Interaction of substrate and catalyst during the formation of oxazolidinones from 2-aminoalcohols and diethyl carbonate using recyclable 1,3-dichlorodistannoxanes

Article abstract of DOI:10.1016/j.molcata.2011.01.022

An efficient synthesis of oxazolidinone (OXZ) using 2-aminoalcohols (2AAs) and diethyl carbonate (DEC) as reagents in the presence of recyclable catalyst 1,3-dichloro-1,1,3,3-tetraalkyldistannoxane, [(RR′SnCl)₂O] ₂ (1) is reported. 0.5 mol% (with respect to 2AA) of 1 provides OXZ quantitatively within 1 h at 80 °C with turnover frequency (TOF) of 200 h^-1. The observed TOF is much higher than the reported value (4 h^-1) of the most convenient and commercially feasible K ₂CO₃ catalyst. Chiral 2AAs produce OXZs with 99% ee. Molar dependency of 1, DEC and 2AA is found to be 1:2:2. Molar conductivities (Ω^-1 cm² mol^-1) in DMSO at 25 °C are 6.41 for 1a (R = R′ = Bu), 5.25 for 1b (R = Bu, R′ = Ph), 2.87 for 1c (R = Ph, R′ = Bu), and 2.21 for 1d (R = R′ = Ph) which reveal the mobility of bridged Cl in 1 during reaction. The study of a broad range of substrates and reaction parameters supports a reaction pathway that begins with initial attack by -OH of the pre-formed 2-ethylcarbamato aminoalcohol (2ECA) of 2AA on Sn^b of 1 displacing the bridged Cl. Change in the reaction rates resulted due to various alkyl and aryl substituents on Sn provides better understanding of the distannoxane catalysis, which has not been attempted before for the said reaction.

Full text of DOI:10.1016/j.molcata.2011.01.022

Journal of Molecular Catalysis A: Chemical 338 (2011) 33–43
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Interaction of substrate and catalyst during the formation of oxazolidinones from
2-aminoalcohols and diethyl carbonate using recyclable
1,3-dichlorodistannoxanes
Sharon Pullaa,1, Vineed Unnikrishnanb,1, Punnamchandar Ramidia, Shane Z. Sullivana,
Anindya Ghosha,∗, Jerry L. Dallasd, Pradip Munshic,∗∗
a Department of Chemistry, University of Arkansas at Little Rock, 2801 South University Avenue, Little Rock, AR 72204, USA
b Department of Applied Chemistry, Cochin University of Science and Technology, Cochin, Kerala 682022, India
c Research Centre, Rubamin Laboratories, Dabhasa, Vadodara, Gujarat 391323, India
d NMR Facility, Med Biochem, 4303 Tupper Hall, University of California, Davis, CA 95616, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 2 October 2010
Received in revised form 19 January 2011
Accepted 22 January 2011
Available online 28 January 2011
An efficient synthesis of oxazolidinone (OXZ) using 2-aminoalcohols (2AAs) and diethyl carbonate
(DEC) as reagents in the presence of recyclable catalyst 1,3-dichloro-1,1,3,3-tetraalkyldistannoxane,
[(RRꢀSnCl)2O]2 (1) is reported. 0.5 mol% (with respect to 2AA) of 1 provides OXZ quantitatively within 1 h at
80 ◦C with turnover frequency (TOF) of 200 h−1. The observed TOF is much higher than the reported value
(4 h−1) of the most convenient and commercially feasible K2CO3 catalyst. Chiral 2AAs produce OXZs with
99% ee. Molar dependency of 1, DEC and 2AA is found to be 1:2:2. Molar conductivities (ꢀ−1 cm2 mol−1) in
DMSO at 25 ◦C are 6.41 for 1a (R = Rꢀ = Bu), 5.25 for 1b (R = Bu, Rꢀ = Ph), 2.87 for 1c (R = Ph, Rꢀ = Bu), and 2.21
for 1d (R = Rꢀ = Ph) which reveal the mobility of bridged Cl in 1 during reaction. The study of a broad range
of substrates and reaction parameters supports a reaction pathway that begins with initial attack by –OH
of the pre-formed 2-ethylcarbamato aminoalcohol (2ECA) of 2AA on Snb of 1 displacing the bridged Cl.
Change in the reaction rates resulted due to various alkyl and aryl substituents on Sn provides better
understanding of the distannoxane catalysis, which has not been attempted before for the said reaction.
© 2011 Elsevier B.V. All rights reserved.
Keywords:
Oxazolidinone
Recyclable distannoxane
Diethylcarbonate
2-Aminoalcohols
Catalyst–substrate interaction
1. Introduction
an important class of molecule are a realistic goal of many research
efforts.
Oxazolidinones (OXZs) are a class of cyclic urethanes that have
various important applications. One of the major applications of
OXZs is as chiral auxiliaries [1] for asymmetric transformations.
Utilizing the versatile properties of OXZs, several important phar-
maceutical products [2], polymers [3], and significant organic
molecules can be synthesized [4]. Thus, in terms of organic syn-
thesis, OXZs belong to a very important class of chemicals. In
particular, 3-substituted 2-OXZs have been reported to be use-
ful as synthetic reagents, inhibitors, and additives, demonstrating
antibacterial and fungicidal activity [5]. Thus, synthetic methods
having high yields and reduction in the production costs for such
The preparation of OXZs is well-documented in the literature
using various starting materials, catalysts, and reaction conditions.
For example, oxidative carbonylation is one of the commonly prac-
ticed methods known to produce OXZs with CO and oxygen as
starting materials on Pd- and Cu-based catalyst systems with high
efficiency [6]. However, costly palladium cannot provide the eco-
nomic viability for the synthesis, and copper has considerable
toxicity. OXZs have also been synthesized using carbon dioxide as
the starting material either in catalytic [7] or non-catalytic path-
ways [8]. But inferior reaction yields and undesirable byproduct
formation make the processes less interesting. Propargylamine and
CO2 produce corresponding OXZs with greater ease and with high
yields in the presence of Pd [9] or Ru [10] and a heterogeneous cat-
alyst [11]. Still, the acetylenic group holds the key for the success
of the reaction and is thus applicable to only a narrow range of sub-
strates. Use of costly palladium or ruthenium at catalyst loading of
5 mol% is another disadvantage of the process. A tosyl derivative
of OXZ can be obtained in high yields from serine using a Grig-
nard/CuX catalyst system [12]. A four-step synthetic method has
been reported by Green and co-workers for a tyrosine-based OXZ
∗
Corresponding author.
Corresponding author. Present address: Research Centre, Reliance Technology
∗∗
Group, Reliance Industries Limited, Vadodara, Gujarat 391346, India.
Tel.: +91 265 669 6049; fax: +91 265 669 63934/3937.
E-mail addresses: axghosh@ualr.edu (A. Ghosh), pradip.munshi@ril.com,
pradip.munshi@gmail.com (P. Munshi).
1
These authors contributed equal amount of work and can be indicated as ‘Unnikr-
ishnan and Pulla et al.’.
1381-1169/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2011.01.022

34
S. Pulla et al. / Journal of Molecular Catalysis A: Chemical 338 (2011) 33–43
Cla
polyurethane [27] synthesis, ring opening polymerization [28],
R
esterification and transesterification [29] reactions, etc. They (1)
show unique advantages in terms of reaction rate, product and
catalyst separation, and high reaction yields when compared to
the conventional Lewis acid-base catalysts like metal alkoxides
or trihalides [29]. We have also demonstrated an efficient green
process of transesterification reaction of polyhydric alcohol by
varying lipophilicity of 1 [30]. Very recently, for the first time,
we have established a synthetic method of producing biodiesel
from triglycerides using 1 in supercritical CO2 [31]. One of the
major features of 1 is that it possesses multi-active catalytic cen-
ters with controllable Lewis acidity, offering several advantages
over other catalysts [29,32]. This unique feature and versatility of 1
have encouraged us to study several unexplored areas such as OXZs
synthesis.
