Surprising isomer preference on IrIII, favoring facile H-C(sp3) bond cleavage

Article abstract of DOI:10.1039/b908954c

Species (PNP)Ir(X)(Y), where PNP = N(SiMe₂CH₂P ^tBu₂)₂^-1 with X and Y halide, have a ^tBu group C-H bond heterolytically split by addition across the Ir-N bond. The Royal Society o

Full text of DOI:10.1039/b908954c

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Surprising isomer preference on IrIII, favoring facile H–C(sp3) bond
cleavagew
Nikolai P. Tsvetkov, Matthew F. Laird, Hongjun Fan, Maren Pink and Kenneth G. Caulton*
Received (in Berkeley, CA, USA) 5th May 2009, Accepted 11th June 2009
First published as an Advance Article on the web 29th June 2009
DOI: 10.1039/b908954c
Species (PNP)Ir(X)(Y), where PNP = N(SiMe2CH2PtBu2)2
vapor diffusion of pentane into benzene shows two indepen-
dent molecules of (PN(H)P*)IrI2 in the asymmetric unit, but
structural parameters in each are very similar and only one will
be discussed here (Fig. 1). The structure is consistent with the
conclusions from spectroscopic data, in that the species is
ꢀ1
with X and Y halide, have a tBu group C–H bond heterolytically
split by addition across the Ir–N bond.
A variety of molecules of the general type (pincer ligand)
Ir(X)(Y) are known, involving X–Y pairs such as alkyl and
hydride, halide and hydride and two hydrides.1–13 All are
5-coordinate, and thus coordinatively unsaturated. Pincers
include as their central donor sp3 carbons, sp2 carbons or
imine nitrogens (in pyridine).14 These find a variety of
catalytic applications, including the especially demanding
dehydrogenation of alkanes. We report here (Scheme 1) how
different the potential energy landscape is for analogous new
t
molecular, with a metallated Bu group. While the hydrogen
originally on iridium was not identified in X-ray difference
maps, the overall structure is consistent with this being located
on the nitrogen, to give a secondary amine ligand on Ir(III).
The geometry around the metal is distorted octahedral, in spite
of the presence of a strained four membered ring including the
metallated carbon (evident especially in the angle Ir–P–C of
90.6(2)1 in that 4-membered ring). The secondary amine is
nonplanar with regard to the Ir and two Si substituents, as it
must be for the proton being on that nucleophile. The Ir–N
bond is long at 2.213(6) A, consistent with protonation at
nitrogen. The Ir–I distance trans to carbon is significantly
longer (by 0.1596(7) A) than the one trans to N.
compounds where the central linking atom is a bis-silyl amide,
ꢀ1
in the pincer ligand PNP
=
N(SiMe2CH2PtBu2)2
.
Surprising differences observed here center on the fact that
the amide nitrogen is a highly reactive functionality, or at least
that the Ir–NSi2 bond is a functionality in its own right. The
identity of ligands X and Y also plays a role.
What is unusual about this result is that there has been no
oxidatively induced reductive elimination16–18 of H with
carbon in 1, to restore a tBu group and form simply (PNP)IrI2,
an isomer of the observed structure. The observed structure
still contains trivalent Ir, so there has been reductive coupling
of Ir–H with NSi2, thus avoiding IrV from oxidant I2; transient
oxidation may be what reverses the polarity of the ‘‘hydride’’
and makes it possible to serve as an acid, protonating amide
nitrogen. Indeed, this reaction may actually occur by hetero-
lytic splitting of I2 by IrIII, to give transient [(PNP*)IrHI+][Iꢀ]
whose hydride is now sufficiently acidic that it protonates the
amide nitrogen. If this proton is cis to N, then migration of H
to N is facile. If this proton is trans to N, then the intimately
ion paired Iꢀ can shuttle this proton to N before that iodide
The product of reaction of [Ir(cyclooctene)2Cl]2 with the
Li(PNP) or (PNP)ClMg reagents of PNP in THF has structure
1, designated (PNP*)IrH (ESIw). This is also true of the Rh
analog.15 We were interested in whether the reactivity of this
molecule would be exclusively that of a d6, Ir(III) center, or
whether it could be a ‘‘source’’ of its reductive elimination
product, (PNP)Ir. In this regard, a key feature of 1 is its
operational unsaturation.
