Cation-specific interactions with carboxylate in amino acid and acetate aqueous solutions: X-ray absorption and ab initio calculations.
ABSTRACT Relative interaction strengths between cations (X = Li (+), Na (+), K (+), NH 4 (+)) and anionic carboxylate groups of acetate and glycine in aqueous solution are determined. These model systems mimic ion pairing of biologically relevant cations with negatively charged groups at protein surfaces. With oxygen 1s X-ray absorption spectroscopy, we can distinguish between spectral contributions from H 2O and carboxylate, which allows us to probe the electronic structure changes of the atomic site of the carboxylate group being closest to the countercation. From the intensity variations of the COO (-) aq O 1s X-ray absorption peak, which quantitatively correlate with the change in the local partial density of states from the carboxylic site, interactions are found to decrease in the sequence Na (+) > Li (+) > K (+) > NH 4 (+). This ordering, as well as the observed bidental nature of the -COO (-) aq and X (+) aq interaction, is supported by combined ab initio and molecular dynamics calculations.
Cation-specific interactions with carboxylate in amino acid and acetate
aqueous solutions: X-ray absorption and ab initio calculations
Emad F. Aziz,* Niklas Ottosson†, Stefan Eisebitt, and Wolfgang Eberhardt
BESSY m.b.H., Albert-Einstein-Strasse 15, 12489 Berlin, Germany,
†also Department of Physics, Uppsala University, Box 530, SE-751 21 Uppsala, Sweden
Barbara Jagoda-Cwiklik#, Robert Vácha, and Pavel Jungwirth*
Institute of Organic Chemistry and Biochemistry, Academy of Sciences of the Czech Republic,
and Center for Biomolecules and Complex Molecular Systems,
Flemingovo nám. 2, 16610 Prague 6, Czech Republic; #also Fritz Haber Institute for Molecular
Dynamics, Hebrew University, Jerusalem, Israel 91904
Max-Born-Institut für Nichtlineare Optik und Kurzzeitspektroskopie, Max-Born-Strasse 2A,
12489 Berlin, Germany
authors: email@example.com (E.F.A.), firstname.lastname@example.org (P.J.), and
Relative interaction strengths between cations (X = Li+, Na+, K+, NH4+) and anionic
carboxylate groups of acetate and glycine in aqueous solution are determined. These
model systems mimic ion pairing of biologically relevant cations with negatively charged
groups at protein surfaces. With oxygen 1s X-ray absorption spectroscopy we can
distinguish between spectral contributions from H2O and carboxylate, which allows to
probe the electronic structure changes of the atomic site of the carboxylic group being
closest to the counter cation. From the intensity variations of the COO-aq O 1s X-ray
absorption peak, which quantitatively correlate with the change in the local partial
density of states from the carboxylic site, interactions are found to decrease in the
sequence Na+ > Li+ > K+ > NH4+. This ordering, as well as the observed bidental nature of
the –COO-aq and X+aq interaction is supported by combined ab initio and molecular
Ion-selective interactions play an important role in many chemical, environmental,
and biological processes occurring in aqueous solution. The cationic interaction with the
protein’s carboxylate groups is of special interest due to its effect on protein association
and enzymatic activity 1, yet the details and consequences of ion pairing on the molecular
level are far from being fully understood. A useful macroscopic measure, though, in
characterizing ion specific interactions is the hydration energy. It has been established
empirically that the tendency for contact ion-pair formation correlates with the match
between the hydration enthalpies, i.e., in the present case the cation and anionic
carboxylate group.2-4 For instance, the hydration enthalpy of sodium matches those of
major intracellular anions or anionic groups, such as carboxylate, better than potassium.
Also, the effects ions have on water are specific, although the existence of long-range
solvation effects (structure-making/breaking) is currently debated 5,6.
