A hafnium-based metal-organic framework as an efficient and multifunctional catalyst for facile CO2 fixation and regioselective and enantioretentive epoxide activation

Article abstract of DOI:10.1021/ja508626n

Porous heterogeneous catalysts play a pivotal role in the chemical industry. Herein a new Hf-based metal-organic framework (Hf-NU-1000) incorporating Hf₆ clusters is reported. It demonstrates high catalytic efficiency for the activation of epoxides, facilitating the quantitative chemical fixation of CO₂ into five-membered cyclic carbonates under ambient conditions, rendering this material an excellent catalyst. As a multifunctional catalyst, Hf-NU-1000 is also efficient for other epoxide activations, leading to the regioselective and enantioretentive formation of 1,2-bifuctionalized systems via solvolytic nucleophilic ring opening.

Full text of DOI:10.1021/ja508626n

Communication
pubs.acs.org/JACS
A Hafnium-Based Metal−Organic Framework as an Efficient and
Multifunctional Catalyst for Facile CO2 Fixation and Regioselective
and Enantioretentive Epoxide Activation
M. Hassan Beyzavi,† Rachel C. Klet,† Samat Tussupbayev,‡ Joshua Borycz,‡ Nicolaas A. Vermeulen,†
Christopher J. Cramer,‡ J. Fraser Stoddart,† Joseph T. Hupp,*,† and Omar K. Farha*,†,§
†International Institute for Nanotechnology (IIN) and Center for the Chemistry of Integrated Systems (CCIS), Department of
Chemistry, Northwestern University, 2145 Sheridan Road, Evanston, Illinois 60208-3113, United States
‡Department of Chemistry, Chemical Theory Center, and Supercomputing Institute, University of Minnesota, 207 Pleasant Street SE,
Minneapolis, Minnesota 55455-0431, United States
§Department of Chemistry, Faculty of Science, King Abdulaziz University, Jeddah 22254, Saudi Arabia
S
* Supporting Information
based on an acid catalyst that activates the epoxide, which can
ABSTRACT: Porous heterogeneous catalysts play a
pivotal role in the chemical industry. Herein a new Hf-
based metal−organic framework (Hf-NU-1000) incorpo-
rating Hf6 clusters is reported. It demonstrates high
catalytic efficiency for the activation of epoxides,
facilitating the quantitative chemical fixation of CO2 into
five-membered cyclic carbonates under ambient condi-
tions, rendering this material an excellent catalyst. As a
multifunctional catalyst, Hf-NU-1000 is also efficient for
other epoxide activations, leading to the regioselective and
enantioretentive formation of 1,2-bifuctionalized systems
via solvolytic nucleophilic ring opening.
then be attacked by a nucleophile cocatalyst to form an alkoxide.
This intermediate can then react with carbon dioxide to
ultimately give the cyclic carbonate.6 However, on account of
the relatively inert nature and low reactivity of CO2, its activation
and incorporation into organic substrates still remains a
formidable challenge.7
Although some homogeneous catalysts8 and several types of
heterogeneous catalysts,9 such as zeolites,10 silica-supported
salts,11 metal oxides,12 titanosilicate,13 a microporous polymer,14
and an organic network,15 have been utilized for the synthesis of
cyclic carbonates, most of the processes demand high pressures
and temperatures, thus requiring high energy and capital costs.
Recently,16 we reported a new Zr-based MOF, NU-1000,
obtained via the solvothermal reaction of ZrCl4 and 1,3,6,8-
tetrakis(p-benzoic acid)pyrene (H4TBAPy) with benzoic acid as
a modulator. The parent-framework node consists of an
octahedral Zr6 cluster capped by eight bridging oxygen-
containing ligands. The 3D NU-1000 structure can be described
as 2D Kagome sheets linked by TBAPy ligands. NU-1000 shares
the same topological features as MOF-54517 and PCN-222,18
comprising triangular and hexagonal channels. The presence of a
high density of acidic sites, the integration of permanent porosity
with ultralarge channels, the high surface area, and the
extraordinary chemical stability of NU-1000 suggested that it
could be an excellent catalyst platform that can overcome the
above-mentioned challenges in CO2 insertion and acid-catalyzed
epoxide activation in general.
