Ligand substitution processes on carbonylmetal derivatives. 3.1 Reaction of hydridopentacarbonylchromates with phosphines

Article abstract of DOI:10.1021/om020130b

The reaction of K⁺[HCr(CO)₅]^- with phosphanes PR₃ (R = Et, Ph, NMe₂) in THF at 65 °C affords the disubstituted complexes trans-Cr(CO)₄(PR₃)₂ isolated in 57-70% yield.

Full text of DOI:10.1021/om020130b

3388
Organometallics 2002, 21, 3388-3394
Liga n d Su bstitu tion P r ocesses on Ca r bon ylm eta l
Der iva tives. 3.1 Rea ction of
Hyd r id op en ta ca r bon ylch r om a tes w ith P h osp h in es
J ean-J acques Brunet,* Ousmane Diallo, Bruno Donnadieu, and
Emmanuel Roblou
Laboratoire de Chimie de Coordination du CNRS, Unite´ No. 8241, Lie´e par Conventions,
a` l’Universite´ Paul Sabatier et a` l’Institut National Polytechnique, 205 Route de Narbonne,
31077 Toulouse Cedex 04, France
Received February 19, 2002
The reaction of K+[HCr(CO)5]- with phosphanes PR3 (R ) Et, Ph, NMe2) in THF at 65 °C
affords the disubstituted complexes trans-Cr(CO)4(PR3)2 isolated in 57-70% yield. The X-ray
crystal structures of trans-Cr(CO)4(PR3)2 derivatives have been determined for R ) Et and
NMe2. These reactions proceed first by exchange of one carbon monoxide ligand, generating
the monosubstituted hydridotetracarbonylchromates K+[HCr(CO)4PR3]-, observed and
characterized by NMR spectroscopy for R ) Et and Ph. The second step involves substitu-
tion of the hydride ligand of K+[HCr(CO)4PR3]- to give the disubstituted derivatives trans-
Cr(CO)4(PR3)2. These ligand exchange processes are discussed and compared with the reaction
-
of phosphanes with the dinuclear bridged K+(µ-H)[Cr(CO)5]2 .
In tr od u ction
tuted Fe(CO)2P3 complexes.12 Furthermore, the reaction
of 1 with phosphines in THF produced a 1/1 mixture of
Fe(CO)3P2 and K2Fe(CO)4, which can be easily sepa-
rated.6 This reaction constitutes the more convenient
preparation of K2Fe(CO)4,13 a non-pyrophoric substi-
tute of the Collman reagent, Na2Fe(CO)4.14 The reaction
mechanisms have been rationalized.1,6,15 In all cases, the
first step is the substitution of a carbon monoxide
ligand of K+[HFe(CO)4]-, leading to K+[HFe(CO)3P]-
complexes. In only one case (P ) P(OMe)3, THF as
solvent) a second CO substitution was observed, leading
quantitatively to the disubstituted hydridoferrate
K+[HFe(CO)2P2]-.1 In other cases, depending on the
nature of P and the nature of the solvent, the
K+[HFe(CO)3P]- complexes either can be isolated or
evolve in situ to selectively afford one of the above
neutral complexes.5
Ligand substitution processes in transition-metal
complexes have attracted much attention, both for the
design of specific syntheses of heteroleptic complexes
and for a contribution to the understanding of reaction
mechanisms, particularly the loss of small molecules or
ions from low-valent metal complexes.2,3 In this respect,
neutral metal carbonyls, especially iron carbonyls, have
been particularly studied.4
In the series of iron carbonyls, we were the first to
report, some years ago, that the anionic hydrido com-
plex, K+[HFe(CO)4]-, 1, is a very versatile material for
the high yield synthesis of a large variety of phosphane-
substituted iron carbonyl complexes, most of which were
described for the first time.5 Indeed, depending on the
characteristics of the phosphorus ligand (cone angle θ,
pKa) and the nature of the solvent (protic or aprotic),
the reaction of 1 with phophites and phosphines (there-
after designated P) highly selectively leads to monosub-
stituted hydrido complexes K+[HFe(CO)3P]- 6 or neutral
disubstituted dihydrido complexes H2Fe(CO)2P2,7-9 but
also to disubstituted Fe(CO)3P2,10,11 and even trisubsti-
As part of our interest in the applications of hydri-
docarbonylmetalates in organic synthesis and homoge-
neous catalysis,16,17 we recently investigated the study
of the reactivity of the group VI analogue K+[HCr(CO)5]-,
2, in organic synthesis.18-20 In addition, in light of the
(7) Brunet, J .-J .; Kindela, F.-B.; Labroue, D.; Neibecker, D. Inorg.
Chem. 1990, 29, 4152-4153.
* Corresponding author. Fax: 33 5 61 555 30 03. E-mail: brunet@
lcc-toulouse.fr.
