7485
N-acylimidazoles, l I 1 8 and Schiff bases. 2 2 These reac- ysis of the dichloroacetate esters is 1.3 calculating it from
tions all involve participation by water in the transition the ko values for the nitro- and methoxy-substituted
state. The observed rate decreases have been explained compounds. It can be seen in Table IV that the value
by the decrease in water activity as acid concentration of p is highly dependent on the ionic strength of the me-
is increasedl3 or by a possible change in rate-deter- dium since the rate decrease produced by increasing acid
mining step. l8 concentrations is much greater for the nitrophenyl ester
The pH-independent reactions are undoubtedly wa- than for the methoxyphenyl ester. This is again very
ter-catalyzed reactions, and, as discussed above, the likely a reflection of the importance of solvation factors
large D 2 0 solvent isotope effect indicates that proton in the hydrolysis of nitrophenyl esters.
transfer is taking place in the transition state. The The magnitude of the water-catalyzed reactions is of
most likely mechanism can therefore be represented by considerable interest since such large water catalysis is
V or a kinetic equivalent. not observed with other nitrophenyl esters (see the ko
values in Table 111). Thus, the carbonate ester is
06- especially susceptible to water catalysis. This indi-
cates a very low Brpnsted coefficient for classical general
base catalysis of carbonate ester hydrolysis. A Brpn-
sted coefficient of 0.3 was observed for bis(4-nitro-
phenyl) carbonate. 2 4 An enhanced water catalysis has
’‘9-H been noted previously with acyl activated esters by a
positive deviation from Brsnsted plots for nucleophilic
‘V .H catalysis. 26
The Hammett p value23 for water-catalyzed hydrol- Acknowledgment. This work was supported by a
(21) J. T. Edward and S. C.R. Meacock, J. Chem. SOC.,2000, 2009 research grant from the National Institutes of Health.
(1957); J. A. Leisten, ibid., 765 (1959).
(22) E. H. Cordes and W. P. Jencks, J. Am. Chem. SOC.,84, 832 (24) The hydrolysis of bis(4-nitrophenyl) carbonate is strongly cata-
(1962). lyzed by the base species of various buffers. Rate constants for these
(23)L. P. Hammett, “Physical Organic Chemistry,” McGraw-Hill reactions will be presented in a subsequent publication.
Book Co., Inc., New York, N. Y.,1940, Chapter VII; H. H. Jaffk, (25) K. Koehler, R. Shora, and E. H. Cordes, J. Am. Chem. SOC.,88,
Chem. Rev., 53, 191 (1953). 3577 (1966).
Diimide Reduction of Porphyrins
H. W. Whitlock, Jr., R. Hanauer, M. Y. Oester, and B. K. Bower
Contributionfrom the Department of Chemistry, University of Wisconsin,
Madison, Wisconsin. Received December 26, 1968
Abstract: A study of the diimide reduction of a,P,y,64etraphenylporphyrinand 1,2,3,4,5,6,7,8-octaethylpor-
phyrin has been carried out. Diimide selectively cis-hydrogenates porphyrins. The stereoselective synthesis of
cis- and trans-octaethylchlorin is described.
T he central role played by oxidation/reduction reac-
tions of porphyrins in photosynthesis and electron-
transport mechanisms’ coupled with the well-recog-
diimide reduction is the best synthetic procedure for
preparing reduced derivatives of the tetraphenyl por-
phyrin ring system. These results are to be compared
nized cryptoolefinic nature of the peripheral double with one electron reduction of metal-free porphyrins
bonds in porphyrin^*-^ has prompted us to investigate which, with the possible exception of photoreduction of
the diimide reduction of meso-tetraphenylporphyrin and tetraphenylporphyrin with benzoin,’ normally8-11 af-
1,2,3,4,5,6,7,8-0ctaethylporphyrin. We have been able ford the isomeric but less stable lo phlorins. (See Figure
to demonstrate that (1) porphyrins and chlorins are in- 1.)
deed readily reduced by diimide produced from the stan- To the extent that one can equate ring currents and
dard diimide precursor p-toluenesulfonylhydrazine,6 c‘aromaticit”’ with lack of reactivity toward nonpolar
(2) reduction of octaethylporphyrin proceeds with a high cycloaddition reagents we view the surprisingly facile
degree of stereoselectivity to cis-octaethylchlorin, (3) diimide reduction of porphyrins as strong evidence for
the picture of them as having marked and mutually dis-
(1) A. A. Krasnovskii, Ann. Rev. Plant Physiol., 11, 363 (1960). tinct regions of ir localization and aromaticity sug-
(2) L. E. Webb and E. B. Fleischer, . I Amer. Chem. SOC.,87, 667
.
