Coordination Chemistry - American Chemical Society

Chapter 14

The Importance of Non-Bonds Michael Laing

Downloaded by PENNSYLVANIA STATE UNIV on July 23, 2013 | http://pubs.acs.org Publication Date: November 4, 1994 | doi: 10.1021/bk-1994-0565.ch014

Department of Chemistry, University of Natal, Durban 4001, South Africa

The term "coordination complex" almost invariably brings to mind a transition metal atom to which are bonded six atoms in an octahedron. We now automatically concentrate our thinking on these metal-to-ligand bonds and tend to assume that they are all-important in the behavior of the compound. In fact, the stoichiometry, coordination geometry, and reactivity of many complexes is dictated not by the metal and the atoms to which it is bonded, but by the bulk of parts of the ligands remote from the bonding center: the Non-Bonds between the ligands.

Alfred Werner was born French in Mulhouse, Alsace; he grew up and received his schooling in German as an outcome of the Franco-Prussian War; and he was a citizen of Switzerland when he became famous. Although he achieved greatness as an inorganic coordination chemist, he began his career as an organic chemist. Thus it seems appropriate to begin this discussion of Non-Bonds with the important contributions of the great Russian crystallographer, A . I . Kitaigorodskii, to an understanding of the principles governing the structure of crystals of organic compounds. Close Packing In 1961 the English translation of Kitaigorodskii's classic book was published (J). In it he emphasized that molecules have a well-defined volume and shape and that the atoms have a "hard" contact surface which prevents two non-bonded atoms from approaching closer than the sum of their contact or van der Waals radii. (The well-known Leybold and C P K models incorporate this principle.) He concluded that it is the thickness and shape of the molecules that dominate their mode of packing in the solid state. They pack together in the crystal by a "bump-in-hollow" process, the projections of one molecule fitting into the hollows of adjacent molecules, invariably

0097-6156/94/0565-0177$08.00/0 © 1994 American Chemical Society In Coordination Chemistry; Kauffman, G.; ACS Symposium Series; American Chemical Society: Washington, DC, 1994.

COORDINATION CHEMISTRY


178

with each molecule in contact with six others in a layer, to give material of maximum density with a minimum energy (2) (Figures 1 and 2). It is important to understand that there is N O essential physical difference between the non-bonded repulsions between atoms within a molecule and these intermolecular repulsions between molecules in the solid state (5). Thus, for example, while two methyl groups within a coordination compound may repel each other, they will still pack together by the bump-in-hollow process, with their C - H bonds acting like gears or knuckles. This process considerably reduces the effective van der Waals volume of the group from the typical value of about 2 À for the radius of a randomly or freely rotating methyl group. The implications of Kitaigorodskii's ideas are enormously wide-ranging because they apply not only to organic molecules and crystals but to every coordination compound. Steric Approach Control In 1957 Dauben described effectively the same concept but from an entirely different point of view — that of the organic chemist looking at the reduction of a carbonyl group to an alcohol (4). He found that the conformation of the O H group produced by reduction of the C = 0 group by the BH4 ion was determined by the pathway of L E A S T hindrance to C(3) (Figure 3). If the approach from above, A , was open, then the H atom went onto C(3) from above and the subsequently formed O H group was axial down, the energetically LESS favoured conformation. The easier the approach from above, the more likely it was to get the thermodynamically less stable axial hydroxyl group below. If bulky groups were present on C(2), C(4), or C(10) above, then the attack came from below and the O H group was formed on the same side as the bulky groups. He termed this phenomenon "Steric Approach Control" — control by kinetics, where the first-formed compound was the thermodynamically less favored isomer. Rate of Reaction Basolo gives a beautiful example of the effect of the physical blocking of the reaction pathway and thus the reduction in reactivity at the metal center (5). In the series of compounds rrûAW-[PtCl(PEt ) L] where L is a phenyl ring, ortho substituted, the rate of nucleophilic substitution of CI by pyridine systematically decreases as the bulk of the ortho substituent increases: 3

L phenyl 0-tolyl mesityl

2

^obs»

s e c

1

4

1.2 χ ΙΟ" 1.7 χ ΙΟ" 3.2 χ 10"

5

6

The cause is easily pictured: the "top" of the metal atom is screened from the incoming nucleophile by the bulk of the methyl group on the phenyl ring, which is remote, chemically inert, and in no way bonded to the metal atom. The methyl groups simply block the pathway for the incoming ligand (Figure 4).

