Harnessing low-valent metal centers through non-bonding orbital interactions

Article abstract of DOI:10.1021/ic1016438

The synthesis, characterization, and computational analysis of a series of low-valent, In(I) complexes bearing the bis(imino)pyridine scaffold, {Ar′N=CPh}₂(NC₅H₃), is reported. A stepwise steric reduction of the aryl group

Full text of DOI:10.1021/ic1016438

Inorg. Chem. 2010, 49, 10635–10641 10635
DOI: 10.1021/ic1016438
Harnessing Low-Valent Metal Centers through Non-Bonding Orbital Interactions
Titel Jurca,† Ilia Korobkov,† Glenn P. A. Yap,‡ Serge I. Gorelsky,*,† and Darrin S. Richeson*,†
†Centre for Catalysis Research and Innovation and Department of Chemistry, University of Ottawa, Ottawa,
Ontario K1N 6N5, Canada, and ‡Department of Chemistry and Biochemistry, University of Delaware, Newark,
Delaware 19716, United States
Received August 12, 2010
The synthesis, characterization, and computational analysis of a series of low-valent, In(I) complexes bearing the
bis(imino)pyridine scaffold, {Ar0NdCPh}2(NC5H3), is reported. A stepwise steric reduction of the aryl groups on the
imine substituents around the coordination site, (Ar0 = 2,5-tBu2C6H3, 2,6-iPr2C6H3, 2,6-(CH3CH2)2C6H3) is explored
through the spectroscopic and crystallographic examination of complexes [{Ar0NdCPh}2(NC5H3)]Inþ(OTf)- (1-3).
Compounds 1-3 displayed long In-N and In-OTf distances indicating only weak or no coordination. Application
of the ligand with Ar0 = 2,6-(CH3)2C6H3 led to an In(III) bis(imino)pyridine complex, [{2,6-Me2C6H3NdCPh}2-
(NC5H3)]In(OTf)2Cl 4 with coordinated ligand, chloride, and triflate groups. Computational analysis of the interactions
between the In cation and the ligands (orbital populations, bond order, and energy decomposition analysis) point to
only minimal covalent interactions of the In(I) cation with the ligands. Although it features three N donor centers, the
bis(imino)pyridine ligand provides little ligand-to-metal donation. A thorough electronic structure analysis revealed a
correlation of compound stability with the reduced contribution of the In(I) 5s lone electron pair to the highest occupied
molecular orbital (HOMO) of the cation. This effect, originating from non-bonding orbital interactions between the metal
and the ligand, is more prominent in sterically crowded environments. The discovery of this correlation may help in
designing new low-valent complexes.
Introduction
lone pair of electrons, three unoccupied valence orbitals, and
no tightly covalently bonded ligands. This unique metalloid
complex apparently possessed insubstantial Ge-N and
Ge-O bonding interactions. These efforts have recently
expanded to demonstrate that azamacrocycles and crown
ethers are suitable ligands for isolation of germanium
dication species.5,6 Our recent isolation and characterization
of [{2,5-tBu2C6H3NdCPh}2(NC5H3)]Inþ(OTf )- 1, bearing
a sterically demanding bis(imino)pyridine ligand, presented a
species with a single long In-N(pyridine) interaction and an
electronic structure for the cation indicating that the Inþ ion
accepts little covalent donor-acceptor interactions from the
ligand.
Our elaboration of these initial investigations is reported
herein. The modular steric features of the bis(imino)pyridine
scaffold are exploited with the synthesis of a systematic series
of these ligands displaying varying steric load. Application of
these ligands to the synthesis of a family of In complexes
revealed a correlation of stability of these In(I) species with
Challenges to conventional metal-ligand bonding interac-
tions are presented by the recent crystallographic characteriza-
tion and computational scrutinization of reactive species that
display only nominal classic Lewis acid/base coordination.1,2
These efforts raise fundamental bonding questions that in turn
inspire the synthesis of increasingly challenging target molecules.
The judicious choice of metal cation environment/ligand array
appears to play an essential role in these issues. In the case of
indium(I), the exceptionally sterically encumbering ortho-
terphenyl ligand, 2,6-(Tripp)2C6H3 (Tripp=2,4,6-iPr3C6H2),
allowed the isolation of the unique example of single-coordi-
nation of this metal cation as 2,6-(Tripp)2C6H3In(I).3 Slightly
reducing the bulk of the ligand to 2,6-(Dipp)2C6H3 (Dipp=
2,6-iPr2C6H3), yielded a “diindene” dimer [2,6-(Dipp)2C6H3In]2
with an In-In bond.4 The isolation of a cryptand encapsu-
lated Ge cation, (Ge cryptand[2.2.2])2þ(O3SCF3)-2, revealed
a species with the dicationic group 14 element possessing a
*To whom correspondence should be addressed. E-mail: darrin@
uottawa.ca (D.S.R.).
(1) Rupar, P. A.; Staroverov, V. N.; Baines, K. M. Science 2008, 322,
1360.
(2) Jurca, T.; Lummis, J.; Burchell, T. J.; Gorlesky, S. I.; Richeson, D. S.
J. Am. Chem. Soc. 2009, 131, 4608.
(3) Haubrich, S. T.; Power, P. P. J. Am. Chem. Soc. 1998, 120, 2202.
(4) Wright, R. J.; Phillips, A. D.; Hardman, N. J.; Power, P. P. J. Am.
Chem. Soc. 2002, 124, 8538.
(5) Cheng, F.; Hector, A. L.; Levason, W.; Reid, G.; Webster, M.; Zhang,
W. Angew. Chem., Int. Ed. 2009, 48, 5152. Rupar, P. A.; Bandyopadhyay, R.;
Cooper, B. F. T.; Stinchcombe, M. R.; Ragogna, P. J.; Macdonald, C. L. B.;
Baines, K. M. Angew. Chem., Int. Ed. 2009, 48, 5155.
