Synthesis

Review

"This is the peer reviewed version of the following article: Synthesis 2016, 48, 3863-3878, which has been published in final form at
DOI: 10.1055/s-0035-1562520. This article may be used for non-commercial purposes in accordance with the Terms and
Conditions for Self-Archiving published by THIEME at https://www.thieme-connect.de/products/ejournals/abstract/10.1055/s-00351562520."

Template for SYNTHESIS © Thieme Stuttgart · New York

2019-05-08

page 1 of 16

Synthesis

Review

Substrate-Assisted Carbon Dioxide Activation as a Versatile Approach for
Heterocyclic Synthesis
Jeroen Rintjemaa
Arjan W. Kleij*a,b
a

Institute of Chemical Research of Catalonia (ICIQ), the
Barcelona Institute of Science and Technology, Av. Països
Catalans 16, 43007 – Tarragona (Spain).
b

Catalan Institute of Research and Advanced Studies (ICREA),
Pg. Lluis Companys 23, 08010 – Barcelona (Spain).
* indicates the main/corresponding author.
akleij@iciq.es

Received:
Accepted:
Published online:
DOI:

Abstract: The CO2 incorporation into substrates containing alcohol or amine
functional groups displays interesting substrate-controlled reactivity, both in
terms of providing milder reaction conditions and the potential towards new
reactivity modes. Various heterocyclic structures can be prepared using such
substrate-driven approach starting from functional derivatives of (homo)allylic
and propargylic alcohols/amines, epoxy alcohols and hydroxy-oxetanes among
others. This review summarizes the recent advances made in this area.
1. Introduction
2. Catalytic conversion of propargylic alcohols
3. Catalytic conversion of propargylic amines
4. (Homo)Allylic alcohol substrate conversions
5. Transformation of allylic amines
6. Epoxy alcohols/amines as substrates
7. Hydroxy oxetanes and azetidines scaffolds
8. Conversion of other substrates
9. Summary and outlook
Key words: amines – alcohols – carbon dioxide – heterocycles - homogeneous
catalysis – substrate activation

be achieved by the enzyme carbonic anhydrase (CA) that
catalyzes the hydration of CO2 with very high efficiency. The
active site of this enzyme consists of a zinc center supported by
three imidazole groups belonging to histidine residues (Figure
1).

Figure 1: Carbonic anhydrase II containing a zinc center supported by three
histidine residues (left) and its position within the enzyme (right).6

1. Introduction
Ever since the beginning of the industrial revolution, the
emission of carbon dioxide and other greenhouse gases into the
atmosphere has increased drastically. The majority of this
increase can be ascribed to industrial processes releasing
annually billions of tons of CO2.1 Increasing efforts are thus being
made to potentially alleviate the CO2 emitted to the environment,
and investigating new technologies to valorize this cheap and
renewable carbon feed stock.2 Until recently, CO2 was mainly
considered as an inert waste material that cannot be easily
utilized in chemical conversions. Over the years, however, the
synthetic potential of carbon dioxide as a carbon building block
has been tremendously explored and many new transformations
have now been developed.3 Such rejuvenated potential may
positively influence the carbon footprint of (part of) our fossilfuel depending chemical processes.4 The utilization of CO2 as a C1
building block is notwithstanding challenging due to its intrinsic
stability, and typically a catalytic method is required for its
activation and conversion.5 In this respect, nature provides an
interesting role model. Carbon dioxide activation can for instance
Template for SYNTHESIS © Thieme Stuttgart · New York

2019-05-08

Scheme 1: Proposed mechanism of activation and hydration of CO2 by
carbonic anhydrase.7

page 2 of 16

Synthesis

Review

Earlier studies inspired by the CO2 activation potential of
carbonic anhydrase focused on mimicking the enzyme´s active
site. Several zinc-based complexes have been prepared
containing a zinc-ligated water molecule and nitrogen-based
ligands that structurally resemble the histidine moieties.8
However, it remains a challenge to effectively mimic CA
activity as it not only depends on the features of the active site
but also on the surrounding enzyme pocket, i.e. the second
coordination sphere. Subsequent research revealed an important
role for the zinc-bound water molecule, which at physiological pH
is partly present as a metal-bound hydroxide ion. This hydroxide
ligand is the key element towards the catalytic hydration of
carbon dioxide (Scheme 1). The first step in the catalytic
hydration of CO2 by CA is the abstraction of a proton from the
water molecule in (A) to form a zinc-bound hydroxide ion (B). In
the next step carbon dioxide coordinates to form intermediate
(C) followed by attack of the hydroxyl group leading to species
(D). Release of the bicarbonate anion and coordination of another
water molecule finally regenerates the starting CA complex (A).

Figure 2: Various HBDs employed in the coupling of epoxides and carbon
dioxide.9, 10b

Subsequent research on the activation of carbon dioxide has
focused more on the particular function of the hydroxyl group,
rather than on synthesizing complicated structural analogues of
CA. Various research groups contributed towards this objective
leading to the synthesis of both metal-9 and organo-based10
catalysts comprising a hydroxyl group (Figure 2) that is directly
or indirectly involved in the conversion of CO2. Compelling
evidence for the interaction/activation of carbon dioxide by
alcohols to form alkyl carbonic acid species comes for instance
from its reaction with diazodiphenylmethane.11 The latter is a
compound that reacts with carboxylic acids forming related
carbonate and ether species. Alcohols were found to react with
diazodiphenylmethane in supercritical CO2 forming the
aforementioned product, which suggests the intermediacy of a
carbonic acid species. More recently the group of Stoddart
investigated the properties of a metal-organic framework (MOF)
consisting of cyclodextrin units.12 This MOF was capable of
reversibly binding CO2 through its hydroxyl groups present in the
cyclodextrin units, forming a carbonic acid species. Evidence for
this formation was provided by solid state 13C NMR, showing a
diagnostic new peak at 158 ppm (cf., formation of a carbonate)
upon exposure of the MOF to CO2.

Template for SYNTHESIS © Thieme Stuttgart · New York

2019-05-08

Many (organo)catalysts in the area of CO2 conversion
catalysis are based on phenol/alcohol derivatives such as 1,2,3trihydroxybenzene (pyrogallol; Figure 2, E) that serve as
hydrogen bond donors (HBDs; other examples include F and G)
or are able to form a metal-hydroxide intermediate that can
activate carbon dioxide (H and I). These systems have
demonstrated high catalytic efficiency in the (indirect) activation
of CO2 in the context of cyclic organic carbonate and carbamate
formation. The preparation of cyclic organic carbonates (COCs) is
among the most intensively studied transformations in the field
of CO2 catalysis.13
Some of the major limitations in CO2 catalysis are the
conversion of organic substrates under mild (ambient)
conditions, broadening of the product scope and the
development of more sustainable catalytic processes. Whereas
the vast majority of the reported catalytic strategies towards CO2
conversion are purely metal-based, the use of substrate-directed
CO2 activation remains to date underdeveloped. The
incorporation of alcohol or amine functionalities in the substrate
may assist in the activation of carbon dioxide preparing it thus
for conversion into value-added organic structures. Not only can
this facilitate the coupling of CO2 with the substrate under milder
conditions, but it also opens up new pathways to access new
products that would be difficult to synthesize by any
conventional methods. Prospectively, substrates equipped with
suitable CO2-activating groups may hold promise to further
widen the scope of functionalized organics that can be derived
from CO2 and help to increase its synthetic value. In this review
the focus will be on the latest developments in this sub-area of
CO2 catalysis describing the categories of substrates that have
been successfully utilized in substrate-controlled CO2 conversion
approaches towards the formation of various heterocyclic
structures. In each separate section, the key contributions are
highlighted together with the mechanistic understanding of the
involved organic transformations.

2. Catalytic conversion of propargylic alcohols
Among the possible scaffolds that can be used for substrate‒
assisted activation of carbon dioxide, propargylic alcohols are by
far the most widely considered. To date many contributions have
been made involving the cyclization of propargylic alcohols in the
presence of CO2 typically leading to the formation of -alkylidene
cyclic carbonates. These carbonates are important building
blocks for several post-synthetic transformations,14 including the
formation of carbamates and ureas.15

2.1 Metal catalysts
The first example of a coupling reaction between a
propargylic alcohol and carbon dioxide was already reported in
196116 followed up by a contribution from Laas et al. twenty
years later.17 Both catalytic processes use copper iodide (CuI) as
the catalyst in combination with triethylamine (NEt 3) as base
(Scheme 2). Other early examples of catalytic systems that can
mediate similar conversions employ different metals such as
palladium,18 cobalt19 and ruthenium.20

page 3 of 16

Synthesis

Review

Scheme 2: Copper catalyzed formation of -alkylidene cyclic carbonates.

was obtained. Another mechanistic clue arose from the reduction
of copper during the catalytic cycle, which is later re-oxidized by
oxygen from the air. If the reaction is performed under anaerobic
conditions, only 15% of 7 is obtained together with formation of
6 as the major product.

