6068
Journal of the American Chemical Society
/ 100:19 / September 13, 1978
Synthesis, Characterization, and Structure of Tri-p-halogeno-hexacarbonyldirhenate( I) Salts of Monocationic Porphyrin Acids C. P. Hrung,la M. Tsutsui,*la D. L. Cullen,lb E. F. Meyer, Jr.,*b and C. N. Morimoto'c Contributionfrom the Department of Chemistry and Department of Biophysics and Biochemistry, Texas Agricultural Experiment Station, Texas A & M Unicersity, College Station, Texas 77843, and Control Data Corporation, Sunnyuale, California 94086. Receiued February 3, 1978
Abstract: Two new compounds, ~-(2,3,7,8,12,13,17,18-octaethylporphyrin)-bis[tricarbonylrhenium(I)],(C36H44N4)[Re(C0)3]2, and a salt of a monocationic porphyrin acid, octaethylporphinium tri-pchloro-hexacarbonyldirhenate(l), (C36H4,N4)+[Re2(COj6cl3]-, have been synthesized by the reaction of a 2.1 mole ratio of rhenium pentacarbonyl chloride and octaethylporphyrin in refluxing decalin under argon. The structure of the latter usas determined from three-dimensional diffractometer data. A total of I O 792 independent reflections were measured. The compound crystallizes in the monoclinic space group P 2 i / a with a unit cell of a = 18.140( 3 ) A, b = 19.847(3) A, c = 13.625 ( 2 ) A, p = 1 1 1.64(2)'. There are four molecules in the unit cell. The structure was solved by heavy-atom methods and refined by least-squares techniques to a conventional R index of 0.045 for the 5995 reflections having I L 3al. The porphyrin monoacid cation has three coplanar pyrrole rings with the plane of the fourth ring tilted by 8.6' from the mean plane of the other three. Evidence indicates that the nitrogen atom i n the tilted ring has sp3 hybridization. The anion consists of two rhenium atoms bridged by three chlorine atoms. Three carbonyl groups complete the octahedral coordination on each metal atom. There is a water molecule of crystallization which is involved in hydrogen bonding with the pyrrole nitrogen atoms of the cation. The salt could also be prepared using a 1.5,' 1 Re(C0)5CI/H20EP mixture with (H-OEP)Re(CO)3 as the other product. Analogous compounds could be made using mesoporphyrin IX dimethyl ester and Re(CO)sBr, but the monocationic porphyrin species could not be obtained when mesotetraphenylporphyrin was used.
Introduction T h e metal-carbonyl insertion method2 has been found useful for the preparation of bimetallic porphyrin complexes of rhodium, rhenium, and t e c h n e t i ~ m . Metal ~ . ~ carbonyl halides a r e more reactive reagents t h a n metal carbonyls for the synthesis of bimetallic complexes of porphyrins as well as bimetallic salts of porphyrin acid^.^-^ For example, dimeric rhodium dicarbonyl chloride reacts with porphyrin in benzene a t ambient temperature or in boiling chloroform to form a dirhodium porphyrin complex or a dirhodium salt of a porphyrin d i a ~ i d . In ~ ,the ~ ~present work it was discovered that dimeric rhenium tetracarbonyl halides a r e also capable of reacting with porphyrin in refluxing decalin to form a monorhenium or a dirhenium porphyrin complex as well a s a salt complex of a porphyrin monoacid.Ii This salt contains a n anionic dimeric halocarbonyl rhenium complex, tri-p-halogeno-hexacarbonyldirhenate( I). T h e structure of the salt containing chlorine as the halogen and octaethylporphinium monoacid as the cation has been elucidated by several means, primary among them being an X-ray structural analysis. A preliminary communication on this work has been pub1ished.l' Only a few porphyrin acid structures have been reported. Most of these have been porphyrin diacids. The earlier studies revealed macrocycles with large deviations from planarity, presumably because of the nonbonded interactions between the four imino hydrogen atoms.' However, a more recent report on the structure of a n H 4 0 E P 2 + salt ( O E P = octaethylporphyrin anion) showed that the macrocycle in that case was nearly planar.l0 Monoacid porpyrin cations a r e uncommon. Samuels, Shuttleworth, and StevensI2 reported the preparation of a triiodo derivative of octaethylporphyrin. A later X-ray structure determination, reported by Hirayama et al. in a short communication, revealed that this compound should be formulated as (H30EP)+13-.5 While the macrocycle in this case shows some deviations from planarity, the H - H contacts inside the porphyrin "hole" a r e still too short if the pyrrole nitrogen
0002-7863/78/1500-6068$01 .OO/O
atoms a r e assumed to be trigonally hybridized. Thus partial sp3 hybridization of the nitrogen atoms has been proposed for both the monocationic and dicationic octaethylporphyrin salt^.^^^^ Such a hybridization would be necessary to explain the proposed hydrogen bonding in the H 4 0 E P 2 + structure.IO In all of the porphyrin acid structures so far reported, the d a t a have not been of sufficient quality to examine these suggestions adequately. However, in the present case, crystals of excellent quality were available. Thus in addition to elucidating the nature of the material obtained in the synthetic procedure, a full X-ray analysis promised to shed some light on the nature and conformations of the porphyrin acids.
Results and Discussion Monocation Porphyrinium Tri-p-halogeno-hexacarbonyldirhenate(1).Rhenium pentacarbonyl chloride, Re(CO)SCI, and octaethylporphyrin, H 2 0 E P , in a 2.1 mole ratio reacted in decalin under argon to form a dirhenium octaethylporphyrin complex, O E P [ R e ( C 0 ) 3 ] 2 , and a new compound, I. T h e structure of the new compound was formulated as (H3OEP)+[Re2(C0)&13]-, monocation octaethylporphinium tri-p-chloro-hexacarbonyldirhenate(I),based on elemental analysis, spectroscopic d a t a , and a single-crystal X-ray diffraction analysis. T h e new compound I is sensitive t o moisture and heat. It is stable in dry methylene chloride, tetrahydrofuran, and ethyl acetate but decomposes immediately in alcohols, acetone, and water to free octaethylporphyrin, H 2 0 E P . It can be kept under argon for several days without decomposition in a vacuum desiccator stored in a refrigerator. However, it decomposes gradually in solution a t room temperature to dirhenium OCtaethylporphyrin as indicated by its visible absorption spect r u m , This behavior was confirmed by heating a purified sample of I in decalin under argon (Figure 2). After cooling, dirhenium octaethylporphyrin formed in the supernatant while the undecomposed burgundy-colored solid of I remained at the
0 1978 American Chemical Society
Tsutsui et al. / Tri-k-halogeno-hexacarbonyldirhenate(I) Salts of Porphyrin Acids
6069
a), A
-+4 2
(H30EPl+ [ReZ(COI 6 C 1 3 ] -
Figure 1. Visible spectra of ( H ~ O E P ) + [ R ~ Z ( C O ) ~ Cand I ~ ] (H4-
OEP)*+2CI-.
Scheme I
H20EP
+ Re(C0)sCI
(H-OEP)Re(C0)3
-
+ HC1+
HzOEP (H3OEP)W-
(H-OEP)Re(C0)3
+ Re(C0)sCI
2Re(CO)5CI [Re(C0)4C1]2
-
OEP [Re(COl,lz
Figure 2. Thermal decomposition of (H3OEP)+[Re2(C0)&13]- to form OEP[Re(C0)3]2 in refluxing decalin. The latter was isolated after cooling to have absorptions as shown in the dotted line.
-
+ HCI + 2 C O t
OEP[Re(C0)3]2
[Re(C0)&1]2
+ HCI + 2 C O f
+ 2 CO?.
