Solid-state 115In and 31P NMR studies of triarylphosphine indium trihalide adducts

Published on Web 03/29/2010

Solid-State ¹¹⁵In and ³¹P NMR Studies of Triarylphosphine

Indium Trihalide Adducts

Fu Chen, Guibin Ma, Guy M. Bernard, Ronald G. Cavell, Robert McDonald,

Michael J. Ferguson, and Roderick E. Wasylishen*

Department of Chemistry, Gunning/Lemieux Chemistry Centre, UniVersity of Alberta, Edmonton,

Alberta, Canada T6G 2G2

Received January 23, 2010; E-mail: Roderick.Wasylishen@ualberta.ca

Abstract: Solid-state ¹¹⁵In and ³¹P NMR spectroscopy, relativistic density functional theory (DFT)

calculations, and single-crystal X-ray diffraction were used to investigate a series of triarylphosphine

indium(III) trihalide adducts, X₃In(PR₃) and X₃In(PR₃)₂(X ) Cl, Br or I; PR₃) triarylphosphine ligand). The

electric ﬁeld gradient tensors at indium as well as the indium and phosphorus magnetic shielding tensors

and the direct and indirect ¹¹⁵In-³¹P spin-spin coupling were characterized; for complexes possessing a

C₃symmetry axis, the anisotropy in the indirect spin-spin coupling, ∆J(¹¹⁵In,³¹P), was also determined.

The ¹¹⁵In quadrupolar coupling constants, C_Q(¹¹⁵In), range from (1.25 ( 0.10 to -166.0 ( 2.0 MHz. For

any given phosphine ligand, the indium nuclei are most shielded for X ) I and least shielded for X ) Cl,

a trend also observed for other group-13 nuclei in M(III) complexes. This experimental trend, attributed to

spin-orbit effects of the halogen ligands, is reproduced by the DFT calculations. The spans of the indium

magnetic shielding tensors for these complexes, δ₁₁- δ₃₃, range from 40 ( 7 to 710 ( 60 ppm; those

1

determined for phosphorus range from 28 ( 1.5 to 50 ( 3 ppm. Values of J(¹¹⁵In,³¹P) range from 550 (

20 to 2500 ( 20 Hz. For any given halide, the ¹J(¹¹⁵In,³¹P) values generally increase with increasing basicity

1

of the PR₃ligand. Calculated values of J(¹¹⁵In,³¹P) and ∆J(¹¹⁵In,³¹P) duplicate experimental trends and

indicate that both the Fermi-contact and spin-dipolar Fermi-contact mechanisms make important

1

contributions to the J(¹¹⁵In,³¹P) tensors.

Chart 1

Introduction

Indium-phosphorus compounds have become important for

numerous disciplines^1,2including materials chemistry.^3-5For

example, semiconductors such as InP have attracted much

attention because of their importance in applications such as

light-emitting diodes.¹The adducts formed by indium trihalide

or indium trialkyl moieties with phosphine ligands, which are

classical Lewis acid-base complexes, are single-source precur-

sors for the preparation of a wide range of InP-based semicon-

ductors.⁴The adducts of indium trihalides with triarylphosphine

(1) (a) O’Brien, P.; Pickett, N. L. Coordination Complexes as Precursors

for Semiconductor Films and Nanoparticles. In ComprehensiVe

Coordination Chemistry II; McCleverty, J. A., Meyer, T. J., Eds.;

Elsevier: Amsterdam, 2004; Vol. 9, Chapter 23, pp 1005-1063. (b)

Grant, I. R. III-V Compounds. In Chemistry of Aluminum, Gallium,

Indium and Thallium; Downs, A. J., Ed.; Blackie: London; 1993;

Chapter 5, pp 292-321.

ligands, ﬁrst prepared by Carty and Tuck over four decades

ago,^6,7possess either approximate tetrahedral X₃In(PR₃) or

trigonal-bipyramidal X₃In(PR₃)₂(X ) Cl, Br or I; PR₃)

triarylphosphine) structures (Chart 1),⁸depending on the bulki-

ness of both the halides and the triarylphosphine ligands.⁸

Sterically demanding substituents generally lead to 1:1 adducts

with approximately tetrahedral structures.

(2) (a) Trost, B. M.; Sharma, S.; Schmidt, T. J. Am. Chem. Soc. 1992,

114, 7903–7904. (b) Trost, B. M.; Sharma, S.; Schmidt, T. Tetrahedron

Lett. 1993, 34, 7183–7186.

(3) van Weert, M. H. M.; Helman, A.; van den Einden, W.; Algra, R. E.;

Verheijen, M. A.; Borgstro¨m, M. T.; Immink, G.; Kelly, J. J.;

Kouwenhoven, L. P.; Bakkers, E. P. A. M. J. Am. Chem. Soc. 2009,

131, 4578–4579.

(4) (a) Wells, R. L.; Aubuchon, S. R.; Kher, S. S.; Lube, M. S. Chem.

Mater. 1995, 7, 793–800. (b) Briand, G. G.; Davidson, R. J.; Decken,

A. Inorg. Chem. 2005, 44, 9914–9920. (c) Malik, M. A.; O’Brien, P.

Top. Organomet. Chem. 2005, 9, 173–204.

(6) Carty, A. J.; Tuck, D. G. J. Chem. Soc. A 1966, 1081–1087.

(7) Carty, A. J. Can. J. Chem. 1967, 45, 345–351.

(5) (a) Tuck, D. G. Indium and Thallium. In ComprehensiVe Coordination

Chemistry; Wilkinson, G., Gillard, R. D., McCleverty, J. A., Eds.;

Pergamon: Oxford, 1987; Vol. 3, Chapter 25.2, pp 153-182. (b) Carty,

A. J.; Tuck, D. G. Prog. Inorg. Chem. 1975, 19, 243–337.

(8) Rasika Dias, H. V. Indium and Thallium. In ComprehensiVe Coordina-

tion Chemistry II; McCleverty, J. A., Meyer, T. J., Eds.; Elsevier:

Amsterdam, 2004; Vol. 3, pp 383-463.

9

10.1021/ja100625p  2010 American Chemical Society

J. AM. CHEM. SOC. 2010, 132, 5479–5493 5479

A R T I C L E S

Chen et al.

Because there are spin-active isotopes of indium and phos-

phorus with high natural abundance, solid-state ¹¹⁵In and ³¹P

NMR spectroscopy may provide insights into these adducts that

are not readily available by other techniques. For example, the

determination of indium and phosphorus chemical shift, CS,

tensors, indirect spin-spin coupling constants between ¹¹⁵In and

³¹P nuclei, ¹J(¹¹⁵In,³¹P), and electric ﬁeld gradient, EFG, tensors

at indium can provide useful complementary structural informa-

tion about the local indium and phosphorus environments. Solid-

state indium NMR spectroscopy has traditionally been consid-

ered challenging because the two naturally occurring NMR-

active isotopes, ¹¹³In and ¹¹⁵In, are quadrupolar nuclei (S ) 9/2)

with the largest nuclear quadrupole moments of the main group

elements (¹¹³In, N.A. ) 4.28%, Q ) 75.9 fm²; ¹¹⁵In, N.A. )

95.72%, Q ) 77.0 fm²).⁹The large Q values result in

proportionally large nuclear quadrupolar coupling constants,

which generally lead to broad NMR line shapes that are difﬁcult

to acquire experimentally. Despite its slightly larger nuclear

quadrupole moment, ¹¹⁵In is the preferred nucleus for NMR

studies because of its much greater natural abundance.

technique.^23-25From the known range of isotropic indium

chemical shift values, δ_iso(In), ∼1100 ppm,²⁶one might expect

the orientation dependence of the chemical shifts to be

signiﬁcant, but such information for solid indium compounds

has not been systematically investigated.

In this contribution, a series of triarylphosphine indium

trihalide adducts, X₃In(PR₃) and X₃In(PR₃)₂(Chart 1) are

investigated by determining and interpreting their indium EFG

and CS tensors; these results are corroborated by the results of

density functional theory (DFT) calculations. Possible causes

for the observed sensitivity of the indium CS tensors to the

nature of the halogen ligands are considered. In addition,

analysis of the ³¹P NMR spectra allowed the determination of

¹J(¹¹⁵In,³¹P) values, and for several of these adducts,

∆J(¹¹⁵In,³¹P), the anisotropy in ¹J(¹¹⁵In,³¹P), was also determined.

Previously, ∆J(¹¹⁵In,³¹P) has only been determined from an

analysis of the ³¹P NMR spectra of Br₃In[P(p-Anis)₃].²⁷The

1

signs for ∆J(¹¹⁵In,³¹P), J(¹¹⁵In,³¹P), and C_Q(¹¹⁵In) were also

determined for high-symmetry adducts. Finally, since knowledge

of the structures of the adducts under study is invaluable for a

proper analysis of the ¹¹⁵In and ³¹P NMR spectra, single-crystal

X-ray diffraction data for several of these compounds are also

presented.

Until recently, solid-state ¹¹⁵In NMR studies have been mainly

conﬁned to compounds such as the indium spinel CdIn₂S₄¹⁰and

11

3-

the hexachloroindate anion, InCl₆

,

for which symmetry

ensures that the EFGs at indium are small, resulting in nuclear

quadruploar coupling constant values, C_Q(¹¹⁵In), that are typi-

cally less than 50 MHz. Larger C_Q(¹¹⁵In) values have been

reported, but these were determined using either nuclear

quadrupolar resonance (NQR)¹²or microwave spectroscopy.^13-15

However, indium CS and spin-spin coupling tensors cannot

be characterized using NQR spectroscopy, and while high-

resolution microwave spectroscopy can in principle provide such

information,¹⁶the technique can only be applied to small

molecules in the gas state, and thus only a few indium

compounds, such as diatomic indium-halides,^13,17,18InOH,¹⁴

and In(C₅H₅)¹⁵have been studied by this technique. Recent work

in our lab has demonstrated that obtaining ¹¹⁵In NMR spectra

of solid samples with C_Q(¹¹⁵In) values as large as 250-300 MHz

is feasible.¹⁹Broad central-transition powder patterns for ¹¹⁵In

NMR spectra were acquired with either the quadrupolar-echo²⁰

or the quadrupolar Carr-Purcell Meiboom-Gill (QCPMG)^21,22

pulse sequences, in conjunction with the stepped-frequency

Theory and Background

Indium NMR Spectroscopy. The ¹¹⁵In nuclei for the com-

plexes considered here are subject to the Zeeman, nuclear

quadrupolar, and nuclear magnetic shielding interactions, and

since they are directly bonded to one or two ³¹P nuclei (I )

1/2, NA ) 100%), they are also subject to the direct dipolar

and indirect nuclear spin-spin coupling interactions.

The nuclear quadrupolar coupling describes the interaction

of the nuclear quadrupole moment, eQ, with the EFG at the

nucleus;²⁸the latter is described by a symmetric traceless

second-rank tensor which, when in its principal axis system

(PAS), is diagonal and may be characterized by two independent

parameters: C_Q) eQV_ZZ/h (in Hz), where |V_ZZ| ) |eq_ZZ| is the

principal component of the EFG tensor with the greatest

magnitude, and the asymmetry parameter, η_Q) (V_XX- V_YY)/

V_ZZwith 0 e η_Qe 1 and |V_XX| e |V_YY| e |V_ZZ|. For the systems

considered here, the quadrupolar interaction can be treated as a

perturbation of the Zeeman interaction (Vide infra); therefore,

it is convenient to discuss the results of both ﬁrst- and second-

order perturbation theory.^28,29To ﬁrst order, the quadrupolar

interaction has no effect on the central transition, m_s) 1/2 f

m_s) -1/2, but deﬁnes the line shape of the satellite transi-

tions.²⁸However, to second order, the quadrupolar interaction

affects all allowed NMR transitions (nine for S ) 9/2 nuclei).²⁸

(9) Pyykko¨, P. Mol. Phys. 2008, 106, 1965–1974.

(10) Breuer, A.; Siebert, D. Ber. Bunsen Ges. Phys. Chem. 1996, 100, 1736–

1739.

(11) Yamada, K.; Kumano, K.; Okuda, T. Solid State Ionics 2005, 176,

823–829.

(12) Bastow, T. J.; West, G. W. J. Phys.: Condens. Matter 2003, 15, 8389–

8406.

(13) Hensel, K. D.; Gerry, M. C. L. J. Chem. Soc., Faraday Trans. 1997,

93, 1053–1059.

(23) (a) Kennedy, M. A.; Vold, R. L.; Vold, R. R. J. Magn. Reson. 1991,

92, 320–331. (b) Bastow, T. J.; Smith, M. E.; Stuart, S. N. Chem.

Phys. Lett. 1992, 191, 125–129.

(14) Lakin, N. M.; Varberg, T. D.; Brown, J. M. J. Mol. Spectrosc. 1997,

183, 34–41.

