Synthesis of cobalt-containing cyclophanes, and the formation of an unprecedented seven-membered cyclic diyne

Article abstract of DOI:10.1039/b310515f

The acid-catalysed Nicholas reaction of the bis-propargyl complex [{Co ₂(CO)₆(μ-η²-HOMe₂CC≡C-) }₂] 1 in the presence of a variety of nucleophiles leads in each case to [{Co₂(CO)₆}₂{cyclo-μ-η²: μ-η²-C(=CH₂)CH₂CMe₂C≡C- C≡}], a complex which contains an unprecedented seven-membered macrocyclic diyne ligand. The reactivity of 1 is compared to that of [{Co ₂(CO)₆}₂{μ-η²:μ- η²-1,4-C₆H₄(C≡CCMe ₂OH)₂}] which does not cyclise on treatment with nucleophiles but instead gives the expected substitution products.

Full text of DOI:10.1039/b310515f

View Article Online / Journal Homepage / Table of Contents for this issue
P A P E R
Synthesis of cobalt-containing cyclophanes, and the formation of an
unprecedented seven-membered cyclic diyney
Vladimir B. Golovko,a Louisa J. Hope-Weeks,b Martin J. Mays,*a Mary McPartlin,c
Anna M. Sloana and Anthony D. Woods*a
a
Department of Chemistry, University of Cambridge, Lensfield Road, Cambridge
UK CB2 1EW. E-mail: adw24@cus.cam.ac.uk; mjm14@cus.cam.ac.uk
University of California, Lawrence Livermore National Laboratory,
Division of Chemistry and Chemical Engineering, Livermore CA 94551, USA
Department of Health and Human Sciences, London Metropolitan University, London,
UK N7 8DB
b
c
Received (in Durham, UK) 29th August 2003, Accepted 14th January 2004
First published as an Advance Article on the web 3rd March 2004
2
=
The acid-catalysed Nicholas reaction of the bis-propargyl complex [{Co (CO) (m-Z -HOMe CC C–)} ] 1
=
2
6
2
2
in the presence of a variety of nucleophiles leads in each case to [{Co2(CO)6}2{cyclo-m-Z2:m-Z2-
C( CH )CH CMe C C–C C}], a complex which contains an unprecedented seven-membered macrocyclic
diyne ligand. The reactivity of 1 is compared to that of [{Co (CO) } {m-Z :m-Z -1,4-C H (C CCMe OH) }]
=
which does not cyclise on treatment with nucleophiles but instead gives the expected substitution products.
=
=
=
2
=
=
2
2
2
2
=
2
6
2
6
4
2
2
Instead, in every case the, complex [{Co2(CO)6}2{cyclo-m-Z2:
Introduction
2
=
=
=
2
=
m-Z -C( CH )CH CMe C C–C C}] 2 was isolated in vir-
=
2
2
In recent years there has been a great deal of interest in the
ability of dicobalt coordinated bis-propargyl alcohols to stabi-
lise dications.1 We have recently reported that the acid-cata-
lysed nucleophilic substitution of such alcohols with dithiols
(often called the Nicholas reaction) leads to the synthesis of
a hitherto inaccessible range of new thio-macrocycles contain-
ing both sulfur atoms and diyne units.1,2
There is considerable current interest both in the synthesis
of polymeric materials with acetylenic backbones (due to
their important electronic properties) and in the synthesis of
small cyclophanes and strained conjugated carbocycles.3 It
seemed possible that a Nicholas reaction between dicobalt-
coordinated bis-propargyl alcohols and arene nucleophiles
might similarly provide a one-pot route both to unsaturated
chains and to cyclophanes. The investigation was stimulated
by the recent report of Vollhardt and co-workers that the
pyrolysis of dicobalt-stabilised conjugated carbocycles leads
to the smooth formation of nano-tubes and nano-onions
whereas the analogous uncoordinated carbocycles decompose
explosively.4
tually quantitative yield (Scheme 1).
Complex 2 contains a seven-membered conjugated diyne
ring which represents, to our knowledge, the smallest such ring
to have been reported. Prior to the synthesis of 2 the smallest
reported cyclic diyne in which the two alkyne units are adja-
cent contained ten carbon atoms.5 The smallest diyne ring of
any kind to have been previously reported is the eight-mem-
bered sym-dibenzo-1,5-cycloctadien-3,7-diyne, although the
benzo linkers in this latter molecule prevent any ring strain
at the alkynic centres.6 The seven-membered diyne-containing
ring present in 2 is particularly noteworthy in view of the con-
jugation of the two alkyne units and the vinyl group. The
severe ring strain that would otherwise have been present in
such a small cyclic diyne is, of course, at least partially alle-
viated by the coordination of the diyne to the dicobalt units.
The reaction presumably involves an intra-molecular self-
cyclisation of an intermediate A (Scheme 2). Intermediate A
stems from the dehydration of one of the propargyl moieties
to give an ene-yne followed by protonation of the remaining
propargyl alcohol and subsequent loss of H2O. Indeed, if the
reaction is quenched after stirring at ꢀ78 ꢁC for 30 mins it is
2
2
=
=
2
=
=
possible to isolate [{Co (CO) } {m-Z :m-Z -HOMe C CC
2
6
2
=
CC( CH2)Me}] 3 which is the precursor to intermediate A.
Results and discussion
The acid-catalysed reactions of the dicobalt-coordinated bis-
=
2
propargylic complex, [{Co (CO) (m-Z -HOMe CC C–)} ] 1,
=
2
6
2
2
with a variety of arene nucleophiles such as PhOH, PhSH,
1,4-C6H4(OH)2 , and PhOMe were investigated. No substitu-
tion products were isolated in any of these reactions. In fact,
even when 1 was treated with more potent nucleophilic
reagents, such as HSC3H6SH, no substitution products were
obtained, despite the fact that HSC3H6SH has been shown
to react readily with the less sterically hindered complex
2
2
=
[{Co (CO) (m-Z -HOCH C C–)} ] to give a thio-macrocycle.
=
2
6
2
2
y Electronic supplementary information (ESI) available: crystallo-
graphic data for complexes 8 and 11. See http://www.rsc.org/
suppdata/nj/b3/b310515f/
Scheme 1 Acid catalysed cyclisation of 1 and 4.
T h i s j o u r n a l i s Q T h e R o y a l S o c i e t y o f C h e m i s t r y a n d t h e
C e n t r e N a t i o n a l d e l a R e c h e r c h e S c i e n t i f i q u e 2 0 0 4
N e w . J . C h e m . , 2 0 0 4 , 2 8 , 5 2 7 – 5 3 4
527

View Article Online
ꢁ
˚
Table 1 Bond lengths (A) and bond angles ( ) for 2, 5 and 7
Unit
2
7
5
C1–C2
C2–C3
1.362(6)
1.439(8)
1.338(5)
1.510(6)
1.533(6)
1.966
1.362(6)
1.438(6)
1.349(6)
1.508(7)
1.516(7)
1.951
1.360(8)
1.42(1)
1.364(9)
1.50(1)
1.538(8)
1.980
C3–C4
C4–C5
C8–C9
Mean Calkyne–Co(1)
Mean Calkyne–Co(2)
Mean Calkyne–Co(3)
Mean Calkyne–Co(4)
C1–C9
1.961
1.965
1.980
1.964
1.975
1.972
1.972
1.985
1.98
1.476(6)
2.4640(9)
2.4593(14)
1.466(6)
2.4831(7)
2.4542(9)
1.475(9)
2.477(1)
2.478(1)
Co2–Co1
Co3–Co4
C9–C1–C2
C5–C4–C3
C4–C5–C6
C4–C5–C7
C4–C5–C8
C9–C8–C5
C1–C9–C8
C1–C9–C10
130.3(4)
132.5(4)
110.0(3)
111.0(4)
109.1(3)
117.7(3)
116.2(4)
122.5(4)
128.8(4)
132.6(4)
109.3(4)
111.8(4)
109.4(4)
117.0(4)
116.2(4)
121.8(5)
128.1(6)
127.3(31)
114.6(9)
116.0(13)
111.7(13)
111.2(6)
111.1(6)
120.4(6)
Scheme 2 Proposed pathway leading to formation of 2.
