Preparation and reactivity of terminal gold(i) amides and phosphides

Article abstract of DOI:10.1039/c2sc21519e

Nucleophilic terminal gold(i) amides have been prepared and their reactivity toward a variety of electrophiles has been explored. For the first time these frequently proposed intermediates were isolated and shown to be unreactive in the amination of π-bonds. The first crystallographically determined terminal group 11 metal phosphide was also synthesized. Preliminary DFT studies have been conducted to understand the structure and reactivity of these complexes.

Full text of DOI:10.1039/c2sc21519e

Chemical Science
View Article Online
View Journal
EDGE ARTICLE
Preparation and reactivity of terminal gold(I) amides
and phosphides†
Cite this: DOI: 10.1039/c2sc21519e
Miles W. Johnson, Sophia L. Shevick, F. Dean Toste* and Robert G. Bergman*
Nucleophilic terminal gold(I) amides have been prepared and their reactivity toward a variety of
electrophiles has been explored. For the first time these frequently proposed intermediates were
isolated and shown to be unreactive in the amination of p-bonds. The first crystallographically
determined terminal group 11 metal phosphide was also synthesized. Preliminary DFT studies have been
conducted to understand the structure and reactivity of these complexes.
Received 13th September 2012
Accepted 28th November 2012
DOI: 10.1039/c2sc21519e
www.rsc.org/chemicalscience
Results and discussion
Introduction
Identifying a suitable neutral ancillary ligand and means of
installing the amido functionality were the rst challenges in
this investigation. The N-heterocyclic carbene IPr (1,3-bis(2,
Terminal transition metal amides represent an important class
of compounds that has been explored throughout the d-block
elements. Late-metal amido complexes are of particular
1
6-diisopropylphenyl)imidazol-2-ylidene) was chosen as a sup-
interest because they are proposed as key intermediates in
porting ligand due to its strong binding and common use in
important transformations such as Buchwald–Hartwig
15,16
2
3
gold catalysis.
A metathetical approach was needed to form
coupling, hydroamination of olens and amination of C–H
4
the Au–N bond as an oxidative route was precluded by the desire
to make a gold(I) complex. Three gold amides were thus
synthesized from IPrAuCl and the corresponding lithium amide
(Scheme 1). The diisopropyl amide 1 (Fig. 1) and anilide 2 were
characterized by X-ray diffraction, revealing that both are
monomeric in the solid state. Compound 3 was only charac-
terized spectroscopically due to decomposition upon purica-
tion. The gold(I) anilide 4 was also accessible by this route using
the sterically less demanding IMes (1,3-bis(2,4,6-trimethyl-
phenyl)imidazol-2-ylidene) ligand.
bonds. Additionally, they are used as model systems for
5
studying biological active sites and are precursors to imido
6
complexes. However, no terminal gold(I) amides have been
7
structurally characterized with the exception of those pos-
8,9
sessing amido ligands that act as loosely coordinating anions.
Nonetheless, these species have been invoked as reactive
1
0,11
intermediates in the addition of amines to p-bonds,
a eld
12
that has undergone tremendous growth in the past decade.
From the perspective of coordination chemistry, gold amides
are peculiar among transition metal amido complexes as they
should possess the most covalent transition metal–nitrogen
The diisopropyl amide 1 was of particular interest. Some
transition metal amides are known to undergo b-hydride
elimination; however, complex 1 undergoes no such decom-
13
bond and yet the greatest hard-so mismatch, presenting a
unique opportunity to understand factors that dictate their
17
14
position pathway even aer prolonged heating. That two
aliphatic groups are present on the amido nitrogen is note-
worthy in that most other group 11 metal amides to our
reactivity. The paucity of data surrounding gold amides
motivated us to pursue their synthesis and chemistry. We now
report examples of terminal gold(I) amides, their reactivity,
and the viability of their participation in catalytic reactions.
Additionally, we were able to synthesize and characterize
crystallographically the rst terminal gold phosphide,
further expanding our understanding of transition metal
pnictogenides.
