Cyclic guanidines as efficient organocatalysts for the synthesis of polyurethanes

Article abstract of DOI:10.1021/ma2026258

A systematic survey of basic/nucleophilic organocatalysts for the polyaddition in bulk of polyols, PEG-600, and PTMO-650, to isophorone diisocyanate (IPDI) has been performed. Guanidines were shown to be very efficient catalysts for the urethane linkage f

Full text of DOI:10.1021/ma2026258

Article
pubs.acs.org/Macromolecules
Cyclic Guanidines as Efficient Organocatalysts for the Synthesis of
Polyurethanes
Jerome Alsarraf,†,‡ Yacine Ait Ammar,†,‡,§ Frederic Robert,†,‡ Eric Cloutet,§,⊥ Henri Cramail,*,§,⊥
́
́
́
and Yannick Landais*,†,‡
†Univ. Bordeaux, ISM, UMR 5255, 33400 Talence, France
‡CNRS, ISM, UMR 5255, 33400 Talence, France
§Univ. Bordeaux, LCPO, UMR 5629, 33600 Pessac, France
⊥CNRS, LCPO, UMR 5629, 33600 Pessac, France
S
* Supporting Information
ABSTRACT: A systematic survey of basic/nucleophilic organo-
catalysts for the polyaddition in bulk of polyols, PEG-600, and
PTMO-650, to isophorone diisocyanate (IPDI) has been
performed. Guanidines were shown to be very efficient catalysts
for the urethane linkage formation. Bicyclic penta-alkylated
guanidines such as MTBD led to polyurethane molecular weight
and dispersity that are in the range of those observed with tin-
based catalysts such as DBTDL. Tetra-alkylated guanidine such
as TBD was shown to be a weaker catalyst as compared to pentaalkylated guanidines, as a result of its high reactivity toward
isocyanate, resulting in the formation of a less nucleophilic urea. Although the mechanism has not yet been firmly established,
these experiments suggest that a nucleophilic-catalysis mechanism, involving the attack of one of the nitrogen of the guanidine
onto the unsaturated system of the isocyanate, should not be totally ruled out with such strong Bronsted base catalysts.
̈
allowing the premixing of the monomers is thus continuing and
has concentrated a lot of efforts recently.8 In the course of our
investigations on the synthesis of biosourced polyurethanes,9
we studied the catalytic activities of several nitrogen-based
organocatalysts. Among the various small organic activators that
were tested, cyclic guanidines were shown to exhibit attractive
catalytic activities, rivaling with the commonly used dibutyltin
dilaurate (DBTDL). We provide here a preliminary account of
these studies, including the selection of the best catalysts using
polymerization in bulk and a comparison with closely related
amidines DBU and DBN.
INTRODUCTION
■
Polyurethanes constitute an important class of polymeric
materials, estimated around 5 wt % of the current world
polymer production1 with numerous technical applications
(foams, coatings, biomaterials, fibers, etc.). Polyurethanes are
usually prepared, based on the original discovery of Bayer2 by
the most straightforward route involving the addition of polyols
to polyisocyanates in the presence of a catalyst. Organometallic
reagents or tertiary amines are commonly employed to mediate
this polyaddition. Dibutyltin dilaurate (DBTDL) is the most
active catalyst currently in use,3 but environmental concerns
should lead, in not too distant a future, to a ban of such toxic
organometallic reagents.4 The presence of heavy metals in the
resulting polymer also has detrimental effects on the aging of
the final material. Recent efforts in polymer synthesis have thus
focused on the design of organocatalysts that could
advantageously replace metal-based catalysts.5 Small organic
molecules, including carbenes, thioureas, and guanidines
(TBD) were thus shown to drive various polymerizations,
enabling good selectivities, relatively high rates6 and excellent
functional group tolerance. Polyurethane synthesis is well-
known to be catalyzed by tertiary amines, such as 1,4-
diazabicyclo[2.2.2]octane (DABCO), 2,2′-bis-
(dimethylaminoethyl ether) (BDMAEE) or amidines such as
1,8-diazabicyclo[5.4.0]undec-7-ene (DBU).7 However, none of
these organocatalysts have permitted yet reaching the reaction
rates attained with tin reagents. The search for more efficient
catalysts that would also display delayed catalytic activity
RESULTS AND DISCUSSION
■
Screening of Organocatalysts in Polyurethane Syn-
thesis Using IPDI and PEG-600 (or PTMO-650) as
Precursors. The synthesis of PU from equimolar amounts of
commercially available isophorone diisocyanate (IPDI) and
dried poly(ethylene glycol) 600 (PEG-600) (Scheme 1) was
first studied using a panel of structurally and functionally
different basic/nucleophic organocatalysts 2−8 (1 mol %)
(Scheme 2). DABCO and DBU, standard catalysts for PU
synthesis, were thus studied, as well as amino-pyridine and
guanidines, including acyclic guanidines 7a,b and guanidine
dimer 8 (obtained by coupling TBD 2a with 1,4-bis-
Received: December 6, 2011
Revised: January 22, 2012
Published: February 21, 2012
© 2012 American Chemical Society
2249
dx.doi.org/10.1021/ma2026258 | Macromolecules 2012, 45, 2249−2256

Macromolecules
Article
Waymouth and Hedrick during catalyzed ring-opening
polymerization of lactides and lactones, where 2a was reported
to be more efficient than 2b.12 Acylic guanidines 7a,b were
slightly less reactive.13 It is worth noticing that the presence of a
free imino group in 7a had no effect on the rate of the reaction
(vide infra). Finally, the dimer 8 showed the same reactivity
than MTBD when used in only 0.5 mol %. PU characteristics,
with respect to molecular weight and molecular weight
distribution obtained from SEC analysis are given in Table 1.
