Theor Chem Acc (2013) 132:1383

DOI 10.1007/s00214-013-1383-3

REGULAR ARTICLE

DFT insight into o-semiquinone radicals and Ca2+ ion interaction:

structure, g tensor, and stability
Maciej Witwicki • Julia Jezierska

Received: 4 April 2013 / Accepted: 12 July 2013

Ó The Author(s) 2013. This article is published with open access at Springerlink.com

Abstract Density functional theory methods were o-semiquinones are known to possess chelating ability
employed to elucidate the interactions between calcium toward metal ions [5–7], which is particularly important for
ions and various o-semiquinone radicals mimicking the the activation of electron transfer through interaction with
interactions occurring in biochemical systems. Predicted cations acting as Lewis acids [5, 8].
changes in the molecular and electronic structures of the Electron paramagnetic resonance (EPR) spectroscopy
radicals on Ca2? coordination were correlated with the has established its important position in investigation of
changes of g tensor and compared with those exerted by semiquinone radicals in laboratory conditions and in their
Mg2? ions (reported by us previously). In order to broaden natural surroundings [8–14]. Also, the formation of a
the insight into the differences between the Mg2? and Ca2? complex between diamagnetic metal ions and semiquinone
complexes, their relative stability was estimated on the radicals can be efficiently investigated using the EPR
basis of theoretically predicted Gibbs energies for the techniques since the g and A tensors are sensitive to the
process of the complex formation. radical–metal ions interaction [5, 6, 15–18].
Recent years have witnessed an increasing interest in the
Keywords EPR  ESR  Semiquinone radicals  application of theoretical methods to chemical and bio-
Paramagnetic complexes  Radical ligands chemical systems [19–21]. One of the most significant
quantum chemical methods employed in this type of
studies are the ones based on density functional theory
1 Introduction (DFT) since these methods can be applied to (nearly) real
chemical systems. Organic radicals (including semiqui-
Organic radical ions play increasingly important roles in nones) have been the subject of successful DFT studies
modern biochemistry and material science [1, 2]. Semiq- covering diverse environmental factors significantly
uinones are typical organic radical anions being the inter- affecting the radical various properties as well as their EPR
mediate form in the redox equilibrium between quinones tensors (g and A) [14, 22–46]. However, far too little
and hydroquinones. These radicals are present in all life attention has been paid to the interaction between the
forms as they act as electron-transfer agents in the mito- radicals and diamagnetic metal ions. Previously, we
chondrial respiratory chain and in the reaction centers reported the results of a detailed DFT study of the influence
of bacterial and plant photosynthesis [3, 4]. Moreover, of Mg2? on the o-semiquinone ligands in the formed
complexes [47]. In the present work, we aimed to charac-
terize theoretically the effects of the Ca2? ion on the
Electronic supplementary material The online version of this
article (doi:10.1007/s00214-013-1383-3) contains supplementary electronic structure of the o-semiquinone radicals and on
material, which is available to authorized users. the molecular geometries of the resulting complexes, in
correlation with both the g tensor components and the
M. Witwicki (&)  J. Jezierska
characteristics of the previously studied Mg2? complexes.
Faculty of Chemistry, Wroclaw University, 14 F. Joliot-Curie
St., 50-283 Wroclaw, Poland In order to make the comparison meaningful, exactly the
e-mail: [email protected] same theory levels and software versions were used here as

123
Page 2 of 13 Theor Chem Acc (2013) 132:1383

All the optimizations of molecular structures were car-

ried out using the Gaussian 09 [54] suite of programs
employing the popular UB3LYP hybrid functional [55–57]
and the TZVP basis set [58]. The initial geometries for the
optimizations of the Ca2? complexes with o-semiquinone
ligands were prepared using the structures determined by
X-ray crystallography for similar diamagnetic Ca2? com-
plexes. The coordination number (c.n.) of Ca2? in these
diamagnetic systems was found to be predominantly 7 [59,
60] and 8 [61, 62]. In order to make the investigation more
complete, also the complexes with c.n. = 6 (octahedral)
and c.n. = 4 (square planar and tetrahedral) were included
in the computational analysis. To complete the coordina-
tion sphere of Ca2?, the optimized structures contained six,
five, four or two acetonitrile molecules in addition to the
chelating semiquinone ligand. The effect of solvation on
Fig. 1 Schematic structures of anionic semiquinone radical ligands geometry was covered by employing the integral equation
derived from o-quinone (sq), 9,10-phenanthrenequinone (psq), and formalism variant (IEFPCM) of Tomasi’s PCM method
1,10-phenanthroline-5,6-dione (ptsq). In addition, the principal
directions of g tensor are shown
[63–65]. No symmetry constraints were imposed on the
optimization procedures. All the open-shell computations
were carried out using the spin-unrestricted formalism. The
before. Moreover, to provide a broader insight into the geometries of the investigated species did not reveal
differences between Mg2? and Ca2? complexes, their rel- imaginary frequencies. The initial square planar structures
ative stability was estimated. This aspect seems to be underwent convergence to tetrahedral. This result was
highly interesting as ubiquinone (Coenzyme Q10) and independent of the o-semiquinone ligand and PCM
2-palmitoylhydroquinone have been shown to transport inclusion.
various metal ions of biological importance (Mg2?and The ORCA electronic structure package [66] was used
Ca2?) [48, 49] or of high toxicity (Sr2? and Ba2?) [50] to calculate the g tensors and to perform the Löwdin
through membranes. Although the reduced forms of the population analysis. In these undertakings, the hybrid
p-quinones correspond directly to the p-semiquinones, the (UB3LYP [55–57] and UPBE0 [67, 68]) and generalized
mechanisms proposed for Q10 assume its transformation gradient approximations functionals (UBP86 [69, 70],
into polyhydroxy forms and the metal ion coordination to UPBE [68], and UOLYP [55, 71]), together with the
the oxygens of the ortho hydroxy groups [49]. Therefore, TZVP basis set [58] were employed. The conductor-like
the theoretical approach to the relative stability of the screening model (COSMO) [72, 73] was the continuum
model complexes with Mg2? and Ca2? ions is expected to solvation model used in the computations. The g tensors
reveal which of the ions can be preferred in the transport were computed using Neese’s CPKS method [74] com-
across membranes. bined with an accurate mean field approximation [RI-
In this study, semiquinone radicals with different aro- SOMF(1X)] [75] to the Breit–Pauli spin–orbit coupling
maticity, derived from o-quinone (sq), 9,10-phenan- operator [76, 77]. In this work, all the computed com-
threnequinone (psq), and 1,10-phenanthroline-5,6-dione ponents of the g tensors are given as g-shifts (Dgij) in
(ptsq) (see Fig. 1), were chosen as the model ligands parts per million (ppm):
coordinating Ca2? ions.
Dgij ¼ ðgij  ge Þ  106 ppm; ð1Þ
where ij = xx, yy, zz, and ge = 2.002319 is the free elec-
2 Computational details tron g value.
In order to examine the relative stability of the radical
In the calculations, acetonitrile was selected as a solvent complexes, Gibbs energies at T = 298.15 K were calcu-
because it was mainly used in the experimental studies [5, lated at the (U)B3LYP/TZVP theory level for the process
8, 51]. Acetonitrile was included in the calculations by of radical complex ([ML(c.n.-2)R]?) formation from a
using the continuum solvation models (PCM and cation complex with acetonitrile ([MLc.n.]2?) and a radical
COSMO), which have been shown to provide an accurate ligand (R-):
and efficient approximation to the aprotic solvent effects
½MLc:n: 2þ þ R ! ½MLðc:n:2Þ Rþ þ 2L; ð2Þ
[22, 28, 52, 53].

123
Theor Chem Acc (2013) 132:1383 Page 3 of 13

molecule; (2) an Arabic numeral following a short name of

an o-semiquinone indicates the c.n. of Ca2?; and (3) the
letter A in the superscript indicates that the continuum
solvation model was included in the calculation. The above
rules were also applied to the Mg2? complexes taken from
our previous work [47], but an asterisk * was used for the
latter to distinguish them from the Ca2? complexes.

3.1 Molecular structure and spin density

From the EPR point of view, it is interesting to see how the

formation of the Ca2? complex affects the spin density of
Fig. 2 Thermodynamic cycle used in the calculations of the Gibbs the semiquinone ligands. The changes are illustrated in this
energies for the complexes formation in acetonitrile article with the Löwdin spin populations (see Table 1).
Since the spin density of semiquinones depends on the
length of the bonds between hydroxyl oxygens and carbon
where L = CH3CN, M = Mg or Ca, and R- = sq, psq or atoms (RC–O) [14, 22, 34, 52], the impact of Ca2? com-
ptsq (all the structures of the Mg2? complexes were taken plexation on RC–O should be first taken under investigation.
from our former work [47]). The Gibbs energies were Regardless of the c.n., the RC–O distances predicted for the
calculated for the reaction taking place in the gas phase Ca2? complex with the o-semiquinones are significantly
(given in Eq. 2) and in acetonitrile using the thermody- larger than the ones for the uncomplexed radicals, e.g.,
namic cycle shown in Fig. 2. The solvation energies were RC–O increased from 1.265 Å for sqA to 1.271 Å for sq8A.
obtained from single-point PCM computations. The struc- The increase in RC–O was less significant for the complexes
tures of the acetonitrile molecules (L) and of the cations with a higher c.n. The RC–O elongation leads to a more
coordinated to the acetonitrile molecules for various c.n. profound spin population on the ipso carbons, as compared
([MLc.n.]2?) were optimized as described above, but the with the uncomplexed radical. At the same time, the spin
restricted formalism (RB3LYP/TZVP) was used for the population on the hydroxyl oxygen atoms becomes sig-
closed-shell species. The DFT methods have been proved nificantly lower. Importantly, the larger RC–O values (and
useful in the prediction of the Ca2? and Mg2? affinity for slightly more significant changes in spin populations) were
nonradical ligands [78–80]. observed in the case of Mg2? coordination [47].
An interesting problem seems to be the concentration of
spin density on the Ca atom. In each of the investigated
3 Results and discussion complexes, the spin population on Ca was found to be
barely noticeable. To ensure that this is not due to the basis
Figure 3 shows optimized geometries of the Ca2? com- set effect, the computations with other basis sets (SVP,
plexes with sq. Similar figures for psq and ptsq are given TZVPPP, QZVP) were conducted. Regardless of the basis
in Supplementary Materials (Figure S1). These figures also set used, the spin population remained insignificant (see
illustrate the rules used for structure naming: (1) The short Table S2 in Supplementary Materials). Thus, Ca2? com-
names of o-semiquinones given in Fig. 1 define the radical plexation generates a change of spin populations similar to

Fig. 3 Optimized structures of the Ca2? complexes with sq; the atoms numbering shown for sq4 is valid for all the structures discussed in the
paper

123
Page 4 of 13 Theor Chem Acc (2013) 132:1383

Table 1 Lengths of the C–O and O–M (M = Ca or Mg) bonds (in Å) as well as Löwdin spin populations (q); all computed at the UB3LYP/
TZVP theory level
RC1–O1 RC2–O2 RO1–M RO2–M qC1 qC2 qO1 qO2 qM

sq 1.252 1.252 n/a n/a 0.060 0.060 0.253 0.253 n/a

A
sq 1.265 1.265 n/a n/a 0.104 0.104 0.243 0.243 n/a
Ca2? complexes
sq4 1.285 1.285 2.320 2.320 0.183 0.183 0.196 0.188 0.008
sq4A 1.274 1.273 2.489 2.486 0.149 0.149 0.212 0.213 0.003
sq6 1.277 1.277 2.381 2.381 0.149 0.149 0.215 0.215 0.003
A
sq6 1.273 1.273 2.498 2.496 0.144 0.144 0.217 0.217 0.002
sq7 1.274 1.274 2.441 2.441 0.139 0.139 0.219 0.219 0.003
sq7A 1.271 1.271 2.488 2.487 0.135 0.135 0.222 0.221 0.002
sq8 1.271 1.271 2.465 2.465 0.128 0.129 0.226 0.226 0.000
sq8A 1.271 1.270 2.551 2.551 0.127 0.126 0.228 0.228 0.000
2?
Mg complexes
sq4*a 1.291 1.291 1.968 1.968 0.182 0.182 0.188 0.188 0.005
sq4A*a 1.288 1.288 2.002 2.002 0.174 0.175 0.194 0.194 0.003
sq6*a 1.280 1.280 2.053 2.053 0.152 0.152 0.214 0.214 0.000
sq6A*a 1.280 1.280 2.080 2.078 0.151 0.151 0.215 0.215 -0.001
Values for the complexes of psq and ptsq are given in Supplementary Materials (Table S1)
a
The values taken from [47]

the one observed by us previously for the Mg2? com- Dg tensors is not to be significantly dependent on the
plexation [47]. The impact of the two cations on spin functional.
populations and RC–O is qualitatively similar to the changes Figure 1 shows principal axes of the g tensor whose
induced by the solvent [14, 22, 31, 34, 52], but it is directions remained unaffected by Ca2? interaction. This
quantitatively far greater. result was independent of the examined model complex
It is essential to compare the lengths of the metal– and the used methodology (various functionals and/or
oxygen bonds for Ca2? and Mg2? complexes. The RO–Ca continuum solvent model inclusion).
values are predicted to increase with the c.n. from 2.489 Å The Dgzz values were found to be far less sensitive to the
for sq4A to 2.498 Å for sq6A and to 2.551 Å for sq8A complexation (see Table 2) than the perpendicular com-
(Table 1), similarly to the change in RO–Mg from 2.002 Å ponents (Dgxx and Dgyy). The Dgzz magnitude tends to rise
for sq4A* (tetrahedral) to 2.080 Å for sq6A* (octahedral). when the radical interacts with the cation. A good example
It is apparent that the RO–Ca values are significantly higher of this is the semiquinone derived from o-quinone. Dgzz
than the RO–Mg values. increases from -125 ppm for sqA to -22 ppm for sq4A,
11 ppm for sq6A, 27 ppm for sq7A, and 44 ppm for sq7A
3.2 g Tensor (employing UB3LYP). It is also clear from this data that
Dgzz becomes more positive for the complexes with the
It is known that the DFT methods might misestimate the larger c.n.
covalent character of metal–ligand bonds [81–83]. As mentioned above, the spin populations on the
Although this is more frequently the case of transition hydroxyl oxygens markedly decrease upon the o-semiqui-
metal coordination compounds, we decided to calculate the none interaction with Ca2?. According to Stone’s qualita-
Dg tensors with a vast array of functionals (UBP86, UPBE, tive model [84, 85], such spin redistribution ought to
UOLYP, UB3LYP, UPBE0). The comparison of the reduce the Dgxx and Dgyy values. Our DFT predictions of
g tensors presented in Table 2 shows that all the methods, the Dg tensor prove the correctness of this hypothesis.
both GGA and hybrid approximations, yield similar out- Regardless of the c.n., the lowering of the perpendicular
comes. Therefore, one can expect that the variation of the components is substantial. In general, the effects of Ca2?
covalency on the functional is small. The reason for this is and Mg2? complex formation on the Dg tensor are similar,
the fact that the interaction between the Ca2? ion and the albeit the interaction between Mg2? and the o-semiqui-
o-semiquinones is mainly electrostatic in nature; therefore, nones, as revealed by the shorter O–Mg bonds and
a minor contribution of the covalency to the computed DG298 (to be discussed below), is clearly stronger

123
Theor Chem Acc (2013) 132:1383 Page 5 of 13

Table 2 Dg tensors (in ppm) calculated using the UB3LYP, UBP86, UPBE, UPBE0, and UOLYP functionals
UB3LYP/TZVP UBP86/TZVP UPBE0/TZVP
Dgxx Dgyy Dgzz Dgiso Dgxx Dgyy Dgzz Dgiso Dgxx Dgyy Dgzz Dgiso

Ca2? complexes
sq4 3610 2249 -24 1945 3620 2455 -15 2020 3591 2480 -12 2020
sq4A 3878 2733 -22 2196 3883 2553 -25 2137 3888 2762 -19 2211
psq4 2903 2215 -75 1681 2784 1993 -79 1566 2897 2249 -72 1691
psq4A 3345 2529 -87 1929 3227 2305 -82 1817 3369 2569 -85 1951
ptsq4 2930 2257 -89 1699 2860 2064 -93 1610 2904 2250 -89 1688
ptsq4A 3436 2621 -102 1985 3344 2403 -99 1883 3444 2634 -101 1992
sq6 4139 2811 9 2320 4071 2583 -4 2216 4116 2821 14 2317
A
sq6 4077 2911 11 2333 4052 2718 7 2259 4082 2927 15 2341
psq6 3474 2554 -70 1986 3294 2291 -78 1836 3463 2573 -66 1990
psq6A 3561 2660 -69 2050 3405 2421 -66 1920 3577 2685 -67 2065
ptsq6 3489 2570 -83 1992 3352 2333 -90 1865 3462 2553 -80 1978
ptsq6A 3622 2720 -84 2086 3495 2489 -83 1967 3624 2718 -84 2086
sq7 4276 2884 21 2394 4190 2645 12 2282 4264 2896 25 2395
sq7A 4220 2968 27 2405 4167 2754 24 2315 4228 2982 31 2414
psq7 3564 2584 -60 2030 3356 2309 -54 1870 3569 2606 -56 2039
psq7A 3744 2681 -56 2123 3547 2430 -50 1976 3767 2709 -54 2141
ptsq7 3583 2591 -73 2034 3418 2340 -70 1896 3571 2579 -73 2026
ptsq7A 3835 2758 -71 2174 3665 2507 -67 2035 3846 2759 -71 2178
sq8 4461 3072 28 2520 4333 2821 20 2391 4486 3053 46 2528
sq8A 4378 3134 44 2519 4296 2919 41 2419 4377 3136 34 2516
psq8 3863 2926 -53 2245 3615 2632 -61 2062 3866 2934 -48 2250
psq8A 3881 2924 -48 2252 3662 2656 -48 2090 3900 2938 -45 2264
ptsq8 3860 2922 -68 2238 3651 2647 -78 2073 3848 2897 -64 2227
ptsq8A 3949 2992 -62 2293 3753 2724 -63 2138 3955 2980 -59 2292
Exptl for Ca2? complexes with
sq in watera 1981 1981 1981
psq in acetonitrileb 1681 1681 1681
ptsq in acetonitrileb 2281 2281 2281
c
PQQ in acetonitrile 3531 2861 -199 2064 3531 2861 -199 2064 3531 2861 -199 2064
o-Semiquinones
sq 5179 4406 -111 3158 4903 4216 -105 3005 5257 4465 -112 3203
sqA 4789 3989 -125 2884 4696 3794 -118 2791 4846 4043 -127 2921
psq 4571 4163 -89 2882 4140 3781 -79 2614 4652 4262 -90 2941
psqA 4295 3841 -94 2681 4041 3513 -84 2490 4352 3935 -96 2730
ptsq 4584 4302 -95 2930 4192 3947 -87 2684 4645 4377 -96 2975
ptsqA 4328 3965 -99 2731 4103 3658 -90 2557 4372 4035 -101 2769
Exptl for
sq in watera 2281 2281 2281
psq in acetonitrileb 2481 2481 2481
b
ptsq in acetonitrile 2681 2681 2681
UPBE/TZVP UOLYP/TZVP
Dgxx Dgyy Dgzz Dgiso Dgxx Dgyy Dgzz Dgiso

Ca2? complexes
sq4 3669 2260 -20 1970 3624 2287 -11 1967
sq4A 3946 2564 -23 2162 3882 2572 -15 2147

123
Page 6 of 13 Theor Chem Acc (2013) 132:1383

Table 2 continued
UPBE/TZVP UOLYP/TZVP
Dgxx Dgyy Dgzz Dgiso Dgxx Dgyy Dgzz Dgiso

psq4 2849 1999 -74 1591 2896 2013 -73 1612

psq4A 3296 2317 -78 1845 3322 2330 -79 1858
ptsq4 2926 2084 -86 1641 2951 2116 -87 1660
ptsq4A 3412 2426 -93 1915 3421 2454 -96 1926
sq6 4145 2610 -2 2251 4081 2637 8 2242
sq6A 4135 2747 9 2297 4081 2766 17 2288
psq6 3377 2315 -75 1872 3412 2341 -75 1893
psq6A 3491 2447 -64 1958 3528 2468 -65 1977
ptsq6 3436 2369 -86 1906 3450 2412 -87 1925
ptsq6A 3580 2526 -79 2009 3598 2564 -82 2027
sq7 4262 2673 13 2316 4207 2702 23 2310
sq7A 4249 2786 26 2353 4207 2808 33 2349
psq7 3436 2328 -53 1904 3479 2357 -54 1927
psq7A 3632 2450 -48 2011 3674 2473 -51 2032
ptsq7 3497 2372 -68 1934 3520 2422 -71 1957
ptsq7A 3748 2541 -65 2075 3772 2583 -70 2095
sq8 4436 2840 31 2436 4361 2852 41 2418
sq8A 4347 2928 27 2434 4294 2931 35 2420
psq8 3699 2662 -60 2100 3695 2662 -57 2100
psq8A 3748 2688 -46 2130 3756 2695 -45 2135
ptsq8 3733 2688 -75 2115 3715 2703 -73 2115
ptsq8A 3836 2768 -60 2181 3829 2787 -61 2185
Exptl for Ca2? complexes with
sq in watera 1981 1981
psq in acetonitrileb 1681 1681
ptsq in acetonitrileb 2281 2281
PQQ in acetonitrilec 3531 2861 -199 2064 3531 2861 -199 2064
o-Semiquinones
sq 4978 4277 -102 3051 4852 4190 -96 2982
A
sq 4794 3850 -115 2843 4686 3797 -113 2790
psq 4210 3814 -75 2650 4191 3742 -80 2618
psqA 4139 3558 -78 2540 4146 3528 -86 2530
ptsq 4262 3987 -82 2722 4230 3906 -87 2683
ptsqA 4197 3706 -85 2606 4179 3664 -90 2584
Exptl for
sq in watera 2281 2281
psq in acetonitrileb 2481 2481
ptsq in acetonitrileb 2681 2681
a
Eaton [16], experimental values obtained in water; therefore, for the sq free radical, somewhat more significant overestimation of calculated
Dgiso is observed; see e.g. [14, 22, 31, 34, 52] for investigation of the hydrogen bonds effect
b
Yuasa et al. [5]
c
Pyrroloquinoline quinone (2,7,9-tricarboxypyrroloquinoline), values taken from [11]

compared to Ca2?. The stronger interaction in the case of criterion for the strength of the interaction between an
Mg2? should be expected to induced more significant o-semiquinone and metal cation.
decrease of Dgxx and Dgyy. This, however, was not Interestingly, the COSMO correction decreases Dg-
observed, suggesting that the g factor is not a sufficient shifts for the solvated semiquinone, as expected [14, 22,

123
Theor Chem Acc (2013) 132:1383 Page 7 of 13

34, 52], but increases them for the complexed semiquinone. DgRMC contributes only to the diagonal components of the
The complex can be considered as comprising of two parts: Dg tensor (s = t), DgDCst is the diamagnetic correction and
the semiquinone ligand and the cation with acetonitrile PSO
Dgst ; is the paramagnetic spin–orbit term. The values of
molecules. The COSMO correction stabilizes the latter part the three terms are given in Table 3.
of the complex, decreasing the strength of the cation– Irrespective of the interactions with the metal ion, the
semiquinone interaction and therefore decreasing the cat- predicted DgRMC and DgDC st values were found to be of
ion effect on the Dg tensor. The weakening of Ca2?– minor magnitude. Moreover, their opposite signs lead to
semiquinone interaction is clearly seen in the COSMO- the mutual cancelation of the two terms whereby the Dgxx
induced RCa–O elongation. and Dgyy components are dominated by DgPSO st . In this case,
Unfortunately, a direct comparison of the calculated an accurate approximation of the atomic contributions to
Dg tensor diagonal components with their experimental the Dg tensor can be obtained via the breakdown of DgPSO st
counterparts is limited as very few high-field experiments
into the contributions from the particular atoms. Since the
have been performed so far. Therefore, we focused on the
mean field approximation to the molecular spin–orbit
values of the Dgiso parameters. Since the calculated Dgiso
coupling operator employed in this work [RI-SOMF(1X)]
values for all the considered models are close to the
[75] takes into account the mulicenter terms (except for the
experimental ones, it is impossible to determine explicitly
exchange part), these terms have to be neglected to obtain
on the basis of Dgiso which coordination sphere is preferred
the atomic contributions. Considering that such an omis-
in real chemical systems. To answer this question, the
sion may cause significant errors [75], in Table 3, the
Gibbs free energies were calculated for the reaction given
DgPSO
st (1c) values calculated in the one-center approxima-
in Eq. 2; the results are presented below. On the other
tion are compared with the ones calculated including the
hand, a general agreement between the calculated Dg ten-
multicenter terms (DgPSO st ), revealing only a limited
sor components and ones experimentally determined for
deviation.
similar systems was expected to be a good additional way
The contributions from all the atoms for sq were cal-
of verifying the quality of the computations. The choice of
culated at the UB3LYP/TZVP theory level and are shown
pyrroloquinoline quinone (2,7,9-tricarboxypyrroloquino-
in Fig. 4. Both perpendicular components are dominated
line, PQQ), a quinone cofactor belonging to a class of
by the contributions from the oxygens. This is in agreement
dehydrogenases known as quinoproteins, seems to be the
with the previous reports for p-semiquinone [34] and the
most appropriate. PQQ is bonded to the Ca2? ion and
phenoxyl radical [35]. The contributions from carbon
exhibits an EPR spectrum with the Dgxx = 3,531,
atoms, even from these in the ipso positions, are consid-
Dgyy = 2,861, and Dgzz = -199 components [10, 11]. The
erably small. Moreover, the contributions from the differ-
magnitude of the Dg-shifts predicted by us for the model
ent carbons to Dgxx have opposite signs, which leads to
o-semiquinone complexes is similar, despite the chemical
their mutual cancelation.
differences. This fact (in combination with the good
The inclusion of the COSMO model (sqA) results in a
agreement between experimental and theoretical Dgiso)
significant decrease in the contributions from the oxygens
suggests the high accuracy of the computations.
to the Dgxx and Dgyy components (see Table 3). After the
The distribution of spin density in the o-semiquinones
attachment of Mg2? or Ca2? to sq, barely noticeable
coordinating Ca2? (and Mg2?) gives rise to an interesting
contributions of the metal atoms were predicted. Thus, the
question about the direct contribution of the Ca2? (and
observed diminution of Dgxx and Dgyy upon the complex
Mg2?) ion to the perpendicular Dg tensor components.
formation is exclusively the result of the reduced contri-
Minor spin populations on the metal cations to a certain
butions from the oxygens. Consequently, the impact of a
degree suggest that such a direct impact should be insig-
diamagnetic metal ion on the Dg tensor of the semiquinone
nificant and only strong indirect effects may be expected.
radical can be described as indirect; a metal ion does not
To investigate this problem, we performed a theoretical
bring any significant direct contribution, but causes a
analysis of the atomic contributions to Dgxx and Dgyy. In
decrease in the contributions from hydroxyl oxygen atoms.
order to make this analysis complete, the calculations were
Table 3 is quite informative in another way. As it was
also done (for the first time) for the representative Mg2?
mentioned above, Dgzz increases on complex formation,
complex investigated by us previously [47].
and this table shows that this increase is related primarily to
In one-component DFT calculations, the total Dg tensor
is given as a sum of three contributions [74, 86, 87]: the rise of DgDCzz and secondarily to the growth of Dgzz
PSO

term.
Dgst ¼ dst DgRMC þ DgDC PSO
st þ Dgst ; ð3Þ
As it was demonstrated in our former systematic study
where DgRMC is the relativistic mass correction to kinetic of Mg2? complexes with o-semiquinone ligands [47],
energy, dst is the Kronecker delta function ensuring that breaking down of the dominant DgPSO st term into the

123
Page 8 of 13 Theor Chem Acc (2013) 132:1383

Table 3 Individual contributions to the Dg tensor components and the direct contributions of selected atoms obtained employing the one-center
approximation
Radical Ca2? complex Mg2? complex
sq sqA sq6A sq6A*
Dgxx Dgyy Dgzz Dgxx Dgyy Dgzz Dgxx Dgyy Dgzz Dgxx Dgyy Dgzz

Dgtotal 5179 4406 -111 4789 3989 -125 4077 2911 11 4045 2806 -46
DgRMC -231 -231 -231 -229 -229 -229 -223 -223 -223 -225 -225 -225
DgDC 154 177 142 149 178 136 169 249 223 164 226 197
DgPSO 5256 4460 -22 4869 4040 -32 4131 2885 11 4106 2805 -18
DgPSO(1c) 5145 4412 -24 4832 3920 -33 4176 2822 0 4098 2701 -29
PSO PSO
Dg - Dg (1c) 112 48 2 37 120 1 -45 63 11 8 104 11
Selected atoms contributions to Dg(1c)
O1 2640 2098 -14 2495 1831 -17 2171 1397 2 2138 1255 0
O2 2639 2099 -14 2495 1831 -17 2171 1397 2 2143 1257 0
C1 -137 40 1 -133 60 0 -107 46 -9 -119 56 -8
C2 -137 40 1 -133 60 0 -107 46 -9 -119 56 -8
Ca or Mg n/a n/a n/a n/a n/a n/a -32 -132 26 -22 -39 9
All calculated at the UB3LYP/TZVP theory level and given in ppm

1 X Xnc X
2M D  ! !  E
 
rst ¼ 2mc nci dki dmi vk  Rk Rm h01
t vm ;
2c c¼a;b i¼1 k;m
s

ð4bÞ
X nc
X   01  
DgPSO;occocc ¼ nci S1;s   ð4cÞ
st 2mc ij Wi ht Wj ;
c¼a;b i;j¼1

X nc
X X
vir   01  
DgPSO;occvir ¼2 nci u1;s  
st 2mc aj Wi ht Wa :
c¼a;b i¼1 a

ð4dÞ
The term rst has been shown to be numerically irrelevant
[86, 89], and it is therefore neglected in the further dis-
cussion; DgPSO;occocc
st are the couplings between occupied
PSO;occvir
orbitals and Dgst between occupied and virtual
ones; h01
t is the paramagnetic spin–orbit operator defined in
[86]; Wi and Wa are occupied and virtual Kohn–Sham
orbitals, respectively; the orbitals are expanded into the set
Fig. 4 Contributions of the particular atoms to Dgxx (green) and Dgyy of 2M basis functions {vk}; the expansion coefficients are
(maroon) for sq; calculated at the UB3LYP/TZVP theory level
the dki; nci is the occupation number of Wi; S1;s 1;s
ij and uaj are
the first-order occupied–occupied and occupied–virtual
coefficients, respectively; and the coefficients 2mc afford
contributions originating from the particular excited states
the correct signs for a and b spins (ma = ‘ and
can be fruitful for the understanding of the Dg tensor
mb = 2‘). All the Dg tensor calculations performed with
changes on metal ions complexation. To provide similar
the ADF package were spin-unrestricted, based on the
insight in this work, the alternative one-component
scalar Pauli Hamiltonian and employing the UBP86 func-
method proposed by Schreckenbach and Ziegler [86], as
tional in concert with the standard all-electron Slater-type
implemented in the ADF package [88], was used. In this
TZP basis set. The ADF program was used because the
method [86]:
implementation included in it allows to analyze
DgPSO
st ¼ rst þ DgPSO;occocc
st þ DgPSO;occvir
st ; ð4aÞ DgPSO;occvir in terms of single excitations.
st

123
Theor Chem Acc (2013) 132:1383 Page 9 of 13

The DgPSO;occvir
st term is usually the most important one isosurfaces of SOMO, HOMO, and HOMO-2 are shown in
[89], and it is shown here to dominate the perpendicular Fig. 5c.
components of the uncomplexed sq radical and the studied
complexes (see Table 4). Dgzz is very small in magnitude 3.3 Relative stability
because of the insignificant coupling between occupied and
virtual orbitals. Nonetheless, DgPSO;occvir is slightly It would be interesting to find which metal cation, Mg2? or
zz
2? 2?
increased after the Ca or Mg complex formation. To Ca2?, forms more stable complexes with o-semiquinone
   
298
meaningfully discuss the contribution of excited states to ligands. The enthalpies DHgas , the entropies DS298
gas ,
the Dgxx and Dgyy components, the possible excitations the Gibbs energies in the gas phase and in acetonitrile
were classified into three groups: (1) from the doubly 298
(DG298
gas and DG , respectively) and the changes in the
occupied orbitals to the SOMO (D ? S); (2) from the P
SOMO to the virtual ones (S ? V); and (3) from the solvation energies DG298
sol calculated for the radical
doubly occupied orbitals to the virtual (D ? V). The last complexes formation (Eq. 2) are reported in Table 5.
group is expected to bring small contributions to the Dg First, it is sensible to compare the Gibbs energies calcu-
tensor of organic radicals as contributions of these excited lated for the complex formation taking place in acetonitrile
states arise from the spin polarization solely. The contri- (DG298) and in the gas phase (DG298 298
gas ). The DGgas values are

butions of the three groups of excited states to DgPSO;occvir

st
significantly more negative, whereby the formation of the
are listed in Table 4 and visualized in Fig. 5. semiquinone complex with Mg2? or Ca2? from the free
The perpendicular components for sq are dominated by anionic radical and cation complexed by acetonitrile mole-
the contributions from D ? S excited states, in accordance cules tends to be energetically more beneficial in the gas
with the report for p-semiquinone [52]. This group of phase. This can be explained by the fact that the stability of
excited states is the preeminent one also after the Ca2? and ions is increased by solvation more significantly than the
Mg2? complexation; however, its contribution is signifi- stability of uncharged molecules. Considering that the
cantly decreased. For the o-semiquinone radical anion sq, complex formation in Eq. 2 is inseparable from the reduction
Dgxx and Dgyy components are mainly prevailed by the of the charge (one 2? cation and one 1- radical anion are the
contributions from HOMO-2 ? SOMO and HOMO ? substrates and one 1? radical complex is the product), the
SOMO excited states, respectively (the both excited states solvation of the substrates is expected to be energetically
are of the D ? S type). The formation of a complex more beneficial than the solvation of the products, which
between Ca2? and sq results in the significantly reduced should lead to the lowering of the radical binding affinity in
contributions of these two excited states, and consequently, the solution. This presumption is clearly corroborated by the
P
Dgxx and Dgyy decrease on the complex formation. Inter- positive change of solvation energy DG298 sol (see Table 5).
P
estingly, after the complex formation, the HOMO ? Despite the positive DGsol values, the resulting DG298 is
298

SOMO contribution becomes nearly negligible. The predicted to stay negative.

Table 4 Contributions of excited states to the g tensor; all calculated at the UBP86/TZP theory level with the ADF package employing the
method proposed by Schreckenbach and Ziegler [86]
Radical Ca2? complex Mg2? complex
sq sq6 sq6*
Dgxx Dgyy Dgzz Dgxx Dgyy Dgzz Dgxx Dgyy Dgzz
total
Dg 6032 5116 -212 4278 2931 -339 4692 3175 -268
DgRMC -216 -216 -216 -208 -208 -208 -212 -212 -212
DC
Dg 82 67 41 80 66 28 80 65 42
DgPSO 6166 5265 -37 4405 3073 -159 4824 3322 -98
DgPSO,occ–occ 258 584 -17 281 556 30 320 599 20
DgPSO,occ–vir
Total 5908 4681 -20 4124 2517 -189 4504 2723 -118
P
(D ? V) -178 37 -20 -206 177 19 -330 -69 -32
P
(S ? V) -108 174 0 -168 -168 64 -169 168 -68
P
(D ? S) 6194 4469 0 4498 2509 -272 5003 2624 -18
HOMO ? SOMO 0 2016 0 0 295 0 0 303 0
HOMO-2 ? SOMO 4980 0 0 3749 0 0 4228 0 0

123
Page 10 of 13 Theor Chem Acc (2013) 132:1383

Fig. 5 Graphical illustration of various excited states contributions to Dgxx (a) and to Dgyy (b). In addition, molecular orbitals connected to the
excited states giving significant contributions to the Dg tensor are shown (c). Labels to the orbitals were given according to the results for sq

Table 5 Selected thermodynamic properties calculated at the (U)B3LYP/TZVP theory level according to the thermodynamic cycle shown in
Fig. 2
P
298
DHgas (kcal mol-1) DS298 -1
gas [cal (mol K) ] DG298 -1 a
gas (kcal mol ) DG298 -1 b
sol (kcal mol )
DG298 (kcal mol-1)c

Ca2? complexes
sq4 -144.1 21.0 -150.4 131.1 -19.3
psq4 -143.2 11.8 -146.7 128.0 -18.7
ptsq4 -135.2 10.8 -138.4 122.1 -16.4
sq6 -130.1 57.9 -147.4 118.8 -28.6
psq6 -126.2 58.5 -143.6 114.8 -28.8
ptsq6 -119.5 57.8 -136.8 110.0 -26.7
sq7 -129.1 49.1 -143.7 117.2 -26.5
psq7 -124.8 45.5 -138.3 112.3 -26.1
ptsq7 -118.2 45.0 -131.6 107.1 -24.5
sq8 -125.5 34.5 -135.8 113.0 -22.9
psq8 -120.4 42.1 -133.0 109.6 -23.3
ptsq8 -114.4 42.3 -127.1 104.3 -22.7
Mg2? complexes
sq4* -154.2 38.5 -165.6 135.5 -30.2
psq4* -152.4 38.8 -163.9 126.9 -37.0
ptsq4* -143.1 36.8 -154.1 133.6 -20.5
sq6* -140.3 33.2 -150.2 123.4 -26.8
psq6* -136.8 32.8 -146.6 114.1 -32.5
ptsq6* -129.7 31.5 -139.1 120.2 -18.9
2?
The asterisks (*) indicate the Mg complexes (structures taken from Ref. [47])
a
DG298
gas ¼ 298
DHgas  TDS298
gas ¼ 298
DHgas  298:15DS298
gas
b P
DG298 sol sol sol sol sol
sol ¼ DG3 þ 2DG4  DG1  DG2 ; DGi are defined in Fig. 2 and were obtained from the single-point PCM calculations
c 298 298
P 298
DG ¼ DGgas þ DGsol

For the Ca2? complexes, the most negative DG298 values c.n. = 6. To understand why the higher c.n. are energetically
were obtained for c.n. = 6, strongly suggesting the forma- beneficial in the case of Ca2? ions, one has to compare the ion
tion of the complexes with this c.n. in real chemical systems. radius of the both cations. Ca2? has significantly greater
On the other hand, for c.n. = 4, the DG298 values are the least radius (1.00 Å) [90] than the Mg2? (0.72 Å) [90]; in con-
negative, indicting a low probability of such Ca2? complex sequence, Ca2? may be surrounded by a larger number of
formation. In contrast, for the Mg2? ion, c.n. = 4 is shown to ligands without significant steric repulsion between them.
be energetically more profitable as the predicted DG298 Another point of concern is the relative stability of the
values are considerably lower than for the complexes with complexes containing various o-semiquinone ligands. In

123
Theor Chem Acc (2013) 132:1383 Page 11 of 13

general, the relative stabilities of Ca2? coordination com- the molecular and electronic structure of the radical species
pounds with sq and psq radical ligands are comparable; interacting with diamagnetic metal ions. In other words,
however, in the case of Mg2?, the stability of the psq this good agreement between the theoretical and experi-
complexes, clearly indicated by the more negative DG298 mental Dgiso values might be treated as an indication that
value, is noticeably increased as compared with sq. The the other predicted properties (spin distribution, structural
coordination of both cations to the ptsq ligand results in parameters as RC–O, RCa–O) properly characterize the real
lower stability than the coordination to sq and psq. This systems.
can be explained by the fact that the interaction between o- The conducted computations revealed that, in general,
semiquinone radicals and Mg2? and Ca2? cations is mainly the effects of Ca2? and Mg2? complex formation on the
electrostatic in nature, whereby the stability of the formed Dg tensor are similar, although the interaction between
complexes decreases as the negative charge located on the Mg2? and the o-semiquinones, as revealed by the shorter
hydroxyl oxygens atoms is reduced. In the case of ptsq, the O–Mg bonds and more negative DG298, is clearly stronger
negative charge on the O atoms should be moderately compared to Ca2?. The stronger interaction in case of
diminished as compared with sq and psq, since the ptsq Mg2? should be expected to induce more significant
molecule contains two N atoms that are additional attrac- decrease of Dgxx and Dgyy, but this was not observed.
tors of the negative charge. This can be illustrated using the Therefore, this study have shown that the g factor is not a
Löwdin atomic charges. For the hydroxyl oxygens of sq reliable criterion for the strength of the interaction between
and psq, they are predicted to be -0.32 and -0.31, an o-semiquinone and diamagnetic metal cation.
respectively, while for ptsq, just -0.27. The calculated atomic contributions to the Dg tensor
Perhaps the most interesting is the relative stability of indicate that the impact of the metal ion (Ca2? or Mg2?) on
the Ca2? and Mg2? complexes with o-semiquinones since the Dg tensor of o-semiquinone radicals is mainly indirect.
the two cations usually coexist in natural systems and so Although the metal ion brings only a barely noticeable
competition between them is expected to occur. As it can direct contribution, it causes a significant decrease in the
be seen in Table 5, the most negative DG298 values are contributions of hydroxyl oxygens to the Dgxx and Dgyy
predicted for Mg2? complexes with c.n. = 4. This fact components. In addition, the contributions of various
strongly suggests that the formation of the o-semiquinone excited states to the Dg tensor were analyzed. It was shown
complexes with this cation is more favorable, and there- that the decrease of Dgxx and Dgyy on the complex for-
fore, Mg2? can be expected as preferred over Ca2? in the mation is the consequence of reduced contributions of
mechanism of cation transport through membranes [49]. HOMO-2 ? SOMO and HOMO ? SOMO excited states.
In our opinion, it is always sensible to confront theo- Another important observation is that the general sta-
retical results with more general ideas, here with the bility of the Mg2? complexes is higher than that of the
Pearson hard and soft acids and bases concept (HSAB) complexes with Ca2?. Therefore, in the transport mecha-
[91]. According to HSAB, certain metal ions (hard Lewis nism through membranes with Q10 playing the role of the
acids) exhibit high affinity for oxygen donor ligands. Thus, transfer agent [49], Mg2? ions can be expected to be
the harder the Lewis acid, the stronger the preference favored over Ca2?.
for O donors. For the complexes of Mg2? and Ca2? with
o-semiquinones, the Mg2? [ Ca2? stability order is Acknowledgments This work was financed from the National
Science Centre (NCN) funds allocated on the basis of decision DEC-
expected as Mg2? is considered to be a moderately harder 2011/03/B/ST5/01742. The computations were performed using the
acid than Ca2? due to the same ?2 charge but noticeably computers belonging to the Wrocław Center for Networking and
smaller size. To summarize, in spite of its limitations, Supercomputing (Grant No. 47).
HSAB gives a qualitative answer being in agreement with
Open Access This article is distributed under the terms of the
the results yielded by DFT methods. Creative Commons Attribution License which permits any use, dis-
tribution, and reproduction in any medium, provided the original
author(s) and the source are credited.
4 Conclusions

This paper has provided a detailed insight into the inter- References
action between o-semiquinone radicals and Ca2? ions.
Good agreement between the calculated and experimental 1. Bauld NL (1997) Radicals, ion radicals, and triplets. Wiley-VCH,
giso parameters, supported by accordance of the calculated Weinheim
2. Todres ZV (2003) Organic ion radicals: chemistry and applica-
g tensors with the experimental data for PQQ, suggests that
tions. Marcel Dekker, New York
DFT methods are suitable not only for theoretical exami- 3. Ohashi S, Iemura T, Okada N, Itoh S, Furukawa H, Okuda M,
nation of this parameter but also may provide insight into Ohnishi-Kameyama M, Ogawa T, Miyashita H, Watanabe T, Itoh

123
Page 12 of 13 Theor Chem Acc (2013) 132:1383

S, Oh-oka H, Inoue K, Kobayashi M (2010) Photosynth Res 43. Barone V, Cimino P (2009) J Chem Theory Comput 5:192
104:305 44. Ciofini I, Adamo C, Barone V (2004) J Chem Phys 121:6710
4. Barber J (2009) Chem Soc Rev 38:185 45. Condic-Jurkic K, Smith A, Hendrik Z, Smith DM (2012) J Chem
5. Yuasa J, Suenobu T, Fukuzumi S (2006) ChemPhysChem 7:942 Theory Comput 8:1078
6. Ueda A, Ogasawara K, Nishida S, Ise T, Yoshino T, Nakazawa S, 46. Pauwels E, Declerck R, Verstraelen T, De Sterck B, Kay CWM,
Sato K, Takui T, Nakasuji K, Morita Y (2010) Angew Chem Int Van Speybroeck V, Waroquier M (2010) J Phys Chem B
Ed 49:6333 114:16655
7. Vostrikova KE (2008) Coord Chem Rev 252:1409 47. Witwicki M, Jezierska J (2011) J Phys Chem B 115:3172
8. Fukuzumi S, Ohkuboa F, Morimoto Y (2012) Phys Chem Chem 48. Bennett IM, Farfano HMV, Bogani F, Primak A, Liddell PA,
Phys 14:8472 Otero L, Sereno L, Silber JJ, Moore AL, Moore TA, Gust D
9. Zech SG, Hofbauer W, Kamlowski A, Fromme P, Stehlik D, (2002) Nature 420:398
Lubitz W, Bittl R (2000) J Phys Chem B 104:9728 49. Bogeski I, Gulaboski R, Kappl R, Mirceski VB, Stefova M, Pe-
10. Kay CWM, Mennenga B, Görisch H, Bittl R (2004) FEBS Lett treska J, Hoth M (2011) J Am Chem Soc 133:9293
564:69 50. Mirceski V, Gulaboski R, Bogeski I, Hoth M (2007) J Phys Chem
11. Kay CWM, Mennenga B, Görisch H, Bittl R (2005) J Am Chem C 111:6068
Soc 127:7974 51. Yuasa J, Suenobu T, Fukuzumi S (2005) J Phys Chem A
12. Jezierski A, Czechowski F, Jerzykiewicz M, Chen Y, Drozd J 109:9356
(2000) Spectrochim Acta A 56:379 52. Ciofini I, Reviakine R, Arbuznikov A, Kaupp M (2004) Theor
13. Jezierski A, Czechowski F, Jerzykiewicz M, Golonka I, Drozd J, Chem Acc 111:132
Bylinska E, Chen Y, Seaward MRD (2002) Spectrochim Acta A 53. Begue D, Carbonniere P, Barone V, Pouchan C (2005) Chem
58:1293 Phys Lett 416:206
14. Witwicki M, Jezierska J, Ozarowski A (2009) Chem Phys Lett 54. Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA,
473:160 Cheeseman JR, Scalmani G, Barone V, Mennucci B, Petersson
15. Witwicki M, Jerzykiewicz M, Jaszewski AR, Jezierska J, Oza- GA, Nakatsuji H, Caricato M, Li X, Hratchian HP, Izmaylov
rowski A (2009) J Phys Chem A 113:14115 AF, Bloino J, Zheng G, Sonnenberg JL, Hada M, Ehara M,
16. Eaton DR (1964) Inorg Chem 3:1268 Toyota K, Fukuda R, Hasegawa J, Ishida M, Nakajima T, Honda
17. Jerzykiewicz M (2012) Spectrochim Acta A 96:127 Y, Kitao O, Nakai H, Vreven T, Montgomery JJA, Peralta JE,
18. Jerzykiewicz M (2013) Chemosphere 92:445 Ogliaro F, Bearpark M, Heyd JJ, Brothers E, Kudin KN, Star-
19. Kukushkin AK, Jalkanen KJ (2010) Theor Chem Acc 125:121 overov VN, Kobayashi R, Normand J, Raghavachari K, Rendell
20. Sinnecker S, Neese F (2007) Top Curr Chem 268:47 A, Burant JC, Iyengar SS, Tomasi J, Cossi M, Rega N, Millam
21. Alberto ME, Marino T, Russo N, Sicilia E, Toscano M (2012) NJ, Klene M, Knox JE, Cross JB, Bakken V, Adamo C, Jara-
Phys Chem Chem Phys 14:14943 millo J, Gomperts R, Stratmann RE, Yazyev O, Austin AJ,
22. Witwicki M, Jezierska J (2010) Chem Phys Lett 493:364 Cammi R, Pomelli C, Ochterski JW, Martin RL, Morokuma K,
23. Leopoldini M, Marino T, Russo N, Toscano M (2004) Theor Zakrzewski VG, Voth GA, Salvador P, Dannenberg JJ, Dapprich
Chem Acc 111:210 S, Daniels AD, Farkas Ö, Foresman JB, Ortiz JV, Cioslowski J,
24. Geldof D, Krishtal A, Blockhuys F, Van Alsenoy C (2012) Theor Fox DJ (2009) Gaussian 09 revision A.02. Gaussian, Inc.,
Chem Acc 131:1243 Wallingford CT
25. De Vleeschouwer F, Geerlings P, De Proft F (2012) Theor Chem 55. Lee C, Yang W, Parr RG (1988) Phys Rev B 37:785
Acc 131:1245 56. Becke AD (1993) J Chem Phys 98:1372
26. Villamena FA, Locigno EJ, Rockenbauer A, Hadad CM, Zweier 57. Stephens PJ, Devlin FJ, Chabalowski CF, Frisch MJ (1994) J
JL (2006) J Phys Chem A 110:13253 Phys Chem 98:11623
27. Villamena FA, Locigno EJ, Rockenbauer A, Hadad CM, Zweier 58. Schaefer A, Horn H, Ahlrichs R (1992) J Chem Phys 97:2571
JL (2007) J Phys Chem A 111:384 59. Barge A, Botta M, Casellato U, Tamburini S, Vigato PA (2005)
28. Rinkevicius Z, Telyatnyk L, Vahtras O (2004) J Chem Phys Eur J Inorg Chem 2005:1492
121:5051 60. Hewitt IJ, Tang JK, Madhu NT, Clérac R, Buth R, Anson CE,
29. Rinkevicius Z, Murugan NA, Kongsted J, Aidas K, Steindal AH, Powell AK (2006) Chem Commun 2650. http://pubs.rsc.org/en/
Ågren H (2011) J Phys Chem B 115:4350 content/articlelanding/2006/cc/b518026k
30. Li X, Rinkevicius Z, Kongsted J, Murugan NA, Ågren H (2012) J 61. Akine S, Kagiyama S, Nabeshima T (2007) Inorg Chem 46:9525
Chem Theory Comput 8:4766 62. Akine S, Taniguchi T, Nabeshima T (2006) J Am Chem Soc
31. Mattar SM (2004) J Phys Chem B 108:9449 128:15765
32. Asher JR, Kaupp M (2008) Theor Chem Acc 119:477 63. Cancès E, Mennucci B, Tomasi J (1997) J Chem Phys 107:3032
33. Asher JR, Doltsinis NL, Kaupp M (2004) J Am Chem Soc 64. Tomasi J, Mennucci B, Cancès E (1999) J Mol Struct Theochem
126:9854 464:211
34. Kaupp M, Remenyi C, Vaara J, Malkina OL, Malkin VG (2002) J 65. Tomasi J, Mennucci B, Cammi R (2005) Chem Rev 105:2999
Am Chem Soc 124:2709 66. Neese F (2007) ORCA-an ab initio, DFT and semiempirical SCF-
35. Malkina OL, Vaara J, Schimmelpfennig B, Munzarova M, Mal- MO package, Version 2.6.35. University of Bonn, Germany
kin VG, Kaupp M (2000) J Am Chem Soc 122:9206 67. Adamo C, Barone V (1999) J Chem Phys 110:6158
36. O’Malley PJ (1998) J Phys Chem A 102:248 68. Perdew JP, Burke K, Ernzerhof M (1996) Phys Rev Lett 77:3865
37. O’Malley PJ (1998) Chem Phys Lett 285:99 69. Becke AD (1988) Phys Rev A 38:3098
38. O’Malley PJ (1998) Chem Phys Lett 291:367 70. Perdew JP (1986) Phys Rev B 33:8822
39. Lin T, O’Malley PJ (2011) J Phys Chem B 115:9311 71. Hoe W, Cohen A, Handy NC (2001) Chem Phys Lett 341:319
40. Martin E, Samoilova RI, Narasimhulu KV, Lin T, O’Malley PJ, 72. Klamt A, Schüürmann G (1993) J Chem Soc Perkin Trans 2:799
Wraight CA, Dikanov SA (2011) J Am Chem Soc 133:5525 73. Sinnecker S, Rajendran A, Klamt A, Diedenhofen M, Neese F
41. Witwicki M, Jezierska J (2012) Geochim Cosmochim Acta (2006) J Phys Chem A 110:2235
86:384 74. Neese F (2001) J Chem Phys 115:11080
42. Improta R, Barone V (2004) Chem Rev 104:1231 75. Neese F (2005) J Chem Phys 122:34107

123
Theor Chem Acc (2013) 132:1383 Page 13 of 13

76. Bethe H, Salpeter E (1957) Quantum mechanics of one- and two- Dickson RM, Ellis DE, van Faassen M, Fan L, Fischer TH,
electron atoms. Springer, Berlin Fonseca Guerra C, Ghysels A, Giammona A, van Gisbergen SJA,
77. Breit G (1929) Phys Rev 34:553 Götz AW, Groeneveld JA, Gritsenko OV, Grüning M, Harris FE,
78. Dinadayalane TC, Hassan A, Leszczynski J (2012) Theor Chem van den Hoek P, Jacob CR, Jacobsen H, Jensen L, van Kessel G,
Acc 131:1131 Kootstra F, Krykunov MV, van Lenthe E, McCormack DA,
79. Russo N, Toscano M, Grand A (2003) J Phys Chem A 107:11533 Michalak A, Mitoraj M, Neugebauer J, Nicu VP, Noodleman L,
80. Remko M, Rode BM (2006) J Phys Chem A 110:1960 Osinga VP, Patchkovskii S, Philipsen PHT, Post D, Pye CC,
81. Neese F (2009) Coord Chem Rev 253:526 Ravenek W, Rodrı́guez IJ, Ros P, Schipper PRT, Schreckenbach
82. Atanasov M, Comba P, Martin B, Muller V, Rajaraman G, G, Seth M, Snijders JG, Solà M, Swart M, Swerhone D, te Velde
Rohwer H, Wunderlich S (2006) J Comput Chem 27:1263 G, Vernooijs P, Versluis L, Visscher L, Visser O, Wang F,
83. Tewary S, Gass IA, Murray KS, Rajaraman G (2013) Eur J Inorg Wesolowski TA, van Wezenbeek EM, Wiesenekker G, Wolff
Chem 2013:1024–1032 SK, Woo TK, Yakovlev AL, Ziegler T (2008) Amsterdam
84. Stone AJ (1963) Mol Phys 6:509 Density Functional (ADF) 2008.01, SCM, Theoretical Chemistry,
85. Stone AJ (1964) Mol Phys 7:311 Vrije Universiteit, Amsterdam, The Netherlands (http://www.
86. Schreckenbach G, Ziegler T (1997) J Phys Chem A 101:3388 scm.com)
87. Kaupp M, Bühl M, Malkin VG (2004) Calculation of nmr and epr 89. Schreckenbach G, Ziegler T (1998) Theor Chem Acc 99:71
parameters: theory and applications. Wiley-VCH, Weinheim 90. Atkins PW (1990) Physical chemistry, 4th edn. W. H. Freeman
88. Baerend EJ, Autschbach J, Bashford D, Bérces A, Bickelhaupt and Co., New York
FM, Bo C, Boerrigter PM, Cavallo L, Chong DP, Deng L, 91. Pearson RG (1963) J Am Chem Soc 85:3533

123