Redox active aluminium(iii) complexes convert CO2 into MgCO 3 or CaCO3 in a synthetic cycle using Mg or Ca metal

Article abstract of DOI:10.1039/c2cc37208h

Redox-active Group 13 molecules possess the unusual combination of concomitant redox and acid-base reactivity. These combined properties enable regeneration of a metal hydroxide complex in a cycle for conversion of CO ₂ into carbonate salts. Reaction of (IP^-) ₂Al(OH) (M = Al, Ga) with 1 atm of CO₂ affords [(IP ^-)₂Al]₂(μ²κ¹: κ²-OCO₂). Subsequent reduction affords MgCO ₃ or CaCO₃ and two equivalents of [(IP^2-) ₂Al]^-, which can be reoxidized to (IP^-) ₂Al(OH) to close a cycle.

Full text of DOI:10.1039/c2cc37208h

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Redox active aluminium(III) complexes convert CO2
into MgCO3 or CaCO3 in a synthetic cycle using Mg
or Ca metal†‡
Cite this: Chem. Commun., 2013,
49, 4175
Received 3rd October 2012,
Accepted 9th November 2012
Thomas W. Myers and Louise A. Berben*
DOI: 10.1039/c2cc37208h
www.rsc.org/chemcomm
Redox-active Group 13 molecules possess the unusual combination complexes, (IPÀ)2M(OH) (where M = Al, 1; or Ga, 2), highly
of concomitant redox and acid–base reactivity. These combined nucleophilic and reactive toward CO2.7,8 In combination with the
properties enable regeneration of a metal hydroxide complex in a relative scarcity of monomeric Al(III)–OH and Ga(III)–OH
cycle for conversion of CO2 into carbonate salts. Reaction of complexes, and hence their relatively unexplored reactivity,9 this
(IPÀ)2Al(OH) (M
= Al, Ga) with 1 atm of CO2 affords unusual combination of properties prompted us to explore the
[(IPÀ)2Al]2(l2j1:j2-OCO2). Subsequent reduction affords MgCO3 reactivity of (IPÀ)2M(OH) with small molecules.
or CaCO3 and two equivalents of [(IP2À)2Al]À, which can be reoxidized
to (IPÀ)2Al(OH) to close a cycle.
Reactions of the zinc-containing carbonic anhydrase enzyme,10
its model complexes,11 and hydroxo complexes of acidic ions such
as Cd(II),12 and Sn(IV),13 with CO2 have demonstrated that for-
The activation and functionalization of CO2 is under investigation mation of a carbonate complexes results from reaction of the basic
for formation of fuels and commodity chemicals such as CO,1 M–OH unit with CO2. Dinuclear m-OH-bridged binuclear transition
methanol,2 and formate or formic acid,3 amongst others. The metal complexes have also been shown to afford carbonate from
formation of carbonate salts from CO2 is another example of a CO2.14 In most of these examples, the metal carbonate complex is
potential means to sequester and store CO2 as a useful product.4 the final product of the reaction.
For example, the aragonite form of CaCO3 is used industrially on
When 1 atm of CO2 was introduced into a hexanes solution
extremely large scales as the major ingredient for concrete and of (IPÀ)2Al(OH) a green precipitate was formed within 3 h (3,
MgCO3 is important, also on large scales, in a variety of applica- Scheme 1). Single crystals of 3 were grown by the diffusion of
tions such as fillers, drying agents, or whiteners. The removal of pentane into a concentrated THF solution of 3 at À25 1C over 1
CO2 from smoke stacks is desirable for the reduction of green- week (Fig. 1, Tables S1, S2, ESI‡). The solid state structure
house gases in the atmosphere, and the use of CO2 as a feedstock revealed that an acid–base reaction occurred and no change to
for fuels and chemical could contribute to its consumption and the IPÀ ligand oxidation state was apparent. Each of the four
recycling. It is well-known that strong bases react with CO2 to form IP ligands has bond lengths and angles consistent with the
carbonate salts as in the Solvay process,5 however methods are
needed to recycle the large quantities of base currently used for
such processes. Thus, the use of a regenerable hydroxide to afford
carbonates from CO2 would be a desirable process.
