Temperature dependent solvent effects in photochemistry of 1-phenylpentan-1-ones

Article abstract of DOI:10.1135/cccc19992007

Temperature dependent solvent effects have been investigated on the Norrish Type II reaction of 1-phenylpentan-1-one and its p-methyl derivative. Efficiencies of the photoreaction were studied in terms of solvent polarity and base addition as a function of temperature. Such a small structure change as the p-methyl substitution in 1-phenylpentan-1-one altered the temperature dependent photoreactivity in presence of weak bases. The experimental results suggest that the hydrogen bonding between the Type II biradical intermediate OH group and the solvent is weaker for 1-(4-methylphenyl)pentan-1-one than that for 1-phenylpentan-1-one at 20°C but the interactions probably vanish in both cases at 80°C.

Full text of DOI:10.1135/cccc19992007

Solvent Effects in Photochemistry
2007
TEMPERATURE DEPENDENT SOLVENT EFFECTS IN
PHOTOCHEMISTRY OF 1-PHENYLPENTAN-1-ONES
1
,
2
Petr KLÁN * and Jaromír LITERÁK
Department of Organic Chemistry, Masaryk University, Kotlářská 2, 611 37Brno,
1
2
Czech Republic; e-mail: klan@sci.muni.cz, literak@chemi.muni.cz
Received Septem ber 27, 1999
Accepted October 18, 1999
Dedicated to Professor Milan Kratochvíl on the occasion of his 75th birthday.
Tem perature depen den t solven t effects h ave been in vestigated on th e Norrish Type II reac-
tion of 1-ph en ylpen tan -1-on e an d its p-m eth yl derivative. Efficien cies of th e ph otoreaction
were studied in term s of solven t polarity an d base addition as a fun ction of tem perature.
Such a sm all structure ch an ge as th e p-m eth yl substitution in 1-ph en ylpen tan -1-on e altered
th e tem perature depen den t ph otoreactivity in presen ce of weak bases. Th e experim en tal re-
sults suggest th at th e h ydrogen bon din g between th e Type II biradical in term ediate OH
group an d th e solven t is weaker for 1-(4-m eth ylph en yl)pen tan -1-on e th an th at for
1
-ph en ylpen tan -1-on e at 20 °C but th e in teraction s probably van ish in both cases at 80 °C.
Key w ord s: Ph otoch em istry; Solven t effects; Keton es; Norrish Type II reaction ;
Valeroph en on e.
Structure an d solven t effects on ph otoreactivity of alkyl aryl keton es h ave
1
been exten sively studied over four decades . Alkyl ph en yl keton es with h y-
drogen on γ-carbon react upon irradiation on th eir alkyl ch ain s accordin g
to Norrish Type II reaction 2 via th e triplet state to produce th e triplet
l
1
,4-biradicals, wh ich in tersystem cross to th e sh ort lived sin glet biradicals
3
th at can cleave, cyclize (Yan g cyclization ) , or disproportion ate back to th e
startin g keton e (Sch em e 1). Scaian o’s group provided tem perature studies of
-ph en ylpen tan -1-on e an d its p-m eth oxy derivative in n on polar solven ts4
1
an d Wagn er studied tem perature effects on ph otoch em istry of m olecules,
5
in wh ich two differen t ph en yl keton es are attach ed to a flexible teth er .
Zepp an d h is coworkers sh owed th at goin g from 20 to 30 °C, th e triplet re-
6
activity of 1-ph en ylpen tan -1-on e in water in creased . Th ere h as been , h ow-
ever, n o detailed study of tem perature effects on th e ph en yl keton e triplet
lifetim e in polar an d protic solven ts yet.
Collect. Czech. Chem. Commun. (Vol. 64) (1999)