atalysts based on permanently porous frameworks have the
C
potential to unify the best features of homogeneous
catalysts (selectivity and ease of modification) and heteroge-
neous catalysts (ease of purification and recyclability). Metal−
organic frameworks (MOFs) represent a special class of porous
materials consisting of easily constructed organic linkers and
metal clusters.1 The modular nature and facile tunability of
MOFs make them ideal heterogeneous catalysts with uniform
active sites through wisely chosen building blocks. In catalysis,
MOFs allow for the same level of structural refinement as
molecular homogeneous catalysts, while their ultrahigh surface
area, pore volume, and heterogeneous nature facilitate good
catalytic activity and rapid purification.2
Catalyst-mediated reactions of CO2 represent one potential
positive contributor to climate-relevant carbon capture and
storage/sequestration (CCS), albeit a far from sufficient one to
satisfy this enormous challenge.3 Well-designed reactions that
utilize waste CO2 in the production of commercially relevant
chemicals are much sought these days. Some of these reactions
include the formation of carbonates, where the carbonyl carbon
obtained from CO2 is isohypsic with its starting material and does
not require reagent-driven changes in oxidation state.4 The acid-
catalyzed cycloaddition of CO2 with an epoxide to form a cyclic
carbonate, a functionality having various important applications,5
is a highly atom-economical reaction that might be catalyzed by
an appropriately designed MOF. Mechanistically, this reaction is
By comparing the dissociation enthalpies of typical Hf−O
versus Zr−O bonds (802 vs 776 kJ mol−1),19 we inferred that Hf
is more oxophilic (i.e., has stronger M−O bonds) than Zr and
should thus function as a stronger Brønsted acid. Therefore, we
hypothesized that a Hf-cluster-based NU-1000 would be an even
more efficient catalyst than the Zr-cluster-based NU-1000 for
acid-catalyzed systems [also see the Supporting Information
(SI)]. In order to test our hypothesis, we prepared the Hf-based
MOF (Hf-NU-1000) with the same topology as NU-1000.20 To
Received: August 25, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja508626n | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
catalyst, no product was obtained (entry 1c). The same reaction
proceeds much faster at elevated temperature (55 °C, 13 h; entry
1d). Hf-NU-1000 was compared with the reported MOFs9d gea-
MOF-1,9c Cr-MIL-101, MOF-5, ZIF-8, Ni(salphen)-MOF, Co-
MOF-74, Mg-MOF-74, [Cu(Hip)2(Bpy)]n [CHB(M)], ZIF-68,
F-IRMOF-3, and MIL68(In)-NH2 (Table S2, entries 1e−o),
which have also been used for the preparation of styrene
carbonate, and Hf-NU-1000 clearly stands out in terms of yield
and milder conditions.
The industrially important epoxide divinylbenzene dioxide
(DVBDO) was also examined, and biscarbonated DVBDO,
which could be a useful epoxy resin monomer24 candidate, was
obtained quantitatively after 19 h at 55 °C (Table 1, entry 2). Hf-
NU-1000 was also compared with other MOF materials for the
conversion of propylene oxide (PO) to propylene carbonate (r.t.,
26 h; entry 3). Hf-NU-1000 again showed shorter reaction time
and higher product yield compared with the previously reported
MOFs9d MMCF-2,9a PCN-224(Co),9b gea-MOF-1,9c HKUST-
1, MOF-505, MMPF-9, Cr-MIL-101, MOF-5, Ni(salphen)-
MOF, CHB(M), and Zn4O(BDC)x(ABDC)3−x based on MOF-
5 (MIXMOF) (Table S2, entries 3b−l). It is worth noting that
PO is more studied than styrene oxide because of its higher
reactivity and ease of isolation. However, its low boiling point
could easily lead to mass loss, which can complicate yield
calculations; thus, careful handling is required.
