Crystal structure of K2[C2O6] - First proof of existence and constitution of a peroxodicarbonate ion

Article abstract of DOI:10.1002/1521-3773(20020603)41:11<1922::AID-ANIE1922>3.0.CO;2-T

By using high-resolution X-ray powder diffraction, we have determined the first crystal structure of a peroxodicarbonate compound, synthesized by electrocrystallization at -20°C. The constitution of the peroxodicarbonate anion was established and all structural details, including its conformation (see picture), can be understood qualitatively in terms of classical concepts of the chemical bond.

Full text of DOI:10.1002/1521-3773(20020603)41:11<1922::AID-ANIE1922>3.0.CO;2-T

COMMUNICATIONS
conclusive evidence for its constitution,[8] and even existence,
has been lacking. Therefore we have reinvestigated a material
with the bulk composition of K2C2O6 which, according to
vibrational spectroscopy,[3e] contains peroxide functions. By
means of a crystal structure determination using high-
resolution X-ray powder diffraction data (Figure 1) we show
that this compound is free of water and hydrogen peroxide of
crystallization, and contains the fully deprotonated peroxodi-
carbonate anion.[9]
Crystal Structure of K2[C2O6]–First Proof of
Existence and Constitution of a
Peroxodicarbonate Ion**
Robert E. Dinnebier, Sascha Vensky,
Peter W. Stephens, and Martin Jansen*
The characteristics of homonuclear bonds between second-
row elements are predominantly determined by the strong p
overlap of 2p-type wave functions. These interactions can, on
one hand, cause considerable strengthening through multiple
bond formation, and on the other, weakening of the bonds
through lone pair/lone pair repulsion, as soon as the number
of p electrons involved per atom exceeds three. The general
trend is significantly modulated by the nature of the
substituents and the conformation across the homonuclear
bond. Those substituents lacking lone pairs and exhibiting
high electronegativities appear to reinforce the X X homo-
nuclear single bonds, by increasing the s contribution to the
sigma bond, and by removing electron density from the
repulsive lone pairs. This is immediately evident upon
À
comparing the O O bond lengths in O2F2, O2(SiMe3)2, and
O22À (in K2O2) which cover the impressively wide range from
122 to 150 and 154 pm.[1] Fully in line with these consider-
ations, among the known diperoxo anions, peroxodisulfate is
more stable than peroxodiphosphate, and similar anions of
Group 14 elements, such as peroxodisilicate, are not known at
all.[2] Since their discovery in the 19th century, percarbonates
have been studied repeatedly, in order to reveal constitutions
and bonding properties, and to inspect their potential for
application as bleaching agents in the detergent industry.[3] In
spite of these efforts, the extent of reliable insights has
remained limited.[4] Major problems arise from the fact that
peroxide functions in the structures can also originate from
hydrogen peroxide of crystallization,[6] and because of the
thermal lability preventing complete removal of water by
drying. It has been shown recently by single-crystal X-ray
structure analysis that KHCO4 ¥ H2O2, and the respective
rubidium compound, unambiguously constitute hydrogen-
monoperoxocarbonates.[7] For a peroxodicarbonate, however,
Figure 1. Scattered X-ray intensity for potassium peroxodicarbonate at
Tˆ À1238C as a function of diffraction angle 2V. Shown are the observed
pattern (diamonds), the best Rietveld-fit profile in P21/c (line a), the
difference curve between observed and calculated profile (line b), and the
reflection markers (vertical bars). The wavelength was l ˆ 1.12074(2) ä.
The higher angle part starting at 2Vˆ 328 is enlarged by a factor of 5.
Potassium peroxodicarbonate was synthesized electro-
chemically from a saturated aqueous solution of K2CO3 at
Tˆ À208C. The samples as obtained are microcrystalline and
white, with a faint sky blue color. We have been able to
prepare single-phase material; however, most charges contain
small amounts of KHCO3, the first decomposition product of
K2C2O6 in an aqueous environment, as an impurity.