18
knowledge require aryl substitution for stability.
The
Department of Chemistry, University of California-Berkeley, Berkeley, CA, 94720, USA.
E-mail: fdtoste@berkeley.edu; rbergman@berkeley.edu; Tel: +1 510-642-2850/2156
†
Electronic supplementary information (ESI) available: Synthetic, spectroscopic,
crystallographic, and computational details. CCDC 906061 and 901602–901607.
For ESI and crystallographic data in CIF or other electronic format See DOI:
1
0.1039/c2sc21519e
Scheme 1 Synthesis of gold(I) amides.
This journal is ª The Royal Society of Chemistry 2012
Chem. Sci.

View Article Online
Chemical Science
Edge Article
3,21
hydroamination by transition metals, to form the net Au–N
insertion product 6. The regiochemistry of this reaction was
veried unambiguously through single crystal X-ray diffrac-
tion. The stability of this complex may provide insight as to
why anti-Markovnikov hydroamination of acrylonitrile is less
22
effective using gold catalysts than copper ones. Complex 1
also reacts with the heteroallenes ethyl isocyanate, di(p-tolyl)
carbodiimide, and carbon dioxide to form complexes 7, 8, and
9
, respectively. Product 9 could only be identied spectro-
23
scopically due to decomposition upon concentration.
A
1
3
Fig. 1 Solid state structure of amido complex 1 (50% probability ellipsoids). labeling experiment using CO allowed for identication of
2
˚
13
Hydrogen atoms have been omitted for clarity. Selected bond lengths (A) and
the carbonyl carbon resonance at d 161 ppm in the C NMR
spectrum. The solution IR spectrum exhibited a strong
angles (deg): Au(1)–N(3), 1.967(4); Au(1)–C(1), 1.995(4); Au(1)–N(3)–C(28),
1
22.0(3); Au(1)–N(3)–C(31), 124.8(3); C(28)–N(3)–C(31), 124.8(3).
ꢀ1
absorbance at n ¼ 1578 cm and the assignment of this band
ꢀ
1
13
was conrmed by its shi to 1527 cm when the C-enriched
analogue was analyzed. Both of these results are in agreement
with the spectral properties of other late transition metal
conjugate acid of 1 was also synthesized and characterized
crystallographically to examine the difference in Au–N bond
lengths between the two compounds (see ESI†). Complex 1 was
found to have a shorter Au–N bond than that of the corre-
sponding amine complex, 2.0921(2) A compared to 1.967(4) A,
consistent with observations made on ruthenium amides.
23
carbamates. This result is interesting in that gold hydrocarbyl
complexes have been shown to be unreactive toward CO2,
suggesting that formation of 9 takes place via nucleophilic
˚
˚
24
attack rather than insertion.
19
Aer gaining an understanding of the reactivity of these
amides toward traditional electrophiles, we investigated the
likelihood that these species and similar complexes intervene as
intermediates in gold-catalyzed reactions. The following exper-
iments suggest that gold amides are not viable intermediates in
the gold-catalyzed addition of amines to p-bonds. Published
reports of the intermolecular reaction of unsaturated substrates
with gold amides are not available to determine if such
complexes are competent in amination. Our compounds offer
the opportunity to probe the mechanism of amination without
exogenous amine present. Combining 1 with both alkynes
Additionally, a calculated gas-phase structure of 1 matched
that found in the solid state (Fig. 2), arguing against the
possibility that crystal packing effects affect the complex's
structure (see the ESI†). These data when taken together
suggest that the planarity of late transition metal amides with
aryl substitution may not be attributed to delocalization alone
20
(vide infra).