These data of PU samples quenched with methanol after 18 h
clearly demonstrate the efficiency of the organocatalysts in
comparison to DBTDL 1 (entries 2−3). In all cases, PUs with
reasonable high molecular weights could be obtained, in
particular when mechanical stirring was used to overcome the
viscosity increase (entries 3, 6, 7, 9). PUs were carefully
analyzed through FT-IR spectroscopy (see Supporting
Information). Whatever the catalyst used, all PUs exhibit
similar molecular IR spectra without detectable isocyanurate,
allophanate or urea bands that prove the high selectivity of the
tested organocatalysts. It is finally worth adding that treatment
of polyurethanes with 2b in the presence of EtOH was tested
Scheme 1. Organocatalyzed Synthesis of Polyurethane (PU)
from IPDI and PEG-600
Scheme 2. Structure of Organocatalysts 2−11
1
and led to no noticeable changes in the H NMR spectrum,
demonstrating that guanidine 2b does not catalyze the
transurethanization reaction.
The scope of applications of the catalyzed PU synthesis was
then further extended to the polymerization of IPDI and
poly(tetramethylene oxide) (PTMO-650) restricting our study
to the most efficient amidine and guanidine catalysts (1 mol %)
(Scheme 3). As above, equimolar amounts of IPDI and PTMO-
650 were reacted in bulk at 60 °C in the presence of various
organocatalysts. Time course of the reaction, summarized in
Figure 2, confirm the general trend observed in the previous
polymerization experiment. While the activity of the tin catalyst
1 does not seem to be affected by traces of water present in
commercial PTMO-650 (entry 1, Table 2), such was not the
case with guanidines. Overlapping FT-IR spectra of PU
prepared using tin catalyst 1 and MTBD 2b showed no
significant difference, except a small enlargement of the band at
1716 cm−1 due to the presence of traces of urea (see
Supporting Information). This could be considerably reduced
through purification by azeotropic distillation (toluene) of
PTMO-650 before use. As indicated by experiments in entry 4
(vs entry 3), PU of higher molecular weights were repeatedly
obtained upon catalysis with 2b, when water was removed from
the diol prior to the reaction. Use of a slight excess of IPDI also
provided higher molecular weight (entry 5), proving that some
NCO function is generally lost through secondary reactions, in
particular reaction with water traces (leading to urea). MTBD
2b was found to be as active as DBTDL, leading to a complete
conversion after 1 h. Interestingly, 2b and its dimeric analogue
8 displayed much higher reactivity than the unsubstituted TBD
2a (entry 3 and 10 vs 2), in contrast with precedent in the
literature.10,11 The analogous DBN 4a (entry 8) showed a
surprising behavior with a very fast initial polymerization rate,
leading to more than 50% of the conversion of NCO functions
after 1 min and then a complete loss of activity, suggesting that
inhibition of the catalyst occurs very rapidly (vide infra). The
superbasic “proton sponge” but poorly nucleophilic naph-
thylbisguanidine (TMGN) 3,14 surprisingly led to a poor
catalytic activity. Considering the high efficiency of the cyclic
penta-alkyl guanidine catalysts 2a−c in the PU synthesis, new
bicyclic guanidines, having different ring sizes, were prepar-
ed12b,15 (Supporting Information) and tested in the con-
(bromomethyl)benzene) (Supporting Information). Dibutyltin
dilaurate 1 was used as a reference. IPDI and PEG-600 were
investigated as to perform the reaction in bulk at 60 °C, the low
viscosity of the polymer formed allowing aliquots sampling and
monitoring of the reaction using FT-IR. Reaction progress was
estimated by measuring the disappearance of the NCO and OH
stretching vibrations at 2252 and 3464 cm−1, respectively, and
the formation of the intense urethane CO bond at 1715
cm−1 as a function of time. Quantification was carried out,
relying on the unchanged absorption of CH2 bonds at 2867
cm−1. From these preliminary experiments, summarized in
Figure 1, most catalysts were found to be active in the
polyaddition of PEG-600 with IPDI, leading to PUs with
complete conversion after 1 h, all catalysts leading to 50%
conversion after less than 45 min. Amidines and guanidines
clearly emerged as superior catalysts. Strikingly, strongly basic
and nucleophilic guanidine MTBD 2b and amidine DBU 4b
were both found to be more active than DBTDL, leading to
complete conversion after only 15 min. They exhibit
remarkable catalytic activities as compared to other less basic,
hence nucleophilic heterocycles including substituted DMAP 5
and DABCO 6, which required almost 5 h to reach completion.
In contrast, guanidine 2a (TBD) which was recently reported
to be an excellent catalyst for polyester synthesis, acting
through an original dual activation,10 and for polyurethane
synthesis11 displayed poor activity as compared to N-alkylated
guanidine 2b. This behavior contrasts with the observations of
2250
dx.doi.org/10.1021/ma2026258 | Macromolecules 2012, 45, 2249−2256

Macromolecules
Article
Figure 1. Screening of various organocatalysts in the synthesis of PU from IPDI and PEG-600.