Reaction (1 : 1) of I2 with (PNP*)IrH in THF gives rapid
and complete conversion to a single diamagnetic product 2I
with an AX 31P{1H} NMR spectrum (large JPP, indicative of
1
trans phosphorus nuclei) and a H NMR spectrum consistent
with no symmetry; altogether these indicate that the Ir–C
bond of 1 remains intact. However, there is no evidence of a
hydride ligand. The compound has low solubility in benzene
and cyclohexane, but is more soluble in CD2Cl2.z
These observations suggest that 2 is a polar compound. This
1
is further supported by the presence of a H NMR resonance
at 4.4 ppm attributed to an N–H functionality; this was
identified as an exchangeable proton by adding a trace of
D2O in CD2Cl2; the resonance at 4.4 ppm disappears within
1 h. The CI mass spectrum of this compound shows the ion of
mass (PNP)IrI+. The X-ray structure of crystals grown by
Department of Chemistry, Indiana University, 800 E. Kirkwood
Avenue, Bloomington, Indiana, USA. E-mail: caulton@indiana.edu;
Fax: +1 812 855 8300; Tel: +1 812 855 4798
w Electronic supplementary information (ESI) available: Full
experimental and computational details are available. CCDC 730042
and 730044. For ESI and crystallographic data in CIF or other
electronic format see DOI: 10.1039/b908954c
Scheme 1
ꢁc
This journal is The Royal Society of Chemistry 2009
4578 | Chem. Commun., 2009, 4578–4580

View Article Online
Fig. 1 ORTEP view (50% probabilities) of the nonhydrogen atoms of
molecule A of [(PtBu2CH2SiMe2)N(H)(SiMe2CH2PtBu{CMe2CH2})]IrI2),
2I, showing selected atom labeling; there is a hydrogen on N1. Unlabeled
atoms are carbons. Selected structural parameters: Ir1–C18, 2.119(6) A;
Ir1–N1, 2.213(5); Ir1–P2, 2.3221(14); Ir1–P1, 2.4337(15); Ir1–I2, 2.6647(4);
Ir1–I1, 2.8243(5); C18–Ir1–N1, 95.3(2)1; C18–Ir1–P2, 67.62(18);
C18–Ir1–I2, 89.72(18); N1–Ir1–I2, 174.94(15); C18–Ir1–I1, 160.92(18);
N1–Ir1–I1, 81.58(16); C15–P2–Ir1, 90.6(2); N1–Ir1–P2, 88.02(13);
C18–Ir1–P1, 102.57(18); N1–Ir1–P1, 87.92(13); P2–Ir1–P1, 168.97(5);
P2–Ir1–I2, 94.63(4); P1–Ir1–I2, 90.23(4); P2–Ir1–I1, 93.41(4); P1–Ir1–I1,
96.13(4); I2–Ir1–I1, 93.945(16).
Fig. 2 ORTEP view (50% probabilities) of the nonhydrogen atoms
of 3, [(PtBu2CH2SiMe2)N(H)(SiMe2CH2PtBuCMe2CH2)IrCl]O3SCF3,
showing selected atom labeling. Unlabeled atoms in the cation are
carbons. Selected structural parameters: Ir1–C11, 2.103(5) A; Ir1–N1,
2.176(4); Ir1–P1, 2.3331(12); Ir1–Cl1, 2.3519(12); Ir1–P2, 2.3682(12);
N1–O1, 3.076(6); C11–Ir1–N1, 91.57(18)1; C11–Ir1–P1, 68.94(14);
N1–Ir1–P1, 87.79(11); C11–Ir1–Cl1, 93.65(14); N1–Ir1–Cl1,
174.73(11);
P1–Ir1–Cl1,
93.42(4);
C11–Ir1–P2,
103.10(14);
then binds to unsaturated Ir. Each mechanism leads to the
observed product, containing IrIII. We have heated the
observed product in THF at 60 1C to test whether it or
(PNP)IrI2 is thermodynamically more stable. We observe no
change after 12 h. Iodine oxidation of (PNP*)IrH is selective
in that it does not halogenate the carbon–Ir bond, to form
[(PtBu2CH2SiMe2)N(SiMe2CH2PtBu{CMe2CH2I})]IrH(I).