Reactivity studies began with a survey of the interactions
between various electrophiles and 1 with the intention of
gauging the nucleophilicity and basicity of this complex
(Scheme 2). The reaction of 1 with benzyl bromide led to
(
diphenylacetylene and 3-hexyne) (eqn (1)) and alkenes (cyclo-
immediate alkylation of the amide ligand, resulting in full
conversion to N,N-diisopropylbenzylamine and isolation of
IPrAuBr (5) in 98% yield. Compound 1 was competent in
ꢁ
pentene and styrene) led to no reaction even at 75 C. Addi-
tionally, the kinetic basicity of 1 is sufficiently high to
25
deprotonate the weak acid uorene (pK 23 in THF) forming
a
the amination of acrylonitrile,
a typical substrate for
26
the alkyl gold complex 10. This demonstrates that a strong
base must be present in order to generate a gold amide from the
corresponding amine complex. These experiments suggest that
gold amides are unlikely intermediates in the insertion of
unsaturated C–C bonds into Au–N bonds, and support the
Fig. 2 Solid state structure of phosphido complex 4 (50% probability ellipsoids).
˚
Hydrogen atoms have been omitted for clarity. Selected bond lengths (A) and
angles (deg): Au(1)–P(1), 2.3195(9); Au(1)–C(1), 2.042(3); Au(1)–P(1)–C(28),
101.2(1); Au(1)–P(1)–C(37), 118.0(1); C(28)–P(1)–C(37), 103.2(2).
Scheme 2 Reactivity of complex 1 with various electrophiles.
Chem. Sci.
This journal is ª The Royal Society of Chemistry 2012

View Article Online
Edge Article
Chemical Science
hypothesis that the hydroamination of p-bonds takes place via Table
an outersphere mechanism in gold catalysis.
1
Charge distribution and population analysis of Au–heteroatom
a
11
complexes
(
1)
b
c
c
d
e
Our ndings concerning gold amides prompted us to explore
the possibility of preparing a terminal gold phosphide. There
are numerous examples of bridging gold phosphides in the
literature but no terminal group 11 metal phosphides have been
X
q (Au)
q (X)
Dq (Au–X)
% Contribution
N(i-Pr)
NH(Ph)
P(Mes)
2
1
2
13
14
16
17
0.25
0.34
0.11
0.04
0.35
0.20
ꢀ0.81
ꢀ0.91
ꢀ0.22
ꢀ0.23
ꢀ0.88
ꢀ1.25
1.06
1.25
0.11
0.19
1.23
1.51
63
39
52
69
71
0
2
27
isolated to date. Salt metathesis proved to be insufficient for
P(t-Bu)
2
the synthesis of the desired compounds. Instead, abstraction of Ot-Bu
the chloride from IPrAuCl in the presence of a phosphine NTf
resulted in the formation of corresponding phosphine salts 11
2
a
Natural charge based on NBO analysis and occupancy on population
b
c
and 12 (Scheme 3). The former could be deprotonated to form analysis d BPV86/LANL2DZ/6-311++G(d.p).
X
¼
pnictogen.
q
¼
charge. Dq ¼ difference in charge between Au and heteroatom.
the dimesityl phosphide 13, which was characterized in the
solid state. Complex 14 could only be characterized in solution
e
Percent contribution of X to the HOMO.
28
due to its instability. However, this compound proved to be
catalytically active in the alkylation of tert-butyl phosphine with
benzyl chloride and the intermediate phosphine complex 15,
29
the catalytic resting state, was fully characterized (Scheme 4).
This marks the rst example of C–P bond formation by a
homogeneous gold catalyst.
A density functional theory study was undertaken in order to
gain insight into the bonding of these gold pnictogenides, in
particular, the interactions leading to the gold–heteroatom Fig. 3 Calculated HOMO (BPV86/LANL2DZ/6-311++G(d,p), 0.06 isocontour) of
complexes 1 and 13.