(6) Crown ether complexes of In(I)-OTf are reported in (a) Andrews,
C. G.; Macdonald, C. L. B. Angew. Chem., Int. Ed. 2005, 44, 7453. (b) Cooper,
B. F. T; Macdonald, C. L. B. J. Organomet. Chem. 2008, 693, 1707.
r
2010 American Chemical Society
Published on Web 10/12/2010
pubs.acs.org/IC

10636 Inorganic Chemistry, Vol. 49, No. 22, 2010
Jurca et al.
to display minimal metal-ligand coordination. Notably, 1 is
monomeric with a long In-N2 bond distance of 2.495(5) A.8,9
The most direct comparisons to 1 can be made with In(I)
complexes of anionic ligands. For example, the In-N bond
distance in the two coordinate β-diketiminate complex
[In{(NDippCMe)2CH}] is 2.27 A.10 Furthermore, for the three
coordinate tris(3,5-di-tert-butylpyrazolyl)hydroborato In(I)
these distances averaged about 2.47 A11 and for the four coordi-
nate amidinato species [DippNC(tBu)NDipp]In(I) an In-N
distance of 2.329(5) A was observed.12 The In center in 1 lies
slightly out of the mean plane defined by the three N centers of
the ligand (26.1°). The In-Nimine distances, In-N1 2.747(5) A
and In-N3 2.689(5) A, are beyond the sum of the covalent radii
for In and N (covalent radius N = 0.71 A, In = 1.42 A).13
Furthermore, the long In-O(triflate) distance (In-O2
7.975 A) was consistent with no or very little coordination
of the anion.6,7 The shortest In/triflate contact observed for 1
is to an F atom of the CF3 group at a distance of 6.485 A.
Within 1, the imine CdN distances (N1-C1 1.281(9) A,
N3-C13 1.283(9) A) are comparable to those for the free
ligand, and the carbon centers of the imine moieties are
planar.2 The SO bonds within the triflate anion are essentially
equivalent and range from 1.406(7) A to 1.425(6) A. These
values are consistent with a non-coordinated anion. All of
these structural features point to the fact that 1 displayed a
very weakly coordinated In center in spite of the presence of
additional sites for coordination. Interestingly, the structural
features for the singly coordinated cation 1 are reminiscent of
the In(terphenyl) monomer 2,6-(Tripp)2C6H3In.3
Figure 1. X-ray structure of the [{2,5-tBu2C6H3NdC(Me)}2(NC5H3)]Inþ
cation of compound 1. Hydrogen atoms, tert-butyl groups, and triflate
counterion have been omitted for clarity.
Scheme 1. Bis(imino)pyridine Ligands Used in This Report
The periphery of the bis(imino)pyridine ligand A can be
readily modified to provide a systematic reduction in steric bulk
via stepwise removal of CH3 groups from the Nimine-Ar0
substituents. The new bis(imino)pyridines, B-D, were pre-
pared using a modified literature synthesis.2,14 The modular
steric features and the relative ease of synthesis for these species,
along with the fact that the application of this ligand frame-
work in main group chemistry is underdeveloped,15,16 further
inspired our efforts.
Scheme 2
Like A, ligands B and C also react smoothly with In-
(OSO2CF3), to provide orange compounds 2 and 3 in isolated
the complexity and steric demand of the ligand substituents.
Our thorough computational assessment of these com-
pounds expands on the concept that sterically encumbered
ligands can provide kinetic stabilization to reactive species
with only weak coordination and electron donation.
(8) For reviews of low valent In chemistry see: (a) Pardoe, J. A. J.; Downs,
A. J. Chem. Rev. 2007, 107, 2. (b) Schmidbaur, H. Angew. Chem., Int. Ed. Engl.
1985, 24, 893.
(9) This distance is similar to that in the metastable complex InBr TME-
3
DA. Green, S. P.; Jones, C.; Stasch, A. Chem. Commun. 2008, 6285.
(10) Hill, M. S.; Hitchcock, P. B. Chem Commun. 2004, 1818.
(11) Kuchta, M. C.; Dias, H. V. R.; Bott, S. G.; Parkin, G. Inorg. Chem.
1996, 35, 943.
Results and Discussion
The first low-valent, main group metal complex of the bis-
(imino)pyridine scaffold was recently prepared when the solu-
ble and readily accessible In(I) synthon, In(O3SCF3),7 was
allowed to react with ligand A (Schemes 1 and 2) to provide an
excellent isolated yield (76%) of the bright orange compound 1.2
The bulky steric environment provided by the 2,5-tBu2C6H3
substituents of the imine function in A were a key feature of the
observed species. The identity of this species as the bis(imino)-
pyridine complex [{2,5-tBu2C6H3NdCPh}2(NC5H3)]-
Inþ(OTf)- 1 was established by single crystal X-ray analysis,
and the results for the cationic component are summarized in
Figure 1 and Tables 1 and 2.
(12) Jones, C.; Junk, P. C.; Platts, J. A.; Rathmann, D.; Stasch, A. Dalton
Trans. 2005, 2497.
ꢀ
ꢀ
(13) Cordero, B.; Gomez, V.; Platero-Prats, A. E.; Reves, M.; Echeverrıa,
´
ꢀ
J.; Cremades, E.; Barragan, F.; Alvarez, S. Dalton Trans. 2008, 2832.
€
€
(14) Kleigrewe, N.; Steffen, W.; Blomker, T.; Kehr, G.; Frohlich, R.;
Wibbeling, B.; Erker, G.; Wasilke, J.-C.; Wu, G.; Bazan, G. C. J. Am. Chem.
Soc. 2005, 127, 13955.
(15) For examples of main group bis(imino)pyridine compounds see: (a)
Martin, C. D.; Le, C. M.; Ragogna, P. J. J. Am. Chem. Soc. 2009, 131, 15126.
