вход по аккаунту


Charge Transfer and Radical Ions in Photochemistry.

код для вставкиСкачать
harge Transfer and Radical Ions in Photochemistry**
By Jochen Mattay’
Bimolecular photoreactions usually occur via several steps involving “complex-type’’ intermediates (excimers or exciplexes). Although physical chemists have investigated them for
nearly thirty years (Fiirster, Weller), and postulated their existence in numerous publications, it is only in the last decade that evidence for an exciplex intermediate in cycloadditions (Culdwell) has been obtained, thereby establishing the link between spectroscopy and
synthetic photochemistry. Since then, many investigations have confirmed the role of exciplexes as intermediates in bimolecular photoreactions. The photochemically induced
charge transfer from a donor to an acceptor substrate determines not only the bond
strengths in the primary exciplex but also influences-via the charge-transfer nature of the
complex- the structure of the exciplex and thereby the selectivity of subsequent reactions.
Such chemical processes between the donor and the acceptor prevent the reverse transfer of
electrons, which usually results in regeneration of the starting materials, thereby only causing the dissipation of energy and occasionally-under specific conditions-the generation
of molecules of the starting material in the triplet state. Ionic photodissociation also reduces the probability of a reverse transfer of electrons. This process, which produces solvated radical ions, is promoted by polar solvents and salt effects, particularly special salt
effects. By this means it is possible to generate specific radical cations from, for example,
alkenes and dienes and to study their reactions. Although in some areas, for example physical and biophysical chemistry, discussions of single electron transfer (SET) processes have
long been found in the basic textbooks, organic chemistry lags far behind, although-and
this is shown by the numerous recent publications-organic chemists are also aware of the
fundamental importance of these processes. In this connection photochemistry plays a central role, since its instruments make it possible to carry out specific SET processes, to analyze them, and to use them in chemical reactions.
1. Introduction
“One of the simplest elementary processes in solution is
the exchange of a n electron between two chemical substrates.” This statement by Ebersod” is of fundamental importance for many areas of chemistry.[’-61 In this elementary process, bonds are neither broken nor formed. However, the transfer of an electron has important consequences for the chemical reactivity of the compounds.“]
For example, a nucleophilic reagent R-Nu is transformed
by oxidation (i.e., the loss of a n electron) into a n electrophile. Correspondingly, the reduction of an electrophilic
substrate enhances its nucleophilic character. Electron
transfer therefore leads to “umpolung” (Fig. 1) with all its
consequences for the ensuing reactions.
+ ee
- NuoQ
The importance of electron transfer in photochemistry
has been known ever since the fundamental work of Weller
et al.‘’] and, for example, the studies of
but it was not until the 1970s that numerous
groups began to approach the problem experimentally.
These investigations paralleled the development of highly
time-resolved methods of transient analysis, such as laser
flash spectroscopy, which made possible the study of
chemical processes in the nsI’o-’31and P S [ ’ ~ - ’ ~ I region, and
chemically induced dynamic nuclear polarization
(CIDNP),”’] which can also be applied in a time-dependent manner.”’. ‘’I Occasionally, time-resolved ESR spectro~copy[’~.~’]
and, most recently, time-resolved Raman
spectroscopyf2’]have been used. These modern analytical
methods can be used to obtain information about the types
of reactive intermediates and their kinetics. Simultaneously, interest has been growing in organic syntheses that proceed via electron-transfer p r o c e ~ s e s . ~ ~ ~ - ~ ~ ~
2. Basic Principles
R -E
2.1. Energetics
Fig. I. Urnpolung by means of electron transfer.
Prof. Dr. J. Mattay
lnstitut fur Organische Chemie der Technischen Hochschule
Prof.-Pirlet-Strasse I , D-5 100 Aachen (FRG)
Radical Ions and Photochemical Charge Transfer Phenomena, Part 17. Part 16: J. Mattay, J. Gersdorf, K. Buchkremer, Chem. Ber. 120 (1987)
Angen,. Chem. Int. Ed. Engl 26 (1987) 825-845
Why is there so much activity in the field of photochemistry at the moment? Because specific photochemically initiated electron-transfer processes can be used to carry out
chemical reactions involving radical ions in homogeneous
solution, and these reactions, in turn, can be studied by the
modern analytical techniques mentioned above. Photo-
0 VCH Verlngsgesellschnji mbH. 0-6940 Weinherm. 1987
0570-OX33/X7/0909-OX25$ 02.50/0
excitation of electron-acceptor (A) or electron-donor (D)
substrates leads to well-defined changes in their redox
properties (Fig. 2). For example, the donor properties of D
tronically excited species, and AE,,,, the coulomb interaction energy of the two radical ions. at a distance a from
one another in a given s o l ~ e n t . [ *The
~ - ~ exact
~ ~ value of
AEcou,is obtained, according to the Born equation, from
Equation (d).IZ9' With F=96489 C, e = 1.602 x
+ +
~ ~ = 8 . 8 5 410-"
C V - ' m-I, and Ns6.023 x loz3 mol-I,
Fig. 2. Photochemical excitation and redox properties of donor (D) and acceptor (A) molecules.
increase proportionally to the excitation energy
(AE(D*)=hv); i.e., the ionization energy, IP(D), is reduced by the promotion of an electron from the HOMO to
the LUMO [Eq. (a)]. The electron affinity of the acceptor,
EA(A), behaves similarly [Eq. (b)].".2'1 Electron transfer
EA (A*) = EA (A) +AE(A*)
between donor and acceptor substrates should therefore
occur more easily after photochemical excitation of one of
the reacting species"1 if either IP(D*)< EA(A) or
IP( D) < EA (A*) holds true (Fig. 3).
one obtains the free energy of formation (AG) of the solvent-separated radical-ion pair A?'D,""(SSIP) in a solvent with dielectric constant F . In general, the distance a
for a solvent-separated radical-ion pair is taken as 0.7 nm.
The Weller equation therefore allows an estimate, to a first
approximation, as to whether electron transfer within a donor-acceptor system is thermodynamically allowed (AG
negative, exergonic process) or not (AG positive, endergonic process). In addition, it shows that the electrontransfer process can be decisively influenced by the polarity of the solvent as well as by the electronic properties of
the reacting species. This gives the chemist a number of
ways to direct reactions involving radical ions in a desired
2.2. Radical-Ion Pairs, Exciplexes
While the fate of a donor-acceptor pair in the exergonic
region of electron transfer (AG<O, Eq. (c)) appears at first
glance to be straightforward-namely, the preferred formation of radical-ion pairs-this is not true for the endergonic region. In addition to energy transfer?] an exciplex
(i.e., a complex in an excited state) may be formed, especially when the energy of the excited species (sensitizer) is
greater than the excitation energy of the second species
(for the possibility of energy transfer via exciplex intermediates, see, for example, Kauarnos and Turro["]).Figure 4
Fig. 3. Photochemically induced electron transfer between donor and acceptor molecules.
Simple energy considerations such as this clarify the dependence of the electron transfer on electronic conditions
in the participating substrates. I P and EA are valid only for
the gas phase. However, since the oxidation and reduction
potentials (E7;2(D) and EE:(A)), which are the corresponding parameters in solution, are easily obtained experimentally'261and are linearly related to IP and EA, respecti~ely,'~'~
the conditions for electron transfer between
D and A after irradiation can be formulated as follows:
In Equation (c)-the
Weller equati~n-EY;~(D) and
E;;$(A) are the oxidation potential of the donor and the
reduction potential of the acceptor, respectively, measured
in acetonitrile; AE,,,,, is the excitation energy of the elec826
Fig. 4. General kinetic scheme for photochem~callyinduced electron transfer
in donor-acceptor systems (according to [30]). ENC complex = collision
complex; EXC = exciplex; CIP = contact ion pair; SSIP = solvent-separated radical-ion pair; FRI = free radical ions.
shows various quenching processes that an excited acceptor molecule A* can undergo with a donor molecule D
in the ground state (spin multiplicities as well as subsequent processes, e.g., the formation of triplet and product
molecules, are not considered here for reasons of simplicity).I3O1The path characterized by k3 shows the direct formation of the SSIP from the collision complex in the exergonic region. However, the electron transfer is also possible via an exciplex intermediate (k2,k4, k5). This possibility
Angew. Chem. Int. Ed. Engl. 26 (1957) 825-545
is supported by the observation that, in contrast to the expectations based on the Marcus theory,l8] reductions in the
rate constants for electron transfers in highly exergonic regions have not been observed.l''I Recently, however, Miller, CIoss et al.I3'] found clear evidence for the so-called
"Marcus inverted region" in intramolecular electron-transfer processes.
According to Weller et a1.)27.32.331
exciplexes form because of a coupling of a local excited state with the socalled charge-transfer (CT) state [(e) and (f)].The weight(A*D) t--)(AD*) H(AOQD0@1
studies have shown that indirect evidence can also be
found. For example, reversibly formed exciplexes have
been recognized by the biexponential decay of their monomer luminescence.1481Similarly, the deviation of fluorescence or phosphorescence quenching from Rehm-WeIler
behavior [correlation between logk, and AG: k , = rate
constant for the bimolecular quenching process;
AG = reaction free energy of the electron transfer, given by
Equation (c)] indicates the existence of additional intermediates [ Further examples are found in numerous
reviews.[6. 23,ZS. 32.34-37.431
w = cI w(A*D)+c, w(AD*)+ c3w ( A @ ~ 1
ing of the wave function v ( A G Q D G Odetermines
the CT
character of the exciplex. For c3> c , and c2,the wave function corresponds to a highly polar exciplex which can easily dissociate in polar solvents to form solvated radical
ions (ionic p h o t o d i s s ~ c i a t i o n [ ~ ~In
~ ) .the extreme case
c3 c , and c2, the exciplex can be considered as a contact
ion ~ a i r . 1 ~ ~ 1
Exciplexes are often discussed as intermediates in photochemical r e a c t i o n ~ ; [ ~ ' . ~this
~ - ~is~ also
the case for several photochemical reactions of arenes and carbonyl compounds which have been investigated in our laboratory.13X-451
If such an exciplex exhibits a bathochromic shift
in emission compared with the monomer luminescencef27-461
(Fig. 5), then the experimental detection of a
Fig. 5 . Schematic potential curves for the formation of exciplexes (from 1271).
AHe = enthalpy of exciplex formation: hvk/hvM = emission energy of exciplex/monomer; E,, = repulsion energy in the ground state; v, = vibrational
quantum number. 1/11 = fluorescence of the exciplex/monomer, 111 = absorption of the monomer.
reactive intermediate in a photochemical reaction is relatively simple. For example, Caidwell et ~ 1 . were
~ ~ ' able
quench the 9-phenanthrenecarbonitrile-anetholeexciplex
specifically with dimethyl acetylenedicarboxylate (quenching of hvE,but not hv,; cf. Fig. 5) and thereby demonstrate
the reactive character of this intermediate. Similar examples have been described by other authors.[
Although the observation of exciplex emission is not the
rule,'22Jthe existence of exciplexes and their role as reactive intermediates is nowadays unchallenged. Numerous
Angew. Chem. Inr. Ed. Engl. 26 0 9 8 7 ) 825-845
2.3. Aim and Scope of This Article
In our studies of the photochemistry of donor-acceptor
systems, we have been able to show that reaction selectivity is strongly influenced by photochemically induced
Reactions involving radical ions
and those involving exciplex intermediates follow different
paths. However, the reactions of the latter species are also
determined by the degree of charge transfer (i.e., the reactivity of highly polar exciplexes is different from that of
weakly polar species). The reaction free energy for the
electron transfer (AG) calculated from the Weller equation
has proved to be an excellent ordering ~ r i t e r i o n . ' ~ ' - ~
I n~ . ~ ' ~
the following discussion, we will present some exampIes
chosen from our own work which demonstrate this correlation between reaction selectivity and AG values. Examples from other areas will make the general validity of this
AG correlation clear. With the help of transient analysis
methods, the intermediates involved can be detected and
their kinetics investigated. It will be shown in detail how,
for example, the equilibrium between contact ion pairs
(CIP) and solvent-separated ion pairs (SSIP) o r free radical ions (FRI) can be influenced by the suitable choice of
the reaction medium and how the course of the reaction
can thereby be directed. Borderline cases such as the involvement of ternary complexes in photoreactions will be
discussed. This article is restricted to the photochemistry
of electron donor-acceptor systems. The generation of radical ions by photoionization["] will not be dealt with here.
