 4, {Yb(THF)4[(μ-CO)2Mn(CO)3]2}∞ appeared (Equation (2.6)).
The latter compound is a chain (Figure 2.7).
53
Et2O [Yb(THF)6][Mn(CO)5]2 {(THF)2(Et2O)2Yb[(μ-CO)2Mn(CO)3]2} (2.5) 1-2 days 1 3
Et2O [Yb(THF)6][Mn(CO)5]2 {(THF)4Yb[(μ-CO)2Mn(CO)3]2} (2.6) 7-10 days 1 4
54
Figure 2.5. Molecular structure of 3. (a) asymmetric unit (25% thermal ellipsoids), (b) ring containing eight metal atoms, Yb1 lies on a 2-fold rotation axis.
55
Figure 2.6. Views of a portion of the 2-D polymeric sheet of 3. (a) top view
(along c axic), (b) cross-section (view along b axis). Ytterbium atoms shown in red, manganese in green, carbon in grey and oxygen in blue.
56
Figure 2.7. Molecular structure of 4. (a) asymmetric unit (25% thermal ellipsoids),
(b) a ‘diamond’ (15% thermal ellipsoids), (c) portion of a 1-D linear chain of 4 (only oxygen atoms of THF ligands are shown for clarity). Yb1 lies on a 2-fold rotation axis.
57
The IR spectrum of a freshly prepared ether solution of 1 (Table 2.1, see Figure
B.8 in Appendix B) consists of one strong sharp absorption in the terminal carbonyl region (2015 cm-1) and three medium to weak bands at 1931, 1908, and 1707 cm-1. In 24 hours the peak at 2015 cm-1 remains almost as strong, while the intensity of the others
diminishes significantly. Such patterns suggest that after dissolving 1 in Et2O, the cationic and anionic species form clusters in solution where the degree of isocarbonyl bonding is pronounced and could be close to what is observed in the solid state structure.
Such clusters will have sufficient amount of CO ligands in the terminal position to give rise to a strong absorption at 2015 cm-1. Possibly the four IR bands arise from a situation
- in which there are free [Mn(CO)5] anions in the freshly prepared ether solution of 1. If such is the case, the two bands at 1931 and 1908 cm-1 can be assigned to that species in its undisturbed trigonal bipyramidal shape, and the band at 1707 cm-1 would correspond
to bridging isocarbonyls. With time, all the free manganese pentacarbonyl condenses to
form clusters, and its IR signals disappear. After more time (4-6 days) two new peaks
begin to emerge, at 2046 an 1978 cm-1. This new spectrum corresponds to the spectrum
of 4, crystals of which grew at about the same time. However, as these two peaks become
stronger, a third absorption occurs, and the final spectrum in 10 days consists of three
absorptions at 2046(m, sh), 2014(s, sh), and 1978(m, sh) cm-1, which coincides entirely
with the spectrum of Mn2(CO)10.
Taken all of the aforesaid into account, we propose the following pathways for
formation of 3 and 4. It is reasonable to assume that when complex 1 is added to Et2O, some THF ligands dissociate from the Yb2+ core, leaving coordination sites vacant for 58
- other Lewis bases. Those vacant sites are then filled with oxygen atoms of [Mn(CO)5] and Et2O, thus forming diamond shaped clusters that combine to form crystals of 3 (Fig.
2.5 and 2.6). The clusters contain both THF and Et2O as ligands coordinated to the
cation. However, complex 3 is unstable in Et2O. It slowly converts to complex 4 (Figure
- 2.7), a reaction in which [Mn(CO)5] is oxidized to form Mn2(CO)10 (Scheme 2.1). Since diethyl ether is a much weaker coordinating agent (a weaker Lewis base) than THF, it is plausible to think that the Yb2+ center is more electron-deficient in the
2+ 2+ [Yb(Et2O)2(THF)2] configuration than in [Yb(THF)4] . In other words, stabilizing
influence of the two Et2O ligands does not make this oxidation state of the cation stable
2+ - enough. The Yb cation then draws electrons from the two [Mn(CO)5] moieties,
· 0 oxidizing them to the [Mn(CO)5] radicals, and getting reduced to Yb (Step 1 in Scheme
2.1). The Yb0 atom cannot hold any ligands around itself, so it releases any ligands that
· were coordinated to it, and precipitates as metallic ytterbium. The two [Mn(CO)5] radicals combine together and produce Mn2(CO)10, giving rise to its signature peaks in the solution IR (Step 2 in Scheme 2.1). The THF molecules that were released in Step 1 create excess of THF in ether solution of 1. These THF ligands then coordinate to another
2+ Yb center, displacing Et2O (Step 3 in Scheme 2.1). Thus one half of the initial amount
of ytterbium precipitates as a metal, and another half forms diamond-shaped chains of 4
(Figure 2.7) that contain no Et2O molecules.
59
Scheme 2.1 Proposed mechanism of formation of complexes 3 and 4 out of parent
compound 1.
8b Earlier, we observed that the solvent separated ion pairs [Ln(THF)6][Co(CO)4]
- (Ln = Yb, Eu) when placed in Et2O undergo reactions in which [Co(CO)4] is oxidized to
- form the previously unreported cluster [Co4(CO)11] that is linked to the lanthanide
through isocarbonyl linkages. As in the present case, a black powder is also produced that
is attributed to the formation of the reduced lanthanide. Unfortunately, this finely divided
powder is dispersed though out the polymeric sample and as in the present case could not
be separated from it for positive identification of Ln0.
Another possibility that has been suggested for the redox reaction in which
- Mn2(CO)10 and the black powder are formed from [Mn(CO)5] , is the disproportionation
- of [Mn(CO)5] to Mn2(CO)10 and Mn. However, we believe this possibility to be remote.
60
- We are unaware of [Mn(CO)5] normally undergoing spontaneous disproportionation to
give Mn and Mn2(CO)10 . Furthermore, there is no evidence for the formation of CO. If
- 2+ the [Mn(CO)5] disproportionates when the THF is replaced by Et2O, Yb is an "innocent
bystander", and another anion is required to compensate for the excess Yb2+ when
- [Mn(CO)5] condenses to Mn2(CO)10.
The redox reaction (Equation (2.7)) can be written as decomposition of 3 to 4
(Equation (2.8); compare to Scheme 1.3, Section 1.3.4.3). The fact that it takes more time
to form crystals of 4 suggests that it is a thermodynamic product, while the structure of 3
is a kinetic one.
Et2O Yb2+ + 2[Mn(CO) ]- 0 5 Yb + Mn2(CO)10 (2.7)
Et2O {(THF)2(Et2O)2Yb[(μ-CO)2Mn(CO)3]2} 3
{(THF)4Yb[(μ-CO)2Mn(CO)3]2} + Yb + Mn2(CO)10 + 4Et2O(2.8) 4
61
Attempted analytical confirmation of the stoichiometry of reaction (2.8)
encountered certain difficulties. First, 3 proved to be difficult to redissolve in Et2O, no
doubt due to its polymeric nature. Also, both 4 and the Yb metal precipitate required
separation that led to loss of products and greater experimental errors. In spite of this,
determination of the amount of ytterbium produced in reaction (2.8) was found to
correspond approximately to what is expected from the process’s stoichiometry48. That, along with the fact that solution IR of mother liquor of 4 reveals the presence of
- Mn2(CO)10, confirming the oxidation of [Mn(CO)5] .
2.6 Molecular Structure of {Yb(THF)2(Et2O)2[(μ-CO)2Mn(CO)3]2}∞ (3)
The structure of 3 is represented in Figures 2.5 and 2.6. Crystallographic data are
given in Table 2.2, selected bond distances and bond angles are given in Table 2.7. In the
structure of 3, one Yb atom, two ligands (THF and Et2O, Fig. 2.4a), and one manganese
pentacarbonylate anion form the asymmetric unit. Symmetry transformations of these
unique atoms generate a non-planar polymeric sheet (Fig. 2.6a). The axial ‘poles’
(constructed of the two axial carbonyls and manganese atom) of the two manganese
pentacarbonylate anions attached to the same ytterbium atom are almost exactly
perpendicular to each other. In 3, four ytterbium and four manganese atoms bound by
isocarbonyl linkages form “eight-member” rings (Fig. 2.5b), similarly with {(Pyr)4Yb[(μ-
8b CO)2Co(CO)2]2}∞ . A cross-section of this zigzag puckered sheet (Fig. 2.6b) shows that
Mn(CO)5 units lie above and below the plane containing Yb atoms, as in the 62
Table 2.7 Selected Bond Lengths (Å) and Bond Angles (deg) for {Yb(THF)2(Et2O)2[(μ-
(CO)2Mn(CO)3)2]}∞ (3) and {Yb(THF)4[(μ-(CO)2Mn(CO)3)2]}∞ (4)
3 4
Mn(1)-C(1) 1.836(9) Mn(1)-C(1) 1.823(6)
Mn(1)-C(2) 1.834(9) Mn(1)-C(2) 1.810(7)
Mn(1)-C(3) 1.800(9) Mn(1)-C(3) 1.830(6)
Mn(1)-C(4) 1.759(8) Mn(1)-C(4) 1.773(4)
Mn(1)-C(5) 1.777(8) Mn(1)-C(5) 1.775(4)
C(1)-O(1) 1.117(8) C(1)-O(1) 1.183(15)
C(2)-O(2) 1.136(9) C(2)-O(2) 1.197(15)
C(3)-O(3) 1.164(9) C(3)-O(3) 1.138(7)
C(4)-O(4) 1.179(8) C(4)-O(4) 1.169(5)
C(5)-O(5) 1.165(8) C(5)-O(5) 1.168(5)
Yb(1)-O(11) 2.470(4) Yb(1)-O(6) 2.456(3)
Yb(1)-O(21) 2.473(5) Yb(1)-O(7) 2.468(3)
Yb(1)-O(4) 2.502(5) Yb(1)-O(4) 2.507(3)
Yb(1)-O(5) 2.486(5) Yb(1)-O(5) 2.521(3)
C(2)-Mn(1)-C(1) 177.5(4) C(2)-Mn(1)-C(3) 179.1(3)
C(5)-O(5)-Yb(1) 158.8(6) C(5)-O(5)-Yb(1) 174.8(4)
C(4)-O(4)-Yb(1) 159.4(5) C(4)-O(4)-Yb(1) 173.5(4)
O(11)-Yb(1)-O(11) 109.6(2) O(6)-Yb(1)-O(6) 93.50(19)
O(21)-Yb(1)-O(21) 111.4(2) O(7)-Yb(1)-O(7) 90.48(15)
63 Continued
Table 2.7 (continued)
O(4)-Yb(1)-O(4) 119.3(3) O(4)-Yb(1)-O(4) 138.32(18)
O(11)-Yb(1)-O(4) 72.60(16) O(6)-Yb(1)-O(4) 72.35(13)
O(21)-Yb(1)-O(5) 72.40(18) O(7)-Yb(1)-O(4) 74.49(12)
compound cited above. The C-O-Yb angles are 158.5(5)º and 159.6(5)º, noticeably less
then 180º, indicating internal constraints in the structure. Significant disorder was present
in the carbon atoms of THF and in some oxygen atoms of CO. The Yb-O bond distances
to THF, 2.472(7) and 2.474(6) Å are longer than those in 1 and 2, corresponding to the
large (8-fold) coordination number of Yb here, and are similar to the (C)O-Yb distances
for the ordered Mn(CO)5 units in 1 (2.49 Å and 2.52 Å).The Mn-C distances in 3 are of two types, terminal (1.784(13) Å) and bridging (1.834(9) Å), with the average C-O
distances of 1.162(3) and 1.128(10) Å, respectively.
2.7 Molecular Structure of {Yb(THF)4[(μ-CO)2Mn(CO)3]2}∞ (4)
The structure of 4 is shown in Figure 2.7. Crystallographic data are given in Table
2.2, selected bond distances and bond angles are given in Table 2.7. As in compound 3,
the asymmetric unit of 4 consists of one Yb atom, two ligands, and one
pentacarbonylmanganese anion. The substantial difference is that the two 64
ligands coordinated to the Yb2+ center are both THF molecules (in 3, one is a THF molecule, and another one is Et2O). Symmetry transformations of the unique atoms
generate a one-dimensional polymeric chain (Fig. 2.7c). Linear chains of 4 are comprised
of four-member diamonds (Fig. 2.7b) that have alternatively two ytterbium and two
manganese atoms at their vertices, the former being the diamonds’ connection points.
Each diamond is planar, with its plane perpendicular to the planes of its two neighbors
(Fig 2.7c). The four THF ligands are located around the Yb2+ cation according to the
principle of least repulsion, thus lying on bisectors of dihedral angles formed by the plane
of one diamond and the edge of another. The C-O-Yb angles (174.7(4) and 173.5(4)º) are
closer to 180º then corresponding angles in 3 suggesting the greater stability of structure
4.
The Yb2+ center in both structures 3 and 4 is 8-coordinated, and possesses a
distorted square-antiprism geometry (Fig. 2.8). Vertices of the square antiprism are
- occupied by oxygen atoms of alternating ligand molecules and [Mn(CO)5] anions. In 3,
Yb-O(THF) and Yb-O(Et2O) distances are 2.467(4) and 2.475(5) Å, respectively; in 4,
Yb-O(THF) bond lengths are 2.457(3) and 2.469(3) Å. In both compounds Yb-O(THF)
bonds are about the same length and longer then those in 1 and 2, which is believed to be
due to donation of some electron density from the manganese pentacarbonylate anion to
ytterbium7a, 8. Each anion is bound to two Yb(II) centers through two of its equatorial carbonyls. Expectedly, Mn-C distances of bridging carbonyls are shorter then terminal ones. C-O distances of bridging ligands (3, 1.168(7), 1.169(8) Å; 4, 1.168(5), 1.169(5) Å) are greater then those of non-bridging equatorial carbonyls (3, 1.162(9) Å; 4, 65
1.152(7) Å) and the two axial ligands (3, 1.125(8), 1.134(9) Å; 4, 1.148(8), 1.138(7) Å).
In both structures, C-Mn-C angles deviate slightly from perfect trigonal bipyramidal geometry.
Figure 2.8. Coordination geometry around YbII center in 4. Yb1 lies on
a 2-fold rotation axis.
66
2.8 Minor Products of Condensation of 1 with Et2O. Molecular Structures of
(THF)2Mn3(CO)10 (5) and [(THF)5Yb(μ-CO)Mn3(CO)13][Mn3(CO)14] (6)
After compound 4 with some mother liquor was allowed to sit in a flask for 3 weeks, ruby-red crystals of 5 formed in the mixture with the original crystals of 4. The structure of 5 is shown in Figure. 2.9, crystallographic data are given in Table 2.8, bond lengths and bond angles are given in Table 2.9. It has a discrete molecular structure in which two Mn(CO)5 units are connected to the central Mn(THF)2 unit, at an angle of
121.76(2)º. This molecule was synthesized and characterized previously49, via the reaction of Mn2(CO)10 with AlMe3 in a mixture of THF and hexanes, which led to several products, including 5. The authors could not reproduce their synthesis, all repeated attempts to isolate compound 5 only led to formation of Mn2(CO)10. Unfortunately, we could not duplicate the synthesis either. Repeated efforts to prepare 5 resulted either in no reaction or formation of [(THF)5Yb(μ-CO)Mn3(CO)13][Mn3(CO)14], 6, (Figure 2.10) which appears to be more stable. Nevertheless we report this result as confirmation of the existence of (THF)2Mn3(CO)10 and proof of its relative stability. This molecule is predicted to be paramagnetic50 and it would be interesting to prepare a sufficiently large quantity of this material to investigate its magnetic properties.
67
Figure 2.9 Molecular structure of 5. 35% probability ellipsoids shown.
68
Table 2.8 Crystallographic Data for (THF)2Mn3(CO)10 (5) and [(THF)5Yb(μ-
CO)Mn3(CO)13][Mn3(CO)14] (6).
5 6 empirical formula C18H16Mn3O12 C48H40Mn6O33Yb formula weight 541.16 1647.28 space group P2(1)/n P2(1)/m a, Å 8.8567(10) 11.6814(10) b, Å 16.6528(10) 22.7064(10) c, Å 15.9064(10) 12.7820(10)
β, deg 94.950(10) 114.186(10)
V, Å3 2337.3(3) 3092.7(4)
Z 4 2
-3 Dcalcd, g⋅cm 1.674 1.769
T, °C -73 -73
μ, mm-1 1.654 2.772
a R1[I>2σ(I)] 0.0330 0.0327
b wR2 (all data) 0.0901 0.0814
GOF on F2 1.027 1.040
69
Table 2.9 Selected Bond Distances (Å) and Bond Angles (deg) for (THF)2Mn3(CO)10
(5).
Bond Distances Bond Angles
Mn(1)-O(3) 2.1460(17) Mn(2)-Mn(1)-Mn(3) 121.767(19)
Mn(1)-O(4) 2.1437(17) O(4)-Mn(1)-O(3) 86.64(7)
Mn(1)-Mn(2) 2.7200(6) C(11)-Mn(2)-Mn(1) 85.84(9)
Mn(1)-Mn(3) 2.7263(6) C(14)-Mn(2)-Mn(1) 73.10(9)
Mn(2)-C(11) 1.826(3) C(13)-Mn(2)-Mn(1) 176.22(9)
Mn(2)-C(12) 1.844(3) C(14)-Mn(2)-C(11) 91.34(13)
Mn(2)-C(13) 1.832(3) C(15)-Mn(2)-C(11) 89.41(13)
Mn(2)-C(14) 1.818(3)
Mn(2)-C(15) 1.820(3)
C(11)-O(11) 1.149(3)
C(12)-O(12) 1.136(3)
C(13)-O(13) 1.133(3)
C(14)-O(14) 1.153(3)
C(15)-O(15) 1.154(3)
70
Ion-paired compound 6 tends to form more predictably than 5, crystallizing in
about 3 to 4 weeks in the flask containing crystals of 4 with some mother liquor from
which the latter was grown. This process is reproducible but every time with a different
degree of success. We were unable to determine what conditions favor formation of 6.
The analogous molecule Mn2Fe(CO)14 has been prepared by UV photolysis of an
51 equimolar solution of Fe(CO)5 and Mn2(CO)10 in n-hexane . Dark-red crystals of 6 are
always mixed with 4, and both 4 and 6 are soluble and insoluble in the same sets of
solvents. For these reasons, no elemental analysis or infrared data were obtained.
Crystallographic data for 6 are given in Table 2.8. Selected bond distances and
bond angles are represented in Table 2.10. Complex 6 (Figure 2.10) is an ionic compound
- that contains a rarely observed anion [Mn3(CO)14] , which is isoelectronic with
51 52 Mn2Fe(CO)14 and has similar geometry. It was first reported by Curtis , and its X-ray
structure was published by Bau and colleagues53. Uniquely, there are two such units in
our structure and one of them is connected to the YbII center through an isocarbonyl
+ linkage, thus forming the complex cation [(THF)5Yb(μ-CO)Mn3(CO)13] (Fig. 2.10a)
- (the other [Mn3(CO)14] anion remains free). In this cation, ytterbium has distorted
octahedral coordination. The isocarbonyl bridge belongs to the central manganese atom
of the [Mn3(CO)14] core, and four out of five THF ligands are disordered. In the free
- [Mn3(CO)14] anion (Fig. 2.10b) the central Mn atom resides on a center of inversion,
thus there is one unique Mn(CO)5 moiety and two carbonyl ligands attached to the central
manganese, the rest of the anion being symmetry generated. While the free anion
possesses crystallographically imposed linear character and is of D4h symmetry, 71
the anion bound to the Yb atom is slightly bent (Mn(1)-Mn(2)-Mn(1)’ angle 172.56(5)
deg), and Mn(2) lies on a mirror plane. In the complex cation, the Mn-Mn distance
54 (2.8843(8) Å) is shorter than that in Mn2(CO)10 (2.9042(8) Å) ; length of the same bond
in the free anion (2.9030(8) Å) is closer to the value for the manganese carbonyl dimer.