Sna
O
R
R/
R'
Snb
Clb
Clb
Snb
//
R'
NNamame
1a
R =
Bu
Bu
Ph
R =
R'
R
R
Bu
Ph
Bu
Ph
O
Sna
Cla
1b
1c
1d
Ph
Dimeric structure of 1
Scheme 1. OXZ formation from 2AA and DEC using 1.
that can be used as a chiral auxiliary [13]. Selective rearrangement
of certain carbamates also generates OXZ [14]. The rearrangement
of specific propargylic tert-butylcarbamates, initiated by an effi-
cient gold catalyst, has been reported for the synthesis of OXZ [15].
N-protected alkynylamines can be converted into alkylidene 2-OXZ
under mild reaction conditions [16]. Cycloaddition of aziridines
with isocyanates produces imino-OXZ [17], but the specific nature
of the substrate and expensive catalyst system restricts the proce-
dure from being widely practiced. Use of microwave irradiation
as an energy efficient and “green” pathway to synthesize OXZ-
2-ones starting with urea and ethanolamine and assisted by a
catalytic amount of nitromethane has also been reported [18].
Trichloromethylchloroformate, a very reactive reagent, has been
found to form OXZ, imidazolidinone and dioxolanone catalyzed
by activated charcoal [19]. However, use of microwave irradiation
and toxic trichloromethylformate severely limits these processes.
Therefore, from the above discussion, it is evident that the synthe-
sis of OXZs is specific to a substrate and catalyst, and lacks a broader
scope.
In this report, we have studied a series of 2AAs, both chiral and
achiral, in order to understand the structure-reactivity relationship
(Scheme 1). Interestingly, recovered catalyst (1) shows efficiency in
producing OXZ similar to that of the original one. Additionally, the
reaction kinetics and the possible mechanism of the catalytic path-
way have been discussed. The present work also addresses for the
first time the effect of substituents on Sn to understand the mech-
anistic route of OXZ formation using 1. The proposed mechanistic
route is different compared to other literature reported distannox-
ane catalyzed [27,29,33] reactions where mostly alkyl and bridging
groups of the catalyst are varied. As diethyl carbonate, a congener
of carbon dioxide, is considered as a green solvent/reagent, the
present findings will also contribute to the field of sustainable
development for OXZ synthesis.
2. Experimental
2.1. General
2-Aminoalcohols (2AAs), which can be derived from their cor-
responding amino acids, are very useful substrates in the context
of OXZ synthesis. They provide a much wider scope, and thus they
are the focus of the present work. The synthesis involves cycliliza-
tion of 2AAs with organic carbonates or equivalent molecules in
the presence of catalysts. The catalyst can be alkali-based or an
equivalent, Lewis acidic, or transition metal-based [20]. Alkoxide
catalysts require drastic reaction conditions that may damage the
chirality, if present, of the final product [21] and may also produce
other undesirable byproducts. Diethyl carbonate (DEC) in the pres-
ence of K2CO3 as a catalyst can produce the product at satisfactory
yields [22]. It is noteworthy that 2AAs can also be converted to
OXZ using alkali catalyst at the expense of cyclic carbonate [23] by
transesterification method. However, use of DMF as solvent, long
reaction time and moderate yields are major drawbacks for this
process [24].
Considering all of the above-mentioned concerns, a simple,
economically viable catalytic method that can be used for syn-
thesizing a broad range of OXZs in good yields is still a major
challenge. We have tried to address these concerns by using
1,1,3,3-tetraalkyl/aryl-1,3-dichloro distannoxane (1, 0.5 mol%) as
an efficient, robust, and recyclable catalyst that can generate OXZ
from the corresponding 2-aminoalcohols (2AA) in the presence of
DEC in short time (1 h), under mild reaction conditions (80 ◦C) and
in the absence of any additional solvent (Scheme 1). Turn over fre-
quency (TOF) of the process is as high as 200 h−1 and the purity of
the product is more than 98%. This is a significant improvement in
comparison to one of the most efficient synthetic methods of OXZ
production where TOF of 4 h−1 is achieved using K2CO3 as catalyst
[20].
Chemicals are purchased from Aldrich Chemical Company, USA
and used without further purification unless otherwise specified.
Solvents are purified by standard purification procedures before
use in the reactions [34]. Reaction products are analyzed by Gas
Chromatography (Shimadzu, GC 2010) using a DB-5 column (J&W
Scientific). 1H NMR spectra are obtained with a 300 MHz Varian
FT spectrometer using deuterated solvent as the lock. The spectra
are collected at 25 ◦C and chemical shifts (ı, ppm) are referenced
to residual solvent peak (CDCl3 ı, 1H, 7.26 ppm). 119Sn NMR has
been recorded using a Bruker AV500 in CDCl3. Electrospray ioniza-
tion mass spectra (ESI-MS) are recorded using a Micromass Q-TOF
mass spectrometer. Infra Red (IR) spectra are recorded using a
Perkin Elmer, Spectrum 100 instrument. The elemental analyses (C,
H) are carried out with a Perkin-Elmer 2400 C elemental analyzer
(accuracy 0.3%). Amount of tin was estimated using inductively
coupled plasma optical emission spectroscopy (ICP-OES), Perkin
Elmer, model 4300 DV (detection limit 1 ppm). Chloride is esti-
mated using an Ion Chromatography (Metrohm) method (detection
limit 1 ppb). Enantiomeric excess is measured in Shimadzu LC 2010
HPLC using Daicel’s CHIRALPAK®AD-H column.
Catalysts are synthesized according to an established procedure
[30]. Chlorine and Sn are determined according to published meth-
ods [35].
2.2. Preparation of oxazolidinone
In a typical reaction, 257 g (2.18 mol) of DEC, 73.3 g (0.49 mol)
(S)-phenylalaninol and 1.2 g (0.001 mol) catalyst 1a are combined
in a three-necked round bottom flask equipped with a 12ꢀꢀ Vigreux
condenser and heated at 80 ◦C for 1 h. The initial heterogeneous
solution slowly becomes homogenous within 10 min. The progress
Catalysts 1 have been used in a number of places in organic
syntheses: namely, in polyester synthesis [25], urethane [26] and

S. Pulla et al. / Journal of Molecular Catalysis A: Chemical 338 (2011) 33–43
35
of the reaction is monitored by a Shimadzu 2010 gas chromato-
graph equipped with a DB-5 column while comparing the reaction
mixture with the authentic sample. After 1 h, the flask is cooled
to room temperature. The excess of DEC and ethanol produced
are distilled. The resultant concentrated mass is diluted with
dichloromethane and analyzed using GC. During the cooling of
the reaction mixture, the catalyst also settles down at the bottom
of the flask, which is easily collected by filtration. The recov-
ered catalyst and DEC are used in the next reaction to test for
recyclability. The procedure also works for 2-aminophenol and 2-
aminophenylethanol.
Other OXZs are synthesized following the similar procedure
except the purification method, which is performed according to
the nature of substrates and products. For example, in case of the
substrates 2-aminobutanol and ethanolamine 1a does not separate
from the reaction mixtures. So at the end of the reaction, mixture
is loaded on a 6ꢀꢀ silica gel column for the separation of the cata-
lyst. OXZ and the catalyst are then eluted with ethylacetate–hexane
mixture with 3:1 and 1:1 ratio, respectively.
Fig. 1. GC chromatogram of (S)-2-benzyloxazolidinone. Phenylalaninol 0.25 mmol,
DEC 1.08 mol, 1a 1.0 mmol, 80 ◦C, 1 h.
2.5. Determination of rate constants (k)
The recovered catalyst is dried under vacuum oven for 4 h at
80 ◦C and characterized by 1H NMR and 119Sn, ESI-MS and the
elemental analysis. The recovered catalyst is used for further exper-
iments. Generally solvent extraction methods are employed for
catalyst recovery. In cases where solvent recovery cannot be used
due to solubility issue, separation using silica-gel column chro-
matography is pursued.
