Photoinduced alkaline pH-jump on the nanosecond time scale

Article abstract of DOI:10.1016/S0009-2614(01)00807-7

The triphenylmethane leucohydroxide salt 4-(dimethylamino)-4^′-(trimethylammonium) triphenylmethanol iodide has been used as a photoactivatable caged hydroxide to rapidly increase the pH of a neutral aqueous solution on the nanosecond time scale

Full text of DOI:10.1016/S0009-2614(01)00807-7

24 August 2001
Chemical Physics Letters 344 +2001) 387±394
www.elsevier.com/locate/cplett
Photoinduced alkaline pH-jump on the nanosecond time scale
Stefania Abbruzzetti a, Mauro Carcelli b, Paolo Pelagatti b,
Dominga Rogolino b, Cristiano Viappiani a,*
a
Dipartimento di Fisica, Universitaꢀ di Parma and Istituto Nazionale per la Fisica della Materia,
Parco Area delle Scienze 7A, I-43100, Parma, Italy
b
Dipartimento di Chimica Generale ed Inorganica, Chimica, Analitica, Chimica Fisica, Universitaꢀ di Parma, Parma, Italy
Received 28 May 2001
Abstract
The triphenylmethane leucohydroxide salt 4-+dimethylamino)-40-+trimethylammonium) triphenylmethanol iodide
has been used as a photoactivatable caged hydroxide to rapidly increase the pH of a neutral aqueous solution on the
nanosecond time scale. Photoexcitation of the compound with a nanosecond ultraviolet laser pulse results in the for-
mation of a colored carbocation with a rate greater than 108 sÀ1; recombination occurs with a multiexponential
kinetics. The dissociation is completely reversible on time scales of 102 s and o€ers a large time window after the pH-
jump, in which the proton transfer reactions can be studied. The e€ectiveness of the photodissociation process to
increase the pH of the solution has been demonstrated by means of the pH indicator bromoxylenol blue. Ó 2001
Elsevier Science B.V. All rights reserved.
1. Introduction
investigating fast events in protein folding/un-
folding triggered by proton transfer reactions. To
this purpose, we have recently reported the use of
photoactivatable caged protons to induce acid
perturbations in solutions containing model poly-
peptides and proteins [5,6,13]. Proton transfer re-
actions are among the fastest known reactions and
their kinetic characterization has become feasible
only after the introduction of photoactivatable
caged compounds that release protons or hydro-
xide ions [4,14]. Light-induced acidi®cation was
generally obtained by using aromatic alcohols or
o-nitrobenzyl derivatives, whereas alkalinization
was obtained by means of N-etherosubstituted
aromatic compounds [3,14,15]. In solution, aro-
matic alcohols and N-etherosubstituted aromatic
compounds a€ord short transient pH pulses,
which disappear within a few microseconds after
the molecules have relaxed to their electronic
There is a general interest in compounds that
are able to release an active species under con-
trolled conditions, the so-called `caged com-
pounds'. In particular, photocleavable caged
compounds are a precious tool in the study of
biological processes [1±3], especially in time-re-
solved studies of cellular functions and of macro-
molecular
properties
[4±6].
The
recent
development of time-resolved techniques based on
pulsed-lasers has opened the possibility of in-
vestigating protein folding reactions on sub-
millisecond time scales [7±12]. We are interested in
* Corresponding author. Fax: +39-521-905223.
E-mail address: cristiano.viappiani@®s.unipr.it +C. Via-
ppiani).
0009-2614/01/$ - see front matter Ó 2001 Elsevier Science B.V. All rights reserved.