(b) Reeske, G.; Cowley, A. H. Chem. Commun. 2006, 1784. (c) Reeske, G.;
Cowley, A. H. Chem. Commun. 2006, 4856. (d) Knijnenburg, Q.; Smits,
J. M. M.; Budzelaar, P. H. M. Organometallics 2006, 25, 1036. (e) Blackmore,
I. J.; Gibson, V. C.; Hitchcock, P. B.; Rees, C. W.; Williams, D. J.; White, A. J. P.
J. Am. Chem. Soc. 2005, 127, 6012. (f) Scott, J.; Gambarotta, S.; Korobkov, I.;
Knijnenburg, Q.; de Bruin, B.; Budzelaar, P. H. M. J. Am. Chem. Soc. 2005, 127,
17204. (g) Bruce, M.; Gibson, V. C.; Redshaw, C.; Solan, G. A.; White, A. J. P.;
Williams, D. J. Chem. Commun. 1998, 2523. (h) Baker, R. J.; Jones, C.; Kloth,
M.; Mills, D. P. New J. Chem. 2004, 28, 2017. (i) Knijnenburg, Q.; Smits,
J. M. M.; Budzelaar, P. H. M. C. R. Chim. 2004, 7, 865.
Through a combined structural investigation and density-
functional theory (DFT) analysis, In(I) complex 1 was revealed
(7) Macdonald, C. L. B.; Corrente, A. M.; Andrews, C. G.; Taylor, A.;
Ellis, B. D. Chem. Commun. 2004, 250.
€
(16) Abram, S.; Maichle-Mossmer, C.; Abram, U. Polyhedron 1997, 16, 2183.

Article
Inorganic Chemistry, Vol. 49, No. 22, 2010 10637
Table 1. Selected Bond Distances for Compounds 1-3 (A)
1
2
3
C(1)-C(8)
C(12)-C(13)
In(1)-N(2)
In(1)-N(1)
In(1)-N(3)
In(1)-O(2)
N(1)-C(1)
N(3)-C(13)
N(2)-C(8)
N(2)-C(12)
N(1)-C(20)
N(3)-C(34)
O(1)-S(1)
O(2)-S(1)
O(3)-S(1)
1.500(10)
1.488(10)
2.495(5)
2.747(5)
2.689(5)
7.975
1.281(9)
1.283(9)
1.329(9)
1.340(9)
1.426(9)
1.432(9)
1.425(6)
1.408(7)
1.419(8)
C(13)-C(20)
C(24)-C(25)
In(1)-N(2)
In(1)-N(1)
In(1)-N(3)
In(1)-O(1)
N(1)-C(13)
N(3)-C(25)
N(2)-C(20)
N(2)-C(24)
N(1)-C(6)
N(3)-C(37)
O(1)-S(1)
1.502(9)
1.495(9)
2.502(5)
2.664(5)
2.679(5)
2.462(6)
1.272(8)
1.272(8)
1.337(8)
1.350(8)
1.437(8)
1.446(8)
1.462(6)
1.421(7)
1.435(7)
C(7)-C(8)
C(12)-C(13)
In(1)-N(2)
In(1)-N(1)
In(1)-N(3)
In(1)-O(1)
N(1)-C(7)
N(3)-C(13)
N(2)-C(8)
N(2)-C(12)
N(1)-C(20)
N(3)-C(30)
O(1)-S(1)
O(2)-S(1)
O(3)-S(1)
1.496(12)
1.497(12)
2.501(8)
2.731(8)
2.637(6)
2.532(7)
1.268(11)
1.280(12)
1.368(11)
1.357(10)
1.433(11)
1.461(12)
1.460(7)
1.475(7)
1.418(8)
O(2)-S(1)
O(3)-S(1)
Table 2. Selected Bond Angles for Compounds 1-3 (deg)
1
2
3
O(2)-In(1)-N(2)
In(1)-O(2)-S(1)
O(1)-In(1)-N(2)
In(1)-O(1)-S(1)
In(1)-N(2)-C(20)
In(1)-N(2)-C(24)
In(1)-N(1)-C(13)
In(1)-N(3)-C(25)
N(1)-In(1)-N(2)
N(3)-In(1)-N(2)
N(1)-C(13)-C(20)
N(3)-C(25)-C(24)
N(2)-C(20)-C(13)
N(2)-C(24)-C(25)
N(2)-C(20)-C(21)
N(2)-C(24)-C(23)
C(13)-N(1)-C(6)
C(25)-N(3)-C(37)
C(20)-N(2)-C(24)
75.8(2)
138.0(4)
120.7(4)
120.6(4)
116.46
115.89
63.62
63.15
118.5(6)
118.6(6)
117.4(5)
116.9(5)
122.7(6)
122.4(6)
120.7(5)
120.2(5)
117.7(5)
118.5(6)
119.7(6)
118.9(6)
O(1)-In(1)-N(2)
In(1)-O(1)-S(1)
In(1)-N(2)-C(8)
In(1)-N(2)-C(12)
In(1)-N(1)-C(7)
In(1)-N(3)-C(13)
N(1)-In(1)-N(2)
N(3)-In(1)-N(2)
N(1)-C(7)-C(8)
N(3)-C(13)-C(12)
N(2)-C(8)-C(7)
N(2)-C(12)-C(13)
N(2)-C(8)-C(9)
N(2)-C(12)-C(11)
C(7)-N(1)-C(20)
C(13)-N(3)-C(30)
C(8)-N(2)-C(12)
C(8)-C(9)-C(10)
76.6(2)
178.3(5)
122.0(6)
119.3(6)
116.65
118.4(6)
63.56
64.7(2)
119.6(9)
117.6(8)
117.6(8)
118.7(8)
121.5(9)
121.8(8)
118.7(8)
119.0(8)
117.7(7)
120.3(9)
118.0(9)
120.6(9)
In(1)-N(2)-C(8)
In(1)-N(2)-C(12)
In(1)-N(1)-C(1)
In(1)-N(3)-C(13)
N(1)-In(1)-N(2)
N(3)-In(1)-N(2)
N(1)-C(1)-C(8)
N(3)-C(13)-C(12)
N(2)-C(8)-C(1)
N(2)-C(12)-C(13)
N(2)-C(8)-C(9)
N(2)-C(12)-C(11)
C(1)-N(1)-C(20)
C(13)-N(3)-C(34)
C(8)-N(2)-C(12)
C(8)-C(9)-C(10)
C(12)-C(11)-C(10)
C(9)-C(10)-C(11)
121.0(4)
118.8(4)
109.71
108.68
61.85
63.09
116.1(6)
116.9(6)
116.7(6)
116.5(6)
122.0(6)
121.8(7)
121.3(6)
122.0(6)
119.2(6)
119.0(7)
118.6(7)
119.2(6)
C(20)-C(21)-C(22)
C(24)-C(23)-C(22)
C(21)-C(22)-C(23)
C(12)-C(11)-C(10)
C(9)-C(10)-C(11)
yields of 63% and 39%, respectively (Scheme 2). A clear
indication of the steric demands of these two ligand systems
was reflected in the NMR spectra of 2 and 3. Specifically, the
appearance of diastereotopic Me groups for 2 and diaster-
eotopic CH2 groups for 3 was consistent with hindered rota-
tion of the Nimine-Ar0 bonds in these compounds. Complex 2
theoretical computational optimization albeit the metric data
must always be interpreted mindful of the larger statistically
estimated experimental errors.