The cyclisation of 1 to 2 proceeds smoothly on addition of
catalytic amounts of HBF4 in the absence of a nucleophile;
the presence of oxygen- or sulfur-based nucleophiles only
serves to increase the rate of the reaction, presumably by facil-
itating proton abstraction (Scheme 2). Alternatively, the rea-
son for the acceleration of the reaction in the presence of
Lewis bases is that it proceeds via a radical mechanism in a
similar way to that reported by Melikyan and co-workers.1c
However, the isolation of 3 when the reaction was quenched
at an early stage, and the lack of other coupled products
suggests that this is not the case.
are very small as compared to those in related acyclic cobalt-
alkyne complexes; the mean C–C C angle in such alkyne com-
=
=
plexes is 141.31ꢁ.9 Indeed, there is only one documented
cobalt-complexed alkyne that contains a lower bend-back angle
than 2.10 Furthermore, the bond angles at C(9) and C(8) show
a certain amount of distortion as a result of the ring strain pre-
sent within 2. Thus the C(1)–C(9)–C(8) angle of 116.2(4)ꢁ is
smaller than expected for an idealised sp2 geometry and the
C(9)–C(8)–C(5) angle of 117.7(3) is significantly greater than
the expected sp3 angle of 109ꢁ.
Although dehydration reactions are well known,7 it is
unusual for them to be the sole outcome of a reaction when
nucleophiles are also present. It should be noted that the
analogous reaction of the mono-propargylic complex
2
=
[Co (CO) (m-Z -HC CCMe OH)] in the presence of a nucleo-
=
2
6
2
phile does not result in dehydration; instead the expected addi-
tion of the nucleophile to the propargylic centre takes place.8
The most striking feature of the diyne reaction reported here
is, however, the intramolecular cyclisation step to give 2, since
it might reasonably have been assumed that such a reaction
would be disfavoured by the strained cycle which is formed.
Obviously the rate of intramolecular addition must be faster
than that of the competing E1 elimination reaction at the sec-
ond Me2Cþ centre but, if this is the case, it is surprising that no
products derived from standard SN1 substitution are obtained
even when a large excess of a sterically unencumbered nucleo-
phile is used.
The small nature of the seven-membered ring forces the Co2
units in these cores to adopt a cisoid configuration relative to
each other as opposed to the transoid configuration adopted
for these units in acyclic or larger macrocyclic 1,3 diyne
complexes.11
In order to ascertain whether increased steric bulk at the Co2
moieties would prevent cyclisation the related complex
2
=
[{Co (CO) (dppm)(m-Z -HOMe CC C–)} ] 4 was treated with
=
range of nucleophiles. In each case
2
4
2
2
HBF4 and
a
2
2
=
=
=
2
[{Co (CO) (dppm)} {cyclo-m-Z :m-Z -C( CH )CH CMe C C–
2
4
2
2
2
=
C C}] 5 was obtained as the sole product (Scheme 1) in
=
good yield. The precursor 4 was synthesised by refluxing a
toluene solution of 1 with 2.2 equivalents of dppm. Interest-
ingly, some [Co2(CO)6(Z1-dppm)2] 8 and [Co2(CO)4(dppm)-
The X-ray crystal structure of 2 (Fig. 1) shows that,
although the average bond-lengths and angles (Table 1) within
the pseudo-tetrahedral Co2C2 cores fall within the normal
range, there is a good deal of strain present within the mole-
cule. Thus the alkyne bend-back angles of 132.5(4)ꢁ and
130.3(4) for C(5)–C(4)–C(3) and C(9)–C(1)–C(2) respectively
2
=
(m-Z -HOMe CC CCCCMe OH)] 9 were obtained during
=
2
2
work up (Scheme 3).12 These two species are presumably
formed by the phosphine promoted decoordination of a
Co2(CO)6 moiety from an alkyne, in a manner similar to
that previously reported by Simpson et al.11
1
Compound 5 was initially identified by H and 13C NMR
spectroscopic studies. In an attempt to grow crystals of 5 we
Fig. 1 Molecular Structure of 2.
Scheme 3 Reaction of 1 with dppm.
T h i s j o u r n a l i s Q T h e R o y a l S o c i e t y o f C h e m i s t r y a n d t h e
C e n t r e N a t i o n a l d e l a R e c h e r c h e S c i e n t i f i q u e 2 0 0 4
528
N e w . J . C h e m . , 2 0 0 4 , 2 8 , 5 2 7 – 5 3 4

View Article Online
also isolated a minute quantity of [{Co2(CO)4(dppm)(m-Z2-
=
H C C(Me)C C–)} 6, which contains an acyclic bis-eneyne
=
=
2
2
backbone. It is not clear whether 6 was initially present as a
minor impurity in the solution of 5 or whether 5 undergoes
an acid-catalysed ring–opening reaction on standing in solu-
tion open to air. Certainly, the original 1H and 13C NMR spec-
tra of 5 show no resonances due to 6. The molecular structures
of both 6 and 5 (Figs. 2,3 respectively) were obtained by single
crystal X-ray analysis. A bulk-powder diffraction study of sev-
eral batches of crystals obtained by various crystallisation
techniques showed that the sample was almost exclusively 5.
It is unclear why 6 should have been formed; certainly the
dppm ligands of 5 do not appear to add any undue steric con-
straints to the core present in cyclic 2 that might warrant an
isomerism to the (presumably) less strained acyclic diyne 6.
The unusual cisoid configuration of the Co2 cores present in
2 and 5 means that the dppm ligands are both orientated
exo to the macrocyclic cavity. We therefore assume that 6
was present as a minor, inseparable impurity in the original
reaction mixture rather than being formed by a reverse Nicho-
las reaction during the crystallisation of 5.
Interestingly, whilst attempting to form 5 by direct reaction
of 2 with two equivalents of dppm, it was noted that, when
only a small excess of dppm was used with a short reflux period
2
2
=
some [{Co2(CO)4dppm}(Co2(CO)6)(cyclo-m-Z :m-Z -C( CH2)-
=
=
2
=
CH CMe C C–C C)] 7 in which 2has undergone a singledppm
=
2
substitution was also obtained. This substitution occurred at the
=
Co2 moiety adjacent to the C CH2 group rather than to the
CMe2 group (Fig. 3).
In order to determine whether this was merely a coincidental
result or whether the dppm substitution was in fact selective
and occurred preferentially at the Co2 unit adjacent to the
Fig. 3 Molecular structures of 7 (top) and 5 (bottom); dppm hydro-
gens removed for clarity.
=
C CH2 group 2 was treated with one equivalent of dppm.