H+ [Re2(C0)6C13]-
+ HCI
--
-
+ 2COf
(H30EP)'CI-
+ H+[Re2(C0)&13](H30EP)+[Re2(C0)6C13]-
+ HCI
(H30EP)+[Re2(CO)gC13]2 H z O E P -t 4 R e ( C 0 ) 5 C I OEP[Re(C0)3]2 HClt, 8COt
+
+
+
bottom of the reaction flask. T h e amount of the dirhenium octaethylporphyrin varied with the heating time and temperature, between 100 and 195 'C for 2-10 h. Prolonged heating of I in refluxing decalin for 1-2 days caused further decomposition to monorhenium octaethylporphyrin, ( H - 0 E P ) R e ( C 0 ) 3 , as indicated by its visible absorption spectrum." By reaction of H 2 M P I X D M E with stoichiometric quantities of R e ( C 0 ) s B r or [Re(CO)dBr]2 in refluxing decalin under argon, a n analogous salt-like complex of I was prepared: ( H ~ M P I X D M E ) + [ R ~ Z ( C O ) ~(11), B ~ ~monocation ]mesoporphyrinium I X dimethyl ester tri-p-bromo-hexacarbonyldirhenate(1). This complex has chemical and spectroscopic properties similar to those of I (Figure 3). Therefore, a structure identical with that of I is proposed for 11. On the basis of the stepwise changes of visible absorption spectra of the reaction mixture and the X-ray crystal structure of I, Scheme I was proposed. Because of the instability of rneso-tetraphenylporphine m o n ~ c a t i o n none , ~ ~ ~of the corresponding monocation tetraphenylporphinium analogues could be prepared; possible reasons for this instability a r e outlined by Fleischer.' X-ray Structural Data. A preliminary study on the singlecrystal X-ray diffraction analysis of I h a s already been reported." It is a n ionic compound with the porphyrin moiety present a s a monocationic species, H 3 0 E P + . In the anionic species [Re2(C0)&13]-, the two rhenium ions have octahedral coordination. They are joined by three bridging chlorine atoms. Three carbonyl groups on each metal ion complete the coordination. ORTEPI3 drawings of the compound a r e shown in Figure 4. Figure 4 a also shows the nomenclature for t h e different types of carbon atoms and the designations for the four rings. A stereoview of the entire formula unit, including a water molecule of crystallization, is shown in Figure 5.
I480
(H-.W)Re(CO)
3
H2MPIKT)t61
Figure 3. Repeated scan spectrophotometry showing progress of the formation of (H-MP)Re(C0)3, MP(Re(C03)]2, and (H3MPlXDME)'[ Re2(CO)sBr31-.
A. Cation. The imino hydrogen atoms appear to be localized on rings A, B, a n d C . Such a localization was postulated for the H3OEP+ cation a s found in the triiodide salt.5 T h e hydrogen atoms in that study were not located directly, but the C,-N-C, angle for one ring was reported as 102' as compared to the values of 109 and 110' for the other two crystallographically independent rings, which presumably bear imino hydrogen atoms. A similar difference is noted in the present case. Bond lengths and angles a r e tabulated in Table I. Some nonbonded contacts of interest a r e also shown. Ring D, the pyrrolenine ring, has a C,-N-C, bond angle of 105.3 (6)', while the analogous angles on the other three rings a r e 110.6 (6), 108.3 (6), and 110.8 (6)'. These values for rings A, B, and D correspond to those tabulated by H o a r d i 4 for pyrrole rings
6070
Journal of the American Chemical Society
/ 100:19 / September 13, 1978
Table I. Bond Lengths (A) and Angles (deg)a N( 1)-C( 1) N(l)-C(4) N(2)-C(6) N(2)-C(9) N(3)-C( 11) N(3)-C( 14) N(4)-C( 16) N(4)-C(19) C(I)-C(2) C( I)-C(20) W)-C(3) C(2)-C(21) C(3)-C(4) C(3)-C(23) C(4)-C(5) C(5)-C(6) C(6)-C(7) C(7)-C(8) C(7)-C(25) C(8)-C(9) C(8)-C(27) c(9) -C( 10) C( lO)-C( 11) C(1 I)-C(12) C(12)-C(13) C(12)-C(29) C( 13)-c( 14) C( 13)-C(31) C( 14)-C( 15) C( 15)-C( 16) C( 16)-C( 17) C(17)-C( 18) C( 17)-C(33) C( 18)-C( 19) C( 18)-C(35) C( 19)-c(20) C( 2 1)-C( 22) C(23)-C(24) C( 25)-C( 26) C(27)-C(28) C (29) -C (30) C(31)-C(32) C(33)-C(34) C(35)-C(36) N( 1)-H( 1) N (2)-H( 2) N (3) - H ( 31 N(I)-N(2) N(I)-N(4) N(2)-N(3) N(3)-N(4) N(l)-N(3) N(2)-N(4) C(5)-C( 15) C(lO)-C(20) (I
A. Cation 1.377 (9) C(l)-N(l)-C(4) 1.351 (9) C(6)-N(2)-C(9) 1.382 (8) C(ll)-N(3)-C(l4) 1.391 (8) C( 16)-N(4)-C( 19) 1.378 (8) N(l)-C(l)-C(2) 1.363 (8) N(l)-C(l)-C(20) 1.367 (8) C(2)-C(I)-C(20) 1.364 (8) C(I)-C(2)-C(3) 1.406 (9) C(I)-C(2)-C(21) 1.366 (10) C(3)-C(2)-C(21) 1.381 (9) C(2)-C(3)-C(4) 1.496 (9) C(2)-C(3)-C(23) 1.407 (9) C(4)-C(3)-C(23) 1.504 ( I O ) N(I)-C(4)-C(3) 1.391 (9) N(I)-C(4)-C(5) 1.381 (9) C(3)-C(4)-C(5) 1.417 (9) C(4)-C(5)-C(6) 1.348 (9) N(2)-C(6)-C(5) 1.523 ( I O ) N(2)-C(6)-C(7) 1.410 (9) C(5)-C(6)-C(7) 1.516 (9) C(6)-C(7)-C(8) 1.381 (9) C(6)-C(7)-C(25) 1.352 (9) C(8)-C(7)-C(25) 1.435 (9) C(7)-C(8)-C(9) 1.358 (9) C(7)-C(8)-C(27) 1.520 (9) C(9)-C(8)-C(27) 1.421 (8) N(2)-C(9)-C(8) 1.520 (9) N(2)-C(9)-C(IO) 1.392 (9) C(8)-C(9)-C(IO) 1.392 (9) C(9)-C(lO)-C(ll) 1.425 (9) N(3)-C(Il)-C(lO) 1.341 ( I O ) N(3)-C(ll)-C(l2) 1.495 ( I O ) C( 10)-C( 1 I)-C( 12) 1.4 19 ( I O ) C( 1 I)-C( 12)-C( 13) 1.698 (20) C(ll)-C(12)-C(29) 1.407 ( I O ) C(13)-C(12)-C(29) 1.458 (13) C(12)-C(13)-C(14) 1.467 (12) C(12)-C(13)-C(31) 1.449 (12) C(14)-C(13)-C(31) 1.480 (1 3) N(3)-C( 14)-C( 13) 1.486 (1 I ) N(3)-C(14)-C(15) 1.472 ( I O ) C( 13)-C( l4)-C( 15) 1.472 (12) C(14)-C( 15)-C(l6) 1.262 (18) N(4)-C( l6)-C( 15) 0.62 (7) N(4)-C(16)-C(17) 0.84 (7) C(ls\-C(16)-C(17) 0.62 (8) C( 16)-C(17)-C(18) 3.002 (9) C(16)-C(17)-C(33) 2.908 (9) C(18)-C(17)-C(33) 2.940 (9) C(17)-C(18)-C(19) 2.945 (9) C(17)-C(18)-C(35) 4.226 (8) C(19)-C(18)-C(35) 4.113 (9) N(4)-C(19)-C(18) 6.782 (10) N(4)-C(19)-C(20) 6.883 ( I O ) C(18)-C(I9)-C(20)
110.6 (6) 108.3 (6) 110.8 (5) 105.3 (6) 106.4 (6) 125.3 (6) 128.2 (6) 107.9 (6) 124.6 (7) 127.5 (7) 107.7 (6) 126.3 (6) 126.0 (6) 107.2 (6) 124.6 (6) 128.1 (7) 131.3 (6) 123.8 (6) 107.4 (6) 128.8 (6) 108.2 (6) 123.8 (6) 128.0 (6) 108.8 (6) 127.6 (6) 123.6 (6) 107.1 (6) 123.5 (6) 129.2 (6) 130.0 (6) 125.6 (6) 105.5 (6) 129.0 (6) 108.7 ( 6 ) 123.3 (6) 127.9 (6) 107.9 (6) 128.3 (6) 123.6 (6) 107.1 (6) 125.2 (6) 127.7 (6) 129.0 (6) 124.1 (6) 110.8 (6) 125.1 (6) 105.9 (6) 125.3 (6) 128.7 (7) 108.1 (6) 127.4 (7) 122.7 (18) 109.9 (6) 125.2 (7) 124.7 (7)
