Mechanistic investigation of the reaction of epoxides with heterocumulenes catalysed by a bimetallic aluminium salen complex

Article abstract of DOI:10.1002/chem.201400007

The bimetallic aluminium(salen) complex [(Al(salen))₂O] is known to catalyse the reaction between epoxides and heterocumulenes (carbon dioxide, carbon disulfide and isocyanates) leading to five-membered ring heterocycles. Despite their apparent

Full text of DOI:10.1002/chem.201400007

DOI: 10.1002/chem.201400007
Full Paper
&
Synthetic Methods
Mechanistic Investigation of the Reaction of Epoxides with
Heterocumulenes Catalysed by a Bimetallic Aluminium Salen
Complex
Christopher Beattie[b] and Michael North*[a, b]
Abstract: The bimetallic aluminium(salen) complex
[(Al(salen))2O] is known to catalyse the reaction between ep-
oxides and heterocumulenes (carbon dioxide, carbon disul-
fide and isocyanates) leading to five-membered ring hetero-
cycles. Despite their apparent similarities, these three reac-
tions have very different mechanistic features, and a kinetic
study of oxazolidinone synthesis combined with previous ki-
netic work on cyclic carbonate and cyclic dithiocarbonate
synthesis showed that all three reactions follow different
rate equations. An NMR study of [Al(salen)]2O and phenyliso-
cyanate provided evidence for an interaction between them,
consistent with the rate equation data. A variable-tempera-
ture kinetics study on all three reactions showed that cyclic
carbonate synthesis had a lower enthalpy of activation and
a more negative entropy of activation than the other two
heterocycle syntheses. The kinetic study was extended to ox-
azolidinone synthesis catalysed by the monometallic com-
plex Al(salen)Cl, and this reaction was found to have a much
less negative entropy of activation than any reaction cata-
lysed by [Al(salen)]2O, a result that can be explained by the
partial dissociation of an oligomeric Al(salen)Cl complex. A
mechanistic rationale for all of the results is presented in
terms of [Al(salen)]2O being able to function as a Lewis acid
and/or a Lewis base, depending upon the susceptibility of
the heterocumulene to reaction with nucleophiles.
Introduction
Bimetallic aluminium(salen) complex 1 was introduced as a cat-
alyst by Jacobsen, who showed that it would catalyse asym-
metric Michael additions.[1] Zhu subsequently reported Passeri-
ni type reactions between aldehydes, isocyanides and hydro-
gen azide catalysed by complex 1,[2] and we demonstrated
asymmetric cyanohydrin synthesis catalysed by a combination
of complex
1
and triphenylphosphane oxide.[3] In recent
papers, we have also reported the use of bimetallic aluminium-
(salen) complex 1 to catalyse the reaction between epoxides 2
and carbon dioxide,[4] carbon disulfide[5] or isocyanates[6] lead-
ing to cyclic carbonates[7] 3, cyclic dithiocarbonates 4,[8] cyclic
Scheme 1. Heterocycle synthesis achieved by use of catalyst 1.
trithiocarbonates[8]
5
and oxazolidinones[9] 6, respectively
(Scheme 1). Each of these studies was accompanied by a mech-
anistic investigation on that particular reaction, which resulted
in the catalytic cycles shown in Scheme 2, 3 and 4 being sug-
gested for the three reactions. It is clear, that despite the ap-
[a] Prof. M. North
Green Chemistry Centre of Excellence
Department of Chemistry, The University of York
Heslington, York, YO10 5DD (UK)
Fax: (+44)01904-322-705
E-mail: Michael.north@york.ac.uk
[b] C. Beattie, Prof. M. North
School of Chemistry, Newcastle University
Newcastle upon Tyne, NE1 7RU (UK)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201400007. It contains details on chemicals
and instrumentations.
Scheme 2. Catalytic cycle for cyclic carbonate synthesis achieved by use of
catalyst 1.
Chem. Eur. J. 2014, 20, 1 – 8
1
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
epoxides with heterocumulenes catalysed by complex 1, and
to provide an overarching mechanistic understanding of all
three processes. In this paper we report the results of this
study.
Results and Discussion
Because the reaction between epoxides and isocyanates cata-
lysed by complex 1 had not previously been studied kinetically,
this was the starting point for this study, to allow the rate
equation to be compared to those for the corresponding reac-
tions with carbon disulfide and carbon dioxide. Our approach
to acquiring kinetic data was to use reaction conditions that
were as close to those used synthetically as possible. There-
fore, the kinetic experiments were carried out at 808C in tolu-
ene using styrene oxide 2a and phenylisocyanate 7 as sub-
strates with 5 mol% catalyst 1 (Scheme 5). The reactions could
Scheme 3. Catalytic cycle for cyclic dithiocarbonate synthesis achieved by
use of catalyst 1.
Scheme 5. Reaction used to study the kinetics of oxazolidinone synthesis.
Scheme 4. Catalytic cycle for oxazolidinone synthesis achieved by use of ca-
talyst 1.
be conveniently monitored by withdrawing samples at regular
1
intervals and analysing them by H NMR spectroscopy to deter-
parent similarities between the three reactions shown in
Scheme 1, there are significant differences between them.