The peroxodicarbonate anion (Figure 2) consists of two
carbonate groups that are connected through two of its
oxygen atoms to form a peroxo group. The molecular
symmetry was found to be C2 within the limits of experimental
error.[13] The carbon atoms are surrounded by oxygen atoms in
a near trigonal-planar arrangement and in general, the
[*] Prof. Dr. M. Jansen, Priv.-Doz. Dr. R. E. Dinnebier,
Dipl.-Chem. S. Vensky
Max-Planck-Institute for Solid-State Research
Heisenbergstrasse 1, 70569 Stuttgart (Germany)
Fax : (‡49)711-689-1502
E-mail: m.jansen@fkf.mpg.de
Prof. Dr. P. W. Stephens
Department of Physics and Astronomy
State University of New York
Stony Brook, NY 11974-3800 (USA)
[**] This work was supported by the Deutsche Forschungsgemeinschaft
(DFG) and the Fonds der Chemischen Industrie. Research was carried
out in part at the National Synchrotron Light Source at Brookhaven
National Laboratory, which is supported by the U.S. Department of
Energy, Division of Materials Sciences and Division of Chemical
Sciences. The SUNY X3 beamline at NSLS is supported by the
Division of Basic Energy Sciences of the US Department of Energy
under Grant No. DE-FG02-86ER45231. Work at Stony Brook was
partially supported by the National Science Foundation. We thank
Michael Knapp from beamline B2 (Hasylab) for the measurements at
Hasylab.
Figure 2. Conformation of the peroxodicabonate anion as found in K2C2O6
at Tˆ À1238C. The torsion angle between the two carbonate groups is
given.
1922
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
1433-7851/02/4111-1922 $ 20.00+.50/0
Angew. Chem. Int. Ed. 2002, 41, No. 11

COMMUNICATIONS
intramolecular bond lengths and angles of the carbonate
group are within the typical range found for this class of
molecules (Table 1). Remarkably, the bond length between
the carbon atom and the peroxo-oxygen atom is only 3%
(Table 1). The potassium ions are connected to seven oxygen
atoms of five different peroxodicarbonate dianions. In the
case of the K(1)O7 polyhedron, one edge is equivalent to a
peroxo bridge, whereas in the case of the K(2)O7 polyhedron,
one edge is equivalent to the triangular edge of a
carbonate group. In both cases another peroxodicar-
bonate group is side-on coordinated (carbonate- and
peroxo-oxygen atoms).
Table 1. Selected bond lengths [pm] and torsion angles [8] for K2C2O6 and related
molecules.
À
À
À
À
Substance
Ref. d(O O) d(K O)
d(C O)
d(C OO) C-O-O-C
The crystal structure of K2[C2O6] is based on a three-
dimensional framework of KO7 polyhedra, intercon-
nected by carbonate groups. For a better understanding,
if only the central carbon atoms of the carbonate groups
are considered, the crystal structure of K2C2O6 can be
viewed as a strongly distorted NaCl arrangement with
mean KC6 and CK6 distances of 350(20) pm (Figure 4).
K2C2O6
KHCO4 ¥ H2O2
H2O2
K2CO3
(C6H11)2C2O6
147(1)
146
146
À
263(1) 290(1) 126(2) 129(2) 131(1)
93(1)
101COOH
93HOOH
À
[7a]
275 283
123 125
À
138
À
[15a]
[15b]
[15c]
À
264 286
127 129
118 132
À
143
À
138
90
À
longer than the lengths of the terminal C O bonds. Therefore,
this bond contains some degree of double-bond character.
The deviations from planarity for the carbonate groups
including the second peroxo-oxygen atom are below the
À
precision of the measurement. The O O bond length in the
peroxo group (147(1) pm) is virtually the same as for the
related peroxomonohydrogencarbonate anion (146 pm).[7]
The C-O-O-C dihedral angle of 93(1)8 corresponds to that
for dicyclohexylperoxodicarbonate (908)[15c] and for solid
H2O2 (938).[15a] According to the experimental findings (short
À
bonds, planarity of OOCO2 moiety) the C O(peroxo) bonds
show double-bond character and the peroxo-oxygen atoms
can be regarded as sp2 hybridized which leads to an occupied
À
pz orbital orthogonal to the O2C OO plane. The strong
À
repulsion of these orbitals across the O O bond favors the
Figure 4. Packing scheme of K2C2O6 at Tˆ À1238C in a view perpendic-
ular to the plane spanned by the c axis and the median of a and b axes. The
structural relationship to the NaCl type of structure is clearly visible.
staggered conformation of O2C-O-O-CO2.