Readers who are interested in electron transfer processes
in microheterogeneous systems, as well as in the relationship between energy transfer and electron transfer, are
referred to other reviews.~5.6.22.s2-s41
Furthermore, reactions of radical ions in homogeneous solution have been
reviewed by Chanon et a1.,(221 Farid et
3. The Concept
3.1. Photochemically Induced Charge Transfer,
Intermediates, and the Course of Reaction
Table 1 shows, in a simplified manner, possible intermediates and the course of reaction for photoreactions between electron-donor and electron-acceptor molecules.
This scheme was developed for the reactions between arenes and a l k e n e ~ , ~ but
~ ' . ~has,
~ ~ as we shall see, proved
valid for other systems. The decisive criterion for the
course of the reaction is the type of intermediate and
hence the degree of charge transfer, which can be correlated with the reaction free energy for the formation of
SSIPs. A positive or negative AG value [calculated by
means of Eq. (c)] should show, to a first approximation,
whether the formation of SSIPs is allowed or not, i.e.,
Table 1. Photoreactions of electron donor-acceptor systems
( A O O D O ~)\
Reactions of
radical ions (0
(A)D D'
whether free, or at least solvent-separated, radical ions
(pairs) are preferentially formed, assuming that the polarity of the reaction medium suffices for the solvation of the
individual species.
The situation in the endergonic region of electron transfer is more complicated. A calculation of the free energy of
formation of the exciplex is only possible when molecular
parameters such as dipole moment, the radius, and the
separation between the components are known.[*'] Although these values are not usually known, Equation (c)
offers a way out of the dilemma. Figure 6 shows curves for
3.2. Influencing Charge Transfer by Solvent Polarity and
Salt Effects (Transient Analysis Studies)
The reaction medium plays a decisive part in processes
that occur via poIar intermediate^."^.'^^ According to
Equations (c) and (d), for example, the formation of solvent-separated ion pairs in acetonitrile (&= 37.5) is favored
by 0.98 eV (=94.4 kJ mol-') over that in n-hexane
( E = 1.89). Ionic photodissociation of donor-acceptor
pairs[341to solvated radical ions therefore requires a polar
solvent. The influence of the solvent can often be followed
by transient analysis because of the specific UV/VIS absorption of the intermediates involved.
Fluorenone 1 is particularly well suited as an acceptor
for such studies since the concentration, solvent, temperature, and cation-radius dependence of the absorption spectra of pairs formed from metal cations and fluorenone
radical anions are already known from earlier
The effect of solvent and of added salts on the photoreactions of fluorenone with the electron-rich alkenes 2 and 3
is shown in Figures 7 and 8,
Whereas the
formation of solvated radical ions is favored in polar sol-
dimer (CIP)
A* +D
Fig. 6. Dependence of exciplex and SSIP potentials on the dielectric constant
of the solvent (from [SS]).
the potential of an exciplex and an SSIP as a function of
the polarity of the solvent, as expressed by the dielectric
constant E. Since the curves cross in the region of E S
the experimentally easily obtainable AG(SS1P) value can
be used as a n approximate estimate of the exciplex free
energy. Equation (c) therefore makes it possible to estimate the CT character of an exciplex [cf. Eqs. (e) and (f)].
This method is particularly useful when the correlations
with AG values for electron transfer are carried out within
a substrate series, since the parameters that determine the
exciplex energy depend on the type of molecule.
Fig. 7. Transient absorption spectra (AOD = absorption difference) 100 ps
after flash photolytic excitation of 1 in the presence of tetramethyl-1,3-dioxole 2 in various solvents (from [60]).a) Acetonitrile. b) Benzene. c) Cyclohexane.
vents, only contact ion pairs or paramagnetic dimers could
be observed in nonpolar media. It should be noted, in particular, that electron transfer from electron-rich alkenes to
1 even takes place in nonpolar solvents through the formation of contact ion pairs.
Following the fundamental work of Hughes, Ingold et
a1.[621and Winstein et al.,@'I salt effects have been frequently investigated in organic chemistry.I5'I The so-called
normal salt effect is largely determined by the ionic
strength; the departure from the "normal" behavior of the
reaction rate at low salt concentrations has been called the
special salt effect by Winstein et al.r631This effect is presumably due to an ion exchange process that causes a shift
in the equilibrium between the SSIP and the CIP. Whereas
Anyen, Cliem In,. Ed. Engl. 26 (1987) 825-845
benzophenone-alkene and benzil-alkene systems, we
could extend the lifetimes by a factor of 15 (as determined
by laser flash spectroscopy) by the addition of lithium
p e r c h l ~ r a t e . ~In
~ ~addition,
the special salt effect occasionally results in a greater yield of radical ions.[a.641In
the investigations of the systems so far mentioned, only the
formation and decay of the long-wavelength-absorbing radical anions could be followed spectroscopically. Simultaneous analysis of both radical-ion species is possible using
p-chloranil 4 as the acceptor and 2-methoxy-1,l-diphenylethene 5 as the d o n ~ r , ~because
~ ~ . ~ 'of~ their favorable absorption characteristics (Fig. 9). Both the radical anion
4 0 ° and the radical cation 5 O @ decay with second-order
kinetics and have almost identical lifetimes in the microsecond region, as expected for a reverse transfer of electrons. The addition of lithium perchlorate increases the
lifetime by a factor of 10 without affecting the coupling of
the two radical ions (Table 2).
1 +Ei>=(OEi
Table 2. Effect of LiC104 o n the lifetimes r,,2 of 40° and So@
Fig. 8. Ocpendencr of the tranbient spectra on added lithrum perchlorate
100 ps after flash photolytic excitation of 1 in the presence of 3 in acetonirile-special salt effect (from 1601). 0, no LiCIO,; 0, 0.1 M LiCIO,.
without LiCIO,
the long-wavelength absorption at ca. 550 nm in Figure 8
can be ascribed to the SSIP formed by l o Qand 3@@, the
addition of lithium perchlorate brings about the formation
of a new ion pair (consisting of l o Q
and Li@).The hypsochromic shift of the absorption shows that this species is
a contact ion pair [Eq. (g)]. The special salt effect can
1 M LiCIO,
Further transient analysis methods such as ESR spectroscopy using in situ photolysis ~ o n d i t i o n s [ ~can
* . ~yield
information about the structure of the radical-ion intermediates involved but provide little information about their
dynamics. It should be noted that salt effects in photochemically induced electron-transfer processes have been
investigated by other groups. For example, Goodman and
were able to show, with the help of picosecond
laser flash spectroscopy, that both the electron back transfer and the dissociation of contact ion pairs formed between trans-stilbene and electron-poor alkenes could be
accelerated by the addition of salts. However, these effects,
therefore be used to influence the equilibrium between the
SSIP and the CIP.
The separation between AOe and Do@ makes for a
longer lifetime for the radical ions, since the reverse transfer of electrons clearly takes longer. For example, using the
6 00
Anyew Chem. l n l . Ed. Engl 26 11987) 825-845
Fig. 9. Transient absorption spectra of 4 and 5
in acetonitrile SO ns (+,
T-T = absorption)
800ns after the start of
the laser flash (from [44,
611). 0 , no LiCIO,; 0 ,
ascribed to ionic strength, affect the dissociation more
strongly, so that ion-pair separation predominates. Goodson and Scb~ster'"~found, moreover, that exciplexes in
nonpolar solvents could also dissociate into free radical
ions upon addition of ammonium salts. This effect was ascribed to the formation of larger salt aggregates. The same
authors showed, in later work, that the lifetimes of 1,4naphthalenedicarbonitrile radical anions could be dramatically increased by the addition of magnesium salts owing
to the formation of complexes.[681Masnoui and Kocbi[691
investigated, in detail, the dynamic behavior of ion pairs
formed by the C T excitation of electron donor-acceptor
complexes formed between anthracene and tetranitromethane, and observed both normal and special salt effects in
the presence of tetra-n-butylammonium trinitromethanide
and tetra-n-butylammonium perchlorate.
In conclusion, it has been established that solvents and
salts strongly influence the ionic dissociation of photochemically excited electron donor-acceptor pairs. Electron
acceptors (for example, mixtures of carbonyl compounds
or arenecarbonitriles and metal salts) are therefore suitable
catalysts for chemical reactions that proceed via radical
cations [Eq. (h)].150.6'1
4.1. Photoreactions of Benzene and Monosubstituted
Benzenes with AIkenes
4.1. I . The a,a,a-TrifluorotolueneExample
Discussion of the mechanism of the cycloaddition of alkenes 7 to photochemically excited benzene or its derivatives 6 has been carried on ever since the discovery of ortho c y ~ l o a d d i t i o n ~ (to
' ~ - 9~ )~and
~ meta c y ~ l o a d d i t i o n [ ~ ~ . ~ ~ ~
(to 8).1761
On the basis of the results of various groups, we
have been able to show, during the last few years, that the
exciplex mechanism most fully explains the current experimental results, especially since it accounts for cycloaddition as well as substitution (to 11 and 12) (Fig. 11).[42.431
4. Photochemical Reactions between Donor and
Acceptor Compounds
Photoreactions between donor and acceptor molecules
proceed mostly via exciplex intermediates (see Section
2.2). A scheme by which numerous reactions can be classified is shown in Figure 10 (cf. Table 1). According to this
Fig. 10. Reaction between donor and acceptor molecules via exciplex and
contact-ion-pair intermediates (alternatively,excitation of the donor is possible)-cf. Table 1.
scheme, the amount of charge transfer determines the type
of intermediate and thereby the way in which the reaction
proceeds. This concept will be exemplified by the photochemical reaction between arenes and alkenes. Three types
of intermediates can be distinguished: polar and nonpolar
exciplexes react to give cycloadducts, although with different degrees of selectivity; exciplexes having complete
charge transfer (contact ion pairs)[291usually show a different reaction behavior-if suitably substituted, they undergo heterolytic bond fission to form radical pairs, which
then couple with one another ("substitution"), or they decay via the formation of triplet states. Dissociation into
solvated radical ions and their synthetic use will be dealt
with in Sections 5.1-5.3.
Fig. 11. Photoreactions of arenes with alkenes
The photoreaction of a,a,a-trifluorotoluene 13 with alkenes results in the formation of all these product types in
good yields (50-72%) (Fig. 12).[4'1It should be noted that
para cycloaddition (to 10) is seldom observed and that 10
in some cases is formed from the orfho adducts 9 via secondary photochemical processes.[431
9, R:CF3
Fig. 12. Photoredctions of 13 w i t h dlkenes. Note that in the system 13/17 the
alkene 17 acts as an acceptor (Table 3).
Angew. G e m . Int. Ed. Engl. 26 (1987) 825-845
Table 3 . Preferred reaction type in the photoreaction of 13 with the olefins 2
and 14-17 as a funcrion of the oxidation potential of the olefin and AG
IEq. (c)l Ial-
- 2.64
[a] E;::(13)=
AG [eV] [b]
+ 1.72
- 3 18 V, E’;;>(13)=2.91 V, AEe,,,,=4.6SeV. Ib] In dioxane
After electronic excitation, 13 preferentially forms the
substitution products 18 with electron-rich alkenes such as
2 and 14. With electron-poor reactants such as 16 and 17,
on the other hand, cycloaddition to give 8, R = C F 3 , is
mainly observed (Table 3). This trend is reflected in the
oxidation potentials of the alkenes. The free energies of
electron transfer calculated from Equation (c) show clearly
that only substitution will be observed at AG=O in the exergonic region and only cycloaddition in the endergonic
region. The reaction of 13 with 15 occupies an intermediate position with respect to both selectivity and charge
transfer. In line with its small positive AG value (in dioxane, ~ = 2 . 2 ) , the exciplex is very polar (for indirect evidence see Ref. [41]), and, in addition to forming the substitution product 18, forms mainly the ortho and mefa cycloadducts 8 and 9 (R=CF3), respectively. According to
Equations (c) and (d), the electron transfer for this system
should be exergonic in a solvent of ~ > 3 Indeed,
the cycloaddition is completely suppressed when the photoreaction is carried out in solvents more polar than dioxane
(Fig. 13). Only the formation of the substitution products
The mechanism that best describes the relationship between photochemically induced charge transfer and the
course of the reaction is shown in Figure 14. The key to the
-0.6 -AG[eV]
Fig. 14. Mechanism of the reactions of 13 w i t h alkenes
reaction is the intermediate. For AG > 0, cycloaddition occurs via an exciplex 19. Only when A G < O is a CIP, 20,
formed; 20 either dissociates in polar media or undergoes
heterolytic bond breaking to form a radical pair 21, which
in turn, finally recombines to form substitution products.
