The catalytic synthesis of thiacrowns from thiiranes by Group VI and VII transition metal carbonyl complexes

Article abstract of DOI:10.1016/S0022-328X(02)01307-4

The synthesis of thiacrowns by the catalytic ring opening cyclooligomerization of thiirane has been investigated by using some Group VI and Group VII metal carbonyl complexes. The Group VI catalyst precursors were transition metal complexes of the form M(

Full text of DOI:10.1016/S0022-328X(02)01307-4

Journal of Organometallic Chemistry 652 (2002) 51Á59
/
www.elsevier.com/locate/jorganchem
The catalytic synthesis of thiacrowns from thiiranes by Group VI and
VII transition metal carbonyl complexes
Richard D. Adams *, Kellie M. Brosius, O.-Sung Kwon
Department of Chemistry and Biochemistry, The University of South Carolina, Columbia, SC 29208 USA
Received 21 November 2001
Abstract
The synthesis of thiacrowns by the catalytic ring opening cyclooligomerization of thiirane has been investigated by using some
Group VI and Group VII metal carbonyl complexes. The Group VI catalyst precursors were transition metal complexes of the form
M(NCCH3)(PR3)x(CO)5ꢀx (xꢁ
/
0, Mꢁ
/
Cr, Mo and W; xꢁ
/
1, Mꢁ
/
Cr, Rꢁ
/
Ph; Mꢁ
/
W, Rꢁ
/
Ph, or OC6H4Áp-Me). For Group
/
VII, the manganese carbonyl cations: [Mn(NCCH3)(L)(CO)4]ꢂ, Lꢁ
/
CO, PPh3, PMe2Ph, PEt3, PBu3 and P(OMe)3 were studied.
The yields of the thiacrowns were significantly improved by the addition of dialkyl acetylenedicarboxylates to the reactions
catalyzed by the Group VI complexes, but these had no beneficial effect for thiacrown formation for the manganese catalysts. The
principal thiacrown products are 1,4,7,10-tetrathiacyclododecane (12S4), 1,4,7,10,13-pentathiacyclopentadecane (15S5) and
1,4,7,10,13,16-hexathiacyclohexadecane (18S6). Some of the manganese catalysts also yield the cyclic polydisulfides (SCH2CH2S)n
(1Á
/
4) (where nꢁ
/
2Á5), but the principal side product with the manganese catalysts is thiirane homopolymer. The phosphine-
/
substituted catalysts give better yields of thiacrowns than the parent carbonyls in all cases. # 2002 Published by Elsevier Science
B.V.
Keywords: Thiacrown, thiirane; Tungsten carbonyl; Chromium carbonyl; Molybdenum carbonyl; Manganese carbonyl; Catalysis
1. Introduction
We have recently shown that some thiacrowns can be
obtained catalytically in good yields from the strained
four membered heterocycles, thietanes, when they are
treated with rhenium or tungsten carbonyl complexes,
e.g. Eq. (2) [3].
Thiacrowns are an emerging class of polydentate
macrocyclic compounds that have been shown to be
effective ligands for low-valent transition metal ions [1].
Studies of the coordination chemistry of thiacrowns
have progressed slowly because of their lack of avail-
ability, due to the low yields obtained by conventional
thiacrown synthetic procedures [1b,d]. One of the most
effective methods to enhance the yields of thiacrowns is
to add cesium carbonate to the standard coupling
reactions, e.g. Eq. (1) [2].
ð2Þ
It would also be desirable to obtain thiacrowns from
the saturated three membered sulfur heterocycles, thiir-
anes. Thiiranes generally react with uncharged metal
carbonyl complexes by desulfurization to yield olefins
and sulfido metal carbonyl complexes [4]. Cationic
metal complexes, on the other hand, generally produce
ring-opening polymerization of thiiranes [5].
ð1Þ
Our initial studies of the reactions of thiiranes with
W(NCCH3)(CO)5 yielded thiirane complexes of W(CO)5
[6,7]. Abel et al. obtained similar complexes some years
* Corresponding author. Tel.: ꢂ
6781.
/
1-803-777-7187; fax: ꢂ1-803-777-
/
E-mail address: adams@mail.chem.sc.edu (R.D. Adams).
0022-328X/02/$ - see front matter # 2002 Published by Elsevier Science B.V.
PII: S 0 0 2 2 - 3 2 8 X ( 0 2 ) 0 1 3 0 7 - 4

52
R.D. Adams et al. / Journal of Organometallic Chemistry 652 (2002) 51Á59
/
earlier [8]. The reactions of W(NCCH3)(CO)5 with an
excess of thiirane yielded the cyclic polydisulfides
2. Results
The results of our investigation of the catalytic
formation of thiacrowns from thiirane by
(SCH2CH2S)n, 1Á
of ring opening and desulfurization [6,7]. The postulated
mechanism of disulfide formation involves a tungstenÁ
dithietane (SCH2CH2S) intermediate [6,7]. To establish
the existence of this intermediate, MeO2CCÅCCO2Me
(DMAD) was added to the reaction, and the six-
membered heterocycle SCH2CH2SC(CO2Me)C-
/
4 (where nꢁ
/
2Á
/
5) by a combination
W(NCCH3)(CO)5 as a function of an added cocatalyst
are listed in Table 1. The ratio of thiirane to cocatalyst
was 2/1, v/v in all cases. The ratios of the products were
/
/
1
determined by integration of their resonances by H-
NMR spectroscopy. The ratio of the total of amount of
thiacrowns T versus the total amount of cyclic disulfides
(CO2Me) (5) was formed by a trapping reaction.