Rate constants are measured for various oxazolidinone forma-
tions as shown by Eq. (1). As observed the reaction is considered
first order with respect to 2AA. Concentration of DEC is kept high
enough in order to maintain a pseudo first order nature of the reac-
tion. For each reaction following the above-mentioned procedure,
concentration of the components including the product varied dur-
ing the course of the reaction is measured by GC. Variation of 2AA
are noted in 10 min time interval and plotted against time following
the equation
2.3. N-(2-hydroxyethyl)ethylcarbamate
[2AA] = [2AA]0 − kt
(1)
Ethanolamine (100 mg, 1.64 mmol), and DEC (871 mg,
7.37 mmol) are taken in a round bottom flask and heated at
80 ◦C for 1 h. The reaction is monitored by GC where no formation
of oxazolidinone is noticed at this reaction condition. Formation of
carbamate is confirmed by indentifying the corresponding m/z (M+
132) of the desired product in the GC–MS. The reaction mixture
is concentrated under reduced pressure and purified by silica
gel column chromatography. N-ethylcarbamatoethanolamine is
eluted with 1:5 ethyl acetate–hexane mixture. The solvent eluted
from the column is concentrated at the reduced pressure to get
the desired product. Yield 65%. 1H NMR; ı 1.24 (3H, t, J = 7.1), 4.23
(2H, q, J = 7.1), 3.49 (2H, t, J = 6.6), 3.48 (2H, t, J = 6.6).
where [2AA] denotes concentration of 2AA at time t while [2AA]0
denotes the initial concentration of 2AA and k denotes the rate
constant. Finally k is obtained from the slope of the plot.
2.6. Estimation of Cl− and Sn
A representative method for the estimation of Cl− and Sn during
4-benzyl-2-oxazolidinone synthesis using catalyst 1a is described
here. 150 mg (0.99 mmol) of 4-benzyl-2-oxazolidinone is digested
with equivalent amount of concentrated HNO3 (69%) in a porcelain
crucible for 5 h. The resultant digested mass is then evaporated to
dryness. The solid obtained after evaporation is dissolved in 10 mL
water and acidified with dilute HNO3 (6.9%) until solution turned
clear. After filtering the solution, the filtrate is transferred into a
100 mL volumetric flask and made up the volume with water. The
solution is then analyzed to estimate Cl− by using a Metrohm col-
umn in ion chromatography method (limit of detection 1 ppb). The
same solution is used to determine Sn content by ICP-OES.
Other carbamates are prepared using the same procedure
with the appropriate amount of reactants. The carbamate of 2-
ethoxyethylamine is prepared in a similar manner.
N-(methyl)(2-hydroxyethyl)ethylcarbamate: 1H NMR: ı 1.24 (3H,
t, J = 6.9), 2.74 (3H), 4.24 (2H, q, J = 7.1), 3.50 (2H, t, J = 6.1), 3.53 (2H,
t, J = 6.1).
N-(2-ethoxyethyl)ethylcarbamate: 1H NMR: ı 1.25 (3H, t, J = 6.9),
1.24 (3H, t, J = 6.9), 3.40 (2H, q, J = 6.9), 4.28 (2H, q, J = 6.9), 3.46 (2H,
t, J = 6.7), 3.45 (2H, t, J = 6.7).
3. Results and discussion
3.1. Synthesis of oxazolidiones
2.4. Oxazolidinone from N-(2-hydroxyethyl)ethylcarbamate
All the catalysts are synthesized according to our previously
reported procedures [30]. For the synthesis of OXZ, typically DEC,
2AA and a catalyst (0.5 mol% with respect to 2AA) are combined in
a round bottom flask while the progress of the reaction is mon-
itored using a GC equipped with a DB-5 column, following the
diminishing peak of 2AA and growing peak of OXZ. As an example,
Fig. 1 shows a typical chromatogram obtained during the for-
mation of (S)-4-benzyl-2-oxazolidinone (retention time 10.6 min)
using (S)-phenylalaninol (retention time 14.2 min) as the start-
ing compound. The purified (S)-4-benzyl-2-oxazolidinone from
the reaction mixture is characterized using 1H NMR, electrospray
ionization mass spectroscopy (ESI-MS) and elemental analysis. (S)-
N-(2-hydroxyethyl)ethylcarbamate (175 mg, 2.0 mmol) is taken
with 20 mL toluene in a round bottom flask and heated at
80 ◦C for 1 h. The reaction is monitored by GC. The peak at
8.95 min of the chromatogram which is different from the N-
(2-hydroxyethyl)ethylcarbamate peak at 10.11 min indicates the
formation of oxazolidinone. Concentrated mass obtained after
evaporation of toluene from the reaction mixture is separated
through a 10 in. silica gel column and oxazolidinone is eluted by
1:3 ethylacetate–hexane. 1H NMR: ı 3.49 (2, 1H, ddd, J = 8.0, J = 4.3,
J = 0.0), 3.491 (2, 1H, ddd, J = 8.0, J = 4.3, J = 0.0), 4.44 (3, 1H, ddd,
J = 8.0, J = 4.3, J = 0.0), 4.44 (3, 1H, ddd, J = 8.0, J = 4.3, J = 0.0). mp 87 ◦C.

36
S. Pulla et al. / Journal of Molecular Catalysis A: Chemical 338 (2011) 33–43
100
100
80
60
40
20
0
1a
1b
1c
1d
90
Variation of 2AA
Phenyl-alaninol
80
2-aminobutanol
70
60
50
(b)
40
30
Variation of DEC
Phenyl-alaninol
20
2-amino butanol
(a)
10
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4
Concentration (mol%)
Catalyst concentration (mol%)
Fig. 3. Variation of (a) 2AA and (b) DEC. 1 h, 80 ◦C. (a) 1a 0.004 mol, 2AA 1.09 mol
and (b) 1a 0.004 mmol, DEC 1.13 mol.
Fig. 2. Variation in yield of formation of (S)-2-benzyloxazolidinone with respect
to catalyst concentration. Temperature 80 ◦C, DEC 1.07 mol, phenylalaniol 0.25 mol,
time 1 h.
amount of catalyst of 0.5 mol% with respect to 2AA is chosen for
most of the reactions.
4-benzyl-2-oxazolidinone shows m/z 150.99, which is in complete
agreement with the literature-reported values [36]. The yield of
OXZ is calculated based on external standard method using in-
house produced authentic OXZ samples. In the GC (Fig. 1), there
is an extra peak at 14.8 min which vanishes with time and is also
different from the product and reactant. The new peak shows mass
spectrum at m/z 180 which corresponds to ethyl carbamate of 2AA,
a possible intermediate of the reaction.
Results indicate that almost 100% yield is obtained within 1 h at
80 ◦C using 0.5 mol% of 1a, while maintaining concentrations of DEC
and 2AA of 5:1 (Table 1). The role of 1 is critical for OXZ formation
as no product formation is observed when reactions are performed
under similar conditions without adding any catalyst. 1 and OXZ
can be separated using short column chromatography, depending
upon the nature of the 2AAs. We have studied ICP-OES and ion
chromatography analyses for detecting Sn and Cl− [35] in order to
check if they have leached from the catalyst during the reaction. The
result indicates that tin or chlorine residues are present at lower
than the detection limit in the product.
A series of 2AAs along with different catalysts have been studied
in order to understand the reactivity patterns and mechanism. It has
been found that the nature of both 2AAs and 1 affects the efficiency
of the reaction, which is addressed in details here.
Table 1 shows the results obtained upon variation of substrates,
catalysts and other reaction parameters. The yield increases with
time and reaches the maximum within 1 h (Table 1, entries 3–8).