The general structural features for the cations in compounds
2 and 3 are evocative of those observed for 1. In particular the
In-Npy distances (2In1-N2 2.502(5) A, 3In1-N2 2.501(8) A)
are the same as was observed in 1. Furthermore, the In-Nimine
distances for 2 (In1-N1 2.664(5) A, In1-N3 2.679(5) A) and 3
(In1-N1 2.731(8) A, In1-N3 2.637(6) A) are greater than the
sum of the covalent radii for these two elements. Again, like
compound 1, the imine CdN distances within 2 and 3 are
comparable to those for the free ligand.
The reduction of the size of the aryl substituents in 2 and 3,
compared to 1, allowed for a closer approach of the OTf-
anions to the In centers. For compound 2 the shortest In-O
(In1-O1) distance was 2.462(6) A while for compound 3 a
slightly longer distance for In1-O1 of 2.532(7) A was obtained.
For comparison, these distances are shorter than those reported
for In(OTf) (In-O 2.579(6) A and 2.589(6) A)7 but consider-
ably longer than in InOTF[18]crown-6 where the In-O dis-
tance was 2.370(2) A.6 The orientation of the triflate anions in
compounds 2 and 3 suggests an alignment of the negative
charge density of the O and F centers with δþ charges on the
imino-carbon ligand sites (i.e., 2 C13, C25; 3 C7, C13). The
specific differences in triflate anion orientation for 2 and 3, as
shown in Figures 2B and 3B, is likely an indication for a low
energy barrier to reorganization of the anion.
i
displayed two broad, equal intensity doublets for the Pr
methyl groups, and the uniqueness of these groups was
confirmed by the appearance of two singlets for these methyl
goups in the 13C NMR spectrum of 2. Attempts to carry out
NMR experiments for the ethyl compound, 3, in CDCl3 were
unsuccessful. These observations can be attributed to traces
of HCl in the CDCl3 or to oxidative addition reactions of the
alkyl halide to the In (I) complex, which resulted in conver-
sion of 3 to an In3þ species.17
The detailed structural features of 2 and 3 were established
by single crystal X-ray analysis studies, which confirmed these
species as the bis(imino)pyridine complexes [{2,6-R2C6H3Nd
CPh}2(NC5H3)]Inþ(OTf)- (R = iPr, 2; Et, 3). Pictorial repre-
sentations for the structures of 2and 3are presented in Figures 2
and 3 with a summary of bonding parameters presented in
Tables 1 and 2. Notwithstanding the low data resolution for
compound 3, the structure obtained is consistent with non-
crystallographic spectroscopic results and with independent
(17) Hill, M. S.; Hitchcock, P. B.; Pongtavornpinyo, R. Inorg. Chem.
2007, 46, 3783.

10638 Inorganic Chemistry, Vol. 49, No. 22, 2010
Jurca et al.
Figure 2. X-ray structure (A: top view, B: side view) of compound 2, with hydrogen atoms and iPr groups omitted for clarity. Selected bond lengths and
angles are given in Tables 1 and 2, respectively.
Figure 3. X-ray structure (A: top view, B: side view) of compound 3, with hydrogen atoms and ethyl groups omitted for clarity. Selected bond lengths and
angles are given in Tables 1 and 2, respectively.
Scheme 3
In striking contrast to the syntheses of the In (I) compounds
1-3, ligand D reacted with In(O3SCF3), to generate a red
reaction mixture (Scheme 3). The only pure material isolated
from this mixture was an In(III) bis(imino)pyridine complex,
[{2,6-Me2C6H3NdCPh}2(NC5H3)]In(OTf)2Cl 4, possessing
two -O3SCF3 anions and one chloride ion. Large yellow
block-like crystals were isolated and subjected to elemental
analysis confirming homogeneity of the crystallized product.
Figure 4. X-ray structure of compound 4, with hydrogen atoms and
methyl groups omitted for clarity. Selected bond lengths and angles are
given in Table 3.
When these reactions were repeated using different precursor
batches, including samples with satisfactory carbon microanal-
ysis, only compound 4 could be isolated from the reaction
mixture. Variable quantities of indium metal powder were
observed in these reactions, indicative of some disproportiona-
tion reaction. We attribute the appearance of the chloride
ligand in compound 4 to a slight impurity arising from the
synthesis of InOTf.
