Hydroxy-group directivity in the regioselective and diastereoselective [2+2] photocycloaddition (Paternò-Büchi reaction) of aromatic carbonyl compounds to chiral and achiral allylic substrates: The preparation of oxetanes with up to three stereogenic centers as synthetic building blocks

Article abstract of DOI:10.1055/s-2001-15070

A set of achiral and chiral allylic alcohols and derivatives 4 is regio- and diastereoselectively converted to the oxetanes 5-7 by the Paternò-Büchi photocycloaddition with the carbonyl partners benzophenone (1), acetophenone (2) and benzaldehyde (3). Chiral allylic alcohols with 1,3-allylic strain yield the oxetanes in high threo diastereoselectivity, which was increased by higher steric interactions. The synergistic interplay between hydrogen bonding and 1,3-allylic strain in the CO-bond-forming step is made responsible for this stereoselectivity, as proven by solvent (methanol versus benzene) effects and masking of the allylic hydroxy group. For the prochiral carbonyl partners 2 and 3, an excellent cis diastereoselectivity is found for the substituents on the oxetane ring. This is rationalized in terms of the preferred orthogonal alignment of the 2p orbitals at the radical sites in the intermediary triplet 1,4-diradical for the intersystem-crossing step (ISC). The present highly regio- and diastereoselective photocycloaddition provides an attractive preparative route to sidechain-functionalized oxetanes with up to three stereogenic centers.

Full text of DOI:10.1055/s-2001-15070

PAPER
1203
Hydroxy-Group Directivity in the Regioselective and Diastereoselective [2+2]
Photocycloaddition (Paternò–Büchi Reaction) of Aromatic Carbonyl
Compounds to Chiral and Achiral Allylic Substrates: The Preparation of
Oxetanes with up to Three Stereogenic Centers as Synthetic Building Blocks
Preparationof
O
xetane
s
a
w
ithupto
l
T
hre
e
S
d
tereogenic
C
e
enter
s
as
S
m
ynthetic Building
B
a
locks r Adam,* Veit R. Stegmann
Institut für Organische Chemie, Universität Würzburg, Am Hubland, 97074 Würzburg, Germany
Fax +49(931)8884756; E-mail: adam@chemie.uni-wuerzburg.de
Received 9 March 2001
Dedicated to Professor Howard E. Zimmerman on the occasion of his 75th birthday
We report herein the highly regio- and diastereoselective
Abstract: A set of achiral and chiral allylic alcohols and derivatives
4 is regio- and diastereoselectively converted to the oxetanes 5–7 by
the Paternò Büchi photocycloaddition with the carbonyl partners
benzophenone (1), acetophenone (2) and benzaldehyde (3). Chiral
oxetane formation which depends on the substitution pat-
tern of the olefinic substrate as well as the nature of the
carbonyl partner. Indeed, our novel methodology allows
allylic alcohols with 1,3-allylic strain yield the oxetanes in high the preparation of sidechain-functionalized oxetanes with
threo diastereoselectivity, which was increased by higher steric in-
teractions. The synergistic interplay between hydrogen bonding and
1,3-allylic strain in the CO-bond-forming step is made responsible
up to three stereogenic centers of defined relative config-
uration.
The results for the [2+2] photocycloaddition of the carbo-
nyl partners 1–3 with the olefin substrates 4 to afford the
oxetanes 5–7 are summarized in the Tables 1– 3. The al-
lylic alcohol 4a with a trisubstituted C=C bond was con-
verted to the oxetane 5a highly (>95:5) regioselectively
(Table 1, entry 1). The corresponding alkene 4b without
an allylic substituent, however, led to a mixture of regioi-
somers 5b and 5b’ in only a ratio of 84:16 (entry 2). The
achiral allylic alcohol 4c with a tetrasubstituted C=C
bond displayed a moderate (61:39) regioselectivity
(Table 1, entry 3). For comparison, the [2+2] photocy-
cloaddition with the analogous acetate 4d as substrate pro-
ceeded expectedly with low (55:45) regioselectivity
(entry 4) since this is known for alkenes with tetrasubsti-
tuted C=C bonds.5
for this stereoselectivity, as proven by solvent (methanol versus
benzene) effects and masking of the allylic hydroxy group. For the
prochiral carbonyl partners 2 and 3, an excellent cis diastereoselec-
tivity is found for the substituents on the oxetane ring. This is ratio-
nalized in terms of the preferred orthogonal alignment of the 2p
orbitals at the radical sites in the intermediary triplet 1,4-diradical
for the intersystem-crossing step (ISC). The present highly regio-
and diastereoselective photocycloaddition provides an attractive
preparative route to sidechain-functionalized oxetanes with up to
three stereogenic centers.
Keywords: [2+2] photocycloaddition, Paternò–Büchi reaction, hy-
drogen bonding, 1,3–allylic strain, diastereoselective synthesis
Since the discovery1 of the [2+2] photocycloaddition of
carbonyl compounds to alkenes (the Paternò–Büchi reac-
tion), this reaction attracted much attention from the
mechanistic as well as synthetic points of view. The syn-
thetic applicability is well established and stereoselective
syntheses have been recently described in several re-
views.2 In this context, hydrogen bonding has been uti-
lized as a controlling factor of regio- and stereoselectivity
in intermolecular [2+2] photocycloadditions only in the
last few years.3 Motivated by our studies on the substrate-
directed, diastereoselective oxyfunctionalization of chiral
allylic alcohols, it was of synthetic interest to assess
whether the hydroxy-directing effect4 applies to the Pater-
nò–Büchi reaction. In order to determine whether an allyl-
ic hydroxy group in the olefin is able to influence the
regio- and diastereoselectivity in the [2+2] photocycload-
dition of the electronically excited aromatic carbonyl part-
ners 1–3, appropriate chiral and achiral allylic substrates
4 with tri- or tetrasubstituted C=C bonds (Scheme 1) were
chosen.
In view of the enhanced regioselectivity caused by the al-
lylic hydroxy group in the trisubstituted alkene 4a com-
pared to the unfunctionalized substrate 4b (entries 1, 2),
the chiral allylic alcohols 4e–h and the allylic silyl ether
4i (Scheme 1) were employed, in order to assess, whether
both a regio- as well as diastereoselective oxetane forma-
tion (threo versus erythro) may be achieved (Scheme 1).
Indeed, in the nonpolar benzene, the regio- and diastereo-
selectivities were very high in the [2+2] photocycloaddi-
Synthesis 2001, No. 8, 18 06 2001. Article Identifier:
1437-210X,E;2001,0,08,1203,1214,ftx,en;C02601ss.pdf.