Similarly the temperature (Table 1, entries 19–21) is optimized
at 80 ◦C. A rise in temperature is required to obtain similar effi-
ciencies in the case of a less efficient catalyst. However, very high
temperature lowers the reaction yield (Table 1, entries 19–24).
We have checked the dependency of DEC for OXZ formation
using catalyst 1a and (S)-phenylalaninol as a substrate. Hexane is
used as an inert solvent during this investigation. It is presumed
that the use of hexane is just limited as a medium in order to per-
form the reaction at very low concentration of DEC. Fig. 3 shows the
effect of DEC concentration on OXZ formation using catalyst 1a. It
is evident from the plot that a 1st order dependency exists on DEC
when the concentration of DEC is low and reactivity reaches a max-
imum when the molar ratio of DEC to catalyst reaches 2:1. Further
increase in the DEC concentration does not effectively improve the
yield. The result has been verified for different 2AAs too.
A similar observation is encountered when percent yields are
determined using different concentrations of 2AAs namely, (S)-
phenylalaninol and 2-aminobutanol. In these cases, addition of
hexane is not necessary. It is evident from Fig. 3 that there is a
linear relationship between yield (%) and concentrations of 2AA
when catalyst (1a) concentration is kept constant. This is a different
observation when compared to those of other reported reactions,
like esterification reaction catalyzed by 1 where saturation of the
reaction yields is observed [27,37]. Therefore, it can be presumed
that the change in Gibbs free energy (ꢁG) for the reaction is highly
negative and this is true at least for the rate determining step (RDS)
of the overall reaction [38]. Stoichiometrically, the present reaction
(Eq. (1)) has positive entropy change (ꢁS = 1 eu), whereas esterifi-
cation and other reported reactions have ꢁS = 0 eu. As a result, in
accord with the formula ꢁG = ꢁH − TꢁS, assuming ꢁH is small, ꢁG
comes out to be negative, resembling a spontaneous reaction [39].
The linear relationship is found to be maintained even under the
low concentrations of 2AAs. With increasing 2AA concentration,
the yield reaches a maximum when 2AA and 1a have a ratio of 2:1.
Thus, it can be concluded that for obtaining an optimum product
yield, maintaining a ratio of 1, DEC and 2AA of 1:2:2, respectively
is extremely important. The fact that 1 has two active Lewis acid
centers and taking into consideration the above-mentioned ratio
of reactants to 1 for obtaining the optimum yield, it can be con-
cluded that two nonequivalent Sn (Sna and Snb) atoms of 1 interact
with both DEC and 2AA during the product formation. The yield
obtained in the case of DEC variation (Fig. 3a) is lower than that of
2AA variation (Fig. 3b). This is due to the dilution by hexane which
lowers the effective concentration of 1. However, the slopes of the
two different plots indicate that the strength of interaction of cat-
alyst with 2AA is higher than with DEC [40]. This can be explained
3.2. Dependence of catalyst, DEC and 2AA
When the amount of 1 is varied (in terms of mol%) in the for-
mation of OXZ using (S)-phenylalaninol as substrate, it is shown in
Fig. 2 that yield is directly proportional to the concentration of 1.
This indicates a first order dependency on the catalyst. However,
when different catalysts are used, depending on the substitution on
the catalysts, the yield of the reaction varies too. As Fig. 2 shows, 1a
is much superior in comparison to other catalysts, namely 1b–1d.
In fact, 1d is found to be the least active among the series. It can be
concluded from Fig. 2 that the introduction of aromatic substituents
in the catalyst structure causes a reduction of activity. An optimum

S. Pulla et al. / Journal of Molecular Catalysis A: Chemical 338 (2011) 33–43
37
Table 1
Reactivity of 1a with 2AA.a
Entry
Substrate
Catalystb
Temperature (◦C)
80
Time (min)
60
Product
% Yieldc
97
% eed
O
O
1
1a
O
N
H2
H
N
H
Ethanolamine
O
O
2
1a
80
60
98.5
99
N
H
O
N
H2
H
(S)-2-Aminobutanol
3
4
5
6
7
1a
1a
1b
1c
1d
80
80
80
80
80
50
40
60
60
60
30
<5
80
76
67
O
O
8
1a
1a
1a
80
80
80
60
60
60
99
99.2
99.5
N
H
O
N
H2
H
(S)-2-Aminopropanol
O
Ph
O
9
100
N
O
N
H2
H
H
Ph
(S)-2-Aminophenylethanol
O
NH2
O
10
80
N
H
OH
2-Aminophenol
11
12
13
1b
1c
1d
80
80
80
60
60
60
60
40
10
O
H
O
N
H
O
N
H2
H
14
15
1a
1a
80
80
60
60
70
99.2
O
Et
cis-4-Aminocyclohexanol
O
N
H2
H
No reaction
-
4-Aminophenol
O
O
O
16
1a
80
60
71.4
98.9
N
H
H
O
N
H2
O
17
18
1a
1a
80
80
60
60
98.2
98.8
99.4
99.5
N
H
O
O
N
N
H
H
H2
H2
O
O
N
2-Aminocyclopentanol
O
Ph
O
Ph
19
1a
80
60
99
99.8
N
H
O
N
H2
H

38
S. Pulla et al. / Journal of Molecular Catalysis A: Chemical 338 (2011) 33–43
Table 1 (Continued)
Entry
Substrate
(S)-phenylalaninol
Catalystb
Temperature (◦C)
Time (min)
Product
% Yieldc
% eed
20
21
22
23
24
1a
1a
1a
1a
1a
60
40
100
110
120
60
60
60
60
60
52
20
95
80
70
99.5
99
95
95
93
O
O
H
O
N
H
25
26
1a
1a
1a
80
80
80
60
60
60
93
N
H
H
O
O
N
No reaction
O
nBu
nBu
N
H
27
86
O
N
a
DEC is used in all cases.
b
c
In all the cases 0.5 mol% catalyst is used with respect to substrate 0.5 mol, unless the value mentioned in parentheses.
Yields are measured in GC by external standard method using authentic samples.
ee is determined by % area obtained in chiral HPLC chromatogram.
d
by considering the nucleophilicity of the reagents. 2AA is actually
a stronger nucleophile than DEC. So it is expected that 2AA will be
the first to attack the most active Lewis acidic site of 1.
100
90
80
70
60
50
40
30
20
HO
H2 N
trans-2-aminocyclohexanol
3.3. Effect of nature of substrates and proximity of substituent
Results listed in Table 1 reveal that catalyst 1 is highly efficient
in promoting the reaction of 2AA and DEC to form OXZs. However,
the nature of the substrate has a significant effect on the reac-
tion yield. In order to understand such an effect a detailed study
using various substrates is performed. Proximity of –OH and –NH2
groups in 2AAs is an important factor in governing the yield of the
reaction, as it is directly related to the feasibility of the OXZ ring
formation. Thus, no products are obtained when p-aminophenol
and p-aminocyclohexanol are combined, as the amine and alco-
hol groups are too far apart in these molecules making the ring
formation difficult. Only ethylcarbamic ester forms in the case of
p-aminocyclohexanol; p-aminophenol does not even form any car-
bamate ester due to its low nucleophilicity (Table 1, entries 14 and
15).
O
H
N
H2
cis-2-aminocyclohexanol
50 60
10
20
30
40
Time (min)
Fig. 4. Relative rate of cis and trans 2-aminocyclohexanols towards formation of
OXZ. 2AA 0.25 mol, DEC 1.15 mol, 80 ◦C, 1a 0.001 mol. H is omitted from the bonds
for clarity.
Subsequently, the rate of formation of OXZ from cis-2-
aminocyclohexanol is lower than that of the trans isomer (Fig. 4).
