ARTICLE IN PRESS

G Model
MOLCAA-8537; No. of Pages 14

Journal of Molecular Catalysis A: Chemical xxx (2012) xxx–xxx

Contents lists available at SciVerse ScienceDirect

Journal of Molecular Catalysis A: Chemical

journal homepage: www.elsevier.com/locate/molcata

Computational and experimental investigations of the urethane formation

mechanism in the presence of organotin(IV) carboxylate catalysts
Ransi Devendra a , Neil R. Edmonds a , Tilo Söhnel a,b,∗
a
School of Chemical Sciences, The University of Auckland, New Zealand
b
Centre for Theoretical Chemistry and Physics, The New Zealand Institute for Advanced Study, Massey University, Auckland, New Zealand

a r t i c l e i n f o a b s t r a c t

Article history: Although organotin carboxylate compounds are widely used in the polyurethane industry as catalysts,
Received 9 November 2011 their reactions mechanisms are not fully understood. The urethane formation between isocyanate and
Received in revised form 25 July 2012 alcohol groups in the presence of organotin carboxylate catalysts is simulated with selected compounds
Accepted 14 September 2012
using computational methods and the results are compared with experimental findings. The results reveal
Available online xxx
that organotin carboxylate catalysis of the reaction between these isocyanate and n-butanol in a non-
polar medium follow two routes depending on the isocyanate structure. The reaction rate for aliphatic
Keywords:
primary isocyanates the concentration of the carboxylate ligand in the catalyst has a significant effect
Organotin compounds
Urethane formation
while for aromatic isocyanates the effect is much less. Based on experimental and computational results
DFT a mechanism is proposed for organotin carboxylate catalysis of urethane formation for these isocyanates.
Catalyst
Isocyanate © 2012 Elsevier B.V. All rights reserved.

1. Introduction Bloodworh and Davies proposed a catalytic mechanism for

organotin compounds in urethane reactions for the first time [4]. In
Organotin catalysts are widely used in the polyurethane indus- this work phenylisocyanate was reacted with tributyltin methoxide
try due to their outstanding efficiency [1]. The function of the as shown in Scheme 2.
catalyst is to promote the reactivity between the isocyanate and The mechanism proposed was based on the reaction products
the alcohol groups to form the urethane linkage. Scheme 1 shows obtained when phenyl isocyanate reacts with trialkyltin alkoxide.
the reaction for the urethane formation. Organotin catalysts are Since then many investigations have carried out and several mech-
generally used in the preparation of polyurethanes based on both anisms for organotin catalysis in urethane formation have been
aliphatic and aromatic isocyanates. The catalytic effect of organotin presented [4–6,10,11]. At present there are three distinct mech-
compounds on the reactivity of aliphatic isocyanates is much more anisms suggested for the organotin catalyzed urethane formation:
significant than with aromatic isocyanates [2]. However, due to the (1) insertion mechanism (organo tin alkoxide intermediate), (2)
wide usage of organotin compounds as catalysts for the industrial Lewis acid mechanism and (3) ionic mechanism. Of these, the inser-
synthesis of polyurethanes the subject has been widely investi- tion mechanism is the most accepted in literature [4,6]. In this
gated to understand the catalytic function in order to gain better mechanism the organotin compounds first undergoes alcoholysis
control of the reactions concerned [3–7]. to form an intermediate, which becomes the dominant catalyst in
Tin in organotin catalysts used in urethane synthesis has a the urethane reaction. The Lewis acid mechanism was proposed
valance state of four. The Sn(IV) is preferentially tetrahedrally coor- based on ebullioscopy data [10]. Less evidence exists for the ionic
dinated in organotin compounds. However, due to extra 5d orbitals mechanism [8] due to the higher catalytic action shown by this
in Sn4+ these compounds could form structures which contain more reaction in non-polar media. The above-mentioned mechanisms
than four coordinated tin [8]. Sometimes these compounds form are derived from indirect data analysis due to a lack of experimental
dimmeric or polymeric structures through extra coordination [9]. techniques that could monitor reactions during experimentation.
In the organotin carboxylate compounds the carboxylate ligand Computational methods have the ability to investigate interac-
such as acetic or lauric acid coordinates to tin forming bipyramidal tions between molecules and hence indicate possible mechanisms.
structures [8]. This approach gives a more accurate scenario for the interaction
than it can be obtained from experimental procedures. Most of
the work reported in literature on the application of computa-
∗ Corresponding author. Tel.: +64 9 373 923 9722; fax: +64 9 373 7422. tional methods for the investigation of isocyanate chemistry is
E-mail address: t.soehnel@auckland.ac.nz (T. Söhnel). for the uncatalyzed reaction between isocyanate and alcohol in

1381-1169/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molcata.2012.09.015

Please cite this article in press as: R. Devendra, et al., J. Mol. Catal. A: Chem. (2012), http://dx.doi.org/10.1016/j.molcata.2012.09.015
 ARTICLE IN PRESS
G Model
MOLCAA-8537; No. of Pages 14

2 R. Devendra et al. / Journal of Molecular Catalysis A: Chemical xxx (2012) xxx–xxx

Scheme 1. Reaction between isocyanate and alcohol to form urethane.

Scheme 2. Reaction between tributyltin methoxide and phenylisocyanate to form tributyltin carbamate.

urethane reactions. These have been done using semi-empirical program packages Gaussian09 [22] and Spartan10 [23]. All cal-
[12], ab initio [13] and DFT [14–16] methods. Investigations on culated energies are reported as electronic energies at 0 K unless
amine catalysis of urethane formation were only reported using otherwise specified. The values reported as Gibbs free energies and
PM3 semi-empirical [17] and DFT methods [18]. Investigation of the enthalpies includes zero point correction at 298 K. The influence
catalytic effect of organo-metallic, metal salts and metal chelates of solvents was obtained by using a CPCM polarizable conducting
using computational methods have received less attention. Here we model [24,25].
report computational work on some of the proposed mechanisms The computational simulations were done using model com-
related to organotin compounds as catalyst in urethane reaction. pounds. Dimethyltin derivatives were used as the model organotin
A new mechanism for urethane catalysis of organotin carboxylate compound and methanol was used as the alcohol and methyl iso-
in lower polar medium is proposed to explain some of the exper- cyanate or phenyl isocyanate were used as isocyanates. The reason
imental observation to get a better understanding of the catalytic to use these compounds was to reduce the computational costs.
mechanism. Table 1 shows all model compounds and the corresponding reac-
tants that are used in experiments and industrial applications. We
2. Experimental details assume that the behavior of these molecules will give us the correct
trends in behavior of actual compounds.
2.1. Computational details
2.2. Experimental work on reactivity of isocyanate and alcohol in
Interactions between molecules were simulated by bringing the the presence of organotin compound as catalyst
interacting centers close together by a step-by-step fashion while
optimizing the total energy of the system at each step. If the total Aliphatic and aromatic isocyanates were reacted with alcohol in
energy of the system went through a maximum, the interaction was various quantities in the presence of organotin compounds as cata-
considered to have different product to reactants. The structure at lyst in xylene. The total weights of the reaction mixtures were made
the maximum energy was considered as a structure that is close to to the same weight. The amounts of tin in the reaction mixtures
a transition state. were maintained at constant levels in all experiments. Samples
Once the near transition state structure for an interaction is were withdrawn at regular intervals and free isocyanate content in
identified it was optimized using the Berny algorithm [19]. The the reaction mix was determined according to the ASTM D2572-
transition state was verified by frequency analysis, which resulted 97 method [26]. In this method isocyanate first reacted with a
in one negative frequency in all cases. It was further verified known amount of dibutylamine then the isocyanate content was
related to products form and reactants using IRC (intrinsic reac- determined by titrating excess amine with standard solution of
tion coordinate) procedure using the transition state structure hydrochloric acid.
[20,21]. The configuration of the molecules on the product side was Desmodur N3390 (isocyanaurettrime of hexamethylenediiso-
used to investigate the subsequent interaction between molecules cyanate in 10% butyl acetate; Bayer Germany) and Desmodure
and search for corresponding transition states. This method was L75 (adduct of trimethylolpropane and 2,4-toluene diisocyanate
repeated until the final products were obtained. The energies of in 25% ethyl acetate; Bayer, Germany) were used as the aliphatic
molecular assemblies at minima were calculated as global minima
through optimization or single point energy calculation if it was a Table 1
selected point at the reaction path. Model compounds used to simulate reactants used in practice.
In the present work DFT (B3LYP and PBE functionals) and Reagent used in experiment Model compound
ab initio MP2 calculations were carried out with basis sets 6-
n-Butanol, polyol Methanol
31G*, 6-31G**, 6-31+G** for the light elements (C, N, O, H) and Dibutytin dilaurate (DBTDL) Dimethyltin diacetate (DMTDA)
LANL2DZ for Sn. However, most of the energy profile calculations Dibutytin diacetate (DBTDA) Dimethyltin diacetate (DMTDA)
were done on DFT/B3LYP/LANL2DZ/6-31G* level of theory in gas Dibutyltin dibutoxide (DBTDButoxide) Dimethyltin dimethoxide (DMTDM)
phase (unless otherwise stated) to keep the computational costs Aliphatic isocyanate Methyl isocyanate
Aromatic isocyanate Phenyl isocyanate
to a minimum. The calculations were performed by using the