The popularity of copper based systems can be ascribed to its
availability, low cost and ease of use in catalysis.21 However, one
of the challenges in these copper catalyzed cyclization reactions
was the conversion of propargylic alcohols incorporating
internal alkynes. This problem was addressed in a recent
contribution from Qi and Jiang22 using a combination of CuI and
N,N-diisopropylethylamine (DIPEA). Interestingly, this recent
work demonstrated not only the conversion of such propargylic
alcohols (Scheme 3, 1) into the targeted cyclic carbonates 2, but
also delivered a new route to iodoalkylidene derivatives 3 under
appropriate and selected reaction conditions.

Scheme 4: Proposed reaction mechanism of the formation of alkylidene cyclic
carbonates (path a) and its iodinated derivatives (path b).

Scheme 3: Synthesis of the conventional -alkylidene cyclic carbonates (left)
and the -iodoalkylidene derivatives (right). Below representative examples
of the product scope achieved in either type of conversion.

Incorporation of iodine into the carbonate structure gives the
(E)-iodoalkylidene cyclic carbonates, of which the synthesis is
related to that reported by Minakata in the formation of
iodocarbonates (see section on allylic alcohols conversion).23 A
different iodine source is used in this case, which is generated in
situ from copper and iodide salts.
Scheme 4 displays the proposed reaction mechanism, which
starts with the activation of CO2 by the propargylic alcohol group
and coordination of copper to the alkyne 4. Cyclization occurs
through attack of the carbonate anion on the activated triple
bond forming intermediate 5. Although this vinyl-metal species
was not isolated, similar intermediates have been proposed to
exist using, for instance, silver-based catalysts.24 Simple
protonation of the copper-bound intermediate gives the cyclic
carbonate 6 through path a. The other pathway involves the prior
formation of an I3 anion coupled with the reduction of Cu(II) to
Cu(I), after which iodine replaces copper to form the iodinated
product 7. Several observations support this proposed reaction
mechanism. When the reaction is performed with I2 (i.e., a
different source of electrophilic iodine) a mixture of products
Template for SYNTHESIS © Thieme Stuttgart · New York

2019-05-08

Under the conditions outlined in Scheme 3 (with below part
of the product scope; compounds 2a-b and 3a-c) and following
the two pathways displayed in Scheme 4, two different types of
products can thus be prepared The -alkylidene cyclic
carbonates are typically obtained in moderate to high yields
depending on the substrate substitution pattern. The iodinated
variants are generally acquired in good to excellent yields of up
to 97% and cover a wide substrate scope. A limitation of this
protocol is the necessity for tertiary propargylic alcohols, as the
use of primary or secondary alcohols (Scheme 3, 2c) only leads
to trace amounts of product. A clear advantage of this method is
the availability and cost of the copper catalyst and iodine salts
that are used. However, the reactions are not really atom- and
waste-efficient producing two equivalents of copper salt and
requiring four equivalents of KI.
Reaction conditions associated to the use of copper based
chemistry discussed above typically involve high temperatures
(around 100ºC) and high pressures of CO2 (40 bar). The quest for
catalytic systems that could convert these propargylic alcohols
under milder conditions led the group of Yamada to use silverbased catalysts. In 2007 they reported on the combined use of
silver acetate (AgOAc) and DBU (1,8-diazabicyclo-undec-7-ene)
in the coupling of CO2 with propargylic alcohols.24 A variety of
substrates containing internal alkynes could be converted under
mild conditions (25ºC, 10 bar CO2). These conversions at ambient
temperature opened up the possibility for enantioselective
carbon dioxide incorporation. Three years after their initial
publication, Yamada et al. reported on the enantioselective
coupling of CO2 with bis-propargylic alcohols (Scheme 5).25
DBU was not used for the asymmetric catalysis reactions, as
it was expected that it strongly coordinates to the silver center
page 4 of 16

Synthesis

Review

and thus would negatively affect the asymmetric conversion of
the propargylic alcohol. After optimization of the reaction
conditions (3 mol% AgOAc, 1 mol% of ligand L at 0ºC) the
product (9) depicted in Scheme 5 was obtained in an excellent
yield of 98% and high ee (92%). Various groups are tolerated in
the position next to the alcohol (R2) giving the respective
products in similar high yields. When too large steric bulk was
introduced, however, the yield dropped to 58% (11c, R2 = tBu)
even when using extended reaction times. Several other
substituted phenyls and allyl-substituted propargylic alcohols 10
yielded the cyclic carbonates 11 in good yields and high ee.
Benzyl groups attached to the alkyne units resulted in fast
reactions though with significant loss of stereoselectivity
resulting in <50% ee.

propargylic alcohols (12) could be converted with good to
excellent yields (cf., products 13a-b). Several electron
withdrawing substituents were tolerated on the aromatic ring
(cf., products 13c-d), though electron donating groups led to very
poor results. Lower yields were obtained when a tolyl group was
present and no product formation was observed in the case of a
para-methoxyaryl substituent (cf., intended formation of 13e).
Changing the substitution pattern near the alcohol also gave
access to cyclic carbonate formation in good yield (compound
13f, R2 = c-hexyl; 74%). However, the substrate scope seems to
be restricted to tertiary alcohols only, as no product was obtained
for primary or secondary alcohols (example: formation of 13g).
This is a common observation in the area of propargylic
alcohol/CO2 couplings, and its origin has been proposed to be a
result of the gem-dialkyl effect.35 This effect describes the rateacceleration of a cyclization between two reactive groups, by
introducing substituents on the carbon atom that links those two
groups. In the case of propargylic alcohols this implies
substitution on the carbon connecting the alcohol with the alkyne
moiety. The gem-dialkyl effect causes angle compression,
bringing the two reactive groups closer together, also known as
the “Thorpe-Ingold” effect.

Scheme 5: Catalytic system based on AgOAc and a chiral Schiff base reported
by Yamada for the asymmetric synthesis of carbonates 11.

In the years following this seminal work, silver has
maintained its importance as a catalyst in the formation of alkylidene cyclic carbonates from propargylic alcohols. Several
systems based on silver have been developed recently,26
including the work from Liu et al. on porous organic polymers27
and contributions from the group of He.28 Other metals have also
been employed successfully, including tungstate29 and zinc.30
Further information on metal catalyzed cyclization of propargylic
alcohols with CO2 can be found in other recent overviews.31

2.2 Organocatalysts
The first example of organocatalyzed CO2 incorporation into
propargylic alcohol substrates was published in 1989 by the
group of Dixneuf.32 They found that simple tributylphosphine
(PBu3) effectively catalyzes the formation of the cyclic carbonate
product at 50ºC giving almost quantitative conversion after 20 h.
In the years after their initial publication, they further developed
the phosphine catalyzed coupling of CO2 and propargylic
alcohols.33 Recently, a contribution from the group of Ikariya
demonstrated the combination of phosphine catalysis and a
supercritical CO2 (sc-CO2) medium to facilitate similar
reactions.34 Scheme 6 reports part of the substrate scope
performed under the optimal conditions (sc-CO2) using 5 mol%
of PBu3 as an organocatalyst. Both terminal and internal
Template for SYNTHESIS © Thieme Stuttgart · New York

2019-05-08

Scheme 6: Organocatalyzed carbonate formation under sc-CO2 conditions.

Concerning cyclization reactions with CO2, there are some
examples where primary and secondary propargylic alcohols can
be used as substrates. However, these reactions typically require
high CO2 pressures to give the cyclic carbonates21b,26a or are
converted in the presence of amines to yield linear carbamates.36
Other organocatalysts that have been used in the same
context include a K2CO3-crown ether,37 bicyclic guanidines,38
alkoxide-functionalized imidazolium betaines39 and Nheterocyclic carbenes (NHCs).40 Closely related to the latter type
of organocatalysts are N-heterocyclic olefins (NHOs), which have
been shown to efficiently activate and bind CO2.41 Scheme 7
depicts the structure of the NHO and subsequent formation of the
adduct upon activation of CO2. An interesting property of these
types of complexes is that the C(carboxylate)-C(NHO) bond
length is significantly longer compared to its NHC-CO2 analogue.
This makes the NHO-CO2 adduct less stable but more suitable for
catalytic turnover as facile decarboxylation takes place under
milder temperature conditions (60ºC). These properties allow
for an improved reactivity in the carboxylative cyclization of
propargylic alcohols.
Two different mechanistic pathways were suggested by the
authors, the first one being based on the mechanism proposed by
page 5 of 16

Synthesis

Review

Ikariya regarding related NHC catalysis.40a In this case the NHOCO2 adduct thus attacks the alkyne triple bond, followed by
hydrogen transfer to form an alkoxide intermediate. This
alkoxide species performs an intramolecular cyclization, which
releases the cyclic carbonate product and regenerates the
catalyst. Due to the different properties of the NHOs there is a
second reaction pathway possible, which is outlined in scheme 8.