0 re1
4.1.2. Generalization of the AG Correlation
Figure IS shows the correlation between mode selectivity (substitution, ortho and meta cycloaddition) and the AG
values for the electron transfer, as exemplified by the photoreaction of benzene 6, R = H, with alkenes 7.[421
Fig. 13. Reaction selectivity (as measured by product quantum yield Ore,)as a
function of the dielectric constant ( E ) in the photoreaction between 13 and
15. x =overall yield; O=yield of 18; .=yield of 8 and 9 (R=CF,)
in the photoreactions of 6 , R = H, with alkenes 7.
18 is observed. This “special” solvent effect is overlaid by
the “normal” effect; i.e., the product formation is suppressed in favor of ionic photodissociation in polar media-an effect that is observed with all reactions in the exergonic region of electron transfer. The Weller equation,
with the inclusion of the coulomb term [Eq. (d)], can therefore be used to control reaction selectivity.
AG 5 1.5 eV) preferentially form ortho cycloadducts, while
meta cycloaddition is mainly observed between reactants
with AG? 1.5 eV). Similar correlations can be established
for other a r e n e ~ . [ ~In
* l the case of benzonitrile, for example, even cycloaddition to the nitrile group can be included. One must however always be aware of the empiri-
Angew Chem. Inr. Ed. Engl. 26 (1987) 825-845
Fig. 15. Mode selectivity as a function of the free energy of electron transl’er
83 1
cal character of these correlations; i.e., their validity may
be confined to only a series of related corn pound^.[^^-^^^
The classification of photocycloadditions of alkenes to
arenes by means of the type of intermediate corresponding
to the AG value also allows some connections to be drawn
concerning stereo- and regioselectivity. Equation (c) can
be used, for example, to determine which reagent acts as
donor and which as acceptor. In many systems vinylene
carbonate is an acceptor since the following relation is valid :[41-43.77]
AG( 17 = A, arene = D) <AG( 17 = D, arene = A)
and by Arnold et al.[xylin the photoreactions of arene carbonitriles with alkenes as well as by Ohashi et a1.@I1
and later Albini et a1.[8z1in the corresponding reactions
The electronic interactions between the components of
such intermediates influence the stereochemical course of
the cycloaddition in a dramatic
(Fig. 16). Further aspects of the stereo- and regioselectivity have been
discussed comprehensively e l s e ~ h e r e . [ ~ ~ ~ ~ ~ ]
22: exo
Fig. 17. Photoreaction of 24 with 25 in acetonitrile
Fig. 16. Direction of electron transfer and selectivity in the photocycloadditions of 1,3-dioxole derivatives 14, 15, and 17 to arenes.
4.2. Photoreactions of Multiply Substituted Benzenes and
Higher Arenes
The photoreactions of higher, especially acceptor-substituted, arenes with donor substrates have been investigated
often. Usually, systems have been dealt with in which electron transfer is thermodynamically allowed according to
Equation (c) and which accordingly yield preferentially
substitution products. Some typical examples will be presented in the following discussion.
4.2.1. Reactions of Excited Acceptor Arenes
As we have already demonstrated in the case of simply
substituted benzene derivatives, the substitution products
are not formed by direct coupling of the radical-ion species of the contact ion pairs. In most cases the reaction occurs via a number of steps involving the heterolytic splitting off of an electrofugal or nucleofugal group and the
final recombination of the radicals thus formed (exceptions include the reactions of 34 and 35 and of anthracene
and N.N-dimethylaniline, see below). Very frequently, deprotonation of the radical cation by the radical anion is
observed [Eq. (i)]. The first evidence for such a reaction
was found in the 1970s by McCullough and co-work832
with alkylarenes as electron donors. Photoexcitation of
terephthalodinitrile 24 in the presence of 2,3-dimethyl-2butene 25 results in the formation of the dimerization
products of 25 (28-30),'801in addition to the substitution
products 26 and 27. The free energies calculated according to Equation (c) for radical-pair formation are
-0.96 eV and thus support a mechanism involving electron and subsequent proton transfer to give 31 and 32, respectively, as the key steps (Fig. 17). Compound 26 is presumably formed via 33.
The corresponding reactions of 1,4-naphthalenedicarbonitrile1801and 9-phenanthrene~arbonitriIe[~~~
give similar
results. Mizuno, Otsuji et al. were able to transform these
variable-yield photosubstitution reactions into efficient allylation methods by the introduction of the trimethylsilyl
group (which is easily split off as a cation) into the olefinic
(Fig. IS).
AIbini et al. have studied in detail the photoreactions of
arenecarbonitriles with alkylbenzenes.'".x61 Also in these
cases, the first step is photochemically induced electron
transfer with the formation of a radical-ion pair, which
[Arylene (CN),
[Arylene (CN)?'
[Arylene (CN):"
' ' ' ~R,C=CRC~iMe3Oo]
~ ~
3 I
Arylene (CN)CH,CR=CR,
Fig. 18. Photochemical allylation or arened~carbon~rr~les
with allylsilanes.
Angew. Chem. Int. Ed. Engl. 26 (1987) 82.11845
then generally undergoes deprotonation and subsequent
recombination of the radicals. With peralkylated toluene
or dibenzyl derivatives, for which no proton transfer is
possible, they even observed heterolytic fission of a C-C
bond.['"' The high stereoselectivity of the addition of cumene 35 to 1,4-naphthalenedicarbonitrile34 to give the
products 36 and 37 (Fig. 19) indicates that the reactions
Fig. 20. Cycloaddition and substitution In the photoreaction of 38 with 39.
Fig. 19. Photochemical addition of 35 to 34
occur within the solvent cage and consequently by means
of radical-ion pairs or'radical pairs. Only dissociated radicals undergo the typical reactions of free and non-correlated intermediates. The AG values calculated by the
Weller equation indicate once again that a n exergonic
electron transfer takes place.[871
Because of their low oxidation potentials, amines are
often used as donor components in photochemical electron-transfer reactions (see also Section 4.3) and also in
reactions with excited acceptor arenes. For example, electronically excited anthracene reacts with N,N-dimethylaniline, by direct coupling of the radical ions, to give 9-(4'dimethylaminophenyl)-9,lO-dihydroanthracene in 70%
yield along with small amounts of dimerization and H addition products.[881 In contrast, trans-stilbene reacts with
triethylamine via a multistep series of electron and proton
transfers to give disproportionation and recombination
products (cage reactions) as well as symmetrical dimerization products (escape reactions).[891
Kochi. Rentzepis et al. have carried out comprehensive
studies on the role of ion pairs in the photoreactions of C T
c o m p l e x e ~ . [Irradiation
~ ~ . ~ ~ ~ in the long-wavelength absorption bands of CT complexes of 9-substituted anthracenes
42 and tetranitromethane 43 results, in accordance with
the Mulliken
in a fast electron transfer ( t < 25 ps)
with the formation of a radical-ion pair. Within a few ps
the tetranitromethylide fragments. Ionic and radical coupling follow in two slower steps to form 44 and the final
product 45,respectively (Fig. 21). The corresponding steps
apply to the nitration of methoxybenzenes['61 and alkene~.['~'
4.2.2. Reactions of Excited Donor Arenes
In 1974 McCullough et al.[yO1
reported on the photoreactions of naphthalene 38 with acrylonitrile 39, which gave,
together with the cycloadduct 40,the substitution products
41 (Fig. 20). In polar solvents, such as acetonitrile, the
electron transfer should, according to Equation (c), be
thermodynamically allowed (for the redox potentials and
excitation energies, see Refs. 142, 901). Here again one can
assume the formation of an ion pair followed by a proton
transfer to the acrylonitrile radical anion (either from the
solvent or, with other donor arenes such as indene, from
the corresponding radical cation). This assumption is supported by the catalytic effect of protic solvents and isotopic labeling experiments.["] For weaker donors, such as
anthracene and phenanthrene, the electron transfer is endergonic and hence no substitution products are
Angew. Chem. Inr. Ed. Engl. 26 (1987) 825-845
Fig. 21. Ionic coupling within a contact ion pair and radical coupling via free
radicals in the photoreactions of 42 and 43.
With tetracyanoethylene (TCNE), the anthracenes 42
form no
Transient analyses in the ps region,
however, clearly show the formation of ion pairs. A proton
transfer from 42'@
to T C N E
or the formation of DielsAlder dimers apparently cannot compete with the rapid
back transfer of electrons. On the other hand, the system
38/N-methyldihydrotriazoledione is reactive: Sheridan et
a1.["' were able to show, however, that here the long-wavelength-absorbing alkene component was excited and
formed, from the triplet o r singlet state, a contact ion pair
with 38 as donor. This gave a Diels-Alder adduct in 40%
yield. The same photochemical Diels-Alder reaction was
observed when the CT complex was directly excited.
Sundberg has a reported, in a comprehensive revie~,'~''
on chloroacetamide cyclization in aromatic systems in
which the initiating step is a photochemically induced
electron transfer and which, in other variations, leads to
lactam formation.
4.3. Photoreactions of Carbonyl Compounds with Alkenes,
Alkylbenzenes, and Amines
The photoreactions of carbonyl compounds have been
studied in detail since the early days of photochemistry.f99.’001In particular, the. [2 21 cycloaddition with alkenes, which leads to oxetanes 48 (Paterno-Biichi reaction), has assumed great synthetic importance[t00-1@21
although, depending on the substituents, a number of byproducts (46,47, 49, 50) can be formed (Fig. 22).
R=H : 18%
R=CH3: 45%
/ \
R=CH3 :
Fig. 22. Products of photoredcttons of ketones wlth alkenes
In general, the exciplex-diradical mechanism[1001
is used
to explain the reaction behavior of simple alkenes. However, owing to the ready accessibility of very electron-rich alkenes such as 2, 14, and 15, electron-transfer processes
have recently been shown to occur as we11.[40.44~4s.491
Whereas with “normal” alkenes a large number of products are often observed[’001(Fig. 22), the photochemically
induced electron transfer can act as a “regulator” for reactions with electron-rich alkenes. Upon photoexcitation,
biacetyl 54 reacts with ketene acetal 53 to form exclusively
the oxetane 51.[45.491Trapping experiments with the radical
cation 530° indicate an electron transfer as the primary
photochemical step. The components of the contact ion
pair, which show high spin densities at their radical centers, recombine through the radical positions to form first a
zwitterion 59 and then 51 (the most important resonance
structures of the radical ions taking part are given in Fig.
