Mechanisms of cyaphide (C≡P-) formation

Article abstract of DOI:10.1002/anie.200702442

(Chemical Equation Presented) Rock 'n' roll cyaphide: Although the kinetic product B is an observable intermediate of nucleophilic attack on the low-coordinate phosphorus atom in A, it is not directly involved in the mechanism leading to cyaphide complex F. Instead, DFT calculations suggest that nucleophilic attack at silicon (→C) initiates decomposition via D to F. In contrast to cyanide complexes, isocyaphide E is only a transition state for C≡P-ligand rotation D?D′.

Full text of DOI:10.1002/anie.200702442

Communications
DOI: 10.1002/anie.200702442
ꢀ
C P Ligands
ꢁ
ꢀ
Mechanisms of Cyaphide (C P ) Formation**
Andreas Ehlers,* Joseph G. Cordaro, Daniel Stein, and Hansjörg Grützmacher*
Many complexes and coordination polymers with unique
properties have been constructed with the cyanide ion.[1]
However, analogous chemistry with the heavier phosphorus
congener of cyanide has not been developed. We reported the
synthesis and X-ray crystallographic characterization of
a complex with a terminal cyaphide ligand, [RuH(dppe)2-
subsequent reaction was monitored by 31P{1H} NMR spec-
troscopy, a new quintet at d = 309.5 (JPP = 27 Hz) ppm and a
doublet at d = 62.7 (JPP = 27 Hz) ppm were seen for inter-
mediate X. Over time, this intermediate species disappeared
and broader resonances at d = 164.8 and 65.3 ppm appeared
for product 2. When two equivalents of NaOPh (relative to 1)
were used, the rate of reaction to form the cyaphide complex
2 was accelerated. A clue to the nature of X was obtained
serendipitously when the reaction of NaOPh and 1 was
performed in wet acetonitrile. From this reaction a new
product was detected in the 31P NMR spectrum (d = 332.0
(qnt, JPP = 28 Hz), 67.7 (d, JPP = 28 Hz) ppm). Further inves-
tigations into this crystalline material by X-ray diffraction
revealed the unique l5s3-phosphaketenylruthenium complex
3 (Figure 1).[4]
(C P)] (2; dppe = bis-1,2-diphenylphosphinoethane).[2,3]
ꢀ
Herein we present experimental and computational results
which suggest a possible mechanism for the formation of 2.
+
ꢀ
When [RuH(dppe)2(P CSiPh3)] (1) was treated with
1.2 equivalents of NaOPh in dry [D8]THF (Scheme 1) and the
Scheme 1. Synthesis of phosphaoximato complex 3.
[*] Dr. A. Ehlers
Faculty of Exact Sciences
Department of Organic and Inorganic Chemistry
Division of Chemistry, Vrije Universiteit
De Boelelaan 1083, 1081 HV Amsterdam (The Netherlands)
Fax: (+31)20-598-74-88
Figure 1. Molecular structure of 3. One acetonitrile molecule in the
crystal lattice and hydrogen atoms (except that bonded to Ru) are
omitted for clarity. Thermal ellipsoids are set at 30% probability.
Selected bond lengths [Š] and angles [8]: O1-P1 1.509(2), P1-C1
1.663(3), C1-Si1 1.823(3), Ru1-P1 2.358(1), Ru1-P2 2.351(1), Ru1-P3
2.336(1), Ru1-P4 2.340(1), Ru1-P5 2.350(1); O1-P-C1 113.5(1),
Ru1-P1-O1 122.7(1), Ru1-P1-C1 123.8(1), P1-C1-Si1 127.4(2),
P1-Ru1-P2 93.9(1), P1-Ru1-P3 89.0(1), P1-Ru1-P4 97.5(1), P1-Ru1-P5
96.8(1), P2-Ru1-P3 81.3(1), P3-Ru1-P4 97.2(1), P4-Ru1-P5 83.5(1), P5-
Ru1-P2 96.8(1).