3. Catalytic conversion of propargylic amines
Closely related to propargylic alcohols are their N-containing
analogues, propargylic amines. Their cyclization with CO2 leads
to the formation of cyclic carbamates and these 5-membered
heterocyclic products are usually described as oxazolidinones.
These compounds are versatile chemical products that have
found application in various fields of chemistry, for instance as
chiral auxiliaries,42 scaffolds for antimicrobial drugs43 or antidepressants,44 and in organic synthesis.45 In contrast to the large
number of reactions based on propargylic alcohols that emerged
during the end of the last century, relatively few examples have
been reported for propargylic amines, probably due to the fact
that the latter have more restricted accessibility. First, only the
intermolecular three-component coupling had been known,
combining an amine, alkyne and CO2 to form the cyclic
carbamates catalyzed by either iron46 or ruthenium complexes.47
The first example of a propargylic amine being directly used in a
carboxylative cyclization was reported in 1996 by the group of
Costa (Scheme 9).48

Scheme 7: Synthesis of an NHO-CO2 adduct from the NHO and CO2 and
subsequent use in the formation of -alkylidene cyclic carbonates.

Scheme 9: Organocatalyzed incorporation of CO2 into propargylic alcohols.

They reported on the formation of 5-membered cyclic
carbamates using the bicyclic guanidinium base MTBD as
catalyst. Various substrates could be converted to their
respective carbamate products under ambient conditions (1 bar
CO2, rt), although primary or secondary propargylic amines did
not provide productive catalysis. When the temperature was
raised to 110ºC and the CO2 pressure to 10 bar, both targeted
products could be obtained in appreciable yields in the range 60‒
70%. This result supports the rate-enhancement caused by the
gem-dialkyl effect discussed previously (see section 2.2 on
propargylic alcohol conversion).

Scheme 8: Proposed reaction mechanism for the NHO-catalyzed cyclization of
propargylic alcohols with CO2.

In the catalytic cycle displayed in Scheme 8, the NHO (14)
first deprotonates the propargylic alcohol, forming an alkoxide
and an imidazolium species (15). CO2 subsequently reacts with
the substrate to form a carboxylate anion (16) that attacks the
alkyne moiety to give intermediate 17. Rapid proton exchange
from the imidazolium intermediate to the carbanionic species
closes the cycle with reformation of 14 and release of the alkylidene cyclic carbonate. Kinetic isotope effect (KIE)
measurements show that the second pathway (Scheme 8) makes
a smaller contribution to the overall reaction compared to the
manifold previously proposed for NHC catalysts.40a

Template for SYNTHESIS © Thieme Stuttgart · New York

2019-05-08

In the following years several groups reported on the
synthesis of oxazolidinones and derivatives from propargylic
amines using catalytic systems based on silver,49 gold,50
palladium,51 superbases52 and ionic liquids.53 Interestingly, this
particular cyclization reaction with CO2 can even be performed in
the absence of catalyst and solvent. Ikariya reported on the
formation of cyclic carbamates from propargylic amines using
supercritical CO2 both as reactant and solvent without the
requirement of any further additive.54 Even more recently, a
contribution towards cyclic carbamate synthesis was presented
by the group of Nevado.55
The latter example work describes a three-component
tandem reaction involving the well-known cyclization step using
CO2 and the propargylic amine (18) followed by an in situ Pdmediated cross-coupling of the -alkylidene cyclic carbonate
intermediate with aryl iodides (Scheme 10). Under optimized
conditions, a series of substituted oxazolidinones was prepared.
Varying the nature of the aryl iodide generally led to the targeted
products in high yields of up to 95% (19a-c); moderate yields
were noted when electron-donating groups (EDGs) such as parapage 6 of 16

Synthesis

Review

OMe (19d) were present in the aryl iodide reagent. Terminal
alkynes could also be converted: remarkably, product 19e was
isolated as the (E)-isomer in contrast to the (Z)-products which
are normally obtained through coupling of internal alkynes with
CO2. Furthermore, various N-substituted propargylic amines
were tested, giving good yields when the N-R substituents are
nBu (19f), iPr (19g) or allyl. When a sterically more congested
tBu group was present, only a moderate yield of the cyclic
carbamate 19h could be obtained even after a 3 day reaction
time.

Assumingly, the propargylic amine containing a terminal alkyne
moiety first undergoes a Sonogashira type cross-coupling. This is
suggested by the formation of an internal alkyne (22) when the
reaction of 20 (Scheme 10, bottom) is performed in the absence
of CO2.

Scheme 11: Schematic representation of the carboxylative cyclization and
reductive elimination steps in the Pd-mediated three-component coupling of
propargylic amines, CO2 and aryl iodides.

Scheme 10: Synthesis of oxazolidinones incorporating tetra-substituted exocyclic double bonds starting from internal (top) or terminal (bottom)
propargylic alcohols.

In the second part of the substrate scope phase, a larger
amount of aryl iodide was added (3 equiv.) as well as an
additional catalytic amount of CuI. This smoothly led to the
formation of cyclic carbamates having a tetra-substituted exocyclic double bond starting from propargylic amines with a
terminal alkyne group (Scheme 10, below). Electronic variation
on the aromatic ring shows a similar effect as noted within the
other series of cyclic carbamate products with good yields of up
to 72% when using the non-substituted aryl iodide (21a) and
substrates carrying EWG substituents (21b), and poorer results
for substrates having EDGs (21c; 48% yield). The chemoselective formation of the cyclic carbamate product derived from
aryl iodides equipped with additional C-Br functions was also
feasible, and the latter functional groups did not participate in the
cross-coupling event (cf., formation of 21e). A mechanism was
tentatively proposed combining both the carboxylative
cyclization and cross-coupling chemistry (Scheme 11).
Template for SYNTHESIS © Thieme Stuttgart · New York

2019-05-08

The second step involves the activation of CO2 by the amine
unit and coordination of palladium to the alkyne triple bond to
give 23. Ring closure leads to the formation of palladium-vinyl
intermediate 24 similar to that described for the copper
intermediate in Scheme 4. The difference here is that the
palladium ion is not replaced by a hydrogen or iodine atom, but
instead is involved in a reductive elimination process to give the
final product 25 containing two aromatic groups on the terminal
end of the double bond. Evidence for this mechanism was
provided by a control experiment that showed the inability of
mono-substituted products such as 19e (Scheme 10, top) to
undergo an additional cross-coupling reaction to afford the tetrasubstituted olefin analogue. Overall this approach displays an
efficient way to access various types of highly substituted
oxazolidinones through an efficient multicomponent reaction.
Although the catalytic system consisting of Pd, Cu and a base is
more complex than others reported in the literature, the wide
product diversity makes it a valuable contribution to this area of
CO2 conversion.

4. (Homo)Allylic alcohol substrate conversions
Rather ideal candidates for the substrate-assisted activation
of CO2 are allylic alcohols. Their main advantages are their
availability and broad range of substitution patterns that can be
accessed. There is a broad range of allylic alcohols known that can
be directly used without the need for prior modification before
catalytic conversion. More complex derivatives can be easily
synthesized, for instance from ketones through the use of
Grignard reagents.56 However, allylic alcohols are less reactive
than propargylic systems in the activation and conversion of CO 2
and therefore these substrates require thus a different approach
to produce effective catalytic turnover.
The first synthesis of cyclic carbonates via the coupling of
allylic alcohols with CO2 was reported by Cardillo in 1981.57 This
paper described the regio- and stereoselective functionalization
of (homo)allylic alcohols. Key step in this process is the cyclofunctionalization of the alcohol-carbonate derivatives,
page 7 of 16

Synthesis

Review

intermediates that are formed through coupling of CO2 with
allylic- or homo-allylic alcohols (Scheme 12, top).

variation of this protocol towards the formation of iodofunctionalized cyclic carbonates.23 Instead of utilizing a strong
base to force the formation of a carbonate intermediate, they
made use of the reaction of CO2 with alcohols that results in the
formation of an unstable carbonic acid monoester (Scheme 13,
37). The formation of this carbonic acid monoester is
thermodynamically highly disfavored. However, the carbonate
intermediate can be trapped by using a reactive iodonium source,
i.e. tert-butyl hypoiodite (tBuOI).59 Subsequent ring-closure
forms the cyclic carbonate (39) and effectively displaces the
equilibrium towards the targeted product. Whereas the reaction
of molecular iodine with olefins usually proceeds through an
iodonium intermediate, a different mechanism is likely operative
when using tBuOI. Formation of an iodonium species with tBuOI
has been reported, though specifically for substrates lacking a
hydroxyl group.60 It was proposed that the iodine atom from
tBuOI initially replaces the proton of the alcohol group to
generate a reactive intermediate (38) that can cyclize to give the
cyclic carbonate (39). A similar result was obtained for amides,
which can be deprotonated by the hypoiodite reagent leading to
the formation of oxazolines.61 This different reaction pathway
can be explained by the potential of tBuOI to readily form
radicals. Whereas molecular iodine can lead to radical pathways
under the influence of light62 or radical initiators,63 tBuOI does
not require any external stimulus.

Scheme 12: Proposed mechanism towards the formation of six-membered
cyclic iodocarbonates from (homo)allylic alcohols. Regioselective cyclization
leading to either 5- or 6-membered cyclic carbonates is shown below with
the carbon centers of the involved substrates highlighted.