There are two inequivalent potassium sites, but every
potassium ion is coordinated to seven oxygen atoms, forming
irregular KO7 polyhedra (Figure 3). The polyhedra can be
described as strongly distorted trigonal prisms with an addi-
tional oxygen atom above one of the rectangular faces, which
forms a long bond to the central cation. The bond lengths
between the potassium atoms and the oxygen atoms are in the
range 2.63 to 2.9 ä, comparable to that of related compounds
Application of peroxodicarbonate as a bleaching compo-
nent in detergents would liberate harmless carbon dioxide and
potassium, besides the desired active oxygen, and would help
to maintain favorable pH conditions. Also the onset of rapid
decomposition occurs at an appreciably high temperature of
1418C. Nevertheless all prospects for application in the
detergent industry seem to be hindered by its reduced
stability during long-term storage.
Experimental Section
K2C2O6 was synthesized by anodic oxidation of a saturated aqueous
solution of potassium carbonate (p.a., Merck) at Tˆ À208C. The electro-
crystallization was carried out under galvanostatic conditions (I ˆ 250 mA,
U > 16 V) with a potentiostat (EG&G Princeton Applied Research, Type
363). A Pt wire (Æ ˆ 0.5 mm, l ˆ 40 mm) was used as an anode, a Pt mesh
(Æcyl. ˆ 40 mm, h ˆ 50 mm, 20 mesh) as cathode. The anodic part was
separated by a glassy membrane. Potassium peroxodicarbonate, obtained
as a microcrystalline, light blue powder, was filtered and washed with
ethanol and diethyl ether. The peroxide content was determined iodo-
metrically.[16] K2C2O6 decomposes rather slowly when stored below À208C
for several weeks (8% per week).
According to the results of thermal analysis (DTA/TG/MS; Metzsch STA
429; 208C 8008C; 28CminÀ1) and in agreement with previous exper-
iments,[3e] K2C2O6 decomposes at 1418C by generation of oxygen and
carbon dioxide.
Figure 3. Representation of the two crystallographically distinct KO7
coordination polyhedra and specially connected peroxodicarbonate groups
of potassium peroxodicarbonate at Tˆ À1238C. The other oxygen atoms
are carbonate-oxygen atoms of end-on coordinated peroxodicarbonate
groups.
Received: January 28, 2002 [Z18360]
Angew. Chem. Int. Ed. 2002, 41, No. 11
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
1433-7851/02/4111-1923 $ 20.00+.50/0
1923

COMMUNICATIONS
[1] a) J. J. Turner, Endeavour 1968, 27, 42 47; b) D. Kˆnigstein, M.
Jansen, Monatsh. Chem. 1996, 127, 1221 1227; c) T. Bremm, M.
Jansen, Z. Anorg. Allg. Chem. 1992, 610, 64 66.
[13] Molecular calculations (MINDO), in which C2 symmetry was assumed
for the peroxodicarbonate group, failed to correctly postulate the
bond lengths and the dihedral angle.[14]
[2] a) D. Y. Naumov, A. V. Virovets, N. V. Podberezskaya, P. B. Novikov,
A. A. Politov, J. Struct. Chem. 1997, 38, 772 778; b) W. P. Griffith,
R. D. Powell, A. C. Skapski, Polyhedron 1988, 7, 1305 1310.
[3] a) E. J. Constam, A. von Hansen, Z. Elektrochem. 1896, 7, 137 144;
b) A. von Hansen, Z. Elektrochem. 1896, 7, 445 448; c) A. K.
