Quenching of a photosensitized dye through single-electron transfer from trivalent phosphorus compounds

Article abstract of DOI:10.1021/jo000448i

Various types of trivalent phosphorus compounds 1 undergo single-electron transfer (SET) to the photoexcited state of rhodamine 6G (Rho⁺(*)) in aqueous acetonitrile to quench the fluorescence from Rho⁺(*). The rate constants k(p) for the overall SET process were determined by the Stern-Volmer method. The rate is nearly constant at a diffusion-controlled limit in the region of E( 1/2 )(1) < 1.3 V (vs Ag/Ag⁺), whereas log k(p) depends linearly on E( 1/2 )(1) in the region of E( 1/2 )(1) > 1.3 V, the slope of the correlation line being -αF/RT with α = 0.2. The potential at which the change in dependence of log k(p) on E( 1/2 )(1) occurs (1.3 V) is in accordance with the value of E( 1/2 )(Rho⁺(*)) (1.22 V) that has been obtained experimentally. Thus, the SET step is exothermic when E( 1/2 )(1) < 1.3 V and endothermic when E( 1/2 )(1) > 1.3 V. The α-value (0.2) obtained in the endothermic region shows that the SET step from 1 to Rho⁺(*)is irreversible in this region. Trivalent phosphorus radical cation 1(·+) generated in the SET step undergoes an ionic reaction with water in the solvent rapidly enough to make the SET step irreversible. In contrast, the SET from amines 2 and alkoxybenzenes 3 to Rho⁺(*) is reversible when the SET step is endothermic, meaning that the radical cations 2(·+) and 3(·+) generated in the SET step undergo rapid 'back SET' in the ground state to regenerate 2 and 3.

Full text of DOI:10.1021/jo000448i

J . Org. Chem. 2000, 65, 4715-4720
4715
Qu en ch in g of a P h otosen sitized Dye th r ou gh Sin gle-Electr on
Tr a n sfer fr om Tr iva len t P h osp h or u s Com p ou n d s
Shinro Yasui,*,† Munekazu Tsujimoto,† Kenji Itoh,‡ and Atsuyoshi Ohno‡,§
Tezukayama College, Gakuen-Minami, Nara, 631-8585, J apan, and Institute for Chemical Research,
Kyoto University, Uji, Kyoto, 611-0011, J apan
yasui@tandai.tezukayama-u.ac.jp
Received March 24, 2000
Various types of trivalent phosphorus compounds 1 undergo single-electron transfer (SET) to the
photoexcited state of rhodamine 6G (Rh o+*) in aqueous acetonitrile to quench the fluorescence
from Rh o+*. The rate constants kp for the overall SET process were determined by the Stern-
Volmer method. The rate is nearly constant at a diffusion-controlled limit in the region of E1/2(1)
< 1.3 V (vs Ag/Ag+), whereas log kp depends linearly on E1/2(1) in the region of E1/2(1) > 1.3 V, the
slope of the correlation line being -RF/RT with R ) 0.2. The potential at which the change in
dependence of log kp on E1/2(1) occurs (1.3 V) is in accordance with the value of E1/2(Rh o+*) (1.22
V) that has been obtained experimentally. Thus, the SET step is exothermic when E1/2(1) < 1.3 V
and endothermic when E1/2(1) > 1.3 V. The R-value (0.2) obtained in the endothermic region shows
that the SET step from 1 to Rh o+* is irreversible in this region. Trivalent phosphorus radical
cation 1•+ generated in the SET step undergoes an ionic reaction with water in the solvent rapidly
enough to make the SET step irreversible. In contrast, the SET from amines 2 and alkoxybenzenes
3 to Rh o+* is reversible when the SET step is endothermic, meaning that the radical cations 2•+
and 3•+ generated in the SET step undergo rapid “back SET” in the ground state to regenerate 2
and 3.
In tr od u ction
way in this model. This model in fact explains well a
kinetic behavior in many types of photoinduced SET
reactions and has eventually been found applicable to
SET reactions occurring in the dark.6 Nevertheless, in
some SET reactions, dependence of SET rate on free-
energy change is weaker than predicted by the Rehm-
Weller’s model at the region where the SET step is
endothermic.7-12 Schuster showed that such a kinetic
behavior is observed when the SET step is followed by a
rapid chemical reaction of the resulting radical pair;13
SET takes place in an irreversible way in this case.
A family of rhodamine dyes is a useful tool to study
SET processes kinetically, because a dye of this family
is intensely fluorescent on photoexcitation, and the
fluorescence is quenched through SET from one-electron
donors.14,15 That is, a rate constant of SET to this dye in
the photoexcited state is determined easily based on the
Stern-Volmer (SV) method. We examined the photore-
Trivalent phosphorus compounds Z3P readily undergo
single-electron transfer (SET) to good electron acceptors,
generating the corresponding trivalent phosphorus radi-
cal cations Z3P•+ 1,2
To elucidate energetics of SET
.
processes from Z3P is of great importance from a mecha-
nistic point of view and also because Z3P•+ has synthetic
potential.3 However, there has been no report so far,
except for the one from our group,4 to discuss such SET
processes quantitatively.