E-mail: ehlers@chem.vu.nl
Dr. D. Stein, Prof. Dr. H. Grützmacher
Laboratory of Inorganic Chemistry
Department of Chemistry and Applied Biosciences
ETH Zürich, 8093 Zürich (Switzerland)
Fax: (+41)44-633-1418
E-mail: gruetzmacher@inorg.chem.ethz.ch
Dr. J. G. Cordaro
Materials Research Laboratory
University of California
Santa Barbara, CA 93106 (USA)
The coordination sphere around phosphorus in this l5s3-
phosphaketenyl complex is trigonal-planar (ꢀP = 3608). Both
ꢁ
ꢁ
the P1 O1 (1.509(2) Š) and P1 C1 (1.663(3) Š) bonds are
[**] We thank the ETH Zürich for financial support.
3
ꢁ
=
short and typical for low-coordinate s -type structures, R P(
Supportinginformation for this article is available on the WWW
under http://www.angewandte.org or from the author.
X)2 (X = CR2, NR, O, S).[5] In analogy with complexes
7878
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 7878 –7881

Angewandte
Chemie
ꢁ
= ꢁ
containing oximate, [R2C N O] , as ligand, 3
can also be viewed as a k1-P-phosphaoximato
complex. While oximato complexes are numer-
ous, only one other report of a ruthenium
phosphaoximato complex exists.[6,7] Curiously,
this species, containing the k -P,C-[P( O)CtBu-
2
=
2ꢁ
=
(C O)] ion, has a markedly low-frequency
resonance in the 31P NMR spectrum (d =
47.0 ppm) in comparison to 3.
Hypothesizing that in the presence of residual
water and phenoxide, hydroxide ions are formed
that then attack at phosphorus to irreversibly
generate 3, we treated 1 with excess LiOH in
THF. Indeed, quantitative formation of 3 was
obtained through this more direct pathway. A P-
metalated phosphoniocarbene I, R2P-C-R’,[8] is a
likely intermediate in this reaction which con-
verts by a 1,3-H shift into the final product 3
(Scheme 1). DFT calculations[9] show that ther-
modynamically the rearrangement of I to 3 is
strongly favored (DHr = ꢁ33.9 kcalmolꢁ1).
The results discussed above serve as a starting
point for DFT calculations aimed at understand-
ing the mechanism leading to the formation of
cyaphide complex 2.[9] We used a simplified
model where each 1,2-diphenylphosphinoethane
(dppe) ligand was replaced by two PH3 mole-
cules, SiPh3 by a SiH3 group, and phenoxide by
methoxide. Based on the experimental results
that suggested hydroxide attack at phosphorus,
we hypothesized that the elimination of
Ph3SiOPh from 1 might likewise occur by initial
phenoxide attack at phosphorus. Therefore, our
computations started with A as a model for
intermediate I which is set at 0.0 kcalmolꢁ1 in the
relative energy diagram shown in Figure 2A. This
P-metalated phosphinocarbene A rearranges via
TS1 with a modest barrier (10.2 kcalmolꢁ1) to the
Figure 2. Calculated reaction paths (DFT) for the interaction of the silylphosphaalkyne
complex [RuH(PH3)4(P C-SiH3)]+ with methoxide, MeOꢁ to give E + H3SiOMe
ꢀ
C-metalated phosphaalkene B, which is 4.1 kcal overall. A) nucleophilic attack at phosphorus; B) nucleophilic attack at silicon.
molꢁ1 more stable than A. Calculated 31P-NMR-
spectroscopic shifts of A and B[10] allow us to
decide which of these two isomers corresponds to the
experimentally detected intermediate X (Scheme 1). While
A has a d(31P)calcd of d = 97 ppm versus H3PO4, which is typical
for phosphinocarbenes,[8] the d(31P)calcd of B is d = 334 ppm,
which is in reasonably good agreement with the experimen-
tally detected resonance at d = 309.5 ppm.[11] Consequently
we propose that X has a structure similar to that of isomer B,
but with [Ru] = [Ru(dppe)2] and OPh and SiPh3 instead of
OMe and SiH3, respectively.
We then inspected the intramolecular elimination of the
silylether MeOSiH3 from B. The first intermediate of this
pathway is the h2-bonded phosphaalkyne silicate complex C
which is obtained via transition state TS2. The slightly
exothermic dissociation of MeOSiH3 from C (DHr (C!D) =
ꢁ1.5 kcalmolꢁ1), leads to the h2-bonded cyaphide complex D
which rearranges via a small barrier (TS3) to the end-on
bonded cyaphide complex E in the most exothermic step (DHr
(D!E) = ꢁ19.5 kcalmolꢁ1). Note that the overall reaction
B!E is almost thermoneutral (DHr (B!E) = 1.1 kcalmolꢁ1),
which would explain the rather long reaction time (ca. 14 h)
for complete conversion of 1 into 2. However, the high barrier
of 38.1 kcalmolꢁ1 calculated for the rate-limiting step via
transition state TS2 makes the intramolecular elimination of
MeOSiH3 very unlikely.