(8) Brunet, J .-J .; Kindela, F.-B.; Neibecker, D. Inorg. Synth. 1992,
29, 156-160.
(1) Part of the Ph.D. Thesis of E. Roblou. For Part 2 of the series,
see: Brunet, J .-J .; Commenges, G.; Kindela, F.-B.; Neibecker, D.
Organometallics 1992, 11, 3023-3030.
(2) Basolo, F.; Pearson, R. G. Mechanisms of Inorganic Reactions;
Wiley: New York, 1968; Basolo, F. Inorg. Chim. Acta 1981, 50, 65-
70.
(3) Albers, M. O.; Coville, N. J . Coord. Chem. Rev. 1984, 53, 227-
259. Luth, T. Y. Coord. Chem. Rev. 1984, 60, 255-276.
(4) Darensbourg, D. J . In Advances in Organometallic Chemistry;
Stone, F. G. A., West, R., Eds.; Academic Press: New York, 1982; Vol.
21, pp 113-150.
(5) For a review, see: Brunet, J .-J .; Chauvin, R.; Diallo, O.; Kindela,
P.; Leglaye, P.; Neibecker, D. Coord. Chem. Rev. 1998, 178-180, 331-
351.
(9) Arion, V.; Brunet, J .-J .; Neibecker, D. Inorg. Chem. 2001, 2628-
2630.
(10) Brunet, J .-J .; Kindela, F.-B.; Neibecker, J . Organomet. Chem.
1989, 368, 209-212.
(11) Brunet, J .-J .; Kindela, F.-B.; Neibecker, D. Inorg. Synth. 1992,
29, 151-156.
(12) Brunet, J .-J .; Kindela, F.-B.; Neibecker, D. Inorg. Synth. 1996,
31, 202-210.
(13) Baby, A.; Brunet, J .-J .; Kindela, F.-B.; Neibecker, D. Synth.
Comm. 1994, 24, 2827-2834.
(14) Collman, J . P. Acc. Chem. Res. 1975, 8, 342-347.
(15) Brunet, J .-J .; Kindela, F.-B.; Neibecker, D. Phosphorus, Sulfur
Silicon 1993, 77, 65-68.
(6) Brunet, J .-J .; Commenges, G.; Kindela, F.-B.; Neibecker, D.
Organometallics 1992, 11, 1343-1350.
(16) Brunet, J .-J . Chem. Rev. 1990, 90, 1041-1059.
(17) Brunet, J .-J . Eur. J . Inorg. Chem. 2000, 1377-1390.
10.1021/om020130b CCC: $22.00 © 2002 American Chemical Society
Publication on Web 06/29/2002

Hydridopentacarbonylchromates
Organometallics, Vol. 21, No. 16, 2002 3389
Ta ble 1. Ma in NMR Ch a r a cter istics of th e Ch r om iu m Ca r bon yl Der iva tivesa ,b
compound
KHCr(CO)5
KHCr(CO)4PEt3
KHCr(CO)4PPh3
KHCr2(CO)10
31P{1H} δ (ppm)
1H δ (ppm), J (Hz)
compound
Cr(CO)5PEt3
Cr(CO4(PEt3)2
Cr(CO)5PPh3
31P{1H} δ (ppm)
-6.97
37.9
51.8
59.0
54.6
80.9
-7.07 (d), 38
-6.25 (d), 35
-19.49
Cr(CO)4(PPh3)2
78.0
KHCr2(CO)9PEt3
KHCr2(CO)8(PEt3)2
KHCr2(CO)9PPh3
-17.85(d), 26
-17.16 (t), 39
-7.23 (d), 24
Cr(CO)5P(NMe2)3
Cr(CO)4[P(NMe2)3]2
Cr(CO)5[HP(NMe2)2]d
159.5
177.8
137.8
c
a
b
In THF-d8 unless otherwise noted. As references, the 31P{1H} chemical shift of the free ligands were the following: PEt3, δ ) -18.5
ppm; PPh3, δ ) -1.5 ppm; P(NMe2)3, δ ) 121.9 ppm. In CD3CN. 1H NMR: δ ) 6.84 ppm (d), J H-P ) 398 Hz.
c
d
1
coordination chemistry developed with 1, we were
obviously interested in examining the reaction of 2 with
phosphanes. In a short comparative study of the coor-
dination chemistry of 1 and 2, we already noticed some
differences, which were tentatively attributed to the
lower acidity of 2 as compared to that of 1.18 We report
here full details of the reaction of 2 with PEt3, PPh3,
and P(NMe2)3. The goal of this study was to examine
the possible synthetic interest of these reactions, but
also to gain more insight into the ligand substitution
processes on hydridocarbonylmetalates.
Resu lts a n d Discu ssion
Contrary to K+[HFe(CO)4]-, which is stable in protic
media, K+[HCr(CO)5]- reacts with alcohols to generate
the stable bridged dinuclear hydride K+(µ-H)[Cr(CO)5]2
,
-
3, with loss of molecular hydrogen.5,21 Thus, only
reactions of 2 in aprotic solvents have been considered.
