Aluminum-Mediated Formation of Cyclic Carbonates: Benchmarking Catalytic Performance Metrics

Article abstract of DOI:10.1002/cssc.201601712

We report a comparative study on the activity of a series of fifteen binary catalysts derived from various reported aluminum-based complexes. A benchmarking of their initial rates in the coupling of various terminal and internal epoxides in the presence of three different nucleophilic additives was carried out, providing for the first time a useful comparison of activity metrics in the area of cyclic organic carbonate formation. These investigations provide a useful framework for how to realistically valorize relative reactivities and which features are important when considering the ideal operational window of each binary catalyst system.

Full text of DOI:10.1002/cssc.201601712

Accepted Article
Title: Aluminum-Mediated Formation of Cyclic Carbonates:
Benchmarking Catalytic Performance Metrics
Authors: Arjan Willem Kleij and Jeroen Rintjema
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemSusChem 10.1002/cssc.201601712
Link to VoR: http://dx.doi.org/10.1002/cssc.201601712
A Journal of
www.chemsuschem.org

10.1002/cssc.201601712
ChemSusChem
FULL PAPER
Aluminum-Mediated Formation of Cyclic Carbonates:
Benchmarking Catalytic Performance Metrics
Jeroen Rintjema,[a] and Arjan W. Kleij*[a][b]
Abstract: We report a comparative study on the activity of a series
of five binary catalysts derived from various reported aluminum-
based complexes. A benchmarking of their initial rates in the
coupling of various terminal and internal epoxides in the presence of
three different nucleophilic additives has been carried out providing
for the first time a useful comparison of activity metrics in the area of
cyclic organic carbonate formation. These investigations provide a
useful framework how to realistically valorize relative reactivities and
which features are important to consider the ideal operational
window of each binary catalyst system.
Particularly relevant for potential commercial applications is the
design of catalysts that can combine low toxicity and cost, easy
handling, abundance, modular features and high activity. In this
context, various groups have focused on the use of Fe-,[13] Zn-
[14]
,
and Al-based catalysts[15] that generally display catalytic
performances that are among the best reported to date. For
example, Ema et al. reported on highly active bifunctional
Zn(porphyrin) catalysts with high turnover numbers at elevated
temperatures (120160ºC).[16] More recently the same authors
described
a synergistic (theoretical) model explaining the
cooperation between the Lewis acid site and peripheral
ammonium halide groups that serve as nucleophiles.[17]
Introduction
The conversion and activation of small molecules such as
carbon dioxide (CO2) is among the most active fields in modern
research.[1] Typically, catalysis has proven to be a key enabling
technology for the conversion of CO2 into organic molecules with
a higher degree of complexity as it allows to reduce the high
kinetic barrier of breaking the rather inert C=O bonds.[2]
Additionally, an energy input is required to thermodynamically
facilitate CO2 conversion, and co-reactants with a relative high
free energy such as strained heterocycles,[3] and reducing
agents including hydrogen,[4] silanes and boron compounds[5]
have been classically used. In all cases, catalyst selection,
development and optimization is paramount to achieve the best
performance in terms of selectivity and/or activity.
One of the most widely studied reactions in CO2 catalysis is
undoubtedly the formation of cyclic organic carbonates
(COCs),[6] and some of these have been commercialized.[7] Both
metal-based[8] as well as organocatalytic approaches[9] have
become popular to mediate the coupling between epoxides and
CO2 and discovery of new catalysts continues to draw broad
attention from the scientific communities. Much less explored is
the development of new reactivity that aims at the discovery of
new/improved selectivity,[10] increasing the overall reactivity
(including efforts towards more sustainable reaction
conditions)[11] and amplification of the substrate portfolio to
provide new opportunities in synthetic and polymer chemistry.[12]
Figure 1. Three different types of Al(III) complexes used as components of
highly active binary catalysts: Kleij´s aminotriphenolate 1, North´s bimetallic
salen system 2 and Wang/Qin´s porphyrin-based complex 3.
[a]
[b]
J. Rintjema, Prof. Dr. A. W. Kleij
However, a direct comparison between (binary/bifunctional)
catalysts active towards the formation of COCs remains difficult
as many systems incorporate different metal centers, are based
on different ratios between metallic and nucleophilic sites, are
operated under different reaction conditions (temperature,
pressure) and catalytic performances are often determined in
different reactor systems. These differences often render claims
of high(est) reactivity unsuitable to serve as absolute reference
points. Therefore, benchmarking relative performances under
Institute of Chemical Research of Catalonia (ICIQ)
The Barcelona Institute of Science and Technology
Av. Països Catalans 16, 43007 Tarragona (Spain)
E-mail: akleij@iciq.es
Prof. Dr. A. W. Kleij
Catalan Institute of Research and Advanced Studies (ICREA)
Pg. Lluís Companys 23, 08010 Barcelona (Spain)
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/cssc.201601712
ChemSusChem
FULL PAPER
identical conditions can help to further determine the impact of a
given catalyst in the context of COC synthesis by carefully
examining the influence of substrate, reaction conditions and
catalyst composition. In this respect, a recent comparative study
on the influence of different hydrogen-bond donor systems on
the overall activity of binary catalysts in the conversion of
propylene oxide and CO2 into its cyclic carbonate derivative has
shown the importance of benchmarking, and revealed that
fluorinated alcohols explicitly are highly potent catalyst
components.[18]
on 1a shows significantly better activity compared to the other
screened binary catalysts that are derived from 2 and 4.
Table 1. Comparison of the reactivities of binary Al-catalysts 14/X (X = Cl,
Br, I) in the conversion of terminal epoxide A into the cyclic carbonate 5.[a]
Here we report on a detailed comparison between a pre-
selected series of Al-based binary catalysts (Figure 1) and have
assessed their initial reactivity towards six different epoxide
partners. Our results discussed herein clearly demonstrate that
the different operative mechanistic manifolds are crucial for the
catalyst performance. Specifically, the nature and amount of the
nucleophile, the operating reaction temperatures and
concentration of the binary couples are important parameters
that define the overall reactivity and the most suitable operation
window for each binary catalyst system. This information should
be carefully considered depending on the type of epoxide/CO2
coupling reaction that is of interest.
Entry
1
Cat

Nu
Substrate
Yield [%][b]
TON[c]

PPNCl
PPNCl
PPNCl
PPNCl
PPNCl
PPNCl
TBAB
TBAB
TBAB
TBAB
TBAB
TBAB
TBAI
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
3
14
5
2
1a
2
1400
250
3500
3300
400