© Georg Thieme Verlag Stuttgart · New York
ISSN 0039-7881
Scheme 1

1204
W. Adam, V. R. Stegmann
PAPER
Table 1 Regioselectivity in the [2+2] Photocycloaddition of Benzophenone (1) with the Achiral Allylic Substrates 4a–d
Entry
Substrate
R1
X
Time
[h]
Convn.a
[%]
mba,b
[%]
Regioselectivitya
5 : 5’
1
2
3
4
4a
4b
4c
4d
H
OH
H
5
4
5
6
58
55
58
59
91
90
88
92
>95 : 05
84 : 16
61 : 39
55 : 45
H
Me
Me
OH
OAc
a Determined by 1H NMR spectroscopy directly on the crude product mixture; error limits 5% of the given values.
b Mass balance (mb) based on the allylic substrates 4.
tion of benzophenone to the substrates 4e–h (Table 2, aligns the substrate conformationally for the regioselec-
entries 1–4); the threo-5e–h oxetanes were the strongly tive and threo diastereoselective attack (Scheme 2, top) of
preferred photoproducts. Moreover, the threo diastereose- the triplet-excited benzophenone, assisted by hydrogen
lectivity increased slightly with the steric demand of the bonding3a in the exciplex.7 Subsequently, the favored
R2 substituent (Me < Et < i-Pr), as is evident in Table 2 threo exciplex transforms to the triplet preoxetane diradi-
(entries 1–3). The t-Bu-substituted oxetane threo-5h was cal threo-3D, which finally cyclizes to the threo-5 oxetane
formed as the sole diastereomer, as revealed by 1H NMR as the essentially exclusive photoproduct. Since hydrogen
spectroscopy. In the presence of the protic methanol, how- bonding is reduced in the presence of the protic methanol
ever, the threo diastereoselectivity decreased significantly (competitive intermolecular hydrogen bonding) and since
for the allylic alcohol 4e (entry 5), and disappeared com- the silyl ether 4i cannot serve as hydrogen-bonding donor,
pletely for the allylic silyl ether 4i even in benzene (entry the diastereoselectivity expectedly decreases (Table 2, en-
6). Additionally, in the latter case, the regioisomer 5i’ was tries 5, 6).
formed as the minor photoadduct, the 5i:5i’ regioselectiv-
ity was 83:17 (Table 2, entry 6), which is in good accord
with the regioselectivity observed for the unfunctional-
ized substrate 4b (Table 1, entry 2).
For the unsymmetrical carbonyl partners acetophenone
(2) and benzaldehyde (3), an additional stereogenic center
is generated during the [2+2] photocycloaddition. Should
it be possible to control the incipient stereogenic center at
These selectivity data may be rationalized in terms of the the carbonyl carbon atom, this would considerably en-
1,3-allylic strain6 in the chiral allylic alcohols 4e–h, which hance the synthetic utility of the present [2+2] photocy-
Table 2 Selectivities in the [2+2] Photocycloaddition of Benzophenone (1) with the Chiral Allylic Substrates 4e–i
Entry
Substrated
X
R
Time
[h]
Convn.a
[%]
mba,b
[%]
Selectivitiesa
Diastereoc
Regio (5 : 5’)
(threo : erythro)
1
2
3
4
5
6
4e
4f
OH
OH
OH
OH
OH
Me
Et
28
20
20
20
32
32
90
90
89
92
85
84
82
95
91
88
81
85
90 : 10
93 : 07
95 : 05
>95 : 05
69 : 31
52 : 48
>95 : 05c
>95 : 05c
>95 : 05c
>95 : 05c
95 : 05c
4g
4h
4e
4i
i-Pr
t-Bu
Me
83 : 17e
t
OSiMe2 Bu Me
a See footnote a in Table 1.
b See footnote b in Table 1.
c The diastereomeric ratio of the major regioisomer 5 is given.
d In C6D6 except entry 5 for which a 1:1 mixture of C6D6 and CD3OD was used as solvent.
e The diastereomeric ratio of the minor regioisomer 5i’ is 78:22, whose relative configurations have not been determined.
Synthesis 2001, No. 8, 1203–1214 ISSN 0039-7881 © Thieme Stuttgart · New York

PAPER
Preparation of Oxetanes with up to Three Stereogenic Centers as Synthetic Building Blocks
1205
Scheme 3
Table 3
Selectivitiesa
Time Convn.a mba,b
Diastereoc
(cis : trans) a : a’
Regio
PhCOR
R
[h]
[%]
91
[%]
63
2
3
Me 24
>95 : 05
>95 : 05
>95 : 05
65 : 35d
H
7
72
90
a See footnote a Table 1.
b See footnote b in Table 1.
Scheme 2
cThe diastereomeric ratio of the major regioisomer 6a or 7a is given.
d The minor regioisomer 7a’ was formed as a pair of diastereomers.
cloaddition. To assess this possibility, the achiral allylic
alcohol 4a was taken as the olefin substrate (Scheme 3),
for which the so-called simple diastereoselectivity8 ap-
plies. Most gratifyingly, complete (cis : trans >95:5) sim-
ple diastereoselectivity was observed for both carbonyl
partners 2 and 3. Additionally, for acetophenone, also
the regioselectivity was very high (Table 3, entry 1),
analogous to benzophenone (Table 1, entry 1), but for
benzaldehyde it was much lower (Table 3, entry 2). It is
known,9 however, that the regioselectivity for the cy-
cloaddition of benzaldehyde to the unfunctionalized alk-
ene 4b is also poor (approx. 60:40), while benzophenone
reacts highly regioselectively (Table 1, entry 2). Presum-
ably, this reduced regioselectivity is general for benzalde-
hyde and cannot be compensated by the hydroxy-group
directivity.
prior to cyclization to the oxetane.14 Relevant conformers
3
of the D, which meet the afore-mentioned criteria, are
shown in Scheme 4 exemplarily for the cis-6a oxetane.
Steric interactions are minimized when the diradical as-
sumes the optimal conformation as depicted in the center
of Scheme 4. Intersystem-crossing and subsequent cy-
clization preferentially leads to the cis-6a oxetane, as is
experimentally observed.
For this pronounced cis diastereoselectivity, the interme-
diate triplet diradical 3D plays a decisive role, since in the
subsequent reaction step, i.e., the cyclization to the oxet-
ane, the pertinent stereogenic center is formed. The fol-
lowing four points need to be considered to rationalize the
cis stereocontrol:10 (i) Although the triplet preoxetane
diradical 3D has a relatively short lifetime ( ~ 1–7 ns),11
it is long-lived enough to allow conformational changes;
(ii) intersystem-crossing (ISC) from the triplet diradical
Scheme 4
3D to the singlet state proceeds predominantly through the The [2+2] photocycloaddition of the unsymmetrical car-
spin-orbit-coupling (SOC) mechanism;12 (iii) the extent of bonyl partners 2 and 3 with the chiral allylic alcohols 4e–
SOC is strongly dependent on the orientation of the 2p or- h provides the opportunity for threo- and cis-diastereose-
bitals at the radical sites, in fact, an orthogonal alignment lective oxetane formation (Table 4). The simple diastere-
is suited best for efficient ISC;13 (iv) once the ISC to the oselectivity was again very high in all cases, i.e., only the
putative singlet diradical has occurred, immediate oxetane cis-diastereomers of the oxetanes 6e–h and 7e–h were
formation (besides cleavage to the starting materials) is formed. For benzaldehyde, additionally the regioisomers
expected without further conformational changes. Hence, 7e’–h’ were obtained as a mixture of all four possible di-
from the product configuration we may conclude the pre- astereomers. The threo diastereoselectivity for the oxet-
ferred conformation of the triplet preoxetane diradical 3D anes 6e–h (Table 4, entries 1 4) and 7e–h (entries 5–8)
Synthesis 2001, No. 8, 1203–1214 ISSN 0039-7881 © Thieme Stuttgart · New York

1206
W. Adam, V. R. Stegmann
PAPER
significantly increases with increasing steric demand of
R2 in the chiral allylic alcohols, analogous to the results
observed before for benzophenone as the carbonyl partner
(Table 2, entries 1 4). The extent of threo diastereoselec-
tivity, however, is lower for the oxetanes 6 and 7 com-
pared to 5. Indeed, the threo diastereoselectivity decreases
for a particular chiral allylic alcohol 4, e.g. 4e, and the re-
spective carbonyl partners in the order 1 > 2 > 3 (compare
entry 1 in Table 2 with entries 1 and 5 in Table 4). Such
steric expression implicates a less pronounced hydroxy-
directing effect for the carbonyl partners 2 and 3 com-
pared to 1. This may be due to the efficacy of the 1,3-al-
lylic strain in the chiral allylic alcohol 4, which should
decrease as the steric interactions between the carbonyl
partners and the substrate diminish (Scheme 5). There-
Scheme 5
fore, increased amounts of the erythro-oxetanes are ob- lylic strain and prochiral carbonyl partners need to be em-
tained in the order 3 > 2 > 1, as observed (Table 4).