PII: S 0 0 0 9 - 2 6 1 4 + 0 1 ) 0 0 8 0 7 - 7

388
S. Abbruzzetti et al. / Chemical Physics Letters 344 ,2001) 387±394
2. Materials and methods
ground states [4,14±16]. On the contrary, com-
pounds as o-nitrobenzaldehyde photorelease irre-
versibly the proton and achieve a step decrease in
pH [17±19]. An interesting class of photosensitive
compounds are triarylmethanes that, depending
on the solvent, undergo either homolytic cleavage
to give radicals [20,21] or heterolytic cleavage [22]
upon ultraviolet illumination. Solutions of trip-
henylmethane leucohydroxide derivatives in polar
solvents undergo heterolytic cleavage within a few
nanoseconds from photolysis, proceeding from the
excited singlet state [23,24]. Mostly, the tertiary
alcohol is colorless, with a peak absorbance at 254
nm, whereas the carbocation formed upon pho-
tolysis is characterized by a green±blue color with
absorption maxima at 420 and 620 nm [25]. The
hydroxide ion is then reversibly bound via a
thermal reaction which is complete in a few min-
utes [26]. In principle, a rapid increase in pH can
be used to refold acid-denatured proteins, refold
basic proteins from neutral solutions [27] or unfold
neutral solutions of native proteins [28,29]. How-
ever, the photodissociation of triarylmethanes can
be induced in aqueous solution at pH above 6
[25,26]. Therefore, the use of these compounds to
photoinduce the alkalinization of the solution is
limited to refolding/unfolding studies starting from
neutral solutions. In this work we have performed
a preliminary study on the kinetics of the recom-
bination reactions that follow the nanosecond
photodissociation of 4-+dimethylamino)-40-+trime-
thylammonium) triphenylmethanol iodide ++1) in
Scheme 1). The utility of the compound in kinetic
investigations is shown for the alkaline induced
deprotonation of the pH indicator bromoxylenol
blue on the submillisecond time scale.
The compound 4,40-bis+dimethylamino) triphe-
nylmethane leucohydroxide was synthesized by
reaction between 4,40-bis+dimethylamino)benzo-
phenone and phenyl lithium in ether and sub-
sequent hydrolysis with methanol. In order to
enhance its solubility in water, the salt 4-+di-
methylamino)-40-+trimethylammonium) triphenyl-
methanol iodide ++1) in Scheme 2) was prepared by
treating in toluene the neutral leucohydroxide with
freshly distilled methyl iodide [26].
In the pH-jump experiments, the solutions were
nitrogen saturated to avoid competition by car-
bonate bu€ering. The pH of the solutions was
adjusted by addition of concentrated aqueous
NaOH or HCl. Concentrations of +1) in ¯ash
photolysis experiments were kept at about 0.6
mM, a€ording an absorbance of about 1 +1 cm
pathlength) at 308 nm.
The transient absorption experiments were
conducted with an experimental setup already
described [30]. Photoexcitation at 308 nm was
obtained with an excimer laser +EMG50, Lambda
Physik, 8 ns FWHM, 25 mJ). The beam was fo-
cused to a line with a 0.5 m cylindrical lens. The
probe light in the pH-jump experiments and in the
determination of the overall time course of the
carbocation formation and recombination was a
cw HeNe laser +Nec Corporation, 633 nm, 10
mW). The monitoring beam was passed inside a
rectangular quartz cuvette +4 mm  10 mm) along
the 10 mm path, at right angle with the pump
beam. The cuvette was held in a temperature
controlled holder. The holder was ¯ushed with
nitrogen to prevent condensation at low tempera-
tures and to minimize CO2 uptake by the solutions
during the kinetic experiments. The transmitted
(2)
(3)
(1)
Scheme 1.
Scheme 2.

S. Abbruzzetti et al. / Chemical Physics Letters 344 ,2001) 387±394
389
intensity of the cw beam was monitored by a large
area Si avalanche photodiode +Hamamatsu,
S2385) and the output voltage was ampli®ed using
two cascaded broadband operational ampli®ers
+Burr Brown, OPA643, gain Â49). The probe
beam was passed through a monochromator +H25,
Jobin Yvon) before being focused onto the Si
avalanche photodiode to remove the high intensity
stray light from the pump laser. The voltage signal
was digitized by a digital sampling oscilloscope
+LeCroy 9370, 1 GHz, 1 Gs/s). Typically 16 traces
were averaged at the repetition rate of 1 shot every
3 min, to get a single signal. Care was taken in
order to avoid thermal lensing e€ects in the mea-
sured signal.