Table 1. Synthesis of Polyurethanes from Bulk Polyaddition
of IPDI and PEG-600 (Stoichiometric Ratio) in the
Presence of Organocatalysts 2−8 at 60°C
densation between IPDI and PTMO-650. The size of the ring
of the different guanidines was found to have an effect on the
polymerization rate. Guanidine 10 thus led to a lower activity
than the larger analogues 9a and 9b (entry 11−13), with
incomplete conversion of NCO moieties even after 3 h,
indicating that an amidyl functional group (N−CHN)
embedded in a 6- or 7-membered ring is required to get
efficient catalysis. As above, dimeric 8 at 0.5 mol % showed the
same reactivity than 2b. Highly basic phosphazene 1115 led to
reasonable catalytic activity, albeit lower than that observed
with guanidines (entry 14). Finally, the reaction afforded the
PU with complete conversion after 6 h at 60 °C using only 0.05
mol % of MTBD 2b, demonstrating the high efficiency of such
guanidines (entry 15). As already discussed with PEG-600 as
diol monomer, Table 2 below indicates that polymerization in
the presence of guanidines provide PUs having satisfying
molecular weights. These studies demonstrate unambiguously
that substituted guanidines exhibit unprecedented catalytic
activities in polymerization of diols and di-isocyanates, whatever
the nature of the monomer with catalyst loading as low as 0.05
mol %.
a
bc
,
bc
,
c
entry
catalyst
Mw (g mol−1
)
Mn (g mol−1
)
D = Mw/Mn
1
2
no catalyst
35 000
48 100
54 000
28 900
33 700
48 000
74 000
29 800
43 000
29 300
31 250
25 000
33 400
34 600
21 700
24 800
33 100
39 800
22 400
30 100
22 000
23 300
1.40
1.44
1.56
1.33
1.36
1.45
1.89
1.33
1.43
1.33
1.34
DBTDL 1
d
3
DBTDL 1
4
5
2a
2b
2b
4b
7a
7a
7b
d
6
d
7
8
d
9
10
11
e
8
a
b
1 mol % of catalyst was used. Aliquots of PU were taken after
quenching the reaction mixture with MeOH (after 18 h of stirring).
c
Estimated through SEC analysis onto PU samples quenched as above;
d
DMF as eluent with PS standards. Performed using a mechanical
stirrer. 0.5 mol % of dimeric catalyst 8 was used.
e
Preliminary Mechanistical Investigations. Two mecha-
nisms have been proposed to rationalize the course of the base/
nucleophile-catalyzed addition of alcohols onto isocyanates,
which are still subject to controversy. The first kinetic studies
by Baker et al.16 suggested that a nucleophilic catalysis was
involved, with the addition of the catalyst onto the NCO
functional group as a preliminary step (nucleophilic catalysis
mechanism). More recent studies17 showed that this mecha-
nism is not general and that activation of the alcohol by the
basic catalyst (general base catalysis mechanism) more
appropriately describes the addition of an alcohol onto an
Scheme 3. Organocatalyzed Synthesis of Polyurethane (PU)
from IPDI and PTMO-650
2251
dx.doi.org/10.1021/ma2026258 | Macromolecules 2012, 45, 2249−2256

Macromolecules
Article
Figure 2. Screening of various organocatalysts in the synthesis of PU from IPDI and PTMO-650.
Table 2. Polyaddition of IPDI and PTMO-650 using
organocatalysts 2a−c, 4a,b, 8−10
proposed. Finally, dual activation of the alcohol and the
isocyanate may also be invoked. Recent reports on this latter
mode of action have been reported, for instance in ROP of
lactides.10,19 In order to get better insights into the mechanism
of the guanidine-catalyzed addition of alcohols onto iso-
cyanates, a series of experiments with guanidine 2a−b was thus
performed using 2,3-dimethoxyphenethyl alcohol and benzyl
isocyanate (BnNCO) as model compounds in solution
(Supporting Information). For instance, 1H NMR of an
equimolar mixture of this alcohol model and 2b revealed
proton chemical shifts for the alcohol hydrogen, which
indicates that hydrogen bonded complex MTBD-2,3-dimethox-
yphenethyl alcohol might be involved at some stage
(Supporting Information).10,20 However, activation through
general base catalysis should also imply that the reaction rate
increases with the basicity of the catalyst.21 As shown above,
catalytic activities of organocatalysts follow the order: DABCO
6 < DMAP 5 (17.95) < TBD 2a (26.03) < TMGN 3 (25.4) <
BEMP 11 (27,5) < TMG 7a (23.3) ∼ BnTMG 7b < MTBD 2b
(25.43) ∼ DBU 4b (24.33), which does not fit with the
corresponding pKa’s (in CH3CN).21,22 It is thus worthy of note
that bis-guanidine 3 (TMGN) and phosphazene 11 led to poor
activity despite their strong basicity (Figure 2). The difference
in catalytic activity between TMGN 3 and MTBD 2b, which
exhibit similar pKa’s (25.4 and 25.43 respectively) during
polymerization of IPDI and PTMO-650 (Figure 2), is also
striking, suggesting that other mechanism(s) than general base
catalysis may operate.10 Additional experiments led us speculate
that a nucleophilic mechanism should not be completely ruled
out (vide infra). For instance, the low nucleophilicity of 3, due
to steric hindrance around the nitrogen centers would explain