Observations of ion-specific effects on proteins date back as early as 1888, when
Hofmeister recognized 7 that inorganic salts can be ranked by their ability to 'salt out' hen
egg white protein in aqueous solution. Hofmeister series have since been observed for
numerous phenomena in aqueous ion solutions, including their interfaces.8 It applies to
many different proteins as well as to polymers.9,10 In a 1985 review on the subject,11 no
less than 38 macroscopic effects following the Hofmeister series of inorganic solutes
were identified in aqueous solutions indicating its universal importance. Understanding
the microscopic origin of this remarkable phenomenon, which is at the heart of aqueous
chemistry, calls for a combined experimental and theoretical effort.
On the theoretical side, a recent combined molecular dynamics and quantum
chemical study has revealed that sodium interacts more strongly with protein surfaces
than potassium,12 and that the ion-specific interaction is mediated by weak ion pairs
(COO-:X+)aq. This finding and interpretation thereof is awaiting firm experimental
confirmation. To that end spectroscopic measurements with high sensitivity to the local
electronic structure, applicable to the aqueous phase, are desirable. Among the most
powerful techniques are electron spectroscopy techniques, ideally in conjunction with
high-intensity and tunable synchrotron radiation from third-generation synchrotron light
facilities. Such experiments, including X-ray absorption (XAS) and emission (XES), or
photoelectron (PES) spectroscopy became feasible only recently for aqueous solutions.13-
17 XAS, used in this work, is highly sensitive to the empty valence states of a given
atomic site, and hence the technique is most suitable for probing orbital changes mediated
by ion-ion interactions. This was previously exploited for the study of ion pairing in salt
18 and amino acid aqueous solutions.19 Here, we report oxygen K-edge XAS
measurements from aqueous solutions of X-acetate (X = Li+, Na+, K+, NH4+), and of
glycine aqueous solutions, with additions of either NaCl or KCl. Recently, a XA study
focused on the carbon signal of the carboxylate in the presence of alkali cations in water.
20 The O1s XA transition investigated here, as opposed to C1s XAS 20, directly probes
the atomic site of the –COO-aq group closest to the cation, and has thus superior
sensitivity to differences in counter-ion interactions. Another important advantage of the
present study is that the bonding geometry can be determined, which is an information
not accessible otherwise. Additionally, certain differences in the O1s XA spectra of neat
water and solutions can be assigned to the distortion of the water network. Both acetate
and glycine solutions were studied as to assure that even the small acetate is a sufficiently
good model system for ion-ion interactions at the protein surface in water.
XAS measurements at the oxygen K-edge of acetate and glycine/salt aqueous
solutions were performed at the U41-PGM undulator beamline, BESSY, Berlin. The
experimental setup has been described in detail previously.21 Briefly, the aqueous
solution is circulated (1L/min) within a stainless steel closed tubing system, inside an
UHV chamber, as to warrant continuous renewal of the irradiated liquid sample. In the
interaction region the X-ray radiation hits the sample flowing behind a 150 nm thick
Si3N4 membrane. X-ray absorption is recorded by total fluorescence yield (FY)
measurements using a 5x5 mm2 GaAsP photodiode. Due to the long attenuation lengths
of X-rays, on the order of a few microns, the method is primarily bulk sensitive. The
setup is readily applicable to more complex molecules than studied here, and for FY
measurements the use of an oxygen-free membrane is equally well suited as a free-
vacuum jet experimental setup. However, when detecting the total electron yield (TEY) a
window-less setup is required 20,22; TEY, moreover preferentially probes the solution
interface, determined by the electron mean free path in aqueous solutions.
The aqueous solutions were prepared freshly before each measurement, and highly
demineralized water, and salts of the highest purity commercially available (Sigma
Aldrich) were used. The pH of the solution was measured within ±0.1 accuracy (pH
meter 766, Knick), and no further pH adjustment was necessary. At the pH values of the
as-prepared aqueous solutions, pH 8.6 (1M Na-acetate), pH 8.7 (1M Li-acetate), pH 8.1
(1M K-acetate), pH 7.0 (1M NH4-acetate), pH 7.4 (1M glycine in 1M NaCl), and pH 7.4
(1M glycine in 1M KCl), in all cases >99.9% of the carboxylate groups are deprotonated,
as calculated from the law of mass action.