Please cite this article in press as: R. Devendra, et al., J. Mol. Catal. A: Chem. (2012), http://dx.doi.org/10.1016/j.molcata.2012.09.015
 ARTICLE IN PRESS
G Model
MOLCAA-8537; No. of Pages 14

R. Devendra et al. / Journal of Molecular Catalysis A: Chemical xxx (2012) xxx–xxx 3

Scheme 3. Formation of tin alkoxide from alcoholysis of organotin diacetate.

Table 2
Energy differences in kJ mol−1 for the formation of tin alkoxide from alcoholysis of organotin diacetate as gas phase reaction and with toluene and water as solvents at 298 K.

LANL2DZ/6-31G(d)

B3LYP PBE MP2

Gas Toluene Water Gas Toluene Water Gas Toluene Water

H 26.775 26.221 25.559 23.386 26.838 33.568 26.612 28.998 26.182

G 24.829 24.346 24.273 33.358 19.916 32.369 25.906 23.772 29.938

LANL2DZ/6-311 + G(2df,2p)

B3LYP PBE

Gas Toluene Water Gas Toluene Water

H 32.395 31.424 30.560 38.278 37.118 36.081

G 28.534 27.486 26.539 33.847 32.699 31.649

and aromatic isocyanate, respectively. Reagent grade n-butanol, (1- The alcoholysis of DMTDA reaction shown in Scheme 3 was
methoxy-2-propyl)-acetate, dibutyltin diacetate, glacial acetic acid, simulated at DFT B3LYP/LANL2DZ/6-31G** level of theory. Because
xylene (all Merck, Germany) and dibutyltin dibutoxide (Gelest, inc. of the self-association of alcohol in non-polar solvents due to
USA) were used without further purification. hydrogen binding [12], a hydrogen bonded polymeric form of alco-
Each isocyanate was reacted with varying amounts of n-butanol hol (three molecules of methanol) was used in the calculations.
were reacted at 25 ◦ C in xylene as the solvent in the presence of Fig. 1 shows the energy profile of this interaction. The polymeric
selected catalysts. In this set of experiments the isocyanate amount form also reduces the bond strain during the reaction. From the
was maintained at 0.274 equivalents and varying amounts of n- energy profile of the interaction it can be seen that the total
butanol were reacted in the presence of 3.768 mg tin metal in the energy of the system does not go through an energy maximum
selected catalyst. The total volume of the starting reaction mix that would characterize a substitution nucleophilic bimolecular
was maintained at 182 mL and 172 mL for Desmodur N3390 and (SN2) type reaction. This would mean that the activation energy
Desmodur L75 systems, respectively. for the transfer of hydrogen from methanol to DMTDA to form a
methoxide–acedic acid complex reaction could be high, which will
3. Results and discussion not allow the reaction to take place at ambient temperature. The
optimized structure for the complex does not show a transition
3.1. Simulation of alkoxide formation of the hydrogen atom from methanol to DMTDA but a relatively
short Sn O distance to a methanol molecule has been formed
In literature, the formation of an alkoxide intermediate is instead (2.35 Å). That leads to a slight elongation of the Sn O dis-
proposed to be the active part of the catalyst for organotin(IV) di- tances of the carbonyl groups by about 0.1 Å. The free formation
substituted catalysis of urethane reactions [4,6]. Scheme 3 indicates energy for the complex obtained after optimization (Fig. 1, inlay)
the proposed alkoxide formation for organotin(IV) carboxylate
catalysis in urethane reaction.
Table 2 shows the Gibbs free energy and enthalpy calculated at
DFT (B3LYP and PBE, respectively) and MP2 level of theory for the
above reaction. The energies were also calculated using CPCM sol-
vent model water and toluene polar and non-polar solvents. They
were all calculated including zero point correction at 298 K. These
values are in reasonable agreement with each other and the pos-
itive values obtained for the reaction free energies and enthalpies
reveal that the reaction is thermodynamically unfavorable at ambi-
ent conditions. By using the LANL2DZ/6-311+G(2df,2p) basis sets
B3LYP and PBE results are within a range 2 kJ mol−1 for the gas
phase, water and toluene calculations (Table 2). At the LANL2DZ/6-
31G(d) level, B3LYP and MP2 calculations results are very similar,
therefore further MP2 calculations at the higher level have not been
conducted. These data also suggest that the influence of the solvent
is very minimal.
To simulate the interaction between organotin carboxylate and
alcohol was done by bringing together oxygen of the alcohol and
tin atom of the organotin carboxylate step by step while optimizing Fig. 1. Energy profile for the alcoholysis of dimethyltin diacetate (DMTDA) with
the energy. three molecules of methanol; optimized minimum confirmation shown as inlay.

Please cite this article in press as: R. Devendra, et al., J. Mol. Catal. A: Chem. (2012), http://dx.doi.org/10.1016/j.molcata.2012.09.015
 ARTICLE IN PRESS
G Model
MOLCAA-8537; No. of Pages 14

4 R. Devendra et al. / Journal of Molecular Catalysis A: Chemical xxx (2012) xxx–xxx

Table 3
Gibbs free energies and molecular structures of the two minima and the transition
state of the interaction between DMTDA and one single molecule of methanol at
298 K.