its low catalytic activity, supporting such a mechanism. Figures
convn
b
Mw
Mn
D =
d
a
cd
,
cd
,
entry
catalyst
(%)
(g mol−1
)
(g mol−1
)
Mw/Mn
e
1
DBTDL 1
>98
82
93 000
21 000
36 000
50 200
85 000
42 700
28 250
-
67 900
14 200
26 700
32 600
58 600
29 400
18 800
-
1.37
1.48
1.35
1.54
1.45
1.45
1.50
-
e
2
2a
2b
2b
2b
2c
3
e
3
>98
>98
>98
>98
>98
86
f
4
g
5
e
6
e
7
e
8
4a
4b
e
9
>98
>98
>98
>98
>98
>98
>98
31 500
32 300
41 600
50 400
33 800
29 600
55 000
20 700
21 500
27 500
32 500
21 900
19 500
34 800
1.52
1.50
1.51
1.55
1.54
1.52
1.58
e
h
10
8
e
11
9a
9b
10
11
e
12
e
13
e
14
e
i
15
2b
a
b
1 mol % of catalyst was used. Estimated by measuring, using FT-IR,
c
the disappearance of the isocyanate NC = O band. Aliquots of PU
were taken after quenching the reaction mixture with MeOH (after 18
d
h of stirring). Estimated through SEC analysis; DMF as eluent with
PS standards. The PTMO-650 used was not dried before polymer-
ization. The PTMO-650 used was carefully dried through azeotropic
distillation in toluene before polymerization. 1.1 equiv of IPDI was
e
f
g
h
used for 1 equiv of commercial PTMO-650. 0.5 mol % of 8 was used.
i
0.05 mol % of 2b was used.
isocyanate.18 Several refinements on the latter mechanism
(complete proton transfer or concerted single step proton
transfer followed by nucleophilic attack) have also been
2252
dx.doi.org/10.1021/ma2026258 | Macromolecules 2012, 45, 2249−2256

Macromolecules
Article
1 and 2 also show that TBD 2a, which is more basic than
MTBD 2b and TMG 7a, is comparatively a weaker catalyst.
This was ascribed to the ability of 2a to react with isocyanates.
We thus observed that reaction of a 1:1 mixture of 2a and
BnNCO, led to the rapid formation of the corresponding urea
12 (Scheme 4).23 Interestingly, the latter was shown to catalyze
Scheme 5. Reactions of 2b and 4a,b with BnNCO
Scheme 4. Reaction of TBD 2a with Benzylisocyanate
the formation of polyurethanes from IPDI and PTMO-650 at
60 °C (Figure 3), with the same efficiency than TBD 2a alone,
suggesting that the formation of PU is probably catalyzed by an
urea such as 12 and not by 2a.11,13 This also indicates that
guanidines are likely to react with isocyanates in the presence of
alcohols in the medium (vide infra). The reduced catalytic
activity of 2a as compared to the alkylated analogue 2b, would
be due to the presence of the electron-withdrawing C(
O)NHR moiety, as in 12, which likely decreases the electron
density on the guanidine framework, reducing both the basicity
and the nucleophilicity of 2a. The high reactivity of TBD
toward electrophiles (including CO2)24 also implies that
impurities in the medium might reduce its catalytic activity.
Repeating the polymerization using freshly sublimated TBD 2a
however led to similar results than those obtained with
unpurified commercial TBD (Figure 3).
Finally, reasoning that amidines 4a,b might catalyze the
above reaction through a mechanism similar to that of
guanidines, the reactivity of 4a,b toward isocyanates was
studied, treating both amidines with BnNCO. Pleasingly, this
led to the formation, in good yields, of tricyclic compounds
13a,b incorporating two molecules of BnNCO (Scheme 5).25
Similarly, guanidine 2b provided under the same conditions,
compound 14 in high yield.26 Interestingly, when 13a, 13b, and
14 were left in THF under reflux in the presence of a primary
alcohol such as 3,4-dimethoxyphenethyl alcohol used pre-
viously, they exhibited contrasting behavior. While 13a and 13b
were stable under these conditions, 14 returned guanidine 2b
along with the corresponding carbamate (Supporting Informa-
tion). The rapid deactivation of 4a during reaction of IPDI and
PTMO-650 (Figure 2) might thus be explained by the
irreversible formation in the medium of an adduct between
4a and IPDI, similar to 13a. The difference in reactivity
between DBN 4a and DBU 4b (Figure 2) is striking and may
be rationalized invoking a much higher reactivity of 4a toward
isocyanates as compared to 4b. Although calorimetric experi-
ments have not been performed, we noticed that addition of
BnNCO onto 4a was much more exothermic than that with 4b.
As the formation of 13a is not reversible, no free amidine 4a
remains in the medium during polymerization, thus explaining
its in situ deactivation after less than 1 min. This also shows that
the free amidine is the real catalyst in this case, in contrast to
TBD 2a where the urea adduct, 12, catalyzes the reaction. This
difference in reactivity between closely related bicyclic amidines
points out again the importance of the ring size in these bicyclic
organocatalysts. The difference in stability between adducts 12
and 14 issued from guanidines and 13a,b generated from
amidines is also of interest and suggests that cyclic N-
alkylguanidines such as 2b should be superior and selective
Figure 3. Organocatalyzed polymerization of IPDI and PTMO-650 in the presence of 2a and urea 12.