(1966). gested by X-ray crystallographers.2-4 An interesting
(3) S.J. Silvers and A. Tulinsky, ibid., 89,3331 (1967).
(4) J. L. Hoard, M. J. Hamor, T. A. Hamor, and W. S. Caughey, (7) G. R. Seely and M. Calvin, J. Chem. Phys., 23, 1068 (1955).
J . Amer. Chem. SOC.,87,2312 (1965). (8) D. Mauzerall, J. Amer. Chem. SOC.,84, 2437 (1962).
(5) R. Grigg, A. W. Johnson, and A. Sweeney, Chem. Commun., 697 (9) H. H. Inhoffen, P. Jaeger, R. Mahlhop, C. D. Mengler, Ann., 704,
(1968). 188 (1967).
(6) E. E. van Tamelen, R. S. Dewey, and R. J. Timmons, J. Amer. (10) G. L. Closs and L. E. Closs, J . Amer. Chem. SOC.,85, 818 (1963).
Chem. SOC.,83,3725 (1961). (11) A. N. Sidorov, Usp. Khim., 35, 366 (1966).
Whitlock, Hanauer, Oester, Bower / Diimide Reduction of Porphyrins
7486
from these reductions are kinetically determined as is
shown by the stability of tetraphenylisobacteriochlorin
and zinc tetraphenylbacteriochlorinunder the reaction
conditions associated with formation of their isomers.
Our present understanding of the factors governing
porphyrin chlorin bacteriochlorin rates of diimide reduction of olefins17predicts that dif-
ferences in n-bond order, localization energies or mag-
nitude of the atomic orbital coefficients of the highest
filled porphyrin molecular orbitals should determine
the difference in reactivity of chlorin and metallochlorin.
Huckel molecular orbital calculations 18,19 to this end
isobacteriochlorin porphyrinogen were inconclusive. *o
There are two other noteworthy features of the pre-
parative aspects of diimide reduction of these porphyrin
derivatives. o-Chloranil dehydrogenation*1,22of tetra-
phenylbacteriochlorin to tetraphenylchlorin is suffi-
ciently faster than dehydrogenation of the chlorin that
Ph
the most efficient chlorin preparation involves reduction
of tetraphenylporphyrin to a chlorin-bacteriochlorin
tetraphenylporphyrin octaethylporphyrin mixture followed by addition of o-chloranil to dehydro-
Figure 1. Nomenclature used for the ring systems discussed in the genate the bacteriochlorin. This technique may also be
paper. used to advantage in the preparation of octaethylchlo-
rin described below. Separation of the pigments from
one another is facilitated by their differing partition co-
corollary of this, however, is that chlorins and metallo- efficientsbetween benzene and phosphoric acid. Tetra-
chlorins differ significantly in this respect (see below). phenylporphyrin may be separated from tetraphenyl-
Reduction of Tetraphenylporphyrin. The behavior chlorin by extraction of a benzene solution of the two
of tetraphenylporphyrin (prepared according to Adler, first with 68% (w/w) phosphoric acid, which removes
et al. 12) tetraphenylchlorin, zinc tetraphenylporphyrin,
the porphyrin into the acid layer, followed by 82%
and zinc tetraphenylchlorin toward p-toluenesulfonyl- (w/w) phosphoric acid. Tetraphenylbacteriochlorin
hydrazine in pyridine is summarized in Scheme I. Our may be separated from the porphyrin and chlorin by
Scheme I. Reduction of Tetraphenylporphyrin (TPP), Tetra- extraction of benzene solutions of the three with 82x
phenylchlorin (TPC), Zinc TPP (ZnTPP), and Zinc TPC (ZnTPC) phosphoric acid: the less basic bacteriochlorin re-
with Diimide in Pyridinea mains in the benzene layer. Separation of tetraphenyl-
TPP + TPC ---t TPBC bacteriochlorin and tetraphenylisobacteriochlorin may
J.Zn*+ J.znZ+ be achieved by selective extraction of the latter from
ZnTPP +ZnTPC +ZnTP-i-BC benzene with phosphoric acid. Dilution of the acid and
extraction enables one to isolate the free porphyrin spe-
TPBC = tetraphenylbacteriochlorin; ZnTP-i-BC = zinc cies. Use of phosphoric acid offers several advantages
tetraphenylisobacteriochlorin. over, e.g., hydrochloric acid in its decreased antiper-
observations as to the sequential nature of these reduc- sonnel properties and minimization of acid-catalyzed
tions (Scheme I) are in good qualitative agreement with autoxidation of the reduced porphyrins. The relative
the spectroscopic studies of Sidorov on the hydrazine- basicities of these porphyrins as inversely equated with
aerobic pyridine reduction of tetraphenylporphyrin. l 3 the strength of phosphoric acid needed to extract them
It seems clear that the processes observed are in fact re- from benzene solution is: porphyrin > chlorin, iso-
ductions of the peripheral double bonds of the porphy- bacteriochlorin > bacteriochlorin.