In these circumstances 7 was obtained in near quantitative
yield and none of the possible alternative in which the dppm
had coordinated to the Co2 moiety adjacent to the CMe2
group. Examination of the crystal structures of 2, 5 and 7
shows that dppm coordination leads to a decrease in the
C(2)–C(1)–C(9) alkyne bend-back angles from 130.3(4) in 2
to 129.0(4) in 7 and 128.1(6) in 5. This apparent decrease is,
however, only just larger than experimental error and should
be treated with some care. The X-ray structures show a
decrease in the C(3)–C(4)–C(5) alkyne bend-back angle from
132.5(4) in 2 to 127.3(31)ꢁ in 5 on dppm substitution i.e. more
marked than the corresponding decrease in the C(2)–C(1)–C(9)
alkyne bend-back angle of 128.8(4) in 7. Due to the high esd’s
in 5 this result is of low significance, however, it is of interest
that this observation is consistent with the anticipated effect
of dppm substitution at Co(4)–Co(3) leading to a greater dis-
tortion of the compound. The C(3)–C(4)–C(5) alkyne
bend-back angle in 5 is the lowest recorded for a complex of
this type and is a result of the unfavourable steric interactions
between the bulky axial P(3)–Ph group and the C(7), C(6)
methyl groups. An inspection of the space-filling diagrams of
5 and 7 (Fig. 4) demonstrates these factors. In 7 the single
=
dppm substitution at the C CH2 end of the molecule of 2
(Co(1)–Co(2) unit) leads to less steric crowding than would
occur were dppm substitution to take place at the other end
of the molecule (i.e. at the Co(3)–Co(4) unit) where the two
methyl groups would be brought into extremely close contact
with a phenyl group dppm ligand. Inspection of the Co(3)–
Co(4) unit in disubstituted 5 shows that substitution at the
Co(3)–Co(4) end of the molecule leads to very close contacts
between the ring methyl groups and the dppm ligand.
It has previously been shown that treatment of bis-Co2(CO)6
coordinated diynes with an excess of phosphine or phosphite
ligand can result in decoordination of one of the Co2 moieties
to yield a mono-coordinated diyne and Co2(CO)4(PR3)4 .11 In
the current case prolonged refluxing of a toluene solution of 2
with an excess of dppm resulted only in the formation of 5. It
seems that decoordination of a Co2 moiety in 5 is disfavoured,
presumably because this would necessarily increase the bend-
back angles at the uncoordinated alkyne carbons from 130ꢁ
in 2 to near linearity. This increase in the bend-back angle
would obviously create a lot more strain within the ring.
Given that a Co2 moiety could not be decoordinated by treat-
ment with dppm, an alternate strategy for the synthesis of a
cycle in which one of the alkynes was uncoordinated was con-
ceived. It has been recently demonstrated that only one alkyne
of a fused diyne (i.e. a diyne in which the two alkyne moieties
are directly bonded to each other) needs to be coordinated to
a Co2 moiety in order for both propargylic centres to be
activated.13 It was therefore hypothesised that treatment
2
=
of [Co (CO) (dppm)(m-Z -HOMe CC CCCCMe OH)] 9 with
=
HBF4 might result in cyclisation. Accordingly 9 was treated
2
4
2
2
Fig. 2 Molecular structure of 6; dppm hydrogens removed for
clarity.
T h i s j o u r n a l i s Q T h e R o y a l S o c i e t y o f C h e m i s t r y a n d t h e
C e n t r e N a t i o n a l d e l a R e c h e r c h e S c i e n t i f i q u e 2 0 0 4
N e w . J . C h e m . , 2 0 0 4 , 2 8 , 5 2 7 – 5 3 4
529

View Article Online
Fig. 4 Space filling diagram of 7 (left) and 5 (right) showing steric clashes between the carbacyclic ring and the dppm groups.
with HBF4 at ꢀ78 ꢁC followed by warming to 0 ꢁC; standard
to 1 and 4 due to the arene spacer. In fact treatment of 11 with
HBF4 did not lead to any cyclisation or dimerisation; instead
work-up then yielded the dimer [{Co2(CO)4dppm}2{m-Z2:
2
2
=
=
=
=
=
=
=
m-Z -HOMe CCCC CCMe CH C( CH )C CCCCMe OH]
[{Co (CO) } {m-Z -1,4-C H (C CC( CH )Me) }] 12 was iso-
=
2 6 2 6 4 2 2
2
2
2
2
2
10 (Scheme 4).
It should be noted that 10 contains a similar bridging
lated as the sole product after standard work-up procedures
(Scheme 5).
=
Me2CCH2C CH2 unit to that in 2 and 5, but now this bridge
The differing reactivity of 11 as compared to 1 and 4 was
further highlighted on treatment of 11 with HBF4 and a range
of sulfur-based nucleophiles HSZSH [Z ¼ (CH2)n (n ¼ 5,6),
C2H4OC2H4 or C2H4SC2H4]. In all cases 11 did react with
the nucleophiles to yield [{Co2(CO)6}2{cyclo-m-Z2-1,4-
is inter- rather than intra-molecular. However, the two uncoor-
dinated propargyl alcohols have not reacted, and cyclisation
has not occurred. It is not clear why cyclisation does not occur
since it has previously been shown that the stabilising effect of
a Co2 unit can be transmitted through an uncoordinated
alkyne bond and thus activation of a remote propargyl alcohol
is possible.
=
=
C H (C CCMe SZSCMe C C)}] (14 to 17) as the sole pro-
=
=
6
4
2
2
duct after standard work up (Scheme 6). In no case did we find
any traces of 12 or of the dimers of compounds 14 to 17, even
though such dimers are among the products of similar
1
Compound 10 has been fully characterised by IR, 13C, H
2
=
NMR and FABm/s spectroscopy. In addition 10 has been
the subject of a single crystal X-ray diffraction study. The
molecular structure of 10 is shown in Fig. 5; relevant bond
lengths and angles are given in Table 2. Interestingly the two
Co2 moieties do not adopt the transoid geometry common in
acyclic diynes, but rather are cisoid. Although it was only pos-
sible to crystallographically locate one of the O–H hydrogen
atoms it seems logical to assume that this cisoid conformation
is due to an intramolecular H bond between the two O–H
reactions involving [{Co (CO) (m-Z -HOH CC C–)} ] and
=
2,15
2
6
2
2
2
=
[{Co (CO) } {m-Z -1,4-C H (C CCH OAc) }].
However,
reaction of 11 with HSC2H4SH did yield a rather unusual pro-
=
2
6
2
6
4
2
2
2
=
duct [{Co (CO) } {m-Z -1,4-C H (C CCMe SC H SSC H S-
=
CMe C C)}] 13 which is derived from substitution of both
2
6
2
6
4
2
2
4
2
4
=
=
2
of the propargylic–OH groups by two molecules of
HSC2H4SH, followed by coupling of the terminal S–H groups.
Such coupling of dithiols to form disulfides is relatively
common.16
groups [O(1)–H(02) 2.107 A]. Regarding the linking C(5)–
Complexes 13 to 17 have been fully characterised by IR, 1H
NMR, 13C NMR and FABm/s spectroscopy. In addition 17
has been the subject of a single crystal X-ray diffraction study.
˚
C(6)–C(9)–C(10) fragment that joins the two diynes it is inter-
esting to note that the formally sp2 hybridised C(10) atom
shows a significant distortion from the idealised bond angles
[e.g. C(12)–C(10)–C(9) 115.6(4)ꢁ]. Likewise, the adjacent C(9)
also shows a significant distortion from its idealised sp3 geome-
try [C(10)–C(9)–C(6) 117.7(4)ꢁ] and the C(9)–C(6) bond length
1
2
1
4
Compound 17 crystallises with equivalent of CHCl3 and
C6H14 within its asymmetric unit; the molecular structure of
17 is shown in Fig. 6 relevant bond angles and lengths are
shown in Table 3. The bond angles and lengths of the
Co2C2 core are all within the expected range. The formation
of a macrocycle does not result in any significant deviation
from the geometry of the parent complex 11; thus the
C(14)–C(13)–C(12) bend-back angle of 141.3(4)ꢁ is identical
˚
[1.504(7)A] is short for a C–C single bond. Within the Co2
cores there are no unusual features and the alkyne bend-back
angles fall within their normal range [C(13)–C(12)–C(10)
144.4(4)ꢁ; C(14)–C(13)–C(12) 141.8(4)ꢁ; C(6)–C(5)–C(4)
139.8(4)ꢁ; C(5)–C(4)–C(3) 142.6(4)ꢁ].