Index Molecular assembly Gibbs free

energy (au)

Minimum 1 −655.795672

Fig. 2. Energy profile of the interaction between DMTDA and one single molecule
of methanol.

is 14.2 kJ mol−1 starting from DMTDA and 3 non hydrogen-bonded

molecules methanol.
The simulation was also performed using toluene as no-polar
and water as polar solvents. These simulations showed similar
behaviors to the gas phase calculations. We also extended the
calculation to MP2 level of theory with LANL2DZ/6-31G** and TS1 −655.793535
LANL2DZ/6-31+G** basis sets. Both gave energy profiles close to
those of the DFT calculations. These investigations revel that the
alcoholysis step in the interaction between DMTDA and methanol
is less likely to happen at ambient conditions.
Similar calculations for other organotin compounds such
as DMTDmercaptide (dimethyltin dimercaptide) and DMTDC
(dimethyltin dichloride), which are commonly used as catalysts in
urethane synthesis, did not show a reaction that goes through an
energy maximum. Alcoholysis of DMTDA (dimethyltin diacetate)
using a single molecule of methanol also shows an energy min-
imum similar to the polymeric form due to the attraction of the
oxygen of methanol to the more positively charged tin atom. Fig. 2
shows the energy profile of this interaction and Table 3 shows the
structures of the molecules associated with the interaction. In this
Minimum 2 −655.794267
simulation DFT B3LYP LANL2DZ/6-31G* level of theory was used
(since in the rest of our investigations we used 6-31G* basis set for
light elements). However, by using DFT B3LYP/LANL2DZ/6-31G**
similar trends in the energy profile were observed as with the 6-
31G* basis set for the light elements. It is interesting to see that the
energy profile for the interaction between DMTDA and methanol
shows two minima, in contrast to the calculations including three
molecules of methanol (Fig. 1).
We also carried out simulations using PCMC solvent model in Level of theory DFT B3LYP/LANL2DZ/6-31G*.
toluene and water at DFT B3LYP LANL2DZ/6-31G* level of theory.
The solvent calculations resulted in similar trends for the energy
profiles as observed for the gas phase. MP2/LANL2DZ/6-31+G**
calculations gave very similar results to DFT results. The TS1 tran- OH group of methanol is associated with the Sn center due to elec-
sition state (Fig. 2) showed one negative frequency at −38 cm−1 . trostatic attraction (2.47 Å); in this complex the alcohol molecule
IRC calculations reproduced the starting and final products in the stays hydrogen-bonded with one of the carboxylic oxygen (1.67 Å).
interaction. Further to the above investigation we also carried out In this situation the chelating carboxylic acid ligand behaves as
the interaction between DMTDA and methanol with other higher a carboxylic salt. Figs. 3 and 4 show the theoretical IR spectra of
energy conformers of DMTDA structures (not reported here). These dimethyltin methoxyacetate and methanol complexed dimethyltin
simulations showed higher energy structures resembles to alkoxide diacetat, respectively.
formation when oxygen atom of the alcohol comes close to the tin Both, dimethyltin methoxyacetate and the DMTDA/methanol
atom. However, these results show less evidence for their existence complex show absorptions around 526 cm−1 . These absorptions
from practical observations, which will be discussed later section. are due to stretching of the Sn O bond as seen in the theoretical
The interaction between DMTDA and methanol is discussed next. spectra. This is in agreement with experimental data (611 cm−1 )
Fig. 2 shows that the interaction between DMTDA and methanol [27]. Fig. 4 (bottom) shows the FTIR-ATR absorption spectrum of a
has two energy minima. In the second minimum the oxygen of the mixture of DBTDA (1 mole) and n-butanol (1 mole). It can be seen

Please cite this article in press as: R. Devendra, et al., J. Mol. Catal. A: Chem. (2012), http://dx.doi.org/10.1016/j.molcata.2012.09.015
 ARTICLE IN PRESS
G Model
MOLCAA-8537; No. of Pages 14

R. Devendra et al. / Journal of Molecular Catalysis A: Chemical xxx (2012) xxx–xxx 5

Fig. 3. Theoretical IR spectrum of dimethyltin methoxyacetate.

Fig. 5. FTIR-ATR spectra of DBTDA + n-butanol (1:2) moles and DBTDbutox-

ide + acetic acid (1:2) moles (top); FTIR-ATR spectra (525–1000 cm−1 ) for different
compounds used in the analysis of Sn O bond formation (bottom).

that the experimental spectrum resembles the theoretical spectra

reasonably well (Fig. 4, top).

3.2. Interaction between various organotin compounds and

organic reagents

By mixing dibutyltin diacetate (DBTDA) and n-butanol in the

molar ratio of 1:2 no significant change in temperature could be
observed experimentally. However, mixing of dibutyltin dibutox-
ide (DBTDbutoxide) and acetic acid at the molar ratio of 1:2 resulted
in an exothermal reaction indicating the formation of dibutyltin
diacetate. The FTIR-ATR spectrum of the mix of dibutyltin dibutox-
ide and acetic acid at a molar ration 1:2 was similar to the spectrum
of a 1:2 mixture dibutyltin diacetate and n-butanol (Fig. 5). The
increase in the OH absorption around 3360 cm−1 in the dibutyltin
butoxide/acetic acid mixture indicates the formation of alcohol
[28], which confirms that the reaction is favored towards the forma-
tion of dibutyltin diacetate rather than to the dibutyltin butoxide
derivative. This is also confirmed by the absence of strong Sn O
band found in DBTDbutoxide. Fig. 5 also shows the FTIR-ATR spec-
tra around the Sn O stretching region for different compounds.
Fig. 4. Theoretical IR spectrum of hydrogen bonded methanol molecule to It can be seen that the DBTDA and n-butanol mix does not show
dimethyltin diacetate molecule (top); FTIR-ATR spectrum of a (1:1) mole mixture
of DBTDA and n-butanol (bottom).
absorption around the Sn O stretching frequency. This may indi-
cate the absence of the formation of tin alkoxide when organotin
carboxylate is mixed with alcohol.

Please cite this article in press as: R. Devendra, et al., J. Mol. Catal. A: Chem. (2012), http://dx.doi.org/10.1016/j.molcata.2012.09.015
 ARTICLE IN PRESS
G Model
MOLCAA-8537; No. of Pages 14