ployed.
In summary, on the basis of the hydroxy-directing effect,
a convenient methodology has been developed for the
highly regio- and diastereoselective synthesis of
sidechain-functionalized oxetanes with up to three stereo-
genic centers through the Paternò–Büchi photocycloaddi-
tion. For this highly stereoselective process, easily
accessible trisubstituted chiral allylic alcohols with 1,3-al-
Irradiations were conducted in Rayonet Photoreactors (RPR 3500 Å
or RPR 3000 Å). 1H and 13C NMR spectra were recorded in CDCl3
on a Bruker AC 200 (1H: 200 MHz; 13C: 50 MHz), a Bruker AC 250
(1H: 250 MHz; 13C: 63 MHz), or a Bruker DMX 600 (1H: 600 MHz)
spectrometer against CDCl3 or C6D6 as reference standard on the
scale (ppm). A Perkin–Elmer 1600 Series FTIR spectrophotometer
Table 4 Selectivities in the [2+2] Photocycloaddition of Acetophenone (2) and Benzaldehyde (3) to the Chiral Allylic Alcohols 4e–h.
Selectivitiesa
Entry
PhCOR
Substrate
R2
Time [h]
Convn.a
[%]
mba,b
[%]
Diastereoc
(threo: erythro)
Regiod
6 : 6’
1
2
3
4
2
2
2
2
4e
4f
Me
Et
12
4
79
34
62
76
83
99
81
80
82 : 18
88 : 12
92 : 08
>95 : 05
>95 : 05
>95 : 05
>95 : 05
>95 : 05
7 : 7’
4g
4h
i-Pr
t-Bu
12
17
5
6
7
8
3
3
3
3
4e
4f
Me
Et
3
3
3
3
72
70
58
52
85
86
96
96
72 : 28
81 : 19
91 : 09
>95 : 05
67 : 33
65 : 35
60 : 40
59 : 41
4g
4h
i-Pr
t-Bu
a See footnote a in Table 1.
b See footnote b in Table 1.
c The diastereomeric ratio of the major regioisomer 6 or 7 is given.
d The minor regioisomers 7e’–7h’ were formed as a set of four diastereomers in about equal amounts.
Synthesis 2001, No. 8, 1203–1214 ISSN 0039-7881 © Thieme Stuttgart · New York

PAPER
Preparation of Oxetanes with up to Three Stereogenic Centers as Synthetic Building Blocks
1207
was used for the IR spectra. TLC analysis was conducted on pre-
coated silica-gel foils 60 F254 (20 20 cm) from Merck, Darmstadt.
Spots were visualized either by UV irradiation (254 nm) or by a 5%
polymolybdic acid in EtOH. Silica gel (32–63 m) from Woelm, Er-
langen was used for flash chromatography. Solvents were dried by
standard methods and purified by distillation before use. Mps (un-
corrected) were determined on a Büchi Melting Point B-545 appa-
ratus. All commercial reagents were used without further
purification, unless otherwise stated.
52°C, 22 torr) yielded 36.1 g (71%) of the allylic alcohol 4e as a
colorless liquid.
1H NMR (250 MHz, CDCl3): = 1.15 (d, 3H, J = 6.3 Hz), 1.62 (d,
3H, J = 1.2 Hz), 1.65 (d, 3H, J = 1.2 Hz), 2.41 (br s, 1H), 4.48 (dq,
1H, J = 8.5 Hz, 6.3Hz), 5.13 (d, 1H, J = 8.5 Hz).
13C NMR (50 MHz, CDCl3): = 17.9 (q), 23.5 (q), 25.5 (q), 64.6 (d),
129.3 (d), 133.7 (s).
5-Methyl-4-hexen-3-ol (4f)17
Under an Ar atm, a solution of ethyl bromide (6.13 g, 56.3 mmol)
in Et2O (12 mL) was added dropwise to a suspension of magnesium
turnings (1.37 g, 56.3 mmol) in Et2O (12 mL) at such a rate that the
solvent refluxed smoothly. The reaction mixture was heated at re-
flux for 30 min and then a solution of 3-methyl-2-butenal (3.79 g,
45.0 mmol) in Et2O (18 mL) was added dropwise. After heating at
reflux for 4 h, the solution was cooled to r.t., followed by hydrolysis
with ice-cold aqueous NH4Cl (50 mL). The phases were separated;
the aqueous phase was extracted with Et2O (3 30 mL); the com-
bined Et2O phases were extracted with H2O (2 75 mL), and dried
(MgSO4). Evaporation (20°C, 15 torr) of the solvent yielded 4.41 g
(86%) of a yellow oil, which on fractional distillation (bp 63°C, 12
torr) afforded 3.05 g (59%) of allylic alcohol 4f as a colorless oil.
1H NMR (200 MHz, CDCl3): = 0.87 (dd, 3H, J = 7.4, 7.4 Hz),
1.36–1.63 (m, 2H), 1.50 (br s, 1H), 1.68 (d, 3H, J = 1.2 Hz), 1.72 (d,
3H, J = 1.2 Hz), 4.25 (ddd, 1H, J = 8.7, 6.4, 6.4 Hz), 5.14 (d, 1H,
J = 8.7 Hz).
13C NMR (50 MHz, CDCl3): = 9.7 (q), 18.2 (q), 25.8 (q), 30.5 (t),
70.1 (d), 127.9 (d), 135.3 (s).
Allylic Substrates 4
2,3-Dimethyl-2-buten-1-yl Acetate (4d)
Caution! Selenium dioxide is toxic. The resulting selenium-contain-
ing organic by-products produce an appalling stench. Therefore, the
reaction and the workup must be carried out in an efficient hood.