Thus, the reaction of epoxides with carbon dioxide requires
the use of tetrabutylammonium bromide as a cocatalyst; for
the reaction with carbon disulfide, the cocatalyst can be
changed to tributylamine, and no cocatalyst is required in the
reaction with isocyanates. The formation of cyclic carbonates 3
and oxazolidinones 6 was shown to preserve the epoxide ste-
reochemistry, whereas the formation of cyclic di- and trithiocar-
bonates 4/5 was found to proceed with inversion of stereo-
chemistry. Only bimetallic complexes were found to be active
for cyclic carbonate synthesis, but monometallic aluminium-
(salen) complexes were also active catalysts for the reaction
between epoxides and carbon disulfide or isocyanates. The
complex 1 catalysed reactions between epoxides and carbon
dioxide or carbon disulfide also had different kinetic equations
represented by rate Equations (1) and (2).
mine the ratio of epoxide 2a to oxazolidinones 6a,b present.
Compounds 6a,b were formed in a 1:1.9 ratio in these reac-
tions.
Reactions carried out under these conditions showed
a good fit to first-order kinetics, and reactions using three dif-
ferent initial concentrations of styrene oxide 2a showed no
change in the rate of reaction, whereas reactions carried out
using three different initial concentrations of phenyl isocyanate
7 showed that the reaction rate increased as the concentration
of isocyanate 7 increased.[10] The only other species present in
this reaction was catalyst 1, and the order with respect to cata-
lyst 1 was determined by carrying out reactions using 5–
13 mol% of the catalyst. These reactions were carried out in
duplicate,[10] and the resulting plot of log[1] versus log(k1avg) is
shown in Figure 1. The data could be fitted to a straight line
rate ¼ k½1ꢀ½Bu4NBrꢀ2½epoxideꢀ½CO2ꢀ for cyclic carbonate synthesis
ð1Þ
rate ¼ k½1ꢀ½epoxideꢀ for cyclic dithiocarbonate synthesis
ð2Þ
The kinetics of the reaction between epoxides and isocya-
nates catalysed by complex 1 had not previously been report-
ed. Therefore, we have undertaken an integrated kinetic study
to investigate the origin of the differences in the reaction of
Figure 1. Plot of log[1] versus log(k1avg) for the synthesis of oxazolidinones
6a and b. Reactions carried out in toluene at 808C with [2a]0 =0.40m and
[7]0 =0.42m.
&
&
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
2
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
with a slope of 1.1, suggesting that the reaction was first order
in complex 1 concentration, and this was confirmed by a plot
of [1] versus k1avg, which also fitted to a straight line.[10] Thus,
the kinetic equation for the reaction shown in Scheme 5 is
given by Equation (3).
Scheme 6. Adduct formation between complex 1 and phenylisocyanate 7.
rate ¼ k½1ꢀ½isocyanateꢀ
ð3Þ
(Scheme 6) in which the two aluminium ions and hence the
two salen ligands are clearly different. This would be consistent
with the 13C NMR data.
This rate equation is very different to those observed for the
reaction of epoxides with carbon dioxide and carbon disulfide
[Eq. (1) and (2)] and suggests that the initial interaction on the
catalytic cycle is between catalyst 1 and the isocyanate. This
was supported by a 13C NMR spectroscopic study of a mixture
of complex 1 and phenylisocyanate 7, which showed both an
increase in the number of peaks and changes in their chemical
shift, consistent with the formation of a new species with
lower symmetry than complex 1. In contrast, no change in the
spectrum of complex 1 was observed when styrene oxide 2a
was added. Figure 2 shows the aliphatic region of spectra re-
Having determined the kinetic equations for all three reac-
tions, a variable-temperature kinetic study was carried out to
allow the enthalpy and entropy of activation as well as the
Gibbs free energy of activation for each reaction to be deter-
mined. For oxazolidinone synthesis, the variable-temperature
study was carried out in duplicate using epoxide 2a and isocy-
anate 7 with 5 mol% catalyst 1 in toluene at temperatures of
70 to 1108C; the resulting rate data were then used to create
an Eyring plot.[10] A variable-temperature kinetic study of the
reaction between styrene oxide 2a and excess carbon dioxide
to form styrene carbonate 9 catalysed by 5 mol% complex
1 and 5 mol% tetrabutylammonium bromide was carried out
in propylene carbonate at temperatures of 20 to 408C
(Scheme 7). The reactions were monitored by removing sam-
ples for analysis by 1H NMR spectroscopy and the rate data
were used to create an Eyring plot.[10]
Scheme 7. Reaction used to study the kinetics of cyclic carbonate synthesis.
Figure 2. Partial 13C NMR spectra of complex 1 (bottom) and a mixture of
complex 1 and isocyanate 7 (top).
For the reaction between epoxides and carbon disulfide, ep-
oxide 2b was used as substrate because this substrate gives
almost exclusively (>96%) dithiocarbonate 10 as the prod-
uct,[5] so complications due to trithiocarbonate formation were
avoided. In this case, the kinetics of reactions carried out at
35–508C in CDCl3 using 5 mol% complex 1 and tetrabutylam-
monium bromide as catalyst and cocatalyst, respectively, were
corded in CDCl3 with signals corresponding to the cyclohexyl
ring carbon atoms and the tert-butyl methyl groups. The spec-
trum of complex 1 shows two signals above 55 ppm corre-
sponding to the CH-N groups. In the spectrum of the mixture,
these signals are shifted to above 60 ppm, and one of them is
resolved into two separate signals. Similarly, the spectrum of
complex 1 shows two peaks for the tert-butyl groups (either
side of 30 ppm) and in the mixture, one of these is resolved
into two separate signals. The other four peaks in the spec-
trum of complex 1 correspond to the four CH2 groups within
the cyclohexyl ring and, again, when phenylisocyanate is
added, two of these signals are resolved into two peaks.