Re( l)-Cl(l) Re( I)-C1(2) Re( 1)-C1(3) Re( l)-C(37) Re( 1) -C( 38) Re( l)-C(39) Re(Z)-CI(I) Re( 2)-CI( 2) Re(2)-CI( 3) Re( 2)-C( 40) Re(2)-C(41) Re(2)-C(42) C(37)-0(1) C (3 8) -0(2) C(39)-0(3) C(40)-0(4) C( 4 1)-O( 5) C( 4 1)-O( 6) Re(l)-Re(2)
2.517 (2) 2.501 (2) 2.493 (2) 1.889 (8) 1.852 ( I O ) 1.877 (8) 2.530 (2) 2.509 (2) 2.501 (2) 1.091 (9) 1.897 (9) 1.867 (9) 1.142 (8) 1.191 (9) 1.138 (8) 1.158 (9) 1.125 ( 8 ) 1.159 (9) 3.375 (1)
C(I)-C(2O)-C( 19) C(2)-C(2I)-C(22) c(3) -c(2 3) -c(24) C(7)-C(25)-C(26) C(8)-C(27)-C(28) C(I2)-C(29)-C(30) C( l3)-C(3 l)-C(32) C( 17)-C(33)-C(34) C( 18)-C(35)-C(36) C( I)-N( 1)-H( 1) C(4)-N( I)-H( 1) C(6)-N(2)-H(2) C(9)-N( 2)-H( 2) C( 11)-N(3)-H(3) C(14)-N(3)-H(3)
127.2 (7) 112.5 (8) 114.3 (7) 113.4 (8) 112.0 (7) 113.3 (6) 113.8 (6) 110.8 (7) 93.4 ( I 5 ) 124 (8) 124 (8) 118 (5) 98 (5) 124 (9) 125 (9)
B. Anion Cl(l)-Re(l)-Cl(2) Cl(l)-Re(l)-Cl(3) Cl(l)-Re(l)-C(37) Cl(l)-Re(l)-C(38) Cl(1)-Re(l)-C(39) C1(2)-Re(l)-C(37) Cl(Z)-Re(l)-C(38) CI(Z)-Re(l)-C(39) C1(3)-Re(l)-C(37) C1(3)-Re(l)-C(38) C1(3)-Re(l)-C(39) C(37)-Re( l)-C(38) C(37)-Re(l)-C(39) C(38)-Re(l)-C(39) Cl(l)-Re(2)-C1(2) Cl(l)-Re(2)-C1(3) Cl(l)-Re(2)-C(40) Cl(l)-Re(2)-C(41) Cl( 1)-Re(2)-C(42) C1(2)-Re(Z)-C1(3) CI( 2)-Re( 2) -C(40) C1(2)-Re(2)-C(41) Cl( 2)-Re( 2)-C (42) C1(3)-Re(2)-C(40) C1(3)-Re(2)-C(41) C1(3)-Re(2)-C(42) C( 40) -Re( 2)-C( 4 I ) C(40)-Re(Z)-C(42) C(41)-Re(Z)-C(42) Re( 1)-CI( 1)-Re(2) Re( I )-C1(2)-Re( 2) Re( l)-CI(3)-Re(2) Re(l )-C(37)-O( 1) Re( l)-C(38)-0(2) Re( l)-C(39)-0(3) Re(2)-C(40)-0(4) Re(2)-C(4 1 )-O(5) Re(2)-C(42)-0(6)
79.06 (5) 79.87 (6) 171.0 (2) 95.6 (3) 97.1 (2) 94.6 (2) 93.8 (2) 175.1 (2) 92.9 (2) 173.4 (3) 95.7 (2) 91.2 (3) 88.9 (3) 89.5 (3) 76.68 ( 5 ) 79.47 (6) 96.57 (2) 170.9 (2) 97.5 (3) 80.37 (6) 174.8 (2) 95.6 (2) 93.9 (3) 96.8 (3) 92.7 (2) 173.9 (3) 88.9 (3) 88.7 (4) 89.9 (3) 83.93 (6) 84.70 (6) 85.02 (6) 177.7 (7) 177.8 (8) 179.2 (7) 179.1 (8) 177.5 (8) 178.7 ( 8 )
Some nonbonded distances of interest are also given.
respectively lacking and carrying a n N - H bond in free base porphyrins. T h e value for ring B is intermediate. Reasons for this apparent anomaly will be discussed below. Unlike the earlier studies, electron density peaks corresponding to the imino hydrogen atoms were found in the present case, thus confirming t h e localization of hydrogen atoms. N o evidence of a n y electron density corresponding to a hydrogen a t o m bonded to N ( 4 ) (ring D) was found. Of particular interest is the planarity of the cation. T h e free base of octaethylporphyrin is essentially planar. T h e structure of meso-tetraphenylporphyrin, H2TPP, has been reported as planar in a triclinic crystalline m o d i f i c a t i ~ nand ~ ~ as nonplanar in a tetragonal crystalline modification.16 Three structures of
divalent cationic porphyrin species have been reported. In the two earlier studies, both of which involved meso-substituted porphyrins, the cations a r e decidedly n ~ n p l a n a r .I~n these nonplanar species, the symmetry of the cations may be described as Sq. T h e nonplanarity of these cations has been postulated as a mechanism for relieving the crowding of the imino hydrogen atoms. In a much more recent study, the H40EP2+cation, as observed in a crystalline salt formulated as (H40EP2+)[Rh(C12)(CO)2-]2, is nearly planar.I0 As in the earlier study,s the monovalent H3OEP+ cation in the present work has three of the four pyrrole rings (A, C, D) approximately coplanar. Information on least-squares planes is given in Table 11. As is normally found the individual pyrrole
Tsutsui et al. / Tri-p-halogeno-hexacarbonyldirhenate(I) Salts of Porphyrin Acids
607 1
Table 11. Least-Squares Planes
N(1)
N(2) N(3) N(4) C(I) C(2) C(3) C(4) (3-5) C(6) C(7) C(8) C(9) C(10) C(11) C(12) C(13) C(14) ~(15) C(16) C(17) C( 18) C(19) C(20) o(7)
plane 1
plane 2
0.007 0.127 -0.013 -0.024 0.040 0.102 0.050 0.035 -0.107 0.002 -0.189 -0.091 0.078 0.094 0.052 0.060 0.009 -0.041 -0.049 -0.023 0.004 -0.046 -0.062 -0.007 2.254
-0.027 0.027 -0.027 0.027 0.044 0.082 -0.020 -0.043 -0.226 -0.133 -0.371 -0.266 -0.047 -0.007 0.001
0.034 0.024 -0.009 0.027 0.064 0.139 0.082 0.014 0.047 2.228 plane 2
plane 1 plane 2 plane 3 plane 4 plane 5 Dlane 6
2.1
A. Deviations (A) from Planes plane 3 plane 4
plane 5
plane 6
plane 7
-0.003 0.208 0.109 0.009 -0.008 0.016 -0.018 0.0 14 -0.100 0.057 -0. IO5 0.036 0.204 0.255 0.213 0.258 0.183 0.094 0.054 0.034 0.033 -0.059 -0.073 -0.055 2.304
-0.298 -0.009 -0.504 -0.677 -0.373 -0.196 -0.072 -0.094 -0.111 0.018 -0.022 0.017 -0.005 -0.108 -0.331 -0.444 -0.674 -0.708 -0.860 -0.830 -0.958 -0.946 -0.77 1 -0.603 1.844
0.106 0.095 -0.002 0.117 0.203 0.273 0.165 0.105 -0.096 -0.040 -0.295 -0.225
0.102 0.183 -0.016 0.013 0.148 0.243 0.199 0.154 0.006 0.088 -0.107 -0.039 0.01 12 0.095 0.036
-0.008 0.092 -0.024 -0.01 5 0.036 0.091 0.026 0.008 -0.147 -0.042 -0.247 -0.147 0.035 0.058 0.030 0.045 0.000 -0.038 -0.035 -0.004 0.036 -0.016 -0.046 0.003 2.241
-0.01 1 -0,001
0.000 0.001 -0.002 0.002 0.057 0.130 0.221 0.198 0.135 0.200 2.31 1
B. Angles (deg) between Least-Squares Planes plane 3 plane 4 plane 5 0.6 1.5
1.9 3.2 2.2
7.7 9.8 8.3 7.9
0.01 1
-0.055 -0.067 -0.07 1 -0.016 0.013 -0.005 -0.004 0.084 2.304
plane 6
plane 7
2.7 1.5 2.2 4.3
I .4 2.9 1.7 3.2 7.1 2.7
9.8
C. Eauations of Planesu plane 1 macrocycle: N(I)-N(4), C(I)-C(20) 11.71 I X + 1 2 . 2 8 6 ~- 8 . 9 0 9 ~= -2.725 plane 2 N(1)-N(4) 1 1 . 9 9 2 ~ 1 1 . 7 0 3 ~- 9.1972 = -2.854 plane 3 N(l), N(3), N(4), C(l)-C(5), C(lO)-C(20) 1 1 . 7 9 6 ~ 1 2 . 1 2 1 ~- 8 . 9 9 0 ~= -2.755 plane 4 ring A: h ' ( I ) , C(l)-C(4) 1 1 . 2 7 4 ~ 12.454,~- 9.0632 = -2.909
plane 5 ring B: N(2), C(6)-C(9) 1 0 . 8 8 3 ~ 34.155~- 7.6042 = -1.625 plane 6 ring C: N(3), C(Il)-C(14) 1 2 . 2 9 6 ~ 11.627~- 9.0322 = -2.812 plane 7 ring D: N(4), C( 16)-C( 19) 1 1 . 8 9 9 ~+ 1 2 . 3 6 ~- 8.6932 = -2.628
+
+
+
+
+
All planes are unweighted. x, y, z are in monoclinic fractional coordinates.