To confirm the heterogeneous nature of the reaction, under
the same conditions as in Table 1, entry 1a, 40 h after the outset
of the reaction, the catalyst was removed and the reaction was
allowed to continue. As expected, no increase in the formation of
carbonate was detected (see the SI). At the end of the reaction,
inductively coupled plasma (ICP) analysis of the reaction
mixture filtrate revealed no Hf leaching, indicating that the
catalytic reaction is indeed heterogeneous in nature. Addition-
ally, in the absence of the ammonium salt cocatalyst, no
conversion was detected for the model reaction. Furthermore,
the catalyst was reused five successive times without a significant
decrease in the efficiency of the catalyst or structural
deterioration as determined by PXRD analysis (see the SI).
After such encouraging results for the cycloaddition reaction of
CO2 with epoxides were obtained, and since mechanistically this
reaction occurs via an acid-catalyzed ring-opening step,6,9a we
became interested in investigating the performance of Hf-NU-
1000 in the activation of epoxides for the preparation of 1,2-
bifuctionalized systems via acid-catalyzed nucleophilic ring
Figure 1. Relevant structural features and representations of Hf-NU-
1000. For simplicity, H atoms except for the cluster are not shown. For
clarity, the carboxylates have been removed from the cluster shown in
the top inset.
confirm that NU-1000 and Hf-NU-1000 have the same overall
crystal structure, periodic density functional theory (DFT)
within the Vienna ab Initio Simulation Package (VASP)21 was
used to optimize the ionic positions of NU-1000 starting from
the validated NU-1000 X-ray diffraction (XRD) data. The Zr4+
ions were then replaced with Hf4+, and the simulated Hf-NU-
1000 structure was used to optimize the ionic positions.
Comparison of the experimental and simulated powder XRD
(PXRD) patterns of Hf-NU-1000 and NU-1000 confirmed that
the Hf and Zr versions of NU-1000 are indeed isostructural
(Figure 1; also see the SI).
The porosity of Hf-NU-1000 was studied by N2 adsorption−
desorption experiments at 77 K, and the resulting isotherm (type
IVc) yielded a Brunauer−Emmett−Teller (BET) surface area of
1780 m2 g−1 and a total pore volume of 1.14 cm3 g−1. DFT pore
size distribution analysis revealed pore diameters of ca. 13 and 29
Å, assignable to the triangular micropores and hexagonal
mesopores, respectively. Thermogravimetric analysis (TGA) of
the activated sample showed no major decomposition up to 500
°C (see the SI). Diffuse reflectance infrared Fourier transform
spectroscopy (DRIFTS) confirmed the presence of −OH groups
and H2O molecules in the Hf6 nodes, with peaks at 3679 cm−1
assigned to the non-H-bonded −OH and bridging −OH
stretches, the peak at 3678 cm−1 assigned to the non-H-bonded
H2O stretch, and the peak at 2752 cm−1 assigned to the H-
bonded H2O and OH stretches (see the SI).22
Given the high stability, porosity, and large channels of Hf-
NU-1000, we decided to investigate its performance as an acid
catalyst in the context of CO2 fixation through reaction with
epoxides to form cyclic carbonates under ambient conditions.
Hf-NU-1000 demonstrates highly efficient catalytic activity for
the quantitative cycloaddition of styrene oxide using CO2 at 1
atm gauge pressure to form styrene carbonate at room
temperature (r.t.) (Table 1, entry 1a). To the best of our
knowledge, this is the mildest and the most efficient catalytic
system for this type of reaction. Since the conversion of epoxide
to carbonate is complete and quantitative, the pure product could
be obtained after a simple aqueous extraction, without the need
for laborious purification steps such as distillation, which can
cause decomposition of the product and the formation of
byproducts.23 Under the same reaction conditions, NU-1000 was
not as efficient as Hf-NU-1000 (entry 1b), confirming our
hypothesis about their relative catalytic activities. When the same
reaction conditions were employed, but in the absence of
Table 1. Cycloaddition Reactions of CO2 with Epoxides
Catalyzed by Hf-NU-1000 Yielding Cyclic Carbonates
a
Reaction conditions: epoxide (0.2 mmol), catalyst (4.0 mol % −OH
active site), and nBu4NBr (10 mol %) under CO2 (1 atm gauge
b
pressure). The same conditions as in footnote a, but without the
c
catalyst. Determined by 1H NMR analysis using 1-bromo-3,5-
difluorobenzene as the internal standard.