As shown in Figure 4, compound 4 displayed a distorted
octahedral coordination geometry for In (III) with the coordi-
nation sphere consisting of a planar, tridentate bis(imino)-
pyridine ligand {2,6-Me2C6H3NdCPh}2(NC5H3) with the
remaining three coordination sites occupied by two, trans-
oriented triflate anions and a chloride anion. The distances
for In-N (In1-N1 2.289(3) A, In1-N2 2.171(4) A) and
In-Cl (In1-Cl1 2.312(2) A) in 4, as well as the ligand features,
are comparable with the metrical parameters for the recently
reported cationic bis(imino)pyridine indium(III) complex
[{2,5-tBu2C6H3NdCPh}2(NC5H3)]InCl2þ where average
In-N and In-Cl distances of 2.27 A and 2.33 A, respec-
tively, were observed.2 The In1-O1 distance of 2.269(3) A
and the observed S-O bond distances of the triflate anion
(S1-O1 1.476 A, S1-O2 1.424 A, S1-O3 1.431 A) are
consistent with coordination of O1.7,18
(18) Andrews, C. G.; Macdonald, C. L. B. J. Organomet. Chem. 2005, 690,
5090.

Article
Inorganic Chemistry, Vol. 49, No. 22, 2010 10639
Table 3. Selected Bond Distances and Bond Angles for Compound and 4
cation and, of these, only 0.10 electrons are transferred to a
cation fragment orbital that has a significantly In-based com-
ponent. This lowest unoccupied fragment orbital (LUFOþ2)
is 85% In 5p in nature. This analysis provided further valida-
tion that the indium center in the cation receives only minor
electron donation from the OTf- anion with the bulk of
electron donation actually going to the ligand component
of the In[bis(imino)pyridine]þ cation. The donation from the
triflate to indium corresponds to an In-OTf bond order of
only 0.17.
selected bond lengths (A)
selected bond angles (deg)
In(1)-N(1)
2.289(3)
2.171(4)
2.269(3)
2.312(2)
1.285(4)
1.340(4)
1.463(4)
1.476(3)
1.425(3)
1.431(3)
In(1)-N(2)-C(2)
In(1)-N(1)-C(1)
N(1)-C(1)-C(2)
N(2)-C(2)-C(1)
N(2)-In(1)-N(1)
N(2)-In(1)-Cl(1)
N(2)-In(1)-O(1)
118.7(2)
In(1)-N(2)
In(1)-O(1)
In(1)-Cl(1)
N(1)-C(1)
N(2)-C(2)
N(1)-C(11)
S(1)-O(1)
S(1)-O(2)
S(1)-O(3)
115.2(2)
117.4(3)
114.8(3)
73.53(7)
180.00(1)
83.46(6)
An optimization for the In[bis(imino)pyridine]þ cation in 2
was also carried out in the absence of the OTf- anion. The
computed structure for this cation yielded In-N distances that
are in agreement with the experimental In-N distances in 2
(In-Npy: 2.57 A vs 2.502(5) A exp and In-Nimine: 2.63 A vs
2.66 A exp). These results also point to electrostatic rather than
orbital interaction for the association of the anion with the
cation in 2.
Computational methods provide a valuable approach to
explore the role of substituents of the aryl group on Nimine-Ar0
sites of the In[bis(imino)pyridine]þ cations. Consequently, a
DFT optimization of the model cation [{C6H5NdCPh}2(NC5-
H3)]Inþ, with unsubstituted phenyl groups, was performed.
This was an attempt to evaluate the role of both ligand
sterics and of the counterion in our computational analysis of
the cation. Interestingly, under these model conditions, the
computed values for the In-Npy (2.52 A) and the In-Nimine
(2.63A) distancesare essentially identicaltothecrystallographic
values obtained for 1-3. Significantly, this model system anal-
ysis again provided only nominal bonding between the bis-
(imino)pyridine ligand and the Inþ center with bond orders for
the In-Npy of 0.27 and for In-Nimine of 0.24.
Computational Evaluation
A DFT computational study with the B3LYP functional
and the mixed DZVP/TZVP basis set was undertaken to obtain
a thorough understanding of the electronic nature of the In[bis-
(imino)pyridine]þ cations and the interactions of the cationwith
the OTf- anion in compounds 1-3. Optimizations for com-
pounds 1-3 gave structures that were in agreement with the
experimental structures. In the case of compound 1, the analysis
of the electronic structure of the cation pointed toward only
nominal covalent interactions of the In cation with the ligand.2
Specifically, the natural population analysis (NPA) for the
valence configuration of the In(I) center was 5s1.935p0.20, and
the NPA-derived charge of the In atom was þ0.86 au. These
results indicated that much of the cationic charge remains on the
In, and only 0.14 electrons are transferred from the ligand to Inþ
inthe complex. Analysisfor the cationcomponentof2gave very
similar results with a charge on the indium center of þ0.83 au
and only 0.17 electrons transferred from the ligand in this case.
For comparison, the NPA-derived charge of the Ge(II) center in
the complex (Ge cryptand[2.2.2])2þ(O3SCF3)-2, which dis-
played minimal coordination between the cryptand ligand and
Ge, was þ1.38 au and corresponded to 0.62 electrons being
donated to the Ge(II) ion.1 This is significantly stronger charge
donation than obtained for 1 or 2.
These results were convincingly supported by the bond
order analysis. In the case of the cation of 1, Mayer bond
order values for In-N1/In-N3 and In-N2 interactions are
only 0.23 and 0.28, respectively. Similarly, for the cation in 2,
the In-Nimine (In-N1/In-N3) interactions corresponded to
a bond order of 0.19 while an In-N2 bond order of 0.22 was
obtained. These values are all subtantially lower than the
expected indices of 1 for single bonds. The Wiberg bond
indices calculated in the natural atomic orbital basis are only
0.07-0.10, also consistent with very weak In-ligand interac-
tions. These observations are also in harmony with the
extremely long In-N distances in these species.
While the triflate anion of 1 was well beyond bonding
distance, the shortest OTf- contact in 2was at an In-Odistance
of 2.462(6) A. A careful examination of the cation/anion pair
alignment observed in both the crystallographic (Figure 3B)
and the computational structures of 2 suggests a significant
electrostatic component to the interaction energy for the ion
pair. Specifically, the negatively charged oxygen atoms of
triflate (the average NPA charge of O is -0.94 au) are posi-
tioned over the sites of positive charge localization of the cation:
In, Cimine (C13, C25=þ0.30au) and Cpy (C20, C24=þ0.17au).