rings a r e planar. For rings A, C, and D the interplanar angles between adjacent pyrrole rings are 3.2 and 2.7'. T h e maximum deviation of an atom in rings A, B, and D from the mean plane of these three rings is 0.09 A (for C(2)). T h e fourth ring (B) is tilted by 8.6' from the plane of the other three rings. By way of comparison, this angle is 14" in the triiodide salt. In the current study, interplanar angles between ring B and the adjacent rings a r e 7.9 and 9.8'. It is believed that this deviation from planarity is due in part to the tight H-H contacts of the imino hydrogens. Ring B, the one which is not coplanar, bears the imino hydrogen a t o m which comes in closest contact with the other two imino hydrogen atoms. However, this deviation from planarity is still not sufficient to relieve the tight contacts completely. In the present case, the H-H contacts a r e still -1.5-1.6 A if we assume trigonal hybridization and a N-H bond length of 1.O A. T h e minimum H-H nonbonded contact is usually taken to be 1.9-2.0 In both the H40EP2+ saltlo and in the triiodide salt of
H30EP+,5sp3 hybridization of the nitrogen atoms has been postulated. Such a hybridization would be advantageous for a number of reasons. Most importantly, such a hybridization would relieve the tight H-H contacts. A second reason would be to explain the hydrogen-bonding arrangement found in some of these compounds. In the case of the H40EP2+ salt,1° hydrogen bonds a r e postulated between the pyrrole nitrogen atoms and one of the chlorine atoms of the anion. For this to occur, the N-H bonds need to point more or less directly toward the chlorine atoms, as would be the case if there were sp3 hybridization of the nitrogen atoms. In the triiodide salt of H30EP+,Sno hydrogen bonding is reported. In the present case' there is hydrogen bonding between the pyrrole nitrogen atoms and the water molecule of solvation. Intermolecular contacts between pyrrole nitrogen atoms and the oxygen atom of the water molecule a r e shown schematically in Figure 6. N 0 distances are 2.97, 2.99, and 3.05 A for N( l ) , N(2), and N(4), respectively. These are short
...
6012
Journal of the American Chemical Society
01
03
02
Figure 4. (a) ORTEPI3 drawing of the porphyrin cation, H30EP+. Numbering scheme is shown. Shown in parentheses are the designations for the various types of carbon atoms. The thermal ellipsoids are drawn for 50% probability, except those of the hydrogen atoms, which are not drawn to scale. H(21) and H(40) are hidden behind C ( 2 6 ) and C(34), respectively. (b) ORTEP drawing of the complex anion.
/ 100:19 / September 13, 1978
Figure 6. Schematic drawing showing the significant contacts between the porphyrins cation and the oxygen atom of the water of crystallization. The lines between nitrogen atoms signify the planes of the macrocycle. Of the hydrogen atoms shown, it is believed that only H(2) is involved in hydrogen bonding. This is indicated by a solid line between H(2) and O(7). The N(2)-H(2)-0(7) bond angle is also shown.
indicate that N ( 2 ) has a t least some degree of sp3 hybridization, whereas N ( 1 ) and N ( 3 ) a r e trigonally hybridized. H ( 1 ) and H ( 3 ) a r e 0.13 ( 7 ) and 0.04 ( 7 ) A, respectively, o u t of planes defined by the pyrrole rings to which they a r e bonded. T h e bond angles also indicate trigonal hybridization. On the other hand H(2) is 0.68 (7)' out of the plane of ring B in the direction of the water molecule. T h e N ( 2 ) - H ( 2 ) O(7) angle is 148'. It is presumed that the hydrogen atoms of the water molecule a r e involved in t h e other two hydrogen bonds. However, these hydrogen atoms could not be located. This is not surprising in view of the difficulty of locating hydrogen atoms in the presence of heavy metal ions. There is also a possibility the positions of these atoms may be disordered. If N ( 2 ) is a sp3-hybridized a t o m , the problem of tight H-H
-..
Figure 5. A stereoview of the formula unit of (H30EP)+[Re2(CO)&l3]--HzO. The water of crystallization is illustrated by a large circle. Imino hydrogen atoms are indicated by small circles. Other hydrogen atoms have been omitted. The notation of the pyrrole rings is shown. Note the tilt of ring B with respect to the macrocycle.
enough t o be considered hydrogen-bonding distances. T h e fourth N 0 distance (N(3) O(7)) is 3.22 A, probably too long to be considered a hydrogen bond. T h e r e a r e also fairly short intermolecular distances between the water molecule and two of the chlorine atoms of the anion ( 0 ( 7 ) - C l ( l ) , 3.43 A; 0(7)-C1(2), 3.46 A) but these a r e probably too long to b e considered 0 - H CI hydrogen bonds. Thus, there are a t least three hydrogen bonds involving the water molecule. A t least one of the hydrogen bonds should involve a n imino hydrogen atom, especially since N ( 4 ) has no hydrogen atom of its own and yet is implicated in the hydrogen bonding scheme. The observed hydrogen atom positions would
contacts is also eliminated. The three observed H-H contacts a r e 2.62 (H(l)-H(2)), 3.00 ( H ( l ) - H ( 3 ) ) , and 2.34 A ( H ( 2 ) - H ( 3 ) ) . If the hydrogen atom positions a r e given idealized values (assuming sp3 hybridization for N(2), trigonal hybridization for N ( 1) and N(3), and a N - H bond length of 1.0 A) the corresponding H-H contacts a r e 2.22, 2.23, and 2.19 A. A caveat should be observed here. T h e accuracy of the hydrogen a t o m positions is quite low. For example, the N - H observed bond lengths in this case are certainly underestimated and the angles involving H ( 2 ) differ considerably from ideal values. Thus the observed hydrogen a t o m positions should be
Tsutsui et al. / Tri-p-halogeno-hexacarbonyldirhenatetI) Salts of Porphyrin Acids
6073
Figure 7. Stereoview of the packing in the unit cell. Circles of increasing size represent rhenium, chlorine, and oxygen atoms, respectively.
viewed as merely a n indication of the nature of the nitrogen atom hybridization. However, other supportive geometric evidence for the suggested hybridization is available. T h e C,-N-C, angle in ring B (108.3 (6)') is significantly less t h a n that found in rings A and C (1 10.6 (6) and 110.8 (5)O). In an N-substituted porphyrin, ethoxycarbonylmethyloctaethylporphyrin,l* the only reported structure which could serve as a valid comparison, the C,-N-C, angle is 107' on the pyrrole ring which has the substituent on the nitrogen a t o m . There is little doubt that the nitrogen atom in this ring has sp3 hybridization. On the other hand, the angle is 109' in the ring in which the nitrogen a t o m is trigonally hybridized and bears a hydrogen atom. Unfortunately the standard deviations in this compound make it difficult to assess the significance of these differences. However, as previously noted, the angle in ring B is significantly smaller than that found in free base porphyrins for pyrrole rings carrying a N-H bond, yet significantly larger than that found in rings lacking a N - H bond.I4 T h e C,-N bond lengths provide inconclusive evidence for this hybridization scheme. One would expect a bond length of -1.41 8, if sp3 hybridization were present, as opposed to -1.38 8, if there were trigonal h y b r i d i z a t i ~ n . ~T.h' ~e C,-N distances a r e slightly longer in ring B t h a n in the other rings, but this difference is not statistically significant. Given the esd of -0.01 8, in a bond length, it is not unexpected that a difference, if there is one, cannot be observed. T h e average C,-Cb distance of 1.42 ( I ) 8, is somewhat shorter t h a n the usual 1.44 8, but once again it is difficult to judge the statistical significance of the difference. T h e Cb-Cb distances fall within the range tabulated by Hoard.14 With the angles, other bond parameters in exception of the C,-C,-C, the macrocyclic skeleton agree well with those tabulated. These angles a r e larger t h a n expected, especially those involving atoms in ring B. Angles of 125-127' a r e normally found.I4 angle would further increase the Opening of the C,-C,-C, distance between the hydrogen atoms. In a n oxodipyrromethene where pyrrole rings bear imino hydrogen atoms, the angle is 133', presumably to relieve the close H - H contacts.I9 Since ring D does not bear a n imino hydrogen atom, the need for an extension of the C,-C,-CI angle is eliminated and the angles involving this ring a r e smaller. The values for terminal C-C bonds are unusually short. This is a common observation for octaethylporphyrin complexes and is probably due to thermal shortening. Previous experience has shown that when these bond distances are corrected for thermal motion assuming a "riding" model, more reasonable values are obtained.'O However, thermal shortening does not completely explain the short C(35)-C(36) distance. This is undoubtedly short because of the unusually long C ( 18)-C(35) distance. N o explanation for this long bond length is readily available. T h e large thermal ellipsoids and standard deviations for C(35) and C(36) may indicate some disorder, but attempts to resolve this
Table 111. Intermolecular Contacts (A) 5 3.5 Af Cl(2)-0(7) C1(3)-O( 7) C(38)-0(7) N(11-W) N(2)-0(7) N(3)-0(7) N(4)-0(7) 0(2)-C(10) 0(2)-C(ll) 0(6)-C(36)
O( 1)-C(33)O O( 1)-C(34)' C(5)-C(3 l ) b C(27)-C(33) O( l)-C(24)C O(2)-0(4) 0(3)-C(2 I ) e
3.43 3.47 3.46 2.9?* 2.99* 3.22 3.05* 3.19 3.24 3.42
3.47 3.20 3.45 3.47 3.30 3.09 3.38
- x , -y, b -x, -y, 1 - + x, 'I2 - y , - 1 + z . d - '12 + x , - y , z . e '/2 - x, + y , -1 + z. fsuperscripts denote the a
-2.
