Energetics of Saddling versus Ruffling in Metalloporphyrins: Unusual Ruffled Dodecasubstituted Porphyrins.

Presented herein is a first major density functional theory (BP86/D3/STO-TZ2P) survey of the energetics of saddling versus ruffling for a wide range of dodecasubstituted metalloporphyrins with M = Ni, Cu, Zn, Pd, and Pt. For the majority of X8TPP (i.e., β-octasubstituted-meso-tetraphenylporphyrin), the calculations indicated a clear preference for the saddled conformation, consistent with a large body of experimental data. The preference for the saddled conformation relative to the ruffled conformation was found to vary from about ∼0.3-0.4 eV for Me8TPP derivatives up to 1 eV for I8TPP and (CF3)8TPP derivatives. For X = Ph, that is, dodecaphenylporphyrins, the saddled and the ruffled conformation are almost equienergetic, with even a slight preference for the ruffled conformation in some cases. This finding provides a satisfactory explanation for the X-ray crystallographic observation of both saddled and ruffled conformations for dodecaphenylporphyrin complexes as well as for spectroscopic evidence for conformational mobility of these complexes in solution. The calculations also indicate near-equienergetic saddled and ruffled conformations for meso-tetraacetylenyltetrabenzoporphyrins, again consonant with key crystallographic findings. By and large, both the energetics and nonplanar distortions of the metalloporphyrin derivatives correlated well with the Charton and Sterimol B1 steric parameters of the peripheral substituents.

Introduction

Despite their aromatic character, metalloporphyrins adopt a variety of nonplanar conformations such as the ruffled, saddled, domed, waved, and various intermediate conformations.[1−3] These deformations result from such factors as a sterically hindered set of substituents, a coordinated atom that is too small or too large, and specific metal–porphyrin orbital interactions. Of the various deformations, ruffling and saddling are the most common. Ruffling, where the meso carbons are alternately displaced above and below the mean porphyrin plane, commonly occurs for a coordinated atom/ion that is too small for a planar porphyrin; a common example is the Ni(II) ion.[4−8] Certain sterically hindered substitution patterns, such as four bulky meso substituents, also result in ruffling. meso-Tetraisopropyl-[9−11] and meso-tetrakis(t-butyl)porphyrin[12] derivatives provide good examples of such ruffled porphyrins. Saddling, where the pyrrole rings are alternately tilted above and below the mean porphyrin plane, is most commonly associated with dodecasubstituted porphyrin derivatives, where it provides relief from peripheral steric overcrowding. Interestingly, the ruffled conformation has been observed for a handful of dodecasubstituted porphyrin derivatives. Thus, Smith and co-workers reported X-ray structures for both the ruffled (CCDC: XAWRUI) and saddled (CCDC: TEZXEB) conformations of Ni dodecaphenylporphyrin[13,14] and underscored the flexibility of the system. There is also significant spectroscopic evidence that Ni dodecaphenylporphyrin is conformationally mobile in solution.[15] A ruffled conformation has been found for a Pt β-octaalkyl-meso-tetraacetylenyl-porphyrin (CCDC: LUTYOO, Figure ).[16] Unlike for ruffled porphyrins[17−19] and hydroporphyrins,[20−23] few major quantum chemical studies have addressed the question of energetics associated with saddling in porphyrins,[24,25] which has left us relatively in the dark about the relative energies of the saddled versus ruffled conformations of various dodecasubstituted metalloporphyrins. Detailed information on this subject should not only allow for a better appreciation of metalloporphyrin structural chemistry and spectroscopy[1,2,15,25−27] but also stimulate the use of nonplanar porphyrins to create novel supramolecular and nanoscale structures.[28,29]

Figure 1

Representative diagrams of a (a) ruffled and (b) saddled NiDPP and (c) ruffled Pt acetylenyl porphyrins. In (a) and (c), the substituents at C2, C3, C12, and C13 have been removed for clarity.

