Heterobimetallic complexes composed of bismuth and lithium carboxylates as polyurethane catalysts - alternatives to organotin compounds

Article abstract of DOI:10.1039/d1gc00446h

Organotin compounds are important catalysts for the synthesis of polyurethanes consisting of aliphatic isocyanates and polyols. Due to their toxicity, however, it has been a long-standing goal to develop more environmentally benign catalysts. Bismuth octoates and neodecanoates are such well-known alternatives, but their catalytic activity is insufficient for many applications. Herein we show that the catalytic activity of bismuth carboxylates can be enhanced significantly by the addition of lithium carboxylates. Structural and spectroscopic results reveal the spontaneous formation of heterobimetallic complexes consisting of two bismuth and four lithium carboxylates that show dynamic behaviour in solution. The mechanism of the bismuth-catalyzed urethane reaction was elucidated in detail using quantum chemical calculations, paving the way for a rational catalyst design.

Full text of DOI:10.1039/d1gc00446h

Green Chemistry
View Article Online
View Journal | View Issue
PAPER
Heterobimetallic complexes composed of
bismuth and lithium carboxylates as polyurethane
catalysts – alternatives to organotin compounds†
Cite this: Green Chem., 2021, 23,
747
2
a
b
a
a
Emre Levent, Oliver Sala, Lukas F. B. Wilm, Pawel Löwe and
a,c
Fabian Dielmann
*
Organotin compounds are important catalysts for the synthesis of polyurethanes consisting of aliphatic
isocyanates and polyols. Due to their toxicity, however, it has been a long-standing goal to develop more
environmentally benign catalysts. Bismuth octoates and neodecanoates are such well-known alternatives,
but their catalytic activity is insufficient for many applications. Herein we show that the catalytic activity of
bismuth carboxylates can be enhanced significantly by the addition of lithium carboxylates. Structural and
spectroscopic results reveal the spontaneous formation of heterobimetallic complexes consisting of two
bismuth and four lithium carboxylates that show dynamic behaviour in solution. The mechanism of the
bismuth-catalyzed urethane reaction was elucidated in detail using quantum chemical calculations,
paving the way for a rational catalyst design.
Received 4th February 2021,
Accepted 22nd March 2021
DOI: 10.1039/d1gc00446h
rsc.li/greenchem
nate and alcohol, but do not promote the isocyanate–water
reaction as this would lead to blistering. Further important
Introduction
Polyurethanes (PUs) represent one of the most important catalyst properties are sufficient stability against light and air
classes of polymer materials with numerous technical appli- as well as tolerance against influences from other ingredients
cations, including elastomers, rigid foams, soft foams, coat- of the coating formulation. Organotin compounds, in particu-
1
ings and adhesives. These materials account for 7.9 wt% of lar dibutyltin dilaurate (DBTL), remain the most important cat-
2
the world polymer production due to their versatile appli- alysts in aliphatic urethane synthesis due to their remarkable
4
,15
cations at low cost. A key advantage of PU technology is the catalytic activity and advantageous properties.
simple construction of material composites in a single proces- owing to the toxicity of organotin compounds and the difficul-
However,
1
6,17
sing step via a polyaddition reaction of diisocyanates and ties to remove them from the final products,
it has been a
1
diols. Depending on the desired end product, the polymeriz- long-standing goal to develop more environmentally benign
ation is controlled by task-specific catalysts. Tertiary amines alternative catalysts. In efforts to replace organotin catalysts,
3
,4
18
and Lewis-acidic metal complexes are commonly employed to numerous Lewis-acidic metal complexes were identified as
1
9–27
mediate this polyaddition reaction. Recent developments in effective catalysts for the isocyanate-hydroxyl reaction,
organocatalytic PU formation include the use of amidines/gua- including some inorganic tin complexes with amino/alkoxide
5
6–9
10–12
28–35
nidines, NHCs
and organic acids.
For coating appli- ligands that exhibit thermolatent behaviour.
However,
1
3,14
cations involving aliphatic polyisocyanate components,
apart from DBTL, only a few metal complexes based on
catalysts with high catalytic activity at very low concentrations bismuth, aluminium, and zirconium are among the promising
are required that accelerate the gelling reaction between isocya- candidates for clear coat applications, as discoloration of the
2
1,26,27
coating formulation is highly undesirable.
While the
practical use of Zr catalyst is hampered by their sensitivity
towards hydrolysis and low tolerance against functional
a
Institut für Anorganische und Analytische Chemie, Westfälische
Wilhelms-Universität Münster, Corrensstraße 30, 48149 Münster, Germany
2
7
b
groups, bismuth carboxylates are recognized as non-toxic
BASF SE, Digitalization of R&D, Carl-Bosch-Straße 38,
1
5,36–39
6
c
7056 Ludwigshafen am Rhein, Germany
alternatives to DBTL catalysts.
In addition, there is evi-
Institut für Allgemeine, Anorganische und Theoretische Chemie,
Leopold-Franzens-Universität Innsbruck, Innrain 80-82, 6020 Innsbruck, Austria.
E-mail: fabian.dielmann@uibk.ac.at
dence that combinations of bismuth and zinc carboxylates or
bismuth and lithium carboxylates can accelerate the drying
40
time of polyurethane coating formulations. Herein, we inves-
†
Electronic supplementary information (ESI) available. CCDC 2057615–2057620.
For ESI and crystallographic data in CIF or other electronic format see DOI:
0.1039/d1gc00446h
tigate the influence of lithium carboxylates in the bismuth-
catalyzed PU reaction.
1
This journal is © The Royal Society of Chemistry 2021
Green Chem., 2021, 23, 2747–2755 | 2747

View Article Online
Paper
Green Chemistry
mode forming a Bi
molecular Bi–O bonds of 2.87 Å.
Next, we explored the influence of lithium carboxylates on
the coordination behaviour of bismuth complexes 1–3.