In Coordination Chemistry; Kauffman, G.; ACS Symposium Series; American Chemical Society: Washington, DC, 1994.


14·

LAING

The Importance of Non-Bonds

179

Figure 1. Packing of randomly shaped molecules by a simple bump-in-hollow process.

Figure 2. Randomly shaped molecules close-packed by "glides" to yield a structure with a rectangular cell. Note how each molecule touches six others.



180


Figure 3. If the axial groups at C(2), C(4), and C(10) are small, the B H will approach along pathway A . If the groups are bulky, the attack is along path B. 4

Figure 4. A schematic diagram of the compound mzAw-[PtCl(PEt ) L], L = mesityl, showing how the bulk of the two methyl groups blocks access to the platinum atom. 3

2


14.

181


LAING

Cone Angle for Phosphine Ligands As a result of his studies of the rates of substitution of metal carbonyls by various phosphines and of the equilibrium constants for decomposition of the substituted complexes, Tolman concluded that in many cases it was not electronic bonding effects that control the reactions but that it was rather the bulk of the phosphine ligand that dictates the lability of the compounds (6). He developed the concept of Cone Angle for a ligand, defined in terms of an idealized M - P X system, where the M-P distance was 2.28 À, and the P X ligand was constructed of CPK models. The Cone Angle θ was the apex angle of the ideal cone, which had its apex at the metal atom and just touched the van der Waals surface of the peripheral atoms of the ligand, usually hydrogens (7). Typical values of Θ, the Cone Angle, are: P F 104°; PMe^j 118°; P P h 145°; P(f-But) 182° (Figure 5). Subsequently, modifications have been suggested for estimating the Cone Angle of asymmetric ligands of the class P X Y Z (8 -11). Also, the effective Cone Angle as observed and measured from crystal structures often is considerably smaller than the ideal. This is a result of the bump-in-hollow gearwheel effect as well as of compression of the Metal-P-R angles. Nevertheless, the principle is correct, and the concept is of great value, as is made clear in a recent review by McAuliffe (12). The importance of Cone Angle in metal-catalyzed reactions in organic synthesis has also been elegantly described (13). 3


3

3

3

3

cis^ or *rans-[PtCl (PR ) ]? 2

3

2

Forty years ago Chatt and Wilkins studied the cis-trans equilibria for these square planar complexes of platinum (14). It was evident that bond energies favored the cis isomer. This is caused by the τ back-bonding from the metal d-orbital being concentrated on only one phosphorus atom. The isomer which has the greater Pt-P double-bond character will have the larger overall bond strength. This strengthening of the Pt-P bond will be favored by the arrangement Cl-Pt-PR . Thus the cis square planar isomers will be favored if bonding effects A L O N E were in control (In the trans case two phosphorus atoms would of necessity be competing for the favors of the same d-orbital). The effect is also clearly seen in the facial and meridional isomers of [IrCl (PMe2Ph) ] (15). In the fac isomer the mean bond lengths are: Ir-Cl 2.46 À; Ir-P 2.29 À. In the mer isomer, the mean length for the trans pair of Ir-P bonds is 2.37 À, while that for the trans pair of Ir-Cl bonds is 2.36 À. On the other hand, the lengths of the unique bonds of the trans Cl-Ir-P moiety are: Ir-Cl 2.43 À; Ir-P 2.28 À (values very close to those found in the facial isomer). The Ir-P bond is considerably shorter and stronger, indicating that M - P bond strengths favor the facial isomer for [ M X ( P R ) ] as well as the cis isomer for [ M X ( P R ) ] , 3

3

3

3

3

3

2

3

2


182


Figure 5. The Cone Angle Θ, as defined for a typical phosphine ligand P X . 3


Trans

Cis

Figure 6. The cis and trans isomers of [PtCl (PPr ) ] showing Pt-P and Pt-Cl bond lengths and how the van der Waals envelopes of the cis phosphines interfere. 2

3

2

Figure 7. The molecule [Pt{PPh(i-But) }2l, showing how the van der Waals surfaces of the hydrogen atoms block access to the platinum atom. 2