6 R. Devendra et al. / Journal of Molecular Catalysis A: Chemical xxx (2012) xxx–xxx

These results indicate that an initial formation of alkoxide from Based on the reported work on Lewis interactions the long-
an interaction between organotin carboxylate and alcohol is less range attraction of the negatively charged isocyanate oxygen and
likely. It can also be said that the subsequent steps suggested are the more electropositive tin atom of the organotin carboxylate
also less likely to happen in the mechanism proposed by Blood- were taken into consideration. Methyl isocyanate was placed direc-
worth and Davies [4] if tin alkoxide is present in an alcohol and ting the isocyanate oxygen towards the tin atom of the organotin
isocyanate containing environment. Possible interactions in a mix- carboxylate. Then the more positively charged isocyanate carbon
ture of isocyanate, alcohol and catalyst are discussed in the next atom, which shows an unoccupied molecular orbital, was inter-
section. acted with the highest energy occupied molecular orbital on the
The energy differences of the formation for selected complexes oxygen atom of the alcohol group. The two centers were brought
from starting materials suggested in literature [4] were calculated together step-by-step. At each step the total energy of the system
using DFT B3LYP/LANL2DZ/6-31G* level of theory at 298 K. Starting was minimized.
materials in the reaction mixture were taken as methanol, methyl In the absence of explanations for some of the experimentally
isocyanate and DMT methoxyacetate (dimethyltin methoxy- observed catalytic effects of organotin carboxylate in urethane
acetate). The selected complexes DMTmethoxyacetate/methyl reactions a new mechanism is proposed to explain the catalysis
isocyanate (the complex has two configurations i.e. O and N mechanism. In this model the initial interaction occurred due to the
coordinated), DMTmethoxyacetate/methanol and dimer of DMT- Lewis effect leading to the formation of a ternary complex by inter-
methoxyacetate show enthalpy differences of −7.2 kJ mol−1 action between the isocyanate and an alcohol complexed organotin
(O-coordinated), 8.8 kJ mol−1 (N-coordinated), −33.8 kJ mol−1 carboxylate. Similar ternary complexes are reported in literature
and −88.5 kJ mol−1 , respectively. However, Gibbs free energy [5,11].
calculations show 35.0 kJ mol−1 (O-coordinated), 42.8 kJ mol−1 (N-
coordinated) and 11.2 kJ mol−1 DMTmethoxy acetate/isocyanate
and DMTmethoxy acetate/methanol, respectively, but
3.3.1. Aliphatic primary isocyanate catalysis
−23.4 kJ mol−1 for the dimer of DMTmethoxy acetate. It indicates
Organotin carboxylate catalysts in an alcohol/isocyanate envi-
that DMTmethoxy acetate forms a more stable dimer compared to
ronment can exist in several molecular assemblies. The possibility
the rest. Fig. 6 shows the electrostatic isopotential maps for DMT-
of forming a complex between methanol and DMTDA has been
methoxy acetate and methyl isocyanate. It may be that the inter-
already discussed. In this complex the alcohol molecule is associ-
acting electrostatic fields interfere with the approaching methyl
ated with the DMTDA and the energy profile indicates the formation
isocyanate molecule when nitrogen of the isocyanate is directed
of a complexe at the first minimum and more stable complex
towards the tin center. This may prevent methyl isocyanate to coor-
at the second minimum. The two complexes show Gibbs free
dinate with the tin atom in spite of the electron pair on the nitrogen
energy and enthalpy differences of 18.6 kJ mol−1 , −19.7 kJ mol−1
atom has a higher polarizability than electrons on the oxygen.
and 22.1 kJ mol−1 , −30.6 kJ mol−1 , respectively, relative to the start-
Although in literature [4,6] it is suggested that once the organ-
ing material (calculated at B3LYP/LANL2DZ/6-31G* level of theory).
otin alkoxide is formed, the isocyanate will interact with organotin
The first complex is considered here since it could undergo fur-
alkoxide to form tin carbamate. From our calculations it is evident
ther reaction with isocyanate. Calculations for the first complex
that complex forming between isocyanate and tin alkoxide is less
also show that the HOMO is positioned on the hydrogen-bonded
probable than the complex forming between organotin alkoxide
methanol OH with 25.2 kJ mol−1 higher energy than the HOMO
and alcohol or dimerization of organotin alkoxide. Therefore, the
located on the unassociated methanol OH. A similar mechanism
organotin alkoxide isocyanate interaction in the urethane forma-
was proposed by Entelis et al. [29] on organotin catalysis of the
tion could proceed through an alternate path. However, since there
reaction between isocyanate and alcohol. The higher energy occu-
is less evidence of the formation of organotin alkoxide from organ-
pied molecular orbital on the OH group will interact with the lowest
otin dicarboxylate an alternate mechanism is proposed, which will
energy unoccupied molecular orbital of the incoming isocyanate
be discussed next.
where the isocyanate oxygen is directed towards the tin atom of
the organotin complex due to Lewis action. The isocyanate car-
3.3. Proposed new mechanism to depict the catalytic action of bon will become more electropositive due to Lewis interaction,
organotin carboxylate catalyst in urethane formation which will lead to the formation of a ternary complex. The struc-
ture of this ternary complex (index 1) is shown in Table 4. However,
To simulate the interaction between isocyanate and alcohol in DMTDA, methanol and methyl isocyanate can also form other com-
the presence of organotin compounds the following method was plexes. The molecular assemblies listed in Table 4 are considered as
adopted. the most probable interactions. To simplify the model used in the

Fig. 6. Electrostatic isopotential map for dimethyltin methoxy acetate and methyl isocyanate with the methyl isocyanate oxygen directed towards tin center (left); electrostatic
isopotential map for dimethyltin methoxy acetate and methyl isocyanate with the methyl isocyanate nitrogen directed towards tin center (right).

Please cite this article in press as: R. Devendra, et al., J. Mol. Catal. A: Chem. (2012), http://dx.doi.org/10.1016/j.molcata.2012.09.015
 ARTICLE IN PRESS
G Model
MOLCAA-8537; No. of Pages 14

R. Devendra et al. / Journal of Molecular Catalysis A: Chemical xxx (2012) xxx–xxx 7

Table 4 as discussed previously. These also will lead to similar subsequent

Gibbs free energies of the two possible molecular assemblies formed by the associ-
interaction between these molecules as discussed in the following
ation of DMTDA, methyl isocyanate and methanol at 298 K.
paragraph.
Index Molecular assembly Gibbs free In the molecular assembly the oxygen atom of the isocyanate
energy (au)
molecule is attracted towards the tin atom due to Lewis action,
which makes the isocyanate carbon more electropositive. This car-
bon then interacts with the oxygen atom of the alcohol. Fig. 7 shows
the energy profile with a possible reaction path for the DMTDA cat-
alyzed reaction between methyl isocyanate and methanol in the
gas phase. The transition state of the interaction between DMTDA,
1 −863.748520
methyl isocyanate and alcohol was determined by a using a coor-
dinates scan as explained in the previous section. In this work
DFT/B3LYP/LANL2DZ/6-31G* model chemistry was used. How-
ever, we have also used DFT B3LYP LANL2DZ/6-31+G** and MP2
LANL2DZ/6-31G* for the energy profile determination. All calcula-
tions resulted in very similar energy profiles.
As shown in Fig. 7, TS1 represents the transition state of
DMTaceate O-carbamate formation from the interaction of DMTDA,
methyl isocyanate and methanol. TS2 is the transition state of the
2 −863.747668 interaction between DMT acetate, O-methyl carbamate and acetic
acid that was formed in the previous reaction. TS3 is the transition
state of the interaction between DMT acetate O-methyl carbamate
and methanol, which is present in the reaction mix. This leads to
the formation of DMTmethoxy acetate and urethane. Similarly, TS4
is the transition state for the interaction between rearranged DMT
Level of theory DFT B3LYP/LANL2DZ/6-31G*.
acetate O-methyl carbamate, i.e. DMT acetate N-carbamate, and
methanol. It can be seen that this reaction has a higher activation
calculations it was assumed that alcohol and isocyanate are present energy compared to the interactions discussed previously. In this
in stoichiometric amounts. diagram the energy profile associated with the rearrangement is
The assemblies indexed as 1 and 2 (Table 4) show Gibbs energies not shown for clarity. TS5 is the reaction path for the interaction
and enthalpies of 48.8, −32.3 and 51.0, −40.8 kJ mol−1 , respec- between DMT acetate N-methyl carbamate and acetic acid, which
tively, obtained from reactions of DMTDA, methyl isocyante and forms DMTDA and urethane. However, in the simulation the acetic
methanol. These complexes have a negative enthalpy and posi- acid molecule tend to move away from the tin center during the
tive Gibbs energy of formation, showing the very strong entropic rearrangement, which makes it less probable for the reaction to
factor, which is observed in all supramolecular complexes and take place in the presence of large amount of unreacted alcohol
which can be described as a classical Curtin–Hammett situation. in the reaction mix. From Fig. 7 it is evident that the formation of
Both assemblies could exist due to their very small energy dif- DMTDA is energetically more favorable than the formation of DMT-
ferences (2.2 kJ mol−1 ). The index 1 assembly was selected for methoxy acetate, which means formation of DMTDA and urethane
further investigation because the molecular arrangement shows a is the most preferred reaction (position marked in Fig. 7; the molec-
possible interaction between methanol and isocyante to undergo ular assembly just after TS1 is shown as inlay). It can be seen that
urethane formation. The assembly index 2 was formed as a result hydrogen atom of the acetic acid molecule that left the DMTDA
of isocyanate interacting with the DMTDA methanol complex that molecule stays close to the nitrogen atom of the DMTacetate
formed at the second minimum. Ternary complexes similar to index O-methyl carbamate. This will lead to the formation of DMTDA.
1 could also be formed from higher energy conformers of DMTDA Since TS4 has a relatively high transition state energy it is possible