1
monitored by H NMR spectroscopy (Scheme 8).[10] The result-
In Scheme 4, the adduct between complex 1 and phenyliso-
cyanate 7 is drawn with both aluminium atoms still bound to
the m-oxygen to emphasise the increased Lewis acidity of the
positively charged ion. In such a structure, the two aluminium
ions and their salen ligands would be expected to be identical,
with the positive charge delocalised between them. However,
this adduct can also be drawn as uncharged species 8
Scheme 8. Reaction used to study the kinetics of cyclic dithiocarbonate syn-
thesis.
ing Eyring plot is given in the Supporting Information and the
thermodynamic parameters extracted from all the Eyring plots
are collected in Table 1.
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
3
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
monometallic species. Therefore, a kinetic study of the reaction
between styrene oxide 2a and phenylisocyanate 7 catalysed
by complex 11 was undertaken under the same conditions
used for reactions catalysed by complex 1 (Scheme 5).
Table 1. Activation parameters for reactions catalysed by complex 1 or
11[a,b]
DH°
[kJmolꢁ1
DS°
[Jmolꢁ1 Kꢁ1
DG°
[kJmolꢁ1
]
Catalyst
Heterocumulene
]
]
Reactions in which the initial concentrations of epoxide 2a
and isocyanate 7 were varied again showed that these reac-
tions were first order in isocyanate concentration and zero
order in epoxide concentration.[10] Compounds 6a,b were
again formed in a 1:1.9 ratio in these reactions. Then, reactions
were carried out (in duplicate) at four different concentrations
of catalyst 11, and the resulting plot of log([11]) versus log(k1)
(Figure 3) indicated that the reaction was third order in catalyst
1
1
1
11
PhNCO
CO2
CS2
58.9ꢂ2.0
25.8ꢂ0.5
61.1ꢂ0.6
80.2ꢂ2.3
ꢁ140ꢂ6
ꢁ185ꢂ2
ꢁ123ꢂ2
ꢁ49ꢂ7
97.0ꢂ3.6
76.3ꢂ1.1
94.7ꢂ1.2
93.7ꢂ4.2
PhNCO
[a] Data based on the average of two data sets with the error limits de-
rived from the two individual data sets. [b] DG° at 273 K.
Table 1 shows that the Gibbs free energy of activation for
cyclic carbonate synthesis is approximately 20 kJmolꢁ1 lower
than that for the synthesis of cyclic dithiocarbonates or oxazo-
lidinones. This is consistent with the facts that complex 1 was
originally optimized for cyclic carbonate synthesis and that
cyclic carbonate synthesis catalysed by complex 1 occurs at
208C, whereas the formation of cyclic dithiocarbonates or oxa-
zolidinones requires temperatures of 50–908C. The enthalpies
and entropies of activation for cyclic dithiocarbonate or oxazo-
lidinone synthesis are also very similar. The rate equations for
cyclic dithiocarbonate synthesis [Eq. (2)] and oxazolidinone syn-
thesis [Eq. (3)] both suggest that the first step in both of the
catalytic cycles shown in Scheme 3 and 4 is the rate-determin-
ing step. In both cases, this involves a homogeneous liquid-
phase reaction between complex 1 and a second species, to
form a single activated complex. Thus, both reactions would
be expected to have similar and negative entropies of activa-
tion.
Figure 3. Plot of log[11] versus log(k1avg) for the synthesis of oxazolidinones
6a and b. Reactions carried out in toluene at 808C with [2a]0 =0.40m and
[7]0 =0.42m.
The enthalpies and entropies of activation for cyclic carbon-
ate synthesis are, however, very different to those for cyclic di-
thiocarbonate or oxazolidinone synthesis. The rate equation
[Eq. (1)] suggests that the rate-determining step is much later
in the catalytic cycle because all components of the reaction
appear in the rate equation. This, combined with the fact that
the carbon dioxide starts as a gas rather than in solution, ac-
counts for the much more negative entropy of activation for
this reaction. However, this is more than offset by the much
lower enthalpy of activation for cyclic carbonate synthesis,
which is consistent with a rate-determining step late in the cat-
alytic cycle, after the epoxide has been ring-opened and its
strain energy released.
concentration. This was confirmed by a plot of [11] versus
(k1·[11]ꢁ2), which was also found to be a straight line passing
through the origin (Figure 4). Thus, the rate equation for oxa-
zolidinone synthesis catalysed by complex 11 is represented
by Equation (4).