For the bridging carbonyl, the Mn-C distance (1.784(7) Å) is significantly shorter than
corresponding distances of equatorial CO ligands (1.836(5) Å average). For comparison,
average Mn-C distances in Mn2Fe(CO)14 are 1.855(10) Å (equatorial) and 1.805(10) Å
(apical). The reasons for such length differences were thoroughly discussed elsewhere53.
In both free and bound [Mn3(CO)14] units, equatorial carbonyls of Mn(CO)5 groups are
bent inward, a phenomenon that has been observed for a number of carbonyl containing
53 systems . In the bound [Mn3(CO)14] unit this is slightly more pronounced for one of the ligands (C(14)-Mn(1)-Mn(2), 78.46(17)º) then for any other (85[2]º, averaged over both bound and free anions). The Yb-O(OC) distance (2.443(4) Å) is longer then any of Yb-
O(THF) (2.374[3] Å).
72
Figure 2.10 Molecular structure of 6 (35% probability ellipsoids shown): (a) the
+ [(THF)5Yb(μ-CO)Mn3(CO)13] cation (only oxygen atoms of THF shown); (b) the
- [Mn3(CO)14] anion. Yb1 and Mn2 lie on a common mirror plane and Mn3 lies on a center of symmetry.
73
Table 2.10 Selected Bond Distances (Å) and Bond Angles (deg) for (THF)2Mn3(CO)10
(5).
Bond Distances Bond Angles
Yb(1)-O(5) 2.378(4) Mn(1)#1-Mn(2)-Mn(1) 172.62(4)
Yb(1)-O(6) 2.369(3) C(11)-Mn(1)-Mn(2) 88.3(2)
Yb(1)-O(7) 2.382(4) C(12)-Mn(1)-Mn(2) 174.22(16)
Yb(1)-O(8) 2.383(4) C(13)-Mn(1)-Mn(2) 83.25(15)
O(24)-Yb(1) 2.437(4) C(14)-Mn(1)-Mn(2) 78.58(15)
Mn(1)-Mn(2) 2.8841(7) C(15)-Mn(1)-Mn(2) 87.43(17)
Mn(3)-Mn(4) 2.9023(7) C(21)-Mn(2)-Mn(1) 86.35(2)
Mn(1)-C(11) 1.828(6) C(22)-Mn(2)-Mn(1) 90.42(2)
Mn(1)-C(12) 1.801(6) C(23)-Mn(2)-Mn(1) 89.50(2)
Mn(1)-C(13) 1.837(5) C(24)-Mn(2)-Mn(1) 93.66(2)
Mn(1)-C(14) 1.824(5) C(15)-Mn(1)-C(11) 89.4(2)
Mn(1)-C(15) 1.834(5) C(12)-Mn(1)-C(11) 94.4(3)
Mn(2)-C(21) 1.826(7) C(24)-Mn(2)-C(21) 179.5(3)
Mn(2)-C(22) 1.829(7) C(22)-Mn(2)-C(21) 88.2(3)
Mn(2)-C(23) 1.837(6) C(45)-Mn(4)-Mn(3) 178.00(15)
Mn(2)-C(24) 1.790(6) C(41)-Mn(4)-Mn(3) 83.43(16)
C(11)-O(11) 1.140(6) C(42)-Mn(4)-Mn(3) 83.52(15)
C(12)-O(12) 1.147(6) C(43)-Mn(4)-Mn(3) 85.76(16)
C(13)-O(13) 1.136(5) C(44)-Mn(4)-Mn(3) 82.11(14)
74 Continued
Table 2.10 (continued)
C(14)-O(14) 1.143(5) C(31)-Mn(3)-Mn(4) 90.60(14)
C(15)-O(15) 1.148(6) C(41)-Mn(4)-C(42) 166.9(2)
C(21)-O(21) 1.151(8) C(41)-Mn(4)-C(43) 90.4(2)
C(22)-O(22) 1.145(8) C(41)-Mn(4)-C(44) 90.7(2)
C(23)-O(23) 1.137(7) C(32)-Mn(3)-C(31) 89.6(2)
C(24)-O(24) 1.171(7)
Mn(3)-C(31) 1.829(5)
Mn(3)-C(32) 1.828(5)
Mn(4)-C(41) 1.837(5)
Mn(4)-C(42) 1.842(5)
Mn(4)-C(43) 1.839(5)
Mn(4)-C(44) 1.842(5)
Mn(4)-C(45) 1.810(5)
C(31)-O(31) 1.148(5)
C(32)-O(32) 1.147(5)
C(41)-O(41) 1.144(6)
C(42)-O(42) 1.141(6)
C(43)-O(43) 1.144(6)
C(44)-O(44) 1.141(5)
C(45)-O(45) 1.140(5)
75
CHAPTER 3
Yb(II) – GROUP 6 TRANSITION METAL CARBONYL COMPLEXES.
- COMPLEXES OF THE [CpM(CO)3] ANIONS.
3.1 Starting Materials
Cyclopentadienyl tricarbonyl dimers of the transition metals of group 6 of the
general formula [CpM(CO)3]2 (M = Cr, Mo, W) were employed as starting materials to
- generate the related carbonyl anions. The resulting [CpM(CO)3] anions contain the
cyclopentadienyl ring, and is therefore substantially different from the TM anions studied in the previous chapter. It was of interest for us to investigate how such a carbonylate ligand will coordinate to the Yb2+ cation, because its symmetry is obviously lower than
- that of the [M(CO)5] anions dealt with in Chapter 2.
The dimers of molybdenum and tungsten are available commercially in 99%
purity, but the chromium analogue is not. For this reason, we prepared it according to the
published procedure (see Experimental Section). Later on we discovered that the mercury
76
- complex Hg[CpCr(CO)3]2 serves just as well as the [CpCr(CO)3] anion source, and is
easier to prepare. Consequently, preparation of the Yb(II)-TM complexes containing the
- [CpCr(CO)3] anion were carried out preferentially via the latter route.
All three [CpM(CO)3]2 dimers have similar molecular structures and IR spectra.
55 The two [CpM(CO)3] units are held together by a single M-M bond . There are no
bridging μ–CO ligands in the structure, and the two halves rotate along the M-M bond in solution55. The solution IR spectra show three bands in the carbonyl stretching region at
-1 -1 2011(w), 1956(s), 1912(s) cm ([CpMo(CO)3]2, in THF); 2008(w), 1953(s), 1907(s) cm
-1 ([CpW(CO)3]2, in THF); 2015(mw), 1948(s), 1910 cm ([CpCr(CO)3]2). The
- [CpM(CO)3] anions (as sodium salts) show two peaks in strongly coordinating solvent
pyridine at 1890, 1773 cm-1 (M = Cr); 1895, 1777 cm-1 (M = Mo); 1889, 1772 cm-1 (M =
W) (see Figure B.22 in Appendix B). The same tungsten salt demonstrates three absorptions in THF at 1895, 1793 and 1742 cm-1, showing the common phenomenon of
formation of the contact ions pairs (see Section 2.3).
The Hg[CpCr(CO)3]2 complex shows three bands in its pyridine IR spectrum at
1981(m), 1952(s), 1880(s) cm-1. Burlich and Ferrari56 reported bands at 1985(m),
1958(vs), 1900(m) and 1883(s) cm-1 for this compound in toluene. Its molecular structure has not been X-ray determined. In a closely analogous structure57 Hg[η5-
EtC5H4Cr(CO)3]2 the Hg-Cr bond length is 2.663[1] Å, the average Cr-C bond distance is
1.852[9] Å, average C-O distance is 1.14[9] Å.
77
- The [CpMo(CO)3] anion has been put to interact with the Ln(III) cations,
yielding discrete molecular type complexes21a, d. It was of interest to investigate in a
- systematic fashion interaction of the [CpM(CO)3] anions with the divalent ytterbium
cation.
3.2 Notes on Syntheses
The molybdenum and tungsten compounds discussed here (8a, 8b, 8c, 9a, 9b, 9c)
were all prepared in similar manner (Scheme 3.1). Refluxing the ytterbium amalgam with
- different [CpM(CO)3]2 dimers has led to the reduction of the dimer to the [CpM(CO)3] anion and formation of the Yb2+ cation. The chromium-containing compounds 7a, 7b and
7c were prepared starting with the Hg[CpCr(CO)3]2 complex via the intermetalation reaction (Scheme 3.2), because the [CpCr(CO)3]2 dimer was not available commercially.
In solid, these complexes are either one-dimensional chains (compounds 7a, 7b, 8a, 8b,
9a), or discrete molecules (7c, 8c, 9b, 9c) where the lanthanide and the transition metal atoms are connected via isocarbonyl bridges. Information on the composition of all of these compounds is given in Table 3.1; the compounds’ structures are shown in Figures
3.1-3.7. Crystallographic information for all structures is given in Table 3.2.
78
{(L) Yb[Cp(μ-CO) M(CO)] } n 2 2 8 8a: L = THF: M = Mo, n = 4 9a: L = THF, M = W, n = 3 8b: L = CH3CN: M = Mo, n = 4
Solvent Yb/Hg + [CpM(CO)3]2
(L)nYb[Cp(μ-CO)M(CO)2]2
9b: L = CH3CN, M = W, n = 6 8c: L = DME: M = Mo, n = 3 9c: L = DME: M = W, n = 3
Scheme 3.1 Preparation of molybdenum- and tungsten-containing complexes 8a, 8b, 8c,
9a, 9b, 9c.
79
{(THF)4Yb[Cp(μ-CO)Cr(CO)2][Cp(μ-CO)2Cr(CO)]} 8
THF 7a
CH3CN
Yb + Hg[CpCr(CO)3]2 {(CH3CN)4Yb[Cp(μ-CO)2Cr(CO)]2}8 7b DME
(DME)3Yb[Cp(μ-CO)Cr(CO)2]2 7c
Scheme 3.2 Preparation of chromium-containing complexes 7a, 7b, 7c
80
Table 3.1 Compositions of complexes 1a, 1b, 1c, 2a, 2b, 2c, 3a, 3b, 3c.
THF CH3CN
Cr {(THF)4Yb[Cp(µ-CO)Cr(CO)2]2}∞ {(CH3CN)4Yb[Cp(μ-CO)2Cr(CO)]}∞
(7a) (7b)
Mo {(THF)4Yb[Cp(μ-CO)2Mo(CO)]}∞ {(CH3CN)4Yb[Cp(μ-CO)2Mo(CO)]}∞
(8a) (8b)
W {(THF)3Yb[Cp(μ-CO)2W(CO)]}∞ (9a) (CH3CN)6Yb[Cp(μ-CO)W(CO)2]2 (9b)
Continued 81
Table 3.1 (continued)
DME
Cr (DME)3Yb[Cp(μ-CO)Cr(CO)2]2 (7c)
Mo (DME)3Yb[Cp(μ-CO)Mo(CO)2]2 (8c)
W (DME)3Yb[Cp(μ-CO)W(CO)2]2 (9c)
82
Table 3.2 Crystallographic Information for Compounds 1a, 1b, 1c, 2a, 2b, 2c, 3a, 3b, 3c.
1a 1b 1c 2a 2b 2c 3a 3b 3c
empirical C32H42Cr2 C24H22Cr2 C28H40 C16H21Mo C48H44 C28H40 C28H34O9 C28H28N6 C28H40O12
formula O10Yb N4O6Yb Cr2O12Yb O5Yb0.50 Mo4N8O12 Mo2O12Yb W2Yb O6W2Yb W2Yb
Yb2
formula 863.70 739.50 845.64 951.58 827.38 933.52 1055.29 1085.30 1109.34
weight
space P21/c C2/c C2/c C2/c C2/c C2/c C2/c C2/c C2/c 83
group
a, Å 10.6731 24.6934 18.7444 20.8974 25.5270 19.0185 23.7189 29.060(6) 18.9249
(10) (10) (10) (10) (10) (10) (10) (10)
b, Å 21.7182 8.3749(10) 9.1234 11.4670 8.4074(10) 9.0016(10) 10.6338 8.2800(17) 8.9939(10) (10) (10) (10) (10)
Continued
83
Table 3.2 (continued)
c, Å 15.0296 13.3449 20.7991 15.5191 13.3612 21.0414 12.2108 13.776(3) 21.0428
(10) (10) (10) (10) (10) (10) (10) (10)
β, deg 101.337 98.245(10) 103.993 111.527 96.383(10) 104.137 95.468(10) 90.21(3) 104.090
(10) (10) (10) (10) (10)
V, Å3 3415.9(4) 2731.3(4) 3451.4(5) 3459.4(4) 2849.7(4) 3493.1(5) 3065.8(4) 3314.7(12 3473.9(5)
84 )
Z 4 4 4 4 4 4 4 4 4
Dcalcd, 1.679 1.798 1.627 1.827 1.928 1.775 2.286 2.175 2.121
g⋅cm-3
T, °C -123 -123 -123 -123 -123 -123 -123 -123 -123
μ, mm-1 3.393 4.221 3.361 3.448 4.164 3.417 10.560 9.769 9.331
Continued
84
Table 3.2 (continued)
a R1 0.0297 0.0296 0.0276 0.0208 0.0247 0.0315 0.0245 0.0297 0.0321
[I>2σ(I)]a
b wR2 0.0431 0.0547 0.0502 0.0484 0.0499 0.0518 0.0582 0.0806 0.0634
(all data)b
GOF on 0.962 1.005 0.979 1.052 1.060 0.992 1.073 1.133 0.998
F2
85
85
3.3 The Structure of {(THF)4Yb[Cp(μ-CO)Cr(CO)2][Cp(μ-CO)2Cr(CO)]}∞
(7a)
- This is a polymeric 1-D chain that contains two types of [CpCr(CO)3] anions, a bridging one and a terminal one (Fig. 3.1). This is the only structure described in this work that contains two different types of anions. Each Yb2+ cation has three
- [CpCr(CO)3] anions coordinated to it through their carbonyl ligands, two of these anions
serving as bridges to the neighboring Yb2+ centers, and one being a terminal anion. Each
bridging anion employs two of its CO ligands to bind to two different Yb2+ cations, while
the terminal anion uses only one carbonyl ligand to coordinate to one Yb2+ center. In
addition, there are four THF molecules around each ytterbium cation that complete its
coordination sphere. Thus the coordination number of the lanthanide cation in this
structure is seven. The metal ‘backbone’ of 7a can be described as zigzag
···Yb···Cr···Yb··· sequence, where the plane of each of the triangles that make the zigzag
is roughly perpendicular to the planes of its two adjacent triangles. In the view down the
- crystallographic a axis, the Cp rings of the terminal [CpCr(CO)3] anions point
alternatively to the ‘right’ and ‘left’ sides of the chain, making this a syntactic polymeric
arrangement. Selected bond distances and angles for this structure are given in Table 3.3.
The O-Yb-O angles, that can be called angles between two [CpCr(CO)3] units
coordinated to the same Yb atom, are 124.92(9)º (O23-Yb1-O22, between two bridging
anions), 83.28(8)º (O13-Yb1-O23) and 133.40(8)º (O22-Yb1-O13) (between terminal
anion and either of two bridging anions coordinated to the same Yb2+ center). Of the four
THF ligands coordinated to the ytterbium cation, three lie below and one above the 86
Figure 3.1 Molecular structure of {Yb(THF)4[Cp(μ-CO)Cr(CO)2][Cp(μ-
CO)2Cr(CO)]}∞ 7a. (a) asymmetric unit (35% probability ellipsoids); (b) view down crystallographic b axis; (c) view down crystallographic a axis. Only oxygen atoms of
THF ligands are shown in (b) and (c).
87
Table 3.3 Selected Bond Distances (Å) and Bond Angles (deg) for 7a {(THF)4Yb[Cp(μ-
CO)Cr(CO)2] [Cp(μ-CO)2Cr(CO)]}∞
Bonds Angles
C11-O11 1.171(4) C13-O13-Yb1 163.9(3)
C12-O12 1.173(4) C23-O23-Yb1 158.2(2)
C13-O13 1.193(4) O13-Yb1-O23 83.28(8)
C21-O21 1.168(4) C11-Cr1-C12 89.17(17)
C22-O22 1.191(4) C11-Cr1-C13 87.87(16)
C23-O23 1.199(4) C12-Cr1-C13 88.63(16)
O13-Yb1 2.388(2) C21-Cr2-C22 89.55(17)
O23-Yb1 2.448(2) C21-Cr2-C23 87.28(16)
O1-Yb1 2.412(2) C22-Cr2-C23 89.95(15)
O2-Yb1 2.422(2)
O3-Yb1 2.435(2)
O4-Yb1 2.398(2)
C11-Cr1 1.808(4)
C12-Cr1 1.806(4)
C13-Cr1 1.769(4)
C21-Cr2 1.816(4)
C22-Cr2 1.780(4)
C23-Cr2 1.777(4)
88
plane that can be drawn through the three oxygen atoms of carbonyl ligands coordinated
to the Yb2+ cation. The terminal C-O distance in the bridging anion (1.168(4) Å) appears
to be less than the corresponding distances for the terminal anions (1.171(4) Å, 1.173(4)
Å), possibly as a result of a more effective electron draining by the two YbII cations attached to each bridging anion. But the differences are within three ESDs and statistically not significant.
3.4 Molecular Structures of the 1-D Polymeric Chain Compounds
{(THF)4Yb[Cp(μ-CO)2Mo(CO)]2}∞ (8a), {(THF)3Yb[Cp(μ-CO)2W(CO)]2}∞ (9a),
{(CH3CN)4Yb[Cp(μ-CO)2Cr(CO)]2}∞ (7b), {(CH3CN)4Yb[Cp(μ-CO)2Mo(CO)]2}∞
(8b).
2+ - In these compounds, each Yb cation is bound to four [CpM(CO)3] anions through isocarbonyl bridges, and also has three or four ligands coordinated around it.
Each anion is then coordinated to two cations via its carbonyl ligands. Thus a polymeric one-dimensional chain is formed, in which the two neighboring Yb2+ cations with their
coordinated ligands are connected through two bridging anions (Figures 3.2 – 3.5). Two
- of the three carbonyl ligands of each [CpM(CO)3] anion serve as isocarbonyl linkages, and the third one remains a terminal ligand. These chains are comprised of 'eight-member rings’ containing four metal atoms (two Yb and two M) and four isocarbonyl linkages.
Two adjacent rings are connected through the common Yb atom. Because of the presence
89
of isocarbonyl links, these ‘rings’ are not planar but look much like the chair
conformation of cyclohexane molecule. Such a ring-like structural motif is not unusual
for the Ln-TM compounds with isocarbonyl bridges, and structures with four- or eight-
member ‘rings’ have been reported previously8, 21.
3.4.1 One-Dimensional Chain Structures 8a and 9a
Both of these complexes have THF molecules as ligands coordinated to the Yb2+ core. In 8a, four THF molecules are bound to the cation (Fig. 3.2), while in 9a there are three of them (Fig. 3.3). This difference in the number of ligands around the Yb2+ center causes the two chains to assume different geometries (see Table 3.4 for bond lengths and angles). Both structures consist of ‘eight-member rings’ as it was described above, and those rings can be said to have a chair conformation. If we connect ytterbium atoms with imaginary lines in these structures, we will see that the Yb atoms in 8a lie almost on a straight line (Yb-Yb-Yb angle of ca. 167º), and in 9a, they lie on a zigzag above and below the chain alternatively (Yb-Yb-Yb angle of ca. 116º). In 8a, the ‘rings’ are almost perpendicular to each other (Fig. 3.2b), and the four THF ligands occupy space around the Yb atom according to the principle of the least repulsion, forming a perfect tetrahedron with the lanthanide atom in its center.
