Influence of Functional Groups on the Ene Reaction of Singlet Oxygen with 1,4-Cyclohexadienes?

Article abstract of DOI:10.1111/php.13422

The photooxygenation of 1,4-cyclohexadienes has been studied with a special focus on regio- and stereoselectivities. In all examples, only the methyl-substituted double bond undergoes an ene reaction with singlet oxygen, to afford hydroperoxides in moderate to good yields. We explain the high regioselectivities by a “large-group effect” of the adjacent quaternary stereocenter. Nitriles decrease the reactivity of singlet oxygen, presumably by quenching, but can stabilize proposed per-epoxide intermediates by polar interactions resulting in different stereoselectivities. Spiro lactams and lactones show an interesting effect on regio- and stereoselectivities of the ene reactions. Thus, singlet oxygen attacks the double bond preferentially anti to the carbonyl group, affording only one regioisomeric hydroperoxide. If the reaction occurs from the opposite face, the other regioisomer is exclusively formed by severe electrostatic repulsion in a perepoxide intermediate. We explain this unusual behavior by the fixed geometry of spiro compounds and call it a “spiro effect” in singlet oxygen ene reactions.

Full text of DOI:10.1111/php.13422

Photochemistry and Photobiology, 20**, **: *–*
Special Issue Research Article
Influence of Functional Groups on the Ene Reaction of Singlet Oxygen
with 1,4-Cyclohexadienes†
€
Werner Fudickar, Melanie Metz, Yasemin Mai-Linde, Tobias Kruger, Alexandra Kelling, Eric Sperlich and
Torsten Linker*
Department of Chemistry, University of Potsdam, Golm, GermanyOpen access funding enabled and organized by
ProjektDEAL.
Received 22 February 2021, accepted 23 March 2021, DOI: 10.1111/php.13422
ABSTRACT
H
O
O
(A)
The photooxygenation of 1,4-cyclohexadienes has been studied
with a special focus on regio- and stereoselectivities. In all
examples, only the methyl-substituted double bond undergoes
an ene reaction with singlet oxygen, to afford hydroperoxides
in moderate to good yields. We explain the high regioselectivi-
ties by a “large-group effect” of the adjacent quaternary stere-
ocenter. Nitriles decrease the reactivity of singlet oxygen,
presumably by quenching, but can stabilize proposed per-
epoxide intermediates by polar interactions resulting in differ-
ent stereoselectivities. Spiro lactams and lactones show an
interesting effect on regio- and stereoselectivities of the ene
reactions. Thus, singlet oxygen attacks the double bond prefer-
entially anti to the carbonyl group, affording only one regioiso-
meric hydroperoxide. If the reaction occurs from the opposite
face, the other regioisomer is exclusively formed by severe elec-
trostatic repulsion in a perepoxide intermediate. We explain
this unusual behavior by the fixed geometry of spiro com-
pounds and call it a “spiro effect” in singlet oxygen ene reac-
tions.
H
ene
1O2
(B)
(C)
[4 + 2]
[2 + 2]
O
O
D
D
O
O
Scheme 1. Reaction pathways of singlet oxygen with p-systems.
with p-systems are the [4 + 2] cycloaddition (A) and the ene
reaction (B), the [2 + 2] cycloaddition (C) is only observed with
electron-rich alkenes (Scheme 1).
A perfect overview of such chemical reactions of singlet oxy-
gen is given by Ed Clennan in chapter 18 of the book on 1O2
(1). Ed is one of the pioneers in singlet oxygen chemistry and
contributed to various mechanistic and synthetic aspects in many
reviews. His research interests range from [4 + 2] cycloadditions
(8), over ene reactions (9,10), to persulfoxides (11), and more
recently zeolites (12).
Our group became interested in singlet oxygen chemistry
when we investigated the photooxygenation of 1,4-cyclohexadi-
enes 1a (R1 = Me, Et, iPr; R2 = CO2H), which are easily avail-
able by Birch reduction on a large scale (13–15). They are
suitable substrates to examine site-, regio- and stereoselectivities
of 1O2 reactions in one system, because two double bonds can
be attacked from different sides and faces and 16 isomeric prod-
ucts might be formed. However, irradiation with visible light and
tetraphenylporphyrin (TPP) as sensitizer afforded hydroperoxides
2 by an ene reaction as sole regioisomer with high stereoselectiv-
ities (Scheme 2a) (16). This unusual behavior was explained by
electronic interactions in initially formed perepoxides. Although
the existence of such labile intermediates was controversially dis-
cussed by theoretical (17,18) and experimental (19) investiga-
tions, they were trapped by a procedure developed by Ed (20).
INTRODUCTION
Singlet oxygen (1O2) is an important oxidant with applications in
chemistry, biology and medicine (1). It can be generated chemi-
cally from hydrogen peroxide and metal hypochlorites, which
was already described by Foote (2,3). More recently, the cleav-
age of organic peroxides offered a milder method (4). However,
the irradiation of molecular oxygen with visible light in the pres-
ence of a sensitizer is still the most convenient method for 1O2
production (5,6). This reactive and short-lived oxygen species
can oxidize various organic compounds under such conditions
(photooxygenation), which was applied in the synthesis of sev-
eral natural products (7). The main reaction pathways of 1O2
*Corresponding author email: linker@uni-potsdam.de (Torsten Linker)
†This article is part of a Special Issue celebrating the career of Dr. Edward Clen-
nan.
© 2021 The Authors. Photochemistry and Photobiology published by Wiley Peri-
odicals LLC on behalf of American Society for Photobiology.
This is an open access article under the terms of the Creative Commons Attribution
License, which permits use, distribution and reproduction in any medium, provided
the original work is properly cited.
1

2
Werner Fudickar et al.
R1
R2
CH3
R1
R2
CH2
O2, TPP, h
a)
OOH
1a: R1 = Me, Et, iPr; R2 = CO2H
2a, 2b
1b: R1 = Me, Et, iPr; R2 = CO2 ,CO2Me, CH2OH 1 regioisomer; de up to > 97%
X
R
NC
O
1O2
1O2
CH3
b)
?
H3C
1c f: R = CO2H, CO2 , CO2Me, CH2OH
1g l: X = O, NH
Scheme 2. Reactions of 1,4-cyclohexadienes 1 with singlet oxygen; a) previous work (16,21); b) this work.
125.6 (d, C-3), 128.3 (s, C-2), 129.0 (d, C-5), 172.6 (s, CO2). IR (film):
v = 3377, 2954, 1727, 1669, 1434, 1237, 1032, 962, 742 cmꢀ1. HRMS
(GC-TOF): m/z Calcd for C11H14NO2 (M + H+): 192.1025; Found:
192.1018.
Synthesis of c-spirolactones 1j –l. Arene carboxylic acid 7 (50 mmol)
was introduced into a three-necked flask (500 mL), equipped with a dry-
ice condenser and cooled to – 78°C by a dry-ice acetone bath. Ammonia
(250 mL) was condensed into the three-necked flask, and lithium
(800 mg, 115 mmol) was added in small pieces to the solution at – 78°C,
until it remained blue and the solution was stirred at – 78°C for 1 h.
Ethylene oxide (10 mL) was condensed into a dropping funnel, which
was cooled by a dry-ice acetone bath, and added within 5 min. The
ammonia was allowed to evaporate overnight at RT. The solid residue
was dissolved in water (100 mL), cooled to 0°C, acidified with 6 M HCl
Later on, we investigated the influence of esters and alcohols on
the photooxygenation of 1,4-cyclohexadienes 1b (R1 = Me, Et,
iPr; R2 = CO2Me, CH2OH) (21). However, the methyl group
was only located in the 2-position of the starting materials (o-
toluic acid as precursor of the Birch reductions), and R1 was
always an alkyl group. Herein, we describe the influence of other
functional groups on the ene reaction of singlet oxygen with 1,4-
cyclohexadienes, with a special focus on nitriles (1c–f) and spiro
compounds (1g–l) (Scheme 2b).
MATERIALS AND METHODS
to pH
1 and extracted with dichloromethane (3 9 100 mL). The
Chemicals and instrumentation. All starting materials were used as
purchased without further purification. Nitrile 1c was synthesized by
Birch reduction according to our previously published procedure (22).