The combined ab initio and molecular dynamics approach is analogous to that
applied in our previous study.23 First, MD simulations of each of the investigated ion
pairs in water (800 SPCE water molecules in a periodic cubic cell) were performed
employing a non-polarizable forcefield.24,25 Each of the systems contained 800 water
molecules in a cubic periodic box with up to six cation/anion pairs. After nanosecond
equilibration, several ns of production runs were performed at 300 K and 1 atm with a 2
fs time step with the program Gromacs 188.8.131.52 Cation-anion radial distribution functions
were then extracted from the simulations.
In the next step, calculations were carried out using a polarizable continuum solvent
model. Geometries of the contact ion pairs were obtained from gas phase ab initio
optimizations, except that in order to get relevant aqueous phase anion-cation distances
we took these values from the previous MD simulations. Namely, for each ion pair it
corresponded to the position of the principal cation-anion oxygen peak of the radial
distribution function. The geometries of all investigated contact ion pairs (highlighting
the employed cation–anion oxygen distances) are depicted in Figure 1. Note the
symmetric bidentate structure of each of the alkali cation-acetate pairs. For ammonium
cation, the optimal structure of the ion pair is slightly asymmentric, however, with a
negligible (below 0.5 kcal/mol) barrier toward symmetrization.
In each case, the free energy of ion pairing was evaluated as the difference between
the energy of the solvated contact ion pair and the energies of the separately solvated
cation and anion in water. Ab initio calculations were performed at the second-order
Møller-Plesset perturbation theory level (MP2), employing the aug-cc-pVTZ basis sets
for acetate with additional core-valence basis functions (cc-pCVTZ) added for C and O
and the cc-pVDZ set for the cations.27 Water was described within a polarizable
continuum solvent using the COSMO model.28,29 All COSMO parameters were taken as
the default ones except for the ionic radius of sodium, which was reduced by 1.3% to
match exactly the experimental difference between hydration free energies of Na+ and K+
amounting to 17.5 kcal/mol.30 The present ab initio calculations were performed using
the Gaussian03 program package.31
3. Results and discussion
Figure 2 contrasts the oxygen K-edge X-ray absorption (XA) spectra of a series of
1M X-acetate aqueous solutions (X = Li+, Na+, K+, NH4+) as well as of pure liquid water,
and Figure 3 shows the analogous data for 1M glycine in 1M NaCl, and 1M glycine in
1M KCl aqueous solutions. All spectra were measured under the same experimental
conditions, and slight changes in the photon flux were accounted for by normalizing the
spectral intensities. This leads to identical intensities in the (background) region >545 eV
photon energy, i.e., sufficiently far away from the absorption edges. The O 1s XA spectra
of the solutions are dominated by the characteristic water features. These are the pre-edge
at 535 eV, the main- and post- edges at 538 eV and 540 eV, respectively, in agreement
with the reported values. 32,33 All XA solution-spectra of Figure 2 almost fully overlap
with the neat water spectrum, the main differences being the occurrence of a new peak
(labeled A) at 532.8 eV photon energy, with a constant width of 1.2 eV (fwhm) for each
X, and some systematic intensity changes in the main- and post-edge region. Peak A is a
sole spectral signature of the –COO-aq group, and arises from the promotion of a O 1s
core-level electron to the lowest unoccupied molecular orbital (LUMO), of π* character.
The peak position and width of A is the same for all acetate solutions and also for the
glycine solutions. Intensities of A, however, change considerably. For acetate solutions
(Figure 2) intensities of A decrease in the sequence Na+ > Li+ > K+ > NH4+, and for the
two glycine solutions (Figure 3), A is more intense for NaCl than for KCl.