3
ð4Þ
rate ¼ k½11ꢀ ½isocyanateꢀ
A possible explanation for the third-order kinetics with re-
spect to complex 11 is that complex 11 exists as chloride
bridged oligomers, giving a structure such as 12. There is
ample literature precedent for chloride to bridge between two
aluminium ions[11,12] and for this bridging to result in the for-
mation of oligomers or clusters.[12] To investigate this aspect of
Monometallic aluminium(salen) chloride 11 is also an active
catalyst for the synthesis of oxazolidinones from epoxides and
isocyanates, and is about half as catalytically active as complex
1.[6] Complex 11 is also known to catalyse the reaction be-
tween epoxides and carbon disulfide, displaying similar catalyt-
ic activity to complex 1,[5] but was inactive for the synthesis of
cyclic carbonates from epoxides and carbon dioxide. The latter
two results are easily explained because the catalytic cycle for
cyclic carbonate synthesis (Scheme 2) requires a bimetallic cat-
alyst, whereas the catalytic cycle for cyclic dithiocarbonate syn-
thesis (Scheme 3) requires only a single metal ion. However,
the catalytic cycle proposed for oxazolidinone synthesis
(Scheme 4) involves interconversion between bimetallic and
Figure 4. Plot of [11] versus k1avg[11]ꢁ2 for the synthesis of oxazolidinones
6a and b. Reactions carried out in toluene at 808C with [2a]0 =0.40m and
[7]0 =0.42m.
&
&
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
4
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
the catalysis further, and to allow a direct comparison with the
results obtained using complex 1, a variable-temperature
study of the synthesis of oxazolidinones 6a,b was carried out
to allow an Eyring plot to be constructed;[10] the thermody-
namic parameters extracted from this Eyring plot are included
in Table 1.
initiate the catalytic cycle as shown in Scheme 4. In contrast,
for reactions with heterocumulenes, which are less susceptible
to attack by nucleophiles (carbon dioxide and carbon disul-
fide), the coordinatively unsaturated aluminium ions of com-
plex 1 can act as Lewis acids and coordinate to the epoxide.
For the reaction with carbon disulfide, this initial epoxide coor-
dination appears to be rate determining, with all subsequent
steps in the catalytic cycle shown in Scheme 3 involving very
good nucleophiles or leaving groups. In contrast, for cyclic car-
bonate synthesis, initial coordination of the epoxide to the alu-
minium ion appears not to be the rate-determining step of the
catalytic cycle, resulting in a more complex rate equation.
The entropies of activation determined from variable-tem-
perature kinetic data also support the above hypothesis. The
reactions of epoxides with carbon disulfide and isocyanates
have very similar negative entropies of activation, consistent
with two species coming together to form a single adduct in
the rate-determining transition state. The reaction with carbon
dioxide has a much more negative entropy of activation, con-
sistent with more species having come together by the time
the reaction reaches the rate-determining transition state.
The overall analysis suggests that for future catalyst design,
for reactions involving heterocumulenes that have relatively
low susceptibility to attack by nucleophiles, the Lewis acidity
of the catalyst should be optimized to facilitate its reaction
with and activation of epoxides, whereas for reactions involv-
ing isocyanates and other heterocumulenes that are highly
susceptible to attack by nucleophiles, the Lewis basicity of the
catalyst should be optimized because this interaction deter-
mines the overall rate of reaction. The data suggest that cata-
lyst 1 is significantly more suitable for reaction with carbon di-
oxide than with carbon disulfide or isocyanates because both
the enthalpy and Gibbs free energy of activation for cyclic car-
bonate synthesis are 20–30 kJmolꢁ1 lower than those for cyclic
dithiocarbonate or oxazolidinone synthesis. This analysis will
guide future catalyst development for these and related reac-
tions.
The most striking feature of the activation parameters for
the reaction catalysed by complex 11 is the entropy of activa-
tion, which is only about one third that of the corresponding
reaction catalysed by complex 1. Because both reactions in-
volve the same substrates reacting to give the same products
in the same solvent and at the same temperatures, this differ-
ence in entropies of activation can be accounted for by disso-
ciation of oligomeric species 12 resulting in a large increase in
entropy during the transition state, which largely offsets the
decrease in entropy associated with the aluminium(salen) unit
and isocyanate coming together. In contrast, the enthalpy of
activation for the reaction catalysed by complex 11 is signifi-
cantly higher than that for the reaction catalysed by complex
1. This presumably reflects the fact that no bridging oxygen is
available in complex 11 to bond to the isocyanate during the
transition state. The net result of these differences in enthalpy
and entropy of activation largely cancels out, so that reactions
catalysed by complexes 1 and 11 have the same Gibbs free
energy of activation, within the experimental error.
Conclusion
Complex 1 catalyses the formation of five-membered ring het-
erocycles from epoxides and carbon dioxide, carbon disulfide
and isocyanates. These three reactions follow three different
rate equations [Eq. (1–3)]. The rate equations for the reactions
with carbon disulfide and isocyanates [Eq. (2) and (3), respec-
tively] are both of the form: rate=k[substrate], but differ in
which substrate the rate does (and does not) depend upon.
This suggests that for the reaction with carbon disulfide, the
initial interaction is between complex 1 and the epoxide,
whereas for the reaction between complex 1 and isocyanates,
the initial interaction is between complex 1 and the isocya-
nate. In both cases, this initial interaction appears to be rate
determining. The 13C NMR spectrum of a mixture of complex
1 and phenylisocyanate was consistent with the latter interac-
tion. In contrast, the rate equation for cyclic carbonate synthe-
sis catalysed by complex 1 [Eq. (1)] has a much more complex
form, involving both substrates and the tetrabutylammonium
bromide cocatalyst, as well as complex 1. This suggests that
for this reaction the rate-determining step is later in the cata-
lytic cycle.