However, i f the alkyl groups on the phosphines are large, the van der Waals repulsions between the groups will cause the square planar compound to isomerize to yield the trans isomer, which does not suffer from inter-ligand compression strain. A good example is [PtCl (PPr ) ] which can be prepared as the pure cis isomer, but on equilibration in benzene for half an hour, it yields an equilibrium mixture that contains 97% of the trans isomer (5). Non-bonds are clearly beating bonds. The larger the Cone Angle of the phosphine ligand, the more the trans isomer is favored (Figure 6). 2

3

2


183


14. LAING

Phosphine Complexes of Platinum and Palladium Compounds like [ P t i P P h ^ ] and [Ni(PMe3) ] are easily made (16). However, in solution [ P t i P P h ^ J dissociates, forming the three-coordinate complex [Pt(PPh ) ]. Determination of the crystal structure of [PtiPPhj)^ shows quite certainly that the metal atom is four- coordinate with P-Pt-P angles close to 109°, i.e., it is tetrahedral (17). This appears to contradict Tolman's value of 145° for the Cone Angle of triphenylphosphine. In fact, it does not, because in this tetrakis case the phenyl rings are locked together like gears. Nevertheless, there are severe compression strains, which cause dissociation of the complex in solution. By the same token, the inter-ligand repulsions in the compound [Pt(P(cyclohexyl) ) ] are surprisingly small, a result of the "intermeshing" of the cyclohexyl groups reducing the effective Cone Angle far below Tolman's estimate of about 170° (78). If the phosphine ligand is large enough, only two ligands can be fitted onto the platinum atom. The compound [Pt{PPh(r-But) }2l is an example (19). The P-Pt-P backbone is linear, and the metal atom is shielded from the outside world by the organic moieties, thus rendering the platinum atom inert to attack. The Cone Angle of PPh(f-But) is 170° (Figure 7). 4


3

3

3

3

2

2

Addition of 0

2

to [Plat inum(0) (phosphine^]

It is possible to prepare a variety of diphosphine complexes of platinum(O) and palladium(O) (19). Some of these compounds are sensitive to oxygen, forming compounds of the class [ P t i P R ^ O J , while others are quite inert to oxygen (11). The explanation is simply given by consideration of the Cone Angles of the phosphine ligands involved. Where the ligand has a sufficiently large Cone Angle, e.g., PPh(i-But) , θ = 170°, the compound is inert because there is neither a pathway for the oxygen molecule to approach the metal atom nor is there place in the coordination shell about the metal atom i f the 0 should come within bonding distance. When the Cone Angle is smaller, the compound is easily formed, e.g., [Pt(PPh ) ] immediately reacts with oxygen to yield [PuTPhj^OJ (20). The compound [Pt{PPh(r-But) } 0 ] can be made by an indirect route (21) and the crystal structure shows beautifully how the anisotropy of the ligand allows the compound to exist. The phenyl rings of the two phosphine ligands lie face-to-face, just touching, in a typical Kitaigorodskii close-packing mode, thus reducing the effective Cone Angles of the ligands to give a P-Pt-P angle of 113°. It is impossible to fit the two ligands onto the metal in any other orientation (Figure 8). 2

2

3

2

Reactive [RuH(PMe Ph) ] 2

2

2

2

+

5

+

A number of cationic hydride complexes of the type [ R u H L ] , L = P(OR) , were prepared in 1974. They were remarkably inert in solution. Subsequently, the compound [RuH(PMe2Ph) ] , with a far bulkier phosphine ligand, was found to be highly reactive in solution and a precursor to a large range of new Ru(II) complexes formed by the substitution of one or more of the phosphine ligands (22). The crystal structure of the PFg salt clearly showed the severe compression strains of the PMe Ph 5