Transient spectra were measured using a cw Xe
lamp as detection source and removing the cy-
lindrical lens from the pump beam path [31].
3. Results and discussion
3.1. Photodissociation and rebinding of the hydro-
xide ion
Fig. 1A and B shows the time course of the
transient absorbance of an aqueous solution of
compound +1), measured at 633 nm, following
photolysis by a 308 nm laser pulse; the pre-pulse
pH was 10.9 and T ˆ 293 K. As expected, pho-
tolysis of +1) at pre-pulse pH above 6 leads to the
formation of a colored species with rate constant
above the experimental resolution of the setup
+ꢀ108 sÀ1). The measured transient spectrum at
the end of the laser pulse +see Fig. 1C) is consistent
with the reported spectrum of the carbocation,
formed upon heterolytic cleavage of the parent
carbinol [24]. The decay of the transient absor-
bance occurs with a multiphasic behavior, ex-
tending from the nanosecond time scale +Fig. 1A)
to the tens of seconds +Fig. 1B). We propose +vide
infra) that the multiphasic decay of the transient
absorbance can be associated with recombination
reactions of the carbocation with hydroxide from
the solution, characterized by di€erent rate con-
stants. The slowest recombination was known
from the original work of Irie [26]. In our experi-
ments we consistently found the presence of ad-
Fig. 1. Transient absorption at 633 nm after photolysis with a
308 nm laser pulse of an aqueous solution of compound +1) +0.6
mM) at pre-pulse pH ˆ 10.9. The trace is the average of 16
single shots. +A) The rise in the absorbance occurs below the
experimental resolution, whereas the decay on the sub-micro-
second time scale can be described by a monoexponential decay
function plus an o€set +solid line). +B) On longer time scales, the
remaining carbocations recombine with hydroxide with
a
complex kinetics, extending from the microseconds to the tens
of seconds. The solid line is the result of a ®t to a three ex-
ponential decay function. +C) Transient spectrum measured at
the end of the laser pulse.

390
S. Abbruzzetti et al. / Chemical Physics Letters 344 ,2001) 387±394
ditional faster reactions, evidenced in Figs. 1A and
B.
The slow recombination allows the investiga-
tion of proton transfer reactions in a time window
extending from about 500 ns to hundreds of mil-
liseconds. The longer accessible time range de-
pends on how tight the photolysis beam is focused
[30]. With our setup the largest time window,
which is not a€ected by di€usional motions of the
solution, extends to about 500 ms.
The solid line in Fig. 1A is the result of a ®t to a
single exponential decay +plus an o€set). The re-
covered rate constant k1 ˆ 1=s1 ˆ ꢁ1:1 Æ 0:1† Â
107 sÀ1 +Table 1) is essentially independent of the
pre-pulse pH in the interval 6.2±11.4. This ®nding
proves that the recombination rate is not de-
termined by the bulk concentration of free hy-
droxide: in that case, an increase of the rate k1 with
pH should have been observed.
Increasing the ionic strength of the solution by
addition of KCl reduces the rate k1, as shown in
Fig. 2 for two representative concentrations. The
data were analyzed according to the equation [32]
Fig. 2. Transient absorbance at 633 nm following photo-
excitation of a solution of +1) at pre-pulse pH ˆ 10.9 in the
presence of two di€erent ionic strengths. Curve A: [KCl] ˆ 1.5
mM; curve B: [KCl] ˆ 84 mM.