(15) Drouin, B. J.; Cassak, P. A.; Briggs, P. M.; Kukolich, S. G. J. Chem.

Phys. 1997, 107, 3766–3773.

(24) Ooms, K. J.; Bernard, G. M.; Kadziola, A.; Kofod, P.; Wasylishen,

R. E. Phys. Chem. Chem. Phys. 2009, 11, 2690–2699.

(25) (a) Lipton, A. S.; Wright, T. A.; Bowman, M. K.; Reger, D. L.; Ellis,

P. D. J. Am. Chem. Soc. 2002, 124, 5850–5860. (b) Bryant, P. L.;

Butler, L. G.; Reyes, A. P.; Kuhns, P. Solid State Nucl. Magn. Reson.

2000, 16, 63–67.

(16) Bryce, D. L.; Wasylishen, R. E. Acc. Chem. Res. 2003, 36, 327–334.

(17) Barrett, A. H.; Mandel, M. Phys. ReV. 1958, 109, 1572–1589.

(18) Walker, N. R.; Francis, S. G.; Rowlands, J. J.; Legon, A. C. J. Mol.

Spectrosc. 2006, 239, 126–129.

(19) Chen, F.; Ma, G.; Cavell, R. G.; Terskikh, V. V.; Wasylishen, R. E.

Chem. Commun. 2008, 5933–5935.

(26) Akitt, J. W. Aluminum, Gallium, Indium and Thallium. In Multinuclear

NMR; Mason, J., Ed.; Plenum Press: New York, 1987; Chapter 9, pp

259-292.

(20) Bodart, P. R.; Amoureux, J.-P.; Dumazy, Y.; Lefort, R. Mol. Phys.

2000, 98, 1545–1551.

(27) Wasylishen, R. E.; Wright, K. C.; Eichele, K.; Cameron, T. S. Inorg.

Chem. 1994, 33, 407–408.

(21) Larsen, F. H.; Jakobsen, H. J.; Ellis, P. D.; Nielsen, N. C. J. Phys.

Chem. A 1997, 101, 8597–8606.

(28) Abragam, A. Principles of Nuclear Magnetism; Clarendon Press:

Oxford, 1961; Chapter VII, pp 216-263.

(22) Lipton, A. S.; Heck, R. W.; Ellis, P. D. J. Am. Chem. Soc. 2004, 126,

4735–4739.

(29) Bain, A. D. J. Magn. Reson. 2006, 179, 308–310.

9

5480 J. AM. CHEM. SOC. VOL. 132, NO. 15, 2010

¹¹⁵In and ³¹P NMR Studies of In-P Complexes

A R T I C L E S

It should be noted that the third-order perturbation term does

not contribute to the line shape of the central transition for half-

integer quadrupolar nuclei.²⁹In the absence of anisotropic

magnetic shielding, the maximum and minimum frequencies

of the central transition for S ) 9/2 nuclei of a stationary sample

spin-spin coupling tensor. When the two coupled nuclei lie

on a C_nsymmetry axis with n g 3, J is axially symmetric with

the unique component coincident with the dipolar vector, r_IS;

36,37

in such cases, the anisotropy of J, ∆J, is J_|- J_⊥

.

The

dipolar interaction describes the through-space interaction

between two spins, I and S. The magnitude of this interaction

is represented by the dipolar coupling constant:

2

with axial symmetry are ν_S+ C_Q/(384ν_S) and ν_S- C_Q/(216ν_S),

respectively,³⁰where ν_Sis the isotropic resonance frequency

for the quadrupolar nucleus S. From the inverse dependence of

these expressions on ν_S(and hence of the breadth of the NMR

powder pattern), it is apparent that obtaining NMR spectra of

the central transition at the highest practical applied magnetic

ﬁeld strength is usually advantageous, particularly for nuclei

with large values of C_Q.

µ₀γ_Iγ_Sp

-3

R_DD

)

〈r_IS

〉

(1)

(_4π)

2π

where r_ISis the internuclear separation.³⁷Experimentally, one

measures an effective dipolar coupling constant, R_eff. Recalling

that we are considering the case where the dipolar and J tensors

are conincident, R_eff) R_DD- ∆J/3.³⁷If r_ISis known, R_DDcan

be calculated from eq 1 and if R_effis determined from NMR

experiments, the magnitude of ∆J may be estimated.

In the nonrelativistic theory proposed by Ramsey,³¹nuclear

magnetic shielding is expressed as the sum of diamagnetic, σ_dia,

and paramagnetic, σ_para, terms. For heavy atoms, spin-orbit

effects on the shielding are substantial, and thus this factor,

Phosphorus-31 NMR Spectroscopy of the ¹¹⁵In-³¹P Spin

Pair. If under MAS conditions the breadth of the central

transition of an ¹¹⁵In NMR spectrum exceeds the maximum

practical spinning frequency, one must obtain and analyze

spectra of stationary samples. This does not preclude the

determination of ¹¹⁵In NMR parameters (Vide infra), but the

methods may be time-consuming. As an alternative, or to

corroborate data obtained by ¹¹⁵In NMR spectroscopy, one may

determine some ¹¹⁵In quadrupolar parameters indirectly if the

¹¹⁵In nuclei are spin-spin coupled to a spin-1/2 nucleus, such

as ³¹P for the samples considered here.³⁸

In the high-ﬁeld approximation (only valid if the ratio

C_Q/[4S(2S - 1)ν_S] is small) the isotropic region of an NMR

spectrum of a ³¹P nucleus spin-spin coupled to ¹¹⁵In consists

of 10 peaks split by J(¹¹⁵In,³¹P) (Figure 1, upper trace).³⁸

However, if this ratio is signiﬁcant, the direct dipolar interaction

for the ¹¹⁵In-³¹P spin pair is not averaged to zero by magic

angle spinning (MAS),³⁸a consequence of the fact that the

indium nuclei are not quantized exactly along B₀.³⁹In this case,

one observes an uneven spacing between the peaks in the ³¹P

NMR spectra, described by:⁴⁰

σ

_spin-orbit, is often included in discussions of their nuclear

magnetic shielding.³²This effect is now accepted to be the

source of the long-known relativistic effect on nuclear magnetic

shielding.^33,34

The magnetic shielding experienced by a nucleus is affected

by the local environment experienced by that nucleus and thus

is generally dependent on the orientation of the molecule relative

to B₀. The shielding is described by a second-rank tensor, σ,

which is diagonal in its PAS with three principal components,

σ₁₁, σ₂₂, and σ₃₃, where σ₁₁e σ₂₂e σ₃₃.³⁵In practice, NMR

spectroscopists measure chemical shifts, δ, the nuclear magnetic

shielding of a given nucleus relative to that for a reference

compound. The chemical shift is related to the magnetic

shielding by δ_sample≈ σ_ref- σ_sample, where the latter two terms

refer to the magnetic shielding of the reference compound and

of the sample, respectively.³⁵When discussing NMR spectra,

it is convenient to use three magnetic shielding or chemical shift

parameters: the isotropic nuclear magnetic shielding, σ_iso

(1/3)Tr(σ), or the isotropic chemical shift, δ_iso) (δ₁₁+ δ₂₂

)

+

δ₃₃)/3, where δ₁₁g δ₂₂g δ₃₃, the span, Ω ) σ₃₃- σ₁₁) δ₁₁

- δ₃₃, describing the maximum orientation dependence of the

magnetic shielding interaction, and the skew, κ ) 3(σ_iso

-

[S(S + 1) - 3m²]

σ₂₂)/Ω ) 3(δ₂₂- δ_iso)/Ω, which is unitless with -1 e κ e

ν_m) ν_L- m_s|J_iso| +

d

(2)

+1.³⁵

S(2S - 1)

The isotropic indirect spin-spin coupling constant, J, between

I and S is given by (1/3)Tr(J),^36,37where J is the indirect

where m_sis the quantum spin-state of the ¹¹⁵In nucleus, ν_Lis

the isotropic NMR frequency of the ³¹P nucleus, and d is the

residual dipolar coupling constant, deﬁned as

(30) Amoureux, J. P.; Fernandez, C.; Granger, P. Interpretation of Qua-

drupolar Powder Spectra: Static and MAS Experiments. In Multi-

nuclear Magnetic Resonance in Liquids and Solids - Chemical

Applications; Granger, P. Harris, R. K., Eds.; NATO ASI Series C:

Mathematical and Physical Sciences; Kluwer Academic Publishers:

Dordrecht; 1990, pp 409-424.

3C_QR_eff

d ) +

[(3 cos²ꢀ^D- 1) + η_Qsin²ꢀ^Dcos 2R^D]

20ν_S

(3)

(31) Ramsey, N. F. Phys. ReV. 1950, 78, 699–703.

Here, ꢀ^Dand R^Dare polar angles describing the orientation

of the ¹¹⁵In-³¹P dipolar vector with respect to the principal

components of the EFG tensor at the ¹¹⁵In nucleus. It should be

noted that in eq 3 the J and dipolar tensors are assumed to be

(32) Pyykko¨, P.; Go¨rling, A.; Ro¨sch, N. Mol. Phys. 1987, 61, 195–205.

(33) (a) Sugimoto, M.; Kanayama, M.; Nakatsuji, H. J. Phys. Chem. 1993,

97, 5868–5874. (b) Takashima, H.; Hada, M.; Nakatsuji, H. Chem.

Phys. Lett. 1995, 235, 13–16.

(34) Autschbach, J.; Ziegler, T. Relativistic Computation of NMR Shield-

ings and Spin-Spin Coupling Constants. In Encyclopedia of Nuclear

Magnetic Resonance; Grant, D. M., Harris, R. K., Eds.; John Wiley

& Sons: Chichester, 2002; Vol. 9, pp 306-323.

(38) (a) Harris, R. K.; Olivieri, A. C. Prog. Nucl. Magn. Reson. Spectrosc.

1992, 24, 435–456, and references cited therein. (b) Harris, R. K.;

Olivieri, A. C. Spinning Sideband Analysis for Spin-1/2 Nuclei. In

Encyclopedia of Nuclear Magnetic Resonance; Grant, D. M., Harris,

R. K., Eds.; John Wiley & Sons: Chichester, 2002; Vol. 9, pp 141-

150. (c) Grondona, P.; Olivieri, A. C. Concepts Magn. Reson. 1993,

5, 319–339.

(35) Mason, J. Solid State Nucl. Magn. Reson. 1993, 2, 285–288.

(36) Buckingham, A. D.; Love, I. J. Magn. Reson. 1970, 2, 338–351.

(37) (a) Wasylishen, R. E. Dipolar and Indirect Coupling Tensors in Solids.

In Encyclopedia of Nuclear Magnetic Resonance; Grant, D. M., Harris,

R. K., Eds.; John Wiley & Sons: Chichester, 1996; Vol. 3, pp 1685-

1695. (b) Wasylishen, R. E. Indirect Nuclear Spin-Spin Coupling

Tensors. In Encyclopedia of Nuclear Magnetic Resonance; Grant,

D. M., Harris, R. K., Eds.; John Wiley & Sons: Chichester, 2002;

Vol. 9, pp 274-282 and references cited therein.

(39) VanderHart, D. L.; Gutowsky, H. S.; Farrar, T. C. J. Am. Chem. Soc.

1967, 89, 5056–5057.

(40) (a) Olivieri, A. C. Solid State Nucl. Magn. Reson. 1992, 1, 345–353.

(b) Olivieri, A. C. J. Magn. Reson. 1989, 81, 201–205.

9

J. AM. CHEM. SOC. VOL. 132, NO. 15, 2010 5481

A R T I C L E S

Chen et al.

Figure 1. Calculated splitting pattern in a ³¹P NMR spectrum of an MAS sample containing an ¹¹⁵In-³¹P spin pair with d ) 0, d > 0, and d < 0 (see eq

1

3 for the deﬁnition of d). The spin states as indicated are those for a positive sign for J(¹¹⁵In,³¹P), but the splitting pattern is invariant to this sign.

Anis)₃]}_0.97·{I₃In[P(m-Anis)₃]₂}_0.03, and {I₃In[P(m-Anis)₃]}_0.19·{I₃In[P(m-

Anis)₃]₂}_0.81; see Supporting Information for a summary of these results.

The preparation of Cl₃In(TMP) was reported in our earlier com-

munication.¹⁹Cl₃In[P(p-Anis)₃]₂,⁷Br₃In[P(p-Anis)₃],^7,27I₃In[P(p-

Anis)₃]₂,⁷Cl₃In(PPh₃)₂,⁴¹Br₃In(PPh₃)₂,⁶I₃In(PPh₃) (reaction in

cyclohexane),⁴²and I₃In(PPh₃)₂· I₃In(PPh₃) (reaction in ethyl

acetate)^42,43were prepared according to literature methods. The

preparations were carried out under an inert atmosphere, but all

synthesized adducts are stable in air for moderate periods of time.