2
O
2
four-membered ring with long inter-
Results and discussion
Synthesis and solid-state structure of heterobimetallic
carboxylate complexes
Based on the superior performance of bismuth neodecanoate Heating a suspension of complex 1–3 and two equivalents of
4
1
among different carboxylates in PU synthesis, three alpha- the corresponding lithium carboxylate in acetonitrile gave a
branched carboxylates were chosen as model systems for the clear solution, which indicates the formation of a heterobime-
present study (Scheme 1). Various methods exist for the syn- tallic compound as Li(piv) is poorly soluble in acetonitrile.
4
2–44
thesis of bismuth carboxylates Bi(O
2
CR)
3
.
The synthesis of The heterobimetallic complexes 7–9 were readily obtained as
bismuth carboxylates via salt metathesis reactions often white crystalline solid in excellent yield after allowing the reac-
results in salt contaminations due to the variable and high tion mixture to cool to room temperature (Scheme 1). Note
4
5
coordination number of bismuth(III) complexes. We therefore that complexes 7–9 also crystallize from an equimolar mixture
prepared bismuth pivalate Bi(piv) (1), bismuth 2,2-dimethyl- of bismuth and lithium carboxylates, which demonstrates the
butanoate Bi(dmb)3 (2) and bismuth 2-phenylisobutyrate Bi propensity for the formation of heterobimetallic complexes.
3
4
4
(
pib)
3
(3) according to the procedure reported by Andrews.
2
The asymmetric and symmetric CO stretching frequencies in
Heating a mixture of BiPh
3
and stochiometric amounts of car- the IR spectra of complexes 1–3 and 7–9 appear at very similar
boxylic acid at 110 °C for 16 hours afforded bismuth carboxy- frequencies (Table 1). The separation Δν of these bands can
lates 1–3 in quantitative yield (Scheme 1). The progress of the provide information about the coordination mode of the car-
1
53
reaction was monitored by H NMR spectroscopy based on the boxylate ligand. For complexes 1–3 and 7–9 Δν values
formation of benzene and the decrease in BiPh3 signals. between 160 and 186 cm− are obtained, which are in the
Complexes 1–3 were obtained as white solids that are soluble typical range of chelating and bridging carboxylates
1
in common polar and nonpolar solvents. In the solid-state (150–200 cm− ).
structures of bismuth carboxylates, the bismuth centres are
typically nine coordinate and comprise the carboxylates in che- mation of isostructural complexes 7, 7′, 8, 9 and 9′ with the
1 53
Single crystals X-ray diffraction studies revealed the for-
4
6,47
lating and bridging (µ
2
and µ
3
) binding modes.
Thus, poly- general
formula
[Bi
2
Li
4
(O
2
CR)
2
(μ-O
2
CR)
4
(μ
3
-O
2
CR)
2
4
(μ -
meric structures are generally obtained with small substituents O CR) (MeCN) ], consisting of two bismuth carboxylates, four
2
2
2
such as in Bi(O
(O CC H NH )
2 6 4 2 3
2
CH)
3
, Bi(O
while the bulky pivalate ligands lead to S62†). The solid-state structure of the pivalate complex is
the tetrameric complex [{Bi(O CtBu)(μ -O CtBu)(μ -O CtBu)} ] depicted in Fig. 1. Compound 7′ and 9′ are two polymorphs
2 3 3 2 6 5 3
CCH ) , Bi(O CC H ) , and Bi lithium carboxylates and two acetonitrile ligands (Fig. S58–
4
8–50
2
2
2
3
2
4
in which the Bi atoms are connected by multiple Bi–O bonds containing one acetonitrile solvate molecules in the asym-
5
1,52
formed by the bridging pivalate ligands.
The almost identi- metric unit (see the ESI† for details). Their geometrical para-
cal steric bulk of the 2,2-dimethylbutanoate ligands and the meters are very similar to those of 7 and 9, respectively, and
observation that complex 2 is soluble in nonpolar solvents, thus are not included in the discussion. Selected bond lengths
suggests that 2 is also a tetramer in the solid state. The intro- for the heterobimetallic complexes 7, 8 and 9 are summarized
duction of a phenyl group in 2-phenylisobutyrate further in Table 2. In the hexanuclear complexes 7–9, two Bi(O CR)
2
3
increases the steric bulk of the carboxylate ligand. An X-ray units are linked by a Li
diffraction study of a single crystal grown from a saturated with crystallographic inversion centre located in the
THF solution revealed that 3 is dimeric in the solid state Li (O CR) six-membered ring. Each Bi is eight coordinate in a
4 2 4 2
(O CR) (MeCN) ladder-type structure
a
2
2
2
(Fig. S57†). Each Bi atom is coordinated by three chelating pib disordered square antiprismatic environment (Fig. 2) while
ligands and one THF molecule. These monomers are con- the lithium adopts a tetrahedral geometry. For each Bi atom,
nected by two pib ligands in tridentate bridging coordination four different types of carboxylate ligands can be distin-
guished. The first carboxylate asymmetrically chelates the Bi
atom with rather short Bi–O distances of 2.20–2.63 Å. Two
further carboxylates are chelating the Bi atom with a short
(
2.21–2.38 Å) and a long (2.72–3.00 Å) Bi–O bond. The oxygen
Table 1 IR stretching frequencies [cm− ] of the carboxylate ligands in
1
complexes 1–3 and 7–9
Compound
Bi(piv) (1)
ν
AS(COO)
ν
s
(COO)
Δν
3
1538
1527
1521
1570, 1536
1572, 1531
1575, 1544
1357
1358
1360
1409, 1360
1399, 1357
1395, 1358
181
169
160
161, 176
173, 174
180, 186
Bi(dmb) (2)
Bi(pib) (3)
Bi Li (piv)10(MeCN)
3
3
2
4
2
(7)
2
Scheme 1 Synthesis of Bi carboxylates and heterobimetallic complexes Bi Li (dmb) (MeCN) (8)
2
4
10
containing bismuth and lithium carboxylates.
2
Bi Li
4 2
(pib)10(MeCN) (9)
2748 | Green Chem., 2021, 23, 2747–2755
This journal is © The Royal Society of Chemistry 2021

View Article Online
Green Chemistry
Paper
Fig. 2 Section of the molecular structure of 7 showing the distorted
square-planar coordination geometry of the Bi atom. Connection of O
atoms via the C atom are illustrated by dashed lines.