The results of energy decomposition analysis support that the
major component of the interaction energy between the two
ions (Eint=-82.9 kcal mol-1) is, in fact, electrostatic with only
approximately 40% of the cation/anion energy because of
orbital interactions (Eorb=-32.7 kcal mol-1).
For the more simplified In[bis(imino)pþyridine]þ cation
with Ar0=H, the bonding between the In cation and the
ligand was analyzed more intimately using a molecular orbital
interaction diagram (Figure 5). Three occupied fragment orbi-
tals (HOFO, HOFO-1 and HOFO-3) of the ligand participate
in the charge donation to the two σ-type 5p orbitals (px and pz)
of the Inþ cation. The In py orbital is non-bonding with the
ligand π-framework. There is no back-donation from the metal
into the extended π-system since In 5s f L π*, is symmetry
forbidden (zero overlap, see Figure 5). Only the ligand HOFO-3
mixes with the occupied In 5s orbital. For the In[bis(imino)-
pyridine]þ cations, the lone pair of electrons on Inþ is localized
in the 5s orbital, and the raising of the energy of these electrons is
a dominant feature in a destabilizing effect on the metal-ligand
bonding (Figure 5). For the In[bis(imino)pyridine]þ cation with
Ar0 =H, the highest occupied molecular orbital (HOMO) of
the cation has 74% contribution from In-localized orbitals
(Figure 5, Table 4). Elaboration of the Nimine-Ar0 groups of
the In[bis(imino)pyridine]þ cations (Ar0 =Ph, 2,6-iPr2C6H3,
2,5-tBu2C6H3) significantlyreducedthecontributionfromIn-
localized orbitals and, most importantly, of the lone electron
pair in In 5s to the HOMO of the cation as summarized in
Table 4. As a result the reactivity of the In(I) lone pair of
electrons is diminished, and we propose that this effectplays a
critical role in leading to the stable complexes 1-3.
The decrease in In 5s contribution to the HOMO arises
primarily from two effects. The first is because changing Ar0
from H to substituted phenyl groups increases the number of
occupiedelectroniclevelsof theligandwhich canmix with the
5s orbital, and this distributes the In orbital over a larger
number of In[bis(imino)pyridine]þ orbitals. The second con-
tribution arises from the fact that as the ligand changes from
Ar0 = Htoligands A-C, theenergyoftheoccupiedorbitalof
the bis(imino)pyridine fragment that overlaps with the In 5s
(i.e., the analogue of the HOFO-3 in Figure 5) rises in energy.
As a result, the mixing between In 5s and this ligand orbital
The precise orbital contributions to the interaction energy
were explored in more detail by examining the specifics of the
electron transfer from the triflate anion to the In[bis(imino)-
pyridine]þ cation through a fragment orbital population anal-
ysis. Only 0.21 electrons are transferred from the anion to the

10640 Inorganic Chemistry, Vol. 49, No. 22, 2010
Jurca et al.
Figure 5. Molecular orbital interaction diagram for the di(imino)pyridine ligand (Ar0 = H) and the Inþ cation. The four highest occupied fragment
orbitals (HOFOs) of the ligand and their contribution to the ligand-to-metal charge transfer interactions in the complex (via the % of electron population
transferred to the Inþ cation) are shown on the right.
Table 4. Energy and % In Character in the HOMO of In[bis(imino)pyridine]þ
Cations
one ligand donor orbital mixes with the In 5s orbital and that
this four electron interaction decreases as the steric demands of
the imino-aryl groups increases, leading to the HOMO that is
dominated by ligand contributions over the In lone pair.
In spite of having three N donor centers, the bis(imino)-
pyridine ligand, while stabilizing a low valent metal center,
provides little charge donation to the metal cation because of
the “mismatch” of donor orbital of the ligand and the sp metal
acceptor orbitals (Figure 5). Using this strategy, we are in the
process of targeting other low-valent compounds.
Ar0
ε
HOMO (eV)
% In (% In s)a character in HOMO
H
Ph
-9.84
-8.82
-8.52
-8.56
74% (66%)
57% (47%)
37% (28%)
19% (14%)
2,6-iPr2C6H3
2,5-tBu2C6H3
a Indium s orbital contribution (%).
leads to the HOMO that has a lower In character and higher
contribution of ligand occupied orbitals.19
Experimental Section
Conclusion
General Methods. Reactions were performed in a glovebox
with a nitrogen atmosphere, with the exception of ligand synthesis,
which was performed using standard Schlenk techniques under a
flow of N2. All solvents were sparged with nitrogen and then dried
by passage through a column of activated alumina using an
apparatus purchased from Anhydrous Engineering. Deuterated
chloroform was dried using activated molecular sieves. Metal
halides were purchased from Strem Chemicals and used as received.
All other chemicals were purchased from Aldrich and used without
further purification. Compounds A, B, and 1 were synthesized
according to literature procedures.2,20 1H and 13C NMR spectra
were run on a Bruker Avance 300 MHz spectrometer with CDCl3,
CD2Cl2, and DMSO-d6 as solvents and internal standards. Ele-
mental analyses for A-D, and 1-4 were performed by Midwest
Microlab LLC, Indianapolis, IN.
The sterically demanding bis(imino)pyridine scaffold readily
reacted with the In(I) synthon, In(O3SCF3), to allow the
preparation, isolation, and analysis of a set of exceptional
compounds in which the Inþ cation is involved in only mini-
mal covalent interactions with the ligand and counterion. The
Inþ centers in these compounds are electron deficient possessing
a 5s lone pair of electrons, and three unoccupied p orbitals.
Structural features of the compounds point to weak coordina-
tion, and computational analysis of the two σ-type 5p orbitals
revealed that they only receive limited electron donation from
the ligands. While the reduction in the steric demands of the
ligand did allow for closer approach of the triflate counterion to
the In centers, the major component of the interaction energy
between the two ions was determined to be electrostatic in
nature with substantial component of the interaction to be
between the ligand and the OTf- anion. The indium center
receives only minor electron donation from the anion.