In an attempt to form further cyclophanes via this metho-
dology we synthesised [{Co2(CO)6}2{m-Z2:m-Z2-1,4-C6H4-
14
(C CCMe OH) }] 11. We hoped that this might form large
=
=
2
2
cyclophanes on treatment with HBF4 via a dehydration/dimer-
isation mechanism akin to that leading to the formation of 10;
it seemed unlikely that it would cyclise in an analogous manner
Scheme 4 Formation of 10.
Fig. 5 Molecular structure of 10; dppm protons removed for clarity.
T h i s j o u r n a l i s Q T h e R o y a l S o c i e t y o f C h e m i s t r y a n d t h e
C e n t r e N a t i o n a l d e l a R e c h e r c h e S c i e n t i f i q u e 2 0 0 4
530
N e w . J . C h e m . , 2 0 0 4 , 2 8 , 5 2 7 – 5 3 4

View Article Online
ꢁ
˚
Table 2 Bond lengths (A) and bond angles ( ) for 10
Co1–Co2
Co3–Co4
2.4751(9)
2.4748(8)
1.969
C6–C9
C9–C10
1.504(7)
1.511(7)
1.478(7)
1.367(6)
1.404(6)
1.198(6)
Mean Calkyne–Co1
Mean Calkyne–Co2
Mean Calkyne–Co3
Man Calkyne–Co4
C1–C2
C10–C12
C12–C13
1.967
1.969
1.960
C13–C14
C14–C15
1.473(7)
1.197(6)
1.399(6)
1.363(7)
1.510(7)
C12–C13–C14
C12–C10–C9
C10–C9–C6
C6–C5–C4
C5–C4–C3
141.8(4)
C2–C3
C3–C4
115.6(4)
117.7(4)
139.8(4)
142.6(4)
Scheme 6 Formation of 13 to 17.
C4–C5
C5–C6
chromatography was performed on Kieselgel 60 (70–230 mesh
ASTM). All products are listed in order of decreasing Rf .
Unless otherwise stated, all reagents were obtained from
commercial suppliers and used without further purifi-
within experimental error to the corresponding bend-back
angle in 11.
2
=
cation. 1,4-C H (C CCMe OH) and [{Co (CO) (m-Z -HOMe -
=
CC C–)} ] were prepared by the literature methods.
6
4
2
2
2
6
2
18,19
=
=
2
Regarding the macrocyclic backbone, the oxygen atom is
not in its preferred anti configuration [C(17)–C(16)–O(1)–
C(15) ꢀ155.1ꢁ; C(16)–O(1)–C(15)–C(14) 129.8ꢁ]17 and results
in the lone pairs of the oxygen atom not being orientated
directly into the macrocyclic cavity. As is to be expected this
deviation from the anti geometry of the oxygen atom also
results in the sulfur atoms not being in their preferred gauche
geometries [C(1)–C(18)–S(2)–C(17) ꢀ56.0ꢁ; C(18)–S(2)–C(17)–
C(16) ꢀ89.5ꢁ; C(15)–C(14)–S(1)–C(11) 108.5ꢁ; C(14)–S(1)–
C(11)–C(10) 54.7ꢁ].
Crystal structure determinations
Single-crystal X-ray diffraction data were collected using a
Nonius-Kappa CCD diffractometer, equipped with an Oxford
Cryosystems cryostream and employing MoKa (0.71069 A)
˚
irradiation from a sealed tube X-ray source. Cell refinement,
data collection and data reduction were performed with the
programs DENZO20 and COLLECT21 and multi-scan absorp-
tion corrections were applied to all intensity data with the pro-
gram SORTAV.22 All structures were solved and refined with
the programs SHELXS97 and SHELXL9723 respectively. The
structure of complex 5 shows disorder about a pseudo C2 axis
located approximately between the mid-point of the C(2)–C(3)
bond and the C(5)–C(9) atoms of the carbacyclic ring. This
leads to two positions of the C(2), C(3), C(6) and C(8) atoms
being observed in an approximately 6:4 ratio. The two posi-
tions were refined with the total occupancy of the site summing
to unity. One of the phenyl rings in compound 6 is disordered
over two sites and the n-hexane solvent molecule is disordered
about an inversion centre. The n-hexane and chloroform sol-
vent molecules in compound 7 are disordered about the same
site in a 1:1 ratio.
It is possible that this distortion arises from an electronic
repulsion between the oxygen lone pairs and the p-cloud of
the benzene ring. The separation between the centroid of the
˚
ring and the oxygen atom is 4.1 A.
Conclusion
We have provided further evidence that the reactivity of fused-
diynes is markedly different from that of diynes which contain
spacer units between the two alkyne moieties. Most notably,
the presence of a direct alkyne–alkyne bond in bis-propargyl
alcohols seems to activate the propargylic centres to elimina-
tion rather than to nucleophilic attack. In one case this has
led to the formation of an unprecedented seven-membered cyc-
lic diyne complex containing a high degree of conjugation. We
believe that this result represents an important step forward in
the synthesis of small, conjugated carbocycles.
The bulk powder diffraction experiment PXRD profiles were
measured at room temperature on a STOE STADI-P
high-resolution powder diffractometer using graphite-mono-
˚
chromated Cu Ka radiation (l ¼ 1.5148 A). Relatively high
backgrounds (plots displaced upwards from zero on the y-scale)
arises from fluorescence of Co in Cu Ka radiation, simulated
patterns are from Cerius2 (Accelrys Inc.).
Experimental
A summary of data collection and data refinement details is
given in Table 4.
Unless otherwise stated all experiments were carried out under
an atmosphere of dry, oxygen-free nitrogen, using conven-
tional Schlenk line techniques, and solvents freshly distilled
from the appropriate drying agent. Except where otherwise
indicated NMR spectra were recorded in CDCl3 using a Bru-
ker DRX 400 spectrometer, with TMS as an external standard
for 1H and 13C spectra and H3PO4 as an external standard for
31P NMR spectra. Infrared spectra were, unless otherwise
stated, recorded in dichloromethane solution in 0.5 mm
NaCl solution cells, using a Perkin Elmer 1710 Fourier Trans-
form spectrometer. FAB mass spectra were obtained using a
Kratos MS 890 instrument, using 3-nitrobenzyl alcohol as a
matrix. Preparative TLC was carried out on 1 mm silica
plates prepared at the University of Cambridge. Column
Scheme 5 Dehydration of 11.
Fig. 6 Molecular structure of 17.
T h i s j o u r n a l i s Q T h e R o y a l S o c i e t y o f C h e m i s t r y a n d t h e
C e n t r e N a t i o n a l d e l a R e c h e r c h e S c i e n t i f i q u e 2 0 0 4
N e w . J . C h e m . , 2 0 0 4 , 2 8 , 5 2 7 – 5 3 4
531

View Article Online
(CH3); 31P NMR (THF-d8) d: 38.53; LSIMSm/s: 1417 MNaþ,
ꢁ
˚
Table 3 Bond lengths (A) and bond angles ( ) for 17
1394 MHþ, Mþ ꢀ nCO (n ¼ 1 to 6).
Co1–Co2
Co3–Co4
2.4561(9)
2.4638(8)
1.374(6)
1.347(6)
1.462(6)
1.459(6)
1.964
C2–C1–C18
C1–C2–C3
141.3(4)
143.0(4)
143.9(4)
143.5(4)
108.5
Data for 8
C1–C2
C9–C10
C8–C9–C10
C9–C10–C11
IR nCO (cmꢀ1): 2083.3 vw, 2045.4 vs, 2011.2 vs, 1985.6 vs,
1820.8 s, 1793.0 s; 1H NMR d: 7.25–7.74 (m, 20 H, C6H5),
3.1 (vb,s, 2H, PCH2P); 31P NMR d: 60.93; LSIMSm/s: 693
MNaþ-dppm, 670 Mþ-dppm; Mþ -dppm ꢀ nCO (n ¼ 1 to
6); 1226 Mþ ꢀ 6CO.