23). Since 53 and 54 also undergo an aldol-type cycloaddition (via dipolar intermediates) to give the other regioisomeric oxetane 52,’49. the reaction here involves “umpolung” of the reactivity of a carbonyl group by means of
photochemically induced electron transfer.’491
The extremely electron-rich alkenes 2 and 55 show another type of reaction behavior. Because of the allylic hydrogen, 2B0 and 55O@, respectively, are deprotonated by
54” with the formation of a radical pair, which then recombines to give the reaciion products 57 and 58 [see also
Eq. (i)].[40.4s1
The dependence of the ratio (57+ 58):56 on
: 25%
Fig. 23. Photoreactions of 54 with electron-rich alkenes.
the oxidation potential of the alkene and transient analysis
investigations by means of ESR spectroscopy support this
extended mechanism for the photoreactions between carbony1 compounds and a l k e n e ~ . [ ~ ~ , ~ ~ I
In an article summarizing recent results, Wagner et
have described the influence of charge transfer on
the photoreactions of phenyl ketones with alkyl benzenes
as donors. The ketones react via both the nn* and the nn*
states by means of a CT mechanism in which the extent of
the charge and electron transfer is determined by the electrochemical redox potentials of the species involved. Although nn* and nn* states are the same in this respect,
they show different reaction behaviors because of the different structures of the exciplexes involved. In addition, H
abstraction (the initiating step in the formation of a photoreduction product) can change from a one-step (H transfer) to a two-step (e’ and H @transfer) mechanism (cf. the
Angew. Chsm. In,. Ed. E n g l 26 (1987) 825-845
4.4. Photoreactions of Phthalimides
photoreactions of 54 with 2 and 55, Fig. 23, and Refs. 140,
Besides ketones, other carbonyl compounds such as carAmines have also been used as powerful electron donors
acids and their derivatives are of photochemical
in photoreactions with ketones. An early e ~ a m p l e " ~ ~
. ~''~~
I t - ' 14] We will restrict ourselves here to the phoshown in Equation 0'). Seebach et al.L'"71
were able to use
toreactions of N-methylphthalimides 66 with alkenes 7 ;
the selectivity of these reactions is similarly determined by
charge transfer (from the alkene to the imide) and the
[ P h Z 8 0 H HGCH,]
products are of synthetic interest.
this method for a n asymmetric photopinacol reaction. The
involvement of contact ion pairs was examined in detail by
Peters et al.['Oxl
The preparative applications of this reaction have been discussed el~ewhere.''~,
Quinones are powerful oxidizing agents, particularly
after photoexcitation, and their reactions have been frequently described.["',
Here, the photoreactions of 1,4naphthoquinones 60 with I,]-diarylethenes, such as 61,
should be briefly mentioned, since a direct dependence of
the course of the reaction on the AG values of the electron
transfer similar to that found in the case of a,a,a-trifluorotoluene (Section 4.1.1) has been observed (Fig. 24)." I'
(ca. 3 ~ 0 % )
+ MeOH
i &I
\' '
m,,, I
+ W
68 ,
(ca. 30%)
(ca. 10-70%)
Fig. 25. Addition and reduction in the photoreactions of 66 with alkenes
(ca. 30-60%)
Fig. 24. Charge transfer and the course of reaction in photoreactions of 60
with ],I-diphenylethene 61.
Electronically excited naphthoquinones 60 form two
types of products with 1,l-diarylethenes of type 61. The
cyclobutanes 62 are formed only when electron transfer is
endergonic according to Equation (c), probably via exciplex intermediates. For thermodynamically allowed electron transfer, one observes the formation of the ethene adducts 63 exclusively. The presumed intermediacy of contact ion pairs and diradicals or zwitterions is supported by
CIDNP studies. A subsequent photoreaction, via 64,
makes physiologically interesting benz[a]anthraquinones
such as 65 available by means of this route.
Angenz. Chem. Inr. Ed. Engl. 26 11987) 825-845
In their fundamental work, Muzzocchi et al.["51investigated the photoreactions of 66 with various alkenes 7.
Two types of reaction were observed (Fig. 25); the regioand stereospecific addition of the alkenes to a carbonyl
group of 66 with ring expansion to the dihydrobenzazepinediones 67['16'(I) and the photoreduction to 68 and 69,
which, in the presence of methanol, leads to 70"'s1(11).
The AG values, calculated according to Equation (c),
clearly demonstrate the regulatory effect of electron transfer: in the endergonic region only the ( 0 2 +TI') addition (I)
was observed and in the exergonic region only the photoreduction (11). Trapping experiments with methanol as a
nucleophile confirmed the existence of radical-ion intermediates. Only recently have Muzzocchi et al. been able to
show, with intramolecular systems, that, here too, a direct
coupling of the radical ions of the contact ion pair is possible through their radical centers." 'I The synthetic potential of such intramolecular reactions, which produce annelated dihydrobenzazepinedioneso r macrocyclic spiro com835
pounds, has recently been demonstrated by Kubo et a1.[”81
and Machida et al.[’‘’I Other synthetic applications have
been described in reviews by Kanaoka,[‘‘*I Mazzocchi,[’1 3 1
and Coyle.[’
4.5. Photoreactions of Alkylideneammonium Salts
Alkylideneammonium salts such as 71 are powerful
electron acceptors even in their ground states. This property should be enhanced in the excited state. Indeed, photochemical electron-transfer processes have even been observed with simple alkenes;[”O1 they lead to products such
as 72-78.
In Figure 26, the probable mechanism is shown; the initiating step is the transfer of an electron from a donor molecule to an alkylideneammonium salt 71 with the formation of an a-aminoradical-radical-cationpair. This recombines either directly through the radical positions or after
trapping of the radical cation by nucleophiles (e.g., methanol). The values of AG for the electron transfer, calculated
’ko (Z)
from Equation (c), can once again be used as a criterion
for the course of the reaction.l’*’. ‘”l
This method gives access to a large variety of N-heterocycles.[24. The synthetic potential of the already widely
used alkylideneammonium salts is thereby enormously increased (see, for example, Refs. [122-1241).
As has already been shown in Section 4.2.1, suitably
substituted radical cations can easily split off electrofugal
groups with the formation of a radical; the simplest case is
deprotonation [see Eq. (i)]. Muriano et al. were able to
transform this procedure into an efficient method for the
allylation of alkylideneammonium salts by means of allylsilanes.1’2s1An intramolecular variant of this method to
form structural units of the isoquinoline alkaloids is shown
in Figure 27.[1261
OMe ( 2 7 % )
Fig. 27. “Diradical cyclization method” for the synthesis of isoquinoline alkaloid building blocks.
5. Reactions of the Radical Cations of Alkenes and
+ MeOH
Fig. 26. Photoreactions of alkylideneammonium
‘ /!?\
In polar solvents the components of a contact ion pair
can be “separated” by solvation to form either a solventseparated radical-ion pair or free radical ions. This process
may be enhanced by the salt effect; in some cases the lifetime and the yield can be greatly increased by ion-exchange reactions [special salt effect: see also Section 3.2
and Eq. (g)]. This hindering of the back electron transfer
often allows, for the first time, the directed production of
radical ions and thereby makes possible the study of their
reactions without the involvement of the radical-ion partner, i.e., without chemical reaction [Eq. (h), Fig. 281.
Angew. Chem. Int. Ed. Engl. 26 (1987) 825-845
Fig. 28. Intermediates and the course of reaction in electron-transfer processes between donor and acceptor molecules. For abbreviations, see caption
to Figure 4.
I1 hv
The following discussion will be restricted to the reactions of radical cations, in particular those of alkenes and
5.1. Dimerization and Cycloaddition
Fig. 30. Chain mechanism (I) and electron-transfer sensitization (11) in radical-cation-catalyzed reactions.
5. I . I . Reactions of Alkenes
Among the first examples of photochemically induced
reactions of alkene radical cations was that of the dimerization of N-vinylcarbazole 79, first discovered by Ellinger11271
and later thoroughly investigated by Ledwith (Fig.
The decisive steps are the oxidation of the al+e*
79 0”
c , R= Aryl
I131 1321
10-30% [133,131]
Fig. 29. Dimerization of alkenes via radical cations.
kene to a radical cation 79a Oe by electron acceptors such
as p-chloranil 4 or halogenated aromatic ketones (primary
electron transfer) and the reduction of an acyclic radicalionic dimer 80a or a cyclobutane radical ion 81a
(secondary electron transfer). Since the quantum yield in these
reactions is greater than 100%(maximum value of 0:66), a
chain reaction is involved ( I in Fig. 30) in which the secondary transfer takes place from monomer 79a to the
dimer 80a or 81a0@.The same is true for the enol ether
79b and the aryl derivative 79c, although here, because of
the lower quantum yieId, the chain length must be
A detailed analysis of the kinetics and thermodynamics of
the elementary processes can b e found in Evans et
Other dimerizations have been described by Farid et
Arnold et aI.,“351Pac et a1.,[136-’37’
and other^.^'^.^^^
The regioselectivity of these reactions is generally greater
Angew Chem Inr Ed Enql 26 (f987) 825-845
than that of the corresponding triplet reactions involving
diradical intermediates, which mostly yield mixtures of reg i o i ~ o m e r s . ~Differences
in the stereoselectivity have also
been observed: for example, 1,3-dioxole 15 dimerizes in
an energy-transfer reaction under triplet conditions to give
a mixture of endo- and exo-bis(methy1enedioxy)cyclobutanes in the ratio 0.6:
whereas, in a reaction via 15”@,
the stereoisomer ratio becomes endu :exo= 1.5 :l.‘s1.5b1
Another mechanism has also been discussed which one can
call “electron-transfer sensitization”‘”’’ (1 1 in Fig. 30).
Here, the secondary electron transfer takes place between
the product radical cation and the radical anion of the acceptor. Since the acceptor is thereby regenerated, it can
also be viewed as the sensitizer (Sens in Fig. 30). This
mechanism plays a role in many electron-transfer reactions, often in modified form such as the so-called “triplex
mechaniSm.”13V, 137. 1391 Quantum yields below 100% are a
necessary but not sufficient criterion for mechanism I f ,
since, because of the complex kinetics, a single inefficient
elementary process can drastically reduce the quantum
yield in mechanism I. Comparison experiments should
therefore b e carried out with thermal electron acceptors,
e.g., ammoniumyl salts R3No@XQ,’129.
1401 for which a chain
process is clearly involved. A further criterion in the case
Fig. 31. Redox photosensitization according to Por et
for “hole transfer,” e.g., phenanthrene).
[I361 (M
of mechanism I would be chain initiation by means of an
independently generated intermediate (for example, by
In some cases the efficiency of the dimerization of alkenes by means of radical cations can be increased by the
use of a mediator (“hole transfer”). This mechanism has
been investigated by several Japanese groups”36.14‘-1431
and, according to Pac et al./1361can be classified as “redox
photosensitization” (Fig. 3 I)-cf. also Ref. 11371.
In a fundamental study, Farid et al.[1441
demonstrated the
different reactivities of radical-ion pairs and free radical
ions in the dimerization of 1,l-diphenylethene 61 to give
the cyclobutane 84 and the “(4+2) dimer” 87 (Fig. 32).
specific oxidation of only one of the reaction species (Fig.
33), which can be easily accomplished by means of Equa-
+am/ k?
Fig. 33. Possible reaction paths in radical-cation-catalyzed reactions between
various donor compounds when only one reactant is oxidized, i.e., when
Equations (k) and (I) are satisfied.
tion (c). In the most favorable case, the conditions of
Equation (k) should apply; i.e., only the oxidation of D
<o <AG(AO~D,O@)
should be thermodynamically allowed. With a sufficiently
high concentration of the reaction species D’, it is often
possible to suppress the undesired dimerization to DD. In
the less favorable case, i.e., when both reacting species can
undergo oxidation (only Equation (I) is valid), the dimeri-
E%(D) < .GL(D’)
Ph Ph
- 82
Ph Ph
P h H
Ph ph
5.1.2. Reactions of Dienes (Diels-Alder Reactions)
Ph Ph
Ph Ph
Fig. 32. Electron-transfer-sensitized dimerization of 61 via radical-ion pairs
or free radicals to give 84 or [4+2] dimers 87,respectively.
The initially formed contact ion pair [61 O 0 820e] reacts
further, by the formation of a C-C bond with a neutral
alkene 61 and subsequent back electron transfer, to give
the 1,4-diradical 83, which cyclizes to give 84. This course
of reaction largely parallels the triplex mechanism.f39.137.1391 J f , on the other hand, the radical-ion pairs
are separated by solvation, the reaction takes a totally different course. In this case, reaction of 61 gives an acyclic
radical-cation adduct, 85, which, because of its longer life,
cyclizes to give first the radical-ionic [4+2] dimer 86 and
then 87 by means of back electron transfer. Concentrationdependence experiments (higher concentrations of 61 favor the cage reaction giving 84) and quenching experiments with radical-cation traps support this mechanism.