Rehm and Weller carried out kinetics on fluorescence
quenching occurring through SET in a number of donor-
acceptor pairs.5 To explain a relationship found between
SET rate and free-energy change, they presented a
mechanistic model by postulating that a radical pair
generated in the SET step disappears mainly through a
rapid back SET; that is, SET takes place in a reversible
* To whom correspondence should be addressed. Tel: +81-742-41-
4764; Fax: +81-742-41-4762.
(6) (a) Schlesener, C. J .; Amatore, C.; Kochi, J . K. J . Am. Chem.
Soc. 1984, 106, 3567-3577. (b) Patz, M.; Kuwahara, Y.; Suenobu, T.;
Fukuzumi, S. Chem. Lett. 1997, 567-568. (c) Baciocchi, E.; Lanza-
lunga, O.; Lapi, A.; Manduchi, L. J . Am. Chem. Soc. 1998, 120, 5783-
5787.
† Tezukayama College.
‡ Kyoto University.
§ Present address: Fukui University of Technology, Fukui, 910-8505,
J apan.
(1) (a) Yasui, S. Rev. Heteroatom Chem. 1995, 12, 145-161 and
references therein. (b) Yasui, S.; Shioji, K.; Tsujimoto, M.; Ohno, A.
Chem. Lett. 1995, 783-784. (c) Yasui, S.; Tsujimoto, M.; Shioji, K.;
Ohno, A. Chem. Ber./ Recueil 1997, 130, 1699-1707. (d) Yasui, S.;
Shioji, K.; Tsujimoto, M.; Ohno, A. J . Chem. Soc., Perkin Trans. 2 1999,
855-862.
(7) Grimshaw, J .; Langan, J . R.; Salmon, G. A. J . Chem. Soc.,
Faraday Trans. 1994, 90, 75-81.
(8) Save´ant, J .-M. Acc. Chem. Res. 1993, 26, 455-461.
(9) Bank, S.; J uckett, D. A. J . Am. Chem. Soc. 1975, 97, 567-573.
(10) Gardner, H. C.; Kochi, J . K. J . Am. Chem. Soc. 1975, 97, 1855-
1865.
(2) (a) Powell, R. D.; Hall, C. D. J . Am. Chem. Soc. 1969, 91, 5403-
5404. (b) Pandey, G.; Hajra, S.; Ghorai, M. K. Tetrahedron Lett. 1994,
35, 7837-7840.
(11) Chan, T. W.; Bruice, T. C.; J . Am. Chem. Soc. 1977, 99, 7287-
7291.
(12) Thomas, M. J .; Foote, C. S. Photochem. Photobiol. 1978, 27,
683-693.
(3) (a) Ganapathy, S.; Dockery, K. P.; Sopchik, A. E.; Bentrude, W.
G. J . Am. Chem. Soc. 1993, 115, 8863-8864. (b) Soma Sekhar, B. B.
V.; Bentrude, W. G. Tetrahedron Lett. 1999, 40, 1087-1090.
(4) Yasui, S.; Itoh, K.; Tsujimoto, M.; Ohno, A. Chem. Lett. 1998,
1019-1020.
(13) Schuster, G. B. J . Am. Chem. Soc. 1979, 101, 5851-5853.
(14) Nakashima, N.; Yoshihara, K.; Willig, F. J . Chem. Phys. 1980,
73, 3553-3559.
(15) Itoh, K.; Chiyokawa, Y.; Nakao, M.; Honda, K. J . Am. Chem.
Soc. 1984, 106, 1620-1627.
(5) Rehm, D.; Weller, A. Isr. J . Chem. 1970, 8, 259-271.
10.1021/jo000448i CCC: $19.00 © 2000 American Chemical Society
Published on Web 06/29/2000

4716 J . Org. Chem., Vol. 65, No. 15, 2000
Yasui et al.
Sch em e 1
Ta ble 1. P h otolysis of Tr iva len t P h osp h or u s
Com p ou n d s Z3P (1) w ith Rh od a m in e 6G (Rh o+)a
F igu r e 1. Stern-Volmer plots for SET quenching of Rh o+*
by 1. For simplicity, only the plots for selected compounds are
shown.
yield, %b
Z3P
time/h
1c
4
MBd
Ster n -Volm er An a lysis. When the solution of Rh o+
in aq MeCN (the concentration of water is about 5.6 ×
10-2 M) was irradiated at the wavelength of its absorp-
tion maximum (525 nm), intense fluorescence was ob-
served at 545 nm. The fluorescence was quenched by 1,
and the quenching rate constant kp (in M-1 s-1) was
determined at 25 °C based on the Stern-Volmer equa-
tion, eq 1.