C2–C3
C6–C9
C11–S1–C14–C15
C10–C11–S1–C14
C18–S2–C17–C16
C1–C18–S2–C17
C17–C16–O1–C15
C16–O1–C15–C14
54.7
ꢀ89.5
Mean Calkyne–Co1
Mean Calkyne–Co2
Mean Calkyne–Co3
Mean Calkyne–Co4
1.980
1.964
1.990
ꢀ56.0
ꢀ155
129.8
Data for 9
IR nCO (cmꢀ1): 2027.2 s, 2000.9 vs, 1973.1 s, 1952.6 sh; H
NMR d: 7.51–6.98 (m, 20H, C6H5), 3.60 (dd, 1H, JH–P 10
1
2
2
Preparation of [{Co (CO) (dppm)(l-g -HOMe CC C–)} ] 4
=
=
2
4
2
2
2
2
2
Hz, JH–P 13 Hz, PCHHP), 3.23 (dd, 1H, JH–P 11 Hz, JH–P
13 Hz, PCHHP), 2.01(s, 1H, OH), 1.77 (s, 1H, OH), 1.66 (s,
6H, CH3) 1.57 (s, 6H, CH3 ; 13C NMR d: 201.94, 206.40
To a solution of 1 (982 mg, 1.33 mmol) in toluene (200 cm3)
was added dppm (1.093 g, 2.15 eq, 2.85 mmol). The mixture
was stirred for 1 hour at 80 ꢁC and the solvent was then
removed on a rotary evaporator. The residue was dissolved
in dichloromethane and the solution was adsorbed onto silica;
the silica was then pumped dry and added to the top of a chro-
matography column. Elution with hexane:ethyl acetate 7:1
afforded an orange band of [Co2(CO)6(Z1-PPh2CH2PPh2)2]
8 (102 mg). Further elution with hexane:ethyl acetate 2:1
afforded red-orange crystalline 9, which was crystallised from
hexane:ethyl acetate (683 mg, 0.88 mmol, 65.8%). Further elu-
tion with ethyl acetate afforded green crystalline 4 (213 mg,
0.15 mmol, 11.5%).
=
(CO), 137.93 to 128.17 (Ph), 113.8, 93.00 (C C coordinated
=
to Co ), 100.00, 83.4 (free C C), 74.304 (C(OH)Me ), 66.4
(C(OH)Me2), 36.15 (t, JP–C 20.2Hz, PCH2P), 31.67 (CH3);
31P NMR d: 38.81; LSIMSm/s 802 MNaþ, Mþ ꢀ nCO
(n ¼ 1,3,4), Mþ ꢀ 4CO ꢀ nH2O (n ¼ 1,2).
=
=
2
2
1
2
2
Preparation of [{Co2(CO)4L2}2{l-g :l-g -(cyclo-C( CH2)-
=
=
=
=
CH CMe C C–C C–)}] L ¼ 2CO 2; L ¼ dppm 5
=
2
2
2
2
To a solution of 1 or 4 (1.36 mmol) in 200 cm3 dichloro-
methane at ꢀ78 ꢁC was added 0.1 ml of 48% HBF4 in ether.
After warming to room temperature and further stirring for
2 hours an excess of NaHCO3 was added to the mixture.
The resulting mixture was filtered through a plug of MgSO4
and the solvent was carefully removed on a rotary evaporator.
Purification by flash chromatography (hexane:ethyl acetate)
Data for 4
IR nCO (cmꢀ1): 2024.0 w, 1999.4 vs, 1986.0 sh; 1H NMR
(CD3COCD3) d: 7.73–7.05 (m, 40H, C6H5), 5.21 (dd, 2H,
2
2
2
2
2JH–P 12 Hz, JH–H 13 Hz, PCHH), 3.49 (dd, 2H, JH–P 11
yielded [{Co2(CO)6}2{m-Z :m-Z -[cyclo-C( CH2)CH2CMe2-
=
Hz, JH–H 13 Hz, PCHH), 1.74 (s, 12H, CH3); 13C NMR
2
=
C C–C C–]}] 2 (930 mg, 1.32 mmol); [{Co (CO) (dppm)} -
=
{m-Z :m-Z -[cyclo-C( CH )CH CMe C C–C C–]}]
=
=
2
4
2
2
2
=
=
=
2
=
=
(THF-d8) d: 202.31, 206.19 (CO), 137.99–125.57 (Ph), 110.05,
5
(528
2
2
=
97.25 (C C), 73.96 (CMe OH), 33.57 (v.br., PCP), 31.24
mg, 43%).
=
2
Table 4 X-Ray crystallographic data for the new complexes
Complex
2
5
6
7
10
17
Empirical formula
Weight
C22H10Co4O12
702.02
C68H54Co4O8P4
1358.71
C75H70Cl2Co4O8P4
1529.81
Triclinic
C45 H32Co4O10P2
1030.37
C79.5H70Cl4Co4O10P4
1686.7
C34H28Cl1.5Co4O13S2
997.58
Triclinic
Crystal system
Crystal size
Space group
Orthorhombic
0.21 ꢂ 0.16 ꢂ 0.09
P212121
9.369(2)
16.409(2)
16.632(2)
90
Monoclinic
0.23 ꢂ 0.07 ꢂ 0.05
Cc
Monoclinic
0.28 ꢂ 0.28 ꢂ 0.23
C2/c
Monoclinic
0.19 ꢂ 0.12 ꢂ 0.05
C2/c
0.35 ꢂ 0.35 ꢂ 0.23
0.12 ꢂ 0.12 ꢂ 0.10
P1-
¯
P1
˚
a (A)
9.9757(2)
35.2755(8)
17.6562(4)
90
11.3551(1)
13.8087(2)
23.8944(3)
73.855(2)
80.770(2)
82.120(2)
3535.32(7)
2
39.7446(5)
15.4562(2)
16.7757(3)
90
40.5806(5)
11.9801(2)
36.1936(5)
90
12.4650(4)
12.9587(5)
13.8733(4)
69.498(2)
73.291(2)
81.816(2)
2008.22(12)
2
˚
b (A)
˚
c (A)
a(ꢁ)
b(ꢁ)
g(ꢁ)
90
90
106.167(9)
90
91.171(1)
90
118.245(8)
90
3
˚
V/A
2556.9(7)
4
1.824
5967.5(2)
4
1.512
10 303.2(3)
8
1.328
15 500.8(4)
8
1.446
Z
Dc Mg/m3
Abs coefficient mmꢀ1
F(000)
1.437
1.143
1572
1.650
1.887
1003
2.611
1384
1.257
2776
1.376
4160
1.119
6904
y range/ꢁ
Index ranges
3.52 to 27.49
7 ꢃ h ꢃ 12
ꢀ21 ꢃ k ꢃ 13
ꢀ18 ꢃ l ꢃ 21
9560
3.59 to 27.49
ꢀ10 ꢃ h ꢃ 12
ꢀ42 ꢃ k ꢃ 45
ꢀ22 ꢃ ll ꢃ 17
22 460
3.52 to 27.45
ꢀ14 ꢃ h ꢃ 14
ꢀ17 ꢃ k ꢃ 17
ꢀ30 ꢃ ll ꢃ 31
43 652
3.54 to 27.49
ꢀ51 ꢃ h ꢃ 50
ꢀ18 ꢃ k ꢃ 20
ꢀ21 ꢃ ll ꢃ 21
45 396
3.55 to 25.05
ꢀ48 ꢃ h ꢃ 47
ꢀ13 ꢃ k ꢃ 14
ꢀ42 ꢃ ll ꢃ 43
52 975
3.70 to 27.46
ꢀ15 ꢃ h ꢃ 16
ꢀ16 ꢃ k ꢃ 16
ꢀ17 ꢃ ll ꢃ 17
18 407
Reflections
measured
Independent
reflections
5431
10 401
16 067
11 771
13 600
9128
Rint
Goodness of
fit on F2
0.0584
1.071
0.0466
1.131
0.0429
1.019
0.0449
1.048
0.1132
1.052
0.0415
1.046
Final R indices R1
wR2
0.0448
0.0804
0.0602
0.0458
0.0997
0.0595
0.0565
0.1404
0.0798
0.0585
0.1882
0.0842
0.0582
0.1535
0.0822
0.0549
0.1278
0.1009
R indices (all data) R1
wR2
Largest diff peak and hole
0.0872
0.795 and ꢀ0.658
0.2710
0.758 and ꢀ0.515
0.1545
1.653 and ꢀ2.592
0.2091
1.711 and ꢀ0.507
0.1813
1.496 and ꢀ1.385
0.1477
0.950 and ꢀ1.253
T h i s j o u r n a l i s Q T h e R o y a l S o c i e t y o f C h e m i s t r y a n d t h e
C e n t r e N a t i o n a l d e l a R e c h e r c h e S c i e n t i f i q u e 2 0 0 4
532
N e w . J . C h e m . , 2 0 0 4 , 2 8 , 5 2 7 – 5 3 4

View Article Online
2 was also the sole product when the reaction was carried
out in a manner identical to the above procedure but with 1
equivalent of a nucleophile added immediately after the addi-
tion of HBF4 .