Experimental Section
Procedure for Measuring the Kinetics of the Reaction be-
tween Epoxide 2a and Phenylisocyanate 7 Catalysed by
Complex 1
Complex
0.84 mmol) were dissolved in toluene (2 mL) and the solution was
heated to 808C, after which phenylisocyanate (0.1 mL,
1 (0.04–0.11 mmol) and styrene oxide 2a (0.1 mL,
7
0.84 mmol) was added. An aliquot of the reaction mixture was re-
moved and immediately quenched with CDCl3 every 30 min for 4–
1
The above results can be explained on the basis of the rela-
tive susceptibility of the three heterocumulenes to attack by
nucleophiles. Isocyanates are particularly susceptible to attack
by nucleophiles, including oxygen-based nucleophiles.[13]
Carbon dioxide and carbon disulfide are less susceptible to
attack by nucleophiles, although they are known to react with
oxygen- and nitrogen-based nucleophiles.[14,15] The bridging
oxygen of complex 1 is potentially a good, hard nucleo-
phile,[16,17] and for the reaction with isocyanates this appears to
5 h. Each sample was analysed by H NMR spectroscopy to deter-
mine the ratio of epoxide 2a to oxazolidinones 6a,b.
Procedure for Measuring the Kinetics of the Reaction be-
tween Epoxide 2a and CO2 Catalysed by Complex 1
Complex 1 (50 mg, 0.04 mmol) and Bu4NBr (12.9 mg, 0.04 mmol)
were placed in a test tube fitted with a side-arm and dissolved in
propylene carbonate (0.5 mL). A round-bottom flask was filled with
dry-ice pellets and attached to the side-arm. The test tube was
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
5
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
m) C. Beattie, M. North, P. Villuendas, C. Young, J. Org. Chem. 2013, 78,
419–426.
[5] a) M. North, P. Villuendas, Synlett 2010, 623–627; b) W. Clegg, R. W. Har-
rington, M. North, P. Villuendas, J. Org. Chem. 2010, 75, 6201–6207.
[6] T. Baronsky, C. Beattie, R. W. Harrington, R. Irfan, M. North, J. G. Osende,
C. Young, ACS Catal. 2013, 3, 790–797.
sealed with a Suba seal pierced with an empty balloon. The CO2
was allowed to flush the system and fill the balloon. The solution
was then heated (20–408C), and styrene oxide 2a (0.1 mL,
0.84 mmol) was added to the solution. An aliquot of the reaction
mixture was removed and immediately quenched with CDCl3 every
30 min for 4–5 h. Each sample was analysed by 1H NMR spectrosco-
py to determine the ratio of epoxide 2a to cyclic carbonate 9.
[7] For recent reviews of other catalysts for the synthesis of cyclic carbo-
nates from epoxides and carbon dioxide, see: a) M. North, R. Pasquale,
C. Young, Green Chem. 2010, 12, 1514–1539; b) A. Decortes, A. M. Cas-
tilla, A. W. Kleij, Angew. Chem. 2010, 122, 10016–10032; Angew. Chem.
Int. Ed. 2010, 49, 9822–9837; c) P. P. Pescarmona, M. Taherimehr, Catal.
Sci. Technol. 2012, 2, 2169–2187; d) X.-B. Lu, D. J. Darensbourg, Chem.
Soc. Rev. 2012, 41, 1462–1484; e) N. Kielland, C. J. Whiteoak, A. W. Kleij,
Adv. Synth. Catal. 2013, 355, 2115–2138; f) M. North, in New and Future
Developments in Catalysis: Activation of CO2 (Ed.: S. L. Suib), Elsevier,
London, 2013, Chapter 13.
[8] For other catalysts for the formation of cyclic products from epoxides
and carbon disulfide, see: a) J. A. Durden, Jr., H. A. Stansbury, Jr., W. H.
Catlette, J. Am. Chem. Soc. 1960, 82, 3082–3084; b) C. G. Overberger, A.
Drucker, J. Org. Chem. 1964, 29, 360–366; c) M. Kyaw, L. N. Owen, J.
Chem. Soc. 1965, 1298–1305; d) S. Hayashi, M. Furukawa, Y. Fujino, T.
Nakao, K. Nagato, Chem. Pharm. Bull. 1971, 19, 1594–1597; e) G. E.
McCasland, A. B. Zanlungo, L. J. Durham, J. Org. Chem. 1974, 39, 1462–
1466; f) M. V. Jesudason, L. N. Owen, J. Chem. Soc., Perkin Trans. 1 1974,
1443–1446; g) Y. Taguchi, K. Yanagiya, I. Shibuya, Y. Suhara, Bull. Chem.
Soc. Jpn. 1988, 61, 921–925; h) Y. Taguchi, M. Yasumoto, I. Shibuya, Y.