We have been interested in redox active iminopyridine (IP)
ligand complexes of the Group 13 metals, and these complexes
possess an unusual combination of properties.6 The complexes
have highly Lewis acidic metal centres combined with a capability
for performing redox reactions. In addition, the high Lewis acidity
of aluminium(III) and gallium(III) should render their hydroxide
Department of Chemistry, University of California, Davis, CA 95616, USA.
E-mail: laberben@ucdavis.edu; Tel: +1 530 752-8475
† This article is part of the ChemComm ‘Emerging Investigators 2013’ themed
issue.
‡ Electronic supplementary information (ESI) available: Preparation of
compounds, IR, CIF and CV data. CCDC 903780. For ESI and crystallographic
Scheme 1
data in CIF or other electronic format see DOI: 10.1039/c2cc37208h
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 4175--4177 4175

View Article Online
Communication
ChemComm
crystals of 4. Crystals of 4 were of sufficient quality to identify
the unit cell parameters as isostructural with 3 although a full
crystal structure determination was not possible.
Further characterization of 3 and 4 was achieved using IR
spectroscopy and magnetic susceptibility measurements. The
OCO2 moiety in 3 was labeled with 13C which enabled assignment
of C–O stretches in the IR spectrum at 1540 and 1519 cmÀ1 for 3
(Fig. 2 left, Table S3, ESI‡). The corresponding 13C–O absorption
bands were observed at 1507 and 1489 cmÀ1. By analogy, the C–O
absorptions for 4 were identified at 1540 and 1516 cmÀ1. The
electronic structures of 3 and 4 were probed using temperature
dependent susceptibility measurements and EPR spectroscopy
and were found to be consistent with a ligand-based antiferro-
magnetically coupled biradical electronic structure at each of the
metal centres (Fig. S1, ESI‡). The room temperature magnetic
moments for 3 and 4 are 3.47 and 3.52 mB, respectively, and are
consistent with this model.18
Fig. 1 Solid state structure of [(IPÀ)2Al]2(m2k1:k2-OCO2) in 3. Pink, grey, blue, and
red, ellipsoids represent Al, C, N, and O atoms, respectively. Ellipsoids at 50%
and H atoms are omitted.
A balanced reaction scheme for the formation of 3 suggested
that the byproduct of the reactions of (IPÀ)2Al(OH) with CO2
should be an equivalent of H2O (Scheme 1). We were unable to
detect any water in the reaction solution and so to probe the
reaction mechanism further, reaction of 3 with one equivalent
of H2O in THF was investigated. The starting complex 1 was
regenerated, as identified by IR spectroscopy, and CO2 was
released in 80% yield as measured by GC-TCD analysis of
the reaction headspace. As further confirmation that H2O is
liberated as a side product, 3 Å molecular sieves were added to
the reaction mixture to absorb any water released. Under these
conditions, the yield of 3 increased from 68% to 82%.
Compound 4 behaved similarly to 3.
Electrochemical measurements were performed on 3 and 4 in
0.3 M Bu4NPF6 THF solutions using cyclic voltammetry (CV). The
CV measurements revealed quasi-reversible oxidation events
associated with the IP ligands (Fig. 2 right, Table S4, ESI‡).
