Direct photolytic route to trans-Cr(CO)4(AsPh3)2 and crystal structures of cis-Mo(CO)4(AsPh3)2 and cis-W(CO)4(AsPh3)2

Article abstract of DOI:10.1016/S0020-1693(01)00404-2

The preparation of trans-Cr(CO)₄(AsPh₃)₂ is accomplished in reasonable yield by the direct photolysis of a solution of Cr(CO)₆ and an excess of AsPh₃ in heptane. The structures of cis-Mo(CO)₄(AsPh₃)₂ and cis-W(CO)₄(AsPh₃)₂ are reported. These are found to be isostructural and consist of distorted octahedral geometries with As-M-As angles of 102.56(2) and 102.62(2)° for M = Mo and W, respectively. The Mo-As distances of 2.6428(9) and 2.6489(9) A? are significantly longer than those in the W analogue at 2.6384(6) and 2.6361(6) A?. On the basis of a comparison to literature data, the AsPh₃ ligands in these molecules can be described as extremely compressed.

Full text of DOI:10.1016/S0020-1693(01)00404-2

www.elsevier.nl/locate/ica
Inorganica Chimica Acta 318 (2001) 77–83
Direct photolytic route to trans-Cr(CO)4(AsPh3)2 and crystal
structures of cis-Mo(CO)4(AsPh3)2 and cis-W(CO)4(AsPh3)2
Christina L. Bergstrom, Rudy L. Luck *
Department of Chemistry, Michigan Technological Uni6ersity, 1400 Townsend Dri6e, Houghton, MI 49931, USA
Received 13 November 2000; accepted 21 February 2001
Abstract
The preparation of trans-Cr(CO)4(AsPh3)2 is accomplished in reasonable yield by the direct photolysis of a solution of Cr(CO)6
and an excess of AsPh3 in heptane. The structures of cis-Mo(CO)4(AsPh3)2 and cis-W(CO)4(AsPh3)2 are reported. These are found
to be isostructural and consist of distorted octahedral geometries with AsꢀMꢀAs angles of 102.56(2) and 102.62(2)° for M=Mo
,
and W, respectively. The MoꢀAs distances of 2.6428(9) and 2.6489(9) A are significantly longer than those in the W analogue at
,
2.6384(6) and 2.6361(6) A. On the basis of a comparison to literature data, the AsPh3 ligands in these molecules can be described
as extremely compressed. © 2001 Elsevier Science B.V. All rights reserved.
Keywords: Crystal structure; Chromium complexes; Carbonyl complexes; Arsine complexes
they afford methods by which the electronic and steric
components of bonding can be determined [5].
1. Introduction
A widely used synthetic strategy to yield cis-
M(CO)4L2, M=Mo and W, species consists of the
prior preparation of a cis-M(CO)4(pip)2 derivative,
pip=piperidine, followed by the substitution accom-
plished by reaction with appropriate quantities of the
required ligand [6]. Curiously enough this method has
not so far been advocated for Cr(CO)6. One of the first
investigations into substitution reactions on Cr(CO)6
involved the direct thermal reactions between hexacar-
bonyl and ligand [7]. This afforded monosubstitued
compounds with ligands such as PPh3 and AsPh3.
Disubstitution was only reported with smaller ligands
such as P(OPh)3 affording Cr(CO)4(P(OPh)3)2. In order
to facilitate the substitution of CO on Group VI metal
carbonyls, Chatt et al. [8] employed sodium borohy-
dride as a catalyst in boiling ethanol. This afforded a
variety of di- and tri-substituted derivatives but only
Cr(CO)5(AsPh3). Therefore it is not surprising that the
thermal route to Cr(CO)4(AsPh3)2 was not successful
[4]. More recently, it is also noteworthy that Balbich
and dos Santos [9] provide IR spectroscopic evidence
for the one-pot synthesis of cis- and trans-
Cr(CO)4(AsPh3)2. The synthetic route consisted of the
reaction of Cr(CO)6 and AsPh3 in pentane under pho-
Over the last few decades many important contribu-
tions were made towards the understanding of substitu-
tion reactions and the stability of Group VI metal
carbonyls with a variety of ligands. Of particular note
were the assessments of steric contributions to the
structures of bis(phosphine) derivatives of molybdenum
tetracarbonyl [1–3]. These studies determined that there
was a direct correlation between the bulkiness of a
ligand and the extent of the distortion in the molecule.
It was also demonstrated that the trans isomer is ther-
modynamically more stable by 2–3 kcal mol−1 over the
cis isomer. An earlier study on complexes of the form
trans-Cr(CO)4L2 concluded that ligand dissociation was
related to the p-accepting character of L [4]. This report
also detailed difficulties in the synthesis of trans-
Cr(CO)4(AsPh3)2 via thermal routes. More recently, a
study on M(CO)5(EPh3) compounds, E=P, As or Sb,
revealed that these ligands are weaker p-bonding lig-
ands than the CO ligand [5]. Studies on metal carbonyl
systems continue to be of interest due to the fact that
* Corresponding author. Tel.: +1-906-487 2309; fax: +1-906-487
2061.