Another side product that results from using DMAD
is 6, 2,3,4,5,8,9,10-heptamethoxycarbonyl-6-methoxy-
3,6-oxabicyclo[2.2.1]benzothiopyran [7].
D (1Á4) is given in column 2. The principal thiacrown
/
products were 12S4 and 15S5. The amounts of 18S6
were always small. In column 3 the ratio of 12S4 to 15S5
is given. In the last two columns, the turnover number
for the 6-h tests (TON, number of moles of product/
mole of catalyst) and turnover frequency (TOFꢁTON/
/
h) for 12S4 formation are listed. The amounts of 12S4
that were formed were determined by integration of the
resonances by 1H-NMR spectroscopy by using a
predetermined amount of hexamethylbenzene as an
internal reference. It can be seen that even in the absence
of cocatalyst, entry 1, significant quantities of thia-
crowns
are
formed
T/Dꢁ/0.22
with
the
W(NCCH3)(CO)5 catalyst. The ratio of 12S4/15S5 is
close to unity (1.2) and the TON for 12S4 formation is
7.8.
When the reaction is conducted in the presence of
dimethyl acetylenedicarboxylate (DMAD) or diethyl
acetylenedicarboxylate (DEAD), the selectivity for thia-
crown formation increased dramatically. The ratio of
thiacrowns to cyclic disulfides was 6.3 and 4.9, respec-
tively. However, the overall amount of the thiacrowns
was not increased significantly (i.e. the TON for 12S4 in
all three cases is similar). Thus, the co-catalyst is not
increasing the amount of thiacrowns formed, but is
instead decreasing the amounts of disulfides formed.
The decrease in disulfides can be attributed in part to the
catalytic formation of the heterocycle SCH2CH2SC-
(CO2Me)C(CO2Me) (5) as observed in the DMAD
reaction. For all other cocatalysts that were studied,
the amounts of thiacrowns that were formed were
actually significantly less than those of these first three
trials, except possibly for the ethyl propiolate run which
was similar to the run in which no co-catalyst was added
(i.e. ethyl propiolate had no beneficial effect). In the
presence of di-tert-butyl acetylenedicarboxylate
(DTAD) the amounts of thiacrowns actually decreased.
To try to ascertain whether the carboxylate or alkyne
function alone was effective in increasing thiacrown
selectivity, tests were made of several simple carbox-
ylates: ethyl acetate, ethyl propiolate, methyl trifluor-
Unexpectedly, it was found that significant amounts
of the thiacrown macrocycles: 1,4,7,10-tetrathiacyclodo-
decane (12S4), 1,4,7,10,13-pentathiacyclopentadecane
(15S5) and 1,4,7,10,13,16-hexathiacyclohexadecane
(18S6) were formed when the reaction was performed
in the presence of dimethyl acetylenedicarboxylate
DMAD, Eq. (3) [7].
ð3Þ
We have now investigated this reaction as a function
of the type of the ‘cocatalyst’ and also with other Group
VI metal catalysts. In addition, we have discovered a
series of cationic manganese catalysts of the form
[M(NCCH3)(L)(CO)4]ꢂ, Mꢁ
/
Mn, LꢁCO, PPh3,
/
PMe2Ph, PEt3, PBu3, and P(OMe)3, that produce
substantial quantities of the thiacrowns 12S4, 15S5
and 18S6 upon reaction with thiirane [9]. Interestingly,
the cocatalysts that enhance the thiacrown selectivity of
the Group VI catalysts do not enhance thiacrown
formation with the manganese catalysts. These results
are reported here in full and will be discussed in
two sections, first those of the Group VI metal
carbonyl complexes and then those of the manganese
cations.
oacetate, diethyl malonate, (ꢂ)-diethyl-(L)-tartrate,
/
dimethyl terephthalate, as well as some slightly more
complex ones: ethoxycarbonyl isocyanate, ethyl isothio-
cyanatoformate, diethyl succinate, and diethyl maleate.
In no case was there any significant improvement in

R.D. Adams et al. / Journal of Organometallic Chemistry 652 (2002) 51Á
/59
53
Table 1
Comparison of the products formed in the reaction of thiirane and W(NCCH3)(CO)5 with various co-catalysts
a
b
c
Co-catalyst
T/D
12S4/15S5
TON12S4
TOF12S4
None
0.22
6.3
1.2
7.8
5.8
8.3
0.68
1.3
Dimethyl acetylenedicarboxylate (DMAD)
Diethyl acetylenedicarboxylate (DEAD)
di-tert-Butyl acetylenedicarboxylate (DTAD)
3-Hexyne
0.92
1.5
0.97
1.4
4.9
0.52
0.00
0.27
0.36
0.31
0.37
NA
0.11
NA
d
d
d
d
NA
Phenyl acetylene
Ethyl acetate
2.1
1.5
4.3
3.5
2.9
6.2
0.59
0.48
1.0
Ethyl propiolate
Diethyl succinate
d
d
d
d
0.0
0.40
NA
1.1
NA
1.6
NA
0.27
Diethyl malonate
Diethyl maleate
d
d
d
d
0.0
NA
NA
NA
Methyl trifluoroacetate
Dimethyl terephthalate
Ethoxycarbonyl isocyanate
Ethyl isothiocyanatoformate
N-methyl formamide
0.17
0.33
0.29
1.6
1.5
3.0
0.25
0.50
e
e
e
e
0.0
0.0
0.0
0.0
NA
NA
NA
NA
NA
NA
e
e
e
e
e
e
e
e
e
NA
NA
NA
e
e
e
(ꢂ
/
)-Diethyl-(
L
)-tartrate
NA
NA
NA
Ratio of thiirane to co-catalyst is 2/1 v/v. NA, not applicable; TON, moles of 12S4/moles catalyst; TOF, TON/h.
a
T/Dꢁ
12S4/15S5ꢁ
/
ratio of thiacrowns to cyclic disulfides.