1b
1d
1h
5
5
7
75.6
77.9
92.6
23.3
18.6
6.5
98.9
96.5
99.1
[1] ) [Rh o+] ) 5.00 × 10-3 M in MeOH at 25 °C under an
a
argon atmosphere. Irradiation at hν > 450 nm with a sharp-cut
glass-filter. Determined on GC. c Recovered. Material balance.
b
d
action of trivalent phosphorus compounds 1a -m (Z3P; Z
) aryl, alkyl, alkoxy) with rhodamine 6G (Rh o+) in
aqueous acetonitrile (aq MeCN) and found that each of
these phosphorus compounds undergoes SET quenching
of fluorescence from Rh o+ in the photoexcited state,
Rh o+*. Then we determined rate constants of the SET
reactions occurring from 1 to Rh o+* based on the SV
method. Meanwhile, we determined half-wave potentials
of 1a -m and Rh o+*; the values for 1 spread over more
than 1 V, which is equivalent to more than 96.5 kJ mol-1
in the free-energy scale.
I0/I ) 1 + kpτ0[1]
(1)
where I0 and I are fluorescence intensities observed in
the absence and presence of 1, respectively, and τ0 is a
lifetime of fluorescence from Rh o+*, the photoexcited
state of Rh o+, in aq MeCN. The ratio I0/I was found to
correlate linearly with the concentration of 1 with the
intercept being unity for each compound as seen in Figure
1, showing validity of our analysis. The slope of each line
gave the value of kpτ0 as listed in Table 2. The fluores-
cence lifetime of Rh o+* in ethanol has been reported to
be 3.0 ns.16 By assuming that the τ0 value is unchanged
with changing solvent from ethanol to aq MeCN, rate
constants kp are estimated from the kpτ0 values, as listed
in Table 2.17 For some compounds, the reaction was
carried out also in MeOH to determine kp in this solvent.
The fluorescence from Rh o+* was quenched also by
amines 2 and alkoxybenzenes 3 in aq MeCN. The Stern-
Volmer analysis gave the values of kp for the quenching
by these quenchers, as summarized in Table 3.
With these values of the half-wave potentials as well
as the rate constants of the SET process in hand, we
examined energetics of the SET process from 1 to Rh o+*.
Dependence of the rate of the overall SET process on the
free-energy change in the SET step shows that the SET
takes place in an irreversible way when the SET step is
endothermic. The irreversibility in the SET results from
a rapid reaction of trivalent phosphorus radical cation
1
•+, which is generated in the SET step, with water in
the solvent. In contrast, the SET quenching of Rh o+* by
amines 2 and alkoxybenzenes 3 occurring when the SET
step is endothermic was found to be completely revers-
ible.
Electr och em ica l Mea su r em en ts. Peak oxidation
potentials Epox of 1-3 were determined by cyclic voltam-
metry (CV) in aq MeCN at 25 °C using Ag/Ag+ as a
reference electrode.18,19 The Epox value of each compound
was hardly changed with changing scan rate, indicating
that the anodic oxidation is nearly reversible. Obviously,
Resu lts
Stea d y-Sta te P h otolysis of Tr iva len t P h osp h or u s
Com p ou n d s w ith Rh od a m in e 6G. The solution of
equimolar amounts (5.00 × 10-3 M) of 1 and rhodamine
6G (Rh o+) in methanol was irradiated with the light of
>450 nm for 5-7 h under an argon atmosphere at 25
°C. The resulting solution was analyzed on GC or GCMS,
which showed formation of the corresponding pentavalent
oxo-compound 4 (Scheme 1). The yield of 4 and the
amount of the recovered starting material 1 are given in
Table 1. The material balance shows that 1 is converted
exclusively to 4 without side reactions taking place under
the conditions. No reaction took place between 1 and
Rh o+ in the dark. Without Rh o+, no photoinduced
decomposition of 1 was observed within the period
examined.
(16) Lessing, H. E.; von J ena, A.; Reichert, M. Chem. Phys. Lett.
1975, 36, 517-522.
(17) Uncertainty in the τ0 value of Rh o+* results only in movement
of all points in Figure 2 in unison, providing little influence on the
discussion here.
(18) The Epox values of some trivalent phosphorus compounds have
been reported in Yasui, S.; Tsujimoto, M.; Okamura, M.; Ohno, A. Bull.
Chem. Soc. J pn. 1998, 71, 927-932.
ox
(19) There is inconsistency in reported values of Ep determined
by CV for trivalent phosphorus compounds. For example, these values
have been reported for 1b: 0.83 V20a and 0.75 V20b in acetonitrile, and
1.15 V in butyronitrile20c (normalized as values vs Ag/Ag+). A value of
Epox could depend largely on purification methods of solvent. To avoid
ambiguity in potentials, we measured the Epox values of all compounds
used in this work by ourselves using the same solvent as for the Stern-
Volmer analysis.