A more direct route to cyaphide complex E is simply
nucleophilic attack at silicon by methoxide followed by
MeOSiH3 elimination (Figure 2B). This alternative mecha-
nism requires that the ion pair and intermediates A and B are
in equilibrium with one another and that only the ion pair,
and not intermediates A and B, is on the pathway leading to
E. Calculations put the pentacoordinate silicon intermediate
ꢁ1
ꢁ ꢁ
F, with a linear O Si C arrangement, 26.1 kcalmol higher
in energy than A. Elimination of MeOSiH3 leads to the h2-
bonded complex D via the phosphorus-coordinated cyaphide
complex G. This isocyaphide complex G is actually calculated
to be a transition state only 4.8 kcalmolꢁ1 higher in energy
Angew. Chem. Int. Ed. 2007, 46, 7878 –7881
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 7879

Communications
than F. Note that attempts to find a less symmetric pathway
for the attack of methoxide on the silicon atom in [RuH-
Keywords: coordination modes · cyaphides ·
density functional calculations · multiple bonds ·
reaction mechanisms
.
+
ꢀ
(PH3)4(P C-SiH3)] directly led to the silicate complex C,
implying that the actual barrier for the F!D conversion is
probably much lower than 26.1 kcalmolꢁ1. An interesting
feature is the large energy difference of 33.9 kcalmolꢁ1
between the cyaphide complex E and isocyaphide transi-
tion-state complex G, which can be understood in terms of the
[1] See for example polynuclear cyano complexes in materials
science: a) M. Verdaguer, A. Bleuzen, V. Marvaud, J. Vaisser-
mann, M. Seuleiman, C. Desplanches, A. Scuiller, C. Train, R.
Garde, G. Gelly, C. Lomenech, I. Rosenman, P. Veillet, C.
Cartier, F. Villain, Coord. Chem. Rev. 1999, 190–192, 1023;
b) K. R. Dunbar, R. A. Heintz, Prog. Inorg. Chem. 1996, 45, 283;
c) W. R. Entley, G. S. Girolami, Science 1995, 268, 397.
[2] J. Cordaro, D. Stein, H. Grützmacher, Angew. Chem. 2006, 118,
6305; Angew. Chem. Int. Ed. 2006, 45, 6159.
ꢁ
Ru–P versus Ru–C s- and p-bonding interactions. The Ru P
ꢁ1
ꢁ
s bond in G (49.9 kcalmol ) is weaker than the Ru C s bond
in E (DEs = 62.1 kcalmolꢁ1), as a consequence of the larger
energy difference between the accepting metal d-orbital and
the donating lone pair on phosphorus (ꢁ3.9 eV) compared to
[3] R. J. Angelici, Angew. Chem. 2007, 119, 334; Angew. Chem. Int.
Ed. 2007, 46, 330.
ꢁ
carbon (0.6 eV). The Ru P p bond in G (DEp = 16.1 kcal
ꢁ1
ꢁ
mol ) is reduced compared to the Ru C p bond of E (DEp =
[4] Structure data: 3: Colorless crystals suitable for X-ray analysis
were obtained directly from the reaction vessel when wet
CH3CN was used as solvent in place of THF;
C71H65OP5RuSi·C2H3N, monoclinic, space group P21/c, a =
15.718(1), b = 22.248(1), c = 19.197(1) Š, V= 6218.5(6) Š3, Z =
23.6 kcalmolꢁ1) as there is less favorable overlap. As a result,
the phosphorus-coordinated isocyaphide complex G is even
higher in energy than the h2-bonded complex D, and thus G is
ꢁ
ꢀ
the transition state for the rotation of the C P moiety.
4, 1calcd = 1.345 Mgmꢁ3
0.09 mm3, Bruker SMART Apex diffractometer with CCD
area detector, MoKa radiation (0.71073 Š), 200 K, 2Vmax
,
crystal dimensions 0.54 ” 0.40 ”
The investigations presented herein are an example of an
observed intermediate not being on the pathway leading to
product formation. The observance of intermediate X merely
reflects the kinetic preference of phenolate addition to
phosphorus.[12] Calculations predict a transition state that is
much too high in energy from intermediate B for path A to be
a reasonable mechanism leading to cyaphide complex 2.