In a 250-mL, round-bottomed flask were placed 2,3-dimethyl-2-
butene (3.55 g, 42.2 mmol) in acetic anhydride (150 mL) and sele-
nium dioxide (2.34 g, 21.1 mmol) as solid. After heating to 80°C for
3 h, the mixture was distilled (56–58°C / 15 mbar) and to the distil-
late, which contained acetic anhydride and product 4d, was added
the same volume of H2O and the mixture was stirred 8–10 h at r.t.
(approx. 20°C). Sat. NaHCO3 (approx. 500 mL) was carefully (gas
formation!) added until the formation of a second phase was ob-
served. The organic phase was separated and the aqueous phase was
extracted with Et2O (3 50 mL). The combined organic phases
were washed with sat. NaHCO3 (4 80 mL) until no further gas for-
mation was observed, dried (MgSO4), and the solvent was evaporat-
ed (20°C, 15 Torr). Flash chromatography on silica gel (120 g;
petroleum ether–Et2O, 20:1) yielded 1.23 g (41%) of the allylic ac-
etate 4d as a colorless oil.
2,5-Dimethyl-4-hexen-3-ol (4g)18
IR (NaCl): = 2980, 2905, 1735, 1445, 1375, 1270, 1230, 1150,
1015, 965, 900 cm–1.
1H NMR (200 MHz, CDCl3): = 1.71 (s, 3H), 1.71 (s, 3H), 1.74 (s,
3H), 2.06 (s, 3H), 4.58 (s, 2H).
13C NMR (50 MHz, CDCl3): = 16.7 (q), 20.2 (q), 20.9 (q), 21.0 (q),
65.7 (t), 122.7 (s), 132.0 (s), 171.4 (s).
The preparation and work-up procedure was conducted analogously
to that of 5-methyl-4-hexen-3-ol (4f) by starting from isopropyl bro-
mide (6.15 g, 50.0 mmol), magnesium turnings (1.22 g, 50.0 mmol),
and 3-methyl-2-butenal (3.28 g, 40.0 mmol). The crude product
(4.13 g, 81%) was fractionally distilled (bp 59 °C, 12 Torr) to yield
3.23 g (63%) of the allylic alcohol 4g as a colorless oil.
Anal. Calcd for C8H14O2 (142.2): C, 67.57; H, 9.92. Found: C,
67.31; H, 9.44.
1H NMR (200 MHz, CDCl3): = 0.84 (d, 3H, J = 6.8 Hz), 0.94 (d,
3H, J = 6.7 Hz), 1.39 (br s, 1H), 1.65 (m, 1H), 1.67 (d, 3H, J = 0.6
Hz), 1.73 (d, 3H, J = 1.1 Hz), 4.03 (dd, 1H, J = 6.8, 9.0 Hz), 5.17 (d,
1H, J = 9.0 Hz).
2,3-Dimethyl-2-buten-1-ol (4c)15
To a solution of 2,3-dimethyl-2-butenyl acetate (4d, 737 mg, 5.18
mmol) in MeOH (50 mL) was added solid K2CO3 (7.16 g, 51.8
mmol). The mixture was stirred for 20 h at r.t. until complete con-
version was indicated by TLC. H2O (50 mL) was added and the
mixture was extracted with Et2O (3 70 mL). The combined organ-
ic phases were washed with H2O (3 50 mL) and sat. NaCl (50 mL),
dried (MgSO4), and evaporated (20°C, 15 Torr). Purification of the
residue by flash chromatography on silica gel (50 g; petroleum
ether–Et2O, 10:1) yielded 281 mg (54%) of the 2,3-dimethyl-2-
buten-1-ol (4c) as a colorless oil.
1H NMR (200 MHz, CDCl3): = 1.08 (s, 1H), 1.69 (s, 3H), 1.74 (s,
6H), 4.13 (s, 2H).
13C NMR (63 MHz, CDCl3): = 16.5 (q), 19.9 (q), 20.8 (q), 63.9 (t),
127.3 (s), 129.3 (s).
13C NMR (63 MHz, CDCl3): = 18.0 (q), 18.3 (q), 18.3 (q), 25.9 (q),
34.4 (d), 73.7 (d), 126.3 (d), 135.6 (s).
2,2,5-Trimethyl-4-hexen-3-ol (4h)16
Under an Ar atm, 3-methyl-2-butenal (2.63 g, 31.3 mmol) were dis-
solved in dry Et2O (60 mL) and cooled to 0°C. Within 30 min, tert-
BuLi (1.5 M, 25 mL) were added dropwise and the reaction mixture
was stirred for 4 h at 0°C. Then, sat. NH4Cl (30 mL) was adminis-
tered carefully, the phases were separated, and the aqueous phase
was extracted with Et2O (3 30 mL). The combined organic phases
were washed with sat. NaCl (3 15 mL), dried (MgSO4), and the
solvent was evaporated (20°C, 15 Torr). The residue was purified
by silica gel chromatography (petroleum ether–Et2O, 10:1) to afford
1.80 g (40%) of the allylic alcohol 4h as pale yellow oil.
1H NMR (200 MHz, CDCl3): = 0.88 (s, 9H), 1.35 (br s, 1H), 1.67
(d, 3H, J = 1.2 Hz), 1.73 (d, 3H, J = 1.2 Hz), 3.99 (d, 1H, J = 9.3 Hz),
5.22 (d, 1H, J = 9.3 Hz).
13C NMR (63 MHz, CDCl3): = 18.4 (q), 25.5 (q), 26.1 (q), 35.3 (s),
76.0 (d), 124.8 (d), 135.9 (s).
4-Methyl-3-penten-2-ol (4e)16
Under a N2 atm, a solution of 4-methyl-3-penten-2-one (50.0 g, 509
mmol) in dry Et2O (100 mL) was added dropwise to a suspension of
LiAlH4 (9.75 g, 257 mmol) in dry Et2O. After stirring for 1 h at r.t.,
the reaction was stopped by carefully adding ice water (15 mL),
15% NaOH (15 mL), and H2O (50 mL). The phases were separated,
the aqueous phase was extracted with Et2O (3 50 mL), the com-
bined organic phases were dried (MgSO4), and the solvent was
evaporated (20°C, 30 Torr). Fractional distillation of the residue (bp
2-(tert-Butyldimethylsilyloxy)-4-methyl-3-pentene (4i)19
At r.t., a solution of 4-methyl-3-penten-2-ol (4e, 8.00 g, 80.0 mmol)
in CH2Cl2 (20 mL) was added slowly to a solution of tert-butyldim-
ethylsilyl chloride (12.8 g, 85.0 mmol) and imidazole (7.60 g, 112
Synthesis 2001, No. 8, 1203–1214 ISSN 0039-7881 © Thieme Stuttgart · New York

1208
W. Adam, V. R. Stegmann
PAPER
mmol) in CH2Cl2 (80 mL). After stirring for 12 h, the reaction mix-
ture was hydrolyzed by adding H2O (100 mL). The phases were sep-
arated, the aqueous phase was extracted with CH2Cl2 (3 50 mL),
the combined organic phases were washed with a 10% HCl (100
mL), H2O (100 mL), and sat. NaCl (10 mL), and dried (MgSO4).