bond and possible causes for the planarity at the amido
nitrogen. Complexes 1, 3, 13, and 14 were analyzed and
30
compared against their well dened and studied t-butoxide (16)
and triimide (17) analogues. Our amides were predicted to nitrogen lone pair, with the respective lone pairs residing in sp
have signicantly polarized Au–N bonds based on natural and p orbitals (Fig. 3). These results contrast with the HOMO of
population analysis, with electrostatic charge differences 17 that consisted of <13% contribution from the amido
approaching that of the gold–triimide nitrogen bond of 17 (see nitrogen. Additionally, natural localized molecular orbital
31
Table 1). The phosphides, however, were found to have little (NLMO) calculations for 1 and 2 indicate that in both cases the
2
difference in charge between the phosphorus and gold centers, Au–N bond consists of 82% contribution from an sp -hybrid-
indicating a highly covalent bond. In all four cases the HOMO ized nitrogen. All second order interactions between gold and
was identied as being predominantly the phosphorus or nitrogen were less than 3% of the primary interactions, signi-
fying that multiple bond character is not present. These data
suggest that the planarity at nitrogen can be attributed to the
electrostatic benet of concentrating the amide nitrogen's
20
electron density at an orbital of high s-character. Phosphides
3 and 14 exhibit much more contribution from gold in the Au–
1
P bond, 30% and 33%, respectively, corroborating the small
difference in electronegativity between the two centers.
Conclusions
Scheme 3 Synthesis of gold(I) phosphides.
In summary, a number of terminal gold amides have been
synthesized and their reactivity explored. High nucleophilicity
was observed with traditional electrophiles, including carbon
dioxide. Despite this nucleophilicity, these amides were
unreactive toward unactivated p-bonds, lending further support
for the outer sphere mechanism for addition of amines to
unsaturated bonds. Additionally, we isolated and identied
crystallographically the rst terminal group 11 metal phosphide
Scheme 4 Catalytic benzylation of di-tert-butyl phosphine.
This journal is ª The Royal Society of Chemistry 2012
Chem. Sci.

View Article Online
Chemical Science
Edge Article
and demonstrated catalytic reactivity with another gold phos-
phide. Computational work was conducted as well to gain an
E. Tkatchouk, K. Kelly, W. A. Goddard, III and F. D. Toste,
Chem. Sci., 2010, 1, 226.
understanding of the bonding and geometries of these unique 12 For recent reviews on gold catalysis, see: (a) A. Corma,
complexes.
A. Leyva-P ´e rez and M. J. Sabater, Chem. Rev., 2011, 111,
657; (b) N. D. Shapiro and F. D. Toste, Synlett, 2010, 5, 675.
1
1
3 Hard-so mismatches for gold(I) complexes are
documented. For examples of a gold uoride and alkoxide,
see: (a) D. S. Laitar, P. M u¨ ller, T. G. Gray and J. P. Sadighi,
Organometallics, 2005, 24, 4503; (b) N. P. Mankad and
F. D. Toste, J. Am. Chem. Soc., 2010, 132, 12959; (c)
E. Tkattchouk, N. P. Mankad, D. Benitez, W. A. Goddard
and F. D. Toste, J. Am. Chem. Soc., 2011, 133, 14293; (d)
N. P. Mankad and F. D. Toste, Chem. Sci., 2012, 3, 72; (e)
For an example of a gold hydride, see: E. Y. Tsui, P. M u¨ ller
and J. P. Sadighi, Angew. Chem., Int. Ed., 2008, 47, 8937.
4 (a) P. L. Holland, R. A. Andersen and R. G. Bergman,
Comments Inorg. Chem., 1999, 21, 115; (b) J. R. Fulton,
A. W. Holland, D. J. Fox and R. G. Bergman, Acc. Chem.
Res., 2002, 35, 44.
Acknowledgements
R. G. B. and F. D. T. thank the U.S. Air Force (ID # 11RSA093) for
a
research grant. M.W.J acknowledges the NSF (no.
DGE1106400) and UC Berkeley for pre-doctoral fellowships. Drs
Neal Mankad, J u¨ rgen Koller, and Antonio DiPasquale are
thanked for helpful discussion and assistance with X-ray crys-
tallographic studies. Dr Erika Cobar, Mark Lipke, Dr Xinzheng
Yang, and Dr Kathleen Durkin provided helpful discussion
concerning the computational portion of this manuscript.
1
Notes and references
1
M. Lappert, P. Power, A. Protchenko and A. Seeber, Metal
Amide Chemistry, John Wiley and Sons, 2009.