2
008
Klán, Literák:
We h ave recen tly studied a sim ultan eous effect of both ultraviolet an d
m icrowave irradiation on valeroph en on e ph otoreactivity as part of our pro-
7
gram on microwave photochemistry an d on e of us also dem on strated an ap-
plication of Norrish Type II reaction in in tram olecular triplet en ergy
8
tran sfer studies . Now we wish to report addition al results from our in vesti-
gation s on ph otoreactivity of 1-ph en ylpen tan -1-on e (valeroph en on e, 1a)
an d 1-(4-m eth ylph en yl)pen tan -1-on e (4-m eth ylvaleroph en on e, 1b) as a
fun ction of tem perature, wavelen gth , an d solven t properties.
H
H
3
O
O
O
OH
*
1
. hν
2. ISC
R
R
R
R
1
1
a, R = H
b, R = CH3
2
3
HO
+
R
4
5
SCHEME 1
EXPERIMENTAL
Equipm en t
1H an d 13C NMR spectra were obtain ed for solution s in CDCl3 on an Avan ce Bruker DRX
00. Ch em ical sh ifts are given in ppm (δ-scale), couplin g con stan ts J in Hz. Gas ch rom ato-
graph y was accom plish ed on a Sh im adzu GC-17A apparatus an d on a GC/m ass system TRIO
000 (FISONS In strum en ts). UV spectra were obtain ed on a Sh im adzu UV-1601 in strum en t
5
1
with m atch ed 1.0 cm quartz cells an d tem perature depen den t absorption spectra were m ea-
sured on a Cary 50 equipped with th e Fibre Optic dip probe an d Cary Win software.
Ch em icals an d Solven ts
1
-Ph en ylpen tan -1-on e (1a) was obtain ed from Aldrich Ch em ical Co. an d was purified by dis-
tillation un der reduced pressure. 1-(4-Meth ylph en yl)pen tan -1-on e (1b) was syn th esized from
9
p-m eth ylben zon itrile an d butylm agn esium brom ide accordin g to th e Wagn er’s procedure
an d was purified by distillation un der reduced pressure. Hexadecan e (99+%) from
Sch uch ardt was used as received. Reagen t grade ben zen e was wash ed with sulfuric acid un til
furth er portion s rem ain ed colourless, th en wash ed with water, dried, an d distilled over
P O , b.p. 80 °C. Reagen t grade cycloh exan e was worked up with sulfuric acid in th e sam e
2
5
way as ben zen e a th en distilled over CaH , b.p. 81 °C. Aceton itrile (an alytical grade) was
2
dried an d distilled over CaCl , b.p. 80–81 °C. Eth an ol (an alytical grade) was dried an d dis-
2
tilled over sodium , b.p. 78 °C. Pyridin e from Aldrich (99+%) was distilled before th e use.
Collect. Czech. Chem. Commun. (Vol. 64) (1999)