1
appears in the range of 190.0–187.6 ppm. By means of
H
DOSY NMR spectroscopy the hydrodynamic radius rH of 1 was
determined, which is in good agreement with the radius of the
tetramer in the solid state (Table 3). However, three sets of
signals would be expected for the carboxylate ligands of the
Fig. 1 Molecular structure of the heterobimetallic cluster 7 in the solid tetramer according to the solid-state structure of 1. Thus, the
state. Complexes 8 and 9 are isostructural and thus depicted in the ESI.† NMR data clearly indicates that in solution complexes 1–3
Hydrogen atoms are omitted for clarity. Thermal ellipsoids are set at
form dynamic aggregates.
50% probability.
Dynamic behavior was also observed for the heterobimetal-
lic complexes 7–9 in dichloromethane solutions, as indicated
by the appearance of only one set of signals for the carboxylate
1
13
13
ligands in the H or C NMR spectra. The carboxylate
C
atom of the latter is additionally coordinated by a Li atom.
Finally, two carboxylates are bridging in (Bi–O8:
.29–2.52 Å) and μ (Bi–O5: 2.69–2.90 Å) coordination mode.
NMR resonances of 7–9 (189.7–187.5 ppm) are significantly
broadened and appear in the same range of those of 1–3. In
the Li NMR spectra two signals were observed for complexes 8
μ
3
2
4
7
Note that well-defined heterometallic Bi–Li complexes have
5
4
55
been described with alkoxide, tert-butylamine and boraa-
5
6
midinate ligands. The latter Bi/Li complexes exhibit fluxional
behaviour in solution.
1
Table 3 Hydrodynamic radius rH from H-DOSY-NMR and averaged
3
radius from the molecular structure of Bi(piv) (1) and heterobimetallic
complex 7 in [Å]
NMR-study of heterobimetallic carboxylate complexes
To study the behavior of the Bi carboxylate complexes 1–3 and
Entry
Bi(piv)
3
(1)
2 4 2
Bi Li (piv)10(MeCN) (7)
1
13
7
1
7
–9 in solution we performed H, C, Li and H DOSY NMR
a
1
13
Averaged solid state radius
Hydrodynamic radius r
7.96
8.58
8.04
experiments. The H and C NMR spectra of Bi complexes 1–3
b
b
H
7.64
in CD Cl show only one set of signals for the carboxylate
2
2
1
3
a
51
b
Calculated from literature crystal structure.
.
Recorded in MeOD.
ligands. The characteristic C signal of the carboxylate groups
Table 2 Selected bond lengths [Å] of heterobimetallic complexes
Entry
2
Bi Li
4
(piv)10 (MeCN)
2
(7)
2
Bi Li
4
(dmb)10 (MeCN)
2
(8)
2 4 2
Bi Li (pib)10 (MeCN) (9)
C–O
Li–O
1.247(3)–1.287(3)
1.877(4)–2.000(9)
3.432(3)
2.236(1)
2.515(1)
2.355(1)
2.718(1)
2.687(1)
2.294(1)
1.233(1)–1.311(4)
1.849(3)–2.011(1)
3.522(3)
2.256(4)
2.420(1)
2.375(2)
2.772(1)
2.889(1)
2.213(1)
1.240(2)–1.287(2)
1.897(4)–2.040(4)
3.545(3)
2.204(1)
2.627(1)
2.318(1)
3.000(1)
2.802(1)
2.254(1)
Bi⋯Li
Bi1–O1
Bi1–O2
Bi1–O3
Bi1–O4
Bi1–O5
Bi1–O6
Bi1–O7
Bi1–O8
Bi1–O9
2.781(1)
2.294(1)
3.163(1)
2.930(1)
2.515(5)
3.355(9)
2.737(1)
2.341(1)
3.181(1)
This journal is © The Royal Society of Chemistry 2021
Green Chem., 2021, 23, 2747–2755 | 2749

View Article Online
Paper
Green Chemistry
(
0.01 ppm, −0.5 ppm) and 9 (−0.1 ppm, −1.5 ppm) while 7
shows only one signal at −0.1 ppm. The hydrodynamic radius
of 7 is significantly larger than that of 1 and agrees with the
radius for the presence of dynamic aggregates in solutions of
r
H
1
7
–9 was provided by low temperature H NMR experiments.
Upon cooling a solution of 7 in CD Cl to 193 K, the singlet for
2
2
the methyl groups split into more than eight resonances, indi-
cating the presence of several different aggregates rather than
the discrete complex 7 in solution (Fig. S30†). Several signals
for the methyl protons were also observed in the low tempera-
1
ture H NMR spectra of complexes 8 and 9 (Fig. S31 and S32†),
which, however, can also be attributed to the presence of
several rotamers.
It should be noted that Li(piv) is very poorly soluble in
3 3
CD CN, but dissolves quickly upon addition of Bi(piv) .
Mixtures of Bi(piv)
3
and Li(piv) with 1 : 1 and 1 : 2 stoichio- Fig. 3 Lewis acid catalysed urethan reaction of Desmodur LD and
1
metry in CD CN both show a singlet at 1.13 ppm in the H ethanol in the presence of Bi complexes 1–3, Bi/Li complexes 7–9 and
3
DBTL.
NMR spectra, which appears downfield compared to that of Bi
(
piv) (1.06) and downfield relative to that of Li(piv) (1.16 ppm).
Collectively, the NMR studies show that the addition of Li car-
boxylates to bismuth carboxylates leads to the spontaneous
formation of heterobimetallic complexes in solution. However,
these species are not well defined, but rather constitute a
highly dynamic mixture of aggregates with an average size
comparable to that of 7–9.
lyzed reaction. Among the bismuth compounds the catalytic
activity was found to increase with the steric bulk of the car-
boxylate ligands pib > dmb > piv. Strikingly, the heterobimetal-
lic complexes 7–9 clearly emerge as superior catalysts com-
pared to their lithium free counterparts and compound 8 was
found to be almost as active as DBTL, leading to 90% conver-
sion after only 15 minutes.