15 N. Marion and S. Nolan, Chem. Soc. Rev., 2008, 37, 1776.
2
S. Shekhar, P. Ryberg, J. F. Hartwig, J. S. Mathew, 16 Phosphine ligands proved to be inadequate in the
D. G. Blackmond, E. I. Strieter and S. L. Buchwald, J. Am.
Chem. Soc., 2006, 128, 3584.
preparation of gold pnictogenides, presumably due to their
higher lability in comparison to N-heterocyclic carbenes.
3
4
C. Munro-Leighton, E. D. Blue and T. B. Gunnoe, J. Am. 17 For examples of b-hydride elimination from transition metal
Chem. Soc., 2006, 128, 1446.
R. T. Gephart, III, D. L. Huang, M. J. B. Aguila, G. Schmidt,
A. Shahu and T. H. Warren, Angew. Chem., Int. Ed., 2012,
amides, see: (a) J. F. Hartwig, S. Richards, D. Baranano and
F. Paul, J. Am. Chem. Soc., 1996, 118, 3626; (b) S. E. Diamond
and F. Mares, J. Organomet. Chem., 1977, 142, C55. It should
be noted b-hydride elimination from a heteroatom is less
facile than from carbon: P. L. Theofannis and
W. A. Goddard, III, Organometallics, 2011, 30, 4941.
51, 6488.
5
6
7
N. P. Mankad, W. E. Antholine, R. K. Szilagyi and J. C. Peters,
J. Am. Chem. Soc., 2009, 131, 3878.
V. M. Iluc and G. L. Hillhouse, J. Am. Chem. Soc., 2010, 132, 18 For key examples, see: (a) E. D. Blue, A. Davis, D. Conner,
1
5148.
T. B. Gunnoe, P. D. Boyle and P. S. White, J. Am. Chem.
Soc., 2003, 125, 9435; (b) Ref. 4; (c) Ref. 6. For a noteworthy
alkyl substituted amide, see: S. Wiese, Y. M. Badiei,
R. T. Gephart, S. Mossin, M. S. Varonka, M. M. Melzer,
K. Meyer, T. R. Cundari and T. H. Warren, Angew. Chem.,
Int. Ed., 2010, 49, 8850.
For an isolated gold(III) anilide, see: M. A. Cinellu,
G. Minghetti, M. V. Pinna, S. Stoccoro, A. Zucca and
M. Manaserro, Eur. J. Inorg. Chem., 2003, 2304.
(a) E. Brul ´e , S. Gaillard, M.-N. Rager, T. Roisnel,
V. Gu ´e rineau, S. P. Nolan and C. M. Thomas,
8
Organometallics, 2011, 30, 2650; (b) N. M ´e zailles, L. Ricard 19 (a) A. W. Kaplan, J. C. M. Ritter and R. G. Bergman, J. Am.
and F. Gagosz, Org. Lett., 2005, 7, 4133; (c) L. Ricard and
F. Gagosz, Organometallics, 2007, 26, 4704.
N-bound heterocycles are well documented ligands for gold
Chem. Soc., 1998, 120, 6828; (b) J. R. Fulton,
M. W. Bouwkamp and R. G. Bergman, J. Am. Chem. Soc.,
2000, 122, 8799.
9
though fall outside of the realm of traditional transition 20 For a thorough analysis on this subject see: L. A. Goj,
metal amide chemistry.
0 (a) N. Nishina and Y. Yamamoto, Angew. Chem., Int. Ed.,
006, 45, 3314; (b) V. Lavallo, G. D. Frey, B. Donnadieu,
E. D. Blue, S. A. Delp, T. B. Gunnoe, T. R. Cundari,
A. W. Pierpont, J. L. Petersen and P. D. Boyle, Inorg. Chem.,
2006, 45, 9032.
1
2
M. Soleilhavoup and G. Bertrand, Angew. Chem., Int. Ed., 21 L. Fadini and A. Togni, Tetrahedron: Asymmetry, 2008, 19,
008, 47, 5224; (c) X. Zeng, M. Soleilhavoup and 2555.