Single-Crystal X-ray Diffraction. Single crystals were grown

by very slow evaporation of the solvents. Suitable crystals were

mounted on glass ﬁbers by means of paratone-N oil, and data were

collected at 193 K using graphite-monochromated Mo KR radiation

(0.71073 Å) on a Bruker PLATFORM/SMART 1000 CCD dif-

fractometer. The structures were solved by Patterson search/structure

expansion (DIRDIF-99)⁴⁴or by direct methods (SHELXS-97)⁴⁵and

reﬁned using full-matrix least-squares on F²(SHELXL-97).⁴⁵All

non-hydrogen atoms in the structure were reﬁned with anisotropic

displacement parameters. Selected crystal data and structure reﬁne-

ment details for the triarylphosphine indium trihalide adducts are

available from Supporting Information.

Solid-State NMR Spectroscopy. Samples were packed in 4 mm

o.d. rotors. Phosphorus-31 NMR spectra were acquired using a

Chemagnetics CMX Inﬁnity 200 (B₀) 4.70 T) as well as Bruker

Avance 300 (B₀) 7.05 T) and 500 (B₀) 11.75 T) NMR

spectrometers using the combination of cross-polarization (CP), ¹H

two-pulse phase modulation (TPPM) decoupling⁴⁶and magic angle

spinning (MAS). Proton 90° pulse widths of 2.5 or 4.0 µs, contact

times of 5-10 ms and recycle delays of 4- 8 s were used to acquire

most ³¹P NMR spectra. Phosphorus-31 chemical shifts were

referenced with respect to 85% aqueous phosphoric acid (δ_iso) 0)

by setting the isotropic peak of an external solid ammonium

dihydrogen phosphate sample to 0.8 ppm.⁴⁷All spectra were

acquired at ambient temperature with a spinning frequency ranging

from 8.0 to 15.0 kHz.

coincident; more general equations are available in the litera-

ture.⁴⁰If J_isoand R_effmake detectable contributions to the ³¹P

NMR spectra, ꢀ^Dand R^Dmay be determined from those of MAS

and stationary samples. As shown in Figure 1, in the presence

of signiﬁcant residual dipolar coupling, the d value can be

obtained experimentally, from which one may obtain informa-

tion about C_Q(eq 3). If |C_Q| is determined from other

experiments (e.g., ¹¹⁵In NMR experiments for the samples

considered here) and if the values for ꢀ^Dand R^Dmay be

surmised from molecular symmetry, the sign for C_Qmay be

determined from the “sense” of the spectrum (i.e., whether the

splitting between peaks increases or decreases with increasing

frequency). Since simulations of NMR spectra of quadrupolar

nuclei are invariant to the sign of C_Q, this is the only NMR

technique that allows the determination of the sign of C_Q. For

several of the samples considered here, d was of sufﬁcient

magnitude to allow the determination of the sign of C_Qor to

corroborate the value for C_Q(¹¹⁵In) determined from ¹¹⁵In NMR.

Note that d describes the center of mass for the individual peaks

of the multiplets. In principle, each of these peaks has a distinct

line shape,³⁸but as shown below, the line shapes are not

detectable for the ³¹P NMR spectra considered here.

Experimental and Computational Details

Sample Preparation. InCl₃, InBr₃, InI₃, and tris(p-methoxyphe-

nyl)phosphine, P(p-Anis)₃; tris(m-methoxyphenyl)phosphine, P(m-

Anis)₃; and tris(2,4,6-trimethoxyphenyl)phosphine, TMP, were

purchased from Strem Chemicals. Triphenylphosphine, PPh₃;

tris(2,6-dimethoxyphenyl)phosphine, TDP; and tris(o-methoxyphe-

nyl)phosphine, P(o-Anis)₃, were purchased from Alfa Aesar. All

commercial compounds were used as received. Owing to the ease

of hydrolysis of the anhydrous indium halides and oxidation of the

phosphine ligands, all syntheses were carried out in a dry glovebox

under argon. Anhydrous solvents were used in all the preparations.

Similar procedures were used for the synthesis of all the

triarylphosphine indium trihalide adducts. Unless otherwise noted,

reactions were carried out in ethyl acetate. All adducts were

prepared by adding a solution of the anhydrous indium(III) trihalide,

dissolved in selected organic solvents, to a stoichiometric amount

of the phosphine ligands in the same solvent. During stirring, white

or light-green powders of these adducts precipitated and were

ﬁltered, washed with a small amount of the same solvent, and dried

in vacuo. Elemental analysis (EA) was carried out for all adducts

synthesized here: Br₃In(TMP), I₃In(TMP), Cl₃In(TDP), Br₃In(TDP),

I₃In(TDP), Cl₃In[P(o-Anis)₃]·P(o-Anis)₃, Cl₃In[P(o-Anis)₃], Br₃In[P(o-

Anis)₃],I₃In[P(o-Anis)₃],Cl₃In[P(m-Anis)₃]₂,Br₃In[P(m-Anis)₃],{I₃In[P(m-

(41) Veidis, M. V.; Palenik, G. J. J. Chem. Soc. D 1969, 586–587.

(42) Brown, M. A.; Tuck, D. G.; Wells, E. J. Can. J. Chem. 1996, 74,

1535–1549.

(43) Godfrey, S. M.; Kelly, K. J.; Kramkowski, P.; McAuliffe, C. A.;

Pritchard, R. G. Chem. Commun. 1997, 1001–1002.

(44) Beurskens, P. T.; Beurskens, G.; de Gelder, R.; Garcia-Granda, S.;

Israel, R.; Gould, R. O.; Smits, J. M. M. The DIRDIF-99 Program

System; Crystallography Laboratory, University of Nijmegen: Nijmegen,

the Netherlands, 1999.

(45) Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112–122.

(46) Bennett, A. E.; Rienstra, C. M.; Auger, M.; Lakshmi, K. V.; Grifﬁn,

R. G. J. Chem. Phys. 1995, 103, 6951–6958.

(47) Eichele, K.; Wasylishen, R. E. J. Phys. Chem. 1994, 98, 3108–3113.

9

5482 J. AM. CHEM. SOC. VOL. 132, NO. 15, 2010

¹¹⁵In and ³¹P NMR Studies of In-P Complexes

A R T I C L E S

Indium-115 NMR spectra of stationary samples were acquired

using either quadrupolar-echo²⁰or QCPMG²¹pulse sequences;

where necessary (typically, spectra whose breadth exceeded 100

kHz), the stepped-frequency technique²³was used. Bruker Avance

300 and 500 NMR spectrometers, with ¹¹⁵In NMR frequencies of

65.8 and 109.6 MHz, respectively, and a Bruker Avance II 900

NMR spectrometer (21.14 T), with a ¹¹⁵In NMR frequency of 197.2

MHz, were used to acquire all spectra. A relaxation delay of 0.5 s

was used for all samples. For the solid-state NMR experiments,

the ¹¹⁵In 90° pulse widths, τ_(sel), that selectively excited the central

transition were set as τ_(sel)) τ_(nonsel)/(S + 1/2) ) τ_(nonsel)/5.²⁰The

polarization functions and is optimized for ZORA calculations, was

used for indium, phosphorus, oxygen, chlorine, bromine and iodine

atoms and the TZP basis set, of triple- quality with one set of

polarization functions, also optimized for ZORA calculations, was

used for the H and C atoms.

Geometries of the X₃In(PR₃)_n(n ) 1 or 2) adducts obtained from

X-ray diffraction were used for the calculations. To reduce

computational time, the PR₃ligands in the trigonal-bipyramidal

X₃In(PR₃)₂adducts were replaced by PMe₃ligands without chang-

ing the geometry of the X₃InP₂core. To investigate the effects of

the halides and of the geometry on the indium magnetic shielding,

CS tensors for model X₃In(PMe₃) and X₃In(PMe₃)₂(X ) Cl, Br,

or I) adducts with a C₃symmetry axis were also calculated. The

apparent correlation between ¹J(¹¹⁵In,³¹P) and the basicity of

the triarylphosphine ligands was investigated by calculating the

¹J(¹¹⁵In,³¹P) and ∆J(¹¹⁵In,³¹P) values as well as their signs for

Cl₃In(PR₃) and Br₃In(PR₃) adducts. The geometries used for the

latter were those obtained from X-ray crystallography, where

available. If such data were unavailable, the structures for

Cl₃In(TMP) or Br₃In(TMP) were used, with the methoxy groups

replaced by H where needed (e.g., the para methoxy groups of TMP

were replaced with hydrogen atoms to model the TDP ligand).

Because of the long computational time, it was impractical to

τ_(sel)value was 0.4 µs for spectra acquired at 7.05 and 11.75 T, and

0.5 µs for spectra acquired at 21.14 T. Proton decoupling was

achieved using the TPPM scheme.⁴⁶Each step of the ¹¹⁵In NMR

spectra acquired at 7.05 or 11.75 T is the sum of 29020 to 77824

scans. Experimental conditions for the QCPMG pulse sequence

were set following literature procedures,²¹with 1024 or 2048 scans

for each step. The individual stepped-frequency spectra were

coadded using the skyline projection method.⁴⁸In-115 NMR

chemical shifts were referenced with respect to an external solution

of 0.1 M In(NO₃)₃in 0.5 M HNO₃(δ_iso) 0). Phosphorus-31 and

indium-115 NMR parameters were determined by visual comparison

of experimental NMR spectra with those simulated using the

WSolids software package.⁴⁹This software includes the quadrupolar

interaction to second-order perturbation theory for simulations of

¹¹⁵In NMR spectra, and includes spin-spin interactions with ¹¹⁵In

for simulations of ³¹P NMR spectra of MAS samples.

1

undertake DFT calculations of J(¹¹⁵In,³¹P) and ∆J(¹¹⁵In,³¹P) for

I₃In(PR₃) or for the trigonal-bipyramidal adducts.

Results and Discussion

Quantum Chemical Calculations. DFT calculations of indium

CS⁵⁰and EFG tensors,⁵¹as well as the ¹J(¹¹⁵In,³¹P) and ∆J(¹¹⁵In,³¹P)

values,⁵²were performed using the Amsterdam Density Functional

(ADF) program.⁵³The Vosko-Wilk-Nusair (VWN) local density

approximation,⁵⁴with the Becke-Perdew generalized gradient

approximation (GGA),⁵⁵was used for the exchange-correlation

functional. All calculations allowed for relativistic effects, including

spin-orbital and scalar corrections, and were carried out using the

zeroth-order regular approximation (ZORA) formalism.⁵⁶The ¹¹⁵In

EFG values determined using this software are based on a value

for Q of 81.0 fm², rather than the more recent value summarized

by Pyykko¨;⁹calculated values reported here have been scaled by a

factor of 0.951 to incorporate the more recent value for Q(¹¹⁵In).

The Fermi-contact (FC), spin dipolar (SD), paramagnetic spin-orbital

(PSO) and diamagnetic spin-orbital (DSO) mechanisms were

included in all J calculations;^36,57a description of these mechanisms,

as implemented in the ZORA formalism, has been presented.⁵²ADF

uses Slater-type basis sets; for the results presented herein, the QZ4P

basis set, which is of quadruple- quality with four sets of

This section is organized as follows: ﬁrst, the molecular

structures of the triarylphosphine indium trihalide adducts are

discussed. Next, solid-state ¹¹⁵In NMR spectra for each adduct

are presented and interpreted. Third, the ¹J(¹¹⁵In,³¹P) values for

all complexes, determined Via ³¹P NMR spectroscopy, are

considered. For complexes with an exact or approximate C₃axis,

the ∆J(¹¹⁵In,³¹P) values and the signs for ∆J(¹¹⁵In,³¹P) and

C_Q(¹¹⁵In) are also presented. Finally, relativistic DFT calculations

of NMR parameters for these compounds are compared with

the experimental values and the causes of the observed sensitiv-

ity of the indium CS tensors to the nature of the directly bonded

halogen ligands are explored.

Structures of the Triarylphosphine Indium Trihalide

Adducts. Structural data obtained Via X-ray crystallography are

important for this study, to relate NMR data with structure.

Single-crystal structures for Cl₃In(PPh₃)₂,⁴¹I₃In(PPh₃),⁴²

I₃In(PPh₃)₂·I₃In(PPh₃),⁴²Cl₃In(TMP),¹⁹and Br₃In[P(p-Anis)₃]²⁷

have been reported previously. The structures for nine additional

adducts were determined as part of this study; see Tables S1

and S2 in Supporting Information for more detailed X-ray

diffraction data and for selected structural information.

(48) (a) Nagayama, K.; Bachmann, P.; Wu¨thrich, K.; Ernst, R. R. J. Magn.

Reson. 1978, 31, 133–148. (b) Blu¨mich, B.; Ziessow, D. J. Magn.

Reson. 1982, 49, 151–154.

