Synthesis of rhenium-platinum complexes bearing carbaborane ligands and their unusual dynamic behavior

Article abstract of DOI:10.1039/b002043p

Treatment of Cs[3,3,3-(CO)₃-closo-3,1,2-ReC₂B₉H ₁₁]1 with [Pt(NCMe)₂{Ph₂P(CH₂)₂PPh ₂}][BF₄]₂ yielded [3,3,3-(CO)₃,3,8-(Pt{Ph₂P(CH₂) ₂PPh₂})-8-(μ-H)-closo-3,1,2-ReC₂B ₉H₁₀][BF₄] 2b as the sole product. The cation of salt 2b is a bimetallic species with a Re-Pt bond bridged by a nido-7,8-C₂B₉H₁₁ ligand pentahapto coordinated to the rhenium and with the β-B-H bond in the ligating CCBBB ring of the carbaborane cage forming a 3-centre 2-electron B-HPt bond. Deprotonation of 2b with strong bases gives an isomeric mixture of the complexes [3,3,3-(CO)₃-3,n-(Pt{Ph₂P(CH₂)2PPh ₂})-closo-3,1,2-ReC₂B₉H₁₀] 3a (n = 4), 3b (n = 8). The isomers are readily separated and the proportion of 3a to 3b formed is solvent- as well as base-dependent. In THF solutions of 2b afford a 1:3:1 equilibrium mixture of 2b, 3a, and 3b. Refluxing THF solutions of 1 with [Pt(H)(THF)(PEt₃)₂[BF₄] for several days yields a separable mixture of the isomers [3,3,3-(CO)₃-3,n)-{Pt(PEt₃) ₂}-closo-3,1,2-ReC₂B₉H₁₀] 4a (n = 4), 4b (n = 8). The same reactants, when stirred together in CH₂Cl₂ at ambient temperatures, give 4a and [3,3,3-(CO)3-3,8-{Pt(H)(PEt₃)}-8-(μ-H)-closo-3,1,2-ReC ₂B₉H₁₀] 5. Complex 4a, not 5, is the precursor to 4b, the interconversion occurring at elevated temperatures in solution by an intramolecular exchange of B-Pt a bonds between α-B and β-B vertices in the coordinating CCBBB face of the cage. Treatment of 1 with [Pt(Me)(THF)-(PMe₂Ph)₂][BF₄] at room temperature initially gives [3,3,3-(CO)₃-3,4-{Pt(PMe₂Ph) ₂}-closo-3,1,2-ReC₂B₉H₁₀] 6a. However, 6a isomerizes in CH₂Cl₂ solutions at ambient temperatures to give a 1:1 mixture of 6a and [3,3,3-(CO)₃-3,8-{Pt(PMe₂Ph) ₂}-closo-3,1,2-ReC₂B₉H₁₀] 6b. The NMR data (¹H, ¹³C, ¹¹B and ³¹P) for the new complexes are discussed. The Royal Society of Chemistry 2000.

Full text of DOI:10.1039/b002043p

View Article Online / Journal Homepage / Table of Contents for this issue
Synthesis of rhenium–platinum complexes bearing carbaborane
ligands and their unusual dynamic behavior†
Dianne D. Ellis, Paul A. Jelliss and F. Gordon A. Stone*
Department of Chemistry & Biochemistry, Baylor University, Waco, TX 76798-7348, USA
Received 14th March 2000, Accepted 25th April 2000
Published on the Web 6th June 2000
Treatment of Cs[3,3,3-(CO) -closo-3,1,2-ReC B H ] 1 with [Pt(NCMe) {Ph P(CH ) PPh }][BF ] yielded
3
2
9
11
2
2
2
2
2
4 2
[
3,3,3-(CO) -3,8-(Pt{Ph P(CH ) PPh })-8-(µ-H)-closo-3,1,2-ReC B H ][BF ] 2b as the sole product. The cation
3 2 2 2 2 2 9 10 4
of salt 2b is a bimetallic species with a Re–Pt bond bridged by a nido-7,8-C B H ligand pentahapto coordinated
2
9
11
to the rhenium and with the β-B–H bond in the ligating CCBBB ring of the carbaborane cage forming a 3-centre
-electron B–H Pt bond. Deprotonation of 2b with strong bases gives an isomeric mixture of the complexes [3,3,3-
CO) -3,n-(Pt{Ph P(CH ) PPh })-closo-3,1,2-ReC B H ] 3a (n = 4), 3b (n = 8). The isomers are readily separated
2
(
3
2
2
2
2
2
9
10
and the proportion of 3a to 3b formed is solvent- as well as base-dependent. In THF solutions of 2b afford a
1
:3:1 equilibrium mixture of 2b, 3a, and 3b. Refluxing THF solutions of 1 with [Pt(H)(THF)(PEt ) ][BF ] for
3
2
4
several days yields a separable mixture of the isomers [3,3,3-(CO) -3,n-{Pt(PEt ) }-closo-3,1,2-ReC B H ] 4a
3
3
2
2
9
10
(
n = 4), 4b (n = 8). The same reactants, when stirred together in CH Cl at ambient temperatures, give 4a and
2 2
[
3,3,3-(CO) -3,8-{Pt(H)(PEt )}-8-(µ-H)-closo-3,1,2-ReC B H ] 5. Complex 4a, not 5, is the precursor to 4b, the
3 3 2 9 10
interconversion occurring at elevated temperatures in solution by an intramolecular exchange of B–Pt σ bonds
between α-B and β-B vertices in the coordinating CCBBB face of the cage. Treatment of 1 with [Pt(Me)(THF)-
(
PMe Ph) ][BF ] at room temperature initially gives [3,3,3-(CO) -3,4-{Pt(PMe Ph) }-closo-3,1,2-ReC B H ] 6a.
2 2 4 3 2 2 2 9 10
However, 6a isomerizes in CH Cl solutions at ambient temperatures to give a 1:1 mixture of 6a and [3,3,3-(CO) -
2
2
3
1
13
11
31
3
,8-{Pt(PMe Ph) }-closo-3,1,2-ReC B H ] 6b. The NMR data ( H, C, B and P) for the new complexes are
2 2 2 9 10
discussed.
In recent years many bimetal compounds containing rhenium
1
and platinum have been reported including the first species
2
incorporating carbaborane ligands. In the metallacarbaborane
complexes, prepared using [N(PPh ) ] [2,2,2-(CO) -closo-2,1-
3
2 2
3
ReCB H ] as the precursor, the Re–Pt bonds are bridged
10
11
by nido-7-CB H fragments. The cage system is pentahapto
10
11
coordinated to the rhenium to give a closo-2,1-ReCB H
10
11
framework and a platinum–phosphine fragment is exopoly-
hedrally attached by a Re–Pt bond and an agostic B–H Pt
bridge. We have recently carried out studies with the more
familiar nido-7,8-C B H group, employing the complex salt
Results and discussion
2
9
11
3
The salt [Pt(NCMe) {Ph P(CH ) PPh }][BF ] , formed in situ
2 2 2 2 2 4 2
Cs[3,3,3-(CO) -closo-3,1,2-ReC B H ] 1 in reactions with
3
2
9
11
by treating [PtCl {Ph P(CH ) PPh }] with 2 mol equivalents of
2
2
2
2
2
mid- to late-transition metal fragments, including iron,
4
Ag[BF
to afford in high yield a complex of composition [(CO)3-
Pt(PEt ) }-closo-ReC B H ] 2. As discussed below, detailed
4
] in MeCN, was treated with salt 1. Reaction occurred
ruthenium, rhodium, copper and silver. The Group 1B metal
reagents gave species having Re–Cu and Re–Ag bonds in their
solid state structures, while the remainder gave a novel class of
exo-closo complexes with no direct Re–M bonds. We have now
further developed this work, describing herein the syntheses
from 1 of rhenium–platinum complexes and demonstrating the
fragility of the Re–Pt bonds formed in some of the products.
This lability leads to a remarkable intramolecular exchange of
exo-B–Pt σ bonds between boron vertices in the coordinating
face of the cage ligand.
{
3
2
2
9
10
examination of the NMR spectra and a single crystal X-ray
crystallographic study revealed that this product was [3,3,3-
(
CO) -3,8-(Pt{Ph P(CH ) PPh })-8-(µ-H)-closo-3,1,2-ReC B -
3 2 2 2 2 2 9
H ][BF ] 2b and contained a bridging carbaborane ligand.
10
4
There was no evidence for formation of an isomer [3,3,3-(CO)3-
,4-(Pt{Ph P(CH ) PPh })-4-(µ-H)-closo-3,1,2-ReC B H ]-
3
2
2
2
2
2
9
10
[
BF ] 2a.
4
†
The compounds described in this paper have rhenium atoms
incorporated into closo-1,2-dicarba-3-rhenadodecaborane structures
and most have exo-PtL2 groups. For complexes with 3,4-PtL2 frag-
ments, it should be noted that such species are racemic and the 3,7-PtL2
notation is equally valid. The former is used in the formulae according
to IUPAC convention.