15S5.
b
c
/
ratio of 12S4Á
/
Reaction time was 6 h. Analysis by 1H- and 13C-NMR spectroscopy.
Only cyclic disulfides formed.
No cyclic disulfides or thiacrown macrocycles formed.
d
e
thiacrown selectivity compared to the run that contained
no cocatalyst, and in fact, in some of these cases, neither
thiacrowns nor disulfides were formed. To test for the
efficacy of the alkyne function, catalytic runs were
performed with the simple alkynes: 3-hexyne and phenyl
acetylene. Again there was no improvement in selectivity
for thiacrown formation, indeed with 3-hexyne, no
thiacrowns were formed at all. We conclude that it is
the combination of alkyne and carboxylate functions
that make this agent effective, see below.
Next we examined the catalytic activity of different
Group VI carbonyl complexes. DMAD was the only
cocatalyst used in this series of tests. Since it was shown
previously that tungsten hexacarbonyl is inactive [7], our
work was focused only on activated complexes, (i.e.
NCCH3 derivatives). The carbonyl complexes
We also investigated the ability of the cationic
manganese complexes [M(NCCH3)(L)(CO)4]ꢂ, Mꢁ
Mn, LꢁCO, PPh3, PMe2Ph, PEt3, PBu3 and P(OMe)3
/
/
to produce thiacrowns. The PPh3 [14,10] and PMe2Ph
[11] complexes were reported previously. We have made
the others by similar procedures. The previous reports
show that the manganese cation has an approximate
octahedral six-coordinate geometry with the NCCH3
and phosphine ligands in cis-related to one another
[10,11]. To minimize the effects of the counter ion on the
reaction and also to increase the solubility of the
catalyst, the counter ion BPh4ꢀ was selected and used
for all catalysts studied. Interestingly, these salts pro-
duced thiacrowns without the need of DMAD, in fact
when DMAD was added the amounts of thiacrowns
formed were not increased (Eq. (4)).
M(NCCH3)(PR3)x(CO)5ꢀx (xꢁ
W; xꢁ1, MꢁCr, RꢁPh; Mꢁ
OC6H4Áp-Me) were tested. These results are listed in
/
0, Mꢁ/Cr, Mo and
/
/
/
/
W, Rꢁ
/
Ph or
/
Table 2. The chromium and molybdenum complexes
produced no significant amounts of thiacrowns in the
absence of DMAD, but both produced considerable
amounts of thiacrowns with a high selectivity for 12S4 in
the presence of DMAD, see Fig. 1.
ð4Þ
The formation of cyclic polydisulfides was minimal in
the manganese-catalyzed reactions. There was, however,
another major side product that was established by a
combination of elemental analysis and solid state NMR
analysis to be the homopolymer of thiirane. The
polymer existed in two forms one that is soluble in
methylene chloride and one that is insoluble. The
For these catalysts, the DMAD can be regarded as a
true cocatalyst because it actually increased the amounts
of thiacrowns compared to when it was absent. Inter-
estingly, when a phosphine ligand was substituted for a
carbonyl ligand in the tungsten and chromium catalysts,
the selectivity for thiacrown formation increased even
further, see Table 2.

54
R.D. Adams et al. / Journal of Organometallic Chemistry 652 (2002) 51Á59
/
Table 2
Comparison of the products formed in the reaction of thiirane with various metal catalysts
a
b
c
c
Metal catalyst
T/D
12S4/15S5
TON12S4
TOF12S4
W(NCCH3)(CO)5ꢂ
W(NCCH3)(CO)5
W(NCCH3)(PPh3)(CO)4
/
DMAD
6.3
0.22
0.92
1.2
5.8
7.8
0.97
1.3
d
d
e
e
e
e
0.0
11
14
0.0
NA
NA
NA
W(NCCH3)(PPh3)(CO)4ꢂ
/DMAD
1.2
1.1
5.4
4.1
0.91
0.68
W(NCCH3){P(OC6H4-p-Me)3}(CO)4ꢂ
/
DMAD
d
f
f
f
f
Cr(NCCH3)(CO)5
Cr(NCCH3)(PPh3)(CO)4
Cr(NCCH3)(CO)5ꢂDMAD
Cr(NCCH3)(PPh3)(CO)4ꢂDMAD
NA
NA
NA
NA
NA
NA
d
e
e
e
e
0.0
3.7
/
19
26
13
1.9
1.2
/
18
7.2
d
Mo(NCCH3)(CO)5
Mo(NCCH3)(CO)5ꢂ
0.19
8.0
1.8
3.4
0.20
6.8
0.038
1.1
/
DMAD
Ratio of thiirane to co-catalyst is 2/1. NA, not applicable; TON, moles of 12S4/moles catalyst; TOF, TON/h.
a
T/Dꢁ
12S4/15S5ꢁ
/
ratio of thiacrowns to cyclic disulfides.
15S5.
b
c
d
e
f
/
ratio of 12S4Á
/
Reaction time was 6 h. Analysis by 1H- and 13C-NMR spectroscopy.
No DMAD used.
No cyclic disulfides or thiacrown macrocycles were formed.
Only cyclic disulfides were formed.
difference we believe is simply a matter of length the
polymer: the insoluble polymer being a long chain form
and the soluble polymers being shorter chain versions of
the same thing. The results of these tests are listed in
Table 3. As with the Group VI complexes, the activity of
all the phosphine and phosphite derivatives is signifi-
cantly higher than that of the parent carbonyl,
[Mn(NCCH3)(CO)5]ꢂ, but the phosphine complexes
also produced much larger amounts of the polymers.