The solvent was evaporated (20°C, 15 Torr) and the residue was
distilled fractionally (bp 73°C, 10 Torr) to yield 14.9 g (87%) of the
allylic substrate 4i as a colorless oil.
1H NMR (200 MHz, CDCl3): = 0.03 (s, 3H), 0.04 (s, 3H), 0.88 (s,
9H), 1.16 (d, 3H, J = 6.2 Hz), 1.61 (d, 3H, J = 1.3 Hz), 1.67 (d, 3H,
J = 1.2 Hz), 4.51 (dq, 1H, J = 8.3, 6.2 Hz), 5.16 (d, 1H, J = 8.3 Hz).
13C NMR (50 MHz, CDCl3): = 21.0 (q), 26.7 (q), 45.0 (s), 63.0 (t),
85.2 (d), 91.6 (s), 124.9 (d), 125.3 (d), 126.5 (d), 126.6 (d), 127.8
(d), 128.0 (d), 144. 0 (s), 144. 8 (s).
Anal. Calcd for C18H20O2 (268.4): C, 80.56; H, 7.51. Found: C,
80.28; H, 7.36.
Irradiation of 2-Methyl-2-butene (4b) and Benzophenone (1)
A solution of 2-methyl-2-butene (4b, 526 mg, 7.50 mmol) and ben-
zophenone (683 mg, 3.75 mmol) in benzene (15 mL) was irradiated
for 38 h according to the above general procedure. Flash chroma-
tography on silica gel (130 g; PE–Et2O, 100:1) yielded first 550 mg
(58%) of 5b (mp 108–110°C) and then 120 mg (13%) of 5b’ (mp
77–78°C) as colorless powders.
13C NMR (50 MHz, CDCl3): = –4.7 (q), –4.4 (q), 17.9 (s), 18.3 (q),
24.8 (q), 25.6 (q), 25.9 (q), 66.0 (d), 130.1 (s), 130.6 (d).
3,3,4-Trimethyl-2,2-diphenyloxetane (5b)
IR (KBr): = 3100, 3020, 2905, 1620, 1510, 1485, 1475, 1460,
1405, 1085, 1005, 935 cm–1.
1H NMR (200 MHz, CDCl3): = 1.07 (s, 3H), 1.09 (s, 3H), 1.29 (d,
Photolyses of the Carbonyl Compounds 1–3 and the Allylic
Substrates 4 on the Analytical Scale; General Procedure
Samples of the allylic substrates 4 (200–250 mol) and the respective
carbonyl compound 1 3 (200–250 mol) were dissolved in of d6-
benzene (1.0 mL) or in a mixture of d6-benzophenone and d4-meth-
anol (1:1, 1.0 mL) and transferred to a NMR tube. In the case of
benzophenone (1), Pyrex NMR tubes were employed while Quartz
ones were used for acetophenone (2) and benzaldehyde (3) as car-
bonyl partners. The samples were degassed for 5 min with a steady
3H, J = 6.3 Hz), 4.58 (q, 1H, J = 6.3 Hz), 7.16–7.68 (m, 10H).
13C NMR (50 MHz, CDCl3): = 16.8 (q), 21.2 (q), 25.6 (q), 45.5 (s),
81.6 (d), 91.0 (s), 125.0 (d), 125.6 (d), 126.2 (d), 126.3 (d), 127.6
(d), 127.9 (d), 144.3 (s), 145.3 (s).
1
Anal. Calcd for C18H20O (252.4): C, 85.67; H, 7.99. Found: C,
85.34; H, 7.77.
stream of Ar gas, the NMR tubes were sealed, and the H NMR
spectrum was recorded. The NMR samples were placed into a vac-
uum-jacketed Pyrex or Quartz tube filled with EtOH, cooled to
10°C by means of a coldfinger, and irradiated in a Rayonet Photo-
reactor (RPR 3500 Å or RPR 3000 Å). The photolysates were di-
rectly analyzed by 1H NMR spectroscopy, to determine the
diastereomeric ratios from the relative areas of the relevant 1H NMR
peaks. The mass balances and conversions were calibrated against
the internal standard or versus the total area of the aromatic signals.
The results are given in Tables 1– 3 and Scheme 3.
3,4,4-Trimethyl-2,2-diphenyloxetane (5b’)
IR (KBr): = 3110, 3010, 2970, 1620, 1510, 1465, 1385, 1170,
1010, 970, 875 cm–1.
1H NMR (200 MHz, CDCl3): = 0.93 (d, 3H, J = 7.5 Hz), 1.33 (s,
3H), 1.39 (s, 3H), 3.40 (q, 1H, J = 7.5 Hz), 7.16–7.47 (m, 10H).
13C NMR (50 MHz, CDCl3): = 12.8 (q), 23.8 (q), 31.2 (q), 47.3 (d),
81.5 (s), 85.5 (s), 124. 9 (d), 125.9 (d), 126.3 (d), 126.4 (d), 127.7
(d), 127.9 (d), 143.8 (s), 149.2 (s).
Photolyses of the Carbonyl Compounds 1–3 and the Allylic
Substrates 4 on the Preparative Scale; General Procedure
The allylic substrates 4 and the appropriate carbonyl compound 1–
3 were dissolved in benzene, the solution transferred to a vacuum-
jacketed Pyrex [for benzophenone (1)] or Quartz [for acetophenone
(2) and benzaldehyde (3)] tube and degassed for at least 15 min with
a steady stream of Ar gas. The reaction mixture was cooled to 10°C
by means of a coldfinger and irradiated in a Rayonet Photoreactor
(RPR 3500 Å or RPR 3000 Å). The solvent was evaporated (20°C,
15 Torr) and the residue was purified by crystallization, Kugelrohr
distillation or by silica gel flash chromatography with mixtures of
petroleum ether (PE) and Et2O as eluent.
Anal. Calcd for C18H20O (252.4): C, 85.67; H, 7.99. Found:C,
85.28; H, 8.00.
Irradiation of 2,3-Dimethyl-2-buten-1-ol (4c) and Benzo-
phenone (1)
A solution of 2,3-dimethyl-2-buten-1-ol (4c, 239 mg, 2.39 mmol)
and benzophenone (435 mg, 2.39 mmol) in benzene (10 mL) was ir-
radiated for 44 h according to the above general procedure. Flash
chromatography on silica gel (70 g; PE–Et2O, 10:1) yielded first 92
mg (14%) of 5c’(mp 144–146°C), then 165 mg (24%) of a mixture
of the regioisomers 5c and 5c’ and, finally, 47.0 mg (7%) of 5c (mp
108–110°C) as white powders.
Irradiation of 3-Methyl-2-penten-1-ol (4a) and Benzophenone
(1)
2,3,3-Trimethyl-4,4-diphenyloxetan-2-ylmethanol (5c)
IR (KBr): = 3445, 3060, 2990, 2870, 1595, 1490, 1465, 1445,
1370, 1045, 940 cm–1.
1H NMR (200 MHz, C6D6): = 0.93 (s, 3H), 1.04 (s, 3H), 1.25 (s,
3H), 3.37 (dd, 1H, J = 10.9, 6.6 Hz), 3.49 (dd, 1H, J = 10.9, 5.0 Hz),
6.94–7.15 (m, 6H), 7.48–7.61 (m, 4H).