2 6-Bis{1-[(2,6-diethylphenyl)imino]-benzyl}pyridine (C). Under
a nitrogen atmosphere, a mixture of 2,6-dibenzoylpyridine (1.0 g,
3.48 mmol), a slight excess of 2,6-diethylaniline (1.143 g, 7.66 mmol),
and p-toluenesulfonic acid (0.1 g) in toluene (150 mL) were placed in
a round-bottom flask that was equipped with a Dean-Stark trap.
The reaction mixture was heated to 160 °C, using a sand bath, for
48 h. The workup was performed under lab atmosphere conditions.
The reaction mixture was allowed to cool to room temperature, and
the solvent was removed under vacuum to give a dark yellow oil.
Cold methanol was added to the oil, and the solution was held at
In line with the observations that ligand sterics play an impor-
tant role in stabilizing these species, by reducing the ligand steric
demands to Ar0 =2,6-Me2C6H3 the In(III) bis(imino)pyridine
complex, [{2,6-Me2C6H3NdCPh}2(NC5H3)]In(OTf)2Cl 4, was
produced. Electronic structure analysis demonstrated that only
(19) Composition of the HOMO in terms of fragment orbitals is provided
in the Supporting Information, Table S2.
(20) Jurca, T.; Dawson, K.; Mallov, I.; Burchell, T.; Yap, G. P. A.;
Richeson, D. S. Dalton Trans. 2010, 39, 1266.

Article
Inorganic Chemistry, Vol. 49, No. 22, 2010 10641
-20 °C for 48 h, over which time a multitude of large yellow rectan-
gular crystals formed. The solid was removed by filtration, and the
crystals were washed with cold methanol and hexanes. The crystals
were then crushed to a fine pale yellow powder, and held under
vacuum for 72 h to remove remaining methanol. A very fine pale
yellow powder was collected in moderate 39% yield (750 mg, 1.36
mmol), but in high purity: Analysis for C39H39N3 Calculated: C,
85.21; H, 7.15; N, 7.64 Found: C, 85.13; H, 7.17; N, 7.63. Room
[In-2,6-Bis{1-[(2,6-diethylphenyl)imino]-benzyl}pyridine][OSO2-
CF3] (3). InOSO2CF3 powder (90 mg, 0.341 mmol) was added
to a clear yellow solution of C (194 mg, 0.353 mmol) in 5 mL of
hexanes. The reaction mixture was sealed and allowed to stir for
18 h. A gradual color change from translucent yellow to opaque
orange was observed, as the InOSO2CF3 slowly went into solution.
The solution was then held at -20 °C for 24 h, and a pale orange
precipitate formed. This solution was filtered, and the precipitate
was washed with 5 ꢀ 2 mL of hexanes, and allowed to dry under
vacuum. The resulting pale orange powder was then dissolved in
toluene, and then the solution was passed through a plug of Celite to
remove impurities. The solution was then dried under vacuum,
resulting in the isolation of a dark orange powder in a modest yield
of 39% (108 mg) because of mechanical loss from purification. Dark
orange rod-like crystals suitable for X-ray analysis were grown by
diffusion of hexanes into a saturated toluene solution of 3 and stor-
ing at -20 °C for several days. 1HNMR(C7D8,23°C): δ7.45(v br s,
4H, Ar-H), 7.26(d, 2H, py, m-CH), 7.09(br s, 2H, Ar-CH), 7.02-
6.82(br m, 11H, Ar-H), 2.79(br m, 4H, CH2), 2.41(br m, 4H, CH2),
1.20(t, 12H, CH3). 13C NMR (C7D8) δ 168.2 (CdN imine),
155.1(py-CH), 144.1(py-C), 140.5(py-CH), 134.9(Ar-CH), 134.1-
(Ar-CH), 130.6(Ar-CH), 129.9(Ar-C), 129.7(Ar-CH), 129.6-
128.7 (Ar-CH overlapped by C7D8 signals), 126.3(Ar-CH), 126.1-
(Ar-C), 123.9(Ar-C), 25.5(CH2), 15.4(CH3). Analysis for C40H39-
F3InN3O3S Calculated: C, 59.05; H, 4.83; N, 5.16 Found: C, 58.87;
H, 4.85; N, 4.98.
1
temperature H NMR (CDCl3, 300 MHz) and 13C NMR NMR
(CDCl3, 75 MHz) showed a number of resonances which could not
be definitively assigned and were attributed to a fluxional species.
1H NMR (T = 115 °C, d6-dimethyl sulfoxide, 300 MHz) δ 7.77(br t,
1 H, py, p-CH), 7.52-7.27 (br m, 12 H, py-CH, Ar-H), 6.93 (br m,
6H, Ar-H), 2.37(br s, 8H, CH2), 1.07(t, 12H, CH3). 13C NMR (T=
115 °C, d6-dimethyl sulfoxide, 75 MHz) δ 164.9 (CdN imine), 155.0-
(py-C), 147.5(Ar-CH), 137.4(Ar-C), 136.8(Ar-CH), 131.0(Ar-
CH), 130.3(Ar-C), 128.7(Ar-CH), 128.2(Ar-CH), 125.5(Ar-
CH), 123.3(Ar-CH), 122.8(Ar-C), 24.4(-CH2), 13.3(-CH3).
2,6-Bis{1-[(2,6-dimethylphenyl)imino]-benzyl}pyridine (D).
Following the procedure described for C, a mixture of 2,6-diben-
zoylpyridine (1.0 g, 3.48 mmol), an excess of 2,6-dimethylaniline
(1.10 g, 9.08 mmol), p-toluenesulfonic acid (0.1 g) in toluene
(150 mL) were placed in a round-bottom flask under nitrogen
atmosphere, and the reaction flask was equipped with a Dean-
Stark trap. The reaction mixture was heated to 160 °C, using a sand
bath, for 48 h. The workup was performed under lab atmosphere
conditions. The reaction mixture was cooled to room temperature,
and the solvent was removed under vacuum to give a dark yellow oil.