3
4
3a
3b
4
35
33
4
Results and Discussion
5
Reactivity towards Terminal Epoxide Conversion
6
The complexes illustrated in Figure 1 (13) were tested under
identical reaction conditions in a parallel reactor system (see
Supporting Information for details). For comparison also the
mono-metallic Al(salen)Cl complex 4 was included in these
studies. In order to examine the reactivities of the complexes
14 upon combination with various nucleophiles, initial kinetics
were determined in most of the following cases reported here.
First, the conversion of three different terminal epoxides (AC)
was investigated at 90ºC, 10 bar and at low complex loading
([Al] = 0.01 mol%), see Tables 1 and 2.[19] As co-catalytic
nucleophiles tetrabutylammonium halides (TBA-X; X = Br, I) and
bis(triphenylphosphine)iminium chloride (PPNCl)[20] were used in
a five-fold excess with respect to the Al complex.
The conversion of 1,2-epoxyhexane A (Table 1) and total
turnover numbers (TONs) were determined for all binary
combinations after 2 h on a per reactive center basis, i.e. per Al
site. The data were compared within each series with the
background conversion mediated by the nucleophilic additive
alone as a reference (entries 1, 7 and 13). Striking differences
were noted for the binary systems based on 14, as well as
significant changes upon variation of the type of nucleophile
(halide). In the presence of PPNCl as nucleophile (entries 16),
the data clearly demonstrate that the Wang and Qin complex 3b
has excellent activity under these conditions (entry 5), in line
with their previous findings.[11c] Interestingly, the non-chlorinated
version of this Al(porphyrin) based binary catalyst (Figure 1, 3a;
entry 4) displays similar activity features suggesting that Lewis
acidity for this system under dilute catalyst conditions does not
play an imperative role. Apart from 3b, the Kleij system based
7

1
8
1a
2
19
3
1900
150
1100
1200
300

9
10
11
12
13
14
15
16
17
18
3a
3b
4
11
12
3

2
1a
2
TBAI
13
2
1300
100
700
700
300
TBAI
3a
3b
4
TBAI
7
TBAI
7
TBAI
3
[a] Reaction conditions: epoxide A (10.0 mmol), Al-complex (0.001 mmol;
0.01 mol%), nucleophile (Nu: 0.005 mmol; 0.05 mol%), mesitylene (1.0
mmol), 90ºC, 10 bar, 2 h; [b] Determined by 1H NMR (CDCl3) using
mesitylene as internal standard; the average of two runs is reported.
Selectivity towards the cyclic carbonate 5 was >99%. [c] TON = total
turnover number observed per Al center; note that complex 2 contains two
Al centers.
A dramatic change in the activity order was observed when
changing the nucleophile from PPNCl to TBAB and TBAI
(entries 718). In the presence of both TBAB and TBAI the most
This article is protected by copyright. All rights reserved.

10.1002/cssc.201601712
ChemSusChem
FULL PAPER
active binary catalyst is the one based on complex 1a, followed
by the porphyrin systems 3a/TBAX and 3b/TBAX (X = Br, I).
Binary couple 1a/TBAB shows the highest TON (entry 8; 1900)
under these conditions and these observations align well with
previous data in the literature that showed that bromide based
nucleophiles give the highest activities when combined with
Al(aminotriphenolate) complexes.[11a]
Additionally, a similar reactivity difference between catalysts
based on 1a and the North system 2 has been observed at a
ten-fold higher loading of both catalyst components (0.01 mol%
[Al], 0.05 mol% TBAI).[15c] The combined results for the
conversion of terminal epoxide A into mono-substituted cyclic
carbonate 5 (Table 1) suggest that matching of the Lewis acid
complex and nucleophilic additive has a pronounced influence
on the overall catalyst performance.
Intrigued by the results presented in Table 1, we then
decided to screen two other terminal epoxides to see whether
the trends observed in the catalytic formation of cyclic carbonate
5 from terminal epoxide A were of a more general nature (see
Table 2). Indeed, when all results are considered it can be
observed that also in the conversion of styrene oxide (B) and
propylene oxide (C) in the presence of PPNCl as nucleophilic
additive, both porphyrin based binary systems 3a/PPNCl and
3b/PPNCl have significantly higher activities compared to the
other binary catalysts derived from 1, 2 and 4 (Table 2, entries
1-6). After combining complexes 14 with either TBAB or TBAI,
the activities noted in the presence of 3a and 3b drop whereas
those for the systems based on 1a specifically become more
competitive and reveal similar to slightly improved activities
(entries 8 and 14) as noted for the binary combinations
comprising of metalloporphyrins 3a or 3b.
The somewhat lower conversion kinetics for the styrene
oxide substrate compared to 1,2-epoxyhexane and propylene
oxide is likely a result of electronic effects as reported
previously.[21] This affects the nucleophilic character of the
alkoxide intermediate after initial ring-opening of the epoxide via
a more pronounced charge delocalization. Overall, the results of
Tables 1 and 2 align well and show clear reproducible trends in
the initial conversion kinetics of terminal epoxides and CO2 at
90ºC catalyzed by binary catalysts derived from Al-complexes
14.
Table 2. Comparison of the reactivities of binary Al-catalysts 14/Nu in the
conversion of terminal epoxides
B and C into their respective cyclic
carbonates 6 and 7.[a]
Entry
Cat
Nu
Substrate
Yield 6, 7
TON 6, 7
[c]
[%][b]
1
2

1a
2
PPNCl
PPNCl
PPNCl
PPNCl
PPNCl
PPNCl
TBAB
TBAB
TBAB
TBAB
TBAB
TBAB
TBAI
B, C
B, C
B, C
B, C
B, C
B, C
B, C
B, C
B, C
B, C
B, C
B, C
B, C
B, C
B, C
B, C
B, C
B, C
5, 3
8, 29
5, 7

800, 2900
250, 350
2200, 4900
2400, 5300
600, 500

3
4
3a
3b
4
22, 49
24, 53
6, 5
5
6
7

2, 3
8
1a
2
12, 43
3, 9
1200, 4300
150, 450
1500, 4000
1300, 4100
300, 600

9
10
11
12
13
14
15
16
17
18
3a
3b
4
15, 40
13, 41
3, 6
Influence of Reaction Temperature
Our next focus was on the influence of the reaction
temperature on the catalyst performance and therefore parallel
catalysis experiments were performed at 25, 50 and 105ºC in a