Quenching of a Photosensitized Dye
J . Org. Chem., Vol. 65, No. 15, 2000 4717
Ta ble 2. F lu or escen ce Qu en ch in g of Rh od a m in e 6G in th e P h otoexcited Sta te (Rh o+*) by Tr iva len t P h osp h or u s
Com p ou n d s Z3P (1)a
Z3P (1)
kpτ0, M-1
10-9kp, M-1 s-1
E1/2, Vb
tri(p-tolyl)phosphine (1a )
triphenylphosphine (1b)
tributylphosphine (1c)
25.1
8.37
0.79
0.91
1.07
23.3 (24.0)c
22.9
7.78 (8.00)c
7.65
methyl diphenylphosphinite (1d )
ethyl diphenylphosphinite (1e)
dimethyl phenylphosphonite (1f)
diethyl phenylphosphonite (1g)
trimethyl phosphite (1h )
triethyl phosphite (1i)
15.4
15.0
5.12
5.01
1.21
1.17
1.49
1.47
3.79 (2.90)c
7.17 (5.61)c
0.229
1.26 (0.967)c
2.39 (1.87)c
0.0762
0.0456
0.159
1.81
0.137
0.477
0.504
1.87 [1.80]d
1.83
triisopropyl phosphite (1j)
tributyl phosphite (1k )
0.168
1.85
triphenyl phosphite (1m )
0.136
0.0453
1.89
a
b
In aqueous acetonitrile under an argon atmosphere at 25 °C. Half-wave potential vs Ag/Ag+. c In parentheses are shown the values
d
obtained in methanol. Determined by rotating disk electrode.
Ta ble 3. F lu or escen ce Qu en ch in g of Rh od a m in e 6G in
th e P h otoexcited Sta te (Rh o+*) by Sever a l Qu en ch er sa
Sch em e 2
10-9kp,
quencher
kpτ0, M-1 M-1 s-1
E1/2, Vb
0.51
0.66
0.82
0.82
0.97
1.11
1.11
N,N-dimethylaniline (2a )
triphenylamine (2b)
N-methylaniline (2c)
tributylamine (2d )
78.0
75.1
71.5
30.9
27.4
14.2
10.1
12.2
6.10
1.66
41.0
27.8
36.2
18.6
0.753
0.207
26.0
25.0
23.8
10.3
9.13
4.72
3.36
4.06
2.03
0.553
13.7
9.25
12.1
6.19
0.251
triethylamine (2e)
dicyclohexylamine (2f)
dipropylamine (2g)
dibutylamine (2h )
diethylamine (2i)
reduction potential Epred of Rh o+ in the ground state was
determined by CV to be -1.13 V vs Ag/Ag+ at 25 °C. Since
the voltammogram showed reversible feature for the
reduction of Rh o+, the half-wave potential of Rh o+ is
given by
1.13
1.17
1.22
0.67
dibenzylamine (2j)
1,2,4-trimethoxybenzene (3a )
1,4-dipentoxybenzene (3b)
1,4-dimethoxybenzene (3c)
1,2-dimethoxybenzene (3d )
1,3-dimethoxybenzene (3e)
4-methylanisole (3f)
0.92 (0.96)c
0.96
E1/2(Rh o+/Rh o•) ) Epred(Rh o+/Rh o•) + 0.03 )
-1.10 V (vs Ag/Ag+) (4)
1.07
1.17
0.0690 1.23 (1.30)c
a
In aqueous acetonitrile under an argon atmosphere at 25 °C.
Half-wave potential vs Ag/Ag+. c Determined by rotating disk
With these values, E1/2(Rh o+*/Rh o•) ) 1.22 V was
obtained.
b
electrode.
one electron is transferred in the oxidation. Under these
Discu ssion
ox
circumstances, Ep is related to half-wave potential E1/2
by eq 2 at 25 °C.21
Mech a n ism of F lu or escen ce Qu en ch in g. Photoly-
sis of trivalent phosphorus compound 1 with R h o+ in
methanol affords the corresponding pentavalent oxo-
compound 4. The observation is evidence for single-
electron transfer (SET) occurring from 1 to Rh o+*.1,2
Thus, as shown in Scheme 2, the SET produces trivalent
phosphorus radical cation 1•+, which is attacked by
methanol nucleophilically to afford phosphoranyl radical
1•-OMe. Since 1•-OMe has oxidation potential lower than
that of Rh o+ in the ground state,23 its oxidation by Rh o+
takes place easily to give the phosphonium cation 1+-
OMe,24 which subsequently collapses to 4. In the quench-
ing experiments performed in aq MeCN, water in the
solvent acts as a nucleophile toward 1•+ to give 1•-OH,
which leads to 4 in a similar way.25
E1/2 ) Epox - 0.03
(2)
The E1/2 values of 1-3 were calculated based on eq 2,
as listed in the last columns of Tables 2 and 3. For some
compounds, the E1/2 values were determined also by
rotating disk electrode (RDE). The values obtained by
both methods are in good agreement with each other.