2.
'12
following equivalent positions relative to positions given in Table V. No superscript indicates that only the identity transformation has been made. Asterisks indicate hydrogen-bonding distances.
possible disorder failed. This bond length was long in separate refinements using both the copper and molybdenum radiation d a t a (see Experimental Section). B. Complex Anion. T h e preparative,2'-26 m e ~ h a n i s t i c , ~ ~ * ~ * and s p e c t r o s ~ o p i c ~properties ~-~~ of halocarbonyls of rhenium(1) have been extensively studied in the past. Four series of halocarbonyl anions of rhenium(1) besides the one found in this study have been reported: [Re(C0)3X3l2-, [Re(C0)4X2]-, IRez(C0)7X3]-, and [Re2(C0)6X4]2-.32-34T h e structures of these anions were deduced from infrared spectroscopic data. The preparation of [Re(CO)&I3]- is mentioned only briefly in a review article.35 T h e structures of two compounds containing this anion have recently been reported.36 These are of the type [Re(CO)3X3] [(CO)3Re(arene)]. In one compound X = Br and the arene is toluene, whereas in the other X = CI and the arene is hexamethylbenzene. In the anion in the present work the average Re-CI distance is 2.51 (1) A, the average Re-C distance is 1.88 ( 2 ) A, while the average C-0 distance is 1.15 (2) A. These numbers agree well with those found in the arene complexes. The Re-C and C - 0 distances also agree with those found in T P P [ R e ( C 0 ) 3 ] 2 . 4 T h e R e ( 1)-Re(2) distance of 3.375 ( 1 ) 8, is too long to postulate any sort of direct metal-metal interactions. None would be expected. C. Molecular Packing. A stereoview of the packing in the unit cell is shown in Figure 7. T h e reason for all eight terminal carbon atoms being pointed in one direction with respect to the porphyrin ring is apparent from this figure. With this arrangement steric interactions with the porphyrin cation related by a center of symmetry are avoided. This was also observed in N-ethoxycarbonylrnethyloctaethylporphyrin.~*T h e interplanar separation between the cation and its centrosymmetrically related nearest neighbor is 3.4 A, which is the approximate layer separation in graphite. However, there are no
6074
Journal of the American Chemical Society
unusually close contacts between atoms in the two macrocycles. A listing of the intermolecular contacts 53.5 8, is given in Table 111. With the obvious exception of the hydrogen-bonding distances between the cation and the water molecule, and with the possible exception of the C l ( 2 ) - 0 ( 7 ) and Cl(3)-0(7) distances which, though unlikely, may correspond to weak 0 - H C1 hydrogen-bonding distances, none of these contacts is expected to have any significant effect on the observed structure.
Experimental Section Materials. Octaethylporphyrin, HzOEP, was purchased from Strem Chemical Co.; dirhenium decacarbonyl, Rez(CO)lo, and rhenium pentacarbonyl chloride, Re(CO)SCl, were purchased from Pressure Chemical Co.; decahydronaphthalene (decalin) was purchased from J. T. Baker Chemical Co., treated with concentrated sulfuric acid, neutralized with sodium bicarbonate solution, washed with distilled water, dried over anhydrous calcium chloride overnight, filtered, and further dried over sodium wire; finally it was distilled under vacuum and stored in a Schlenk tube under argon before use. Other organic solvents were reagent grade chemicals, dried over type 4A molecular sieve and distilled under argon before use. Rhenium pentacarbonyl bromide, Re(CO)sBr, and dimeric rhenium tetracarbonyl bromide, [Re(CO)4Br]2, were prepared by literature procedure^.^^-^^ Physical Measurements. Elemental analyses were performed by Schwarzkopf Microanalytical Laboratory, Woodside, N.Y. Visible spectra were measured with a Cary 14 spectrophotometer. Infrared spectra were measured with a Beckman IR-8 spectrophotometer. Mass spectra were obtained on a CEC-21-104 mass spectrometer. Proton magnetic resonance spectra were obtained using Varian T-60 spectrometers. Monocation Octaethylporphinium Tri-p-chloro-hexacarbonyldirhenate(I), (H30EP)+[Re2(COj&13]- (I). A mixture of 95 mg (0.26 mmol) of Re(C0)sCI and 65.5 mg (0.13 mmol) of HzOEP was refluxed in decalin under argon. The progress of the reaction was followed by visible spectroscopy. The reaction was complete after 20 h of refluxing. The decalin solution was then allowed to stand at room temperature. A large quantity of burgundy-colored crystals was collected by centrifugation and washed with decalin and n-pentane. Finally, the crude product was recrystallized from dichloromethane/ cyclohexane to give 84 mg (0.07 mmol) of dark red crystals (I), mp 21 5-220 "C. Anal. Calcd for Re2C42H47N406C13: Re, 31.50; N, 4.74; CI, 8.98. Found: Re, 26.25; N , 4.81; CI, 8.91. The supernatant contained only dirhenium octaethylporphyrin complex, OEP[Re(CO)3] 2, as indicated by its visible spectrum. About 5 5 mg (0.05 mmol) of the dirhenium octaethylporphyrin complex was isolated from the supernatant solution. The new complex I was also prepared by heating HzOEP and Re(C0)5CI in a 1: 1.5 mole ratio in decalin under argon for approximately 5 h. The supernatant contained only the monorhenium octaethylporphyrin complex, (H-OEP)Re(C0)3, as indicated by its visible absorption spectrum. When equimolar quantities of H20EP and Re(COj5CI were refluxed in decalin under argon for approximately 3 h, only the monorhenium octaethylporphyrin complex was formed. Compound I has visible absorptions in dichloromethane at 390 (Soret band), 530,555,570, and 602 nm. The visible spectrum is extremely similar to that of monocation octaethylporphinium triiodide, (H30EP)+.I3- (the only previously known porphyrin monocation salt),5 and is distinctly different from that of the dication salt, (H40EP)2+.2CI- (Figure When ethanol is added to the dichloromethane solution of I , the visible spectrum reverts to that of metal free octaethylporphyrin H20EP. The infrared spectrum of I in the solid state (KBr pellet) has two broad peaks at 3350 and 3375 cm-' attributed to stretching vibrations of the two chemically independent N-H bonds, and three strong metal-carbonyl stretching bands at 1950, 2020, and 2050 ern-'. The far-infrared spectrum of I (in Nujol) has two broad metal-halide stretching peaks at 200 and 250 cm-I. The ' H N M R solution of I in deuteriochloroform shows two sharp peaks at 6 1.90 (t) and 4.20 (9) for the ethyl substituents and two broad peaks at 6 10.70 and -3.50 for the bridged methine protons and the pyrrolic N-H protons, respectively. Monocation Mesoporphyrinium IX Dimethyl Ester Tri-pbromohexacarbonyldirhenate(I), (H3MPIXDME)+[Re*(CO)6Br3]- (11). The
100:19
/ September 13, 1978
Table IV. Crystal Data for (C~6H4,N4)+[Re2(C0)6c13]-.H20a a = 18.140 (3) 8, b = 19.847 (3) A c = 13.625 (2)
A
@ =111.64(2)' v = 4559 A3
systematic absences: space group:
mol wt = 1200.6 z = 4 dc&d = 1.75 g/cm3 = 58.4 cm-I (Mo K a radiation) hOl ( h odd); OkO ( k odd) P21/a
a Estimated standard deviation of least significant figures shown in parentheses.
