Rhodium - Hydrogen - Tin three-center bonds. NMR (1H, 31P, 103Rh, 119Sn) study of [Rh(Cl)(H)(SnPh3)(PPh3)(py)] and related compounds

Article abstract of DOI:10.1021/ic991054g

The complex trans-[Rh(NCBPh₃)(H)(SnPh₃)(PPh₃)₂] (1) reacts with pyridine and substituted pyridines (L) in dichloromethane at 22 °C to give [Rh(NCBPh₃)(H)(SnPh₃)(PPh₃)(L)] (2a) and at -25 °C to give trans-[Rh(NCBPh₃)(H)(SnPh₃)(PPh₃)₂(L)] (3a). These complexes and numerous analogues can be prepared (the majority in solution only) by the reactions of [Rh(X)(PPh₃)₃] (X = NCBPh₃ (a), N(CN)₂ (b), NCS (c), N₃ (d), NCO (e), O₂CCF₃ (f), Cl (g)) with Ph₃SnH in solutions containing pyridines (4-Rpy; R = CO₂Me, H, NMe₂), 1-methylimidazole (1-Meim), and benzonitriles (4-RC₆H₄CN; R = COMe, H, NMe₂) (L). NMR data for the series of complexes 2 in which ligands X, L, and, for X = Cl, the phosphine P(4-C₆H₄R)₃ (R = F, H, Me) were varied independently show systematic changes in the parameters δ(¹¹⁹Sn), δ(¹⁰³Rh), J(¹¹⁹Sn-¹H), and J(¹⁰³Rh¹¹⁹Sn), which are related to the electron-donating properties of X, L, and the phosphine. Plots of J(¹¹⁹Sn-¹H) against δ(¹¹⁹Sn), δ(¹⁰³Rh), and J(¹⁰³Rh-¹¹⁹Sn) are approximately linear and show δ(¹⁰³Rh) and J(¹⁰³Rh-¹¹⁹Sn) increasing with J(¹¹⁹Sn-¹H) and δ(¹¹⁹Sn) decreasing. Complexes 3 give higher values of J(¹¹⁹Sn-¹H) and lower values of δ(¹¹⁹Sn) than found for the less electron-rich 2, with data for 3 continuing the trends in J(¹¹⁹Sn-¹H) and δ(¹¹⁹Sn) observed for 2. Values of δ(¹⁰³Rh) and J(¹⁰³Rh-¹¹⁹Sn) for 3 do not match the pattern found for 2; nor do data for an isomeric form of 3, cis-[Rh(Cl)(H)(SnPh₃)(PR₃)₂(L)] (L = benzonitriles) (4). The plot of J(¹¹⁹Sn-¹H)/δ(¹¹⁹Sn) for 3 shows discontinuities at high values of J(¹¹⁹Sn-¹H), with the trend in δ(¹¹⁹Sn) toward more negative values (as the ligands become more nucleophilic) being transformed into an increase and changes in J(¹¹⁹Sn-¹H) becoming smaller. These patterns of NMR data are interpreted in terms of the weakening of an Rh-(H-Sn) three-center interaction and changes in the coordination geometry of tin as the electron density on rhodium is increased.

Full text of DOI:10.1021/ic991054g

4510
Inorg. Chem. 2000, 39, 4510-4519
31
Rhodium-Hydrogen-Tin Three-Center Bonds. NMR (1H, P, 103Rh, 119Sn) Study of
[Rh(Cl)(H)(SnPh3)(PPh3)(py)] and Related Compounds
Laurence Carlton
Centre for Molecular Design, Department of Chemistry, University of the Witwatersrand,
Johannesburg, Republic of South Africa
ReceiVed August 31, 1999
The complex trans-[Rh(NCBPh3)(H)(SnPh3)(PPh3)2] (1) reacts with pyridine and substituted pyridines (L) in
dichloromethane at 22 °C to give [Rh(NCBPh3)(H)(SnPh3)(PPh3)(L)] (2a) and at -25 °C to give trans-[Rh-
(NCBPh3)(H)(SnPh3)(PPh3)2(L)] (3a). These complexes and numerous analogues can be prepared (the majority
in solution only) by the reactions of [Rh(X)(PPh3)3] (X ) NCBPh3 (a), N(CN)2 (b), NCS (c), N3 (d), NCO (e),
O2CCF3 (f), Cl (g)) with Ph3SnH in solutions containing pyridines (4-Rpy; R ) CO2Me, H, NMe2),
1-methylimidazole (1-Meim), and benzonitriles (4-RC6H4CN; R ) COMe, H, NMe2) (L). NMR data for the
series of complexes 2 in which ligands X, L, and, for X ) Cl, the phosphine P(4-C6H4R)3 (R ) F, H, Me) were
varied independently show systematic changes in the parameters δ(119Sn), δ(103Rh), J(119Sn-1H), and J(103Rh-
119Sn), which are related to the electron-donating properties of X, L, and the phosphine. Plots of J(119Sn-1H)
against δ(119Sn), δ(103Rh), and J(103Rh-119Sn) are approximately linear and show δ(103Rh) and J(103Rh-119Sn)
increasing with J(119Sn-1H) and δ(119Sn) decreasing. Complexes 3 give higher values of J(119Sn-1H) and lower
values of δ(119Sn) than found for the less electron-rich 2, with data for 3 continuing the trends in J(119Sn-1H)
and δ(119Sn) observed for 2. Values of δ(103Rh) and J(103Rh-119Sn) for 3 do not match the pattern found for 2;
nor do data for an isomeric form of 3, cis-[Rh(Cl)(H)(SnPh3)(PR3)2(L)] (L ) benzonitriles) (4). The plot of
J(119Sn-1H)/δ(119Sn) for 3 shows discontinuities at high values of J(119Sn-1H), with the trend in δ(119Sn) toward
more negative values (as the ligands become more nucleophilic) being transformed into an increase and changes
in J(119Sn-1H) becoming smaller. These patterns of NMR data are interpreted in terms of the weakening of an
Rh-(H-Sn) three-center interaction and changes in the coordination geometry of tin as the electron density on
rhodium is increased.
Introduction
a change occurs in the magnitude of the carbon-hydrogen spin-
coupling constant, which falls to a value significantly lower than
that of the uncomplexed hydrocarbon but substantially higher
than that of the full oxidative addition product in which the
C-H bond is broken. Agostic bonds have since commonly been
reported on the basis of NMR characterization alone.5
Three-center bonded “agostic”1 interactions, involving a
transition metal, hydrogen, and carbon, have been the focus of
much research into the activation of C-H bonds and the search
for catalytic systems that can usefully derivatize alkanes.2
Although the characterization of such bonds has relied heavily
on diffraction studies,3,4 a very useful role for NMR spectros-
copy in the identification of three-center bonds was recognized
some years ago by Green and co-workers.1,3 On complexation
of a C-H bond to a transition metal to form an agostic bond,
Many three-center-bonded complexes, particularly those in
which the bond is unsupported, are unstable, reactive, and
(4) (a) Schultz, A. J.; Williams, J. M.; Schrock, R. R.; Rupprecht, G. A.;
Fellman, J. D. J. Am. Chem. Soc. 1979, 101, 1593. (b) Brown, R. K.;
Williams, J. M.; Schultz, A. J.; Stucky, G. D.; Ittel, S. D.; Harlow, R.
L. J. Am. Chem. Soc. 1980, 102, 981. (c) Beno, M. A.; Williams, J.
M.; Tachikawa, M.; Muetterties, E. L. J. Am. Chem. Soc. 1981, 103,
1485. (d) Schultz, A. J.; Teller, R. G.; Beno, M. A.; Williams, J. M.;
Brookhart, M.; Lamanna, W.; Humphrey, M. B. Science 1983, 220,
197. (e) Cracknell, R. B.; Orpen, A. G.; Spencer, J. L. J. Chem. Soc.,
Chem. Commun. 1984, 326. (f) Carr, N.; Dunne, B. J.; Orpen, A. G.;
Spencer, J. L. J. Chem. Soc., Chem. Commun. 1988, 926. (g) Jordan,
R. F.; Bradley, P. K.; Baenziger, N. C.; LaPointe, R. E. J. Am. Chem.
Soc. 1990, 112, 1289. (h) Feng, S. G.; White, P. S.; Templeton, J. L.