could be envisioned as a `geminate' recombination
of the hydroxide still in the electrostatic cage of the
p
‡
carbocation, situation indicated as +C Á Á Á OHÀ†
À Á
log ꢁk1† ˆ log k10 ‡ 1:02zAzB I;
ꢁ1†
+Eq. +2)). The geminate ion pair can either evolve
into the ground state carbinol or into the separated
ions di€using freely in solution:
where k10 and k1 are the rates observed in the ab-
sence and in the presence of KCl, respectively, I is
the ionic strength of the solution, and zA and zB are
the charges of the reacting ions. A plot of
logꢁk1=k10) as a function of the square root of the
ionic strength shows a linear behavior +Fig. 3) with
a slope of À0:96 Æ 0:05, indicating that the reac-
tion being monitored is characterized by charged
reactants with an e€ective charge product
zA Â zB ꢀ À1. The ionic strength dependence sug-
gests that the reaction being observed is the re-
combination between the carbocation +charge +1)
and the hydroxide ion +charge )1). This process
kb
ka
‡
‡
COH ¢ ꢁC Á Á Á OHÀ† ¢ C ‡ OHÀ
ꢁ2†
k
Àa
k
Àb
In our picture the forward rate ka is much smaller
than the backward rate kÀa, since at pH above
neutrality the equilibrium ground state carboca-
tion concentration is negligible. The forward es-
cape rate kb is likely much larger than kÀb, since
from our data there is no evidence of pH depen-
dence of the overall recombination rate even at the
highest pH investigated +pH ˆ 11.4). The geminate
Table 1
Rate constants and activation parameters for the recombination reactions in aqueous solutions at pre-pulse pH ˆ 10.9:1 is the geminate
rebinding, 2 and 3 are the pH dependent second-order processes
1
2
3
a
ki ˆ 1=si ꢁsÀ1
†
ꢁ1:1 Æ 0:1†  107
14 Æ 4
ꢁ2:7 Æ 0:3†  104
16 Æ 5
ꢁ3:0 Æ 0:3†  103
Ea ꢁkJ molÀ1
†
ꢀ0
DHz ꢁkJ molÀ1
†
12 Æ 4
13 Æ 5
ꢀ0
DSz ꢁJ KÀ1molÀ1
a At T ˆ 293 K.
†
À68 Æ 12
À116 Æ 17
À181 Æ 10

S. Abbruzzetti et al. / Chemical Physics Letters 344 ,2001) 387±394
391
and s3, plus an o€set; the corresponding rates,
k2 ˆ 1=s2 and k3 ˆ 1=s3, measured at T ˆ 293 K
and pre-pulse pH ˆ 10.9 are reported in Table 1.
These two decays are well separated +approxi-
mately one order of magnitude). Both k2 and k3
increase with increasing pH suggesting that this
biexponential relaxation is associated with the re-
combination reaction between hydroxide and
carbocation. Attempts to ®t the data with a sec-
ond-order reaction scheme [32] did not lead to
satisfactory results and therefore we present only a
qualitative description of the kinetics of this reac-
tion. Plots of the apparent rates k2, and k3, mea-
sured at di€erent pre-pulse pH values +not shown),
were linear for both parameters. From the slopes
we could estimate the biÀm1 olecular rate constants
Fig. 3. Plot of logꢁk1=k10† as a function of the square root of the
ionic strength. k10 and k1 are the rates observed with di€erent
ionic strengths. The ionic strength was varied by addition of
KCl +T ˆ 293 K, pre-pulse pH ˆ 10.9).
k2b ˆ ꢁ1:3 Æ 0:5†  107 M
s
À1, and k3b ˆ ꢁ1:6 Æ
0:5†  106 MÀ1 sÀ1
.
Finally, the concentration of the carbocation
vanishes within a few minutes in a further re-
combination phase, that is coupled with di€usional
processes mixing the solution +Fig. 1B). This phase
is consistent with the slow recombination pre-
viously reported [26]. There is no residual absor-
bance change at very long times, indicating that
the carbocations have completely reacted with the
hydroxide to regenerate compound +1). This last
phase cannot be properly described in terms of
rate constants, since the measured apparent rate is
mixed with the contributions from di€usional
motions of the solution through the illuminated
volume.
ion pair disappears through two concomitant, es-
sentially unidirectional, reactions. In terms of the
resulting rate constant, the measured rate k1 can be
written as:
k1 ꢀ k ‡ kb;
ꢁ3†
Àa
where the rates ka and k have been neglected.
Àb
The possibility that the nanosecond decay is
due, at least partly, to a triplet±triplet absorbance
cannot be completely ruled out. However, the io-
nic strength dependence of the rate constant
strongly suggests that the ionic reaction dominates
this decay. In addition, we did not observe di€er-
ences in the lifetime of this transient when ex-
periments were conducted either in nitrogen or in
air saturated solutions. On the contrary, collisional
quenching by molecular oxygen is generally ob-
served for triplet states.