Prompted by the above considerations, we undertook a dispersion-corrected density functional theory (DFT) study (BP86-D3/STO-TZ2P) of a wide range of dodecasubstituted metalloporphyrins. As shown in Figure , three broad classes of complexes were examined (a) X8TPP (TPP = tetraphenylporphyrin), where the meso substituent is phenyl and the β substituent X = Me,[30−32] Cl,[33−36] Br,[37−39] Ph,[13,14,40,41] I,[42] and CF3;[43] (b) Y4TBP (TBP = tetrabenzoporphyrin), where the meso substituent Y = C≡C-SiMe3 (hereafter abbreviated as A) and Ph,[44−47] and (c) X8TAP (TAP = meso-tetrakis(trimethylsilylacetylenyl)porphyrin), where X = Me[16] and Br. For each porphyrin ligand, five different divalent metals were examined, namely, Ni,[13,14,16,35,36,38,39,42,43,45,46] Cu,[33,38,40,42,44] Zn,[13,34,37,41] Pd, and Pt.[16,47] For each metalloporphyrin, the ruffled and saddled conformations were optimized. Table lists various geometry parameters and the energy difference between the saddled and ruffled conformations (Esadd – Eruff). The key geometry parameters of interest are the ruffling (ψ) and saddling dihedrals (χ), defined in Figure , and the out-of-plane displacements of meso (z), α (zα), and β (zβ) carbons.

Figure 2

Molecules studied in this work.

Table 1

Electronic Energy Differences (eV) between the Saddled and Ruffled Optimized Geometries, M–N Distances (Å), Ruffling (Ψ) and Saddling (χ) Angles (deg), and α, β and meso Carbon Displacements (Å) above the Mean N4 Plane

		ruffled			saddled
complex	M	(M–N)av	Ψ	zmeso	(M–N)av	χ	zα	zβ	Esadd – Eruff
TPP	Cu	2.008	1.1	0.019	2.012	15.7	0.075	0.244	–0.05
	Ni	1.933	31.3	0.514	1.945	14.7	0.239	0.573	–0.03
	Pd	2.025	1.0	0.017	2.027	13.6	0.083	0.248	–0.03
	Pt	2.024	0.7	0.013	2.026	11.3	0.070	0.208	–0.03
	Zn	2.043	0.3	0.005	2.048	16.3	0.033	0.163	–0.06
Me8TPP	Ni	1.890	56.0	0.890	1.903	36.0	0.507	1.223	–0.32
	Cu	1.979	43.0	0.695	1.987	44.9	0.429	1.106	–0.40
	Zn	2.040	31.5	0.517	2.033	45.4	0.384	1.018	–0.38
	Pd	2.014	37.5	0.608	2.012	43.5	0.407	1.056	–0.38
	Pt	2.015	36.9	0.602	2.010	39.0	0.401	1.020	–0.36
Cl8TPP	Ni	1.894	56.0	0.893	1.912	35.5	0.492	1.193	–0.42
	Cu	1.990	40.6	0.656	1.996	44.0	0.412	1.069	–0.47
	Zn	2.060	22.1	0.362	2.037	38.3	0.365	0.944	–0.37
	Pd	2.027	32.1	0.520	2.018	41.7	0.388	1.009	–0.43
	Pt	2.028	30.8	0.501	2.015	35.7	0.381	0.965	–0.39
Br8TPP	Ni	1.888	59.1	0.932	1.906	38.2	0.521	1.261	–0.57
	Cu	1.979	46.2	0.740	1.992	48.4	0.446	1.154	–0.65
	Zn	2.045	33.4	0.541	2.034	44.2	0.407	1.055	–0.58
	Pd	2.019	39.2	0.629	2.016	46.5	0.424	1.102	–0.62
	Pt	2.020	38.2	0.614	2.011	40.6	0.420	1.065	–0.58
I8TPP	Ni	1.873	64.7	1.010	1.897	44.3	0.557	1.357	–0.81
	Cu	1.955	55.3	0.874	1.987	56.8	0.490	1.274	–0.92
	Zn	2.017	46.3	0.740	2.028	54.2	0.457	1.196	–0.93
	Pd	1.999	49.4	0.785	2.011	54.7	0.470	1.225	–0.95
	Pt	2.002	48.3	0.770	2.005	48.5	0.468	1.193	–0.90
(CF3)8TPP	Ni	1.842	73.8	1.167	1.880	67.9	0.652	1.615	–0.56
	Cu	1.919	67.0	1.071	1.984	88.1	0.599	1.586	–0.88
	Zn	1.977	60.8	0.981	2.026	87.7	0.575	1.541	–1.00
	Pd	1.966	62.2	1.003	2.000	80.1	0.580	1.522	–0.91
	Pt	1.973	61.0	0.986	1.993	62.7	0.547	1.398	–0.85
DPP	Ni	1.860	64.9	1.048	1.933	25.9	0.423	1.011	0.15
	Cu	1.946	53.4	0.887	1.995	30.3	0.336	0.850	0.02
	Zn	2.013	41.2	0.710	2.048	22.9	0.240	0.615	0.01
	Pd	1.990	46.5	0.786	2.024	26.6	0.536	0.768	0.00
	Pt	1.994	45.3	0.770	2.023	22.0	0.283	0.700	0.05
Ph4TBP	Ni	1.889	60.2	0.946	1.928	25.2	0.485	1.132	–0.15
	Cu	1.983	47.3	0.753	1.998	28.6	0.415	1.003	–0.25
	Zn	2.049	34.9	0.563	2.045	26.5	0.368	0.894	–0.26
	Pd	2.018	41.6	0.665	2.024	31.0	0.385	0.954	–0.24
	Pt	2.018	40.8	0.654	2.022	28.3	0.379	0.927	–0.24
A4TBP	Ni	1.890	60.9	1.005	1.893	13.5	0.356	0.808	0.02
	Cu	1.987	46.9	0.805	1.987	20.0	0.417	0.960	–0.03
	Zn	2.059	31.5	0.576	2.039	21.0	0.412	0.949	–0.10
	Pd	2.021	40.8	0.716	2.020	24.3	0.399	0.944	–0.02
	Pt	2.021	40.4	0.711	2.019	22.6	0.391	0.918	–0.01
Me8TAP	Ni	1.890	55.7	0.931	1.903	21.5	0.522	1.184	0.18
	Cu	1.981	41.8	0.731	1.982	26.1	0.442	1.038	0.05
	Zn	2.047	27.1	0.506	2.031	25.9	0.390	0.929	0.00
	Pd	2.016	35.6	0.636	2.014	28.7	0.404	0.974	0.07
	Pt	2.016	35.4	0.636	2.015	26.4	0.395	0.943	0.08
Br8TAP	Ni	1.885	59.8	1.019	1.900	28.0	0.568	1.306	0.25
	Cu	1.975	47.4	0.845	1.979	29.3	0.431	1.036	0.03
	Zn	2.040	35.0	0.661	2.027	31.9	0.455	1.085	–0.01
	Pd	2.013	41.1	0.752	2.015	35.7	0.457	1.110	0.06
	Pt	2.013	40.5	0.745	2.013	31.5	0.448	1.073	0.10