rins by diimide and do not involve poorly defined Reduction of 1,2,3,4,5,6,7,8-0ctaethylporphyrin. Re-
complexes between porphyrin or hydrazine and oxy- duction of octaethylporphyrin with diimide in hot
gen. l 3 pyridine or @-picolineaffords a single isomer of octa-
A remarkable feature of these reductions is the in- ethylchlorin (OEC), mp 216-217". Reduction of iron-
fluence on the course of reaction of the presence of (111) octaethylporphyrin chloride by a modification of
chelated zinc. Diimide reduction of metal-free tetra- the procedures of Eisner, et and Schlesinger, et
phenylchlorin affords tetraphenylbacteriochlorin con- - * ~ in 72% yield an isomeric octaethylchlo-
~ l . , ~ * affords
taminated by no more than 2-4x of tetraphenyliso- (17) E. W. Garbisch, S. M. Schildcrout, D. B. Patterson, and C. M.
bacteriochlorin as determined by its uv-visible spec- Sprecher, J . Amer. Chem. SOC.,87, 2932 (1965).
trum. Reduction of zinc tetraphenylchlorin affords the (18) G. W. Wheland and D. E. Mann, J . Chem. Phys., 17,264 (1949).
(19) W. E. Kurtin and P-S Song, Tetrahedron, 24, 2255 (1968).
zinc complex of tetraphenylisobacteriochlorin with a (20) R. C. Dougherty, H. H. Strain, and J. J. Katz, J . Amer. Chem.
similar degree of selectivity. The product mixtures Soc., 87, 104 (1965).
(21) U. Eisner and R. P. Linstead, J . Chem. SOC.,3749 (1965).
(12) A. D. Adler, F. R. Longo, and J. D. Finarellis, J . Org. Chem., 32, (22) J. R. L. Smith and M. Calvin, J . Amer. Chem. Soc., 88, 4500
476 (1967). (1966).
(13) A. N. Sidorov, Biofirika, 10, 226 (1965). Reduction of por- (23) U. Eisner, A. Lichtarowicz, and R. P. Linstead, J . Chem. Soc.,
phyrins to chlorins by hydrazine has also been observed by E. W. Baker, 733 (1957).
A. H. Corwin, E. Klesper, and P. E. Wei, J . Org. Chem., 33, 3144 (1968). (24) W. Schlesinger, A. H. Corwin, and L. J. Sargent, J. Amer. Chem.
(14) E. J. Corey, W. L. Mock, and D. J. Pasto, Tetrahedron Letters, Soc., 1 2 , 2867 (1950).
347 (1961), and succeeding papers. (25) H. Fischer and F. Balaz, Ann., 553, 166 (1942).
(15) S. Hunig, H. R. Milles, and W. Thier, ibid., 353 (1961). (26) H. Fischer, I<. Platz, H. Helbzrger, and H. Niemer, ibid., 479,
(16) S. Hunig and R. Muller, Angew. Chem., 75, 298 (1963). 26 (1930).
Journal of the American Chemical Society 91:26 J December 17, 1969
7487
rin, mp 231.8-232'. Both reactions are completely
stereoselective within the limits of detection of nmr (ca.
95 %). This is clearly seen on comparison of the bridg- t ram-octaethvlchlorin
ing methine region of the two spectra: the spectrum
of the former OEC shows peaks due to the methine hy-
drogens as two singlets at 6 9.66 and 8.81 (CDC13, 100
MHz) while that of the latter OEC exhibited a pair of
singlets at 6 9.72 and 8.86 (CDC13, 100 MHZ).~*A syn-
thetic mixture of the two showed a resolved doublet for
each methine peak. Scarcity of material prevented de-
termining the ultimate limits of detection of contamina-
tion of one isomer with the other, other than to note
that each reduction gave a chlorin that had but a pair of
peaks in the methine region. The remainders of the
spectra of the two isomers (Figure 2) are also different
in a nonconcentration dependent manner.