13C NMR (50 MHz, CDCl3): = 20.1 (q), 22.0 (q), 24.5 (q), 46.3 (s),
67.1 (t), 84.9 (s), 90.0 (s), 125.0 (d), 125.2 (d), 126.2 (d), 126.4 (d),
127.7 (d), 127.9 (d), 145.6 (s), 145.7 (s).
A solution of 4-methyl-3-penten-2-ol (4a, 1.29 g, 15.0 mmol) and
benzophenone (1.37 g, 7.50 mmol) in benzene (30 mL) was irradi-
ated for 38 h according to the above general procedure. The crude
photolysate was submitted to Kugelrohr distillation (50°C, 0.06
Torr) to remove the excess of 4a. Further distillation (170°C, 0.06
Torr) yielded 1.15 g (57%) of 5a as a colorless oil, which solidified
upon standing. Recrystallization from CH2Cl2–PE (2:5) afforded
855 mg (42%) of the oxetane 5a as a white powder (mp 133–
135°C).
Anal. Calcd for C19H22O2 (282.4): C, 80.82; H, 7.85. Found: C,
80.47; H, 7.99.
3,3-Dimethyl-4,4-diphenyloxetan-2-ylmethanol (5a)
IR (KBr): = 3500, 3055, 1595, 1485, 1465, 1445, 1035, 975, 835,
770, 755, 710 cm–1.
1H NMR (200 MHz, CDCl3): = 1.10 (s, 3H), 1.14 (s, 3H), 1.89 (br
s, 1H), 3.65–3.90 (m, 2H), 4.47 (dd, 1H, J = 6.9, 4.4 Hz), 7.15–7.60
(m, 10H).
2,2,3-Trimethyl-4,4-diphenyloxetan-3-ylmethanol (5c’)
IR (KBr): = 3475, 3065, 2990, 2870, 1595, 1490, 1475, 1445,
1390, 1030, 920, 715 cm–1.
Synthesis 2001, No. 8, 1203–1214 ISSN 0039-7881 © Thieme Stuttgart · New York

PAPER
Preparation of Oxetanes with up to Three Stereogenic Centers as Synthetic Building Blocks
1209
1H NMR (200 MHz, C6D6): = 1.13 (s, 3H), 1.18 (s, 3H), 1.25 (s,
3H), 3.19 (dd, 1H, J = 10.7, 4.6 Hz), 3.50 (dd, 1H, J = 10.7, 4.6 Hz),
6.98–7.18 (m, 6H), 7.66–7.82 (m, 4H).
13C NMR (50 MHz, CDCl3): = 17.9 (q), 25.1 (q), 26.4 (q), 50.6 (s),
65.7 (t), 82.9 (s), 88.6 (s), 125.5 (d), 125.6 (d), 126.2 (d), 126.5 (d),
127.6 (d), 127.9 (d), 145. 1 (s), 145.3 (s).
(R*,S*)-1-(3,3-Dimethyl-4,4-diphenyloxetan-2-yl)-ethanol
(erythro-5e)
1H NMR (200 MHz, C6D6): = 1.07 (s, 3H), 1.18 (s, 3H), 1.21 (d,
3H, J = 5.9 Hz), 3.73–3.88 (m, 1H), 3.91 (d, 1H, J = 9.0 Hz), 7.00–
7.20 (m, 6H); only the resolved signals are reported.
13C NMR (50 MHz, CDCl3): = 19.5 (q), 21.3 (q), 27.0 (q), 45.3 (s),
67.5 (d), 87.9 (d), 90.4 (s), 124.9 (d), 125.4 (d), 126.3 (d), 126.4 (d),
127.7 (d), 127.9 (d), 144.0 (s), 144.9 (s).
Anal. Calcd for C19H22O2 (282.4): C, 80.82; H, 7.85. Found: C,
80.81; H, 7.85.
For the mixture of diastereomers (92:8):
Irradiation of 2,3-Dimethyl-2-buten-1-yl acetate (4d) and
Benzophenone (1)
IR (KBr): = 3455, 3060, 2970, 1600, 1450, 1390, 1070, 1000, 930,
885 cm–1.
A solution of 2,3-dimethyl-2-buten-1-yl acetate (4d, 356 mg, 2.50
mmol) and benzophenone (456 mg, 2.50 mmol) in benzene (10 mL)
was irradiated for 38 h according to the above general procedure.
Flash chromatography on silica gel (130 g; PE–Et2O, 20:1) yielded
427 mg (53%) of a regioisomeric mixture of 5d and 5d’ which was
not separable.
Anal. Calcd for C19H22O2 (282.3): C, 80.82; H, 7.85. Found: C,
80.56; H, 7.83.
Assignment of the Relative threo Configuration
(R*,R*)-1-[3,3-Dimethyl-4,4-diphenyloxetan-2-yl]-ethyl-4-
nitrobenzoate (p-Nitrobenzoate of threo-5e)
For the full set of NMR spectroscopic data, authentic samples of 5d
and 5d’were prepared from 5c and 5c’by acetylation with acetic an-
hydride / pyridine in CH2Cl2.
A diastereomeric mixture (approx. 92:8) of threo,erythro-5e (332
mg, 1.18 mmol) was dissolved in pyridine (1.2 mL) and 4-nitroben-
zoyl chloride (219 mg, 1.18 mmol) was added as a solid. After stir-
ring the reaction mixture for 4 h at r.t., toluene (2 3 mL) was added
and the pyridine was removed by azeotropic distillation (40°C, 15
torr). The residue was dissolved in CHCl3 (5 mL), silica gel (2 g)
was added (to facilitate the transfer to the chromatography column),
and the solvent was evaporated (20°C, 15 Torr). The residue was
purified by flash chromatography on silica gel (60 g; PE–Et2O,
10:1) to yield 430 mg (84%) of diastereomerically pure (by NMR
analysis) of the p-nitrobenzoate (Rf = 0.23) as a pale yellow solid
(mp 132–133°C). Recrystallization from EtOAc afforded pale yel-
low prisms, which were submitted to X-ray analysis.20
2,3,3-Trimethyl-4,4-diphenyloxetan-2-ylmethyl acetate (5d)
1H NMR (200 MHz, CDCl3): = 1.13 (s, 3H), 1.16 (s, 3H), 1.33 (s,
3H), 2.07 (s, 3H), 4.10 (d, 1H, J = 11.6 Hz), 4.17 (d, 1H, J = 11.6
Hz), 7.16–7.32 (m, 6H), 7.50–7.56 (m, 4H).
13C NMR (50 MHz, CDCl3): = 20.6 (q), 20.9 (q), 22.3 (q), 24.1 (q),
46.5 (s), 68.5 (t), 83.1 (s), 90.0 (s), 125.2 (d), 125.2 (d), 126.3 (d),
126.4 (d), 127.7 (d), 127.8 (d), 145.2 (s), 145.4 (s), 171.0 (s).