After the first radical-cation-catalyzed dimerizations of
alkenes, there soon followed examples of mixed cycloadditions.[130. 148. 1461 A d ecisive criterion for selectivity is the
zation to D D can at least be minimized by the optimal
choice of concentration^^'^^^ (see also Section 5.1.2). Since
the reactions occur via several steps, the course of the reaction can be influenced by other effects, e.g., stabilization of
acyclic intermediates by appropriate substituents, the effects of counterion, o r even other mechanisms (cf. Section
P h R P h
One of the most important reactions of dienes is the
Diels-Alder reaction.”471Its scope of application, however,
is limited by one of the fundamental rules of thermal
(4x 2n) cycloadditions: namely, a sufficiently high reaction rate is only achieved if the frontier orbitals of the
starting materials (HOMO and LUMO) are energetically
Such ideal conditions are not always fulfilled,
and there has therefore been no lack of attempts to find
improvements for Diels-Alder reactions that yield poor results. Examples include reactions under high-pressure conditionS,[180-1821 catalysis by Lewis acids,[153-1881
and the exploitation of the hydrophobic effect for reactions in aqueous media.”56.
A further approach makes use of the following principle: oxidation of one of the reacting species results in an
increase in the frontier-orbital interaction, which should in
turn lead to an acceleration of the Diels-Alder reaction. As
Freeman et al.L1a7b1
and Hammond et al.[16”l have shown,
the radiation-induced dimerization of cyclohexadiene, reported by Schenck et al.)16’a1can be explained in this way.
A similar effect was also observed by Mizuno et a1.[188Jin
the Diels-Alder reaction of furan with indene in the presence of naphthalenecarbonitriles as photochemical electron acceptors; this reaction probably occurs by means of
mechanism IJ (Fig. 30). Later, Bauld et a1.[1591
applied this
principle to thermal electron acceptors and coined the
term “cation-radical catalyzed Diels-Alder reaction”; this
Anqew. Chem. Int. Ed. Engl. 26 (1987) 825-845
reaction is based on a chain mechanism (Fig. 30, I). ACcording to theoretical considerations, the radical-cationcatalyzed Diels-Alder reaction should obey the "principle
of role selectivity," which states that the cycloaddition
ockurs preferably by means of the radical cation of the dienophile.""O1 Since then, many groups have concerned
themselves with this problem. Although no comprehensive
picture has yet emerged, some of the mechanistic aspects
will be dealt with here since they are directly related to the
subject of this review (cf. Section 3).
The dimerization of simple dienes to Diels-Alder adducts via photochemically generated radical cations does
not usually present any great difficulties as long as the
diene is oxidizable, which, in turn, can be estimated from
Equation (c). One exception is 2-acetoxy-1,3-cyclohexadiene (see Fig. 39 and accompanying text). Various acceptors have been used: arene~arbonitriles,"~~.
k etones
and quinones in the presence of lithium perchlorate (specia1 salt e f f e ~ t ) , [ ~ '162-1641
. ~ ' . and pyrylium ~ a l t s . ~162-1651
The dimerization of 1,3-cycIohexadiene 88 has been investigated under various conditions (Fig. 34, Table 4). Under thermal conditions, in spite of long reaction times, the
[4+2] dimers 89 and 90 are formed only in poor
yield.i166.16'] The dimerization is only better under highpressure conditions (up to 7 kbar); however, the cyclobutane adducts 91 and 92 as well as some acyclic dimers,
formed by nonconcerted processes, are also produced.[1681
The dimerization by means of radical-cation catalysis, on
the other hand, occurs very efficiently to give, preferentially, the Diels-Alder adducts 89 and 90.16'.65.139.
Here, it is unimportant whether one employs thermal acceptors via a chain mechanism (Table 4, No. 8) or whether
one uses photochemical acceptors (Nos. 5-7). Only the stereoselectivity ( e n d d e x o ratio) changes, which indicates
that different mechanisms are operating. It should be
noted that the dimerization under energy-transfer conditions yields mainly the cyclobutane adducts 91 and 92
(No. 2, 3).[139.166.1671
The question of the validity of the "role selectivity" of
radical-cation-catalyzed Diels-Alder reactions cannot be
answered by product studies of the dimerization of dienes.
For example, the mixed cycloaddition between 88 and 1,3dioxole 15 provided clear evidence that the Diels-Alder
reaction can take place via either the radical cation of the
dienophile (here 15) or the radical cation of the diene
(here 88).[1621
Whereas the reaction via 150° forms the
endo isomer 97 preferentially, both reaction paths are almost equally important in the corresponding cycloaddition
via 880° (Fig. 35).
98 H H
Fig. 35. Reaction paths in the radical-cation-catalyzed Diels-Alder reaction
less preferred).
between 88 and 15 ( 3 preferred,
This shows at least that the cycloaddition via the diene
radical cation cannot be concerted. The results obtained
for the reactions between furan 99 and indene 1001'5"1631
or I,4-dioxene 103"631
confirm a multistep mechanism in
which the cycloadduct 101 as well as the substitution
products 102 and 104 are formed via acyclic radical-ion
intermediates (Fig. 36).
Other authors have also found clear evidence for a multistep mechanism: Gross et al.['691in the case of the dimerization of 1,3-butadiene to 4-vinylcyclohexene in the gas
phase, which he investigated by means of the collision-activated decomposition (CAD) method in a tandem mass
spectrometer; and recently Roth et al.L1791
in the case of the
Fig. 34 Dimerization of 88 under thermal conditions and energy- and electron-transfer conditions. 93-96: electron-acceptor molecules. See also
Table 4.
Table 4. Reaction conditions and product ratios in the dimerization of 88.
No. Reaction
hv, direct
hv, triplet
hv, 93, benzene
hv, 93
hv, 94, CH,CN
hv, R2C0 [a], LiCIO,, CH,CN
A, 95, CHZC12
Yield [%]
1166, 1671
[166, 1671
main products
minor products
[61, 162-1651
[61, 162-1651
[61, 159, 162-1641
[a] RZCO= 1 , 4 , 96.
Chem. fnr. Ed. Engl. 26 (1987) 82s-845
101 (35%)
102 (20%)
pairs: both TMB and methanol efficiently quench the
type-I I reaction but affect the type-I reaction involving
contact ion pairs (or exciplexes according to Weller, cf.
Section 2.2) far less. We have observed similar effects in
the dimerization of l-acetoxy-1,3-cyclohexadiene.
The mechanism described here for the dimerization of
dienes via photochemically induced charge transfer resembles the one proposed by Furid et a1.1'441for the dimerization of 1,l-diphenylethene 61."441Whether so-called triplex intermediates are involved in this case, as suggested
b y Schusrer et al.,'f391cannot be decided at present. We also
observed the involvement of ternary complexes, in a different context,[391either as reactive intermediates or simply as
transition states; such complexes therefore appear to have
general significance in p h o t ~ c h e m i s t r y .' ~3 7~. 139.
~ . i441
The stereoselectivity in radical-cation-catalyzed DielsAlder reactions can also be influenced in other ways, e.g.,
by the choice of the acceptor (Fig. 38, Table 5). The radi-
Fig. 36 Muillstep reaction between 99 and 100 or 103; possible radicalcation intermediates involved in the formation of 102 or 104, respectively.
dimerization of spiro[2.4]heptadiene in solution, which he
studied by means of C I D N P techniques.
We have observed different selectivities for contact ion
pairs, on the one hand, and solvent-separated ion pairs o r
free radical ions, on the other hand, in the dimerization of
A high concentration of the diene
and the absence of salts favor the formation of both the
cyclobutanes 91 and 92 and the ex0 Diels-Alder adduct
90 even in polar solvents (20% conversion, I in Fig. 37).
Addition of lithium perchlorate or a low concentration of
the diene, however, result in preferential formation of the
endo Diels-Alder adduct 89 (I1 in Fig. 37).
(Aoo 88@@88),
A + 89+(90)
Fig. 37. Dimerization of 88 via contact ion pairs (exciplexes) (I) or via solvated radical ions (11). A: 4 and 93 (LiCIO,), 94; main product 89.
These effects are even more pronounced in the dimerization of 1-acetoxy- 1,3-~yclohexadieneto the corresponding
[2 + 21 and [4 + 21 adducts. Quenching experiments using
1,2,4-trimethoxybenzene (TMB, E';;, =0.82 V versus Ag/
AgN03 in CH3CN1'"31)and methanol confirm our assumption of the involvement of differently solvated radical-ion
o C)
Fig. 38. Diels-Alder reaction between 88 and 103.
Table 5. Influence of acceptor on the product ratio in the reaction of 88 with
hv, S I
105 : 106
2 : l
10 : 1
Yield [Oh]
29% ( 409" 89
29% ( + 50% 89
+ 90)
+ 90)
cal-cation chain mechanism in the presence of 95 shows
only a small endo selectivity to form 105 ; the selectivity is
increased by the use of the photochemical electron acceptor 94 (under singlet conditions). Here, also, an electron-transfer process is probably involved, as is shown by
1,2,4-trimethoxyben~ e n e , ~I7l1
' ~which,
because of its low oxidation potential
(see above) can trap radical-cation intermediates. On the
other hand, a smaller endo selectivity is observed under triplet conditions with chloranil 4 as sensitizer. Whether in
this case a Diels-Alder reaction via energy transfer is operating o r whether contact ion pairs are involved, as we have
observed in the dimerization of 1,3-cyclohexadiene (see
Fig. 37 and accompanying text), is not yet known.
Although the radical-cation-catalyzed Diels-Alder reaction can take place via various reaction mechanisms,
whose relationships are not yet fully understood, radicalcation catalysis offers a new possibility for influencing selectivity and, in addition, often makes it possible to carry
out, for the first time, reactions that are impossible under
thermal conditions. For example, no Diels-Alder addition
occurs between 88 and 103 (Fig. 38) even under high pressure."72'
Also noteworthy is the so-called triplex mechanism,[39. 137. 1391 which could, in the future, offer a further
Angew. Chem. h i . Ed. Engl. 26 (1987) 825-845
method of directing selectivity: here, a highly polar exciplex, which does not dissociate in nonpolar solvents into
radical ions, is trapped by a second donor molecule with
the formation of the product [possibly via a ternary complex as reactive intermediate, see Eq. (m)]. Such exciplexes
stabilizes acyclic radical-ion intermediates more efficiently
than one in position 2 (112). Consequently, the cycloaddition of 111 and 103 to form the Diels-Alder adducts 114
and 115 (yield: 45%) now competes efficiently with the dimerization of 111. The carboxonium ion structure of the
intermediate 113 apparently favors this mixed reaction.
are characterized by relatively strong bonding energies (up
to 80 kJ mol-I), a marked C T character (cf. Section 3.1),
and a high dipole moment.
Th ey are often structured, as, for example, our studies in the field of arene
photochemistry have shown, and have a decisive influence
on the course of the reaction.'431 By this means it should be
possible, even for the [2+2] and [4+2] cycloadditions, to
make use of the different diastereomeric interactions in the
exciplex formation to direct the reaction. Recently Lemaire
et al.L1771
reported very successful applications-using
charge-transfer complexes, however.
Further cycloadditions of 1,3-cyclohexadienes have
been investigated by Bauld et al.['731and Steckhan et al."651
as well as in our own laboratory16'.'63.164.'741
with special
emphasis on the correlation between substitution and
mechanism, on the one hand, and efficiency and selectivity, on the other. I-Acetoxy-1,3-cyclohexadiene107 dimerizes, with high efficiency, to give the Diels-Alder adducts
109 and 110 ( A = 9 4 , 47%; A = 4 , 63%). In contrast, the
corresponding reaction of the isomer 111 proceeds very
unselectively and in low yield, affording products similar
to those obtained in the triplet-sensitized reaction (energytransfer conditions) (Fig. 39). Assuming a multistep mechanism, a substituent at position 1 of the allylic unit (108)
5.2. Reactions with Nucleophiles
Alkene radical cations can be captured in nucleophilic
solvents as neutral products. This process has been extensively investigated by Arnold et al. and extended to the
counterpart, i.e., the reactions of the corresponding radical
anions with e l e ~ t r o p h i l e s "l7*]~ ~ .(Fig. 40). For example,
1 16 (70%)
117 (64%)
Fig. 40. Reactions of photochemically generdted radical cdtions and aniona
with alcohols.