Which chlorin is the cis and which is the trans? The
simplest (but suspect) reasoning, that the chlorin from
alkali metal reduction of the hemin chloride is the trans,
is correct. Oxidation of this isomer by the procedure
of Ficken, et U I . , afforded
~~ racemic 2,3-diethylsuccinic
acid.30 It appears then that diimide in P-picoline re-
duces octaethylporphyrin to cis-octaethylchlorin and
sodium/alcohol reduces iron octaethylporphyrin chlo-
ride to the trans-chlorin, both reactions proceeding with
a high degree of stereoselectivity. 3 1 Figure 2. "r (CDC13) at 100 MHz of cis and truns-octaethyl-
Consideration of the cause of stereoselective sodium chlorin. The chemical shifts in cycles per second are relative to
isoamyl alcohol reduction of the hemin chloride to iron tetramethylsilane. Peak assignments for the two isomers are 6 1.05,
CHsCHzCH; 1.75, CHsCHz(3=; 2.0-2.5, CH3CHzCH; CU. 3.9,
trans-octaethylchlorin reveals an interesting facet of the CH~CHZC=;4.5, C H ~ C H Z C H .The two singlets for the bridging
chemistry of the iron chlorin. Treatment of iron(II1) methine hydrogens are off scale to the left.
trans-octaethylchlorin chloride with sodium isoamyl-
oxide in isoamyl alcohol-0-d under conditions approxi-
mating those of the sodium and alcohol reduction re- tive reduction of the metalloporphyrin by sodium in
sults in the selective exchange of two of the 42 aliphatic alcohol is a reflection of the greater stability of the trans-
hydrogens32as in eq 1. This conclusion follows from chlorin, regardless of the stereochemistry of the chlorin
initially produced.
The role played by iron in the reduction and hydro-
gen-deuterium exchange process is poorly understood.
Fischer and G i b i a ~have~ ~ shown
~ that certain optically
active (by virtue of substituents on the dihydro ring)
chlorins are racemized under Wolff-Kishner reaction
conditions, although their conditions are somewhat
harsher than ours, and there is, of course, no indication
the low voltage mass spectrum of the derived chlorin, in their work of the positional specificity of deprotona-
64% d2,34and its nmr spectrum which showed virtual tion of metal-free chlorins, if in fact racemization pro-
disappearance of the band assigned to the two methine ceeds in that way. It is clear from the work of Schles-
hydrogens. Inasmuch then as H/D exchange is a suffi- inger24and Corwin" that the path of alkali metal reduc-
cient condition for cisltrans isomerization stereoselec- tion of porphyrins may be influenced by the particular
metal complexed by the ligand. Two mechanistic pos-
(27) H.Fischer and H. Helberger, Ann., 471, 285 (1929). sibilities for the role of iron in the hydrogen-deuterium
(28) Assignment of the upfield peak to the pair of hydrogens labeled exchange are the usual superacid role of the metal in
76 (Figure 1) is based on hydrogen-deuterium exchange experiments in
which the faster exchanging hydrogens give rise to the peak at 6 8.86. facilitating attack of nucleophilic species on the por-
(29) L.Schotte and A. Rosenberg, Arkiu Kemi, 8,551 (1966). phyrin and a coupling of deprotonation of the porphyrin
(30) G. E. Ficken, R. B. Johns, and R. P. Linstead, J. Chem. SOC.,
2272 (1956). species with redox reactions37of the metal. In this re-
(31) Since this paper was submitted, H. H. Infoffen, J. W. Buchler, spect it is interesting to note that sodium/alcohol reduc-
and R. Thomas [Tetrahedron Lett., 1145 (1969)l have reported the low- tion of iron octaethylchlorin chloride affords the iron
yield reduction of octaethylporphyrin to a 5 :1 mixture of cis- and truns-
octaethylchlorin by diborane in tetrahydrofuran. The properties of the complex of octaethyl-cis-bacteriochlorin.
two isomers quoted by them are in reasonable agreement with those of
ours. Experimental Section
(32) Exchange of two labile vinyl hydrogens on the bridging methy-
lenes20.Sa also occurs but is not considered since any deuterium thus truns-1,2,3,4,5,6,7,8-Octaethylchlorin. A solution of 1 .O g (6.17
introduced is washed out in subsequent acid demetalation of the metal mmoles) of anhydrous ferric chloride and 0.5 g of anhydrous so-
chlorin. dium acetate in 200 ml of glacial acetic acid contained in a 250-ml
(33) R. B. Woodward and V. Sklric, J . Amer. Chem. SOC.,83, 4676 round-bottomed flask fitted with magnetic stirrer and Sohxlet
(1961).