The selectivity for 12S4 formation is also higher for the
phosphine-substituted derivatives. To demonstrate this,
increased significantly (greater than a factor of 2)
relative to the amounts of polymer that were formed.
3. Discussion
A proposed mechanism for thiacrown macrocycle
formation is shown in the Scheme. The activation of
thiirane occurs via formation of a sulfur-coordinated
metalÁthiirane complex A formed by displacement of
/
the labile NCCH3 ligand from the catalyst precursor. In
the case of the reaction with W(NCCH3)(CO)5 it was
possible to isolate and characterize the thiirane complex
[6,7]. Ring opening via nucleophilic attack by a second
equivalent of thiirane to the intermediate A results in
formation of a thiiranium/thiolato zwitterionic ligand
intermediate such as B [6,7]. This is followed by a series
of ring opening chain growth steps passing through
intermediates C and D that are terminated by a
cyclization step D to E resulting in formation of a
complexed 12S4 thiacrown. This same mechanism can
explain the formation of the 15S5 and 18S6 by extending
the chain growth process by one and two steps,
1
the H-NMR spectra of the product mixtures showing
the 12S4 formation obtained from thiirane reactions
catalyzed by [Mn(NCCH3)(PMe2Ph)(CO)4]ꢂ and
[Mn(NCCH3)(CO)5]ꢂ are compared in Fig. 2. As the
catalyst gets more selective in terms of 12S4 production,
the workup is greatly simplified. Filtration, solvent
extraction, precipitation and chromatography can pro-
vide significant amounts of pure 12S4.
The most active catalyst [Mn(NCCH3)(PPhMe2)-
(CO)4][BPh4] was also tested at a 4 8C, see last entry
in Table 3. As expected the catalytic activity was lower,
but the amounts of the thiacrown products were
Fig. 1. (a) 1H-NMR spectrum of the reaction of thiirane and Cr(NCCH3)(CO)5 in the CH2 region showing the resonances of the cyclic polydisulfides
1Á
4. (b) 1H-NMR spectrum of the reaction of thiirane with Cr(NCCH3)(CO)5 in the presence of DMAD.
/

R.D. Adams et al. / Journal of Organometallic Chemistry 652 (2002) 51Á
/59
55
Table 3
Comparison of the products formed in the reaction of thiirane with various metal catalysts
a
c
Metal catalyst
Ratio of 12S4/15S5/18S6/polymer
Ratio of soluble b/insoluble product (mg)
TON for 12S4
[Mn(NCCH3)(CO)5][BPh4]
20/24/9.9/46
13/2.5/1.5/83
36/5.4/2.5/57
7/14/8/71
104/140
673/393
493/492
2.5
15
[Mn(NCCH3)(PPh3)(CO)4][BPh4]
[Mn(NCCH3)(PPhMe2)(CO)4][BPh4]
[Mn(NCCH3)(PEt3)(CO)4][BPh4]
[Mn(NCCH3)(PBu3)(CO)4][BPh4]
[Mn(NCCH3)(P(OMe)3)(CO)4][BPh4]
[Mn(NCCH3)(PPhMe2)(CO)4][BPh4]
[Mn(NCCH3)(PPhMe2)(CO)4][BPh4]
25
221.1/249.2
222.1/210.6
395/224.8
10.5/241.1
130/42.1
9.6
4.4
6.3
1.18
12.9
13/19/9/51
17/22/5/56
d
f
e
20/1.6/0/3
29/15/5.4/31
TON, moles of 12S4/moles catalyst.
a
From the CH2Cl2 soluble product mixture.
b
c
d
e
f
Soluble or insoluble with respect to methylene chloride.
Reaction time was 6 h. Analysis by 1H- and 13C-NMR spectroscopy.
Includes DMAD as a co-catalyst. Ratio of thiirane to co-catalyst is 2/1.
Ratio of 12S4/15S5/18S6/6; no polymer formation was observed.
Run at 4 8C
respectively. Notably, there was no evidence for forma-
tion of any 1,4,7-trithiacyclononane 9S3 or 1,4-dithia-
cyclohexane 6S2 in these reactions. The formation of
these heterocycles would be inhibited by this mechanism
because the chains of the precursors to these rings are
too short to allow for a cyclization step by a backside
ring-opening addition of the thiolato sulfur atom to the
thiiranium ring. The thiacrowns are released and the
catalytic cycle is closed by substitution of the thiacrown
ligand with another equivalent of thiirane, step E to A.
Polymer formation, as observed for the manganese
catalysts, occurs when the cyclization step fails to occur.
This may be because the positive charge on the
manganese atom reduces the nucleophilicity of the
negatively-charged thiolato sulfur atom which is central
to the cyclization step D to E. Competing elimination of
ethylene from the intermediate B leads to precursor F
and on to the disulfide products [6,7]
interaction such as G formed with the alkyne and the
intermediate B. This should inhibit the ethylene elim-
ination step B to F that leads to the disulfide products.
In the case of the manganese cation catalysts, the
positive charge on the manganese atoms should decrease
the nucleophilicity of the negatively charged thiolato
sulfur atom in the intermediate B. This should lower the
probability of transformation to the disulfide precursors
F. Thus, the chain growth steps become dominant even
in the absence of a co-catalyst.