was crystallised from hexane:dichloromethane (1.673 g,
1.23 mmol, 93.2%).
Alternative preparation of
[{Co2(CO)4(dppm)}{Co2(CO)6}(cyclo-l-g :l-g -C( CH2)
2
2
=
=
CH CMe C C–C C)] 7
=
2 2
=
Data for 2
=
1
=
H NMR: d: 5.40 (s,br, 1H, C CHH), 5.35 (s,br, 1H,
To a solution of 2 (1.824 g, 2.58 mmol) in toluene (200 cm3)
was added dppm (1.013 g, 2.64 mmol, 1.02 eq.). The mixture
was stirred at 70 ꢁC for 2 hours and allowed to cool down to
room temperature. The solvent removed in vacuo and the resi-
due dissolved in the minimum of dichloromethane and
adsorbed onto silica. The silica was pumped dry and added
to the top of a chromatography column. Elution with hexane
afforded first a green band, which corresponded to the
unreacted starting material (372 mg, 0.53 mmol). Further elu-
tion with hexane:dichloromethane 3:1 afforded the deep green-
brown major product 7 (1.572 g, 1.52 mmol, 74.15% yield
based on reacted starting material) and green 5 (620 mg,
0.46 mmol, 22.44% yield on reacted starting material).
=
C CHH), 2.93 (s, 2H, Me2CCH2), 1.55 (s, 3H. CMeCH3),
1.35 s, 3H. CCH3Me); 13C NMR: d: 198.9 (CO), 116.7
=
=
=
=
=
=
(C CH ), 110.2 (C CH ), 93.4 (br, external C C–C C),
2
2
=
=
=
=
88.5 (br, internal C C–C C), 60.9 (Me2C–CH2), 58.2
(Me2C–CH2), 44.2 (Me), 38.2 (Me); FAB m/s: 702 (Mþ),
Mþ ꢀ nCO (n ¼ 1 to 12); Analytical calculated for
C22H10O12Co4 : C 37.64, H 1.44; Found C 38.01, H 1.62
Data for 5
IR nCO (cmꢀ1): 2029.9 w, 2002.9 vs, 1979.5 s, 1957.0 w; H
1
=
NMR d 7.49–7.00 (m, 40 H, C6H5), 5.29 (s, 2H, C CH2),
3.79 (dd, JH–P 10 Hz, JH–P 13 Hz, 1H PCHHP), 3.59 (dd,
2
2
IR nCO (cmꢀ1): 2084.8 vs, 2068.4 w, 2048.1 vs, 2027.0 vs,
2JH–P 10 Hz, JH–P 12 Hz, 1H, PCHHP),3.39 (dd, JH–P 10
2013.4 vs, 1990.1 sh, 1970.2 s, 1955.8 sh.; H NMR d: 7.30–
2
2
1
2
2
2
=
7.05 (m, 20 H, C6H5), 5.44 (broad s, 2H, C CH2), 3.50 (broad
Hz, JH–P 12 Hz, 1H PCHHP), 3.20 (dd, JH–P 10 Hz, JH–P
12 Hz, 1H PCHHP), 2.37 (s, 2H, CH2), 1.18 (s, 6H, CH3);
13C NMR d: 207.57, 207.35, 205.47, 203.49 (CO), 150.54
s, 1H, PCH2P), 3.31 (broad s, 1H, PCH2P), 2.41 (broad s, 2H,
C(CH3)2CH2C CH2), 1.32 (broad s, 6H, CH3); 13C NMR d:
=
=
=
=
206.02, 202.68, 200.20 (CO), 149.03 (C CH2), 138–128 (aro-
(C CH2), 136–128 (aromatic C of dppm), 119.29 (C CH2),
=
=
matic, dppm), 120.88 (C CH2), 106.52, 100.10, 96.25, 87.61
105.99, 103.90, 102.79, 86.87 (C C), 47.04 (CH ), 39.57
2
=
1
1
=
(C C), 45.53 (CH ), 39.80 (PCH P), 38.14 (CMe ), 32.93
=
2 2 2
(CMe2), 39.03 (t, JP–C 16.1Hz, PCH2P), 37.77 (t, JP–C
16.1Hz, PCH2P), 33.32 (CH3); 31P NMR d:38.44, 34.68; MS
(LSIMS): 1381 MNaþ, 1359 MHþ, Mþ ꢀ nCO (n ¼ 1, 3, 5–
8); Analytical Calculated. for C75H70O8P4Cl2Co4 (as for crys-
tals from DCM: hexane): C, 58.88; H, 4.61; P, 8.10; Found: C,
59.33; H, 4.15; P, 5.13.
(CH3); MS (LSIMS): Mþ ꢀ nCO (n ¼ 3–10); Analytical Calcu-
lated for C45H32O10P2Co4 : C, 52.45; H, 3.13; P, 6.01; Found:
C, 52.36; H, 4.05; P, 3.49.
Preparation of [{Co2(CO)4dppm}2{l-g2:l-g2-
=
=
=
=
HOMe CCC CCMe CH C( CH )C CCCCMe OH] 10
=
2
2
2
2
2
2
2
=
=
2
=
=
Preparation of [{Co (CO) } {l-g :l-g -HOMe C CC CC-
=
( CH2)Me}] 3
To a solution of the complex 9 (495 mg, 0.21 mmol) in 200 m3
dichloromethane at ꢀ78 ꢁC under argon were added three
drops of 48% HBF4 in ether. After slow warming to 0 ꢁC an
excess of NaHCO3 was added to the mixture. The resulting
mixture was filtered through a plug of MgSO4 . The solvent
was removed under reduced pressure and the residue absorbed
on silica and applied to the top of a column. Elution with hex-
ane: ethyl acetate 4:1 produced a first band of the orange-red
crystalline complex 10 (108 mg, 0.07 mmol, 34.3%), and a
second band proved to be the starting material 9 (334 mg).
Complex 10 was crystallised as red-orange crystals by slow
evaporation of a dichloromethane:hexane solution.