Complexes 3 and 4 were oxidized in two, successive quasi-
reversible events at À0.56 V and À0.16 V for 3, and at very similar
potentials for 4. The two events presumably each represent a
two-electron process, i.e. one IPÀ ligand at each metal centre of
the dinuclear molecule is oxidized in each event. The complexes
also each display two two-electron reduction events associated with
the IP ligands at À1.61 V and À1.95 V for 3 and at À1.52 V and
À1.82 V for 4. In 3 both of the two-electron reduction events are
quasi-reversible, but neither of these events is reversible in 4.19
These redox events correspond to the processes:
Chart 1
one-electron reduced oxidation state,6a and there are two
inequivalent aluminium centres in 3: one is five-coordinate
and the other is six coordinate. Overall, the molecular formula
for 3 is {(IPÀ)2Al}2(m2k1:k2-OCO2). One unusual feature associated
with the five-coordinate aluminium centre is the very low t value
of 0.552, as compared with our previously reported work on
five-coordinate complexes of the form (IPÀ)2AlX.15 We think this
is likely an effect of strain imposed by the bridging carbonate
arrangement. The bridging carbonate ligand itself has C–O
bond lengths of 1.279(6), 1.257(9), and 1.301(9) Å. These bond
lengths agree well with those observed in the handful of other
structurally characterized complexes containing the unsymme-
trical m2k1:k2-OCO2 moiety observed herein (Chart 1, left). These
other examples involve the metal ions Zn(II),14 Ti(IV), Rh(I), Co(II),
Fe(II), and U(IV).16 The other common coordination mode for a
m2-OCO2 ligand is the symmetrical bridging mode described as
m2k2:k2-OCO2 (Chart 1, right).14,17
The analogous gallium complex, (IPÀ)2Ga(OH),8 displayed
similar reactivity with CO2 and that reaction afforded green
[(IPÀ)2M]2(m2k1:k2-OCO2) 2 [(IPÀ)(IP2À)M]2(m2k1:k2-OCO2)
and
[(IPÀ)(IP2À)M]2(m2k1:k2-OCO2) 2 [(IP2À)2M]2(m2k1:k2-OCO2).
The accessible reduction events observed electrochemically in 3
and 4 implied that chemical reduction of the complexes should be
possible. As discussed earlier, the formation of both magnesium
and calcium carbonates from CO2 is desirable, and so the
reduction of 3 was attempted with both Mg and Ca metal.
In THF solution, a mixture of 3 and Mg turnings held at 60 1C
for 12 h produced MgCO3 and the magnesium salt of [(IP2À)2Al]À
Fig. 2 (left) IR spectra for 3 (blue), 3C13 (green), and 4 (red). Absorptions for
2À
CO3 labeled with %. (right) CV’s for 3 (blue) and 4 (red), recorded in 0.3 M
Bu4NPF6 THF solution. Glassy carbon electrode 100 mV sÀ1
.
c
4176 Chem. Commun., 2013, 49, 4175--4177
This journal is The Royal Society of Chemistry 2013

View Article Online
ChemComm
Communication
(5) in good yield. We have previously reported [(IP2À)2Al]À as a
sodium salt. Similarly, reaction of 3 with Ca metal at 60 1C for 1 h
afforded CaCO3 and the calcium salt of 5. The carbonate salts were
identified by IR spectroscopy and could be filtered off from the
reaction mixture. We have previously demonstrated that
(IPÀ)2Al(OH) are formed by the two-electron oxidation of four-
electron reduced [(IP2À)2Al]À. Therefore, the reactions described
above close a synthetic cycle that generates MgCO3 or CaCO3 from
CO2 with regeneration of the metal hydroxide necessary for CO2
activation. The cycle incorporates compounds 1, 3, and 5
(Scheme 1). A similar cycle was demonstrated for the analogous
gallium complexes 2, 4, and [(IP2À)2Ga]À (6).
To further illustrate the utility of the aluminium-based synthetic
cycle for interconversion of 1, 3, and 5 and the formation of MgCO3
or CaCO3 from CO2 we performed the reaction in a one-pot
procedure. By successive addition of CO2 to 1, removal of CO2
in vacuo, addition of Mg (or Ca), followed by Bu4NI and then
pyridine-N-oxide (pyO), MgCO3 (or CaCO3) could be generated from
CO2 in one system (Scheme 1).20 These cycles could be repeated in
the same vessel at least three times to generate 1.4 turnovers of
MgCO3, or 2.3 turnovers of CaCO3 as an isolated white solid. We
have already found that addition of 3 Å molecular sieves to the
reaction increases the isolated yield of MgCO3 from 1.1 to
1.4 turnovers, and the isolated yield of CaCO3 from 1.7 to
2.3 turnovers. We anticipate that further optimization of the reaction
conditions will provide even greater improvements to the isolated
yields of MgCO3 and CaCO3. A related synthetic cycle for conversion
of CO2 has recently been reported by Meyer and coworkers using a
uranium-mediated process. In that case the CO2 activating complex
consists of a U–O–U core which affords a uranium carbonate
complex along with CO gas upon reaction with CO2.17c
Alternative synthetic pathways for the functionalization of
the trapped CO2 in complexes 3 are of interest and so we also
investigated the reactivity of 3 with Me3SiCl. Upon addition of
Me3SiCl to a solution of 3, formation of (IPÀ)2AlCl along with
Me3SiOC(O)OSiMe3 was observed as confirmed by IR spectro-
scopy and GC-MS.