90
Figure 3.2 Molecular structure of {Yb(THF)4(Cp(μ-CO)2Mo(CO)}∞ 8a. (a) asymmetric unit (35% probability ellipsoids); (b) view of two ‘eight-member rings’ down crystallographic a axis; (c) view down crystallographic b axis. Only oxygen atoms of THF ligands are shown in (b) and (c). One of THF ligands is disordered, see
Appendix A.
91
Figure 3.3 Molecular structure of {Yb(THF)3(Cp(μ-CO)2W(CO)}∞ 9a. (a) asymmetric unit (35% probability ellipsoids); (b) view down crystallographic a axis;
(c) view down crystallographic b axis. Only oxygen atoms of THF ligands are shown in (b) and (c).
92
Table 3.4 Selected Bond Distances (Å) and Bond Angles (deg) for {(THF)nYb[Cp(μ-
CO)2M(CO)]2}∞ (8a, M = Mo, n = 4; 9a, M = W, n = 3)
Bonds 8a 9a
C1-O1 1.170(4) 1.175(6)
C2-O2 1.184(3) 1.195(7)
C3-O3 1.183(3) 1.208(6)
O11-Yb1 2.4675(19) 2.455(4)
O21-Yb1 2.498(2) 2.446(5)
O3-Yb1 2.5012(19) 2.370(4)
C1-M1 1.944(4) 1.947(6)
C2-M1 1.913(3) 1.896(6)
C3-M1 1.914(3) 1.897(5)
Angles
C3-O3-Yb1 148.2(2) 141.7(3)
O3-Yb1-O3 121.87(9) 178.10(19)
C1-M1-C2 86.98(12) 89.1(2)
C1-M1-C3 86.68(12) 87.0(2)
C2-M1-C3 85.48(11) 87.0(2)
O3-Yb1-O11 73.76(7) 83.82(14)
O3-Yb1-O21 77.91(8) 90.95(10)
93
It is instructive to observe that under the same conditions, substitution of Yb for
La leads to formation of a discrete molecule (THF)5La[(μ-CO)Mo(CO)2Cp]3 (E)
(Scheme 3.3, Figure 3.4)21a. In this structure, lanthanum has +3 oxidation state, as opposed to ytterbium in 8a, where the latter forms the Yb2+ ion. This difference in charge
does not cause difference in coordination number around the lanthanide cation (eight in
both 8a and E), but prompts change in its environment. Apparently, THF is a harder
Lewis base, and the harder Lewis acid (La3+) prefers it to the carbonyl ligand of the
- [CpMo(CO)3] anion. Such an environment around the La(III) core eliminates the
possibility of formation of the polymeric structure.
Ln = Yb {(CH CN) Yb[Cp( -CO) M(CO)] } 3 4 μ 2 2 8 (8a)
THF [CpMo(CO)3]2 + Ln/Hg in excess
(THF)5La[(μ-CO)Mo(CO)2Cp]3*THF (E) Ln = La
Scheme 3.3 Reduction of the [CpMo(CO)3]2 with ytterbium and with lanthanum.
94
Figure 3.4 Molecular structure of the discrete molecule (THF)5La[(μ-
21a CO)Mo(CO)2Cp]3 obtained by Beletskaya et al . Only oxygen atoms of THF ligands are shown for clarity.
95
A side view of the structure of 9a can be described as ‘valleys and hills’ (Fig.
3.3b). In this compound, two of the three THF ligands coordinated to the Yb2+ center are
directed outside of the chain (placed ‘on top of the hill’), and one is located ‘in the
valley’, in the ‘pocket’ between two adjacent chair-like ‘eight-member rings’. In both
compounds the bridging carbonyl ligands are bound to the Yb atom at an angle, C-O-Yb
angles being 148.2(2)º and 156.4(2)º in 8a, 141.7(3)º and 159.6(4)º in 9a. If we draw a
plane through the four oxygen atoms of the bridging carbonyls within the ‘eight-member
ring’, we will see that in 9a the tungsten atoms deviate from that plane much more than
the molybdenum atoms do in 8a. This is perhaps due to the slightly larger atomic radius
of tungsten. Structure 8a has minor disorder in one of its THF ligands (see Appendix A).
3.4.2 One-Dimensional Chain Structures 7b and 8b
These two structures are similar to one another (Fig. 3.5 and 3.6; selected bond
lengths and angles are given in Table 3.5). In each of them, there are four acetonitrile
ligands coordinated to the Yb2+ cation. In the ‘eight-member rings’, the shape is
reminiscent of the ‘chair’ conformation of cyclohexane, Yb atoms occupy corners
pointing up and down the ring’s plane. The chair conformation of the ‘rings’ means that
the isocarbonyl bridges are significantly bent, the C-O-Yb angles being 134.8(3)º and
142.7(3)º (7b), 133.4(2)º and 142.8(2)º (8b). Inequality of the two angles within the same
structure may arise due to the fact that the Cp rings and the terminal carbonyls of the two
- [CpM(CO)3] anions of the same ‘eight- member ring’ point opposite from each 96
other, thus disturbing what could otherwise be a perfect cyclohexane-like geometry. If we
draw a plane through the four oxygen atoms of the bridging carbonyls within the ‘four-
member ring’, then one atom of the transition metal will lie above and the other one
below that plane. This effect is more pronounced in 8b, probably because of the larger
size of the molybdenum atom. As expected, the C–O distances for the bridging carbonyl
ligands (1.185(4) Å and 1.183(4) Å) in 8b are longer than those of the terminal carbonyl
(1.166(4) Å). In 7b, however, the length of the C–O bond of the terminal carbonyl
(1.170(5) Å) is close to that of one of the bridging ligands (1.175(5) Å), and the second carbonyl bridge has a C–O distance noticeably longer then the other two (1.187(5) Å). At
the same time, the corresponding Yb-O distances are comparable in both compounds (see
Table 4). This observation suggests that the two isocarbonyl ligands in 7b interact with unequal strength with the Yb2+ cation. The fact that the C-O distance in the bridging carbonyl ligand is close to that in the terminal one is the evidence of weaker π*-back-
bonding from chromium atom to the CO orbitals. Such back-bonding is responsible for
the elongation of the bridging carbonyl C-O bond lengths30, 58. This feature seems to
- anticipate that in the structure of 7a, there are two types of the [CpCr(CO)3] anions
incorporated into the 1-D chain, a bridging one and a terminal one.
97
Figure 3.5 Molecular structure of {Yb(CH3CN)4[Cp(μ-CO)2Cr(CO)]2}∞ 7b. (a) asymmetric unit (35% probability ellipsoids); (b) view down crystallographic a axis; (c) view down crystallographic b axis. Only nitrogen atoms of acetonitrile ligands are shown in (b) and (c).
98
Figure 3.6 Molecular structure of {Yb(CH3CN)4(Cp(μ-CO)2Mo(CO))2}∞ 8b. (a) asymmetric unit (35% probability ellipsoids); (b) view down crystallographic a axis; (c) view down crystallographic b axis. Only nitrogen atoms of acetonitrile ligands are shown
in (b) and (c).
99
Table 3.5 Selected Bond Distances (Å) and Bond Angles (deg) for {(CH3CN)4Yb[Cp(μ-
CO)2M(CO)]2}∞ (7b, M = Cr; 8b, M = Mo)
Bonds 7b 8b
C1-O1 1.170(5) 1.166(4)
C2-O2 1.187(5) 1.185(4)
C3-O3 1.175(5) 1.183(4)
O3-Yb1 2.543(3) 2.542(3)
Yb1-N1 2.523(4) 2.515(3)
Yb1-N2 2.546(4) 2.539(4)
M1-C1 1.815(5) 1.939(4)
M1-C2 1.782(4) 1.903(4)
M1-C3 1.800(5) 1.924(4)
Angles
C3-O3-Yb1 142.7(3) 142.8(2)
C11-N1-Yb1 172.4(4) 171.9(3)
C21-N2-Yb1 177.9(4) 178.5(3)
O3-Yb1-O3 135.68(13) 136.16(12)
C1-M1-C2 90.22(17) 88.52(15)
C1-M1-C3 89.17(18) 87.10(15)
C2-M1-C3 88.69(18) 85.50(15)
100
3.5 Molecular Structures of the Discrete Molecular Compounds
(CH3CN)6Yb[Cp(μ-CO)W(CO)2]2 (9b) and (DME)3Yb[Cp(μ-CO)M(CO)2]2 (M = Cr,
7c; M = Mo, 8c; M = W, 9c).
These compounds are not polymeric solids, but exist as molecular crystals (Figs.
3.7, 3.8). For structural information, see Table 3.6. In all of these structures there are two
- 2+ [CpM(CO)3] anions, each coordinated to the Yb cation through an isocarbonyl bridge.
The cation is also surrounded by a ‘belt’ of ligands (CH3CN in 9b, DME in others). Each
- 2+ [CpM(CO)3] anion thus has one bridging and two terminal carbonyl ligands. The Yb cation has coordination number 8, which in the case of 9b amounts to an almost perfect square antiprysmatic geometry. The C-O-Yb angle varies from 155.9(4)º in 9c to
- 161.7(5)º in 9b (Table 3.6). In these compounds, the [CpM(CO)3] anions are rotated
relative to each other so that the bisectors drawn between the two terminal CO ligands are
nearly perpendicular (in 9b and 9c), or form a sharp angle close to 80º (in 7c and 8c).
One of the three DME ligands in 7c, 8c and 9c is disordered (see Appendix A). More discussion on structural features of these compounds is presented below.
101
Figure 3.7 Molecular structure of Yb(CH3CN)6(Cp(μ-CO)W(CO)2)2 9b (35% probability ellipsoids)
102
Figure 3.8. Molecular structure of compounds Yb(DME)3(Cp(μ-CO)M(CO)2)2 7c (M =
Cr), 8c (M = Mo), 9c (M = W); 35% probability ellipsoids. (a) Entire molecule, only oxygen atoms of DME ligands are shown for clarity; (b) The Yb(II) center with the three
DME ligands. In compounds 7c, 8c, 9c one of the three DME ligands is disordered, only major portion is shown (see Appendix B).
103
Table 3.6 Selected Bond Distances (Å) and Bond Angles (deg) for 9b
(CH3CN)6Yb[Cp(μ-CO)W(CO)2]2 and (DME)3Yb[Cp(μ-CO)M(CO)2]2 (M = Cr, 7c; M =
Mo, 8c; M = W, 9c)
Bonds 7c 8c 9c 9b
C1-O1 1.163(3) 1.171(5) 1.182(9) 1.160(11)
C2-O2 1.177(4) 1.171(5) 1.178(8) 1.179(10)
C3-O3 1.196(3) 1.193(5) 1.212(7) 1.194(10)
M1-C1 1.814(4) 1.943(6) 1.935(9) 1.948(10)
M1-C2 1.797(3) 1.935(5) 1.924(8) 1.920(9)
M1-C3 1.767(3) 1.891(4) 1.876(7) 1.906(9)
O3-Yb1 2.3999(18) 2.400(3) 2.393(4) 2.456(5)
O4-Yb1 2.5358(19) 2.517(3) 2.532(4) ------
O5-Yb1 2.4977(19) 2.493(3) 2.499(5) ------
O6-Yb1 2.639(14) 2.606(19) 2.621(11) ------
O7-Yb1 2.425(15) 2.458(19) 2.475(12) ------
N11-Yb1 ------2.547(8)
N21-Yb1 ------2.624(8)
N31-Yb1 ------2.594(7)
Angles
C3-O3-Yb1 159.8(2) 156.6(3) 155.9(4) 161.7(5)
O3-Yb1-O3 148.88(10) 148.10(14) 148.7(2) 144.0(2)
C11-N11-Yb1 ------177.3(7)
Continued 104
Table 3.6 (continued)
C21-N21-Yb1 ------136.1(7)
C31-N31-Yb1 ------159.9(7)
C1-M1-C2 89.55(13) 87.52(18) 87.3(3) 87.6(4)
C1-M1-C3 91.10(13) 88.53(18) 89.4(3) 89.0(3)
C2-M1-C3 86.82(12) 84.82(17) 85.2(3) 86.6(3)
3.6 Discussion of the Solution Infrared Spectra.
As one can see from Table 3.7, IR spectra of all compounds crystallized from the
same solvent are similar to each other. This is an interesting observation, because we
might expect that the differences in crystal structures between 7a (which has two
different types of anions) and 8a, 9a (two anions of the same type) would be reflected in
the spectra. Structural difference between 7c, 8c (1-D chains) and 9c (a discrete molecular structure) are not detectable in the compounds’ IR spectra. If we group spectra by the solvent, we can see that each of them consists of three major peaks, and a weak feature at ca. 2010-2050 cm-1 (this feature is much less pronounced for chromium
compounds than for the molybdenum and tungsten analogs). Some of these bands appear
to be split for the compounds of the DME group, but it is unclear whether this is a real
splitting or two independent absorptions close together.
105
Table 3.7 Solution Infrared Data for Compounds 7a, 7b, 7c, 8a, 8b, 8c, 9a, 9b, 9c in the
Carbonyl Stretching Frequency Region
-1 Compound Structure Type Medium υCO, cm
7a 1-D chain THF 2010 (vw), 1910 (s), 1815 (s),
1745 (vw), 1679 (s)
8a 1-D chain THF 2022 (vw), 1912 (s), 1817 (s),
1678 (s)
9a 1-D chain THF 2018 (vw), 1907 (s), 1812 (s),
1677 (s)
7b 1-D chain CH3CN 2056 (vw), 1945 (ms), 1893 (s),
1775 (s)
8b 1-D chain CH3CN 2023 (vw), 1930 (m), 1897 (s),
1777 (s)
9b discrete molecule CH3CN 2019 (m), 1922 (s), 1891 (s),
1772 (s)
7c discrete molecule DME 2009 (vw), 1906 (s), 1812 (s),
1792 (m), 1687 (s)
8c discrete molecule DME 2022 (wv), 1931 (vw), 1910 (s),
1814 (s), 1791 (m), 1687 (s)
9c discrete molecule DME 2018 (m), 1923 (ms), 1905 (s),
1809 (s), 1686 (s)
Na[CpM(CO)3] THF 1895(s), 1793(s), 1742(s)
106
- The isolated [CpM(CO)3] anion would possess local C3v symmetry. For such a
species, two peaks are expected in the IR spectrum in the carbonyl stretching region. The fact that three peaks are observed in solution IR spectra of all compounds studied here
suggests the lowering of their symmetry in the solution. Such lowering of the symmetry
most probably occurs via formation of the contact ion pairs8, 9 (see Section 2.3). For
complexes described in the present work, lowest frequency absorptions, located around
-1 -1 -1 1677 cm (THF group, 7a, 8a, 9a), 1774 cm (CH3CN group, 7b, 8b, 9b), and 1687 cm
(DME group, 7c, 8c, 9c), can be taken as evidence of the presence of species with bridging carbonyl ligands, since it falls in the characteristic IR region of absorption for such CO ligands. The two other peaks can be assigned to absorption by the terminal carbonyls. Further corroboration for the existence of contact pairs can be found in comparison of the IR spectra of 9a and its sodium analogue, Na[CpW(CO)3] (Table 3.7).
The sodium cation bears a single positive charge, and thus is less electron-withdrawing
than the Yb2+ cation. Consequently, it causes lesser shift of the electron density to the π*-
antibonding orbitals of the bridging CO, and its IR signal appears at higher wavenumbers
-1 -1 (Na[CpW(CO)3], 1742 cm ; 9a, 1677 cm ). It also causes less electron density to be
withdrawn from the π*-antibonding of the terminal carbonyl ligands, and their bands arise
-1 at lower frequencies than those in 9a (Na[CpW(CO)3], 1895, 1796 cm ; 9a, 1907, 1812
cm-1).
Spectra of the compounds of the THF and the DME ligands have absorptions at
almost exactly coinciding positions, and those of the CH3CN group are shifted to the
-1 higher wavenumbers by 80-100 cm (in the CH3CN group, the distance between the 107
two bands on the left is ca. 40 cm-1, much less than that in the THF and DME groups,
-1 where it is ca. 95 cm ). The CH3CN molecule is a stronger electron donor than THF or
DME, and it supplies more electron density to the Yb2+ cation. The cation, stabilized in
- this manner, draws less electron density from the [CpM(CO)3] anions attached to it,
which in turn results in less electron density being back-donated by the transition metal to
the π*-orbitals of the carbonyl ligands. This is why carbonyl stretching frequencies of
compounds of the CH3CN group are located in higher wavenumbers than those of the
compounds prepared in THF and DME. The CO ligands in structures 7b, 8b and 9b have
less electrons density on their anti-bonding orbitals than their THF and DME analogues
do, which causes C-O bonds to absorb at higher frequencies.
Taking all these data into account, it is plausible to think that when reaction
occurs (Schemes 3.1 and 3.2), molecular-type species Yb(L)n[CpM(CO)3]2 form in
solution first. Because the IR spectra of the molecular compounds 8c, 9a, 9b, 9c are very close to those of the polymeric 1-D chains 7a, 7b, 7c, 8a, 8b, we can say that extended polymers are not formed in solution. Upon evaporation of solvent, these species crystallize either ‘as is’, without further polymerization (8c, 9a, 9b, 9c), or condense into
1-D polymers (7a, 7b, 7c, 8a, 8b). This explains why the peculiarities of crystal structures of 7a (two different types of anions in the solid) and 9c (a discrete molecular
structure when 9a and 9b are polymeric chains), that are somewhat unexpected for their
solvent groups, do not manifest themselves in the solution IR spectra.
108
It was reasonable to believe that the relative Lewis basic strength of the
- [CpM(CO)3] anions is reflected in their coordinating ability through the carbonyl
ligands, and that it would also be reflected in the IR spectra: the more basic the anion is,
the more its carbonyl stretching bands are shifted to the right. In order to test this, we
collected the solution IR spectra of the salts Na[CpM(CO)3]2 (M = Cr, Mo, W) in
pyridine, the solvent that efficiently surrounds the Na+ cation, blocking interactions
between the cation and the anion8, 9 and so preventing formation of the isocarbonyl bridges. The spectra of all three compounds consist of two absorptions (local C3v symmetry) at almost exactly the same positions for all three salts, 1891 and 1774 cm-1.
This result means either that the differences in Lewis bacisities of the anions do not translate into the differences of π*-orbital population of their CO ligands, which can be
detected by the IR, or that such translation only happens in an anisotropic environment,
when a cation for coordination is available.
3.7 Discussion of the Molecular Structures.
The obvious question is how to account for the formation of polymeric chains in
some cases and discrete molecules in others. It is hard to propose a definitive explanation
here. We tried to rationalize the formation of structures on the basis of the relative
nucleophilicities of the species present in solution, but that led to counterintuitive results.
