J O U R N A L OF THE AMERICAN CHEMICAL SOCIETY (Registered in U. S. Patent Office)
(0Copyright, 1960, b y the American Chemical Society) NUMBER 10
hL4Y 31, 1960
VOLUME S2
PHYSICAL AND INORGANIC CHEMISTRY [CONTRIBUTION FROM THE CHEMISTRY DEPARTMIZNT O F
BRANDEIS UNIVERSITY]
Studies on Metastable States of Porphyrins. 11. Spectra and Decay Kinetics of Tetraphenylporphine, Zinc Tetraphenylporphine and Bacteriochlorophyll1 BY LXURIPEKKARIKEN AND HENRY LINSCHITZ RECEIVED JULY 23, 1959 The flash technique used earlier to study the metastable states of chlorophyll2 has been applied to the above compounds in toluene and pyridine solution. The spectra all show a main band just below the Soret peak, as well as other regularities. k2( C*)2 ks( C*)( C,) (C* = excited state; C, = ground state). T h e respective T h e decay law is again -dC*/dt = klC* rate constants for all the compounds are almost the same, except for a much faster first-order decay in bacteriochlorophyll Little or no flash-bleaching is observed with the Cu-+, Co++and S i + + complexes of tetraphenylporphine
+
+
The absorption spectra and decay kinetics of metastable (triplet) states of chlorophyll-a and b have been studied recently in this Laboratory, using flash-excitation apparatus with a time-resolusecond.2 We have now extended this tion of investigation to bacteriochlorophyll, tetraphenylporphine (TPP) and zinc tetraphenylporphine (ZnTPP) and have also examined other metal complexes of TPP. In addition, absorption spectra of the metastable states of these compounds, as well as chlorophyll-a and b, have been followed into the infrared, as far as 10,500 These new data permit us to begin systematic, comparative study of the triplet states of these spectroscopically interesting and biochemically significant compounds.
A.
Experimental 1. Method.-The method used in this work involves the almost complete conversion of dissolved dye t o its metastable state, by means of a brief and intense light flash, and the measurement of the magnitude and time course of resulting spectral changes in the solution using a scanning beam, monochromator, photo-cell and oscilloscope. A full description of the a p p a r a t ~ soperational ,~ techniques and mode of calculation of results has been given previqusly.2 For measurements in the region from 7000 to 10,500 A . , theusual photomultiplier was replaced by a n inirared sensitive tube (DuMont, Type 6911). At 10,500 A . using a 100-watt zirconium arc source and red filter, this tube gave a signal-
s.
(1) T h i s work was assisted b y a grant from t h e U. Atomic Energy Commission to Brandeis University (Contract No. A T (30-1)-2003). (2) H. Linschitz and K. Sarkanen, THISJOURNAL, 8 0 , 4826 (1958). (3) We wish to thank t h e Chemistry Department and Instrumentation Section of t h e Brookhaven Sational Laboratory for kind permission t o use components of the 5ash apparatus t h a t were developed and constructed there.
to-noise ratio ,of better than 10, at a monochromator slit widbh of 200 A. At 7000 8. slit widths could be reduced to 40 A . All test solutions were prepared on the vacuum line from stock solutions of the pigments in toluene.2 2. Spectrum Measurements.-The determination of the spectra of the metastable states of the porphyrins is rendered difficult by the large variation in intensity of the several triplet and singlet absorption bands. The change in absorbance in each region of the spectrum therefore was measured a t a concentration giving the optimum precision for the particular band in question, up to a point at which the initial fractional conversion began to fall off seriously. The region under the extremely sharp Soret bands of TPP and ZnTPP, which is especially sensitive to slit-width errors, was studied at concentrations near 2 X lo-’ X , while the relatively weak triplet bands in the near infrared required measurements near 10-5 it{. Overlapping data at suitable wave lengths were taken a t intermediate concentrations, so that, assuming Beer’s law, fractional conversions could be checked over the full concentration range. The per cent. conversions, extrapolated back t o “zero time” of the flash,$ were estimated from the height of residual peaks under the original bands. At the lowest concentrations, using the Soret band, conversions were (within 5Yc error) 100 and 927, for TPP and ZnTPP, respectively, and using the far red peak, 90% for bacteriochlorophyll. I n all cases, these remained at 90% or higher up to about 1 X M ,as judged by the constancy (within 10%) of the ratio ADh/C over this interval ( A D A = change in absorbance a t a convenient wave length; C = dye concentration). Only these runs were used to establish the o are small, exspectra so that corrections to 1 0 0 ~ conversion cept under the Soret bands. I n kinetic runs on TPP, a t 1.4 X M , the initial conversion was about 70yocomplete. 3 . Kinetic Measurements.-For convenience in this and the following paper, we briefly review here the kinrtic analysis given earlier.* We let
2407
( CO) = total concn. of dye (C”) = concn. of metastable state (C,) = concn. of ground state
2408
LAURIPEKKARINEN
AND
ADOX= obsd. change in absorbance a t wave length X (referred to the original s o h . ) , at some arbitrary time, t = 0, after the flash ADA = obsd. change in absorbance a t wave length A, at time t after the arbitrary starting point ax*,a,x = molar extinction co-eficients of metastable and ground state 1 = cell length
ax
I
=
-
(ax*
_a,A)L
Thc rate law found to describc the dccay of the triplet state of chlorophyll is2
-"* dt Since
C* =
= klC*
+ kQ(C*)z + k3(C*)(Cg)
U X A D Athis , becomes d - In ( A D o / A D ) x = A dt
in which
A and
k1
+ BxADx
(1)
(2)
+ k,Co
BA = eA(ks
- ks)
Tangents drawn t o plots of log ( A D , / A D ) hus. t give thederivative directly, and the values of A and BA are then found from plots of this derivative against ADA. K1 and ka are separated by measuring the change in A with CO. The spectrum measurements give ah for the wave length at which ADA is observed, and kz is thus evaluated. Our results to date on a number of compounds all indicate t h a t kz > > ka. Moreover, in dilute solutions k , > > k3Co, and the latter term can be neglected.
r-
1
,
._ ,\
j .
,
,,'
~ - - , '
-.__,
4co
spectrum of our bacteriochlorophyll in ether checked within experimental error with that of Holt and J e ~ o b s with , ~ regard to location and relative height of the various peaks, but our molar extinction coefficients were uniformly low by about 60%. However, we assume t h a t the absorption of the metastable state relative to the ground state is still reliable. The rate constant K 1 is independent of the absolute extinction, in any case. I n passing from ether to pyridine, an especially large shift of the "orange peak" occurs, as noted by Weigls for other solvents. The main bacterioclilorophyll peaks (our preparation) in pyridine lie at 780, 612 and 373 m p , compared with 770, 574 and 358 m p in ether. The subsidiary pefk in ether, a t 392 mp, is not resolved in pyridine (Fig. 1). r P P was obtained by condensing pyrrole and benzaldehyde in a sealed tube7*8and was chromatographed several times. The metal complexesQwere purified similarly. The spectra of TPP and ZnTPP checked well with previous measurements.9 Solvents were reagent-grade. Ppridine was refluxed and fractionated over KOI-I. Toluene was washed with HJSOI, water, dried and fractionated over sodium.
Results and Discussion Absorption Spectra.-Figures 1-3 give the absorption spectra of the metastable states of bacteriochlorophyll, T P P and ZnTPP. Peak loca1.
a,
Fig. 2.--Absorption spectra of tetraphenylporphine {dashed line) and its triplet state (solid line), corrected t o 100% conversion, in toluene.
:\ , - '
ic3