Is this preference for the metallated structure specific for
iodide, or is it also true for other halides? In a synthetic
approach to (PNP)IrCl2 starting from intact tBu groups,
(Scheme 1) paramagnetic, C2v (PNP)IrCl reacts with equi-
molar N-chlorosuccinimide (NCS) over 10 min at 25 1C in
benzene to form a diamagnetic product with no symmetry, not
a hydride, and in general wholly inconsistent with the expected
product, (PNP)IrCl2. Three tBu doublets and two single
methyl groups are consistent with (PN(H)P*)IrCl2, structure
2Cl. The N–H proton is detected at 4.47 ppm. What is most
surprising is the absence of (PNP)IrCl2 as the product.
N1–Ir1–P2, 89.01(10); P1–Ir1–P2, 171.33(4); Cl1–Ir1–P2, 90.49(4);
C8–P1–Ir1, 88.05(16); C8–C11–Ir1, 106.8(3).
ligands is not defined by the spectroscopic data, but a single
crystal X-ray structure determination shows (Fig. 2) that this
is a salt, with outer sphere triflate and a square pyramidal
structure around Ir. The empty coordination site in this
unsaturated species is trans to the metallated carbon, and
hydrogen bonding connects the amine proton to triflate
oxygen. It is noteworthy that this salt shows enough solubility
in benzene to yield a useful NMR spectrum.
Is this unexpected tendency of three examples to prefer the
heterolytic splitting of an H–C(sp3) bond due to the amide
nitrogen functionality, or due to our being the first to explore
dihalide versions of (pincer)Ir(X)(Y)? To address this latter
idea, we have synthesized (PNP)IrHCl by dehydrohalogena-
tion of (PN(H)P)IrHCl2 with LiNiPr2 (Scheme 1) and find it to
have Cs symmetry, with one hydride ligand, indicating the
structure to be the previously conventional one, (pincer)IrHCl.
To address kinetic vs. thermodynamic stabilities, we have
carried out DFT(PBE) calculations of both (PNP)IrCl2 and
the observed product 2Cl and find 2Cl to be more stable by
17.8 kcal molꢀ1. For comparison, comparable calculations
show (PNP)IrHCl to be more stable than the isomeric
(PN(H)P*)IrHCl but by only 2.1 kcal molꢀ1; this indicates
that the phosphine R groups even on (pincer)IrH(halide) are
vulnerable (e.g. to H–D exchange). This means that the
observed dihalide products 2 are thermodynamically, not
merely kinetically, preferred. The DFT result indicates
unexpectedly high reactivity (instability) for the (PNP)IrCl2
structure, in particular that the metal complex is sufficiently
reactive to attack an H–C(sp3) bond. Key to this C–H bond
We also investigated outer sphere 1-electron oxidation of
C
2v symmetric (PNP)IrCl using [Cp2Fe]O3SCF3. This reaction
(Scheme 1) proceeds to completion over 12 h at 25 1C even in
nonpolar benzene, to form ferrocene (detected by H NMR)
1
and a single product 3, diamagnetic and with no symmetry.