It is a common explanation that the stronger Lewis bases win the competition for the Lnn+ cation, which is a Lewis acid9. Species with the greatest Lewis basicity coordinate to 109
the cation and thus determine the final structure. The solvent molecules and the
- [CpM(CO)3] anions are the two bases present in solution at the time of the crystal
formation, and their relative basic strength must influence the outcome. The relative
nucleophilicities (Lewis basicities) of various anions were determined by Dessy and
42 - coworkers . According to their scale, where nucleophilicity of the [Co(CO)4] anion is
- taken as 1, the values for the [CpM(CO)3] anions are 4, 67 and 500 for M = Cr, Mo and
W, respectively. It was also noted that the nucleophilicities of THF and DME are
- approximately the same as that of the [Co(CO)4] anion, i.e. close to 1. It is generally
agreed that CH3CN is a stronger coordinating molecule than either THF or DME, and
thus it should be a stronger Lewis base than either of the two. Given this poorer basic
strength of THF and DME, one might think that they will always be ousted by the
- 2+ [CpM(CO)3] anions, the latter coordinating to the Yb cation via isocarbonyl bridges.
This appears to be applicable to the compounds prepared in THF (7a, 8a, 9a).
Nonetheless, molecular crystals were grown in all cases from DME (7c, 8c, 9c). It can possibly be explained by the bulkiness of this molecule and its bidentate nature; these ligands surround the Yb2+ center and do not allow formation of more then two
- isocarbonyl bridges. Also recall that the [CpM(CO)3] anions are bulky, too, and their
combination with the DME creates a sterically confined environment.
As for the remaining two groups of compounds, prepared in THF (7a, 8a, 9a) and
acetonitrile (7b, 8b, 9b), it seems that their structural features are determined by the
change of the Lewis basicity of the anion, which tunes the electron donating ability of the
carbonyl ligands. In both groups the bridging C-O distances increases while 110
going down from Cr to W (although 7a deviates from this trend) (Table 3.8), which is
probably a result of increase in π*-back-bonding. This is consistent with the series of
- 42 thenucleophilic strength for the [CpM(CO)3] anions . We can say that the more
nucleophilic anion stabilizes the Yb2+ cation better by donating more electron density to
- it. Thus in the THF group, the [CpW(CO)3] anion, which has the greatest nucleophilicity
value of the three, feeds the cation with the most electron density, to the extent that three
THF ligands are necessary for complete stability of the structure. Its molybdenum analog
is less electron donating, and the system now requires four auxiliary THF molecules. In
the case of 7a, four isocarbonyl links with two chromium-containing anions would
probably be not adequate to the task, and one of the links is replaced with a THF ligand.
Another possible explanation for the existence of the terminal anion in 7a is that the size
- of the [CpCr(CO)3] anion is smaller due to the smaller covalent radius of Cr (1.27 Å) compared to Mo (1.45 Å) and W (1.46 Å)*. This results in a shorter M-C(bridging CO)
bond in 7a (1.780(4), 1.777(4) and 1.769(4) Å) than in 8a (1.913(3), 1.914(3) Å) or 9a
- (1.896(6), 1.897(5) Å), which draws the whole of the [CpM(CO)3] anion closer to the
Yb2+ cation in 7a, creating a more crowded environment around it. Had 7a had four bridging anions around the cation, as is the case in 8a and 9a, they would be too close to each other, creating excessive repulsion.
The same approach can be applied to the acetonitrile group, compounds 7b, 8b
and 9b. Compound 9b contains the most nucleophilic (the most electron-donating) anion
- 2+ [CpW(CO)3] , and two such anions provide enough electron density to the Yb cation to complement electronic effects of the six CH3CN ligands. In 8b, the Mo- 111
containing anion is not so powerful, and the ratio of four acetonitrile to four isocarbonyl
ligands works best. For 7b the same ratio seems to be just enough to hold the structure
together, coming close to a break. This is reflected in the fact that one of the bridging C-
O distances (C3-O3, 1.175(5) Å) is noticeably shorter than another (C2-O2, 1.187(5) Å)
and is close to that of the terminal carbonyl (C1-O1, 1.170(5) Å). Expectedly, the Yb-O distance is shorter for the longer C-O bridge (Yb1-O3, 2.543(3) Å vs. Yb1-O2, 2.461(3)
Å), indicating stronger bonding.
112
Table 3.8 Bond distances (Å) for the bridging carbonyl ligands.
Compound Bonds Distance
7a C13-O13* 1.193(4)
C23-O23** 1.199(4)
C22-O22** 1.191(4)
8a C3-O3 1.183(3)
C2-O2 1.184(3)
9a C3-O3 1.208(6)
C2-O2 1.195(7)
7b C3-O3 1.175(5)
C2-O2 1.187(5)
8b C3-O3 1.183(4)
C2-O2 1.185(4)
*Terminal anion
**Bridging anion
113
CHAPTER 4
Yb(II) – TUNGSTEN CARBONYL COMPLEXES DERIVED FROM W(CO)6
4.1 Starting Materials
Tungsten hexacarbonyl W(CO)6 was primarily employed in this portion of
research to generate the transition metal carbonyl species. Some work was also done with the chromium hexacarbonyl Cr(CO)6, but the latter compound proved to be unreactive in the types of processes we tried to use it in. No attempt was made to investigate the chemistry of Mo(CO)6 in the analogous reactions. Both Cr(CO)6 and W(CO)6 are
available commercially in 99% purity, and can easily be purified further by sublimation on water bath.
Hexacarbonyls of the group 6 transition metals have been so thoroughly
investigated during the past century that it would be superfluous to cite literature on their
chemical properties here. Their structural properties and IR spectral data are relevant for
our purposes in this work. The chromium and tungsten hexacarbonyls exhibit one strong
114
band in their solution IR with a weak feature on its shoulder. The bands’ positions are
-1 -1 2020(w), 1978(s) cm for Cr(CO)6 (in pyridine) and 2014(w), 1975(s) cm for W(CO)6
(in THF). These absorptions show some fine structure when the spectrum is recorded in solid (KBr pellets), but in solution the fine division is not detectable. These solid compounds possess an octahedral geometry. The average Cr-C bond length is 1.9137[18]
59 Å, the average C-O bond length is 1.1399[14] Å . In W(CO)6 these values are 2.045[6]
Å and 1.15[9] Å, respectively60.
4.2 Reactions of the Ytterbium Amalgam with the Tungsten Hexacarbonyl in
Various Solvents. Overview of Preparation of [Yb(DMF)7][W2(CO)10] (10),
{(CH3CN)6Yb(W2(CO)10)·CH3CN}∞ (11) and {(pyr)5Yb[Hg(W(CO)5)2]}∞ (12).
All complexes considered in this chapter were prepared using the same modus operandi. The tungsten hexacarbonyl was refluxed with the ytterbium amalgam in one of the following three solvents: DMF, CH3CN or pyridine. In the course of reaction, the
-I -II 0 W(CO)6 was reduced to the W or W species, accompanied by the oxidation of Yb to
YbII. The results are summarized in Scheme 4.1. Depending on the solvent, the product of
reaction was either a solvent-separated ion pair compound [Yb(DMF)7][W2(CO)10] (10), or 1-D chain-like polymers {(CH3CN)6Yb(W2(CO)10)·CH3CN}∞ (11) and
{(pyr)5Yb[Hg(W(CO)5)2]}∞ (12). The last complex contains a novel anion
2- [Hg(W(CO)5)2] , which to the best of our knowledge was not reported previously.
115
All syntheses were carried out at room temperature. Reduction of W(CO)6 with the Yb amalgam proceeds rather slowly. Preparation of 11 can be accomplished in four days, and preparation of 10 and 12 takes two days.
DMF [Yb(DMF)7][W2(CO)10] (10)
CH3CN
Yb/Hg + W(CO)6 {(CH3CN)6Yb(W2(CO)10)·CH3CN}8 (11)
pyridine
{(pyr)5Yb[Hg(W(CO)5)2]} 8 (12)
Scheme 4.1 Preparation of compounds [Yb(DMF)7][W2(CO)10] (10),
{(CH3CN)6Yb(W2(CO)10)·CH3CN}∞ (11) and {(pyr)5Yb[Hg(W(CO)5)2]}∞ (12).
116
4.3 Preparation, Molecular Structure and the Solution IR Spectrum of the
Solvent-Separated Ion Pairs Complex [Yb(DMF)7][W2(CO)10] (10)
Reaction of the W(CO)6 with the ytterbium amalgam in DMF (Scheme 4.1) leads
2- to reduction of the hexacarbonyl molecule and formation of the [W2(CO)10] anion, accompanied with loss of two CO molecules per one anion formed. It is known that this anion can be prepared via reaction of alkali amalgam with the W(CO)6 in solution, and
2- that upon prolonged contact with excess amalgam the [W2(CO)10] anion can be further
2- 36 reduced to the [W(CO)5] anion . The ytterbium amalgam is not as powerful a reducing
agent as that of an alkali metal, and after three weeks of refluxing the solution of 10 over
2- the Yb/Hg, no signs of the [W(CO)5] anion could be observed in the IR spectra.
2+ 2- The Yb cation that forms simultaneously with the [W2(CO)10] anion has seven
2- DMF molecules coordinated around it as ligands. Both the [W2(CO)10] anion and the
DMF molecules appear to be strong Lewis bases (due to their highly nucleophilic oxygen
atoms), and thus both could compete for binding to the oxophilic Yb2+ cation. However,
the DMF proves to be a stronger coordination agent, and the cation is surrounded with it
completely, leaving no room for coordination for the carbonyl ligands of the
2- [W2(CO)10] anion. Thus the final structure belongs to Type II (Section 1.1), the solvent-
separated ion pairs (Figure 4.1).
117
Figure 4.1 Molecular structure of the [Yb(DMF)7][W2(CO)10] (10) (35% probability
2+ ellipsoids). (a) the [Yb(DMF)7] cation, only oxygen atoms of the DMF ligands are
2- shown for clarity; (b) the [W2(CO)10] anion.
118
Table 4.1. Crystallographic Data for the [Yb(DMF)7][W2(CO)10] (10),
{(CH3CN)6Yb(W2(CO)10)·CH3CN}∞ (11), {(pyr)5Yb[Hg(W(CO)5)2]}∞ (12) and
(pyr)3Cr(CO)3 (13).
10 11 12 13
empirical C31H49N7O17W2 C24H21N7O10W2 C35H25HgN5O10 C18H15CrN3O3
formula Yb Yb W2Yb
formula 1332.51 1108.22 1416.93 373.33
weight
space group P 21/c P 21/c P 21/n C 2/c
a, Å 13.7058(10) 10.2281(10) 9.4532(10) 15.7183(10)
b, Å 19.2789(10) 23.0066(12) 13.4015(10) 14.3505(10)
c, Å 17.1981(10) 15.5465(10) 30.3031(16) 14.7449(10)
β, deg 91.161(10) 108.782(10) 94.521(10) 93.919(10)
V, Å3 4543.4(5) 3463.5(5) 3827.1(5) 3318.2(4)
Z 4 4 4 8
- Dcalcd, g⋅cm 1.948 2.125 2.459 1.495
3
T, °C -123 -123 -123 -123
μ, mm-1 7.164 9.361 12.473 0.712
a R1[I>2σ(I)] 0.0396 0.0396 0.0405 0.0326
Continued
119
Table 4.1 (continued) wR2 (all 0.0916 0.0968 0.0559 0.0736 data)b
GOF on F2 1.068 1.028 0.976 1.022
120
Table 4.2. Bond Distances (Å) and Bond Angles (deg) for the [Yb(DMF)7][W2(CO)10]
(10) and the {(CH3CN)6Yb(W2(CO)10)·CH3CN}∞ (11)
10 11 10 11
Yb1-O11 (-N1) 2.416(4) 2.598(7) C3-O3 1.164(9) 1.150(10)
Yb1-O12 (-N2) 2.479(4) 2.573(7) C4-O4 1.148(9) 1.161(11)
Yb1-O13 (-N3) 2.400(4) 2.568(7) C5-O5 1.165(9) 1.186(9)
Yb1-O14 (-N4) 2.373(5) 2.526(8) C6-O6 1.153(8) 1.158(11)
Yb1-O15 (-N5) 2.469(5) 2.543(8) C7-O7 1.153(8) 1.165(12)
Yb1-O16 (-N6) 2.440(5) 2.522(8) C8-O8 1.153(8) 1.149(10)
Yb1-O5 2.479(5) C9-O9 1.157(8) 1.145(11)
Yb1-O10 2.500(5) C10-O10 1.170(8) 1.209(9)
Yb1-O17 2.415(5) W2-W1-C1 176.4(2) 83.2(2)
W1-W2 3.1491(4) 3.1331(5) W2-W1-C2 85.2(2) 82.7(3)
W1-C1 1.946(8) 2.016(9) W2-W1-C3 85.5(2) 88.1(2)
W1-C2 2.023(9) 2.024(10) W2-W1-C4 82.4(2) 85.9(2)
W1-C3 2.022(8) 2.044(9) W2-W1-C5 84.5(2) 174.9(2)
W1-C4 2.026(8) 2.033(11) W1-W2-C6 89.2(2) 86.4(2)
W1-C5 2.022(8) 1.926(8) W1-W2-C7 80.7(2) 85.6(2)
W2-C6 2.020(7) 2.046(10) W1-W2-C8 84.4(2) 83.4(2)
W2-C7 2.009(8) 2.008(10) W1-W2-C9 86.19(19) 83.2(2)
W2-C8 2.028(8) 2.029(10) W1-W2-C10 174.0(2) 176.6(2)
W2-C9 2.021(8) 2.041(11) Yb1-O5-C5 140.1(5)
121 Continued
Table 4.2 (continued)
W2-C10 1.950(7) 1.906(8) Yb1-O10-C10 141.3(6)
C1-O1 1.167(8) 1.156(10) O5-Yb1-O10 68.6(2)
C2-O2 1.153(9) 1.158(11)
The crystallographic information for 10 is given in Table 4.1, the selected bond
lengths and angles are given in Table 4.2. The Yb2+ cation has coordination number
seven, and the geometry of this complex can be approximately described as a square and
a trigonal pyramids with a common apex, where the Yb2+ is located (Figure 4.1a). The
Yb-O distances fall within the range 2.373(5) to 2.479(4) Å. This is a normal range for
the Yb-O distance when the oxygen atom is part of a ligand coordinated to the ytterbium
atom (see Section 2.4).
To the best of our knowledge, the X-ray structure determination for the
2- [W2(CO)10] anion (Figure 4.1b) was not accomplished before. This anion was first synthesized in the 1960s61, and the preparation has been repeated and improved since
36a 2- 2- then . The X-ray structures of the analogous [Cr2(CO)10] and [Mo2(CO)10] anions were published in 1970, rather shortly after these anions were first prepared62, 63. Several
complexes are known that contain the [W2(CO)10] unit and some bridging ligand of the μ-
64 65 type, such as the (μ-CHPh)W2(CO)10 , (μ-CHCH=CMe2)W2(CO)10 , and (μ-
66 SiPh2)W2(CO)10 . The W-W bond lengths in these compounds are reported to be 122
66a 2- 3.118(1), 3.1450(5) and 3.2256(8) Å, respectively. In the free [W2(CO)10] anion as it
is in 10, the W-W distance is 3.1491(4) Å. This distance is within the range reported
above, and is almost one angstrom longer than the length of the quadruple W-W bond that ranges from 2.155(2) to 2.375(1) Å58.
2- The [W2(CO)10] anion is isoelectronic and isostructural with the M2(CO)10 (M =
2- 2- Mn, Tc, Re) molecules, as well as the [Cr2(CO)10] and [Mo2(CO)10] anions. It is also
2- isoelectronic with the [Fe2(CO)8] and Co2(CO)8 complexes. It possesses approximately
D4d symmetry, a staggered geometry with the torsion angles between 42.4º and 49.4º. The
axial carbonyl ligands are slightly bent away from the equatorial ones, the C(eq)-W-
2- C(ax) angles in the range of 92.0(3)º to 98.6(3)º (Table 4.2). For the [Mo2(CO)10] anion,
63c 2- this range is 92.5(5)º to 98.8(5)º , and for the [Cr2(CO)10] anion, it is 90.7(2)º to
62c 2- 62d 98.4(2)º (the [Cr2(CO)10] anion also exists in the eclipsed conformation , where the
angles lie in the range of 90.72(6)º to 94.32(7)º, more narrow than in case of the
staggered structure). The reasons for such bending have been discussed in the literature53.
The axial CO ligands do not form quite a straight line with the W-W bond, the W-W-
C(ax) angles being 174.0(2)º and 176.4(2)º. The corresponding values for the
2- 2- [Cr2(CO)10] and [Mo2(CO)10] anions are 175.76(13)º, 174.59(14)º and 175.1(3)º,
173.7(3)º. The W-C(ax) bond lengths (1.946(8) Å and 1.950(7) Å) are predictably shorter
than the W-C(eq) ones (2.009(8) Å to 2.028(8) Å). The C-O distances in the axial
carbonyl ligands (1.167(8) Å and 1.170(8) Å) are also predictably longer than their
equatorial counterparts (1.153(8) Å to 1.165(9) Å), although these ranges overlap. The
axial ligands’ location in the anion is such that they are more susceptible to 123
accepting the electron density from the metal onto their π*-orbitals, which leads to the
elongation of the carbon to oxygen bond, and the shortening of the W-C distance (see
Section 1.1). The axial C-O distance appears to increase smoothly as one goes down the
2- 2- group of the periodic table ([Cr2(CO)10] , 1.178(4) Å and 1.176(4) Å; [Mo2(CO)10] ,
2- 1.17(1) Å and 1.19(1) Å; [W2(CO)10] , 1.167(8) Å and 1.170(8) Å). The trend is the
2- same for the equatorial C-O bond lengths, albeit not quite as definitive ([Cr2(CO)10] ,
2- 2- 1.139(5) to 1.162(4); [Mo2(CO)10] , 1.14(1) to 1.20(1); [W2(CO)10] , 1.153(8) Å to
1.165(9) Å).
The solution IR spectrum of 10 in DMF consists of four bands at 1956(m),
1939(m), 1887(s) and 1790(s) cm-1 (Table 4.3). This coincides with the data reported
36a -1 earlier for the Na2[W2(CO)10] in HMPA , except for the peak at 1956 cm . A peak at
-1 similar position (1965 cm ) was reported for the THF solution of the Na2[W2(CO)10],
along with the peaks at 2020(w), 1935(s), 1890(vs), 1825(s), 1800(s), 1740(s) cm-1 61b.
This is obviously a more complex pattern than that observed in HMPA, or the one we observe for 10 in DMF. It can probably be attributed to interactions between the cation and the anion in solution, with formation of the contact ion pairs (see Section 2.3). If the peak at 1956 cm-1 in the spectrum of 10 is really due to the weak
(DMF)nYb···(OC)W2(CO)9 interaction, then we should see its counterpart in lower
wavenumbers. Unfortunately, the DMF solvent absorbs at about 1700 cm-1, which prevents collection of meaningful IR data in that region. The HMPA is one of the most powerful coordinating molecules known, so the spectrum in this solvent can probably be
2- taken as the spectrum of the undisturbed [W2(CO)10] anion. The solution IR 124
2- spectra of all three [M2(CO)10] anions (M = Cr, Mo, W) are almost identical, showing the most intense absorption at 1885 – 1890 cm-1, and the accompanying bands at 1910 –
-1 -1 61b 1940 cm and 1740 – 1790 cm . The M2(CO)10 molecules (M = Mn, Tc, Re), have three bands in their IR spectra at about 2045(m), 2010(s) and 1980(m) cm-1, manifesting their similar structures (D4d/D4h symmetries in solution; three active IR bands are expected for both) and the lesser amount of the electron density on the π*-orbitals of their carbonyl ligands (hence the bands that are located in higher energy region).