2
6 2
To a solution of 1 (740 mg, 1.0 mmol) in dichloromethane (200
cm3) at ꢀ78 ꢁC under argon was added three drops of 48%
HBF4 in ether. The mixture was stirred at ꢀ78 ꢁC for 9 h. After
slow warming to ca. 0 ꢁC during 1 h an excess of NaHCO3 was
added to the mixture. The resulting mixture was filtered
through a plug of MgSO4 and the solvent was carefully
removed on a rotary evaporator. The residue was dissolved
in the minimum amount of dichloromethane and the solution
was applied to the base of TLC plates. Elution with hexane:
dichloromethane 2:1 afforded complex 2 (137 mg, 0.20 mmol,
31.3%), deep green [{Co2(CO)4dppm}2{m-Z2:m-Z2-MeC-
IR nCO (cmꢀ1): 2183.4 vw, 2023.4 s, 2000.2 s, 1968.5 s,
1603.1 w; 1H NMR d: 7.48–7.03 (m, 40 H, C6H5), 6.00 (d,
=
=
=
=
( CH )C CC CMe OH]}] 3 (150 mg, 0.21 mmol, 33.4%),
=
and starting material 1 (280 mg).
2
2
2
2
=
JH–H 1.91 Hz, 1H, C CHH), 5.54 (d, JH–H 1.91 Hz, 1H,
2
C CHH), 3.74 (dd, JH–P 11 Hz, JH–P 13 Hz, 1H, PCHHP)
2
3 IR nCO (cmꢀ1): 2099.7 s, 2079.4 vs, 2059.1 vs, 2032.5 vs,
=
1
2
2
=
2023.1 vs, 1975 sh.; H NMR d: 5.31 (s, 2H, C CH2), 2.18 (s,
3.56 (dd, JH–P 11 Hz, JH–P 13 Hz, 1H, PCHHP), 3.46 (s,
1H, OH), 3.25 (m, 2H, PCH2P) 2.90 (s, 2H, CCH2C), 2.41
(s, 1H, OH, 1.59 (s, 6H, CH3), 1.50 (s, 6H, CH3), 1.43 (s,
6H, CH3); 13C NMR d: 207.20, 202.25 (CO), 147.68
(C CH2), 138.69–129 (m, aromatic C of dppm) 120.97
(C CH ), 101.63, 100.25, 84.91, 84.54 (C C), 66.26 (C–OH),
66.06 (C–OH), 49.34 (CCH2C), 42.03 (CCMe2), 35.18
(PCH2P), 33.73 (PCH2P), 31.45 (CH3), 31.34 (CH3), 30.28
(CH3); 31P NMR d: 38.99 (s, br), 38.85 (s, br); LSIMSm/s
1548.1 MNaþ, 1526.4 MHþ, Mþ ꢀ nCO (n ¼ 1–8): 1497–
1301, 1507 Mþ ꢀ H2O; Analytical Calculated for
C78H69O10P4Co4 : C, 61.39; H, 4.56; P, 8.11; Found: C,
61.70; H, 4.70; P, 5.21%.
3H, CH3), 1.86 (s, 1H, OH), 1.71 (s, 6H, CH3 ,); 13C NMR d:
=
198.9 (CO), 142.4 (C CH2), 116.4 (C CH2), 111.8, 102.9,
=
=
=
92.2, 89.9 (C C), 73.6 (CMe OH), 30.9 (C(CH ) OH), 26.7
3 2
2
þ
þ
=
=
(C( CH2)CH3); FABm/s 742.6 MNa ; 702.66 MH ꢀ H2O;
MNaþ ꢀ nCO (n ¼ 1–3, 5), MHþ ꢀ H2O ꢀ nCO (n ¼ 1,3,4).
=
=
=
2
Alternative preparation of 5
To a solution of 2 (930 mg, 1.32 mmol) in toluene (200 cm3)
was added dppm (1.046 mg, 2.72 mmol, 2.06 eq). The mixture
was stirred for 3 hours at 70 ꢁC, the solvent removed in vacuo
and the residue dissolved in dichloromethane and adsor-
bed onto silica. The silica was pumped dry and added to the
top of a chromatography column. Elution with hexane:
dichloromethane 6:4 afforded green [Co2(CO)4(dppm)(cyclo-
Preparation of [{Co2(CO)6}2{l-g2-1,4-
=
=
C H (C CCMe OH) }] 11
2
2
2
6
4
2
=
=
=
=
m-Z :m-Z -C( CH )CH CMe C C–C C)Co (CO) ] 7 (70 mg).
=
2
2
2
2
6
=
To a solution of 1,4-C H (C CCMe OH) (1 eq., 0.0146 mol)
=
6 4 2 2
Further elution with hexane:dichloromethane 6:4 afforded
complex 5 as a deep dark green crystalline material which
in toluene (100 cm3) and DCM (40 cm3) was added Co2(CO)8
T h i s j o u r n a l i s Q T h e R o y a l S o c i e t y o f C h e m i s t r y a n d t h e
C e n t r e N a t i o n a l d e l a R e c h e r c h e S c i e n t i f i q u e 2 0 0 4
N e w . J . C h e m . , 2 0 0 4 , 2 8 , 5 2 7 – 5 3 4
533

View Article Online
(10 g, 0.029 mol). The reaction mixture was stirred under N2 at
RT for 3 h. The solvent was removed in vacuo and the residue
purified by flash chromatography on silica to yield red
crystalline 11 (10 g, 85%).
Elution afforded red solid 11 (0.040 g, 25%) and dark brown
solid 17 (0.093 g, 50%).
IR nCO (cmꢀ1): 2088.0(m), 2054.8(s), 2028.0(s); 1H NMR d
7.4 (s, 4H, Ph), 3.2 (t3JHH 8Hz, 4H, CH2O), 2.4 (t3JHH 8Hz,
4H, CH2S), 1.7 (s, 12H, CH3); 13C NMR d199.4 (CO), 139.0
IR nCO (cmꢀ1): 2090.0 m, 2054.2 s, 2026 vs; 1H NMR
d:7.30 (s, 4H, C6H4), 1.6 (s, 12H, CH3); 13C NMR d: 199
(C6H4), 129.3 (C6H4), 111.9 (C C), 94.9 (C C), 69.5
=
=
=
=
=
(CO), 131.50, 122.6 (C6H4), 95.4, 81.7 (C C), 65.51
(CH2O), 49.1 (CMe2S), 33.0 (CH3), 30.1 (CH2S); FABm/s
Mþ ꢀ nCO (n ¼ 3–12).
=
(CMe2OH), 31.4 (CH3); LSIMS m/s: 814 Mþ; Mþ ꢀ nCO
(n ¼ 1 to 12).
Acknowledgements
2
Reactions of [{Co (CO) } {l-g -1,4-C H (C CCMe OH) }] 11
=
=
2
6 2
6
4
2
2
We acknowledge the financial support of the EPSRC
(L.J.H.-W, A.D.W and for the purchase of the Nonius Kappa
CCD diffractometer) and of Dstl, Fort Halstead (L.J.H-W).
A.D.W. is grateful to St Catharine’s College, Cambridge, for
the award of a Research Fellowship. We thank the Cambridge
Overseas Trust for financial support to V.B.G. The help of the
EPSRC mass spectrometry service, Swansea is gratefully
acknowledged. Dr John E Davies and Dr Andrew D Bond
are thanked for crystal structure determinations and for the
powder X-ray diffraction studies.
To a solution of 11 (0.17 g, 0.21 mmol) in DCM (50 cm3) was
added 54%wt HBF4ꢄOEt2(0.05 ml, 0.0034 mmol) at ꢀ78 ꢁC
and the appropriate dithiol HSZSH (0.24 mmol, 1.2 eq.).
The resultant mixture was stirred for 10 mins at ꢀ78 ꢁC, then
allowed to warm to RT and stirred for a further 2 hours. The
reaction was quenched with an excess of NaHCO3 and dried
with MgSO4 . The mixture was filtered through a silica plug,
the solvent removed in vacuo, the residue redissolved in the
minimum DCM and applied to the base of TLC plates. The
reaction mixtures were separated as follows:
a) No thiol:
Elution with 3:1 DCM:hexane afforded brown crystalline 12
(0.13 g, 92%).