Taken together, the chemistry described herein demon-
strates that redox-active Group 13 complexes can possess both
Lewis acidic properties and redox activity, and these properties
can be utilized in a synthetic cycle for CO2 conversion into
either MgCO3 or CaCO3. In future work we will take further
advantage of the combined redox and Lewis acid reactivity
afforded us by these complexes to effect transformations on
small molecules. We will also extend this demonstrated
formation of C–O bonds to investigate the formation of C–N
and C–S bonds with CO2 through substitution of the OH ligand
in (IPÀ)2M(OH) with S- and N-donor ligands.
J. Am. Chem. Soc., 1996, 118, 1769; (d) J. W. Raebiger, J. W. Turner,
B. C. Noll, C. J. Curtis, A. Miedaner, B. Cox and D. L. DuBois,
Organometallics, 2006, 25, 3345.
2 For example: E. B. Cole, P. S. Lakkaraju, D. M. Rampulla,
A. J. Morris, E. Abelev and A. B. Bocarsly, J. Am. Chem. Soc., 2010,
132, 11539.
3 For example: (a) H. Ishida, H. Tanaka, K. Tanaka and T. Tanaka,
Chem. Commun., 1987, 131; (b) W. Leitner, Angew. Chem., Int. Ed.
Engl., 1995, 34, 2207; (c) C. Arana, S. Yan, K. M. Keshavarz,
K. T. Potts and H. D. Abruna, Inorg. Chem., 1992, 31, 3680;
(d) P. Kang, C. Cheng, Z. Chen, C. K. Schauer, T. J. Meyer and
M. Brookhart, J. Am. Chem. Soc., 2012, 134, 5500; (e) M. D. Rail and
L. A. Berben, J. Am. Chem. Soc., 2011, 133, 18577.
4 For example: (a) G. M. Bond, J. Stringer, D. K. Brandvold,
F. A. Simsek, M. G. Medina and G. Egeland, Energy Fuels, 2001,
15, 309; (b) R. M. Enick, E. J. Beckman, C. Shi, J. Xu and L. Chordia,
Energy Fuels, 2001, 15, 256; (c) A. M. Appel, R. Newell, D. L. DuBois
and M. Rakowski DuBois, Inorg. Chem., 2005, 44, 3046.
5 For example: H. P. Huang, Y. Shi, W. Li and C. S. Chang, Energy
Fuels, 2001, 15, 263.
6 (a) T. W. Myers, N. Kazem, S. Stoll, R. D. Britt, M. Shanmugam and
L. A. Berben, J. Am. Chem. Soc., 2011, 133, 8662; (b) K. Kowolik,
M. Shanmugam, T. W. Myers, C. D. Cates and L. A. Berben, Dalton
Trans., 2012, 7969.
7 T. W. Myers and L. A. Berben, J. Am. Chem. Soc., 2011, 133, 11865.
8 C. D. Cates, T. W. Myers and L. A. Berben, Inorg. Chem., 2012,
51, 11891.
9 For example: (a) R. J. Wehmschulte, J. M. Steele and M. A. Khan,
Organometallics, 2003, 22, 4678; (b) V. Jancik, L. W. Pineda,
A. C. Stu¨ckl, H. W. Roesky and R. Herbst-Irmer, Organometallics,
2005, 24, 1511; (c) S. Singh, V. Jancik, H. W. Roesky and R. Herbst-
Irmer, Inorg. Chem., 2006, 45, 949.