Fig. 7. Energy diagram for the DMTDA catalyzed reaction between methyl isocyanate and methanol; inlay: molecular assembly just after TS1.

Please cite this article in press as: R. Devendra, et al., J. Mol. Catal. A: Chem. (2012), http://dx.doi.org/10.1016/j.molcata.2012.09.015
 ARTICLE IN PRESS
G Model
MOLCAA-8537; No. of Pages 14

8 R. Devendra et al. / Journal of Molecular Catalysis A: Chemical xxx (2012) xxx–xxx

isocyanates. Fig. 9 shows the atomistic arrangement and the energy

diagram for the DMTDA catalyzed reaction between phenyl iso-
cyanate and methanol (the molecular assembly just after TS1 is
marked in Fig. 9). It can be seen that the acetic acid and the carba-
mate nitrogen are on the opposite side of the molecular assembly.
Therefore the interaction between these two centers will be much
less compared to the same of aliphatic system.
Since the acetic acid and the nitrogen of the oxygen-substituted
carbamate are in opposite sides of the molecular assembly
the oxygen-substituted carbamate could rearrange to nitrogen-
substituted carbamate or interact with alcohol in the mixture.
TS3 and TS4 are the transition states associated in the interaction
between DMTacetate O-phenylcarbamate and methanol. TS2 rep-
resents the transition state associated with the interaction between
DMTacetate, N-phenylcarbamate, and methanol. This transition
state has a higher energy compared to previously discussed reac-
tion path. TS5 is the transition state for the interaction between
DMTacetate N-phenylcarbamate and acetic acid. However, as seen
Fig. 8. Energy diagram for the DMTDA catalyzed reaction between methyl isocyante in the aliphatic system during the rearrangement of the carbamate
and methanol in non-polar (toluene) and polar (water) solvents. the de-attached acetic acid molecule tent to move away from the
tin center. This will make this reaction less probable in an envi-
to say that this reaction path is less probable in a medium rich in ronment with high concentration of alcohol. Since the amount of
alcohol concentration. The high concentration alcohol will prevent catalyst in the system is much less than the reactant, the acetic acid
formation of N substituted methyl-carbamate. Table S1 shows the will not come into the reaction after TS1. Therefore, only the inter-
corresponding transition state structures. action between the DMT acetate O-phenylcarbamate and alcohol
Reaction pathways for more probable reactions were calcu- was considered. Table S2 shows the corresponding transition state
lated using a CPCM solvent model at DFT B3LYP LANL2DZ/6-31G* structures.
level of theory. Fig. 8 shows the energy diagram for the Fig. 10 shows the energy diagram for the DMTDA catalyzed
selected reaction in non-polar (toluene) and polar (water) sol- reaction between phenyl isocyanate and methanol in non-polar
vents. The final energy difference for the starting material and (toluene) and polar (water) solvents. It can be seen that the polar
the products are different for toluene and water by 2.8 kJ mol−1 solvent shows a smaller energy change than the non-polar solvent
for the formation of DMTDA + urethane (−115.6 kJ mol−1 and with higher activation energy (70.7 kJ mol−1 and 65.8 kJ mol−1 ). It
112.8 kJ mol−1 , respectively) and 1.1 kJ mol−1 for DMTmethoxy can also be seen that DMTDA catalyzed overall reaction in polar
acetate + urethane (−86.8 kJ mol−1 and −85.7 kJ mol−1 , respec- solvents shows a lesser energy difference from reactant to product
tively). The gas phase calculation results in reaction enthalpies than in non-polar solvent (−86.0 kJ mol−1 and −85.3 kJ mol−1 ). This
of −119.4 kJ mol−1 (DMTDA + urethane) and −92.3 kJ mol−1 (DMT- indicates the possibility of slower reaction rate in polar solvents.
methoxy acetate + urethane). Similar effects are observed experimentally.
By comparing the atomic possitions of the two molecular essam-
3.3.2. Aromatic isocyanate catalysis blies shown in Figs. 7 and 9 it can be seen that the aliphatic
The mechanism for the interaction of phenyl isocyanates system has a higher digree of probability than the aromatic system
with organotin carboxylate catalysts is similar to that of methyl for the hydrogen transfer from carboxylic acid to tin carbamate

Fig. 9. Energy diagram for the DMTDA catalyzed reaction between phenyl isocyanate and methanol; inlay: molecular assembly of DMTDA catalyzed reaction between phenyl
isocyanate and methanol just after TS1.

Please cite this article in press as: R. Devendra, et al., J. Mol. Catal. A: Chem. (2012), http://dx.doi.org/10.1016/j.molcata.2012.09.015
 ARTICLE IN PRESS
G Model
MOLCAA-8537; No. of Pages 14

R. Devendra et al. / Journal of Molecular Catalysis A: Chemical xxx (2012) xxx–xxx 9

Fig. 11. Energy diagram for the dimethyltin dimethoxy catalyzed reaction between
Fig. 10. Energy diagram for the DMTDA catalyzed reaction between phenyl iso-
phenyl isocyanate and methanol.
cyanate and methanol.