B
dx.doi.org/10.1021/ja508626n | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
opening. As the initial step, we chose to focus on the synthesis of
vicinal azidohydrins, which are important precursors of α-amino
alcohols25 (known as β-blockers). Additionally, vicinal azidohy-
drins are present in various bioactive natural products26 and in
the chemistry of carbohydrates,27 nucleosides,28 lactams,29 and
oxazolines.30 Although azidohydrins are generally synthesized
from epoxides by reaction with alkali azides, this classical method
is often accompanied by side reactions such as isomerization,
epimerization, and rearrangement.27,31 To the best of our
knowledge, there is no report of highly regioselective azidolysis
of epoxides with a high yield and complete conversion of
substrate.32 Tanabe and Cohen33 and Song et al.34 studied this
reaction using a series of MOFs, but they obtained only moderate
conversions and/or required stepwise postsynthetic modifica-
tions.
Figure 2. Optimized geometries of the TS structures TSa and TSb for
the epoxide ring-opening steps leading to the products A and B,
respectively. H atoms of the phenyl rings have been removed for clarity.
Under the optimized conditions, Hf-NU-1000 catalyzed the
solvolytic reaction of trimethylsilyl azide (TMS-N3) with styrene
oxide with high regioselectivity for substitution at the benzylic
position and complete conversion of substrate (Table S3, entry
1). The catalyst was also reused five successive times, and neither
a considerable decrease in the efficiency of the catalyst nor
structural deterioration as determined by PXRD analysis was
observed (see the SI). In order to gain a better understanding of
the origin of this unique regioselectivity, a few control reactions
were conducted. First, hafnium(IV) oxychloride octahydrate was
utilized as a Lewis acid catalyst to examine whether the
regioselectivity originates from HfIV; the conversion decreased
to 75%, and the co-occurrence of a side-product (30%) was
observed (entry 2). When dehydrated Hf-NU-1000 was used,
both the conversion (60%) and regioselectivity (85:15)
decreased (entry 3). Additionally, aqueous HCl (an alternative
homogeneous catalyst) was used to catalyze this reaction, but it
showed poor conversion (48%) and the co-occurrence of a side
product (26%) (entry 4).
In order to elucidate the mechanism of the reaction, we turned
to computational modeling. Quantum-chemical calculations
were carried out using DFT at the M06-L/def2-SVP level. The
SMD continuum solvent model35 was employed to account for
solvation with TMS-N3 as the solvent. Hf-NU-1000 serves as a
proton donor in the ring-opening reaction because of the
presence of aqua and OH ligands in the Hf6 node. Our
calculations show that styrene oxide forms an association
complex with the Hf6 node via both hydrogen-bonding and π-
stacking interactions (see the SI). The main criteria defining the
reaction selectivity are the difference in the activation barriers of
the epoxide ring-opening step (kinetic factor) and the relative
stability of the products (if the reaction operates under
thermodynamic control). The optimized geometries of the
transition state (TS) structures TSa (azide attack at the benzylic
position) and TSb (azide attack at the less hindered side) leading
to products A and B, respectively, are depicted in Figure 2. In
both TS structures, the aromatic ring of styrene oxide engages in
a T-shaped π-stacking interaction with the aromatic rings of the
Hf6 node linker ligands. The calculations show that for the
reaction catalyzed by the Hf6 node, the difference in activation
free energies is 3.0 kcal/mol, while in the case of HCl the
difference is about 2 kcal/mol. The difference in the free energies
of the products [ΔfG2°98(b′) − ΔfG2°98(a′)] is 0.1 kcal/mol, while
with HCl it is −2.3 kcal/mol. These numbers track the variations
in the experimental ratios of the products A and B. To assess the
effect of the predicted π-stacking interaction, PO was reacted
with TMS-N3 under the same conditions as mentioned in Table
S3 for styrene oxide, and a dramatic decrease in regioselectivity
was detected. In this case, a 25:75 ratio of the two regioisomers
was observed, with the major product resulting from azide attack
at the less hindered side (see the SI).