Instead, direct attack of phenolate at silicon is proposed.
While kinetically less favorable than the addition of pheno-
late to phosphorus, the nucleophilic substitution reaction at
silicon is the product-forming step. Two important properties
associated with phenolate as the nucleophile in the reaction
with 1 are notable. First, the lack of a-hydrogen means
rearrangement to a phosphaoximato complex is not possible
if phenolate attacks at phosphorus. Second, despite being
kinetically more favorable, the initial attack at the phospho-
=
56.748, 67870 reflections, 15535 independent (Rint = 0.0268),
direct methods; refinement against full matrix (versus F2) with
SHELXTL (v.6.12) and SHELXL-97; 744 parameters, R1 =
0.0459 and wR2 (all data) = 0.1290, max./min. residual electron
density 1.499/ꢁ0.539 eŠꢁ3. The main residual electron density is
located at the heavy atom ruthenium. All non-hydrogen atoms
were refined anisotropically. The contribution of the hydrogen
atoms, in their calculated positions, was included in the refine-
ment using a riding model. CCDC-634845 (3) contains the
supplementary crystallographic data for this paper. These data
can be obtained free of charge from The Cambridge Crystallo-
graphic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
[5] Multiple bonds and Low Coordination in Phosphorus Chemistry
(Eds.: M. Regitz, O. J.Scherer), Thieme, Stuttgart, 1990.
=
=
[6] [Ru(PPh3)2(CNtBu)2{P( O)CtBu(C O)}]: A. F. Hill, C. Jones,
A. J. P. White, D. J. Williams, J. D. E. T. Wilton-Ely, Chem.
Commun. 1998, 367.
ꢀ
rus center of P C-SiPh3 must be reversible. While we could
not compute the proposed ion-pair intermediate with suffi-
cient accuracy, we assume that the barrier to phenolate
dissociation from phosphorus is low. The irreversible elimi-
[7] R. B. Bedford, A. F. Hill, C. Jones, A. J. P. White, D. J. Williams,
J. D. E. T. Wilton-Ely, Organometallics 1998, 17, 4744.
[8] a) N. Merceron-Saffon, A. Bacereido, H. Gornitzka, G. Ber-
trand, Science 2003, 301, 1223; b) C. Buron, H. Gornitzka, V.
Romanenko, G. Bertrand, Science 2000, 288, 834; c) D. Bour-
issou, O. Guerret, F. P. Gabbaï, G. Bertrand, Chem. Rev. 2000,
100, 39.
ꢁ
nation of Ph3Si OPh by phenolate attack at silicon subse-
quently drives the reaction to completion.
[9] DFT calculations were performed using ADF2006.01, E. J.
Baerends, J. Autschbach, A. BØrces, F. M. Bickelhaupt, C. Bo,
P. M. Boerrigter, L. Cavallo, D. P. Chong, L. Deng, R. M.
Dickson, D. E. Ellis, M. van Faassen, L. Fan, T. H. Fischer, C.
Fonseca Guerra, S. J.A. van Gisbergen, J.A. Groeneveld, O. V.
Gritsenko, M. Grüning, F. E. Harris, P. van den Hoek, C. R.
Jacob, H. Jacobsen, L. Jensen, G. van Kessel, F. Kootstra, E.
van Lenthe, D. A. McCormack, A. Michalak, J. Neugebauer,
V. P. Osinga, S. Patchkovskii, P. H. T. Philipsen, D. Post, C. C.
Pye, W. Ravenek, P. Ros, P. R. T. Schipper, G. Schreckenbach,
J. G. Snijders, M. Solà, M. Swart, D. Swerhone, G. te Velde, P.
Vernooijs, L. Versluis, L. Visscher, O. Visser, F. Wang, T. A.
Wesolowski, E. van Wezenbeek, G. Wiesenekker, S. K. Wolff,
T. K. Woo, A. L. Yakovlev, and T. Ziegler, SCM, Theoretical
Chemistry, Vrije Universiteit, Amsterdam, The Netherlands.
The exchange-correlation potential is based on the GGA
exchange functional OPTX(N. C. Handy, A. J. Cohen, Mol.