Figure 3

Definition of ruffling (χ) and saddling (ψ) dihedrals.

Results and Discussion

Table shows that the majority of the complexes prefer a saddled conformation by a clear margin of energy, consistent with the large body of available crystallographic data.[1] Also, as shown in Figure , with the possible exception of Ni(II), the smallest metal ion considered, Esadd – Eruff is essentially independent of the metal, for the metal ions considered.

Figure 4

Energy difference (eV) between saddled and ruffled conformations as a function of metal ion ionic radius. Ionic radii (Å): Cu 0.71, Ni 0.63, Pd 0.78, Pt 0.74, and Zn 0.74. The horizontal dotted lines represent the average value off Esadd – Eruff for Cu, Pt, Zn, and Pd, underscoring that only Ni deviates significantly from this value.

For the X8TPP complexes and for a given metal ion, Esadd – Eruff exhibits a strong, linear dependence on the steric bulk of the β substituent X, as measured by either the Charton[48−51] or the Sterimol B1[52−57] parameters (Figure ), with one key exception, X = Ph. Thus, the preference for the saddled conformation relative to the ruffled conformation ranges from about ∼0.3–0.4 eV for Me8TPP derivatives up to 1 eV for I8TPP and (CF3)8TPP derivatives. Interestingly, for X = Ph, that is, dodecaphenylporphyrins, the saddled and the ruffled conformations are almost equienergetic, with even a slight preference for the ruffled conformation in some cases. In other words, the phenyl groups in dodecaphenylporphyrins exert a much lower steric effect than that implied by their Charton and Sterimol B1 parameters. The anomalously low steric effects of the phenyl groups are most reasonably ascribed to the manner in which they stack in a circular arrangement around the porphyrin periphery. These results explain the experimental observation of both the saddled and ruffled conformations of nickel dodecaphenylporphyrin.[15]