125
Table 4.3. Infrared Data in the Carbonyl Stretching Frequency Region for Complexes
[Yb(DMF)7][W2(CO)10] (10), {(CH3CN)6Yb(W2(CO)10)·CH3CN}∞ (11),
{(pyr)5Yb[Hg(W(CO)5)2]}∞ (12) and (pyr)3Cr(CO)3 (13)
-1 Compound Medium υCO, cm
10 DMF 1956(ms), 1939(ms), 1887(vs), 1790(s)
11 CH3CN 1998(w), 1960(m), 1938(s), 1876(s)
12 Pyridine 2042(vw), 1934(vs, sh), 1877(m)
13 Pyridine 1902(s, sh), 1776(s, sh)
W(CO)6 THF 2014(w), 1975(s, sh)
Cr(CO)6 Pyridine 2020(w), 1980(s, sh)
126
4.4 Preparation, Molecular Structure and the Solution IR Spectrum of the 1-
D Polymer {(CH3CN)6Yb(W2(CO)10)·CH3CN}∞ (11)
Synthesis of 11 does not differ from that of 10 (Scheme 4.1), except that acetonitrile is used as a solvent. Tungsten hexacarbonyl is reduced with the ytterbium
2- amalgam to produce the [W2(CO)10] anion. However, in this instance the anion does not
stand free but is coordinated to two Yb2+ cations via its axial carbonyl ligands, forming a
one-dimensional polymeric chain (Figure 4.2). The CH3CN molecule is a weaker Lewis
2- base than DMF in structure 10, and the [W2(CO)10] anion competes with the solvent for
coordination sites around the lanthanide cation. As it was pointed out above (Section 4.2),
2- the C-O distance in axial carbonyl ligands in the uncoordinated [W2(CO)10] anion in 10
is longer than those in equatorial CO, which reflects the fact that the axial ligands are
more electron-rich than the equatorial ones. Consequently, they have more Lewis basic
nature, and that is why coordination to the cation occurs via them (apart from steric
considerations). One molecule of acetonitrile co-crystallizes per monomeric unit of the
chain, completing the structure.
The crystallographic information for structure 11 is given in Table 4.1. Selected
bond lengths and angles are given in Table 4.2. The polymeric chains of 11 appear as a
somewhat almost flat zigzag arrangement (Figure 4.2b), with the Yb-Yb-Yb angle of ca.
2+ 2- 117º. Each Yb cation has six CH3CN ligands and two [W2(CO)10] anions coordinated to it. With the total coordination number eight, the geometry around the cation is a distorted square antiprism, the distortion arising from different types of ligands 127
2- coordinated to the cation. It is curious that the oxygen atoms of the [W2(CO)10] anions coordinated to the lanthanide center occupy two neighboring positions in the cation’s coordination sphere, instead of positioning themselves trans to each other (the O5-Yb1-
O10 angle is 68.6(2)º). One would think that the trans configuration would create less steric strain, and the structure would be more stable in that case. Each point of the zigzag
2+ chain (where the Yb cation is located) is ‘crowned’ with the six CH3CN ligands, the
‘crowns’ pointing in the opposite direction along the a crystallographic axis. The linear
2- alignment of the two [W2(CO)10] anions along the Yb-O direction is impossible at such
a sharp angle as it would create too much steric repulsion between the two bulky anions.
2- As a consequence, the two [W2(CO)10] anions ‘bend away’, and the C-O-Yb angles are
140.1(5)º and 141.3(6)º. The Yb-O distances are slightly unequal, 2.479(5) Å and
2.500(5) Å.
2- The [W2(CO)10] anion in this compound has essentially the same structure as the one in 10, but there are some quantitative differences. It possesses approximate D4d symmetry (staggered conformation) with torsion angles ranging from 41.8º to 48.9º. Just like in the structure 10, the axial CO ligands do not perfectly align with the W-W bond, the W-C-O angles are 174.9(2)º and 174.0(2)º. But in 10, both axial ligands are bent away in the same direction from the W-W axis, the torsion angle between the two CO ligands being ca. 7.9º. In 11 they are bent in opposite directions away from each other,
2+ 2- due to coordination to two Yb cations that lie on opposite sides from the [W2(CO)10] anion. Here, the torsion angle between the axial ligands is ca. 162.7º.
128
Figure 4.2 Molecular structure of the {(CH3CN)6Yb[W2(CO)10]•CH3CN}∞ (11) (a) the asymmetric unit (35% probability ellipsoids), only oxygen atoms of the CH3CN ligands are shown for clarity, the co-crystallized molecule of CH3CN is omitted; (b) view of the 1-D chain of 11.
129
2- The W-W bond length in the [W2(CO)10] anion in 11 is 3.1331(5) Å, which is slightly shorter than that of the same bond in structure 10 (3.1491(4) Å). The axial W-C distances are 1.926(8) Å and 1.906(8) Å, and the corresponding axial C-O distances are
1.186(9) Å and 1.209(9) Å. The axial W-C and C-O bond lengths are respectively shorter
2- and longer than those in the [W2(CO)10] anion in 11 (Table 4.2), because some electron
density is drawn from the anion to the Yb2+ cation (see Section 1.1). The W-C(eq) bond
distances are between 2.008(10) Å and 2.046(10) Å, and the C-O bond length in
equatorial ligands lie in the range of 1.145(11) Å to 1.165(12) Å. This is comparable to
2- the corresponding values for the free [W2(CO)10] anion in 11 (2.009(8) Å to 2.028(8) Å
and 1.153(9) Å to 1.165(9) Å, respectively). It can be seen that coordination to the cation
only affects the axial dimension of the anion, leaving the equatorial features virtually
unchanged.
The solution IR spectrum of 11 in pyridine (Table 4.3) in the stretching carbonyl region consists of four absorptions at 1998 (w), 1960(m), 1938(s) and 1876(s) cm-1. This
2- is in agreement with the data collected earlier for the [W2(CO)10] anion (Section 4.2),
except that the low-energy absorption at ca. 1740 – 1790 cm-1 is missing. This absence is
rather singular, because strong absorption in that region is reported by all other
researchers of this anion. One might speculate that contact ion pair formation would be
more intense for 11 than for 10, because acetonitrile is a weaker coordination agent
compared to DMF, as is evidenced by the presence of the Yb-OC bond in the solid
structure. Then the (CH3CN)nYb···OCW2(CO)9 pairs would be forming more easily, resulting in a more complex IR spectrum. However, this is not observed 130
experimentally. As of now, we do not have a satisfactory explanation for this
phenomenon.
4.5 Preparation, Molecular Structure and the Solution IR Spectrum of the 1-
D Polymer {(pyr)5Yb[Hg(W(CO)5)2]}∞ (12)
Preparation of 12 is similar to that of 10 and 11 (Scheme 4.1), with the key
difference in that pyridine is employed as a solvent. This apparently is not a very
significant change, but it leads to a drastically different outcome in the synthesis,
2- producing the mercury-containing anion [Hg(W(CO)5)2] as the product of reduction of
the W(CO)6 with the ytterbium amalgam. This is surprisingly different from what we observed during the preparation of 10 and 11, where the tungsten hexacarbonyl was
2- simply reduced to the dinuclear homometallic [W2(CO)10] anion. The ytterbium metal is
oxidized to the Yb2+ cation, which adds up five pyridine ligands, and the final structure is
a one-dimensional polymeric chain {(pyr)5Yb[Hg(W(CO)5)2]}∞ (Figure 4.3).
2- To the best of our knowledge, anions of the type [Hg(M(CO)5)2] , where M is a
group 6 transition metal, were not structurally characterized before. Moreover, the only
representative of this group of anion that has been reported is the chromium analog
2- 67 [Hg(Cr(CO)5)2] . It was prepared by Behrens et al. in 1971 via heating the
Na2Cr2(CO)10 with metallic mercury in THF at 130-140º for four weeks. It was
characterized by elemental analysis and the IR spectroscopy. No X-ray data were 131
provided. No further reports dealing with this anion could be found in the literature. In
light of the lengthy procedure at elevated temperatures that was employed by Behrens to
2- prepare the [Hg(Cr(CO)5)2] , it is remarkable how readily its tungsten analog forms in
our syntheses.
Crystallographic information for 12 is given in Table 4.1. Selected bond lengths and angles are given in Table 4.4. The [Hg(W(CO)5)2] units in this structure are
connected to the Yb2+ cations through their axial carbonyl ligands. Each cation has two
CO ligands coordinated to it, and five pyridine molecules, resulting in a coordination
number of seven. The coordination geometry around the Yb2+ center can be roughly described as bicapped square prism with one more atom lying above one of the prism’s edges. The Yb-O distances in 12 are 2.479(5) and 2.487(5) Å, and the Yb-N distances
2- range from 2.513(7) Å to 2.581(6) Å. The [Hg(W(CO)5)2] anion does not coordinate to
the Yb2+ cation in a strictly linear a fashion, the Yb-O-C(ax) angles being 151.3(6)º and
145.7(6)º. This gives the polymeric chain a somewhat wavy appearance. In the solid, the
chains reside in two groups, so that two chains are parallel to each other, one above
another (if viewed down the c crystallographic axis), in the [110] direction, and another
two are directed in the [-1-10] direction. Thus when viewed down the c crystallographic
axis (the [001] direction), these pairs of chains seem to cross at an angle of ca. 70º.
2- The [Hg(W(CO)5)2] anion in 12 is nearly linear, the W-Hg-W angle is
177.27(2)º. Its geometry is close to the eclipsed one, with the torsion angles between the
equatorial carbonyl ligands at different tungsten atoms in the range from ca. 132
11.9º to ca. 16.3º. Such a spatial arrangement is similar to what is found in the
39 isoelectronic and isostructural molecule of Hg[Mn(CO)5]2 . The W-Hg bond lengths are
2.8359(5) Å and 2.8519(5) Å. As it is usual in systems such as this one, the axial C-O
distances (1.188(8) Å and 1.191(8) Å) are somewhat longer than the equatorial ones
(1.136(9) Å to 1.160(9) Å). We can note that this difference is not as clear-cut as it is in
10 or 11 (Sections 4.2 and 4.3, respectively). This is likely due to the fact that in the
2- [Hg(W(CO)5)2] anion the same charge is spread over the larger system of atoms as
2- compared to the [W2(CO)10] in 10 and 11. The axial W-C bond lengths in 12 are
1.935(9) Å and 1.928(9) Å, slightly shorter than their equatorial counterparts (2.013(9) Å to 2.040(9) Å). The equatorial CO ligands are bent away from the axial ones, the Hg1-W-
C(eq) angles range from 75.8(2)º to 91.0(2)º.
The solution IR spectrum of 12 in pyridine in the carbonyl stretching region
(Table 4.3) consists of three absorptions at 2042 (wv), 1934 (vs), 1877 (m) cm-1. The
band at 1934 cm-1 has a slight feature at ca. 1970 cm-1 that may be another weak
absorption. If this feature indeed is a band, than these three absorptions – at ca. 1970,
-1 1934 and 1877 cm – make the envelope of three peaks expected for D4d or D4h species.
2- The [Hg(W(CO)5)2] anion’s symmetry is close to D4h in the solid state, but in solution
the W(CO)5 moieties must rotate freely along the W-Hg-W axis. For the
2- 67 [Hg(Cr(CO)5)2] anion, Behrens reports absorptions at 1946 (ms), 1870 (vs) and 1835
-1 (ms) cm (in CH3CN). These absorptions are attributed to the B2, E1 and B2 modes,
respectively. This is consistent with the D4d/D4h symmetry of Mn2(CO)10 and Re2(CO)10 which exhibit similar patterns (see Section 2.1) covering about the same range as 133
2- does the [Hg(W(CO)5)2] anion, although the distribution of the peaks over the range is
2- different. The [Hg(Cr(CO)5)2] anion seems to absorb at lower wavenumbers than does
2- * the [Hg(W(CO)5)2] anion. This may indicate more intense π -back-donation (Section
2- 1.1) onto the carbonyl anti-bonding orbitals in the [Hg(Cr(CO)5)2] anion. Finally, we can point out that the contact ion pair formation is below the detection limit in the IR spectrum of 12, because pyridine is a strong coordinating solvent, precluding any significant interaction between the anion and the Yb2+ cation (Section 2.3).
134
Figure 4.3 Molecular structure of the {(pyr)5Yb(Hg(W(CO)5)2)}∞ (12). (a) asymmetric unit (35% probability ellipsoids), only nitrogen atoms of pyridine ligands are shown for clarity; (b) view of the 1-D chain of 12.
135
Table 4.4. Bond Distances (Å) and Bond Angles (deg) for the 1-D Polymeric Chain
{(pyr)5Yb[Hg(W(CO)5)2]}∞ (12)
Yb1-O5 2.479(5) C4-O4 1.150(9)
Yb1-O10 2.487(5) C5-O5 1.188(8)
Yb1-N11 2.581(6) C6-O6 1.146(8)
Yb1-N21 2.555(7) C7-O7 1.150(9)
Yb1-N31 2.513(7) C8-O8 1.160(9)
Yb1-N41 2.559(6) C9-O9 1.136(9)
Yb1-N51 2.570(7) C10-O10 1.191(8)
W1-Hg1 2.8359(5) Hg1-W1-C1 82.7(2)
W2-Hg1 2.8519(5) Hg1-W1-C2 91.0(2)
W1-C1 2.013(9) Hg1-W1-C3 82.9(2)
W1-C2 2.030(9) Hg1-W1-C4 86.3(2)
W1-C3 2.037(9) Hg1-W1-C5 175.4(2)
W1-C4 2.025(9) Hg1-W2-C6 75.8(2)
W1-C5 1.935(9) Hg1-W2-C7 88.3(2)
W2-C6 2.034(8) Hg1-W2-C8 89.0(2)
W2-C7 2.028(9) Hg1-W2-C9 85.2(2)
W2-C8 2.028(9) Hg1-W2-C10 174.6(3)
W2-C9 2.040(9) Yb1-O5-C5 145.7(6)
W2-C10 1.928(9) Yb1-O10-C10 151.3(6)
C1-O1 1.158(9) O5-Yb1-O10 152.14(18)
136 Continued
Table 4.4 (continued)
C2-O2 1.152(9) W1-Hg1-W2 177.27(2)
C3-O3 1.161(9)
137
4.6 Reduction of {(pyr)5Yb[Hg(W(CO)5)2]}∞ (12) with Sodium Amalgam
2- It was our hope that the [Hg(W(CO)5)2] anion could be reduced with ytterbium
amalgam in the fashion of the isoelectronic and isostructural Hg[M(CO)5]2 (M = Mn, Re)
2- complexes (see Chapter 2), yielding the Yb - W(CO)n species, where the [W(CO)5] anion would be present. The latter anion was prepared via reduction of the W(CO)6 by an alkali metal36a. Its presence could easily be detected by solution IR spectroscopy.
However, after 5 days of refluxing the solution of 12 over excess ytterbium amalgam no
new absorptions were visible in the IR spectrum. Thus we concluded that the ytterbium
2- amalgam could not reduce the [Hg(W(CO)5)2] anion any further. Neither does this anion react with pure mercury, which was confirmed by running similar experiments with the mercury metal instead of an amalgam.
2- Next, we attempted to reduce the [Hg(W(CO)5)2] anion with sodium amalgam.
2- If successful, the synthesis would provide complexes containing the [W(CO)5] anion and mixed Yb2+ and Na+ anions. A solution of 12 in pyridine was refluxed over an excess
of the sodium amalgam overnight, changing color from nearly black to deep-red. The IR
revealed that the spectrum of the reaction mixture was identical with that of solution
2- [Yb(DMF)7][W2(CO)10] (10) (Table 4.3), indicating the presence of the [W2(CO)10] anion. This result was unexpected, as it implied elimination of mercury and recombination of the two [W(CO)5] moieties. The elemental analysis of the product
showed that sodium was present and ytterbium was almost absent from the formula68.
This allows us to formulate the product as [Na(pyr)n]2[W2(CO)10], and the whole 138
reaction as the following (Equation (4.1))
pyridine ((pyr)5Yb)[Hg(W(CO)5)2] + 2Na/Hg Na2[W2(CO)10] + Yb + Hg (4.1)
This course of reaction is markedly different from the one that takes place when a
neutral Hg[M(CO)n]2 complex is engaged in a trasmetalation reaction (see Chapters 2 and
- 3). In the latter case, two [M(CO)n] moieties turn into the separate [M(CO)n] anions and
combine with the cation that comes to substitute mercury, which in turn is reduced to its
2- metallic state. In reduction of the [Hg(W(CO)5)2] anion, two [W(CO)5] units combine to
2- produce the [W2(CO)10] anion. Similar course of reduction was reported for the
69 [MFe(CO)4]x species (M = Zn, Cd, Hg) , where the [MFe(CO)4]x complex is probably
2- first reduced to the [MFe(CO)4] with sodium amalgam, and then oxidized to the
2- [Fe2(CO)8] anion (Equation (4.2)). One substantial difference between reactions (4.1) and (4.2) is that in the former there is also the Yb2+ cation, which is evidently reduced
0 along with mercury to Yb metal. Mechanism of transformation of the [MFe(CO)4]x into
2- 69 the [Fe2(CO)8] anion was proposed (Scheme 4.1) , although no compelling evidence
for it was shown. In particular, authors suggest that base-induced homolysis of the M-Fe
bond takes place, but never positively identify what plays the role of the base in reaction
(4.2) or Scheme 4.1. Perhaps is could be a molecule of the solvent.
139
THF 2- [MFe(CO)4]x + 2Na/Hg [Fe2(CO)8] (4.2)
2- In the [Hg(W(CO)5)2] anion, the mercury atom can be said to possess a formal
2+ oxidation state. Then the formal charge of 2- can be ascribed to each of the tungsten
2- 2- 2+ 2- atoms, and thus the anion contains two [W(CO)5] ‘sub-anions’ in it, (OC)5W -Hg -W
(CO)5. The actual partial charges borne by these atoms must be much less in their
absolute values. The difference in electronegativities between these two elements (Hg,
2.00; W, 2.36, Pauling scale41) is less than that between mercury and carbon (C, 2.55,
Pauling scale), and the Hg-C bond is considered essentially covalent. Furthermore, the
tungsten atoms are reduced, which probably lowers their affinity toward the mercury
electrons. It is safe to say that the Hg-W bond is a covalent one. Then we could consider the mercury atom as being neutral, and each of the tungsten atoms having -1 charge, thus
- 0 - (OC)5W -Hg -W (CO)5. In any case, it is yet to be explained why the two [W(CO)5] units
2- 2- prefer to merge into the [W2(CO)10] anion instead of forming two [W(CO)5] anions.
Finally, we may note that reduction of the W(CO)6 with sodium amalgam in
pyridine cleanly produces the Na2[W2(CO)10] (as evidenced by the IR spectroscopy), and
subsequent refluxing of this salt over the ytterbium amalgam does not lead to any change
in the IR spectrum, indicating that no further reaction occurs.
140
nB [MFe(CO)4]x [(M{B}n)Fe(CO)4](1)
2e- 2- (2) [(M{B}n)Fe(CO)4] [Fe(CO)4] -M
-B 2- 2- (3) [Fe(CO)4] + [(M{B}n)Fe(CO)4] [M(Fe(CO)4)2] fast
+2e- (Na/Hg) + B 2- [Fe(CO)4] [Fe(CO)4{B}] slow (4)
2- 2- [Fe(CO)4{B}] + [Fe(CO)4] [Fe2(CO)8] (5)
Scheme 4.1 Proposed reaction pathway for transformation of the [MFe(CO)4]x into the
2- 69 [Fe2(CO)8] anion .