F igu r e 1. Molecular structure of complex 4 with atom-
labeling scheme. Ellipsoids represent 50% probability. Se-
lected bond lengths (Å) and angles (deg): Cr-P 2.329(1);
Cr-C(1) 1.863(3); Cr-C(2) 1.869(3); P-C(11) 1.839(3);
P-C(21) 1.845(3); P-C(31) 1.836(3); C(1)-Cr-C(2)
91.1(1); C(1)-Cr-C(2)′ 88.9(1) [Symmetry transformations
used to generate equivalent atoms: -x+1, -y, -z].
-
Reactions of (µ-H)[Cr(CO)5]2 (Et4N+ salt) with phos-
phines have been reported before22,23 and will be dis-
cussed comparatively.
Rea ction of 2 w ith Tr ieth ylp h osp h in e (θ ) 132°).
The reaction of 2 with 1 equiv of PEt3 in THF at 55 °C
was monitored by 31P{1H} NMR. After 2 h reaction,
besides the signal of unreacted PEt3 (δ ) - 18.5 ppm),
two new signals were observed at 54.6 ppm (major) and
51.8 ppm (minor). After 4 h, the signal at 51.8 ppm had
increased, and after 23 h, all the phosphine had been
consumed to leave only the signal at 51.8 ppm, together
with very small signals at 40.9 and 35.5 ppm. None of
the above signals corresponded to the monosubstituted
complex Cr(CO)5PEt3 (δ ) 37.9 ppm), prepared inde-
pendently according to a literature procedure.24 The
signal at 51.8 ppm could be attributed to the disubsti-
tuted complex trans-Cr(CO)4(PEt3)2, 4 (vide infra). It
must be noted that, in this reaction, the fate of half of
the chromium species has not been determined. How-
ever, when the reaction was performed with 2 equiv of
PEt3 in THF at 65 °C, a similar progress was observed
and 4 was the only product formed (quantitative con-
sumption of PEt3). This reaction can thus be formally
represented by eq 1. The reaction product 4 (yellow
solid) could be isolated analytically pure in 70% yield
and fully characterized (Table 1). Moreover, suitable
single crystals were obtained, and the structure of 4 was
established by X-ray diffraction (Figure 1, Table 2).
This result is in contrast with the chemistry developed
with the hydridocarbonylferrate 1. Indeed, the reaction
of 1 with phosphines (PPh3, PBu3) in THF occurs as
shown by eq 2, and the yield of the disubstituted
complex Fe(CO)3(PR3)2 cannot exceed 50%.6
(18) Brunet, J .-J .; Chauvin, R.; Leglaye, P. J . Organomet. Chem.
1998, 571, 7-13.
(19) Brunet, J .-J .; Chauvin, R.; Leglaye, P. Eur. J . Inorg. Chem.
1999, 713-716.
(20) Andrieu, B.; Brunet, J .-J .; Diallo, O.; Donnadieu, B.; Lienafa,
J .; Roblou, E. J . Organomet. Chem. 2002, 643-44, 27-31.
(21) Marko, L.; Nagy-Magos, N. J . Organomet. Chem. 1985, 285,
193-203.
In reaction 2, the monosubstituted hydride [HFe-
(CO)3PR3]- complex formed in the first step is suf-
ficiently basic to deprotonate [HFe(CO)4]-, yielding the
dianionic species [Fe(CO)4]2- and a neutral dihydride
H2Fe(CO)3PR3. The latter is not stable at the reaction
temperature and loses H2 to generate the unsaturated
(22) Darensbourg, M. Y.; Walker, N. J . Organomet. Chem. 1976, 117,
C68-C74.
(23) Darensbourg, M. Y.; Walker, N.; Burch, R. R. Inorg. Chem.
1978, 17, 52-56.
(24) Vincent, E.; Verdonk, L.; Van der Kelen, G. P. J . Mol. Struct.
1980, 69, 33-40.

3390 Organometallics, Vol. 21, No. 16, 2002
Ta ble 2. Cr ysta l Da ta a n d Str u ctu r e Refin em en t for Com p lexes 4 a n d 12
Brunet et al.