nitrogen. This means the hydrogen transfer step of the aliphatic as gas phase reaction. It can be seen that after TS1 the reaction
system will be more effecient. It is also intererasting to note that becomes spontaneous. Table S3 shows the corresponding transition
the atomic charge on the nitrogen atoms of the tin carbamate after state structures. Fig. 12 shows the energy diagram for the reaction
the first transition state for the methyl isocyanate system has a in non-polar and polar solvents. From this figure it can be seen
slightly higher negative charge (−0.613; NBO) than the phenyl iso- that the energy difference for non-polar (toluene) solvent is slightly
cyante (−0.580; NBO). The higher negative charge should attract higher (1.3 kJ mol−1 ) than for polar (water) solvent, which indicates
the carboxylate hydrogen more effectively. This situation creastes that reaction will be marginally faster in non-polar solvents.
a condition where the hydrogen transfer will take place depending Next the catalytic effect of organotinalkoxy derivatives in
on the type of isocyanate used. Therefore it is possible for this reac- the urethane formation is discussed. The both forms, organotin
tion to follow two different routes depending on the environment dimethoxide and organotin methoxyacetate, will be considered
and the isocyanate as follows: in interactions with methyl isocyanate/methanol and phenyl iso-
cyanate/methanol.
Step 1 Transfer of the hydrogen to nitrogen of the carbamate group The organotin methoxyacetate will function as a catalyst to
to liberate the urethane molecule and then reforming of the form urethane by forming a ternary molecular assembly. The
catalyst by combining the organotin methoxy carboxylate ternary complex is formed from a organotin methoxy acetate with
with the carboxylic acid group (Scheme 4). methanol, which subsequently interact with a methyl isocyanate to
Step 2 Interaction of organotin carbamate(I) group to interact with give a ternary complex. It is interesting to see that the tin atom of
alcohol to form a tin alkoxide (Schemes 5 and 6). the alcohol complexed organotin methoxyacetate has higher pos-
itive charge (2.319; NBO) than the charge of the tin atom of the
As shown in step 1 the carboxylic acid will influence the catalysis un-complexed organotin methoxyacetate (2.283; NBO).
function by transferring hydrogen atom to nitrogen of the organ-
otin carbamate to liberate the urethane. In contrast, the carboxylic
acid will not influence the function of the catalysis and the subse-
quent steps will change the organotin carbamate(I) to an organotin
alkoxide by reacting with an alcohol molecule as shown in step 2.
This will convert all organo carboxylate derivatives to organotin
alkoxide derivatives.
In the subsequent section the conversion of organotin carba-
mate to organotin alkoxide derivatives will be discussed, which will
be the dominant catalyst for the aromatic isocyanate in urethane
formation.
Scheme 7 shows the interaction between organotin methoxy-
carbamate and alcohol to form urethane and the corresponding
organotin alkoxide derivatives. In this reaction organotin methoxy
O-phenylcarbamate was considered, because experimental inves-
tigations have shown that the catalytic reaction of DBTDA was
similar to DBTDbutoxide in xylene as discussed in subsequent sec-
tion. Based on this result we assume that organotin carboxylate
converts to organotin alkoxide in non-polar solvents.
Fig. 11 shows the energy diagram for the dimethyltin dialkox-
ide catalyzed reaction between phenyl isocyanate and methanol
to form urethane. In this diagram we have only considered the Fig. 12. Energy diagram for the dimethyltin dimethoxy catalyzed reaction between
reaction between dimethyltin methoxy O-carbamate and methanol phenyl isocyanate and methanol in non-polar (toluene) and polar (water) media.

Please cite this article in press as: R. Devendra, et al., J. Mol. Catal. A: Chem. (2012), http://dx.doi.org/10.1016/j.molcata.2012.09.015
 ARTICLE IN PRESS
G Model
MOLCAA-8537; No. of Pages 14

10 R. Devendra et al. / Journal of Molecular Catalysis A: Chemical xxx (2012) xxx–xxx

Table 5
Structures for different stages of the reaction between different isocyanate and
methanol with DMTmethoxy acetate and DMTDmethoxide used as catalyst.

Index Structure Molecular

assembly

1 Ternary
complex

2 Transition
state:
DMTmethoxy
Fig. 13. The energy profile of the reactions for urethane formation when DMT-
acetate + methanol
methoxy acetate; DMTDmethoxide is used as catalyst for the reaction between
+ methyl
different isocyanates and methanol.
isocyanate

Fig. 13 shows the energy profile for the formation of urethanes

from both methyl isocyanate and phenyl isocyanates by interac-
ting with methanol when organotin dimethoxide and organotin
methoxyacetate compounds are used as catalysts. Since the reac-
tion mixture contains methanol the catalysis is considered to take
place through alcohol complexed organotin alkoxide similar to the 3 Transition
state:
catalysis of urethane formation with DMTDA as discussed previ-
organotin
ously. The structures shown in Table 5 are for the DMTmethoxy carba-
acetate, methyl isocyanate and methanol products obtained during mate + methanol
the reaction. However, details of all species formed are not shown
in the energy diagram.
From the calculated energy profile it is evident that both aro-
matic and aliphatic isocyanates show similar trends in the urethane
reaction if these catalysts were used and if they were to follow the
same mechanism. We have also calculated the energy profiles by
4 DMTmethoxy
first considering the initial interaction being between the tin alkox-
acetate + urethane
ide and the isocyanate. The results show that the energy profile has
similar trends as before for both types of isocyanates. However, the
experimental results indicate different trends for aromatic and pri-
mary aliphatic isocyanates, which will be discussed in a subsequent
Level of theory DFT B3LYP/LANL2DZ/6-31G*.
section.

3.3.3. Prediction of reactivity between isocyanate and alcohol in

non-polar and polar media the reaction rate will increase when moving from non-polar to
Fig. 8 shows the reaction profile for the DMTDA catalyzed polar media for aliphatic systems due to the formation of DMT-
reaction between methyl isocyante and methanol in non-polar methoxy acetate.
(toluene) and polar (water) solvents at 0 K. The optimized min- However, for aromatic systems the reaction first follows con-
imum structures and transition states were used to calculate verting the original catalyst DMTDA to DMTmethoxy acetate as
differences in free energies at 298 K (including zero-point correc- explained previously. The catalytic action will only depend on the
tion). It is interesting to see that the free energy for reactants and activation free energy of the reaction. Calculations show that the
products for path 1, where the reaction regenerates the original activation free energies for TS1 in this reaction are 130.7 kJ mol−1
catalyst DMTDA, are −54.4 kJ mol−1 and −52.6 kJ mol−1 in toluene and 141.2 kJ mol−1 in non-polar and polar solvents. Similarly, if
and water, respectively. This means that the reaction rate should DMTDmethoxide is functioning as the catalyst the free energies
become less in polar media for path 1. However, for path 2, where of activation are 139.0 kJ mol−1 and 147.0 kJ mol−1 in non-polar
DMTmethoxy acetate is formed, free energies of −30.0 kJ mol−1 and and polar solvents. The free energies of activation for TS1, when
−28.3 kJ mol−1 for toluene and water media were obtained. This DMTDA is used as a catalyst, are 175.7 kJ mol−1 and 182.2 kJ mol−1
means that in polar media it seems more likely to form more DMT- in non-polar and polar media. This means that the reaction rate
methoxy acetate intermediate. The free energy of activation for TS1 may decrease in polar solvents.
in non-polar and polar media (by using DMTmethoxy acetate as cat- The reduction of the catalytic effect of organoalkyltin alkokxy
alyst) is 141.8 and 132.9 kJ mol−1 , respectively. By using DMTDA derivatives formed when organotin alkoxide is reacted with dif-
as catalyst the free energies of activation for TS1 are 164.9 and ferent molar quantities of carboxylic acid may be due to the poor
175.5 kJ mol−1 in non-polar and polar media, respectively. This solubility of the complex formed at transition state in non-polar
means using DMTmethoxy acetate as catalyst lowers the activation media for aliphatic isocyanates. The transition state has a higher
energy in comparison to using DMTDA as the catalyst. Therefore, charge separation for organotin alkoxy derivatives.