Alcohol 1f and lactams 1g–i were synthesized according to our very
recently published procedure (23). Melting points were determined by
using a Mel-Temp from Electrothermal. TLC was performed using TLC
Silica gel 60 F254 aluminum sheets from Merck. 1H NMR and 13C
NMR spectra were measured by using a Bruker NEO 400 (400 MHz,
100 MHz), NEO 500 (500 MHz, 125 MHz), or a Bruker Avance 600
(600 MHz, 150 MHz) NMR spectrometer. Signals were assigned by two-
dimensional methods (HSQC). The IR spectra were recorded in film by
using a Nicolet Avatar 370 FT-IR spectrometer from Thermo Electron
Corporation. Elemental analysis was performed on a Vario EL III
elemental analyzer. HRMS spectra were measured at a GC-MS Trace
DSX II spectrometer.
combined organic layers were dried over magnesium sulfate, and the
solvent was removed in vacuo. The crude product was dissolved in dry
pyridine (100 mL) and cooled to 0°C. Thionyl chloride (5 mL, 70 mmol)
was added at this temperature within 10 min, and the mixture was stirred
at room temperature for 20 h. The reagents were removed in vacuo, water
(100 mL) was added and the mixture was extracted with dichloromethane
(3 9 80 mL). The combined organic phases were washed with brine
(50 mL), dried over magnesium sulfate and concentrated in vacuo. The
crude products were purified by column chromatography (MeOtBu/
hexane = 1:1) to obtain the lactones 1g–j in pure form.
o-Lactone 1j. Colorless oil (7.5 g, 91%). Rf = 0.39 (MeOtBu/hex-
ane = 1:1). 1H NMR (500 MHz, CDCl3): d = 1.71 (quart., J = 1.7 Hz,
3H, 7-H), 2.11 (ddd, J = 13.1, 7.5, 4.7 Hz 1H, 8-H), 2.55 (dt, J = 13.1,
8.5 Hz 1H, 8´-H), 2.65 (dddquint., J = 23.3, 3.5, 2.1, 1.7 Hz, 1H, 4-H),
2.77 (dddquint., J = 23.3, 3.5, 2.1, 1.7 Hz, 1H, 4´-H), 4.34 (ddd,
J = 9.2, 8.5, 4.7 Hz 1H, 9-H), 4.38 (ddd, J = 9.2, 8.5, 7.5 Hz 1H, 9´-H),
5.61 (dt, J = 9.9, 2.1 Hz, 1H, 6-H), 5.71 (sext.d, J = 1.7, 1.3 Hz, 1H, 3-
H), 5.92 (dtd, J = 9.9, 3.5, 1.3 Hz, 1H, 5-H),. 13C{1H} NMR (125 MHz,
CDCl3): d = 19.5 (q, C-7), 26.8 (t, C-4), 34.8 (t, C-8), 48.5 (s, C-1),
65.6 (t, C-9), 124.2 (d, C-3), 124.6 (d, C-6), 126.4 (d, C-5), 129.5 (s, C-
2), 178.2 (s, CO2). IR (film): v = 2972, 2870, 1760, 1452, 1371, 1153,
1022, 703 cmꢀ1. HRMS (GC-TOF): m/z Calcd for C10H13O2 (M + H+):
165.0916; Found: 165.0912. Anal. Calcd for C10H12O2 (164.20): C,
73.15; H, 7.37. Found: C, 73.11; H, 7.31.
Synthesis of ester 1e. Powdered potassium hydroxide (280 mg,
5.0 mmol) was suspended in dry DMSO (15 mL) under nitrogen
atmosphere. A solution of acid 1c (890 mg, 5.0 mmol) in dry DMSO
(8 mL) was added dropwise within 30 min, and the solution was cooled
to 0°C. Iodomethane (1.2 mL, 19 mmol) was added within 5 min at this
temperature, and the mixture was stirred for 18 h at room temperature.
Water (150 mL) was added, and the mixture was extracted with
dichloromethane (3 9 100 mL). The combined organic phases were
washed with a dil. solution of sodium thiosulfate (50 mL), dried over
magnesium sulfate and concentrated in vacuo affording 850 mg (89%) of
the pure ester 1e as a colorless oil.
m-Lactone 1k. White solid (7.5 g, 91%). Rf = 0.36 (MeOtBu/hex-
ane = 1:1). Mp = 55–56°C. 1H NMR (500 MHz, CDCl3): d = 1.74 (d,
J = 1.6 Hz, 3H, 7-H), 2.21 (dt, J = 12.8, 7.0 Hz 1H, 8-H), 2.24 (dt,
J = 12.8, 7.0 Hz 1H, 8´-H), 2.56 (dm., J = 22.7 Hz, 1H, 4-H), 2.70 (dm,
J = 22.7 Hz, 1H, 4´-H), 4.34 (t, J = 7.0 Hz, 2H, 9-H), 5.30 (quart.,
J = 1.6 Hz, 1H, 2-H), 5.57 (dq, J = 9.9, 2.1 Hz, 1H, 6-H), 5.97 (dt,
J = 9.9, 3.4 Hz, 1H, 5-H),. 13C{1H} NMR (125 MHz, CDCl3): d = 23.2
(q, C-7), 30.8 (t, C-4), 37.8 (t, C-8), 46.5 (s, C-1), 65.3 (t, C-9), 118.8
(d, C-2), 124.1 (d, C-6), 127.6 (d, C-5), 135.2 (s, C-3), 178.5 (s, CO2).
Rf = 0.49 (MeOtBu/hexane = 1:1). 1H NMR (500 MHz, CDCl3):
d = 1.67 (quart., J = 1.7 Hz, 3H, 7-H), 2.75 (dddquint., J = 23.3, 3.4,
2.1, 1.7 Hz, 1H, 4-H), 2.81 (d, J = 16.9 Hz 1H, 8-H), 2.83 (dddquint.,
J = 23.3, 3.4, 2.1, 1.7 Hz, 1H, 4´-H), 2.87 (d, J = 16.9 Hz 1H, 8´-H),
3.71 (s, 3H, OMe), 5.57 (dt, J = 9.8, 2.1 Hz, 1H, 6-H), 5.83 (sext.d,
J = 1.7, 1.3 Hz, 1H, 3-H), 6.10 (dtd, J = 9.8, 3.4, 1.3 Hz, 1H, 5-H),.
13C{1H} NMR (125 MHz, CDCl3): d = 19.2 (q, C-7), 25.8 (t, C-8), 27.0
(t, C-4), 49.7 (s, C-1), 53.1 (q, OMe), 117.2 (s, CN), 124.3 (d, C-6),
IR (KBr): v = 3668, 2975, 2910, 1755, 1373, 1152, 1020, 717 cmꢀ1
.

Photochemistry and Photobiology
3
HRMS (GC-TOF): m/z Calcd for C10H13O2 (M + H+): 165.0916; Found:
165.0916. Anal. Calcd for C10H12O2 (164.20): C, 73.15; H, 7.37. Found:
C, 73.06; H, 7.67.
Methyl ester hydroperoxide regio-3e. Colorless oil (15 mg, 7%).
Rf = 0.35 (MeOtBu/hexane = 1:1). 1H NMR (400 MHz, CDCl3):
d = 1.45 (s, 3H, 7-H), 2.96 (d, J = 16.4 Hz, 1H, 8-H), 3.18 (d,
J = 16.4 Hz, 1H, 8´-H), 3.86 (s, 3H, OMe), 5.64 (dm, J = 9.2 Hz, 1H,
6-H), 6.04–6.14 (m, 3H, 3-H, 4-H, 5-H), 8.09 (br, 1H, OOH). 13C{1H}
NMR (150 MHz, CDCl3) d = 20.1 (q, C-7), 29.6 (t, C-8), 51.8 (s, C-1),
53.2 (q, OMe), 85.5 (s, C-2), 118.3 (s, CN), 124.0, 128.7, 129.1, 132.0
(4 d, C-3, C-4, C-5, C-6), 172.6 (s, CO2). IR (film): v = 3374, 2953,
2255, 1732, 1231, 1072, 927, 734 cmꢀ1. HRMS (GC-TOF): m/z Calcd
for C11H14NO4 (M + H+): 224.0923; Found: 224.0908.
Alcohol hydroperoxide cis-2f (main isomer). Colorless oil (110 mg,
56%). Rf = 0.11 (MeOtBu/hexane = 1:1). 1H NMR (500 MHz, CDCl3):
d = 2.19 (br, 1H, OH), 2.42 (dddd, J = 18.9, 5.3, 2.4, 1.1 Hz, 1H, 4-H),
2.59 (dddd, J = 18.9, 5.0, 2.7, 2.2 Hz, 1H, 4´-H), 2.60 (d, J = 16.6 Hz,
1H, 8-H), 2.82 (d, J = 16.6 Hz, 1H, 8´-H), 3.53 (dd, J = 10.7, 8.4 Hz,
1H, 9-H), 4.81 (d, J = 10.7 Hz, 1H, 9´-H), 4.71 (dd, J = 5.0, 2.4 Hz,
1H, 3-H), 5.31 (s, 1H, 7-H), 5.51 (ddd, J = 10.0, 2.7, 1.7 Hz, 1H, 6-H),
5.59 (s, 1H, 7´-H), 5.96 (ddd, J = 10.0, 5.3, 2.2 Hz, 1H, 5-H), 8.57 (br,
1H, OOH). 13C{1H} NMR (125 MHz, CDCl3): d = 25.8 (t, C-8), 29.3
(t, C-4), 45.1 (s, C-1), 68.6 (t, C-9), 84.1 (d, C-3), 117.6 (s, CN), 120.0
(t, C-7), 127.1 (d, C-6), 128.4 (t, C-5), 141.5 (s, C-2). IR (film):
v = 3371, 2929, 2254, 1660, 1420, 1026, 923, 741 cmꢀ1. HRMS (GC-
TOF): m/z Calcd for C10H14NO3 (M + H+): 196.0974; Found: 196.0976.