(49) Eichele, K.; Wasylishen, R. E. WSOLIDS, Version 2.0.18; University

of Alberta: Edmonton, Canada, 2000.

We were unable to obtain crystals suitable for X-ray diffrac-

tion studies for adducts containing the P(m-Anis)₃or TDP

ligands. However, EA combined with solid-state ³¹P NMR (Vide

infra) established whether 1:1 or 2:1 adducts were formed and

conﬁrmed the In-P connectivity. Existing structural data⁸as

well as that presented here have shown that 1:1 and 2:1

triarylphosphine indium(III) trihalide adducts adopt approximate

tetrahedral or trigonal-bipyramidal geometries about the indium

atom, respectively. In the ensuing discussion, adducts are

assumed to adopt one of these two structures, depending on

the EA and NMR data, if X-ray diffraction data are unavailable.

Adducts formed by InCl₃with less crowded phosphine ligands

adopt 2:1 trigonal-bipyramidal structures: Cl₃In[P(p-Anis)₃]₂,⁷

Cl₃In[P(m-Anis)₃]₂, and Cl₃In(PPh₃)₂;⁴¹in contrast, crowded

phosphine ligands such as TMP or TDP form 1:1 adducts which

have approximately tetrahedral structures. See Figures 2a and

2b for examples of tetrahedral and trigonal-bipyramidal

(50) Schreckenbach, G.; Ziegler, T. J. Phys. Chem. 1995, 99, 606–611.

(51) van Lenthe, E.; Baerends, E. J. J. Chem. Phys. 2000, 112, 8279–8292.

(52) (a) Autschbach, J.; Ziegler, T. J. Chem. Phys. 2000, 113, 9410–9418.

(b) Autschbach, J.; Ziegler, T. J. Chem. Phys. 2000, 113, 936–947.

(53) (a) ADF, 2006.01; Theoretical Chemistry, Vrije Universiteit: Amster-

dam, 2006; http://www.scm.com. (b) te Velde, G.; Bickelhaupt, F. M.;

Baerends, E. J.; Fonseca Guerra, C.; van Gisbergen, S. J. A.; Snijders,

J. G.; Zeigler, T. J. Comput. Chem. 2001, 22, 931–967. (c) Fonseca

Guerra, C.; Snijders, J. G.; te Velde, G.; Baerends, E. J. Theor. Chem.

Acc. 1998, 99, 391–403.

(54) Vosko, S. H.; Wilk, L.; Nusair, M. Can. J. Phys. 1980, 58, 1200–

1211.

(55) (a) Becke, A. D. Phys. ReV. A 1988, 38, 3098–3100. (b) Perdew, J. P.

Phys. ReV. B 1986, 33, 8822–8824. (c) Perdew, J. P. Phys. ReV. B

1986, 34, 7406.

(56) (a) van Lenthe, E.; Baerends, E. J.; Snijders, J. G. J. Chem. Phys.

1993, 99, 4597–4610. (b) van Lenthe, E.; Baerends, E. J.; Snijders,

J. G. J. Chem. Phys. 1994, 101, 9783–9792. (c) van Lenthe, E.; Ehlers,

A.; Baerends, E. J. J. Chem. Phys. 1999, 110, 8943–8953.

(57) Vaara, J.; Jokisaari, J.; Wasylishen, R. E.; Bryce, D. L. Prog. Nucl.

Magn. Reson. Spectrosc. 2002, 41, 233–304.

9

J. AM. CHEM. SOC. VOL. 132, NO. 15, 2010 5483

A R T I C L E S

Chen et al.

Figure 2. Molecular structures of (a) Cl₃In(TMP),¹⁹(b) Cl₃In[P(p-Anis)₃]₂, (c) Cl₃In[P(o-Anis)₃] and (d) Cl₃In[P(o-Anis)₃]·P(o-Anis)₃.

structures, respectively. EA results indicate that the adducts of

InCl₃with P(o-Anis)₃can adopt either the 1:1 or 2:1 structures,

depending on the ratio of InCl₃to P(o-Anis)₃used in the

synthesis; the former adopts a tetrahedral structure (Figure 2c).

Surprisingly, single-crystal X-ray diffraction data indicate that

the 2:1 adduct does not have a trigonal-bipyramidal structure,

but instead forms tetrahedral Cl₃In[P(o-Anis)₃] with an additional

uncoordinated P(o-Anis)₃molecule packed in the same unit cell

(Figure 2(d)); thus, the molecular formula is Cl₃In[P(o-

Anis)₃]·P(o-Anis)₃.

For the InBr₃adducts considered here, only Br₃In(PPh₃)₂

adopts the trigonal-bipyramidal structure; the remaining ad-

ducts, Br₃In[P(o-Anis)₃], Br₃In[P(p-Anis)₃],²⁷Br₃In[P(m-Anis)₃],

Br₃In(TDP), and Br₃In(TMP), are tetrahedral, presumably due

to the bulkiness of the Br atoms.

Adducts formed by InI₃with P(o-Anis)₃, P(p-Anis)₃, TDP,

and TMP are approximately tetrahedral; the structures of the

InI₃adducts with PPh₃or P(m-Anis)₃depend on the solvents

used for the synthesis. Those with PPh₃can crystallize either

as I₃In(PPh₃) alone, or as an equimolar mixture of I₃In(PPh₃)

and I₃In(PPh₃)₂packed in the same unit cell.⁴²Similarly for

InI₃and P(m-Anis)₃prepared in cyclohexane, one obtains a

mixture that is {I₃In[P(m-Anis)₃]}_0.97·{I₃In[P(m-Anis)₃]₂}_0.03, but

when prepared in ethyl acetate, EA indicates that one obtains a

mixture that is {I₃In[P(m-Anis)₃]}_0.19· {I₃In[P(m-Anis)₃]₂}_0.81

.

Because of the relationship between molecular symmetry and

NMR parameters, this aspect of the structural analyses is of

particular interest. X-ray crystallographic studies of two tetra-

hedral adducts, Br₃In[P(p-Anis)₃] (Supporting Information) and

I₃In(PPh₃),⁴²indicate that there is a C₃axis along the In-P bond,

as for I₃In(PPh₃) in I₃In(PPh₃)₂· I₃In(PPh₃).⁴²Among the

trigonal-bipyramidal adducts, there is a C₃axis along the

P-In-P axis of I₃In(PPh₃)₂in I₃In(PPh₃)₂·I₃In(PPh₃).⁴²The two

phosphorus sites in each unique X₃In(PR₃)₂molecular unit are

distinct for Cl₃In(PPh₃)₂and Br₃In(PPh₃)₂, with coplanar indium

and chlorine atoms; Cl₃In[P(p-Anis)₃]₂has a similar structure,

with a C₂symmetry axis along an In-Cl bond.

Solid-State ¹¹⁵In NMR. Experimental ¹¹⁵In NMR parameters

determined for the triarylphosphine indium trihalide adducts

investigated here are listed in Table 1. With the exception of

the spectra for Br₃In[P(o-Anis)₃] and Br₃In[P(m-Anis)₃], the

breadths of the central ¹¹⁵In NMR transitions (m_I) 1/2 f m_I

) -1/2) for these complexes exceeded 100 kHz at B₀) 21.14

T, precluding the acquisition of NMR spectra of MAS samples.

1:1 Adducts. Simulation of the central transition powder

pattern for the ¹¹⁵In NMR spectrum of Cl₃In(TDP), shown in

9

5484 J. AM. CHEM. SOC. VOL. 132, NO. 15, 2010

¹¹⁵In and ³¹P NMR Studies of In-P Complexes

Table 1. Experimental ¹¹⁵In NMR Parameters

A R T I C L E S

C_Q/MHz

η_Q

δ

_iso/ppm

δ₁₁/ppm

δ₂₂/ppm

δ₃₃/ppm

Ω/ppm

κ

R/deg

ꢀ/deg

γ/deg

a

-165.0 ( 2.0 0.14 ( 0.02 420 ( 30 470 ( 30 431 ( 30 360 ( 30 110 ( 40 0.30 ( 0.10 90 ( 15 0 ( 15 0 ( 10

-166.0 ( 2.0 0.28 ( 0.04 400 ( 30 443 ( 30 413 ( 30 343 ( 30 100 ( 40 0.40 ( 0.10 90 ( 15 0 ( 15 0 ( 10

Cl₃In(TMP)

Cl₃In(TDP)

Cl₃In[P(o-Anis)₃] · P(o-Anis)₃-122.0 ( 2.0 0.17 ( 0.04 390 ( 30 442 ( 30 387 ( 30 342 ( 30 100 ( 40 -0.10 ( 0.10 90 ( 10 50 ( 25 0 ( 15

Cl₃In[P(o-Anis)₃]

Cl₃In[P(p-Anis)₃]₂

Cl₃In[P(m-Anis)₃]₂

Cl₃In(PPh₃)₂

(106.0 ( 2.0 0.60 ( 0.04 400 ( 30 439 ( 30 403 ( 30 359 ( 30 80 ( 40 0.10 ( 0.20 5 ( 10 75 ( 15 50 ( 20

(86.0 ( 2.0 0.35 ( 0.08 300 ( 50 600 ( 50 180 ( 50 120 ( 50 480 ( 70 -0.75 ( 0.20 15 ( 10 85 ( 5

5 ( 5

(54.0 ( 4.0 0.75 ( 0.10 310 ( 30 649 ( 30 182 ( 30

99 ( 30 550 ( 40 -0.70 ( 0.20 0 ( 15 90 ( 15 0 ( 10

(44.0 ( 2.0 0.66 ( 0.04 310 ( 30 622 ( 30 166 ( 30 142 ( 30 480 ( 40 -0.90 ( 0.10 85 ( 5

2 ( 2 12 ( 4

Br₃In(TMP)

-166.0 ( 2.0 0.23 ( 0.05 260 ( 40 335 ( 40 290 ( 40 155 ( 40 180 ( 60 0.50 ( 0.20 90 ( 15 0 ( 15 0 ( 10

-166.0 ( 2.0 0.30 ( 0.05 260 ( 40 338 ( 40 284 ( 40 158 ( 40 180 ( 60 0.50 ( 0.20 90 ( 15 0 ( 15 0 ( 10

(72.0 ( 2.0 0.62 ( 0.06 280 ( 40 370 ( 40 280 ( 40 190 ( 40 180 ( 60 0.00 ( 0.20 88 ( 10 75 ( 15 15 ( 20

Br₃In(TDP)

Br₃In[P(o-Anis)₃]

Br₃In[P(p-Anis)₃]

Br₃In[P(m-Anis)₃]

Br₃In(PPh₃)₂

(1.25 ( 0.10 0.00

323 ( 5

360 ( 5

332 ( 5

305 ( 5

292 ( 5

305 ( 5

292 ( 5

55 ( 7 -1.00

40 ( 7 -1.00 ( 0.10

s

90

0

(5.0 ( 1.0 0.00 ( 0.10 305 ( 5

(32.0 ( 2.0 0.27 ( 0.05 120 ( 30 424 ( 30

-150.0 ( 4.0 0.30 ( 0.05 -100 ( 30 83 ( 30

-8 ( 30 -56 ( 30 480 ( 40 -0.80 ( 0.10 80 ( 10 40 ( 20 80 ( 20

33 ( 30 -417 ( 30 500 ( 40 0.80 ( 0.10 90 ( 15 0 ( 15 0 ( 10

I₃In(TMP)

I₃In(TDP)

-148.0 ( 2.0 0.25 ( 0.05 -200 ( 30 -17 ( 30 -67 ( 30 -517 ( 30 500 ( 40 0.80 ( 0.10 90 ( 15 0 ( 15 0 ( 10

-75.0 ( 2.0 0.30 ( 0.10 -80 ( 30 147 ( 30 -95 ( 30 -293 ( 30 440 ( 40 -0.10 ( 0.10 84 ( 10 65 ( 10 8 ( 10

I₃In[P(o-Anis)₃]

I₃In[P(p-Anis)₃]

-40.0 ( 4.0 0.15 ( 0.15 -50 ( 40 270 ( 40

21 ( 40 -440 ( 40 710 ( 60 0.30 ( 0.10 78 ( 10 64 ( 20 14 ( 10

b

+24.0 ( 2.0 0.00

+36.0 ( 2.0 0.00

(120.0 ( 4.0 0.00

-115 ( 30

-140 ( 50

58 ( 30

33 ( 50

58 ( 30 -462 ( 30 520 ( 40 1.00

33 ( 50 -487 ( 50 520 ( 70 1.00

s

0

I₃In(PPh₃)

c

I₃In(PPh₃)

d

-450 ( 50 -343 ( 50 -343 ( 50 -663 ( 50 320 ( 70 1.00

I₃In(PPh₃)₂

^aFrom ref 19. ^bFree I₃In(PPh₃). ^cI₃In(PPh₃) in I₃In(PPh₃)₂·I₃In(PPh₃). ^dI₃In(PPh₃)₂in I₃In(PPh₃)₂·I₃In(PPh₃).