109 :endo
110 :ex0
114 :endo
115 . e x o
Fig. 39. Dimerizations and cycloadditions of 107 and 111. A: 1 (LiC104), 4
93 (LICIO,), 94: main products, 109 and 114.
Int. Ed. Engl.
26 (1987)825-845
whereas the radical cation generated from 61 by electrontransfer sensitization reacts with an alcohol to give the
anti-Markovnikov product 116, the corresponding radical
anion shows normal behavior, giving 117.
Other solvents can also react as nucleophiles with alkene
radical cations: e.g., the acyclic radical-ion intermediate in
the dimerization of enol ethers can be captured by acetonitrile with the formation of pyridine derivatives.'i321
Nitromethane acts, in addition, as an oxygen-transfer
agent.1179' Finally, Arnold et al. applied this method to
other alkene radical-cation capture reactions of synthetic
interest, such as the photosensitized cyanidation of alkenes.l'*'] Other examples can be found in the reviews by
Farid et al.[231
and M ~ r i a n o . [ ~ ~ ~
Attempts to capture radical cations with nucleophiles
are not always successful. We have found differences in
the reactivity of 1,3-dioxole radical cations on the one
hand with OHe and HPOiQ and on the other with alcoh o l ~ . ' ~"I1' . Whereas the charged nucleophiles reacted, in
part, with the radical cations to form neutral products, this
did not occur with alcohols. Although Arnold et al. were
successful in similar experiments[*'. 1781 and thus clearly
conflicting results have been obtained, there is an interesting parallel between our results and the theoretical considerations of pros^.^"^^ According to the latter, a direct attack
84 I
hv. [ A ]
Fig. 41. Radical-cationcatalyzed isornerizations
and rearrangements of alkenes
(0= quantum
yield). a) cis/fmns isomerization. b) Valence
isornerization. c) Rearrangement.
'R = CH,-CH-CH,
of a nucleophile on a radical cation is "forbidden" and
therefore has a high activation energy.
5.3. Isomerization and Rearrangement
Two examples of the rearrangement of cyclopropene
derivatives (124) via radical cations, leading to the formation of 125 and 126, are shown in Figure
The corresponding energy-transfer reactions, on the other hand,
lead to intramolecular I2 21 cycloadducts (triplet) or indene derivatives via the intermediate formation of carbenes (singlet). Roth et al.['931and Gassman et al.['941have
reported on the rearrangements of bicyclopropenyl- and
bicyclo( 1.1 .O]butane derivatives.
Since the double-bond character of the alkene has been
removed in the radical cation, isomerization and rearrangements are observed, as in the case of the corresponding photoreactions involving energy transfer. Recently, the
cidtrans isomerization of alkenes has been intensively
studied. In Figure 41a, two examples are shown, both of
5.4. Back Electron Transfer with the Formation of Triplets
which occur by means of a chain
solvent polarity and the addition of salts increase the effiAccording to Weller,[301
back electron transfer can occur
ciency of this isomerization and thus confirm that the
from various intermediates. For example, a contact ion
mechanism involves solvated radical ions."s41 Here again,
pair in the singlet state should give the starting material
the analogous reaction via alkene radical anions is known,
predominantly in the singlet (ground) state unless intersysbut it occurs via an electron-transfer sensitization (cf. I1 in
tem crossing by means of a hyperfine mechanism can comFig. 30)."s51
pete with the dissociation of the radical-ion pair. Whereas
Valence isomerizations by means of radical cations have
this process takes place rapidly (ns region), the back elecbeen observed with Dewar benzene,""] in the intramolecutron transfer by recombination of the radical ions takes
and in the
lar cycloaddition of "cage corn pound^,"^'^^.
place slowly (ps region) and gives, corresponding to the
norbornadiene (123)/quadricyclane (120) ~ y s t e m . [ ' * ~ - ' ~ ' ~ number of spin sublevels, 75% of the starting material in
The latter, in particular, is of fundamental importance in
the triplet state. Triplet formation is favored by back elecconnection with the storage of solar energy and has theretron transfer when the energy of the radical-ion pair lies
fore been intensively investigated. The valence-isomeric rahigher than the triplet energy of one of the starting materidical cations 121 and 122 can be distinguished by CIDNP
als. Farid et al.f'951have reported such an example in the
methods (Fig. 41b).f'89.1901The energetically unfavorable
electron-transfer-sensitized reactions of methyl 1,2-dipheisomerization 123 120 can only be achieved by means of
nylcyclopropene-3-carboxylate.In nonpolar solvents, it
strong electron acceptors. In such a case, the contact ion
reacts with 9,lO-anthracenedicarbonitrile to form the empair dissociates with the formation of 123 in a triplet state,
Diels-Alder adduct via a luminescent exciplex, whereas in
which then isomerizes to 120''901[Eq. (n), cf. Section 5.4).
acetonitrile the cyclopropene dimerizes to give an anfi-tricycl0[3.1 .0.02.4]hexane derivative by means of a back-electron-transfer triplet mechanism.
Angew. Chem. Inr. Ed. Engl. 26 (1987) 825-845
We have observed similar effects in the photoreactions
of electron-poor arenes with 1,3-dioxoles, which often, in
the exergonic region of electron transfer, also formed dioxole dimer~.["~l
Here, solvent polarity plays a decisive role
since it strongly affects the energy of the ion pairs [Eqs. (c)
and (d)] in a different way than it does the triplet energy of
the reactants. The studies of BauU et al.r1961
on the dimerization of 1,3-cyclohexadiene 88 and 2,4-dimethyl- 1,3-pentadiene in different solvents can be interpreted on this basis. In dichloromethane ( E = 8.9) the proportion of cyclobutane, which is formed via the triplet reaction path according to Bauld, is dramatically increased in comparison to
the amount formed in acetonitrile ( E = 37.5). Sometimes,
however, the specific interaction between the radical ions
and the solvent can also influence the competition between
the radical-ion and the triplet reaction
sensitizers that are also strong electron acceptors are used,
on the other hand, the involvement of differently solvated
radical ions (CIP or SSIP) may also lead to the product
ratios observed here (cf. Section 5.1.1 and 5.1.2).
6. Conclusions and Prospects
The electrochemical redox potentials and electronic excitation energies of reactants as well as the polarity of the
reaction medium determine the amount of charge transfer
between electron-donor and electron-acceptor compounds.
The free energies of formation calculated according to the
Weller equation [Eq. (c)] for solvent-separated radical-ion
pairs have proved to be suitable as criteria for the estimation of photochemically induced charge transfer. In the exergonic region of electron transfer, solvated radical ions
are formed preferentially after the excitation of donor-acceptor systems (Fig. 42, left). Back electron transfer, al-
radical ions
Fig. 42. "t'dge" and "escape" processes in photochemical electron-transfer
reactions between donor and acceptor molecules.
Angew. Chem. I n ( . Ed. Engl. 26 11987) 825-845
ready reduced in this way, can be further suppressed by
means of salt effects (particularly the special salt effect).
This gives rise to new possibilities in the generation of radical ions and the study of their reactions. Impressive
proof of this is provided by examples from the currently
intensively investigated field of radical-cation-catalyzed
cycloadditions of olefins and dienes. Interactions with the
acceptor radical anion-either via differently solvated radical ion pairs or by means of the so-called triplex mechanism-the influence of counterions, and the participation
of triplet reaction paths can alter the course of the reaction
markedly, particularly the stereoselectivity. In this, and in
the fact that many reactions are only possible under the
conditions of radical-cation catalysis, lies the synthetic potential of these new methods.
A further possible method of hindering the back transfer
of electrons is offered by the reactions of donor and acceptor radical ions among themselves. In the most common cases, these reactions are initiated by the heterolytic
splitting off of a nucleofugal or electrofugal group (Fig. 42,
middle). The recombination of the radicals thus formed
terminates the process. If the recombination is prevented
by the choice of suitable substrates, then a new method for
the generation of free radicals by means of photochemically induced electron transfer is
(Fig. 42,
I n the endergonic region, the free energies of electron
transfer can be used to estimate the CT character and to
determine the direction of the charge transfer within the
exciplex. The AG values calculated by means of Equation
(c) correlate, at least within a substrate series, with the
course of the reaction (Figs. 10 and 15). In other words,
photochemically induced charge transfer has a regulatory
effect on the reactions between electron-donor and electron-acceptor molecules. Here also, it is possible to influence charge transfer and stereoselectivity by a suitable
choice of reaction medium.
Many photoreactions in which charge transfer has a decisive influence on the course of the reaction could not be
dealt with in this article. These include the diverse photoreactions of metal coordination compounds,f2-6.22,
the photoreactions involved in the extensive studies of the
storage of solar energycs1(this area has recently been reviewed by Fendter[2021),
and many other organic reaction~.[~.~~]
The progress made in the field of photochemistry would
not have been possible without interaction between individual chemical disciplines. This is particularly true for the
study of electron-transfer processes, which play a decisive
role in many areas of chemistry, whether they be in biological, inorganic, or organic systems, o r whether they take
place under thermal conditions or under the influence of
light. The enormous increase, in recent years, in the number of publications dealing with electron-transfer processes is less a sign of a short-lived fashion than an indication of growing awareness of chemists that the questions
concerning these processes need to be answered. The development of more refined analytical methods, the investigation of fundamental questions such as the applicability
of the Marcus theory or Weller's empirical relationship to
the kinetics of electron transfer, and synthetic studies by
preparatively oriented organic chemists will give a further
impulse to research in this important area.
The personal work referred to here has only been possible
thanks to the enthusiastic help of my co-workers Karl Buchkremer, Dr. Joachim Gersdorf; Christel Dittmer, Jurgen
Mertes, Cisela Trampe, and Gabi Weber and the cooperation of colleagues working in other areas. I would specially
like to thank Dr. H . Gorner and Dr. S . Steenken (MaxPlanck-Institut fur Strahlenchernie, Miilheim) for help with
the fundamental transient-analysis studies and Dr. J . Runsink (RWTH Aachen) for help with the N M R analyses. I
thank the Deutsche Forschungsgemeinschaft and the Fonds
der Chemischen Industrie for their generous financial help
and, finally, Bayer and BASF for gifrs of materials.
Received: December 2, 1986;
supplemented: May 11, 1987 [A 632 IE]
German version: Angew. Chem. 99 (1987) 849
Translated by Dr. Douglas Mans, Kronberg, Taunus
[I] L. Eberson, Adv. Phys. Org. Chem. 18 (1982) 79.
[2] H. Taube, Angew. Chem. 96 (1984) 315; Angew. Chem. I n t . Ed. Engl. 23
(1984) 329.
131 S . J. Lippard (Ed.): Prog. Inorg. Chem. 30 (1983) 141, 273, 389, 441;
ihrd. 32 (1984) 443.
[4] N. Sutin, J . Photochem. 10 (1979) 19.
[5] M. Gratzel (Ed.): Energy Resources through Photochemistry and Catalysis, Academic Press, New York 1983.
[6] G. J. Kavarnos, N. J. Turro, Chem. Rev. 86 (1986) 401.
[7] H. Leonhardt, A. Weller, Ber. Bunsenges. Phys. Chem. 67 (1963) 791.
181 R. A. Marcus, Annu. Rev. Phys. Chem. I5 (1964) 155.
[9] N. S. Hush, Trans. Faraday Soc. 57 (1961) 557.
[lo] N. Mataga, M. Mitiga, T. Nishimura, J. Mol. Struct. 47 (1978) 199.
[ 1 I] J. C. Scaiano, Acc. Chem. Res. I5 (1982) 252.
[I21 N. J. Turro, Tetrahedron 38 (1982) 809.
[ 131 J. F. Rabeck: Experimental Methods in Photochemistry and Photophysics. Parf 2. Wiley, New York 1982, p. 822ff.
[I41 N. Mataga, Pure Appl. Chem. 56 (1984) 1255.
[I51 J. D. Simon, K. S . Peters, Ace. Chem. Res. 17 (1984) 277.
[I61 E. F. Hilinski, P. M. Rentzepis, Acc. Chem. Res. 16 (1983) 224.
[17] H. D. Roth in L. T. Muus (Ed.): Chemically Induced Magnetic Polari.sation, Reidel, Dordrecht 1977, p. 39.
[I81 G. L. Closs, R. J. Miller, 0. D. Redwine, Acc. Chem. Res. 18 (1985)
[I91 N. J. Turro, J. Photochem. 25 (1984) 69.
[20] P. J. Hore, C. G. Joslin, K. A. McLauchlan, Chem. SOC.Rev. 8 (1979)
[21] J. Terner, M. A. El-Sayed, Arc. Chem. Res. 18 (1985) 331.
[22] M. Julliard, M. Chanon, Chem. Rev 83 (1983) 425.
I231 S . L. Mattes, S. Farid in A. Padwa (Ed.): Organic Photochemistry, Yo/.
6. Dekker, New York 1983, p. 233ff.
(241 P. S. Mariano in W. M. Horspool (Ed.): Synthetic Organic Photochemisfry. Plenum, New York 1984, p. 145-257.
[2S] N. J. Turro: Modern Molealar Photochemisrry, Benjamin/Cummings,
Menlo Park, CA, USA 1978.
I261 J. Heinze, Angew. Chem. 96 (1984) 823; Angew. Chem. Int. Ed. Engl. 23
( 1984) 83 I .
[27] D. Rehm, A. Weller, Z . Phys. Chem. (Munich) 69 (1970) 183.
[28] D. Rehm, A. Weller, Isr. J . Chem. 8 (1970) 259.
[29] A. Weller, Z . Phys. Chem. (Munich) 133 (1982) 93.
[30] A. Weller, Z . Phys. Chem. (Munich) 130 (1982) 129; Pure Appl. Chem. 54
(1982) 1885.
1311 J. R. Miller, L. T. Calcaterra, G. L. Closs, J . Am. Chem. SOC.106 (1984)
[32] A. Weller in M. Gordon, W. R. Ware (Ed.): rite Exciplex. Academic
Press, New York 1975, p. 23ff.
[33] H. Beens, H. Knibbe, A. Weller, J. Chem. Phys. 47 (1967) 1183; H
Beens, A. Weller, Phys. Chem. Lett. 2 (1968) 140.
1341 H. Masuhara, N. Mataga,Acc. Chem. Res. 14 (1981) 312.
[35] F. D. Lewis, Acc. Chem. Res. 12 (1979) 152.
1361 R. A. Caldwell, Acc. Chem. Res. 13 (1980) 45.
1371 S . L. Mattes, S . Farid, Acc. Chem. Res. 15 (1982) 80.
[38] J. Mattdy, H. Leismann, H:D. Scharf, Chem. Ber 112 (1979) 577; Mol.
Pholochem. 9 (1979) 119.
[39] H. Leismann, J. Mattay, H.-D. Scharf, J. Am. Chem. SOC. 106 (1984)
1401 J. Mattay, J. Gersdorf, H. Leismann. S. Steenken, Angew. Chem. 96
(1984) 240; Angew. Chew. Int. Ed. Engl. 23 (1984) 249.
1411 J. Mattay, J. Runsink, J. Gersdorf, T. Rumbach, C. Ly, J. Am. Chem.
SOC. 107 (1985) 2557; Helu. Chim. Acta 69 (1986) 442.
[42] J. Mattdy, Tetrahedron 41 (1985) 2393, 2405.
1431 J. Mattay, J. Photochem. 37 (1987) 167.
[44] J. Gersdorf, J. Mattay, H. Corner, J . Am. Chew. SOC.109 (1987) 1203.
1451 J. Mattay, J. Gersdorf, K. Buchkremer, Chern. Ber. I20 (1987) 307.
[46] H. Leismann, J. Mattay, Tetrahedron Lett. 1978, 4265.
1471 R. A. Caldwell, L. Smith, J . Am. Chem. SOC.96 (1974) 2994.
[48] G. Schrnitt, Dissertanon. Technische Hochschule Aachen 1986.
1491 J. Mattay, J. Gersdorf, U. Freudenberg, Tetrahedron Lett. 25 (1984)
[SO] J . Mattay, J. Gersdorf, J. Mertes, J. Chem. SOC.Chem. Commun. 1985.
1511 J. Mattay, G. Trampe, S. Steenken, Proc. 11th lUPAC-S.vmp. Photochem. (Lissabon 1986), pp. 114-115.
(521 J. H. Fendler: Membrane Mimetic Chemistry. Wiley, New York 1982;
Chem. Eng. News 62 (1984) No. 1. p. 25.
(531 M. Gratzel, Pure Appl Chem. 54 (1982) 2369.
154) H. Gerischer, J. J. Katz (Eds.): Light-Induced Charge Separation in Biology and Chemistry Systems (Life Sci. Res. Rep. 12). Verlag Chemie,
Weinheim 1979.
I551 Y. Taniguchi, Y. Nishina, N. Mataga, Bull Chem. SOC.Jpn. 45 (1972)
156) L. Reichardt: Solvent Effprts in Organic Chemistry, Verlag Chemie,
Weinheim 1979; Angew. Chem. 77 (1965) 30; Angew. Chem. lnt. Ed.
Engl. 4 (1965) 29; ibid. 91 (1979) 119 and 18 (1979) 98.
[57] M. Szwarc (Ed.): Ions and Ion Pairs in Organic Reacfions. Yo/. I . 2.
Wiley, New York 1974.
[58] N. Hirotd, S. J. Weissmann, J. Am. Chem. SOC.86 (1964) 2538; N . Hirota in E. T. Kaiser, L. Kevan (Eds.): Radical Ions. Wiley, New York
1968, p. 35.
1591 K. Nakamurd, J. Am. Chem. SOC.102 (1980) 7846.
1601 J. Gersdorf, J. Mattay, J . Photochem. 28 (1985) 405.
[61] J. Mattay, G. Trampe, J. Mertes, J. Gersdorf, Proc. 11th 1UPAC-Symp.
Photochem. (Lisbon 1986), pp. 209-210.
1621 L. C. Bateman, M. G. Church, E. D. Hughes, C. K. Ingold, N. A. Taher, J . Chem. SOC.1940. 979.
1631 A. H. Fainberg, S. J. Winstein, J. Am. Chem. SOC.78 (1956) 2767.
1641 J. Mattay, J. Gersdorf, H. Gorner, Proc. 10th IUPAC-Symp. Photochem.
(Interlaken, Switzerland 1984), pp. 153-154.
1651 J. Gersdorf, Dissertation. Technische Hochschule Aachen 1985.
[66] J. L. Goodman, K. S Peters, J . Am. Chem. SOC.107 (1985) 6459.
[67] B. E. Goodson, G. B. Schuster, J . A m . Chem. SOC.106 (1984) 7254.
[68] B. E. Goodson, G. B. Schuster, Tetrahedron Lett. 27(1986) 3123.
[69] J. M. Masnovi, J. K. Kochi, J . Am. Chem. SOC.I07 (1985) 7880.
I701 D. E. Ayer, G. H. Biichi, US Pat. 2805242 (1957). E. 1. duPont d e Nemours;Chem. Abstr. 52 (1958) 2904a.
[71] H. J . F. Angus, D. Bryce-Smith, Proc. Chew. SOC (London) 1959. 326; J .
Chem. SOC. 1960, 4791.
[72] G. 0. Schenck, R. Steinmetz, Tetrahedron Lett. 1960, I .
1731 E. Grovenstein, D. V. Rao, J. W. Taylor, J. Am. Chem. SOC.83 (1961)
[74] D. Bryce-Smith, A. Gilbert, B. H. Orger, Chem. Commun. 1966, 512.
1751 K. E. Wilzbach, L. Kaplan, J . Am. Chem. Soc. 88 (1966) 2066.
[76] A. Gilbert in D. Bryce-Smith (Ed.): Photochemistry (Spec. Period. Rep.
Photochem.) 1-17 (1970- 1986), The Royal Society of Chemistry, London.
1771 J. Mattay, J. Runsink, R. Hertel, J. Kalbe, 1. Schewe, J . Photochem. 37
(1987) 335.
1781 J . J. McCullough, R. C. Miller, D. Fung, W. S . Wu, J. Am. Chem. SOC.
97 (1975) 5942.
[79] J. J. McCullough, R. C. Miller, W. S. Wu, Can. J . Chem. 55 (1977)
[80] D. R. Arnold, P. C. Wong, A. J. Maroulis, T. S. Cameron, Pure Appl.
Chem. 52 (1980) 2609.
[Sl] A. Yoshino, K. Yamasaki, T. Yonezawa, M. Ohashi, J . Chem. SOC.Perkin Trans. I 1975. 735.
[82] A. Albini, E. Fasani, R. Oberti, Tetrahedron 38 (1982) 1027.
[83] F. D. Lewis, R. J DeVoe, Tetrahedron 38 (1982) 1069.
I841 K. Mizuno, M. Ikeda, Y. Otsuji, Tetrahedron Lett. 26 (1985) 461.
[851 K. Mizuno, K. Terasaka, M. Ikeda, Y . Otsuji, Tetrahedron Lett. 26
(1985) 5819.
1861 A. Albini, E. Fasani, M. Mella, J. A m . Chem. SOC.I08 (1986) 41 19.
[87] A. Albini, E. Fasani, A. Sulpizio, J . A m . Chem. SOC.I06 (1984) 3562.
1881 M. Yasuda, C. Pac, H. Sdkurai, Bull. Chem. SOC.Jpn. 54 (1981) 2352.
I891 F. D. Lewis, T. I. Ho, J. A m . Chem. SOC.102 (1980) 1751.
[90] R. M. Bowman, T. R. Chamberlain, C. W. Huang, J J. McCullough, J.
A m . Chem. SOC.96 (1974) 692.
1911 T. Miyamoto, T. Mori, Y. Odaira, Chem. Commun. 1970. 1598.
1921 J . M. Masnovi, J. K. Kochi, E. F. Hilinski, P. M. Rentzepis, J . Am.
Chem. SOC. 108 (1986) 1126.
[931 E. F. Hilinski, J. M. Masnovi, J. K. Kochi, P. M. Rentzepis, J. Am.
Chem. SOC. 106 (1984) 8071.
Angew. Chew. In!. Ed. Engl. 26 (1987) 825-845
[94] J. M. Masnovi, J. K. Kochi, J . Org. Chem. SO (1985) 5245; R e d . J . R.
Netli. Cliem Soc. 105 (1986) 286.
[YS] G. Briegleb. Elekrronen-Donor-Acceptor-Komplexe. Springer, Berlin
[96] S. Sankararaman, J. K. Kochi, Recl. J . R. Neth. Chem. SOC. 10.5 (1986)
[97] D. P. Kjell, R. S. Sheridan, J . Photochem. 28 (1985) 205.
[YSl R. J. Sundberg in A. Padwa (Ed.): Organrc Phorochemistry. Yo/. 6. Dekker, New York 1983, pp. 121-176.
Chem. 66 (1976) I .
[99] P. J. Wagner. T c J ~Curr.
[loo] G. Jones I I in A. Padwa (Ed.): Organic Photochemistry. Yo/. 5, Dekker,
New York 1981, pp. 1-122.
[loll M. Braun, Nachr. Chem. Tech. Lab. 33 (1985) 213.
[I021 H.-D. Scharf. H. Herzog, R. Pelzer, unpublished.
[I031 1. Mattay, J. Runsink, J . Org. Chem. 50 (1985) 2815.
[I041 P. J. Wagner, R. J. Truman, A. E. Puchalski, R. Wake, J. Am. Chem.
Soc. I08 (1986) 7727 We thank P. J . Wagner for a preprint of this
[IOS] S. G. Cohen, R. J. Baumgarten, J . Am. Chem. SOC.89 (1967) 3471.
11061 S. G. Cohen, A. Parola, G. H. Parsons, Chem. Rev. 73 (1973) 141.
[I071 D. Seebach, H. Daum, J . Am. Chem. SOC.93 (1971) 2795.
[I081 J. D. Simon, K. S. Peters, J . Am. Chem. SOE.I03 (1981) 6403; ibid. 105
(1983) 4875.