E-mail address: rluck@mtu.edu (R.L. Luck).
0020-1693/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved.
PII: S0020-1693(01)00404-2

78
C.L. Bergstrom, R.L. Luck / Inorganica Chimica Acta 318 (2001) 77–83
tolysis and in the presence of g-alumina for 60 min.
However, no details were given regarding the isolation
of pure isomer. Aroney et al. [10] also employed a
photolysis route to synthesize M(CO)5(AsPh3), M=Cr,
Mo or W. These reactions were accomplished by first
photolysing the metal carbonyls in THF and then
adding the desired ligand to substitute the ligand THF
in the compound M(CO)5(THF) formed by photolysis.
This contained lengthy purification procedures consist-
ing of sublimation to remove unreacted hexacarbonyl,
rinsings to remove free ligand and hexane elution on a
silica column to afford the monosubstituted derivatives.
It was also noted that elution with benzene afforded
cis-[M(CO)4(EPh3)2] in trace amounts.
It was with these facts in mind that we decided to
re-explore the chemistry of the Group VI metal car-
bonyls with the ligand triphenylarsine. Was there a
simple route to the disubstituted Cr(CO)4(AsPh3)2 com-
pounds? We were also interested in obtaining crystals
of the cis-M(CO)4(AsPh3)2 molecules, M=Cr, Mo and
W, as the results from single crystal X-ray determina-
tions would allow for assessments of the steric interac-
tions involved in accommodating these ligands in cis
geometries. This could easily be observed not only by a
consideration of the deviation from octahedral geome-
try of the ligands binding to the central metal atom but
also by distortions in the AsPh3 ligand itself.
1915, 1932, 1953, 1986 cm−1 were obtained. These are
probably indicative of cis- and trans-Cr(CO)4(pip)2 and
the monosubstituted Cr(CO)5(pip).
2.2.2. Photolysis route
Approximately 0.5 g of Cr(CO)6 and 3 mol equiva-
lents of triphenylarsine were suspended/dissolved in 450
ml heptane under dinitrogen and photolyzed for 3 h
with a 450 W UV immersion lamp (4.3¦ arc length) and
a 100–125 V, 60 Hz 7830 power supply by Ace Glass
Incorporated. Periodically, when the rate of reaction
slowed, as evident by a decrease in the rate of evolution
of CO which was being monitored, the system was
evacuated and photolysis resumed again under a dini-
trogen atmosphere. Roughly 77% of the theoretical
volume of CO (g) was collected over water for the
duration of photolysis. The compound in the form of a
bright yellow precipitate was filtered and dried under
vacuum. The mass of the solid product was 1.02 g, a
yield of 54%. The IR spectrum in the form of a Nujol
mull consisted of a very large stretch at 1875 cm−1
.
This is at a significantly lower frequency in the carbonyl
stretching region than that obtained for materials pro-
duced by reacting Cr(CO)6 with NHC5H10 at reflux
temperature in heptane (see above). Cr(CO)4(AsPh3)2
was recrystallized by dissolution in benzene and precip-
itation with hexane. Anal. Found: C, 61.28; H, 4.09.
Calc. for C40H45As2CrO4: C, 61.87; H, 3.89%. The low
percentage of carbon in the analysis may be at-
tributable to the presence of silicone grease, which was
2. Experimental
1
1
evident in the H NMR spectra. H NMR (C6H6-d6): l
6.93 (t, 2H, J=7.63 Hz, m-Ph), 7.05 (t, 1H, J=7.02
Hz, p-Ph), 7.41 (d, 1H, J=7.32 Hz, o-Ph), 7.73 (d, 1H,
J=7.63 Hz, o-Ph). IR (Nujol mull) 1875 (very large),
2.1. Materials
Heptane, piperidine and M(CO)6, M=Cr, Mo and
W, were used as purchased from commercial sources.
Dichloromethane was dried over calcium hydride and
distilled under dinitrogen. Benzene and hexane were
dried over sodium benzophenone ketyl and distilled
under dinitrogen. Methanol was dried over magnesium
methoxide. Reactions were carried out under inert at-
mospheres, either Ar or N2, using Schlenk techniques.
The IR spectra were recorded using a KVB/Analect IR
1933 (small), 2007 and 2059 (bumps) cm−1
.