reaction was carried out in essentially the same manner as that described above, with rhenium pentacarbonyl bromide, Re(CO)sBr, substituted for Re(C0)sCI as the metal source. A 120-mg sample of HzMPIXDME (0.21 mmol) and 180 mg (0.44 mmol) of Re(C0)sBr in about 20 mL of decalin were heated under argon in an oil bath. The reaction was observed to proceed at a temperature of ca. 120 "C and the progress of the reaction was followed by visible absorption spectroscopy. A stepwise change in the absorption spectra (Figure 3) and a continuous change of color in the reaction mixture was observed. After 5 h of heating, the decalin solution was cooled to room temperature. A large quantity of burgundy-colored substance crystallized out and was collected by centrifugation, washed with decalin and n-pentane, and finally recrystallized from ethyl acetate/cyclohexane/dichloromethane to give 186 mg (0.14 mmol) of dark red crystals (II), mp 180-185 "C. Anal. Calcd for Re~C42H43N4010Br3:C, 36.63; H, 3.13; Re, 27. IO; N, 4.07; Br, 17.82. Found: C , 35.74; H , 3.26; Re, 29.74; N , 4.31; Br, 19.58. The supernatant contained only the dirhenium mesoporphyrin complex, MP[Re(C0)3]2, as indicated by its visible absorption spectrum. About 75 mg (0.067 mmol) of the dirhenium mesoporphyrin complex was isolated from the supernatant solution. Compound I1 can also be prepared in essentially the same manner as that described above, with dimeric rhenium tetracarbonyl bromide, [Re(C0)4Br]2, substituted for Re(C0)sBr as the metal source. H2MPIXDME (198 mg, 0.33 mmol) and [Re(CO)dBr]2 (304 mg, 0.40 mmol) were mixed in about 20 mL of deca!in and heated under argon in an oil bath. The reaction was observed to proceed at a temperature of ca. 105 "C and the progress of the reaction was followed by visible absorption spectroscopy. Attempted Preparation of meso-TetraphenylporphyrinMonocation Salt. The reaction of stoichiometric quantities of Re(C0)sCI with H2TPP in refluxing decalin under argon, in a manner similar to that used in preparing the salt-like complex I, resulted in a mixture of the meso- tetraphenylporphyrin dication salt, (H4TPP)2+.2C1-, and the monorhenium tetraphenylporphyrin complex, (H-TPP)Re(C0)3. X-ray Study. The crystals of complex I were grown from a solvent pair of dichloromethane and cyclohexane (1:l) in a Schlenk tube which was kept slightly open in a vacuum desiccator over blue Drierite under argon atmosphere and stored in a refrigerator for 5 days. The crystals of I grew as parallelepipeds elongated along c and bounded by 11 101. The crystal chosen for intensity measurements measured 0.23 X 0.34 X 0.44 mm. The crystal was mounted in a capillary at an arbitrary ~ r i e n t a t i o nbut , ~ ~with the c axis approximately parallel to the spindle axis. The crystal has remained stable for over 2 years in contact with mother liquor in a sealed capillary. Crystal data are listed in Table IV. Cell dimensions were determined by least squares, minimizing the differences between observed and calculated 20 values. Twenty-four Cu Ka (A = 1.541 78 A) reflections, all having 28 in the range 50-52.5", were measured at both and -20. A Syntex-Datex automated diffractometer equipped with a graphite monochromator was used. The systematic absences uniquely determined the space group as P21/a. A set of intensity data was collected on the Datex-Syntex diffractometer using Cu K a radiation, using the 8-28 method. A total of 4946 independent reflections were measured out to ca. 100" in 28, the machine limit. Of these 3463 were considered observed and were used in the analysis. They were corrected for absorption effects using a Gaussian integration method. From these data the structure was elucidated, which provided the basis for the preliminary communication.' However, for a number of reasons it was decided to re-collect the data using Mo Ka radiation ( A = 0.710 69 A). At the machine limit of 100" in 20, a large percentage of the reflections still had intensities which were considered
+
Tsutsui et al. / Tri-p-halogeno-hexacarbonyldirhenate(I)Salts of Porphyrin Acids
6075
Table V. Fractional Coordinates and Thermal Motion Parameters Derived from the Least-Squares Refinementa x
11
0.21431
I1
0.16351
0.2259(
41 41
0.05361 0.13981 0.05291 -0.02751
0.08011
0.01681 0.16331 0.28881
0.34681 0.31641 0.24121 0.18921 0.11551
0.06091 -0.00291 0.00651 -0.04671 -0.04371 -0.10021 -0.07291 0.0015( 0.05231 0.1271l 0.1761( 0.24011 0.23281 0.2916( 0.42711 0.48621 0.35721 0.39161 0.07491 0.1275l -J.0751( -0.0636( -0.17491 -0.1644(
21 2)
I1
4)
41 51 41
4) 41 41 51
-0.01451
4)
51
-0.06071
5)
-0.11381 -0.11431 -0.06111
41 41 E) 51
4)
51 51 51 5 ) 51 41 4) 4)
61 61
41 51 41 41 4) 41
41
-0.1918(
51
-0.06641 71 -0.0767(lIl -0.19961 81 -0.1793110l -0.41281 71 -0.4557( 1 0 1 -0.78041 71 -0.8559( 81 -a.85~91 71 -0.93801 71 -0.71751 81 -0.79481 91
0.18441
71 51
51 €1
-0.1551(
-0.11901 -0.17761 -O.I~OO( 0.1783( 0.21191 0.28661 0.12171
5)
71 51 61
I01 51
61 51 51 51
0 . 3 4 0 0 1 61 -0~02121 41
41 41 41 41
51 41 31 31
3)
61 5 )
el 51 51 41
51
51
ei 51 41 5) 51
31
0.2252( 0.13861 0.34351 -0.02721 0.30051 0.OF871 0.22731 0.38771
0.01501
41
0.4191(
61 5) 6) 71 61 6)
0.12361 0.049 1 0.149 I 0.052 I
3) 31
0.4706(
5 )
41 4)
31 41
21
31
49(
521 501 62( 8Y( 55( Y2(
-0.456 -0.555
I 41 i 41
4)
41 41
41 41
51 5) 61 51
8)
321 4 ) 67( 61 411 4 ) 831 61 bll 6 ) 644 6 ) 1631 I I I 2461 1 5 1 541 5 1 961 61 451 4 1 55( 51 491 5 ) 761 61 61( 4 1
81
(
41
51 8 ) 611 5 ) 911 81
7)
ra.358
3)
651 6 ) 60( 5 1
81
51
e,
II 4)
341 4 1 341 4 1 38( 4 1 481 4 1 481 41 521 51
71 7) 7)
0.04681 0.1670( 0.22371 0.34191 0.1377( -0.0237(
0.02011
561 691 751 76( 781 681
11 11
35( 4 ) 341 4 1 351 4 )
I I I
0.12241