In conclusion, the overall recombination pro-
cess of the freely di€using ions can be schemati-
cally summarized as
C ‡ OHÀ ¢ COH
ꢁHeterogeneous recombination†
‡
ꢁ4†
The observed multiple exponential rebinding
shows that a heterogeneous kinetics characterizes
the reaction between hydroxide and the carboca-
tion. The source of this heterogeneity is not clear at
the moment, but the well-known presence of me-
someric forms of the carbocation ++2) and +3) in
Scheme 2) could be implicated. The kinetics of
OHÀ rebinding could be modulated by the di€erent
electrophilic centers of the di-cation: the central
A pH-dependent recombination phase is de-
tected in the long microseconds, as shown in a
semi-log plot in Fig. 1B. The transient spectrum at
the end of the geminate recombination phase is
identical to the spectrum measured at the end of
the laser pulse +reported in Fig. 1C), suggesting
that the species with longer lifetimes can be iden-
ti®ed with the carbocation. This phase is best de-
scribed by a biexponential decay, with lifetimes s2

392
S. Abbruzzetti et al. / Chemical Physics Letters 344 ,2001) 387±394
carbon +2), the dimethylamino nitrogen +3) or the
trimethylammonium group ++2) or +3)). Further
experiments will be necessary to understand the
complete mechanism underlying the reported data.
The temperature dependence of the rebinding
kinetics was measured in order to estimate the
activation parameters of the processes. Table 1
reports the activation parameters determined ex-
perimentally for the ®rst three decays through the
Arrhenius +Eq. +5)) or the Eyring +Eq. +6)) rela-
tions:
an increase in the absorbance at 633 nm is ex-
pected to occur. Fig. 4 reports the transient ab-
sorbance at 633 nm for a solution of the caged
hydroxide +1) alone +A) and in the presence of
bromoxylenol blue +B) at pre-pulse pH ˆ 7.2.
When the solution is ¯ashed with the UVlaser, the
hydroxide release increases the pH. This increase
in [OHÀ] can be estimated from the absorbance at
t ˆ 1 ls by using the carbocation molar extinction
coecient eꢁ633 nm† ˆ 215 MÀ1 cmÀ1. The ab-
sorbance, for trace A, represents the change in
carbocation concentration at that time. When the
indicator is present, the increase in [OHÀ] shifts
the equilibrium of bromoxylenol blue towards the
deprotonated form +InÀ) according to the kinetic
scheme:
À Á
Ea
RT
ln ꢁki† ˆ ln ki0
À
ꢁ5†
ꢁ6†
ꢀ
ꢁ
kih
kBT
DSz DHz
ln
ˆ
À
R
RT
In these equations ki ˆ 1=si is the rate constant of
process i +i ˆ 1, 2 or 3), h is Planck's constant, kB is
Boltzmann's constant, R is the gas constant and T
is the temperature +K). When using Eq. +5), the
activation energy +Ea) was obtained from the slope
of the linear plot of lnꢁki† vs 1=T +not shown).
When the Eyring relation was used, the activation
entropy and enthalpy +DSz and DHz† were ob-
tained from the intercept and the slope of the lin-
ear plot of ln ꢁkih=kBT) vs 1=T +not shown). The
enthalpic barriers are relatively small +processes 1
and 2) or even negligible +process 3) for all of the
processes. On the contrary, the entropic barriers
are relevant and all negative, indicating that a
substantial entropic barrier must be crossed to
restore compound +1).