2.3. Preparation of Mo(CO)4(AsPh3)2, 2, and
W(CO)4(AsPh3)2, 3
Mo(CO)4(NHC5H10)2 and W(CO)4(NHC5H10)2 were
synthesized under Ar by refluxing the appropriate metal
hexacarbonyl with an excess of piperidine in heptane
for approximately 4 h following the literature prepara-
tion [6]. The piperidine compounds were then refluxed
for varying lengths with 3 mol equivalents of AsPh3 in
dichloromethane under argon resulting in red-brown
solutions. Mo(CO)4(pip)2 was refluxed for between 15
and 20 min. The tungsten analog was allowed a longer
refluxing period for substitution — about 3 h. The IR
spectra of these two compounds agreed with what was
previously reported [8]. Crystallizing tubes were pre-
pared without isolating the compound from solution,
since excess triphenylarsine has been shown to stabilize
solutions of these species. The dichloromethane solu-
tions obtained from the reaction mixtures were layered
with equal volumes of dried methanol and the tubes
1
instrument. The H NMR spectrum was recorded on a
Varian XL400 spectrometer in benzene-d6.
2.2. Preparation of trans-Cr(CO)4(AsPh3)2, 1
2.2.1. Attempts with piperidine
Our initial attempts to produce a disubstituted
Cr(CO)4(AsPh3)2 compound involve the attempted syn-
thesis of Cr(CO)4(pip)2, pip=piperidine, using the well
known procedure for Mo and W analogues [1]. These
reactions were not successful in yielding only the cis-
Cr(CO)4(pip)2 complex as evident in the solution IR
(CH2Cl2) of the products in the w(CO) region. Peaks at

C.L. Bergstrom, R.L. Luck / Inorganica Chimica Acta 318 (2001) 77–83
79
Table 1
were kept in a cupboard. Suitable crystals were
obtained in 1 to 2 weeks.
Crystal data and structure refinement details for 2 and 3
2
3
2.4. X-ray diffraction studies
Empirical formula
Formula weight
Temperature (K)
Habit
Crystal system
Space group
C
820.42
291
yellow prism
triclinic
40H30As2Mo1O4 C40H30As2O4W1
908.33
291
yellow prism
triclinic
In both cases, crystals were removed from the mother
liquor to a microscope slide where they were stabilized
by immersion in a mixture of the mother liquor and
mineral oil. The crystals were then mounted using
epoxy onto the head of a thin glass fiber, which was
anchored in a goniometer mounting pin. The pin-
mounted crystal was then inserted into the goniometer
head of the X-ray diffractometer and centered in the
beam path. Standard CAD4 centering, indexing, and
data collection programs were utilized [11]. Twenty-five
reflections between 10 and 15° in q were located by a
random search pattern, centered, and used in indexing.
Final cell constants and orientation matrix were ob-
tained by collecting appropriate preliminary data and
refined by a least-squares fit. The scan width for each
reflection and the scan rate were dependent on the
intensity of the observed diffraction signal. During data
collection, three intensity standards and three orienta-
tion standards were measured at regular intervals to
measure the rate of decay of the crystal and to accom-
modate for crystal movement.
Data were first reduced and corrected for absorption
using psi-scans [12] and then solved using the program
SIR-97 [13], which afforded a complete solution for the
non-H atoms in both cases. These programs were uti-
lized using the WinGX interface [14]. The models were
then refined using SHELXL-97, [15] first with isotropic
and then anisotropic thermal parameters to conver-
gence. The positions and isotropic thermal parameters
of H-atoms were constrained and set to 1.5 times the
isotropic equivalent of the atoms they were attached to,
respectively. This constituted the final model for both
compounds and appropriate crystallographic data is
listed in Table 1. Selected bond distances and angles are
given in Table 2 for 2 and Table 3 for 3.
Unit cell dimensions
,
a (A)
9.716(3)
11.670(2)
16.997(3)
100.63(1)
98.04(2)
109.79(2)
1739.0(6)
0.71073
2
1.57
820
graphite
Enraf-Nonius
(EN) CAD
9.689(2)
11.653(2)
16.998(3)
100.68(1)
97.93(1)
109.77(1)
1732.4(5)
0.71073
2
,
b (A)
,
c (A)
h (°)
i (°)
k (°)
3
,
V (A )
u (Mo Ka radiation)
Z
Dx (g cm−3
F(000)
)
1.74
884
graphite
Monochromator
Diffractometer
EN CAD
4
4
Diffraction geometry
Type of scan
k-geometry
diffractometer
ꢀ–2q scans
2.31
k
ꢀ–2q scans
5.27
0.30×0.20×0.05
v (Mo Ka) (mm−1
)
Crystal size (mm)
0.35×0.30×0.28
Cell dimensions refinement
Number of reflections
2q Range (°)
25
25
20.42–29.80
20.42–29.80
Semi-empirical absorbance correction a
Number of „-scans
Number of reflections
Max./min. transmission
coefficient
qmax (°)
Index range
h
k
296
8
0.565, 0.499
259
7
0.779, 0.301
24.97
24.97
0 to 11
0 to 11
−13 to 13
−20 to 19
6120
−13 to 13
−20 to 19
6098
l
Number of reflections
measured
Number of observed data, 5314
5204
ꢀIꢀ\2|(I)
Number of parameters
424
424
Rint
0.013
0.026
0.071 d
1.03
0.016
0.023
0.054 e
1.03
b
R(F)
3. Discussion
wR(F2) c
Goodness-of fit
3.1. Synthesis of Compound 1
Max. (D/|)
0.001
0.001
−3
,
Residual DQ (e A
)
Max.