J. Am. Chem. Soc. 1990, 112, 8192. (i) Hyla-Kryspin, I.; Gleiter, R.;
Kru¨ger, C.; Zwettler, R.; Erker, G. Organometallics 1990, 9, 517. (j)
Conroy-Lewis, F. M.; Mole, L.; Redhouse, A. D.; Lister, S. A.;
Spencer, J. L. J. Chem. Soc., Chem. Commun. 1991, 1601. (k) Albe´niz,
A. C.; Schulte, G.; Crabtree, R. H. Organometallics 1992, 11, 242.
(l) Ku¨ndig, E. P.; Amurrio, D.; Bernardinelli, G.; Chowdhury, R.
Organometallics 1993, 12, 4275. (m) Kuhlman, R.; Streib, K.; Caulton,
K. G. J. Am. Chem. Soc. 1993, 115, 5813. (n) Poole, A. D.; Williams,
D. N.; Kenwright, A. M.; Gibson, V. C.; Clegg, W.; Hockless, D. C.
R.; O’Neill, P. A. Organometallics 1993, 12, 2549. (o) Braga, D.;
Grepioni, F.; Biradha, K.; Desiraju, G. R. J. Chem. Soc., Dalton Trans.
1996, 3925.
(1) Brookhart, M.; Green, M. L. H. J. Organomet. Chem. 1983, 250, 395.
(2) (a) Muetterties, E. L. Chem. Soc. ReV. 1983, 12, 283. (b) Jones, W.
D.; Feher, F. J. J. Am. Chem. Soc. 1984, 106, 1650. (c) Saillard, J.-
Y.; Hoffmann R. J. Am. Chem. Soc. 1984, 106, 2006. (d) Eisenstein,
O.; Jean, Y. J. Am. Chem. Soc. 1985, 107, 1177. (e) Crabtree, R. H.
Chem. ReV. 1985, 85, 245. (f) Lichtenberger, D. L.; Kellog, G. E. J.
Am. Chem. Soc. 1986, 108, 2560. (g) Crabtree, R. H.; Hamilton, D.
G. AdV. Organomet. Chem. 1988, 28, 299. (h) Morse, J. M.; Parker,
G. H.; Burkey, T. J. Organometallics 1989, 8, 2471. (i) Crabtree, R.
H. Angew. Chem., Int. Ed. Engl. 1993, 32, 789. (j) Song, J.; Hall, M.
B. Organometallics 1993, 12, 3118. (k) Nicholls, J. C.; Spencer, J. L.
Organometallics 1994, 13, 1781. (l) Bengali, A. A.; Schultz, R. H.;
Moore, C. B.; Bergman, R. G. J. Am. Chem. Soc. 1994, 116, 9585.
(m) Arndtsen, B. A.; Bergman, R. G.; Mobley, T. A.; Peterson, T. H.
Acc. Chem. Res. 1995, 28, 154. (n) Arndtsen, B. A.; Bergman, R. G.
Science 1995, 270, 1970. (o) Schneider, J. J. Angew. Chem., Int. Ed.
Engl. 1996, 35, 1068. (p) Hall, C.; Perutz, R. N. Chem. ReV. 1996,
96, 3125. (q) Ashbury, J. B.; Ghosh, H. N.; Yeston, J. S.; Bergman,
R. G.; Lian, T. Organometallics 1998, 17, 3417.
(3) Brookhart, M.; Green, M. L. H.; Wong, L.-L. Prog. Inorg. Chem.
1988, 36, 1.
10.1021/ic991054g CCC: $19.00 © 2000 American Chemical Society
Published on Web 09/08/2000

Rhodium-Hydrogen-Tin Three-Center Bonds
Inorganic Chemistry, Vol. 39, No. 20, 2000 4511
difficult to isolate, and for these compounds, NMR spectroscopy
may be the only readily available method by which these bonds
can be detected. An understanding of how the coupling constant
is influenced by changes (in ligands and/or coordination
geometry) occurring at the metal is likely to be of value in the
design of complexes that are effective in activating C-H bonds.
The activation of Si-H6 and Sn-H7 bonds occurs in a
manner apparently similar to that of C-H bonds but is, in many
cases, more readily observable (the presence of d orbitals
stabilizes the interaction to some extent). The decrease in J(Si-
H), on complexation of R3SiH to a transition metal, to a value
intermediate between those of free R3SiH and fully complexed
R3SiH (i.e., R3Si-M-H) has been noted by Corriu6e and
Schubert,6i and there is clear evidence that a similar relationship
holds in the case of tin.7 The advantages of working with tin
are the higher stability of its three-center-bonded complexes
relative to those of carbon and silicon and the higher natural
abundance of its NMR-active isotopes (117Sn, 119Sn).
An E-H (E ) C, Si, Sn) spin-coupling constant (potentially)
provides information of a more specific nature than that of
simply the presence or absence of a three-center bond: the
strength of a three-center bond is undoubtedly reflected in some
way by J(E-H). Here problems immediately arise, since an
NMR spin-coupling constant is a function of a number of
variables not all of which are related to the strength of the bond-
(s) between the coupled nuclei. It would be useful to know how
the value of J(E-H) responds to changes made (in, for example,
the nucleophilicity of ligands) to the coordination sphere of the
transition metal and whether the effects of changes in electron
density available to the three atoms forming the three-center
bond can be distinguished from other factors (e.g., the s character
of the bond between the coupled nuclei and the energy
difference between electronic ground and excited states) in
determining the value of J.
A step in this direction was taken in a previous study in this
area,7h in which an activation enthalpy (∼96 kJ mol-1) was
measured for the dissociation of Ph3SnH from the complex
[Rh(NCBPh3)(H)(SnPh3)(PPh3)(4-Me2Npy)], for which J(Sn-
H) ) 106 Hz. Complexes with unsubstituted pyridine and
4-MeO2Cpy in place of the more nucleophilic 4-Me2Npy gave
J(Sn-H) values of 99 and 95 Hz, respectively. The related
complex [Rh(NCBPh3)(H)(SnPh3)(PPh3)2] was shown by an
X-ray study, by observation of its thermal stability, and by its
tin-hydrogen coupling constant (J(Sn-H) ) 29 Hz) to be much
closer to the full oxidative addition limit.
The E-H coupling constant is only one of a number of NMR
parameters that are potentially accessible for a three-center-
bonded system. In the present study, chemical shifts and
1
coupling constants for the nuclei H, 31P, 103Rh, and 119Sn are
measured for over 50 compounds in an attempt, within a fairly
narrow range of rhodium chemistry, to correlate NMR data with
chemical properties that might be expected to influence the
strength of a three-center bond.
Experimental Section
Materials. [Rh(Cl)(PPh3)3],8 [Rh(Cl){P(4-C6H4F)3}3],8 [Rh2-
(Cl)2(C2H4)4],9 [Rh(NCBPh3)(PPh3)3],10a [Rh(N(CN)2)(PPh3)3],10b
[Rh(N3)(PPh3)3] (using NaN3 in place of LiN3),11 and [Rh(O2-
CCF3)(PPh3)3] (reaction in hexane rather than ethanol)12 were
prepared by published methods. [Rh(NCO)(PPh3)3] and [Rh-
(NCS)(PPh3)3] were prepared from [Rh(Cl)(PPh3)3] by metath-
esis with KOCN in ethanol or KSCN in benzene/ethanol (with
a large excess of PPh3) and shown by 15N NMR spectroscopy
to have the NCX ligand bound to rhodium via nitrogen.13 [Rh-
(Cl){P(4-C6H4Me)3}3] was prepared in situ in ∼0.04 M
concentration in dichloromethane (0.5 mL) from [Rh2(Cl)2-
(C2H4)4] (4 mg) and P(4-C6H4Me)3 (18 mg); the solution was
centrifuged to remove traces of undissolved solid. For NMR
measurements, the following were prepared in situ in 0.03-
0.04 M concentrations in ∼5% CD2Cl2/CH2Cl2 (0.5 mL): trans-
[Rh(NCBPh3)(H)(SnPh3)(PPh3)2(L)] (3a), from trans-[Rh-
(NCBPh3)(H)(SnPh3)(PPh3)2] (1) (20 mg) and L (∼0.7 M or
10% volume if liquid), and trans-[Rh(X)(H)(SnPh3)(PPh3)2(L)]
(3), from [Rh(X)(PPh3)3] (15-20 mg), L (∼0.7 M or 10%
volume), and Ph3SnH (∼10 mg, ∼0.05 M), both with cooling
in a xylene slurry at ∼ -50 °C, and [Rh(X)(H)(SnPh3)(PPh3)-
(L)] (2), from [Rh(X)(PPh3)3], L, and Ph3SnH, with cooling in
an ice/water bath. For X ) Cl, complexes with P(4-C6H4F)3
and P(4-C6H4Me)3 in place of PPh3 were prepared in a similar
way, as was cis-[Rh(Cl)(H)(SnPh3)(PR3)2(L)] (4).