Catalysis of the isocyanate/alcohol reaction
We next wanted to explore the influence of lithium carboxy-
lates on the catalytic performance of the bismuth complexes
in more detail. Catalyst mixtures containing different amounts
of bismuth and lithium carboxylates were prepared by adding
the respective carboxylates to the xylene/ethanol mixture
immediately before catalysis, one after the other (Fig. 4).
Owing to the sufficient solubility of Li(pib) in xylene, the pib
based catalyst system was chosen for this study. As expected, Li
The catalytic performance of complexes 1–3 and 7–9 in the
urethane reaction was examined using ethanol and the ali-
phatic isocyanate 2-ethylhexyl-(6-isocyanatohexyl)-carbamates
(Desmodur LD). The latter was chosen as model for 1,6-hexa-
methylene diisocyanate (HDI) owing to its lower volatility and
toxicity. The urethane reactions were performed at ambient
temperature with 0.05 mol% catalyst loading using xylene as
solvent (Scheme 2). DBTL was included in our study as a refer-
ence. Using ATR-IR spectroscopy the reaction progress was
monitored by measuring the decreasing intensity of the isocya-
(
pib) does not accelerate the isocyanate conversion at room
−
1
nate band at 2260 cm . The selectivity of the urethane reac-
tion was verified by the simultaneous increase of the character-
−
1
−1
istic ν(CvO) (1692 cm ) and ν(CHN) (1519 cm ) frequencies
5
7
of the alkyl urethane group.
From these experiments, summarized in Fig. 3, the Bi car-
boxylates 1–3 show moderate catalytic activity and accelerate
the urethane reaction significantly compared to the non-cata-
Fig. 4 Lewis acid catalysed urethan reaction of Desmodur LD and
Scheme 2 Catalysis conditions for isocyanate alcohol reaction. Catalyst
concentration is based on the amount of bismuth.
ethanol in the presence of different amounts of Bi(pib)
with complex 9 for comparison.
3
and Li(pib), and
2750 | Green Chem., 2021, 23, 2747–2755
This journal is © The Royal Society of Chemistry 2021

View Article Online
Green Chemistry
Paper
temperature. However, the catalytic activity of Bi(pib)
3
is
slightly increased upon adding 0.5 equiv. of Li(pib), while the
addition of two or more equiv. of Li(pib) resulted in reaction
rates that are identical to that of complex 9, which suggests
that complex 9 is the major species in these reaction mixtures
and heterobimetallic compounds with higher Li content are
not formed. Surprisingly, the best catalytic performance was
observed using a 1 : 1 mixture of Bi(pib)3 and Li(pib).
Collectively, these results reveal the superior performance of
the 1 : 1 mixture while the stability of complex 9 might be
responsible for the lower reaction rate with higher lithium
content.
3
Fig. 5 Isocyanate conversion over time utilizing Bi(piv) . This curve
initially follows pseudo first-order kinetics, deviating though after a
while from the simulated first-order kinetics profile and appearing
th
Mechanistic investigation and density functional theory
calculations
similar to 0 -order kinetics. We assume complex kinetics because of
this change of kinetic profile.
In an attempt to evaluate the reactivity of the urethane cata-
lysts with alcohol and isocyanate functions, complexes 1 and 7
were mixed with stoichiometric amounts of ethanol or 1,6- briefly addressing the uncatalyzed isocyanate-hydroxyl reaction
1
hexamethylene diisocyanate (HDI) and the resulting H NMR we will take a closer look at its mechanistic spectrum in the
spectra were compared to those of the individual compounds context of bismuth catalysis.
1
(
Fig. S33–S36†). The H NMR spectrum of the mixtures of 1 or
The uncatalyzed reaction between aliphatic alcohol and ali-
7
with HDI show the resonances of the individual components phatic or aromatic isocyanate requires a proton shuttle in the
and thus do not indicate a reaction. The same is observed for 6-membered ring transition-state (TS) structure to give
the mixture of 1 and ethanol. However, the CH
resonance of urethane product in reasonable time (Scheme 3). In the
2
ethanol in the mixture with 7 is significantly broadened and is absence of a suitable proton shuttle in the reaction mixture,
not shifted to lower/higher frequency, suggesting that ethanol the alcohol itself acts as such. This stoichiometry – two alco-
is involved in the dynamic formation of coordination com- hols and one isocyanate – leads to pseudo second-order kine-
pounds in solution, although the formation of free pivalic acid tics with a strong dependence on the concentration of the
5
8,59
was not detected.
bismuth pivalate catalysed urethane reaction reveal a 0.4 order are 110 and 100 kJ mol− for aliphatic and aromatic isocya-
kinetics with respect to ethanol (see the ESI†).
nates, respectively, in the infinite dilution scenario. The reac-
Preliminary kinetic studies on the alcohol. The computed reaction barriers for our model system
1
So far, mechanistic studies on metal complex catalyzed PU tion rates are consistent with the experimental reaction times,
reaction have been performed mainly for organotin which are in the range of days and hours, respectively, at
5
8,60–65
catalysts,
leaving critical questions still unanswered. It ambient temperature. Note that the effective rate of the
remains for example unclear what factors determine the urethan bond forming reaction can be influenced by various
selectivity and speed of the isocyanate-hydroxyl reaction when phenomena such as catalysis, autocatalysis, solvent effects and
6
3,66–69
catalyzed by tin, zirconium, bismuth, or zinc catalysts. We molecular interactions.
investigated the Bi-catalyzed urethane formation by means of
Computations reveal that cyclic, tetrameric bismuth tricar-
quantum chemical computations, providing the basis for boxylate clusters form thermodynamically driven by a free
further mechanistic investigations. These computations energy difference of −40 and −18 kJ mol− at 25 °C and 60 °C,
confirm that bismuth carboxylates successfully catalyze respectively (Fig. 6). Contrast this with the endergonic for-
urethane formation for both, aliphatic and aromatic isocya- mation of di- and trimeric bismuth-carboxylate complexes,
1
4
5
nates, speeding up the reaction by a factor of 10 to 10 . their free energy differences amounting to roughly +35 kJ
−
1
Complex kinetics govern the bismuth pivalate catalysis, albeit mol , starting from the monomer at 60 °C. Because of the
th
partially following pseudo 0 -order kinetics (cf. Fig. 5). The huge computational effort of using the tetramer as catalyst, we
turnover-limiting step in the catalytic cycle is the migratory
insertion of the isocyanate into the Bi–O bond of the alkoxy
ligand, finally forming the carbamate that remains co-
ordinated to the metal center in a κ
1
2
- or κ -fashion. The
urethane product is released from the catalyst by protonation
at the carbamate nitrogen- or carbonyl oxygen atom. Bismuth
pivalate serves as model for gaining conceptional mechanistic
insight into the Bi-catalysed urethane formation. Propyl isocya-
nate and ethyl alcohol serve as model substrates. Further infor-
mation on the condensed phase representation and compu-
Scheme 3 Uncatalyzed model reaction between propyl isocyanate and
ethyl alcohol. A urethane species containing also an ester functionality
represents the reaction medium. On the right side the 6-membered ring
tational details in general can be found in the ESI.† After transition-state structure is shown.