A homoallylic alcohol (26) was treated with nBuLi after
which CO2 was added to form the linear, open chain carbonate
intermediate (27). Subsequent treatment with iodine led,
assumingly, to an iodonium intermediate (28) thereby activating
the double bond for intramolecular nucleophilic attack. Although
the authors did not comment on the exact role of iodine in this
reaction sequence, it is likely to form an iodonium species under
these conditions.58 Intramolecular attack of the carbonate moiety
gives the six-membered cyclic carbonate (29) and installs a
synthetically useful alkyliodide group on the terminal carbon
center of the carbonate side chain. The cyclization step is highly
regioselective, producing exclusively 5-membered cyclic
carbonates (5MCCs) from allylic alcohols and six-membered
(6MCCs) derivatives from the respective homo-allylic alcohols
(Scheme 12; below). Several iodocarbonates can be easily
prepared by this methodology. Interestingly, the best carbonate
yield was obtained when nerolidol was used as substrate, a
natural product containing three non-conjugated double bonds.
Only the double bond in the allylic position from the alcohol
reacts, giving the corresponding 5MCC in 90% yield (32). The
main disadvantage of this method is, however, the need for a
strong and air-sensitive base (nBuLi) for the reaction to proceed.
This limits the scope of the reaction as many functional groups
are not compatible under these conditions.
Nearly thirty years after the initial work from Cardillo was
reported,57 Minakata and co-workers presented an elegant
Template for SYNTHESIS © Thieme Stuttgart · New York

2019-05-08

Scheme 13: Proposed mechanism of trapping of a carbonic acid monoester
with tert-butyl hypoiodite (tBuOI). Below the stereo- and regio-selective
formation of cyclic carbonates with tBuOI (selected examples) under mild
conditions (1 bar, ‒20ºC).

This approach can be applied to the conversion of a wide
variety of allylic and homo-allylic alcohols leading to the
formation of 5MCCs and 6MCCs, respectively (Scheme 13;
below). The products are generally obtained in high yields under
very mild conditions (1 bar, ‒20ºC) with reaction times ranging
from 3 to 72 h. This methodology can also be applied to the
conversion of propargylic alcohols forming 5MCCs with an exocyclic double bond (see section 2).
An enantioselective approach towards the formation of cyclic
iodocarbonates was reported by Johnston.64 Their work centers
on the use of an organocatalytic system employing dual Brønsted
acid/base catalysis in combination with N-iodosuccinimide
page 8 of 16

Synthesis

Review

(NIS). The mechanism for this reaction is similar to that described
above, with the iodine forming an iodonium intermediate that
activates the alkene for a nucleophilic attack by the carbonate
species. The difference lies in the role of the organo-catalyst
abbreviated as StilbPBAM (Scheme 14, 46) that serves as both a
Brønsted base to stabilize the carbonate moiety and creates a
chiral environment during the cyclization step.

primary alkyl iodide have been investigated including
nucleophilic substitutions and cross-coupling reactions.65

Scheme 15: Transformation of cyclic iodocarbonate structures into various
useful molecules.

5. Transformation of allylic amines

Scheme 14: Enantioselective coupling of homo-allylic alcohols and CO2.

Initial screening showed that the best yields were obtained
with the chiral PBAM catalyst over several conventional aminebases and HBDs such as DMAP, DBU and TBD. Addition of a
Brønsted acid (HNTf2) increases both the yield and ee of the
product, which is ascribed to the occurrence of a hydrogenbonding network during catalysis. Under the optimized
conditions a range of 19 substrates could be converted to their
chiral cyclic carbonate products (Scheme 14; selected examples).
Para- and meta-substitution on the aromatic ring of the homoallylic alcohol substrate gave the products in excellent yield of up
to 96% and ee´s of up to 91% (45a-b). Ortho-substituted aryl
groups were not tolerated as no product formation was observed
even after long (>96 h) reaction times (45c). The
enantioselectivity remained fairly constant around 90% upon
changing substituents, though with a drop in yield when using
halogenated aryl-substituted homo-allylic alcohols (45d-e). Even
challenging substrates bearing aliphatic substituents showed
fairly good results both in terms of yield and enantioselectivity
(45f-g).
The general development of the coupling reaction between
CO2 and (homo)-allylic alcohols has provided access to a variety
of synthetically interesting organic molecules. However, due to
the rather low reactivity of olefinic double bonds, a
stoichiometric amount of iodine is typically required for the
reaction to proceed. As a consequence, an iodide is incorporated
in the product leading to the formation of cyclic iodocarbonates
in all cases. This is not necessarily a disadvantage, as the iodine
moiety opens up possibilities for post-modification reactions as
shown in Scheme 15. Useful transformations of the cyclic
iodocarbonate include the partial and full reduction leading to
the (chiral) 6MCC (47) and 1,3-diol (48). Basic hydrolysis of the
carbonate gives the epoxide compound 49 through initial
iodohydrin formation followed by cyclization. It is important to
note here that the initial ee of the cyclic iodocarbonate (91%) is
retained in carbonate derived compounds 47-49. Other possible
modifications of this type of cyclic iodocarbonates containing a

Template for SYNTHESIS © Thieme Stuttgart · New York

2019-05-08

Similar to the carboxylative cyclization of allylic alcohols is
the formation of cyclic carbamates from allylic amines and CO2.
Early investigations described the three-component coupling
reaction of an amine, CO2 and an olefinic substrate.66 Allylic
amines were used for the first time in a related transformation
utilizing a carbonate anion,67 whereas the first example using
carbon dioxide was published in 1987 by the group of Toda.68 In
the latter example the formation of 5- and 6-membered cyclic
carbamates (oxazolidinones and oxazinanones) from allylic- and
homo-allylic amines, respectively, was reported. This approach
follows the same mechanistic pathway as described for the
conversion of allylic alcohols with CO2 and iodine (section 4). A
minor drawback of this approach is the need for longer reaction
times and the requirement of a base additive (Cs2CO3) to achieve
high yields. Improved versions of the coupling of allylic amines
with CO2 have been reported by Muñoz,69 (using a guanidine
based catalyst) and later on by Minakata.70 In the latter case an
organic additive is used (tBuOI) thereby excluding the need for a
metal or base to catalyze this reaction.
The (homo)allylic amines are transformed into their
respective products following the same mechanistic steps as
depicted in Scheme 13. A variety of cyclic carbamates can be
obtained through this method with yields ranging from 54-94%.
An interesting application for this method was found in the
synthesis of 3-amino-5-morpholinomethyl-2-oxazolidinone
(AMOZ, Scheme 16). AMOZ is a metabolite of several larger drug
molecules (for instance of furaltadone)71 which has been used as
an antibiotic.72 A simple three-step process transforms allyl
amine (50) into AMOZ (53), through initial coupling with CO2 in
the presence of tBuOI to form the iodinated oxazolidinone 51.
The alkyl iodide can then be converted by simple nucleophilic
substitution chemistry to install the morpholine unit (52). A
hydrazide is formed in the last step, giving the final product
AMOZ (53).

page 9 of 16

Synthesis

Review
a very simple way to form cyclic carbamates from allylic amines
in good yields without the need for any additives. However, this
approach is applicable to a smaller range of substrates, thereby
limiting to some extent the synthetic potential of this protocol.

6. Epoxy alcohols/amines as substrates

Scheme 16: Three-step synthesis of the AMOZ drug precursor.

Other examples of cyclic carbamate formations from allylic
alcohols include a palladium catalyzed CO2 “recycling” reaction
reported by Yoshida.73 The substrates are allylic amines bearing
a methyl carbonate group in the allylic position. The first step in
the
proposed
mechanism
involves
a
Pd-mediated
decarboxylation with subsequent formation of a -allyl-Pd
complex. Carbon dioxide is concomitantly trapped by DBU (a
bicyclic base) and transferred to the amine group. Intramolecular
nucleophilic attack of the newly formed carbamate species
results then into the final oxazolidinone product. An effective
alternative approach towards the carboxylative cyclization of
allyl amines 54 (Scheme 17) was recently developed by the group
of Della Ca´.74 They reported on the synthesis of oxazolidinones
55 from allyl amines in the absence of a metal or base catalyst.
However, the presence of a strong EWG attached to the double
bond was required to facilitate the cyclization step by the initially
formed carbamate anion. An EWG such as an ester can be used to
stabilize the negative charge that forms after intramolecular
attack followed by rapid proton transfer to give the final product.
Initially, the reaction was performed under 45-60 bar of CO2 at
90ºC to obtain the benchmark product in high yield; however,
other substrates gave much poorer results in terms of
conversion/yields. This problem could be eventually overcome
by performing the reactions in scCO2 thereby significantly
improving the yields in most cases.