Figure 3. Experimental and calculated ¹¹⁵In NMR spectra of stationary

powder samples of (a) Cl₃In(TDP), (b) Cl₃In[P(o-Anis)₃], and (c) Cl₃In[P(o-

Anis)₃]·P(o-Anis)₃acquired at 21.14 T.

Figure 4. Experimental and calculated central transition powder patterns

from ¹¹⁵In NMR spectra of (a) Br₃In(TMP), (b) Br₃In[P(o-Anis)₃], and (c)

from all ¹¹⁵In NMR transitions of Br₃In[P(p-Anis)₃].

Figure 3(a), yields a C_Q(¹¹⁵In) value of ( 166.0 ( 2.0 MHz,

equal, within experimental error, to that for Cl₃In(TMP);¹⁹the

CS tensors for these compounds are also similar. The EFG tensor

for the former is almost axially symmetric with V_ZZcoincident

with δ₃₃; these components are aligned along the approximate

C₃axis. Thus, although crystallographic data for Cl₃In(TDP)

are unavailable, these results suggest that the indium coordina-

tion environment for this compound must be similar to that for

Cl₃In(TMP). The ¹¹⁵In NMR spectra of Cl₃In[P(o-Anis)₃] and

Cl₃In[P(o-Anis)₃]·P(o-Anis)₃are shown in Figures 3(b) and 3(c),

respectively. The adducts have similar indium CS tensors, but

the corresponding EFG tensors are different (Table 1), despite

the fact that X-ray crystallography data (Supporting Information)

indicate similar environments about the indium. The Euler angles

for these adducts suggest that the relative orientations of the

EFG and CS tensors are also different, but given the large

uncertainties in the spans of the latter, this conclusion must be

regarded as tentative. These observations are attributed to the

high sensitivity of the indium EFG to minor structural differ-

ences, either short or long-range, a consequence of its large

nuclear quadrupole moment.

Five Br₃In(PR₃) adducts were examined in this study.

Analyses of the ¹¹⁵In NMR spectra for Br₃In(TMP) (Figure 4(a))

and Br₃In(TDP) indicate that these adducts have similar indium

EFG and CS tensors (Table 1). The C_Q(¹¹⁵In) value for

Br₃In[P(o-Anis)₃] determined from the ¹¹⁵In NMR spectrum,

shown in Figure 4(b), is less than that for Cl₃In[P(o-Anis)₃].

For Br₃[P(p-Anis)₃], all ¹¹⁵In NMR transitions can be observeds

see Figure 4(c). The C_Q(¹¹⁵In) value for this complex, 1.25 (

9

J. AM. CHEM. SOC. VOL. 132, NO. 15, 2010 5485

A R T I C L E S

Chen et al.

There is a C₃symmetry axis along the In-P bond for

I₃In(PPh₃) and for I₃In(PPh₃) in I₃In(PPh₃)₂·I₃In(PPh₃); the η_Q

and κ values, 0.0 and 1.00, respectively, indicate that V_ZZand

δ

₃₃are along the C₃axis for both adducts. Within experimental

error, the values for C_Q(¹¹⁵In) are the only ¹¹⁵In NMR parameters

to differ between these two adducts. As for Cl₃In[P(o-Anis)₃]

and Cl₃In[P(o-Anis)₃]·P(o-Anis)₃discussed above, this observa-

tion is attributed to the high sensitivity of the indium EFG

tensors to relatively subtle differences in the environment about

the indium (for EFG tensors, long-range effects can have a

signiﬁcant effect).

The indium magnetic shielding anisotropy for the I₃In(PR₃)

adducts has a much greater effect on the ¹¹⁵In NMR spectra

(Figure 5) than do those for Cl₃In(PR₃) and Br₃In(PR₃). In fact,

for spectra acquired at 21.14 T, the contribution from the

anisotropy in the indium magnetic shielding for I₃In[P(p-Anis)₃]

and I₃In(PPh₃) exceeds that from the second-order term of the

indium quadrupolar interaction (Figures 5(b) and 5(c), respec-

tively).

2:1 Adducts. NMR parameters obtained from the simulation

of the experimental ¹¹⁵In NMR spectra for the X₃In(PR₃)₂

adducts with trigonal-bipyramidal structures, Cl₃In[P(p-

Anis)₃]₂, Cl₃In[P(m-Anis)₃]₂, Cl₃In(PPh₃)₂, Br₃In(PPh₃)₂and

I₃In(PPh₃)₂in I₃In(PPh₃)₂·I₃In(PPh₃), are summarized in Table

1; see Figure 6 for some representative spectra. The C_Q(¹¹⁵In)

values range from ( 32.0 ( 2.0 to ( 120.0 ( 4.0 MHz; there

is a signiﬁcant contribution to the ¹¹⁵In NMR spectra from the

indium magnetic shielding anisotropy, particularly at 21.14 T.

The spans of the 2:1 InCl₃and InBr₃adducts are much greater

than those for the corresponding 1:1 adducts. For NMR spectra

of Cl₃In[P(p-Anis)₃]₂, Cl₃In[P(m-Anis)₃]₂, Cl₃In(PPh₃)₂and

I₃In(PPh₃)₂acquired at 21.14 T, the anisotropic magnetic

shielding is comparable to the second-order quadrupolar per-

turbation (in frequency units); for the latter, the values η_Q, 0,

and κ, 1.00, indicate that V_ZZand δ₃₃are along the C₃axis. For

Br₃In(PPh₃)₂, the anisotropic magnetic shielding dominates at

higher applied magnetic ﬁeld strengths.

Figure 5. Experimental and calculated ¹¹⁵In NMR spectra of stationary

powder samples of (a) I₃In[P(o-Anis)₃], (b) I₃In[P(p-Anis)₃] and (c)

I₃In(PPh₃) acquired at 21.14 T. The upper traces show the individual

contributions from the indium CSA and EFG.

0.10 MHz, was determined by simulation of all transitions of

the ¹¹⁵In NMR spectrum; this value is consistent with a previous

prediction based on an analysis of ³¹P NMR spectra.²⁷The

asymmetry parameter, η_Q) 0, and the skew, κ ) -1.00, are

also consistent with the crystallographic structure which has a

C₃axis along the In-P bond. Thus, the unique components of

the EFG and CS tensors, V_ZZand δ₁₁, respectively, are along

the C₃axis of the molecule. The ¹¹⁵In NMR spectrum for

Br₃In[P(m-Anis)₃] (not shown), is similar to that for Br₃In[P(p-

Anis)₃]; the indium EFG is also very small. Although X-ray

data are unavailable for the latter, the values for η_Qand κ suggest

that within experimental error, this adduct also has a C₃

symmetry axis along the In-P bond. Since the geometry about

the indium for Br₃In[P(p-Anis)₃] is approximately tetrahedral,

the small C_Qvalue suggests that contributions to the EFG

from the P(p-Anis)₃ligand are comparable to those from three

Br atoms;⁵⁸a similar conclusion seems probable for Br₃In[P(m-

Anis)₃], but in this case the structure was not determined.

The indium NMR parameters for the I₃In(PR₃) adducts are

also summarized in Table 1. The indium EFG and CS tensors

for I₃In(TMP) are similar to those for I₃In(TDP), suggesting

that the indium coordination environment for these two adducts

are similar, as was found for the indium trichloride and

tribromide adducts. Analysis of the ¹¹⁵In NMR spectra for

I₃In[P(o-Anis)₃], shown in Figure 5a, indicate that the C_Q(¹¹⁵In)

value for this adduct is similar to that for Br₃In[P(o-Anis)₃];

that for I₃In[P(p-Anis)₃], shown in Figure 5b, indicates that

C_Q(¹¹⁵In) ) -40.0 ( 4.0 MHz and η_Q) 0.15 ( 0.15.

Summary of ¹¹⁵In NMR Results. The C_Q(¹¹⁵In) values for the

indium complexes investigated here cover a wide range, from

1.25 ( 0.10 to -166.0 ( 2.0 MHz (Table 1). There is no

obvious relationship between C_Q(¹¹⁵In) and the structures of the

adducts. A comparison of adducts with the same triarylphos-

phine ligand indicates that the indium nuclei are most and least

shielded for X ) I and Cl, respectively, as shown in Figure 7

and summarized in Table 1; this has been attributed to the

spin-orbit effect of the halogen ligand.³³The spans of the

indium magnetic shielding tensors for the adducts considered

here range from 40 ( 7 to 710 ( 60 ppm. For X ) Cl or Br,

these are greatest for the 2:1 adducts. For the 1:1 adducts, Ω

increases as X changes from Cl to I, but for X₃In(PPh₃)₂, Ω

decreases on going from Cl to I (Figure 7).

Solid-State ³¹P NMR. The experimental ³¹P NMR parameters

determined for the triarylphosphine indium trihalide adducts are

summarized in Table 2. Figures 8-10 provide examples of ³¹

P

NMR spectra for several X₃In(PR₃) and X₃In(PR₃)₂adducts. A

common feature of the spectra of MAS samples for all adducts,

as observed previously,²⁷is the unequal relative heights of the

peaks in each multiplet, although the integrated intensities of

each of the 10 peaks are equal, within experimental error. In

particular, the outer peaks associated with the transitions

|1/2,9/2〉 f |-1/2,9/2〉 and |1/2,-9/2〉 f |-1/2,-9/2〉 are sharper

than those associated with indium spin states of (7/2. Similar

observations have been reported for solution NMR studies (e.g.,

(58) Brill, T. B. Inorg. Chem. 1976, 15, 2558–2560.

9

5486 J. AM. CHEM. SOC. VOL. 132, NO. 15, 2010

¹¹⁵In and ³¹P NMR Studies of In-P Complexes

A R T I C L E S

Figure 6. Experimental and calculated ¹¹⁵In NMR spectra of stationary powder samples of (a) Cl₃In[P(p-Anis)₃]₂, (b) Cl₃In[P(m-Anis)₃]₂, (c) Cl₃In(PPh₃)₂

and (d) Br₃In(PPh₃)₂acquired at 21.14 T. The upper traces show the individual contributions from the indium CSA and EFG.

Figure 7. Spans of the CS tensors for the X₃In(TMP), X₃In(TDP), X₃In[P(o-Anis)₃] and X₃In(PPh₃)₂.

¹⁹F NMR spectra of BiF₆^-, I(²⁰⁹Bi) ) 9/2;⁵⁹¹⁷O NMR spectra

of TcO₄^-, I(⁹⁹Tc) ) 9/2),⁶⁰and is attributed to different

relaxation rates between the spin states of the coupled quadru-

polar nucleus.⁶¹Expressions for the relative probabilities of

single- and double-quantum transitions of spin-9/2 nuclei in the

solid state are not known. Spin-spin interactions with quadru-

polar nuclei may also give rise to differential broadening of

the peaks in the spectra of spin-1/2 nuclei (see Figure S1 and

the accompanying text in Supporting Information for a detailed

discussion of this effect). However these interactions are not

responsible for the patterns seen in Figures 8-10.

1

The J(¹¹⁵In,³¹P) and R_eff(¹¹⁵In,³¹P) values for the adducts

considered here are too small to be extracted from ¹¹⁵In NMR

spectra of stationary samples, but as discussed in the theoretical

section, one might expect to obtain these parameters from ³¹P

NMR spectra. ¹J(¹¹⁵In,³¹P) can be obtained from an analysis of

³¹P NMR spectra of MAS samples and R_eff(¹¹⁵In,³¹P) may be

obtained from those of either stationary or MAS samples. The

latter value along with that for R_DD(

¹¹⁵In,³¹P), available from

(59) Morgan, K.; Sayer, B. G.; Schrobilgen, G. J. J. Magn. Reson. 1983,

52, 139–142.

the In-P bond length, may be used to estimate ∆J(¹¹⁵In,³¹P) if

the J and dipolar tensors are coincident with a C₃symmetry

axis along the In-P bond. In such cases, the EFG tensors are

axially symmetric with the unique component along the In-P

(60) Buckingham, M. J.; Hawkes, G. E.; Thornback, J. R. Inorg. Chim.

Acta 1981, 56, L41–L42.

(61) Pople, J. A. Mol. Phys. 1958, 1, 168–174.

9

J. AM. CHEM. SOC. VOL. 132, NO. 15, 2010 5487

A R T I C L E S

Chen et al.