2253
dx.doi.org/10.1021/ma2026258 | Macromolecules 2012, 45, 2249−2256

Macromolecules
Article
catalysts, whatever the reactivity of isocyanates, as their adducts
form reversibly, thus allowing the catalysis, which is not the
case with amidine adducts such as 13a,b.
In order to obtain further mechanistical insights, we finally
varied the order of addition of the reagents and catalysts (2b)
and compare these new conditions (conditions B and C, Figure
4) with those used previously (conditions A, Figure 1 and 2).
pairs are properly aligned to favor lone pairs delocalization.27
Preliminary mechanistical investigations provide some original
and unexpected features of this catalysis with the isolation of
adducts issued from the nucleophilic addition of guanidines
onto isocyanates. Similar behavior was observed with amidines.
Although the presence of these adducts in the medium in the
presence of the alcohol partner has not been firmly proved
(vide supra), these investigations suggest that the nucleophilic
catalysis mechanism should not be a priori ruled out with these
strongly basic catalysts. This would involve an initial
nucleophilic attack of the guanidine onto the isocyanate C
O functional group28,29 (nucleophilic monomer activation),
followed by the reaction of the activated zwitterionic acyl
intermediate with the alcohol partner to provide the urethane
linkage (Figure 5). Activation of the alcohol by a general base
Figure 4. Organocatalyzed polymerization of IPDI and PTMO-650
varying the order of addition. Condition A: IPDI and PTMO-650 are
premixed at 20 °C, then MTBD 2b is added, and the reaction mixture
is heated to 60 °C. Condition B: catalyst 2b is premixed with IPDI at
20 °C before the addition of PTMO-650. The reaction mixture is then
heated to 60 °C. Condition C: catalyst 2b is premixed with PTMO-
650 at 20 °C before the addition of IPDI. The reaction mixture is then
heated to 60 °C.
Figure 5. Catalytic cycles for the formation of the urethane linkage
using guanidines as organocatalysts.
catalysis mechanism may also operate as suggested by 1H NMR
studies on model alcohols. It is also worth adding that the order
of addition of the different partners might also influence the
mechanistical outcome of the process as shown by experiments
summarized in Figure 4 and Table 3. Further kinetic
investigations along with theoretical studies at the DFT level
will thus be needed to definitely establish the mechanism of
these guanidine-catalyzed PU synthesis. Studies along these
lines are now underway in our laboratory.
For instance, when guanidine 2b was premixed with IPDI
before the addition of the alcohol (condition B), one observes a
rather slow initiation period, that may be attributed to the in
situ formation of an adduct such as 14, which decomposition
would be necessary to restore the catalytic activity. In turn,
when 2b was premixed with the alcohol (PTMO, conditions
C), the reaction was found to be slightly faster as compared to
conditions A and also led to PU with higher molecular weight
(Table 3). This may be explained by an H-bonding activation
CONCLUSION
■
We have reported along these lines the efficiency of cyclic
guanidines as catalysts in the polyaddition of diols onto
diisocyanates. Relatively high molecular weight polyurethanes
with a good control of the molecular structure are obtained
without the formation of allophanates or isocyanurates
moieties. A straightforward preparation of guanidines has
been devised enabling an easy access to a wide range of new
guanidine structures. Elaboration of new guanidines based on
readily available TBD 2a and synthesis of 5-membered ring
guanidines such as 9 and 10 offers an alternative route to
further extend the screening. Efforts are now focusing on
kinetic studies to get insights into the mechanism of this
catalyzed PU synthesis and on the design of novel and more
efficient catalysts.
Table 3. Organocatalyzed Polymerization of IPDI and
PTMO-650 Varying the Order of Addition
convn
a
Mw
Mn
D =
c
bc
,
bc
,
conditions
(%)
(g mol−1
)
(g mol−1
)
Mw/Mn
conditions A
conditions B
conditions C
>98
>98
>98
36 000
28 700
61 300
26 700
18 400
38 800
1.35
1.56
1.58
a
Estimated by measuring, using FT-IR, the disappearance of the
b
isocyanate NCO band. Aliquots of PU were taken after
quenching the reaction mixture with MeOH (after 18 h of stirring).
c
Estimated through SEC analysis; DMF as eluent with PS standards.
of the alcohol by the guanidine (similar to the MTBD-3,4-
phenethyl alcohol complex above), that would accelerate the
addition of the alcohol onto the isocyanate, and in the same
time, prevent the formation of an adduct such as 14 mentioned
above.