Alcohol hydroperoxide trans-2f (minor isomer). Colorless oil (45 mg,
23%). Rf = 0.10 (MeOtBu/hexane = 1:1). 1H NMR (400 MHz, CDCl3):
d = 1.72 (br, 1H, OH), 2.40 (dddd, J = 17.8, 5.5, 4.1, 1.9 Hz, 1H, 4-H),
2.56 (dddd, J = 17.8, 5.2, 3.9, 2.2 Hz, 1H, 4´-H), 2.75 (s, 2H, 8-H), 3.71
(s, 2H, 9-H), 4.69 (dd, J = 5.5, 5.2 Hz, 1H, 3-H), 5.29 (s, 1H, 7-H),
5.55 (s, 1H, 7´-H), 5.58 (ddd, J = 10.0, 2.2, 1.9 Hz, 1H, 6-H), 5.94
(ddd, J = 10.0, 4.1, 3.9 Hz, 1H, 5-H), 8.13 (br, 1H, OOH). 13C{1H}
NMR (100 MHz, CDCl3): d = 25.5 (t, C-8), 30.1 (t, C-4), 45.0 (s, C-1),
66.9 (t, C-9), 82.9 (d, C-3), 115.1 (t, C-7), 118.0 (s, CN), 127.6 (d, C-5),
128.0 (t, C-6), 142.5 (s, C-2). IR (film): v = 3372, 2970, 2936, 2259,
1738, 1366, 1025, 920, 741 cmꢀ1. HRMS (GC-TOF): m/z Calcd for
C10H1NO3Na (M+Na+): 218.0793; Found: 218.0791.
o-Lactam hydroperoxide 2g (main isomer). White solid (130 mg,
67%). Rf = 0.14 (EtOAc). Mp = 134–135°C. 1H NMR (500 MHz,
CD3OD): d = 2.18 (dt, J = 13.0, 8.3 Hz, 1H, 8-H), 2.39 (dddd,
J = 18.1, 4.0, 3.0, 1.6 Hz, 1H, 4-H), 2.52 (dddd, J = 18.1, 5.1, 4.7,
2.4 Hz, 1H, 4´-H), 2.54 (ddd, J = 13.0, 6.2, 3.7 Hz, 1H, 8´-H), 3.34
(ddd, J = 9.8, 8.3, 6.2 Hz, 1H, 9-H), 3.37 (ddd, J = 9.8, 8.3, 3.7 Hz,
1H, 9´-H), 4.72 (dd, J = 5.1, 4.0 Hz, 1H, 3-H), 5.01 (s, 1H, 7-H), 5.30
(s, 1H, 7´-H), 5.45 (ddd, J = 9.9, 2.4, 1.6 Hz, 1H, 6-H), 5.79 (ddd,
J = 9.9, 4.7, 3.0 Hz, 1H, 5-H), 6.00 (br, 1H, NH), 8.02 (br, 1H, OOH).
13C{1H} NMR (125 MHz, CD3OD): d = 29.9 (t, C-4), 35.3 (t, C-8),
38.9 (t, C-9), 51.6 (s, C-1), 82.8 (d, C-3), 111.8 (t, C-7), 125.2 (d, C-5),
128.4 (d, C-6), 143.7 (s, C-2), 180.2 (s, CONH). IR (film): v = 3668,
3246, 2987, 2903, 1677, 1412, 1285, 1059, 904, 700, 671 cmꢀ1. HRMS
(GC-TOF): m/z Calcd for C10H14NO3 (M + H+): 196.0974; Found:
196.0972. Anal. Calcd for C10H13NO3 (195.22): C, 61.53; H, 6.71; N,
7.17. Found: C, 61.24; H, 6.81; N, 7.13.
o-Lactam hydroperoxide 3g (minor isomer). Colorless oil (35 mg,
18%). Rf = 0.33 (EtOAc). 1H NMR (400 MHz, CDCl3): d = 1.47 (s, 3H,
7-H), 2.15 (dt, J = 13.5, 5.8 Hz, 1H, 8-H), 2.86 (dt, J = 13.5, 8.1 Hz,
1H, 8´-H), 3.44 (dd, J = 8.1, 5.8 Hz, 2H, 9-H), 5.61 (d, J = 9.6 Hz, 1H,
3-H), 5.81 (d, J = 9.4 Hz, 1H, 6-H), 6.01 (dd, J = 9.4, 5.4 Hz, 1H, 5-
H), 6.25 (dd, J = 9.6, 5.4 Hz, 1H, 4-H), 6.54 (br, 1H, NH), 8.18 (br, 1H,
OOH). 13C{1H} NMR (100 MHz, CDCl3): d = 19.6 (q, C-7), 27.8 (t, C-
8), 39.3 (t, C-9), 52.3 (s, C-1), 83.2 (s, C-2), 122.6 (d, C-5), 126.7 (d, C-
4), 127.0 (d, C-3), 130.8 (d, C-6), 178.1 (s, CONH). IR (film): v = 3249,
2996, 2883, 1675, 1285, 1030, 904, 701 cmꢀ1. HRMS (GC-TOF): m/z
Calcd for C10H14NO3 (M + H+): 196.0974; Found: 196.0964.
2,3-Dimethyllactone 1l. Pale yellow oil (7.4 g, 83%). Rf = 0.43
(MeOtBu/hexane = 1:1). 1H NMR (600 MHz, CDCl3): d = 1.63, 1.68
(2s, each 3H, CH3), 2.03–2.08 (m, 1H, 8-H), 2.52–2.57 (m, 1H, 8´-H),
2.60 (dm., J = 22.8 Hz, 1H, 4-H), 2.70 (dm, J = 22.8 Hz, 1H, 4´-H),
4.31–4.38 (m, 2H, 9-H), 5.62 (dt, J = 9.9, 1.8 Hz, 1H, 6-H), 5.87 (dt,
J = 9.9, 3.2 Hz, 1H, 5-H),. 13C{1H} NMR (150 MHz, CDCl3):
d = 14.8, 19.1 (2q, 2 9 CH3), 32.8 (t, C-4), 35.1 (t, C-8), 50.2 (s, C-1),
65.5 (t, C-9), 121.9 (s, C-2), 124.4 (d, C-6), 126.5 (d, C-5), 129.2 (s, C-
3), 178.8 (s, CO2). IR (film): v = 3669, 2989, 2917, 1760, 1372, 1167,
1023, 708 cmꢀ1. HRMS (GC-TOF): m/z Calcd for C11H15O2 (M + H+):
179.1072; Found: 179.1073. Anal. Calcd for C11H14O2 (178.23): C,
74.13; H, 7.92. Found: C, 74.22; H, 7.97.
Photooxygenations. A solution of 1,4-cyclohexadiene 1 (1.0 mmol)
and sensitizer tetraphenylporphyrin (TPP, 3 mg) in CDCl3 (or Rose
Bengal (3 mg) in CD3OD for carboxylate 1d) (10 mL) was cooled to –
10°C, oxygen was slowly bubbled through the solution, and the mixture
was irradiated externally with two sodium lamps (1000 and 400 W) for
4–72 h. 1H NMR spectra (400 MHz) were directly measured from the
crude products to determine complete conversion, regioselectivities and
diastereomeric ratios (dr) (see Supporting Information for NMrs of crude
products). The solvent was removed in vacuo and the crude products
were directly purified by column chromatography (MeOtBu/hexane 1:1,
for polar compounds + 1% HOAc) and subsequent HPLC for difficult
separations, to obtain hydroperoxides 2 and 3 in pure form.
Acid hydroperoxide trans-2c. Colorless oil (155 mg, 74%). Rf = 0.16
(CHCl3 + 1% HOAc). 1H NMR (400 MHz, CDCl3): d = 2.51 (dtd,
J = 18.1, 4.5, 1.3 Hz, 1H, 4-H), 2.63 (dddd, J = 18.1, 4.5, 3.25 2.5 Hz,
1H, 4´-H), 2.94 (d, J = 16.6 Hz, 1H, 8-H), 3.01 (d, J = 16.6 Hz, 1H, 8´-
H), 4.80 (t, J = 4.5 Hz, 1H, 3-H), 5.41 (s, 1H, 7-H), 5.60 (s, 1H, 7´-H),
5.71 (ddd, J = 9.8, 2.5, 1.3 Hz, 1H, 6-H), 6.04 (ddd, J = 9.8, 4.5,
3.5 Hz, 1H, 5-H), 8.82 (br, 2H, OOH CO2H). 13C{1H} NMR (100 MHz,
CDCl3): d = 27.2 (t, C-8), 29.2 (t, C-4), 50.5 (s, C-1), 82.2 (d, C-3),
117.5 (s, CN), 118.6 (t, C-7), 124.9 (d, C-6), 128.0 (d, C-5), 140.9 (s, C-
2), 176.7 (s, CO2). IR (film): v = 3424, 3356, 3036, 2932, 2256, 1720,
1262, 1235, 925 cmꢀ1. HRMS (GC-TOF): m/z Calcd for C10H12NO4
(M + H+): 210.0766; Found: 210.0763.
o-Tolylacetonitrile side product. Colorless oil (40 mg, 30%).
Rf = 0.31 (CH2Cl2). 1H NMR (400 MHz, CDCl3): d = 2.37 (s, 3H,
CH3), 3.68 (s, 2H, CH2CN), 7.23–7.32 (m, 3H, H-Ar), 7.31–7.40 (m,
1H, H-Ar). 13C{1H} NMR (100 MHz, CDCl3): d = 19.3 (q, CH3), 21.9
(t, CH2CN), 117.6 (s, CN), 126.8, 128.4, 128.6, 130.7 (4d, C-Ar), 136.0,
136.1 (2s, C-Ar). IR (film): v = 2979, 2920, 2248, 1605, 1414, 1225,
. HRMS (GC-TOF): m/z Calcd for C9H10N
1049, 714 cmꢀ1 (M+):
132.0813; Found: 132.0812.