Phys. 2001, 99, 403) in combination with the non-empirical PBE
(J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 1996, 77,
Experimental Section
ꢀ
3: [RuH(dppe)2(P C-SiPh3)][OTf] (1; 0.135 g, 0.10 mmol) and LiOH
(0.018 g, 1.0 mmol) were combined in THF (10 mL) under argon for
4 days. The reaction mixture was filtered through diatomaceous earth
and the volatile materials were removed under reduced pressure. The
resulting off-white solid was washed with CH3CN and residual solvent
removed in vacuo. Yield 0.065 g of 3 (53.3%). Selected NMR
spectroscopy data: 1H (300 MHz, [D8]THF): d = 2.91 (d, JHP
=
2
2
=
9.14 Hz, (O)P CH, 1H), ꢁ10.17 ppm (dq, trans- JHP = 69.9 Hz, cis-
2JHP = 20.5 Hz, RuH, 1H). 13C{1H} (75.5 MHz, [D8]THF): d =
110.5 ppm (d, 1JCP = 5.3 Hz, (O)P CH-SiPh3). 31P{1H} (121.5 MHz,
=
[D8]THF): d = 332.0 (quintet, 2JPP = 28 Hz, (O)P CHSiPh3),
=
67.7 ppm (d, 2JPP = 28 Hz, dppe). For further details see the Support-
ing Information.
Received: June 5, 2007
Published online: September 4, 2007
7880 www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 7878 –7881

Angewandte
Chemie
3865) (OPBE) and an uncontracted triple-zeta valence-plus-
polarization STO basis set is used for all the atoms including
relativistic effects by the ZORA approximation (E. van Lenthe,
A. W. Ehlers, E. J. Baerends, J. Chem. Phys. 1999, 110, 8943).
Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D.
Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K.
Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul,
S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P.
Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A.
Al-Laham, C. Y. Peng, A. Nanayakkara, M. Challacombe,
P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong, C. Gonzalez,
J. A. Pople, Gaussian, Inc., Wallingford, CT, 2004.
[10] A comparison of the calculated versus experimental 31P NMR
31
= =
chemical shifts of [RuH(PH3)4{O P CH(SiH3)}], d( P)calcd
=
345 ppm versus d(31P)exp. = 332 ppm in 3, and [RuH(PH3)4(C
P)], d(31P)calcd = 140 ppm versus d(31P)exp. = 165 ppm in 2, make
us confident that the chosen level of theory, B3LYP/GIAO 6-
311 + G* (C,H,O,Si,P), LACV3P(Ru), is sufficient. The shield-
ings are referenced against the calculated value of P(CF3)3 which
was set to d = ꢁ3 ppm versus 30% H3PO4. The GIAO calcu-
lations were performed with Gaussian03, RevisionC.02, M. J.
Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, J. A. Montgomery, Jr., T. Vreven, K. N. Kudin,
J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V. Barone, B.
Mennucci, M. Cossi, G. Scalmani, N. Rega, G. A. Petersson, H.
Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J.
Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H.
Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian, J. B. Cross,
V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli,
J. W. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P.
ꢀ
[11] This chemical shift is typical for C-metallophosphaalkenes,RP =
C[MLn]R. L. Weber, Angew. Chem. 1996, 108, 292; Angew.
Chem. Int. Ed. Engl. 1996, 35, 271. Nucleophilic alkoxide attack
at the carbon atom of 1 leads to P-metallophosphaalkenes;
=
calculations show that the more stable Z-[RuH(PH3)4(P
C(OR)SiPh3)] is only 2.1 kcalmolꢁ1 less stable than B but the
theoretically predicted d(31P)calcd of 522 ppm (in good agreement
with experimentally detected values) rule out this isomer as a
possible structure for intermediate X. The E isomer (d31Pcalcd
=
599 ppm) is 8.6 kcalmolꢁ1 higher in energy than the Z isomer.
[12] The nucleophilic attack on phosphaalkynes at phosphorus is well
documented, see: a) J. F. Nixon, Coord. Chem. Rev. 1995, 145,
201; b) F. Mathey, Angew. Chem. 2003, 115, 1616; Angew. Chem.
Int. Ed. 2003, 42, 1578;and references therein.
Angew. Chem. Int. Ed. 2007, 46, 7878 –7881
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7881