Table 2. Experimental ³¹P NMR Parameters

δ

_iso/ppm

δ₁₁/ppm

δ₂₂/ppm

δ₃₃/ppm

Ω/ppm

κ

R_eff(¹¹⁵In,³¹P)/Hz R_DD

(

¹¹⁵In,³¹P)/Hz^a¹J(¹¹⁵In,³¹P)/Hz ∆J(¹¹⁵In,³¹P)/Hz

Cl₃In(TMP)

Cl₃In(TDP)

-49.8 ( 2.0 -22.7 ( 2.0 -60.1 ( 2.0 -66.7 ( 2.0 44.0 ( 3.0 -0.70 ( 0.10 +120 ( 50

-64.5 ( 2.0 -32.0 ( 2.0 -79.5 ( 2.0 -82.0 ( 2.0 50.0 ( 3.0 -0.90 ( 0.10 +120 ( 50

+647 ( 20

+636 ( 30

+639 ( 20

+2500 ( 20 +1581 ( 200

+2410 ( 20 +1548 ( 300

+2080 ( 20 +1557 ( 200

b

Cl₃In[P(o-Anis)₃]·P(o-Anis)₃-38.5 ( 2.0 -16.9 ( 2.0 -46.7 ( 2.0 -51.9 ( 2.0 35.0 ( 3.0 -0.70 ( 0.20 +120 ( 50

-23.2 ( 1.0

-30.0 ( 2.0

-0.7 ( 1.0

-5.8 ( 1.0

-0.2 ( 1.0

-3.7 ( 1.0

-2.3 ( 1.0

s

Cl₃In[P(o-Anis)₃]

Cl₃In[P(p-Anis)₃]₂

+2120 ( 20

+1230 ( 20

+1220 ( 20

+1060 ( 20

+1070 ( 20

+940 ( 20

Cl₃In[P(m-Anis)₃]₂

Cl₃In(PPh₃)₂

Br₃In(TMP)

Br₃In(TDP)

Br₃In[P(o-Anis)₃]

Br₃In[P(p-Anis)₃]

Br₃In[P(m-Anis)₃]

Br₃In(PPh₃)₂

-49.8 ( 2.0 -30.3 ( 2.0 -58.8 ( 2.0 -60.3 ( 2.0 30.0 ( 3.0 -0.90 ( 0.10 +130 ( 50

-59.8 ( 2.0 -39.0 ( 2.0 -69.4 ( 2.0 -71.0 ( 2.0 32.0 ( 3.0 -0.90 ( 0.10 +150 ( 50

+631 ( 20

+615 ( 30

s

+615 ( 20

+615 ( 30

s

+2220 ( 20 +1503 ( 200

+2150 ( 20 +1395 ( 300

c

-20.4 ( 1.0

-20.9 ( 0.5

-14.6 ( 1.0

-8.5 ( 0.5

-6.0 ( 0.5

s

+230 ( 50

+240 ( 50

s

+1590 ( 20

+1112 ( 20

+1125 ( 20 +1125 ( 300

s

d

3.1 ( 0.5 -32.9 ( 0.5 -32.9 ( 0.5 36.0 ( 0.7 -1.00

9.4 ( 1.0 -26.6 ( 1.0 -26.6 ( 1.0 36.0 ( 1.5 -1.00

s

+1155 ( 150

s

+680 ( 20

+550 ( 20

s

e

I₃In(TMP)

I₃In(TDP)

I₃In[P(o-Anis)₃]

I₃In[P(p-Anis)₃]

-59.6 ( 0.5 -40.6 ( 0.5 -67.6 ( 0.5 -70.6 ( 0.5 30.0 ( 0.7 -0.80 ( 0.10 +150 ( 50

-68.0 ( 2.0 -43.9 ( 2.0 -78.1 ( 2.0 -81.9 ( 2.0 38.0 ( 3.0 -0.80 ( 0.10 +240 ( 50

-37.2 ( 1.0 -19.5 ( 1.0 -44.7 ( 1.0 -47.5 ( 1.0 28.0 ( 1.5 -0.80 ( 0.10 +220 ( 50

-25.1 ( 1.0 -3.7 ( 1.0 -35.0 ( 1.0 -36.7 ( 1.0 33.0 ( 1.5 -0.90 ( 0.10 +150 ( 50

+1720 ( 20 +1329 ( 300

+1620 ( 30 +1059 ( 300

+1080 ( 20 +1119 ( 300

+820 ( 20 +1398 ( 200

+593 ( 30

+616 ( 20

s

f

-25.2 ( 1.0

-23.5 ( 0.5 -0.8 ( 0.5 -34.8 ( 0.5 -34.8 ( 0.5 34.0 ( 0.7 -1.00

-30.0 ( 2.0

s

+240 ( 50

s

+620 ( 20

+680 ( 20 +1059 ( 200

+620 ( 20

s

I₃In[P(m-Anis)₃]

g

+593 ( 20

s

I₃In(PPh₃)

h

s

^aCalculated from the In-P bond length according to eq 1. A correction of -2.5% has been applied to R_DD, see text. ^bCalculated as described in

footnote a, assuming that the In-P bond length is equal to that for Cl₃In[P(o-Anis)₃]. ^cCalculated as described in footnote a, assuming that the In-P

bond length is equal to that for Br₃In[P(p-Anis)₃]. ^dSee ref 27; the sign for the alternative value (see text) in this reference is incorrect. ^eCalculated as

described in footnote a, assuming that the In-P bond length is equal to that for I₃In(PPh₃). ^fI₃In[P(m-Anis)₃] in a mixture of I₃In[P(m-Anis)₃] and

I₃In[P(m-Anis)₃]₂, see text. ^gFree I₃In(PPh₃). ^hI₃In(PPh₃) in I₃In(PPh₃)₂·I₃In(PPh₃).

Figure 8. Experimental and calculated ³¹P NMR spectra of (a) MAS and (b) stationary powder samples of I₃In(PPh₃), acquired at 7.05 T.

bond, ꢀ^D) 0 and d ) (3C_QR_eff)/(10V_S) (eq 3). If the sign of

R_eff(¹¹⁵In,³¹P) is known, the sign of C_Q(¹¹⁵In) can also be

determined. See Table S4 in Supporting Information for a

summary of the d values. Since the magnitudes of C_Q(¹¹⁵In) for

the adducts investigated here were determined from analyses

of the ¹¹⁵In NMR spectra, the R_eff(¹¹⁵In,³¹P) values can be

determined from the analyses of ³¹P NMR spectra of MAS or

stationary samples. A comparable treatment is possible for

molecules possessing an approximate C₃axis that incorporates

the indium nucleus.

yield bond lengths that are larger than those obtained from X-ray

or neutron diffraction experiments.⁶³Earlier we suggested a

correction of 2.5 ( 1.5% to R_effdue to librational effects on

C-N bonds.⁶⁴Librational effects are expected to be signiﬁcantly

less for In-P bonds than for C-H, N-H or C-N bonds

because the former is more rigid. A correction of 2.5%, applied

to R_DD(

¹¹⁵In,³¹P) values used in the analysis (see below) is

thought to be an upper limit to the correction.

As an example, a detailed discussion of the analysis of the

data for the 1:1 adduct I₃In(PPh₃) is presented. Analysis of the

When considering errors in experimental values of R_effor

R_DD, one should be aware that librational effects⁶²tend to reduce

the direct dipolar interaction; hence values determined from

room-temperature NMR measurements are also reduced and

(62) Manz, J. J. Am. Chem. Soc. 1980, 102, 1801–1806.

(63) (a) Ishii, Y.; Terao, T.; Hayashi, S. J. Chem. Phys. 1997, 107, 2760–

2774. (b) Zilm, K. W.; Grant, D. M. J. Am. Chem. Soc. 1981, 103,

2913–2922.

9

5488 J. AM. CHEM. SOC. VOL. 132, NO. 15, 2010

¹¹⁵In and ³¹P NMR Studies of In-P Complexes

A R T I C L E S

Figure 9. Experimental and calculated ³¹P NMR spectra of MAS powder samples of (a) Cl₃In[P(o-Anis)₃], (b) Cl₃In[P(o-Anis)₃]·P(o-Anis)₃, (c) Br₃In[P(o-

Anis)₃] and (d) I₃In[P(o-Anis)₃], acquired at 11.75 T. The peak with the higher intensity seen in (b) is attributed to uncoordinated P(o-Anis)₃in Cl₃In[P(o-

Anis)₃]·P(o-Anis)₃.

∆J(¹¹⁵In,³¹P) and ¹J(¹¹⁵In,³¹P) therefore have similar magnitudes

and are both positive. Note that the value for ∆J(¹¹⁵In,³¹P) may

be considered a minimum, since in the absence of the correction

for librational motion, a larger value would be obtained, even

if the lesser of the two possible solutions were used. Comparable

1

∆J(¹¹⁵In,³¹P) and J(¹¹⁵In,³¹P) values were also observed for

InP.⁶⁸Because the effects of residual dipolar coupling are

apparent in ³¹P NMR spectra of I₃In(PPh₃) acquired with MAS

at lower ﬁelds, a value for C_Q(¹¹⁵In) had to be included in the

analysis, which indicated that C_Q(¹¹⁵In) is positive, consistent

with the results obtained from DFT calculations (Vide infra).

A comparable analysis was performed for Cl₃In[P(o-

Anis)₃]·P(o-Anis)₃, Br₃In(TMP), Br₃In[P(p-Anis)₃], and I₃In[P(p-

Anis)₃]; the values for ∆J(¹¹⁵In,³¹P) are summarized in Table

2. The analysis outlined in the previous paragraph assumes that

the J and dipolar tensors are coincident and axially symmetric.

With the exception of Br₃In[P(m-Anis)₃], the asymmetry in the

indium EFG and shielding tensors for the latter compounds

indicate that this may not be the case, but X-ray data (Supporting

Figure 10. Experimental and calculated ³¹P NMR spectra of MAS powder

Information) indicate that these have near C₃symmetry. The

samples of (a) Cl₃In[P(p-Anis)₃]₂and (b) Cl₃In[P(m-Anis)₃]₂, acquired at

large difference between R_effand R_DDmust be a consequence

7.05 T.

of ∆J(¹¹⁵In,³¹P), and since the analysis yields comparable values

of ∆J(¹¹⁵In,³¹P) for closely related adducts, the deviation from

axial symmetry is not thought to be sufﬁcient to signiﬁcantly

spectra of MAS and stationary samples (Figure 8) yielded values

of 680 ( 20 and 240 ( 50 Hz for ¹J(¹¹⁵In,³¹P) and R_eff,

respectively. Since there is a C₃symmetry axis along the In-P

affect the values obtained. These analyses assumed that the signs

bond of this adduct,⁴²we may estimate the value for

of ¹J(¹¹⁵In,³¹P) for these adducts are positive; DFT calculations

∆J(¹¹⁵In,³¹P) from the relationship ∆J ) 3(R_DD- R_eff). From

(Vide infra) support this assumption. For Br₃In[P(p-Anis)₃], the

the known In-P bond length (2.603 Å),⁴²an uncorrected value

of 608 ( 20 Hz is obtained for R_DD; this is reduced to 593 (

20 Hz after applying the correction for librational motion

discussed in the preceding paragraph. Hence, values for ∆J of

+1059 ( 200 or +2499 ( 200 Hz are obtained for ∆J,

depending on the sign of R_eff. Since an analysis of ³¹P NMR

spectra of stationary samples showed that ¹J(¹¹⁵In,³¹P) and

R_eff(¹¹⁵In,³¹P) have the same sign, and since one-bond spin-spin

coupling constants in analogous compounds are positive,^65,66

¹J(¹¹⁵In,³¹P) and R_effare assumed to be positive;^27,67thus, the

smaller of the two possible values for ∆J is more probable.

ten peaks in the ³¹P MAS NMR spectrum due to the coupling

between ³¹P and ¹¹⁵In are equally spaced, as expected from the

C_Q(¹¹⁵In) values of ( 1.25 ( 0.10 MHz determined from an

analysis of the ¹¹⁵In spectra; in this case the small value for d

precludes a determination of a sign for C_Q(¹¹⁵In).

We were unable to obtain X-ray crystallography data for some

1:1 adducts, and thus the value for the In-P bond length was

not determined. Examination of the data for other 1:1 adducts

(67) (a) Jameson, C. J. Spin-Spin Coupling. In Multinuclear NMR; Mason,

J., Ed.; Plenum Press: New York, 1987; Chapter 4, pp 89-131. (b)

Jameson, C. J. Theoretical Considerations: Spin-Spin Coupling. In

Phosphorus-31 NMR Spectroscopy in Stereochemical Analysis, Or-

ganic Compounds and Metal Complexes; Verkade, J. G.; Quin, L. D.,

Eds.; Methods in Stereochemical Analysis, Vol. 8; VCH Publishers,

Inc.: Deerﬁeld Beach, FL, 1987; Chapter 6, pp 205-230.

(68) (a) Tomaselli, M.; deGraw, D.; Yarger, J. L.; Augustine, M. P.; Pines,

A. Phys. ReV. B 1998, 58, 8627–8633. (b) Iijima, T.; Hashi, K.; Goto,

A.; Shimizu, T.; Ohki, S. Chem. Phys. Lett. 2006, 419, 28–32.