Cold methanol was added to this oil, and the solution was held at
-20 °C for 24 h, over which time a bright yellow precipitate formed.
The precipitate was removed by filtration, and washed with cold
methanol and hexanes to afford a fine yellow powder. The powder
was further purified by flashing through a silica plug with 15:1
hexanes/ethyl acetate. The solvent was removed under the vacuum
to afford a very fine yellow powder in moderate 47% yield (809 mg,
1.64 mmol), but in high purity. Analysis for C35H31N3 Calculated:
C, 85.16; H, 6.33; N, 8.51 Found: C, 85.30; H, 6.47; N, 8.41. Room
[In(OSO2CF3)2Cl-2,6-Bis{1-[(2,6-dimethylphenyl)imino]-
benzyl}pyridine] (4). InOSO2CF3 powder (52.5 mg, 0.199 mmol)
was added to a clear yellow solution of D (100 mg, 0.203 mmol) in
5 mL of toluene. The reaction mixture was sealed and allowed to
stir for 4 h. An immediate color change from translucent yellow
to opaque dark red was observed. The solution was then held at
-20 °C for 5 days, and a dark red precipitate formed. This solution
was filtered, and the precipitate was washed with 5 ꢀ 2 mL hexanes,
and allowed to dry under vacuum. A red/bronze powder was isolated
in 103 mg. Attempted crystallization in a range of solvents from three
separate reactions using different purified precursor batches consis-
tently yielded the same decomposition product 4. Large block-like
crystals could be consistently isolated, and were subjected to
elemental analysis, confirming a relative homogeneity of crystallized
product. Analysis for C37H31F6InClN3O6S2 Calculated: C 47.17, H
3.32, N 4.46, Found C 46.15, H 3.60, N 4.22.
1
temperature H NMR (CDCl3, 300 MHz) and 13C NMR NMR
(CDCl3, 75 MHz) showed a number of resonances which could
not be properly assigned and were attributed to a fluxional species.
1H NMR (T=115 °C, d6-dimethyl sulfoxide, 300 MHz) δ 8.20-
7.65(br m, 3 H, py, CH), 7.39 (v br s, 10 H, Ar-H), 6.89 (br m, 6H,
Ar-H), 1.99(m, 12H, CH3).13CNMR(T=115 °C, d6-dimethyl sulf-
oxide, 75 MHz) δ 165.7 (CdN imine), 155.2(py-C), 148.6(Ar-CH),
137.5(Ar-C), 136.9(Ar-CH), 130.3(Ar-C), 128.5(Ar-CH),
128.2(Ar-CH), 127.6(Ar-CH), 125.3(Ar-CH), 122.9(Ar-CH),
122.7(A-C), 18.1(-CH3).
Structural Determination of Compounds 1-4. Single crystals
weremountedonathinglassfiberandheldinplaceusingviscousoil.
They were subsequently cooled to data collection temperature.
Crystal data and details of the measurements are summarized in
the Supporting Information, Table S1. Data were collected on a
Bruker AX SMART 1k CCD diffractometer using 0.3 ω-scans at 0,
90, 180 in Φ. Unit-cell parameters were determined from 60 data
frames collected at different sections of the Ewald sphere. Semiem-
pirical corrections based on equivalent reflections were applied
(Blessing, R., Acta Crystallogr. 1995, A51, 33-38). The structures
were solved and refined using the SHELXTL program suite
(Sheldrick, G. M. AXS, Madison, WI, 1997). Direct methods yielded
all non-hydrogen atoms which were refined with anisotropic thermal
parameters. All hydrogen atom positions were calculated geometri-
cally and were riding on their respective carbon atoms. Despite
repeated attempts, compound 3 consistently yielded highly mosaic
crystals that diffracted weakly. The results presented correspond to
the best of several trials. No diffraction was observed at 2θ greater
than 44.3°, and the data set was truncated accordingly to support a
reasonable signal-to-noise ratio.
[In-2,6-Bis{1-[(2,6-diisopropylphenyl)imino]-benzyl}pyridine]-
[OSO2CF3] (2). InOSO2CF3 powder (87 mg, 0.331 mmol) was
added to a clear yellow solution of B (200 mg, 0.331 mmol) in 5 mL
of toluene. The reaction mixture was sealed and allowed to stir for
6 h. An immediate color change from translucent yellow to translu-
cent red-orange was observed, with a gradual change to dark red.
The solution was then held at -20 °C for 24 h, and a small amount of
orange precipitate formed. This solution was filtered, and the pre-
cipitate was washed with 2 ꢀ 5 mL of hexanes, and allowed to dry
under vacuum. An orange powder of 2 was isolated in 63% yield.
Yellow-orange needle like crystals suitable for X-ray analysis were
grown by diffusion of hexanes into a saturated tetrahydrofuran
(THF) solution of 2and storing at -20 °C for several days. 1HNMR
(CDCl3, 23 °C): δ 7.99(t, 1H, py, p-CH), 7.68(d, 2H, py, m-CH),
7.46-7.28(br m, 10H, Ar-H), 7.16-7.02(br m, 6H, Ar-H) 3.00(br
m, 4H, iPr-CH), 1.20(br d, 12H, CH3), 0.91(br d, 12H, CH3). 13
C
Acknowledgment. We thank NSERC for funding.
NMR (CDCl3). δ 167.4(CdN imine), 154.6(py-C), 141.7(py-CH),
140.3(Ar-C), 138.7(py-CH), 133.7(Ar-CH), 130.6(Ar-C),
130.2(Ar-C), 129.6(Ar-CH), 128.7(Ar-CH), 126.3(Ar-CH),
123.8(Ar-CH), 28.6(CH3), 26.3(CHMe2), 22.7(CH3). Analysis
for C44H47F3InN3O3S Calculated: C, 60.76; H, 5.45; N, 4.83
Found: C, 60.61; H, 5.29; N, 4.82.
Supporting Information Available: Crystallographic data (cif
files) for compounds 2-4. Details of computational methods.
This material is available free of charge via the Internet at http://
pubs.acs.org.