Solvent Effects in Photochemistry
2009
1
-(4-Methylphenyl)pentan-1-one. M.p. 19–20 °C. 1H NMR (500 MHz, CDCl ): 0.95 (t, J = 7.3,
3
3
7
1
H); 1.41 (m , 2 H); 1.71 (m , 2 H); 2.40 (s, 3 H); 2.92 (t, J = 7.5, 2 H); 7.24 (d, J = 8.1, 2 H);
.85 (d, J = 8.1, 2 H). 13C NMR (125 MHz, CDCl ): 13.79, 21.42, 22.39, 26.52, 38.08, 128.08,
3
29.10, 134.57, 143.43, 200.15.
Irradiation Procedures
All solution s were prepared directly by weigh in g th e desired m aterial in to volum etric flasks
or by dilution of stock solution s. Solution m ixtures (3 m l) of both 1a an d 1b of th e sam e
con cen tration in 13 × 100 m m Sim ax (Pyrex equivalen t) tubes were degassed by bubblin g
th e solution s with argon for 15 m in . Magn etically stirred sam ples were irradiated in tubes
equipped with a con den ser. All ph otoch em ical experim en ts were accom plish ed un der argon
at 20 °C or at boil (≈80 °C; for all solven ts used), usin g a con ven tion al m edium -pressure
m ercury lam p (400 W, Con rad–Han ovia) with 2 m m th ick Sim ax filter (>280 n m ) or 5 m m
1
0
th ick solution filters . Th e solution cut-off filter for >313 n m was prepared by dissolvin g
NaBr (400 g) an d Hg(NO3)2 (1 g) in water (1 000 m l) an d th e filter for >366 n m by dissolv-
in g NaBr (650 g) an d Pb(NO3)2 (3 g) in water (1 000 m l). Ph otoreaction con version s in all
experim en ts were always kept below 20%. Tim e of irradiation experim en ts was usually 1 h .
An alysis of th e Irradiated Sam ples
Th e on ly ph otoproducts followed for all reactan ts in th is work were acetoph en on es (4) an d
cyclobutan ols (5). Form ation of a m in or ph otoproduct (<2% yield, assum in g th e sam e GC
respon se factor as valeroph en on e) was also observed an d was believed to be 1-arylcyclopen -
6
tan ol as described elsewere . Iden tification of acetoph en on es an d startin g valeroph en on es,
1
13
isolated by flash ch rom atograph y, was based on H NMR, C NMR, an d on GC com parison s
with auth en tic sam ples (Aldrich ). Cyclobutan ols were iden tified as described in th e literature
6
,11
usin g GC an d GC/m ass in strum en ts
. Th e ph otoproducts are stable un der all GC con di-
–
4
tion s used. For com parison , 4 · 10 M solution valeroph en on e in aceton e was irradiated us-
in g Sim ax-filtered radiation from Hg lam p un til th e reaction reach ed th e first h alf-tim e. Th e
products were an alyzed by GC/m ass spectrom etry an d th e results were successfully com -
6
pared to th ose described in th e literature . Th e sam e an alysis was accom plish ed for
4
-m eth ylvaleroph en on e. Con cen tration s of th e ph otoproducts were calculated from peak in -
tegration s of data obtain ed with flam e ion ization detection , assum in g th at th e cyclobutan ols
h ad th e sam e respon se as valeroph en on es.
Each sam ple was an alyzed th ree tim es; in case th at th e values differed by m ore th an 5%
th e m easurem en t was repeated. Sam ples were an alyzed by GC with respon se factors for
products, startin g keton es, an d h exadecan e as in tern al stan dard. Th e reaction efficien cies are
n ot corrected on m olar absorption coefficien ts. Sin ce th e coefficien ts of both keton es in all
solven ts were in depen den t on tem perature th e calculation s of reaction efficien cy are n ot af-
fected as discussed in th e followin g ch apter. Molar absorption coefficien ts of both 1a an d 1b
for th ree wavelen gth s are listed in Table I.
We tested th e stability of cyclobutan ols in boilin g solven ts. Th e reaction m ixtures after
th e irradiation at 20 °C were h eated to 80 °C for 4 h (four tim es as lon g as an average ph o-
toch em ical experim en t) an d re-an alyzed. Con cen tration s of th e ph otoproducts rem ain ed un -
ch an ged in all cases.
Collect. Czech. Chem. Commun. (Vol. 64) (1999)

2
010
Klán, Literák:
TABLE I
Molar absorption coefficien ts of valeroph en on e (1a) an d 4-m eth ylvaleroph en on e (1b)
εa (1a)
εa (1b)
λ, n m
ben zen e
eth an ol
aceton itrile
ben zen e
eth an ol
aceton itrile
2
3
3
80
13
66
908.2b
55.2
4.0
990.2
69.3
891.3
60.3
764.9b
60.6
3.8
1 687.4
106.2
1 091.2
84.7
a
Molar absorption coefficien ts in l m ol–1 cm –1. b In cycloh exan e.
TABLE II
a
Solven t an d wavelen gth effects on th e ph otolysis of 1a an d 1b m ixtures
F/Cb
Wavelen gth ,
T, °C
Solven t
Rc
n m
1
a
1b
2
8
2
8
2
8
0
0
0
0
0
0
>280
>313
>366
ben zen e
4.05
4.16
4.03
4.06
3.79
4.07
5.11
5.14
4.60
4.66
4.59
4.89
1.22
1.03
1.25
1.10
1.23
1.07
2
8
2
8
0
1
0
1
>280
>313
aceton itrile
4.50
5.10
4.44
4.89
5.87
6.82
5.87
7.06
1.93
1.08
2.33
1.21
2
7
2
7
0
9
0
9
>280
>313
eth an ol
6.00
6.15
6.04
6.65
5.65
7.73
7.59
7.66
1.93
1.40
1.93
1.59
a
Th e keton e con cen tration s were 0.01 m ol l–1. Th e reproducibility for each m easurem en t
was ±5%. Th e Type II fragm en tation /cyclization ratio for both 1a an d 1b. c Ph otoch em ical
b
efficien cy ratios R = ([4a] + [5a]/[4b] + [5b]); n ot corrected on ε values.
Collect. Czech. Chem. Commun. (Vol. 64) (1999)