Melnikov, T. P. Firsova, A. N. Molodkina, Russ. J. Inorg. Chem.
[14] C. Glidewell, J. Mol. Struct. 1980, 67, 35 44.
[15] a) J.-M. Savariault, M. S. Lehmann, J. Am. Chem. Soc. 1980, 102,
1298 1303; b) Y. Idemoto, J. W. jr. Richardson, N. Koura, S. Kohara,
C.-K. Loong, J. Phys. Chem. Solids 1998, 59, 363 376; A. Y. Kosnikov,
V. L. Antonovsky, S. V. Lindeman, Y. T. Struchkov, I. P. Zyatkov,
Kristallografiya 1988, 33, 875 877.
¡
1962, 7, 637 640; d) P. A. Gigue re, D. Lemaire, Can. J. Chem. 1972,
[16] G. Jander, K. F. Jahr, Ma˚analyse, 15th ed., de Gruyter, Berlin, 1989,
pp. 197 198.
50, 1472 1477; e) D. P. Jones, W. P. Griffith, J. Chem. Soc. Dalton
Trans. 1980, 2526 2532.
[4] Early X-ray powder diffraction erperiments have been reported,
which failed to find the correct unit cell and X-ray density.[5]
[5] V. I. Sokol, V. M. Bakulina, E. Y. Filatov, T. P. Firsova, Russ. J. Inorg.
Chem. 1968, 13, 1211 1212.
[6] a) E. H. Riesenfeld, B. Reinhold, Ber. Dtsch. Chem. Ges. 1909, 42,
4377 4383; b) J. M. Adams, R. G. Pritchard, Acta Crystallogr. Sect. B
1977, 33, 3650 3653.
[7] a) A. Adam, M. Mehta, Angew. Chem. 1998, 110, 1457 1459; Angew.
Chem. Int. Ed. 1998, 37, 1387 1388; b) A. Adam, M. Mehta, Z.
Kristallogr. 1998, Supplement Issue 15, 46.
[8] Interpretation of the IR data was based on the wrong molecular
symmetry; however, the postulation of a peroxodicarbonate anion was
correct.[3e]
Self-Assembled Receptors for Enantioselective
Recognition of Chiral Carboxylic Acids in a
Highly Cooperative Manner
Tsutomu Ishi-i, Mercedes Crego-Calama,
Peter Timmerman,* David N. Reinhoudt,* and
Seiji Shinkai
[9] a) Crystal structure data of K2C2O6: monoclinic, space group P21/c
(no. 14), a ˆ 838.05(1), b ˆ 1076.41(2), c ˆ 711.67(1) pm, b ˆ
111.24(0)8 Vˆ 598.4(2)  106 pm3, 1calcd ˆ 2.200 gcmÀ3
,
1measured
ˆ
2.14(1) gcmÀ3 (He pycnometer, Micrometritics AccuPyc 1330 GB),
Z ˆ 4, m ˆ 55.04 cmÀ1. b) Devices for X-ray diffraction experiments:
Suny X3B1 beamline at National Synchroton Light Source, Brook-
haven National Laboratory, USA, double Si(111) monochromator,
and Ge(111) crystal analyzer, l ˆ 115.011(2) pm, Na(Tl)I scintillation
counter with pulse height discriminator, Tˆ À738C, 5.08 < 2V<
43.6798 in steps of 0.00382V, glass capillary of 0.7mm diameter; due
to decomposition even at low temperature, a second measurement for
structure refinement was performed at beamline B2 at the Hamburger
Synchrotronstrahlungslabor (HASYLAB) immediately after synthe-
sis of the material: Three-circle Huber goniometer, Ge(111) double
crystal monochromator, and Ge(111) crystal analyzer, l ˆ
112.074(2) pm, Na(Tl)I scintillation counter, Tˆ À1238C, 9.08 <
2V< 50.58 in steps of 0.00382V, glass capillary of 0.7mm diameter.