(34) This reflects the isotopic composition of the recovered solvent,
the isotope dilution presumably being caused by a proto-Guerbetas re- (36) H.Fischer and H. Bibian, Ann., 550, 208 (1942).
action. (37) A. H. Corwin and 0. D. Collins, 111, J. Org. Chem., 27, 3060
(35) M.Guerbet, C. R. Acud. Sci., Paris, 128, 511 (1899). (1962).
Whitlock, Hanauer, Oester, Bower 1 Diimide Reduction of Porphyrins
7488
extractor containing 1.0 g (1.87 mmoles) of octaethylp~rphyrin~~ to afford on evaporation 183 mg of a brown oil. Chromatography
was heated under reflux until the porphyrin was extracted from the of this oil on silica gel followed by sublimation and preparative thin
thimble. The solution was cooled and the bulk of the acetic acid ~ ~the material eluted with 5 % ether in
layer c h r ~ m a t o g r a p h yof
was removed under reduced pressure. Water (200 ml) was added chloroform afforded 57 mg (22% yield) of diethylmaleimide, and
and the resulting suspension was filtered to afford 1.18 g of crude 23 mg of a solid that was heated overnight at 100" with 2 ml of
iron(II1) octaethylporphyrin chloride. 80% sulfuric acid. The acid mixture was diluted and continuously
Anal. Calcd for C3eH44N4FeCl:C, 69.29; H , 7.11; N , 8.98; extracted with ether to afford 6.5 mg of crystals, mp 112-113.5".
C1, 5.68. Found: C, 69.12; H, 7.24; N, 8.39; Cl, 5.37. Repeated sublimation and recrystallization afforded approximately
Crude iron(II1) porphyrin from above was dissolved in 118 ml 1 mg of dl-cup'-diethylsuccinic acid, mp 128-130" (lit.Zg mp 131-
of isoamyl alcohol in a 250-ml round-bottomed three-necked 133"), mixture melting point with an authentic samplez9of mp 130-
flask fitted with a stirrer and condenser. The mixture was heated 131.5", 130.5-132.0". The infraredspectra of the twosamples were
under reflux for 15 min (oil bath, 150") and then 11.8 g (0.51 g- identical and clearly distinct from that of meso-diethylsuccinic
atom) of sodium was added t o the refluxing solution (the color of a ~ i d . 2 ~Subjecting a sample of meso-diethylsuccinic acid to the
the solution went from red-brown to red to a dark green). The above set of conditions did not isomerize it.
mixture was refluxed for an additional 15 min (oil bath, 175"), the Iron(1II) Octaethylchlorin Chloride. A mixture of 300 mg (0.56
stirring bar was raised above the surface, and the mixture was mmole) of trans-octaethylchlorin, 1.0 g ( 5 mg formula weight) of
cooled in an ice bath. After solidification had occurred 25 ml of ferrous chloride tetrahydrate, and 1.4 g (17 mg formula weight) of
methanol followed by 100 ml of water was added and the mixture sodium acetate in 35 ml of glacial acetic acid was heated under
was stirredat room temperature until the bulk of the reaction mixture reflux with stirring under nitrogen for 5 min. Solvent was removed
had dissolved. The mixture was extracted with 300 ml of benzene in cucuo from the cooled reaction mixture and the residue was slur-
and the organic layer was washed successively with water, one por- ried with water and filtered t o afford 337 mg of the iron(II1) chlorin.
tion of 150 ml of concentrated hydrochloric acid, and water until It should be recrystallized from hexane-chloroform: h z 376 (e
neutral, and then concentrated under reduced pressure to a volume 89,000), 471 (8060), 510 sh (5580), 559 (5960), 603 (24,200), 751
of 75 ml. The resulting dark green solution was added to 500 ml of (2860).41
acetic acid under a nitrogen atmosphere in a 1-1. flask fitted with Anal. Calcd for C36H4aN4FeCl: C, 69.06; H , 7.41; N, 8.95;
stirrer. With rapid stirring 50 ml of a saturated solution of ferrous Fe, 8.92; C1, 5.66. Found: C, 68.80; H, 7.18; N, 9.14; Fe,
sulfate in concentrated hydrochloric acid was added over 2 min, 9.12; C1, 5.83.