Methyl ester hydroperoxide trans-2e (main isomer). Colorless oil
(120 mg, 54%). Rf = 0.26 (MeOtBu/hexane = 1:1). 1H NMR (500 MHz,
CDCl3): d = 2.49 (dddd, J = 18.1, 4.8, 4.3, 1.4 Hz, 1H, 4-H), 2.60
(dddd, J = 18.1, 5.3, 3.2, 2.4 Hz, 1H, 4´-H), 2.95 (d, J = 16.6 Hz, 1H,
8-H), 3.00 (d, J = 16.6 Hz, 1H, 8´-H), 3.77 (s, 3H, OMe), 4.75 (dd,
J = 5.3, 4.3 Hz, 1H, 3-H), 5.26 (s, 1H, 7-H), 5.52 (s, 1H, 7´-H), 5.67
(ddd, J = 9.9, 2.4, 1.4 Hz, 1H, 6-H), 5.98 (ddd, J = 9.9, 4.8, 3.2 Hz,
1H, 5-H), 8.37 (br, 1H, OOH). 13C{1H} NMR (125 MHz, CDCl3):
d = 27.4 (t, C-8), 29.2 (t, C-4), 50.7 (s, C-1), 53.3 (q, OMe), 82.3 (d, C-
3), 117.7 (s, CN), 118.1 (t, C-7), 125.3 (d, C-6), 127.3 (d, C-5), 140.6 (s,
C-2), 172.3 (s, CO2). IR (film): v = 3665, 3380, 2982, 2907, 1731, 1252,
1072, 804, 741 cmꢀ1. HRMS (GC-TOF): m/z Calcd for C11H14NO4
(M + H+): 224.0923; Found: 224.0932.
Methyl ester hydroperoxide cis-2e (minor isomer). Colorless oil
(50 mg, 22%). Rf = 0.29 (MeOtBu/hexane = 1:1). 1H NMR (500 MHz,
CDCl3): d = 2.44 (ddt, J = 19.0, 5.1, 1.6 Hz, 1H, 4-H), 2.59 (dddd,
J = 19.0, 5.1, 2.8, 2.6 Hz, 1H, 4´-H), 2.98 (s, 2H, 8-H), 3.83 (s, 3H,
OMe), 4.69 (dd, J = 5.1, 1.6 Hz, 1H, 3-H), 5.44 (s, 1H, 7-H), 5.68 (s,
1H, 7´-H), 5.73 (dd, J = 9.9, 2.6 Hz, 1H, 6-H), 5.97 (ddd, J = 9.9, 5.1,
2.8 Hz, 1H, 5-H), 9.53 (br, 1H, OOH). 13C{1H} NMR (125 MHz,
CDCl3): d = 26.7 (t, C-8), 29.9 (t, C-4), 49.0 (s, C-1), 53.9 (q, OMe),
83.4 (d, C-3), 116.2 (s, CN), 120.2 (t, C-7), 125.0 (d, C-6), 128.6 (d, C-
5), 140.0 (s, C-2), 173.9 (s, CO2). IR (film): v = 3379, 2954, 2255,
1728, 1234, 1069, 928, 737 cmꢀ1. HRMS (GC-TOF): m/z Calcd for
C11H14NO4 (M + H+): 224.0923; Found: 224.0923. Anal. Calcd for
C11H13NO4 (223.23): C, 59.19; H, 5.87; N, 6.27. Found: C, 58.36; H,
5.86; N, 6.28.
m-Lactam hydroperoxide 2h (main isomer). White solid (110 mg,
56%). Rf = 0.35 (CHCl3/MeOH = 95:5). Mp = 71–72°C. 1H NMR
(400 MHz, CDCl3): d = 1.88 (ddd, J = 13.6, 8.8, 6.7 Hz, 1H, 8-H), 2.42
(ddd, J = 13.6, 8.3, 4.2 Hz, 1H, 8´-H), 2.88 (ddd, J = 19.6, 4.4, 2.0 Hz,
1H, 4-H), 3.08 (ddd, J = 19.6, 52.8, 2.6 Hz, 1H, 4´-H), 3.35 (ddd,
J = 9.4, 8.8, 4.2 Hz, 1H, 9-H), 3.43 (ddd, J = 9.4, 8.3, 6.7 Hz, 1H, 9´-
H), 5.02 (s, 1H, 2-H), 5.10 (s, 1H, 7-H), 5.25 (s, 1H, 7´-H), 5.50 (ddd,
J = 9.6, 2.6, 2.0 Hz, 1H, 6-H), 5.80 (ddd, J = 9.6, 4.4, 2.8 Hz, 1H, 5-
H), 6.43 (br, 1H, NH), 10.14 (br, 1H, OOH). 13C{1H} NMR (100 MHz,
CDCl3): d = 28.6 (t, C-4), 34.4 (t, C-8), 40.1 (t, C-9), 52.1 (s, C-1), 85.9
(d, C-2), 108.9 (t, C-7), 127.9 (d, C-5), 128.2 (d, C-6), 140.3 (s, C-3),
180.5 (s, CONH). IR (film): v = 3259, 2988, 2947, 2893, 1682, 1429,

4
Werner Fudickar et al.
1276, 1054, 915, 729 cmꢀ1
.
HRMS (GC-TOF): m/z Calcd for
J = 9.6, 2.4, 2.0 Hz, 1H, 6-H), 5.86 (dt, J = 9.6, 3.3 Hz, 1H, 5-H), 8.61
(br, 1H, OOH). 13C{1H} NMR (125 MHz, CDCl3): d = 30.0 (t, C-8),
34.1 (t, C-4), 51.0 (s, C-1), 66.7 (t, C-9), 85.5 (d, C-2), 110.0 (t, C-7),
126.3 (d, C-6), 128.5 (d, C-5), 138.9 (s, C-3), 180.5(s, CO2). IR (film):
v = 3382, 2923, 1752, 1375, 1169, 1024, 908, 727 cmꢀ1. HRMS (GC-
TOF): m/z Calcd for C10H13O4 (M + H+): 197.0814; Found: 197.0815.
m-Lactone hydroperoxide 3k (minor isomer, bis peroxide). Colorless
oil (35 mg, 15%). Rf = 0.52 (EtOAc). 1H NMR (500 MHz, CDCl3):
d = 1.53 (s, 3H, 7-H), 1.90 (ddd, J = 13.7, 7.0, 3.9 Hz, 1H, 8-H), 2.75
(dt, J = 13.7, 8.9 Hz, 1H, 8´-H), 4.40 (td, J = 8.9, 7.0 Hz, 1H, 9-H),
4.46 (td, J = 8.9, 3.9 Hz, 1H, 9´-H), 4.94 (dd, J = 6.2, 1.0 Hz, 1H, 6-
H), 6.40 (dd, J = 8.3, 1.0 Hz, 1H, 4-H), 6.80 (dd, J = 8.3, 6.2 Hz, 1H,
5-H), 8.49 (br, 1H, OOH). 13C{1H} NMR (125 MHz, CDCl3): d = 17.6
(q, C-7), 27.2 (t, C-8), 50.5 (s, C-1), 65.7 (t, C-9), 73.6 (d, C-6), 74.7 (s,
C-3), 86.2 (t, C-2), 132.3 (d, C-5), 134.3 (d, C-4), 178.5 (s, CO2). IR
(film): v = 3668, 3363, 2982, 2904, 1741, 1374, 1172, 1043, 1023,
C10H14NO3 (M + H+): 196.0974; Found: 196.0979.
m-Lactam hydroperoxide 3h (minor isomer). Colorless oil (30 mg,
15%). Rf = 0.17 (CHCl3/MeOH = 95:5). 1H NMR (500 MHz, CDCl3):
d = (s, 3H, 7-H), 2.26 (t, J = 6.7 Hz, 1H, 8-H), 3.48 (t, J = 6.7 Hz, 1H,
9-H), 4.87 (d, J = 3.1 Hz, 1H, 2-H), 5.32 (br, 1H, NH), 5.70 (ddd,
J = 3.3, 3.1, 2.0 Hz, 1H, 4-H), 5.99 (dd, J = 9.9, 2.0 Hz, 1H, 6-H), 6.19
(dd, J = 9.9, 3.3 Hz, 1H, 5-H), 7.55 (br, 1H, OOH). 13C{1H} NMR
(125 MHz, CDCl3): d = 20.1 (q, C-7), 34.9 (t, C-8), 39.0 (t, C-9), 48.1
(s, C-1), 78.3 (d, C-2), 126.4, 126.9, 131.7 (3d, C-4, C-5, C-6), 133.0 (s,
C-3), 177.7 (s, CONH). IR (film): v = 3248, 2989, 2900, 1676, 1285,
1031, 904, 702 cmꢀ1. HRMS (GC-TOF): m/z Calcd for C10H14NO3
(M + H+): 196.0974; Found: 196.0973.
2,3-Dimethyllactam 2i (main isomer). Colorless oil (125 mg, 60%).
Rf = 0.15 (EtOAc/MeOH = 98:2). 1H NMR (500 MHz, CDCl3):
d = 1.57 (s, 3H, CH3), 2.30 (ddd, J = 15.2, 8.0, 7.3 Hz, 1H, 8-H), 2.40
(ddd, J = 18.4, 2.6, 2.4 Hz, 1H, 4-H), 2.51 (ddd, J = 15.2, 5.2, 4.2 Hz,
1H, 8´-H), 2.53 (dd, J = 18.4, 5.4 Hz, 1H, 4´-H), 3.36 (ddd, J = 9.7,
7.3, 5.2 Hz, 1H, 9-H), 3.41(ddd, J = 9.7, 8.0, 4.2 Hz, 1H, 9´-H), 5.23 (s,
1H, 7-H), 5.47 (s, 1H, 7´-H), 5.59 (dd, J = 9.9, 2.6 Hz, 1H, 6-H), 5.82
(ddd, J = 9.9, 5.4, 2.4 Hz, 1H, 5-H), 6.33 (br, 1H, NH), 7.54 (br, 1H,
OOH). 13C{1H} NMR (125 MHz, CDCl3): d = 22.0 (q, CH3), 35.7 (t,
C-4), 37.2 (t, C-8), 39.0 (t, C-9), 52.0 (s, C-1), 82.3 (s, C-3), 114.3 (t, C-
7), 124.7 (d, C-5), 128.7 (d, C-6), 145.6 (s, C-2), 180.2 (s, CONH). IR
(film): v = 3668, 3246, 2987, 2903, 1677, 1285, 1059, 1034, 904,
700 cmꢀ1. HRMS (GC-TOF): m/z Calcd for C11H16NO3 (M + H+):
210.1130; Found: 210.1134.