According to Paizs’s calculation, the trans isomer remains in the low
energy conformation where both –OH and –NH2 are in equatorial
positions due to internal hydrogen bonding [41]. As both groups are
locked into equatorial positions they are in close proximity to each
other and can interact with 1 and subsequently form the OXZ ring
very easily. However, this is not the case for the cis isomer, making
the reaction rate low. In the cis isomer the axial hydrogen comes
within close proximity of the axial chlorine, Cla, of 1 making the
relevant transition state (TS) unstable. In the case of SN2 attack on
the Snb for replacing Cl from Cl–Snb–O moiety, the angle of attack
named ‘deviation angle’, is too narrow to allow the axial hydrogen
of the cis isomer to be accommodated in the intermediate structure
and thus explaining the slower reaction rate. As described in Fig. 5,
deviation angle of Clb–Snb–O in 1a is 38◦ and it is shortened to 29◦
in 1d. The value for 1a is actually the value of tetraethyldichlorodis-
tannoxane. This, we have approximated based upon the fact that
they are structurally similar, also 119Sn NMR and Mössbauer iso-
meric shift almost do not change while changing from ethyl group
to butyl group in 1 [42].
Substituents on the nitrogen atom of 2AA can also influence the
reaction yield. Results indicate that a 1◦ amine reacts more effi-
ciently than a 2◦ amine even though the basicity and nucleophilicity
of a 2◦ amine is higher than that of a 1◦ amine [43]. The lowered
reactivity of a 2◦ amine can be attributed to its steric bulk [44]. Thus,
2-aminoethanol (a 1◦ amine) provided a higher yield of OXZ than N-
Cl
R
Sn
O
R
R/
R'
Sn
Cl
Sn
-OH
R'
R'
38o, 1a
29.3o, 1d
162o, 1a
150.7o, 1d
O
Deviation angle
Fig. 5. ‘Deviation angle’, the angle of deviation from linearity, showing availability
of bigger room for 1a than 1d for SN2 displacement of Cl by incoming –OH of 2AA.

S. Pulla et al. / Journal of Molecular Catalysis A: Chemical 338 (2011) 33–43
39
and precludes the interaction of –NH2 with 1 (Scheme 2). Absence
of –OH stretching at 3400 cm−1 also supports the transformation
[47]. Following the previously reported mechanism using 1 as a
catalyst [29,30,33] for various reactions, it can be argued that 2AA
attacks 1 by displacing the bridged Cl, which we also have demon-
strated and described later. So it can be concluded that –OH of 2AA
interacts with 1 displacing Cl [33]. Minor lower yield encountered
in case of aminoethanol than 2-aminobutanol is probably due to
higher nucleophilicity of 2-aminobutanol.
100
90
80
70
60
50
40
30
20
HO
N
H2
O
H
NH2
3.4. Nature of the catalyst
The crystal structures of distannoxanes reported by a number
of authors [42,48,49] reveal that 1 has a stable dimeric ladder type
structure in a distorted trigonal bipyramidal environment. This
makes two Sn atoms non equivalent which is also reflected in 119Sn
NMR studies [50]. Our observation, as mentioned above (Fig. 2) also
supports the non equivalency of Sna and Snb. The unique features of
the catalysts, which differ from conventional single site Lewis acidic
catalysts, are due to two dissymmetric Sn centers (Sna and Snb) in
1. In order to study the effect and role of Sn centers of 1 on reactiv-
ity, substituents on Sn are varied. Alkyl and aromatic substituents
further control the varying electronic and steric environment and,
in turn, the different activity of 1. The catalysts with all the butyl
groups (1a) are found to be the most superior and progressively
show lower activity when substituted with phenyl groups (1b–1d)
as shown in Fig. 7. Position of the phenyl substituent on Sn is one
of the deciding factors on the reactivity of 1 too. This is noticed in
the variation of activities between 1b and 1c, once again proving
the non-degeneracy of Sn. This is in agreement with our earlier
observation regarding the formation of glycerol carbonate from
DEC and showing 1a and 1d are most and least reactive catalysts,
respectively [30].
HO
N
H2
10
20
30
40
50
60
Time (min)
Fig. 6. Effect of substituents in 1 and 2 positions in 2AA. 2AA 0.23 mol, DEC 1.06 mol,
1a 0.001 mol, 80 ◦C, 1 h.
ethyl-2-aminoethanol (a 2◦ amine) (Table 1, entries 1 and 25). This
observation is further verified when N-butyl-aminoethanol, a more
sterically hindered 2◦ amine, yielded much less OXZ (Table 1 entry
27). Basically, substituents on nitrogen do not allow the intermedi-
ate or TS to be accommodated in the deviation angle for displacing
Clb.
As seen in Fig.
6 the rates of OXZ formation using 2-
aminobutanol and 1-amino-2-butanol are significantly different.
A lower reaction yield is obtained in case of the latter. Substituent
effects at the 2 and 1 positions of 2AA may imply a possible influ-
ence in the transition state. In this case too, the deviation angle
can help to explain the effect. Interestingly, Fig. 6 also reveals that
aminoethanol has a similar reaction rate to that of 2-aminobutanol,
proving that the substituents next to –NH2 in 2AA have little effect.
This may be due to the fact that substituents at the 1 position of 2AA,
which is next to the –OH, cause more steric hindrance (Scheme 2)
in the TS. On the other hand, substituents at the 2 position in 2AA,
next to –NH2, are too far away to create an obstacle for product
formation, resulting in a higher reaction rate and yield. It can be sur-
mised from this information that the involvement of the –OH group
in the TS during product formation has more influence on the RDS.
The presence of a IR band at 582 cm−1 and the absence of Sn–N
stretching at 428 cm−1 [45] confirm the formation of Sn–OR [46]
Studies on distannoxanes by a number of groups imply that its
reactivity lies in the displacement of bridged Cl [42,48,49,51,52].
Due to the dimeric structure of distannoxanes the bridged Sn–Cl
bond is longer than the axial bond, making it more susceptible
to displacement. This is shown in many other crystal structures
of distannoxanes as well [53–55]. According to the theoretical
calculations proposed by Wakamatsu and co-workers, additional
stabilization of 1 is gained upon dimerization through the bridg-
ing of Cl with another molecule [56]. This makes the Sn–Cl bond
long enough to be partially ionic in nature. For example, dis-
tances of Sna–Clb and Snb–Clb are 2.697 A and 2.43 A, respectively
˚
˚
in 1d [49]. This difference in distances is even more promi-
nBu
nBu
N
Sn
O
Slower
OH
Sn
Ph
Ph
Cl-
+
HO
N
H2
nBu
nBu
1-Amino-2-butanol
Cl
Sn
O
Cl
Sn
Cl
nBu
nBu
+
nBu
nBu
N
Faster
Sn
O
O
Sn
Cl
Cl-
nBu
nBu
+
HO
N
H2
2-Aminobutanol
Part of catalyst
Probable transient
intermediates
Scheme 2. Effect of the position of substituents in 2AA.

40
S. Pulla et al. / Journal of Molecular Catalysis A: Chemical 338 (2011) 33–43
Cl atom. As the Ph on Sna shortens the Sna–Clb bond distance, 1b
120
100
80
60
40
20
0
should be more active than 1c. In other words, a Ph on Sna has a
stronger influence than on Snb. It is apparent that the displacement
of bridged chloride from terminal Sn is of prime importance in gov-
erning catalytic activity. Although there is not much information
available for 1b and 1c, the sudden drop in molar conductivity from
5.25 (1b) to 2.87 (1c) clearly demonstrates the mobility difference
in bridged chlorines between these two catalysts. The conclusion
drawn above is valid as it has been found that the rate constants
measured for the four catalysts (Table 2) show the same trend as
that of the conductivity.
1a
1b
1c
1d
Relative chlorine replacement strength of a substrate can be
observed from the fact that 2-aminophenol exhibits lower activity
than 2-aminocyclohexanol when catalyzed by 1a (Table 1, entries
10, 16 and 17). Thus apart from the ionic character, Snb–Clb bond
distance of bridged Cl is equally important that differs for all the cat-
alysts. Activity can be simply correlated to the nucleophilic strength
of –OH groups, which is less in 2-aminophenol [58]. Thus, the rate
of reaction varies depending on both substituents on the catalyst
and the substrate.