The observed A-peak intensity changes (Figures 2, 3) can be attributed to the change
of the local density of unoccupied states at the oxygen site of –COO-aq, induced by
interaction with the cations. Since the measurements were performed at identical
concentrations (1M), and also the stoechiometric anion-to-cation ratios are always the
same, the observed intensity changes quantitatively correlate with the strength of ion
pairing. Intensities of the acetate-specific O1s absorption signal are hence proportional to
the total number of empty states of p symmetry in the integration interval. Higher A-peak
intensities result from electron withdrawal from the carboxyl group by the cations, which
is directly connected with an increase of the empty p DOS from the carboxyl group. The
effect scales with the strength of the COO-aq to cation(aq) interaction. A similar
observation was recently reported for ion pairing in alkali-halide salt aqueous solutions
between the Na+ and OH-, 18 where also different pairing situations, contact pairs vs
shared solvent, have been considered explicitly. Note that electron correlation
contributions to peak A, beyond a density of states approximation, can be neglected
because the associated redistribution of spectral intensity occurs typically within 10 eV of
the absorption edge.34-36
Although the solute concentration in the present study is not higher than 1 M
(compared to 55 M of water) changes of the water hydrogen-bonding network can be
identified, which is manifested by intensity variations in the main/post-edge region of the
sodium and potassium acetate solution spectra (Figure 2). Contrary to the A-peak
intensity changes, increase of the main/post-edges intensity correlates with the reverse
order of X, i.e., Na+ < Li+ < K+ < NH4+. Consistent with this behaviour, higher main/post-
edge intensities are observed for glycine/KCl than for gylcine/NaCl aqueous solutions
(Figure 3). A similar intensity increase of the main edge was observed when adding NaCl
to the water 37, and has been attributed to strong perturbation of the electronic structure of
water molecules within the anions hydration shell. An experimental detail, deserving
further consideration, is the observation that the A-peak intensity increases from K+ to
Na+ is slightly larger for acetate than for glycine, indicating a somewhat different
chemical environment of the carboxylate group within these two species.
In discussing the XAS intensity changes of Figures 2 and 3 in terms of –COO--ion
interaction strengths we have not yet explicitly addressed the molecular interaction
geometry. The important issue is whether or not the cation binds equally to the two
oxygen atoms of the –COO-aq group. Our results show that the width of peak A is
independent of the cation used. Hence, for each cation studied the interaction geometry is
essentially of the same bidental nature, in agreement with the calculated structures
presented in Figure 1. All alkali cations are observed to take middle positions between
the two carboxylate oxygens. In case of calculations of the NH4+, there is a tendency of a
weakly preferred interaction with one of the oxygen of the carboxylate. However, this
preference for an asymmetric structure is very weak (below 0.5 kcal/mol) and will be
smeared out by vibrational motions, therefore, it is not observed in the experiment. Note
also that in contrast to cations studied here, in –COOHaq the two oxygens are clearly
distinguishable, giving two distinct peaks in the photoelectron spectrum.38
We now compare the observed ion pairing ordering with results from MD
calculation, and we also discuss the reason why Li+ departs from the Hofmeister series.
Computational results of the relative strength (with respect to Na+) of cation pairing with
acetate in water are summarized in Table 1. We see that sodium forms the most stable ion
pair with aqueous acetate, followed by lithium which is only marginally less stable. The
tendency for ion pairing with potassium and particularly with the ammonium cation is
weaker, the former result being observed and discussed already in our previous studies.
12,23 The fact that lithium binds slightly less strongly than sodium to acetate is in accord
with the empirical law of matching water affinities.39 It states that ions prefer to pair with
counter-ions or ionic groups which have comparable hydration enthalpies which can be
translated in a simple Born solvation picture to surface charge densities.40 From this point
of view Na+ matches most closely the COO- group, followed by Li+, K+, and NH4+.