(64) Bryce, D. L.; Wasylishen, R. E. Inorg. Chem. 2002, 41, 4131–4138.

(65) (a) No¨th, H.; Wrackmeyer, B. NMR Basic Princ. Prog. 1978, 14, 1–

462. (b) McFarlane, H. C. E.; McFarlane, W.; Rycroft, D. S. J. Chem.

Soc., Faraday Trans. 2 1972, 68, 1300–1305.

(66) (a) Rudolph, R. W.; Schultz, C. W. J. Am. Chem. Soc. 1971, 93, 6821–

6822. (b) Cowley, A. H.; Damasco, M. C. J. Am. Chem. Soc. 1971,

93, 6815–6821.

9

J. AM. CHEM. SOC. VOL. 132, NO. 15, 2010 5489

A R T I C L E S

Chen et al.

(Table S2 in Supporting Information) indicates that there is little

variation in the In-P bond lengths for a given halide. Hence,

values for ∆J(¹¹⁵In,³¹P) have been estimated based on the

smallest corrected value of R_DD(and hence the smallest value

for ∆J(¹¹⁵In,³¹P)) predicted from the In-P bond lengths for a

given halide and from the measured R_effvalues; these are

summarized in Table 2.

the P-In-P bond, and thus we cannot assume that ꢀ^D≈ 0.

Hence, signs for C_Q(¹¹⁵In) could not be determined for these

adducts.

Investigators have reported relationships between the electron-

donating ability of phosphines and various molecular proper-

ties.⁷⁰For example, early studies found a relationship between

1

the base strength of phosphines toward BH₃and J(³¹P,¹¹B);⁶⁶

a similar relationship has also been reported between

¹J(¹⁹⁹Hg,³¹P) and the basicity of the phosphine ligand.⁷¹Wada

and Higashizaki found that the basicity of triarylphosphines

increases with the level of methoxy substitution.⁷²Hence, the

basicity of the ligands considered in this study is expected to

increase according to: PPh₃< P(o-Anis)₃≈ P(m-Anis)₃≈ P(p-

Anis)₃< TDP < TMP. For the adducts investigated here, the

¹J(¹¹⁵In,³¹P) values generally increase in the same order.

Relationships between one-bond spin-spin coupling and the

nature of the bond between the nuclei of interest have been

explained by the assumption that the former is dominated by

the Fermi-contact mechanism;⁷³this conclusion is consistent

with the results of DFT calculations for the adducts investigated

here (Vide infra). However, one must consider that the Fermi

contact term contributes only to J_iso;^36,37the observation of

∆J(¹¹⁵In,³¹P) for several of these adducts indicates that mech-

anisms other than the Fermi-contact term also make signiﬁcant

contributions to the In-P J tensor.^27,37

Phosphorus-31 NMR spectra of MAS samples for several

X₃In(PR₃) adducts are shown in Figure 9; the ³¹P NMR

parameters for these are summarized in Table 2. The ³¹P NMR

peaks for Cl₃In[P(o-Anis)₃]·P(o-Anis)₃, Br₃In[P(o-Anis)₃] and

I₃In[P(o-Anis)₃] are signiﬁcantly broader than those for Cl₃In[P(o-

Anis)₃]; this is attributed to the more rapid relaxation of the

spin states for the ¹¹⁵In nuclei and has been reported previously

for ¹³C and ³¹P coupled to ⁵⁹Co.^24,69X₃In(PR₃) adducts with

values for η_Qof approximately 0 must have an approximate C₃

axis along the In-P bond with V_ZZapproximately aligned with

this bond (i.e., ꢀ^D≈ 0). In these cases, the signs for C_Q(¹¹⁵In)

could be determined from analyses of the ³¹P MAS NMR

spectra; these are listed in Table 1. DFT calculations accurately

predicted these signs (Vide infra).

As expected, the isotropic regions of ³¹P NMR spectra of

X₃In(PR₃)₂adducts with two nonequivalent phosphorus nuclei

contain twenty peaks. An interesting example is that for

Cl₃In[P(p-Anis)₃]₂acquired at 7.05 T, shown in Figure 10a,

which contains 20 approximately equally spaced peaks; if the

nuclear magnetic properties of ¹¹⁵In were not known, one might

conclude from this spectrum that ¹¹⁵In has a spin quantum

number I ) 19/2! Of course, from the known properties of ¹¹⁵In,

it is apparent that this unusual pattern is a consequence of the

fortuitous overlap of the peaks from two distinct ³¹P NMR sites,

with a chemical shift difference (in Hz) between them of either

0.5(¹J(¹¹⁵In,³¹P)) or 10(¹J(¹¹⁵In,³¹P)). Analysis of ³¹P NMR

spectra acquired at different magnetic ﬁelds indicates that the

former is the case (i.e., every second peak in Figure 10a is

assigned to one ³¹P site coupled to ¹¹⁵In); the slight differential

broadening of the peaks, discussed above, is consistent with

this assignment. The observation of two distinct ³¹P NMR sites

for this adduct is contrary to what is expected from the X-ray

diffraction data, which indicates that the two phosphorus atoms

are related by C₂symmetry and hence should have identical

δ_isovalues. The single crystal observed Via X-ray diffraction

included a methylene chloride molecule of solvation (see

Supporting Information), but there is no indication of its

presence in the EA data. Hence, the methylene chloride is

thought to have been lost from the sample during its preparation

for the EA and NMR analyses. The solid-state structure of the

sample analyzed by ³¹P NMR is clearly different from that

observed by X-ray diffraction.

1

Summarizing, J(¹¹⁵In,³¹P) values have been determined for

all X₃In(PR₃) and X₃In(PR₃)₂adducts investigated here. In

addition, when sufﬁcient data were available, ∆J(¹¹⁵In,³¹P)

values were estimated, and signs for this parameter and for

1

¹J(¹¹⁵In,³¹P) were proposed. The J(¹¹⁵In,³¹P) and ∆J(¹¹⁵In,³¹P)

values reported here are of comparable magnitude and positive.

Signs for C_Q(¹¹⁵In) have also been determined for those adducts

that have high-order symmetry and for which the residual dipolar

coupling was observable.

DFT Calculation of ¹¹⁵In EFG Tensors. Calculated ¹¹⁵In EFG

tensors are summarized in Table 3; a comparison of experi-

mental and calculated C_Qvalues is shown in Figure 11. Although

some calculated values deviate signiﬁcantly from the corre-

sponding experimental values, the sign for C_Qwas accurately

predicted for those samples for which it was determined

experimentally. Hence, these results suggest that the experi-

mental procedure used here reliably predicts the sign for

C_Q(¹¹⁵In). The calculated η_Qvalues are also qualitatively in

agreement with experiment, as are the calculated relative

orientations between the EFG and CS tenors, deﬁned by the

Euler angles R, ꢀ, γ (Tables 1 and 3).

DFT Calculation of Indium CS Tensors. Because an absolute

magnetic shielding scale for indium has not been established,

to compare calculated and experimental indium CS tensors, the

calculated indium magnetic shielding value for In(acac)₃, 3840

ppm, was arbitrarily set to its experimental value, δ_iso(In) ) 35

ppm;¹⁹all other calculated magnetic shielding values were

converted to chemical shifts according to δ_iso(In) ) 3840 -

σ_iso(In), see Tables 3 and 4. Plots of experimental versus

calculated δ_iso(In), κ, and Ω are shown in Figure 12; these

indicate that calculated CS parameters are qualitatively in

Elemental analysis indicates that the molecular formula of

the adduct formed by InCl₃and P(m-Anis)₃is Cl₃In[P(m-

Anis)₃]₂. Although single-crystal X-ray diffraction data for this

adduct are unavailable, the presence of only 10 peaks in the

³¹P NMR spectrum of an MAS sample (Figure 10b) indicates

that the two ³¹P nuclei must be crystallographically equivalent,

1

or very nearly so. The J(¹¹⁵In,³¹P) values for the two ³¹P sites

of Cl₃In(PPh₃)₂and Br₃In(PPh₃)₂are also summarized in Table

2. Since the indium EFG tensors for the X₃In(PR₃)₂adducts

(Table 1) are not axially symmetric, there is no C₃axis along

(70) Ku¨hl, O. Coord. Chem. ReV. 2005, 249, 693–704.

(71) (a) Allman, T.; Goel, R. G. Can. J. Chem. 1984, 62, 615–620. (b)

Allman, T.; Goel, R. G. Can. J. Chem. 1984, 62, 621–627.

(72) Wada, M.; Higashizaki, S. J. Chem. Soc., Chem. Commun. 1984, 482–

483.

(69) Schurko, R. W.; Wasylishen, R. E.; Nelson, J. H. J. Phys. Chem. 1996,

100, 8057–8060.

(73) Chen, F.; Oh, S.-W.; Wasylishen, R. E. Can. J. Chem. 2009, 87, 1090–

1101.

9

5490 J. AM. CHEM. SOC. VOL. 132, NO. 15, 2010

¹¹⁵In and ³¹P NMR Studies of In-P Complexes

Table 3. Calculated ¹¹⁵In NMR Parameters

A R T I C L E S

C_Q/MHz^aη_Q

σ

_iso/ppm

σ_dia/ppm

σ_para/ppm

σ

_spin-orbit/ppm

δ

_iso/ppm^bδ₁₁/ppm δ₂₂/ppm δ₃₃/ppm Ω/ppm

κ

R/deg ꢀ/deg γ/deg

In(acac)₃

Cl₃In(TMP)

+97.3 0.21 3875

-112.3 0.17 3394

s

-35

446

439

450

225

244

314

-191

-295

308

4

473

466

494

118

318

362

83

-49

453

438

454

68 -111

258

290

75 -732

-3 -880

124

218

26

-59

412

414

402

63 -0.67 57

61 0.36 75

52 -0.08 81

86

17

49

87

23

57

90

19

0

82

88

86

0

15

5

39

26

0

21

0

1

4899

4891

4895

4897

4896

4897

4895

4892

4896

4897

-2086

-2083

-2094

-2180

-2225

-2299

-2407

-2386

-1970

-2031

-2093

-2138

581

593

900

926

928

1542

1624

607

595

881

1653

Cl₃In[P(o-Anis)₃] · P(o-Anis)₃-63.0 0.51 3401

Cl₃In[P(o-Anis)₃]

Br₃In(TMP)

Br₃In[P(o-Anis)₃]

Br₃In[P(p-Anis)₃]

I₃In[P(p-Anis)₃]

I₃In(PPh₃)

Cl₃In(PMe₃)₂

Br₃In(PMe₃)₂

I₃In(PMe₃)₂

-42.5 0.86 3390

-124.8 0.13 3615

-41.6 0.37 3596

92

229

163

0.13

4

0.56 77

0.26 84

155

290

+27.5

-20.0 0.04 4031

+67.5 4135

0

3526

72 -1.00 90

815 0.98 58

877

666 -0.83

0

-3

1.00

0

8

c

+82.2 0.14 3532

-46.4 0.83 3456

-89.2 0.26 3684

733

772

467

67

163

-25

d

384

156

-572

609 -0.82 87

492 -0.79 55

4

0

e

f

+89.1

0

4412

-445 -445 -827

382

1.00

0

^aCalculated C_Qvalues have been scaled by a factor of 0.951 to correspond to a more recent reported value for Q,⁹see text. ^bδ_iso(In)/ppm ) 3840 -

_iso(In)/ppm, see ref 19. ^cThe structure for the Cl₃InP₂core is that for Cl₃In[P(p-Anis)₃]₂. ^dThe structure for the Cl₃InP₂core is that for Cl₃In(PPh₃)₂.

σ

^eThe structure for the Br₃InP₂core is that for Br₃In(PPh₃)₂. ^fThe structure for the I₃InP₂core is that for I₃In(PPh₃)₂.

X₃In(PMe₃) and 2:1 X₃In(PMe₃)₂(X ) Cl, Br or I) adducts, as

shown in Figure 13 and summarized in Table 4.

The increasing indium isotropic shielding for the In(III)

complexes with increasing halogen atomic number (normal

halogen dependence) is in contrast to what has been observed

for InX monohalides, where the shielding decreases, a result

that is consistent with DFT calculations (Table S4 of Supporting

Information). The latter, referred to as the inverse halogen

dependence, IHD, has been reported for B-, Al-, and Ga-halides,

where the metal is in the +1 oxidation state.⁷⁴Kaupp and co-

workers have undertaken DFT calculations of heavy atom effects

on nuclear magnetic shielding.⁷⁵The IHD of the isotropic

chemical shifts for the monohalides is a consequence primarily

of the far greater deshielding effect of the paramagnetic term,

but the trend in the spin-orbit term is also opposite to that for

the In(III) adducts, decreasing as the halogen atomic number

increases.