In summary, we have shown that addition of diols onto
diisocyanates is efficiently catalyzed by guanidines, leading to
polyurethanes with high molecular weight. Cyclic guanidines
are slightly more reactive than acyclic ones as in the former, the
ring constrains the nitrogen into a conformation where the lone
EXPERIMENTAL SECTION
■
General Information. All reactions were carried out under an
argon atmosphere with dry solvents under anhydrous conditions,
unless otherwise noted. Yields refer to chromatographically and
spectroscopically (1H and 13C NMR) homogeneous materials, unless
otherwise stated. Commercial reagents were used without further
purification, unless otherwise stated. H NMR and 13C NMR were
1
recorded on a Bruker AC-300 FT and (1H: 300 MHz, 13C: 75.46
̈
2254
dx.doi.org/10.1021/ma2026258 | Macromolecules 2012, 45, 2249−2256

Macromolecules
Article
MHz) and a Bruker ARX-400 FT (1H: 400 MHz, 13C: 100.6 MHz)
̈
This material is available free of charge via the Internet at
http://pubs.acs.org
using CDCl3 as internal reference unless otherwise indicated. The
chemical shifts (δ) and coupling constants (J) are expressed in ppm
and Hz, respectively. The following abbreviations were used to explain
the multiplicities: s = singlet, d = doublet, t = triplet, q = quartet, quint
= quintet, m = multiplet, br = broad. IR spectra were recorded on a
Perkin-Elmer 1710 spectrophotometer, on a Perkin-Elmer Paragon
500 FT-IR spectrophotometer or on a Perkin-Elmer Mattson Unicam
500 16PC FT-IR using a ZnSe crystal ATR accessory. High resolution
mass spectra (HRMS) were recorded with a Q-TOF 2 spectrometer in
the electrospray ionization (ESI) mode. Melting points were not
AUTHOR INFORMATION
Corresponding Author
■
*(H.C.) E-mail: cramail@enscbp.fr. Fax: (+ 33) 05 40 00 84
87. (Y.L.) E-mail: y.landais@ism.u-bordeaux1.fr. Fax: (+ 33) 05
40 00 62 86.
Notes
The authors declare no competing financial interest.
corrected and determined by using a Buchi Totolli apparatus. Merk
̈
silica gel 60 (70−230 mesh) was used for flash chromatography. Size
exclusion chromatography (SEC) analyses were performed at room
temperature in DMF at 80 °C with a setup consisting of a PL-GPC 50
plus Integrated GPC from Polymer laboratories-Varian and a series of
three columns PLgel 5 μm MIXED-D. The elution of the filtered
samples was monitored using simultaneous UV and refractive index
detections. The elution times were converted to molar mass using a
calibration curve based on low dispersity (Mw/Mn) polystyrene (PS)
standards.
General Procedure for the Synthesis of Polyurethanes. IPDI
(0.8 mL, 3.78 mmol) was added to PTMO-650 (2.45 g, 3.78 mmol) at
room temperature. A solution of catalyst (75.5 μmol, 0.02 mmol) in
THF (0.5 mL) was then added and the mixture stirred at 60 °C.
Aliquots were taken at various period of time and FT-IR spectra
recorded to monitor the time course of the reaction (see Figure 2 and
Supporting Information). After 18 h, the reaction mixture was
quenched with methanol and the polymer analyzed using SEC (Table
2).
Synthesis of Urea 12. Benzylisocyanate (200 μL, 1.63 mmol) was
added dropwise to a solution of TBD 2a (227 mg, 1.63 mmol) in THF
(6 mL) leading to an exothermic reaction. The homogeneous reaction
mixture was stirred 1 h at room temperature (ca. 17 °C). The solvent
was then removed under vacuum affording a viscous oil. A small
amount of ether was then added to precipitate traces of unreacted 2a.
The solid was removed by filtration and the filtrate was concentrated
under vacuum to afford the expected product 12 as a pale yellow oil
(397 mg, 89%). IR (neat): υ = 2932, 2852, 1664, 1543, 698 cm−1. 1H
NMR (C6D6, 300 MHz): δ = 12.2 (br s, 1H), 7.44−7.37 (m, 2H),
7.19−7.09 (m, 2H), 7.07−6.99 (m, 1H),4.68 (d, J = 5.6 Hz, 2H),
3.83−3.75 (m, 2H), 3.19 (t, J = 5.7 Hz, 2H), 2.42 (t, J = 6.2 Hz, 2H),
2.23 (t, J = 6.5 Hz, 2H), 1.29 (qt, J = 6.4 Hz, 2H), 1.25−1.15 ppm (m,
2H). 13C NMR (C6D6, 75 MHz): δ = 157.0, 149.8, 141.2, 127.9, 126.8,
49.0, 48.4, 44.7, 43.2, 40.1, 22.8, 22.2 ppm. HRMS (ESI): m/z calcd
for C15H21N4O [M + H]+, 273.1715; found, 273.1712.
General Procedure for the Synthesis of Adducts 13a,b, 14.
Benzylisocyanate (1.9 mmol) was added dropwise to a solution of
amidine (or guanidine) (1 mmol) in Et2O (1 mL) at 0 °C. The
homogeneous reaction mixture was then stirred 5 min at 0 °C. After
recrystallization of the crude mixture at −40 °C, the crystals were
filtered and washed with a small amount of ether affording the
expected compound as colorless solid.
ACKNOWLEDGMENTS
■
We thank the French “Agence Nationale de la Recherche”
(ANR-09-CP2D-15), and the French Ministry of Research and
Technology for financial support.