identification code
empirical formula
fw
4
12
C8H15Cr0.5O2P
200.17
C16H36CrN6O4P2
490.45
temperature
wavelength
cryst syst
space group
a, Å
180(2) K
0.71073 Å
triclinic
P1h
7.540(5)
7.719(5)
9.138(5)
180(2) K
0.71073 Å
orthorhombic
Pbca
15.4663(13)
13.7007(11)
23.544(2)
90°
b, Å
c, Å
R, deg
108.086(5)
96.328(5)
â, deg
90°
90°
γ, deg
91.565(5)
volume
501.4(5) Å3
4989.0(7) Å3
8
Z
2
density (calcd)
abs coeff
F(000)
cryst size
θ range for data collection
index ranges
1.326 Mg/m3
0.744 mm-1
212
1.306 Mg/m3
0.618 mm-1
2080
0.5 × 0.37 × 0.13 mm3
3.82° to 24.38°
-8 e h e 8, -8 e k e 8,
-10 e l e 10
4037
0.37 × 0.25 × 0.13 mm3
3.27° to 20.81°
-15 e h e 13, -13 e k e 13,
-22 e l e 23
18 034
2600 [R(int) ) 0.1760]
99.4%
empirical (DIFABS)
0.6826 and 0.2171
full-matrix least-squares on F2
2600/0/274
no. of reflns collected
no. of ind reflns
completeness to θ ) 24.38°
abs corr
max. and min. transmn
refinement method
no. of data/restraints/params
goodness-of-fit on F2
final R indices [I>2σ(I)]
R indices (all data)
1520 [R(int) ) 0.0217]
92.3%
empirical (DIFABS)
0.8720 and 0.5770
full-matrix least-squares on F2
1520/0/49
1.071
1.150
R1 ) 0.0455, wR2 ) 0.1098
R1 ) 0.0482, wR2 ) 0.1118
0.658 and -0.520 e Å-3
R1 ) 0.0733, wR2 ) 0.1216
R1 ) 0.1118, wR2 ) 0.1376
0.250 and -0.280 e Å-3
largest diff peak and hole
species “Fe(CO)3(PR3)”, which is finally scavenged by
the phosphine to afford Fe(CO)3(PR3)2 in e50% yield.
Formation of 4 in 70% isolated yield (eq 1) thus
suggests that, if it is generated, the monosubstituted
complex K+[HCr(CO)4PEt3]- does not react as its ana-
logue in the iron series. To try to gain more insight on
this point, the reaction of 2 with PEt3 (2 equiv) was
conducted in THF-d8 in a sealed NMR tube and moni-
tored by both 31P and 1H NMR. After 10 h at room
temperature, the 31P{1H} NMR spectrum indicated the
appearance of the signals at 54.6 and 51.6 ppm, while
the major signal was that of unreacted PEt3. Interest-
ingly, the 1H NMR spectrum (hydrides region) exhibited
two signals, a major one at - 6.97 ppm (KHCr(CO)5)
and a minor one (ca. 10%) at - 7.12 ppm. In fact, the
latter corresponded to a doublet, one branch of which
is buried beneath the signal at - 6.97 ppm. Indeed,
selective irradiation experiments evidenced the coupling
of this hydride (δ ) - 7.05 ppm, d, 2J H-P ) 38 Hz) with
the phosphorus atom, giving the 54.6 ppm signal (d) in
the 31P NMR spectrum. These signals are attributed to
the monosubstituted hydridotetracarbonylchromate
K+[HCr(CO)4PEt3]-, 5 (eq 3). Indeed, the 1H NMR
chemical shift of the substituted hydride 5 is close to
that of 2, as already observed in the tungsten series
([HW(CO)5]-, δ ) - 4.2 ppm; [HW(CO)4PMe3]-, δ )
- 3.6 ppm).25 Moreover, this assignment is also sup-
ported by the fact that the possible mono- and disub-
stituted dinuclear complexes K+(µ-H)[Cr(CO)5][Cr-
observed in the molybdenum series.23 Heating the NMR
tube at 65 °C allowed observation of the slow consump-
tion of 2, 5, and PEt3 to give 4 (part of which precipitated
as yellow crystals) (eq 4), contaminated by some traces
of an unidentified compound exhibiting a 31P NMR
signal at 35.5 ppm. The overall results can be sum-
marized by eqs 3 and 4.
During these experiments, we could observe that 2
was stable for at least 2 days in THF-d8 at 65 °C (in
the presence of liberated carbon monoxide) without
observable transformation to K+(µ-H)[Cr(CO)5]2-. Fur-
thermore, the reaction seemed slow to the end, an
observation that could be related to the fact that, in a
sealed tube, the displaced carbon monoxide cannot be
exhausted and may decrease the rate of the substitution
processes. Last, we believed that 5 did not react with 2
at a significant rate since they could be observed
together (1H NMR analysis) for long reaction times. This
observation became still more obvious when the reaction
was conducted in CD3CN. In this case, we could observe
(CO)4PEt3]- and K+(µ-H)[Cr(CO)4PEt3]2 have been
-
observed in independent experiments (vide infra) and
exhibit 1H NMR chemical shifts close to that of the
parent unsubstituted dinuclear hydride 3, as previously
1
by H NMR a mixture of 2 and 5 for several hours at
65 °C, whereas the consumption of PEt3 was slow.
Formation of disubstituted M(CO)4(PPh3)2 complexes
(M ) Cr, Mo, W) is known to occur when the corre-
(25) Slater, S. G.; Lusk, R.; Schuman, B. F.; Darensbourg, M. Y.