Please cite this article in press as: R. Devendra, et al., J. Mol. Catal. A: Chem. (2012), http://dx.doi.org/10.1016/j.molcata.2012.09.015
 ARTICLE IN PRESS
G Model
MOLCAA-8537; No. of Pages 14

R. Devendra et al. / Journal of Molecular Catalysis A: Chemical xxx (2012) xxx–xxx 11

Fig. 14. Depletion of NCO during the reaction of Desmodur N3390 and n-butanol. Fig. 16. Depletion of NCO for interaction between different molar amounts of n-
DBTDL – dibutyltin dilaurate, DBTDA – dibutyltin diacetate, BBTDbutoxide – butanol and Desmodur L75 using DBTDL as catalyst.
dibutyltin dibutoxide are used as catalysts.

isocyanate no significant influence of the carboxylic group content

in the catalyst can be observed (Fig. 15).
3.4. Experimental investigations on reactivity between different
Figs. 16 and 17 show the effect of alcohol concentration on
isocyanates and alcohol in the presence of organotin compounds
the reaction rate for the two types of isocyanates when an organ-
as catalyst
otin carboxylate is used as a catalyst. The reaction rate increases
with increasing alcohol concentration in the reaction mixture for
The following section will discuss the interaction of selected iso-
aromatic isocyanate. However, for aliphatic primary isocyanates
cyanate of primary aliphatic and aromatic nature, i.e. Desmodur
increasing alcohol concentration in the reaction mixture has a neg-
N3390 (isocyanaurettrime of hexamethylenediisocyanate in 10%
ative effect on the reaction rate as shown in Fig. 17.
butyl acetate) and Desmodure L75 (adduct of trimethylolpropane
The reduction in reaction rate for aliphatic isocyanate with
and 2,4-toluene diisocyanate in 25% ethyl acetate) with n-butanol
increasing alcohol concentration may be due to the association of
in the presence of organotin derivatives as catalyst in xylene as
the alcohol with DBTDL with excess alcohol will push the equi-
solvent.
librium to the second minimum (see Fig. 2). This will retard the
Figs. 14 and 15 show the depletion of the isocyanate group (NCO)
urethane formation due to the formation of the ternary complex
content of the urethane during the reaction of Desmodur N3390 and
that not suitable for urethane reaction as discussed before. In case
Desmodur L75, respectively, with n-butanol by using a set of differ-
of aromatic isocyanate with increasing alcohol concentration the
ent catalysts. It can be seen that for primary aliphatic isocyanates
conversion of organotin carbamate to organotin alkoxide and the
the carboxylic group content in the organotin catalyst has a sig-
urethane formation will increase.
nificant effect on the reaction rate (Fig. 14), whereas for aromatic
From Figs. 14–17 it is evident that the organotin carboxylate
catalysis of primary aliphatic and aromatic isocyanates in urethane

Fig. 15. Depletion of NCO during the reaction of Desmodur L75 and n-butanol.
DBTDL – dibutyltin dilaurate, DBTDA – dibutyltin diacetae, BBTDButoxide – Fig. 17. Depletion of NCO for interaction between different molar amounts of n-
dibutyltin dibutoxide are used as catalysts. butanol and Desmodur N3390 using DBTDL as catalyst.

Please cite this article in press as: R. Devendra, et al., J. Mol. Catal. A: Chem. (2012), http://dx.doi.org/10.1016/j.molcata.2012.09.015
 ARTICLE IN PRESS
G Model
MOLCAA-8537; No. of Pages 14

12 R. Devendra et al. / Journal of Molecular Catalysis A: Chemical xxx (2012) xxx–xxx

Scheme 4. Hydrogen transfer from carboxylic acid to nitrogen of the organotin carbamate for the aliphatic system.

Scheme 5. Rearrangement of organotin carbamate for the aromatic system.

Scheme 6. Interaction between alcohol and organotin acetate O phenyl carbamate.

Scheme 7. Interaction between alcohol and organotinmethoxy O phenyl carbamate.

Please cite this article in press as: R. Devendra, et al., J. Mol. Catal. A: Chem. (2012), http://dx.doi.org/10.1016/j.molcata.2012.09.015
 ARTICLE IN PRESS
G Model
MOLCAA-8537; No. of Pages 14

R. Devendra et al. / Journal of Molecular Catalysis A: Chemical xxx (2012) xxx–xxx 13

Aliphac alcohol Isocyanate

R2OH R1NCO
(R3)2Sn(OOCR4)(OCOR2NR1) + R4COOH

(R3)2Sn(OOCR4)2
Organon(IV) carboxylate Organon(IV) carbamate intermediate

R1NHCOOR2 Urethane

Scheme 8. Suggested mechanism for primary aliphatic isocyanate.

Aromac R2OH, R1NCO

(R3)2Sn(OOCR4)2 (R3)2Sn(OR 2)2 + R1NHCOOR2
iniaon
- R4COOH Organon(IV) alkoxide

R2OH R1NCO
Organon(IV) alkoxide
(R3)2Sn(OR2)(OCOR2NR1) + R2OH
(R3)2Sn(OR2)2
Organon(IV) carbamate intermediate
propagaon
Urethane R1NHCOOR2

R4COOH
(R3)2Sn(OR2)2 (R3)2Sn(OOCR4)2

Scheme 9. Suggested mechanism for aromatic isocyanate.

formation in a low polar medium takes place by two different (1-methoxy-2-propyl)-acetate, which is in agreement with results
mechanisms. It is also possible to say that for primary aliphatic obtained from computational work.
isocyanates in urethane formation with increasing carboxylic acid The reaction mechanism for primary aliphatic and aromatic
ligand content in the organotin catalyst will increase the reaction isocyanates in urethane formation in the presence of dialkyltin
rate. However, for aromatic isocyanates the carboxylic acid content dicarboxylate catalyst in non-polar media can be summarized as
does not significantly affect the reaction rate. shown in Schemes 8 and 9, respectively.
Based on computational and experimental results it can be con- The reaction mechanisms for organotin carboxylate catalysis of
cluded that the reaction between primary aliphatic isocyanate and secondary and tertiary aliphatic isocyantes in urethane formation
alcohol takes place as shown in Scheme 4 and for aromatic iso- will be discussed in a separate paper.
cyanatethe reaction will follow a mechanism where the initial
catalysis is due to organotin carboxylate as in Scheme 5 and sub-
sequently catalysis is due to organotin alkoxide that is formed
as an intermediate as suggested in Schemes 6 and 7. However,
the situation for primary aliphatic isocyanates may change with
increasing polarity in the medium, which will drive the mechanism
similar to what is shown in Scheme 5. The increase in polar-
ity will prevent the acetic acid molecule staying close to the tin
center.

3.5. The reactivity of aliphatic and aromatic isocyanate in

non-polar and polar solvents

Desmodur N3390 and Desmodur L75 was reacted with

n-butanol at stoichiometric amounts in xylene and (1-methoxy-2-
propyl)-acetate using DBTDL as catalyst. The reaction was carried
out as explained previously starting from constant volume for
both solvents. Figs. 18 and 19 show the depletion of NCO con-
tent in the reaction mix with time for Desmodur N3390 (aliphatic)
and Desmodur L75 (aromatic), respectively. From these figures it
can be seen that for aliphatic system reactivity increases and for Fig. 18. Depletion of NCO in the reaction between Desmodur N3390 and n-butanol
aromatic system the reactivity decrease changing from xylene to in different solvents.