Encouraged by the above results, we employed Hf-NU-1000
to catalyze alcoholytic epoxide ring opening. The opening of
epoxides with alcohols is an important transformation in the
synthesis of β-alkoxy alcohols.36 However, the application of
such reactions generally is limited because of the poor
nucleophilicity of alcohols, which requires harsh and/or strongly
acidic conditions, usually leading to the formation of a mixture of
regioisomers and polymerization.37 Therefore, the regioselective
alcoholysis of epoxides has been the subject of extensive
studies.38 On the basis of the above optimized geometries of
the transition states, we speculated that the approach of
methanol to hydrogen-bonded epoxide would show good
stereoselection. Therefore, under our optimized conditions,
(+)-(R)-styrene epoxide was reacted with methanol in the
presence of Hf-NU-1000 (4.0 mol % of −OH active sites). The
regioisomer corresponding to attack at the benzylic position with
inversion of the epoxide stereogenic center was obtained
selectively and in an enantioretentive fashion (Scheme 1, top).
Additionally, (+)-(R)-styrene epoxide was also reacted with
TMS-N3 in the presence of Hf-NU-1000 (4.0 mol % of −OH
active sites). Again, the preferred regioisomer was obtained with
inversion of the epoxide stereogenic center in an enantio-
retentive fashion (Scheme 1, bottom). These selective inversions
of the stereogenic center indicate an SN2-type mechanism in
which the carbocation is not formed as an intermediate. Hf-NU-
1000, an achiral catalyst, yielded the enantiomerically pure
product, which makes it a promising catalyst in asymmetric
catalysis on account of its simple design.
In summary, a new polyoxohafnium-cluster-based MOF
isostructural with Zr-based NU-1000 has been prepared and
characterized. It demonstrates excellent performance in the
context of chemical fixation of CO2 as an inexpensive,
Scheme 1. Regioselective and Enantioretentive Methanolysis
(top) and Azidolysis (bottom) of Styrene Oxide Catalyzed by
Hf-NU-1000 To Give β-Methoxy- and β-Azido Alcohols
C
dx.doi.org/10.1021/ja508626n | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
2615. (b) Feng, D. W.; Chung, W. C.; Wei, Z. W.; Gu, Z. Y.; Jiang, H. L.;
Chen, Y. P.; Darensbourg, D. J.; Zhou, H. C. J. Am. Chem. Soc. 2013, 135,
17105. (c) Guillerm, V.; Weselinski, L. J.; Belmabkhout, Y.; Cairns, A. J.;
D'Elia, V.; Wojtas, L.; Adil, K.; Eddaoudi, M. Nat. Chem. 2014, 6, 673.
(d) See the SI and Table S2 for the complete list of references reporting
MOFs as catalysts for the synthesis of cyclic carbonates.
(10) Doskocil, E. J.; Bordawekar, S. V.; Kaye, B. G.; Davis, R. J. J. Phys.
Chem. B 1999, 103, 6277.
environmentally benign, ubiquitous, and sustainable carbon
source for the preparation of cyclic carbonates under ambient
conditions. In addition, this multifunctional catalyst is efficient
for solvolytic reactions and activates epoxides for the
regioselective and enantioretentive synthesis of 1,2-bifunction-
alized systems. We have further shown that the catalyst can be
recycled and reused for multiple successive runs without
considerable loss of activity or structural deterioration. We
speculate that this catalyst will be useful for the fixation of CO2,
the regioselective synthesis of enantiomerically enriched β-
azidohydrins and β-alkoxy alcohols, and related reactions of
epoxides with suppressed formation of byproducts while
adhering well to the principles of green chemistry and atom
economy.
(11) Wang, J. Q.; Kong, D. L.; Chen, J. Y.; Cai, F.; He, L. N. J. Mol.
Catal. A: Chem. 2006, 249, 143.
(12) (a) Yasuda, H.; He, L. N.; Sakakura, T.; Hu, C. W. J. Catal. 2005,
233, 119. (b) Yasuda, H.; He, L. N.; Sakakura, T. J. Catal. 2002, 209, 547.