Organometallics 1982, 1, 1662-1666.

Hydridopentacarbonylchromates
Organometallics, Vol. 21, No. 16, 2002 3391
Sch em e 1
sponding hexacarbonyl M(CO)6 is reacted with the
phosphine (5 equiv) in the presence of NaBH4 (4 equiv)
in ethanol.26 Although the exact mechanism involved
in these reactions is not known, Darensbourg et al. have
reported convincing data indicating that the dinuclear
that, besides path A, the suggested path B, through
[HCr(CO)5]- (resulting from the disruption of the di-
nuclear monosubstituted hydride), effectively occurs. It
must be noted that, as suggested by one reviewer, the
likelihood that hydride would be extracted from the
binuclear complex 7 is small. In contrast, the disruption
of 7 into 5 and the 16-electron “Cr(CO)4PEt3” species
seems more reasonable than generating two 16-electron
“Cr(CO)4PEt3” species simultaneously.
-
hydride Na+(µ-H)[Cr(CO)5]2 may be the intermediate
on which the carbonyl substitution occurs, either in
ethanol or in THF.23 Since K+(µ-H)[Cr(CO)5]2-, 3, may
result from the evolution of 2,17 we were interested in
comparing the reaction of PEt3 with 2 and 3 under
the same conditions. Thus, the reaction of 3 (free of
any trace of 2) with 4 equiv of PEt3 was performed in
CD3CN in a NMR sealed tube at room temperature. The
1H NMR monitoring of the reaction showed interesting
features. After 10 h, we could observe unreacted 3 (s, δ
) -19.49 ppm), but also significant amounts of 2 (s, δ
) -6.97 ppm). Later on (78 h), we could observe
simultaneously 3 and 2, but also the monosubstituted
mononuclear derivative K+[HCr(CO)4PEt3]-, 5 (vide
supra), the monosubstituted dinuclear complex K+(µ-
H)[Cr(CO)4PEt3][Cr(CO)5]-, 6 (d, δ ) - 17.85 ppm,
2J H-P ) 26 Hz), and the disubstituted dinuclear complex
The reaction of PEt3 with [HCr(CO)5]- thus occurs by
direct substitution of a carbonyl ligand by the phos-
phine, generating the observable intermediate 5, fol-
lowed by substitution of the hydride ligand by another
molecule of the phosphine. Since [HCr(CO)5]- is hexa-
coordinate, the first step probably involves dissociation
of the CO ligand favored as the temperature increases.
For neutral carbonylmetals, another well-known way
to promote carbon monoxide dissociation is photochemi-
cal activation.27 To the best of our knowledge, such an
activation has never been reported for anionic carbon-
ylmetal species. When the reaction of 2 with 2 equiv of
PEt3 was conducted at room temperature under irradia-
tion (365 nm), a slow reaction occurred during which
we could observe the intermediate 5 (NMR monitoring).
The latter is slowly consumed to give the expected
complex 4, together with the trisubstituted derivative
mer-Cr(CO)3(PEt3)3, 8, identified on the basis of its
31P{1H} NMR characteristics (δtrans ) 48.2 ppm, d, 2J P-P
K+(µ-H)[Cr(CO)4PEt3]2-, 7 (t, δ ) - 17.16 ppm, J H-P
2
) 39 Hz). Unexpectedly, the 31P NMR spectra indicated
the formation of both the mono- and the disubstituted
neutral complexes Cr(CO)5PEt3 and trans-Cr(CO)4-
(PEt3)2 in nearly equivalent amounts. This observation
may be related to the fact that the reaction was
performed in a sealed tube, so that the displaced carbon
monoxide was still present, and thus might compete
with the phosphine in the reaction with the unsaturated
“Cr(CO)4PEt3” intermediate (Scheme 1). A similar ob-
servation has been mentioned by Darensbourg et al.22
We have verified independently that the monosubsti-
tuted complex Cr(CO)5PEt3 did not afford significant
amounts of 4 when reacted with 2 equiv of PEt3 for 3
days in THF at 65 °C.
2
) 22 Hz and δcis ) 35.1 ppm, t, J P-P ) 22 Hz).
Rea ction of 2 w ith Tr ip h en ylp h osp h in e P P h 3 (θ
) 145°). The reaction of 2 with 2 equiv of triphen-
ylphosphine in THF proceeds as that with triethylphos-
phine. The reaction is extremely slow at room temper-
ature, but proceeds in refluxing THF. Monitoring the
reaction by 31P NMR allowed the observation of the
formation of the expected intermediate K+[HCr(CO)4-
2
PPh3]-, 10 (THF-d8, δ ) 80.9 ppm, J P-H ) 35 Hz),
These observations strongly support the mechanism
already fully characterized as the cis H-Cr-P isomer.18
proposed by Darensbourg et al.23 (Scheme 1) and prove
Classical workup after 72 h reaction in refluxing THF
(26) Chatt, J .; Leigh, G. J .; Thankarajan, N. J . Organomet. Chem.