Min.
0.72
−0.57
0.50
−0.42
The widely used route [1] to cis bis-substituted
M(CO)4L2, M=Mo or W, compounds, consisting of
the prior synthesis of M(CO)4(pip)2 is not useful if
extended to Cr(CO)6. The problem is that the reaction
yields cis- and trans-Cr(CO)4(pip)2 and also the mono-
substituted Cr(CO)5(pip). Clearly, unless attempts are
made to separate these compounds, further substitution
reactions would only afford mixtures. However,
difficulties were already noted in using an alumina
column to separate Cr(CO)4L2 derivatives [4] produced
Scattering factors from International Tables for Crystallography
(Vol. C)
a North [12].
b R=ꢁ(Fo−Fc)/ꢁ(Fo).
c Rw=[ꢁ[w(Fo 2−Fc 2)2]/ꢁ[w(Fo 2)2]]1/2
.
d w=1/[|2(Fo 2)+(0.0434×P)2+0.5955P], where P=(Max(Fo2,
0)+2×Fc2)/3.
e w=1/[|2(Fo 2)+(0.0246×P)2+0.4445P], where P=(Max(Fo2,
0)+2×Fc2)/3.

80
C.L. Bergstrom, R.L. Luck / Inorganica Chimica Acta 318 (2001) 77–83
Table 2
in reactions consisting of heating up Cr(CO)6 and
AsPh3 in refluxing diglyme for 5 h.
Selected bond distances and angles for 2
The synthesis of trans-Cr(CO)4(AsPh3)2, 1, can be
accomplished by the direct photolysis of Cr(CO)6 and
AsPh3 in heptane. A reasonable elemental analysis re-
sult allows us to conclude that a material having the
stoichiometry of 1 was produced. Evidence for the trans
disposition of the arsine ligands is readily apparent in
the simplicity of the IR spectra of the product (as a
Nujol mull) in the CO stretching region, which con-
sisted of a large absorption at 1875 cm−1. Clearly the
assignment at 1910 cm−1 for what was believed to be
trans-Cr(CO)4(AsPh3)2 should now not be considered
as accurate [4]. The cis isomer would have resulted in a
more complex IR pattern as is evident in a comparison
of cis and trans-Mo(CO)4(PPh3)2 [1]. It is noteworthy
that this photolysis method does not yield the cis
isomer. Given the fact that this compound is not stable
in solution for prolong periods, it is likely that any cis
material produced may have rapidly isomerize to the
trans form. Our attempts to produce crystals of this
compound were not successful.
Our method of direct synthesis by photolysis stands
in contrast to the photolysis of Cr(CO)6 with AsPh3 in
the presence of g-alumina [9] which yielded mixtures of
cis and trans-Cr(CO)4(AsPh3)2. This difference may be
attributable to the fact that a 125 W light source was
used for these experiments [9] whereas we used a (brand
new) 450 W bulb. Presumably the greater energy in our
system was sufficient to accomplish a trans- product as
substitution reactions with trans-Mo(CO)4L2 were
shown to be more facile than with cis-Mo(CO)4L2 [2].
It remains to be seen if our conditions for the synthesis
of 1 can be applied to other Cr(CO)4L2 derivatives,
including those with small enough tertiary phosphine
and arsine ligands which may more readily afford cis
geometries.