Suhara, Bull. Chem. Soc. Jpn. 1989, 62, 474–478; i) N. Kihara, Y. Nakawa-
ki, T. Endo, J. Org. Chem. 1995, 60, 473–475; j) S. Motokucho, D. Takeu-
chi, F. Sanda, T. Endo, Tetrahedron 2001, 57, 7149–7152; k) Y.-M. Shen,
W.-L. Duan, M. Shi, Eur. J. Org. Chem. 2004, 3080–3089; l) S. Motokucho,
Y. Itagaki, A. Sudo, T. Endo, J. Polym. Sci. Part A 2005, 43, 3711–3717;
m) J.-Y. Wu, Z.-B. Luo, L.-X. Dai, X.-L. Hou, J. Org. Chem. 2008, 73, 9137–
9139; n) R. Maggi, C. Malmassari, Ch. Oro, R. Pela, G. Sartori, L. Soldi,
Synthesis 2008, 53–56; o) I. Yavari, M. Ghazanfarpour-Darjani, Z. Hossai-
ni, M. Sabbaghan, N. Hosseini, Synlett 2008, 889–891; p) A. Z. Halimeh-
jani, F. Ebrahimi, N. Azizi, M. R. Saidi, J. Heterocycl. Chem. 2009, 46, 347–
350.
Procedure for Measuring the Kinetics of the Reaction be-
tween Epoxide 2b and CS2 Catalysed by Complex 1
Complex 1 (50 mg, 0.04 mmol), Bu4NBr (12.9 mg, 0.04 mmol) and
3-phenoxypropylene oxide 2b (0.113 mL, 0.84 mmol) were dis-
solved in CDCl3 (0.37 mL). The solution was added to an NMR tube
followed by addition of carbon disulfide (0.09 mL, 1.5 mmol). The
tube was inserted into the NMR spectrometer with the probe
1
heated (35–508C). A H NMR spectrum was collected every 30 min
for 12.5 h and used to determine the ratio of epoxide 2b to cyclic
dithiocarbonate 10.
Procedure for Measuring the Kinetics of the Reaction be-
tween Epoxide 2a and Phenylisocyanate 7 Catalysed by
Complex 11
Complex 11 (5–13 mol%) and styrene oxide 2a (0.1 mL,
0.84 mmol) were dissolved in toluene (0.5–2 mL) and the solution
was heated to 808C, after which phenylisocyanate 7 (0.1 mL,
0.835 mmol) was added. An aliquot of the reaction mixture was re-
moved and immediately quenched with CDCl3 every 30 min for 4–
1
5 h. Each sample was analysed by H NMR spectroscopy to deter-
mine the ratio of epoxide 2a to oxazolidinones 6a,b.
Acknowledgements
[9] For other catalysts for the formation of oxazolidinones from epoxides
and isocyanates, see: a) G. P. Speranza, W. J. Peppel, J. Org. Chem. 1958,
23, 1922–1924; b) J. E. Herweh, J. Heterocycl. Chem. 1968, 5, 687–690;
c) J. E. Herweh, T. A. Foglia, D. Swern, J. Org. Chem. 1968, 33, 4029–
4033; d) J. E. Herweh, W. J. Kauffman, Tetrahedron Lett. 1971, 12, 809–
812; e) H. Siegel, H. Wittmann, Monatsh. Chem. 1982, 113, 1005–1017;
f) C. Qian, A. Baba, M. Fujiwara, H. Matsuda, Tetrahedron Lett. 1986, 27,
77–80; g) I. Shibata, A. Baba, H. Iwasaki, H. Matsuda, J. Org. Chem.
1986, 51, 2177–2184; h) M. Fujiwara, A. Baba, Y. Tomohisa, H. Matsuda,
Chem. Lett. 1986, 1963–1966; i) M. Fujiwara, A. Baba, H. Matsuda, J. Het-
erocycl. Chem. 1988, 25, 1351–1357; j) M. Fujiwara, A. Baba, H. Matsuda,
Bull. Chem. Soc. Jpn. 1990, 63, 1069–1073; k) A. Baba, K. Seki, H. Matsu-
da, J. Heterocycl. Chem. 1990, 27, 1925–1930; l) K. Yano, N. Amishiro, A.
Baba, H. Matsuda, Bull. Chem. Soc. Jpn. 1991, 64, 2661–2667; m) D. Zhu,
Synlett 1994, 129–130; n) I. Javni, A. Guo, Z. S. Petrovic, J. Am. Oil
Chem. Soc. 2003, 80, 595–600; o) H.-Y. Wu, J.-C. Ding, Y.-K. Liu, J. Indian
Chem. Soc. 2003, 80, 36–37; p) L. Aroua, A. Baklouti, Synth. Commun.
2007, 37, 1935–1942; q) M. T. Barros, A. M. F. Phillips, Tetrahedron:
Asymmetry 2010, 21, 2746–2752; r) X. Zhang, W. Chen, C. Zhao, C. Li, X.
Wu, W. Z. Chen, Synth. Commun. 2010, 40, 3654–3659.
The authors thank EPSRC for a studentship to C. B.
Keywords: aluminium · carbon dioxide · carbon disulfide ·
epoxides · kinetics · ring-opening · reaction mechanisms
[1] a) M. S. Taylor, E. N. Jacobsen, J. Am. Chem. Soc. 2003, 125, 11204–
11205; b) I. T. Raheem, S. N. Goodman, E. N. Jacobsen, J. Am. Chem. Soc.