141
4.7 Attempted Reduction of the Chromium Hexacarbonyl with the
Ytterbium Amalgam
Reduction of the chromium hexacarbonyl was attempted in the manner described
in Scheme 4.1. The solvent chosen was pyridine. It was reasonable to anticipate that such
a process would go via a pathway similar to that involving the W(CO)6, and lead to the
chromium analog of the complexes 10, 11 or 12. Unfortunately, this proved to be untrue,
and the product yielded was a well-known compound (pyr)3Cr(CO)3 (13) (Equation
(4.3)).
pyridine Cr(CO)6 + Yb/Hg (pyr)3Cr(CO)3 (4.3) -3CO
The (pyr)3Cr(CO)3 complex is widely used in organometallic synthesis as a
starting material for further preparations70, and is readily prepared by refluxing the
71 Cr(CO)6 in pyridine with heating . Somewhat strangely, the X-ray structure of
(pyr)3Cr(CO)3 was never determined, and we report it herein. The crystallographic
information for 13 is given in Table 4.1, the selected bond lengths and angles are given in
Table 4.5, solution IR data are listed in Table 4.3. Its structure is depicted in Figure 4.4.
This is a molecular complex having the fac-conformation (stool-type geometry). The
chromium coordination environment is nearly perfectly octahedral. The Cr-C 142
bond lengths (1.814(2) to 1.823(2) Å) are slightly shorter than those in Cr(CO)6
(1.9116(19) to 1.9180(13) Å)59, and the C-O bond lengths are slightly longer (1.172(3) to
1.181(3) Å in 13 vs. 1.1380(19) to 1.1426(26) Å in Cr(CO)6). These distances are
consistent with the electron donation effects from pyridine ligands to the chromium atom.
143
Figure 4.4 Molecular structure on the (pyr)3Cr(CO)3 (13) (35% probability ellipsoids)
144
Table 4.5. Bond Distances (Å) and Bond Angles (deg) for the (pyr)3B Cr(CO)B 3B B (13)
Cr1-C1 1.814(2) C1-Cr1-C2 82.35(10)
Cr1-C2 1.814(2) C1-Cr1-C3 87.54(10)
Cr1-C3 1.823(2) C2-Cr1-C3 86.15(10)
C1-O1 1.177(3) N11-Cr1-N21 89.10(7)
C2-O2 1.181(3) N11-Cr1-N31 86.68(6)
C3-O3 1.172(3) N21-Cr1-N31 89.55(6)
Cr1-N11 2.2045(17) C1-Cr1-N11 178.16(9)
Cr1-N21 2.1657(18) C2-Cr1-N21 174.13(8)
Cr1-N31 2.1971(18) C3-Cr1-N31 179.58(9)
145
CHAPTER 5
CONCLUSIONS
1. A series of new Yb-TM carbonyl complexes was prepared employing systematic
synthetic approaches, namely reduction of transition metal carbonyl complexes with
ytterbium amalgam and condensation of the solvent-separated ion pair compounds
into polymeric structures.
2. New instances of condensation of solvent-separated ion pair Ln-TM carbonyl
compounds into polymeric complexes have been observed, providing a valuable
synthetic method. A possible mechanism of the condensation reaction was proposed.
3. Reduction of the [CpM(CO)3]2 dimers, where M = Cr, Mo, W with ytterbium
amalgam was carried out systematically in coordinating solvents THF, CH3CN and
DME. Effect of the solvent on the outcome of the reaction and the products’ structure
is discussed.
2- 4. X-ray structures of the [W2(CO)10] anion and the (pyr)3Cr(CO)3 have been
determined. Although both of these species were previously known, no X-ray
structures had been reported in the literature.
146
2- 5. The previously unknown anion [Hg(W(CO)5B )B 2B ]B P P has been prepared and characterized.
Its structure was confirmed by means of single crystal X-ray crystallography.
Solution IR data were collected and compared to those of the analogous complex
2- [Hg(Cr(CO)5)2] , and its redox chemistry was briefly investigated.
147
CHAPTER 6
EXPERIMENTAL SECTION
6.1 General Procedures
All manipulations were carried out on a standard high vacuum line or in a drybox
under a nitrogen or argon atmosphere. Diethyl ether, tetrahydrofuran, pyridine, toluene,
acetonitrile and 1,3-dimethoxyethane (DME) were dried over sodium/benzophenone and
freshly distilled prior to use. N,N-Dimethylformamide (DMF) was shaken over type 4A
Linde molecular sieves for 48 hours. The sieves, and also celite, were dried under
dynamic vacuum for 18 hours at 130ºC prior to use. Ytterbium powder (Strem) and
metallic mercury (Bethlehem Apparatus, Inc.; quadruple distilled) were used as received.
Dimanganese decacarbonyl, dirhenium decacarbonyl (Aldrich) and tungsten
hexacarbonyl (Strem) were purified by vacuum sublimation at 100ºC. η5-
Cyclopentadienyltricarbonyl dimers of molybdenum and tungsten were purchased from
Aldrich and washed with hexanes to remove impurities. η5-Cyclopentadienylchromium
tricarbonyl dimer was prepared according to the literature procedure72. Mercury bis-
di(cyclopentadienylchromium tricarbonyl) Hg[CpCr(CO)3]2 was prepared according to a published procedure28a. Elemental analyses were performed by Galbraith Laboratories,
148
Inc (Knoxville, TN). Prolonged pumping on crystalline samples caused loss of
solvent ligands. Therefore, samples of compounds 1 and 2 were sent to the analytical
laboratory packed in dry ice, and analyses were calculated taking loss of ligands into
account. Infrared spectra were recorded on a Mattson Polaris Fourier transform
spectrometer with 2 cm-1 resolution. All solution IR spectra were recorded at the highest
concentration that allowed all bands to be nicely resolved, i.e. the most intense peak
reaching nearly zero transmittance. With the cell path length 0.1 mm, this concentration is estimated to be (1-3)⋅10-3 M. Yields of all reactions reported herein are apparently quantitative (no absorptions of starting materials are seen in IR), and practical yields
differ from 100% due to imperfections in collection procedures.
6.2 X-ray Structure Determination
From the X-ray studies it was found that many structures possess disordered THF
or Et2O molecules as well as disordered Mn in 1 and disordered CO in 1, 2 and 4 (see
Appendix B for disorder information). Recognition of disorder was based on the
suggestion in SHELX97 of its possibility for an atom if U1, the largest of its principal thermal ellipsoid axes, is greater than 0.2 Å 2 and if U1 is more than 2.5 times as large as its second largest such axis. For THF the splitting was extended over the C atoms and the occupancies estimated using PART 1 and PART 2 options in SHELX97. The temperature factors are large and only isotropic temperature factors were used in these disordered regions. In addition, THF molecules that lie close to the 2-fold axis 149
in 6 were assigned occupancies of 0.5. All but two of these atoms were refined
anisotropically. DFIX restraints were applied to atoms in the disordered regions: Mn-O
(1.81 A), C-C (1.52 A), C-O (1.43 A) for THF and C-O (1.13 A) for CO, all with esd =
0.02 Å. The ordered regions of the structures were not seriously affected by various
changes in the models of the disordered regions and their bond distances and angles
values appear to be reliable although of larger than desirable esd’s. The latter effect
seems to be due to the very heavy atoms (many electrons) present which contribute most
of the X-ray scattering as well as to the disorder.
Single crystals of compounds discussed in this thesis were grown using the slow
evaporation of solvent technique and kept in their mother liquor until just before they
were mounted on the X-ray diffractometer. Compound 2 was difficult to crystallize, and our best attempts only led to crystals of relatively low quality. Single crystal X-ray
diffraction data were collected using graphite monochromated Mo-Kα radiation (λ =
0.71073 Χ) on a Nonius KappaCCD diffraction system. A single crystal was mounted on
the tip of a glass fiber coated with Fomblin oil (a perfluoro polyether) that provided
protection from the atmospheric oxygen. Unit cell parameters were obtained by indexing
the peaks in the first 10 frames and refined employing the whole data set. All frames
were integrated and corrected for Lorentz and polarization effects using Denzo-SMN
package (Nonius BV, 1999)73. Absorption corrections were applied to all structures
except 2 and 4 using the SORTAV program74 provided by MaXus software75. Absorption
corrections for structures 2 and 4 were accounted for by using SCALEPACK. All of the
structures were solved by direct methods and refined using the SHELXTL-97 150
(difference electron density calculation, full matrix least-squares refinements) structure
solution package, and some other programs76. For each structure, all the nonhydrogen
atoms were located and refined anisotropically with the exception of carbon atoms of the
Et2O and THF ligands in 3 that were refined isotropically. Hydrogen atoms on solvent ligands were calculated assuming standard geometries.
6.3 Preparation of [Yb(L)n][M(CO)5]2, 1, 1a, 1b, 1c, 2. Syntheses of these
compounds are all similar (except for 2), and will be given here using
[Yb(THF)6][Mn(CO)5]2 as an example. Note that single crystals and X-ray analyses were
obtained for compounds 1 and 2 only. There are two routes for preparation of 1, 1a, 1b,
29 1c: (a) via Hg[M(CO)5]2 , and (b) via ytterbium amalgam. Compound 2 could only be
prepared using route (a); preparation of 2 via route (b) was attempted several times and
failed repeatedly.
6.3.1 Route (a). A 50 mL flask was charged with 296 mg (0.5 mmol) of
Hg[Mn(CO)5]2 and 87 mg (0.5 mmol) of Yb. Approximately 20 mL of THF was condensed into the flask at -78ºC. The mixture was warmed to room temperature and
stirred overnight, during which time the solution became dark-red in color and Hg
appeared. Infrared spectroscopy was used to monitor the reaction. Filtration of the reaction mixture through Celite gave a dark-red colored filtrate. Dark-red chunk-like crystals appeared in 1 day after slow evaporation of the solvent at room temperature until
3 mL of solution remained. It was found that in order to obtain the best X-ray quality
crystals, the initially crystallized material should be left in the mother liquor for 151
about a week at room temperature. The mother liquor was then removed, and the crystals
were washed with 3 mL of THF and 10 mL of hexanes. The crystals were dried under
vacuum for 5 min, yielding 338 mg of 1 (68% yield based on Hg[Mn(CO)5]2). Anal.
Calcd. for YbMn2C26H32O14 (-2 THF): C, 36.7; H, 3.80. Found: C, 36.25; H, 3.90. For 2,
the yield was somewhat lower (340 mg, 54% based on Hg[Re(CO)5]2). For 1c, Anal.
Calcd. for YbMn2C30H20O10N4 (-2 pyr): C, 40.9; H, 2.29. Found: C, 39.6; H, 2.68. For 2,
Anal. Calcd. for YbRe2C18.8H17.6O12.2 (-3.8 THF): C, 22.94%; H, 1.81%. Found: C,
22.56%; H, 2.00%.
6.3.2 Route (b). A 50 mL flask was charged with ytterbium amalgam prepared from ca. 5 mL of Hg and 95 mg (0.55 mmol) Yb, and 195 mg (0.5 mmol) of Mn2(CO)10.
Approximately 20 mL of THF was condensed into the flask at -78ºC. The mixture was warmed to room temperature and stirred overnight, during which time the solution became dark-red in color. Subsequent treatment of the solution led to a crop of crystals that were the same as those obtained from route (a). Yield: 1, 358 mg (72% based on
Mn2(CO)10).
6.4 Preparation of {(THF)2(Et2O)2Yb[(μ-(CO)2Mn(CO)3)2]}∞, 3. In a 50 mL
flask compound 1 was prepared from 118 mg (0.20 mmol) of Hg[Mn(CO)5]2 and 35 mg
(0.20 mmol) of Yb. The THF solvent was removed under vacuum and ca. 30 mL of Et2O was condensed into the flask at -78ºC. The mixture was warmed to room temperature and stirred overnight, during which time the solution became green in color. Filtration of the reaction mixture through Celite gave a green colored filtrate. Emerald-green crystals 152
of 3 appear immediately when evaporation of the solvent begins. Good X-ray quality
material can be obtained at an evaporation rate of ca. 3.5 - 4 mL per hour. Yields 57 mg
(33% based on Hg[Mn(CO)5]2). Dry crystals of 3 lose coordinated Et2O easily. Anal.
Calcd. for YbMn2C18H16O12 (-2Et2O): C, 30.6; H, 2.28. Found: C, 30.4; H, 2.69.
6.5 Preparation of {(THF)4Yb[(μ-CO)2Mn(CO)3)2]}∞, 4. In a 50 mL flask
compound 1 was prepared from 591 mg (1.0 mmol) of Hg[Mn(CO)5]2 and 173 mg (1.0
mmol) of Yb. The THF solvent was removed under vacuum and ca. 35 mL of Et2O was condensed into the flask at -78ºC. The mixture was warmed to room temperature and stirred overnight, during which time the solution became green in color. Filtration of the
reaction mixture through Celite gave a green colored filtrate. The filtrate was allowed to
remain in the flask for 3 days, during which time its color changed from green to yellow.
Dark Yb precipitate formed on the bottom of the flask and crystals of 4 began to appear
on the flask’s walls. The Et2O solvent was then evaporated in the course of 4 more days
until ca. 3 mL of it remained, producing well-shaped yellow crystals of 4 stuck to the
walls. The remaining solvent was pipetted out, and the crystals were carefully collected with a spatula. Yield 238 mg (28% based on Hg[Mn(CO)5]2). Anal. Calcd. for
YbMn2C18H16O12 (-2 THF): C, 30.6; H, 2.26. Found: C, 28.9; H, 2.81.
6.6 Preparation of (THF)2Mn3(CO)10, 5. A 50 mL flask containing crystals of 4
and ca. 2 mL of the mother liquor was allowed to remain for 5 days, during which time
the material in the flask started to turn from yellow to red. Investigation under a
microscope revealed that yellow crystals of 4 were mixed with some ruby-red 153
crystals, that were identified as compound 5. All subsequent attempts to repeat this
preparation led either to the formation of 6, or did not yield results. For this reason, no
analyses except the X-ray structure determination were performed on this compound.
6.7 Preparation of [(THF)5Yb(μ-CO)Mn3(CO)13][Mn3(CO)14], 6. Crystals of 4
and ca. 2 mL of the mother liquor was allowed to remain for 7 days in a 50 mL flask, at
the end of which period large, needle-shaped ruby-red crystals of 6 formed. The only way
crystals of 6 could be separated from the remaining crystals of 4 was by means of hand-
picking them, since both 4 and 6 are soluble and insoluble in the same solvents. The
amount of the material formed never exceeded 4 crystals at a time. This, and also the
difficulty of a clean reproduction of the synthesis prevented us to perform analyses other
than the X-ray on this compound.
6.8 Preparation of compounds 7a, 7b, 7c, 8a, 8b, 8c, 9a, 9b, 9c. Complexes 8a,
8b, 8c, 9a, 9b, 9c were prepared according to Scheme 3.1 (the ytterbium amalgam route).
Complexes 7a, 7b and 7c were prepared according to Scheme 3.2 (the mercury bis-
di(chromiumcyclopentadienyltricarbonyl) route).
6.8.1 Preparation via the ytterbium amalgam route (Scheme 3.1). Synthesis of
9b will be used here as an example. A 50 mL flask was charged with ytterbium amalgam prepared from ca. 5 mL of Hg and 100 mg (0.58 mmol) Yb, and 333 mg (0.5 mmol) of
[CpW(CO)3]2. Approximately 20 mL of acetonitrile was condensed into the flask at -
78ºC. The mixture was warmed to room temperature and stirred overnight, during 154
which time the solution became yellow-orange in color. Subsequent slow evaporation of
the solution led to a crop of crystals suitable for X-ray analysis. Yield: 369 mg (67.5%).
Yields of other compounds are: 8a, 51%; 8b, 51%; 8c, 65%; 9a, 53%; 9c, 4% (this
compound is poorly soluble in DME, and thus most of the product is left as a precipitate in a mixture with finely divided ytterbium amalgam).
6.8.2 Preparation via the mercury bis-
di(chromiumcyclopentadienyltricarbonyl) route (Scheme 3.2). A 50 mL flask was
charged with 301 mg (0.5 mmol) of Hg[CpCr(CO)3]2 and 100 mg (0.6 mmol) of
ytterbium metal. Approximately 20 mL of THF (to prepare 7a) or CH3CN (to prepare 7b)
or DME (to prepare 7c) was condensed into the flask at -78ºC. The mixture was warmed
to room temperature and stirred overnight, during which time the solution became green- yellow in color and Hg appeared. Infrared spectroscopy was used to monitor the reaction.
Filtration of the reaction mixture through Celite gave a green-yellow colored filtrate.
Yellow chunk-like crystals appeared in 1 day after slow evaporation of the solvent at room temperature until 3 mL of solution remained. Yield: 7a, 13%; 7b, 42%; 7c, 17%.
6.9 Elemental analyses for compounds 7a, 7b, 7c, 8a, 8b, 8c, 9a, 9b, and 9c were calculated taking into account the loss of solvent ligands (THF, DME or CH3CN)
that occurs when a sample is pumped upon as a part of preparation for shipment to
analytical laboratory. The results are as following. For 7a, Calcd. for C26H30O8.5YbCr2 (-
1.5 THF): C, 41.3%; H, 4.01%. Found: C, 39.1%; H, 4.64%. For 7b, Calcd. for
C22H19N3O6YbCr2 (-1 CH3CN): C, 37.8%; H, 2.75%. Found: C, 36.9%; H, 3.05%. 155
For 7c, Calcd. for C23H30O9YbCr2 (-1 DME, -1 CO): C, 37.9%; H, 4.16%. Found: C,
37.7%; H, 4.75%. For 8a, Calcd. for C28H34O9YbMo2 (-1 THF): C, 38.2%; H, 4.13%.
Found: C, 37.3%; H, 4.17%. For 8b, Calcd. for C24H22O6N4YbMo2 (no loss of CH3CN):
C, 34.8%; H, 2.66%. Found: C, 35.4%; H, 2.95%. For 8c, Calcd. for C26H35O11YbMo2 (-
½ DME): C, 35.1%; H, 3.98%. Found: C, 33.8%; H, 4.15%. For 9a, Calcd. for
C24H26O8YbW2 (-1 THF): C, 29.3%; H, 2.85%. Found: C, 29.5%; H, 2.81%. For 9b,
Calcd. for C28H28O6N6YbW2 (no loss of CH3CN): C, 30.1%; H, 2.61%. Found: C, 29.4%;
H, 2.71%. For 9c, Calcd. for C20H20O8YbW2 (-2 DME): C, 25.8%; H, 2.15%. Found: C,
24.4%; H, 2.44%.
6.9 Preparation of compounds 10, 11 and 12 was carried out according to
Scheme 4.1.
6.9.1 Preparation of [Yb(DMF)7][W2(CO)10] (10). A 50 mL flask was charged
with ytterbium amalgam prepared from ca. 5 mL of Hg and 100 mg (0.58 mmol) Yb, and
176 mg ( 0.50 mmol) of W(CO)6. A stirring bar and approximately 20 mL of DMF were
placed in the flask, and the contents were refluxed for ca. 2 days. The color of the solution changed from colorless to brick-red over that time. Infrared spectroscopy was used to monitor the reaction. The solution was then filtered through celite, and the solvent was slowly evaporated under vacuum on vacuum line. Dark-red crystals of 10 suitable for
X-ray analysis appeared when solution volume was reduced to ca. 2 mL. Yield: 280 mg
(42%). Anal. Calcd. for C28H42N6O16W2Yb (-1 DMF): C, 26.7%; H, 3.37%. Found: 156
C, 26.3%; H, 3.04%.