3, 1
suitable
high-throughput
reactor
platform
(Supporting
1a
2
TBAI
11, 26
3, 6
1100, 2600
150, 300
700, 2000
800, 1800
300, 500
Information for details; see Table 3). TBAB was used as
nucleophile as most of the catalysts studied here showed
appreciable turnover in the presence of this additive. At 25ºC
(entries 18), the two porphyrin based catalysts show the best
performance with in both cases (entries 6 and 7) nearly
quantitative or quantitative conversion, and high turnover
numbers. Under these conditions, the North system 2/TBAB also
performs comparatively well (entry 5) whereas the
aminotriphenolate complexes 1a1c show lower activity (entries
24). Upon raising the reaction temperature to 50ºC (entries
916), the most notable changes are observed for the binary
systems based on 1a1c, with the combination of Al-complex 1a
and TBAB (entry 10) now being the third most efficient catalyst.
The results obtained at 105ºC ([Al] = 0.001 mol%, TBAB = 0.005
mol%) reveal that at elevated temperatures the binary
combination 1a/TBAB shows the best catalytic turnover (entry
TBAI
3a
3b
4
TBAI
7, 20
8, 18
3, 5
TBAI
TBAI
[a] Reaction conditions: epoxide B or C (10.0 mmol), Al-complex (0.001
mmol; 0.01 mol%), nucleophile (Nu: 0.005 mmol; 0.05 mol%), mesitylene
(1.0 mmol), 90ºC, 10 bar, 2 h, Nu = nucleophile; [b] Determined by 1H NMR
(CDCl3) using mesitylene as internal standard, first number refers to yield
when using B as substrate; the average of two runs is reported. Selectivity
towards the cyclic carbonates 6 and 7 was >99%. [c] TON = total turnover
number observed per Al; note that complex 2 contains two Al centers.
This article is protected by copyright. All rights reserved.

10.1002/cssc.201601712
ChemSusChem
FULL PAPER
Table 3. Comparison of the reactivities of binary Al-catalysts 14/TBAB in
the conversion of terminal epoxide A into cyclic carbonate 5 at different
reaction temperatures.[a]
Table 4. Comparison of the reactivities of binary Al-catalysts 14/Nu in the
conversion of internal epoxide DF into cyclic carbonates 810 at 90º.[a]
Entry
1
Cat

Nu
T [ºC]
25
Yield [%][b]
<1
19
13
19
30
97
>99
18
2
TON[c]

TBAB
TBAB
TBAB
TBAB
TBAB
TBAB
TBAB
TBAB
TBAB
TBAB
TBAB
TBAB
TBAB
TBAB
TBAB
TBAB
TBAB
TBAB
TBAB
TBAB
TBAB
TBAB
TBAB
TBAB
Entry
Cat
Nu
Sub
Yield
[%][b]
TON
Sel.
2
1a
1b
1c
2
25
190
130
190
150
970
1000
180

[c]
[%][d]
3
25
1[e]
2[e]
3[e]
4[e]
5[e]

[Cl]/[Br]
[Cl]/[Br]
[Cl]/[Br]
[Cl]/[Br]
[Cl]/[Br]
D
D
D
D
D
<1, <1
8, 12
1, 1

ND
>99
>99
>99
4
25
1a
2
200, 300
13, 13
5
25
6
3a
3b
4
25
3a
3b
5, 3
125, 75
325, 250
7
25
13, 10
24:76[f]
7:93[g]
8
25
6[e]
7
4

[Cl]/[Br]
[Cl]/[Br]
[Cl]/[Br]
[Cl]/[Br]
[Cl]/[Br]
[Cl]/[Br]
[Cl]/[Br]
[Cl]/[Br]
[Cl]/[Br]
[Cl]/[Br]
[Cl]/[Br]
[Cl]/[Br]
[Cl]/[Br]
D
E
E
E
E
E
E
F
F
F
F
F
F
1, 1
4, 2
25, 25
‒
>99
ND
9

50
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
1a
1b
1c
2
50
63
43
38
49
96
97
36
6
630
430
380
245
960
970
360

8
1a
2
12, 19
6, 6
300, 475
75, 75
900, 375
925, 375
125, 100
‒
>99
>99
>99
>99
>99
ND
50
9
50
10
11
12
13
14
15
16
17
18
3a
3b
4
36, 15
37, 15
5, 4
50
3a
3b
4
50
50

1, 1[h]
6, 21[h]
5, 8[h]
20, 12[h]
25, 18[h]
2, 4[h]
50
1a
2
10, 35
4, 7
>99
>99
>99
>99
>99

105
105
105
105
105
105
105
105
1a
1b
1c
2
76
47
51
12
44
52
8
3800
2350
2550
300
2200
2600
400
3a
3b
4
33, 20
42, 30
3, 7
3a
3b
4
[a] Reaction conditions: epoxide (5.0 mmol), Al-complex (0.002 mmol; 0.04
mol%), nucleophile (Nu: 0.010 mmol; 0.2 mol%), mesitylene (1.0 mmol), 10
bar, 2 h; Sub = substrate, Sel. = diastereo-specificity, i.e. cis/trans ratio in
the product. ND stands for not determined, [Cl] = PPNCl, [Br] = TBAB. [b]
Determined by 1H NMR (CDCl3) using mesitylene as internal standard, first
number refers to yield when using PPNCl; the average of two runs is
reported. Selectivity towards the cyclic carbonates was >99%, stereo-
selectivity indicated is towards the trans (8 and 10) or cis (9) product. [c]
TON = total turnover number observed per Al; note that complex 2 contains
two Al centers. [e] Total pressure was 15 bar, partial CO2 pressure was 10
bar. [f] Chemo-selectivity was 81%, 19% of diol product was formed. [g]
Chemo-selectivity was 91%, 9% of diol product was formed. [h] epoxide
(1.0 mmol), Al-complex (0.006 mmol; 0.12 mol%), nucleophile (0.030 mol;
0.6 mol%), MEK (0.5 mL), 2 h (for PPNCl) and 14 h (for TBAB).
[a] Reaction conditions at 25 and 50ºC: epoxide A (5.0 mmol), Al-complex
(0.005 mmol; 0.1 mol%), nucleophile (Nu: 0.025 mmol; 0.5 mol%),
mesitylene (0.5 mmol), 10 bar, 8 h (25ºC) and 3 h (50ºC); Reaction
conditions at 105ºC: epoxide A (5.0 mmol), Al-complex (0.001 mmol; 0.02
mol%), nucleophile (Nu: 0.005 mmol; 0.1 mol%), mesitylene (0.5 mmol), 10
bar, 2 h. [b] Determined by 1H NMR (CDCl3) using mesitylene as internal
standard; the average of two runs is reported. Selectivity towards the cyclic
carbonate 5 was >99%. [c] TON = total turnover number observed per Al;
note that complex 2 contains two Al centers.
This article is protected by copyright. All rights reserved.