0
10
20
30
40
50
60
70
Time (min)
Fig. 7. Relative efficiencies of 1. Phenylalaninol 0.5 mol, 1 1.1 mmol, DEC 1.08 mol,
80 ◦C.
2.51 Å
3.37 Å
nBu
2.43Å
2.697 Å
nBu
Ph
nBu
Snb
Ph
3.5. Mechanism
Snb
O
Clb
Clb
Sna
Cl
Few attempts have already been made by various groups in
describing distannoxane catalysis. But in our case bifunctionality
of 2AA presents complications, which made us to look into the
mode of interaction of 2AA (through –NH2 or –OH) with the specific
catalyst active site, Sna or Snb.
Ph
Ph
Sna
O
nBu
Cl
As explained earlier (Scheme 2), FTIR band at 582 cm−1 (1a) indi-
cates that it is –OH functionality, and not the –NH2 group, most
likely interacts with 1. This also points out that the –OH of 2AA
interacts with Sna not Snb, otherwise the peak would have appeared
at 605 cm−1 [46]. Findings of past studies involving transesterifica-
tion reaction catalyzed by 1 resulted in the displacement of bridged
Cl by alcohol –OH [33]. The chloride replacement perhaps occurs
in our case as well. Addition of AgNO3 to the water extract of reac-
tion mixture of 2-aminobutanol and DEC in the presence of 1a at
30 min (incomplete reaction) produced immediate precipitate of
AgCl, indicating the presence of free chloride (Cl−). Similar test on
1d does not produce AgCl at all. Only a faint white coloration is
observed in case of 1b and 1c. Above results dictate that 1a has got
highest and 1d has got least ability to displace Cl.
External treatment of DEC with ethanolamine (Table 3, entry
1) and N-methylethanolamine (Table 3, entry 2) at 100 ◦C (higher
than the OXZ reaction temperature) provides the correspond-
ing carbamic esters, namely ethyl-2-ethanolcarbamate (EEC) and
ethyl-2-ethanol-N-methylcarbamate (EENMC). But when N,N-
dimethylaminoethanol is treated with DEC, no carbamic ester
can be isolated (Table 3, entry 3), perhaps due to the reason of
unavailability of nitrogen proton that hinders in OXZ ring forma-
tion. EEC and EENMC when heated with 1 form a stoichiometric
amount of the corresponding OXZs (Table 3, entries 4 and 5), evi-
dencing carbamate formation as one of the intermediate steps.
Mentioned earlier, the peak at m/z 180 (Fig. 1) detected due
to ethyl-4-benzylcarbamate strongly support aforesaid statement.
2-Ethoxyethylamine produces corresponding carbamic ester ethyl-
2-ethoxyethylcarbamate when reacted with DEC (Table 3, entry 6),
but no OXZ formation is observed, even when heated separately
with 1 and an excess DEC (Table 3, entry 7). Therefore the presence
of free OH group is necessary for the formation of OXZ which is pro-
tected in the case of 2-ethoxyethylamine. DEC is known to be more
susceptible to attack by -NH2 than by –OH group [40,59]. Hence,
only the –OH (from 2AA) group is involved in interacting with 1 as
free amine groups have already reacted with DEC to form the car-
bamate. It is examined that the carbamate formation is much faster
1a
1d
Scheme 3. Longer Snb–Clb and Sna–Clb distances caused by nBu substituents in 1a
for Cl displacement with greater ease in comparison to Ph substituted 1d. Data for
1a is taken from dichlorotetraethyldistannoxane assuming that Et and Bu has no
effect in structural change as confirmed by Mossbauer and 119Sn NMR.
˚
˚
nent in the case of 1a, which are 3.37 A and 2.51 A, respectively
(Scheme 3) [42]. In this case too, the data mentioned for 1a is
actually of tetraethyldichlorodistannoxane, due to the same reason
spelled in Section 2.3 [51]. Measurements of the molar conductivity
(ꢀ−1 cm2 mol−1) of the catalysts in DMSO (Table 2) show a similar
downward trend: 6.41 for 1a, 5.25 for 1b, 2.87 for 1c and 2.21 for 1d.
The value for 1a is in accordance with the literature reported value
of 6.36 ꢀ−1 cm2 mol−1 [57]. Therefore, the more ionic character of
bridged chlorine would be favorable for producing the reaction
with more ease. The distance of the Sna–Clb bond is 3.37 A for 1a
and 2.697 A (Scheme 3) for 1d. This clearly indicates more labil-
˚
˚
ity of Cl in 1a. The longer bond length in 1a can be explained by
isomeric shift (IS) values, in Mössbauer spectrum as reported by
Harrison and co-workers, 1.46 (mm/s) and 1.26 (mm/s) for 1a and
1d, respectively. Due to the –I effect of the phenyl group, the IS
value of 1d is as close as a methyl derivative (1.28). As indicated by
a large IS, a higher electron density on Sn in 1a is responsible for
making a weaker bridged chlorine, rendering more lability to the
Table 2
Characteristics of 1 that influence reactivity.
Catalyst Sna–Clb (Å) Snb–Clb (Å) ꢂm (s cm2 mol−1
)
k × 10−3 (s−1
2-ABa
)
Phalb
1a
1b
1c
1d
2.697
3.37
2.43
2.51
6.41
5.25
2.87
2.21
51.9333
41.4217
18.930
6.9943
49.721
40.003
12.232
5.890
a
2AB = 2-aminobutanol.
Phal = (S)-phenylalaninol, 80 ◦C, 1 h.
b

S. Pulla et al. / Journal of Molecular Catalysis A: Chemical 338 (2011) 33–43
41
Table 3
Rate constants of the reactions studied to evaluate the mechanism.
Entry
Reactions
Catalyst
No
O
O
k1
H
O
N
H
1
+
+
H
H
O
O
NH2
Et
O
O
Et
Et
O
O
k2
O
H
O
N
2
3
No
No
N
H
Et
O
O
Et
Et
O
O
No reaction
+
H
H
O
O
N
Et
O
O
Et
O
O
k3
O
N
H
4
5
1a
1a
N
H
Et
O
O
N
k4
O
O
H
O
N
Et
O
O
O
Et
k5
Et
O
N
H
+
6
7
No
O
NH2
Et
Et
O
O
Et
O
1
O
No OXZ
Et
O
N
H
1a
Et
O
O
O
k6
O
8
1a
+
NH2
H
O
Et
N
H
O
O
Et
than the corresponding OXZ formation (Table 3, entries 1 and 4),
indicating the formation of carbamate prior to OXZ ring.
On the other side, the effect of the concentration of DEC on yield
(Fig. 3b) designate the activation of the ester group of carbamate
occurs by 1. Most likely Snb of 1 is responsible for this activation, as
Sna is in interaction with –OH. Therefore, it can be concluded that
Sna activates –OH and Snb activates the ester (EtO–CO–) groups of
the preformed 2-ethyl carbamato alcohol.
We may now be in a better position to explain the lower activity
of 1c for OXZ formation when compared to 1b. Degeneracy in the
Cl
R
O
Ph
Sna
O
R
+
Et
O
O
Et
O
N
H2
H
R'
Clb
Snb
R'
O
N
Ph
O
H
O
O
N
H
H
Ph
O
Et
+
+
EtOH
Cl
R
EtOH
Sna
R
O
Et
H
O
N
R'
Clb
Snb
O
R'
O
Ph
Fig. 8. Possible mechanism is shown by presenting one half of the catalyst.