2004, 126, 706–707; c) E. P. Balskus, E. N. Jacobsen, J. Am. Chem. Soc.
2006, 128, 6810–6812; d) M. S. Taylor, D. N. Zaltan, A. M. Lerchner, E. N.
Jacobsen, J. Am. Chem. Soc. 2005, 127, 1313–1317; e) G. M. Sammis, H.
Danjo, E. N. Jacobsen, J. Am. Chem. Soc. 2004, 126, 9928–9929; f) C. D.
Vanderwal, E. N. Jacobsen, J. Am. Chem. Soc. 2004, 126, 14724–14725;
g) M. Gandelman, E. N. Jacobsen, Angew. Chem. 2005, 117, 2445–2449;
Angew. Chem. Int. Ed. 2005, 44, 2393–2397.
[2] T. Yue, M.-X. Wang, D.-X. Wang, J. Zhu, Angew. Chem. 2008, 120, 9596–
9599; Angew. Chem. Int. Ed. 2008, 47, 9454–9457.
[10] Data given in the Supporting Information.
[3] M. North, C. Williamson, Tetrahedron Lett. 2009, 50, 3249–3252.
[4] a) J. Melꢁndez, M. North, R. Pasquale, Eur. J. Inorg. Chem. 2007, 3323–
3326; b) J. Melꢁndez, M. North, P. Villuendas, Chem. Commun. 2009,
2577–2579; c) M. North, P. Villuendas, C. Young, Chem. Eur. J. 2009, 15,
11454–11457; d) M. North, R. Pasquale, Angew. Chem. 2009, 121, 2990–
2992; Angew. Chem. Int. Ed. 2009, 48, 2946–2948; e) I. S. Metcalfe, M.
North, R. Pasquale, A. Thursfield, Energy Environ. Sci. 2010, 3, 212–215;
f) W. Clegg, R. W. Harrington, M. North, R. Pasquale, Chem. Eur. J. 2010,
16, 6828–6843; g) M. North, C. Young, Catal. Sci. Technol. 2011, 1, 93–
99; h) M. North, C. Young, ChemSusChem 2011, 4, 1685–1693; i) J. Me-
lꢁndez, M. North, P. Villuendas, C. Young, Dalton Trans. 2011, 40, 3885–
3902; j) M. North, B. Wang, C. Young, Energy Environ. Sci. 2011, 4, 4163–
4170; k) M. North, P. Villuendas, ChemCatChem 2012, 4, 789–794; l) M.
North, P. Villuendas, C. Young, Tetrahedron Lett. 2012, 53, 2736–2740;
[11] For examples in which the aluminium ion is also bound to nitrogen or
oxygen ligands, see: a) G. H. Robinson, M. F. Self, S. A. Sangokoya, W. T.
Pennington, J. Am. Chem. Soc. 1989, 111, 1520–1522; b) M. D. Healy,
J. W. Ziller, A. R. Barron, Organometallics 1992, 11, 3041–3049; c) H.-D.
Hausen, J. Tçdtmann, J. Weidlein, J. Organomet. Chem. 1994, 466, C1–
C4; d) A. Fischer, J. Organomet. Chem. 1996, 525, 291–294; e) A. Cotto-
ne III, M. J. Scott, Organometallics 2000, 19, 5254–5256; f) A. Cottone III,
M. J. Scott, Organometallics 2002, 21, 3610–3627; g) J. Scott, S. Gambar-
otta, I. Korobkov, Q. Knijnenburg, B. de Bruin, P. H. M. Budzelaar, J. Am.
Chem. Soc. 2005, 127, 17204–17206; h) I. Vidyaratne, J. Scott, S. Gam-
barotta, R. Duchateau, Organometallics 2007, 26, 3201–3211; i) D.
Zhang, Eur. J. Inorg. Chem. 2007, 3077–3082; j) A. Jabri, C. B. Mason, Y.
Sim, S. Gambarotta, T. J. Burchell, R. Duchateau, Angew. Chem. 2008,
&
&
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
120, 9863–9867; Angew. Chem. Int. Ed. 2008, 47, 9717–9721; k) A. Del
Grosso, R. G. Pritchard, C. A. Muryn, M. J. Ingleson, Organometallics
2010, 29, 241–249; l) A. Hernꢂn-Gꢃmez, A. Martꢄn, M. Mena, C. Santa-
marꢄa, Inorg. Chem. 2010, 49, 8401–8410; m) A. Kraft, J. Beck, I. Kross-
ing, Chem. Eur. J. 2011, 17, 12975–12980.
and Mesoporous Mater. 2006, 90, 314–326; k) A. Dibenedetto, M.
Aresta, P. Giannoccaro, C. Pastore, I. Pꢂpai, G. Schubert, Eur. J. Inorg.
Chem. 2006, 908–913; l) Z. J. Dijkstra, A. R. Doornbos, H. Weyten, J. M.