Figure 5

Energy difference (eV) between saddled and ruffled conformations of X8TPP (X Me, Cl, Br, I, CF3, or Ph) complexes of different metal ions as a function of the Charton (H 0.00, Cl 0.55, Br 0.65, I 0.78, Me 0.52, CF3 0.90, and Ph 0.57) and Sterimol B1 (H 1.00, Me 1.52, Cl 1.80, Br 1.95, I 2.15, CF3 1.98, and Ph 1.71) steric parameters.

The acetylenyl-substituted metalloporphyrins considered here, Y4TBP complexes with Y = C≡C-SiMe3 and X8TAP complexes, behave similar to dodecaphenylporphyrins in that they too exhibit essentially equienergetic saddled and ruffled conformations. We view this finding to be quite reasonable because acetylenyl and phenyl substituents are expected to exhibit similar minimum widths and similar Sterimol B1 parameters. Energetics considerations thus provide a rationale for the experimental observation of a ruffled platinum β-octaalkyl-meso-tetraacetylenylporphyrin.[16]

Table S1, Supporting Information shows key experimental structural data for selected saddled porphyrins relevant to this study. The reader may verify that the present calculations generally do an excellent job of reproducing the experimentally observed saddling distortions. For the full set of complexes studied, the degree of saddling (as measured by either zβ or χ, Figure ) or ruffling (as measured by either z or ψ, Figure ) also shows an excellent correlation with the Charton parameter and a somewhat worse correlation with Sterimol B1.

Figure 6

Degree of saddling (zβ) of X8TPP (X Me, Cl, Br, I, CF3, or Ph) complexes of different metal ions as a function of the Charton (H 0.00, Me 0.52, Cl 0.55, Ph 0.57, Br 0.65, I 0.78, and CF3 0.90) and Sterimol B1 (H 1.00, Me 1.52, Ph 1.71, Cl 1.80, Br 1.95, I 2.15, and CF3 1.98) steric parameters of X.

Figure 7

Degree of ruffling (z) of X8TPP (X Me, Cl, Br, I, CF3, or Ph) complexes of different metal ions as a function of the Charton (H 0.00, Me 0.52, Cl 0.55, Ph 0.57, Br 0.65, I 0.78, and CF3 0.90) and Sterimol B1 (H 1.00, Me 1.52, Ph 1.71, Cl 1.80, Br 1.95, I 2.15, and CF3 1.98) steric parameters of X.

Conclusions

Dispersion-corrected DFT calculations indicate a clear preference for the saddled conformation for the majority of X8TPP complexes, consistent with a large body of experimental data. For X8TPP complexes, where X = Ph (i.e., dodecaphenylporphyrins), or Y4TBP complexes, where Y = C≡C-SiMe3, as well as for X8TAP complexes, however, the saddled and ruffled conformations are found to be nearly equienergetic, which explains the experimental observation of the ruffled conformation for a few dodecasubstituted metalloporphyrins. In general, the degree of saddling or ruffling exhibits a clear correlation with the Charton or Sterimol B1 steric parameters of the substituents in question, except for dodecaphenylporphyrins and acetylenyl-substituted porphyrins. For these compounds, both the Charton and Sterimol B1 parameters appear to greatly overestimate the actual steric effects exerted by phenyl and acetylenyl substituents.

Experimental Section

All DFT calculations were carried out with the ADF (Amsterdam Density Functional) 2013 program system,[58] the BP86 functional in conjunction with Grimme’s D3[59] dispersion corrections, Slater-type TZ2P basis sets, a fine mesh for numerical integration, and full geometry optimizations with tight convergence criteria. D2d or D2 symmetry constraints were used to derive the saddled and ruffled optimized conformations for each metalloporphryin. Free energy differences between the two conformations were calculated for several selected complexes and were found to be very similar (to well within 0.1 eV) to the electronic energy differences, and accordingly only the latter have been reported in Table .