2,2,3-Trimethyl-4,4-diphenyloxetan-3-ylmethyl acetate (5d’)
1H NMR (200 MHz, CDCl3): = 1.16 (s, 3H), 1.30 (s, 3H), 1.36 (s,
3H), 2.07 (s, 3H), 4.03 (d, 1H, J = 11.4 Hz), 4.24 (d, 1H, J = 11.4
Hz), 7.13–7.34 (m, 6H), 7.50–7.62 (m, 4H).
13C NMR (50 MHz, CDCl3): = 18.5 (q), 21.0 (q), 25.5 (q), 26.3 (q),
48.9 (s), 67.5 (t), 82.7 (s), 88.0 (s), 125.3 (d), 125.4 (d), 126.4 (d),
126.7 (d), 127.7 (d), 127.9 (d), 144.5 (s), 145.0 (s), 170.8 (s).
IR (KBr): = 3055, 3000, 2870, 1725, 1530, 1345, 1280, 1125, 990,
975, 875, 845 cm–1.
1H NMR (200 MHz, CDCl3): = 1.15 (s, 3H), 1.17 (s, 3H), 1.30 (d,
3H, J = 6.3 Hz), 4.42 (d, 1H, J = 7.9 Hz), 5.41 (dq, 1H, J = 6.3, 7.9
Hz), 7.17–7.63 (m, 10H), 8.15–8.31 (m, 4H).
13C NMR (50 MHz, CDCl3): = 16.4 (q), 21.3 (q), 26.7 (q), 45.5 (s),
72.1 (d), 85.6 (d), 90.5 (s), 123.5 (d), 125.1 (d), 125.5 (d), 126.6 (d),
126.7 (d), 127.8 (d), 128.0 (d), 130.9 (d), 136.0 (s), 143.6 (s), 144.1
(s), 150.4 (s), 164.3 (s).
For the mixture of regioisomers 5d and 5d’:
IR (KBr): = 3085, 2980, 1740, 1490, 1475, 1380, 1245, 1035,
1010, 710 cm–1.
Anal. Calcd for C21H24O3 (324.4): C, 77.75; H, 7.46. Found: C,
77.47; H, 7.27.
Anal. Calcd for C26H25NO5 (431.5): C, 72.37; H, 5.84; N, 3.25.
Found: C, 71.99; H, 5.85; N, 3.15.
Irradiation of 4-Methyl-3-penten-2-ol (4e) and Benzophenone
(1)
Irradiation of 5-Methyl-4-hexen-3-ol (4f) and Benzophenone (1)
A solution of 5-methyl-4-hexen-3-ol (4f, 400 mg, 3.50 mmol) and
benzophenone (638 mg, 3.50 mmol) in benzene (15 mL) was irra-
diated for 24 h according to the above general procedure. Flash
chromatography on silica gel (70 g; PE–Et2O, 5:1) yielded 757 mg
(73%) of a diastereomeric mixture (approx. 94:6) of threo,erythro-
5f as a colorless, sticky oil, which solidified within 3 days (mp 76–
77 °C) on standing at r.t. (approx. 20°C). An enriched sample of the
erythro-diastereomer (minor diastereomer) was obtained upon re-
peated silica gel chromatography.
A solution of 4-methyl-3-penten-2-ol (4e, 350 mg, 3.50 mmol) and
benzophenone (638 mg, 3.50 mmol) in benzene (15 mL) was irra-
diated for 18 h according to the above general procedure. Flash
chromatography on silica gel (150 g; PE–Et2O, 2:1) yielded 679 mg
(69%) of a diastereomeric mixture (approx. 92:8) of threo,erythro-
5e as a colorless, sticky oil, which solidified within 3 days (mp 85–
86 °C) on standing at r.t. An enriched sample of the erythro diaste-
reomer (minor isomer) was obtained upon repeated silica gel chro-
matography.
(R*,R*)-1-(3,3-Dimethyl-4,4-diphenyloxetan-2-yl)-propanol
(threo-5f)
(R*,R*)-1-(3,3-Dimethyl-4,4-diphenyloxetan-2-yl)-ethanol (threo-
5e)
1H NMR (200 MHz, C6D6): = 0.84 (s, 3H), 0.95 1.02 (m, 6H),
1.15–1.26 (m, 2H), 2.46 (br s, 1H), 3.52–3.63 (m, 1H), 4.10 (d, 1H,
J = 7.9 Hz), 6.96–7.20 (m, 6H), 7.40–7.45 (m, 2H), 7.62–7.66 (m,
2H).
1H NMR (200 MHz, C6D6): = 0.82 (s, 3H), 0.89 (s, 3H), 0.90 (d,
3H, J = 6.1 Hz), 2.53 (br s, 1H), 3.73–3.88 (m, 1H), 4.00 (d, 1H,
J = 8.4 Hz), 6.99–7.20 (m, 6H), 7.39–7.43 (m, 2H), 7.60–7.64 (m,
2H).
13C NMR (50 MHz, CDCl3): = 9.7 (q), 21.3 (q), 25.9 (t), 26.6 (q),
45.0 (s), 72.4 (d), 86.8 (d), 90.8 (s), 124.8 (d), 125.5 (d), 126.5 (d),
126.6 (d), 127.8 (d), 128.0 (d), 143.7 (s), 144.6 (s).
13C NMR (50 MHz, CDCl3): = 18.2 (q), 21.2 (q), 26.5 (q), 44.7 (s),
67.5 (d), 88.4 (d), 90.7 (s), 124.8 (d), 125.5 (d), 126.5 (d), 126.6 (d),
127.8 (d), 128.0 (d), 143.7 (s), 144.5 (s).
Synthesis 2001, No. 8, 1203–1214 ISSN 0039-7881 © Thieme Stuttgart · New York

1210
W. Adam, V. R. Stegmann
PAPER
(R*,S*)-1-(3,3-Dimethyl-4,4-diphenyloxetan-2-yl)-propanol
(erythro-5f)
(R*,R*)-1-(3,3-Dimethyl-4,4-diphenyloxetan-2-yl)-2,2-
dimethylpropan-1-ol (threo-5h)
1H NMR (200 MHz, C6D6): = 0.57 (br s, 1H), 0.92 (dd, 3H, J = 7.5,
7.3 Hz), 1.07 (s, 3H), 1.18 (s, 3H), 1.25–1.42 (m, 1H), 1.78–1.98
(m, 1H), 3.49–3.62 (m, 1H), 4.00 (d, 1H, J = 9.3 Hz), 6.96–7.21 (m,
6H), 7.45–7.50 (m, 2H), 7.64–7.68 (m, 2H).
IR (KBr): = 3570, 3085, 2955, 1595, 1480, 1390, 995, 965, 950
cm–1.
1H NMR (200 MHz, C6D6): 0.84 (s, 3H), 0.99 (s, 9H), 1.22 (s, 3H),
3.12 (d, 1H, J = 7.6 Hz), 3.19 (dd, 1H, J = 7.6, 1.5 Hz), 4.58 (d, 1H,
J = 1.5 Hz), 6.96–7.20 (m, 6H), 7.31–7.36 (m, 2H), 7.60–7.65 (m,
2H).