Our studies show that the most effective cocatalysts
for thiacrown formation with the Group VI catalysts are
molecules that contain a combination of adjacent alkyne
and carboxylate functions, e.g. DMAD and DEAD. We
believe that these molecules probably interact with the
nucleophilic thiolato sulfur atom via a Michael-like
The manganese phosphine complexes are consider-
ably more active than the parent carbonyl. Phosphine
Fig. 2. (a) 1H-NMR spectrum of soluble product mixture from the transformation of thiirane by [Mn(NCCH3)(PMe2Ph)(CO)4]ꢂ in the CH2 region.
(b) 1H-NMR spectrum of soluble product mixture from the transformation of thiirane by [Mn(NCCH3)(CO)5]ꢂ
.

56
R.D. Adams et al. / Journal of Organometallic Chemistry 652 (2002) 51Á
/59
Scheme 1.
ligands are better s-donors and poorer p-acceptors [12]
than carbon monoxide, thus the phosphine ligand
increases the electron density on the transition metal
yield significant quantities of thiacrowns from thiiranes
in the absence of dialkyl acetylene dicarboxylates, and
the addition of dialkyl acetylene dicarboxylates does not
increase their yields. Thiirane homopolymer is a major
side product in the manganese catalyzed reactions. In
both systems the yields of thiacrowns and the selectivity
for 12S4 formation are improved when a carbonyl
ligand is replaced with a phosphine ligand. This is
attributed to a combination of steric and electronic
effects.
atom which should result in a stronger metalÁsulfur
/
bond to the thiirane in the intermediate A (see Scheme
1). Since it is the coordination to the metal that activates
the thiirane, then a stronger metalÁ
/
sulfur bond should
produce greater activation.
The phosphine substitution also produces an increase
in the formation of thiirane polymer in the manganese
complexes. Sterically, phosphines are much larger than
carbon monoxide, and the bulky phosphine might
inhibit the cyclization D to E resulting in a relative
increase in polymer formation, as observed.
5. Experimental
It is known that thiirane polymer can decompose to
produce 12S4 [13]. To rule this out as a mechanism of
formation of the thiacrowns in our experiments, we
treated our thiacrown free polymer with fresh manga-
nese catalyst under the same reaction conditions. No
thiacrowns were formed from this polymer.
5.1. General data
Unless specified otherwise, all reactions were carried
out under an atmosphere of nitrogen. Reagent grade
˚
solvents were stored under 4-A molecular sieves. The
compounds: dimethyl acetylenedicarboxylate (DMAD),
diethyl acetylenedicarboxylate (DEAD), di-tert-butyl
acetylenedicarboxylate (DTAD), 3-hexyne, ethyl pro-
piolate, diethyl succinate, diethyl malonate, diethyl
4. Conclusions
maleate, (ꢂ)-diethyl-(L)-tartrate, dimethyl terephtha-
/
Dialkyl acetylene dicarboxylates can enhance signifi-
cantly the selectivity for the catalytic formation of
thiacrowns from thiiranes by activated Group VI metal
carbonyl complexes when the alkyl groups are not
bulky. The Group VII cationic manganese catalysts
late, ethoxycarbonyl isocyanate, ethyl isothiocyanato-
formate, methyl trifluoroacetate, and phenyl acetylene
were all obtained from Aldrich Chemical Company and
were used as received. Ethyl acetate was obtained from
Fisher Chemical Company and was used as received. N-

R.D. Adams et al. / Journal of Organometallic Chemistry 652 (2002) 51Á
/59
57
methyl formamide was obtained from Mallinckrodt Inc.
and was used without further purification. 1,2-bis(Di-
phenylphosphino)ethane, dimethylphenylphosphine, tri-
butylphosphine, trimethylphosphite and triethyl-
phosphine were obtained from Aldrich Chemical Com-
pany and were used as received. Triphenylphosphine
was obtained from Columbia Organic Chemicals Co.,
Inc. and was used without further purification. Thiirane
was purchased from Aldrich Chemical Company and
was vacuum distilled before use. Trimethylene N-oxide
dihydrate (Aldrich) was dehydrated azeotropically with
benzene and the anhydrous reagent was stored under
nitrogen. Mo(CO)6 was purchased from Climax Mo-
lybdenum Company and was also used without further
5.1.2. Preparation of Cr(NCCH3)(PPh3)(CO)4
Cr(NCCH3)(PPh3)(CO)4 was made by a similar
manner as described above, yield 13%. IR (CH2Cl2
cmꢀ1): 2016 (m), 1901 (s, br), 1855 (m). Anal. Calc.
(Found): C, 61.7 (60.96); H, 3.85 (3.97)%. 1H-NMR
(degassed CD2Cl2)ꢁ
/
1.64d (s, 3H, CH3CN), 7.42d (m,
15H, Ph). 31P (degassed CD2Cl2)ꢁ
/58.06d (s).
5.1.3. Preparation of [Mn(NCCH3)(CO)5][BPh4]
This compound was prepared by taking
[Mn(NCCH3)(CO)5][PF6] [20], dissolving it in ethanol,
adding a solution of ethanol containing an excess of
NaBPh4 and collecting the precipitate.
purification. Cr(CO)6, W(CO)6, P(OC6H4Áp-Me)3 and
/
Mn2(CO)10 were purchased from Strem Chemical Com-
pany and were used without further purification. The
5.1.4. Preparation of
[Mn(NCCH3)(PPh3)(CO)4][PF6]
[Mn(NCCH3)(PPh3)(CO)4][PF6] [11] was prepared in
an analogous manner to the published procedures
[11,14b].