3

+

5

2


184


ligands, the "interlocking" of the methyl and phenyl groups of adjacent ligands to effectively reduce the Cone Angle, and (in some senses more interesting) the way in which the H atom on the ruthenium atom was shielded from the outside world by the van der Waals bulk of the phenyl and methyl groups of four of the PMe^Ph ligands. What should be a labile H atom is rendered inert by non-bonds (Figure 9). "Umbrella" Effect in the Purple Isomer of [Ru^CHMPMe^h)^* Reaction of [RuH(PMe2Ph) ]PF with C S yields an orange dithioformato complex [Ru(S CH)(PMe2Ph)4] (25), which rearranges in boiling methanol to a purple isomer (24). This isomer had three PMe Ph ligands arranged facially on the ruthenium atom, while the Ρ atom of the fourth was bonded to the carbon of the CS^ group. The effective Cone Angles of the PMe Ph groups are larger in the purple isomer with a coordination number of 5 than in the orange isomer with a coordination number of 6, indicating that relief of the non-bonded repulsions accompanied the transfer of the unique PMe Ph group. This rearrangement is reversible if the purple isomer is heated with P(OR) ligands of smaller Cone Angle, giving an orange dithioformato compound with the two smaller ligands as the equatorial cis pair, trans to the two S atoms of the S C H moiety. Reducing the steric compressions in the equatorial plane thus allows the Ru-P bonds to become favored (Figure 10). This example shows how delicate is the balance between bonds and non-bonds; how fine the line is between the electronic covalent bonding forces and the repulsions due to the bulk of inert groups. More interesting, however, is the inertness of the vacant coordination site in the purple isomer. The reason is simple: the Ru atom is protected by a phenyl ring "umbrella," which completely blocks access to the Ru atom. The R u - · * C distances exceed the sum of their van der Waals radii, but the gap is nevertheless small enough to exclude approach of either solvent or reactant molecules (Figure 11). 5

6

2

+


2

2

2

2

3

2

16-Electrons or 18-Electrons? Reduction of [{RhCl(CO)(PhO) PN(Et)P(OPh) } ] with Zn/Hg in methanol under C O gas yields the unexpected: NOT the symmetrical structure with four C O and two 18-electron rhodium atoms but [Rh (CO) {(PhO) PN(Et)P(OPh) } ], a compound with only three C O groups, all terminal (25). The crystal structure shows that one Rh atom is 5-coordinate and obeys the ΕΑΝ rule, while the other is square planar with a 16-electron configuration. The back of this Rh atom is completely shielded by a phenyl ring which blocks all access, thus preventing reaction with the C O molecules in solution. The 16-electron configuration is thus protected by a distant phenyl ring "umbrella", remote from the rhodium atom (Figure 12). 2

2

2

3

2

2

2

2

Complexes of CoCl with 4-Substituted Pyridines — Tetrahedral or Tetragonal? 2

Ethanolic solutions of compounds [CoCl (4-Rpy) ] are deep blue, but for some substituents, R, violet-pink crystals are formed, while for other substituents the crystals are deep blue (suggesting that only in these cases is the species in the crystal identical 2

2



14.

LAING

185


Figure 8. The molecule [Pt{PPh(i-But) } 0 ]. touching. 2

2

2

The two phenyl rings lie face to face, just

[RuH ( P M e P h ) I 2

5

+

Figure 9. The cation [Rur^PMe^Ph^]*, showing how the van der Waals surfaces of the hydrogen atoms on the two phenyl rings protect the H atom that is bonded to the ruthenium.


186


[Ru(PMe Ph) S CH] ORANGE 2

4

2

+

—

PMe Ph

Ph e P Me

2

~\ Me P


< h

2

,

Θ

PHP

< A

>

8 0 , v e n t

/ ι M

M

y s Ru— S'

2

PURPLE ISOMER

f

S 0 , V e n t

Ru^

:c

/I

Le Ph

Me Me

\ !>Me Ph

2

2

P(OMe)

3

PMe Ph 2

l/S

© >-

H

(OMe) P—Ru — S ' 3

(OMe) p/ I PMe Ph 3

2

Figure 10. A schematic pathway showing how reaction of the "purple" isomer with phosphite ligands of small Cone Angle will cause reversal of the reaction and thus yield the "orange" isomer. PHENYL UMBRELLA

P2 S1

P1C

"CI

Figure 11. (a) Top view and (b) Side view of the "purple" isomer of [Ru(S CH)(PMe2Ph) ] , showing how the phenyl ring bonded to P4 protects the back of the ruthenium atom. 2


4

+

14.

187


LAING

with that in solution) (26). Violet crystals are formed where R = H , C N , vinyl, or phenyl. Determination of the crystal structures gives an explanation (27, 28). In the crystals of the violet compounds the molecules are stacked to give a polymeric ladder of - C o C l C o C l - atoms. Each Co atom is bonded to four CI atoms in a square plane, with the pyridine ring perpendicular to this plane, giving the cobalt a coordination number of 6 with a tetragonal geometry. The Co · · · Co repeat distance is about 3.7 Â, which can accommodate the pyridine ring whose thickness is about 3.4 À. Thus this polymeric stacking will occur for any of these [CoCl (4-Rpy) ] complexes as long as the van der Waals diameter of the group R is 3.7 À or less. If R is larger, then it becomes impossible to stack the groups R while maintaining the infinite C o C l chain ladder. Whatever bonding advantages the chain of C o C l atoms may have in the tetragonally coordinated violet structures, it is insufficient to overcome the non-bonded repulsions between the groups R, once their diameter exceeds the critical value of about 4 À. As a result, the compound simply crystallizes as individual blue tetrahedral molecules, with the cobalt atom having the coordination number of four, which is what occurs for [CoCl (4Me-py) ] (27). Interestingly, these compounds are isostructural with the zinc analogues in which the zinc atom is always tetrahedrally coordinated in both solution and in the crystal, independent of the bulk of the group R (Figure 13). 2

2


2

2

2

2

2

2

Acetylacetonate Derivatives of Nickel(H) 8

The bonding and stereochemistry of d nickel(H) is susceptible to the bulk of groups remote from the metal atom. The anhydrous acetylacetonate is trimeric, [ N i ( A C A C ) ] , green, and paramagnetic, with each nickel atom bonded to six oxygen atoms (29) (Figure 14). If the methyl groups of the A C A C ligands are replaced by tertiary butyl groups, the anhydrous compound is red, diamagnetic, and monomelic, [Ni(i-ButACAC) ] (30) (Figure 15). It is the bulk of the tertiary butyl groups that prevents the formation of the complicated arrangement of ligands that is required for the "trimeric" structure (Figure 16). Of course, the mode of packing of the individual trimeric molecules in the crystal is exactly as predicted by Kitaigorodskii: Bump-in-Hollow. 3

6

2

Neocuproin : 2,9-Dimethyl-l,10-phenanthroline and Copper A final example of how bulk can make a ligand specific, not only to the metal but also to its oxidation state, is the case of neocuproin and copper(I). This ligand cannot coordinate with C u to give a square planar complex [CuLJ " " because of the collisions between the methyl groups of the two ligands. However, reduction of d C u to d C u immediately and quantitatively yields a colored product [CuLJ"*", in which the four Cu-N bonds are arranged in a tetrahedron with the two ligands lying in mutually perpendicular planes. In this arrangement the methyl groups no longer collide and the complex is stable (57), so much so that it can be used for the quantitative determination of copper in solution (Figure 17). 2 +

2

1

9

2 +

1 0

+



188


Figure 12. The structure of [Rh (CO)3{(PhO) PN(Et)P(OPh) } ], showing how the square planar Rh atom is protected by the phenyl ring, which blocks the approach of any potentially reactive species. 2

2

2

2

Figure 13. A schematic diagram of the crystal structure of a typical violet [CoCl (4-Rpy) ] compound. When the van der Waals diameter of the group R* is about 4 À, they will have to interpenetrate if the molecules pack in this chain-like mode. This is impossible. 2


2

14.

LAING

(CH ) C 3


189


3

^C(CH3)

V

3

HC

Figure 14. The trimeric complex [ N i ( A C A C ) ] , with the planar A C A C groups represented by the curved lines and cross-hatching. 3

6

Figure 15. The square planar molecule [Ni(f-ButACAC) ]. 2



190


Figure 16. The hypothetical trimeric isomer of [Ni(M*utACAC) ] cannot form because of collision of the tertiary butyl group on ring A with ring Β and the tertiary butyl group on ring D with ring C . 2

Figure 17. The complex [Cu(neocuproin) ] , showing the tetrahedral coordination about the Cu(I) ion and the manner in which the methyl groups are accommodated. +

2


14.

LAING


191


Conclusion The examples presented above are not "recent" but were chosen deliberately with the old Russian adage in mind: "The best of the new can often be found in the long-forgotten past, which in turn must inspire the future." From these simple examples can be learned many valuable lessons, all of which are encompassed by one simple concept: it is the bulk of the groups NOT directly bonded to the metal atom that can cause decomposition to relieve compression, can speed up reactions, and can cause inertness by blocking the reactant's pathway. Thus the non-bonded repulsions between those parts of ligands remote from the metal center can have a critical influence on the behavior of the metal atom and can control the structure and reactivity at the metal center of that coordination compound. This is the importance of Non-Bonds. Literature Cited 1. 2. 3. 4. 5.

6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.

17. 18.

Kitaigorodskii, A. I. Organic Chemical Crystallography; Consultants Bureau: New York, 1961. Laing, M . S. Afr. J. Science 1975, 71, 171-175. Kitaigorodsky, A. I. Chem. Soc. Rev. 1978, 133-163. Dauben, W. G.; Fonken, G. J.; Noyce, D. S. J. Am. Chem. Soc. 1956, 78, 2579-2582. Basolo, F.; Pearson, R. G. Mechanisms of Inorganic Reactions, 2nd ed.; John Wiley: New York, 1967; pp 371, 387, 424. Tolman, C. A. J. Am. Chem. Soc. 1970, 92, 2956-2965. Tolman, C. A. Chem. Rev. 1977, 77(3), 313-348. Ferguson, G.; Roberts, P. J.; Alyea, E. C.; Khan, M . Inorg. Chem. 1978, 17, 2965-2967. Immirzi, Α.; Musco, A. Inorg. Chim. Acta 1977, 25, L41-L42. Porzio, M . ; Musco, Α.; Immirzi, A. Inorg. Chem. 1980, 19, 2537-2540. Yamamoto, Y.; Aoki, K.; Yamazaki, H. Inorg. Chem. 1979, 18, 1683-1687. McAuliffe, C. A. In Comprehensive Coordination Chemistry; Wilkinson, G. Ed.; Pergamon: Oxford, 1987; Vol. 2, pp 1015-1029. Bartik, T.; Heimbach, P.; Schenkluhn, H. Kontakte (Merck): Darmstadt 1983(1), 16-27. Chatt, J.; Wilkins, R. G. J. Chem. Soc. 1952, 273, 4300; 1956, 525. Robertson, G. B.; Tucker, P. A. Acta Cryst. 1981, B37, 814-821. Malatesta, L . ; Ugo, R.; Cenini, S. In Werner Centennial; Kauffman, G . B . , E d . ; American Chemical Society Advances in Chemistry Series No. 62: Washington, 1967; p 318. Andrianov, V. G.; Akhrem, I. S.; Chistovalova, Ν. M . ; Struchkov, Yu. T. Russ. J. Struct. Chem. 1976, 17, 111-116. Immirzi, Α.; Musco, Α.; Mann, Β. E. Inorg. Chim. Acta 1977, 21, L37-L38.


192 19. 20. 21.


22. 23. 24. 25. 26. 27. 28. 29. 30. 31.


Otsuka, S.; Yoshida, T.; Matsumoto, M . ; Nakatsu, Κ. J. Am. Chem. Soc. 1976, 98, 5850-5857. Kashiwagi, T.; Yasuoka, N.; Kasai, N.; Kakudo, M . ; Takahashi, T.; Hagihara, N. Chem. Commun. 1969, 743. Yoshida, T.; Tatsumi, K.; Matsumoto, M . ; Nakatsu, K.; Nakamura, Α.; Fueno, T.; Otsuka, S. Nouveau J. de Chim. 1979, 3, 761-774. Ashworth, T. V.; Nolte, M . J.; Singleton, E . ; Laing, M . J. Chem. Soc. (Dalton) 1977, 1816-1822. Laing, M . Acta Cryst. 1978, B34, 2100-2104. Ashworth, T. V.; Singleton, E . ; Laing, M . Chem. Commun. 1976, 875-876. Haines, R. J.; Meintjies, E . ; Laing, M . ; Sommerville, P. J. Organometallic Chem. 1981, 216, C19-C21. Graddon, D. P.; Heng, K. B.; Watton, E. C. Aust. J. Chem. 1968, 21, 121-135. Laing, M . ; Carr, G. Acta Cryst. 1975, B31, 2683-2684. Laing, M . ; Carr, G. J. Chem. Soc. A. 1971, 1141-1144. Bullen, J. G.; Mason, R.; Pauling, P. Inorg. Chem. 1965, 4, 456. Cook, C. D.; Cheng, P.-T.; Nyburg, S. C. J. Am. Chem. Soc. 1969, 91, 2123. Cheng, K. L . ; Ueno, K.; Imamura, T. Handbook of Organic Analytical Reagents; CRC Press: New York, 1982; pp 331-333.

RECEIVED March 15, 1994


Coordination Chemistry - American Chemical Society

Recommend Documents