REFERENCES
■
(1) (a) Oertel, G. Polyurethane Handbook; Hanser Publishers:
Munich, Germany, 1985. (b) Wirpska, Z. Poly(urethane)s: Chemistry,
Technology, and Application: Ellis Horwood: London, 1993. (c) Becker,
R. L. Thiele, Isocyanate reactions in Polyurethane catalysis: CRC Press,
Inc.: Boca Raton, FL, 1996; p 6940. (d) Hepburn, C. Polyurethane
Elastomers; Springer: Berlin, 1992.
(2) (a) Bayer, O. Angew. Chem. 1947, 59, 257−272. (b) Bayer, O.;
Muller, E. Angew. Chem. 1960, 72, 934−939.
̈
(3) (a) Lipatova, T. E.; Bakalo, L. A.; Niselsky, Yu N.; Sirotinskaya, A.
L. J. Macromol. Sci.Chem. 1970, A4 (8), 1743−1758. (b) Frisch, K.
C.; Rumao, L. P. J. Macromol. Sci.Rev. Macromol. Chem. 1970, 5c,
103−149. (c) Luo, S. G.; Tan, H. M.; Zhang, J. G.; Wu, Y. J.; Pei, F.
K.; Meng, X. H. J. Appl. Polymer Sci. 1997, 65, 1217−1225.
(d) Majundar, K. K.; Kundu, A.; Das, I.; Roy, S. Appl. Organomet.
Chem. 2000, 14, 79−85.
(4) For an overview see: (a) Toxicological Profile for Tin and Tin
compounds, ATSDR (Agency for Toxic Substances and Disease
Registry) , 2005. (b) Revised assessment of the risks to health and the
environment associated with the use of the four organotin compounds
TBT, DBT, DOT and TPT, The Scientific Committee on Health and
Environmental Risks (SCHER), 2006.
(5) (a) Kamber, N. E.; Jeong, W.; Waymouth, R. M.; Pratt, R. C.;
Lohmeijer, B. G. G.; Hedrick, J. L. Chem. Rev. 2007, 107, 5813−5840.
(b) Raynaud, J.; Ciolino, A.; Baceiredo, A.; Destarac, M.; Bonnette, F.;
Kato, T.; Gnanou, Y.; Taton, D. Angew. Chem., Int. Ed. 2008, 47,
5390−5393. (d) Raynaud, J.; Gnanou, Y.; Taton, D. Macromolecules
2009, 42, 5996−6005.
(6) For the observations of similar reaction rates between organo-
and metal-based catalysts, see: McLaughlin, M.; Garcia Rubio, S.;
Muthyala, R.; Antunes, O. A. C.; Tilstam, U.; Zlota, A.; Yadav, G. D.;
Laird, T. Org. Process Res. Dev. 2006, 10, 853−865.
(7) Silva, A. L.; Bordado, J. C. Catal. Rev. 2004, 46, 31−51.
(8) (a) Bantu, B.; Manohar Pawar, G.; Decker, U.; Wurst, K.;
Schmidt, A. M.; Buchmeiser, M. R. Chem.Eur. J. 2009, 15, 3103−
3109. (b) Bantu, B.; Manohar Pawar, G.; Wurst, K.; Decker, U.;
Schmidt, A. M.; Buchmeiser, M. R. Eur. J. Inorg. Chem. 2009, 15,
1970−1976.
(9) (a) Cramail, H.; Boyer, A.; Cloutet, E.; Bakhiyi, R.; Alfos, C. WO
2011030076, FR2950051. (b) Cramail, H.; Boyer, A.; Cloutet, E.;
Alfos, C. WO 2011030075, FR2950052. (c) Cramail, H.; Boyer, A.;
Cloutet, E.; Alfos, C. WO 2011045536, FR2951166. (d) Palaskar, D.
V.; Boyer, A.; Cloutet, E.; Alfos, C.; Cramail, H. Biomacromolecules
2010, 11, 1202−1211. (e) Boyer, A.; Cloutet, E.; Tassaing, T.;
Gadenne, B.; Alfos, C.; Cramail, H. Green Chem. 2010, 12, 2205−
2213. (f) Palaskar, D. V.; Boyer, A.; Cloutet, E.; Le Meins, J. F.;
Gadenne, B.; Alfos, C.; Farcet, C.; Cramail, H. J. Polym. Sci., Part A:
Polym. Chem. 2012, DOI: 10.1002.
MTBD-2 BnNCO Adduct (14). Mp = 96−97 °C (THF/pentane).
IR (neat): υ = 2925, 1683, 1645, 1470, 1377, 751, 700 cm−1. 1H NMR
(C6D6, 300 MHz): δ = 7.79−7.73 (m, 2H), 7.56−7.50 (m, 2H), 7.25−
7.01 (m, 6H), 5.31 (dd, J = 11.7 Hz, J = 13.7 Hz, 2H), 5.06 (d, J = 14.1
Hz, 1H), 4.63−4.52 (m, 1H), 4.35 (d, J = 14.1 Hz, 1H), 2.55−2.42
(m, 2H), 2.30−2.08 (m, 4H), 1.79 (s, 3H), 1.72−1.47 (m, 2H), 1.25−
1.08 ppm (m, 3H). 13C NMR (C6D6, 75 MHz): δ = 153.7, 151.4,
140.5, 139.1, 130.0, 129.4, 128.5, 127.5, 126.9, 97.3, 47.8, 46.8, 45.7,
45.1, 42.2, 36.4, 35.7, 23.3, 20.5 ppm. HRMS (ESI): m/z calcd for
C24H30N5O2 [M + H]+, 420.23995; found, 420.2396.