2,3-Dimethyllactam 3i (side isomer). Colorless oil (60 mg, 29%).
Rf = 0.22 (EtOAc/MeOH = 98:2). 1H NMR (500 MHz, CDCl3):
d = 1.67 (s, 3H, CH3), 2.00 (ddd, J = 13.8, 8.9, 6.5 Hz, 1H, 8-H), 2.62
(ddd, J = 13.8, 8.2, 4.2 Hz, 1H, 8´-H), 2.95 (ddd, J = 19.7, 3.5, 2.2 Hz,
1H, 4-H), 3.02 (ddd, J = 19.7, 3.5, 2.2 Hz, 1H, 4´-H), 3.31 (ddd,
J = 9.2, 8.9, 4.2 Hz, 1H, 9-H), 3.41 (ddd, J = 9.2, 8.2, 6.5 Hz, 1H, 9´-
H), 5.81 (br, 1H, NH), 5.16 (s, 1H, 7-H), 5.19 (s, 1H, 7´-H), 5.50 (dt,
J = 9.7, 2.2 Hz, 1H, 6-H), 5.86 (dt, J = 9.7, 3.5 Hz, 1H, 5-H), 8.01 (br,
1H, OOH). 13C{1H} NMR (125 MHz, CDCl3): d = 17.2 (q, CH3), 27.0
(t, C-8), 33.6 (t, C-4), 39.7 (t, C-9), 53.4 (s, C-1), 86.7 (s, C-2), 111.0 (t,
C-7), 127.5 (d, C-5), 128.9 (d, C-6), 143.6 (s, C-3), 178.7 (s, CONH). IR
(film): v = 3265, 2987, 2915, 2391, 1664, 1421, 1301, 1119, 1052, 888,
696 cmꢀ1. HRMS (GC-TOF): m/z Calcd for C11H16NO3 (M + H+):
210.1130; Found: 210.1132.
704 cmꢀ1
.
HRMS (GC-TOF): m/z Calcd for C10H13O6 (M + H+):
229.0712; Found: 229.0716.
2,3-Dimethyllactone 2l (main isomer). Colorless oil (105 mg, 50%).
Rf = 0.43 (EtOAc). 1H NMR (500 MHz, CD3OD): d = 1.52 (s, 3H,
CH3), 2.28 (ddd, J = 18.5, 2.8, 2.3 Hz, 1H, 4-H), 2.35 (ddd, J = 13.0,
9.6, 8.1 Hz, 1H, 8-H), 2.49 (ddd, J = 18.5, 5.4, 1.0 Hz, 1H, 4´-H), 2.83
(ddd, J = 13.0, 6.4, 3.0 Hz, 1H, 8´-H), 4.28 (ddd, J = 9.6, 9.0, 6.4 Hz,
1H, 9-H), 4.44 (ddd, J = 9.0, 8.1, 3.0 Hz, 1H, 9´-H), 5.01 (s, 1H, 7-H),
5.44 (s, 1H, 7´-H), 5.52 (ddd, J = 9.9, 2.8, 1.0 Hz, 1H, 6-H), 5.79 (ddd,
J = 9.9, 5.4, 2.3 Hz, 1H, 5-H), 7.56 (br, 1H, OOH). 13C{1H} NMR
(125 MHz, CD3OD): d = 21.2 (q, CH3), 36.0 (t, C-4), 36.5 (t, C-8), 52.0
(s, C-1), 66.0 (t, C-9), 80.6 (s, C-3), 113.1 (t, C-7), 125.5 (d, C-5), 126.9
(d, C-6), 145.4 (s, C-2), 181.0 (s, CONH). IR (film): v = 3363, 2994,
2922, 2254, 1741, 1375, 1179, 1032, 908, 727 cmꢀ1. HRMS (GC-TOF):
m/z Calcd for C11H15O4 (M + H+): 211.0970; Found: 211.0962.
2,3-Dimethyllactone 3l (side isomer). Colorless oil (90 mg, 43%).
Rf = 0.34 (EtOAc). 1H NMR (500 MHz, CDCl3): d = 1.67 (s, 3H, CH3),
2.11 (ddd, J = 13.5, 8.9, 8.2 Hz, 1H, 8-H), 2.69 (ddd, J = 13.5, 8.0,
4.0 Hz, 1H, 8´-H), 2.91 (dm, J = 20.0 Hz, 1H, 4-H), 3.04 (dm,
J = 20.0 Hz, 1H, 4´-H), 4.27 (ddd, J = 9.0, 8.9, 4.0 Hz, 1H, 9-H), 4.44
(ddd, J = 9.0, 8.2, 8.0 Hz, 1H, 9´-H), 5.06 (s, 1H, 7-H), 5.12 (s, 1H, 7´-
H), 5.54 (ddd, J = 9.7, 4.4, 2.3 Hz, 1H, 6-H), 5.82 (ddd, J = 9.7, 6.0,
3.5 Hz, 1H, 5-H), 8.24 (br, 1H, OOH). 13C{1H} NMR (125 MHz,
CDCl3): d = 17.5 (q, CH3), 33.1 (t, C-8), 33.8 (t, C-4), 52.3 (s, C-1),
66.1 (t, C-9), 86.1 (s, C-2), 110.9 (t, C-7), 127.5 (d, C-5), 128.2 (d, C-6),
142.7 (s, C-3), 178.7 (s, CONH). IR (film): v = 3362, 2929, 2254, 1742,
1419, 1071, 1053, 921, 741 cmꢀ1. HRMS (GC-TOF): m/z Calcd for
C11H15O4 (M + H+): 211.0970; Found: 211.0968.
o-Lactone hydroperoxide 2j (main isomer). Colorless oil (135 mg,
69%). Rf = 0.28 (EtOAc). 1H NMR (500 MHz, CDCl3): d = 2.34 (ddd,
J = 13.1, 9.6, 8.1 Hz, 1H, 8-H), 2.44 (dddd, J = 18.6, 4.7, 3.4, 1.6 Hz,
1H, 4-H), 2.60 (dddd, J = 18.6, 5.1, 2.9, 2.5 Hz, 1H, 4´-H), 2.67 (ddd,
J = 13.1, 6.3, 3.0 Hz, 1H, 8´-H), 4.32 (ddd, J = 9.6, 9.1, 6.3 Hz, 1H, 9-
H), 4.41 (ddd, J = 9.1, 8.1, 3.0 Hz, 1H, 9´-H), 4.79 (dd, J = 5.1, 3.4 Hz,
1H, 3-H), 5.11 (s, 1H, 7-H), 5.40 (s, 1H, 7´-H), 5.51 (ddd, J = 9.8, 2.5,
1.6 Hz, 1H, 6-H), 5.86 (ddd, J = 9.8, 4.7, 2.9 Hz, 1H, 5-H), 8.09 (br,
1H, OOH). 13C{1H} NMR (125 MHz, CDCl3): d = 30.2 (t, C-4), 36.2
(t, C-8), 50.2 (s, C-1), 65.8 (t, C-9), 83.2 (d, C-3), 114.9 (t, C-7), 126.3
(d, C-5), 126.7 (d, C-6), 141.1 (s, C-2), 178.6 (s, CO2). IR (film):
RESULTS AND DISCUSSION
Photooxygenation of nitriles
We started our investigations with 1,4-cyclohexadiene 1c, which
is easily available by Birch reduction of o-toluic acid (22). Pho-
tooxygenation was performed in deuterated chloroform (CDCl3)
with TPP as sensitizer, and 1H NMR spectra were measured at
different times (see Figure S1). Indeed, the reaction proceeded
very slowly and only 10% conversion of the starting material
was observed after 2 h, 42% after 8 h and >97% only after 72 h
(Table 1, entries 1–3). We could quantify the relative kinetics of
the photooxygenations by an NMR competition experiment (see
below). Additionally, due to the long reaction time and the
slightly acidic reaction conditions in CDCl3, o-tolylacetonitrile
was formed by decarbonylation as side product, in accordance
with our previous studies (16,24). Thus, we could only isolate
the main hydroperoxide trans-2c by column chromatography in
low yield besides o-tolylacetonitrile (entry 3).
v = 3668, 3365, 2980, 2904, 1742, 1376, 1174, 1048, 1020, 704 cmꢀ1
Anal. Calcd for C10H12O4 (196.20): C, 61.22; H, 6.16. Found: C, 61.04;
H, 6.21.