The phosphorus nuclei couple strongly (362 Hz) indicating
they are mutually trans in the product. The 1H NMR spectrum
t
shows three Bu doublets, four SiMe singlets, and two Me
doublets due to a tBu group which has had one methyl group
metallated. This product is [(PN(H)P*)IrCl](OTf), thus
trivalent iridium. It is interesting and surprising that this does
not have the structure (PNP)IrCl(OTf), but that instead there
has been heterolytic cleavage of an H–C(sp3) bond. The
detailed location of the amine H and the Clꢀ and triflate
ꢁc
This journal is The Royal Society of Chemistry 2009
Chem. Commun., 2009, 4578–4580 | 4579

View Article Online
0.0316, wR2 = 0.0771 (5559 observed reflections with I 4 2s(I) and
scission is that it is heterolytic, and so there is no oxidation of
IrIII. This requires development of nucleophilic character at
the amide nitrogen,19 at least at the transition state, as the
C–H bond interacts with an empty iridium orbital in
(PNP)IrCl2. This will clearly depend on the more numerous
4-electron repulsions20 present in the (PNP)IrCl2 structure
than in (PNP)IrHCl.
338 parameters), GOF = 1.026, largest difference peak = 2.462 eAꢀ3
;
peak and hole are of the same magnitude, located near iridium.
CCDC 730044.
1 H. A. Y. Mohammad, J. C. Grimm, K. Eichele, H.-G. Mack,
B. Speiser, F. Novak, M. G. Quintanilla, W. C. Kaska and
H. A. Mayer, Organometallics, 2002, 21, 5775–5784.
2 M. Kanzelberger, B. Singh, M. Czerw, K. Krogh-Jespersen and
A. S. Goldman, J. Am. Chem. Soc., 2000, 122, 11017–11018.
3 K. Krogh-Jespersen, M. Czerw, K. Zhu, B. Singh,
M. Kanzelberger, N. Darji, P. D. Achord, K. B. Renkema and
A. S. Goldman, J. Am. Chem. Soc., 2002, 124, 10797–10809.
4 K. B. Renkema, Y. V. Kissin and A. S. Goldman, J. Am. Chem.
Soc., 2003, 125, 7770–7771.
5 F. Liu, E. B. Pak, B. Singh, C. M. Jensen and A. S. Goldman,
J. Am. Chem. Soc., 1999, 121, 4086–4087.
6 M. Gupta, C. Hagen, R. J. Flesher, W. C. Kaska and C. M. Jensen,
Chem. Commun., 1996, 2083–2084.
7 Z. E. Clarke, P. T. Maragh, T. P. Dasgupta, D. G. Gusev,
A. J. Lough and K. Abdur-Rashid, Organometallics, 2006, 25,
4113–4117.
8 S. A. Kuklin, A. M. Sheloumov, F. M. Dolgushin,
M. G. Ezernitskaya, A. S. Peregudov, P. V. Petrovskii and
A. A. Koridze, Organometallics, 2006, 25, 5466–5476.
9 I. Goettker-Schnetmann, P. White and M. Brookhart, J. Am.
Chem. Soc., 2004, 126, 1804–1811.
10 L. Fan, S. Parkin and O. V. Ozerov, J. Am. Chem. Soc., 2005, 127,
16772–16773.
11 O. V. Ozerov, C. Guo, V. A. Papkov and B. M. Foxman, J. Am.
Chem. Soc., 2004, 126, 4792–4793.
12 E. Ben-Ari, G. Leitus, L. J. W. Shimon and D. Milstein, J. Am.
Chem. Soc., 2006, 128, 15390–15391.
13 E. Ben-Ari, R. Cohen, M. Gandelman, L. J. W. Shimon, J. M.
L. Martin and D. Milstein, Organometallics, 2006, 25, 3190–3210.
14 V. C. Gibson and S. K. Spitzmesser, Chem. Rev., 2003, 103,
283–315.
The small size of the DG1 favoring the (PNP)IrHCl structure
shows that it is easy to reverse the isomer preference for
(PNP)IrX(halide) species (X = hydride or halide), yet the fact
(Scheme 1) that (PNP)Ir(O2) does not show the C–H metallated
structure (ESIw) is also significant. The competition between
the two IrIII isomers evaluated here for X = hydride vs. halide
cannot be attributed to ligand reducing power, nor to the trans
influence (both structures have one high trans influence ligand)
but may find contributions from the fact that the C–H
addition structure has 18 valence electrons, while (PNP)IrHCl
has 16 + d, d being the pi donation from the amide,21–24 and
that hydride is clearly most compatible with the 16 + d
configuration (no ligand trans to hydride).