Please cite this article in press as: R. Devendra, et al., J. Mol. Catal. A: Chem. (2012), http://dx.doi.org/10.1016/j.molcata.2012.09.015
 ARTICLE IN PRESS
G Model
MOLCAA-8537; No. of Pages 14

14 R. Devendra et al. / Journal of Molecular Catalysis A: Chemical xxx (2012) xxx–xxx

References

[1] M. Szycher, Szycher’s Handbook of Polyurethanes, 10, CRC Press, New York,
1999, pp. 1–20.
[2] J.W. Britain, P.G. Gemeinhart, J. Appl. Polym. Sci. 11 (1960) 207–211.
[3] K. Wonghamol, J.E. Kresta, Polyurethane ACS Symposium Series, Vol. 270, 1985,
pp. 111–121.
[4] A.J. Bloodworth, A.G. Davies, J. Chem. Soc. (1965) 5238–5244.
[5] J. Robins, J. Appl. Polym. Sci. 9 (1965) 821–838.
[6] R.P. Houghton, A.W. Malvaney, J. Organomet. Chem. 518 (1996) 21–27.
[7] F.W. van Der Weij, J. Polym. Sci., Polym. Chem. Ed. 19 (1981)
381–388.
[8] Tin Chemistry Fundamental, Frontiers and Applications, John Wiley & Sons,
West Sussex, United Kingdom, 2008, pp. 93–116, 269–283, 681–700.
[9] W.P. Neumann, The Organic Chemistry of Tin, John Wiley & Sons, New York,
USA, 1970, pp. 160–167.
[10] S.L. Reegen, K.C. Frisch, J. Polym. Sci. 8 (1970) 2883–2898.
[11] H.A. Smith, J. Appl. Polym. Sci. 7 (1963) 85–95.
[12] M. Hatanaka, Bull. Chem. Soc. Jpn. 82 (2009) 1149–1151.
[13] G. Raspoet, M.T. Nguyen, M. McGarraghy, A.F. Hegarty, J. Org. Chem. 63 (1998)
6878–6885.
[14] A.Ya. Samuilov, L.A. Zenitova, Ya.D. Samuilov, A.I. Koanovalov, Russ. J. Org.
Chem. 44 (2008) 1316–1322.
[15] A.Ya. Samuilov, L.A. Zenitova, Ya.D. Samuilov, A.I. Koanovalov, Russ. J. Org.
Chem. 45 (2009) 68–73.
[16] A.Ya. Samuilov, L.A. Zenitova, Ya.D. Samuilov, A.I. Koanovalov, Russ. J. Org.
Fig. 19. Depletion of NCO in the reaction between Desmodur L75 and n-butanol in Chem. 46 (2010) 1452–1460.
different solvent using DBTDL as catalyst. [17] N. Malwitz, CAChe Connection, 1, CAChe Scientific, P.O. Box 500, MS 13-400,
Beaverton, OR, 1992, 6–8.
[18] M. Hatanaka, Bull. Chem. Soc. Jpn. 84 (2011) 933–935.
[19] H.B. Schlegel, J. Comp. Chem. 3 (1982) 214–218.
4. Conclusions [20] H.P. Hratchian, H.B. Schlegel, J. Chem. Phys. 120 (2004) 9918–9924.
[21] H.P. Hratchian, H.B. Schlegel, J. Chem. Theory Comput. 1 (2005)
61–69.
Computational methods provide a better explanation of exper- [22] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman,
imental results for the organotin carboxylate catalysis of urethane G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H. Nakatsuji, M. Caricato,
X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M. Hada,
formation. Depending on the type of isocyanate used it is evident M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
that the carboxylic acid ligand contributes to the catalytic func- Kitao, H. Nakai, T. Vreven, J.A. Montgomery, J.E.J. Peralta, F. Ogliaro, M. Bearpark,
tion of the organotin dicarboxylate in urethane reactions in less J.J. Heyd, E. Brothers, K.N. Kudin, V.N. Staroverov, R. Kobayashi, J. Normand, K.
Raghavachari, A. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi, M. Cossi, N. Rega,
polar media. In such environments organotin carboxylate catalysis
J.M. Millam, M. Klene, J.E. Knox, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo,
of aliphatic isocyanates in urethane formation follows a different R. Gomperts, E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W.
mechanism to aromatic isocyanates. Aromatic isocyanates are cat- Ochterski, R.L. Martin, K. Morokuma, V.G. Zakrzewski, G.A. Voth, P. Salvador,
J.J. Dannenberg, S. Dapprich, A.D. Daniels, Ö. Farkas, J.B. Foresman, J.V. Ortiz, J.
alyzed by organotin alkoxide formed as an intermediate while for
Cioslowski, D.J. Fox, Gaussian 09, Gaussian Inc., Wallingford, CT, 2009.
aliphatic isocyanates the catalysis is mainly due to organotin car- [23] W.J. Hehre, Spartan‘10. A Guide to Molecular Mechanics Quantum Chemical
boxylate. Calculations, Wavefunction Inc., Irive, CA, 2006;
Y. Shao, L.F. Molnar, Y. Jung, J. Kussmann, C. Ochsenfeld, S.T. Brown, A.T.B.
Gilbert, L.V. Slipchenko, S.V. Levchenko, D.P. O’Neill, R.A. DiStasio Jr., R.C.
Acknowledgement Lochan, T. Wang, G.J.O. Beran, N.A. Besley, J.M. Herbert, C.Y. Lin, T. Van Voorhis,
S.H. Chien, A. Sodt, R.P. Steele, V.A. Rassolov, P.E. Maslen, P.P. Korambath, R.D.
Adamson, B. Austin, J. Baker, E.F.C. Byrd, H. Dachsel, R.J. Doerksen, A. Dreuw,
The authors thank Uroxsys Ltd., Auckland, New Zealand, for B.D. Dunietz, A.D. Dutoi, T.R. Furlani, S.R. Gwaltney, A. Heyden, S. Hirata, C.-P.
their support of this project. The authors are also very grateful for Hsu, G. Kedziora, R.Z. Khalliulin, P. Klunzinger, A.M. Lee, M.S. Lee, W.Z. Liang,
I. Lotan, N. Nair, B. Peters, E.I. Proynov, P.A. Pieniazek, Y.M. Rhee, J. Ritchie, E.
the support provided by the Computational Chemistry Group at
Rosta, C.D. Sherrill, A.C. Simmonett, J.E. Subotnik, H.L. Woodcock III, W. Zhang,
the School of Chemical Sciences, The University of Auckland and A.T. Bell, A.K. Chakraborty, D.M. Chipman, F.J. Keil, A. Warshel, W.J. Hehre, H.F.
the technical support teams at Gaussian and Wavefuction for their Schaefer, J. Kong, A.I. Krylov, P.M.W. Gill, M. Head-Gordon, Phys. Chem. Chem.
Phys. 8 (2006) 3172.
valuable help. The authors also thank the referees for their help in
[24] V. Barone, M. Cossi, J. Phys. Chem. A 102 (1998) 1995–2001.
correcting of the paper. [25] M. Cossi, N. Rega, G. Scalmani, V. Barone, J. Comp. Chem. 24 (2003)
669–681.
[26] ASTM-D2572-97, Standard Test Method for Isocyanate Group in urethane
Appendix A. Supplementary data Materials or Prepolymers, ASTM International, West Conshohocken, Pennsyl-
vania 19428-2959, United States.
[27] F.H. Butcher, W. Gerrard, E.F. Mooney, R.G. Rees, H.A. Willis, Spectrochim. Acta
Supplementary data associated with this article can be 20 (1961) 51–61.
found, in the online version, at http://dx.doi.org/10.1016/ [28] S.B. Rai, P.K. Srivastava, Spectrochim. Acta Part A 55 (1999) 2793–2800.
j.molcata.2012.09.015. [29] S.G. Entelis, O.V. Nesterov, V.B. Zabrodin, Kinet. Catal. 7 (1966) 627–631.

Please cite this article in press as: R. Devendra, et al., J. Mol. Catal. A: Chem. (2012), http://dx.doi.org/10.1016/j.molcata.2012.09.015