371 521 471 451 521 471
601 4 ) 39( 3 1 4 0 ( 31 631 4 ) 60( 5 1 651 51 461 4 1 49( 4 1 431 4 ) 491 4 ) 461 4 1 571 5 1
561 37t 421 381
101
0.0446( 0.24741
31 481 4 1 381 4 1 32( 4 1
691 5 ) 851 51 641 4 1 431 4 1 631 5 1 611 5 1
511 561
8) 81 14)
-0.4844( -O.5161(
Ul33l 6261 21 642( 2 1 571 I 1 611 I 1
4861 2 1
81 91 81
71
51
71
-0.62001
E l 51
71
51
0.16057I3l 0.219831 31 a.3048( 2) 0.21351 2 ) 0.05941 21 -0.32401 SI -a.41031 5 1 -0.59761 51 -0.51471 51 -0.30491 71 -0,20821 71 -0.1693( 61 -0.24481 6) -0.2402( 61 -0.3111( 61 -0.29711 6) -0.38791 61 -0.e5961 61 -0.55764 61 -0.62041 6 ) -0.72181 61 -0.75581 61 -0.67751 61 -0.68111 61 -0.60811 7) -0.53051 -0.46531 -0.36841 -0.15961
0.03201 0.11921
-0.00131
61 91
0.29931 0.37031 0.18821 0.193 ( 0.100 I 0.046 (
el
-0.04231
6)
0.45611
51
61
6)
0.03561 0.11561
51 51
51 5)
0.1841(
0.3942l 0.29641
31
I 1
-5.07441 0.1596(
0.12051
31 41 41
51
9.3570(
11 11 11 4) 41
0.01021
-0.10941
0.3073(
2I
0.03001 0.0831( 0.0987( 0.14901 0.1678( 0.2144( 0.21741 0.17021 0.15121 0.09931 0.08031 0.02481 0.00731
0.25221 0.3095( 0.26251 0.3285( 0.12051 0.1702( -0.01161 0.00681
0.0347( 0.0735(
2I
U( 22 I 517( 21 5241 2 ) 521 I 1 49( 1 1 641 I 1
u1111 4351 I 1
2
Y
0.1952 I ( 0.09642( 0.20221 0.07411
0.129591 0.291681 0.26471 0.15391
5 )
6) 61
61 51 B O 1 61 711 5 ) 601 5 1 671 51 551 5 1 481 4 1 4R( 41 421 4 1
571 5 1 661 5 1 631 5 )
761 61 74( 61 1l01I0l
97( 71 IO21 91 8 6 ( 7) 92( 81 8 0 1 61 1021 8 ) 95( 6 ) 90( 7) 631 5 1 82( 6 1 631 6 1 891 71 217114l 115111l 671 61 531 5 1 681 6 1 6 0 ( 5)
591 51 561 5 1 1071
SI
103(
4)
I081 5 1
103(
51
531 41
561 4 1 85( 41 1 4 0 1 7)
79( 4 ) 40
4)
5)
113(
61
731 41 9 9 1 61 79( 4 1
521
11
481
41
44(
41
521 41 041 41
V I IT1
-461 31 2 1 1 31 -171 2) -19(
2( 21 171 7) IS( 71 - 8 1 61
hl
71
-201
71
61 9) 1 2 1 9) 281 81 331 R I 221 31 271 81 251 R l 1 3 1 81 1 ( 71
-131 O(
39(
71 71 81 81 81
81
711 9 1
781 € ) 8 1 1 61 801 61 2561 1 6 1 651 5 1
3 3 ( 91 54( 91 I 3 ( 101
291 1 3 1
91
411111 - 3 6 1 151
751 6 )
15113)
115(
I131 E9( IO2(
91
11(13l
61
271 1 0 ) 691 I41 241 9 1 741101 - I s ( 81
91
631 5 1
901 6 1 C3( 611
51 51
e31 71 1031 9 1
el(
81
134(10l 701 5 1 69( 5 ) 1031 € 1 931 7 ) 831 € 1
21
-131 6 ) 291 5 1 51 51
41
01
I l l (
21
01 -31
-101
621 5 1 771 € )
31
21
61 81
41 4)
511
35(
21
31 81
91
I81
461 4 1 451
71
U113)
299( 1651 22( 311
-38111) 201 101 231 1 I I -130(191
4(22) -31
-591 751 9( -II(
81
-7( 71
24(
7) 101 71 241 61 2 4 1 7) 301 71 261 6 ) l 3 ( 61 2 0 1 61
15( E ) I61 61 1 4 1 61
I l l
71
17( 71 -IO( 91 -52112l -61 I O ) -191 91 -21110) 581161 2 3 1 9) -171 1 5 )
221
9) 37( 13)
2 4 1 91 -121141 71 71 191101 9 1 71 3 7 1 91 -7(10) 41111
VI231 -211 3) -991 3 1 -211 21 -61 21 -21 2, -14(
7 )
71 61 7) -71 91 71 91
-13( -71 -41
-181 -4t
R l
9)
-17( 81 -51 91 -211 8 1 -I)(
81 1 3 1 81
J(
31
81 91 4 1 81 O( 7 1 I 1 9) - 3 1 31 -281 91 4( O(
-='(
11 1
-291 101 -251 I I I -22c I I I -1(221 7111) -24( 161 4(11l -53( 1 4 1 -241101 9( 151 -291 1 0 1
571111 I 1 91 -431 1 0 1
173115l
-231111 -66112) -191 1 8 ) 17(18l
5e(151 91
141 91
AI
291 9 ) 621 7 )
- I ( 91 -711Ol
9)
2 2 1 9)
91 9)
31(
7 ) - 1 4 1 IO1 -341 81 31 5 )
71 0 ) i)lIIl -351 81
61
32( 7 )
IF31 6 ) 1211 51 1041 C l A l l 41
-441 9 1 I( 7 )
571 81 69( 7 1
-341131 -281 I J I -91111 181 9) -51 81 251 9 1 -541101
52(
-92( 7 )
89(
71
1071 6 1 IIYI 4 1 1541
28111) -21 7 1
142(
-3€(
El
71 111
731 5 1
611101 O ( 71
40 40
u l n this and subsequent tables estimated standard deviations for the least significant figure are in parentheses. The Debye-Waller factor is defined as T = exp[-2r2(Ul1a*2h2 U22b*2k2 U33c*212 UI2a*b*hk + Ul3a*c*hl+ U33b*c*kl)]. Thevalues for U have been multiplied by IO3, except for those of Re, which have been multiplied by IO4. For those atoms refined isotropically, the values for B (multiplied by a factor of 10) are given in the column labeled U11.Isotropic temperature factors are defined by T = exp[-B(sin2 @/A2].
+
+
+
to be observed. Collection of the data to a higher sin e / k limit would increase the accuracy of the structural parameters. Also, the high linear absorption coefficient for Cu Ka radiation ( 1 19.5 cm-I) introduces a large systematic error which an absorption correction could only approximately correct. Low-order reflections are the most affected by absorption effects. It is these same reflections which contain a significant contribution from the scattering power of the hydrogen atoms. One of the hopes of this study was to locate the imino hydrogen atoms. I n the presence of rhenium and chlorine atoms, which have a high scattering power, locating the hydrogen atoms would require a good set of data as free from systematic error as possible. Therefore the data were recollected on a Syntex P21 diffractometer using Mo Koc radiation. The same crystal was used. Intensity data for I O 792 unique reflections (20,,, = 5 5 O , (sin B/X),,, = 0.65) were obtained: the w scan method was employed with a scan range of 0.6a. The scan rate varied from 2'/min to 12O/min, depending on the number of counts accumulated in a rapid preliminary scan. The intensities were normalized to counts/min. Background measurements were taken at both ends of the scan with w displaced by 0 . 7 O from the Ka peak; each measurement was made for one-half of the scan time. The intensities of the four standard reflections were monitored after every 89 reflections. Only statistical variations were observed and no corrections were applied.
Only the 5995 reflections with I I3ul were used in the analysis. The standard deviation u1 was determined in terms of the statistical 0.031)2 where variances of the counts as u12 = o12(count) u12(count) is the standard deviation based solely on counting statis-
+
tics.