Ernsting, C. J. Elsevier, J. T. F. Keurentjes, J. Supercrit. Fluids 2007, 41,
109–114; m) J. Alauzun, E. Besson, A. Mehdi, C. Reyꢁ, R. J. P. Corriu,
Chem. Mater. 2008, 20, 503–513; n) F. S. Pereira, E. R. deAzevedo, E. F.
da Silva, T. J. Bonagamba, D. L. da Silva Agostꢄni, A. Magalh¼es, A. E. Job,
E. R. P. Gonzꢂlez, Tetrahedron 2008, 64, 10097–10106; o) L. Phan, J. R.
Andreatta, L. K. Horvey, C. F. Edie, A.-L. Luco, A. Mirchandani, D. J. Dare-
nsbourg, P. G. Jessop, J. Org. Chem. 2008, 73, 127–132; p) R. Vaidhyana-
than, S. S. Iremonger, K. W. Dawson, G. K. H. Shimizu, Chem. Commun.
2009, 5230–5232.
[12] For examples in which chloride bridging results in oligomer or cluster
formation, see: a) W. Uhl, M. Layh, J. Organomet. Chem. 1991, 415, 181–
190; b) J. Vollet, R. Burgert, H. Schnçckel, Angew. Chem. 2005, 117,
7117–7121; Angew. Chem. Int. Ed. 2005, 44, 6956–6960; c) F. Marchetti,
G. Pampaloni, C. Pinzino, J. Organomet. Chem. 2006, 691, 3458–3463.
[13] J. H. Saunders, R. J. Slocombe, Chem. Rev. 1948, 43, 203–218.
[14] For reviews, see: a) S. Schenk, J. Notni, U. Kçhn, K. Wermann, E. Anders,
Dalton Trans. 2006, 4191–4206; b) T. Yu, R. Cristiano, R. G. Weiss, Chem.
Soc. Rev. 2010, 39, 1435–1447; c) Z.-Z. Yang, L.-N. He, J. Gao, A.-H. Liu,
B. Yu, Energy Environ. Sci. 2012, 5, 6602–6639.
[15] a) H. B. Wright, M. B. Moore, J. Am. Chem. Soc. 1948, 70, 3865–3866;
b) Y. Taguchi, K. Yanagiya, I. Shibuya, Y. Suhara, Bull. Chem. Soc. Jpn.
1987, 60, 727–730; c) M. Aresta, D. Ballivet-Tkatchenko, D. B. Dell’Ami-
co, M. C. Bonnet, D. Boschi, F. Calderazzo, R. Faure, L. Labella, F. Marche-
tti, Chem. Commun. 2000, 1099–1100; d) E. M. Hampe, D. M. Rudkevich,
Chem. Commun. 2002, 1450–1451; e) M. George, R. G. Weiss, J. Am.
Chem. Soc. 2001, 123, 10393–10394; f) E. M. Hampe, D. M. Rudkevich,
Tetrahedron 2003, 59, 9619–9625; g) E. R. Pꢁrez, R. H. A. Santos, M. T. P.
Gambardella, L. G. M. de Macedo, U. P. Rodrigues-Filho, J.-C. Launay,
D. W. Franco, J. Org. Chem. 2004, 69, 8005–8011; h) K. Masuda, Y. Ito, M.
Horiguchi, H. Fujita, Tetrahedron 2005, 61, 213–229; i) D. J. Heldebrant,
P. G. Jessop, C. A. Thomas, C. A. Eckert, C. L. Liotta, J. Org. Chem. 2005,
70, 5335–5338; j) R. Srivastava, D. Srinivas, P. Ratnasamy, Microporous
[16] For an aluminium(salen) example, see: G.-P. Wu, W.-M. Ren, Y. Luo, B. Li,
W.-Z. Zhang, X.-B. Lu, J. Am. Chem. Soc. 2012, 134, 5682–5688.
[17] For closely related aluminium complexes, see: a) P. Wei, D. A. Atwood,
Polyhedron 1999, 18, 641–646; b) T. D. Nixon, S. Dalgarno, C. V. Ward,
M. Jiang, M. A. Halcrow, C. Kilner, M. Thornton-Pett, T. P. Kee, C. R. Chim.
2004, 7, 809–821; c) M. Cametti, A. D. Cort, M. Colapietro, G. Portalone,
L. Russo, K. Rissanen, Inorg. Chem. 2007, 46, 9057–9059; d) M. Bouyahyi,
E. Grunova, N. Marquet, E. Kirillov, C. M. Thomas, T. Roisnel, J.-F. Carpent-
ier, Organometallics 2008, 27, 5815–5825; e) A. C. Gledhill, N. E. Cos-
grove, T. D. Nixon, C. A. Kilner, J. Fisher, T. P. Kee, Dalton Trans. 2010, 39,
9472–9475.
Received: February 1, 2014
Published online on && &&, 0000
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
7
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
FULL PAPER
One ring to rule them all: Kinetic stud-
ies on the reaction of epoxides with
carbon dioxide, carbon disulfide and
phenylisocyanate catalysed by
&
Synthetic Methods
C. Beattie, M. North*
&& – &&
[Al(salen)]2O provide an overarching
mechanistic understanding of these re-
actions, which follow three different
rate equations (see scheme). The results
highlight the potential of [Al(salen)]2O
to act as both a Lewis acid and a Lewis
base with the relative importance of
these determined by the heterocumu-
lene.
Mechanistic Investigation of the
Reaction of Epoxides with
Heterocumulenes Catalysed by
a Bimetallic Aluminium Salen Complex
&
&
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
8
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!