42
S. Pulla et al. / Journal of Molecular Catalysis A: Chemical 338 (2011) 33–43
Table 4
Recovery and recycle test of 1 in OXZ formation.a
Sub
2AB
Cycl
1a (g)
DECb (g)
U
1d (g)
DECc (g)
Ud
Re
R
U
R
U
R
1
2
3
1.2
1.15
1.14
1.12
257.25 250.2
247.8
1.22
1.20
1.19
1.18
257
256.1
255
240.9
252.1
Phal
1
2
3
1.16
1.2
1.16
260
254.2
247.3
240.4
1.20
1.17
1.18
1.18
1.17
259
258
250.9
243.8
236.4
2ABf
1
2
3
0.7
0.4
0.0
258.5
251.7
244.2
237.3
1.16
1.14
1.13
251.8
247.4
241.2
a
2AA 2.1 mol, 1 2.16 mmol, temperature 80 ◦C, time 1 h.
DEC used with 1a.
DEC used with 1b.
U = used amount. Amount recovered from last cycle.
R = recovered amount, 2AB = 2-aminobutanol, PhAl = phenyl alaninol.
Recovery was performed by silica gel chromatography.
b
c
d
e
f
catalyst’s structure implies that Sna has more positive influence
over the reaction parameter than Snb. Although the presence of
butyl groups on Sna in 1b helps the Cl replacement, phenyl attach-
ment on Snb trims down the strength needed to activate ester
group, subsequently making the catalyst comparatively inefficient.
On the other hand, in the case of 1c, the presence of Ph on Sna
greatly reduces the activity such that it is not possible to compen-
sate by the butyl group on Snb. Ph groups on both Sna and Snb in
1d have accelerated neither Cl replacement nor the activation of
the ester group, performing with least efficiency. We have tried
to accumulate all the information needed to elucidate the possi-
ble reaction pathway as shown in Fig. 8. The mechanism described
above in view to explicate the specific interactions of 2AA and dif-
ferent Sn centers in 1 is however, as a whole in agreement with the
mechanism of transesterification proposed in literature [33]. The
proposed mechanism holds the reason for retention of chirality of
chiral substrates.
Fig. 9. 119Sn NMR (CDCl3) of recovered 1a.
Although Table 4 shows up to 90% catalyst recovery, depending
on the nature of the catalysts, short silica gel (12ꢀꢀ) column chro-
matography can be employed as a general way of separating the
catalyst.
The recovered catalyst is found to be identical to the initial
catalyst and this is confirmed by elemental (C, H) analysis, mass
spectrometry and 1H NMR. Identity of the recovered catalyst is also
manifested in the activity (Table 4) to produce OXZ. To confirm the
retention of Sn core after reaction, we have checked 119Sn NMR
of the recovered catalyst. A representative 119Sn spectrum is given
for 1a (Fig. 9). The distinct peaks (ı, ppm) at −90.2 corresponding
to exocyclic (Sna) and at −137.8 corresponding to endocyclic (Snb)
Sn atoms match very well with the literature values of −92.0 and
−139.8, respectively [50]. The identity of 1a in 119Sn NMR is indica-
tive to support the proposed mechanism [60,61]. Almost no loss in
the catalytic efficiency is observed after three recycles. The catalyst
is stable under reaction conditions as reflected by its activity in the
recovered material, proving the finding of past studies in regard
to retention of identity [62]. Tests indicate that free chloride or Sn
may be present at lower than the detection limit in the product,
confirming the robustness of the catalyst. The above findings show
that it may be suitable to use 1 in continuous batch reactions thus
making it economically feasible to produce OXZ.
3.6. Recovery and recycling of 1 and DEC
Generally, simplicity of separation depends on the differences
in the polarity of the components present in any reaction mixture.
Components can be separated even by simple solvent treatment
using hexane if polarity differs greatly. The reaction mixture
obtained from the reactions carried out with the substrates 2-
aminophenol and phenyl alaninol using catalyst 1a is one of
these situations. On the other hand, substrates like ethanolamine,
2-aminobutanol and 2-aminocyclopentanol require column chro-
matography to separate the components. It is noteworthy that
1c and 1d can be separated from the substrates ethanolamine,
2-aminobutanol and 2-aminocyclopentanol once diluted with hex-
ane.
Solvent extraction does not work if the polarity difference
between the two components is small. Silica gel chromatography is
then the general method of recovering the catalyst with acceptable
amount of loss. Table 4 lists the results obtained during recovery of
1 and DEC when hexane extraction is employed in the recovery pro-
cess of oxazolidinone formations using 2-aminobutanol and phenyl
alaninol, respectively.
4. Conclusions
In conclusion, we have shown that dichlorodistannoxanes are
very effective catalysts for synthesizing oxazolidinone from 2AA
and DEC within 1 h quantitatively. The recyclable catalyst is used
with a variety of chiral and achiral substrates. No chirality loss
is encountered during catalytic reaction. The recovered catalyst
is found to be identical to the original and showed similar cat-
alytic efficiency. The ratio of catalyst, DEC and 2AA is required to
be maintained at 1:2:2 for optimum yield. Both –OH and –NH2
groups within reasonable proximity are necessary to form an oxa-
zolidinone ring. The effect of substituents in the 1 position (next
to –OH) compared to 2 position in 2AA had more influence on the
reaction rate. The catalysts are dimeric and contain two dissym-
metric Sn centers, namely Sna and Snb, with varying reactivity.
The reaction proceeds with carbamate formation as a prior step
through amino group followed by chloride replacement in the cat-
The polarity difference between 1a and phenyl alaninol or 2-
aminophenol is large enough to separate them efficiently. Addition
of hexane further facilitates the separation. But substrates like
ethanolamine, 2-aminobutanol and 2-aminocyclopentanol have
polarities closer to that of 1a and thus need column chromatog-
raphy for separation. 1c and 1d are sufficiently polar with respect
to the reaction medium itself which facilitates easy separation.

S. Pulla et al. / Journal of Molecular Catalysis A: Chemical 338 (2011) 33–43
43
alyst by –OH. The mechanistic study indicates that Sna activates
–OH and Snb activate –O–CO– of a preformed ethylcarbamate from
2AA and DEC. Phenyl substituents reduce the catalytic activity by
shortening the Sn–Cl bond and the influence is most prominent
when attached to the terminal Sna atom of the catalyst. Polar-
ity differences between substrate and catalyst facilitate an easy
separation and recovery for further recycling. As the catalyst is
almost fully recoverable and preparation of the catalyst is quan-
titative with respect to their starting materials, contamination of
the product and the environment is merely negligible. Neverthe-
less, while triorganotin compounds are known to be toxic, there
is no such toxicity is reported so far for tetraorganodistannox-
ane compounds. The findings as a whole provide a broad way of
synthesizing OXZs while retaining their chirality. Depending upon
the need, one can design the process for practical feasibility with-
out separating the catalyst. Consequently, overall the process has
promise to contribute economically as well as environmentally.
[20] J.R. Gage, D.A. Evans, Org. Synth. Coll. Vol. 8 (1993) 528;
J.R. Gage, D.A. Evans, Org. Synth. Coll. Vol. 68 (1990) 77.
[21] M. Kottenhahn, K. Drauz, H. Hilpert, US Patent, 5,744,611 April (1998).
[22] D.A. Evans, A.E. Weber, J. Am. Chem. Soc. 108 (1986) 6757.
[23] L.F. Xiao, L.W. Xu, C.G. Xia, Green Chem. 9 (2007) 369.
[24] J. Otera, J. Nishikido, Esterification: Methods, Reactions and Applications, 2nd
ed., Wiley–VCH, 2009.
[25] H. Takahashi, T. Hayakawa, M. Ueda, Chem. Lett. (2000) 684.
[26] R.P. Houghton, A.W. Mulvaney, J. Organomet. Chem. 517 (1996) 107.