6.9.2 Preparation of {(CH3CN)6Yb(W2(CO)10)·CH3CN}∞ (11). A 50 mL flask
was charged with ytterbium amalgam prepared from ca. 5 mL of Hg and 100 mg (0.58 mmol) Yb, and 176 mg (0.50 mmol) of W(CO)6. A stirring bar was placed in the flask,
and approximately 20 mL of CH3CN were condensed into the flask at -78ºC. The mixture
was warmed to room temperature and stirred for ca. 4 days. The color of the solution
changed from colorless to yellow-red over that time. Infrared spectroscopy was used to
monitor the reaction. The solution was then filtered through celite, and the solvent was
slowly evaporated under vacuum on vacuum line. Dark-red crystals of 11 suitable for X- ray analysis appeared when solution volume was reduced to ca. 2 mL. Yield: 130 mg
(24%). Anal. Calcd. for C16H9N3O10W2Yb (-4 CH3CN): C, 20.4%; H, 0.96%. Found: C,
20.6%; H, 1.19%.
6.9.3 Preparation of {(pyr)5Yb[Hg(W(CO)5)2]}∞ (12). A 50 mL flask was
charged with ytterbium amalgam prepared from ca. 5 mL of Hg and 100 mg (0.58 mmol)
Yb, and 176 mg (0.50 mmol) of W(CO)6. A stirring bar was placed in the flask, and
approximately 20 mL of pyridine were condensed into the flask at -78ºC. The mixture
was warmed to room temperature and stirred for ca. 2 days. The color of the solution
changed from colorless to dark-brown (nearly black) over that time. Infrared
spectroscopy was used to monitor the reaction. The solution was then filtered through
celite, and the solvent was slowly evaporated under vacuum on vacuum line. Dark-red
crystals of 12 suitable for X-ray analysis appeared when solution volume was 157
reduced to ca. 2 mL. Yield: 168 mg (48%). Anal. Calcd. for C35H25O10N5W2HgYb: C,
29.7%; H, 1.78%. Found: C, 29.3%; H, 1.92%.
6.9.4 Preparation of (pyr)3Cr(CO)3 (13). A 50 mL flask was charged with
ytterbium amalgam prepared from ca. 5 mL of Hg and 100 mg (0.58 mmol) Yb, and 110
mg (0.50 mmol) of Cr(CO)6. A stirring bar was placed in the flask, and approximately 20
mL of pyridine were condensed into the flask at -78ºC. The mixture was warmed to room
temperature and stirred for ca. 2 days. The color of the solution changed from colorless to dark-brown (nearly black) over that time. Infrared spectroscopy was used to monitor the reaction. The solution was then filtered through celite, and the solvent was slowly evaporated under vacuum on vacuum line. Dark-red crystals of 13 suitable for X-ray analysis appeared when solution volume was reduced to ca. 4 mL. Yield: 35 mg (19%).
No elemental analysis was performed for this compound.
158
APPENDIX A
DISORDER IN COMPOUNDS STUDIED
159
In the structures presented here the anisotropic displacement parameters (ADP) for the ligands (THF, DME and CO) frequently are large and anisotropic. The temperature at which the data were collected (200 K) is fairly low so it seems likely that the disorder present is not primarily therma l. Under these circumstances the model chosen to represent many of these atoms is one in which the atoms are split. Using
Shelx97 these atoms are split into Part 1 and PART 2 when they are highly anisotropic and the possible separation is estimated to be greater than (0.2)0.5 Å. Problems occur in the refinement of these positions. The contribution to scattering in these structures from relatively “light” atoms (C, O) is small compared to that of “heavy” atoms (Yb, Re and even Mn). In addition, splitting the “light” atom means that it contributes in each PART only a fraction of its entire, and relatively small, contribution to scattering. For this reason it is not desirable to represent the scattering from PART 1 and PART 2 with individual ADP. In the model used to refine these structures, with one exception, the fractional atoms were kept isotropic but were extended in the disordered THF molecules to include all of the C atoms. In one case in structure 2 it was not possible to do so and the one atom suspected to be disordered was not split.
The compounds are air sensitive. The crystals are chosen for their good appearance. They are coated quickly with an inert oil and cooled as rapidly as possible in a chilled nitrogen atmosphere. Several crystals of each compound are examined, their mosaicity determined and their suitability for data collection established before the collection of the X-ray data itself. Despite all this, disordered crystal regions give unreliable structural parameters. DFIX is used to improve their quality. We focus 160
our interpretation on those regions of the crystal where disorder is not indicated. The bond distances and angles obtained for atoms present in these regions seem to yield reasonable bond distances and angles.
Example: The least satisfactory results of the diffraction experiments occurred with structure 1, C34 H48 Mn2 O16 Yb. Data were collected for three different crystals and the best data were used for the structure reported. In space group P-1 the Yb atoms
(atomic number 70) occupy special positions (0, 0, 0) and (1/2, 1/2, 1/2) and thus contribute only to reflections with h+k+1 = 2n. A comparison of the data for h+k+1 even and odd show that 80% of the total intensity measured is due to Yb. Of the remaining
20% approximately 7% is contributed by Mn and 13% by all the light atoms, O, C and H.
DFIX was used to improve the disordered regions.
Unsuccessful attempts were made to solve the structure in the non- centrosymmetric space group P1. No indication of twinning was found. There are two abnormally short intermolecular contacts calculated between atoms in the disordered
[Mn(CO)5 group in adjacent unit cells, namely O8’...O8’ = 2.1 Å and O10’…O10 = 1.6
Å. Our interpretation is that when this disordered unit is present in one unit cell, the appropriate sites in adjacent cell must have the other one of the disordered pair present.
There are no other abnormal contacts calculated in the structure.
161
Figure A.1 Molecular structure (25% probability thermal ellipsoids) showing the
disordered THF ligand in 1. The site occupancy factors are 0.47 for C21, C22, C23, C24;
0.53 for C21’, C22’, C23’, C24’.
162
Figure A.2 Molecular structure (25% probability thermal ellipsoids) showing the disordered THF ligands in 2 (asymmetric unit shown). The site occupancies are 0.29 for
C21, C22, C23, C24; 0.71 for C21’, C22’, C23’, C24’; 0.42 for C51, C52, C53, C54;
0.58 for C51’, C52’, C53’, C54’; 0.46 for C61, C62, C63, C64; 0.54 for C61’, C62’,
C63’, C64’.
163
Figure A.3 Molecular structure (15% probability thermal ellipsoids) showing the
disordered THF and Et2O ligands in 3. The occupancy factors are 0.53 for C11, C12,
C13, C14; 0.47 for C11’, C12’, C13’, C14’; 0.27 for C21, C22, C23, C24; 0.73 for C21’,
C22’, C23’, C24’.
164
Figure A.4 Molecular structure (15% probability thermal ellipsoids) showing the disordered THF ligand and disordered oxygen atoms of the carbonyl groups in the
asymmetric unit of 4. The site occupancy factors are 0.56 for O1; 0.44 for O1’; 0.74 for
O2; 0.26 for O2’; 0.54 for C16, C26, C36, C46; 0.46 for C16’, C26’, C36’, C46’; 0.39 for
C17, C27, C37, C47; 0.61 for C17’, C27’, C37’, C47’.
165
Figure A.5 Molecular structure (25% probability thermal ellipsoids) showing the disordered THF ligand in 5. The site occupancy factors are 0.36 for C31, C32, C33, C34 and 0.64 for C31’, C32’, C33’, C34’.
166
Figure A.6 Molecular structure (15% probability thermal ellipsoids for two top ligands,
35% for the bottom ligand) showing the disordered THF ligands in 6. The site occupancy factors are 0.59 for C61, C62, C63, C64; 0.41 for C61’, C62’, C63’, C64’.
167
Figure A.7 Molecular structure (15% probability thermal ellipsoids for two top ligands,
35% for the bottom ligand) showing the disordered THF ligands in 6. The site occupancy factors are 0.50 for C71 C72, C73, C74, C71A, C72A, C73A, C74A, C81, C82, C81A,
C82A.
168
Figure A.8 Disordered THF ligand in {(THF)4Yb[Cp(μ-CO)2Mo(CO)]2}∞ (8a), 35% probability ellipsoids. This is the only disordered ligand in the structure; all other THF
molecules do not show disorder. Atoms C21, C22, C23, C24 have occupancy factor 0.55.
Atoms C21’, C22’, C23’, C24’ have occupancy factor 0.45.
169
Figure A.9 Disordered DME ligand in structures (DME)3Yb[Cp(μ-CO)M(CO)2]2 (M =
- Cr, 7c; M = Mo, 8c; M = W, 9c), 35% probability ellipsoids. The [Cp(μ-CO)M(CO)2] anions are omitted for clarity. Atoms O6, O7, O6A, O7A, C13, C14, C15, C16, C13A,
C14A, C15A, C16A have occupancy factor 0.50 imposed by a 2-fold rotation axis.
170
APPENDIX B
INFRARED SPECTRA OF COMPOUNDS STUDIED
171
Figure B.1 IR spectrum of [Yb(THF)6][Re(CO)5]2 (2) in THF.
172
Figure B.2 IR spectrum of [Yb(THF)6][Mn(CO)5]2 (1) in THF prepared via the
Hg[Mn(CO)5]2 route.
173
Figure B.3 IR spectrum of NaRe(CO)5 in pyridine.
174
Figure B.4 IR spectrum of [Yb(THF)6][Mn(CO)5]2 (1) in THF prepared via ytterbium amalgam route.
175
Figure B.5 IR spectrum of [Yb(DME)n][Mn(CO)5]2 (1a) in DME.
176
Figure B.6 IR spectrum of [Yb(DMF)n][Mn(CO)5]2 (1b) in DMF.
177
Figure B.7 IR spectrum of [Yb(pyr)6][Mn(CO)5]2 (1c) after stirring in Et2O overnight; the spectrum is recorded in diethyl ether.
178
Figure B.8 IR spectrum of [Yb(THF)6][Mn(CO)5]2 (1) freshly dissolved in Et2O; initial stages of formation of 3.
179
Figure B.9 IR spectrum of the solution of [Yb(THF)6][Mn(CO)5]2 (1) in Et2O in ca. 18 hours.
180
Figure B.10 IR spectrum of {(THF)3(Et2O)[(μ-CO)2Mn(CO)3]2}∞ (3) in KBr pellet.
181
Figure B.11 IR spectrum of {(THF)4Yb[(μ-CO)2Mn(CO)3]2}∞ in KBr pellet.
182
Figure B.12 IR spectra of NaMn(CO)5 (dotted line) and [Yb(pyr)n][Mn(CO)5]2 (1) (solid line) in pyridine.
183
Figure B.13 IR spectrum of {(THF)4Yb[Cp(μ-CO)Cr(CO)2][Cp(μ-CO)2Cr(CO)]}∞ (7a) in THF.
184
Figure B.14 IR spectrum of {(THF)4Yb[Cp(μ-CO)2Mo(CO)]2}∞ (8a) in THF.
185
Figure B.15 IR spectra of Na[CpW(CO)3] (dotted line) and {(THF)3Yb[Cp(μ-
CO)2W(CO)]2}∞ (9a) (solid line) in THF.
186
Figure B.16 IR spectrum of {(CH3CN)4Yb[Cp(μ-CO)2Cr(CO)]2}∞ (7b) in CH3CN (also showing spectrum of a very dilute solution of [CpCr(CO)3]2 in CH3CN, dotted line).
187
Figure B.17 IR spectrum of {(CH3CN)4Yb[Cp(μ-CO)2Mo(CO)]2}∞ (7b) in CH3CN.
188
Figure B.18 IR spectrum of (CH3CN)6Yb[Cp(μ-CO)W(CO)2]2 (9b) in CH3CN.
189
Figure B.19 IR spectra of Hg[CpCr(CO)3]2 in pyridine (dotted line) and
(DME)3Yb[Cp(μ-CO)Cr(CO)2 ] 2 (7c) in DME.
190
Figure B.20 IR spectrum of (DME)3Yb[Cp(μ-CO)Mo(CO)2]2 (8c) in DME.
191
Figure B.21 IR spectrum of (DME)3Yb[Cp(μ-CO)W(CO)2]2 (9c) in DME.
192
- Figure B.22 IR spectra of sodium salts of the [CpCr(CO)3] (solid line), the
- - [CpMo(CO)3] (dashed line) and the [CpW(CO)3] (dotted line) recorded in pyridine.
193
LIST OF REFERENCES
1. James, C.; Willand, P.S. J.Am. Chem. Soc. 1916, 38, 1497.
2. (a) Bailey, W.E.; Williams, R.J.; Milligan, W.O. Acta Cryst. 1973, B29, 1365. (b) Mullica, D.F.; Milligan, W.O.; Kouba, W.T. Inorg. Nucl. Chem. 1979, 41, 967.
3. (a) Deng, H.; Shore, S.G.; J. Am. Chem. Soc. 1991, 113, 8538. (b) Beletskaya, I.P.; Voskoboynikov, A.Z.; Chuklanova, E.B.; Kirillova, N.I.; Shestakova, A.K.; Parshina, N.I.; Gusev, A.I.; Magomedov, G.K.-I. J. Am. Chem. Soc. 1993, 115, 3156. (c) Deng, H.; Chun, S.-H.; Florian, P.; Grandinetti, P.J.; Shore, S.G. Inorg. Chem. 1996, 35, 3891.
4. (a) Traversa, E.; Matsushima, S.;Okada, G.; Sadaoka, Y.; Sakai, Y.; Watanabe, K. Sens. Actuators, B 1995, 25, 661. (b) Sadaoka, Y.; Traversa, E.; Sakamoto, M. J. Mater. Chem. 1996, 6(8), 1355. (c) Sakamoto, M.; Matsuki, K.; Ohsumi, R.; Nakayama, Y.; Matsumoto, A.; Okawa, H. Bull. Chem. Soc. Jpn. 1992, 65, 2278. (d) Sadaoka, Y.; Aono, H.; Traversa, E.; Sakamoto, M. J. Alloys Compd. 1998, 278, 135.
5. (a) Imamura, H.; Miura, Y.; Fujita, K.; Sakata, Y.; Tsuchiya, S. J. Mol. Catal. A 1999, 140, 81. (b) Imamura, H.; Igawa, K.; Sakata, Y.; Tsuchiya, S. Bull. Chem. Soc. Jpn. 1996, 69, 325. (c) Imamura, H.; Igawa, K.; Kasuga, S.; Sakata, Y.; Tsuchiya, S. J. Chem. Soc., Faraday Trans. 1994, 90, 2119. (d) Rath, A.; Aceves, E.; Mitome, J; Liu, J.; Ozkan, U.S.; Shore, S.G. J. Mol. Catal. Commun. 2001, 165, 103. (e) Shore, S.G.; Ding, E.; Park, C.; Keane, M.A. Catal. Commun. 2002, 3, 77. (f) Jujjuri, S.; Ding, E.; Shore, S.G.; Keane, M.A. Appl. Organomet. Chem. 2003, 17, 493. (g) Satyakrishna, J.; Errun, D.; Hommel, E.L.; Shore, S.G.; Keane, M.A. J. Cat. 2006, 239, 486.
6. (a) Marianelli, R.S.; Durney, M.T. J. Organomet. Chem. 1971, 32, C41. (b) Marks, T.J.; Kristoff, J.S.; Alich, A.; Shriver, D.F. J. Organomet. Chem. 1971, 33, C35. (c) Crease, A.E.; Legzdins, P. J. Chem. Soc., Chem. Commun. 1972, 268. (d) Crease, A. E.; Legzdins, P. J. Chem. Soc., Chem. Commun. 1973, 775.
7. (a) Boncella, J.M.; Andersen, R.A. Inorg. Chem. 1984, 23, 432. (b) Tilley, T.D.; Andersen, R.A. J. Chem. Soc., Chem. Commun. 1981, 985. (c) 194
8. Tilley, T.D.; Andersen, R.A. J. Am. Chem. Soc. 1982, 104, 1772. (d) Boncella, J.M.; Andersen, R.A J. Chem. Soc., Chem. Commun. 1984, 809.
9. (a) Plecnik, C.E.; Liu, S.; Liu, J.; Chen, X.; Meyers, E.A.; Shore, S.G. Inorg. Chem. 2002, 41, 4936. (b) Plecnik, C.E.; Liu, S.; Liu, J.; Chen, X.; Meyers, E.A.; Shore, S.G. J. Am. Chem. Soc. 2004, 126, 204. (c) Plecnik, C.E.; Liu, S.; Shore, S.G. Acc. Chem. Res. 2003, 36, 499. (d) Liu, S.; Plecnik, C.; Shore, S. C. R. Chimie, 2005, 8, 1827. (e) Plecnik, C.E. Ph.D. Dissertation, The Ohio State University, 2003.
10. (a) Edgell, W.F.; Yang, M.T.; Koizumi, N. J. Am. Chem. Soc. 1965, 87, 2563. (b) Edgell, W.F.; Lyford, J. IV; Barbetta, A.; Jose, C.I. J. Am. Chem. Soc. 1971, 93, 6403. Edgell, W.F.; Lyford, J. IV J. Am. Chem. Soc. 1971, 93, 6407. (c) Darensbourg, M.Y. Prog. Inorg. Chem. 1985, 33, 221.
11. (a) Knoeppel, D. W.; Shore, S. G. Inorg. Chem. 1996, 35, 5328. (b) Knoeppel, D. W.; Shore, S. G. Inorg. Chem. 1996, 35, 1747. (c) Knoeppel, D. W.; Liu, J.; Meyers, E. A.; Shore, S. G. Inorg. Chem. 1998, 37, 4828. (d) Liu, J.; Knoeppel, D. W.; Liu, S.; Meyers, E. A.; Shore, S. G. Inorg. Chem. 2001, 40, 2842. (e) Plecnik, C.; Liu, S.; Shore, S.G. Acc. Chem. Rev. 2003, 36, 499.
12. (a) Du, B.; Meyers, E. A.; Shore, S. G. Inorg. Chem. 2001, 40, 4353. (b) Liu, S.; Meyers, E. A.; Shore, S. G. Angew. Chem., Int. Ed. 2002, 41, 3609.
13. (a) Du, B. Ph.D. Dissertation, The Ohio State University, 2000. (b) Zhao, H.; Prosvirin, A.V.; Bacsa, J.; Avendano, C.; Dunbar, K.R. Abstracts of Papers, 229th ACS National Meeting, San Diego, CA, United States, March 13-17, 2005, INOR-663. (c) Mishra, A.; Wernsdorfer, W.; Abboud, K.A.; Christou, G. J. Am. Chem. Soc. 2004, 126, 15648. (d) Kou, H.-Zh.; Zhou, B.C.; Wang, R.-J. Inorg. Chem. 2003, 42, 7658. (e) Winpenny, R.E.P. Chem. Soc. Rev. 1998, 27, 447-452. (f) Brechin, E. K.; Harris, S. G.; Parsons, S.; Winpenny, R. E. P. J. Chem. Soc., Dalton Trans.: Inorg. Chem. 1997, 10, 1665. (g) Benelli, C.; Gatteschi, D. Chem. Rev. 2002, 102(6), 2369.
14. (a) Sakamoto, M.; Matsuki, K.; Ohsumi, R.; Nakayama, Y.; Matsumoto, A.; Οkawa, H. Bull. Chem. Soc. Jpn. 1992, 65, 2278. (b) Fang, J.; You, H.; Chen, J.; Lin, J.; Ma, D. Inorg. Chem. 2006, 45(9), 3701.