kc
OHÀ ‡ H ¢ H2O
ꢁ7†
ꢁ8†
ꢁ9†
‡
k
Àc
kd
HIn ¢ H ‡ InÀ
‡
k
Àd
ke
HIn ‡ OHÀ ¢ H2O ‡ InÀ
k
Àe
The kinetics of the deprotonation reaction is re-
¯ected in the rise of the absorbance in trace B. The
3.2. Deprotonation of the pH indicator bromoxyle-
nol blue
The photoinduced release of hydroxide ions can
be used to induce deprotonation of a weak acid
with suitable pKa. To show that this reaction can
be e€ectively performed, a neutral aqueous solu-
tion of the caged hydroxide +1) was photo-
dissociated in the presence of the pH indicator
bromoxylenol blue ꢁpKa ˆ 7:2†. Transient ab-
sorption was monitored at 633 nm, where bro-
moxylenol blue exhibits an absorption band
characteristic of the deprotonated form. The acid
form has a very low absorbance at this wave-
length. Thus, upon deprotonation of the indicator,
Fig. 4. Transient absorbance at 633 nm following photo-
excitation at 308 nm of a solution containing compound +1) at
pre-pulse pH ˆ 7.2 +trace A). Trace B reports the signal ob-
served when bromoxylenol blue is present.

S. Abbruzzetti et al. / Chemical Physics Letters 344 ,2001) 387±394
393
be used to photoinduce the release of OHÀ ions on
submicrosecond time scale. The reaction is re-
versible on long time scales with no apparent fa-
tigue for the compound. This peculiarity allows
the use of the compound in experiments where
several laser shots must be used to reduce the noise
level. The relaxation kinetics of proton transfer
reactions after the pH-jump can be followed in a
time window extending from about 500 ns to
hundreds of milliseconds.
increase in absorbance is characterized by a single
exponential relaxation and the measured lifetime is
s ˆ 8:6 Æ 0:2  10À6 s, resulting in an apparent
rate constant k ˆ 1:16 Æ 0:03  105 sÀ1. Equilib-
rium +7), with rate constants kc ˆ 1:4  1011 MÀ1
s
À1, and k ˆ kc  10ꢁÀ15:76† plays a minor role in
Àc
the observed relaxation. It is known that the rate
constants for the equilibrium +8) are kd ˆ 2:7  103
sÀ1 and k ˆ 4:3  1010 MÀ1 sÀ1 [30]; relaxation
Àd
of equilibrium +9) occurs with rates ke ˆ 8:9  109
MÀ1 sÀ1 and k ˆ ke  10ꢁpKaÀ15:76† [14,30]. The
Àe
measured kinetics is not compatible with the sim-
ple relaxation of equilibrium +8), since the resulting
deprotonation rate would be much smaller than
the measured one. At the present [OHÀ]
+ꢀ13 lM), the dominant reaction is due to equi-
librium +9). A simple estimate of the bimolecular
rate constant using the kinetic schemeÀo1 utlined
Acknowledgements
The authors acknowledge INFM +PRA CADY)
and CNR +Progetto Strategico Biosensori) for the
®nancial support.
above gives a value of ke ˆ 8:9  109 M
s
À1, in
perfect accordance with the previously reported
data [30].
References
On time scales longer than 100 ls the re-
combination of the carbocation with hydroxide
leads to the protonation of bromoxylenol blue and
to the observed decrease in absorbance.
[1] G. Marriott +Ed.), Caged compounds. Methods in
Enzymology, vol. 291, 1998, Academic Press, New York.
[2] J.E.T. Corrie, Y. Katayama, G.P. Reid, M. Anson, D.R.
Trentham, Philos. Trans. R. Soc. Lond. A 340 +1992) 233.
[3] J.A. McCray, D.R. Trentham, Ann. Rev. Biophys.
Biophys. Chem. 18 +1989) 239.
The main kinetic feature in trace +A) of Fig. 4 is
the pH dependent recombination phase +transients
with lifetimes s2 and s3) occurring between 100 ls
and a few milliseconds. This phase is much less
evident, if at all, in trace +B). The decrease in ab-
sorbance at 633 nm for bromoxylenol blue +trace
+B) in Fig. 4) is the result of the relaxation of the
coupled chemical equilibria described in Eqs.
+4),+7)±+9). The coupling between reactions makes
the carbocation-OHÀ recombination rate much
smaller than in the absence of other competitive
processes [4,30,33]. In particular, at alkaline pH,
the rate of formation of HIn is mostly determined
[4] M. Gutman, E. Nachliel, Ann. Rev. Phys. Chem. 48 +1997)
329.