Absorption corrections were made to the data_using a $ scan technique, observing the change of intensity of the 521 reflection. The minimum intensity observed for this reflection was 62% of the maximum. A scan of several other reflections gave similar results. Structure factors were calculated in the usual way assuming a 50% ideally perfect-5W ideally imperfect monochromator mounted in a parallel orientation. Determinationand Refinement of the Structure. Because there are four molecules in the unit cell of space group P 2 l / a , no crystallographic symmetry is imposed on the molecule. The positions of the rhenium atoms were found from a Patterson synthesis, using the data collected with Cu K a radiation. The positions of the remaining 56 nonhydrogen atoms were found from a series of difference Fourier maps. Included among these atoms was a peak attributable to a water molecule of crystallization. Least-squares refinements using block diagonal and finally fullmatrix methods were carried out. The functioq minimized was Zw(F, - Fc)2 where w = 1 / u F ~ Refinement . using the Cu K a data was not
6016
Journal of the American Chemical Society / 100:19 / September 13, 1978
carried beyond the stage where isotropic temperature factors were calculated structure factors (29 pages). Ordering information is given used. At this stage R = ZIlF,I - IF,II/ZF, was 0.106. on any current masthead page. Further refinement utilized the Mo Ka data. The thermal parameters of the nonhydrogen atoms were refined anisotropically. The References and Notes large number of parameters required refining the formula unit in (1) (a) Department of Chemistry, Texas A&M University; (b) Department of blocks, alternately refining the anion and then the cation and water Biochemistry and Biophysics, Texas A&M University; (c) Control Data molecule. Corooration. After several cycles of refinement, AF syntheses were calculated (2) M. isutsui, M. Ichikawa, F. Vohwinkel, and K. Suzuki, J. Am. Chem. SOC., in an effort to locate hydrogen atoms. The three imino hydrogen atoms 88. 854 119661. (3) H. Ogoshi, J. Setsune. T. Omura, and Z. Yoshida, J. Am. Chem. Soc., 97, were thus found. Peaks corresponding to these three atoms were 6461 (1975). 0.3-0.4 e/A3. However, many of the hydrogens on the peripheral (4) M. Tsutsui, C. P. Hrung, D. Ostfeld, T. S. Srivastava, D. L. Cullen, and E. atoms were not located or were in chemically unreasonable positions. F. Meyer, Jr., J. Am. Chem. Soc., 97, 3952 (1975). The positional parameters of the imino hydrogen atoms were refined. (5) N. Hlrayama, A. Takenaka, Y. Sasada, E. Watanabe, H. Ogoshi, and Z. Yoshida, J. Chem. Soc., Chem. Commun., 330 (1974). The thermal parameters for these three atoms were not refined, but (6) H. Ogoshi, E. Watanabe, and 2. Yoshida, Tetrahedron. 29, 3241 (1973). were set at B = 4.0 A2, as was the case for all hydrogen atoms. Ide(7) A. Stone and E. B. Fleischer, J. Am. Chem. SOC.,90, 2735 (1968). alized positions for the other hydrogen atoms were calculated except (8) F. Hibbert and K. P. P. Hunte, J. Chem. Soc., Chem. Commun., 728 (1975). for those on the water molecule. A C-H bond length of 1.0 8, and a (9) A. Takenaka, Y. Sasada, H. Ogoshi, T. Omura, and Z. Yoshida, Acta staggered configuration for the methyl hydrogens were assumed. The Crystallogr., Sect. 6, 31, l(1975). contributions from these were added to the calculated structure factors (IO) E. Cetinkaya, A. W. Johnson, M. F. Lappert, G. M. McLaughlin, and K. W. but the positions were not refined. However, before the final cycles Muir, J. Chem. Soc., Dalton Trans., 1236 (1974). (11) (a) C. P. Hrung, M. Tsutsui, D. L. Cullen, and E. F. Meyer, Jr., J. Am. Chem. of refinement, new positions were calculated and used. The refinement Soc.,98, 7878 (1976); (b) C. P. Hrung, Dissertation, Texas A&M University, converged with R = 0.046 and R , = (ZwilFol - IFcII/Z~Fo2)1/2 = 1975. 0.047. The final value of the standard deviation of an observation of (12) E. Samuels, R. Shuttleworth, and T. S. Stevens, J. Chem. SOC.C, 145 was 1.72 unit weight, defined as IZwiiFoI - IFcii2/(N,- Nv)11/2, (1968). (13) C. K. Johnson, "ORTEP", Report ORNL-3794, Oak Ridge National Labofor N o = 5995 reflections and N , = 532 variables. In the last cycle of ratory, Oak Ridge, Tenn., revised 1965. refinement all shifts were considerably less than one standard devia(14) J. L. Hoard in "Porphyrins and Metalloporphyrins", K. M. Smith, Ed , Elsevier, tion. There were five peaks on the final difference Fourier which were Amsterdam, 1975, see especially p 337. above 1 e/A3. These were all close to the rhenium atoms. Neither they (15) S.Silvers and A. Tulinsky, J. Am. Chem. SOC., 89, 3331 (1967). M. J. Hamor, T. A. Hamor, and J. L. Hoard, J. Am. Chem. Soc., 88, 1938 (16) nor any other peak in the final difference Fourier were believed to have (1964). any physical significance. (17) 0. Ermer and J. D. Dunitz, Chem. Commun., 178 (1971). Correction for anomalous dispersion was made for all nonhydrogen (18) G. M. McLaughlin, J. Chem. SOC., Perkin Trans. 2. 136 (1974). (19) D. L. Cullen, P. S. Black, E. F. Meyer, Jr., D. A. Lightner, G. B. Quistad, and atoms.41 Scattering factors were from ref 42. The rhenium and C. S. Pak, Tetrahedron, 33, 477 (1977). chlorine atoms were assumed to be in the zero ionization state. No (20) D. L. Cullen and E. F. Meyer, Jr., Acta Ctysta//ogr., Sect. B, 32, 2259 evidence of secondary extinction was found. (1976). The structure was solved and the initial refinement performed on (21) E. W. Abei, G. B. Hargreaves, and G. Wilkinson, J. Chem. Soc., 3149 (1958). the IBM 360/65 and its successor at Texas A&M, the Amdahl470 v/6. Most of the programs used have been listed e l ~ e w h e r e .In ~ ~ . ~(22) ~ B. N. Storhoff, J. Organomet. Chem., 43, 197 (1972). (23) D. Vitali and F. Calderazzo, Gazz.Chim. /tal., 102, 587 (1972). addition the data collection and data reduction programs used for the (24) F. Calderazzo and D. Vitali, Coord. Chem. Rev., 18, 13 (1975). molybdenum data are those supplied by Syntex Analytical Instru(25) R. Colton and J. E. Garrard, Aust. J. Chem., 28, 529 (1973). ments. Most of the final refinements and calculations were performed (26) R. Colton and J. E. Garrard, Aust. J. Chem., 28, 1781 (1973). (27) F. Zingales, U. Sartorelli, F.Canziani, and M. Ravegli, lnorg. Chem.. 8, 154 on a PDP 11/40 computer using an Enraf-Nonius structure deter(1967). mination package (SDP). (28) F. Zingales, U. Sartorelli, and'A. Trovati, lnorg. Chem., 8, 1246 (1967). The final positional and thermal parameters for the nonhydrogen (29) M. A. Bennett and R. J. H. Clark, J. Chem. SOC.,5560 (1964). (30) W. A. McAllister and A. L. Marston. Spectrochim. Acta, Part A, 27, 523 atoms and the three imino hydrogen atoms are given in Table V. The (1971). final calculated positions of the other hydrogen atoms are given in (31) M. S. Wrighton, D. L. Morse, and L. Pdungsap, J. Am. Chem. Soc., 97,2073 Table VI. Tables VI1 and VI11 contain the root mean square compo(1975). nents of thermal displacement along the principal axes of the thermal (32) E. W. Abel, I. S. Butler, M. C. Ganorkar, C. R. Jenkins, and M. H. B. Stiddard, Inorg. Chem., 5, 25 (1966). ellipsoids and the observed and calculated structure factors, respec(33) M. J. Hawkes and A. P. Ginsberg, Inorg. Chem., 8, 2189 (1969). tively. Tables VI-VI11 are available as supplementary material. (34) R. Colton and J. E. Knapp, Aust. J. Chem., 25, 9 (1972).
Acknowledgment. T h e authors wish to thank Professor Shinzi Kat0 for many helpful and stimulating discussions. W e wish to acknowledge support of this research project by the Office of Naval Research, the Robert A. Welch Foundation (A-328), and the Texas Agricultural Experiment Station (H-1668). In the later phases support was also given by the National Institutes of Health (AM20051) and the National Science Foundation (CHE76-15557). Supplementary Material Available: Tables VI-VI11 containing the calculated hydrogen atom positions, the rmt mean square components along the principal axes of the thermal ellipsoids, and observed and
1351 H. C. Lewis and B. N. Storhoff. J. Oroanomet. Chem.. 43. 38 11972). i36j R. L. Davis and N. C. Baenziger, 'inorg. Nucl. Chem. .Lett:, 13,' 475 (1977). (37) W. Hieber and H. Schulter, Z.Anorg. Allg. Chem., 243, 164 (1939). (38) E. W. Abel and G. Wilkinson, J. Chem. Soc., 1501 (1959). (39) E. W. Abel, G. B. Hargreaves, and G. Wilkinson, J. Chem. Soc., 3149 (1958). (40) E. F. Meyer, Jr., J. Appl. Crysta//ogr., 8, 45 (1973). (41) D. T. Cromer and D. Liberman, J. Chem. Phys., 53, 1891 (1970). (42) D. T. Cromer and J. T. Waber in "International Tables for X-ray Crystallography", Vol. IV, J. A. lbers and W. C. Hamilton, Ed., Kynoch Press, Birmingham, England, 1974, pp 72-101. (43) D. L. Cullen, E. F. Meyer, Jr., and K. M. Smith, Inorg. Chem., 18, 1179 (1977). (44) D. L. Cullen, E. F. Meyer, Jr., F. Eivazi, and K. M. Smith, J. Chem. Soc.. Perkin Trans. 2,259 (1977).