1971, 297, 105-110.
(27) For a recent paper, see: Bergstrom, C. L.; Luck, R. L. Inorg.
Chim. Acta 2001, 318, 77-83, and references therein.

3392 Organometallics, Vol. 21, No. 16, 2002
Brunet et al.
Sch em e 2
allowed the recovery of unreacted triphenylphosphine
(60%) and the disubstituted complex trans-Cr(CO)4-
(PPh3)2, 9 (analytically pure, 25% yield) (eq 5), which
could be fully characterized (Table 1). The X-ray struc-
ture of 9 has been reported recently.28
90/10 ratio.30 However, even long reaction times (100
h) did not promote complete consumption of P(NMe2)3.
These results suggest that at room temperature the
major reaction is the nucleophilic attack of the hydride
ligand of 2 (already hexacoordinated) at the phosphorus
atom of the phosphine, a reaction that is greatly
accelerated if performed on a protonated form of
P(NMe2)3 (eq 7). In contrast, at higher temperatures,
dissociation of a carbonyl ligand of 2 may be favored,
and the direct ligand exchange at the chromium site
proceeds to form 12. Note that, contrary to what was
observed with PEt3 and PPh3, we could not detect the
expected intermediate K+[HCr(CO)4P(NMe2)3]-, which
might be rapidly converted to the neutral disubstituted
derivative trans-Cr(CO)4[P(NMe2)3]2. Interestingly, when
2 was reacted with P(NMe2)3 (2 equiv) under irradiation
(365 nm), a quantitative reaction was observed, within
38 h, affording 12 selectively (57% isolated yield) (eq
8). Single crystals of 12 were obtained, and the structure
was determined by X-ray analysis (Figure 2, Table 2).
Rea ction of 2 w ith Tr is(d im eth yla m in o)p h os-
p h in e (θ ) 157°). The reaction of K+[HFe(CO)4]- with
P(NMe2)3 does not proceed like that with PR3 or PAr3.
Indeed, instead of promoting the substitution of a
carbonyl ligand of 1 by P(NMe2)3, the reaction occurs
by attack of 1 on the phosphorus atom (substitution of
a “NMe2” group) to give the monosubstituted secondary
phosphine complex Fe(CO)4[HP(NMe2)2] with simulta-
neous formation of an equivalent molecular amount of
K2Fe(CO)4 (eq 6).29
We have briefly reported that the reaction of 2 with
P(NMe2)3 is very slow at room temperature, but that if
P(NMe2)3 is first treated (independently) by AcOH, the
reaction rapidly affords the corresponding monosubsti-
tuted complex Cr(CO)5[HP(NMe2)2], 11 (eq 7).18 These
observations have been explained by the lower acidity
of 2 as compared to that of 1 (toward P(NMe2)3).
Gen er a l Com m en ts
Recently reported theoretical studies31 on the effect
of ligands X on the CO labilization in [XM(CO)5]-
complexes (M ) Cr, Mo, W) indicate that CO dissocia-
tion energies increase in the series X ) NH2 < OH, Cl
< Me < H (i.e., π-donor ligands favor CO dissociation
much more than σ-donor ones) and that CO loss is easier
from the cis position. Although these studies do not take
into account the well-known selective site ion-pairing
interactions,32,33 the calculations are in agreement with
experimental data. As a matter of fact, CO substitution
on [ClCr(CO)5]- by phosphines and phosphites easily
occurs at room temperature (eq 9).34
Studying the reaction of P(NMe2)3 with 2 in more
detail indicated that the reaction at room temperature
in THF very slowly generates 11 (31P NMR, δ ) 137.8
1
ppm, d, J P-H ) 398 Hz, 1H NMR, δ ) 6.84 ppm, d,
1J H-P ) 398 Hz) together with smaller amounts of trans-
Cr(CO)4[P(NMe2)3]2, 12 (31P{1H} NMR, δ ) 177.8 ppm)
(Scheme 2). In contrast, if the reaction was performed
in THF at 65 °C, 11 was not formed (Scheme 2). The
reaction slowly afforded a mixture of 12 and, unexpect-
edly, the known monosubstituted derivative Cr(CO)5P-
(NMe2)3, 13 (31P{1H} NMR, δ ) 159.5 ppm), in a ca.
The results obtained in the present study show that,
in agreement with the theoretical study, the CO sub-
(30) Kroshefsky, R. D.; Verkade, J . G.; Pipal, J . R. Phosphorus Sulfur
1979, 6, 377-389.
(31) Macgregor, S. A.; MacQueen, D. Inorg. Chem. 1999, 38, 4868-
4876.