stirring was continued for 2 min, and the reaction mixture was Hydrogen Exchange of Iron(II1) Octaethylchlorin. To a solution
poured into a cold mixture of 300 ml of saturated aqueous sodium of 0.48 g (0.021 mg-atom) of sodium in 15 ml of isoamyl alcohol-
acetate and 200 ml of benzene. The benzene layer was separated, 0-dl (98% OD by nmr) in a glass tube was added 137 mg (0.212
washed and dried, and evaporated to afford 850 mg of a dark solid. mmole) of iron(I1) chlorin prepared as above. The mixture was
The solid was dissolved in benzene and the benzene solution was freeze-thaw degassed mm), sealed, and heated at 135" for
extracted with 50% (w/w) phosphoric acid until the acid extracts 1 hr. The cooled tube was opened and its contents were poured
were colorless. The organic layer was then extracted with 75% into water. The aqueous mixture was extracted with benzene and
(w/w) phosphoric acid (the extract was dark blue) until the extracts the combined benzene extracts were washed with concentrated
were pale. The remaining benzene layer was discarded, and the hydrochloric acid and water and evaporated. The residue was
75% phosphoric acid extracts were combined, diluted with water dissolved in 50 ml of acetic acid, 5 ml of a saturated solution of
to approximately 60% phosphoric acid content, and extracted with ferrous sulfate in hydrochloric acid was added, and the mixture
benzene. The combined organic extracts were washed and evap- was stirred at room temperature for 10 min and poured into 100 ml
orated to afford on crystallization from ethanol-chloroform 715 mg of saturated aqueous sodium acetate. The mixture was extracted
(72 % yield) of trans-octaethylchlorin: mp (evacuated capillary) with benzene and the combined benzene extracts were extracted
231.8-232" ( M 2 3mp232"); uv-visible X:$391 mp (E 188,500),487 with 8 5 % (w/w) phosphoric acid until the extracts were colorless.
(12,900), 496 (13,400), 520 (4070), 544 (16201,593 (4030), 617 (4480), The benzene layer was washed with water and evaporated to afford
647 (73,150). Combustion analysis of both trans- and cis-octa- 51 mg (37% yield) of iron(II1) octaethylporphyrin, identified by its
ethylchlorin persistently gave results low in carbon. uv-visible spectrum with that of an authentic specimen prepared as
cis-Octaethylchlorin. In a 100-ml three-necked flask fitted with above. The combined phosphoric acid extracts were diluted with
stirrer, condenser, and addition funnel was placed a mixture of 3.75 an equal volume of water (to ca. 50% w/w concentration) and ex-
g of anhydrous potassium carbonate, 0.50 g (0.94 mmole) of octa- tracted with benzene. The benzene extracts were washed with
ethylporphyrin, and 25 ml of freshly distilled P-picoline. The mix- water and evaporated, and the residue was crystallized from chloro-
ture was placed under a nitrogen atmosphere and brought to reflux, form/methanol to afford 27 mg (24% yield) of trans-octaethyl-
and a solution of 5.0 g (25.8 mmoles) of p-toluenesulfonylhydrazine chlorin. Its identity was established by its nmr which showed that
in 15 ml of P-picoline was added dropwise over a period of 2.5 hr. the signal at 6 4.5 (CHCH2CH3)corresponded in area to 0.24 H (88 %
The reaction mixture was refluxed a further 2 hr and then cooled deuteration at this position). Its mass spectrum (determined at 12
and partitioned between benzene and water. The benzene layer V nominal on an A.E.I. MS-902, using a direct inlet probe at 100")
was extracted with cold 10% hydrochloric acid followed by three ~ do, 16.2% d,,
indicated the following isotope c o m p ~ s i t i o n : ~3.5%
35-ml portions of 8 5 % (w/w) phosphoric acid. The combined 63.5% d2, 16.1 % da, 0% d4,0.7% dj.
phosphoric acid extracts were diluted to a 60% (w/w) phosphoric Octaethyl cis-Bacteriochlorin. In a 50-ml three-necked round-
acid concentration and extracted with benzene. The combined bottomed flask fitted with a condenser and magnetic stirrer were
benzene extracts were then washed with 60% phosphoric acid placed under nitrogen 100 mg (0.16 mmole) of iron(II1) octaethyl-
until the washes were colorless, with saturated bicarbonate solution porphyrin chloride and 650 mg (28.2 mg-atoms) of sodium in 20
and with water, and were evaporated to afford, after recrystalliza- ml of isoamyl alcohol. The mixture was refluxed with stirring for
tion from methanol-chloroform, 56 mg (1 1 % yield) of cis-octa- 1.5 hr and cooled to 5". Methanol, 1 ml, was added followed by
ethylchlorin: mp 21C217" (vac); 393 mp ( E 182,800), 489 5 ml of a saturated hydrochloric acid solution of ferrous sulfate.