The half-wave potential E1/2 of Rh o+* was calculated
from eq 3.22
E1/2(Rh o+*/Rh o•) ) E1/2(Rh o+/Rh o•) + ∆E0,0 (3)
The value of ∆E0,0 in eq 3, the zero-zero excitation
energy of Rh o+ to Rh o+*, was estimated to be 2.32 eV
from the absorption maximum of Rh o+ and the wave-
length of the emission from R h o+*. Meanwhile, peak
Singlet-singlet energy transfer from Rh o+* to 1 is
impossible to occur judging from absorption maxima of
Rh o+ and 1. Triplet-triplet energy transfer to quench
Rh o+* is unlikely because the triplet excited state of 1
would be higher in energy than the triplet excited state
(20) (a) Schiavon, G.; Zecchin, S.; Cogoni, G.; Bontempelli, G.
Electroanal. Chem. Interfacial Electrochem. 1973, 48, 425-431. (b)
Ohmori, H.; Nakai, S.; Masui, M. J . Chem. Soc., Perkin Trans. 1 1978,
1333-1335. (c) Culcasi, M.; Berchadsky, Y.; Gronchi, G.; Tordo, P. J .
Org. Chem. 1991, 56, 3537-3542.
(21) See, for example, Oevering, H.; Paddon-Row: M. N.; Heppener,
M.; Oliver, A. M.; Cotsaris, E.; Verhoeven, J . W.; Hush, N. S. J . Am.
Chem. Soc. 1987, 109, 3258-3269.
(22) Ballardini, R.; Varani, G.; Indelli, M. T.; Scandola, F.; Balzani,
V. J . Am. Chem. Soc. 1978, 100, 7219-7223.
(23) Horner, L.; Ro¨ttger, F.; Fuchs, H. Chem. Ber. 1963, 96, 3141-
3147.
(24) Phosphoranyl radicals are oxidized by diaryliodonium and
triarylsulfonium salts to give the corresponding phosphonium cations.
Kampmeier, J . A.; Nalli, T. W. J . Org. Chem. 1993, 58, 943-949.
(25) Yasui, S.; Shioji, K.; Ohno, A.; Yoshihara, M. J . Org. Chem.
1995, 60, 2099-2105.

4718 J . Org. Chem., Vol. 65, No. 15, 2000
Yasui et al.
Sch em e 3
kp of the overall SET process is expressed by
k12k23k30
kp )
(5)
(k32 + k30)(k21 + k23) - k32k23
When the SET step (k23-step) is exothermic, k23 . k21
holds, and then eq 5 reduces to
of Rh o+.26 Besides, the quantum yield of intersystem
crossing from Rh o+* to its triplet excited state has been
reported to be quite low (0.002).27
There is no evidence for formation of a covalent adduct
between 1 and Rh o+. No change in a UV-visible
spectrum of Rh o+ is observed upon the addition of 1
either in the dark or under irradiation. In contrast, when
tris(2,6-dimethoxyphenyl)phosphine (5), a trivalent phos-
kp ) k12
(6)
which predicts a constant value of kp at a diffusion-
controlled limit, because k12 is a rate constant for
encounter of a donor and an acceptor. Meanwhile, when
the SET step is endothermic, k23 , k21 holds. In this case,
eq 5 becomes
kp ) K12k23k30/(k32 + k30)
(7)
with K12 ) k12/k21. Let us assume that the SET step is
irreversible, i.e., k32 , k30. Then, eq 7 becomes
kp ) K12k23
(8)
Given the Horiuchi-Polanyi equation, increase in
activation free-energy of the SET step (∆G23‡) is propor-
tional to increase in free-energy change of this step (∆G23)
(eq 9).
phorus compound that is highly nucleophilic,28 is added
to the solution of Rh o+, the UV-visible spectrum of
Rh o+ changes dramatically. The addition of 5 causes also
significant decrease in the intensity of emission from
Rh o+*; if this decrease results only from the SET
quenching by 5, the quenching rate constant would be
larger than 1012 M-1 s-1, which is highly unlikely for
reactions in solution. In addition, upon irradiation with
Rh o+, 5 does not give the corresponding phosphine oxide,
the product that would be produced if the SET from 5
takes place to give the radical cation 5•+. Hence, 5 forms
a covalent adduct with Rh o+,29 preventing this phospho-
rus compound from undergoing SET, whereas 1a -m do
not form covalent adducts with Rh o+ and undergo SET
to Rh o+*.