(28) Bao, J .; Geib, S. J .; Cooper, N. J . J . Organomet. Chem. 2001,
631, 188192.
(29) Brunet, J . J .; Chauvin, R.; Neibecker, D. Synth. Commun. 1997,
27, 1433-1437.
(32) Darensbourg, M. Y. Prog. Inorg. Chem. 1985, 33, 222-274.
(33) Kao, S. C.; Darensbourg, M. Y.; Schenk, W. Organometallics
1984, 3, 871-876.
(34) Wolkulich, M. J .; Atwood, J . D. Inorg. Synth. 1985, 23, 37-41.

Hydridopentacarbonylchromates
Organometallics, Vol. 21, No. 16, 2002 3393
of the H-M- bond, thus favoring H- exchanges. The
observed reactivities of 2 and of the intermediates
K+[HCr(CO)4PR3]- toward phosphines are in agreement
with the above considerations.
Exp er im en ta l Section
Meth od s a n d Ma ter ia ls. All sample manipulations were
carried out under argon using standard Schlenk tube and
vacuum techniques. Solvents and solutions were transferred
via syringe or cannula. Tetrahydrofuran (SDS) was distilled
over Na/benzophenone. Acetonitrile was distilled over P2O5
under argon and stored over 3 Å molecular sieves. KHCr-
(CO)518 and KHCr2(CO)1037 were prepared according to reported
procedures. Chromium hexacarbonyl (Strem Chemicals) and
phosphines (Strem Chemicals or Fluka) were used without
further purification. Argon U (L’Air Liquide) was used after
passage through molecuar sieves.
In stu m en ta tion . Irradiations were performed with an
external 2 × 15 W lamp (312-365 nm). Infrared spectra were
recorded on a Perkin-Elmer 1725 IRFT spectrometer using
CaF2 windows (0.05 mm). H, 31P, and 13C NMR spectra were
recorded on Bru¨cker AC 200, AM 250, DPX 300, or AMX 400
apparatus.
1
F igu r e 2. Molecular structure of complex 12 with atom-
labeling scheme. Ellipsoids represent 50% probability.
Selected bond lengths (Å) and angles (deg): Cr(1)-C(1)
1.854(9); Cr(1)-C(3) 1.858(8); Cr(1)-C(4) 1.863(8); Cr(1)-
C(2) 1.888(8); Cr(1)-P(1) 2.349(2); Cr(1)-P(2) 2.356(2);
P(1)-N(13) 1.674(6); P(1)-N(12) 1.677(6); P(1)-N(11) 1.698-
(6); P(2)-N(22) 1.682(6); P(2)-N(23) 1.689(6); P(2)-N(21)
1.691(6); C(1)-Cr(1)-C(3) 168.9(3); C(1)-Cr(1)-C(4) 82.0-
(3); C(3)-Cr(1)-C(4) 89.7(3); C(1)-Cr(1)-C(2) 89.7(3);
C(3)-Cr(1)-C(2) 99.4(3); C(4)-Cr(1)-C(2) 169.3(3); C(1)-
Cr(1)-P(1) 89.0(2); C(3)-Cr(1)-P(1) 84.5(2); C(4)-Cr(1)-
P(1) 95.1(2); C(2)-Cr(1)-P(1) 91.3(2); C(1)-Cr(1)-P(2)
94.4(2); C(3)-Cr(1)-P(2) 92.6(2); C(4)-Cr(1)-P(2) 88.3(2);
C(2)-Cr(1)-P(2) 85.7(2); P(1)-Cr(1)-P(2) 175.53(8).
Rea ction s. Th er m a l Activa tion . The reaction procedures
is exemplified in the case of triethylphosphine: In a Schlenk
tube, triethylphosphine (3.41 mmol) was added to KHCr(C0)5
(solid, 1.66 mmol), and freshly distilled THF (30 mL) was
added. The solution was magnetically stirred for 3 days at 67
°C (oil bath). The solvent was then evaporated under reduced
pressure and the reaction product, Cr(CO)4(PEt3)2, extracted
with 5 × 40 mL of hexane (yield 70%). X-ray suitable crystals
were obtained by slow crystallization from a THF/hexane (1/
4) solvent mixture.
P h otoch em ica l Activa tion . The reaction procedure is
exemplified in the case of tris(dimethylamino)phosphine: In a
quartz Schlenk tube, tris(dimethylamino)phosphine (2.57 mmol)
was added to KHCr(C0)5 (solid, 1.28 mmol), and freshly
distilled THF (24 mL) was added. The solution was irradiated
(365 nm) for 38 h. The solvent was then evaporated under
reduced pressure and the reaction product extracted with large
amounts (13 × 40 mL) of hexane (yield 57%). X-ray suitable
crystals were obtained by slow crystallization from a THF/
hexane (1/10) solvent mixture.