compounds, [MnL2(CO)8] where LꢁPPh3, PMe2Ph,
/
PBu3, P(OMe)3 and PEt3 were made by known methods
[10,11,14]. W(NCCH3)(CO)5 [15], Cr(NCCH3)(CO)5
[15], W(PPh3)(CO)5 [16], W(NCCH3)(PPh3)(CO)4 [17],
Mo(NCCH3)(CO)5 [15], Cr(PPh3)(CO)5 [18,19],
[Mn(NCCH3)(CO)5][PF6] [20], [Mn(NCCH3)(dppe)-
(CO)3][BF4] [21], [Mn(NCCH3)(PPhMe2)(CO)4][BPh4]
5.1.5. Preparation of
[Mn(NCCH3)(PEt3)(CO)4][BPh4]
[11b] and W{P(OC6H4Áp-Me)3}(CO)5 [22] were synthe-
/
[Mn(NCCH3)(PEt3)(CO)4][BPh4] was prepared in an
analogous manner to the published procedures [11,14b]
(4% yield). IR (CH2Cl2 cmꢀ1): 2109 (w), 2040 (m), 2025
sized by known procedures. TLC separations were
˚
performed in air by using silica gel (60 A, F254) on
plates (Analtech, 0.25 and 0.50 mm). Mass spectra were
collected using a VG SE-70 in the direct inlet mode
using electron impact ionization. 1H- and 13C-NMR
spectra were obtained either on a Bruker AM-300 or a
WH-400 spectrometer operating at 300 or 400 MHz,
respectively.
1
(s), 1991 (m). H-NMR (CDCl3) 7.45d (m, 15H, Ph),
1.19d (s, 3H, CH3), 1.57d (s, 2H, CH2) 2.08d (s, 3H,
CH3CN). 31P (CDCl3)ꢁ
1757.9Hz). Anal. Calc. (Found): C, 63.3 (65.2); H,
5.89 (5.47)%.
/
ꢀ
/
143.36d (sextet; JPÁMn
ꢁ
/
5.1.6. Preparation of
[Mn(NCCH3)(PBu3)(CO)4][BPh4]
5.1.1. Preparation of W(NCCH3){P(OC6H4Á
Me)3}(CO)4
A 81.0-mg sample of W{P(OC6H4Á
/
p-
[Mn(NCCH3)(PBu3)(CO)4][BPh4] was prepared in an
analogous manner to the published procedures [11,14b]
(11.5% yield). IR (CH2Cl2 cmꢀ1): 2108 (m), 2040 (m,
/
p-Me)3}(CO)5
(0.120 mmol) was dissolved in 10 ml each of distilled,
degassed acetonitrile and methylene chloride. To this
solution was added 9.8 mg of anhydrous trimethyla-
mine-N-oxide (0.132 mmol) in 10 ml of acetonitrile
dropwise over 15 min. The mixture was stirred at room
temperature (r.t.) for 22 h. The volatiles were removed
in vacuo, and the product was purified via TLC (3:1
hexane:CH2Cl2) and recrystallized by dissolving in a
minimum of methylene chloride and adding hexane until
the product precipitated out as light yellow fluffy
1
sh), 2025 (s), 1996 (m). H-NMR (CDCl3) 7.20d (m,
15H, Ph), 0.86d (m, 3H, CH3), 1.25d (m, 6H, (CH2)3),
2.15d (s, 3H, CH3CN). 31P (CDCl3)ꢁ
30.32d (s). Anal.
Calc. (Found): C, 69.2 (66.6); H, 6.86 (7.48).
/
5.1.7. Preparation of
[Mn(NCCH3){P(OMe3)}(CO)4][BPh4]
[Mn(NCCH3)[P(OMe3)](CO)4][BPh4] was prepared in
an analogous manner to the published procedures
[11,14b] (14% yield). IR (CH2Cl2 cmꢀ1): 2077(w),
crystals (32.1 mg, 39% yield). IR (CH2Cl2 cmꢀ1): 2016
1
(m), 1897 (s), 1849 (m). H-NMR (degassed CDCl3)ꢁ
/
1
1.99d (s, 3H, CH3CN), 3.80d (s, 9H, OCH3) 7.38d,
6.91d (m, 15H, C6H5). 31P (degassed CDCl3)ꢁ
25.76 (s;
with 183W satellites, JWP
297.2Hz). Anal. Calc.
(Found): C, 47.06 (47.24); H, 3.48 (3.36)%.
1944 (s), 1972(sh), 1916(m). H-NMR (CDCl3) 7.42d
(m, 15H, Ph), 3.82d (s, 9H, OCH3), 2.36d (s, 3H,
/
CH3CN). 31Pꢁ
Anal. Calc. (Found): C, 83.0 (80.3); H, 6.71 (6.36)%.
/
ꢀ
/
143.3d (septet; JPÁMn
ꢁ1758 Hz).
/
ꢁ
/

58
R.D. Adams et al. / Journal of Organometallic Chemistry 652 (2002) 51Á59
/
5.2. Catalytic studies
dry and weighed (208.3 mg). This residue was then
dissolved in CH2Cl2 and filtered through a glass frit. The
insoluble portion weighed 120.7 mg. To the CH2Cl2
soluble portion (83.1 mg) was added hexane until a
white precipitate formed. This precipitate was subjected
to further re-crystallizations (CH2Cl2/hexane) until pure
12S4 remained (51.3 mg). Catalytic transformations
with the other Group VI metal complexes were
performed similarly. The results are listed in Table 2.
5.2.1. Catalytic transformations of thiirane by
W(NCCH3)(CO)5 in the presence of MeO2CCÅ
/
CCO2Me
A typical procedure was reported previously [7]. It
was found via integration of the appropriate peaks in
1
the H-NMR spectrum that the 12S4 to 15S5 ratio was
0.91. The ratio of polythioether macrocycles to cyclic
disulfides was 6.3. Catalytic transformations with the
other cocatalysts were performed similarly. The results
are listed in Table 1.