This journal is © The Royal Society of Chemistry 2021
Green Chem., 2021, 23, 2747–2755 | 2751

View Article Online
Paper
Green Chemistry
The initial ligand exchange most likely follows an associat-
ive mechanism because of the sterically accessible bismuth
center in Bi(piv) . There is no significant electronic stabiliz-
3
ation of the pivalate’s negative charge formed upon initial
departure from the bismuth center, which additionally sup-
ports this mechanism. Conversely, conjugate base stabilization
hints at
a
dissociative ligand exchange mechanism.
Fig. 6 Computed Bi(piv)
3
-dimer, -trimer, and -tetramer structures. The
Computations additionally confirm that the associative mecha-
nism is dominant. Hence, the ligand exchange rate law is first
order in the entering alcohol ligand. The expected large nega-
tive entropy of activation is the reason for an additional contri-
bution to the solvent dependence of the hydroxyl-isocyanate
highly symmetric Bi(piv)
tetrahedral a geometry and agrees with the solid-state structure.
3
-tetramer structure consists of Bi atoms in
42
limit our discussion to the monomer. Consider that the
concentration of Bi(piv) in solution is low (0.05 mol%)
and that entropy (and diffusion) hamper the formation of
clusters.
3
reaction. It turns out that Bi(piv) prefers exchanging pivalate
3
for alcohol only once (see Fig. 8 and 9). The concerted 1,2-
migratory insertion of the alkoxide ligand is identified by
quantum chemical computations as predominant reaction
mechanism (compare the reaction barriers in Fig. 7).
The efficiency of a catalytic cycle depends on the barrier of
the turnover-limiting step but most importantly though on the
Considering the multiple coordination forms of Bi, we first
examined the alcohol–isocyanate coupling in more detail. In
general, the coupling reaction is accelerated by facilitating the
nucleophilic attack of the alcohol at the carbon atom of the
isocyanate thereby lowering the activation barriers.
Coordination of alcohol to bismuth lowers the alcohol’s pKa,
facilitating its deprotonation and enhancing its nucleophilic
character. Several pathways of urethane formation have been
considered with the insertion reaction either proceeding con-
certedly with protonated or unprotonated alcohol (coordinated
to the catalyst’s metal center) or following an outer-sphere type
of mechanism (cf. Fig. 7). Clearly, the higher nucleophilicity of
coordinated alkoxides in concert with the higher electrophili-
city of the isocyanate carbon atom by interaction of the isocya-
nate nitrogen atom with a Lewis acid is most effective in lower-
ing the reaction barrier. This reaction, by taking place on a
metal center, is reminiscent of a 1,2-migratory insertion.
Migratory insertion rates are strongly dependent on the solvent
Fig. 8 Energy diagram of the Bi(piv)
is turnover limiting. The resting state consists of Bi(piv)
(
3
catalytic cycle. Migratory insertion
(OEt)(PrNCO)
cat-II). The effective equilibria are shifted towards the resting state.
2
7
0
or reaction matrix. Consequently, the polarity of ligands
forming the local matrix around the Bi center) affects the rate
(
of 1,2-migratory insertion.
Fig. 7 Potential modes of reaction between isocyanate and alcohol on
the bismuth center. These modes are shown in increasing order of the
effective barrier height. Concerted protonation of the forming carba-
mate can be ruled out. Protonated alcohol also is unlikely to react with
isocyanate in a Bi complex.
3
Fig. 9 Catalytic cycle starting with Bi(piv) as catalyst precursor. EtO-Bi
2
(piv) is the active form of the catalyst, its being regenerated by the
alcohol component. Highest efficiency is achieved by following this
catalytic cycle (heavy arrows).
2752 | Green Chem., 2021, 23, 2747–2755
This journal is © The Royal Society of Chemistry 2021

View Article Online
Green Chemistry
Paper
absence of easily accessible off-cycle species with siphoning appears to be the non-active form of the catalyst. In this latter
character. Favorable equilibria towards off-cycle species retard scenario, ligand exchange with alcohol must take place, com-
the catalytic cycle and open the door to the possible formation peting with off-cycle equilibria already at the beginning of the
of side products such as allophanates (Fig. 9). Clearly, the Bi catalytic cycle or perhaps before forming the active catalyst
(
piv) catalytic cycle consists of easily accessible off-cycle equi- species cat-I. Pre-equilibria including alcohol corroborate the
3
libria. These efficiency-lowering equilibria have been computa- experimental finding of the non-integer partial reaction order
tionally identified (Fig. 8 and 9). Their presence additionally is of the alcohol reagent in the rate law.
supported by the overall kinetic profile, obtained by experi-
ments, hinting at complex kinetics (cf. Fig. 5).