15. Shuk, P.; Vecher, A.; Kharton, V.; Tichonova, L.; Wiemhöfer, H. D.; Guth, U.; Göpel, W. Sens. Actuators, B 1993, 16, 401.
16. Traversa, E.; Matsushima, S.; Okada, G.; Sadaoka, Y.; Sakai, Y.; 195
Watanabe, K. Sens. Actuators, B 1995, 25, 661.
17. (a) Baiker, A.; Marti, P.E.; Keusch, P.; Fritsch, E.; Reiller, A. J. Catal. 1994, 146, 268. (b) Lago, R.; Bini, G.; Pina, M.A.; Fierro, J.L.G. 3rd World Congress on Oxidation Catalysis Studies in Surface Science and Catalysis 1997, 110, 721. (c) Yoshihko, S.; Enrico, T.; Sakamoto, M. J. Alloys Compd. 1996, 240, 51. (d) Buassi-Monroy, O.S.; Luhrs, C.C.; Chavez-Chavez, A.; Michel, C.R. Mat. Letters 2004, 58, 716.
18. Voskoboynikov, A.Z.; Beletskaya, I.P. Russ. Chem. Bull. 1997, 46, 1789.
19. (a) Suleimanov, G.Z.; Khandozhko, V.N.; Shifrina, R.R.; Abdullaeva, L.T.; Kolobova, N.E.; Beletskaya, I.P. Dokl. Akad. Nauk SSSR 1984, 277, 1407. (b) Suleimanov, G.Z.; Beletskaya, I.P. Dokl. Akad. Nauk SSSR 1981, 261, 381. (c) Magomedov, G.K.-I.; Voskoboynikov, A.Z.; Chuklanova, E.B.; Gusev, A.I.; Beletskaya, I.P. Metalloorg. Khim. 1990, 3, 706. (d) Beletskaya, I.P.; Suleimanov, G.Z.; Shifrina, R.R.; Mekhdiev R.Y.; Agdamskii, T.A.; Khandozhko, V.N.; Kolobova, N.E. J. Organomet. Chem. 1986, 299, 239.
20. Wells, A. F. Structural Inorganic Chemistry, 5th ed.; Oxford University Press: New York, 1984; p 1288.
21. Cannon, J. F.; Robertson, D. L.; Hall, H. T. Mater. Res. Bull. 1972, 7, 5.
22. (a) Pasynskii, A.A.; Eremenko, I.L.; Suleimanov, G. Z.; Nuriev, Yu.A.; Beletskaya, I. P.; Shklover, I.E.; Struchkov, Yu.T. J. J. Organomet. Chem. 1984, 266, 45. (b) Hazin, P.N.; Huffman, J.C.; Bruno, J.W. J. Chem. Soc., Chem. Commun. 1988, 1473. (c) Recknagel, A.; Steiner, A.; Brooker, S.; Stalker, D.; Edelmann, F.T. Chem. Ber. 1991, 124, 1373. (d) Hou, Z.; Aida, K.-i.; Takagi, Y.; Wakatsuki, Y. J. Organomet. Chem. 1994, 473, 101. (e) Lin, G.; Wong, W.-T. J. Organomet. Chem. 1996, 522, 271. (f) Hillier, A.C.; Sella, A.; Elsegood, M.R.J. J. Chem. Soc., Dalton Trans. 1998, 3871.
23. White, J. P., III; Deng, H.; Boyd, E. P.; Gallucci, J.; Shore, S. G. Inorg.Chem. 1994, 33, 1685.
24. (a) Suleimanov, G. Z.; Khandozhko, V. N.; Abdullaeva, L. T.; Shifrina, R. R.; Khalilov, K. S.; Kolobova, N. E.; Beletskaya, I. P. J. Chem. Soc., Chem. Commun. 1984, 191. (b) Suleimanov, G. Z.; Khandozhko, V. N.; Petrovskii, P. V.; Mekhdiev, R. Y.; Kolobova, N. E.; Beletskaya, I. P. J. Chem. Soc., Chem. Commun. 1985, 596. (c) Lin, G.; Wong, W.-T. J. Organomet. Chem. 1996, 522, 271. 196
25. (a) Deng, D.; Zheng, X.; Qian, C.; Sun, J.; Dormond, A.; Baudry, D.; Visseaux, M. J. Chem. Soc., Dalton Trans. 1994, 1665. (b) Hillier, A. C.; Sella, A.; Elsegood, M. R. J. J. Chem. Soc., Dalton Trans. 1998, 3871. (c) Beletskaya, I. P.; Voskoboynikov, A. Z.; Chuklanova, E. B.; Kirillova, N. I.; Shestakova, A. K.; Parshina, I. N.; Gusev, A. I.; Magomedov, G. K.-I. J. Am. Chem. Soc. 1993, 115, 3156.
26. (a) Onaka, S.; Furuichi, N. J. Organomet. Chem. 1979, 173, 77. (b) Onaka, S. Inorg. Chem. 1980, 19, 2132.
27. (a) Beletskaya, I. P.; Suleimanov, G. Z.; Mekhdiev, R. Y.; Khandozhko, V. N.; Petrovskii, P. V.; Kolobova, N. Dokl. Akad. Nauk SSSR 1986, 289, 96. (b) Suleimanov, G. Z.; Khandozhko, V. N.; Mekhdiev, R. Y.; Petrovskii, P. V.; Kolobova, N. E.; Beletskaya, I. P. Izv. Akad. Nauk SSSR, Ser. Khim. 1986, 1210. (c) Suleimanov, G. Z.; Khandozhko, V. N.; Mekhdiev, R. Y.; Petrovskii, P. V.; Yanovskaya, I. M.; Lependina, O. L.; Kolobova, N. E.; Beletskaya, I. P. Izv. Akad. Nauk SSSR, Ser. Khim. 1988, 685.
28. Evans, W. J.; Bloom, I.; Grate, J. W.; Hughes, L. A.; Hunter, W. E.; Atwood, J. L. Inorg.Chem. 1985, 24, 4620.
29. (a) Burlitch, J. M.; Ferrari, A. Inorg. Chem. 1970, 9, 563. (b) Hartley, F. R. The Chemistry of the Metal-Carbon Bond; Wiley: New York, 1987; Vol. 4, pp 71-89. (c) Crabtree, R. H. The Organometallic Chemistry of the Transition Metals, 2nd ed.; Wiley: New York, 1994; p 51.
30. Hsieh, A.T.T.; Mays, M.J. J. Chem. Soc. (A), 1971, 2648.
31. Cotton, F. A.; Wilkinson, G. Advanced Inorganic Chemistry: A Comprehensive Text; Wiley: New York, 1980; pp 1001-1002.
32. (a) Evans, W.J. J. Organomet. Chem. 2002, 652, 61. (b) Fagin, A.A.; Bochkarev, M.N.; Kozimor, S.A.; Ziller, J.W.; Evans, W.J. Z. Allorg. Allg. Chem. 2005, 631, 2848.
33. A.J. Bard, R. Parsons, and J. Jordan, Standard Potentials in Aqueous Solutions, IUPAC (Marcel Dekker), New York, USA, 1985.
34. Beletskaya, I. P.; Voskoboinikov, A. Z.; Magomedov, G. K.-I. Metalloorg. Khim. 1989, 2, 814. 35. 197
36. a) Baiker, A.; Marti, P.E.; Keusch, P.; Fritsch, E.; Reiller, A. J. Catal. 1994, 146, 268. (b) Lago, R.; Bini, G.; Pina, M.A.; Fierro, J.L.G. 3rd World Congress on Oxidation Catalysis Studies in Surface Science and Catalysis 1997, 110, 721. (c) Yoshihko, S.; Enrico, T.; Sakamoto, M. J. Alloys Compd. 1996, 240, 51. (d) Buassi-Monroy, O.S.; Luhrs, C.C.; Chavez-Chavez, A.; Michel, C.R. Mat. Letters 2004, 58, 716.
37. Szymanska-Buzar, T. Coord. Chem. Rev. 2005, 249(21-22), 2195.
38. (a) Anastassiou, A.G.; Elliott, R.L.; Reichmanis, E. J. Am. Chem. Soc. 1974, 96, 7825. (b) Darensbourg, M.Y.; Slater, S. J. Am. Chem. Soc. 1981, 103, 5914. (c) Behrens, H.; Weber, R. Z. Allorg. Allgem. Chem. 1957, 241, 122. (d) Behrens, H.; Vogl, J. Z. Allorg. Allgem. Chem. 1957, 291, 123. (e) Ellis, J.E.; Hagen, G.P. Inorg. Chem. 1977, 16, 1357.
39. Ellis, J.E.; Flom, E.A. J. Organomet. Chem. 1975, 99, 263.
40. Burlitch, J.M. Chem. Commun. 1968, 887.
41. (a) Clegg, W.; Wheatly, P.J. J. Chem. Soc. (A), 1971, 3572. (b) Florke, U.; Haupt, H. Acta Cryst., Sect. C. 1992, 48, 1709.
42. Slater, C. J. Chem. Phys. 1964, 39, 3199.
43. (a) James, A.M.; Lord, M.P. Macmillan's Chemical and Physical Data, Macmillan, London, UK, 1992. (b) www.webelements.com
44. (a) Dessey, R.E.; Pohl, R.L.; King, R.B. J. Am. Chem. Soc. 1966, 88, 5121. (b) King, R.B. Acc. Chem. Res. 1970, 3, 417.
45. By the solution IR spectrum of the ‘pure’ metal pentacarbonyl anion we mean the spectrum which was recorded in strongly coordinating solvent, such as pyridine or DMF. Those solvents bind tightly to the cation and preclude any interaction between it and the carbonyl ligands of the anion, so that the latter’s geometry is unaffected. Compare spectra of 1, 1c and KMn(CO)5 in pyridine, Appendix B.
46. (a) Manning, M.J.; Knobler, C.B.; Khatter, R.; Hawthorne, M.F. Inorg. Chem. 1991, 30, 2009. (b) Deng, D.; Qian, C.; Song, F.; Wang, Z.; Wu, G.; Zheng, P. J. Organomet. Chem. 1993, 443, 79. (c) Tilley, T.D.; Andersen, R.A.; Spencer, B.; Ruben, H.; Zalkin, A.; Templeton, D.H. Inorg. Chem. 1980, 19, 2999. (d) Zinnen, H.A.; Pluth, J.J.; Evans, W.J. J. Chem. Soc., Chem. Commun. 198
1980, 810.
47. (a) Frenz, B.A.; Ibers, J.A. Inorg. Chem. 1972, 11, 1109. (b) Corraine, M.S.; Lai, C.K.; Zhen, Y.; Churchill. M.R.; Buttrey, L.A.; Ziller, J.W.; Atwood, J.D. Organometallics 1992, 11, 35. (c) Kong, G.; Harakas, G.N.; Whittlesey, B.R. J. Am. Chem. Soc. 1995, 117, 3502.
48. (a) Kaska, W.C. J. Am. Chem. Soc. 1968, 90, 6340. (b) Anastassiou, A.G.; Elliot, R.L.; Reichmanis, E. J. Am. Chem. Soc. 1974, 96, 7825. (c) Darensbourg, M. Y.; Slater, S. J. Am. Chem. Soc. 1981, 103, 5914. (d) Maher, J.M.; Beatty, R.P.; Cooper, N.J. Organometallics 1985, 4, 1354. (e) Seidel, R.; Schautz, B.; Henkel, G. Angew. Chem., Intl. Ed. Engl. 1996, 35, 1710.
49. (a) Hanson, A.W. Acta Crystallogr. 1962, 15, 930. (b) Donohue, J.; Caron, A. Acta Crystallogr. 1964, 17, 663.
50. 240 mg of 3 was dissolved in 50 ml of diethyl ether, and the solution was allowed to sit in the flask for 7 days. Black precipitate appeared, along with the crystals of 4. The remaining solution was then removed, and the precipitate was washed with THF in order to separate ytterbium metal (compound 4 dissolves in THF, apparently producing 1). 69 mg of the black material was collected. Analysis performed by Galbraith Laboratories, Inc. showed 42.42% ytterbium content, which amounts to 29.3 mg of ytterbium metal. On the basis of stoichiometry of reaction (8), 24.2 mg is expected. The explanation of this difference could possibly be that not all of 4 was washed off, thus contributing to ytterbium content of the sample.
51. Kong, G.; Harakas, G.N.; Whittlesey, B.R. Am. Chem. Soc. 1995, 117, 3502.
52. Xu, Z., Lin, Z. Chem. Eur. J. 1998, 4, 28.
53. Agron, P.A.; Ellison, R.D.; Levy, H.A. Acta Cryst. 1967, 23, 1079.
54. Curtis, M.D. Inorg. Chem. 1972, 11, 802.
55. Bau, R.; Kirtley, S.W.; Sorrell, T.N.; Winarko, S. J. Am. Chem. Soc., 1974, 96, 988.
56. Bianchi, R.; Gervasio, G.; Marabello, D. Inorg. Chem. 2000, 39, 2360
199
57. (a) Adams, R.D.; Collins, D.M.; Cotton, F.A. J. Am. Chem. Soc. 1974, 96, 749. (b) Collins, D.M.; Cotton, F.A. Inorg. Chem. 1974, 13, 1086. (c) Gould, R.O.; Barker, J.; Kilner, M. Acta Cryst., Sect. C 1988, 44, 461.
58. Burlitch, J.; Ferrari, A. Inorg. Chem. 1970, 9, 563.
59. Song, L.-C.; Tao, Z.-F.; Hu, Q.-M.; Wang, R.-J.; Wang, H.-J. Polyhedron, 1994, 13, 2179.
60. Cotton, F.A.; Murillo, C.A.; Walton, R.A. (editors) Multiple Bonds between Metal Atoms, 3rd Ed., Springer Science and Business Media, Inc., 2005.
61. (a) Whitaker, A.; Jeffery, J.W. Acta Crystallogr. 1967, 23, 977. (b) Jost, A.; Rees, B.; Yelon, W.B. Acta Crystallogr., Sect. B 1975, 31, 2649. (c) Rees, B.; Mitschler, A. J. Am. Chem. Soc. 1976, 98, 7918.
62. (a) Heinemann, F.; Schmidt, H.; Peters, K.; Thiery, D. Z. Kristallogr. 1992, 198, 123. (b) Grevels, F.-W.; Jacke, J.; Klotzbucher, W.E.; Mark, F.; Skibbe, W.; Scheffner, K.; Angermund, K.; Kruger, C.; Lehmann, C.W.; Ozkar, S. Organometallics 1999, 18, 3278.
63. (a) Behrens, H.; Vogl, J. Chem. Ber. 1963, 96(8), 2220. (b) Hayter, R.G. J.Am. Chem. Soc. 1966, 88, 4376.
64. (a) Handy, L.B.; Ruff, J.K.; Dahl, L.F. J. Am. Chem. Soc. 1970, 92, 7312. (b) Hey-Hawkins, E.; von Schnering, H.G. Chem. Ber. 1991, 124, 1167. (c) Lee, I.; Geib, S.G.; Cooper, N.J. Acta Crystallogr., Sect. C 1996, 52, 292. (d) Borrmann, H.; Pirani, A.M.; Schrobilgen, G.J. Acta Crystallogr., Sect. C 1997, 53, 19. (e) Borrmann, H.; Pirani, A.M.; Schrobilgen, G.J. Acta Crystallogr., Sect. C 1997, 53, 1007.
65. (a) Handy, L.B.; Ruff, J.K.; Dahl, L.F. J. Am. Chem. Soc. 1970, 92, 7312. (b) Bachman, R.E.; Whitmire, K.H. Acta Crystallogr., Sect. C 1993, 49, 327. (c) Petz, W.; Neumuller, B.; Lorberth, J.; Megges, K.; Massa, W. Z. Naturforsch, B: Chem. Sci. 2001, 56, 306.
66. Fischer, H.; Zeuner, S.; Ackermann, K.; Schmid, J. Chem. Ber. 1986, 119, 1546.
67. Parlier, A.; Rudler, M.; Rudler, H.; Daran, J.C. J. Organomet. Chem. 1987, 323, 353. 68. (a) Butts, M.D.; Bryan, J.C.; Luo, X.-L.; Kubas, G.J. Inorg. Chem. 200
1994, 36, 3341. (b) Gadek, A.; Kochel, A.; Szymanska-Buzar, T. Organometallics 2003, 22, 4869.
69. Behrens, H.; Feilner, H.-D.; Lindner, E.; Uhlig, D. Zeitschrift fur Naturforschung, B, 1971, 26(10), 990.
70. Product of reduction of 12 with sodium amalgam was analyzed for presence of 2- both sodium and ytterbium in it. The IR spectrum showed that the [W2(CO)10] anion was present, thus the product’s formula could be [Yb(pyr)6][W2(CO)10] or [Na(pyr)x]2[W2(CO)10]. Sodium content was found at 5.05%, and ytterbium content was found at 2.11%. The figure for sodium content is in agreement with formula [Na(pyr)1.5]2[W2(CO)10] (Na, 4.94%). If the product was the [Yb(pyr)6][W2(CO)10], then ytterbium content must have been at least 13.4%. From this, we conclude that the product of reduction of 12 with sodium amalgam is [Na(pyr)x]2[W2(CO)10], and ytterbium present in analysis is due to minor contamination.
71. Sosinsky, B.A.; Shong, R.G.; Fitzgerald, B.J.; Norem, N.; O’Rourke, C. Inorg. Chem. 1983, 22, 3124.
72. (a) Oprunenko, Yu. F.; Malyugina, S. G.; Ustynyuk, Yu. A.; Ustynyuk, N.A. Izvestiya Akademii Nauk SSSR, Seriya Khimicheskaya 1985, 9, 2156. (b) Akhmedov, N.G.; Malyugina, S.G.; Mstislavsky, V.I.; Oprunenko, Yu. F.; Roznyatovsky, V.A.; Ustynyuk, Yu. A.; Batsanov, A.S.; Ustynyuk, N.A. Organometallics 1998, 17, 4607. (c) Zabalov, M.V..; Gloriozov, I. P.; Oprunenko, Yu. F.; Lemenovskii, D. A. Russ. Chem. Bul. (Transl. Of Izv. Akad. Nauk., Ser. Khim) 2003, 52, 1567. (d) Hannesschlager, P.; Brun, P. Appl. Organometal. Chem. 2000, 14, 686. . (e) Perez-Encabo, A.; Perrio, S.; Slwin, A.M.Z.; Thomas, S.E.; Wierzchleyski, A.T.; Williams, D.J. J. Chem. Soc., Perkins Trans. 1 1994, 629.
73. Brauer, G.; Verlag, F.E. (editors) Handbook of Preparative Inorganic Chemistry, v. 2 1965, New York, Academic Press, 1963-65.
74. Birdwhistell, R.; Hackett, P.; Manning, A.R. J. Organomet. Chem. 1978, 157, 239.
75. Otwinowski, Z.; Minor, W. Methods in Enzymology; (Eds.: C.W. Carter Jr., R.M. Sweet), Academy Press, New York, 1997, Vol. 276(A), 307.
76. Blessing, R.H. J. Appl. Crystallogr. 1997, 30, 421.
201
77. Mackay, S.; Gilmore, C.J.; Edwards, C.; Tremayne, M.; Stuart, N.; Shankland, K. MaXus; University of Glasgow: Scotland, Nonius BV: Delft, The Netherlands, and MacScience Co. Ltd.: Yokohama, Japan, 1998.
78. (a) Sheldrick, G.M. SHELXTL-97. University of Göttingen, Germany, 1998. (b) WinGX 1.70.01. L.G. Farrugia, J. Appl. Cryst., 1999, 32, 837-838.
202