5.2.4. Control test: transformations of thiirane by
Cr(CO)6 in the presence of MeO2CCÅ
/CCO2Me
A 1.9-mg (0.00863 mmol) sample of Cr(CO)6 was
placed in a 10-ml round-bottom flask; then 250 ml each
of thiirane (4.19 mmol) and DMAD (2.09 mmole) were
added. The reaction was covered with aluminum foil
and was stirred at 21.5 8C for 20 h. A 1H-NMR
spectrum at this time showed only unreacted thiirane
and DMAD.
5.2.2. Catalytic transformations of thiirane by
W(NCCH3)(PPh3)(CO)4 in the presence of MeO2CCÅ
CCO2Me
/
A solution of thiirane (500 ml, 8.39 mmol), DMAD
(516 ml, 4.19 mmol), hexamethylbenzene (6.8 mg, 0.0419
mmol) and W(NCCH3)(PPh3)(CO)4 (4.3 mg, 0.00718
mmol) was stirred at 25 8C for 6 h in a 10-ml round-
bottom flask that was covered with aluminum foil. A
1H-NMR spectrum after this time showed 12S4, 15S5,
and lesser amounts of 1, 2, 3 and 5. By using the
hexamethylbenzene as a calibration standard, it was
found via integration that the 12S4 to 15S5 ratio was
5.2.5. Reaction of thiirane and
[Mn(NCCH3)(CO)5][BPh4]
A solution of degassed thiirane (4 ml, 67.2 mmol) and
[Mn(NCCH3)(CO)5][BPh4] (19.9 mg, 0.0358 mmole)
was stirred under nitrogen at r.t. for 2 days (Fig. 2b).
Then, the volatiles were removed in vacuo. The dry
residue weighed 239.3 mg. The residue was extracted
with warm methylene chloride to yield 103.7 mg of
‘soluble’ product after removal of the solvent. 139.6 mg
of insoluble residue remained. The insoluble residue was
shown to be thiirane polymer. The soluble portion was
shown by 1H-NMR and 13C-NMR to be 20% 12S4, 24%
15S5, 10% 18S6 and 46% soluble; a mixture of low
molecular weight polymer, organic sulfide compounds,
traces of 1, 2, 3 and 4. Reactions with the phosphine
derivatives of the manganese complexes were performed
similarly. The results are listed in Table 3.
1.17. T/Dꢁ/11.1. By integration, TON with respect to
12S4 formation (TON12S4)ꢁ
/
4.5; TOF12S4 0.75/h.
ꢁ
/
The unreacted thiirane was recovered via vacuum
distillation and the DMAD was removed by washing the
sticky residue with 8ꢃ2 ml methanol washes. The
/
remaining residue was allowed to dry and was dissolved
in CH2Cl2 and filtered through a glass frit. Hexane was
added to the CH2Cl2 soluble portion was added hexane
until a white precipitate formed. This precipitate was
subjected to further re-crystallizations (CH2Cl2/hexane)
until pure 12S4 remained (37.0 mg). The reactions with
the other catalysts were performed similarly. The results
are given in Table 2.
5.2.6. Reaction of thiirane and
[Mn(NCCH3){P(OMe3)}(CO)4][BPh4]
5.2.3. Catalytic transformations of thiirane by
Mo(NCCH3)(CO)5 in the presence of MeO2CCÅ
/
A solution of degassed thiirane (4.0 ml, 67.2 mmol)
and [Mn(NCCH3){P(OMe3)}(CO)4][BPh4] (20.9 mg,
0.0438 mmol) was stirred at r.t. for 48 h, then the
CCO2Me
Thoroughly degassed thiirane (0.50 ml, 8.39 mmol)
and 500 ml (4.07 mmol) of degassed DMAD were added
to a 5-ml side-arm flask that contained 4.5 mg (0.0163
mmol) Mo(NCCH3)(CO)5. This mixture was covered
1
volatiles were evaporated. A H-NMR spectrum after
this time showed 12S4, 15S5 polymer and trace amounts
of 18S6. The dry crude product weighed 619.8 mg. The
crude product was extracted with several warm methyl-
ene chloride. The methylene chloride soluble portion
weighed 395.0 mg (224.8 mg insoluble) and was found to
be 17% 12S4 (remainder consists of 22% 15S5, 5% 18S6
and 56% polymer/organic sulfides). 178 mg of nearly
pure 12S4 was obtained by passing the CH2Cl2 soluble
portion over a 4-inch column of silica gel and eluting
with 200 ml of CH2Cl2 followed by 200 ml of acetone.
Recrystallization from a CH2Cl2/hexane solvent mixture
gave analytically pure 12S4.
1
with aluminum foil and stirred at 19 8C for 6 h. A H-
NMR spectrum at this time showed 12S4 as the main
product as well as smaller amounts of 1, 4, 5 and 15S5.
By integration of the appropriate peaks in the 1H-NMR
spectrum, it was found that the 12S4 to 15S5 ratio was
4.69. The ratio of polythioether macrocycles to cyclic
disulfides was 9.82. The unreacted thiirane was recov-
ered via vacuum distillation and the DMAD was
removed by washing the sticky residue with 8ꢃ2 ml
methanol washes. The remaining residue was allowed to
/

R.D. Adams et al. / Journal of Organometallic Chemistry 652 (2002) 51Á
/59
59
(g) M. Dele´pine, P. Jaffeux, Compt. Rend. 171 (1920) 36;
Acknowledgements
(h) M. Dele´pine, P. Jaffeux, Bull. Soc. Chim. Fr. 27 (1920) 740;
(i) S. Boileau, P. Sigwalt, Compt. Rend. 252 (1961) 882.