DFT Calculations of ¹J(¹¹⁵In,³¹P) and ∆J(¹¹⁵In,³¹P). The

calculated ¹J(¹¹⁵In,³¹P) and ∆J(¹¹⁵In,³¹P) values for the

Cl₃In(PR₃) and Br₃In(PR₃) adducts, as well as their signs, are

listed in Table 5. These values differ from those measured

experimentally but qualitatively reproduce their experimental

Figure 11. Experimental vs calculated C_Q(¹¹⁵In) values for the triarylphos-

phine indium trihalide adducts listed in Table 3, as well as for previously

reported¹⁹In(trop)₃and I₃In[OP(CH₃)₃]₂(the two points in the upper right);

the diagonal line designates perfect agreement between experimental and

calculated values. The solid circles indicate points where the sign for C_Q

was determined experimentally; those with open circles are points where

the sign was assumed to be that of the calculated value. With this

assumption, the standard deviation (s ) [∑(calc - exp)²/(n - 1)]^1/2where

n is the number of data points) is 38 MHz.

1

trends. Speciﬁcally, calculated J(¹¹⁵In,³¹P) and ∆J(¹¹⁵In,³¹P)

values are of similar magnitudes and are positive, in agreement

with experimental results. In addition, the correlation between

1

the J(¹¹⁵In,³¹P) values and the basicity of the triarylphosphine

ligands of Cl₃In(PR₃) and Br₃In(PR₃) is reproduced by calcula-

1

tions. The calculated contribution to J(¹¹⁵In,³¹P)_isofrom the

agreement with the corresponding experimental values. The

experimental trends for δ_iso(In) and Ω for X₃In(PR₃) and

X₃In(PR₃)₂, shown in Figure 7, are reproduced in the calculated

results listed in Table 3; in particular, the calculated spans for

the indium CS tensors of the 1:1 adducts increase from X ) Cl

to X ) I, but the latter has the smallest span for the

trigonal-bipyramidal structures. To investigate the causes of

the observed trend in the indium magnetic shielding, the

diamagnetic, paramagnetic, and spin-orbital contributions to

the indium isotropic magnetic shielding were calculated (Table

3). These indicate that as X changes from Cl to I, the

diamagnetic term is virtually constant and the paramagnetic term

decreases, but the spin-orbital term makes the greatest contri-

bution to the increased magnetic shielding and is the most

variable, accounting for the halogen effect³³discussed above.

Trends in the experimental indium CS tensors (Figure 7) are

also reproduced by the calculated values for the model 1:1

Fermi-contact mechanism for these model adducts is about 99%,

in agreement with the assumption that this mechanism domi-

nates. Calculations also indicate nonzero ∆J(¹¹⁵In,³¹P) values;

these values are mainly due to the spin-dipolar Fermi-contact

(SD × FC) cross term; the contribution of this term to

∆J(¹¹⁵In,³¹P) is also about 99%. Thus, both Fermi-contact and

spin-dipolar Fermi-contact mechanisms make important con-

(74) Gee, M.; Wasylishen, R. E. Aluminum Magnetic Shielding Tensors

and Electric Gradients for Aluminum(I) Hydride, Aluminum(I) Iso-

cyanide, and the Aluminum(I) Halides: Ab Initio Calculations. In

Modeling NMR Chemical Shifts. Gaining Insights into Structure and

EnVironment; Facelli, J. C., de Dios, A. C., Eds.; ACS Symposium

Series 732; American Chemical Society: Washington, DC, 1999;

Chapter 19, pp 259-276.

(75) (a) Kaupp, M.; Malkina, O. L.; Malkin, V. G.; Pyykko¨, P. Chem.sEur.

J. 1998, 4, 118–126. (b) Kaupp, M.; Malkina, O. L.; Malkin, V. G.

Chem. Phys. Lett. 1997, 265, 55–59.

9

J. AM. CHEM. SOC. VOL. 132, NO. 15, 2010 5491

A R T I C L E S

Chen et al.

Table 4. Calculated Indium NMR Parameters for Model 1:1 X₃In(PMe₃) and 2:1 X₃In(PMe₃)₂(X ) Cl, Br, and I) Adducts

σ

_iso/ppm

σ

_dia/ppm

σ

_para/ppm

σ_spin-orbit/ppm

δ

_iso/ppm^a

δ₁₁/ppm

δ₂₂/ppm

δ₃₃/ppm

Ω/ppm

κ

Cl₃In(PMe₃)

Br₃In(PMe₃)

I₃In(PMe₃)

Cl₃In(PMe₃)₂

Br₃In(PMe₃)₂

I₃In(PMe₃)₂

3292

3526

4183

3438

3677

4037

4895

4897

4898

4893

4896

4895

-2198

-2297

-2418

-2046

-2091

-2196

595

928

1703

591

872

1338

548

314

-343

402

163

-197

613

361

-102

782

498

-142

515

290

-102

212

-5

-142

515

290

-825

212

-5

-306

98

71

723

570

503

164

-1.00

1.00

-1.00

1.00

^aδ_iso(In)/ppm ) 3840 - σ_iso(In)/ppm, see ref 19.

-166.0 ( 2.0 MHz; the signs of C_Q(¹¹⁵In) for some adducts

with an exact or approximate C₃axis were determined from

analyses of the ³¹P NMR spectra. For any given phosphine

ligand, the indium nuclei are most shielded for X ) I and least

shielded for X ) Cl, a trend also observed for other group-13

elements in the +3 oxidation state. The anisotropic indium

magnetic shielding is measurable for all adducts, with spans as

large as 710 ( 60 ppm. For adducts with X ) I or possessing

trigonal-bipyramidal structures, Ω values are signiﬁcant; at

21.14 T, the contribution of anisotropic shielding to the ¹¹⁵In

NMR spectra is of the same magnitude as the contribution from

1

C_Q(¹¹⁵In). J(¹¹⁵In,³¹P) values were determined for all adducts

investigated here; ∆J(¹¹⁵In,³¹P) values and their signs were also

determined for several of these. The ¹J(¹¹⁵In,³¹P) and

∆J(¹¹⁵In,³¹P) values are comparable in magnitude and positive.

The former generally increase with increasing basicity of the

triarylphosphine ligands for X₃In(PR₃) and X₃In(PR₃)₂for any

given X.

Theoretical calculations qualitatively reproduce the available

experimental ¹¹⁵In EFG and CS tensors, including their relative

orientations. The theoretical calculations also reproduce the

observed relationship between the isotropic indium magnetic

shielding and the halogen ligands, attributed to the spin-orbital

1

effect of the halogen ligand. Calculated values of J(¹¹⁵In,³¹P)

and ∆J(¹¹⁵In,³¹P) are signiﬁcantly lower than the experimental

values; however, calculations undertaken for the Cl₃In(PR₃) and

1

Br₃In(PR₃) adducts conﬁrm that J(¹¹⁵In,³¹P) and ∆J(¹¹⁵In,³¹P)

are comparable in magnitude with positive signs and reproduce

1

the apparent relation between J(¹¹⁵In,³¹P) and the basicity of

the triarylphosphine ligands. Calculations indicate that both the

Fermi-contact and spin-dipolar Fermi-contact mechanisms are

important factors when considering the J tensor between coupled

indium and phosphorus nuclei; however, J_isovalues are governed

by the efﬁciency of the Fermi-contact mechanism.

The present study has provided an interpretation of solid-

state ¹¹⁵In NMR spectra for general indium-phosphine com-

plexes. Analyses of these spectra allowed the determination of

indium quadrupolar and CS tensors. Our study has shown that

the effect of indium magnetic shielding anisotropy cannot be

ignored when interpreting ¹¹⁵In NMR spectra, particularly for

spectra acquired at high magnetic-ﬁeld strengths. We also have

demonstrated that relativistic DFT calculations are helpful in

the analysis of ¹¹⁵In NMR spectra. In addition, NMR investiga-

tions of spin-1/2 nuclei such as ³¹P coupled to indium comple-

ment ¹¹⁵In NMR data; these can yield important NMR infor-

Figure 12. Experimental vs calculated indium CS tensor values (a) δ_iso

,

(b) κ, and (c) Ω for the triarylphosphine indium trihalide adducts X₃In(PR₃)

and X₃In(PR₃)₂; the diagonal line indicates perfect agreement between

experimental and calculated results. The standard deviations (see the caption

to Figure 11) are 96 ppm, 0.2, and 123 ppm for δ_iso, κ, and Ω, respectively.

tributions to the J tensors between coupled indium and

phosphorus nuclei.

Conclusions

A series of triarylphosphine indium trihalide adducts,

X₃In(PR₃) and X₃In(PR₃)₂(X ) Cl, Br or I, PR₃) triarylphos-

phine ligand), have been investigated using solid-state ¹¹⁵In and

³¹P NMR spectroscopy and single-crystal X-ray diffraction. To

complement the experimental results, relativistic density func-

tional theory (DFT) calculations have also been performed. The

C_Q(¹¹⁵In) values for these adducts range from (1.25 ( 0.10 to

1

mation, such as J(¹¹⁵In,³¹P), ∆J(¹¹⁵In,³¹P), and the signs for

∆J(¹¹⁵In,³¹P) and C_Q(¹¹⁵In), which are difﬁcult or impossible to

determine from the analyses of the experimental ¹¹⁵In NMR

spectra, and they may also provide estimates of C_Q, corroborat-

ing conclusions reached from analyses of the ¹¹⁵In NMR spectra.

Finally, this study clearly demonstrates the advantages of

9

5492 J. AM. CHEM. SOC. VOL. 132, NO. 15, 2010

¹¹⁵In and ³¹P NMR Studies of In-P Complexes

A R T I C L E S

Figure 13. Calculated spans for the indium CS tensors for model adducts.

1

Table 5. Calculated J(¹¹⁵In,³¹P) and ∆J(¹¹⁵In,³¹P) Values for

(Ottawa, Canada), funded by the Canada Foundation for Innovation,

the Ontario Innovation Trust, Recherche Que´bec, the National

Research Council of Canada, and Bruker BioSpin and managed

by the University of Ottawa (www.nmr900.ca). R.G.C. and R.E.W.

thank the Natural Sciences and Engineering Research Council of

Canada (NSERC) for ﬁnancial support through the Discovery Grant

Program and a Major Resources Support grant. R.E.W. also thanks

the Canada Research Chairs program and the University of Alberta

for research support.

X₃In(PR₃) (X ) Cl and Br) Adducts

¹J(¹¹⁵In,³¹P)/Hz

∆J(¹¹⁵In,³¹P)/Hz

Cl₃In(TMP)

Cl₃In(TDP)

+1488

+1329

+1082

+822

+597

+604

+1225

+1042

+708

+456

+351

+446

+1250

+1183

+1066

+1033

+923

Cl₃In[P(o-Anis)₃]

Cl₃In[P(p-Anis)₃]

Cl₃In(PPh₃)

Cl₃In[P(m-Anis)₃]

Br₃In(TMP)

+916

+1165

+1085

+951

Br₃In(TDP)

Br₃In[P(o-Anis)₃]

Br₃In[P(p-Anis)₃]

Br₃In(PPh₃)

Supporting Information Available: Elemental analysis results;

selected X-ray experimental and structural data, as well as the

CIFs, for Br₃In(TMP), Cl₃In[P(o-Anis)₃]·P(o-Anis)₃, Cl₃In[P(o-

Anis)₃], Br₃In[P(o-Anis)₃], Cl₃In[P(p-Anis)₃]₂, Br₃In[P(p-Anis)₃],

I₃In[P(p-Anis)₃], Cl₃In(PPh₃)₂and Br₃In(PPh₃)₂; summary of d

values; experimental and computational indium magnetic shield-

ing parameters for InX monohalides; discussion of the dif-

ferential line broadening of the ³¹P NMR spectra; and additional

¹¹⁵In and ³¹P NMR spectra for all adducts. This material is

available free of charge via the Internet at http://pubs.acs.org.

+891

+832

Br₃In[P(m-Anis)₃]

+881

performing ¹¹⁵In NMR experiments at the highest possible

magnetic ﬁeld strength.

Acknowledgment. We thank Victor Terskikh and Eric Ye for

acquiring some ¹¹⁵In NMR spectra. We acknowledge Klaus Eichele,

Gang Wu, and Jason Clyburne for some preliminary ³¹P NMR work

on this project. Access to the 900 MHz NMR spectrometer was

provided by the National Ultrahigh-Field NMR Facility for Solids

JA100625P

9

J. AM. CHEM. SOC. VOL. 132, NO. 15, 2010 5493

Article Doi

Solid-state 115In and 31P NMR studies of triarylphosphine indium trihalide adducts

DOI: 10.1021/ja100625p

Source and publish data:

Authors:

Article abstract of DOI:10.1021/ja100625p

Full text of DOI:10.1021/ja100625p

Products guided by the article

R&D Labs maybe for 1235485-99-0

Relevant to this article

Hot Product