(12,590), 497 (12,600), 522 (3540), 548 (1635), 596 (4050), 620 (4440), The mixture was stirred for 0.5 hr and the blue solution was par-
651 (69,100). Extending reaction times in attempts to raise the titioned between 200 ml of benzene and 20 ml of aqueous ammonia.
yield of cis-chlorin inevitably resulted in bleaching of the reaction The benzene solution was washed with water, dried over anhydrous
mixture with accompanying formation of octaethylporphyrinogen. sodium sulfate, and evaporated. 4 4 Recrystallization of the residue
Oxidation of rrans-Octaethylchlorin. Racemic Dimethylsuccinic
Acid. To a solution of 296 mg (0.55 mmole) of trans-octaethyl-
chlorin in a mixture of 200 ml of 60% sulfuric acid and 120 ml of (40) Thin and thick layer chromatography was carried out on a silica
85 phosphoric acid was added a solution of 625 mg of chromium gel P F ~ u(Brinckmann) with chloroform as eluent.
trioxide in 10 ml of water. The reaction mixture was stirred at (41) The spectrum of this substance by Eisner42is different from this.
0" for 3.5 hr after which time it was diluted with 1 1. of ice water However, washing benzene solutions of the iron chlorin prepared as
and extracted with five 250-ml portions of ether. The combined above with aqueous ammonia produces a material having spectra similar
ether extracts were dried and evaporated. The dark residue was to that reported by Eisner.
percolated in ether solution through a small column of silica gel (42) U. Eisner, J . Chem. Soc., 3461 (1957).
(43) From the raw data (m/e, relative intensity): 536, 5 ; 537, 25;
+
538, 100; 539, 62; 540, 17; 541, 3. The values of (P l ) / P = 41.2%
(38) H. W. Whitlock and R. Hanauer, J . Org. Chem., 33, 2169 and (P + 2)/P = 9.0% were calculated aia the formulas in "Mass and
(1968). Abundance Tables for Use in Mass Spectroscopy," J. H. Beynon and
(39) R. Hanauer, unreported work. A. E. Williams, Ed., Elsevier Publishing Co., New York, N. Y., 1963.
Journal of the American Chemical Society 1 91:26 1 December 17, 1969
7489
from methanol afforded 72 mg (84%) of octaethyl cis-bacterio- 500 ml of water and digested on a steam bath for 1 hr. The cooled
chlorin, mp 130-133" (lit.42 mp 138"). Its uv-visible spectrum benzene layer was washed with 500 ml of cold 3 N hydrochloric
agreed with that reported. acid and was then extracted with four 500-ml portions of 82%
Diimide Reduction of meso-Tetraphenylporphyrin. A mixture of (w/w) phosphoric acid (to remove the porphyrin and chlorin),
2 g (3.2 mmoles) of meso-tetraphenylporphyrin, 1.2 g of ptoluene- aqueous sodium bicarbonate solution, and water, and evaporated.
sulfonylhydrazine, 4.0 g of anhydrous potassium carbonate, and Recrystallization of the residue from benzene afforded 0.50 g (50%
150 ml of dry pyridine was heated with stirring at 105' under ni- yield) of tetraphenylbacteriochlorin:: : A 356 ( E 130,000), 378
trogen. After 2 hr, a solution of 1.29 g of toluenesulfonylhydrazine (160,000), 520 (60,000), 742 (130,000) [lit.47 Amax 742 (120,000),
in 4 ml of pyridine was added. This was repeated at the end of 520 (60,000)]; nmr (CDClJ 6 1.3 (singlet, NH, 2 H), 3.92 (singlet,
4 hr. After heating a total of 6.5 hr the reaction mixture was added 8 H , -CH2CH2-), 7.52 (multiplet, ArH, 12 H), 7.85 (doublet, Au
to a mixture of 1 1. of benzene and 500 ml of water and the mixture = 2 Hz, 4 H, bacteriochlorin HC=CH).
was digested for 1 hr on a steam bath, and cooled, and the separated Anal. Calcd for CHAH38N4: C, 85.39; H, 5.55; N, 8.92.