∆G23‡ ) ∆G23‡(0) + R∆G23
(9)
Accordingly,
k23 ) kiexp(-R∆G23/RT)
(10)
In eqs 9 and 10, ∆G23‡(0) and ki are the activation free-
energy and the rate constant, respectively, at ∆G23 ) 0,
and R is a proportionality constant that takes a value
from 0 to 1. Combination of eqs 8 and 10 leads to
kp ) K12kiexp(-R∆G23/RT)
(11)
Certainly, the quenching of Rh o+* by amines 2 and
alkoxybenzenes 3 results from SET from these quenchers;
2 and 3 undergo SET to quench the photoexcited state
of many types of substrates.5,22,30 In conclusion, quenchers
examined here, 1, 2, and 3, undergo SET to Rh o+* to
quench the fluorescence from this dye.
Depen den ce of SET Rate on Fr ee-En er gy Ch an ge.
Fluorescence quenching through SET is described by a
sequence shown in Scheme 3.5,13,22 Thus, an encounter
complex is formed between a quencher (Q) and a photo-
excited sensitizer (S*) (k12),31 and then SET takes place
within the complex to give a radical pair (k23), which
disappears in some way (k30). With a steady-state ap-
proximation with respect to concentrations of the en-
counter complex and the radical pair, the rate constant
Free-energy change ∆G23 is related to difference in
standard potentials of a donor (D) and an acceptor (A)
by eq 12 with F as the Faraday constant () 96.5 kJ mol-1
V-1), which in turn is nearly equal to difference in half-
wave potentials.
∆G23 ) F∆E° ) F{E° (D) - E° (A)} ≈
F{E1/2(D) - E1/2(A)} ) F∆E1/2 (12)
Finally, eqs 11 is rewritten as
kp ) K12kiexp(-RF∆E1/2/RT)
(13)
which predicts a linear correlation between log kp and
∆E1/2 with the slope being -RF/(2.3RT) (0 < R < 1).
In Figure 2 is plotted log kp for the SET from 1-3
against E1/2 of 1-3, in place of ∆E1/2, since an acceptor is
the same (Rh o+*) throughout the present study (i.e.,
E1/2(A) in eq 12 is constant). Clearly, the SET from 1
follows eq 6 in the region of E1/2 < 1.3 V; log kp for 1 is
constant at a diffusion-controlled limit. Meanwhile, in the
(26) trans-Stilbene, which has a higher ET value (210 kJ mol-1) than
that of rhodamine 6G (201 kJ mol-1),27 has no influence on photode-
composition of triphenylphosphine (1b) occurring under the irradiation
of light from a high-pressure mercury lamp. Yasui, S.; Itoh, K.; Ohno,
A. Unpublished result.
(27) Korobov, V. E.; Shubin, V. V.; Chibisov, A. K. Chem. Phys. Lett.
1977, 45, 498-501.
(28) Wada, M.; Hayase, S.; Fujiwara, M.; Kawaguchi, T.; Iwasaki,
T.; Uo, A.; Erabi, T. Bull. Chem. Soc. J pn. 1996, 69, 655-664.
(29) The reciprocal of difference in the absorbances of Rh o+ in the
absence and presence of 5 depends linearly on the reciprocal of the
square of a concentration of 5, suggesting formation of a 1:2 adduct
between Rh o+ and 5. From the slope and the intercept of the plot, the
(31) Throughout this paper, we use the term “encounter complex”
to refer to a weakly bound complex that results from diffusional
encounter between a donor and an acceptor, as in a pioneering report
by Rehm and Weller.5 Recently, Kochi and collaborators have given
this term an alternative definition, according to which it designates a
tightly bound complex such as an exciplex.32
stability constant is calculated to be K ) 1.62 × 104 M-2
.
(30) Kayser, R. H.; Young, R. H. Photochem. Photobiol. 1976, 24,
395-401.

Quenching of a Photosensitized Dye
J . Org. Chem., Vol. 65, No. 15, 2000 4719
Sch em e 4
Several SET processes have been claimed to include
irreversible SET steps.7-12,34,35 In these processes, SET
steps are followed by chemical reactions such as cleav-
age7-10,34 or formation12 of a covalent bond as well as H+-
transfer.11 That is, if the “back SET”36 is overcome by a
chemical reaction taking place followed by the SET step,
the SET step is irreversible.13 We have found, in the
present study, that trivalent phosphorus radical cation
F igu r e 2. Plot of log kq against E1/2 of the quencher. O, 9, b,
and 2 represent the reactions with trivalent phosphorus
compounds (1), aromatic amines (2a -c), aliphatic amines (2d -
j), and alkoxybenzenes (3), respectively. The solid broken line
obtained based on eqs 6 and 13, together with a dotted curve
made by eq 15, traces the points for 1. The dashed curve
obtained based on eq 16 traces the points for 2 and 3.