By means of X-ray absorption and combined ab initio and molecular dynamics
simulations we have determined the ordering of a series of cations (X = Li+, Na+, K+,
NH4+) in terms of the strenght of interaction with anionic carboxylate groups of acetate
and glycine in aqueous solutions. The strength of ion pairing with the COO-aq decreases
in the sequence Na+ > Li+ > K+ > NH4+ both in the experiment and calculation. The
cations thus follow, with the exception of lithium, the Hofmeister series. The observed
cationic ordering can be qualitatively rationalized in terms of the empirical law of
matching water affinities within which Na+ matches best the hydration enthalpy of –
COO-aq, followed by Li+, K+, and NH4+. The structure of all the alkali cation –
carboxylate ion pairs is bidentate. Calculations present the ammoinum – carboxylate ion
pair as slightly asymmetric with a marginal barrier toward symmetrization, which
explains why asymmetric structure is not observed in the experiment. The present
systems mimic the process of ion pairing between cations of biological relevance with
charged basic groups at protein surfaces.
Figure 1: Geometries of contact ion pairs of a) lithium-acetate, b) sodium-acetate, c)
potassium-acetate, and d) ammonium acetate.
Figure 2: Oxygen 1s XA spectra for different X-acetate (X = Li+, Na+, K+, NH4+)
solutions. Enlargement of the region between 531 - 534.5 eV presented on the top left
and correlated with the affinity.
Figure 3: Oxygen 1s XA spectra of 1M solution of glycine in 1M NaCl, and 1M glycine
in 1M KCl. Enlargement of the region between 531 - 534.5 eV presented on the top left
and correlated with the affinity.
Table 1: Free energy change upon replacing sodium with lithium, potassium, or
ammonium in a contact ion pair with acetate in water.
of the United States of America 2007, 104, 11167.
(7) Lewith, S. Arch Exp Pathol Pharmakol 1888, 23, 1.
(8) Cacace, M. G.; Landau, E. M.; Ramsden, J. J. Quarterly Reviews of
Biophysics 1997, 30, 241.
(9) Baldwin, R. L. Biophysical Journal 1996, 71, 2056.
(10) Inouye, K.; Kuzuya, K.; Tonomura, B. Journal of Biochemistry 1998, 123,
(11) Collins, K. D.; Washabaugh, M. W. Quarterly Reviews of Biophysics
1985, 18, 323.
(12) Vrbka, L.; Vondrasek, J.; Jagoda-Cwiklik, B.; Vacha, R.; Jungwirth, P.
Proceedings of the National Academy of Sciences of the United States of America 2006,
(13) Aziz, E. F.; Eisebitt, S.; de Groot, F.; Chiou, J.; Dong, C.; Guo, J.;
Eberhardt, W. J. Phys. Chem. B 2007, 111, 4440.
(14) Aziz, E. F.; Zimina, A.; Freiwald, M.; Eisebitt, S.; Eberhardt, W. Journal
of Chemical Physics 2006, 124, 114502.
(15) Nolting, D.; Aziz, E. F.; Ottosson, N.; Faubel, M.; Hertel, I. V.; Winter, B.
Journal of the American Chemical Society 2007, 129, 14068.
(16) Guo, J. H.; Augustsson, A.; Kashtanov, S.; Spangberg, D.; Nordgren, J.;
Hermansson, K.; Luo, Y.; Augustsson, A. Journal of Electron Spectroscopy and Related
Phenomena 2005, 144-147, 287.
(17) Guo, J. H.; Luo, Y.; Augustsson, A.; Rubensson, J. E.; Sathe, C.; Agren,
H.; Siegbahn, H.; Nordgren, J. Physical Review Letters 2002, 89, 137402.
(18) Aziz, E. F.; Eisebitt, S.; Eberhardt, W.; Cwiklik, L.; Jungwirth, P. J. Phys.
Chem. B 2008.
(19) Aziz, E. F.; Eberhardt, W.; Eisebitt, S. Z. Phys. Chem. 2008, Accepted.
(20) Uejio, J. S.; Schwartz, C. P.; Duffin, A. M.; Drisdell, W. S.; Cohen, R. C.;
Saykally, R. J. Proceedings of the National Academy of Sciences of the United States of
America 2008, 105, 6809.