stitution on [HCr(CO)5]- is more difficult than that on
[ClCr(CO)5]- (vide supra), but does occur selectively (i.e.,
without substitution of the hydride ligand, which would
have generated the Cr(CO)5PR3 derivatives). Under the
conditions used, the monosubstituted hydridochromate
[HCr(CO)4PR3]- further evolves by substitution of the
hydride ligand by the phosphine, a reaction that is
related to that reported for [ClCr(CO)4PR3]- (eq 10)34
or for [AcOMo(CO)4PPh3]-.35
X-r a y Cr ysta llogr a p h ic Stu d y. Data for compound 4 were
collected on a Stoe IPDS diffractometer, whereas an Oxford-
Diffraction Xcalibur diffractometer was used for 12. The final
unit cell parameters were obtained by the least-squares
refinement of 8000 and 18 034 reflections for 4 and 12,
respectively. For both compounds, only statistical fluctuations
were observed in the intensity monitors over the course of the
data collections.
The structure was solved by direct methods (SIR97)38 and
refined by least-squares procedures on F2. In both compounds,
all H atoms were introduced at calculated positions as riding
atoms [d(CH) ) 0.99-0.98 Å], using AFIX23 for CH2 and
AFIX137 for CH3 groups, with a displacement parameter equal
to 1.2 (CH2) or 1.5 (CH3) times that of the parent atom. Least-
squares refinements were carried out by minimizing the
function ∑w(Fo2 - Fc2)2, where Fo and Fc are the observed and
calculated structure factors. The weighting scheme used in the
last refinement cycles was w ) 1/[σ2(Fo2) + (aP)2 + bP], where
It has long been established that for the reaction of
hydridometalates [HM(CO)5]- (M ) Cr, Mo, W) with
primary alkyl halides, the tendency of hydride (H-)
transfer vs SET processes is increased by substitution
of a carbonyl ligand by a less effective π-acceptor one
like phosphines or phosphites.36 This effect has been
explained by the decreased ability of the hydride ligand
to give electrons to the metal in the presence of a good
σ-donor ligand such as a phosphane. This reactivity
order also parallels the increased tendency for alkali
metal cations to interact with the hydride metal site
rather than with a carbonyl oxygen site. Such an
interaction with the hydride site results in a decrease
2
P ) (Fo + 2Fc2)/3. Models reached convergence with R )
∑(||Fo| - |Fc||)/∑(|Fo|) and wR2 ) {∑w(Fo - Fc2)2/∑w(Fo2)2}1/2
,
2
having values listed in Table 2.
(37) Grillone, N. D.; Palmisano, L. Transition Met. Chem. 1989, 14,
(35) Cotton, F. A.; Darensbourg, D. J .; Kolthammer, B. W. S. J . Am.
Chem. Soc. 1981, 103, 398-405.
(36) Kao, S. C.; Spillett, C. T.; Ash, C.; Lusk, R.; Park, Y. K.;
Darensbourg, M. Y. Orgnometallics 1985, 4, 83-91.
81-89.
(38) Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G. L.;
Giacovazzo, C.; Guagliardi, A.; Moliterni, A. G. G.; Polidori, G.; Spagn,
R. J . Appl. Crystallogr. 1999, 32, 115-119.

3394 Organometallics, Vol. 21, No. 16, 2002
Brunet et al.
The calculations were carried out with the SHELXL-97
program39 using the integrated system WINGX(1.63).40 Mo-
lecular view was realized with the help of ORTEP.41 Fractional
atomic coordinates and anisotropic thermal parameters for non
hydrogen atoms and atomic coordinates for H atoms have been
deposited at the Cambridge Crystallographic Data Centre.
Copies of the data can be obtained free of charge, on applica-
tion to the Director, CCDC, 12 Union Road Cambridge CB2
IE2, UK.
nancial support. The authors also wish to thank S.
Assie´, Y. Coppel, F. Lacassin, and G. Commenges for
registration of NMR spectra. Dr. J . C. Daran is also
gratefully acknowledged for interesting discussions
about crystallographic data.
Su p p or tin g In for m a tion Ava ila ble: This material is
available free of charge via the Internet at http://pubs.acs.org.
OM020130B
Ack n ow led gm en t. The Centre National de la Re-
cherche Scientifique (France) is acknowledged for fi-
(40) Farrugia, L. J . J . Appl. Crystallogr. 1999, 32, 837-838.
(41) (a) Burnett, M. N.; J ohnson, C. K. ORTEP-III, Report ORNL-
6895; Oak Ridge National Laboratory: Oak Ridge, TN, 1996. (b)
Farrugia, L. J . ORTEP3 for Windows. J . Appl. Crystallogr. 1997, 30,
565.
(39) Sheldrick, G. M. SHELX97, Programs for Crystal Structure
Analysis (Release 97-2); Institu¨t fu¨r Anorganische Chemie der Uni-
versita¨t: Tammanstrasse 4, D-3400 Go¨ttingen, Germany, 1998.