(c) Yamaguchi, K.; Ebitani, K.; Yoshida, T.; Yoshida, H.; Kaneda, K. J.
Am. Chem. Soc. 1999, 121, 4526. (d) Yano, T.; Matsui, H.; Koike, T.;
Ishiguro, H.; Fujihara, H.; Yoshihara, M.; Maeshima, T. Chem. Commun.
1997, 1129.
(13) Srivastava, R.; Srinivas, D.; Ratnasamy, P. J. Catal. 2005, 233, 1.
(14) Xie, Y.; Wang, T. T.; Liu, X. H.; Zou, K.; Deng, W. Q. Nat.
Commun. 2013, 4, 1960.
(15) Chun, J.; Kang, S.; Kang, N.; Lee, S. M.; Kim, H. J.; Son, S. U. J.
Mater. Chem. A 2013, 1, 5517.
(16) Mondloch, J. E.; Bury, W.; Fairen-Jimenez, D.; Kwon, S.;
DeMarco, E. J.; Weston, M. H.; Sarjeant, A. A.; Nguyen, S. T.; Stair, P.
C.; Snurr, R. Q.; Farha, O. K.; Hupp, J. T. J. Am. Chem. Soc. 2013, 135,
10294.
(17) Morris, W.; Volosskiy, B.; Demir, S.; Gandara, F.; McGrier, P. L.;
Furukawa, H.; Cascio, D.; Stoddart, J. F.; Yaghi, O. M. Inorg. Chem.
2012, 51, 6443.
ASSOCIATED CONTENT
* Supporting Information
Syntheses, characterization data, and experimental methods.
This material is available free of charge via the Internet at http://
pubs.acs.org.
■
S
AUTHOR INFORMATION
■
Corresponding Authors
j-hupp@northwestern.edu
o-farha@northwestern.edu
(18) Feng, D. W.; Gu, Z. Y.; Li, J. R.; Jiang, H. L.; Wei, Z. W.; Zhou, H.
C. Angew. Chem., Int. Ed. 2012, 51, 10307.
Notes
The authors declare no competing financial interest.
(19) Kerr, J. A. In CRC Handbook of Chemistry and Physics, 81st ed.;
Lide, D. R., Ed.; CRC Press: Boca Raton, FL, 2000.
(20) It is not surprising that Hf-NU-1000 and Zr-NU-1000 have
similar structures, since Hf4+ and Zr4+ cations have ionic radii of 0.78 and
0.79 Å, respectively, as well as similar coordination chemistry. See: Bon,
V.; Senkovskyy, V.; Senkovska, I.; Kaskel, S. Chem. Commun. 2012, 48,
8407−8409.
(21) Vermoortele, F.; Bueken, B.; Le Bars, G.; Van de Voorde, B.;
Vandichel, M.; Houthoofd, K.; Vimont, A.; Daturi, M.; Waroquier, M.;
Van Speybroeck, V.; Kirschhock, C.; De Vos, D. E. J. Am. Chem. Soc.
2013, 135, 11465.
(22) Planas, N.; Mondloch, J. E.; Tussupbayev, S.; Borycz, J.; Gagliardi,
L.; Hupp, J. T.; Farha, O. K.; Cramer, C. J. J. Phys. Chem. Lett. 2014, 5,
3716.
(23) Kruper, W. J.; Dellar, D. V. J. Org. Chem. 1995, 60, 725.
(24) Unnikrishnan, K. P.; Thachil, E. T. Des. Monomers Polym. 2006, 9,
129.
ACKNOWLEDGMENTS
■
The DFG is gratefully acknowledged for the Postdoctoral
Research Fellowship Award to M.H.B. We also thank ISEN,
DOE, Office of Basic Energy Sciences, Division of Chemical
Sciences, Geosciences and Biosciences under Award DE-FG02-
12ER16362 and the Joint Center of Excellence in Integrated
Nano-Systems (JCIN) at KACST and Northwestern University
(Project 34-937). All NMR runs were performed at IMSERC
funded by the NSF (NSF CHE-1048773), IIN, and NU. We
thank Dr. W. Bury, Dr. J. E. Mondloch, Dr. Y. Zhang, Dr. Y. Wu
and Dr. A. A. Sarjeant for helpful discussions.