.
o-Lactone hydroperoxide 3j (minor isomer). Colorless oil (35 mg,
18%). Rf = 0.39 (EtOAc). 1H NMR (500 MHz, CDCl3): d = 1.51 (s, 3H,
7-H), 2.27 (ddd, J = 13.6, 7.8, 4.8 Hz, 1H, 8-H), 3.05 (ddd, J = 13.6,
8.5, 7.5 Hz, 1H, 8´-H), 4.38 (ddd, J = 9.0, 7.8, 7.5 Hz, 1H, 9-H), 4.49
(ddd, J = 9.0, 8.5, 4.8 Hz, 1H, 9´-H), 5.60 (dt, J = 9.7, 1.0 Hz, 1H, 3-
H), 5.85 (dt, J = 9.4, 1.0 Hz, 1H, 6-H), 6.05 (ddd, J = 9.4, 5.4, 1.0 Hz,
1H, 5-H), 6.21 (ddd, J = 9.7, 5.4, 1.0 Hz, 1H, 4-H), 7.78 (br, 1H,
OOH). 13C{1H} NMR (125 MHz, CDCl3): d = 19.2 (q, C-7), 29.0 (t, C-
8), 51.5 (s, C-1), 65.9 (t, C-9), 83.3 (s, C-2), 123.5 (d, C-5), 126.0 (d, C-
4), 127.3 (d, C-3), 129.4 (d, C-6), 177.2 (s, CO2). IR (film): v = 3371,
2990, 2918, 1747, 1374, 1167, 1021, 704 cmꢀ1. HRMS (GC-TOF): m/z
Calcd for C10H13O4 (M + H+): 197.0814; Found: 197.0817.
m-Lactone hydroperoxide 2k (main isomer). Colorless oil (135 mg,
69%). Rf = 0.62 (EtOAc). 1H NMR (500 MHz, CDCl3): d = 2.06 (ddd,
J = 16.3, 8.7, 7.6 Hz, 1H, 8-H), 2.57 (ddd, J = 16.3, 5.2, 4.9 Hz, 1H
1H, 8´-H), 2.91 (ddd, J = 19.8, 3.3, 2.0 Hz, 1H, 4-H), 3.09 (ddd,
J = 19.8, 3.3, 2.4 Hz, 1H, 4´-H), 4.35 (td, J = 8.7, 4.9 Hz, 1H, 9-H),
4.41 (ddd, J = 8.7, 7.6, 5.2 Hz, 1H, 9´-H), 5.02 (d, J = 1.1 Hz, 1H, 2-
H), 5.12 (d, J = 1.1 Hz, 1H, 7-H), 5.18 (s, 1H, 7´-H), 5.52 (ddd,
The low reactivity can be explained by the planar conforma-
tion of 1,4-cyclohexadienes and steric hindrance by the adjacent
quaternary stereocenter (16,25). However, the nitrile group might
1
additionally quench O2 and influence the reaction electronically
(26), although it is not in conjugation to the double bonds. To

Photochemistry and Photobiology
5
Table 1. Photooxygenation of 1,4-cyclohexadienes 1a–f.*
R1
R2
CH3
R1
R2
CH2
R1
R2
CH2
R1
R2
CH3
OOH
O2, TPP, h
+
+
OOH
OOH
trans-2a f
CDCl3 or CD3OD
1a f
cis-2a f
regio-3a f
cis : trans†
Entry
Comp.
1c
R1
R2
Time
Conv.†
cis-2
trans-2
regio-3
1
2
3
4
5
6
7
8
9
CH2CN
CH2CN
CH2CN
Et
CH2CN
Et
CO2H
CO2H
CO2H
CO2H
2 h
8 h
10%
42%
–
–
–
–
1c
1c
44 : 56
45 : 55
70 : 30
7 : 93
25 : 75
32 : 68
70 : 30
70 : 30
> 96 : 4
–
–
–
72 h
24 h
48 h
24 h
48 h
24 h
48 h
24 h
>97%
>97%
>97%
>97%
>97%
>97%
>97%
>97%
30%‡
50%
–
29%
22%
74%
51%
54%
25%
23%
–
<5%
<5%
<5%
<5%
7%
<5%
5%
1a§
1d
1a§
1e
–
CO2–
CO2
17%
22%
65%
56%
86%
CH2CN
Et
CO2Me
CO2Me
CH2OH
CH2OH
1b¶
1f
CH2CN
Et
10
1b¶
<5%
*For procedure see Materials and Methods. †Determined from 1H NMR of the crude reaction mixtures (Supporting Information). ‡o-Tolylacetonitrile
¶
was isolated in 30% as side product. §From (16). From (21).
substantiate this effect, we photooxygenated a 1:1 mixture of 1,4-
cyclohexadiene 1c and the simple ethyl-substituted derivative 1a
(R1 = Et, entry 4) (16) in an NMR competition experiment (see
Supporting Information for details). Indeed, nitrile 1c reacts about
4 times slower, which cannot be explained by steric hindrance,
because ethyl and CH2CN groups have almost the same steric
demand (27), and thus, electronic effects must be operative.
To suppress the formation of o-tolylacetonitrile and to study
electronic effects, we deprotonated the acid group to carboxylate
1d by addition of potassium carbonate and conducted the pho-
tooxygenation in deuterated methanol with rose Bengal as sensi-
tizer. Such conditions gave better results for simple 1,4-
cyclohexadienes 1b in our previous work as well (16). Indeed,
now full conversion was achieved after 48 h, and the hydroperox-
ide trans-2c was formed with high stereoselectivity and isolated
by column chromatography in 74% yield (Table 1, entry 5).
However, nitrile 1d reacted again slower than the corresponding
ethyl-substituted 1,4-cyclohexadiene 1b under the same condi-
tions (entry 6), corroborating that this functional group decreases
the reactivity of organic compounds with singlet oxygen.
4c. Both compounds could be easily separated by column chro-
matography and were isolated in pure form in good yields (Sup-
porting Information). Thus, the main product in the
photooxygenation of acid 1c must be hydroperoxide trans-2c.
Interestingly, the reaction of deprotonated acid 1d affords one
hydroperoxide trans-2d with high site-, regio- and stereoselectiv-
ity. This might be explained by perepoxides 6d (Scheme 4).
Such labile intermediates exhibit dipolar or zwitterionic char-
acter, they have been postulated to explain regioselectivities of
1O2 reactions (17,28), and they can be trapped chemically (20).
Thus, singlet oxygen can attack cyclohexadiene 1d from two
faces, resulting in the reversible formation of perepoxides cis-6d
and trans-6d. In principle, both double bonds can react; how-
ever, due to the planarity of 1,4-cyclohexadienes (25), the pro-
tons of the methylene group are far away and in an unfavorable
orientation to undergo an efficient ene reaction. Additionally, the
methyl group is located adjacent to the sterically demanding qua-
ternary stereocenter, increasing its reactivity due to the “large-
group effect,” established by Ed (9) and Michael Orfanopoulos
(29,30).
To obtain more stable products and to determine the relative
configuration of the newly formed stereocenter, we photooxy-
genated the carboxylic acid 1c in CDCl3 for 72 h again, and
reduced directly the crude hydroperoxides 2c with dimethyl sul-
fide to the corresponding alcohols 4c, which showed the same
product ratio of 45:55 (Scheme 3, Supporting Information).
Acidification to pH 4 by 1N HCl resulted in the cyclization to
lactone 5c, whereas the trans isomer remained as alcohol trans-
Additionally, the stereoselectivities can be explained by per-
epoxide intermediates 6d as well. If the carboxylic acids are
deprotonated (Table 1, entries 5 and 6), a severe electrostatic
repulsion with the negatively charged oxygen is operative in cis-
6d but not in trans-6d (Scheme 4). This explains the high selec-
tivity in favor of hydroperoxide trans-2d (entry 5), in accordance
with our previous studies with simple ethyl-substituted 1,4-cyclo-
hexadiene 1b (entry 6) (16). However, the trans selectivities with
O
CH2
O
CO2H
CH2
CO2H
CO2H
NC
NC
NC
NC
CH2
CH2
Me-S-Me
> 95%
HCl
+
OOH
cis/trans-2c
dr 45:55
Scheme 3. Transformation of the photooxygenation mixture cis/trans-2c into lactone 5c and hydroxy acid trans-4c.
OH
OH
cis/trans-4c
dr 45:55
5c (40%)
trans-4c (43%)

6
Werner Fudickar et al.
column chromatography (Materials and Methods), and it results
from the ene reaction with the methylene and not methyl group.
This surprising behavior will be discussed in the next chapter,
where such regioisomers 3 are formed in larger amounts with
spiro compounds.
Overall, we found a new effect of nitriles on photooxygena-
tions of alkenes. This functional group decreases the reactivity,
presumably by quenching of singlet oxygen. On the other hand,
perepoxide intermediates are stabilized by attractive intramolecu-
lar polar interactions. Although the influence of various func-
tional groups on the ene reaction of 1O2 has been discussed
intensively (21,32), and Ed investigated the effect of allyl nitriles
on the geminal selectivity (33), to the best of our knowledge
such interesting deactivating/stabilizing behavior was hitherto
unknown.
O
O
O
O
H
H
O
CH2
H
H
O
CH2
H
H
O
O
H
C
H
H
H
N
C
N
cis-6d
trans-6d
Scheme 4. Proposed perepoxide intermediates 6d in the reaction of sin-
glet oxygen with 1,4-cyclohexadiene 1d.
nitriles 1c and 1d are always higher compared to 1a (entries 3–
6). This cannot be rationalized by steric repulsion, because the
CH2CH3 has almost the same steric demand as the CH2CN
group (27). We postulate an attractive electronic interaction
between the negatively charged oxygen of the perepoxide and
the positively charged nitrile group, which is only operative in
intermediate trans-6d (Scheme 4). Thus, repulsion by the car-
boxylate and attraction by the nitrile reinforce each other,
explaining the high stereoselectivity with 1,4-cyclohexadiene 1d
(entry 5). An intramolecular interaction of a nitrile group with
negatively charged oxygen atoms has been discussed previously
(31), but to the best of our knowledge not in perepoxide interme-
diates of singlet oxygen ene reactions.
To further prove such a “nitrile effect” and to obtain more
stable hydroperoxides 2, we synthesized the methyl ester 1e and
alcohol 1f and compared their photooxygenation with the corre-
sponding ethyl derivatives (Table 1, entries 7–10). In both cases,
the nitriles react again slower and with higher trans selectivity.