The main difference between epoxy alcohols/amines and the
previously described allylic and propargylic systems is that there
are two different ways for the substrate to react with CO2.
Conversion of epoxides to cyclic carbonates by binary catalytic
systems has been well documented over the past few decades.75
Binary catalytic systems typically consist of a Lewis or Brønsted
acid (a metal complex or hydrogen bond donor (HBD)) and a
nucleophilic additive (Scheme 18; left). The Lewis acid activates
the epoxide through initial coordination followed by ringopening by the external nucleophile to give a metal-alkoxide
species, followed by insertion of CO2 into the M‒O bond and
subsequent ring-closure producing the cyclic carbonate product.
In the case of epoxy alcohols or amines, a carbonate anion can be
formed that acts an internal nucleophile. This allows for a
substrate-assisted pathway leading to a different product
(Scheme 18; right) which is discussed here in more detail.76

Scheme 18: Two different mechanistic manifolds for the reaction between
epoxy alcohols/amines and CO2.

Scheme 19: Synthesis of hydroxyl-methyl carbonates from epoxy alcohols and
CO2 under basic conditions leading to a mixture of 58 and 59.
Scheme 17: Synthesis of oxazolidinone 55 derivatives from allyl amines 54 in
scCO2.

Substrates bearing an ester group could be efficiently
converted producing the carbamates in excellent yields (55a-b).
No product formation was observed when R1 is hydrogen, alkyl
or phenyl. However, when an additional nitro group was present
on the aryl ring, the oxazolidinone product was obtained in high
yield (cf., formation of 55c-e). Note that under non-supercritical
conditions, only 51% yield of product 55c was obtained and less
than 20% in the case of 55d and 55e. Thus, this method provides
Template for SYNTHESIS © Thieme Stuttgart · New York

2019-05-08

An intermolecular variation on this reaction was reported by
Inoue in 1986, employing an aluminum porphyrin complex to
catalyze the coupling of an amine, CO2 and an epoxide.77 A more
recent contribution relevant to these conversions came from the
group of Deng who reported the formation of linear hydroxy
carbamates catalyzed by a heterogeneous Au/Fe2O3 catalyst
system.78 The intramolecular version was first described in
1987,79 although the cyclic carbonate product was not obtained
in this case. One year later Myers et al. reported on the successful

page 10 of 16

Synthesis

Review

carboxylative cyclization utilizing epoxy alcohols to form cyclic
carbonates80 (Scheme 19) mediated by cesium carbonate.
The mechanism of the reaction proceeds in a similar way as
that described for the allylic and propargylic systems. Carbon
dioxide is activated by the hydroxyl group of 56 and the newly
formed carbonate anion (57) performs an intramolecular attack
on the epoxide followed by proton transfer to furnish the cyclic
carbonate product (58). In addition, the authors observed the
formation of 59 with a final product distribution of 1:3 with 59
being the major product. In 2005, the first example of an epoxy
amine/CO2 coupling was reported by Plettner,81 who observed
the formation of a hydroxyl substituted oxazolidinone instead of
the cyclic carbonate product.
A large step forward in substrate-controlled cyclic carbonate
formation from epoxy alcohols and amines was demonstrated
very recently by Kleij et al.76 The authors showed that two
different products can be formed from a single substrate (i.e.,
product divergence) and catalyst depending on the reaction
conditions that are used (Scheme 20; AltBu as catalyst). Epoxy
alcohols bearing an alcohol group in the -position could be
converted into their respective cyclic carbonate products to
provide unusual and challenging substitution patterns that are
difficult to synthesize via conventional epoxide/CO2 coupling
reactions.

different carbonate products (Schemes 18 and 19). Nonetheless,
products 61f and 61g were obtained in good yields under the
optimized conditions. A second part of the substrate scope
focused on the conversion of epoxy amines (62) that form
oxazolidinones highly selectively in the presence of CO2 and the
Al based catalyst (Scheme 21).

Scheme 21: Use of epoxy amines and their conversion into cyclic carbamates
or oxazolidinones. Ts = tosyl, DIPEA = diisopropylethylamine, TBAB =
tetrabutylammonium bromide.

Alkyl substitutions on the epoxy amine result in relatively
fast reactions under mild temperature conditions (30-40ºC; cf.
formation of 63a-b). Under the same conditions, no reaction was
observed for the epoxy amines with an N-tosyl or N-aryl
fragment. A base (DIPEA) was thus necessary to facilitate
deprotonation of the epoxy amine, thereby favoring the
formation of a linear carbamate intermediate and product
formation (compound 63c). An aniline based epoxy amine can be
converted with complete selectivity for the carbamate product
when using DIPEA, though the reaction is relatively slow in this
case. Addition of a small amount of nucleophile significantly
speeds up the process, with only a minor loss in chemoselectivity (cf., 63d). Interestingly, the oxazolidinone product
formed here is known as Toloxatone, a drug molecule that is used
as an antidepressant.44

7. Hydroxy oxetanes and azetidines scaffolds

Scheme 20: Synthesis of highly substituted cyclic carbonates through a
substrate-controlled CO2 activation process. DIPEA = diisopropylethylamine,
TBAB = tetrabutylammonium bromide; TBAC is the chloride variant of TBAB.

Both terminal and internal epoxy alcohols (60) could be
efficiently converted leading to the formation of mono- and disubstituted cyclic carbonates in excellent yields of up to 95% (cf.,
formation of 61a-c). Sterically crowded tertiary alcohols
required higher reaction pressures. In addition, a base (DIPEA)
was added for some substrates to mediate the deprotonation of
the alcohol group and favor the formation of the linear carbonate
intermediate; in these cases the conversions led to the formation
of elusive tri-substituted carbonates (cf., formation of
compounds 61d-e). Other challenging substrates required the
addition of a small amount of nucleophile to increase overall
kinetics. The chemoselectivity, however, is slightly compromised
in those cases due to two competing pathways leading to
Template for SYNTHESIS © Thieme Stuttgart · New York

2019-05-08

Larger ring-size cyclic ethers such as oxetanes also provide
interesting substrates for CO2 activation; azetidines are the
nitrogen-containing analogues. When substituted with hydroxyl
groups, these substrates provide opportunities for the activation
of CO2 in a similar way as described for epoxy alcohols (section
6). A three-component coupling reaction between an amine, CO2
and an oxetane has been described in 1999 by Yoshida et al82
showing the potential of amines to activate CO2 in situ. Various
linear hydroxyalkyl carbamates were synthesized in the absence
of a catalyst though under relatively high temperature and
pressure conditions (120ºC and 40 bar). An intramolecular
variant was reported by the group of Murakami, describing the
cyclization of a hydroxyl-azetidine (azetidinol) with CO2.83 The
initial synthesis of the azetidinol is achieved through a solar light
driven photolysis of -amido ketones.84 Subsequent coupling of
the azetidinol and CO2 is proposed to take place according to the
mechanism depicted in Scheme 22.
The first step is the deprotonation of 64 by cesium carbonate
to give the alkoxide species 65. Reaction with CO2 gives the
carbonate anion intermediate 66, which enables ring-opening of
the azetidine through an intramolecular nucleophilic attack
page 11 of 16

Synthesis

Review

leading to 67. Final protonation yields the cyclic carbonate
product 68. Both the photolysis and carboxylative cyclization can
be combined to make this reaction an efficient one-pot process.
The scope of the one-pot reaction was investigated by changing
the substitution on the aryl and amine group (Scheme 22; below).
First the influence of the aryl substitution was tested, leading to
moderate to high yields (compounds 70a-c). Also different
substituted tosyl groups were investigated, which all afforded the
cyclic carbonate products in good yields (compounds 70d-f).
Overall this method provides an efficient and simple way to
synthesize amino-substituted cyclic carbonates though still with
a limited functional group diversity.

some cases) may speed up the reaction without compromising
the chemoselectivity.

Scheme 23: Exemplary conversions of hydroxy- and amino-functionalized
oxetanes into cyclic carbonates and carbamates.

8. Conversion of other substrates
Various other types of scaffolds can be considered for
substrate-assisted activation of carbon dioxide including diols
(see introduction) or halohydrins. The three-component
coupling of an amine/alcohol with CO2 and a halide can be used
to obtain linear carbamates86 and carbonates,87 respectively.
Combination of a diol and a dihalide in the presence of CO2 leads
to polycarbonates.88 Even cyclic carbonates can be formed
through this approach when the alcohol and halide
functionalities are both present in the same substrate.89 For
instance, the group of Zhang reported the formation of both 5and 6-membered cyclic carbonates from 1,2- and 1,3-haloalcohols catalyzed by cesium carbonate.89b

Scheme 22: Proposed mechanism for the coupling of an azetidinol with CO2
top) and the investigated substrate scope (below).