[6] R.D. Adams, J.A. Queisser, J.H. Yamamoto, J. Am. Chem. Soc.
118 (1996) 10674.
This material is based upon work supported by the
National Science Foundation under grant no. 9909017.
Any opinions, findings, and conclusions or recommen-
dations expressed in this material are those of the
author(s) and do not necessarily reflect the views of
the National Science Foundation. This work was
[7] R.D. Adams, J.H. Yamamoto, A. Holmes, B.J. Baker, Organo-
metallics 16 (1997) 1430.
[8] E. Abel, N.A. Cooley, K. Kite, K.G. Orrell, V. Sik, M.B.
Hursthouse, H.M. Dawes, Polyhedron 8 (1989) 887.
[9] R.D. Adams, K.M. Brosius, O.S. Kwon, Inorg. Chem. Commun.
4 (2001) 671.
supported in part by a grant from the USÁIsrael
/
Binational Science Foundation. We thank Professor
James Haw for obtaining the solid state NMR spectra
for our thiirane polymer.
[10] P.J. Harris, S.A.R. Knox, R.J. McKinney, G.A. Stone, J. Chem.
Soc. Dalton Trans. (1978) 1009.
[11] (a) R.H. Reimann, E. Singleton, J. Chem. Soc. Dalton Trans.
(1973) 2658;
(b) R.H. Reimann, E. Singleton, J. Chem. Soc. Dalton Trans.
(1974) 808
References
[12] (a) F.A. Cotton, G. Wilkinson, C.A. Murillo, M. Bochmann,
Advanced Inorganic Chemistry, 6th ed., Wiley, New York 1999,
p. 642;
[1] (a) S.R. Cooper, in: S.R. Cooper (Ed.), Crown Compounds:
Toward Future Applications, Ch. 15, VCH Publishers, New York
1992;
(b) W.W. Porterfield, Inorganic Chemistry: A Unified Approach,
2nd ed., Academic Press, San Diego 1993, p. 620, 653-654.
[13] J.L. Lambert, D. van Ooteghem, E.J. Goethals, J. Poly. Sci. Part
A-1 9 (1971) 3055.
(b) S.R. Cooper, S.C. Rawle, Struct. Bond. 72 (1990) 1;
(c) A.J. Blake, M. Schroder, Adv. Inorg. Chem. 35 (1990) 1;
¨
(d) S.R. Cooper, Acc. Chem. Res. 21 (1988) 141.
[2] R.M. Kellogg, in: S.R. Cooper (Ed.), Crown Compounds:
Toward Future Applications, Ch. 14, VCH Publishers, New
York 1992.
[14] (a) R.A. Jackson, A. Poe, Inorg. Chem. 17 (1978) 997;
¨
(b) A.G. Osborne, M.H.B. Stiddard, J. Chem. Soc. (1964) 634;
(c) H. Masuda, T. Taga, T. Sowa, T. Kawamura, T. Yonezawa,
Inorg. Chim. Acta 101 (1985) 45.
[3] (a) R.D. Adams, Acc. Chem. Res. 33 (2000) 171;
(b) R.D. Adams, S.B. Falloon, Chem. Rev. 95 (1995) 2587;
(c) R.D. Adams, S.B. Falloon, J. Am. Chem. Soc. 116 (1994)
10540;
[15] B.F. Ross, J.G. Grasseli, W.M. Ritchey, H.D. Kaesz, Inorg.
Chem. 2 (1963) 1023.
[16] M.J. Aroney, I.E. Buys, M.S. Davies, T.W. Hambley, J. Chem.
Soc. Dalton Trans. (1994) 2827.
(d) R.D. Adams, S.B. Falloon, Organometallics 14 (1995) 1748.
[4] (a) R.D. Adams, J.E. Babin, M. Tasi, Inorg. Chem. 25 (1986)
4514;
[17] E.O. Fischer, R. Aumann, Chem. Ber. 101 (3) (1968) 963.
[18] G. Schwenzer, M.Y. Darensbourg, D.J. Darensbourg, Inorg.
Chem. 11 (1972) 1967.
(b) R.B. King, Inorg. Chem. 2 (1963) 326.
[5] (a) G. Mengoli, G. Vidotto, J. Electronanal. Chem. Interfacial
Electrochem. 75 (2) (1977) 595; Chem. Abstr. 86 (1977) 90339v;
(b) C.C.J. Culvenor; W. Davies; N.S. Heath, J. Chem. Soc. (1949)
282;
[19] (a) Y. Huang, H.L. Uhm, D.F.R. Gilson, I.S. Butler, Inorg.
Chem. 36 (1997) 435;
(b) H.J. Plastas, J.M. Stewart, S.O. Grim, Inorg. Chem. 12 (1973)
265.
[20] N.G. Connelly, L.F. Dahl, Chem. Commun. (1970) 880.
[21] D. Drew, D.J. Darensbourg, M.Y. Darensbourg, Inorg. Chem. 14
(1975) 1579.
(c) M. Dele´pine, S. Eschenbrenner, Bull. Soc. Chim. Fr. 33 (1923)
703;
(d) O. Masson, J. Chem. Soc. 49 (1986) 234;
(e) K. Endo, H. Kojima, Japanese Patent 21,496 (1963); Chem.
Abstr. 60 (1964) 3122d;
[22] L. Hirsivaara, L. Guerricabeitia, M. Haukka, P. Suomalainen,
R.H. Laitinen, T.A. Pakkanen, J. Puriainen, Inorg. Chim. Acta
307 (2000) 47.
(f) G.M. Bennett, J. Chem. Soc. 121 (1922) 2139;