[27] A. Nishio, A. Mochizuki, J.I. Sugiyama, K. Takeuchi, M. Asai, K. Yonetake, M. Ueda,
J. Polym. Sci., Part A: Polym. Chem. 39 (2001) 416.
[28] R. Nagahata, J.I. Sugiyama, M. Goyal, M. Asai, M. Ueda, K. Takeuchi, J. Poly. Sci.,
Part A: Polym. Chem. 38 (2000) 3360.
[29] J. Otera, Chem. Rev. 93 (1993) 1449.
[30] Y. Patel, J. George, S.M. Pillai, P. Munshi, Green Chem. 11 (2009) 1056.
[31] P. Munshi, A. Ghosh, E.J. Beckman, Y. Patel, J. George, S.Z. Sullivan, S. Pulla, P.
Ramidi, V. Malpani, Green Chem. Lett. Rev. 3 (2010) 319.
[32] H.E. Hoydonckx, D.E.D. Vos, S.A. Chavan, P.A. Jacobs, Top. Catal. 27 (2004) 83.
[33] J. Otera, T. Yano, R. Okawara, Chem. Lett. (1985) 901.
[34] D.D. Perrin, W.L.F. Armarego, Purification of Laboratory Chemicals, 4th ed.,
Butterworth-Heinemann, 1997.
[35] J. George, Y. Patel, S. Pillai, P. Munshi, J. Mol. Catal. A: Chem 304 (2009) 1.
[36] D.A. Evans, J. Bartroli, T.L. Shih, J. Am. Chem. Soc. 103 (1981) 2127.
[37] E.N. Suuciu, B. Kuhlmann, G.A. Knudsen, R.C. Michaelson, J. Organomet. Chem.
556 (1998) 41.
Acknowledgements
[38] P.W. Atkins, Physical Chemistry, 3rd ed., Oxford University Press, 1985.
[39] C. Raymond, B. Cruickshank, Entropy Free Energy and Equilibrium in Chemistry,
8th ed., McGraw-Hill, Inc., 2005, p. 765.
[40] J.M. Harris, S.P. McManus, Nucleophilicity Advances in Chemistry Series, vol.
215, American Chemical Society, Washington, DC, 1987.
[41] B. Paizs, I. Pinter, J. Kovcs, W. Viviani, A. Marsura, P. Friant-Michel, I.G. Csizma-
dia, J. Mol. Struc. Theochem. 395 (1997) 41.
[42] P.G. Harrison, K. Molloy, J. Organomet. Chem. 152 (1978) 63.
[43] V.E. Bel’skii, L.S. Novikova, L.A. Kudryavtseva, B.E. Ivanov, Russ. Chem. Bull. 26
(1977) 1188.
AG likes to thank UALR faculty start up grant and the Department
of Energy (Grant number DE-FG36-06GO86072) for financial assis-
tances to complete the work. JLD greatly acknowledges financial
supports from the National Science Foundation, USA (Grant number
NSF DBIO722538) and the National Institute of Health, USA (Grant
number NIH RR11973).
[44] J.B. Conant, R.E. Hussey, J. Am. Chem. Soc. 47 (1925) 476.
[45] G. Socrates, Infrared and Raman Characteristic Group Frequencies, 3rd ed., John
Wiely & Sons, 2004.
[46] R. Okawara, M. Wada, in: F.G.A. Stone, R. West (Eds.), Structural Aspects of
Organotin Compounds in Advances in Organometallic Chemistry, vol. 5, Aca-
demic Press, New York/London, 1967.
[47] G.D. Yadav, A.D. Murkute, Adv. Synth. Catal. 346 (2004) 389.
[48] Y. Lu, L.H. Weng, Z.L. Lan, J. Li, Q.L. Xie, Chin. Chem. Lett. 13 (2002) 185.
[49] J.F. Vollano, R.O. Day, R.H. Holmes, Organometallics 3 (1984) 745.
[50] T. Yano, K. Nakashima, J. Otera, R. Okawars, Organometallics 4 (1985) 1501.
[51] D.C. Gross, Inorg. Chem. 28 (1989) 2355.
[52] R. Okawara, M. Wada, J. Organomet. Chem. 1 (1963) 81.
[53] S.G. Teoh, L.Y. Kok, E.S. Looi, J. Coord. Chem. 55 (2002) 697.
[54] A. Orita, J. Xiang, K. Sakamoto, J. Otera, J. Organomet. Chem. 624 (2001) 287.
[55] D. Ballivet-Tkatchenko, R.A. Ligabue, L. Plasseraud, Braz. J. Chem. Eng. 23 (2006)
111.
[56] K. Wakamatsu, A. Orita, J. Otera, Organometallics 27 (2008) 1092.
[57] J. Otera, T. Yano, R. Okawara, Organometallics 5 (1986) 1167.
[58] R.G. Pearson, J. Songstad, J. Am. Chem. Soc. 89 (1967) 1827.
[59] F.A. Carey, R.J. Sundburg, Advanced Organic chemistry Part A: Structure and
Mechanisms, 5th ed., Springer, 2007.
[60] V. Pinoie, K. Poelmans, H.E. Miltner, I. Verbruggen, M. Biesemans, G.V. Assche,
B.V. Mele, J.C. Martins, R. Willem, Organometallics 26 (2007) 6718.
[61] D. Deshayes, F.A.G. Mercier, P. Degee, I. Verbruggen, M. Biesemans, R. Willem,
P. Dubois, Chem. Eur. J. 9 (2003) 4346.
References
[1] S. Natesan, S. Deekonda, K.K. Manoj, et al., J. Med. Chem. 45 (2002) 3953.
[2] D.J. Diekema, R.N. Jones, Drugs 59 (2000) 7.
[3] R. Hoogenboom, F. Wiesbrock, U.S. Schubert, Polym. Mater.: Sci. Eng. 93 (2005)
894.
[4] S. Yian, US patent, 7,153,971 (2006).
[5] M.G. Madariaga, S. Swindells, Antimicrob. Agents Chemother. 51 (2007) 1130.
[6] B. Gabriele, G. Salerno, D. Brindisi, M. Costa, G.P. Chiusoli, Org. Lett. 2 (2000)
625.
[7] H. Matsuda, A. Baba, R. Nomufa, M. Korl, S. Ogawa, Ind. Eng. Chem. Prod. Res.
Dev. 24 (1985) 239.
[8] B.M. Bhanage, S. Fujita, Y. Ikushima, M. Arai, Green Chem. 5 (2003) 340.
[9] M. Shi, S. Yu-Mei, J. Org. Chem. 67 (2002) 16.
[10] M. Costa, G.P. Chiusoli, D. Taffurelli, G. Dalmonego, J. Chem. Soc., Perkin Trans.
1 (1998) 1541.
[11] R. Maggi, C. Bertolotti, E. Orlandini, C. Oro, G. Sartori, M. Selva, Tetrahedron Lett.
48 (2007) 2131.
[12] M.P. Sibi, D. Rutherford, R. Sharma, J. Chem. Soc., Perkin Trans. 1 (1994) 1675.
[13] R. Green, P.J.M. Taylor, S.D. Bull, T.D. James, M.F. Mahon, A.T. Merritt, Tetrahe-
dron: Asymmetry 14 (2003) 2619.
[14] T. Francis, M.P. Thorne, Can. J. Chem. 54 (1976) 24.
[15] A. Buzas, F. Gagosz, Synlett (2006) 2727.
[16] R.R. Machin, J. Adrio, J.C. Carretero, J. Org. Chem. 71 (2006) 5023.
[17] T. Munegumi, I. Azumaya, T. Kato, H. Masu, S. Saito, Org. Lett. 8 (2006) 379.
[18] G. Bratulescu, Synthesis (2007) 3111.
[62] B. Jousseaume, C. Laporte, M.C. Rascle, T. Toupance, Chem. Commun. (2003)
1428.
[19] N. Alouane, A. Boutier, C. Baron, E. Vrancken, P. Mangeney, Synthesis (2006)
860.