The selectivity of this catalyzed isocyanate-hydroxyl reaction
depends on the height of the barrier of the turnover-limiting
3
The kinetics of Bi(piv) -catalysed urethane formation is step compared to the ones that lead to side products.
mainly determined by which catalyst intermediate is regener- Formation of allophanate requires conquering a barrier of as
ated after the last step of the catalytic cycle, that is the step much as 46 kJ mol− , compared to 41 kJ mol for the turn-
1
−1
th
releasing the product from the catalyst. The intermittent 0 - over-limiting migratory insertion. The extent of allophanate
order kinetic character hints at a resting state located at Bi formation mainly depends on the relative height of the reac-
(
piv)(OEt)(PrNCO) (cat-II). Up to this point we have not con- tion barrier forming allophanates compared to that of carba-
sidered the computed energy diagram in Fig. 8. The relative mate protonation. As outlined in Fig. 8, isocyanate may coordi-
energy levels of Bi(piv)3 (cat), Bi(piv) (OEt) (cat-I), and cat-II nate to cat-P, competing with the coordination of alcohol (or
2
seem to contradict this hypothesis. However, taking into pivalic acid). Latter coordination is clearly favored by 10–15 kJ
−
1
account the uncertainty of quantum chemical computations in mol over that of isocyanate. Moreover, the barrier to form
−
1
condensed phase, the energy level of cat-II might be lower allophanate amounts to as much as 46 kJ mol and, thus, lies
than the one of cat-I. Seeing this, we should not be surprised higher than the ones leading to product release. These results
to find cat-I being regenerated in the catalytic cycle. corroborate the absence of allophanates in experiments.
Computations show, however, that both the alcohol and free
pivalic acid ligand can release the urethane product from the resting state of the Bi(piv)
Because the 1,2-migratory insertion is turnover limiting the
catalyzed reaction is cat-II.
3
catalyst via proton transfer. Whether the alcohol or the car- Consequently, the barrier of the liberation of the product from
boxylic acid coordinates to the catalyst depends on their the catalyst is not turnover limiting. Generally spoken, if com-
instantaneous concentration. Off-cycle equilibria become putations result in this scenario, the overall kinetics most
th
important in case of the regeneration of cat, retarding the probably exhibits pseudo 0 -order. The case in which product
th
efficiency of the catalytic cycle (Fig. 9). The most efficient cata- release is turnover limiting also leads to pseudo 0 -order kine-
lytical path, having the least probability of off-cycle intermedi- tics. But now the resting state consists of the catalyst-product
ates, follows the regeneration of cat-I. As the reaction proceeds, (cat-P) intermediate, raising the probability of side-product for-
urethane increasingly forms and may immobilize the catalyst mation. Let us assume now that the barrier of ligand exchange
ligand pivalic acid, as outlined in Fig. 10. This immobilization with alcohol is higher compared to the one of the insertion
st
redirects the catalytic cycle towards from cat being regenerated step. In this scenario, the reaction will follow pseudo 1 -order
to the more efficient pathway regenerating only cat-I.
kinetics and the resting state would be the catalyst precursor
Computations reveal that protonation of carbamate is ther- species (cat).
moneutral and, in addition, does not discriminate between
Often humidity cannot be excluded during the isocyanate-
carboxylic acid and alcohol as protonating agents in terms of hydroxyl reaction, depending on the application. Therefore,
kinetics (cf. Fig. 8). The corresponding barriers amount to 30 water may be present during synthesis, raising the question of
and 33 kJ mol− , respectively. The alcohol and carboxylic acid the hydrolytic stability of Bi(piv)
1
. Water coordinates favorably
3
concentrations determine – next to steric demand – which to cat and cat-I (Fig. 8). However, the formation of bismuth
species favorably coordinates prior to protonating the carba- hydroxides is unlikely because the reaction pathway is consist-
mate to release the product. Let us have a closer look at the ently uphill in energy. The computations further show that the
two scenarios. Firstly, when the alcohol protonates the carba- barrier for the reaction of water with isocyanate is higher than
mate (via associative ligand exchange), cat-I is regenerated, that for the reaction with alcohol. This high barrier explains
3
leading to the catalytic cycle consisting of less off-cycle equili- why Bi(piv) does not promote the isocyanate–water reaction.
bria (cf. Fig. 9). Secondly, when free carboxylic acid takes over Apparently, water blocks the catalyst and removes it from the
protonation, catalyst precursor cat is regenerated, which catalytic cycle by siphoning. Unfortunately, this bismuth-aqua
complex is no longer available for catalyzing the hydroxyl-iso-
cyanate reaction.
The mechanistic concepts can be applied (in a first approxi-
mation) to multi-nuclear bismuth complexes (e.g. to Bi–Li clus-
ters). Preliminary quantum chemical computations reveal that
in contrast to mononuclear bismuth catalysis, isocyanate
ligand association is thermoneutral or even exergonic when
lithium is involved in the catalyst complex. The substrate iso-
Fig. 10 Possible immobilization of pivalic acid as protonating agent for
product release in the catalytic cycle.
This journal is © The Royal Society of Chemistry 2021
Green Chem., 2021, 23, 2747–2755 | 2753

View Article Online
Paper
Green Chemistry
cyanate then preferably coordinates to lithium, whereas the by the Emmy Noether Program of the DFG (DI 2054/1-1). We
deprotonated alcohol substrate is bridging between lithium thank Prof. F. E. Hahn for his generous support. We thank
and bismuth. Presumably, the higher Lewis acidity of lithium Klaus Bergander for performing and evaluating the DOSY
compared to that of the bismuth species leads to a higher pro- NMR experiments.
tonating potential of alcohols and free carboxylic acids, result-
ing in a more efficient release of product from the dinuclear
complex.
References
1
2
3
H.-W. Engels, H.-G. Pirkl, R. Albers, R. W. Albach,
J. Krause, A. Hoffmann, H. Casselmann and J. Dormish,
Angew. Chem., 2013, 125, 9596–9616.
PlasticsEurope – Association of Plastic Manufacturers,
available at: https://www.plasticseurope.org/de/resources/
market-data.
Conclusions
In conclusion, a series of Bi carboxylates 1–3 and heterobime-
tallic complexes 7–9 have been synthesized and characterized.
The isostructural clusters 7–9 consist of two bismuth carboxy-
lates and four lithium carboxylates and are spontaneously
formed upon mixing bismuth and lithium carboxylates in
E. Delebecq, J.-P. Pascault, B. Boutevin and F. Ganachaud,
Chem. Rev., 2013, 113, 80–118.