Even the alcohol, affording for ethyl derivative 1b exclusively
the cis isomer (entry 10), shows for nitrile 1f a selectivity of
70:30 (entry 9). Thus, the OH stabilizing effect for perepoxides
(21) is stronger than the attractive interaction to the nitrile group,
but in 1,4-cyclohexadiene 1f still operative.
Photooxygenation of spiro compounds
Although various examples of photooxygenations of bicyclic sys-
tems are known as summarized in Ed´s excellent review (9), to
the best of our knowledge, spiro compounds have not been stud-
ied until now. The advantage of such substrates is that they pro-
vide a quaternary stereocenter and are conformationally more
rigid than nitriles 1a–f, which might allow to control the regio-
and stereoselectivities. We published an easy entry to c-spirolac-
tams starting from toluic acids 7 via nitriles (22), and to unsatu-
rated derivatives very recently (Scheme 5a) (23). To compare
lactams 1g–i with lactones, we developed herein a similar strat-
egy by Birch reduction in the presence of ethylene oxide, afford-
ing c-spirolactones 1j–l in high yields and pure form
(Scheme 5b, Supporting Information).
We started the photooxygenations with o-derivatives 1g and
1j under the same conditions like for nitriles 1a–f, in CDCl3 as
solvent and with TPP as sensitizer (Scheme 6). Indeed, hydroper-
oxides 2g and 2j with an exo double bond are again the main
products, in accordance with the “large-group effect,” but regioi-
somers 3g and 3j are formed to a higher extend compared to
nitriles. Even more interesting and surprising are the stereoselec-
tivities, which differ from the monocyclic 1,4-cyclohexadienes
1a–f. Thus, for both regioisomers, we observe only one
Interestingly, we observe for ester 1e and alcohol 1f the for-
mation of regioisomers regio-3 in the NMR spectra of the crude
products to a small extend of <10% (Table 1, Supporting Infor-
mation). Furthermore, we could isolate this minor product 3e by
H
N
CO2H
O
NC
1. Ni / H2
2. SOCl2
R1
1. Li / NH3
2. ClCH2CN
COOH
R2
CH3
1g: R1 = CH3; R2 = H (92%)
1h: R1 = H; R2 = CH3 (95%)
1i: R1 = CH3; R2 = CH3 (95%)
a)
CH3
7
b)
HO
O
CO2H
CH3
O
1. Li / NH3
R1
2.
O
SOCl2
R2
1j: R1 = CH3; R2 = H (91%)
1k: R1 = H; R2 = CH3 (91%)
1l: R1 = CH3; R2 = CH3 (83%)
Scheme 5. Synthesis of c-spriolactams 1g–i and lactones 1j–l from toluic acids 7. a) Ref. (23); b) this work.

Photochemistry and Photobiology
7
X
X
X
O
O
O
CH3
CH3
CH2
O2, TPP, h
CDCl3, 24 h
+
OOH
OOH
1g
1j
X = NH
X = O
2g (67%)
2j (69%)
3g (18%)
3j (18%)
Scheme 6. Photooxygenation of o-lactam 1g and o-lactone 1j.
Figure 1. X-ray structure of hydroperoxide 2g.
O
H
O
X
H
O
X
O
H8
O
CH2
H
O
O
O
O
O
H8
H
H
CH3
X
H
H
O
H
H
O
X
X
O
2g,j
3g,j
CH3
CH3
CH3
O
favored-6g,j
disfavored-6g,j
minor-6g,j
Scheme 7. NOESY interactions in hydroperoxides 2 and 3.
O
Scheme 8. Proposed perepoxide intermediates 6g,j in the photooxygena-
tion of spiro compounds 1g,j.
diastereomer in the 1H NMR spectra (see Supporting Informa-
tion), and only two hydroperoxides were isolated in good yields
in pure form (Materials and Methods).
After the successful photooxygenations of o-lactams and o-
lactones, we became interested in the reactions of m-isomers 1h
and 1k with singlet oxygen (Scheme 9). Now a “large-group
effect” is not operative, full conversion was only achieved after
4 d compared to 1 d for the o-isomer, and therefore, the NMR
spectra of the crude products show some decomposition (Sup-
porting Information). However, hydroperoxides 2h,k with an exo
double bond are again the main products, which were isolated in
pure form. The attack of singlet oxygen occurs for these isomers
anti to the carbonyl group, in accordance with the o-lactams and
o-lactones.
To determine the relative configurations, we measured
NOESY spectra, which are especially valuable for spiro com-
pounds, due to their rigid structure (Scheme 7, Supporting Infor-
mation). Thus, for the main regioisomers 2, we observe a cross-
peak between OOH and 8-H in the lactam/lactone rings, indicat-
ing that singlet oxygen attacks anti to the carbonyl group, in
accordance with reactions of nitriles 1a–e. Furthermore, we
obtained an X-ray structure of hydroperoxide 2g, confirming its
relative configuration unequivocally (Fig. 1). On the other hand,
regioisomers 3 exhibit a distinctive cross-peak between the
methyl group and 8-H, and thus, singlet oxygen attacks the 1,4-
cyclohexadiene syn to the carbonyl group.
On the other hand, minor regioisomers 3h,k result from selec-
tive photooxygenation of the opposite face and ene reaction with
the methylene group, speaking again for a “spiro effect.” Inter-
estingly, such products are 1,3-dienes with an electron-donating
methyl group, which might react further by a [4 + 2] cycloaddi-
We explain this unusual behavior again by perepoxide inter-
mediates 6g,j (Scheme 8). Because lactams and lactones give the
same selectivities, hydrogen bonding of singlet oxygen to the
NH group can be excluded. Thus, this reactive species attacks
the double bond preferentially anti to the carbonyl group, afford-
ing the perepoxide favored-6g,j and the main hydroperoxides 2g,
1
tion of O2. Indeed, for the m-lactam 1h, we obtained the 1,3-di-
ene 3h in 15% yield, whereas for the m-lactone 1k, we could
isolate bisperoxide 3k in 15% in pure form as a single diastere-
omer with 4 stereogenic centers (Materials and Methods)
1
j, controlled by the “large-group effect.” On the other hand, O2
reacts to a smaller extend from the syn face, but now perepoxide
disfavored-6g,j is too much destabilized by electrostatic repulsion
with the lone pairs of the carbonyl group and the resulting syn-
hydroperoxide is not observed. Therefore, the negatively charged
oxygen is pushed away from the spiro center to react with the
methylene and not methyl group, affording the regioisomeric
hydroperoxides 2g,j via intermediate minor-6g,j. Additionally,
the lone pairs of the carbonyl group might stabilize the positive
charge of this intermediate by anchimeric assistance, as discussed
by Ed in photooxygenations of allyl sulfides (34). The main dif-
ference to the monocyclic 1,4-cyclohexadienes (Table 1) is to
our opinion the rigid structure of spiro compounds. Thus, the
carbonyl group is fixed in close proximity and in almost the
same plane to the perepoxide oxygens, and its influence becomes
stronger. Although herein we will only investigate up to six spiro
compounds (see below), we call this a hitherto unknown “spiro
effect,” which overcomes the well-known “large-group effect” in
our studies.
(Scheme 9). The preferential attack of singlet oxygen anti to the
hydroperoxy group is in accordance with bis oxidation of simple
1,4-cyclohexadiene (7,35).
Finally, we photooxygenated the o,m-dimethyl derivatives 1i
and 1l to distinguish between two methyl groups at different
positions (Scheme 10). Indeed, the reactions proceed now much
faster with full conversion after 4 h, because the double is elec-
tron richer and more reactive. Again, we observe only two
regioisomers with excellent stereoselectivity in the 1H NMR of
the crude reaction mixture (Supporting Information). All
hydroperoxides 2 and 3 were isolated by column chromatogra-
phy in good yields in pure form (Materials and Methods). The
main products 2i,l result from reaction of the o-methyl group
due to the “large-group effect.” However, this is again only pos-
1
sible if O2 attacks anti to the carbonyl group because otherwise
the negatively charged oxygen of the perepoxide intermediate
would be in too close contact to this functional group (see
Scheme 8). Thus, if the oxidation occurs from the same face like

8
Werner Fudickar et al.
X
X
X
O
O
O
O
H
H
O
1O2
O
OOH
O2, TPP, h
CDCl3, 4 d
X = NH
X = O
+
H
CH2
H
OOH
H3C
O
CH3
CH3
OOH
1h
1k
2h (56%)
2k (69%)
3h (15%)
3k (15%)
Scheme 9. Photooxygenation of m-lactam 1h and m-lactone 1k.
for the photooxygenations of six spiro compounds, we believe
that such a “spiro effect” might be a general motive for other
singlet oxygen reactions.
X
X
X
O
O
O
OOH
CH3
CH2
O2, TPP, h
CDCl3, 4 h
X = NH
X = O
+
CH3
OOH
CH3
CH3
CH2
Acknowledgements—This work was supported by the University of
€
Potsdam. We thank Angela Krtitschka and the group of Prof. Moller for
1i
1l
2i (60%)
2l (50%)
3i (29%)
3l (43%)
NMR and HRMS measurements. Open Access funding enabled and
organized by Projekt DEAL.
Scheme 10. Photooxygenation of o,m-dimethyllactam 1i and o,m-
dimethyllactone 1l.
SUPPORTING INFORMATION
the carbonyl group, the perepoxide is more stable if its oriented
toward the m-methyl group, affording regioisomers 3i,l, whose
relative configurations were confirmed by NOESY measurements
(Supporting Information). Overall, the stereoselectivity of the sin-
glet oxygen attack controls the regioselectivity of the ene reac-
tions. Because we observe this behavior for all
photooxygenations of spiro compounds 1g–l, we believe that the
“spiro effect” might be a general motive for other singlet oxygen
reactions.