benzene layer was washed with cold 3 N hydrochloric acid, water, Found: C, 85.06; H,5.87; N,8.92.
and saturated sodium bicarbonate solution. Analysis of the visible meso-Tetraphenylisobacteriochlorin. Zinc meso-tetraphenyl-
spectrum of the benzene solution showed it to be a mixture of chlorin was prepared by refluxing with stirring under nitrogen a
meso-tetraphenylchlorin, 62%, and meso-tetraphenylbacterio- mixture of 1 g of the chlorin and 1 g of zinc acetate dihydrate in 75
chlorin, 38 %. T o the benzene solution was added in one portion ml of pyridine for 15 min. To the reaction flask was then added
400 mg (1.64 mmoles) of tetrachloro-o-quinone and the mixture was 6 g of potassium carbonate and 0.3 g of toluenesulfonylhydrazine.
stirred at room temperature for 1 hr. The benzene to solution was Heating and stirring was continued for 26 hr, 0.3 g of toluenesul-
then washed with 5 % aqueous sodium bisulfite solution, 5 aque- fonylhydrazine being added every hour (the reaction mixture was
ous sodium hydroxide solution, 500 ml of 6 8 x (w/w) phosphoric allowed to stand under nitrogen at room temperature for 8 hr after
acid (to remove residual tetraphenylporphyrin), water, and saturated the first 12 hr of heating). The cooled reaction mixture was par-
aqueous sodium bicarbonate solution, and was dried over anhy- titioned between 200 ml of chloroform and 200 ml of water, and
drous sodium sulfate. Removal of solvent gave 1.8 g of a residue the chloroform layer was washed first with cold 3 N hydrochloric
that was recrystallized from 200 ml of benzene to afford 1.45 g (72 % acid and then with concentrated hydrochloric acid until the acid
yield) oftetraphenylchlorin: A$:' 419 mp (e 190,000), 517 (16,0001, washes were almost colorless (three 200-ml washes). The chloro-
542 (12,000), 598 (6100), 652 (42,000) [lit.46348A, 419 (e 190,OaO), form layer was then washed with a 200-ml portion of aqueous
517 (16,000), 542 (11,OOO), 598 (6000), 652 (42,000)]; nmr (CDCla) sodium bicarbonate solution and then with 500 ml of 85% (w/w)
6 1.3 (broad, 2 H , NH), 4.10 (singlet, 4 H,-CHZCHZ-), 7.6-8.5 phosphoric acid. The deeply colored acid layer was separated
(multiplet, 26 H). This latter band could be resolved into a singlet, and the faintly colored chloroform layer was discarded. The cooled
area 2 H at 6 8.34 and AB quartet of area 4 H ( 6 8.10, ~ 8~ 8.49, phosphoric acid layer was washed with chloroform, diluted with
JAB= 4.5 Hz) assigned to the chlorin ring protons. water to 50% (w/w) concentration, and extracted with chloroform,
meso-Tetraphenylbacteriochlorin. A mixture of 1 g (1.63 mmoles) and the colored chloroform extracts were washed, dried, and con-
of tetraphenylporphyrin, 0.60 g (3.2 mmoles) of toluensulfonyl- centrated to a volume of 30 ml. Methanol, 10 ml, was added to
hydrazine, 2.0 g of anhydrous potassium carbonate, and 75 ml of the hot chloroform and the solution was cooled to - 10" for 2 hr.
dry pyridine were heated at 100" with stirring under nitrogen for Filtration afforded 575 mg (57 % yield) of dark red needles : A$:
12 hr. Every 1.5 hr 0.6 g of toluenesulfonylhydrazine was added. 390 mp ( E l00,000), 408 (sh, 79,000), 516 (lO,ooO), 552 (19,000), 594
The cooled reaction mixture was diluted with 1 1. of benzene and (28,000)[lit.s9X z 6 5 9 4 ( c29,000), 552(22,000), 516 (12,000)].
Acknowledgment. Partial support of this work by
(44) At this stage the material was contaminated by 5 % octaethyl- the National Science Foundation and National In-
chlorin and 2 octaethylbacteriochlorin. stitutes of Health is acknowledged.
(45) G.D.Dorough and F. M. Huennekens, J . Amer. Chem. Soc., 74,
3974 (1952). (47) G.D. Dorough and J. R. Miller, J . Amer. Chem. SOC.,74,6106
(46) J. R. Miller and G. D. Dorough, ibid,, 14,3977 (1952). (1952).
Whitlock, Hanauer, Oester, Bower 1 Diimide Reduction of Porphyrins