ASSOCIATED CONTENT
■
S
* Supporting Information
General experimental procedures and information, synthesis
(10) (a) Lohmeijer, B. G. G.; Pratt, R. C.; Leibfarth, F.; Logan, J. W.;
Long, D. A.; Dove, A. P.; Nederberg, F.; Choi, J.; Wade, C.;
Waymouth, R. M.; Hedrick, J. L. Macromolecules 2006, 39, 8574−
and spectroscopic data for guanidines 9a,b, 10, time course
1
experiments, and the corresponding FT-IR and H NMR data.
2255
dx.doi.org/10.1021/ma2026258 | Macromolecules 2012, 45, 2249−2256

Macromolecules
Article
8583. (b) Chuma, A.; Horn, H. W.; Swope, W. C.; Pratt, R. C.; Zhang,
L.; Lohmeijer, B. G. G.; Wade, C. G.; Waymouth, R. M.; Hedrick, J. L.;
Rice, J. E. J. Am. Chem. Soc. 2008, 130, 6749−6754.
(11) Tang, D.; Mulder, D.-J.; Noordover, B. A. J.; Koning, C. E.
Macromol. Rapid Commun. 2011, 32, 1379−1385.
(12) (a) Zhang, L.; Pratt, R. C.; Nederberg, F.; Horn, H. W.; Rice, J.
E.; Waymouth, R. M.; Wade, C. G.; Hedrick, J. L. Macromolecules
2010, 43, 1660−1664. (b) Kiesewetter, M. K.; Scholten, M. D.; Kirn,
N.; Weber, R. L.; Hedrick, J. L.; Waymouth, R. M. J. Org. Chem. 2009,
74, 9490−9496.
(13) Maliverney, C.; Saint-Jalmes, L. Patent, WO 2010/043353 A1.
(14) Raab, V.; Kipke, J.; Gschwind, R. M.; Sundermyer, J. Chem.
Eur. J. 2002, 8, 1682−1693.
(15) Ishikawa, T. Guanidines in Organic Synthesis in Guanidines,
amidines, phosphazenes and related organocatalysts. Superbases for organic
synthesis, Ed. J. Wiley & Sons: Chichester, U.K., 2009 p Chapter 4, pp
93−143.
(16) Baker, J.; Holsworth, J. B. J. Chem. Soc. 1947, 713−726.
(17) Schwetlick, K.; Noak, R.; Stebner, F. J. Chem. Soc., Perkin Trans.
2 1994, 599−608.
(18) Similar mechanisms have been proposed for base-catalyzed
alcoholysis of anhydrides, see: Chen, Y.; McDaid, P.; Deng, L. Chem.
Rev. 2003, 103, 2965−2983.
(19) Bonduelle, C.; MartinVaca, B.; Cossio, F. P.; Bourissou, D.
Chem.Eur. J. 2008, 14, 5304−5312.
(20) Rappo-Abiuso, M.; Llauro, M.-F.; Chevalier, Y.; Le Perchec, P.
Phys. Chem. Chem. Phys. 2001, 3, 99−106.
(21) Bacaloglu, R.; Cotarga, L.; Marcu, N.; Tolgyi, S. J. Prakt. Chem.
1988, 330, 530−540.
(22) For an exhaustive list of pKa of amines in CH3CN, see:
Kajjurand, I.; Kutt, A.; Soovali, L.; Rodima, T.; Maemets, V.; Leito, I.;
̈
̈
̈
Koppel, I. A. J. Org. Chem. 2005, 70, 1019−1028.
(23) Deutsch, J.; Eckelt, R.; Kockritz, A.; Martin, A. Tetrahedron
̈
2009, 65, 10365−10369.
(24) Pereira, F. S.; deAzevedo, E. R.; da Silva, E. F.; Bonagamba, T. J.;
́
da Silva Agostini, D. L.; Magalhaes, A.; Job, A. E.; Perez Gonzalez, E.
R. Tetrahedron 2008, 64, 10097−10106.
(25) Oediger, H., Moeller, F. Ger. Offen. DE 2640964, 1978.
(26) Richter, R. Tetrahedron Lett. 1968, 48, 5037−5039.
(27) Coles, M. P. Chem. Commun. 2009, 3659−3676.
(28) Similar activation of CO2 by MTBD and TBD has recently been
proposed, see: (a) Barbarini, A.; Maggi, R.; Mazzacani, A.; Mori, G.;
Sartori, G.; Sartorio, R. Tetrahedron Lett. 2003, 44, 2931−2934.
́
(b) Villers, C.; Dognon, J.-P.; Pollet, R.; Thuery, P.; Ephritikhine, M.
Angew. Chem., Int. Ed. 2010, 49, 3465−3468. (c) Das Neves Gomes,
C.; Jacquet, O.; Villers, C.; Thuery, P.; Ephritikhine, M.; Cantat, T.
́
Angew. Chem., Int. Ed. 2012, 51, 187−190.
(29) Nederberg, F.; Connor, E. F.; Moller, M.; Glauser, T.; Hedrick,
̈
J. L. Angew. Chem., Int. Ed. 2001, 40, 2712−2715.
2256
dx.doi.org/10.1021/ma2026258 | Macromolecules 2012, 45, 2249−2256