,
Bond distances (A)
Mo1ꢀAs1
Mo1ꢀAs2
Mo1ꢀC1
Mo1ꢀC2
Mo1ꢀC3
Mo1ꢀC4
As1ꢀC111
As1ꢀC121
2.6490(9)
2.6428(9)
2.058(3)
1.976(3)
2.017(3)
1.973(4)
1.956(3)
1.949(3)
As1ꢀC131
As2ꢀC211
As2ꢀC221
As2ꢀC231
O1ꢀC1
O2ꢀC2
O3ꢀC3
1.960(3)
1.947(3)
1.955(3)
1.950(3)
1.136(4)
1.152(4)
1.145(4)
1.152(5)
O4ꢀC4
Bond angles (°)
As1ꢀMo1ꢀAs2
As1ꢀMo1ꢀC1
As1ꢀMo1ꢀC2
As1ꢀMo1ꢀC3
As1ꢀMo1ꢀC4
As2ꢀMo1ꢀC1
As2ꢀMo1ꢀC2
As2ꢀMo1ꢀC3
As2ꢀMo1ꢀC4
C1ꢀMo1ꢀC2
C1ꢀMo1ꢀC3
C1ꢀMo1ꢀC4
C2ꢀMo1ꢀC3
C2ꢀMo1ꢀC4
C3ꢀMo1ꢀC4
102.56(2)
90.47(9)
174.61(9)
84.48(10)
92.32(11)
93.39(9)
81.04(10)
89.44(10)
165.01(11)
93.34(12)
174.64(13)
88.18(14)
91.59(13)
83.99(14)
90.26(14)
C121ꢀAs1ꢀC131 100.98(12)
Mo1ꢀAs2ꢀC211 126.03(10)
Mo1ꢀAs2ꢀC221 116.02(8)
Mo1ꢀAs2ꢀC231 110.46(8)
C211ꢀAs2ꢀC221
C211ꢀAs2ꢀC231 101.95(12)
C221ꢀAs2ꢀC231 100.63(11)
Mo1ꢀC1ꢀO1
Mo1ꢀC2ꢀO2
Mo1ꢀC3ꢀO3
Mo1ꢀC4ꢀO4
98.13(13)
177.2(3)
177.3(3)
177.1(3)
172.4(3)
Mo1ꢀAs1ꢀC111 118.05(10)
Mo1ꢀAs1ꢀC121 120.01(9)
Mo1ꢀAs1ꢀC131 115.06(9)
C111ꢀAs1ꢀC121 101.46(13)
C111ꢀAs1ꢀC131
97.64(12)
Table 3
Selected bond distances and angles for 3
,
Bond distances (A)
W1ꢀAs1
W1ꢀAs2
W1ꢀC1
W1ꢀC2
W1ꢀC3
W1ꢀC4
As1ꢀC111
As1ꢀC121
2.6361(6)
2.6284(6)
2.046(4)
1.987(4)
2.006(5)
1.965(5)
1.957(4)
1.948(4)
As1ꢀC131
As2ꢀC211
As2ꢀC221
As2ꢀC231
O1ꢀC1
O2ꢀC2
O3ꢀC3
1.961(4)
1.944(4)
1.949(3)
1.951(4)
1.139(5)
1.140(5)
1.151(6)
1.157(7)
4. Structural results
O4ꢀC4
Bond angles (°)
As1ꢀW1ꢀAs2
As1ꢀW1ꢀC1
As1ꢀW1ꢀC2
As1ꢀW1ꢀC3
As1ꢀW1ꢀC4
As2ꢀW1ꢀC1
As2ꢀW1ꢀC2
As2ꢀW1ꢀC3
As2ꢀW1ꢀC4
C1ꢀW1ꢀC2
C1ꢀW1ꢀC3
C1ꢀW1ꢀC4
C2ꢀW1ꢀC3
C2ꢀW1ꢀC4
C3ꢀW1ꢀC4
W1ꢀAs1ꢀC111
Representations of 2 and 3 are illustrated in Figs. 1
and 2, respectively. An inspection of the packing dia-
gram of the unit cells of these molecules does not
indicate any intermolecular interactions that would re-
sult in distortions in molecular geometry. It is interest-
ing that the two MoꢀAs distances at 2.6490(9) and
102.62(2)
90.30(11)
174.64(12)
84.22(13)
92.37(14)
93.35(11)
80.92(12)
89.09(13)
164.92(14)
93.54(15)
174.37(17)
88.16(19)
91.86(16)
84.01(19)
90.80(19)
118.10(13)
W1ꢀAs1ꢀC121
W1ꢀAs1ꢀC131
C111ꢀAs1ꢀC121
C111ꢀAs1ꢀC131
C121ꢀAs1ꢀC131
W1ꢀAs2ꢀC211
W1ꢀAs2ꢀC221
W1ꢀAs2ꢀC231
C211ꢀAs2ꢀC221
C211ꢀAs2ꢀC231
C221ꢀAs2ꢀC231
W1ꢀC1ꢀO1
119.73(11)
114.90(13)
101.75(19)
97.44(17)
101.36(17)
125.66(10)
115.99(11)
110.71(11)
98.08(16)
102.30(15)
100.53(14)
177.8(3)
,
2.6428(9) A are significantly longer than the WꢀAs ones
,
which are at 2.6361(6) and 2.6284(6) A. A small signifi-
cant difference was observed for the metal to arsenic
atom distances in
Mo(CO)5(AsPh3) (dMoꢀAs=2.612(0) A) and W(CO)5-
(AsPh3) (dMoꢀAs=2.617(1) A) though the distances for
the PPh3 analogue contained the same trend noted
a
crystallographic study of
,
W1ꢀC2ꢀO2
W1ꢀC3ꢀO3
W1ꢀC4ꢀO4
178.0(4)
176.2(4)
173.5(4)
,
,
above with dMꢀP=2.560(1) A for Mo(CO)5(PPh3) and
,
2.545(1) A for W(CO)5(PPh3) [5]. This difference was

C.L. Bergstrom, R.L. Luck / Inorganica Chimica Acta 318 (2001) 77–83
81
previously attributed [5] to ‘variations in metal–phos-
phorus overlap’ and, in our cases, ‘variations in metal–
arsenic overlap’ may well be applicable to the data
observed for 2 and 3.