1
solution. H DOSY NMR experiments and variable temperature
4
5
A. L. Silva and J. C. Bordado, Catal. Rev., 2004, 46, 31–51.
J. Alsarraf, Y. A. Ammar, F. Robert, E. Cloutet, H. Cramail
and Y. Landais, Macromolecules, 2012, 45, 2249–2256.
B. Bantu, G. M. Pawar, U. Decker, K. Wurst, A. M. Schmidt
and M. R. Buchmeiser, Chem. Eur. J., 2009, 15, 3103–
NMR studies reveal that solutions of 1–3 and 7–9 contain a
dynamic mixture of aggregates with an average size compar-
able to that of their solid-state structures.
Studies on the catalytic performance of the Bi compounds in
the urethane forming reaction show that the Bi-catalyzed reac-
tion can be significantly accelerated by using heterobimetallic
mixtures of Bi and Li carboxylates, with 8 almost reaching the
activity of DBTL without promoting the isocyanate–water reac-
tion. Theoretical studies show that urethane formation utilizing
6
3109.
7
8
H.-J. Noh, T. Sadhasivam, M. Han, K. Lee, J.-Y. Kim and
H.-Y. Jung, J. Catal., 2020, 382, 77–85.
O. Coutelier, M. El Ezzi, M. Destarac, F. Bonnette, T. Kato,
A. Baceiredo, G. Sivasankarapillai, Y. Gnanou and D. Taton,
Polym. Chem., 2012, 3, 605.
Bi(piv)
3
follows complex kinetics that exhibit intermittent
th
pseudo 0 -order kinetics. The turnover-limiting step in the cata-
lytic cycle is the migratory insertion of the isocyanate into the
Bi–O bond of the alkoxy ligand, eventually forming the carba-
mate. The urethane product is released from the catalyst by
proton transfer from either the alcohol substrate or the free
pivalic acid ligand. Collectively, heterobimetallic complexes
composed of bismuth and lithium carboxylates are a promising
alternative to the toxic organotin-containing catalysts in PU syn-
thesis. Successive studies on the synthesis of even more active
bismuth catalysts are under investigation in our laboratories.
9
E. Peris, Chem. Rev., 2018, 118, 9988–10031.
1
0 H. Sardon, A. C. Engler, J. M. W. Chan, J. M. García,
D. J. Coady, A. Pascual, D. Mecerreyes, G. O. Jones,
J. E. Rice, H. W. Horn and J. L. Hedrick, J. Am. Chem. Soc.,
2013, 135, 16235–16241.
1
1
1 H. Sardon, A. Pascual, D. Mecerreyes, D. Taton, H. Cramail
and J. L. Hedrick, Macromolecules, 2015, 48, 3153–
3165.
2 F.-Z. Belmokaddem, J. Dagonneau, J. Lhomme, R. Blanc,
A. Garduno-Alva, C. Maliverney, A. Baceiredo, E. Maerten,
E. Fleury and F. Méchin, Des. Monomers Polym., 2016, 19,
3
47–360.
Author contributions
1
3 F. E. Golling, R. Pires, A. Hecking, J. Weikard, F. Richter,
K. Danielmeier and D. Dijkstra, Polym. Int., 2019, 68, 848–
E.L. carried out the synthetic experiments and analysed the
experimental data. O.S. performed the computational investi-
gations. L.W. and P.L. performed the X-ray single-crystal struc-
ture analyses. F.D., E.L. and O.S. wrote the manuscript. F.D.
directed the investigations.
8
55.
4 D. K. Chattopadhyay and K. V. S. N. Raju, Prog. Polym. Sci.,
007, 32, 352–418.
1
1
2
5 D. Guhl, in Polyurethanes for High Performance Coatings,
European Coatings Conference, vol. 5, pp. 119–137.
6 M. Hoch, J. Appl. Geochem., 2001, 16, 719–743.
1
1
7 Y. Yang and M. W. Urban, Angew. Chem., Int. Ed., 2014, 53,
Conflicts of interest
12142–12147.
1
8 Y. Schellekens, B. van Trimpont, P.-J. Goelen,
K. Binnemans, M. Smet, M.-A. Persoons and D. de Vos,
Green Chem., 2014, 16, 4401–4407.
There are no conflicts to declare.
1
9 Q. Han and M. W. Urban, J. Appl. Polym. Sci., 2002, 86,
Acknowledgements
2322–2329.
Financial support was generously provided by the BASF 20 L. Rand, B. Thir, S. L. Reegen and K. C. Frisch, J. Appl.
Coatings GmbH and BASF SE. Funding for F. D. was provided Polym. Sci., 1965, 9, 1787–1795.
2754 | Green Chem., 2021, 23, 2747–2755
This journal is © The Royal Society of Chemistry 2021

View Article Online
Green Chemistry
Paper
2
2
2
2
2
1 W. J. Blank, Z. A. He and E. T. Hessell, Prog. Org. Coat., 45 G. G. Briand and N. Burford, in Main Chemistry Group, ed.
1
999, 35, 19–29.
2 S. Petrak, V. Shadurka and W. H. Binder, Prog. Org. Coat.,
009, 66, 296–305.
A. H. Cowley and A. G. Sykes, Elsevier textbooks, s.l., 1st
edn, 2000, pp. 285–357.
46 H. Suzuki, Elsevier Science, Amsterdam, New York, 2001.
2
3 R. Gertzmann and C. Gürtler, Tetrahedron Lett., 2005, 46, 47 G. G. Briand and N. Burford, in Advances in Inorganic
659–6662. Chemistry, Academic Press, 2000, vol. 50, pp. 285–357.
4 I. Ahmad, J. H. Zaidi, R. Hussain and A. Munir, Polym. Int., 48 C.-I. Stålhandske, A. Kjekshus, S. Svensson, T. Holme,
6
2
007, 56, 1521–1529.
A. A. Lindberg, G. Jansen, B. Lamm and B. Samuelsson,
Acta Chem. Scand., 1969, 23, 1525–1533.
49 S. I. Troyanov and A. P. Pisarevskii, Koord. Khim., 1991, 909.
50 L. Wrobel, T. Rüffer, M. Korb, H. Lang and M. Mehring,
Eur. J. Inorg. Chem., 2017, 2017, 1032–1040.