(5) (a) Barreto, R. D.; Fehlner, T. P. J. Am. Chem. Soc. 1988, 110, 4471.
(b) Cox, D. N.; Roulet, R. J. Chem. Soc., Chem. Commun. 1989, 175.
(c) Bennett, M. A.; Nicholls, J. C.; Rahman, A. K. F.; Redhouse, A.
D.; Spencer, J. L.; Willis, A. C. J. Chem. Soc., Chem. Commun. 1989,
1328. (d) Brookhart, M.; Lincoln, D. M.; Volpe, A. F.; Schmidt, G.
F. Organometallics 1989, 8, 1212. (e) Erker, G.; Zwettler, R.; Kru¨ger,
C.; Hyla-Kryspin, I.; Gleiter, R. Organometallics 1990, 9, 524. (f)
Brookhart, M.; Lincoln, D. M.; Bennett, M. A.; Pelling, S. J. Am.
Chem. Soc. 1990, 112, 2691. (g) Bullock, R. M.; Lemke, F. R.; Szalda,
D. J. J. Am. Chem. Soc. 1990, 112, 3244. (h) Bower, D. K.; Keister,
J. B. Organometallics 1990, 9, 2321. (i) Etienne, M.; White, P. S.;
Templeton, J. L. J. Am. Chem. Soc. 1991, 113, 2324. (j) Ogasawara,
M.; Aoyagi, K.; Saburi, M. Organometallics 1993, 12, 3393. (k)
Spencer, J. L.; Mhinzi, G. S. J. Chem. Soc., Dalton Trans. 1995, 3819.
(l) Geftakis, S.; Ball, G. E. J. Am. Chem. Soc. 1998, 120, 9953.
(6) (a) Hart-Davis, A. J.; Graham, W. A. G. J. Am. Chem. Soc. 1971, 93,
4388. (b) Schubert, U.; Ackermann, K.; Wo¨rle, B. J. Am. Chem. Soc.
1982, 104, 7378. (c) Schubert, U.; Scholz, G.; Mu¨ller, J.; Ackermann,
K.; Wo¨rle, B.; Stansfield, R. F. D. J. Organomet. Chem. 1986, 306,
303. (d) Colomer, E.; Corriu, R. J. P.; Marzin, C.; Vioux, A. Inorg.
Chem. 1982, 21, 368. (e) Carre´, F.; Colomer, E.; Corriu, R. J. P.;
Vioux, A. Organometallics 1984, 3, 1272. (f) Graham, W. A. G. J.
Organomet. Chem. 1986, 300, 81. (g) Schubert, U.; Bahr, K.; Mu¨ller,
J. J. Organomet. Chem. 1987, 327, 357. (h) Schubert, U.; Mu¨ller, J.;
Alt, H. G. Organometallics 1987, 6, 469. (i) Schubert, U. AdV.
Organomet. Chem. 1990, 30, 151. (j) Luo, X.-L.; Kubas, G. J.; Bryan,
J. C.; Burns, C. J.; Unkefer, C. J. J. Am. Chem. Soc. 1994, 116, 10312.
(k) Luo, X.-L.; Kubas, G. J.; Burns, C. J.; Bryan, J. C.; Unkefer, C.
J. J. Am. Chem. Soc. 1995, 117, 1159. (l) Kim, Y.-J.; Lee, S.-C.; Park,
J.-I.; Osakada, K.; Choi, J.-C.; Yamamoto, T. Organometallics 1998,
17, 4929.
Dichloromethane was distilled from P2O5, and pyridine and
toluene were dried over CaH2; other solvents and reagents were
of the highest available purity and were used without further
treatment. All solutions were prepared under an argon atmo-
sphere.
NMR Spectroscopy. Spectra were recorded at 248 and 213
K on a Bruker DRX 400 spectrometer equipped with a 5 mm
triple-resonance inverse probe with a dedicated 31P channel and
(7) (a) Cardin, C. J.; Cardin, D. J.; Parge, H. E.; Power, J. M. J. Chem.
Soc., Chem. Commun. 1984, 609. (b) Schubert, U.; Kunz, E.; Harkers,
B.; Willnecker, J.; Meyer, J. J. Am. Chem. Soc. 1989, 111, 2572. (c)
Piana, H.; Kirchga¨ssner, U.; Schubert, U. Chem. Ber. 1991, 124, 743.
(d) Schubert, U.; Schubert, J. J. Organomet. Chem. 1992, 434, 169.
(e) Schubert, U.; Gilbert, S.; Mock, S. Chem. Ber. 1992, 125, 835. (f)
Carlton, L.; Weber, R. Inorg. Chem. 1993, 32, 4169. (g) Gilbert, S.;
Knorr, M.; Mock, S.; Schubert, U. J. Organomet. Chem. 1994, 480,
241. (h) Carlton, L.; Weber, R.; Levendis, D. C. Inorg. Chem. 1998,
37, 1264.
(8) Osborn, J. A.; Wilkinson, G. Inorg. Synth. 1967, 10, 67.
(9) Cramer, R. Inorg. Synth. 1974, 15, 14.
(10) (a) Carlton, L.; Weber, R. Inorg. Chem. 1996, 35, 5843. (b) Carlton,
L.; Weber, R. Magn. Reson. Chem. 1997, 35, 817.
(11) Beck, W.; Fehlhammer, W. P.; Po¨llmann, P.; Scha¨chl, H. Chem. Ber.
1969, 102, 1976.
(12) Robinson, S. D.; Uttley, M. F. J. Chem. Soc., Dalton Trans. 1973,
1912.
(13) Carlton, L. Paper in preparation.

4512 Inorganic Chemistry, Vol. 39, No. 20, 2000
Carlton
Chart 1. Complexes Formed by Reaction of 1 with L (P )
PPh3)
Scheme 1. Reactions of [RhX(PPh3)3] with L and Ph3SnH
in Dichloromethanea
extended decoupler range, operating at 400.13 MHz (1H), 161.98
MHz (31P), 12.65 MHz (103Rh), and 149.21 MHz (119Sn). Two-
dimensional 103Rh-31P spectra were obtained using the pulse
sequence π/2(31P)-1/[2J(103Rh-31P)]-π/2(103Rh)-τ-π(31P)-
τ-π/2(103Rh)-Acq(31P).14 A spectral width in f2 (31P) of 8 ppm
and an acquisition time of 0.396 s gave a digital resolution of
1.26 Hz/point; in f1 (103Rh), a spectral width of 40 ppm
(increased in a few cases, where the signal was poorly resolved,
to 100 or 200 ppm) and a time domain of 256 (reduced, in some
cases, to ∼100) gave, after zero filling, a digital resolution of
0.49 Hz/point. With a relaxation delay of 1 s and 4 scans per
increment (in some cases increased to 8 or 16), data collection
required 27 min. To eliminate the possibility of a folded signal
in f1, spectra were first recorded with a spectral width of 2000
ppm.
Chemical shifts were referenced to the generally accepted
standards of H3PO4, ¥(103Rh) ) 3.16 MHz,15 and SnMe4, with
negative values indicating shielding. The chemical shifts of H3-
PO4 (85%, 300 K) and SnMe4 (neat liquid, 248 K) with a CD2-
Cl2 external lock correspond to frequencies of 161.975 500 and
149.210 995 MHz, respectively, in a field (4.395 T) in which
the protons of TMS (in CD2Cl2 at 300 K) resonate at
400.130 020 MHz.
a (i) Complexes 2 at 0 or 22 °C: X ) NCBPh3 (a), N(CN)2 (b),
NCS (c), N3 (d), NCO (e), O2CCF3 (f), Cl/PPh3 (g), Cl/P(C6H4F)3 (g′),
Cl/P(C6H4Me)3 (g′′); L ) 1-Meim (w), 4-MeO2Cpy (x), py (y),
4-Me2Npy (z); P ) PPh3. (ii) Complexes 3 at -25 °C: X ) a, b, c, e,
g, g′, g′′; L ) x, y, z (x, y only for g-g′′), 4-MeCOC6H4CN (r),
C6H5CN (s), 4-Me2NC6H4CN (t). (iii) Complexes 4 at 0 °C: X ) g,
g′, g′′; L ) r, s, t.