G. Bertrand, Org. Lett., 2009, 11, 3166; (d) K. Rei, 22 R. Cober ´a n, S. Marrot, N. Dellus, N. Merceron-Saffon,
2
B. Donnadieu and G. Bertrand, Angew. Chem., Int. Ed.,
011, 50, 5560; (e) K. D. Hesp and M. Stradiotto, J. Am.
T. Kato, E. Peris and A. Baceiredo, Organometallics, 2009,
28, 326.
2
Chem. Soc., 2010, 132, 18026.
23 Similar difficulties were experienced with platinum and
ruthenium carbamate complexes: (a) R. L. Cowan and
W. C. Trogler, J. Am. Chem. Soc., 1989, 111, 4570; (b)
F. L. Joslin, M. P. Johnson, J. T. Mague and
D. M. Roundhill, Organometallics, 1991, 10, 2781.
1
1 For proposed outersphere mechanisms, see: (a) Z. J. Wang,
D. Benitez, E. Tkatchouk, W. A. Goddard, III and
F. D. Toste, J. Am. Chem. Soc., 2010, 132, 13064; (b)
Z. Zhang, C. Liu, R. E. Kinder, X. Han, H. Qian and
R. A. Widenhoefer, J. Am. Chem. Soc., 2006, 128, 9066; (c) 24 M. T. Johnson, v. R. J. M. Janse, M. Axelsson,
R. L. LaLonde, W. E. Brenzovich, Jr, D. Benitez,
M. S. G. Ahlquist and O. F. Wendt, Chem. Sci., 2011, 2, 2373.
Chem. Sci.
This journal is ª The Royal Society of Chemistry 2012

View Article Online
Edge Article
Chemical Science
2
5 A. Streitweiser, D. Z. Wang, M. Stratakis, A. Faccheti,
R. Gareyev, A. Abbotto, J. A. Krom and K. V. Kilway, Can. J.
Chem., 1998, 76, 765.
D. V. Partyka, M. P. Washington, J. B. Updegraff, III,
R. A. Woloszynek and J. D. Protasiewicz, Angew. Chem.,
Int. Ed., 2008, 47, 7489.
2
6 It should be noted that complex 4 undergoes ligand 28 Less sterically demanding groups on phosphorus resulted in
exchange with d -aniline to generate free aniline (see ESI†). the formation of m-phosphides.
Additionally, complex is not sufficiently basic to 29 For enantioselective catalytic
deprotonate triphenylmethane, indicating that its
conjugate acid has a pK of lower than 31.5 in THF.
5
1
examples
of
this
transformation: (a) V. S. Chan, M. Chiu, R. G. Bergman
and F. D. Toste, J. Am. Chem. Soc., 2009, 131, 6021; (b)
M. F. Cain, R. P. Hughes, D. S. Glueck, J. A. Golen,
C. E. Moore and A. L. Rheingold, Inorg. Chem., 2010, 49,
7650; (c) D. S. Glueck, Synlett, 2007, 17, 2627.
a
2
7 For a terminal copper phosphide that was observed
spectroscopically: M. F. Cain, R. P. Hughes, D. S. Glueck,
J. A. Golen, C. E. Moore and A. L. Rheingold, Inorg.
Chem., 2010, 49, 7650, For a review on bridging gold 30 For computational work concerning late transition metal–
phosphides: E. J. Fern ´a ndez, A. Laguna and M. E. Olmos,
heteroatom bonding, see: D. Devarajan, T. B. Gunnoe and
J. Chil. Chem. Soc., 2007, 52, 1200, See also:
D. H. Ess, Inorg. Chem., 2011, 51, 6710.
D. M. Stefanescu, H. F. Yuen, D. S. Glueck, J. A. Golen, 31 Computationally determined charges are highly qualitative
L. N. Zakharov, C. D. Incarvito and A. L. Rheingold, Inorg.
Chem., 2003, 42, 8891, For gold(I) phosphinidines:
and the data produced should only be used to establish
trends in similar compounds.
This journal is ª The Royal Society of Chemistry 2012
Chem. Sci.