10.1002/cssc.201601712
ChemSusChem
FULL PAPER
18, conversion 76%, TON = 3800) of all binary catalyst
combinations.
with respect to the porphyrin based catalysts 3a/TBAB and
3b/TBAB (cf., entries 14, 16 and 17). However, overall the
reactivity found with 3a/PPNCl and 3b/PPNCl stands out for the
conversion of this particular substrate, in line with the turnover
data observed when using cyclopentene oxide (E).
Whereas the aminotriphenolate complexes 1a1c in the
presence of TBAB steadily increase their performance at higher
reaction temperatures, from the date in Table 3 it is clear that
the opposite trend is noted for the porphyrin based catalysts
3a/3b and the North system 2 (entries 2123). The highest
turnover number (3800) was noted for the binary catalyst
1a/TBAB at 105ºC among the series of catalyst systems tested.
Similar order turnovers are reached with 3a (3500) or 3b (3300)
when combined with PPNCl at 90ºC (see Table 1). The reaction
temperature is thus one of the decisive process parameters for
selection of the most efficient catalyst system.
Mechanistic Considerations
The initial activities here determined for all binary catalysts
based on Al-complexes 14 using terminal (A‒C) and internal
epoxides (D‒F) demonstrate a number of interesting features.
First, higher activities are generally achieved when
aminotriphenolate Al-complexes 1a‒1c and the North complex 2
are combined with TBAB; however, the binary catalysts based
on 1a‒1c show significantly higher reactivities at relatively low
complex loadings (0.01 mol% for the terminal epoxide
conversions, 0.04‒0.60 mol% in case of the internal epoxides).
Conversion of Internal Epoxides
After examination of terminal epoxides AC as coupling partners
for CO2 and observing clear trends for the reactivity of binary
catalysts derived from complexes 14, we then turned our focus
on the use of more challenging internal epoxides DF (Table 4)
and their conversion into cyclic carbonates 810 to see whether
the reactivity and performance of the various binary catalyst
systems would follow a similar tendency.
Interestingly, most of the features noted in the conversion of
the terminal epoxides are maintained when using internal
epoxides as substrates. The porphyrin based catalysts derived
from 3a/3b and PPNCl typically give the highest conversions
(and consequently TON values; cf. entries 5, 11 and 17),
whereas in the presence of TBAB the aminophenolate Al-
complex 1a becomes more competitive displaying similar
turnover numbers (entries 2, 8 and 14). However, apart from the
similarities, there are also some other observations that add to
the overall performance of the binary catalyst systems. For
substrate
D
(trans-2,3-epoxybutane) the chemo-selectivity
towards the cyclic carbonate was >99% for all cases except for
the binary catalysts 3b/PPNCl and 3b/TBAB. For these systems
there was significant formation of a diol product (rac-butane-2,3-
diol) of up to 19% of the total amount of product formed.[22] Also,
the stereo-specificity was not quantitative in these cases, and
loss of stereo-information was noted to be more pronounced
when PPNCl was present (dr = 24:76, trans isomer major
component). Therefore, despite the higher activity noted for
3b/PPNCl and 3b/TBAB, the lower chemo-selectivity and stereo-
specificity makes the binary system based on 1a the most
attractive catalyst alternative.
The other two internal epoxides E (cyclopentene oxide) and
F
(trans-2,3-diphenyloxirane) gave rise to stereospecific
conversions as expected since the cyclic carbonate products 9
and 10 are known to be exclusively formed with a preferred cis
(9)[23] or trans (10) configuration.[24] For the porphyrin-based
binary catalysts based on 3a and 3b higher turnovers are again
reached in the presence of PPNCl as additive (entries 10, 11, 16
and 17). For the sterically most crowded substrate F the use of a
smaller nucleophile (Cl) is advantageous as much shorter time
frames (PPNCl: 2 h cf. TBAB: 14 h) are required for appreciable
conversion.[25] Upon changing to bromide as nucleophile, the
binary system 1a/TBAB provides higher conversion and turnover
Scheme 1. Part of the reported mechanistic manifolds reported for the binary
catalyst 2/TBAB and 2 alone.
This article is protected by copyright. All rights reserved.

10.1002/cssc.201601712
ChemSusChem
FULL PAPER
At ambient temperature (25ºC) the conversion of 1,2-
epoxyhexane mediated by 2/TBAB works most efficiently and
shows comparable turnover/Al center as observed for 1a‒1c in
the presence of the same nucleophile. However, both types of
catalyst systems show clear opposite temperature effects. The
North catalyst shows decreased turnovers whereas the
aminotriphenolate complexes perform better at higher reaction
temperature with turnover numbers of up to 3800/h/Al center.
These results seem to be in line with the reported mechanisms
for the aminotriphenolate complex 1a[15c] and North complex 2[26]
in the conversion of epoxides and CO2 into cyclic carbonates.
The mechanism proposed by North et al. (Scheme 1)[26a,b]
revealed a second order dependence on TBAB, meaning that
catalysis mediated by the binary combination 2/TBAB should be
sensitive towards the loading of TBAB, and particularly slowed
down at low concentration of the nucleophile. The loadings of
complex 2 (0.010.04 mol%) and TBAB (0.050.20 mol%) were
rather low (apart from those used for substrate F) throughout the
studies performed here as compared to the much higher
loadings reported in the literature (typically 1.02.5 mol% for 2
and 2.5 mol% for TBAB).[26b] Therefore, it is not surprising that
initial catalytic turnover at these lower loadings was rather
modest.
Another important aspect is that both Al centers in 2 are not
exceptionally Lewis acidic and recent work[26c] revealed that 2
itself in the absence of TBAB preferably acts as a Lewis base
rather than a Lewis acid, activating CO2 via the bridging oxygen
atom to form an interesting carbonate-bridging Al2-complex. This
alternative pathway starts off with CO2 insertion into the Al‒O‒Al
bond, followed by epoxide activation through coordination to one
of the metal centers. Subsequent intramolecular cyclization then
affords the cyclic carbonate product. In both mechanisms, the
activation of an epoxide by the Al2-complex 2 is weak, and
therefore reactions performed at higher reaction temperatures
and at lower loading of 2 should increase the dynamics of this
coordinative interaction. This results in a much lower population
of the epoxide-activated intermediate, slowing down the catalytic
transformation as was indeed observed experimentally.
As opposed to 2/TBAB the activity for the binary couple
1a/TBAB increases at higher reaction temperatures. Previously
the mechanism operative for a similar binary couple (1a/TBAI;
Scheme 2) was investigated in detail by DFT and the initial
ligand exchange was confirmed by X-ray analysis.[15c] These
studies revealed that initial formation of an epoxide-ligated
complex occurs with a free energy change of 9.9 kcal/mol
(substrate was propylene oxide), thus demonstrating the
favorable Lewis acidic character of the metal center. Compared
to 2 forming a hexa-coordinated intermediate with the epoxide
substrate, 1a exchanges a THF molecule (see Figure 1, Scheme
2) for an epoxide and initially remains five-coordinate, and
simply forms a thermodynamically more stable coordination
complex. At higher reaction temperatures it is expected that the
axial coordination site in the trigonal bipyramidal coordination
sphere remains occupied by an epoxide substrate despite a
faster ligand exchange rate with bulk epoxide. Therefore, the
observation of faster turnover for 1a/TBAB (Table 3) at higher
reaction temperatures is in line with this prediction, and the
Lewis acidity for 1a plays thus a dominant role in this
mechanism. Beside the Lewis acidity, in order to create the most
powerful binary combination, a bromide would be preferred over
chloride nucleophile since bromide has a higher nucleophilic
character and better leaving group ability in the ring-opening and
ring-closing step, respectively. This is indeed noted in the results
of Tables 1, 2 and 4 for substrates AE; only in the case of a
high degree of steric crowding in the epoxide (i.e., trans-2,3-
diphenyloxirane F), chloride-based nucleophiles are providing a
better alternative as steric features at the epoxide ring-opening
stage dominate.
Scheme 3. Proposed mechanism mediated by the binary catalyst 3b/PPNCl
based on literature data and experimental findings herein.
The porphyrin based binary catalysts derived from 3a and 3b
in general show fairly similar results in terms of initial rates in the
conversion of epoxide substrates AF, and thus it seems that
here the porphyrin substitution does not play a decisive role.
More importantly, these Al(porphyrin)s show the highest activity
when combined with PPNCl, and do not follow the typical
Scheme 2. Part of the reported mechanism mediated by the binary catalyst
1a/TBAI.
This article is protected by copyright. All rights reserved.

10.1002/cssc.201601712
ChemSusChem
FULL PAPER
reactivity trend observed for many binary catalysts comprising of
halide nucleophiles. The fact that also for the porphyrins 3a and
3b combined with TBAB at high reaction temperature (105ºC,
Table 3) lower turnover numbers are observed seems to
suggest the occurrence of a pre-equilibrium (Scheme 3).
Al(porphyrin)s have been previously shown to form hexa-
coordinated species when combined with nucleophilic additives
including DMAP (4-dimethylaminopyridine), azides and chloride
salts.[27]
epoxide ring-opening occurring at a di-chloro-Al(III)porphyrin
anion, facilitating the formation of the cyclic carbonate.
Therefore, the experimental findings reported in Tables 1, 2 and
4 follow agreeably this envisioned trend.Also, at 105ºC and
lower catalyst loading (0.02 mol% [Al]) the relative rate increase
for epoxide conversion compared to 1a/TBAB is attenuated (see
Table 3), which may be caused by a more dynamic chloride
coordination to 3b.
Darensbourg reported on the use of Al(salen) complexes
such as 4 (see Figure to Table 1) combined with NBu4X as
binary catalysts for the copolymerization of cyclohexene oxide
(CHO) and CO2.[29] Significantly higher copolymerization
activities were obtained using salen ligands equipped with
electron-withdrawing substituents that resulted in more
electrophilic Al complexes. Thus, it is reasonable to assume that
the bimetallic North catalysts 2 and Al(salen) complex 4 having
peripheral electron-donating tert-butyl groups are not Lewis
acidic enough to give high relative activity for cyclic carbonate
formation at comparatively low loadings of [Al]. This trend is
essentially observed in the conversion of all terminal and internal
epoxides AF, and complexes 2 and 4 give generally the lowest
catalytic efficiencies among the binary combinations, apart from
the reactions carried out at 25ºC using 1,2-epoxyhexane A as
substrate (Table 3).
The two most efficient catalyst systems at low loading of [Al]
and nucleophile (i.e., 1a and 3b) were then evaluated in the
synthesis of propylene carbonate (PC) using propylene oxide
(PO) C as substrate (Table 5). Previously the porphyrin catalyst
3b displayed very high turnover numbers of >100.000 in short
time frames (0.5 h) for the coupling of PO and CO2 at 120ºC/3.0
MPa using PPNCl as nucleophilic additive.[11c] Thus, we decided
to benchmark the performance of 1a and 3b under similar
conditions using this specific substrate,[30] and investigated the
influence of the relative loading of the halide on the overall
performance. Both 1a and 3b were combined with TBAB and
PPNCl in different ratios (1:5 and 1:120) in standard autoclave
reactors. The conversion was determined after a pre-set time
interval, i.e. 0.5 h for the reactions performed with a 1:120
[Al]/Nu ratio, and 18 h for those reactions performed with 1:5
combinations of Al-complex and nucleophile. In all cases a clear
positive effect of the addition of the Al-complex on the total
conversion of PO into PC was noted, as the blank experiments
in the absence of the Al-complexes gave much lower conversion
(entries 14).
Interestingly, under these more dilute conditions the relative
performance of complex 1a is improved and high turnovers are
achieved in the presence of 5 equiv. of TBAB or PPNCl (entries
5 and 7; TON up to 16.1  103). When the relative loading of
TBAB or PPNCl is increased (120 equiv., entries 6 and 8),
however, a sharp decrease in overall performance is noted
which seems to suggest that a too large excess of halide
additive may result in competition for coordination to the Al
center and thus block turnover of the epoxide substrate. This
seems to be in line with the reported mechanism for 1a/TBAI,[15c]
where initial epoxide coordination to the metal center is a key
step towards the formation of the cyclic carbonate.
Table 5. Comparison of the reactivities of binary Al-catalysts 1a/Nu and
3a/Nu (Nu = TBAB, PPNCl) in the conversion of propylene oxide C into
cyclic carbonate 5 at different metal-to-nucleophile ratios.[a]
Entry
1
Cat

Nu
ratio Nu/Al
Yield [%][b]
TON[c]

PPNCl[d]
PPNCll[e]
TBABl[d]
TBABl[e]
TBAB


4
5
2


3


3

4


4

5
1a
1a
1a
1a
3b
3b
3b
3b
5:1
32
14
26
7
16.1
7.0
13.0
3.5
19.1
16.6
16.1
20.6
6
TBAB
120:1
5:1
7
PPNCl
PPNCl
TBAB
8
120:1
5:1
9
38
33
32
41
10
11
12
TBAB
120:1
5:1
PPNCl
PPNCl
120:1
[a] Propylene oxide (28.6 mmol), Al-complex (0.00057 mmol; 0.002 mol%),
nucleophile (Nu: 0.069 mmol or 0.0029 mmol; 0.24 or 0.01 mol%),
mesitylene (1.0 mmol), 15 bar, 90ºC; 1:5 [Al]/Nu ratios for 18 h, 1:120
[Al]/Nu ratios for 0.5 h. [b] Determined by 1H NMR (CDCl3) using mesitylene
as internal standard; the average of two runs is reported. Selectivity towards
propylene carbonate was >99%. [c] TON = total turnover number ( 103)
observed per Al center. [d] Blank reaction using 0.0029 mmol Nu, 18 h. [e]
Blank reaction using 0.069 mmol Nu, 0.5 h.
Importantly, the copolymerization of propylene oxide (PO)
and CO2 can be mediated by a Al(porphyrin)/PPNCl combination.
The reaction was found to be first order in [Al] and the
nucleophilic ring-opening of the coordinated epoxide was
suggested to occur on the same face of the porphyrin scaffold.
According to the Pearson acid base concept,[28] a hard Lewis
acid (the Al-complex) binds more strongly with hard ligands such
as chloride than with softer donors such as bromide or iodide.
Such a behavior would therefore increase the probability for
This article is protected by copyright. All rights reserved.

10.1002/cssc.201601712
ChemSusChem
FULL PAPER
The porphyrin binary catalysts based on 3b, however, do not
show this trend, and effectively at higher PPNCl loading (120
equiv.) the catalytic performance is further improved (cf., entries
10 and 12) with a significantly higher turnover (20.6  103)
reported under these conditions. This results also aligns well
with the mechanistic proposal that an initial equilibrium exists
where the Al(porphyrin)Cl coordinates a chloride anion before it
interacts with the epoxide substrate to facilitate its ring-opening
(Scheme 3). All together, the data presented in Table 5
illustrates that both 1a and 3b provide highly active binary
catalysts under dilute conditions, and result in high turnover
numbers in the conversion of PO into PC at different [Al]/Nu
ratios.[31]
Experimental Section
General catalytic procedure
The respective epoxide, Al-complex, co-catalyst (PPN-Cl, TBAB
or TBAI) and internal standard (mesitylene) were charged into a
stainless steel autoclave. The autoclave was then sealed and
heated to the required temperature while stirring. After reaching
the selected reaction temperature, the autoclave was
pressurized with CO2 to the desired pressure and left stirring. At
the end of the chosen time interval, the autoclave was cooled to
rt and then carefully depressurized. An aliquot of the reaction
mixture was taken for analysis and the conversion was
1
determined by H NMR spectroscopy in CDCl3. The identities of
the cyclic carbonate products were confirmed by comparison to
literature data. For specific details on the used (parallel) reactor
systems, their volumes and amount of substrates, Al complexes
and additives, see the Supporting Info (SI). Details concerning
the Al-complexes used in this work and their characterization are
provided in the SI.
Conclusions
Herein we have described the detailed comparison between a
series of Al-based binary catalysts derived from complexes 14
and their (initial) efficiencies to mediate the coupling between
various terminal and internal epoxides, and CO2 to produce their
corresponding cyclic carbonates. Under more dilute catalysis
conditions using low loadings of [Al] and [Nu] while fixing their
relative ratio (1:5), the binary catalysts 1a/TBAB and 3b/PPNCl
display the highest reactivities among all the binary
combinations tested, a trend that was observed for all substrates
tested (AF); in the conversion of highly sterically crowded
trans-2,3-diphenyloxirane, however, the use of a chloride-based
nucleophile is more beneficial.
Acknowledgements
We thank ICIQ, ICREA, and the Spanish Ministerio de
Economía y Competitividad (MINECO) through project CTQ-
2014-60419-R and the Severo Ochoa Excellence Accreditation
2014–2018 through project SEV-2013-0319. JR thanks ICIQ for
a predoctoral fellowship.
These studies also demonstrate that benchmarking
studies[30] are vital to provide insight into the relative reactivities
of binary catalyst systems, and to determine the best operating
window (temperature, metal-to-nucleophile ratio, solvent, scale)
for each individual catalyst. The known mechanistic manifolds
for epoxide conversion into cyclic carbonate can be used a
directing blue-prints to select the most appropriate catalyst
system depending on the reaction temperature, the nature of the
substrate (focusing on chemo-selectivity, cf. conversion of
substrate D in Table 4) and the accessibility of the catalyst
structures. The latter feature can be of high importance in those
cases where scale up of the catalytic process is desired and
larger quantities of catalyst are thus required.
Keywords: aluminum • benchmarking • carbon dioxide • cyclic
organic carbonates • homogeneous catalysis
[1]
For some general reviews on CO2 valorization/conversion: a) J.
Klankermayer, S. Wesselbaum, K. Beydoun, W. Leitner, Angew. Chem.
Int. Ed. 2016, 55, 7296-7343; b) Q. Liu, L. Wu, R. Jackstell, M. Beller,
M. Nat. Commun. 2015, 6, 5933; c) N. Kielland, C. J. Whiteoak, A. W.
Kleij, A. W. Adv. Synth. Catal. 2013, 355, 2115-2138; d) M. Cokoja, C.
Bruckmeier, B. Rieger, W. A. Herrmann, F. E. Kühn, Angew. Chem. Int.
Ed. 2011, 50, 8510-8537; e) X.-B. Lu, D. J. Darensbourg, Chem. Soc.
Rev. 2012, 41, 1462-1484; f) B. Yu, L.-N. He, ChemSusChem 2015, 8,
52-62; g) C. Maeda, Y. Miyazaki, T. Ema, Catal. Sci. Technol. 2014, 4,
1482-1497.
Detailed benchmarking studies for similar type binary
catalysts such as described in this work offer a way to compare
performance metrics under identical reaction conditions and to
unravel intrinsic and relative reactivities. These benchmarking
approaches are essential to support general claims on catalyst
performance in a wider context, thereby stimulating careful
assessment of other considerations that are also important to
select the most appropriate catalyst for CO2 conversion.[31]
[2]
For descriptive contributions relating to CO2 chemistry and utilization: a)
M. Aresta, A. Dibenedetto, A. Angelini, Chem. Rev. 2014, 114, 1709-
1742; b) M. Peters, B. Köhler, W. Kuckshinrichs, W. Leitner, P.
Markewitz, T. E. Müller, ChemSusChem 2011, 4, 1216-1240; c) E. A.
Quadrelli, G. Centi, J.-L. Duplan, S. Perathoner, ChemSusChem 2011,
4, 1194-1215, d) M. Aresta, Carbon Dioxide as Chemical Feedstock,
Wiley-VCH, Weinheim, Germany, 2010.
[3]
[4]
Earlier accounts on the use of epoxides as coupling partners for CO2:
a) A.-A. G. Shaik, S. Sivaram, Chem. Rev. 1996, 96, 951-976; b) M.
North, R. Pasquale, C. Young, Green. Chem. 2010, 12, 1514-1539; c)
T. Sakakura, J.-C. Choi, H. Yasuda, Chem. Rev. 2007, 107, 2365-
2387; d) A. Decortes, A. M. Castilla, A. W. Kleij, Angew. Chem. Int. Ed.
2010, 49, 9822-9837.
For some examples using hydrogen as
a reducing agent in the
synthesis of MeOH and/or formic acid from CO2: a) W.-H. Wang, Y.
Himeda, J. T. Muckerman, G. F. Manbeck, E. Fujita, Chem. Rev. 2015,
This article is protected by copyright. All rights reserved.

10.1002/cssc.201601712
ChemSusChem
FULL PAPER
115, 12936-12973; b) W. Wang, S. Wang, X. Ma, J.Gong, Chem. Soc.
C. Staudt, O. Walter, T. A. Zevaco, E. Dinjus, Catal. Sci. Technol. 2014,
4, 1658-1673; c) Y. Ren, J. Chen, C. Qi, H. Jiang, ChemCatChem 2015,
7, 1535-1538; d) A. Decortes, A. W. Kleij, ChemCatChem 2011, 3, 831-
834; e) A. Decortes, M. Martínez Belmonte, J. Benet-Buchholz, A. W.
Kleij, Chem. Commun. 2010, 46, 4580-4582.
Rev. 2011, 40, 3703-3727; c) G. A. Olah, A. Goeppert, S. G. K.
Prakash, Beyond Oil and Gas: The Methanol Economy, Wiley-VCH,
Weinheim, Germany, 2009.
[5]
[6]
A. Tlili, E. Blondiaux, X. Frogneux, T. Cantat, Green Chem. 2015,17,
157-168.
[15] Al-based catalysis: a) D. Tian, B. Liu, Q. Gan, H. Li, D. J. Darensbourg,
ACS Catal. 2012, 2, 2029-2035; b) M. North, S. C. Z. Quek, N. E.
Pridmore, A. C. Whitwood, X. Wu, ACS Catal. 2015, 5, 3398-3402; c) C.
J. Whiteoak, N. Kielland, V. Laserna, F. Castro-Gómez, E. Martin, E. C.
Escudero-Adán, C. Bo, A. W. Kleij, Chem. Eur. J. 2014, 20, 2264-2275;
d) J. Martínez, J. A. Castro-Osma, A. Earlam, C. Alonso-Moreno, A.
Otero, A. Lara-Sánchez, M. North, A. Rodríguez-Diéguez, Chem. Eur. J.
2015, 21, 9850-9862; e) M. Cozzolino, K. Press, M. Mazzeo, M.
Lamberti, ChemCatChem 2015, 8, 455-460; f) M. North, S. C. Z. Quek,
N. E. Pridmore, A. C. Whitwood, X. Wu, ACS Catal. 2015, 5, 3398-3402.
[16] a) T. Ema, Y. Miyazaki, J. Shimonishi, C. Maeda, J.-Y. Hasegawa, J.
Am. Chem. Soc. 2014, 136, 15270-15279; b) C. Maeda, T. Taniguchi,
K. Ogawa, T. Ema, Angew. Chem. Int. Ed. 2015, 54, 134-138.
[17] C. Maeda, J. Shimonishi, R. Miyazaki, J.-Y. Hasegawa, T. Ema, Chem.
Eur. J. 2016, 22, 6556-6563.
For recent reviews on cyclic carbonates: a) C. Martín, G. Fiorani, A. W.
Kleij, ACS Catal. 2015, 5, 1353-1370; b) J. W. Comerford, I. D. V.
Ingram, M. North, X. Wu, Green Chem. 2015, 17, 1966-1987; c) G.
Fiorani, W. Guo, A. W. Kleij, Green Chem. 2015, 17, 1375-1389; d) P.
P. Pescarmona, M. Taherimehr, Catal. Sci. Technol. 2012, 2, 2169-
2187.
[7]
a) A. Tullo, Chem. Eng. News 2008, 86, 21-22; b) W. J. Peppel, Ind.
Eng. Chem. 1958, 50, 767−770. For general information about the
application potential of COCs: c) B. Schꢀffner, F. Schꢀffner, S. P.
Verevkin, A. Bꢁrner, Chem. Rev. 2010, 110, 4554−4581.
a) A. Decortes, A. M. Castilla, A. W. Kleij, Angew. Chem. Int. Ed. 2010,
49, 9822-9837; b) C. J. Whiteoak, A. W. Kleij, Synlett 2013, 24, 1748-
1756. See also references 6a,b,d.
[8]
[9]
For a recent review on the use of organocatalysis in COC formation
see: M. Cokoja, M. E. Wilhelm, M. H. Anthofer, W. A. Herrmann, F. E.
Kühn, ChemSusChem 2015, 8, 2436-2454. See also reference 6c.
[18] As far as we know, benchmarking studies such as detailed herein have
not been reported in the literature, and we are aware of only one similar
study where several hydrogen-bond donor type organocatalysts have
been compared in the coupling of PO and CO2, see: M. Alves, B.
Grignard, S. Gennen, R. Mereau, C. Detrembleur, C. Jerome, T.
Tassaing, Catal. Sci. Technol. 2015, 5, 4636-4643.
[10] For recent examples of new enantio-, diastereo- or regioselective
approaches towards or using COCs see: a) W. Guo, V. Laserna, E.
Martin, E. C. Escudero-Adán, A. W. Kleij, Chem. Eur. J. 2016, 22,
1722-1727; b) S. Sopeña, V. Laserna, W. Guo, E. Martin, E. C.
Escudero-Adán, A. W. Kleij, Adv. Synth. Catal. 2016, 358, 2172-2178;
c) V. Laserna, G. Fiorani, C. J. Whiteoak, E. Martin, E. Escudero-Adán,
A. W. Kleij, Angew. Chem. Int. Ed. 2014, 53, 10416-10419; d) Y. Liu,
W.-M. Ren, K.-K. He, W.-Z Zhang, W.-B. Li, M. Wang, X.-B. Lu, J. Org.
Chem. 2016, 81, 8959-8966.
[19] Initial reactivities were examined at 90ºC as both the Kleij and
Wang/Qin systems have previously been studied at these elevated
temperatures showing high turnover, see references 11a, 11c and 15c.
[20] Note that we here preferably used PPNCl and not TBAC
(tetrabutylammonium chloride) as the latter is much more hygroscopic
and thus more difficult to handle when using minimal amounts of
reagents.
[11] Highly (re)active systems: a) C. J. Whiteoak, N. Kielland, V. Laserna, E.
C. Escudero-Adán, E. Martin, A. W. Kleij, J. Am. Chem. Soc. 2013, 135,
1228-1231; b) T. Ema, Y. Miyazaki, S. Koyama, Y. Yano, T. Sakai,
Chem. Commun. 2012, 48, 4489-4491; c) Y. Qin, H. Guo, X. Sheng, X.
Wang, F. Wang, Green Chem. 2015, 17, 2853-2858; d) J. Qin, P. Wang,
Q. Li, Y. Zhang, D. Yuan, Y. Yao, Chem. Commun. 2014, 50, 10952-
10955; e) F. Della Monica, S. V. C. Vummaleti, A. Buonerba, A. De Nisi,
M. Monari, S. Milione, A. Grassi, L. Cavallo, C. Capacchione, Adv.
Synth. Catal. 2016, 358, 3231-3243. Recently a binary organocatalyst
was reported that showed initial turnover frequencies of up to 500/h,
see: f) L. Martínez-Rodríguez, J. Otalora Garmilla, A. W. Kleij,
ChemSusChem 2016, 9, 749-755; g) J. A. Castro-Osma, K. J. Lamb, M.
North, ACS Catal. 2016, 6, 5012-5025.
[21] F. Castro-Gomez, G. Salassa, A.W. Kleij, C. Bo, Chem. Eur. J. 2013,
19, 6289-6298.
[22] The formation of
a diol product was independently checked and
confirmed by comparison with the 1H NMR spectrum of an authentic
sample.
[23] Previous work from the Darensbourg group has experimentally and
computationally shown that the formation of cyclopentene carbonate
trans-9 is thermodynamically not feasible due to a high level of ring
strain, see: a) D. J. Darensbourg, A. D. Yeung, Macromolecules 2013,
46, 83–95; b) D. J. Darensbourg, A. D. Yeung, S.-H. Wei, Green Chem.
2013, 15, 1578-1583.
[12] For some recent relevant examples: a) G. L. Gregory, L. M. Jenisch, B.
Charles, G. Kociok-Köhn, A. Buchard, Macromolecules 2016, 49,
7165-7169; b) F. Auriemma, C. de Rosa, M. Rosaria Di Caprio, R. di
Girolamo, W. Chadwick Ellis, G. W. Coates, Angew. Chem. Int. Ed.
2014, 54, 1215-1218; c) L. Peña Carrodeguas, J. González-Fabra, F.
Castro-Gómez, C. Bo, A. W. Kleij, Chem. Eur. J. 2015, 21, 6115-6122;
d) C. Martín, A. W. Kleij, Macromolecules 2016, 49, 6285-6295; e) O.
Hauenstein, M. Reiter, S. Agarwal, B. Rieger, A. Greiner, Green Chem.
2016, 18, 760-770.
[24] The cis-carbonate 10 has been reported by reacting an appropriate diol
precursor with CO2/BuBr mediated by excess DBU (DBU
= ,8-
diazabicyclo[5.4.0]undec-7-ene), see: a) T. Kitamura, Y. Inoue, T.
Maeda, J. Oyamada, Synth. Commun. 2016, 46, 39-45; See also: b) M.
A. Casadei, S. Cesa, M. Feroci, A. Inesi, New J. Chem. 1999, 23, 433-
436. However, as far as we know, the formation of the trans carbonate
10 from the pure trans epoxide F is typically quantitative because of the
higher ring strain in the cis carbonate product: c) C. Beattie, M. North, P.
Villuendas, C. Young, J. Org. Chem. 2013, 78, 419-426. See also
references 11a, 11e, 11g, 13b and 17.
[13] Fe-based catalysis: a) A. Buonerba, A. De Nisi, A. Grassi, S. Milione, C.
Capacchione, S. Vagin, B. Rieger, Catal. Sci. Technol. 2015, 5, 118-
123; b) C.J. Whiteoak, E. Martin, E. Escudero-Adán, A.W. Kleij, Adv.
Synth. Catal. 2013, 355, 2233-2239; c) C. J. Whiteoak, E. Martin, M.
Martínez Belmonte, J. Benet-Buchholz, A. W. Kleij, Adv. Synth. Catal.
2012, 354, 469-476; d) A. Buchard, M. R. Kember, K. G. Sandeman, C.
K. Williams, Chem. Commun. 2011, 47, 212-214; e) M. Adolph, T. A.
Zevaco, C. Altesleben, O. Walter, E. Dinjus, Dalton Trans. 2014, 43,
3285-3296; f) M. A. Fuchs, T. A. Zevaco, E. Ember, O. Walter, I. Held,
E. Dinjus, M. Döring, Dalton Trans. 2013, 42, 5322-5329.
[25] The catalysis data obtained with substrate F seems to suggest that for
sterically hindered internal epoxide conversion chloride nucleophiles
become competitive alternatives, and the nucleophilic ring-opening of
the epoxide is more easily facilitated by these weaker though smaller
nucleophiles.
[26] Mechanistic details for the binary catalysts based on 2/TBAB (25a+b)
and 2 only (25c), respectively: a) M. North, R. Pasquale, Angew. Chem.
Int. Ed. 2009, 48, 2946-2948; b) W. Clegg, R. W. Harrington, M. North,
R. Pasquale, Chem. Eur. J. 2010, 16, 6828-6843; c) J. A. Castro-Osma,
M. North, W. K. Offermans, W. Leitner, T. E. Müller, ChemSusChem
2016, 9, 791-794.
[14] Zn-based catalysis: a) J. Hu, J. Ma, Q. Zhu, Q. Qian, H. Han, Q. Mei, B.
Han, Green Chem. 2016, 18, 382-385; b) M. A. Fuchs, C. Altesleben, S.
This article is protected by copyright. All rights reserved.

10.1002/cssc.201601712
ChemSusChem
FULL PAPER
[27] C. Chatterjee, M. H. Chisholm, Inorg. Chem. 2011, 50, 4481-4492.
[28] R. G. Pearson, J. Am. Chem. Soc. 1963, 85, 3533–3539.
[29] Donald J. Darensbourg, D. R. Billodeaux, Inorg. Chem. 2005, 44, 1433-
1442.
[30] Note that the scale of the reaction (28.6 mmol of PO) was significantly
higher in these cases compared to the results reported in Tables 1-4,
which can positively influence the gas/liquid phase equilibrium for the
CO2 reagent. Consequently, higher turnover numbers may be obtained.
[31] A comparison of catalysts 1a (Kleij), 2 (North) and 3b (Wang and Qin)
and the cost of their synthesis is provided in the Supporting Information,
Tables S1S3. From this general comparison, the use of complex 1a
versus
2 and 3b seems attractive as the cost (per mmole) is
(significantly) lower, whereas the synthetic difficulty towards the
preparation of all three Al complexes is fairly similar in terms of total
steps involved (two) and the use of a pyrophoric Al-reagent upon
metalation of the ligand structure. Alternatively, when sustainable
conditions
are
specifically
pursued
(low
or
ambient
temperature/pressure), the somewhat higher cost of catalyst 3b is
offset by its high activity at 25ºC, see Table 3. However, since catalyst
3b works significantly better in the presence of PPNCl (being more
expensive than TBAB), the type of nucleophile should also be a
determining factor in the choice for a suitable catalyst.
This article is protected by copyright. All rights reserved.

10.1002/cssc.201601712
ChemSusChem
FULL PAPER
Entry for the Table of Contents
FULL PAPER
Take a pick: The reactivities of a
small series of binary catalysts
derived from various combinations
of Al complexes and nucleophiles
have been examined. These
Jeroen Rintjema and Arjan W. Kleij*
benchmarking studies provide
a
Page No. – Page No.
useful comparison of their overall
performances in the coupling of both
terminal and internal epoxides, and
CO2
carbonates.
to
afford
The
their
results
cyclic
are
explained from a mechanistic point
of view, and the crucial role of the
nucleophile, its amount and catalyst
concentration are discussed.
Aluminium-Mediated Formation of
Cyclic Carbonates: Benchmarking
Catalytic Performance Metrics
This article is protected by copyright. All rights reserved.