c) Structure determination process: Data reduction and background
modeling by using the GUFI program,[10] crystal structure solution by
global optimization with 15 parameters by using the DASH program
package,[11] two phases of Rietveld refinements by employing a model
for anisotropic micro-strain with potassium hydrogencarbonate as
second phase by using the GSAS program package,[12] Rp ˆ 0.193,
Rwp ˆ 0.246, RF ˆ 0.080, c2 ˆ 0.84, number of reflections 281, number
of variables 35, number of refined atoms 10. The relatively high
weighted-profile R factor is due to the statistics of the observed step
scan intensities caused by limited time for measurement. d) Further
details on the crystal structure investigation may be obtained from the
Fachinformationszentrum Karlsruhe, 76344 Eggenstein-Leopoldshaf-
en, Germany (fax: (‡49)7247-808-666; e-mail: crysdata@fiz-karls-
ruhe.de), on quoting the depository number CSD-412335.
One of the ultimate aims in molecular recognition is to
understand fully and mimic the exquisite selectivities so
eminently observed in natural receptors, such as antibodies
and enzymes. A first approach is to build synthetic receptors
by using rigid covalent scaffolds, to which are attached
functional groups that can bind their guest molecules by
multiple noncovalent interactions, such as hydrogen bonding,
p
p stacking, or metal coordination.[1] However, covalent
systems are often too rigid and unable to adapt their shape to
that of the guest, which can result in less than optimal binding
affinities and selectivities. Another important drawback of
covalent receptor molecules is their labor-intensive synthesis,
which leaves little potential for structural variations in the
scaffold.[2]
A different approach to shaping the binding site of an
artificial receptor is to bring together the different compo-
nents by multiple noncovalent interactions.[3, 4] This approach,
which more closely resembles Nature×s strategy, is currently
being investigated as a potential alternative to covalent
receptor molecules. A number of systems based on coordina-
tive metal ligand interactions[5] or hydrogen-bonding inter-
actions[6] have been investigated, and some show significant
structural selectivities.[7] However, in the majority of cases,
substrate selectivity is the result of shape complementarity
[10] R. Dinnebier, GUFI a program for measurement and evaluation of
powder patterns, Version 5.0, Heidelberger Geowiss. Abh. 68, Hei-
delberg, 1997.
[11] a) W. I. F. David, K. Shankland, N. Shankland, Chem. Commun.
1998, 931 932; b) J. W. Visser, J. Appl. Crystallogr. 1969, 2,
89 95; c) A. Le Bail, H. Duroy, J. L. Fourquet, Mater. Res. Bull.
1988, 23, 447 452; d) J. Rodriguez-Carvajal, Abstracts of the Satellite
Meeting on Powder Diffraction of the XV Congress of the IUCR, 1990,
p. 127.
[12] a) A. C. Larson, R. B. von Dreele, GSAS 1990, Version Sept. 1997, Los
Alamos National Laboratory Report LAUR 86 748; b) H. M.
Rietveld, J. Appl. Crystallogr. 1969, 2, 65 71; c) P. Thompson,
D. E. Cox, J. B. Hastings, J. Appl. Crystallogr. 1987, 20, 79 83;
d) L. W. Finger, D. E. Cox, A. P. Jephcoat, J. Appl. Crystallogr.
1994, 27, 892 900; e) P. W. Stephens, J. Appl. Crystallogr. 1999,
281 289.
[*] Dr. P. Timmerman, Prof. Dr. Ir. D. N. Reinhoudt, Dr. T. Ishi-i,
Dr. M. Crego-Calama
Laboratory of Supramolecular Chemistry and Technology
‡
MESA Research Institute, University of Twente
PO Box 217, 7500 AE Enschede (The Netherlands)
Fax : (‡31)53-4894645
E-mail: d.n.reinhoudt@ct.utwente.nl
Prof. Dr. S. Shinkai
Chemotransfiguration Project
Japan Science and Technology Corporation (Japan)
Supporting information for this article is available on the WWW under
http://www.angewandte.com or from the author.
1924
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
1433-7851/02/4111-1924 $ 20.00+.50/0
Angew. Chem. Int. Ed. 2002, 41, No. 11