The reaction of CO2 with hydroxy- or amino-oxetanes was
recently described by the group of Kleij. A series of oxetanes
having a hydroxy or amine substituent in the 3-position were
smoothly converted to cyclic carbonates and carbamates
respectively. (Scheme 23).85 Cyclic carbonates derived from 3hydroxy-oxetanes were obtained under very mild conditions
with temperatures as low as 50ºC depending on the substitution
(cf., 71 and 74). In a similar fashion, 3-amino-oxetanes react with
CO2 to give the 5- and 6-membered cyclic carbamate products
(products 72 and 73). Direct CO2 insertion into the oxetane to
give the 6-membered cyclic carbonate was not observed in
neither of these cases. The latter type of reaction typically
requires higher onset temperatures of >60ºC, while the
substrates displayed in Scheme 23 react readily at temperatures
between 25‒50ºC. The proposed mechanism of the reaction
follows the manifold reported in Scheme 22. Carbon dioxide is
activated by the pendant alcohol group of the oxetane, to
generate an internal nucleophile being a linear carbonate or
carbamate anion. Intramolecular nucleophilic attack enables
ring-opening of the oxetane to give the cyclic carbonate or
carbamate product. In general, this reaction does not require an
external nucleophile such as TBAB, but it was found that this (in
Template for SYNTHESIS © Thieme Stuttgart · New York

2019-05-08

Scheme 24: Proposed mechanism for the Cr(salen) catalyzed formation of 4and 5-substituted oxazolidinones.

Aziridines also allow for activation of CO2 leading to the
formation of oxazolidinones. Synthesis of the latter via
aziridine/CO2 coupling was discovered in 1976 by the group of
Ikeda, who reported an iodine mediated synthesis of
oxazolidinones.90 Subsequent reports on this topic include
catalytic systems based on ionic liquids,91 iron,92 chromium93 and
organocatalysts.94 Pinhas et al. showed that the reaction can even
be performed in the absence of solvent and catalyst.95 An
interesting contribution to this area was presented by Baik,
Nguyen and coworkers who reported on the synthesis of 5substituted oxazolidinones from aziridines.93 Normally the
formation of the 4-substituted product is favored, as ringopening occurs on the least substituted position of the aziridine
ring.94a,95 In this specific case, however, the regio-selectivity is not
dominated by sterics but by electronic effects. Scheme 24
displays the proposed mechanism leading to the formation of
both the 4- and 5-substituted isomers, respectively.
page 12 of 16

Synthesis

Review

Instead of initial ring-opening of the aziridine moiety by the
Cr(salen) complex, the metal here serves to activate CO2.
Nucleophilic attack of the aziridine (75) on CO2 leads to the
formation of a carbamate intermediate (76). The newly formed
carbamate anion acts as an internal nucleophile that can open the
aziridine via two possible pathways. Attack on the substituted
carbon (path a) leads to the major (5-substituted) oxazolidinone
77 and path b gives rise to formation of the minor (4-substituted)
product 78. It was suggested that the reaction proceeds through
elongation of one of the C‒N bonds in the aziridine prior to
intramolecular nucleophilic ring-opening of the aziridine. The
more substituted carbon center bearing the phenyl group has
more cationic character leading to elongation of the
corresponding C‒N bond. As a result, the ring-opening
preferentially occurs at the more substituted carbon as was
supported by DFT calculations.

programs and CO2 based polymers. This review clearly
demonstrates the significant progress that has been made in the
area of substrate-controlled CO2 conversion in terms of catalyst
activity and development of alternative chemo-selectivity
allowing for CO2 conversions at lower temperature and pressure,
and diversification of the products that can be derived from this
carbon renewable. However, there is still room for improvement
for many of the discussed substrate classes, especially in terms of
product scope and functionality. In many cases there is a
requirement for the presence of a specific functional group
within the substrate, thereby limiting its applicability in organic
transformations. Most preferably, simple organic scaffolds that
incorporate alcohol or amine functions will be preferred to lift
the potential of substrate-controlled CO2 conversion, and the
ubiquitous presence of such groups in basic chemicals provides a
hopeful starting point. Another important aspect is the
development of suitable organocatalysts within this field of
chemistry. While organocatalysts have already been successfully
implemented in several (simple) CO2 transformations,10c other
sub-areas in CO2 conversion are still mainly dominated by metal
based systems. In addition, the range of products that can be
obtained through CO2 activation by the substrate needs to be
expanded to mature this conceptual approach. The synthesis of
quinazolines, iso-benzofurans and ureas are nice examples of
further product diversification that can be achieved. The
development of new catalysts and methodologies will certainly
add further exciting examples to this rapidly developing area of
CO2 valorization.

Scheme 25: Substrate-assisted activation of CO2 with 2-aminobenzonitriles
(I), o-alkynylanilines (II) or allenyl methylamines (III) leading to various
heterocyclic products.

Acknowledgment

In addition, a Hammet-correlation study was performed to
investigate the effect of electronic substitution at the phenyl ring.
Electron-donating substituents were found to give higher regioselectivity in line with their increased ability to stabilize the
carbocation intermediate. When an aryl group was used having a
para-methoxy substituent, the selectivity increased to 80:1
(versus 40:1 for simple phenyl) in favor of the 5-substituted
oxazolidinone. Other possible N-containing substrates include
(aromatic) amines with additional nitrile, alkyne or allene
functionalities that may give rise to various types of cyclization
products after reaction with CO2 (Scheme 25).96-98

9. Summary and outlook
The substrate-assisted activation of CO2 has become a
powerful strategy to convert a waste molecule into value-added
organic matter. Various scaffolds have been shown to produce
heterocyclic structures with interesting (post)synthetic
potential. Typical scaffolds that provide new reactivity patterns
include propargylic compounds, (homo)allylic compounds,
epoxy alcohols/amines and other cyclic ethers such as oxetanes:
it can be anticipated that new combinations of CO2 activating
groups and functional groups that assist cyclization or any other
type of irreversible transformation will be developed in the near
future providing new synthons for pharmaceutical development
Template for SYNTHESIS © Thieme Stuttgart · New York

2019-05-08

We thank ICIQ, ICREA, and the Spanish Ministerio de Economía y
Competitividad (MINECO) through project CTQ2014-60419-R, and
further support through Severo Ochoa Excellence Accreditation 2014–
2018 (SEV-2013-0319). JR thanks ICIQ for a predoctoral fellowship.

References
(1) Friedlingstein, P.; Andrew, R. M.; Rogelj, J.; Peters, G. P.; Canadell, J.
G.; Knutti, R.; Luderer, G.; Raupach, M. R.; Schaeffer, M; van Vuuren,
D. P.; Le Quere, C. Nat. Geosci. 2014, 7, 709.
(2) (a) Lim, X. Nature 2015, 520, 628. (b) van den Assen, N.; Müller, L.
J.; Steingrube, A.; Voll, P.; Bardow, A. Environ. Sci. Technol. 2016, 50,
1093. (c) Liu, Q.; Wu, L.; Jackstell, R.; Beller, M. Nat. Commun. 2015,
6, 5933.
(3) Aresta, M.; Dibenedetto, A.; Angelini, A. Chem. Rev. 2014, 114, 1709.
(4) Aresta, M.; Dibenedetto, A.; Angelini, A. J. CO2 Utiliz. 2013, 3–4, 65.
(5) Omae, I. Coord. Chem. Rev. 2012, 256, 1384.
(6) Biochemistry, 5th ed., eds. Berg, J.M., Tymoczko, J. L. and Stryer, L.,
W. H. Freeman, NY 2002.
(7) Kelsey, R. A.; Miller, D. A.; Parkin, S. R.; Liu, K.; Remias, J. E.; Yang,
Y.; Lightstone, F. C.; Liu, K.; Lippert, C. A.; Odom, S. A. Dalton Trans.
2016, 45, 324.
(8) (a) Zhang, X.; van Eldik, R.; Koike, T.; Kimura, E. Inorg. Chem. 1993,
32, 5749. (b) Zhang, X.; van Eldik, R. Inorg. Chem. 1995, 34, 5606.
(9) (a) Moller, F.; Merz, K.; Herrmann, C.; Apfel, U. P. Dalton Trans.
2016, 45, 904. (b) Bandeira, N. A. G.; Garai, S.; Müller, A.; Bo, C.
Chem. Commun. 2015, 51, 15596.
(10) (a) Werner, T.; Tenhumberg, N. J. CO2 Utiliz. 2014, 7, 39. (b) Alves,
M.; Grignard, B.; Gennen, S.; Mereau, R.; Detrembleur, C., Jerome, C.;
Tassaing, T. Catal. Sci. Technol. 2015, 5, 4636. (c) Fiorani, G.; Guo,
W.; Kleij, A. W. Green Chem. 2015, 17, 1375.

page 13 of 16

Synthesis

Review

(11) West, K. N.; Wheeler, C.; McCarney, J. P.; Griffith, K. N.; Bush, D.;
Liotta, C. L.; Eckert, C. A., J. Phys. Chem. A 2001, 105, 3947.
(12) Gassensmith, J. J.; Furukawa, H.; Smaldone, R. A.; Forgan, R. S.;
Botros, Y. Y.; Yaghi, O. M.; Stoddart, J. F. J. Am. Chem. Soc. 2011, 133,
15312.
(13) Martín, C.; Fiorani, G.; Kleij, A. W. ACS Catal. 2015, 5, 1353.
(14) (a) Hara, Y.; Onodera, S.; Kochi, T.; Kakiuchi, K. Org. Lett. 2015, 17,
4850. (b) Toullec, P.; Carbayo Martin, A.; Gio-Batta, M.; Bruneau, C.;
Dixneuf, P. H. Tetrahedron Lett. 2000, 41, 5527.
(15) (a) Gendre, P. L.; Thominot, P.; Bruneau, C.; Dixneuf, P. H. J. Org.
Chem. 1998, 63, 1806. (b) Ohe, K.; Matsuda, H.; Morimoto, T.;
Ogoshi, S.; Chatani, N.; Murai, S. J. Am. Chem. Soc. 1994, 116, 4125.
(16) DE1098953, Dimroth, P.; Pasedach, H. 1961.
(17) Laas, H.; Nissen, A.; Nürrenbach, A. Synthesis 1981, 958.
(18) Iritani, K.; Yanagihara, N.; Utimoto, K. J. Org. Chem. 1986, 51, 5499.
(19) Inoue, Y.; Ishikawa, J.; Taniguchi, M.; Hashimoto, H. Bull. Chem. Soc.
Jpn. 1987, 60, 1204.
(20) Sasaki, Y. Tetrahedron Lett. 1986, 27, 1573.
(21) (a) Kim, H.-S.; Kim, J.-W.; Kwon, S.-C.; Shim, S.-C.; Kim, T.-J. J.
Organomet. Chem. 1997, 545–546, 337. (b) Jiang, H.-F.; Wang, A. Z.;
Liu, H.-L.; Qi, C.-R. Eur. J. Org. Chem. 2008, 2309. (c) Gu, Y.; Shi, F.;
Deng, Y. J. Org. Chem. 2004, 69, 391.
(22) Ouyang, L.; Tang, X.; He, H.; Qi, C.; Xiong, W.; Ren, Y.; Jiang, H. Adv.
Synth. Catal. 2015, 357, 2556.
(23) Minakata, S.; Sasaki, I.; Ide, T. Angew. Chem. Int. Ed. 2010, 49, 1309.
(24) Yamada, W.; Sugawara, Y.; Cheng, H. M.; Ikeno, T.; Yamada, T. Eur.
J. Org. Chem. 2007, 2604.
(25) Yoshida, S.; Fukui, K.; Kikuchi, S.; Yamada, T. J. Am. Chem. Soc. 2010,
132, 4072.
(26) (a) Tang, X.; Qi, C.; He, H.; Jiang, H.; Ren, Y.; Yuan, G. Adv. Synth. Catal.
2013, 355, 2019. (b) Cui, M.; Qian, Q.; He, Z.; Ma, J.; Kang, X.; Hu, J.;
Liu, Z.; Han, B. Chem. Eur. J. 2015, 21, 15924.
(27) (a) Yang, Z.-Z.; Zhao, Y.; Zhang, H.; Yu, B.; Ma, Z.; Ji, G.; Liu, Z. Chem.
Commun. 2014, 50, 13910. (b) Yang, Z.; Yu, B.; Zhang, H.; Zhao, Y.;
Chen, Y.; Ma, Z.; Ji, G.; Gao, X.; Han, B.; Liu, Z. ACS Catal. 2016, 6,
1268.
(28) (a) Song, Q.-W.; Chen, W.-Q.; Ma, R.; Yu, A.; Li, Q.-Y.; Chang, Y.; He,
L.-N. ChemSusChem 2015, 8, 821. (b) Song, Q.-W.; He, L.-N. Adv.
Synth. Catal. 2016, 358, 1251. (c) Song, Q.-W.; Yu, B.; Li, X.-D.; Ma,
R.; Diao, Z.-F.; Li, R.-G.; Li, W.; He, L.-N. Green Chem. 2014, 16, 1633.
(29) Kimura, T.; Kamata, K.; Mizuno, N. Angew. Chem. Int. Ed. 2012, 51,
6700.
(30) Hu, J.; Ma, J.; Zhu, Q.; Qian, Q.; Han, H.; Mei, Q.; Han, B. Green Chem.
2016, 18, 382.
(31) (a) Dixneuf, P. H. Catal. Lett. 2014, 145, 360. (b) Qing-Wen, S.; He,
L.-N. in Advances in CO2 Capture, Sequestration, and Conversion,
American Chemical Society, 2015, 1194, 47. (c) S. Kikuchi, T.
Yamada, The Chemical Record 2014, 14, 62.
(32) Fournier, J.; Bruneau, C.; Dixneuf, P. H. Tetrahedron Lett. 1989, 30,
3981.
(33) (a) Joumier J. M.; Fournier, J.; Bruneau, C.; Dixneuf, J. Chem. Soc.,
Perkin Trans. 1 1991, 3271. (b) Bruneau, C.; Dixneuf, P. H. J. Mol.
Catal. 1992, 74, 97.
(34) Kayaki, Y.; Yamamoto, M.; Ikariya, T. J. Org. Chem. 2007, 72, 647.
(35) Jung, M. E.; Gervay, J. J. Am. Chem. Soc. 1991, 113, 224.
(36) Qi, C.; Huang, L.; Jiang, H. Synthesis 2010, 1433.
(37) Uemura, K.; Kawaguchi, T.; Takayama, H.; Nakamura, A.; Inoue, Y. J.
Mol. Catal. A: Chem. 1999, 139, 1.
(38) Ca, N. D.; Gabriele, B.; Ruffolo, G.; Veltri, L.; Zanetta, T.; Costa, M. Adv.
Synth. Catal. 2011, 353, 133.
(39) Wang, Y.-B.; Sun, D.-S.; Zhou, H.; Zhang, W.-Z.; Lu, X.-B. Green Chem.
2014, 16, 2266.
(40) (a) Kayaki, Y.; Yamamoto, M.; Ikariya, T. Angew. Chem. Int. Ed.
2009, 48, 4194. (b) Tommasi, I.; Sorrentino, F. Tetrahedron Lett.
2009, 50, 104.
(41) Wang, Y.-B.; Wang, Y.-M.; Zhang, W.-Z.; Lu, X.-B. J. Am. Chem. Soc.
2013, 135, 11996.

Template for SYNTHESIS © Thieme Stuttgart · New York

2019-05-08

(42) (a) Heravi, M. M.; Zadsirjan, V. Tetrahedron: Asymmetry 2013, 24,
1149. (b) Evans, D. A.; Ennis, M. D.; Le, T.; Mandel, N.; Mandel, G. J.
Am. Chem. Soc. 1984, 106, 1154.
(43) Sarkar, A.; Bhattacharyya, S.; Dey, S. K.; Karmakar, S.; Mukherjee, A.
New J. Chem. 2014, 38, 817.
(44) Giovanni, Z.; Pilar, M.; Giuliano Delle, M.; Domenico, M.; Laura, N.;
Bruno, B. Mini-Reviews in Med. Chem. 2007, 7, 389.
(45) Aurelio, L.; Brownlee, R. T. C.; Hughes, A. B. Chem. Rev. 2004, 104,
5823.
(46) Kim, T.-J.; Kwon, K.-H.; Kwon, S.-C.; Baeg, J.-O.; Shim, S.-C.; Lee, D.H. J. Organomet. Chem. 1990, 389, 205.
(47) (a) Mahé, R.; Dixneuf, P. H.; Lécolier, S. Tetrahedron Lett. 1986, 27,
6333. (b) Sasaki, Y.; Dixneuf, P. H. J. Chem. Soc., Chem. Commun.
1986, 790. (c) Sasaki, Y.; Dixneuf, P. H. J. Org. Chem. 1987, 52, 314.
(48) Costa, M.; Chiusoli, G. P.; Rizzardi, M. Chem. Commun. 1996, 1699.
(49) Yoshida, S.; Fukui, K.; Kikuchi, S.; Yamada, T. Chem. Lett. 2009, 38,
786.
(50) Hase, S.; Kayaki, Y.; Ikariya, T. Organometallics 2013, 32, 5285.
(51) (a) Shi, M.; Shen, Y.-M. J. Org. Chem. 2002, 67, 16. (b) Bacchi, A.;
Chiusoli, G. P.; Costa, M.; Gabriele, B.; Righi, C.; Salerno, G. Chem.
Commun. 1997, 1209.
(52) Costa, M.; Chiusoli, G. P.; Taffurelli, D.; Dalmonego, G. J. Chem. Soc.,
Perkin Trans. 1 1998, 1541.
(53) Hu, J.; Ma, J.; Zhang, Z.; Zhu, Q.; Zhou, H.; Lu, W.; Han, B. Green Chem.
2015, 17, 1219.
(54) Kayaki, Y.; Yamamoto, M.; Suzuki, T.; Ikariya, T. Green Chem. 2006,
8, 1019.
(55) Garcia-Dominguez, P.; Fehr, L.; Rusconi, G.; Nevado, C. Chem. Sci.
2016, 7, 3914.
(56) Deng, Z.; Wei, J.; Liao, L.; Huang, H.; Zhao, X. Org. Lett. 2015, 17,
1834.
(57) G. Cardillo, M. Orena, G. Porzi, S. Sandri, J. Chem. Soc., Chem.
Commun. 1981, 465.
(58) Wu, X.; Gao, Q.; Lian, M.; Liu, S.; Wu, A. RSC Adv. 2014, 4, 51180.
(59) Tanner, D. D.; Gidley, G. C.; Das, N.; Rowe, J. E.; Potter, A. J. Am. Chem.
Soc. 1984, 106, 5261.
(60) Glover, S. A.; Goosen, A. Tetrahedron Lett. 1980, 21, 2005.
(61) Minakata, S. Acc. Chem. Res. 2009, 42, 1172.
(62) Usami, K.; Nagasawa, Y.; Yamaguchi, E.; Tada, N.; Itoh, A. Org. Lett.
2016, 18, 8.
(63) Tang, S.; Liu, K.; Long, Y.; Gao, X.; Gao, M.; Lei, A. Org. Lett. 2015, 17,
2404.
(64) Vara, B. A.; Struble, T. J.; Wang, W.; Dobish, M. C.; Johnston, J. N. J.
Am. Chem. Soc. 2015, 137, 7302.
(65) Jensen, A. E.; Knochel, P. J. Org. Chem. 2002, 67, 79.
(66) (a) Yoshida, Y.; Inoue, S. Chem. Lett. 1977, 6, 1375. (b) McGhee, W.
D.; Riley, D. P. Organometallics 1992, 11, 900.
(67) Cardillo, G.; Orena, M.; Sandri, S. J. Org. Chem. 1986, 51, 713.
(68) Toda, T.; Kitagawa, Y. Angew. Chem. Int. Ed. Engl. 1987, 26, 334.
(69) (a) Fernández, I.; Muñoz, L. Tetrahedron: Asymm. 2006, 17, 2548.
(b) García‐Egido, E.; Fernández, I.; Muñoz, L. Synth. Commun. 2006,
36, 3029.
(70) Takeda, Y.; Okumura, S.; Tone, S.; Sasaki, I.; Minakata, S. Org. Lett.
2012, 14, 4874.
(71) Shu, J.; He, L.; Ding, H.; Wang, L.; Guo, H.; Gao, Y.; Dzakah, E. E.; Zeng,
Z. Anal. Meth. 2014, 6, 2306.
(72) Barbosa, J.; Freitas, A.; Moura, S.; Mourão, J. L.; Noronha da Silveira,
M. I.; Ramos, F. J. Agric. Food. Chem. 2011, 59, 11927.
(73) Yoshida, M.; Ohsawa, Y.; Sugimoto, K.; Tokuyama, H.; Ihara, M.
Tetrahedron Lett. 2007, 48, 8678.
(74) Soldi, L.; Massera, C.; Costa, M.; Della Ca’, N. Tetrahedron Lett. 2014,
55, 1379.
(75) (a) Decortes, A.; Castilla, A. M.; Kleij, A. W. Angew. Chem. Int. Ed.
2010, 49, 9822. (b) Kathalikkattil, A. C.; Babu, R.; Tharun, J.;
Roshan, R.; Park, D.-W. Catal. Surv. Asia 2015, 19, 223. (c) Cokoja,
M.; Wilhelm, M. E.; Anthofer, M. H.; Herrmann, W. A.; Kühn, F. E.
ChemSusChem 2015, 8, 2436. (d) Comerford, J. W.; Ingram, I. D. V.;
North, M.; Wu, X. Green Chem. 2015, 17, 1966.

page 14 of 16

Synthesis

Review

(76) Rintjema, J.; Epping, R.; Fiorani, G.; Martín, E; Escudero-Adán, E. C.;
Kleij, A. W. Angew. Chem. Int. Ed. 2016, 55, 3972.
(77) Kojima, F.; Aida, T.; Inoue, S. J. Am. Chem. Soc. 1986, 108, 391.
(78) Shang, J.; Guo, X.; Li, Z.; Deng, Y. Green Chem. 2016, 18, 3082.
(79) McCombie, S. W.; Metz, W. A. Tetrahedron Lett. 1987, 28, 383.
(80) (a) Myers, A. G.; Widdowson, K. L. Tetrahedron Lett. 1988, 29, 6389.
(b) Myers, A. G.; Proteau, P. J.; Handel, T. M. J. Am. Chem. Soc. 1988,
110, 7212.
(81) Inkster, J. A. H.; Ling, I.; Honson, N. S.; Jacquet, L.; Gries, R.; Plettner,
E. Tetrahedron: Asymm. 2005, 16, 3773.
(82) Ishii, S.; Zhou, M.; Yoshida, Y.; Noguchi, H. Synth. Commun. 1999,
29, 3207.
(83) Ishida, N.; Shimamoto, Y.; Murakami, M. Angew. Chem. Int. Ed.
2012, 51, 11750.
(84) Gold, E. H. J. Am. Chem. Soc. 1971, 93, 2793.
(85) Rintjema, J.; Guo, W.; Martin, E.; Escudero-Adán, E. C.; Kleij, A. W.
Chem. Eur. J. 2015, 21, 10754.
(86) Yoshida, Y.; Ishii, S.; Yamashita, T. Chem. Lett. 1984, 13, 1571.
(87) (a) Salvatore, R. N.; Chu, F.; Nagle, A. S.; Kapxhiu, E. A.; Cross, R. M.;
Jung, K. W. Tetrahedron 2002, 58, 3329. (b) Shi, M.; Shen, Y.-M.
Molecules 2002, 7, 386.
(88) Oi, S.; Nemoto, K.; Matsuno, S.; Inoue, Y. Macromol. Rapid Commun.
1994, 15, 133.

Template for SYNTHESIS © Thieme Stuttgart · New York

2019-05-08

(89) (a) Yan, P.; Tan, X.; Jing, H.; Duan, S.; Wang, X.; Liu, Z. J. Org. Chem.
2011, 76, 2459. (b) Reithofer, M. R.; Sum, Y. N.; Zhang, Y. Green
Chem. 2013, 15, 2086.
(90) Soga, K.; Hosoda, S.; Nakamura, H.; Ikeda, S. J. Chem. Soc., Chem.
Commun. 1976, 617.
(91) Kawanami, H.; Matsumoto, H.; Ikushima, Y. Chem. Lett. 2005, 34,
60.
(92) Gao, J.; Song, Q.-W.; He, L.-N; Liu, C.; Yang, Z.-Z.; Han, X.; Li, X.-D.;
Song, Q.-C. Tetrahedron 2012, 68, 3835.
(93) Adhikari, D.; Miller, A. W.; Baik, M.-H.; Nguyen, S. T. Chem. Sci. 2015,
6, 1293.
(94) (a) Kathalikkattil, A. C.; Tharun, J.; Roshan, R.; Soek, H.-G.; Park, D.W. Appl. Catal. A: Gen. 2012, 447–448, 107. (b) Takeda, Y.; Kawai,
H.; Minakata, S. Chem. Eur. J. 2013, 19, 13479. (c) Wu, Y.; Liu, G.;
Tetrahedron Lett. 2011, 52, 6450. (d) Jiang, H.-F.; Ye, J.-W.; Qi, C.-R.;
Huang, L.-B. Tetrahedron Lett. 2010, 51, 928.
(95) C. Phung, R. M. Ulrich, M. Ibrahim, N. T. G. Tighe, D. L. Lieberman, A.
R. Pinhas, Green Chemistry 2011, 13, 3224-3229.
(96) Nale, D. B.; Saigaonkar, S. D.; Bhanage, B. M. J. CO2 Utiliz. 2014, 8,
67.
(97) Ishida, T.; Kikuchi, S.; Tsubo, T.; Yamada, T. Org. Lett. 2013, 15, 848.
(98) Yamashita, K.; Hase, S.; Kayaki, Y.; Ikariya, T. Org. Lett. 2015, 17,
2334.

page 15 of 16

Synthesis

Review

Biosketches
Arjan W. Kleij received his MSc (1996) and PhD (2000) from the University of Utrecht (The
Netherlands) under the supervision of Professor Gerard van Koten working on the application of
metallodendrimers in homogeneous catalysis. In the next six years, he held various positions
including two in industry at Avantium Technologies as a project leader in its pharma division, and at
Hexion Specialty Chemicals as a R&D scientist concerned with process chemistry in the area of epoxy
resins and precursors. He also held two postdoctoral positions at the UAM (Madrid, 2002) working
with Professor Javier de Mendoza in the area of calixarene chemistry, and with Professor Joost Reek
at the University of Amsterdam (20032005) focusing on supramolecular catalysis. In 2006, he took
a position as Group Leader at the Institute of Chemical Research of Catalonia (ICIQ) and ICREA
fellow, and was promoted to ICREA professor in 2011. His research focuses on the use of carbon
dioxide/functional carbonates as reagents in organic and new (bio)polymer synthesis using metaland organo-catalyzed strategies.
Jeroen Rintjema studied chemistry at the University of Amsterdam (The Netherlands) where he
received his BSc (2009) and his MSc (2011). During this time, he developed new supramolecular gold
chemistry under the supervision of Prof. Joost Reek. Since 2013, he joined the group of Arjan Kleij in
ICIQ as a PhD student working on carbon dioxide conversion catalysis. His research specifically focuses
on the development of new synthetic methods utilizing substrate-mediated CO2 activation strategies.

Template for SYNTHESIS © Thieme Stuttgart · New York

2019-05-08

page 16 of 16