1
•+ generated in the SET step undergoes an ionic reaction
with water in the solvent, which certainly results in
irreversibility of the SET step. The radical cation 1•+
undergoes the ionic reaction with water so rapidly that
“back SET” from Rh o• to 1•+ in the ground state is
practically negligible (Scheme 4). In fact, when MeOH
was used as a neat solvent in place of aq MeCN, no
acceleration of the overall reaction was observed (Table
2),37 showing that k30-step (reaction of 1•+ with MeOH in
this case) is too rapid to be rate-determining.
In analogy with a transfer coefficient in electrolysis as
well as the Brφnsted coefficient, an R-value in eq 13 is
thought to be a measure of the position of transition state
in an SET step.7 Thus, as the transition state moves to
the reactant side, the R-value becomes smaller. On the
other hand, with this movement of the transition state,
the “back SET” becomes less favorable, and as a result,
the SET step becomes less reversible. Consequently, an
R-value can be taken also as an index of “reversibility”
of SET step; the smaller the R-value, the less reversible
the SET step and vice versa. Interestingly, the R-value
observed here (0.2) is comparable to or smaller than
values observed in dissociative SET reactions between
arene radical anions and alkyl halides where the ir-
reversibility in the SET step results from cleavage of a
carbon-halide covalent bond in the intermediary radical
anion RX•-.7,9 This fact suggests that an ionic reaction
of 1•+ with water is as facile as cleavage of a carbon-
halide covalent bond in RX•-.
region of E1/2 > 1.3 V, log kp for 1 shows negative linear
33
dependence on E1/2
,
indicating that the SET in this
region is endothermic. More importantly, the dependence
is accommodated by eq 13, thus showing that the SET
step from 1 to Rh o+* is irreversible in this region. The
slope of the correlation line is given as -RF/(2.3RT) with
R ) 0.2. If one assumes that the SET step is reversible,
i.e., k32 . k30, then eq 7 finally affords
kp ) K12k30exp(-F∆E1/2/RT)
(14)
This equation requires the slope of a linear correlation
between log kp and ∆E1/2 be -F/(2.3RT). This formula of
the slope would be given also by eq 13 if R ) 1 (-17 at
25 °C), but this is not the case for the SET from 1,
confirming that the SET step from 1 to Rh o+* is
irreversible.
The potential at which the change in dependence of
log kp on E1/2 occurs (1.3 V) is a boundary to determine
whether the endothermic SET or the exothermic SET
takes place. In other words, this value of the potential
determines the value of E1/2(Rh o+*/Rh o•) to be 1.3 V.
Taking into consideration experimental uncertainties,
this value of E1/2 is in good agreement with the experi-
mentally determined one () 1.22 V, See the Results
section).
SET fr om Am in es a n d Alk oxyben zen es. In contrast
to the SET from trivalent phosphorus compounds 1, the
SET from 2 and 3 to Rh o+* follows the Rehm-Weller
equation, eq 16, which is well-known to reproduce a
kinetic behavior in SET occurring in many types of
donor-acceptor pairs.5
Around E1/2 ) 1.3 V (i.e., when ∆G23 ≈ 0), the rate
constant experimentally observed kexp is better repro-
duced with a correction according to eq 15.7
1/kexp ) 1/kp + 1/k12
(15)
kp )
k12
(16)
The curve made by eq 15 is shown by a dotted line in
Figure 2, which in fact traces the experimental points
better than the broken line made by eqs 6 and 13 does.
1 + (k21/k30){exp(∆G23‡/RT) + exp(∆G23/RT)}
Thus, an iterative calculation based on eq 16 under a
reasonable assumption5 that k12 ) 2.0 × 1010 M-1 s-1
eventually affords the best fit between theoretical and
experimental values of the rate constants for amines 2
(32) (a) Rathore, R.; Hubig. S. M.; Kochi, J . K. J . Am. Chem. Soc.
1997, 119, 11468-11480. (b) Hubig, S. M.; Kochi, J . K. J . Am. Chem.
Soc. 1999, 121, 1688-1694.
(33) Since eq 1 has been derived by a series of approximations,
sufficiently accurate quenching constants cannot be determined when
kpτ0 < 0.1 M-1. In the present study, kpτ0 values for the phosphites
1h -k ,m are very close to this border, making the points for these
quenchers scattered. See ref 5.
(35) Pelizzetti, E.; Mentasti, E.; Pramauro, E.; Inorg. Chem. 1976,
15, 2898-2900.
(36) In photochemical processes such as fluorescence quenching,
“back SET” takes place in the ground state to regenerate the starting
materials. Thus, the term “back SET” used here does not mean the
reverse reaction in terms of the principle of microscopic reversibility.
(34) (a) Koo, J .-Y.; Schuster, G. B.; J . Am. Chem. Soc. 1977, 99,
6107-6109. (b) Koo, J .-Y.; Schuster, G. B.; J . Am. Chem. Soc. 1978,
100, 4496-4503.