[5] S. Abbruzzetti, E. Crema, L. Masino, A. Vecli, C.
Viappiani, J.R. Small, L.J. Libertini, E.W. Small, Biophys.
J. 78 +1) +2000) 405.
[6] S. Abbruzzetti, C. Viappiani, J.R. Small, L.J. Libertini,
E.W. Small, Biophys. J. 79 +2000) 2714.
[7] W.A. Eaton, V. Munoz, S.J. Hagen, G.S. Jas, L.J. Lapidus,
E.R. Henry, J. Hofrichter, Ann. Rev. Biophys. Biomol.
Struct. 29 +2000) 327.
[8] J. Ervin, J. Sabelko, M. Gruebele, J. Photochem. Photo-
biol. B: Biol. 54 +2000) 1.
[9] H. Roder, M.C.R. Shastry, Curr. Opinion Struct. Biol. 9
+1999) 620.
‡
by the low concentration of H , reacting with InÀ.
[10] M. Gruebele, Ann. Rev. Phys. Chem. 50 +1999) 485.
[11] K.C. Hansen, R.S. Rock, R.W. Larsen, S.I. Chan, J. Am.
Chem. Soc. 112 +2000) 11567.
Under the present experimental conditions this
implies apparent lifetimes in the tens of milli-
seconds [30].
[12] T. Okuno, Biochemistry 39 +2000) 7538.
[13] C. Viappiani, S. Abbruzzetti, J.R. Small, L.J. Libertini,
E.W. Small, Biophys. Chem. 73 +1998) 13.
[14] M. Gutman, E. Nachliel, Biochim. Biophys. Acta 1015
+1990) 391.
4. Conclusions
[15] P. Wan, D. Shukla, Chem. Rev. 93 +1993) 571.
[16] L.G. Arnaut, S.J. Formosinho, J. Photochem. Photobiol.
A: Chem. 75 +1993) 1.
The
compound
4-+dimethylamino)-40-+tri-
methylammonium) triphenylmethanol iodide can
[17] M.V. George, J.C. Scaiano, J. Phys. Chem. 84 +1980) 492.

394
S. Abbruzzetti et al. / Chemical Physics Letters 344 ,2001) 387±394
[18] P. Pelagatti, M. Carcelli, C. Viappiani, Isr. J. Chem. 38
+1998) 213.
[26] M. Irie, J. Am. Chem. Soc. 105 +1983) 2078.
[27] Y. Goto, Y. Hagihara, Biochemistry 31 +1992) 732.
[28] F.I. Rosell, J.C. Ferrer, A.G. Mauk, J. Am. Chem. Soc.
120 +1998) 11234.
[19] G. Bonetti, A. Vecli, C. Viappiani, Chem. Phys. Lett. 269
+1997) 268.
[20] G.N. Lewis, D. Lipkin, T.T. Nagel, J. Am. Chem. Soc. 66
+1944) 1579.
[29] C.J. Nelson, B.E. Bowler, Biochemistry 39 +2000) 13584.
[30] C. Viappiani, G. Bonetti, M. Carcelli, F. Ferrari, A.
Sternieri, Rev. Sci. Instrum. 69 +1) +1998) 270.
[31] B. Pispisa, L. Stella, M. Venanzi, A. Palleschi, C.
Viappiani, A. Polese, C. Toniolo, Macromolecules 33 +3)
+2000) 906.
[21] G. Porter, E. Strachan, Trans. Faraday Soc. 54 +1958) 1595.
[22] L. Harris, J. Kaminsky, R.G. Simard, J. Am. Chem. Soc.
57 +1935) 1151.
[23] R.N. Manchair, Potochem. Photobiol. 6 +1967) 779.
[24] L.E. Manring, K.S. Peters, J. Phys. Chem. 88 +1984) 3516.
[25] A. Granzow, A. Wilson, F. Ramirez, J. Am. Chem. Soc. 96
+1974) 2454.
[32] K.J. Laidler, Chemical Kinetics, HarperCollins Publishers,
Inc, New York, 1987.
[33] Y. Marantz, E. Nachliel, Isr. J. Chem. 39 +1999) 439±445.