Solvent Effects in Photochemistry
2011
RESULTS AND DISCUSSION
We in vestigated th e Norrish Type II (Sch em e 1) reactivity of two ph en yl ke-
ton es, valeroph en on e (1a) an d 4-m eth ylvaleroph en on e (1b), as a fun ction
of tem perature, wavelen gth , an d solven t properties. Equim olar m ixtures of
both keton es were irradiated an d ch em ical yields of ph otoproducts were
obtain ed; such experim en tal arran gem en t guaran teed iden tical ph otoch em -
ical con dition s for both com poun ds. Type II cleavage to acetoph en on e (4a,
R = H), 4-m eth ylacetoph en on e (4b, R = CH ), respectively, an d cyclization
3
to two isom eric 2-m eth yl-1-ph en ylcyclobutan -1-ols an d its p-m eth yl deriva-
tives (5a an d 5b , respectively) were th e dom in an t path ways for th e
ph otoreaction of both valeroph en on es un der all experim en tal con dition s.
1
Acetoph en on es were, as expected, m ajor detectable ph otoproducts . We
h ave n ot directly m easured th e quan tum yields because of tech n ical diffi-
culties in experim en ts with boilin g solution s.
Table II displays th e data from irradiation of equim olar keton e m ixtures
at differen t tem peratures, wavelen gth s, an d in th ree differen t solven ts. Th e
fragm en tation /cyclization ratio was m on itored for both 1a an d 1b. Th e ra-
tio of ph otoproduct con cen tration s ([4a] + [5a]/[4b] + [5b]) from th e sam e
experim en t is th e ph otoch em ical efficien cy ratio. Th e results from irradia-
tion experim en ts, in wh ich pyridin e or water as Lewis base additives were
added in to th e m ixtures are sh own in Table III.
Solven t an d tem perature effects on th e ph otoreaction efficien cy of alkyl
ph en yl keton es an d ph otoproduct form ation depen d on several factors th at
relate to th e n ature of th e lowest triplet as well as in teraction s of th e
biradical in term ediate with th e solven t. Sch em e 2 exem plifies th e Norrish
Type II reaction an d possible solven t effects on th e course of th is reaction .
In th e followin g discussion , we will con sider (i) h ydrogen bon din g capabili-
ties of th e carbon yl group with th e solven t, (ii) h ydrogen bon din g capabili-
ties of th e biradical OH group with th e solven t – (k an d k ), an d (iii)
S
–S
com petition between cleavage an d cyclization (k or k ).
E
C
Type of the Excitation
It is well kn own th at alkyl ph en yl keton es h ave th e lowest n ,π* an d π,π*
1
2
excited states close in en ergy an d vibron ically coupled . An electron -
releasin g substituen t such as p-m eth yl group in th e arom atic rin g stabilizes
th e π,π* triplet. Th is effect also occurs in th e presen ce of a solven t with
h igh er perm ittivity an d better h ydrogen -bon din g abilities (1 (solvated) an d 2
S
H
(
solvated); Sch em e 2) causin g slower h ydrogen atom abstraction (k ). Such
Collect. Czech. Chem. Commun. (Vol. 64) (1999)

2
012
Klán, Literák:
stabilization h as been eviden ced by red sh ifts of th e π–π* electron ic tran si-
1
2a
tion s . At th e sam e tim e, th e n ,π* tran sition is in creased in en ergy because
6
th e lowest n ,π* state is less polar th an th e groun d state .
O H
1. hν
. ISC * R
O H
3
H
H
1
2
R
CH3
CH3
kd
1
2
kH
S
k-H
k-
H
S
S
H
H
O H
O H
k-S
HO H
H
R
R
CH3
3 (solvated)
kE + kC
R
CH3
CH3
kS
1
(solvated)
3
S
H
S
S
1
2
. hν
k
kE + kC
kd
. ISC
S
3
*
H
R
O
H
CH3
O H
H
R
R
+
CH3
HO
2
(solvated)
4
5
SCHEME 2
We m easured th e solven t sh ifts in water an d eth an ol for keton es 1a an d
b as well as for 1-(4-m eth oxyph en yl)pen tan -1-on e (4-m eth oxyvalero-
1
ph en on e) as a fun ction of tem perature. Th e 4-m eth oxy derivative displayed
th e largest effect an d so it is in cluded to th is discussion even if it was n ot
used in th e ph otoch em istry experim en ts. Th e UV spectrum of 4-m eth oxy-
valeroph en on e in water sh owed a blue sh ift wh en goin g from 20 to 80 °C.
–
1
Th e observed 4 n m sh ift correspon ds to 2.1 kcal m ol (valeroph en on e
1
S– L tran sition in water is lowered by approxim ately 8 n m com pared with
a
n on polar solven ts), wh ich m ay cause larger n ,π* population . Th is m ay be
explain ed by th e weaken in g (but n ot breakin g of) th e h ydrogen bon din g
with th e carbon yl group resultin g in an in crease of th e h ydrogen abstrac-
tion rate con stan t. 4-Meth ylvaleroph en on e exh ibited on ly 2 n m sh ift an d
th ere was practically n o ch an ge in absorption spectrum of 1a, obviously
th an ks to th e lack of a stron gly electron don atin g group in con jugation
with th e carbon yl group. We observed th e sam e effect also in eth an ol but
m uch weaker. No tem perature depen den t sh ift of th e n ,π* tran sition ban d
in all com poun ds h ave been detected. Th ose ch aracteristics would h ave lit-
tle effect on th e data sh owed in Tables II an d III, h owever, such tem pera-
Collect. Czech. Chem. Commun. (Vol. 64) (1999)

Solvent Effects in Photochemistry
2013
ture depen den ce expresses tren ds furth er discussed in th e followin g
ch apter.
Biradical Solvation
Earlier studies sh owed th at th e Type II cleavage an d cyclization quan tum
yields reach n early un ity in water or oth er h igh ly polar solven ts, in con trast
1
,6,13
to m uch lower quan tum yields in n on polar organ ic solven ts
quan tum yield in water was foun d essen tially tem perature- an d wave-
. Th e
6
,13
len gth -in depen den t wh en th e n ,π* absorption ban d was irradiated
. Basic
polar solven ts slow th e disproportion ation of biradicals (k–H, Sch em e 2)
wh ich m ean s th at en h an cin g biradical lifetim e raises th e overall quan tum
yield. Sch em e 2 depicts a differen t reaction path way of eith er n ot solvated
or solvated biradical (3). We assum e an equilibrium between th ose in term e-
diates, wh ich certain ly depen ds on both solven t an d biradical properties.
Quan tum yields of th e Type II reaction of alkyl ph en yl keton es (Sch em e 2)
1
are given by th e followin g equation s :
TABLE III
a
Effect of addition of a base on th e Type II reaction of 1a an d 1b m ixtures at wavelen gth
>
313 n m
F/Cb
T, °C
Solven t
Rc
1
a
1b
2
0
ben zen e + pyridin ed
eth an ol + pyridin ed
aceton itrile +
7.02
5.47
8.06
6.38
1.21
1.19
boil
2
0
6.90
6.81
7.90
7.18
1.62
1.44
boil
2
0
5.12
4.96
6.04
6.09
1.34
1.23
d
pyridin e
boil
2
0
aceton itrile + watere
4.32
4.45
4.41
5.41
1.92
1.25
boil
a
–1
Th e keton e con cen tration s were 0.01 m ol l . Th e reproducibility for each m easurem en t
b
Th e Typ e II fragm en tation / cyclization ratio for both 1a an d 1b . c Ph oto-
was ±5%.
d
ch em ical efficien cy rat io s R = ([4a ] + [5a ]/ [4b ] + [5b ]); n ot corrected on ε valu es. 0.5 M
Collect. Czech. Chem. Commun. (Vol. 64) (1999)

2
014
Klán, Literák:
Φ = Φ
Φ Pp
(1)
II
ISC
R
ΦISC = 1 (alkyl ph en yl keton es)
(2)
(3)
(4)
Φ = k /(k + k )
R
H
H
d
P = (k + k )/(k + k + k ) = (k + k ) τ = k τ ,
BR
p
E
C
E
C
–H
E
C
BR
P
wh ere Φ , ΦISC, an d ΦR are quan tum yields of th e Type II reaction , th e
II
in tersystem crossin g from sin glet to triplet, an d th e product form ation , re-
spectively. P is th e probability th at th e biradical will collapse in to products
p
an d τBR is th e biradical lifetim e. Th e value k refers to th e excited state de-
d
cay oth er th an h ydrogen abstraction . Sim ilarly, th e probability for th e
solvated biradical can be expressed by Eq. (5).
S
S
E
S
C
S
E
S
C
S
–H
Pp = k + k
k + k + k
(5)
(
) (
)
Th us, our goal was to add som e m ore in form ation to th is well-kn own
ph otoch em ical subject to reveal h ow p-m eth yl substitution in valero-
ph en on e affects tem perature depen den t solvation . We h ave n ot m easured
quan tum yields, h owever, em ploym en t of a m ixture of two valeroph en on es
(
1a, 1b), wh ere sam e ph otoch em ical con dition for both com poun ds were
an ticipated, brough t an in terestin g in sigh t in to th eir ph otoreactivity.
Wh en we assum e th at triplet en ergy tran sfer is n egligible at 0.01 M con cen -
8
1
tration an d kH >> k (ref. ) th e quan tum yield ratio for th ose two keton es
d
can be expressed usin g Eqs (1)–(4) as:
Φ (1a)/Φ (1b) = P (1a)/P (1b) = [k (1a) τ (1b)]/[k (1a) τBR(1b)] ≈
II
II
p
p
P
BR
P
≈
([4a] + [5a]/[4b] + [5b]) = R ,
(6)
wh ere R is th e con version (ph otoch em ical efficien cy) ratio proportion al to
th e quan tum yield ratio.
Collect. Czech. Chem. Commun. (Vol. 64) (1999)

Solvent Effects in Photochemistry
2015
Table II sh ows tem perature studies of th e m ixture in various solven ts. In
all exam ples th e R ratio dropped wh en tem perature was raised. Th e differ-
en ce was foun d to be quite sm all in ben zen e but large in polar aceton itrile
an d eth an ol. Wavelen gth was foun d to h ave on ly n egligible effect on R. All
solven ts were purposely ch osen to boil at tem perature very close to 80 °C
for easier com parison (R20 is assign ed for th e ph otoch em istry at 20 °C an d
5
4
R80 at 80 °C). Wagn er as well as Scaian o in th eir tem perature studies rea-
son ed th at in p-substituted ph en yl keton es with an electron -don atin g
group, upper (reactive) n ,π* triplet is reach ed from th e lower-lyin g π,π* level
an d th e gap between both states is reflected as an in crease in th e activation
en ergy. In oth er words, n ,π* triplets are th erm ally populated from lower
π,π* levels. Th e fact th at R20 was always larger th an R80 in our results is th en
in accord with th eir results.
Th e ratio R /R , for a correspon din g solven t, con cen tration , an d wave-
2
0
80
len gth , is defin ed in Eq. (7):
R /R ≈ [Φ (1a)/Φ (1b)] /[Φ (1a)/Φ (1b)] ,
80
(7)
2
0
80
II
II
20
II
II
in wh ich R /R ratio, calculated from Eq. (6), is proportion al to th e quan -
2
0
80
tum yield ratio. Th e startin g keton es (1a an d 1b) did n ot h ave sam e m olar
absorption coefficien ts (ε) in all solven ts used (Table I) but th ese coeffi-
cien ts were tem perature-in depen den t. Sin ce it was difficult to obtain pre-
cise coefficien ts in th e m ixtures with pyridin e an d th e calculation s usin g
Eq. (7) are n ot affected by th e correction s on ε, on ly un corrected values R
are listed in Tables II an d III.
Table II clearly depicts th at th e R /R ratio is h igh est in polar aceto-
2
0
80
n itrile, lower in eth an ol, but n egligible in ben zen e (we obtain ed sim ilar re-
sults in cycloh exan e; n ot sh own ). In addition , Table III lists results from ex-
perim en ts in wh ich m ore basic com poun ds (pyridin e, water) were added
in to keton e m ixtures, wh ich caused dram atic ch an ges in values of th e ratio.
Th e ratios R /R , also displayed in Fig. 1, reflect a relation of th e ratio
2
0
80
m agn itude with basicity of th e solven t used (pyridin e, pKaBH
pKaBH
= –1.7; eth an ol, pKaBH = –1.9; aceton itrile, pKaBH = –12.0). It is obvi-
ous from th e Eq. (6) th at k an d τ are th e on ly relevan t values th at can af-
+
= 5.2; water,
+
+
+
p
BR
fect th e ratio. Scaian o m easured biradical lifetim es in various solven ts an d
3
c
foun d th at in creasin g solven t polarity results in lon ger τBR (ref. ). It is also
1
,13
kn own th at m eth yl substitution in ph en yl keton es en h an ces P (refs ).
p
Collect. Czech. Chem. Commun. (Vol. 64) (1999)

2
016
Klán, Literák:
We assum e th at m eth yl substitution in 1b en h an ces P but OH group in
p
th e biradical is less acidic th an th at in 1a; th us, th e h ydrogen bon din g to a
base is weaker. Sch em e 3 sh ows a rin g-substituen t effect in th e biradical,
1
h
which is best visualized by the valence bond representation . An electron-
don atin g group defin itely lowers th e probability of th e radical-cation form .
Steen ken m easured basicity of various radical cation s wh ich was foun d dra-
m atically h igh er th an in th at of n on -ion ized com poun ds (e.g., pK of
a
+
•
[
Ph –OH] is ≈2, wh ich is about 8 orders of m agn itude h igh er th an th at of
1
4
ph en ol) .
OH
OH
R
R
SCHEME 3
Our data suggest th at th e h ydrogen bon din g between th e biradical OH
group an d th e solven t is very specific an d tem perature depen den t. We pro-
pose th e followin g:
1
. High values of R /R ratios reflect weaker in teraction s of th e 1b
20 80
biradical (biradical form ed from 1a is m ore acidic) with weaker bases (eth a-
n ol, aceton itrile) at 20 °C but in teraction s in both derivatives van ish at
1
b
8
0 °C. Th us, weaker biradical in teraction s m ean suppression of P (ref. )
p
(
larger k–H) resultin g in a larger R /R ratio (Eq. (6)).
20 80
1
.8
.6
R
/R
2
0
80
1
1
1
1
.4
.2
.0
benzene
benzene
pyridine
EtOH
EtOH
+ pyridine
MeCN
MeCN
+ water
MeCN
+ pyridine
+
FIG. 1
Tem perature-depen den t solven t effect on ph otoch em istry of 1-ph en ylpen tan -1-on es at 313 n m .
Th e R20/R80 ratios were calculated from Tables II an d III
Collect. Czech. Chem. Commun. (Vol. 64) (1999)

Solvent Effects in Photochemistry
2017
2
. Th e R /R ratio is close to un ity wh en a m uch stron ger base –
2
0
80
pyridin e – was used: it in teracts eith er n ot selectively (pyridin e is a too
stron g base) or both keton e biradicals retain h ydrogen bon din g even at
8
0 °C. Such observation resem bles th e beh avior of keton es in n on polar sol-
ven ts, i.e., in th e absen ce of biradical solvation .
Th us, we probably observed an un expected beh aviour of th e Type II
biradical in wh ich such a sm all structure ch an ge, as th e p-m eth yl substitu-
tion , alter solven t effect in presen ce of weak bases. A weak in teraction (h y-
drogen bon din g) between solven t an d th e Type II in term ediate is obviously
reduced at a h igh er tem perature. Th is is an exam ple of h ydrogen -bon din g
weaken in g with in creasin g tem perature sim ilar to th ose reported in th e pre-
1
5
vious part of th e discussion as well as in th e literature . Th e selectivity was
lost wh en a stron ger base was used.
Th e fact th at th e fragm en tation /cyclization (F/C) ratio often in creased
with tem perature (Table II) probably reflects ch an ged con form ation al fac-
tors in th e course of reaction .
We thank A. and I. Bader for funding through the Foundation for a Civil Society.
REFERENCES
1
. a) Wagner P. J.: Acc. Chem. Res. 1971, 4, 168; b) Wagner P. J., ParkB.-S. in: Organic
Photochemistry (A. Padwa, Ed.), Vol. 11, p. 227. Dekker, New York 1991; and references
therein; c) Wagner P. J., Hammond G. S.: J. Am. Chem. Soc. 1966, 88, 1245; d) Wagner
P. J.: Acc. Chem. Res. 1983, 16, 461; e) Zimmt M. B., Doubleday C. E., Jr., Turro N. J.:
Chem. Phys. Lett. 1987, 134, 549; f) Wagner P. J.: Acc. Chem. Res. 1989, 22, 83; g)
Wagner P. J., Kelso P. A., Kemppainen A. E., McGrath J. M., Schott H. N., Zepp R. G.: J.
Am. Chem. Soc. 1972, 94, 7506; h) Wagner P. J., Kemppainen A. E., Schott H. N.: J. Am.
Chem. Soc. 1973, 95, 5604; i) Scaiano J. C.: Acc. Chem. Res. 1982, 15, 252.
2
3
4
. Norrish R. G. W., Appleyard M. E. S.: J. Chem. Soc. 1934, 874.
. Yang N. C., Yang D. H.: J. Am. Chem. Soc. 1958, 80, 2913.
. Encina M. V., Lissi E. A., Lemp E., Zanocco A., Scaiano J. C.: J. Am. Chem. Soc. 1983,
1
05, 1856.
5. Wagner P. J., Nakahira T.: J. Am. Chem. Soc. 1973, 95, 8474.
6. Zepp R. G., Gumz M. M., Miller W. L., Gao H.: J. Phys. Chem. A 1998, 102, 5716.
7. Klán P., LiterákJ., HájekM.: J. Photochem. Photobiol., A, in press.
8. a) Klán P., Wagner P. J.: J. Am. Chem. Soc. 1998, 120, 2198; b) Wagner P. J., Klán P.:
J. Am. Chem. Soc., in press.
9
. Wagner P. J., Thomas M. J., Harris E.: J. Am. Chem. Soc. 1976, 98, 7675.
1
1
1
0. Zimmerman H. E.: Mol. Photochem. 1971, 3, 281.
1. LaCount R. B., Griffin C. E.: Tetrahedron Lett. 1965, 21, 1549.
2. a) Wagner P. J., Kemppainen A. E., Schott H. N.: J. Am. Chem. Soc. 1973, 95, 1856;
b) Lamola A. A.: J. Chem. Phys. 1967, 47, 4810.
Collect. Czech. Chem. Commun. (Vol. 64) (1999)

2
018
Klán, Literák:
1
1
1
3. a) Doubleday C. E., Jr., Turro N. J., Wang J. F.: Acc. Chem. Res. 1989, 22, 199; b) Scaiano
J. C.: Tetrahedron 1982, 38, 819.
4. a) Steenken S.: Free Radical Res. Commun. 1992, 16, 349; b) Steenken S.: Private
communication.
5. a) Dougherty R. C.: J. Chem. Phys. 1998, 109, 7372; b) Dorsett H. E., Watts R. O.,
Xantheras S. S.: J. Phys. Chem. A 1999, 103, 3351; c) CzeslikC., Jonas J.: Chem. Phys.
Lett. 1999, 302, 633.
Collect. Czech. Chem. Commun. (Vol. 64) (1999)