Additional supporting information may be found online in the
Supporting Information section at the end of the article:
Appendix S1. NMR spectra of newly synthesized starting
materials.
NMR competition experiment
Determination of the relative configurations.
Reactions of nitriles 1c–f and NMR spectra of hydroperoxides
2,3c–f.
Reactions of lactams 1g–i and NMR spectra of hydroperox-
ides 2,3g–i.
Reactions of lactones 1j–l and NMR spectra of hydroperox-
ides 2,3j–l.
CONCLUSION
In conclusion, we have found interesting effects of functional
groups on the photooxygenation of 1,4-cyclohexadienes. Nitriles
decrease their reactivity with singlet oxygen compared to alkyl-
substituents, which we substantiated by a competition experiment
and might be explained by quenching of this reactive species.
On the other hand, if a perepoxide intermediate is formed, it can
be stabilized by the cyano group, resulting in different stereose-
lectivities. The regioselectivities are controlled by the planar con-
formation of 1,4-cyclohexadienes and the “large-group effect,”
due to the adjacent quaternary stereocenter. Other functional
groups like acids, esters or alcohols show the usual behavior in
singlet oxygen ene reactions by polar repulsion or attraction of
perepoxide intermediates. Because of the high polarity and labil-
ity of hydroperoxides, products have been isolated in only mod-
erate yields but pure form. Furthermore, we could determine the
relative configurations unequivocally by transformations into
stable alcohols and lactones.
Spiro lactams and lactones show an interesting effect on
regio- and stereoselectivities of the ene reactions. Thus, singlet
oxygen attacks the double bond preferentially anti to the car-
bonyl group affording only one regioisomeric hydroperoxide. If
the reaction occurs from the opposite face, the other regioisomer
is exclusively formed by severe electrostatic repulsion in a
perepoxide intermediate. We explain this unusual behavior by
the fixed geometry of spiro compounds and the close proximity
of the carbonyl to the negatively charged oxygen. Overall, the
stereoselectivity of the singlet oxygen attack controls the regio-
selectivity of the ene reaction. Because we observe this behavior
Details of X-ray structure analysis.
REFERENCES
1. S. Nonell and C. Flors (eds.) (2016) Singlet oxygen. The Royal Soci-
ety of Chemistry. https://doi.org/10.1039/9781782622208.
2. Foote, C. S. (1968) Photosensitized oxygenations and the role of sin-
glet oxygen. Acc. Chem. Res. 1, 104–110.
3. Greer, A. (2006) Christopher Foote’s discovery of the role of singlet
oxygen [1O2 (1Dg)] in photosensitized oxidation reactions. Acc.
Chem. Res. 39, 797–804.
4. Fudickar, W. and T. Linker (2018) Release of singlet oxygen from
organic peroxides under mild conditions. ChemPhotoChem 2, 548–
558.
5. Ogilby, P. R. (2010) Singlet oxygen: There is indeed something new
under the sun. Chem. Soc. Rev. 39, 3181–3209.
6. Pibiri, I., S. Buscemi, A. Palumbo Piccionello and A. Pace (2018)
Photochemically produced singlet oxygen: Applications and perspec-
tives. ChemPhotoChem 2, 535–547.
7. Ghogare, A. A. and A. Greer (2016) Using singlet oxygen to synthe-
size natural products and drugs. Chem. Rev. 116, 9994–10034.
8. Clennan, E. L. (1991) Synthetic and mechanistic aspects of 1,3-diene
photooxidation. Tetrahedron 47, 1343–1382.
9. Clennan, E. L. (2000) New mechanistic and synthetic aspects of sin-
glet oxygen chemistry. Tetrahedron 56, 9151–9179.
10. Clennan, E. L. and A. Pace (2005) Advances in singlet oxygen
chemistry. Tetrahedron 61, 6665–6691.
11. Clennan, E. L. (2001) Persulfoxide: Key intermediate in reactions of
singlet oxygen with sulfides. Acc. Chem. Res. 34, 875–884.
12. Clennan, E. L. (2003) Molecular oxygenations in zeolites. Mol.
Supramol. Photochem. 9, 275–308.

Photochemistry and Photobiology
9
13. Birch, A. J. (1950) The reduction of organic compounds by metal-
ammonia solutions. Q. Rev. Chem. Soc. 4, 69–93.
14. Rabideau, P. W. and Z. Marcinow (1992) The birch reduction of
aromatic compounds. Org. React. 42, 1–334.
25. Rabideau, P. W. (1989) Conformational analysis of cyclohexenes,
cyclohexadienes, and related hydroaromatic compounds / edited by
Peter W. Rabideau. VCH, New York, NY.
26. Gorman, A. A. (1992) The bimolecular reactivity of singlet molecu-
lar oxygen. Advances in Photochemistry, pp. 217–274.John Wiley &
Sons, New York, NY.
27. Eliel, E. L. and S. H. Wilen (2008) Stereochemistry of Organic
Compounds. Wiley India Pvt. Limited, New Delhi.
28. Malek, B., W. Fang, I. Abramova, N. Walalawela, A. A. Ghogare
and A. Greer (2016) ‘ene’ Reactions of singlet oxygen at the air-wa-
ter interface. J. Org. Chem. 81, 6395–6401.
29. Stratakis, M. and M. Orfanopoulos (2000) Regioselectivity in the
Ene reaction of singlet oxygen with alkenes. Tetrahedron 56, 1595–
1615.
30. Alberti, M. N. and M. Orfanopoulos (2010) Recent mechanistic
insights in the singlet oxygen ene reaction. Synlett 2010, 999–1026.
31. Mowbray, C. E., W. Skranc and J. D. Wallis (1999) Attractive inter-
actions between functional groups in crowded molecules: The solid
state structures of diethyl 2-cyano(20-nitrophenyl)methyltartronate
and diethyl (20-,60-dinitrophenyl)methyltartronate. J. Chem. Crystal-
logr. 29, 335–341.
32. Prein, M. and W. Adam (1996) The Schenck Ene reaction: Diastere-
oselective oxyfunctionalization with singlet oxygen in synthetic
applications. Angew. Chemie (International Ed. English) 35, 477–
494.
33. Clennan, E. L. and X. Chen (1988) Geminal selectivity in singlet
oxygen reactions. J. Org. Chem. 53, 3124–3125.
15. van Bekkum, H., C. B. van Den Bosch, G. van Minnen-Pathuis, J.
C. de Mos and A. M. van Wijk (1971) Reduction of alkyl-substi-
tuted benzoic acids with lithium in liquid ammonia: Synthesis of 1-
alkyl-2,5-cyclohexadiene-1-carboxylic acids. Recl. des Trav. Chim.
des Pays-Bas 90, 137–149.
€
16. Linker, T. and L. Frohlich (1994) Regio- and diastereoselective pho-
tooxygenation of chiral 2,5-cyclohexadiene-l-carboxylic acids.
Angew. Chemie Int. Ed. English 33, 1971–1972.
17. Singleton, D. A., C. Hang, M. J. Szymanski, M. P. Meyer, A. G.
Leach, K. T. Kuwata, J. S. Chen, A. Greer, C. S. Foote and K. N.
Houk (2003) Mechanism of Ene reactions of singlet oxygen. A two-
step no-intermediate mechanism. J. Am. Chem. Soc. 125, 1319–
1328.
18. Sheppard, A. N. and O. Acevedo (2009) Multidimensional explo-
ration of valleyꢀridge inflection points on potential-energy surfaces.
J. Am. Chem. Soc. 131, 2530–2540.
19. Alberti, M. N. and M. Orfanopoulos (2008) The cyclopropyl group
as a hypersensitive probe in the singlet oxygen Ene reaction mecha-
nism. Org. Lett. 10, 2465–2468.
20. Clennan, E. L., M.-F. Chen and G. Xu (1996) New potent trapping
agents for the peroxidic intermediates formed in the reactions of sin-
glet oxygen. Tetrahedron Lett. 37, 2911–2914.
€
21. Linker, T. and L. Frohlich (1995) Substituent effects in the highly
regioselective and diastereoselective ene reaction of singlet oxygen
with chiral cyclohexadienes. J. Am. Chem. Soc. 117, 2694–2697.
22. Kr€uger, T., A. Kelling, U. Schilde and T. Linker (2017) Simple syn-
thesis of c-spirolactams by birch reduction of benzoic acids. Euro-
pean J. Org. Chem. 2017, 1074–1077.
23. Kr€uger, T. and T. Linker (2021) Synthesis of c-Spirolactams by
birch reduction of arenes. European J. Org. Chem. 1585–1591.
24. Vorndran, K. and T. Linker (2003) Simple two-step ipso substitution
of aromatic carboxylic acids by alkyl halides. Angew. Chemie - Int.
Ed. 42, 2489–2491.
34. Clennan, E. L., X. Chen and J. J. Koola (1990) Steric and electronic
effects on the conformations and singlet oxygen ene regiochemistries
of substituted tetramethylethylenes. The origin of the geminal effect.
J. Am. Chem. Soc. 112, 5193–5199.
35. Salamci, E., H. Secßen, Y. S€utbeyaz and M. Balci (1997) A concise
and convenient synthesis of dl-proto-quercitol and dl-gala-Quercitol
via Ene reaction of singlet oxygen combined with [2 + 4] Cycloaddi-
tion to Cyclohexadiene. J. Org. Chem. 62, 2453–2457.