4720 J . Org. Chem., Vol. 65, No. 15, 2000
Yasui et al.
from Rh o+* was monitored on a Shimadzu RF-5000 spectro-
fluorophotometer. GC analysis was done with a Shimadzu GC-
14A gas chromatograph. Cyclic voltammetry (CV) was carried
out on a Cypress Systems OMNI90 potentiostat using a
platinum and Ag/Ag+ as a working and a reference electrode,
respectively, and with tetraethylammonium tetrafluoroborate
(0.10 M) as the supporting electrolyte. Rotating disk electrode
(RDE) was carried out with a BAS RDE-1 with the rotation
rate being 1000 rpm using the same potentiostat and condi-
tions as those for CV.
and alkoxybenzenes 3 with ∆G23‡(0) ) 7.0 kJ mol-1 and
E1/2(Rh o+*/Rh o•) ) 1.12 V (vs Ag/Ag+) (a dashed curve
in Figure 2). Emphasized here is that the value of the
half-wave potential of Rh o+* “theoretically” predicted in
this calculation is again in satisfactory agreement with
the “experimentally” evaluated one (1.22 V). The points
for aromatic amines 2a -c appear above the “theoretical”
curve. Such a discrepancy in SET rates between aromatic
and aliphatic amines is not unusual; a similar kinetic
behavior of amines has been observed in SET quenching
of photoexcited states of transition metal complexes.22
Equation 16 has been derived from eq 5 by assuming a
highly exothermic “back SET” contributing predomi-
nantly to k30-step.5 So, the observation that the SET from
2 and 3 to Rh o+* follows eq 16 indicates that cation
radicals generated from 2 and 3 in the SET step undergo
practically no reaction with a nucleophile such as water
in the solvent. Neither is significant the reactivity of
these radical cations as a free-radical in the present
system, possibly because there is no powerful “free-
radical trapping agent” available here.
Ma ter ia ls. Trivalent phosphorus compounds 1 and 5,
amines 2, and alkoxybenzenes 3 as well as rhodamine 6G
(Rh o+) were commercially available (Tokyo Chemical Industry
Co. or Aldrich). These materials were purified through recrys-
tallization, column chromatography, or distillation, if neces-
sary. Acetonitrile (Nacalai Tesque) was refluxed over calcium
hydride for 5-6 h and distilled. The distillate (referred to as
aq MeCN throughout the text) was used for the Stern-Volmer
analysis and electrochemical measurements without further
attempts for dehydration. The concentration of water in this
solvent was estimated to be 5.6 × 10-2 M by the Karl Fischer
method.38
P h ot or ea ct ion of 1 w it h R h o+. A solution of 1 in
methanol (5.00 × 10-3 M) was irradiated with a light from an
Ushio xenon arc short lamp UXL-500D-0 through a Toshiba
sharp-cut filter Y-47 (irradiation at hν >450 nm) in the
presence of Rh o+ under an argon atmosphere for 5-7 h. The
resulting solution was analyzed on GC to determine the
recovered amount of 1 and the yield of 4.
Con clu sion
In this study, we have shown that SET from trivalent
phosphorus compounds Z3P to a photoexcited dye takes
place in an irreversible way even though the SET step
is endothermic. The irreversibility results from high
reactivity of radical cation Z3P•+ as an electrophile toward
a nucleophile. The finding here provides a further ques-
tion; what endows Z3P•+ with such high reactivity as an
electrophile? It is our next target to find an answer to
this question.
Ster n -Volm er An a lysis. In a quartz cell equipped with
a septum and filled with an argon gas, was placed quencher
1, 2, or 3 in an appropriate amount. The solution of Rh o+ in
aq MeCN (5.00 × 10-6 M) was added to the cell through a
syringe to resolve the quencher. The solution was excited at
525 nm, and the intensity of the resulting emission was
recorded at 545 nm on a spectrofluorophotometer.
Exp er im en ta l Section
In str u m en ts. UV/visible spectra were recorded on a Shi-
madzu UV-2200A recording spectrophotometer. Fluorescence
Ack n ow led gm en t. This work was supported in part
by a Grant-in-Aid for Scientific Research (C) (No.
09640654) from the Ministry of Education, Science,
Sports, and Culture, Japan. One of the authors acknowl-
edges the financial support of a Tezukayama Research
Grant in 1999.
(37) Instead, the values of kp for the reactions of 1f and 1g obtained
in MeOH were slightly smaller than those obtained in aq MeCN. This
is attributable to small changes of the redox potentials of 1 and Rh o+
as well as ∆E0, of Rh o+, which could result from the change of the
0
solvent.
(38) Yasui, S.; Shioji, K.; Ohno, A.; Yoshihara, M. Chem. Lett. 1993,
1393-1396.
J O000448I