5 C. Korah Bina, K. G. Kannan and K. N. Ninan, J. Therm.
Anal., 2004, 78, 753–760.
2
2
6 W. J. Blank, Macromol. Symp., 2002, 187, 261–270.
7 H. Sardon, L. Irusta and M. J. Fernández-Berridi, Prog. Org.
Coat., 2009, 66, 291–295.
51 T. Hatanpää, M. Vehkamäki, M. Ritala and M. Leskelä,
Dalton Trans., 2010, 39, 3219–3226.
2
8 J. Krause, S. Reiter, S. Lindner, A. Schmidt, K. Jurkschat,
M. Schürmann and G. Bradtmöller, WO Pat., 2009132784(A1), 52 S. I. Troyanov and A. P. Pisarevsky, J. Chem. Soc., Chem.
009. Commun., 1993, 335.
9 K. Jurkschat, L. Lovkova-Berends, T. Zöller and M. Schürmann, 53 G. Deacon, Coord. Chem. Rev., 1980, 33, 227–250.
2
2
3
3
3
WO Pat., 2011113926(A1), 2011.
54 M. Mehring, D. Mansfeld, S. Paalasmaa and
M. Schürmann, Chem. Eur. J., 2006, 12, 1767–1781.
55 A. J. Edwards, M. A. Beswick, J. R. Galsworthy, M. A. Paver,
P. R. Raithby, M.-A. Rennie, C. A. Russell,
K. L. Verhorevoort and D. S. Wright, Inorg. Chim. Acta,
1996, 248, 9–14.
0 F. Richter, K. Jurkschat, L. Lovkova-Berends and T. Zöller,
EP Pat., 2772496(A1), 2013.
1 F. Richter, K. Jurkschat, M. Wagner, V. Abeyawarathan and
M. Gock, EP Pat., 2990381(A1), 2014.
2 L. Iovkova-Berends, T. Berends, C. Dietz, G. Bradtmöller,
D. Schollmeyer and K. Jurkschat, Eur. J. Inorg. Chem., 2011, 56 J. Konu, M. S. Balakrishna, T. Chivers and T. W. Swaddle,
011, 3632–3643. Inorg. Chem., 2007, 46, 2627–2636.
3 L. Iovkova-Berends, T. Berends, T. Zöller, G. Bradtmöller, 57 G. G. Suchkova and L. I. Maklakov, Vib. Spectrosc., 2009, 51,
2
3
3
3
3
S. Herres-Pawlis and K. Jurkschat, Eur. J. Inorg. Chem.,
012, 2012, 3191–3199.
4 L. Iovkova-Berends, T. Berends, T. Zöller, D. Schollmeyer,
333–339.
2
58 A. J. Bloodworth and A. G. Davies, Chem. Commun., 1965,
24.
G. Bradtmöller and K. Jurkschat, Eur. J. Inorg. Chem., 2012, 59 R. P. Houghton and A. W. Mulvaney, J. Organomet. Chem.,
012, 3463–3473. 1996, 518, 21–27.
5 B. Glowacki, R. Pallach, M. Lutter, F. Roesler, H. Alnasr, 60 J. Otera, T. Yano and R. Okawara, Chem. Lett., 1985, 14,
2
C. Thomas, D. Schollmeyer and K. Jurkschat, Chem. – Eur.
J., 2018, 24, 19266–19279.
6 T. Vanbergen, I. Verlent, J. de Geeter, B. Haelterman,
901–904.
61 M. V. Zabalov, R. P. Tiger and A. A. Berlin, Russ. Chem.
Bull., 2012, 61, 518–527.
L. Claes and D. de Vos, ChemSusChem, 2020, 13, 3835– 62 R. Devendra, N. R. Edmonds and T. Söhnel, J. Mol. Catal. A:
843. Chem., 2013, 366, 126–139.
7 G. Yin, D. Zhao, X. Wang, Y. Ren, L. Zhang, X. Wu, S. Nie 63 W. Cheikh, Z. B. Rózsa, C. O. Camacho López, P. Mizsey,
and Q. Li, RSC Adv., 2015, 5, 79070–79080. B. Viskolcz, M. Szőri and Z. Fejes, Polymers, 2019, 11, 1543.
8 B. Xu, L. Li, K. Zhang, P. M. Macdonald, M. A. Winnik, 64 A.-C. Draye and J.-J. Tondeur, J. Mol. Catal. A: Chem., 1999,
3
3
3
R. Jenkins, D. Bassett, D. Wolf and O. Nuyken, Langmuir,
997, 13, 6896–6902.
138, 135–144.
65 A. Draye, J. Mol. Catal. A: Chem., 1999, 140, 31–40.
1
3
9 B. Jousseaume, C. Laporte, T. Toupance and J.-M. Bernard, 66 G. Raspoet, M. T. Nguyen, M. McGarraghy and
Tetrahedron Lett., 2002, 43, 6305–6307.
0 P. Hoffmann and B. Schnier, EP Pat., 3247757(B1), 2016.
1 D. Guhl, FAPU, 2008, 49, 30–33.
A. F. Hegarty, J. Org. Chem., 1998, 63, 6878–6885.
67 W. Sultan and J. P. Busnel, J. Therm. Anal., 2006, 83, 355–
359.
4
4
4
2 J. D. Donaldson, J. F. Knifton and S. D. Ross, Spectrochim. 68 M. Muuronen, P. Deglmann and Ž. Tomović, J. Org. Chem.,
Acta, Part A, 1965, 21, 275–277. 2019, 84, 8202–8209.
3 S. I. Troyanov and A. P. Pisarevsky, J. Chem. Soc., Chem. 69 C. O. C. López, Z. Fejes and B. Viskolcz, J. Flow Chem, 2019,
Commun., 1993, 335. 9, 199–204.
4 P. C. Andrews, G. B. Deacon, P. C. Junk, I. Kumar and 70 J. F. Hartwig and L. M. Stanley, Acc. Chem. Res., 2010, 43,
M. Silberstein, Dalton Trans., 2006, 4852–4858. 1461–1475.
4
4
This journal is © The Royal Society of Chemistry 2021
Green Chem., 2021, 23, 2747–2755 | 2755