[I091 L. M. Bruce in S. Patai (Ed.): The Chemistry ofQuinonoid Compounds,
Vol. I , Wiley, New York 1974, pp. 465-538.
[ I lo] K. Maruyama, T. Otsuki, S. Tai, J . Org. Chem. SO (1985) 52.
[ I I I ] J. D. Coyle, Chem. Rev. 78 (1978) 97.
[ I 121 Y. Kanaoka, Acc. Chem. Res. I 1 (1978) 407.
[ I 131 P. H. Mazzocchi in A. Padwa (Ed.): Orgonic Phofochemisrry. Yo/. 5.
Dekker, New York 1981, pp. 421-471.
[ I 141 J. D. Coyle in W. M. Horspool (Ed.): Synthetic Organic Photochemistry.
Plenum, New York 1984, pp. 259-284.
[ I IS] P. H. Mazzocchi, S. Minamikawa, P. Wilson, J . Org. Chem. SO (1985)
268 I .
[ I 161 P. H. Mazzocchi, P. Wilson, F. Khachik, L. Klingler, S. Minamikawa, J .
Org. Chem. 48 (1983) 2981.
[ I 17) P. H. Mazzocchi, G. Fritz, J . Am. Chem. SOC.108 (1986) 5362.
[ I 181 K. Maruyama, T. Ogawa, Y. Kubo, T. Araki, J . Chem. SOC. Perkin
Truns. I 1985. 2025.
[I191 M. Machida, K. Oda, Y. Kanaoka, Tetrahedron 41 (1985) 4995.
[I201 P. S. Mariano, Tetrahedron 39 (1983) 3845; Acc. Chem. Res. 16 (1983)
[I211 P. S. Mariano, J. Stavinoha, E. Bay, Tefrahedron 37 (1981) 3385.
[I221 H. Bohme, H. G. Viehe: lmminium Salts in Organic Chemistry. Wiley,
New York 1976.
11231 W. N. Speckamp, H. Hiemstra, Tetrahedron 41 (1985) 4367.
[I241 S. D. Larsen, P. A. Grieco, W. F. Fobare, J . Am. Chem. SOC.108 (1986)
3512, and references cited therein.
[I251 K. Ohga, U. C. Yoon, P. S. Mariano, J . Org. Chem. 49 (1984) 213.
[I261 G. Dai-Ho, A. J. Y. Lan, P. S. Mariano, Tetrahedron Letr. 26 (1985)
[I271 L. P. Ellinger, Polymer 5 (1964) 599.
[I281 A. Ledwith in M. Gordon, W. R. Ware (Eds.): The Exciplex. Academic
Press, New York 1975, pp. 209-245.
[I291 A. Ledwith, Acc. Chem. Res. 5 (1972) 133.
[I301 S. Farid, S. E. Hartman, T. R. Evans in M. Gordon, W. R. Ware (Eds.):
The Exciplex. Academic Press, New York 1975, pp. 327-343.
[I311 S. Kuwata, Y. Shigernitsu, Y. Odaira, J . Org. Chem. 38 (1973) 3803.
[I321 T. R. Evans, R. W. Wake, 0. Jaenicke in M. Gordon, W. R. Ware
(Eds.): The Exciplex. Academic Press, New York 1975, pp. 345-358.
[I331 T. Asanuma, M. Yarnamoto, Y. Nishijima, J . Chem. Suc. Chem. Commun. 1975. 53, 56, 608.
[I341 M. Kojima, H. Sakuragi, K. Tokumaru, Tefrahedron Lett. 22 (1981)
[I351 R. A. Neunteufel, D. R. Arnold, J . Am. Chem. SOC.95 (1973) 4080.
[I361 T. Majima, C. Pac, A. Nakasone, H. Sakurai, J . Am. Chem. Soc. 103
( I98 1 ) 4499.
[I371 C . Pac, Pure Appl. Chem. 58 (1986) 1249.
[I381 A. Albini, Synthesis 1981. 249.
11391 G. C. Calhoun, G. 8. Schuster, J . Am. Chem. SOC.108 (1986) 8021.
[I401 N. L. Bauld, R. Pabon, J. Am. Chem. Soc. IOS (1983) 633.
[I411 T. Asanuma, T. Gotoh, A. Tsuchida, M. Yamamoto, Y. Nishijima, J .
Chem. SOC.Chem. Commun. 1977. 485.
[I421 T. Gotoh, M. Kato, M. Yarnamoto, Y. Nishijima, J . Chem. Soc. Chem.
Commun. 1981, 90.
[I431 S. Tazuke, N. Kitamura, J . Chem. SOC. Chem. Commun. 1977, 515.
[I441 S. L. Mattes, S. Farid, J . Am. Chem. Soc. 108 (1986) 7356.
[I451 K. Mizuno, R. Kaji, Y. Otsuji, Chem. Lerr. 1977. 1027.
[I461 K. Mizuno, H. Ueda, Y. Otsuji, Chem. Left. 1981. 1237.
[I471 K. N. Houk. Acc. Chem. Res. 8 (1975) 361.
[I481 J. Sauer. R. Sustmann, Angew. Chem. 92 (1980) 773; Angew. Chem. Inr.
Ed. Engl. 19 (1980) 779.
[ 1491 1. Fleming: Grenzorbitale und Reaktionen organischer Verbmdungen.
Verlag Chemie, Weinheim 1979.
Anyew. Chem. In(. Ed. Engl. 26 (1987) 825-845
[I501 W. J. le Noble, H. Kelm, Angew. Chem. 92 (1980) 887; Angew Chem.
I n t . Ed. Engl. 19 (1980) 841.
[lSl] W. J. le Noble, Chem. Unserer Zeit 17(1983) 152.
[I521 K. Matsumoto, A. Sera. T. Uchida, Synthesis 198.5. I, 999.
11531 P. Yates, P. Eaton, J. Am. Chem. SOC.82 (1960) 4436.
[ I541 H. Wollweber in Houben- Weyl-Miiller: Methoden der Organischen
Chemie. Band V / l c , Thieme, Stuttgart 1970, p. 1092.
[I551 P. Welzel, Nachr. Chem. Tech. Lab. 31 (1983) 979.
[I561 D. C. Rideout, R. Breslow, J . Am. Chem. SOC.102 (1980) 7816.
[I571 P. A. Grieco, P. Galatsis, R. F. Spohn, Tetrahedron 42 (1986) 2847.
[ 1581 K. Mizuno, R. Kaji, H. Okada, Y. Otsuji, J. Chem. SOC.Chem. Commun. 1978, 594.
[I591 D. J. Beleville, D. D. Wirth, N. L. Bauld, J. Am. Chem. SOC.103 (1981)
[I601 N. L. Bauld, D. J. Beleville, R. Pabon, R. Chelsky, G. Green, J . Am.
Chem. SOC.105 (1983) 2378.
(1611 C. R. Jones, B. J. Allman, A. Mooring, B. Saphic, J . Am. Chem. SOC.
105 (1983) 652.
[I621 J. Mattay, J. Gersdorf, J. Mertes, J . Chem. SOC.Chem. Commun. 198s.
11631 G . TramDe. Dissertation. Technische Hochschule Aachen 1987.
1. Mattay, J. Gersdorf, G. Trampe, Chemiedozenrentagung (Aachen
1985). Proceedings p. 42; 20. GDCh-Hauptversammlung (Heidelberg
1985). Proceedings p. 1 I I .
I. Mlcoch, E. Steckhan, Angew. Chem. 97 (1985) 429; Angew. Chem.
Inr. Ed. Engl. 24 (1985) 412.
D. Valentine, N. J. Turro, G. S. Hammond, J . Am. Chem. SOC.86 (1964)
a) G. 0. Schenck, S. P. Mannsfeld, G. Schomburg, C. H. Krauch, 2.
Naturforsch. B19 (1964) 18; b) R. Schutte, G. R. Freeman, J . Am.
Chem. SOC.91 (1969) 3715; c) T. L. Palmer, D. G. Whitten, G. S. Hammond, ibid. 92 (1970) 286 I .
F.-G. Klarner, B. M. J. Dogan, 0. Errner, W. von E. Doering, M. P.
Cohen, Angew. Chem. 98 (1986) 109; Angew. Chem. Inr. Ed. Engl. 25
(1986) 108.
G. S . Groenewald, M. L. Gross, J . Am. Chem. SOC.106 (1984) 6569.
H. D. Roth, M. L. M. Schilling, C. J. Abelt, J . Am. Chem. Soc. 108
(1986) 6098.
1. Mertes, Diplumarbert. Technische Hochschule Aachen 1986.
1. Mattay, G. Trampe, F:G. Klarner, unpublished.
R. A. Pabon, D. J. Beleville, N. L Bauld, J . Am. Chem. SOC.10.5 (1983)
G. Trampe, J . Mattay, 2nd Postgraduate Symp. Photochem. (Miilheim a.
d. Ruhr 1986). Proceedings pp. 34-35.
G. Schmidt, Dissertation. Technische Hochschule Aachen 1986.
W. Baumann, H. Bischof, J. C. Frohling, 1. Lumin. 24/25 (1981) 555.
M. Lemaire, A. Guy, D. Imbert, J.-P. Guette, J . Chem. SOC.Chem. Cummun. 1986. 741.
A. J. Maroulis, D. R. Arnold, Synthesis 1979. 819.
S. L. Mattes, S. Farid, J . Chem. SOC.Chem Commun. 1980, 126.
A. J. Maroulis, Y. Shijemitsu, D. R. Arnold, J . Am. Chem. SOC. 100
(1978) 535.
G. Weber, R. Runsink, J. Mattay, J . Chem. SOC. Perkin Trans. 1. in
A. Pross, J . Am. Chem. SOC.108 (1986) 3537.
H. D. Roth, M. L. M. Schilling, J . Am. Chem. SOC.101 (1979) 1898;
ibid. 102 (1980) 4303.
F. D. Lewis, J. R. Petisce, J. D. Oxman, M. J. Nepras, J . Am. Chem.
SOC. 107 (1985) 203.
G. A. Lee, S. H. Israel, J . Org. Chem. 48 (1983) 4557.
T. R. Evans, R. W. Wake, M. M. Sifain, Tefrahedron Left. 1973. 701.
G. Jones 11, W. G. Becker, S. H. Chiang, J . Photochem. 19 (1982) 245.
K. Okada, K. Hisamitsu, T. Miyashi, T. Mukai, 1. Chem. SOC.Chem.
Commun. 1982, 974.
H . D. Roth, M. L. M. Schilling, G. Jones 11, J . Am. Chem. Suc. 103
(1981) 1246.
H. D. Roth, M. L. M. Schilling, J . Am. Chem. SOC.103 (1981) 7210.
G. Jones 11, S. H. Chiang, W. G. Becker, A. J. Welch, J . Phys. Chem. 86
(1982) 2805.
A. Padwa, C . S. Chou, W. F. Rieker, J . Org. Chem. 45 (1980) 4555.
C. J. Abelt, H. D. Roth, J . Am. Chem. SOC.107 (1985) 3840.
P. G. Gassman, B. A. Hay, J . Am. Chem. SOC. 108 (1986) 4227.
K. A. Brown-Wensley, S. L. Mattes, S. Farid, J . Am. Chem. SOC.100
(1978) 4162.
D. W. Reynolds, N. L. Bauld, Tefrahedron Lett. 26 (1985) 2539.
D. F. Eaton, Pure Appl. Chem. 56 (1984) 1191.
M. Chanon, M. L. Tobe, Angew. Chem. 94 (1982) 27; Angew. Chem. Int.
Ed. Engl. 21 (1982) I .
L. G. Vanquickenborne, A. Ceulemans, Coord. Chem. Rev. 48 (1983)
M. A. Fox, Adu. Photochem. 13 (1986) 237.
G. L. Geoffroy, M. S. Wrighton: Organometallic Photochemistry. Academic Press, New York 1979.
J. H. Fendler, J . Phys. Chem. 89 (1985) 2730.
Без категории
Размер файла
1 869 Кб
photochemistry, transfer, radical, ions, charge
Пожаловаться на содержимое документа