The distances from the metal center to the carbon
atom in the trans-carbon monoxide ligands are signifi-
cantly longer than those to the cis-CO ligands, clearly
reflective of the fact that the cis-ligands are involved in
more backbonding, see Tables 2 and 3. The CꢀO
,
distances range from 1.136(4) to 1.152(4) A for 2 and
,
1.139(5) to 1.157(7) A for 3 and are thus not signifi-
cantly different.
As expected, both molecular arrangements are iso-
morphous and they contain distorted octahedral ge-
ometries reflective of the reduction of steric strain
induced by the cis-AsPh3 ligands. The AsꢀMꢀAs angles
of 102.56(2) and 102.62(2)° for 2 and 3, respectively, are
good indicators of this distortion as these ligands situ-
ate away from one another to relieve steric interaction.
An even larger angle of 104.62(7)° was observed for the
PꢀMoꢀP angle in cis-Mo(CO)4(PPh3)2 [3]. It is possible
that in 2 and 3, the larger AsꢀM bond distances as
Fig. 1. An ORTEP3 for Windows [27] representation of 2. H-atoms are
represented by circles of arbitrary radii and ellipsoids are drawn at
the 50% probability level.
,
compared to dMoꢀP of 2.576(2)
A
in cis-
Mo(CO)4(PPh3)2 [3] accounts for the smaller angles
observed. These two ligands, PPh3 and AsPh3, have
similar Tolman cone angles of 145 [16] and 147° [17],
respectively.
The large AsꢀMꢀAs angles result in CeqꢀMꢀCeq an-
gles of 83.99(14) and 84.01(19)° for 2 and 3, respec-
tively. This equatorial distortion is not equivalent in
that the angles As1ꢀMꢀC4 of 92.32(11) and 92.37(14)°
are significantly different to those of As2ꢀMꢀC2 of
81.04(10 and 80.92(12)° for 2 and 3, respectively. It is
noteworthy that the sum of the four (equatorial) angles
is very close to 360° (359.91° for 2 and 359.92° for 3)
which means that these atoms are in planar arrange-
ments. Another consequence of the steric crowding is
the fact that the MꢀC4ꢀO4 angle is 172.4(3) and
173.5(4)° for 2 and 3, respectively. This bend in the
MꢀCꢀO angle can be attributed to a close approach to
the carbonyl group by a H atom (H135) on a phenyl
group. In fact, H135 is located at 2.820(4) and 2.630(4)
Fig. 2. An ORTEP3 for Windows [29] representation of 3. H-atoms are
represented by circles of arbitrary radii and ellipsoids are drawn at
the 50% probability level.
,
A from the O4 and C4 atoms in 2 and 2.87(4) and
,
2.67(5) A from the O4 and C4 atoms in 3.
Comparative evidence of the steric strain in the ter-
tiary arsine ligands themselves is presented in Fig. 3.
Essentially, the longer the AsꢀC bond length, the
smaller the angle that the ligand should be able to
accommodate. The data in Table 4, compiled from
crystallographic data on various compounds containing
AsPh3 fragments, clearly indicate that complexes 2 and
3 contain among the longest dAsꢀC bond distances at
,
1.95 A and the smallest CipsoꢀAsꢀCipso (CꢀAsꢀC) bond
angles at 100°. Only the related M(CO)5(AsPh3),
M=Cr, Mo or W, (labeled as p, q and r in Fig. 3)
contain similar average distances and slightly larger
Fig. 3. Graph of average CꢀAsꢀC angle versus average AsꢀC distance
using data in Table 4.

82
C.L. Bergstrom, R.L. Luck / Inorganica Chimica Acta 318 (2001) 77–83
Table 4
Average AsꢀC bond distances and average CꢀAsꢀC angles for various compounds containing the AsPh3 ligand
,
Compound
Average AsꢀC
Average
d(AsꢀX) (A)
Reference
Illustration number in
Fig. 3
a
CꢀAsꢀC (°) a
,
distance (A)
Br(AsPh3)Au
1.94(7)
1.904(5)
1.895(1)
1.927(2)
1.94(1)
1.912(2)
1.923(3)
1.892(1)
1.87
1.98
1.912
1.933
1.947(1)
104(3)
2.342(5)
1.652(4)
1.666(5)
2.412(2)
2.331(1)
2.490(1)
2.502
[19]
[20]
[20]
[21]
[22]
[23]
[24]
[24]
[25]
[25]
[25]
[23]
[23]
[23]
[26]
[27]
a
b
c
d
e
f
g
h
i
j
k
l
m
n
o
p
q
r
b
(OAsPh3)4[Ni(OH2)6]Br2·2H2O
108.4(5)
107.8(4)
105(1)
(OAsPh3)4[Ni(OH2)6]Br2·2H2O b
c
Ph3AsAuGeCl3
Ph3AsAuCl
Ph3AsGaI3
105(1)
105.49(8)
103.3(9)
105.21(3)
111.84
106.15
103.81
[WS4Ag3I(As*Ph3)3](SAsPh3) c
[WS4Ag3I(AsPh3)3](SAs*Ph3) c
[As*Ph3(m2-I3)(AsPh3)] [I3Co(AsPh3)] d
[AsPh3(m2-I3)(As*Ph3)] [I3Co(AsPh3)] d
[AsPh3(m2-I3)(AsPh3)] [I3Co(As*Ph3)] d
[(As*Ph3)InI3(AsPh3)][InI3(AsPh3)]·H2O
[(AsPh3)InI3(As*Ph3)][InI3(AsPh3)]·H2O
2.011
2.554(3)
2.508(5)
2.519(8)
2.728(13)
2.990(2)
2.926(2)
2.485(1)
2.4972(5)
2.612(0)
2.617(1)
2.6490(9)
2.6428(9)
2.6361(6)
2.6284(6)
105.15
102.62(3)
103.17(3)
109(1)
101.4(9)
102(1)
101(1)
100(2)
100(2)
100(2)
[(AsPh3)InI3(AsPh3)] [InI3(As*Ph3)]·H2O d 1.951(1)
[Ph3AsI][GaI4]
Cr(CO)5(AsPh3)
Mo(CO)5(AsPh3)
W(CO)5(AsPh3)
2 (As1)
2 (As2)
3 (As1)
3 (As2)
1.915(4)
1.947(0)
1.947(6)
1.950(4)
1.955(6)
1.951(4)
1.955(7)
1.948(4)
[5]
[5]
this work
this work
this work
this work
s
t
u
v
100(2)
a Numbers in parentheses represent the standard deviation in the data listed for the three distances and angles.
b Two distinct Ph3AsO entities constitute the asymmetric unit in the green isomer reported for this compound.
c Data obtained from the Cambridge Data Base [28].
d Atom(s) considered were located on threefold axes. Hence, there were no differences in the bond angles or lengths.
the Bi to M bond than the P atom in the P to M bond.
angles, clearly related to the fact that there is less steric
hindrance than in the cases of 2 and 3.
An examination of the data in Fig. 3 would indicate
that the AsꢀL interaction also plays a role in determin-
ing the nature of the hybridization at the As atom. If
there is a strong interaction, such as in the case of the
As=O bond in the Ph3As=O molecule (as represented
by data points b and c), then larger CꢀAsꢀC angles
result. If the interaction is longer, such as in the case of
2 and 3 then smaller CꢀAsꢀC angles pertain. This
suggests that the arsenic atom is capable of accommo-
dating a variety of bond angles and distances and, as
such, may be capable of varying degrees of s and p
orbital hybridization. This conclusion is relevant as a
recent theoretical study on BF3 concluded that resis-
tance of the molecule to rehybridization may better
account for the lack of Lewis acidity as opposed to the
FpꢀBp p-bond being responsible [18]. Clearly, the data
depicted in Fig. 3 suggest that AsPh3 is flexible, does
not resist rehybridization and thus can adopt a range of
geometries depending on what the As atom is coordi-
nated to. Therefore, using a ‘cone angle’ assessed using
one criteria (such as a fixed AsꢀM bond length of 2.28
It is curious that data points i and j in Fig. 3 which
contain the smallest and largest AsꢀC distances has a
large CꢀAsꢀC angle for i but does not have a corre-
sponding small CꢀAsꢀC angle for j. It should be noted
that these AsꢀC distances were not determined accu-
rately (i.e. determined to two decimal places 1.87(3) and
,
1.98(2) A) and thus this may be the source of this
discrepancy. Further these distances were determined
from AsPh3 ligands on the [Ph3AsꢀIꢀIꢀIꢀAsPh3]− an-
ion which contained unequal AsꢀI distances of 2.554(3)
,
and 2.508(5) A and point j in Fig. 3 was from the
AsPh3 ligand which had the shorter AsꢀI distance. This
may affect the hybridization on the As atom leading to
larger CꢀAsꢀC angles.
The amount of s-orbital contribution to the ground-
state description of the MꢀE bond has been stated to
increase from E=P down to E=Bi [27]. In fact mea-
surements of CꢀEꢀC bond angles in complexes
Cr(CO)5(EPh3), E=P, As, Sb or Bi, revealed a slight
decrease from P to Bi (102.6(8), 101.4(8), 100(2), 99(2)°,
respectively) [27]. These data were used to suggest that
the Bi atom contained more s-orbital contribution in
,
A [16]) to rationalize complexation of AsPh3, is clearly
not constructive when applied to different AsꢀM bond
lengths.

C.L. Bergstrom, R.L. Luck / Inorganica Chimica Acta 318 (2001) 77–83
83
[9] I.M. Balbich, J.H.Z. dos Santos, Can. J. Appl. Spectrosc. 39
(1994) 18.
5. Supplementary material
[10] M.J. Aroney, I.E. Buys, M.S. Davies, T.W. Hambley, J. Chem.
Soc., Dalton Trans. (1994) 2827.
[11] Enraf-Nonius, CAD4 Express Software, Enraf-Nonius, Delft, The
Netherlands, 1994.
[12] A.C.T. North, D.C. Phillips, F.S. Mathews, Acta Crystallogr.,
Sect. A. 24 (1968) 351.
[13] A. Altomare, M.C. Burla, M. Camalli, G.L. Cascarano, C.
Giacovazzo, A. Guagliardi, A.G.G. Moliterni, G. Polidori, R.
Spagna, J. Appl. Crystallogr. 32 (1999) 115.
[14] L.J. Farrugia, J. Appl. Crystallogr. 32 (1999) 837.
[15] G.M. Sheldrick, SHELXL97, Program for crystal structure refine-
ment, University of Go¨ttingen, Germany, 1997.
[16] C.A. Tolman, Chem. Rev. 77 (1977) 313.
[17] C.A. McAuliffe, in: G. Wilkinson, R.D. Gillard, J.A. McClev-
erty (Eds.), Comprehensive Coordination Chemistry, vol. 2,
Pergamon Press, New York, 1987, p. 989.
Crystallographic data for the structural analysis has
been deposited with the Cambridge Crystallographic
Data Centre, CCDC No. 150411 for 2 and CCDC No.
150412 for 3. Copies of this information may be ob-
tained free of charge from The Director, CCDC, 12
Union Road, Cambridge, CB2 1EZ, UK (fax: +44-
1223-336033; e-mail: deposit@ccdc.cam.ac.uk or www:
http://www.ccdc.cam.ac.uk). Structure factor tables are
available by E-mail from author RLL.
Acknowledgements
[18] B.D. Rowsell, R.J. Gillespie, G.L. Heard, Inorg. Chem. 38
(1999) 4659.
[19] F.W.B. Einstein, R. Restivo, Acta Crystallogr., Sect. B 31 (1975)
624.
[20] G. Oliva, E.E. Castellano, J. Zukerman-Schpector, A.C. Mass-
abni, Inorg. Chim. Acta 89 (1984) 9.
R.L.L. thanks Michigan Technological University
(MTU) for support. C.L.B. thanks the Dow Chemical
Company for a summer internship at MTU.
[21] U.M. Tripathi, G.L. Wegner, A. Schier, A. Jockisch, H. Schmid-
baur, Z. Naturforsch., Teil B 53 (1998) 939.
[22] B. Weissbart, L.J. Larson, M.M. Olmstead, C.P. Nash, D.S.
Tinti, Inorg. Chem. 34 (1995) 393.
References
[1] D.J. Darensbourg, R.L. Kump, Inorg. Chem. 17 (1978) 2680.
[2] D.J. Darensbourg, A.H. Graves, Inorg. Chem. 18 (1979) 1257.
[3] F.A. Cotton, D.J. Darensbourg, S. Klein, B.W.S. Kolthammer,
Inorg. Chem. 21 (1982) 294.
[23] L.-J. Baker, L.A. Kloo, C.E.F. Rickard, M.J. Taylor, J.
Organomet. Chem. 545 (1997) 249.
[24] G. Sakane, T. Shibahara, H. Hou, Y. Lin, X. Xin, Transition
Met. Chem. 21 (1996) 398.
[4] M.J. Wovkulich, S.J. Feinberg, J.D. Atwood, Inorg. Chem. 19
(1980) 2608.
[25] S.M. Godfrey, H.P. Lane, A.G. Mackie, C.A. McAuliffe, R.G.
Pritchard, Chem. Commun. (1993) 1190.
[5] M.J. Aroney, I.E. Buys, M.S. Davies, T.W. Hambley, J. Chem.
Soc., Dalton Trans. (1994) 2827.
[26] L.-J. Baker, C.E.F. Rickard, M.J. Taylor, J. Organomet. Chem.
C4 (1994) 464.
[6] D.J. Darensbourg, R.L. Kump, Inorg. Chem. 17 (1978) 2680.
[7] C.N. Matthews, T.A. Mager, J.H. Wotiz, J. Am. Chem. Soc. 81
(1959) 2273.
[8] J. Chatt, G.J. Leigh, N. Thankarajan, J. Organomet. Chem. 29
(1971) 105.
[27] A.J. Carty, N.J. Taylor, A.W. Coleman, M.F. Lappert, J. Chem.
Soc., Chem. Commun. (1979) 639.
[28] F.H. Allen, O. Kennard, R. Taylor, Acc. Chem. Res. 16 (1983)
146.
[29] L.J. Farrugia, J. Appl. Crystallogr. 30 (1997) 565.
.