References
1
IR nCO (cmꢀ1): 2087.6 m, 2055.8 vs, 2026.1 s; H NMR d
1
(a) S. C. Bennet, M. A. Philpis and M. J. Went, J. Chem. Soc.,
Chem. Commun., 1994, 225; (b) H. Amouri, C. Da Silva, B.
=
7.50 (s, 4H, C6H4), 5.50 (s, 4H, C CH2), 2.20 (s, 6H, CH3);
´
´
Malezieux, R. Andres, J. Vaissermann and M. Gruselle, Inorg.
Chem., 2000, 39, 5053; (c) G. G. Melikyan, A. Deravakian, S.
Myer, S. Yadegar, K. I. Hardcastle, J. Ciurash and P. J. Toole,
J. Organomet. Chem., 1999, 578, 68.
13
=
C NMR d199.26 (CO), 141.92 (C CH2), 138.02, 129.56
=
=
=
(C H ), 116.96 (C CH ), 95.47, 92.19 (C C), 23.94 (CH );
MS (LSIMS): Mþ ꢀ2nCO (n ¼ 3–6): Mþ ꢀ nCO ꢀ 2Co,
6
4
3
2
3
4
J. E. Davies, L. J. Hope-Weeks, M. J. Mays and P. R. Raithby,
Chem. Commun., 2000, 1411; L. J. Hope-Weeks, M. J. Mays
and A. D. Woods, J. Chem. Soc., Dalton Trans., 2002, 1812.
(a) R. Guo and J. R. Green, Chem. Commun., 1999, 2503; (b) F.
Diederich and Y. Rubin, Angew. Chem., Int. Ed. Engl., 1992,
31, 1101.
(n ¼ 7–12).
b) Z ¼ C2H4
Elution in 9:1 hexane:ethyl acetate afforded red solid 11
(0.017 g, 10%) and brown solid 13 (0.04 g, 22%).
IR nCO (cmꢀ1): 2088.2(m), 2054.2(s), 2025.9(s); 1H NMR d:
7.5 (s, 4H, C6H4), 3.2 (m, 4H, SCH2), 2.8 (m, 4H, CH2), 1.8 (s,
12H, CH3); 13C NMR d: 199.1 (b, CO), 138.6, 129.6 (C6H4),
P. I. Dosa, C. Erben, V. S. Iyer, K. P. C. Vollhardt and I. M.
Wasser, J. Am. Chem. Soc., 1999, 21, 10 430.
5
6
U. H. F. Bunz and V. Enkelmann, Chem. Eur. J., 1999, 263.
Y.-M. Man, T. C. W. Mak and H. N. C. Wong, J. Org. Chem.,
1990, 55, 3214.
=
=
110.4, 94.2 (C C), 49.5 (Me CS), 32.6 (CH ), 29.6 (CH CH S);
2
3
2
2
FABm/s 966 Mþ; Mþ ꢀ nCO (n ¼ 6, 12).
c) Z ¼ C5H10
7
8
K. M. Nicholas, Acc. Chem. Res., 1987, 20, 207.
R. F. Lockwood and J. M. Nicholas, Tettrahedron Lett., 1977,
4163.
Elution in 8:1 hexane:ethyl acetate afforded red solid 11
(0.031 g, 20%) and brown solid 14 (0.059 g, 31%).
IR nCO (cmꢀ1): 2087.3(m), 2054.2(s), 2025.6(s).
1H NMR d: 7.4 (s, 4H, C6H4), 2.2 (m, 4H CH2S), 1.75 (s,
12H, CH3), 1.15 (m, 4H, CH2), 1.1 (m, 2H, CH2); 13C NMR
=
9
Calculated from a CCDC search . F. H. Allen and O. Kennard,
Chemical Design Automation News, 1993, 8, 31.
10 N. A. Bailey and R. Mason, J. Chem. Soc. A, 1968, 1293.
11 C. J. McAdam, N. W. Duffy, B. H. Robinson and J. Simpson,
Organometallics, 1996, 15, 3935.
d: 199.5 (br, CO), 138.8, 129.4 (C H ), 111.8, 94.5 (C C),
=
6
4
=
=
48.7 (C CCMe S), 33.1 (SCH CH ), 32.8 (CH ) 30.6 (CH ),
2
2
2
3
2
12 Crystallographic details of the oxidised analogue of 8,
[Co2(CO)6(PPh2CH2POPh2)2] can be found in the supplementary
informationy.
28.0 (CH2); FABm/s 915 MHþ, Mþ ꢀ nCO (n ¼ 1–4, 11).
d) Z ¼ C6H12
13 V. B. Golovko, M. J. Mays and A. D. Woods, New J. Chem.,
2002, 1706.
Elution in 4:1 hexane:ethyl acetate afforded brown solid 15
(0.1 g, 50%).
14 Complex 11 has also been crystallographically characterised.
Details are included in the supplementary informationy.
15 R. Gibe and J. R. Green, Chem. Commun., 2002, 1550.
16 T. Chivers, M. Parvez, I. Vargas-Baca and G. Schatte, Can. J.
Chem., 1998, 76, 1093.
17 R. E. Wolf, J. R. Hartman, J. M. E. Storey, B. M. Foxman and
S. R. Cooper, J. Am. Chem. Soc., 1987, 109, 4328.
18 H. Greenfield, H. W. Sternberg, R. A. Friedel, J. H. Wotiz,
R. Markby and I. Wender, J. Am. Chem. Soc., 1956, 78, 120.
19 J. MacBride and K. Wade, Synth. Commun., 1996, 26, 2309.
20 Z. Otwinowski and W. Minor, Methods Enzymol., 1997,
276, 307.
21 R. Hooft, COLLECT, Nonius BV, Delft, The Netherlands, 1998.
22 R. H. Blessing, Acta Crystallogr., 1995, A51, 33.
23 G. M. Sheldrick, SHELXS-97, Program for solution of crystal
structures, University of Go¨ttingen, Germany, 1997; G. M.
Sheldrick, SHELXL-97, Program for refinement of crystal struc-
tures, University of Go¨ttingen, Germany, 1997.
IR nCO (cmꢀ1): 2087.4(m), 2053.4(s), 2025.4(s); 1H NMR d
7.4 (s, 4H, C6H4), 2.7 (m, 8H, CH2S), 2.25 (m 4H, CH2S), 1.8
(s, 12H, CH3); 13C NMR d 200.0 (s, CO), 138.5 (C6H4), 129.6
=
=
=
=
=
=
(C H ), 110.6 (C C), 94.0 (C C), 49.2 (C CCMe S), 38.9
6
4
2
(SCH2), 31.0 (CCH3), 28 (br, CH2); FABm/s 929 MHþ,
Mþ ꢀ nCO (n ¼ 1,3–10,12).
e) Z ¼ (C2H4)2S
Elution afforded red solid 11 (0.075 g, 40%) and dark brown
solid 16 (0.099 g, 53%).
IR nCO (cmꢀ1): 2088.0(m), 2055.0(s), 2027.0(s); 1H NMR d
7.6 (s, 4H, Ph), 3.1 (m, 4H, CH2), 2.4 (m, 4H, CH2S), 1.75 (s,
12H, CH3); 13C NMR d 199.2 (CO), 141.9 (C6H4), 129.3
=
=
=
=
(C H ), 110.9 (C C), 94.1 (C C), 45.1 (CMe S), 33.2 (CH ),
6
4
2
3
31.4 (CH2) 30.1(CH2); FABm/s Mþ ꢀ nCO (n ¼ 4,5,7–11).
f) Z ¼ (C2H4)2O
T h i s j o u r n a l i s Q T h e R o y a l S o c i e t y o f C h e m i s t r y a n d t h e
C e n t r e N a t i o n a l d e l a R e c h e r c h e S c i e n t i f i q u e 2 0 0 4
534
N e w . J . C h e m . , 2 0 0 4 , 2 8 , 5 2 7 – 5 3 4