REFERENCES
■
(1) Furukawa, H.; Cordova, K. E.; O’Keeffe, M.; Yaghi, O. M. Science
2013, 341, 974.
(25) Cho, B. T.; Kang, S. K.; Shin, S. H. Tetrahedron: Asymmetry 2002,
13, 1209.
(2) (a) Liu, J.; Chen, L.; Cui, H.; Zhang, J.; Zhang, L.; Su, C.-Y. Chem.
Soc. Rev. 2014, 43, 6011. (b) Lee, J.; Farha, O. K.; Roberts, J.; Scheidt, K.
A.; Nguyen, S. T.; Hupp, J. T. Chem. Soc. Rev. 2009, 38, 1450. (c) Ma, L.
Q.; Abney, C.; Lin, W. B. Chem. Soc. Rev. 2009, 38, 1248.
(3) Aresta, M.; Dibenedetto, A. Dalton Trans. 2007, 2975.
(4) Since CO2 is the most oxidized state of carbon, thermodynamically
carbonates would be the most suitable candidates, as the four chemical
bonds of the carbonyl carbon are connected to oxygen. To validate this
hypothesis, as a model reaction we calculated ΔG for CO2 insertion into
PO in water at 298 K using multiple functionals and determined that the
free energy is approximately 8−15 kcal/mol in favor of the product (see
SI).
(26) Smith, J. G. Synthesis 1984, 629.
(27) Scriven, E. F. V.; Turnbull, K. Chem. Rev. 1988, 88, 297.
(28) Coe, D. M.; Myers, P. L.; Parry, D. M.; Roberts, S. M.; Storer, R. J.
Chem. Soc., Chem. Commun. 1990, 151.
(29) Smith, B. T.; Gracias, V.; Aube, J. J. Org. Chem. 2000, 65, 3771.
(30) Badiang, J. G.; Aube, J. J. Org. Chem. 1996, 61, 2484.
(31) The Chemistry of the Azido Group; Patai, S., Ed.; Wiley: London,
1971.
(32) Kas’yan, L. I.; Okovityi, S. I.; Kas’yan, A. O. Russ. J. Org. Chem.
2004, 40, 1.
(33) Tanabe, K. K.; Cohen, S. M. Inorg. Chem. 2010, 49, 6766.
(34) Song, F. J.; Wang, C.; Lin, W. B. Chem. Commun. 2011, 47, 8256.
(35) Marenich, A. V.; Cramer, C. J.; Truhlar, D. G. J. Phys. Chem. B
2009, 113, 6378.
(36) Baylon, C.; Prestat, G.; Heck, M. P.; Mioskowski, C. Tetrahedron
Lett. 2000, 41, 3833.
(37) (a) Olah, G. A.; Fung, A. P.; Meidar, D. Synthesis 1981, 280.
(b) Reeve, W.; Christoffel, I. J. Am. Chem. Soc. 1950, 72, 1480.
(38) See the SI for an extensive list of references for the synthesis of β-
methoxy alcohols from styrene oxide using different catalysts.
(5) Sakakura, T.; Kohno, K. Chem. Commun. 2009, 1312.
(6) North, M.; Pasquale, R. Angew. Chem., Int. Ed. 2009, 48, 2946.
(7) Markewitz, P.; Kuckshinrichs, W.; Leitner, W.; Linssen, J.; Zapp, P.;
Bongartz, R.; Schreiber, A.; Muller, T. E. Energy Environ. Sci. 2012, 5,
̈
7281.
(8) See the SI for an extensive list of references reporting
homogeneous catalysts for the synthesis of cyclic carbonates.
(9) (a) Gao, W.-Y.; Chen, Y.; Niu, Y.; Williams, K.; Cash, L.; Perez, P.
J.; Wojtas, L.; Cai, J.; Chen, Y.-S.; Ma, S. Angew. Chem., Int. Ed. 2014, 53,
D
dx.doi.org/10.1021/ja508626n | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX