0 downloads
0 Views
158KB Size

J. Phys. Chem. A 2007, 111, 12011-12018

12011

Molecular Structures of Phthalocyaninatozinc and Hexadecafluorophthalocyaninatozinc Studied by Gas-Phase Electron Diffraction and Quantum Chemical Calculations Tatyana Strenalyuk, Svein Samdal,* and Hans Vidar Volden Department of Chemistry, Centre for Theoretical and Computational Chemistry, UniVersity of Oslo, P.O. Box 1033 Blindern, NO-0315 Oslo, Norway ReceiVed: June 8, 2007; In Final Form: September 4, 2007

The molecular structures of phthalocyaninatozinc (HPc-Zn) and hexadecafluorophthalocyaninatozinc (FPcZn) are determined using the gas electron diffraction (GED) method and high-level density functional theory (DFT) quantum chemical calculations. Calculations at the B3LYP/6-311++G** level indicate that the equilibrium structures of HPc-Zn and FPc-Zn have D4h symmetry and yield structural parameters in good agreement with those obtained by GED at 480 and 523 °C respectively. The calculated force fields indicate that both molecules are flexible. Normal coordinate calculations on HPc-Zn yield five vibrational frequencies (one degenerate) in the range 22-100 cm-1, and ten vibrational frequencies ranging from 13 to 100 cm-1 (three degenerate) for FPc-Zn. The high-level force field calculations confirm most of the previous vibrational assignments, and some new ones are suggested. The out-of-plane vibration of the Zn atom in HPc-Zn was studied in detail optimizing models in which the distance from the Zn atom to the two symmetry equivalent diagonally opposed N atoms (h) was fixed. The calculations indicate that the vibrationally activated vertically displacement of the Zn atom is accompanied by distortion of the ligand from D4h to C2V symmetry. The average height, h, at the temperature of the GED experiment was calculated to be 14.5 pm. Small structural changes indicate that a full F substitution on the benzo-subunits do not significantly alter the geometry, however there are indications that the benzo-subunits may shrink slightly with perfluorination.

Introduction Metal phthalocyanines (MPcs) represent a class of organic materials that attracts great attention in several different fields of science and industry. Their high thermal and chemical stability have made them suitable for many technological applications such as dyes, pigments, semiconductors, energy conversion in photovoltaic and solar cells, photosensitizers, electrophotography, molecular electronics, gas sensors, electrochromism in display devices, liquid crystals, LangmuirBlodgett films, nonlinear optics, electrocatalytic reagents, and photodynamic therapy. Due to the similarity in structure to biologically important molecules such as chlorophyll and hemoglobin they have caught special interest. There are several excellent books about MPcs where the above-mentioned properties and applications are discussed.1-6 SciFinder gives about 39 000 hits on phthalocyanines. In spite of this huge amount of information about Pcs, there is almost nothing known about their molecular structures in the gas phase. MPcs are thermally stable compounds, and sublimation is used both for purification and for creation of thin films for different purposes. It should therefore be possible to get useful information about the molecular structures of these compounds in the gas phase using the gas electron diffraction (GED) method. GED is the only method suitable to determine the molecular geometry for such large molecules in the gas phase. The only GED investigations so far have been carried out by Fink and his coworkers who have studied HPc-Zn7 in 1993, HPc-Sn and HPcMg and reinvestigated HPc-Zn8 in 1999, and HPc-Cu and HPcNi9 in 2000. The reasons MPcs have not been more extensively * Corresponding author. E-mail: [email protected] Phone: 47 2285 5458. Fax +47 2285 5441.

studied by the GED method might be related first to the lack of equipment to heat the effusion cell to a sufficiently high temperature to evaporate the compound and second to the fact that most GED laboratories have used photographic plates to record data that might make it more difficult to obtain good data than using image plates. Fink and his co-workers avoided this problem by using an electron counting device,7 but this is a time-consuming process and the collection of a data set may take as much as 24 h and the observed scattering range is usually small compared to conventional GED investigations. The data range for HPc-Zn was up to s ) 145 pm-1, which made the precision and accuracy of the obtained structural parameters less reliable than usually obtained by the GED method. Since 2000 we have been using the image plate (IP) technology to record the GED data. The IPs are more sensitive, have higher resolution, have a more linear response that makes blackness correction unnecessary, and have a much larger dynamic range than photographic plates. Diffraction data from gaseous benzene, which is used for calibration of the electron wavelength have shown that the reproducibility and accuracy obtained by using IPs are better than for photographic plates,10 which should make the IPs particular suitable for recording data at high temperatures. We have recently successfully investigated F-SubPc with molecular formula C24BClF12N6, systematic name chloro[1,2,3,4,8,9,10,11,15,16,17,18-dodecafluoro-7,12:14,19diimino-21,5-nitrilo-5H-tribenzo[c,h,m]1,6,11 triazacyclopentadecinato(2-)κN22,κN23,κN24]boron(III) and trivial name chlorododecafluorosubphthalocyanatoboron(III) by GED,11 and a natural extension would be to determine the molecular geometry of MPcs in the gas phase. There is a rather limited amount of high-level quantum chemical calculations available on MPcs. Mostly semiempirical

10.1021/jp074455n CCC: $37.00 © 2007 American Chemical Society Published on Web 11/03/2007

12012 J. Phys. Chem. A, Vol. 111, No. 47, 2007 and molecular mechanics methods have been used, obviously because of the molecular size and time-consuming calculations. For HPc-Zn there are conflicting predictions of the equilibrium geometry, one B3LYP/6-31G* calculation predicts a planar D4h geometry,8 another B3LYP/6-31G* calculation predicts a nonplanar C4V geometry12 where the Zn atom is 7.5 pm above the Pc plane. The first GED investigation7 of HPc-Zn concluded with a nonplanar C4V geometry where the Zn atom is 34(13) pm above the Pc plane, whereas the reinvestigation concluded with a planar D4h geometry8 with the Zn atom in the symmetry plane. Moreover, there is a substantial disagreement between some of the C-C bond distances determined by GED7,8 for HPcZn on one hand and by the quantum chemical calculations or the X-ray diffraction structure13 on the other. There are also a limited number of force field calculations and vibrational assignments for MPcs. B3LYP/6-31G** calculations have been used to calculate IR and Raman intensities for HPc-Zn12,14 and to assign the fundamental frequencies. However, larger basis sets including diffuse and polarization functions are probably needed to give good predictions of the IR intensities15,16 or Raman intensities.17,18 MPcs have been considered as rather rigid molecules, but force field calculations predict several low-frequency modes, and this might have serious consequences for a GED analysis. The calculated root-meansquare amplitudes of vibration (u-values) from the molecular force field are nowadays always used in GED investigations to get more reliable u-values than those transferred from smaller fragments of the molecules. The previous investigations on HPcZn have neither used calculated u-values from forces field calculations nor explored the affect of the shrinkage correction, which might affect the determination of the molecular geometry. We present here our results from GED and high-level quantum chemical studies on the structures of the molecules HPc-Zn with molecular formula C32H16N8Zn, systematic name [29H,31H-phthalocyaninato(2-)-κN29,κN3O,κN31,κN32]zinc(II) and trivial name phthalocyaninatozinc, and FPc-Zn with molecular formula C32F16N8Zn, systematic name [1,2,3,4,8,9,10,11,15,16,17,18,22,23,24,25-hexadecafluoro-29H,31H-phthalocyaninato(2-)-κN29,κN3O,κN31,κN32]zinc(II) and trivial name hexadecafluorophthalocyaninatozinc. The latter has been included because it has been claimed that the properties of MPcs are little influenced by a peripheral substitution. Experimental Section Both compounds, HPc-Zn and FPc-Zn, were obtained from Aldrich and used without further purification. The gas electron diffraction (GED) data were recorded on the Balzers KD-G2 unit19 at the University of Oslo with use of an accelerating voltage of about 42 kV and a high-temperature stainless steel inlet system.20,21 Experimental data were recorded on FujiFilm BAS-III image plates, and they were scanned with a FujiFilm BAS-1800II scanner. The image plate has a rectangular shape of approximately 13 × 18 cm2. The data are collected and averaged over sectors along the positive and negative y-axis (short axis) and x-axis (long axis). More details about handling10 of the experimental data and their processing22 are given elsewhere. The measured temperature represents the oven temperature and is measured with a Cu/constantan thermocouple. The thermocouple is placed outside the metal oven, and the dimension of the oven is approximately 7 × 2 × 2 cm3. The nozzle opening is connected to the oven and is approximately 4 cm from the thermocouple. The measured oven temperature is assumed to be the temperature of the gas sample.

Strenalyuk et al.

Figure 1. Molecular numbering of XPc-Zn (X ) H or F) together with some auxiliary points. The figure below shows possible bending of isoindole unit under C4V symmetry.

The modification and scattering functions were computed from tabulated atomic scattering factors23 for the proper wavelength and s values. The experimental backgrounds were computed with the program KCED12,24 where the coefficients of a chosen degree of a polynomial function are determined by the least-squares method by minimizing the differences between the experimental intensity and the currently best geometrical model on the modified form. A background is subtracted from data obtained for each individual sector. An average experimental intensity curve is obtained along the y- and x-axis, respectively giving two average experimental intensity curves for each camera distance as shown in Figure 2and 3. The experimental conditions are summarized in Table 1. Quantum Chemical Calculations. The quantum chemical calculations were performed with the GAUSSIAN03 program package25 running on the HP “superdome” facilities in Oslo. The functional used for the calculations was the Becke threeparameter (B3LYP) hybrid functional,26,27 which has been widely used in theoretical studies of pyrrole macrocycles.28-38 Our previous experience with quantum chemical calculations on F-SubPc with different methods and basis sets confirms that DFT gives good agreement with the experimental results.11 The basissetswehaveusedare6-31G*,39 6-31++G**,6-311++G**,40,41 cc-pVTZ(H,C,N,F),42 and cc-pVTZ-NR(Zn).43 To our knowledge, the present calculations represent the highest level performed so far on HPc-Zn and FPc-Zn. The molecular geometry was optimized by assuming different symmetries. The numbering of the atoms is shown in Figure 1. The most important structural parameters are given in Table 2. The Cartesian coordinates from the B3LYP/cc-pVTZ calculations for HPc-Zn and FPc-Zn are given as Supporting Information (Table S1 and S2). Molecular force field calculations were carried out for all basis sets except for DFT/cc-pVTZ, which requires too much computer time, to ensure that the stationary points represent a minimum. The u-values and shrinkage correction values,

Molecular Structures of Phthalocyaninatozincs

Figure 2. Average molecular intensity curves for HPc-Zn. The first two curves are the average from eight sectors from the long camera distance along the y- and x-axis, respectively. The two next curves are the corresponding average curves from eight sectors from the middle camera distance. Dots are experimental points and a full line is a theoretical curve. The four curves below are difference curves.

Figure 3. Average molecular intensity curves for FPc-Zn. The first two curves are the average from eight sectors from the long camera distance along the y- and x-axis, respectively. The next two curves are the corresponding average curves from eight sectors from the middle camera distance. Dots are experimental points, and a full line is the theoretical curve. The four curves below are difference curves.

k-values, used for shrinkage correction in the GED analysis were calculated from the B3LYP/6-311++G** force field. The proposed assignment of the IR and Raman spectra have mainly been based on calculated IR and Raman intensities from the quantum chemical calculations.12,14 Because the reliability of the IR and Raman intensities are improved by introduction of diffuse and polarization functions,15-18 and because there is little information about how these intensities change with inclusion of such functions, we have included these intensities together with the calculated frequencies as Supporting Information (Table S3). Structure Analysis. The electron diffraction technique is described elsewhere.44,45 A brief description that should be particularly relevant for this investigation to understand the assumptions made, to understand the precision and accuracy obtained for the structural parameters, and to judge the experimental results is given in the study of F-SubPc.11

J. Phys. Chem. A, Vol. 111, No. 47, 2007 12013 HPc-Zn and FPc-Zn have similar molecular structures with a symmetrical Pc macrocycle composed of four isoindole units and a central cavity accommodating a Zn atom. In the case of FPc-Zn all H atoms are replaced with F atoms. Owing to the structural similarity of these two molecules, structural analysis was processed in the same manner for both molecules. The molecular symmetry was assumed to be C4V, which can easily be constrained to D4h during the analysis. The numbering of the atoms and auxiliary points is shown on Figure 1. The molecular geometry of HPc-Zn and FPc-Zn can be modeled by rotating the isoindole fragment (N2, C6, C7, C14, C15, C22, C23, C30, C31, X38, X39, X46, X47, where X ) H or F) 90.0°, 180°, and 270° about the 4-fold symmetry axis. The following set of independent parameters was composed to describe the molecular geometry: bond lengths Zn1-N2, N2C6, C6-C14, C14-C15, C14-C22, C22-C30, C30-C31, C22-X38 C30-X46, bond angles C6-N2-C7, C15-C14C22, C14-C22-X38, C22-C30-X46; dihedral angles C15C14-C22-X38, C14-C22-C30-X46, which are all assumed 180.0°. In addition five more parameters shown in Figure 1 were used to test whether average molecular structure possessed C4V symmetry or D4h symmetry. These parameters are the angle Zn1-N2-A1 (a1), where A1 is an auxiliary point used to define the position of the Zn atom out of central cavity, the angle Zn1N2-A2 (a2) that defines the pyramidality of the N2 atom, the angle N2-A2-A3 (a3) defining the planarity of the pyrrole ring, and the angle A2-A3-A4 (a4) that defines the folding of the isoindole unit about the C14-C15 bond, and finally the angle N54-A6-A7 which defines the conformation of the ring consisting of the atoms Zn1, N2, C7, N54, C8, N3. If N54 is above the plane through the atoms N2, C7, C8, N3, the ring has a boat conformation, and if N54 is below, it has a chair conformation. It is difficult to determine accurate structural parameters by the GED method where H atoms are involved due to the low scattering power of the H atoms. It is also difficult to determine accurate differences between bond lengths if the bond type and bond lengths are similar. This is the reason for introducing the following constraints: r(C14C22) ) r(C14C15) + ∆1, r(C22C30) ) r(C14C15) + ∆2, r(C30C31) ) r(C14C15) + ∆3, r(C30X46) ) r(C22X38) + ∆4 and r(N54C7) ) r(N2C6) + ∆5. The differences ∆1-5 for HPc-Zn were taken from the quantum chemical calculation at the B3LYP/6-311++G** level, and were -1.57, -1.72, -0.37, +0.10, and -4.21 pm, respectively. The differences ∆1-5 for FPc-Zn were taken from the B3LYP/6-311++G** calculation and were -2.51, -2.38, -1.56, +0.15, and -4.33 pm, respectively. With these parameters and restrictions the molecular geometry can adapt D4h symmetry, C4V symmetry with a planar Pc and the Zn atom above the Pc plane, or C4V symmetry with a concave shape of macrocycle and a square pyramidal coordination of a central Zn atom. The shrinkage effect can be corrected for by using either a rectilinear46,47 or a curvilinear48,49 treatment of the vibrating atoms. The program SHRINK48,49 was used to calculate these shrinkage correction terms kh0 (rectilinear) and kh1 (curvilinear), u-values, and the ra - re differences from the B3LYP/6311++G** molecular force field. The curvilinear approach is generally considered to be the more realistic. Both the rectilinear and curvilinear approaches were tested, and the final results are given in Table 3 for HPc-Zn and in Table 4 for FPc-Zn. Comparison between experimental and theoretical results is given in Table 5and 6. The u-values were refined in groups.

12014 J. Phys. Chem. A, Vol. 111, No. 47, 2007

Strenalyuk et al.

TABLE 1: Experimental Conditions for the GED Investigation of HPc-Zn and FPc-Zn HPc-Zn nozzle temp/ °C nozzle-to-plate distance/mm number of plates s-range: x-direction/nm-1 y-direction/nm-1 ∆s/nm-1 accelerating voltage/kV electron wave length/pm degree of polynomiala a

FPc-Zn

480 498.73 4

480 248.75 4

523 498.67 4

523 248.87 4

17.50-145.00 17.50-125.00 1.25 42 5.82 8

40.00-302.50 40.00-242.50 1.25 42 5.82 10

17.50-150.00 17.50-133.75 1.25 42 5.82 9

40.00-302.50 43.75-227.50 1.25 42 5.82 12

Degree of polynomial used in the background subtraction

TABLE 2: Calculated Molecular Geometrya (D4h) Using B3LYP for HPc-Zn and FPc-Zn HPc-Zn 6-31G*

6-31++G**

FPc-Zn

6-311++G**

cc-pVTZ

6-31G*

6-311++G**

cc-pVTZ

199.97 136.73 145.85 132.59 140.62 139.01 138.85 140.20 108.10 108.19

199.35 137.16 145.95 132.75 141.59 139.16 139.43 140.15 133.26 133.55

200.42 137.10 146.21 132.82 141.75 139.28 139.40 139.91 133.70 133.82

200.25 136.58 145.81 132.31 141.18 138.65 138.89 139.69 132.94 133.13

125.00 110.00 108.39 127.44 106.61 121.00 117.85 121.14 120.66 119.64 125.11

125.09 109.83 108.59 127.79 106.50 120.62 118.48 120.90 122.57 120.15 124.25

124.91 110.18 108.40 127.69 106.51 120.48 118.71 120.82 122.54 120.23 124.80

124.80 110.40 108.24 127.65 106.56 120.53 118.64 120.83 122.56 120.19 125.10

Bond Lengths Zn1-N2 N2-C6 C6-C14 C6-N57 C14-C15 C14-C22 C22-C30 C30-C31 C22-X38 C30-X46

199.06 137.30 146.07 132.98 141.02 139.52 139.42 140.79 108.56 108.66

200.12 137.28 146.23 133.13 141.20 139.66 139.57 140.90 108.51 108.61

200.11 137.06 146.12 132.85 140.94 139.37 139.22 140.57 108.32 108.42

Zn1N2C6 C6N2C7 N2C6C14 N2C6N57 C6C14C15 C15C14C22 C14C22C30 C22C30C31 C14C22X38 C22C30X46 C6N57C13

125.29 109.41 108.77 127.53 106.53 121.08 117.76 121.16 120.63 119.60 124.36

125.14 109.72 108.59 127.44 106.55 121.06 117.79 121.15 120.68 119.61 124.84

125.04 109.92 108.44 127.50 106.60 121.02 117.82 121.15 120.67 119.63 124.92

Bond Angles

a

Numbering is given in Figure 1. Distances in pm and angles in degrees.

The KCED2550 least-squares fitting program was used to obtain the experimental parameters. The intensity and radial distribution curves are shown in Figures 2 and 4 for HPc-Zn and in Figures 3 and 5 for FPc-Zn. Results and Discussion Optimized Theoretical Geometry. The optimized geometry parameter values are given in Table 2. As can be seen, there are only small changes in the geometry for the different basis sets, which indicates that good predictions of the geometry may be obtained by the 6-31G* basis set compared to the far more time-consuming cc-pVTZ calculations. The agreement between the calculated geometry and the experimental derived geometry is very good, as can be seen in Tables 5 and 6. DFT force field calculations predict that both HPc-Zn and FPc-Zn have D4h symmetry. Comparison of the structural parameters for the different basis sets in Table 2 reveals small changes between HPc-Zn and FPcZn. The changes in bond lengths are usually less than 0.5 pm and smaller than 0.5° for bond angles. The only parameter that has a clear change is the bond angle C14-C22-X38, which increases by about 2° from HPc-Zn to FPc-Zn. This increase is probably due to the electrostatic repulsion between the electronegative atoms F38 and N57 bridge atom. This confirms the assumption that Pcs are little affected by peripheral substitution.

Molecular Vibrations and Flexibility. The MPcs have been commonly regarded as rigid molecules, but this is a rather doubtful description because of five vibrational frequencies (one degenerate) for HPc-Zn and ten vibrational frequencies (three degenerate) for FPc-Zn below 100 cm-1. The lowest vibrational mode, B2u, is 22 cm-1 for HPc-Zn and as low as 13 cm-1for FPc-Zn. It is important to understand that the GED structure represents a molecular average structure which is the weighted (Boltzman) average over all vibrational states. Due to the low vibrational mode at 22 cm-1 (263 J mol-1) and an experimental temperature of 750 K, which give RT ) 6240 J mol-1 a large number of the vibrationally excited states are populated, and averaging over these states should give an average nonplanar molecular structure to be observed in GED if shrinkage is not included in the GED analysis. It is important to realize that the D4h symmetry obtained by the DFT calculations (re structure) and a C4V symmetry obtained by the GED method without shrinkage correction do not represent a serious disagreement. It simply represents two different molecular structures that are not directly comparable. Looking at the atomic displacements for the different vibrational modes for HPc-Zn shows that the atomic displacements of the Zn atom are small for the lowest mode except for the 124 cm-1 A2u mode, which shows a displacement of the Zn atom out of the symmetry plane. This mode has been simulated by calculating the potential energy as

Molecular Structures of Phthalocyaninatozincs

J. Phys. Chem. A, Vol. 111, No. 47, 2007 12015

TABLE 3: Experimental Parameters for HPc-Zna rectilinear approach

curvilinear approch

parameter

C4V

D4h

D4h + kh0

C4V

D4h

D4h + kh1

Zn1-N2 N2-C6 C6-N57 C6-C14 C14-C15 C14-C22 C22-C30 C30-C 31 C22-H38 C30-H46 C7N2C6 C15C14C22 C14C22C30b C22C30C31b C14C22H38 C22C30H46 a1d a2d a3d a4d N54A7A6e h(Zn)e R,f %

198.8(7) 138.5(8) 134.3(8) 145.5(6) 140.9(6) 139.3(6) 139.2(6) 140.5(6) 107.8(13) 107.9(13) 107.6(9) 121.6(3) 116.7(7) 121.7(3) 120.6c 119.6c 9.3(15) 73.8(49) 98.0(68) 96.7(53) 75.3(46) 32.3(51) 5.06

197.8(6) 137.0(5) 132.8(5) 145.2(6) 141.9(4) 140.3(4) 140.2(4) 141.5(4) 106.7(12) 106.8(12) 108.9(5) 122.0(2) 115.9(5) 122.1(2) 120.6c 119.6c 0.0c 90.0c 90.0c 90.0c 90.0c 0.0 5.42

198.8(6) 138.3(5) 134.1(5) 145.9(6) 141.0(5) 139.5(5) 139.3(5) 140.7(5) 108.3(12) 108.4(12) 107.6(7) 121.1(3) 117.5(6) 121.4(3) 120.6c 119.6c 0.0 90.0c 90.0c 90.0c 90.0c 0.0 5.12

198.8(7) 138.4(8) 134.2(8) 145.6(6) 140.9(6) 139.4(6) 139.2(6) 140.6(6) 107.9(12) 108.0(12) 107.8(10) 121.6(3) 116.7(7) 121.7(3) 120.6c 119.6c 9.1(15) 74.5(49) 97.3(69) 97.3(50) 75.7(47) 31.4(50) 4.96

197.8(6) 137.0(5) 132.8(5) 145.3(6) 141.9(4) 140.3(4) 140.1(4) 141.5(4) 106.8(12) 106.9(12) 108.9(5) 122.0(2) 115.9(5) 122.1(3) 120.6c 119.6c 0.0c 90.0c 90.0c 90.0c 90.0c 0.0 5.31

198.9(6) 138.2(5) 134.0(5) 146.0(6) 141.2(5) 139.4(5) 139.5(5) 140.7(5) 108.6(12) 108.7(12) 108.6(5) 121.1(3) 117.7(8) 121.2(4) 120.7c 119.6c 0.0c 90.0c 90.0f 90.0c 180.0c 0.0 5.10

a ra parameters. Parenthesized values are estimated error limits defined as 2.5(σ2lsq + (0.001r)2)1/2, for distances and as 2.5σlsq for angles, where σlsq is one standard deviation taken from the least-squares refinements using a diagonal weight matrix and, the 0.001r term represents 0.1% uncertainty in the electron wavelength. The errors are in the units of the last digits. b Dependent parameter. c Fixed to their DFT calculated values. d See Figure 1 for definition. Zn1N2A1 ) a1 (a1 is negative if the Zn atom is below the plane through the cavity N atoms), Zn1N2A2 ) a2 + 90°, N2A1A3 ) a3 + 90°, A1A2A3 ) a4 + 90°. e The dihedral angle Zn1A6A7N54 is 0°, which gives the six-member ring consisting of the atoms Zn1, N2, calc 2 obs 2 C7, N54, C8, N3 a boat conformation. f R ) {∑sw(Iobs s - Is ) }/{∑sw(Is ) } × 100%, where w is a weight function usually equal to 1 and I is the molecular modified intensity.

TABLE 4: Experimental Parameters for FPc-Zna rectilinear approach

curvilinear approch

parameter

C4V

D4h

D4h + kh0

C4V

D4h

D4h + kh1

Zn1-N2 N2-C6 C6-N57 C6-C14 C14-C15 C14-C22 C22-C30 C30-C31 C22-F38 C30-F46 C7N2C6 C15C14C22 C14C22C30b C22C30C31b C14C22F38 C22C30F46 a1d a2d a3d a4d N54A7A6e h(Zn)e R,f %

201.0(9) 138.9(5) 134.6(5) 149.1(8) 142.2(4) 139.7(4) 139.9(4) 140.7(4) 132.3(4) 132.5(4) 105.8(12) 120.3(5) 119.1(9) 120.6(5) 121.2(9) 118.4(10) -2.3(30) 110.7(29) 55.4(33) 109.8(26) 90.7(53) -7.9(86) 4.55

199.6(8) 137.8(5) 133.5(5) 145.3(10) 142.6(4) 140.0(4) 140.2(4) 141.0(4) 132.5(5) 132.7(5) 112.3(8) 120.5(4) 118.7(8) 120.8(4) 123.1(10) 119.4(12) 0.0c 90.0c 90.0c 90.0c 90.0c 0.0 6.06

200.0(8) 138.1(6) 133.8(5) 147.9(8) 142.6(4) 140.1(4) 140.2(4) 141.0(4) 132.6(5) 132.7(5) 111.2(7) 120.3(3) 118.9(6) 120.9(4) 123.4(7) 118.5(9) 0.0c 90.0c 90.0c 90.0c 90.0c 0.0 4.84

200.8(8) 139.0(5) 134.6(5) 149.1(8) 142.2(4) 139.7(4) 139.8(4) 140.7(4) 132.3(4) 132.5(4) 105.7(12) 120.5(5) 118.7(9) 120.8(5) 121.3(9) 118.5(11) -1.9(25) 164.2(28) 54.5(32) 109.6(26) 91.0(51) -6.7(81) 4.31

199.4(8) 137.8(5) 133.5(5) 145.4(10) 142.5(4) 140.0(4) 140.2(4) 141.0(4) 132.5(5) 132.7(5) 112.1(7) 120.5(4) 118.7(8) 120.8(4) 123.1(8) 119.3(11) 0.0c 90.0c 90.0c 90.0c 90.0c 0.0 5.85

200.0(8) 138.1(5) 133.8(5) 147.9(8) 142.6(4) 140.1(4) 140.2(4) 141.0(4) 132.6(5) 132.7(5) 111.1(7) 120.5(3) 118.8(6) 120.8(4) 123.2(6) 118.5(9) 0.0c 90.0c 90.0c 90.0c 90.0c 0.0 4.57

a ra parameters. Parenthesized values are estimated error limits defined as 2.5(σ2lsq + (0.001r)2)1/2, for distances and as 2.5σlsq for angles, where σlsq is one standard deviation taken from the least-squares refinements using a diagonal weight matrix and, the 0.001r term represents 0.1% uncertainty in the electron wavelength. The errors are in the units of the last digits. b Dependent parameter. c Fixed to their DFT calculated values. d See Figure 1 for definition. Zn1N2A1 ) a1 (a1 is negative if the Zn atom is below the plane through the cavity N atoms), Zn1N2A2 ) a2 + 90°, N2A1A3 ) a3 + 90°, A1A2A3 ) a4 + 90°. e The dihedral angle Zn1A6A7N54 is 0°, which gives the six-member ring consisting of the atoms Zn1, N2, calc 2 obs 2 C7, N54, C8, N3 a boat conformation. f R ) {∑sw(Iobs s - Is ) }/{∑sw(Is ) } × 100%, where w is a weight function usually equal to 1 and I is the molecular modified intensity.

function of the distance, h, above the plane through the atoms N2 and N4 at the DFT/6-311++G** level. All structural parameters were optimized for fixed values of h. Starting with C4V symmetry created serious convergence problems but reduc-

ing the symmetry to C2V removed these problems. The potential energy function is shown in Figure 6. This clearly shows that the movement of the Zn atom out of the symmetry plane goes through C2V symmetry and not C4V symmetry. C2V symmetry

12016 J. Phys. Chem. A, Vol. 111, No. 47, 2007

Strenalyuk et al.

TABLE 5: Comparison of the Structure Parameters for HPc-Zna GED8

GED, present work reb

uexp

uh1b

Zn1-N2 N2-C6 C6-N57 C6-C14 C14-C15 C14-C22 C22-C30 C30-C31 C22-H38 C30-H46

198.2(6) 138.0(5) 133.6(5)d 145.5(6) 140.7(5) 139.0(5)d 138.9(5)d 140.2(5)d 106.9(12) 107.0(12)d

9.4(5) 5.9(2) 5.5(2) 6.1(2) 5.7(2) 5.6(2) 5.6(2) 5.7(2) 8.2(2) 8.3(2)

8.2 5.2 4.8 5.5 5.0 4.9 4.9 5.0 7.6 7.6

C6N2C7 C15C14C22 C14C22C30 C22C30C31 C14C22H38 C22C30H46 R,g %

108.6(5) 121.1(3) 117.7(8)e 121.2(4)e 120.7f 119.6f 5.10

parameter

cc-pVTZ re

kh1b Bond Lengths 0.169 0.069 0.029 0.012 0.003 -0.002 0.049 0.003 0.073 0.077

199.1 137.3 133.0 146.1 141.0 139.5 139.4 140.8 108.6 108.7

r 195.4(29)c 137.4(23) 142.0(27) 155(10) 137.0(14) 136.9 139.9 112.2(82) av

u 5.1(44)

X-ray13 198.0 136.9 133.3 145.5 140.0 139.3 139.1 139.6

Bond Angles 109.4 121.1 117.8 121.2 120.7 119.6

106.9(46) 117.2(22) 121.6

109.1 121.3 117.3 121.4

4.68

Parenthesized values are estimated error limits defined as 2.5(σ lsq + (0.001r) ) , for distances and as 2.5σlsq for angles and amplitudes, where σlsq is one standard deviation taken from the least-squares refinements using a diagonal weight matrix and, the 0.001r term represents 0.1% uncertainty in the electron wavelength. The errors are in the units of the last digits. b Curvilinear treatment of the vibrating atoms (by SHRINK program). c Uncertainties are estimates of 2σ. d This value was calculated by using constraints between parameters (see text). e Dependent parameter. f This calc 2 obs 2 parameter was fixed at its B3LYP/cc-pVTZ calculated values. g R ) {∑sw(Iobs s - Is ) }/{∑sw(Is ) } × 100%, where w is a weight function usually equal to 1 and I is the molecular modified intensity. a

2

2 1/2

TABLE 6: Calculated and Experimental Parameters of FPc-Zna GED, present work parameter

reb

Zn1-N2 N2-C6 C6-N57 C6-C14 C14-C15 C14-C22 C22-C30 C30-31C C22-F38 C30-F46

199.4(8) 137.9(5) 133.2(6)c 147.3(8) 142.1(4) 139.6(4)c 139.9(4)c 141.7(4)c 131.8(5) 132.0(5)c

C7N2C6 C15C14C22 C14C22C30 C22C30C31 C14C22F38 C22C30F46 R,e %

111.1(7) 120.5(3) 118.8(6)d 121.4(4)d 123.2(6) 118.5(9) 4.57e

uexp

uh1b

Bond Length 10.0(7) 8.4 5.3 5.3 4.8 4.8 5.6 5.6 5.2 5.2 5.0 5.0 5.0 5.0 5.0 5.0 4.9 4.9 4.9 4.9

kh1b 0.282 0.097 -0.039 -0.002 -0.122 -0.041 0.091 0.103 -0.079 -0.027

cc-pVTZ re 200.4 136.9 132.5 146.1 141.5 139.0 139.1 139.9 133.2 133.4

Bond Angles 110.4 120.5 120.8 120.9 122.6 120.2

a Parenthesized values are estimated error limits defined as 2.5(σ2lsq + (0.001r)2)1/2, for distances and as 2.5σlsq for angles and amplitudes, where σlsq is one standard deviation taken from the least-squares refinements using a diagonal weight matrix and, the 0.001r term represents 0.1% uncertainty in the electron wavelength. The errors are in the units of the last digits. b Curvilinear treatment of the vibrating atoms (by SHRINK program). c This value was calculated by using constraints between parameters (see text). d Dependent parameter. e R obs 2 2 ) {∑sw(Iobs - Icalc s s ) }/{∑sw(Is ) } × 100%, where w is a weight function usually equal to 1 and I is the molecular modified intensity.

allows two different N-Zn bond lengths and the splitting of these bond lengths, ∆r1 ) r(N2-Zn) - 200.11 and ∆r2 ) r(N3-Zn) - 200.11 are also shown in Figure 6. When the potential energy function is obtained the average height 〈h〉 ) ∫h exp(-∆E/RT) dh × [∫exp(-∆E/RT) dh]-1 can be calculated.

Figure 4. Radial distribution (RD) curves for HPc-Zn. The experimental points (dots) are the average of the four curves shown in Figure 2, and theoretical points are used for the unobserved region s < 17.50 nm-1. The damping coefficient for the RD function is 35 pm2. The curve below is the difference curve.

This gives 〈hN2N3〉 ) 14.5 pm for the height from the Zn atom to the plane containing atoms N2 and N4, and correspondingly 〈hN3N5〉 ) 6.8 pm. This explains the results obtained in the first GED investigation of HPc-Zn where h ) 34(13) pm.7 However, it is not clear if shrinkage corrections have been applied in the second GED investigations of HPc-Zn. Several IR and Raman investigations12,14,51-59 on HPc-Zn have been made, and assignments of the vibrational frequencies have been discussed. Calculated frequencies at different level of theory have also been carried out and assignments to observed frequencies have been attempted.12,56,57 Our high-level B3LYP/ 6-311++G** force field calculations together with calculated IR and Raman intensities give a good basis for assignments of the observed frequencies. A comparison of the observed vibrational frequencies from several different investigations have

Molecular Structures of Phthalocyaninatozincs

Figure 5. Radial distribution (RD) curves for FPc-Zn. The experimental points (dots) are the average of the four curves shown in Figure 3, and theoretical points are used for the unobserved region s < 17.50 nm-1. The damping coefficient for the RD function is 35 pm2. The curve below is the difference curve.

Figure 6. ∆E is the increase in energy as a function of the hight of the Zn atom, h, above the plane through the nitrogen atoms N2 and N4 as calculated by B3LYP/ 6-311++G**. For h ) 0.0 pm the symmetry is D4h and for all other values of h is the symmetry C2V. ∆r1 ) r(Zn1-N2) - r(Zn1-N2)h)0 and ∆r2 ) r(Zn1-N3) - r(Zn1N3)h)0. If the movement of the Zn atom had gone through C4V symmetry ∆r1 would be equal to ∆r2.

been made and summarized in Table S3 and given as Supporting Information. The agreement between calculated and the proposed observed frequencies is very good. There is little information about vibrational frequencies for FPc-Zn; however, the B3LYP/6-311++G** calculated frequencies together with the calculated IR and Raman intensities should be a good basis for the assignment of the vibrational spectra for FPc-Zn. Experimentally Derived Structure. As outlined above, it is important to understand the meaning of the determined structures and the importance of including the shrinkage effect particularly for planar molecules. The shrinkage effect can be included by using either a rectilinear approach (kh0) or a curvilinear approach (kh1). The ra structure is transferred to a kh0 structure or a kh1 structure, which represents a geometrically consistent structure of averaged nuclear positions that is transferred back again to an ra structure for a proper calculation of the molecular intensities. The re structure can be estimated from the ra structure by using the ra - re correction term

J. Phys. Chem. A, Vol. 111, No. 47, 2007 12017 calculated from the SHRINK program using the B3LYP/6311G++G** force field. In Tables 3 and 4 are the experimental ra structures given without correcting for shrinkage, correcting for shrinkage using kh0 values and kh1 values. As expected, the GED ra structure without shrinkage correction gives a nonplanar structure. When shrinkage is included, the h parameter or angle a1 is approaching zero and start oscillating. Therefore D4h symmetry was adopted in the final refinements. This strongly supports that the equilibrium structure has D4h symmetry. It should be noted that the goodness-of-fit parameter, R, is slightly larger when shrinkage is included, compared to when shrinkage is not included. There might be two reasons for this, first that the C4V model has more adjustable parameters, but more likely that the correction terms are based on the assumption that the atoms perform small amplitude motions about their equilibrium position, and this is probably not a very good approximation. However, the shrinkage corrected structure represents the best geometrical structure obtainable at the moment. It should also be noted that the very long C14-C15 bond length of 155(10) pm obtained in the previous GED investigations is not reproduced in our GED investigation. This is most likely due to the limited experimental data obtained for HPc-Zn. It is of interest to compare the molecular structure of HPcZn and FPc-Zn, and in particular the changes in the benzo subunits. The C-C bond length in benzene10 is 1.3975 pm and in hexafluorobenzene60,61 139.4(7) pm and 139.0 pm, a shortening of the C-C bond length is indicated. Perfluorination also causes shortening of the CdC bond length in ethylene.62 A possible explanation for this effect is that the electronegative F atom draws electron density from the p-orbitals of the C atom, which leaves more s-character in the bonds between the C atoms. Because s-orbitals are more compact than p-orbitals the C-C bond lengths shortens.61 The average C-C bond length calculated from B3LYP/cc-pVTZ given in Table 2 is 139.67 and 139.60 pm, respectively, for HPc-Zn and FPc-Zn. No overall shrinkage of the benzo subunits is predicted, where actually an increase is predicted for the C14-C15 bond length. However, comparing the experimental results indicates a small shortening of the bond length with 0.8 pm. This confirms both theoretically and experimentally that there are small changes in the molecular geometry with perfluorination of the benzo subunits. The only significant change is the increase of 2.5° for the C14-C22-X38 bond angle, which is expected due to the electrostatic repulsion between the electron negative N57 and F38 atoms. Concluding Remarks. Our GED investigation is consistent with both HPc-Zn and FPc-Zn having D4h equilibrium geometries as also is predicted by high-level quantum chemical calculations. There is a good agreement between the geometry obtained from the DFT calculations, our GED investigation, and the X-ray structure for HPc-Zn. There is also good agreement between the geometry obtained from the DFT calculations and our GED investigation for FPc-Zn. Calculated frequencies are valuable tools to assign the vibrational frequencies. It seems as the movement of the Zn atom out of the symmetry plane goes through C2V symmetry and not C4V symmetry. Both HPc-Zn and FPc-Zn are not rigid molecules but flexible molecules performing large movement about their equilibrium geometry. There are small structural differences between HPc-Zn and FPc-Zn; however, there are indications that the benzo subunits shrinks slightly when substituted with F atoms. Acknowledgment. S. Gundersen is acknowledged for workup of the experimental data. The University of Oslo is thanked for a generous amount of computer time. T.S. thanks the

12018 J. Phys. Chem. A, Vol. 111, No. 47, 2007 International Student Quota Program of the University of Oslo for financial support. Supporting Information Available: Tables of Cartesian coordinates, frequencies, and IR and Raman intensities. This material is available free of charge via the Internet at http:// pubs.acs.org. References and Notes (1) Kadish, K. M.; Smith, K. M.; Guilard R. Application of Phthalocyanines; The Porphyrin Handbook, Vol. 19; Academic Press: San Diego, CA, 2003. (2) Kadish, K. M.; Smith, K. M.; Guilard, R. Phthalocyanines: Properties and Materials; The Porphyrin Handbook, Vol. 17; Academic Press: San Diego, CA, 2003. (3) Kadish, K. M.; Smith, K. M.; Guilard, R. Phthalocyanines: Spectroscopic and Electrochemical Characterization; The Porphyrin Handbook, Vol. 16; Academic Press: San Diego, CA, 2003. (4) Engel, M. K. Single-crystal structures of phthalocyanine complexes and related macrocycles; The Porphyrin Handbook, Vol. 20; Academic Press: San Diego, CA, 2003. (5) McKeown, N. B. Phthalocyanine Materials: Structure, Synthesis and Function; Cambridge University Press: Cambridge, U.K., 1998. (6) Leznoff, C. C.; Lever, A. B. P. Phthalocyanines - Properties and Applications; VCH Publishers: New York; Vol. 1 (1989), Vol. 2 (1992), Vol. 3 (1993). Vol. 4 (1996). (7) Mihill, A.; Buell, W.; Fink, M. J. Chem. Phys. 1993, 99, 6416. (8) Ruan, C.-y.; Mastryukov, V.; Fink, M. J. Chem. Phys. 1999, 111, 3035. (9) Mastryukov, V.; Ruan, C. y.; Fink, M.; Wang, Z.; Pachter, R. J. Mol. Struct. 2000, 556, 225. (10) Gundersen, S.; Samdal, S.; Strand, T. G.; Volden, H. V. J. Mol. Struct. 2007, 832, 164. (11) Samdal, S.; Volden, H. V.; Ferro, V. R.; Garcia de la Vega, J. M.; Gonzalez-Rodriguez, D.; Torres, T. J. Chem. Phys. 2007, 111, 4542. (12) Tackley, D. R.; Dent, G.; Smith, W. E. Phys. Chem. Chem. Phys. 2000, 2, 3949. (13) Scheidt, W. R.; Dow, W. J. Am. Chem. Soc. 1977, 99, 1101. (14) Tackley, D. R.; Dent, G.; Smith, W. E. Phys. Chem. Chem. Phys. 2001, 3, 1419. (15) Porezag, D.; Pederson, M. R. Phys. ReV. B 1996, 54, 7830. (16) Halls, M. D.; Schlegel, H. B. J. Chem. Phys. 1998, 109, 10587. (17) Johnson, B. G.; Florian, J. Chem. Phys. Lett. 1995, 247, 120. (18) Halls, M. D.; Schlegel, H. B. J. Chem. Phys. 1999, 111, 8819. (19) Zeil, W.; Haase, J.; Wegmann, L. Z. Instrumentenkunde 1966, 74, 84. (20) Martinsen, K.-G., Volden, H. V. Annual Report 1993 from the Norwegian GED-Group. 1993. (21) Martinsen, K.-G. Dissertations submitted to the Faculty of Matematics and Natural Sciences, University of Oslo, 1997. (22) Gundersen, S.; Samdal, S.; Seip, R.; Strand, T. G. J. Mol. Struct. 2004, 691, 149. (23) Ross, A. W.; Fink, M.; Hilderbrandt, R. International Tables for Crystallography; Kluwer Academic Publishers: Dordrecht, The Netherlands, 1992. (24) Gundersen, G., Samdal, S. Annual Report 1976 from the Norwegian GED-Group. 1976. (25) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels,

Strenalyuk et al. A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; 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.; Gonzalez, C.; Pople, J. A. Gaussian 03, revision B.03; Gaussian, Inc.: Pittsburgh, PA, 2003. (26) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (27) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785. (28) Ferro, V. R.; Poveda, L. A.; Claessens, C. G.; Gonzalez-Jonte, R. H.; Garcia de la Vega, J. M. Int. J. Quantum Chem. 2003, 91, 369. (29) Gong, X. D.; Xiao, H. M.; Gao, P.; Tian, H. THEOCHEM 2002, 587, 189. (30) Ferro, V. R.; Garcia de la Vega, J. M.; Claessens, C. G.; Poveda, L. A.; Gonzalez-Jonte, R. H. J. Porphyrins Phthalocyanines 2001, 5, 491. (31) Ferro, V. R.; Garcia de la Vega, J. M.; Gonzalez-Jonte, R. H.; Poveda, L. A. THEOCHEM 2001, 537, 223. (32) Ryeng, H.; Ghosh, A. J. Am. Chem. Soc. 2002, 124, 8099. (33) Vangberg, T.; Lie, R.; Ghosh, A. J. Am. Chem. Soc. 2002, 124, 8122. (34) Wasbotten, I. H.; Wondimagegn, T.; Ghosh, A. J. Am. Chem. Soc. 2002, 124, 8104. (35) Ghosh, A. J. Porphyrins Phthalocyanines 2000, 4, 380. (36) Nguyen, K. A.; Pachter, R. J. Phys. Chem A 2000, 104, 4549. (37) Parusel, A. B. J.; Ghosh, A. J. Phys. Chem. A 2000, 104, 2504. (38) Parusel, A. B. J.; Wondimagegn, T.; Ghosh, A. J. Am. Chem. Soc. 2000, 122, 6371. (39) Hariharan, P. C.; Pople, J. A. Mol. Phys. 1974, 27, 209. (40) Krishnan, R.; Binkley, J. S.; Seeger, R.; Pople, J. A. J. Chem. Phys. 1980, 72, 650. (41) Frisch, M. J.; Pople, J. A.; Binkley, J. S. J. Chem. Phys. 1984, 80, 3265. (42) Dunning, T. H., Jr. J.Chem. Phys. 1989, 90, 1007. (43) Balabanov, N. B.; Peterson, K. A. J. Chem. Phys. 2005, 123, 064107/1. (44) Seip, H. M. Specialist Periodical Reports: Molecular Structure by Diffraction Methods; Sim, G. A., Sutton, L. E., Eds.; The Chemical Society, London, 1973; Vol. 1, p 7. (45) Hargittai, I., Hargittai, M., Eds. Stereochemical Applications of GasPhase Electron Diffraction, Pt. A: The Electron Diffraction Technique. Methods Stereochem. Anal. 1988, 10, 1988. (46) Hedberg, L.; Mills, I. M. J. Mol. Spectrosc. 1993, 160, 117. (47) Hedberg, L.; Mills, I. M. J. Mol. Spectrosc. 2000, 203, 82. (48) Sipachev, V. A. THEOCHEM 1985, 22, 143. (49) Sipachev, V. A. J. Mol. Struct. 2001, 567-568, 67. (50) Gundersen, G.; Samdal, S.; Seip, H.-M.; Strand, T. G. Annual Report 1977, 1980, 1981 from the Norwegian Gas Electron Diffraction Group. (51) Kobayashi, T. Spectrochim. Acta 1970, 26, 1313. (52) Kobayashi, T.; Kurokawa, F.; Uyeda, N.; Suito, E. Spectrochim. Acta 1970, 26, 1305. (53) Terzian, G.; Moubaraki, B.; Mossoyan-Deneux, M.; Benlian, D. Spectrochim. Acta. 1989, 45A, 675. (54) Aroca, R.; Zeng, Z. Q.; Mink, J. J. Phys. Chem. Solids 1990, 51, 135. (55) Gobernado-Mitre, I.; Aroca, R.; DeSaja, J. A. Spectrochim. Acta 1994, 50A, 1243. (56) Ding, H.; Wang, S.; Xi, S. J. Mol. Struct. 1999, 475, 175. (57) Gladkov, L. L.; Shkirman, S. F.; Sushko, N. I.; Konstantinova, V. K.; Sokolov, N. A.; Solovyov, K. N. Spectrosc. Lett. 2001, 34, 709. (58) Seoudi, R.; El-Bahy, G. S.; El Sayed, Z. A. J. Mol. Struct. 2005, 753, 119. (59) Szybowicz, M.; Runka, T.; Drozdowski, M.; Bala, W.; Wojdyla, M.; Grodzicki, A.; Piszczek, P.; Bratkowski, A. J. Mol. Struct. 2007, 830, 14. (60) Almenningen, A.; Bastiansen, O.; Seip, R.; Seip, H. M. Acta Chem. Scand. 1964, 18, 2115. (61) Braden, D. A.; Hudson, B. S. J. Phys. Chem. A 2000, 104, 982. (62) Carlos, J. L., Jr.; Karl, R. R., Jr.; Bauer, S. H. J. Chem. Soc., Faraday Trans. 2 1974, 70, 177.