(21) Aziz, E. F.; Freiwald, M.; Eisebitt, S.; Eberhardt, W. Physical Review B
2006, 73, 75120.
(22) Winter, B.; Faubel, M. Chemical Reviews 2006, 106, 1176.
(23) Jagoda-Cwiklik, B.; Vacha, R.; Lund, M.; Srebro, M.; Jungwirth, P.
Journal of Physical Chemistry B 2007, 111, 14077.
(24) Berendsen, H. J. C.; Grigera, J. R.; Straatsma, T. P. Journal of Physical
Chemistry 1987, 91, 6269.
Waigh, T. A. Applied Biophysics; WILEY & SONS, 2007.
Collins, K. D. Biophysical Journal 1997, 72, 65.
Collins, K. D. Methods 2004, 34, 300.
Collins, K. D. Biophysical Chemistry 2006, 119, 271.
Omta, A. W.; Kropman, M. F.; Woutersen, S.; Bakker, H. J. Science 2003,
(6) Laage, D.; Hynes, J. T. Proceedings of the National Academy of Sciences
2001, 7, 306.
Transactions 2 1993, 799.
Physics 2000, 2, 97.
A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A.; Stratmann, R. E.; Burant, J.
C.; Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.;
Tomasi, J.; Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.;
Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.;
Salvador, P.; Dannenberg, J. J.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Baboul, A. G.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith,
T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.;
Johnson, B.; Chen, W.; Wong, M. W.; Andres, J. L.; Gonzalez, C.; Head-Gordon, M.;
Replogle, E. S.; Pople, J. A. Gaussian 03; Windows 03 ed. Pittsburgh PA, 2003.
(32) Wernet, P.; Nordlund, D.; Bergmann, U.; Cavalleri, M.; Odelius, M.;
Ogasawara, H.; Naslund, L. A.; Hirsch, T. K.; Ojamae, L.; Glatzel, P.; Pettersson, L. G.
M.; Nilsson, A. Science 2004, 304, 995.
(33) Smith, J. D.; Cappa, C. D.; Wilson, K. R.; Messer, B. M.; Cohen, R. C.;
Saykally, R. J. Science 2004, 306, 851.
(34) Hawlicka, E.; Swiatla-Wojcik, D. Journal of Physical Chemistry A 2002,
(35) Zaanen, J.; Sawatzky, G. A.; Allen, J. W. Physical Review Letters 1985,
(36) Zaanen, J.; Sawatzky, G. A.; Fink, J.; Speier, W.; Fuggle, J. C. Physical
Review B 1985, 32, 4905.
(37) Cappa, C. D.; Smith, J. D.; Wilson, K. R.; Messer, B. M.; Gilles, M. K.;
Cohen, R. C.; Saykally, R. J. J. Phys. Chem. B 2005, 109, 7046.
(38) Jungwirth, P.; Winter, B. Annual Review of Physical Chemistry 2008, 59,
(39) Collins, K. D.; Neilson, G. W.; Enderby, J. E. Biophysical Chemistry
2007, 128, 95.
(40) Born, M. Zeitschrift fur Physik 1920, 1, 45.
(25) Jungwirth, P.; Finlayson-Pitts, B. J.; Tobias, D. J. Chemical Reviews 2006,
(26) Lindahl, E.; Hess, B.; van der Spoel, D. Journal of Molecular Modeling
Dunning, T. H. Journal of Chemical Physics 1989, 90, 1007.
Klamt, A.; Schuurmann, G. Journal of the Chemical Society-Perkin
Barone, V.; Cossi, M. Journal of Physical Chemistry A 1998, 102, 1995.
Schmid, R.; Miah, A. M.; Sapunov, V. N. Physical Chemistry Chemical
Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M.
∆G (sodium → other cation) [kcal/mol]
lithium + 0.3
potassium + 2.5
ammonium + 3.4