2a-4a are less stable than 1, all studies being carried out on
samples prepared in situ at temperatures of 0 °C and below.
Complexes 2a-4a can also be prepared from [Rh(NCBPh3)-
(PPh3)3] and Ph3SnH in solutions containing L, again without
isolation in solid form. This method was used to prepare a series
of analogues of 2 (also of 3 and 4; see below) in which ligands
X, L, and, for X ) Cl, the phosphine were independently varied
(Scheme 1, Figure 1). The low solubility of (dimethyamino)-
pyridine in toluene precluded studies of 2fz (X ) trifluoro-
acetate). Reactions with methylimidazole (1-Meim, w) give, with
X ) N(CN)2, NCO, and N3, a second, unidentified, product in
which group X has been lost. With X ) Cl and O2CCF3, this
is the only major tin-containing product; the desired products,
2fw and 2gw, are not formed.
In solutions prepared at temperatures of -25 °C or below,
product 3 is formed (Scheme 1) in some, but not all, cases.
With L ) 4-Me2Npy, 3 is formed only with X ) NCBPh3,
N(CN)2, NCS, and NCO (complexes 3az-cz,ez). With L )
1-Meim, a product (3aw) is formed in good yield only in the
case for X ) NCBPh3 (starting from 1) where precipitation (as
a white microcrystalline powder) prevented the recording of an
119Sn spectrum. With L ) py, 3ay is isolated as a very pale
yellow powder at -50 °C using hexane. 3aw and 3ay are
obtained in slightly impure form in 60-70% yield and are stable,
in the solid state, at room temperature. The concentration of 3
can be increased by the addition of triphenylphosphine; data
for 3bz and 3cz were obtained under this condition. With X )
NCS and L ) 4-MeO2Cpy (3cx), a thick cream-colored
precipitate is slowly formed.
With the exception of complexes containing trifluoroacetate,
which, for reasons of stability, were dissolved in toluene, all
compounds were studied in solution in dichloromethane. To
allow a meaningful comparison between the data obtained using
the two solvents, a correction was applied to the trifluoroacetate
data based on measurements from [Rh(NCO)(H)(SnPh3)-
(PPh3)(py)], for which δ(119Sn) and δ(103Rh) are -132.4 and
1267 ppm, respectively, in dichloromethane and -130.7 and
1287 ppm, respectively, in toluene (all at -25 °C).
The effect on chemical shifts of varying the concentration of
L was measured for [Rh(Cl)(H)(SnPh3)(PPh3)(py)] (2gy). Upon
an increase in the concentration of pyridine from 10% to 30%,
the changes observed were δ(1H) + 0.14, δ(31P) - 0.09,
δ(103Rh) + 0.6, and δ(119Sn) - 0.05 ppm. No changes were
detectable on reducing the concentration of pyridine to 5%.
Results and Discussion
Complexes Studied. The complex trans-[Rh(NCBPh3)(H)-
(SnPh3)(PPh3)2] (1), which can be readily prepared in good
yield,10 is a precursor of a number of compounds containing
pyridine and substituted pyridines. In solution in dichlo-
romethane or toluene at room temperature, 1 reacts with L (L:
4-carbomethoxypyridine, 4-MeO2Cpy, x; pyridine, py, y; 4-(di-
methylamino)pyridine, 4-Me2Npy, z) to give 2a7h (Chart 1). In
solutions prepared at ∼ -50 °C, 3a is formed as the only
product, which, on standing at -25 °C, undergoes slow
conversion to 4a and other products.7h In solution, complexes
In spectra of 3 recorded at -25 °C, the 1H signal (Figure 1a)
shows slight to moderate broadening and the 119Sn signal severe
broadening (Figure 1d) where X ) N(CN)2, NCO, NCS, and
Cl. At -60 °C, the spectral lines are well resolved (Figure 1e).
(14) Bax, A.; Griffey, R. H.; Hawkins, B. L. J. Magn. Reson. 1983, 55,
301.
(15) Kidd, R. G.; Goodfellow, R. J. In NMR and the Periodic Table; Harris,
R. K., Mann, B. E., Eds.; Academic Press: London, 1978; pp 244-
249.

Rhodium-Hydrogen-Tin Three-Center Bonds
Inorganic Chemistry, Vol. 39, No. 20, 2000 4513
single scale exists by which to quantify the electron-donating/-
withdrawing power of the ligands and functional groups used
in the present study. The species -Me, -F, -CO2Me, -COMe,
and -NMe2 are included in the Hammett scale;16 other scales
dealing more specifically with ligands in metal complexes are
based on carbonyl stretching frequencies17 or redox potentials.18
To avoid the problems associated with quantification of this
variable, the data in the present study are presented in graphs
with NMR parameters on both axes. More specifically, J(Sn-
H) is plotted as a function of δ(119Sn), δ(103Rh), and J(103Rh-
119Sn). The electron-donating/-withdrawing dimension emerges
of its own accord from this treatment of the data.
Factors Influencing Chemical Shifts. The rhodium chemical
shift is determined principally by two terms in the Ramsey
equation:19,20 ∆E-1, where ∆E is the average energy difference
between filled and empty d orbitals, and r- 3 where r is the
d
average radius of a valence level d orbital. ∆E is influenced by
weak/strong properties of ligands and increases according to
the spectrochemical series; r is influenced by hard/soft properties
and increases according to the nephelauxetic series.20,21 Weak
hard ligands cause a shift in δ(103Rh) to high δ, and strong soft
ligands cause a shift to low δ. For weak soft ligands, where the
two effects oppose each other, and for ligands that occupy a
position toward the middle of each scale, such as pyridine, the
influence on δ(103Rh) may be difficult to predict. In general,
3
for weak ligands the r- term is dominant and for strong
ligands the ∆E-1 term is dominant.21
Subject to these constraints, a change in the nucleophilicity
of one or more of the ligands (X, L, P) attached to rhodium
would be expected to have an influence on δ(103Rh) in the
absence of any interaction with triphenyltin hydride.22 The
presence of such an interaction, which is itself sensitive to
changes in the electron density on rhodium, would further
modify the influence (of changes in X, L, and P) on δ(103Rh).
The influences on the tin chemical shift are less numerous
in the systems of the present study. The substituents on tin are
limited to rhodium, hydrogen, and three phenyl groups, the
rhodium fulfilling the role of a substituent of variable electron
density. The effect on δ(119Sn) of an increase in the electron
density on tin has been shown for the series of compounds
Me3SnCH3-nCln (n ) 0-3) to be a decrease from +85 to 0
ppm23 (the authors use the opposite sign convention) and for
[Rh(SnCl3)2(1,5-cod){P(C6H4R)3}]- a decrease from +44.1 to
1
Figure 1. 1H (a), H{31P} (b), 31P{1H) (c), 119Sn{1H} (d-f) spectra.
Spectra a-e were recorded for a mixture of [Rh(N(CN)2)(PPh3)3] (0.03
M), 4-MeO2Cpy (0.5 M), and Ph3SnH (0.05 M) in dichloromethane
(prepared at <-25 °C) giving [Rh(N(CN)2)(H)(SnPh3)(PPh3)(4-MeO2-
Cpy)] (2bx) and trans-[Rh(N(CN)2)(H)(SnPh3)(PPh3)2(4-MeO2Cpy)]
(3bx); spectra a-d were recorded at -25 °C, and spectrum e was
recorded at -60 °C. Spectrum f was recorded at -25 °C for a mixture
of [Rh(Cl){P(C6H4F)3}3] (0.03 M), 4-Me2NC6H4CN (0.5 M), and Ph3-
SnH (0.05 M) in dichloromethane giving cis-[Rh(Cl)(H)(SnPh3)-
{P(C6H4F)3}2(4-Me2NC6H4CN)] (4g′t).
+34.8 ppm as R is varied in the order F, Cl, H, OMe.24
A
change in the tin coordination number from 4 to 5 also causes
a negative shift in δ(119Sn).25
Signs of Coupling Constants. In a two-dimensional (X-Y
correlated) spectrum showing “passive” coupling to a third
nucleus (Z), the tilt of the cross-peaks provides information
about the relative signs of J(X-Z) and J(Y-Z). Where nuclei
At -60 °C, tin-hydrogen coupling constants for 3 are up to 8
Hz larger than those measured at -25 °C (no further increase
was found on going to -70 °C), indicating that ligand L is quite
labile (tin has a large trans effect), and at the higher temperature,
J(Sn-H) is a weighted average of the values for complexes
with and without ligand L.
(16) (a) Jaffe, H. H. Chem. ReV. 1953, 53, 191. (b) McDaniel, D. H.; Brown,
H. C. J. Org. Chem. 1958, 23, 420. (c) Wells, P. R. Chem. ReV. 1963,
63, 171.
(17) Tolman, C. A. Chem. ReV. 1977, 77, 313.
The reaction of [RhCl(PR3)3] with Ph3SnH in the presence
of benzonitrile gives, not 2 or 3, but 4 (Scheme 1, Figure 1f),
formed via what appears to be trans-[RhCl(H)(SnPh3)(PR3)2]
(J(Sn-H) ∼ 30 Hz). Attempts to prepare 2 and 3 with L )
PMePh2, PMe3, and P(OMe)3 from 1 with e1 equiv of L were
unsuccessful, giving mixtures of unidentified products.
NMR Parameters. A listing of NMR data for the compounds
studied is given in Table 1. The nuclei most informative about
changes in Rh-H-Sn bonding are 103Rh and 119Sn, and
discussion will be restricted to the parameters δ(103Rh),
δ(119Sn), J(119Sn-1H), and J(103Rh-119Sn). Unfortunately no
(18) Chatt, J.; Kan, C. T.; Leigh, G. J.; Pickett, C. J.; Stanley, D. R. J.
Chem. Soc., Dalton Trans. 1980, 2032.
(19) Ramsey, N. F. Phys. ReV. 1950, 78, 699; 1952, 86, 243.
(20) Mason, J. Chem. ReV. 1987, 87, 1299.
(21) Mann, B. E. In Transition Metal NMR; Pregosin, P. S., Ed.; Elsevier:
Amsterdam, 1991; pp 177-215.
(22) (a) Elsevier, C. J.; Kowall, B.; Kragten, H. Inorg. Chem. 1995, 34,
4836. (b) Carlton, L. Magn. Reson. Chem. 1997, 35, 153.
(23) Davies, A. G.; Harrison, P. G.; Kennedy, J. D.; Mitchell, T. N.;
Puddephatt, R. J.; McFarlane, W. J. Chem. Soc. C 1969, 1136.
(24) Kretschmer, M.; Pregosin, P. S.; Garralda, M. J. Organomet. Chem.
1983, 244, 175.
(25) Wrackmeyer, B. Annu. Rep. NMR Spectrosc. 1985, 16, 73.

Table 1. NMR Data
δ(1H)b
δ(31P)c
δ(103Rh)d
(
119Sn)e
213K J(P-H)f J(Rh-H)f J(Sn-H)f,g J(Rh-P)f J(Sn-P)f,h J(Rh-Sn)f
complexa
248K 213K 248K 213K 248K 213K 248K
[Rh(NCBPh3)(H)(SnPh3)(PPh3)(1-Meim)] (2aw)
[Rh(NCBPh3)(H)(SnPh3)(PPh3)(4-MeO2Cpy)] (2ax)i
[Rh(NCBPh3)(H)(SnPh3)(PPh3)(py)] (2ay)
[Rh(NCBPh3)(H)(SnPh3)(PPh3)(4-Me2Npy)] (2az)
[Rh(N(CN)2)(H)(SnPh3)(PPh3)(1-Meim)] (2bw)
[Rh(N(CN)2)(H)(SnPh3)(PPh3)(4-MeO2Cpy)] (2bx)
[Rh(N(CN)2)(H)(SnPh3)(PPh3)(py)] (2by)
[Rh(N(CN)2)(H)(SnPh3)(PPh3)(4-Me2Npy)] (2bz)
[Rh(NCS)(H)(SnPh3)(PPh3)(1-Meim)] (2cw)
[Rh(NCS)(H)(SnPh3)(PPh3)(4-MeO2Cpy)] (2cx)
[Rh(NCS)(H)(SnPh3)(PPh3)(py)] (2cy)
-16.25 -16.10 54.47 54.97 1067 1041 -119.7 -116.5
-16.34 -16.17 54.20 54.66 1171 1146 -115.9 -113.3
-16.31 -16.16 54.08 54.44 1184 1160 -120.4 -117.7
-16.39 -16.27 54.01 54.54 1204 1176 -128.9 -126.0
23.2
22.7
22.7
22.7
23.4
23.0
23.3
23.3
23.5
23.1
23.3
23.4
23.5
14.4
15.0
15.1
15.5
15.7
16.2
16.5
16.7
14.9
15.2
15.3
15.7
18.4
16.9
18.8
19.4
14.0
15.2
15.2
15.1
20.4
20.7
21.5
21.8
22.3
22.1
22.0
22.5
21.6
22.0
21.7
11.9
11.6
11.6
10.2
10.7
10.4
10.4
11.9
11.6
12.2
10.9
11.5
10.9
10.7
11.0
9.8
102
95
99
106
106
98
134.9
135.3
134.6
132.6
135.6
135.6
135.1
133.6
136.0
136.0
135.4
133.8
137.9
137.7
137.8
136.2
137.0
136.8
136.5
135.0
141.2
140.0
137.0
137.5
135.7
139.4
138.5
136.9
136.2
136.2
134.8
107.6
107.7
107.8
105.2
106.1
106.0
105.8
106.1
106.1
105.9
106.0
106.1
105.8
106.7
106.7
106.6
107.4
107.5
108.4
224
205
211
220
232
214
220
229
237
216
224
232
244
230
232
241
250
229
235
244
225
230
230
237
244
226
234
242
232
238
245
143
145
146
154
152
153
157
156
158
163
161
163
172
168
169
173
172
168
167
356
355
361
367
365
355
359
367
369
356
363
370
381
367
370
381
379
370
371
382
370
378
370
375
384
361
367
375
376
382
390
336
337
340
-17.01
55.17
1094
-126.0
-17.02 -16.95 54.58 55.25 1203 1172 -120.1 -116.6
-17.03 -16.95 54.65 55.26 1212 1182 -124.3 -120.9
-17.07 -16.97 54.48 55.29 1231 1195 -131.6 -127.4
101.5
107
107
100
103
109
114
104.5
108
114
112
105
108.5
115.5
110
114
100
105
113
97
-16.92
55.11
1108
-125.7
-16.92 -16.88 54.47 55.10 1219 1189 -120.4 -117.5
-16.94 -16.88 54.61 55.21 1226 1195 -124.5 -121.0
-17.02 -16.94 54.48 55.38 1243 1207 -132.1 -128.0
[Rh(NCS)(H)(SnPh3)(PPh3)(4-Me2Npy)] (2cz)
[Rh(N3)(H)(SnPh3)(PPh3)(1-Meim)] (2dw)
[Rh(N3)(H)(SnPh3)(PPh3)(4-MeO2Cpy)] (2dx)
[Rh(N3)(H)(SnPh3)(PPh3)(py)] (2dy)
-17.80
-17.59
-17.68
-17.91
55.18
55.60
55.72
55.34
1196
1267
1281
1311
-136.3
-129.2
-134.1
-142.6
∼23
23.0
23.3
21.2
22.3
22.5
22.2
22.9
24.8
22.2
22.3
22.6
22.1
22.5
22.8
22.6
22.4
22.7
10.8
10.9
10.5
13.0
12.2
12.3
11.9
11.8
11.9
11.2
11.7
12.6
11.0
11.7
12.0
12.0
12.6
12.7
13.0
[Rh(N3)(H)(SnPh3)(PPh3)(4-Me2Npy)] (2dz)
[Rh(NCO)(H)(SnPh3)(PPh3)(1-Meim)] (2ew)
[Rh(NCO)(H)(SnPh3)(PPh3)(4-MeO2Cpy)] (2ex)
[Rh(NCO)(H)(SnPh3)(PPh3)(py)] (2ey)
[Rh(NCO)(H)(SnPh3)(PPh3)(4-Me2Npy)] (2ez)
[Rh(O2CCF3)(H)(SnPh3)(PPh3)(4-MeO2Cpy)] (2fx)j
[Rh(O2CCF3)(H)(SnPh3)(PPh3)(py)] (2fy)j
[Rh(Cl)(H)(SnPh3)(PPh3)(4-MeO2Cpy)] (2gx)
[Rh(Cl)(H)(SnPh3)(PPh3)(py)] (2gy)
[Rh(Cl)(H)(SnPh3)(PPh3)(4-Me2Npy)] (2gz)
-17.18 -17.10 55.68 56.35 1162 1130 -134.6 -130.4
-17.11 -17.08 55.66 56.30 1258 1226 -128.2 -125.0
-17.15 -17.12 55.71 56.32 1267 1236 -132.4 -129.0
-17.26 -17.22 55.38 56.19 1290 1253 -141.0 -137.2
-20.48
-20.53
54.36
54.84
1405
1411
-137.7
-143.1
-17.21 -17.26 53.79 54.47 1262 1234 -137.6 -135.1
-17.35 -17.36 53.76 54.36 1277 1250 -142.7 -139.9
-17.65 -17.65 52.98 53.56 1320 1292 -153.5 -150.9
-17.23 -17.31 51.85 52.52 1248 1220 -132.6 -130.2
-17.37 -17.43 52.02 52.60 1261 1234 -137.0 -134.3
-17.69 -17.75 51.64 52.27 1297 1267 -147.5 -144.6
-17.35 -17.43 52.24 53.01 1277 1249 -140.5 -138.4
-17.49 -17.52 52.21 52.92 1290 1263 -145.6 -143.1
-17.79 -17.81 51.54 52.32 1331 1300 -155.9 -153.3
[Rh(Cl)(H)(SnPh3){P(C6H4F)3}(4-MeO2Cpy)] (2g′x)
[Rh(Cl)(H)(SnPh3){P(C6H4F)3}(py)] (2g′y)
101
109
102
106
113
96
[Rh(Cl)(H)(SnPh3){P(C6H4F)3}(4-Me2Npy)] (2g′z)
[Rh(Cl)(H)(SnPh3){P(C6H4Me)3}(4-MeO2Cpy)] (2g′′x)
[Rh(Cl)(H)(SnPh3){P(C6H4Me)3}(py)] (2g′′y)
[Rh(Cl)(H)(SnPh3){P(C6H4Me)3](4-Me2Npy)] (2g′′z)
trans-[Rh(NCBPh3)(H)(SnPh3)(PPh3)2(4-MeCOC6H4CN)] (3ar)
trans-[Rh(NCBPh3)(H)(SnPh3)(PPh3)2(C6H5CN)] (3as)
trans-[Rh(NCBPh3)(H)(SnPh3)(PPh3)2(4-Me2NC6H4CN)] (3at)
trans-[Rh(NCBPh3)(H)(SnPh3)(PPh3)2(1-Meim)] (3aw)
trans-[Rh(NCBPh3)(H)(SnPh3)(PPh3)2(4-MeO2Cpy)] (3ax)
trans-[Rh(NCBPh3)(H)(SnPh3)(PPh3)2(py)] (3ay)
trans-[Rh(NCBPh3)(H)(SnPh3)(PPh3)2(4-Me2Npy)] (3az)
trans-[Rh(N(CN)2)(H)(SnPh3)(PPh3)2(4-MeO2Cpy)] (3bx)
trans-[Rh(N(CN)2)(H)(SnPh3)(PPh3)2(py)] (3by)
trans-[Rh(N(CN)2)(H)(SnPh3)(PPh3)2(4-Me2Npy)] (3bz)
trans-[Rh(NCS)(H)(SnPh3)(PPh3)2(4-MeO2Cpy)] (3cx)
trans-[Rh(NCS)(H)(SnPh3)(PPh3)2(py)] (3cy)
trans-[Rh(NCS)(H)(SnPh3)(PPh3)2(4-Me2Npy)] (3cz)
trans-[Rh(NCO)(H)(SnPh3)(PPh3)2(4-MeO2Cpy)] (3ex)
trans-[Rh(NCO)(H)(SnPh3)(PPh3)2(py)] (3ey)
trans-[Rh(NCO)(H)(SnPh3)(PPh3)2(4-Me2Npy)] (3ez)
trans-[Rh(Cl)(H)(SnPh3)(PPh3)2(4-MeO2Cpy)] (3gx)
trans-[Rh(Cl)(H)(SnPh3)(PPh3)2(py)] (3gy)
-15.46
-15.61
-15.79
40.60
40.43
40.46
422
430
451
-115.9
-115.9
-117.9
97
99
-16.30
40.31
454
124
121
124
126
122
126
128
124
128
130
130
134
136
128
133
122
-16.26
-16.34
-16.31
-16.27
-16.30
-16.28
-16.21
-16.23
-16.24
-16.44
-16.48
-16.48
-16.65
-16.69
-16.65
39.18
39.67
40.40
38.99
39.47
40.21
38.90
39.41
40.19
38.62
39.13
39.85
37.83
38.34
35.46
481
481
488
500
501
508
512
512
518
561
563
570
573
578
576
-145.2
-147.3
-147.2
-142.5
-144.8
-144.0
-142.6
-144.5
-143.5
-150.0
-152.1
-151.7
-163.7
-166.3
-157.3
337
336
336
342
340
341
346
344
342
355
352
354
342
351
347
17.6
17.7
18.3
trans-[Rh(Cl)(H)(SnPh3){P(C6H4F)3}2(4-MeO2Cpy)] (3g′x)

Rhodium-Hydrogen-Tin Three-Center Bonds
Inorganic Chemistry, Vol. 39, No. 20, 2000 4515
X and Y both have positive gyromagnetic ratios (γ), a positive
tilt (from lower left to upper right, the high-δ peaks in the X
dimension correlating with the high-δ peaks in the Y dimension)
indicates that J(X-Z) and J(Y-Z) have the same sign. In the
case where γX and γY have opposite signs, the situation is
complicated by the fact that the precessional frequencies of X
and Y also have opposite signs, a difference which is not
reflected in the chemical shift scale. In a two-dimensional X-Y
correlated spectrum obtained from nuclei having opposite signs
of γ, a positive tilt of cross-peaks will indicate that J(X-Z)
and J(Y-Z) have opposite signs. The signs of NMR frequencies
and phases in relation to γ have been discussed by Levitt,26
and the determination of the signs of coupling constants in
transition metal complexes using 2D spectra has been described
by Otting et al.27
The nuclei observed in the 2D spectra shown in Figure 2 are
1H and 31P (positive γ) and 119Sn and 103Rh (negative γ).
Accordingly, the positive tilt of the cross-peaks in Figure 2a
(
119Sn-1H) indicates that J(103Rh-119Sn) and J(103Rh-1H) have
opposite signs as do J(119Sn-31P) and J(31P-1H); the positive
tilt in Figure 2b (31P-1H) indicates that J(103Rh-31P) and
J(103Rh-1H) have the same sign as do J(119Sn-31P) and
J(119Sn-1H); the negative tilt in Figure 2c (103Rh-31P) indicates
that J(103Rh-119Sn) and J(119Sn-31P) also have the same sign
as do J(103Rh-1H) and J(31P-1H). The signs of both
1J(103Rh-1H) and 1J(103Rh-31P) have been measured as nega-
tive.28 From this and the data above, it is most likely that, for
complexes
2
and 3, 1J(103Rh-1H), 1J(103Rh-31P), and
2J(31P-1Hcis) are negative and 2J(119Sn-1Hcis), 2J(119Sn-31Pcis),
1
and J(103Rh-119Sn) are positive.
J(119Sn-1H) as a Function of δ(119Sn). For complexes 2,
changes occur in J(119Sn-1H) and δ(119Sn) as ligand X is
varied: these changes are shown in Figure 3 for pyridine-
containing derivatives. Data points for complexes 2 define a
gradient with J(119Sn-1H) increasing and δ(119Sn) decreasing,
the sequence of points from lower left to upper right matching
the order of increasing electron-donating ability of X. Data for
2, X ) Cl, lie ∼ 14 ppm toward lower δ(119Sn) values than the
data obtained with X ) N-donor. A weak interaction between
Cl and Sn could account for this difference. With L ) 1-Meim
and varied X, the data points for 2 give a line (not shown; the
Meim data are described briefly elsewhere29), of very similar
gradient, lying ∼6 ppm toward higher δ(119Sn) values than the
line for X ) NCBPh3 etc. A change of the phosphine (X ) Cl
only) has an effect similar to that of changes in X, as shown
also in Figure 3, where, with X ) Cl, L ) py, and varied
phosphine (complex 2), the three data points in order of
increasing phosphine nucleophilicity lie approximately on a line
parallel to that for varied X.
When the effects of varying ligand L are superimposed on
this picture, lines shown in the lower portion of Figure 4b are
obtained. With ligands X ) NCBPh3, N(CN)2, NCS, N3, and
NCO, these lines are, to a good approximation, straight, having
a gradient very similar, but not identical, to the gradient in Figure
3. With X ) Cl, the lines show slight curvature but otherwise
resemble the lines obtained with X ) NCBPh3, N(CN)2, etc. In
(26) Levitt, M. H. J. Magn. Reson. 1997, 126, 164.
(27) Otting, G.; Soler, L. P.; Messerle, B. A. J. Magn. Reson. 1999, 137,
413.
(28) (a) Goggin, P. L.; Goodfellow, R. J.; Knight, J. R.; Norton, M. G.;
Taylor, B. F. J. Chem. Soc., Dalton Trans. 1973, 220. (b) Hyde, M.
E.; Kennedy, J. D.; Shaw, B. L.; McFarlane, W. J. Chem. Soc., Dalton
Trans. 1977, 1571.
(29) Carlton, L. Appl. Organomet. Chem., in press.

4516 Inorganic Chemistry, Vol. 39, No. 20, 2000
Carlton
Figure 3. Plot of J(119Sn-1H) vs δ(119Sn) for the complexes [Rh(X)-
(H)(SnPh3)(PPh3)(L)] (2) with L ) py and with X and phosphine (X
) Cl only) varied: X ) NCBPh3 (a), N(CN)2 (b), NCS (c), N3 (d),
NCO (e), O2CCF3 (f), Cl/PPh3 (g), Cl/P(C6H4F)3 (g′), Cl/P(C6H4Me)3
(g′′). Complexes were prepared in situ; the solvent was dichloromethane
except for 2f (toluene) where δ(Sn) is corrected. Data were recorded
at -25 °C.
leads to a steeper J(119Sn-1H)/δ(119Sn) gradient as compared
to that for 2 with the trend toward lower δ(119Sn) values that
accompanies increasing nucleophilicity of L reversed on ex-
change of 4-Me2Npy for py. For a given ligand X, the data
points for 3 (L ) 4-MeO2Cpy) lie on, or close to, the projection
of the gradient for the relevant complexes 2. The data for
complexes 3ar-t (X ) NCBPh3, L ) benzonitriles) shown in
Figure 4b closely match those for 2ax-z (X ) NCBPh3, L )
pyridines). However, upon variation of ligand X (work currently
in progress), this apparent accord is lost.
The general trend in δ(119Sn) for complexes 2 and 3 (Figure
3) is reversed for 4 (not shown), which gives signals with δ
values ∼150 ppm higher than those for 2 or 3. As the
nucleophilicity of L (benzonitrile) is increased, both J(119Sn-
1H) and δ(119Sn) increase. The effects of varying the phosphine
are less clear. The similarities between the data for 2 and 3
clearly do not extend to 4.
J(119Sn-1H) as a Function of δ(103Rh). A plot of J(119Sn-
1H) vs δ(103Rh) is shown in Figure 5a for complex 2 with L )
py and X varied. With L ) Meim and X varied, a line (not
shown) of similar gradient, lying ∼150 ppm toward lower δ
values, is obtained. The data point for the chloro complex (2gy)
lies close to the line defined by the other X variants, in contrast
to the findings shown in Figure 3 for δ(119Sn) data. Data for 3
(L ) py, X varied; Figure 5b) give a line approximately parallel
to that found for 2 (Figure 5a), indicating very similar responses
to changes in X for the two series of compounds.
The rhodium chemical shifts for 2 are shown in Figure 5c
(where the δ(103Rh) scale has been expanded relative to that in
Figure 5a) for each ligand X, grouped according to the variation
in pyridine. The gradient J(119Sn-1H)/δ(103Rh) associated with
changes in pyridine is much steeper (i.e., a smaller change in
δ(103Rh) relative to J(119Sn-1H)) than that associated with
changes in X, in contrast to the J(119Sn-1H)/δ(119Sn) data
(Figure 3), where the effects of changing X and changing
pyridine are very similar. The differing effects of X and L on
Figure 2. 119Sn-1H (a), 31P-1H (b) and 103Rh-31P (c) correlated
spectra for trans-[Rh(NCBPh3)(H)(SnPh3)(PPh3)2(4-Me2Npy)] (3az)
(a and c) at -60 °C and for [Rh(NCBPh3)(H)(SnPh3)(PPh3)(py)] (2ay)
(b) at -25 °C, showing passive coupling. The solvent was CD2Cl2.
The main signal (center section) of spectrum c is shown at a much
higher contour level than the Sn satellites. 1D spectra are external
projections.
every case, the sequence in which the data points are ordered
(from lower to upper) is from the least to the most nucleophilic
pyridine.
When data (J(119Sn-1H)/δ(119Sn)) from the six-coordinate
bis(phosphine) complexes 3 are included, lines in the upper
portion of Figure 4b are obtained. Complexes 3 give higher
J(119Sn-1H) and lower (i.e., more negative) δ(119Sn) values than
2. Variation of ligand X and phosphine (X ) Cl) causes a greater
dispersion of data points in the δ(119Sn) dimension for 3 (as
compared to 2), and there is no clear relationship between J(119
-
Sn-1H)/δ(119Sn) and the electron-donating properties of X, in
contrast to what is found for 2. Variation of ligand L (pyridine)

Rhodium-Hydrogen-Tin Three-Center Bonds
Inorganic Chemistry, Vol. 39, No. 20, 2000 4517
Figure 4. (a) Plot of J(119Sn-1H) vs δ(119Sn) for the complexes [Rh(X)(H)(SnPh3)(PPh3)(L)] (2) and trans-[Rh(X)(H)(SnPh3)(PPh3)2(L)] (3) with
X, phosphine (X ) Cl only), and L varied: X ) NCBPh3 (a), N(CN)2 (b), NCS (c), NCO (e), Cl/PPh3 (g), Cl/P(C6H4F)3 (g′), CP/P(C6H4Me)3 (g′′);
L ) 4-MeO2Cpy (x), py (y), 4-Me2Npy (z), 4-MeCOC6H4CN (r), C6H5CN (s), 4-Me2NC6H4CN (t). Complexes were prepared in situ; the solvent
was dichloromethane. Data were recorded at -60 °C. (b, Inset) Plot of J(119Sn-1H) vs δ(119Sn) for complexes 2a and 3a (dichloromethane, -60
°C).
(pyridines) are very small, amounting to no more than 0-2 ppm
on replacing 4-MeO2Cpy by py (X ) NCBPh3, N(CN)2, NCS,
NCO) and 6-7 ppm on replacing py by 4-Me2Npy (Figure 6a).
Much larger changes in δ(103Rh) are found for 3ar-t
(X ) NCBPh3, L ) benzonitriles) on varying L (Figure 6b).
J(119Sn-1H) as a Function of J(103Rh-119Sn). A graph
showing variations in J(119Sn-1H) and J(103Rh-119Sn) recorded
for 2 and 3 in response to changes in ligand X (L ) py; X
varied) is given in Figure 7. The linear relationship holds also
for 2 with L ) 1-Meim and X varied but is rather less clear for
L ) MeO2Cpy and Me2Npy. Plots (not shown) of J(119Sn-1H)
vs J(103Rh-119Sn) for 2 with L varied show gradients similar
to those found with X varied. For 3, the effects due to variations
in L form no clear pattern.
Figure 5. (a) Plot of J(119Sn-1H) vs δ(103Rh) for the complexes [Rh-
NMR Data and Rh-H-Sn Bonding. Three features of the
(X)(H)(SnPh3)(PPh3)(L)] (2) with L ) py and X ) NCBPh3 (a), N(CN)2
data from plots of J(119Sn-1H) vs δ(119Sn) and δ(103Rh) are
(b), NCS (c), N3 (d), NCO (e), O2CCF3 (f). Data were recorded at
-25 °C. (b) Plot of J(119Sn-1H) vs δ(103Rh) for the complexes trans-
[Rh(X)(H)(SnPh3)(PPh3)2(L)] (3) with L ) py and X varied. Data were
recorded at -60 °C. (c) Plot of J(119Sn-1H) vs δ(103Rh) for 2 with X,
phosphine (X ) Cl only), and L varied: L ) 4-MeO2Cpy (x), py (y),
4-Me2Npy (z). Data were recorded at -60 °C. The solvent was
dichloromethane except for 2f (toluene) where δ(Rh) is corrected. All
complexes were prepared in situ.
particularly relevant to an interpretation of the Rh-H-Sn
bonding.
(1) The response of δ(119Sn) to changes in complexes 2 and
3: A near-linear decrease in δ(119Sn) (Figure 4a) as the electron
density on 2 is increased leads, for the more electron-rich 3, to
a reversal of the direction of change of δ(119Sn).
(2) The response of δ(103Rh) to changes in complex 3: When
ligand L has low nucleophilicity (benzonitrile), variations in
electron-donating ability cause fairly large changes in δ(103Rh);
δ(103Rh) are likely to reflect differing influences on ∆E and
r-3 (see above). Changes in δ(103Rh) for 3 on varying L
d

4518 Inorganic Chemistry, Vol. 39, No. 20, 2000
Carlton
in which the bond between tin and hydrogen no longer exists.
The direction of movement along the trajectory defines a process
as being one of oxidative addition or reductive elimination. In
a finely balanced system, the three-center bond will be stable.
On moving along the trajectory, the tin atom, in a complex
containing Rh, H, and Sn, will undergo a change in geometry
from four-coordinate (in the full oxidative addition product) to
five-coordinate as the interaction between Sn and H becomes
stronger and, as the Rh-Sn and Rh-H bonds become weaker,
will then return to four-coordinate geometry (free R3SnH). The
results described in (1), above, agree with this. The tin chemical
shift is sensitive to Sn geometry, becoming more negative as
the coordination number is changed from 4 to 5.25 However,
the facts that the reversal of the direction of change of δ(119Sn)
is observed only in 3 with L ) Me2Npy and that this change
occurs at values of J(119Sn-1H) that differ by as much as 10
Hz (Figure 4a) might suggest that the 4-Me2Npy ligand itself,
rather than the increased electron density that it conveys to the
rhodium, is in some way responsible for the effect. Evidence
against such an interpretation is found in the results for 2
(Figures 3 and 4a), which show no anomalies for complexes of
4-Me2Npy. Studies involving a wider range of pyridines might
help to clarify the situation.
Figure 6. (a) Plot of J(119Sn-1H) vs δ(103Rh) for the complexes trans-
[Rh(X)(H)(SnPh3)(PPh3)2(L)] (3) with X and L varied: X ) NCBPh3
(a), N(CN)2 (b), NCS (c), NCO (e); L ) 4-MeO2Cpy (x), py (y), 4-Me2-
Npy (z). The complexes were prepared in situ; the solvent was
dichloromethane. Data were recorded at -60 °C. Dotted line: X varied.
Solid line: L varied. (b, Inset) Plot of J(119Sn-1H) vs δ(103Rh) for 3a:
L ) 4-MeCOC6H4CN (r), C6H5CN (s), 4-Me2NC6H4CN (t). The solvent
was dichloromethane, and data were recorded at -60 °C.
The results described in (2) and (3), above, can be rationalized
in terms of a transfer of electron density from rhodium to tin
and hydrogen. As the electron density on rhodium is increased,
a progressively larger fraction is transferred to Sn and H, which
together act as an acceptor of electron density that is not readily
taken up elsewhere. When the source of the increased electron
density is the ligand positioned trans to tin (as in complex 3),
this transfer should be particularly efficient. If a buildup of
electron density on rhodium is avoided in this way, the fact
should be reflected in the rhodium chemical shift, which, as is
observed, should then be relatively insensitive to such changes.
With lower overall electron density on rhodium, any increase
should become more evenly distributed and the influence on
δ(103Rh) should be proportionately stronger (as observed).
Electron density transferred to tin can strengthen the Sn-H
interaction and increase the contribution of 1J(119Sn-1H) to the
1
observed coupling constant. J(119Sn-1H) has a negative sign
2
and J(119Sn-1H) has a positive sign; thus any increase in the
1J component will subtract from changes in 2J, causing
2J(119Sn-1H) to appear smaller than might otherwise be
expected. An effect matching this is observed for the most
electron-rich complexes, 3, with L ) 4-Me2Npy.
Interpretation of J(119Sn-1H). The value of J(119Sn-1H)
correlates with the electron density on rhodium. With increasing
J(119Sn-1H), ligands X appear in the order NCBPh3, N(CN)2,
NCS, Cl, N3. NCO, O2CCF3, ligands L in the order 4-MeO2-
Cpy, py, 4-Me2Npy, and phosphine ligands in the order P(4-
C6H4F)3, PPh3, P(4-C6H4Me)3, i.e., according to their electron
donating abilities.
Figure 7. Plot of J(119Sn-1H) vs J(103Rh-119Sn) for the complexes
[Rh(X)(H)(SnPh3)(PPh3)(L)] (2) and trans-[Rh(X)(H)(SnPh3)(PPh3)2-
(L)] (3) with L ) py and X varied: X ) NCBPh3 (a), N(CN)2 (b),
NCS (c), N3 (d), NCO (e), O2CCF3 (f). The complexes were prepared
in situ; the solvent was dichloromethane (toluene for f). Data for 2
were recorded at -25 °C; data for 3, at -60 °C.
As 3 is made increasingly electron-rich, so it becomes more
difficult to prepare (in situ), the yield diminishing until, with
highly electron-donating combinations of ligands, no product
3 is formed. For 3, the highest observed value of J(119Sn-1H)
is 136 Hz (3cz); this appears to represent a limiting region
beyond which the complexes (at -60 °C) are unstable.
when L has higher nucleophilicity (pyridine), similar variations
cause much smaller changes in δ(103Rh) (Figure 6).
(3) The response of J(119Sn-1H) to changes in complexes 2
and 3: When py is replaced by 4-Me2Npy in complex 2 the
average increase in J(119Sn-1H) is ∼7 Hz; when py is replaced
by 4-Me2Npy in complex 3, the average increase in J(119Sn-
1H) is only 2 Hz. The corresponding increases in J(119Sn-1H)
on exchanging 4-MeO2Cpy for py are 3-4 Hz for both 2 and
3.
Conclusion
The parameters that are most informative regarding potential
influences on the three-center bond in complexes 2 and 3,
A three-center bond can be visualized as lying on a trajec-
tory30 leading from the individual starting compounds, here Ph3-
SnH and a rhodium complex, to an oxidative addition product
(30) Crabtree, R. H.; Holt, E. M.; Lavin, M.; Morehouse, S. M. Inorg.
Chem. 1985, 24, 1986.

Rhodium-Hydrogen-Tin Three-Center Bonds
Inorganic Chemistry, Vol. 39, No. 20, 2000 4519
J(119Sn-1H), δ(119Sn) and δ(103Rh), are insensitive to a variety
of changes in ligands, suggesting that, within limits, these
changes do not significantly alter either the electronic balance
of the bond (J(Sn-H), δ(Sn)) or the electronic environment of
rhodium (δ(Rh)). This can be seen for complexes 2 with X an
reversal of the direction of change of δ(119Sn) for complexes
having the most nucleophilic pyridine is a predictable conse-
quence of a transition from five- to four-coordination for tin.
For 3, the diminishing responses of 2J(119Sn-1H) to variations
in L are likely to reflect increased contributions from
1J(119Sn-1H) as the Sn-H bond becomes stronger.
N-donor ligand and L a pyridine, where the responses of J(119
-
Sn-1H) and δ(119Sn) to variations in X and L are largely
independent of whether X or L is varied. For complexes 3a-
c,e, where the effects on δ(103Rh) of variations in L (pyridines)
are minimal, the changes caused by varying L appear to be
transmitted almost entirely to Sn and H. For the most electron-
rich complexes (3), the data show trends consistent with events
that are expected to occur close to the dissociation limit. The
Acknowledgment. Thanks are expressed to the University
of the Witwatersrand and the NRF for financial support and to
Professor J. Jeener and Professor M. H. Levitt as well as an
anonymous reviewer for advice concerning signs of NMR data.
IC991054G