10 For example: (a) S. Lindskog, Pharmacol. Ther., 1997, 74, 1;
(b) D. W. Christianson and J. D. Cox, Annu. Rev. Biochem., 1999,
68, 33; (c) B. C. Tripp, K. Smith and J. G. Ferry, J. Biol. Chem., 2001,
276, 48615.
11 For example: (a) G. Parkin, Chem. Rev., 2004, 104, 699;
(b) H. Vahrenkamp, Acc. Chem. Res., 1999, 32, 589.
12 R. A. Allred, L. H. McAlexander, A. M. Arif and L. M. Berreau, Inorg.
Chem., 2002, 41, 6790.
´
13 E. Simon-Manso and C. P. Kubiak, Angew. Chem., Int. Ed., 2005,
44, 1125.
14 N. Kitajima, S. Hikichi, M. Tanaka and Y. Moro-oka, J. Am. Chem.
Soc., 1993, 115, 5496.
15 A. W. Addison, T. N. Rao, J. J. Van Rijn and G. C. Verschoor, J. Chem.
Soc., Dalton Trans., 1984, 1349.
16 For example: (a) S. Krogsrud, S. Komiya, T. Ito, J. A. Ibers and
A. Yamamoto, Inorg. Chem., 1976, 15, 2798; (b) V. V. Burlakov,
F. M. Dolgushin, A. I. Yanovsky, Y. T. Struchkov, V. B. Shur,
U. Rosenthal and U. Thewalt, J. Organomet. Chem., 1996, 522, 241;
(c) Z. W. Mao, F. W. Heinemann, G. Liehr and R. V. Eldik, J. Chem.
Soc., Dalton Trans., 2001, 3652; (d) C. Bergquist, T. Fillebeen,
M. M. Morlok and G. Parkin, J. Am. Chem. Soc., 2003, 125, 6189;
(e) A. R. Sadique, W. W. Brennessel and P. L. Holland, Inorg. Chem.,
2008, 47, 784; ( f ) U. P. Singh, P. Babbar and P. Tyagi, Transition Met.
Chem., 2008, 33, 931; (g) O. T. Summerscales, A. S. P. Frey, F. G.
N. Cloke and P. B. Hitchcock, Chem. Commun., 2009, 198;
(h) O. P. Lam, S. C. Bart, H. Kameo, F. W. Heinemann and
K. Meyer, Chem. Commun., 2010, 46, 3137.
17 For example: (a) D. A. Palmer and R. V. Eldik, Chem. Rev., 1983,
83, 651; (b) R. A. Allred, L. H. McAlexander, A. M. Arif and
L. M. Berreau, Inorg. Chem., 2002, 41, 6790; (c) A. C. Schmidt,
A. V. Nizovtsev, A. Scheurer, F. W. Heinemann and K. Meyer, Chem.
Commun., 2012, 48, 8634.
18 No attempts were made to fit the data. These results are consistent
with our previous work on biradical complexes of the form
We are grateful to the University of California Davis and the
Alfred P. Sloan Foundation for support of this work and the
National Science Foundation (Grant 0840444) for a Dual source
X-ray diffractometer.
(IPÀ)2MX6,7
.
19 Reduction peak heights are higher than oxidation peak heights
as observed for other complexes in this class. We ascribe this
to partial loss of the IP ligand upon oxidation. Moreover, free
IP in solution is reduced concomitant with the reduction events
for 3 and 4.
Notes and references
1 For example: (a) B. Fisher and R. Eisenberg, J. Am. Chem. Soc., 1980, 20 Control experiments demonstrated that pyO reacts only very
102, 7363; (b) K. S. Ratliff, R. E. Lentz and C. P. Kubiak, Organo-
metallics, 1992, 11, 1986; (c) I. Bhugun, D. Lexa and J. M. Saveant,
slowly with Mg metal, presumably because the reaction is
heterogeneous.
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 4175--4177 4177