Electronic supplementary information (ESI) available: fully coupled
1
and selectively decoupled H NMR spectra of complex 5. See http://
www.rsc.org/suppdata/dt/b0/b002043p/
DOI: 10.1039/b002043p
J. Chem. Soc., Dalton Trans., 2000, 2113–2122
This journal is © The Royal Society of Chemistry 2000
2113

View Article Online
Table 1 Analytical and physical data
c
Analysis (%)
Yield
(%)
a
b
Ϫ1
Compound
ν
˜
(CO) (cm
)
C
H
max
2
3
3
4
4
5
6
b [3,3,3-(CO) -3,8-(Pt{Ph P(CH ) PPh })-8-(µ-H)-closo-3,1,2-ReC B H ][BF ]
86
50
25
2053vs, 1993s, 1968s
2022vs, 1952s, 1933s
2024vs, 1955s, 1929s
2020vs, 1949s, 1921s
2018vs, 1945s, 1927s
2034vs, 1962s, 1950s
2021vs, 1950s, 1920s
34.4 (35.1)
36.9 (37.4)
37.4 (37.4)
24.8 (24.5)
24.5 (24.5)
18.7 (18.4)
28.0 (28.5)
3.3 (3.3)
3.5 (3.4)
3.5 (3.4)
4.8 (4.8)
4.8 (4.8)
3.8 (3.8)
3.6 (3.5)
3
2
2
2
2
2
9
10
4
a [3,3,3-(CO) -3,4-(Pt{Ph P(CH ) PPh })-closo-3,1,2-ReC B H ]
3
2
2
2
2
2
9
10
b [3,3,3-(CO) -3,8-(Pt{Ph P(CH ) PPh })-closo-3,1,2-ReC B H ]
3
2
2
2
2
2
9
10
d
a [3,3,3-(CO) -3,4-{Pt(PEt ) }-closo-3,1,2-ReC B H ]
21
3
3
2
2
9
10
b [3,3,3-(CO) -3,8-{Pt(PEt ) }-closo-3,1,2-ReC B H ]
14
42
12
3
3
2
2
9
10
[3,3,3-(CO) -3,8-{Pt(H)(PEt )}-8-(µ-H)-closo-3,1,2-ReC B H ]
3
3
2
9
10
e
f
[3,3,3-(CO) -3,n-{Pt(PMe Ph) }-closo-3,1,2-ReC B H ] (n = 4 or 8)
3
2
2
2
9
10
a
b
Ϫ1
All compounds are yellow. Measured in CH Cl ; medium-intensity broad bands observed at ca. 2550 cm in the spectra of all the compounds are
2
2
c
d
due to B–H absorptions. Calculated values are given in parentheses. Yield reported for preparation in CH Cl at 25 ЊC. Yield when prepared in
refluxing THF is 7%. Compounds 6a and 6b cannot be isolated separately (see text). Therefore the overall yield, IR spectrum and microanalytical
data for the 1:1 mixture are reported. Contains 1.0 mol equivalent CH Cl , confirmed by NMR.
2
2
e
f
2
2
a
Table 2 Hydrogen-1 and carbon-13 NMR data
1
b
13
c
Compound
H
C
2
b
7.48–7.81 (m, 20 H, Ph), 3.63 [br s, 2 H, cage CH, J(PtH) = 38],
190.0 (CO × 1), 186.6 [CO × 2, J(PtC) = 47], 134.6–124.9
(Ph), 38.1 (cage CH), 31.1 [dd, CH2, J(PC) = 44, 8,
2
.98 [ddt, 2 H, CH , J(PH) = 32, 7, J(HH) = 7], 2.61 [ddt, 2 H,
2
CH , J(PH) = 29, 8, J(HH) = 8], Ϫ4.86 [br m, 1 H, B–H Pt,
J(PtC) = 74], 23.6 [dd, CH , J(PC) = 40, 6, J(PtC) = 59]
2
2
J(PtH) ≈ 550]
3
a
7.72–7.37 (m, 20 H, Ph), 3.79, 3.36 (br s × 2, 2 H, cage CH),
192.7, 192.6, 192.5 (CO × 3), 133.9–128.8 (Ph), 44.4, 36.2
(cage CH × 2), 31.3 [dd, CH , J(PC) = 19, 19], 26.9 [dd, CH ,
2
.35–2.11 (br m, 4 H, CH × 2)
2
2
2
J(PC) = 9, 24]
3
b
a
7.75–7.37 (m, 20 H, Ph), 3.37 (br s, 2 H, cage CH), 2.34–2.11
br m, 4 H, CH × 2)
192.1 (br, CO), 134.2–128.6 (Ph), 34.4 (cage CH), 31.0 [dd,
CH , J(PC) = 20, 38], 27.1 [dd, CH , J(PC) = 8, 31]
(
2
2
2
d
e
4
3.87, 3.28 (br s × 2, 2 H, cage CH), 1.92 (br m, 12 H, CH ),
196.9, 192.7, 191.7 (CO × 3), 44.5, 34.2 (cage CH × 2), 17.4
[d, CH , J(PC) = 32], 17.3 [d, CH , J(PC) = 20], 8.4, 8.2
2
1
.01 [dt, 9 H, Me, J(PH) = 8, J(HH) = 7], 0.89 [dt, 9 H, Me,
2
2
J(PH) = 11, J(HH) = 7]
3.31 (br s, 2 H, cage CH), 1.90, 1.72 (br m × 2, 12 H, CH ),
(Me × 2)
4
b
197.2 (CO), 191.0 (CO × 2), 33.3 (cage CH), 17.6–17.1 (br m,
2
0
.98 [dt, 9 H, Me, J(PH) = 7, J(HH) = 6], 0.84 [dt, 9 H, Me,
CH × 2), 8.3, 7.9 (Me × 2)
2
J(PH) = 9, J(HH) = 6]
3.42 (br s, 2 H, cage CH), 1.97 [dq, 6 H, CH , J(PH) = 11,
5
194.2 (CO), 186.6 (CO × 2), 35.4 (cage CH), 22.5 [d, CH2,
J(PC) = 39, J(PtC) = 70], 9.3 [d, Me, J(PC) = 2, J(PtC) = 39]
2
J(HH) = 7, J(PtH) = 50], 1.09 [dt, 9 H, Me, J(PH) = 19,
J(HH) = 7], Ϫ4.84 [dq, 1 H, B–H Pt, J(HH) = 15, J(BH) =
6
0, J(PtH) ≈ 465], Ϫ11.03 [br s, 1 H, PtH, J(PtH) = 1090]
f
f
6
a
b
7.43–6.95 (m, 20 H, Ph), 4.05, 3.36 (br s × 2, 2 H, cage CH),
197.5, 197.1, 193.4 (CO × 3), 131.5–128.5 (Ph), 45.9 [cage
CH, J(PtC) = 45], 41.4 (cage CH), 18.7, 16.9 [d × 2, PMe,
J(PC) = 39, 25]
192.2 (CO), 191.8 (CO × 2), 131.0–128.9 (Ph), 34.5 (cage
CH), 17.8 [dd, PMe, J(PC) = 40, 4], 16.7 [d, PMe, J(PC) = 24]
1
.74 [dd, 6 H, PMe, J(PH) = 8, 3], 1.65 [dd, 6 H, PMe,
J(PH) = 12, 11]
6
7.43–7.21 (m, 20 H, Ph), 3.41 (br s, 2 H, cage CH), 1.77–1.55
(
br m, 12 H, PMe)
a
195
Chemical shifts (δ) in ppm, coupling constants (J) in Hz, measurements at room temperature in CD Cl , unless otherwise stated. Often Pt
2
2
b
satellites could not be observed due to poor compound solubility. They are noted where unambiguously assignable. Resonances for terminal BH
protons occur as broad unresolved signals in the range δ ca. Ϫ2 to 3. Resonances for cage CH and B–H Pt protons are described as broad,
c
corresponding to approximate ν values of 10 and 300 Hz, respectively. Hydrogen-1 decoupled, chemical shifts are positive to high frequency of
1
/2
d
e
f
SiMe4. Measured at Ϫ73 ЊC. Measured at Ϫ90 ЊC. Measured at Ϫ80 ЊC.
molecule is bisected by a mirror plane of symmetry, reflected
in the observation of one signal both in the H NMR spectrum
Complex 2b constitutes a rare example of a cationic metalla-
carbaborane without a charge-compensating group bound to
the polyhedral carbaborane framework. Much earlier we
reported the syntheses of the bimetallic cationic complexes [1,2-
Me -3,3-(CO) -3-(L)-3,4-(µ-H) -3,4-{Pt(PEt ) }-8-(CH C H -
1
for the cage CH groups [δ 3.63, J(PtH) = 38 Hz] and also in the
1
3
1
C-{ H} NMR spectrum (δ 38.1). The molecular C symmetry
s
11
1
is supported by the 1:1:3:2:2 pattern seen in the B-{ H}
2
2
2
3
2
2
6
4
11
t
NMR spectrum (Table 3). The peak at δ 22.2 is split in the B
195
Me-4)-closo-3,1,2-WC B H ][BF ] (L = CO, CNBu or PMe ) in
2
9
7
4
3
NMR spectrum into a doublet with Pt satellites [J(HB) = 52,
which a carbaborane ligand spans a W–Pt bond by virtue of a
5,7
5
J(PtB) = 140 Hz] as is typical for a B–H Pt moiety. The cis-
B–H Pt bridge. Other groups have synthesized a handful of
positional rigidity of the Ph P(CH ) PPh ligand about the
monometallic cationic metallacarbaboranes with and without
2
1
2
2
2
3
1
6
platinum is revealed in the P-{ H} NMR spectrum by the
195
charge-compensating groups bound to a carbaborane vertex.
appearance of two signals (δ 62.3 and 55.9) with Pt satellite
couplings of similar magnitude to each other [J(PtP) = 2955
and 3393 Hz, respectively]. The broadness of the latter reson-
ance implies that the phosphorus nucleus from which it derives
lies transoid with respect to the boron atom of the B–H Pt
group. All the NMR data strongly support the structure
The IR spectrum of 2b (Table 1) displays νmax(CO) bands at
Ϫ1
2
053, 1993 and 1968 cm , indicative of adduct formation
Ϫ1 3,4
between the monoanion of 1 [νmax(CO) 2005 and 1905 cm ]
and the dicationic fragment [Pt{Ph P(CH ) PPh }] . The H
2ϩ
1
2
2
2
2
NMR spectrum (Table 2) revealed the presence of a 3-centre
-electron B–H Pt agostic bond by displaying a broad reson-
2
11
1
ance at δ Ϫ4.86, the signal showing no resolvable B– H coup-
ling but clearly straddled by broad Pt satellites [J(PtH) ≈ 550
depicted, where a β-B atom in the CCBBB face coordinat-
ing the rhenium is participating in the agostic B–H Pt
group. To confirm the structure of the cation, especially the
presence of the Re–Pt bond, single crystals of 2b were
grown and an X-ray diffraction study was carried out. The
195
Hz]. The presence of the Pt-chelating Ph P(CH ) PPh ligand is
2
2
2
2
13
1
confirmed in this and the C-{ H} NMR spectrum with peaks
for the phenyl groups as well as the CH CH moiety. The
2
2
2
114
J. Chem. Soc., Dalton Trans., 2000, 2113–2122

View Article Online
a
Table 3 Boron-11 and phosphorus-31 NMR data
11
b
31
c
Compound
B
P
2
3
3
4
4
5
6
b
a
b
a
b
22.2 [1 B, B–H Pt, J(HB) = 52, J(PtB) = 140], Ϫ0.5 (1 B), Ϫ1.2
1 B, BF ), Ϫ7.7 (3 B), Ϫ13.1 (2 B), Ϫ15.5 (2 B)
37.7 [1 B, B–Pt, J(PtB) ≈ 485], 0.5 (1 B), Ϫ0.9 (1 B), Ϫ3.2 (1 B),
Ϫ7.2 (1 B), Ϫ16.5 (2 B), Ϫ19.2 (2 B)
62.3 [J(PtP) = 2955], 55.9 [br, J(PtP) = 3393]
(
4
64.8 [br, J(PtP) = 2060], 48.0 [d, J(PP) = 6, J(PtP) = 4709]
65.4 [br, J(PtP) = 2050], 49.0 [d, J(PP) = 17, J(PtP) = 5158]
49.6 [1 B, B–Pt, J(PtB) ≈ 440], Ϫ4.1 (1 B), Ϫ8.2 (2 B), Ϫ10.7
(
2 B), Ϫ16.3 (2 B), Ϫ23.3 (1 B)
d
34.5 [d, 1 B, B–Pt, J(PB) = 76, J(PtB) ≈ 580], 1.1 (1 B), Ϫ1.3
1 B), Ϫ2.9 (1 B), Ϫ6.9 (1 B), Ϫ16.3 (2 B), Ϫ19.8 (2 B)
45.6 [1 B, B–Pt, J(PtB) ≈ 490], Ϫ3.9 (1 B), Ϫ8.0 (2 B), Ϫ10.4
2 B), Ϫ16.6 (2 B), Ϫ23.5 (1 B)
18.1 [1 B, B–H Pt, J(HB) = 60], Ϫ1.9 (1 B), Ϫ8.8 (2 B), Ϫ12.5
2 B), Ϫ18.3 (2 B), Ϫ19.9 (1 B)
22.6 [J(PtP) = 4652], 22.3 [br, J(PtP) = 2093]
(
e
25.4 [br, J(PtP) = 2141], 24.3 [J(PtP) = 5060]
(
39.1 [J(PtP) = 3798]
(
f
a
b
35.9 [d, 1 B, B–Pt, J(PB) = 65, J(PtB) ≈ 530], 1.3 (1 B), Ϫ1.0
8.00 [br, J(PtP) = 2148], Ϫ7.4 [J(PtP) = 4724]
(
(
1 B), Ϫ2.8 (1 B), Ϫ6.7 (1 B), Ϫ16.1 (2 B), Ϫ19.3 (1 B), Ϫ20.0
1 B)
f
6
47.3 [d, 1 B, B–Pt, J(PB) = 76, J(PtB) ≈ 485], Ϫ3.6 (1 B), Ϫ7.9
11.1 [br, J(PtP) = 2193], Ϫ5.4 [J(PtP) = 5121]
(
2 B), Ϫ10.3 (2 B), Ϫ16.4 (2 B), Ϫ23.3 (1 B)
a
b
Chemical shifts (δ) in ppm, coupling constants (J) in Hz, measurements at room temperature in CD Cl , unless otherwise stated. Hydrogen-1
2
2
decoupled, chemical shifts are positive to high frequency of BF ؒOEt (external). The B–H Pt and B–Pt assignments were made from fully coupled
3
2
1
1
c
B spectra. Hydrogen-1 decoupled, chemical shifts are positive to high frequency of 85% H PO (external). Peaks denoted as broad have ν = 200–
50 Hz. Measured at Ϫ73 ЊC. Measured at Ϫ90 ЊC. Measured at Ϫ80 ЊC.
3
4
1/2
d
e
f
2
Table 4 Selected internuclear distances (Å) and angles (Њ) for the cation of [3,3,3-(CO) -3,8-(Pt{Ph P(CH ) PPh })-8-(µ-H)-closo-3,1,2-
3
2
2
2
2
ReC B H ][BF ] 2b
2
9
10
4
Pt–P(1)
Re–C(4)
Re–B(5)
P(1)–C(6)
C(5)–O(5)
2.252(4)
1.56(3)
2.33(2)
1.83(2)
1.11(2)
Pt–P(2)
2.268(4)
1.92(3)
2.335(13)
1.832(14)
1.50(2)
Pt–B(4)
2.34(2)
1.96(2)
2.34(2)
1.16(3)
Pt–Re
Re–B(4)
Re–C(1)
2.8126(10)
2.29(2)
2.35(2)
Re–C(3)
Re–C(2)
P(2)–C(7)
C(6)–C(7)
Re–C(5)
Re–B(3)
C(3)–O(3)
C(4)–O(4)
1.41(3)
P(1)–Pt–P(2)
P(1)–Pt–Re
C(4)–Re–C(3)
C(4)–Re–Pt
B(4)–Re–Pt
84.6(2)
166.95(11)
106.7(10)
77.7(7)
B(3)–Re–Pt
77.4(4)
176(2)
115.2(6)
108.42(12)
81.3(9)
C(3)–Re–Pt
B(5)–Re–Pt
C(1)–Re–Pt
O(5)–C(5)–Re
P(2)–Pt–B(4)
74.8(5)
86.7(4)
123.3(5)
176(2)
B(4)–Pt–Re
C(3)–Re–C(5)
C(5)–Re–Pt
C(2)–Re–Pt
O(3)–C(3)–Re
51.9(5)
82.8(7)
143.1(5)
118.1(4)
179(2)
O(4)–C(4)–Re
P(1)–Pt–B(4)
P(2)–Pt–Re
53.4(4)
C(4)–Re–C(5)
159.8(5)
Re(CO) unit from a tripodal piano stool arrangement found in
3
the anion of 1 is perceptible with C(3)–Re–C(4) 106.7(10),
C(3)–Re–C(5) 82.8(7) and C(4)–Re–C(5) 81.3(9)Њ, i.e. the first
exceeds the last two by more than 20Њ. Comparison should be
made with the complexes [3,3,3-(CO) -3,8-{M(PPh )}-8-(µ-H)-
3
3
closo-3,1,2-ReC B H ] (M = Cu or Ag) where the OC–Re–CO
2
9
10
angles bisected by the ReMP plane are no greater than 10Њ more
obtuse than the remaining two OC–Re–CO angles of each of
4
those complexes. In both the copper and silver complexes only
one phenyl group of the M(PPh ) unit is disposed towards the
3
Re(CO) group. The likely cause of this enhanced perturbation
3
in 2b is the steric pressure from both the lower phenyl groups of
the Ph P(CH ) PPh ligand (Fig. 1) which are forced to flank
2
2
2
2
the Re–Pt bond, driving the C(3)O(3) and C(4)O(4) ligands
further apart and therefore closer to the C(5)O(5) group.
When complex 2b is dissolved in MeCN it immediately
affords [Pt(NCMe) {Ph P(CH ) PPh }][3,3,3-(CO) -closo-3,1,
2
2
2
2
2
3
Fig. 1 Structure of the cation of [3,3,3-(CO) -3,8-(Pt{Ph P(CH ) -
3
2
2 2
2
-ReC B H ][BF ], hence the importance of removing all
PPh })-8-(µ-H)-closo-3,1,2-ReC B H ][BF ] 2b, showing the crystallo-
2 9 11 4
2
2
9
10
4
graphic labeling scheme. Thermal ellipsoids are shown at the 40%
probability level. Only the agostic hydrogen atom is shown for clarity.
traces of MeCN when isolating 2b, usually by washing with
copious quantities of Et O. The behaviour of complex 2b in
2
THF (tetrahydrofuran), however, is at variance with its simple
dissociation in MeCN. This phenomenon will be analysed later
pending discussion of the results of treating 2b with various
strong bases.
cation is shown in Fig. 1 and salient bond lengths and angles
are given in Table 4.
The Re–Pt bond [2.8126(10) Å] is slightly longer than that
found in the neutral complex [2,2,2-(CO) -2,7-{Pt(PPh ) }-7-
Solutions of complex 2b in CH Cl or THF were treated with
3
3
2
2
2
2
(
µ-H)-closo-2,1-ReCB H ] [2.7931(4) Å], suggesting that the
various strong bases under the conditions listed in Table 5. It
should be noted that the sterically encumbered reagents Proton
Sponge [1,8-bis(dimethylamino)naphthalene] and DBU (1,8-di-
azabicyclo[5.4.0]undec-7-ene) instigated complete dissociation
10
10
molecular charge has little influence on this connectivity. The
presence of the β-B–H Pt linkage is substantiated [B(4)–Pt
2
.34(2) Å] with the agostic hydrogen atom located. The corre-
sponding distance observed in [2,2,2-(CO) -2,7-{Pt(PPh ) }-7-
of the platinum fragment from the anion [3,3,3-(CO) -closo-
3
3
2
3
–
(
µ-H)-closo-2,1-ReCB H ] is 2.355(3) Å. The distortion of the
3,1,2-ReC B H ] , as did solutions of Na[BH ] (entries 1–3).
10
10
2
9
11
4
J. Chem. Soc., Dalton Trans., 2000, 2113–2122
2115

View Article Online
Table 5 Reaction of the salt 2b with base under varying conditions
a
Entry
Base
Solvent
CH Cl
T/ЊC
Ratio 3a:3b
1
2
3
4
5
6
7
8
Proton Sponge
DBU
Na[BH4]
K-Selectride
K-Selectride
LiPh
Ϫ80
dissoc.
dissoc.
dissoc.
2:1
1:1
2:3
2
2
2
2
2
CH Cl
Ϫ80
Ϫ80
Ϫ80
Ϫ80
Ϫ80
Ϫ80
25
2
CH Cl
2
CH Cl
2
THF
THF
THF
THF
t
LiBu
LiBu
1:3
1:4
t
a
dissoc. denotes complete dissociation of the platinum fragment from
the rhenacarborane fragment.
Clearly the bridged Re–Pt system of 2b is sensitive to even
mildly nucleophilic species. However, careful treatment with
s
K-Selectride K[BHBu ] in CH Cl at Ϫ80 ЊC (Table 5, entry 4)
3
2
2
Fig.
2
Structure of [3,3,3-(CO) -3,4-(Pt{Ph P(CH ) PPh })-closo-
3 2 2 2 2
gave a mixture of two isomeric complexes [3,3,3-(CO) -3,n-
3
3
,1,2-ReC B H ] 3a, showing the crystallographic labeling scheme.
2
9
10
(
Pt{Ph P(CH ) PPh })-closo-3,1,2-ReC B H ] 3a (n = 4), 3b
2 2 2 2 2 9 10
Thermal ellipsoids are shown at the 40% probability level. Hydrogen
atoms are omitted for clarity.
(
n = 8) which were separable by column chromatography. The
9
Pt–P bond distances. To test this notion, single crystals of 3a
were grown for the purpose of carrying out an X-ray diffraction
study. Selected bond lengths and angles are given in Table 6 and
the structure is shown in Fig. 2.
The Re(CO)3 unit is less distorted from
a pseudo-
symmetrical piano stool arrangement [C(3)–Re–C(4) 93.5(7),
C(3)–Re–C(5) 84.5(8), C(4)–Re–C(5) 85.7(7)Њ] than in complex
2
b, most likely because the Pt{Ph P(CH ) PPh } fragment has
2 2 2 2
been lifted away from the Re(CO) moiety upon completion of
3
species 3a and 3b shared almost identical IR spectra (Table 1).
Microanalyses confirmed that these were isomers of the same
basic molecular formula, although complex 3b was decidedly
the conversion from the B–H Pt system to the B–Pt σ bond.
This process involves an effective contraction of the B–Pt dis-
tance from 2.34(2) Å [Pt–B(4) in 2b] to 2.061(18) Å [Pt–B(3) in
3a]. The disparity between Pt–P bond lengths in 3a is evident
with the Pt–P(1) bond [2.202(4) Å] transoid to the Re atom
shorter than that in 2b [2.252(4) Å] and the Pt–P(2) bond
1
13
1
more soluble than 3a in CH Cl solutions. The H and C-{ H}
2
2
NMR spectra displayed all the expected signals for the cage,
CO and Ph P(CH ) PPh ligands for 3a and 3b with the excep-
2
2
2
2
tion of the CO signals for 3b which appear as a single broad
[
2.314(4) Å] transoid to the α-B atom longer than in 2b [2.268(4)
11
1
unresolved peak. The B-{ H} NMR spectra are particularly
informative with singlet resonances at δ 37.7 (3a) and 49.6 (3b),
lying isolated in the low field region away from the remaining
Å]. While neither structure is of exemplary quality, making it
necessary to treat the data with a little caution, the comparisons
would seem to support the NMR evidence discussed above.
Thus the σ-bonded carbaborane cage might be considered to
exert a considerably stronger trans influence on a Pt–P bond
than a B–H moiety, which would be expected when it is
recalled that σ effects are most significant when evaluating trans
195
peaks. Additionally, these signals, which both bear broad Pt
satellites [J(PtB) ≈ 485 (3a) and 440 Hz (3b)], remain as singlets
11
in the fully coupled B NMR spectrum. This is indicative of the
7
a–c,8
presence of exopolyhedral B–Pt σ bonds in both 3a and 3b.
11
1
The B-{ H} NMR spectrum of 3a reveals a 1:1:1:1:1:2:2
pattern of peaks synonymous with the absence of any formal
molecular symmetry. The corresponding pattern for 3b is
9
influences. The shorter Pt–P(1) bond in complex 3a as com-
pared with that in 2b may be accounted for by the decrease in
positive charge on the platinum, leading to increased π back
donation to both phosphine ligands. This effect is more than
offset for the Pt–P(2) bond in 3a by the trans influence exerted
by the B–Pt σ bond. It should be noted that similar features for
1
:1:2:2:2:1 suggesting the presence of a molecular plane of
symmetry through the Re, Pt, the apical B and both P atoms.
This information enabled 3a to be assigned as the α isomer,
195
31
where an α-B atom in the CCBBB ring participates in σ bond-
the Pt– P NMR coupling constants and the Pt–PEt bond
3
31
ing to the platinum, and likewise 3b as the β isomer. The P-
lengths were reported for the complex [1,2-Me -3-(µ-H)-3,3-
2
1
{
H} NMR spectra of 3a and 3b display interesting features.
Signals at δ 64.8 (3a) and 65.4 (3b) are broad (ν1/2 ≈ 220 and
50 Hz, respectively) most likely due to the transoid location of
these P atoms with respect to the σ-bonded quadrupolar boron
(CO) -3-(PMe )-3,4-{Pt(PEt ) }-8-(CH C H Me-4)-closo-3,1,2-
2
3
3
2
2
6
4
WC B H ], which has a B–Pt bond trans to one phosphine and
2
9
7
2
cis to another, though these properties were not previously
5
discussed.
2,5,7,8
195
31
atoms.
Their Pt– P coupling constants [J(PtP) = 2060
One curious aspect of the reaction of complex 2b with
(
3a) and 2050 Hz (3b)] are diminished by ca. 1300 Hz in
K-Selectride in CH Cl at Ϫ80 ЊC is how the majority of the
2
2
comparison with the corresponding signal for 2b [δ 55.9,
J(PtP) = 3393 Hz]. Furthermore, the P nuclei transoid to the
Re–Pt bonds resonate at δ 48.0 (3a) and 49.0 (3b) with J(PtP)
values which are ca. 1750 and 2200 Hz, respectively, greater
than that in complex 2b [δ 62.3, J(PtP) = 2955 Hz] (Table 3).
The implication is that upon deprotonation of the agostic
B–H Pt bridge with formation of a B–Pt bond the B-transoid
phosphorus atom of the phosphine arm becomes more weakly
bound, while the Re-transoid P atom (cisoid to B) becomes
more strongly bonded. This inference is based on the premise
product formed could be the α isomer 3a (by a ratio of 2:1 over
3b) when the starting material exists solely as the β isomer. The
product distribution would seem to rule out a simple acid–base
exchange where the Selectride anion abstracts a proton from the
B–H Pt group. It is more likely that the Selectride delivers
a hydride anion to the platinum center, where most of the
cationic charge of 2b logically resides (Scheme 1). This might be
accompanied by the displacement of one of the phosphine lig-
and P atoms, giving a neutral intermediate A with a terminal
ꢀ
hydride. Support for such a structure will be presented later
when compound 5 is discussed. It is at this point that isomeriz-
195
31
that Pt– P one-bond coupling constants are sensitive to the
2
116 J. Chem. Soc., Dalton Trans., 2000, 2113–2122

View Article Online
Table 6 Selected internuclear distances (Å) and angles (Њ) for [3,3,3-(CO) -3,4-(Pt{Ph P(CH ) PPh })-closo-3,1,2-ReC B H ] 3a
3
2
2
2
2
2
9
10
Pt–B(3)
2.061(18)
1.89(2)
2.296(14)
1.15(2)
Pt–P(1)
2.202(4)
1.93(2)
2.307(14)
1.18(2)
Pt–P(2)
2.314(4)
1.95(2)
2.373(14)
1.18(2)
Pt–Re
Re–B(3)
Re–B(4)
2.7583(9)
2.28(2)
2.39(2)
Re–C(4)
Re–C(1)
C(3)–O(3)
Re–C(5)
Re–C(2)
C(4)–O(4)
Re–C(3)
Re–B(5)
C(5)–O(5)
B(3)–Pt–P(1)
B(3)–Pt–Re
C(4)–Re–C(5)
C(4)–Re–Pt
B(3)–Re–Pt
97.0(4)
54.2(4)
85.7(7)
81.6(5)
47.1(4)
B(5)–Re–Pt
119.3(4)
176(2)
173.4(4)
150.75(11)
93.5(7)
C(5)–Re–Pt
C(1)–Re–Pt
B(4)–Re–Pt
O(5)–C(5)–Re
P(1)–Pt–P(2)
147.9(6)
117.8(3)
78.2(4)
177(2)
P(2)–Pt–Re
C(5)–Re–C(3)
C(3)–Re–Pt
C(2)–Re–Pt
O(3)–C(3)–Re
122.29(10)
84.5(8)
67.2(5)
79.4(3)
174.6(14)
O(4)–C(4)–Re
B(3)–Pt–P(2)
P(1)–Pt–Re
C(4)–Re–C(3)
86.9(2)
Scheme 1 Possible pathways in the formation of complexes 3a and 3b by addition of K-Selectride to 2b in CH Cl at low temperature.
2
2
1
95
ation to A can occur, since there clearly is no such rearrange-
peaks were too weak to be able to observe Pt satellites. When
the sample is warmed these signals are seen to broaden and
disappear as conversion into 2b is completed. We attribute these
ꢁ
ment in the starting material 2b. This would involve a trans-
formation from the β-B–H Pt to the α-B–H Pt system,
which has previously been reported for several related mole-
signals to the cationic species C (Scheme 2) where the Pt atoms
ꢁ
2,10
cules having B–H M groups, and it is here that A would be
ꢁ
favoured over A by a ratio of 2:1. A dual pathway may then
ꢀ
swing into operation with both molecules A and A undergoing
ꢁ
ꢀ
oxidative addition reactions of their B–H Pt groups to give
B–Pt–H systems. The platinum() centers so generated in the
anti
syn
intermediates Bꢁ (or Bꢁ ) and B can be stabilized by recoord-
ꢀ
ination of the pendant phosphine arm in a trans position with
anti
respect to the newly generated hydride ligand. The isomers B
ꢁ
syn
and Bꢁ arise from the fact that there would be two enantio-
meric forms of A and the second hydride in B could lie anti or
ꢁ
ꢁ
syn to the cage carbon atoms. Now the hydrogen atoms of the
PtH group are suitably located in a cisoid fashion to undergo
2
reductive elimination processes to give the observed products
3
a and 3b. Such oxidative addition–reductive elimination
processes involving exopolyhedral platinum fragments on
metallacarbaborane cages are well documented.
5,7a–c,8
Both 3a and 3b can be protonated with HBF ؒOEt2 to
4
regenerate complex 2b quantitatively. Reverse α → β isomeriz-
ation must therefore occur in the conversion of 3a into 2b. In an
attempt to make a closer examination of the protonation of 3a,
a sample was dissolved in CD Cl in an NMR tube and cooled
Scheme 2 Protonation of complex 3a to reform salt 2b.
2
2
to Ϫ80 ЊC in the spectrometer probe, then removed briefly to
bear terminal hydride ligands cisoid to both the phosphorus
nuclei. These structures can be appreciated by considering
1
treat it with 1 mol equivalent of HBF ؒOEt . A H NMR spec-
4
2
trum run at Ϫ80 ЊC within a few seconds of reintroducing the
sample displayed a weak apparent triplet at δ Ϫ16.4 [J(PH) =
that the platinum dz
2
orbital in 3a is the most conceivable site
ϩ
for attack by H . Rehybridization of the platinum center upon
increase in formal oxidation number (Pt → Pt ) could
II
IV
7
and 7 Hz]. A further weaker broad resonance was observed at
δ Ϫ19.2 and the ratio of these signals was 3:1 respectively. The
be accompanied by coordination and stabilization by Et O,
2
J. Chem. Soc., Dalton Trans., 2000, 2113–2122
2117

View Article Online
conveniently delivered with the acid reagent and necessarily
bonded trans to the hydride ligand. Depending on which side
of the molecule the acid attacked, two isomers are possible
Reactions between complex 1 and other platinum–phosphine
combinations [Pt(NCMe) (PR ) ][BF ] [PR = PEt , (PR ) =
2
3
2
4 2
3
3
3 2
(C H ) P(CH ) P(C H ) ] proved too difficult to work up to
6
11
2
2
2
6
11 2
(
syn and anti), though identification of the major species is,
give pure products. To circumvent this the salt 1 was treated
syn
anti
however, indeterminable. In both Cꢁ and Cꢁ the hydride is
in position (cisoid to the α-boron) to undergo formation of an
α-B–H Pt group by reductive elimination and concomitant
loss of Et O to give 2a. Instantaneous rearrangement to give
the β-B–H Pt group would alternatively yield 2b.
with [Pt(H)(THF)(PEt ) ][BF ], readily generated in situ from
3
2
4
[PtCl(H)(PEt ) ] and Ag[BF ] in THF. This procedure has the
3
2
4
advantage of pre-equipping the platinum fragment with a
hydrogen atom, so that the product might be expected to be
both neutral and lead to the elimination of molecular hydrogen.
When the reaction was carried out in refluxing THF for 4 d
(Scheme 4) two neutral products were indeed isolated after
2
The behavior of salt 2b in THF was alluded to above and
revealed to be unlike that in MeCN where complete dissoci-
ation occurred. In fact in d -THF the complex is present in its
8
cationic form at a level of only ca. 20%. The remainder was
identified as a mixture of the complexes 3a and 3b in a ratio of
1
:3. Thus THF itself is capable of acting as a deprotonating
agent setting up an equilibrium mixture of 2b, 3a and 3b. To
1
support this a H NMR spectrum of a d -THF sample showed
8
a broad peak at δ 10.88 which corresponds with that for solu-
ϩ
tions of protonated THF molecules, [C H OH] , as shown
4
8
by treating a sample of d -THF with a small amount of
8
HBF ؒOEt . The mechanism of the THF deprotonation is most
4
2
likely similar to that described in Scheme 1, except that cationic
intermediates are involved where THF is bound to the Pt atoms
(
Scheme 3). The first of these to form could be D , akin to A
ꢀ
ꢀ
Scheme 4 Reactions of salt 1 with [Pt(H)(THF)(PEt ) ][BF ] under
3
2
4
differing conditions.
chromatographic purification, [3,3,3-(CO) -3,n-{Pt(PEt ) }-
3
3 2
closo-3,1,2-ReC B H ] 4a (n = 4), 4b (n = 8), in a ratio of
2
9
10
1
:2, respectively. The yields (Table 1) of both were poor
and it should be noted that the synthesis of 4a and 4b was
accompanied by the formation of significant amounts of
ϩ
1
the cation [Pt(H)(PEt ) ] . This was identified from the H
3
3
31
1
and P-{ H} NMR spectra of the reaction mixture but not
isolated.
The IR spectra (Table 1) of both complexes 4a and 4b are
very similar to those of 3a and 3b, respectively, immediately
11
1
suggesting analogous structures. The B-{ H} NMR spectrum
Table 3) of 4a contains a diagnostic doublet resonance with
(
195
broad Pt satellites at δ 34.5 [J(PB) = 76, J(PtB) ≈ 580 Hz] to
be compared with the corresponding signal in the same spec-
trum of 3a [δ 37.7, J(PtB) ≈ 485 Hz]. The signal for 4a remains
11
unchanged in a fully coupled B NMR spectrum, confirming
11
1
the presence of a B–Pt linkage. Likewise the B-{ H} NMR
spectrum of 4b displays a peak at δ 45.6 [J(PtB) ≈ 490 Hz] to be
1
compared with that of 3b [δ 49.6, J(PtB) ≈ 440 Hz]. The H and
C-{ H} NMR spectra were also supportive of the proposed
structures, though it was necessary to carry out data accumu-
13
1
Scheme 3 THF as a base in the formation of complexes 3a and 3b
from 2b.
lations at low temperatures (–73 ЊC for 4a and Ϫ90 ЊC for 4b).
13
1
(
and to complex 5 discussed presently). An exchange might
The C-{ H} NMR spectra showed three CO resonances for 4a
[δ 196.9, 192.7, 191.7 (1:1:1)] reflecting the lack of symmetry in
the molecule, and two for 4b [δ 197.2, 191.0 (1:2)] as a result of
then occur between D and D , though under these conditions
ꢀ
ꢁ
the former β isomer is favored, accounting for the higher per-
centage of 3b in the mixture. The intermediates E as shown bear
resemblance to B (Scheme 1) and also to C (Scheme 2), noting
that the THF molecules in E would be situated trans to the cage
boron atom. The D /D equilibrium would go some way to
explaining why most of the deprotonations with strong bases in
THF (Table 5, entries 5–8) result in a product distribution
favouring 3b over 3a. Most of the base (R ) merely abstracts H
from the protonated THF molecule, though some will still react
with the small amount of 2b present in THF solution. Hence
the outcome of the reactions in Table 5 (entries 5–8) is dictated
as much by the solvent used as the strong base added.
the molecular C symmetry, a plane encompassing the Re, Pt,
s
31
1
apical B and P atoms. The peaks in the P-{ H} NMR spectra
of 4a and 4b measured at room temperature were perceptibly
broader than their counterparts in the spectra of 3a and 3b.
Running the spectral measurements at Ϫ70 and Ϫ80 ЊC,
respectively, led to more discernible signals (Table 3) which
were now similar to those of 3a and 3b at room temperature,
ꢁ
ꢀ
–
ϩ
195
31
especially in terms of the wide disparity between the Pt– P
coupling constants of the two phosphorus nuclei in any one
species. Apparently the onset of some dynamic process is being
observed at room temperature (see below).
2
118
J. Chem. Soc., Dalton Trans., 2000, 2113–2122

View Article Online
Repeating the reaction of 1 with [Pt(H)(THF)(PEt ) ][BF ]
starting with a sample of 4b. It is therefore merely a coincidence
that the CH Cl reaction at room temperature between 1 and
3
2
4
but in CH Cl at ambient temperatures yielded complex 4a and
2
2
2
2
a second compound which was easily separated by chromato-
graphic purification and fractional crystallization. This second
species was identified as [3,3,3-(CO) -3,8-{Pt(H)(PEt )}-8-
[Pt(H)(THF)(PEt ) ][BF ] (Scheme 4) produces 4a and 5 in a
3 2 4
ratio of 1:2. The formation of 4b can solely be accounted for
by the isomerization of 4a at a temperature as low as that of
refluxing THF (65 ЊC).
3
3
(
µ-H)-closo-3,1,2-ReC B H ] 5 and the ratio of 4a:5 was found
2 9 10
to be 1:2. There was no evidence of the formation of 4b under
We have synthesized many dimetallic carbaborane complexes
where one of the metals is exopolyhedrally σ bonded to a boron
these conditions. Compound 5 was easily identified from
1
physical and spectroscopic data (Tables 1–3). The H NMR
atom in the CCBBB ring of a 1,2-R -closo-3,1,2-MC B H
8
2
2
9
spectrum displays two notable high field resonances. The first
12
(
R = H or Me) cage. Hitherto there has been no precedent in
at δ Ϫ4.84 shows the distinctive characteristics of a B–H1 Pt
this chemistry for σ-bonded exo-metal fragments isomerizing
in such a manner as to transfer from α-B to β-B vertices, or
vice versa. There was not even any evidence of such exchange
occurring between complexes 3a and 3b in this study.
95
system with J(BH) = 60 Hz and satellites due to the
Pt
5,7
nucleus [J(PtH) ≈ 465 Hz]. An additional doublet splitting
[J(HH) = 15 Hz] results from a trans coupling across the plat-
inum with a terminal hydride, which gives rise to the second
To account for the broadness of the peaks in the NMR spec-
tra of complexes 4a and 4b at ambient temperatures a dynamic
process involving heterolytic Re–Pt bond cleavage may be
taking place. This would be feasible because the local Re(CO)3
moiety is perfectly stable in exo-closo complexes such as [3,3,3-
(CO) -8-(µ-H)-8-{M(CO) (η-C H )}-closo-3,1,2-ReC B H ]
high field resonance at δ Ϫ11.03. The latter signal is quite broad
3
1
1
1
1
with irresolute P– H or H– H couplings, but with a large
195
1
Pt– H coupling for the satellites [J(PtH) = 1090 Hz], strongly
indicating the terminal nature of this hydride ligand.
7a,11
This is
reminiscent of the complex [1,2-Me -3-(µ-CC H Me -2,6)-3,3-
2
6
3
2
3
2
5
5
2
9
10
(
CO) -3,8-{Pt(PEt )}-8-(µ-H)-closo-3,1,2-WC B H ] which has
2 3 2 9 8
(M = Fe or Ru) and [3,3,3-(CO) -8,9,12-(µ-H) -8,9,12-{RuCl-
3
3
a terminal platinum hydride [δ Ϫ10.22, J(PtH) = 1165 Hz] and a
4
(
1
PPh ) }-closo-3,1,2-ReC B H ] as it is in the parent anion of
. The cationic platinum center of the pendant exo-[Pt(PEt ) ]
3
2
2
9
8
1
trans B–H Pt proton (δ Ϫ4.7), though the H NMR spectrum
ϩ
3 2
from which these resonances were taken lacks the fine detail
group could readily be stabilized by the adjacent β-B–H bond
starting from 4a) or one of the two available α-B–H bonds
(starting from 4b) to give zwitterionic intermediates F or F ,
7a
observed for that of complex 5. The employment of the β-B
(
atom (CCBBB) in the B–H Pt agostic system is evident from
ꢁ
ꢀ
1
13
1
11
1
the H, C-{ H} and B-{ H} NMR spectra (Tables 2 and 3).
The first of these shows a single resonance for the cage CH
protons (δ 3.42), implying the presence of a mirror plane bisect-
ing the molecule which would not be associated with the use of
respectively (Scheme 5). Interconversion processes involving
a boron atom α to the cage carbons in the CCBBB ring. This is
11
1
supported by the signal count in the B-{ H} NMR spectrum
and their integrals (1:1:2:2:2:1). The peak at δ 18.1 in this
spectrum is in the region expected for B–H Pt groups with a
1
11
5,7
diagnostic H– B coupling [J(HB) = 60 Hz]. The presence
of only one PEt ligand on the platinum was realized in the
3
31
1
P-{ H} NMR spectrum (Table 3) with a single sharp reson-
1
ance at δ 39.1 [J(PtP) = 3798 Hz] and particularly in the H
NMR spectrum where peak integrals were particularly
informative. The 15 Hz coupling between the agostic and
1
terminal hydrogen nuclei was verified by a H NMR measure-
ment with selective decoupling of the broad hydride resonance
at δ Ϫ11.03 (Supporting Information). With the exception of
the satellites, the doublet of quartets at δ Ϫ4.84 was reduced to
11
a simple B-coupled quartet, as expected.
Thus the loss of a PEt ligand is confirmed in the formation
3
of complex 5, the structure of which would seem to support the
postulated intermediates A in Scheme 1, and perhaps even D in
Scheme 3. Phosphine dissociation would also be consistent with
ϩ
the observed formation of the [Pt(H)(PEt ) ] cation. Since the
3
3
product ratio of 4a:5 (1:2) matches that of 4a:4b (1:2) it
might have been expected that 5 was in fact the logical precur-
sor to 4b, noting that in their exopolyhedral interactions with
the platinum atom both involve β-B vertices. It was further
Scheme
5
Possible mechanism for the interconversion between
isomers 4a and 4b.
thought that addition of 1 mol equivalent of PEt to 5 would
3
afford 4b. However, this did not occur (Scheme 4). An NMR
examination of the reaction mixture revealed a ca. 50% decom-
position of 5 to [Pt(H)(PEt ) ][3,3,3-(CO) -closo-3,1,2-ReC B -
metal–metal bond fission/formation and which give such exo-
closo species with no M–M connectivity have been documented
for solutions of the complexes [3,3,3-(CO) -8,9-(µ-H) -8,9-
3
3
3
2
9
H ]. The question then arises as to how 4b forms and perhaps,
1
1
3
2
therefore, a shadow of doubt is thrown on the intermediacy of
A (Scheme 1) in the synthesis of the compounds 3. It was previ-
ously mentioned that 4a and 4b may be undergoing some
dynamic process at room temperature (see above). NMR spec-
{Rh(PPh ) }-closo-3,1,2-ReC B H ] and [3,3,3-(CO) -3,8-
3 2 2 9 9 3
4
{M(PPh )}-8-(µ-H)-closo-3,1,2-ReC B H ] (M = Cu or Ag).
3
2
9
10
At elevated temperatures (in refluxing THF or above) a full
oxidative addition of the α- or β-B–H bond might occur to give
a platinum() center. In the resulting intermediate G this more
electrophilic Pt atom could be stabilized by a B–H bond from
tra were measured on a sample of 4a in d -toluene at 90 ЊC, as it
8
was suspected that spectral measurements at an elevated tem-
perature would reveal something about this dynamic process.
To our great surprise the formation of 4b was clearly observed
to occur until a ratio of 4a to 4b of 1:2 was reached, this taking
ca. 3 h at this temperature. A similar result was obtained when
the non-metal-coordinating B belt. Once again this can satis-
5
factorily be rationalized because these B belt boron vertices are
5
occasionally found to participate in agostic bonding to exo-
4,13
metals. Intermediate G may then transform into either of the
J. Chem. Soc., Dalton Trans., 2000, 2113–2122 2119

View Article Online
intermediates F, which can then undergo reformation to 4a or
b. The ground-state structures for 4a and 4b at the low temper-
4
atures of the NMR measurements are unlikely to be in the form
of F, as the NMR spectra are most similar to those of their 3a
and 3b congeners. Additionally, there is no actual direct spectral
evidence for the intermediates F or for G.
Doubts were raised above as to the viability of the postulated
intermediates A and D in Schemes 1 and 3, since the complex 5
did not proceed to give 4b as had been expected. An alternative
system was therefore scrutinized: the reaction of 1 with the salt
[
Pt(Me)(THF)(PMe Ph) ][BF ], generated in situ by treating
2 2 4
[
PtCl(Me)(PMe Ph) ] with 1 mol equivalent of Ag[BF ] in
2
2
4
THF. The reaction was evidently much more rapid than that
between 1 and [Pt(H)(THF)(PEt ) ][BF ] as a single product,
3
2
4
[
3,3,3-(CO) -3,4-{Pt(PMe Ph) }-closo-3,1,2-ReC B H ]
6a,
3
2
2
2
9
10
was isolated after 1 h reaction time at room temperature. Com-
pound 6a was characterized from spectroscopic data (Tables 2
and 3) and found to be manifestly similar to 4a. Most notable
11
1
was the doublet resonance in the B-{ H} NMR spectrum at
δ 35.9 [J(PB) = 65, J(PtB) ≈ 530 Hz] to be compared with that
in 4a [δ 34.5, J(PB) = 76, J(PtB) ≈ 580 Hz]. The room temper-
Scheme 6 Formation of complex 6a and the facile interconversion
31
1
with its isomer 6b.
ature P-{ H} NMR spectrum for 6a was also similar to that
for 4a, implying the existence of an analogous dynamic process
the establishment of an equilibrium with a second isomer. This
1
13
1
and it was necessary to measure this and the H and C-{ H}
was identified as [3,3,3-(CO) -3,8-{Pt(PMe Ph) }-closo-3,1,2-
3
2
2
NMR spectra at Ϫ80 ЊC to observe distinguishable signals.
ReC B H ] 6b from new signals that appeared in the NMR
2
9
10
spectra of the mixture. The final equilibrium ratio of 6a to 6b at
5 ЊC was found to be 1:1 and furthermore 6b could not be
2
isolated free of 6a. However, because all relevant signals for 6a
were identified from near pure solutions of this complex, those
for 6b in the mixture were easily abstracted from the spectra of
11
1
the mixture. The B-{ H} NMR spectrum of 6b displayed a
195
doublet peak at δ 47.3 with the customary broad Pt satellites
J(PB) = 76, J(PtB) ≈ 485 Hz]. These data are very close to
[
1
13
those observed for 4b [δ 45.6, J(PtB) ≈ 490 Hz]. The H and C-
{
1
H} NMR spectra measured at Ϫ80 ЊC both reflected the
presence of an apparent mirror plane bisecting the molecule
through the Re, Pt, apical B and both P atoms with one signal in
1
the H NMR spectrum at δ 3.41 for the cage CH protons and
13
1
also in the C-{ H} NMR spectrum at δ 34.5 for the cage CH
carbon nuclei. In the latter spectrum the CO nuclei give rise to
two signals at δ 192.2 and 191.8 (1:2) as expected with the
molecular C symmetry. Clearly 6b is an analog of 3b and 4b.
s
31
1
The room temperature P-{ H} NMR spectrum of the 6a/6b
mixture comprised poorly resolved peaks which became
sharper at Ϫ80 ЊC. The same dynamic equilibrium that pro-
duced 4b from 4a must be occurring for the complexes 6, except
that for the latter it is an ambient temperature phenomenon, a
feature entirely without precedent in this sphere of metalla-
carbaborane chemistry. The reason for the difference in the
facility of isomeric interconversion, or lack thereof, for the
compounds 3, 4 and 6 is at the present time unknown.
The reaction that produced complex 6a was monitored by IR
and NMR spectroscopy and after about 30 min reaction time
the IR and NMR data clearly indicated the presence of a
species, designated H (Scheme 6), akin to complex 5. The IR
Ϫ1
spectrum of H showed peaks at 2034, 1964 and 1949 cm
Conclusion
which are almost identical with those of 5 (2034, 1962, 1950
As we expand our crop of new metallacarbaboranes incorpor-
Ϫ1
1
cm ). The H NMR spectrum of H displayed a diagnostic
ating 1,2-R -closo-3,1,2-MC B H (R = H or Me) and closo-2,1-
195
2
2
9
9
quartet at δ Ϫ5.07 [J(BH) = 81 Hz] with a Pt satellite set
MCB H cages, it is apparent to us that our understanding of
II
10 11
[
J(PtH) = 209 Hz], in addition to a peak at δ 0.88, where Pt Me
their subsequent reactivity patterns is still limited. In particular,
14
groups typically resonate. Also detected in the mixture were
signals due to free PMe Ph and the cation [Pt(Me)(PMe Ph) ] ,
ϩ
the activation of B–H vertices in the CCBBB coordinating face
of the carbaborane remains currently unpredictable. This effect
is clearly dependent on the physical conditions under which it is
made to occur (as in the syntheses of complexes 3, 4 and 5),
as well as the intrinsic structure of the metallacarbaboranes
2
2
3
verified by independent synthesis of this species. This would
support the idea of phosphine ligand dissociation from one
platinum center to be scavenged by another. As the reaction
proceeded beyond 30 min, IR and NMR peaks due to H dimin-
ished and those for 6a grew accordingly. Thus H, which must
possess an α-B–H Pt unit, is likely to be an intermediate on
the path to 6a, and the existence of A and D in Schemes 1 and 3
is placed on a firmer footing.
(
cf. complexes 4 and 6).
Experimental
General
Continued stirring of solutions of complex 6a in CH Cl for
2
2
2
4 h accompanied by analysis by NMR spectroscopy revealed
All experiments were conducted under an atmosphere of dry
2
120 J. Chem. Soc., Dalton Trans., 2000, 2113–2122

View Article Online
nitrogen using Schlenk tube techniques. Solvents were freshly
distilled under nitrogen from appropriate drying agents before
use. Light petroleum refers to that fraction of boiling point 40–
and [3,3,3-(CO) -3,4-{Pt(PEt ) }-closo-3,1,2-ReC B H ] 4a
(0.02 g) in order of elution.
3
3
2
2
9
10
6
0 ЊC. Chromatography columns (ca. 20 cm long and 2 cm in
Reaction of compound 1 with [Pt(H)(THF)(PEt ) ][BF ] at
3
2
4
diameter) were packed with silica gel (Acros, 60–200 mesh).
Celite pads used for filtration were ca. 3 cm long and 2 cm in
diameter. The NMR measurements were recorded at the follow-
room temperature. The procedure was similar to that described
above for the compounds 4 ([PtCl(H)(PEt ) ] (0.09 g, 0.19
3
2
mmol), Ag[BF ] (0.04 g, 0.21 mmol), 1 (0.10 g, 0.19 mmol)),
4
1
13
11
3
ing frequencies: H at 360.13, C at 90.56, B at 115.55, and
except that the reaction was carried out in CH Cl (20 cm ) at
2
2
31
4
15
P at 145.78 MHz. The reagents 1, [PtCl(H)(PEt ) ], and
3
2
25 ЊC for 24 h. Column chromatography at Ϫ30 ЊC gave a
single yellow fraction upon elution with CH Cl –light petrol-
8a
[PtCl(Me)(PMe Ph) ] were made as previously described. The
2 2
2
2
platinaphosphine complex [PtCl {Ph P(CH ) PPh }] was syn-
thesized by adding Ph P(CH ) PPh to CH Cl solutions of
2
2
2
2
2
eum (2:3). This was composed of a mixture of compounds 4a
and 5. The former was considerably more soluble than 5 and
was readily washed away upon crystallization from CH Cl –
2
2
2
2
2
2
16
[PtCl (NCPh) ]. K-Selectride, Proton Sponge, DBU, Na[BH4]
2 2
2
2
3
and Ag[BF ] were purchased from Aldrich.
4
light petroleum (8 cm , 1:1) giving yellow microcrystals of
3,3,3-(CO) -3,8-{Pt(H)(PEt )}-8-(µ-H)-closo-3,1,2-ReC B H ]
[
3
3
2
9
10
Preparations
5 (0.05 g). The compound 4a (0.03 g) was isolated from the
washings and crystallized from CH Cl –light petroleum (10
2
2
[
3,3,3-(CO) -3,8-(Pt{Ph P(CH ) PPh })-8-(ꢂ-H)-closo-3,1,2-
3 2 2 2 2
3
cm , 1:4).
ReC B H ][BF ]. To the compounds [PtCl {Ph P(CH ) PPh }]
2
9
10
4
2
2
2
2
2
(
0.40 g, 0.60 mmol) and Ag[BF ] (0.24 g, 1.23 mmol) was
4
3
[3,3,3-(CO) -3,n-{Pt(PMe Ph) }-closo-3,1,2-ReC B H ] (n ؍
or 8). A similar procedure to that described for complexes 4a
3
2
2
2
9
10
added MeCN (20 cm ). A precipitate appeared almost instant-
aneously, though stirring was maintained for ca. 2 h. The sus-
pension was filtered through a Celite pad to remove AgCl using
MeCN as solvent. The volume of the filtrate was then reduced
to ca. 5 cm and a solution of 1 (0.32 g, 0.60 mmol) in MeCN
(
in vacuo and CH Cl (20 cm ) added to the residue. The sus-
pension, now containing undissolved CsBF , was once again
filtered through Celite and solvent then removed in vacuo.
Crystallization from CH Cl –Et O (20 cm , 1:4), followed by
4
and 5 was used ([PtCl(Me)(PMe Ph) ] (0.15 g, 0.29 mmol),
2
2
Ag[BF ] (0.06 g, 0.31 mmol), 1 (0.15 g, 0.28 mmol)), but with
4
3
just 1 h stirring in CH Cl at 25 ЊC. Chromatographic purifi-
2
2
3
cation at Ϫ30 ЊC (CH Cl –light petroleum, 1:1) gave a single
2 2
10 cm ) added and stirred for 5 min. Solvent was then removed
3
yellow fraction of [3,3,3-(CO) -3,4-{Pt(PMe Ph) }-closo-3,1,2-
3 2 2
2
2
ReC B H ] 6a (0.03 g). Stirring in CH Cl for 24 h led to 50%
2
9
10
2
2
4
isomerization to [3,3,3-(CO) -3,8-{Pt(PMe Ph) }-closo-3,1,2-
3
2
2
3
ReC B H ] 6b.
2 9 10
2
2
2
3
washing with Et O (4 × 20 cm ) and drying in vacuo, yielded
2
Crystallography
[3,3,3-(CO) -3,8-(Pt{Ph P(CH ) PPh })-8-(µ-H)-closo-3,1,2-
3 2 2 2 2
ReC B H ][BF ] 2b (0.55 g).
2
9
10
4
Crystals of complex 2b were grown by evaporation from a
CH Cl solution, while those of 3a were grown by diffusion of
2
2
[
3,3,3-(CO) -3,n-(Pt{Ph P(CH ) PPh })-closo-3,1,2-ReC B -
3 2 2 2 2 2 9
light petroleum into CH Cl2 solutions. Both complexes co-
2
H ] (n ؍ 4 or 8). Various alkali metal bases can be used in this
1
0
crystallized with a molecule of CH Cl in the asymmetric unit.
2
2
reaction in addition to varying the solvent between CH Cl and
2
2
A crystal of 2b was mounted on a glass fiber and data were
collected on a Siemens SMART CCD area-detector three-circle
diffractometer. A full sphere of data were set to collect, but
unfortunately the crystal decomposed before the data collection
could be completed and the raw data were corrected accordingly.
However there was still a substantial redundancy in data allow-
THF (see text). The use of one particular base (K-Selectride)
and solvent (CH Cl ) is documented here. Complex 2b (0.10 g,
2
2
3
0
.09 mmol) was dissolved in CH Cl (10 cm ) and cooled to
2 2
Ϫ80 ЊC. A solution of K-Selectride (1 M in THF, 13 µl, 0.09
mmol) was added and stirring maintained while warming to
room temperature. After ca. 1 h at room temperature, solvent
was removed in vacuo and CH Cl –light petroleum (5 cm , 1:1)
added to the residue. This was chromatographed at Ϫ30 ЊC,
eluting initially with the same solvent mixture to yield a yellow
fraction. Removal of solvent in vacuo followed by crystalliz-
17
ing for empirical absorption corrections (SADABS ) to be
3
2
2
applied using multiple measurements of equivalent reflections.
17
The data frames were integrated using SAINT. For 3a dif-
fracted intensities were collected on an Enraf-Nonius CAD-4
diffractometer. No crystal decay was detected during the data
collection. The data were corrected for Lorentz, polarization
and X-ray absorption effects, the latter by a semi-empirical
method based on azimuthal scans of ψ data of several
Euclerian angles (χ) near 90Њ.
3
ation from CH Cl –light petroleum (10 cm , 1:4) gave yellow
2
2
microcrystals of [3,3,3-(CO) -3,8-(Pt{Ph P(CH ) PPh })-closo-
3
2
2
2
2
3
,1,2-ReC B H ] 3b (0.02 g). A second yellow fraction was
2 9 10
eluted from the column using a 3:2 mixture of CH Cl –light
petroleum. Removal of solvent in vacuo and crystallization
2
2
Both structures were solved by conventional direct methods
3
2
from CH Cl –light petroleum (10 cm , 2:3) produced yellow
2
2
and refined by full-matrix least squares on all F data using
microcrystals of [3,3,3-(CO) -3,4-(Pt{Ph P(CH ) PPh })-closo-
3
2
2
2
2
Siemens SHELXTL version 5.03 or SHELXTL 93 and
17
3
,1,2-ReC B H ] 3a (0.04 g).
2 9 10
SHELXTL 97. For 3a the non-hydrogen atoms were
refined with anisotropic thermal parameters. For 2b all non-
hydrogen atoms were refined anisotropically (with restrictions
placed to generate sensible ellipsoids) with the exception of
cage boron atoms B(3) and B(5) and the phenyl carbon atom
C(106) which were refined with isotropic thermal parameters.
All four phenyl rings in this structure were constrained to be
planar. Cage carbon atoms were identified from the magnitudes
of their anisotropic thermal parameters and from a comparison
of the bond lengths to adjacent boron atoms. The agostic
B–H Pt hydrogen atom H(4) in complex 2b was located
[
3,3,3-(CO) -3,n-{Pt(PEt ) }-closo-3,1,2-ReC B H ] (n ؍ 4
3 3 2 2 9 10
or 8). To the compounds [PtCl(H)(PEt ) ] (0.17 g, 0.36 mmol)
3
2
3
and Ag[BF ] (0.08 g, 0.41 mmol) was added THF (15 cm ) and
4
the mixture stirred for 2 h. The suspension was filtered through
Celite onto a solution of 1 (0.20 g, 0.37 mmol) in THF (5 cm ).
3
This was then refluxed for 4 d. Solvent was removed in vacuo
3
and CH Cl (10 cm ) added. The mixture was filtered through
2
2
Celite followed by removal of solvent in vacuo. The residue
3
was taken up in CH Cl –light petroleum (5 cm , 2:3). Chrom-
2
2
atographic separation at Ϫ30 ЊC, eluting with CH Cl –light
(Uiso = 1.5U [B(4)]), but its position not refined. All other
2
2
iso
petroleum (2:3), afforded two yellow fractions. Removal of
solvent from either followed by crystallization from CH Cl –
hydrogen atoms were included in calculated positions and
allowed to ride on the parent boron or carbon atoms with iso-
tropic thermal parameters (Uiso = 1.2Uiso of the parent atom
2
2
3
light petroleum (10 cm ; 1:4) gave yellow microcrystals of
3,3,3-(CO) -3,8-{Pt(PEt ) }-closo-3,1,2-ReC B H ] 4b (0.04 g)
[
except for the B–H hydrogen atoms in 2b where Uiso = 1.5U ).
3
3
2
2
9
10
iso
J. Chem. Soc., Dalton Trans., 2000, 2113–2122
2121

View Article Online
Table 7 Data for crystal structure analyses of complexes 2b and 3a
2
b
3a
Chemical formula
M
T/K
C H B F O P PtReؒCH Cl
1167.85
292(2)
C H B O P PtReؒCH Cl
2
1080.03
293(2)
3
1
35 10
4
3
2
2
2
31 34
9
3
2
2
Colour, habit
Crystal size/mm
Crystal system
Space group
Yellow prism
0.55 × 0.35 × 0.30
Orange-yellow parallelopiped
0.24 × 0.22 × 0.18
Triclinic
Triclinic
¯
¯
P1
P1
a/Å
b/Å
c/Å
α/Њ
β/Њ
γ/Њ
12.122(3)
13.084(2)
15.155(3)
110.42(2)
106.75(2)
92.59(2)
2128.5(8)
2
11.302(2)
13.0493(12)
15.508(4)
66.198(14)
80.29(2)
66.735(10)
1922.4(5)
2
3
U/Å
Z
µ(Mo-Kα)/mm
Ϫ1
6.374
7.035
No. reflections collected
No. independent reflections
R(int)
8730
6816
0.061
0.146, 0.065
0.0432, 0.0
2.047, Ϫ1.781
5329
5019
0.037
0.126, 0.053
0.0517, 0.5597
1.936, Ϫ1.296
a
b
Final residuals wR2, R1
a
Weighting factors a, b
Ϫ3
Largest difference peak, hole/e Å
a
b
2
2
2
2
2
2
1/2
Ϫ1
2
2
2
2
2
Structure was refined on Fo using all data: wR2 = [Σ[w(Fo Ϫ F ) ]/Σw(F ) ] where w = [σ (F ) ϩ (aP) ϩ bP] and P = [max(F ,0) ϩ 2F ]/3.
c
o
o
o
c
The value is given for comparison with refinements based on F with a typical threshold of F > 4σ(F ) and R1 = Σ |F | Ϫ |F | /Σ|F | and
o
o
o
o
c
o
Ϫ1
2
2
w
= [σ (F ) ϩ gF ].
o o
Large residual peaks due to absorption effects were located
close to the metal centers in both structures. All calculations
were carried out on Silicon Graphics Iris, Indigo or Indy or
Dell computers and the experimental data are summarized in
Table 7.
CCDC reference number 186/1953.
See http://www.rsc.org/suppdata/dt/b0/b002043p/ for crystal-
lographic files in .cif format.
Dalton Trans., 1993, 1917; Y.-K. Yan and M. P. Mingos, Chem.
Soc. Rev., 1995, 203; G. Ferguson, J. Pollock, P. A. McEneany,
D. P. O’Connell, T. R. Spalding, J. F. Gallagher, R. Macias and
J. D. Kennedy, Chem. Commun., 1996, 679.
(a) D. D. Devore, J. A. K. Howard, J. C. Jeffery, M. U. Pilotti and
F. G. A. Stone, J. Chem. Soc., Dalton Trans., 1989, 303; (b) J. E.
Goldberg, D. F. Mullica, E. L. Sappenfield and F. G. A. Stone,
J. Chem. Soc., Dalton Trans., 1992, 2693; (c) S. J. Dossett,
D. F. Mullica, E. L. Sappenfield, F. G. A. Stone and M. J. Went,
J. Chem. Soc., Dalton Trans., 1993, 281; (d) J. E. Goldberg, J. A. K.
Howard, H. Müller, M. U. Pilotti and F. G. A. Stone, J. Chem. Soc.,
Dalton Trans., 1990, 3055; (e) J. C. Jeffery, P. A. Jelliss and F. G. A.
Stone, Inorg. Chem., 1993, 32, 3943.
7
Acknowledgements
We thank the Robert A. Welch Foundation for support (Grant
AA-1201), and Dr John C. Jeffery (Bristol University) for
assistance in acquiring the crystallographic data for complex
8 (a) N. Carr, M. C. Gimeno and F. G. A. Stone, J. Chem. Soc., Dalton
Trans., 1990, 2617; (b) S. Anderson, D. F. Mullica, E. L. Sappenfield
and F. G. A. Stone, Organometallics, 1995, 14, 3516.
9
R. J. Blau and J. H. Espenson, Inorg. Chem., 1986, 25, 878.
2
b.
1
0 J. C. Jeffery, M. A. Ruiz, P. Sherwood and F. G. A. Stone, J. Chem.
Soc., Dalton Trans., 1989, 1845; S. L. Hendershot, J. C. Jeffery,
P. A. Jelliss, D. F. Mullica, E. L. Sappenfield and F. G. A. Stone,
Inorg. Chem., 1996, 35, 6561.
References
1
R. J. Puddephatt and J. Xiao, Coord. Chem. Rev., 1995, 143, 457
and refs. cited therein; M. Bergamo, T. Beringhelli, G. Ciani,
G. D’Alfonso, M. Moret and A. Sironi, Organometallics, 1996, 15,
11 D. M. Roundhill, Adv. Organomet. Chem., 1975, 13, 273; H. D.
Kaesz and R. B. Saillant, Chem. Rev., 1972, 72, 231.
12 F. G. A. Stone, Adv. Organomet. Chem., 1990, 31, 53; S. A. Brew
and F. G. A. Stone, Adv. Organomet. Chem., 1993, 35, 135;
P. A. Jelliss and F. G. A. Stone, J. Organomet. Chem., 1995, 500,
307.
13 J. A. Long, T. B. Marder, P. E. Behnken and M. F. Hawthorne,
J. Am. Chem. Soc., 1984, 106, 2979; I. T. Chizhevsky, I. A.
Lobanova, V. I. Bregadze, P. V. Petrovskii, V. A. Antonovich,
A. V. Polyakov, A. I. Yanovskii and Y. T. Struchkov, Mendeleev
Commun., 1991, 48; J. C. Jeffery, P. A. Jelliss and F. G. A. Stone,
J. Chem. Soc., Dalton Trans., 1993, 1073; A. Franken, J. Kautz,
F. G. A. Stone and P.-Y. Yu, Organometallics, 2000, 20, 1993.
14 F. R. Hartley, in Comprehensive Organometallic Chemistry,
eds. E. W. Abel, F. G. A. Stone, G. Wilkinson, Pergamon (Elsevier
Science Ltd), Oxford, 1982, vol. 6, sect. 39.5.1.5(i).
1
637 and refs. cited therein; C. P. Casey, E. W. Rutter Jr. and K. J.
Haller, J. Am. Chem. Soc., 1987, 109, 6886; S. W. Carr, X. L. R.
Fontaine, B. L. Shaw and M. Thornton-Pett, J. Chem. Soc., Dalton
Trans., 1988, 769.
2
3
I. Blandford, J. C. Jeffery, P. A. Jelliss and F. G. A. Stone,
Organometallics, 1998, 17, 1402.
M. F. Hawthorne and T. D. Andrews, J. Am. Chem. Soc., 1965, 87,
2
496; T. D. Andrews, M. F. Hawthorne, D. V. Howe, R. L. Pilling,
D. Pitts, M. Reintjes, L. F. Warren Jr., P. A. Wegner and D. C.
Young, J. Am. Chem. Soc., 1968, 90, 879.
4
5
6
D. D. Ellis, P. A. Jelliss and F. G. A. Stone, Organometallics, 1999,
1
8, 4982.
M. J. Attfield, J. A. K. Howard, A. N. de M. Jelfs, C. M. Nunn and
F. G. A. Stone, J. Chem. Soc., Dalton Trans., 1987, 2219.
15 G. W. Parshall, Inorg. Synth., 1970, 12, 28.
16 G. K. Anderson and M. Lin, Inorg. Synth., 1990, 28, 60.
17 Siemens X-Ray Instruments, Madison, WI, 1995.
H. C. Kang, S. S. Lee, C. B. Knobler and M. F. Hawthorne, Inorg.
Chem., 1991, 30, 2024; N. L. Douek and A. J. Welch, J. Chem. Soc.,
2
122
J. Chem. Soc., Dalton Trans., 2000, 2113–2122