The unusually facile H–C(sp3) bond cleavages reported here
are all heterolytic in character (i.e. constant metal oxidation
state) and thus depend on developing Brønsted basicity at
amide nitrogen, and on the electron rich character of the
Ir–amide bond. This is the generality to be derived from our
work, and it is analogous to the heterolytic splitting of H2
by d6 ruthenium amides which has been widely exploited
recently.25,26
This work was supported by the National Science Founda-
tion (NSF CHE-0544829).
15 A. Y. Verat, M. Pink, H. Fan, J. Tomaszewski and K. G. Caulton,
Organometallics, 2008, 27, 166–168.
16 T. T. Tsou and J. K. Kochi, J. Am. Chem. Soc., 1978, 100,
1634–1635.
17 J. K. Kochi, Organometallic Mechanisms and Catalysis,
John Wiley, New York, 1978.
18 W. Lau, J. C. Huffman and J. K. Kochi, Organometallics, 1982, 1,
155–169.
Notes and references
z Crystal data for 2I: orange block, 0.15 ꢂ 0.12 ꢂ 0.12 mm,
C22H52I2IrNP2Si2, M = 894.77, monoclinic space group P21/c,
Z = 8, T = 150 K, a = 16.4974(9) A; b = 12.9874(7) A;
19 D. Conner, K. N. Jayaprakash, T. R. Cundari and T. B. Gunnoe,
Organometallics, 2004, 23, 2724–2733.
20 K. G. Caulton, New J. Chem., 1994, 18, 25–41.
21 C. C. Bickford, T. J. Johnson, E. R. Davidson and K. G. Caulton,
Inorg. Chem., 1994, 33, 1080–1086.
22 T. J. Johnson, K. Folting, W. E. Streib, J. D. Martin,
J. C. Huffman, S.A. Jackson, O. Eisenstein and K. G. Caulton,
Inorg. Chem., 1995, 34, 488–499.
23 D. R. Tyler, Acc. Chem. Res., 1991, 24, 325–331.
24 D. E. Wigley, Prog. Inorg. Chem., 1994, 42, 239–482.
25 S. E. Clapham, A. Hadzovic and R. H. Morris, Coord. Chem. Rev.,
2004, 248, 2201–2237.
c
= 30.8640(18) A; a = 901; b = 105.052(10)1, g = 901,
V = 6386.0(6) A3; rcalc. = 1.861 Mg mꢀ3, m = 6.302 mmꢀ1, MoKa,
2ymax = 551. Total reflections collected = 81 937. 14 579 (Rint = 0.0396)
were unique. Final residues: R1 = 0.0363, wR2 = 0.0736 (11 532 observed
reflections with I 4 2s(I) and 571 parameters), GOF = 1.023, largest
difference peak = 4.621 eAꢀ3; peak and hole are of same magnitude,
located near iodine. CCDC 730042.
Crystal data for 3: red block, 0.15 ꢂ 0.13 ꢂ 0.10 mm, C23H52ClF3-
IrNO3P2SSi2, M = 825.49, monoclinic P21/n, Z = 4, T = 150 K,
a = 12.9232(9) A; b = 20.6668(15) A; c = 13.5066(9) A; a = 901;
b = 105.573(1)1; g = 901, V = 3474.9(4) A3, rcalc. = 1.578 Mg mꢀ3
,
m = 4.181 mmꢀ1, MoKa, 2ymax = 511. Total reflections collected =
28 585. 6581 (Rint = 0.0400) were unique. Final residues were R1
=
26 R. Noyori and S. Hashiguchi, Acc. Chem. Res., 1997, 30, 97–102.
ꢁc
This journal is The Royal Society of Chemistry 2009
4580 | Chem. Commun., 2009, 4578–4580