13C NMR (50 MHz, CDCl3): = 21.5, 25.8 (q), 26.3, 34.8, 46.2 (s),
74.8 (d), 82.7 (d), 92.0 (s), 124.5 (d), 125.3 (d), 126.5 (d), 126.6 (d),
128.0 (d), 128.1 (d), 143.9 (s), 144.4 (s).
13C NMR (50 MHz, CDCl3): = 9.3 (q), 21.4 (q), 26.0 (t), 27.0 (q),
45.4 (s), 72.6 (d), 86.4 (d), 90.4 (s), 124.9 (d), 125.4 (d), 126.3 (d),
126.5 (d), 127.7 (d), 128.0 (d), 144.1 (s),
144.9 (s).
For the mixture of diastereomers (94:6):
IR (KBr): = 3310, 3055, 2960, 2875, 1600, 1490, 1465, 1385, 995,
970, 900 cm .
Anal. Calcd for C22H28O2 (324.5): C, 81.44; H, 8.70. Found: C,
81.18; H, 8.49.
1
Anal. Calcd for C20H24O2 (296.4): C, 81.04; H, 8.16. Found: C,
81.00; H, 7.96.
Irradiation of 2-(tert-Butyldimethylsilyloxy)-4-methyl-3-
pentene (4i) and Benzophenone (1)
A solution of 2-(tert-butyldimethylsilyloxy)-4-methyl-3-pentene
(4i, 536 mg, 2.50 mmol) and benzophenone (901 mg, 4.94 mmol)
in benzene (15 mL) was irradiated for 24 h according to the above
general procedure. Flash chromatography on silica gel (160 g; PE–
Et2O, 100:1) yielded 550 mg (55%) of a diastereomeric mixture
(approx. 50:50) of threo,erythro-5i (Rf = 0.15) as a colorless oil.
Further elution yielded 60.1 mg (6%) of the regioisomer 5i .
Irradiation of 2,5-Dimethyl-4-hexen-3-ol (4g) and Benzo-
phenone (1)
A solution of 2,5-dimethyl-4-hexen-3-ol (4g, 449 mg, 3.50 mmol)
and benzophenone (765 mg, 4.20 mmol) in benzene (15 mL) was ir-
radiated for 24 h according to the above general procedure. Flash
chromatography on silica gel (70 g; PE–Et2O, 10:1) yielded 759 mg
(70%) of a diastereomeric mixture (approx. 95:5) of threo,erythro-
5g as a colorless, sticky oil, which solidified within 3 days (mp 95–
96 °C) on standing at r.t. (approx. 20°C). An enriched sample of the
erythro-diastereomer (minor isomer) was obtained upon repeated
silica gel chromatography.
(R*,R*)-[1-(3,3-Dimethyl-4,4-diphenyloxetan-2-yl)ethoxy](1,1-
dimethylethyl)dimethylsilane (threo-5i)
1H NMR (200 MHz, C6D6): = 0.24 (s, 3H), 0.43 (s, 3H), 0.82 (s,
3H), 0.85 (d, 3H, J = 6.1 Hz), 0.96 (s, 3H), 1.13 (s, 9H), 3.94 (dq,
1H, J = 9.0, 6.1 Hz), 4.13 (d, 1H, J = 9.0 Hz), 6.90–7.20 (m, 6H),
7.46–7.50 (m, 2H), 7.70–7.75 (m, 2H).
(R*,R*)-1-(3,3-Dimethyl-4,4-diphenyloxetan-2-yl)-2-methyl-
propan-1-ol (threo-5g)
1H NMR (200 MHz, C6D6): = 0.85 (s, 3H), 0.96 (d, 3H, J = 6.6
Hz), 0.99 (d, 3H, J = 6.4 Hz), 1.06 (s, 3H), 1.41–1.56 (m, 1H), 2.49
(d, 1H, J = 5.2 Hz), 3.49 (m, 1H), 4.34 (d, 1H, J = 6.1 Hz), 7.01–7.18
(m, 6H), 7.37–7.41 (m, 2H), 7.61–7.66 (m, 2H).
13C NMR (50 MHz, CDCl3): = –4.7 (q), –4.2 (q), 18.4 (s), 19.7 (q),
21.0 (q), 25.3 (q), 25.9 (q), 45.2 (s), 68.9 (d), 87.7 (d), 89.2 (s),
125.0 (d), 125.9 (d), 126.2 (d), 126.4 (d), 127.6 (d), 128.0 (d), 143.7
(s), 145.0 (s).
13C NMR (50 MHz, CDCl3): = 16.2 (q), 19.5(q), 21.5 (q), 26.4 (q),
31.4 (d), 45.5 (s), 73.8 (d), 85.0 (d), 91.1 (s), 124.7 (d), 125.5 (d),
126.5 (d), 126.6 (d), 127.9 (d), 128.1 (d), 143.8 (s), 144.5 (s).
(R*,S*)-[1-(3,3-Dimethyl-4,4-diphenyloxetan-2-yl)ethoxy](1,1-
dimethylethyl)dimethylsilane (erythro-5i)
1H NMR (200 MHz, C6D6): = –0.06 (s, 3H), 0.00 (s, 3H), 0.89 (s,
9H), 1.12 (s, 3H), 1.20 (s, 3H), 1.32 (d, 3H, J = 5.8 Hz), 3.97 (dq,
1H, J = 9.0, 5.8 Hz), 4.11 (d, 1H, J = 9.0 Hz), 6.95–7.22 (m, 6H),
7.47–7.52 (m, 2H), 7.68–7.72 (m, 2H).
(R*,S*)-1-(3,3-Dimethyl-4,4-diphenyloxetan-2-yl)-2-methyl-
propan-1-ol (erythro-5g)
1H NMR (200 MHz, C6D6): = 0.94 (d, 3H, J = 7.0 Hz), 0.99 (d, 3H,
J = 6.9 Hz), 1.08 (s, 3H), 1.18 (s, 3H), 2.04–2.18 (m, 1H), 3.55 (dd,
1H, J = 9.4, 3.0 Hz), 4.23 (d, 1H, J = 9.4 Hz), 6.96–7.21 (m, 6H),
7.46–7.50 (m, 2H), 7.66–7.70 (m, 2H).
13C NMR (50 MHz, CDCl3): = –4.6 (q), –3.1 (q), 17.8, 19.7, 21.3,
25.8, 26.4, 45.4, 68.1, 87.7, 89.9, 124.9, 125.6, 126.2, 126.4, 127.6,
127.9, 144.1, 145.0
13C NMR (50 MHz, CDCl3): = 15.5, 19.2, 21.3, 26.8, 28.9, 45.5,
74.8, 84.3, 90.4, 124.9, 125.4, 126.5, 127.7, 128.0, 144.0, 144.9;
only the fully resolved signals are reported.
For the mixture of diastereomers (approx. 50:50):
IR (neat): = 3060, 3025, 2955, 2855, 1600, 1465, 1255, 1105,
1000, 970, 835 cm–1.
For the mixture of diastereomers (95:5):
Anal. Calcd for C25H36O2Si (396.7): C, 75.70; H, 9.14. Found: C,
75.44; H, 9.41.
IR (KBr): = 3485, 3065, 3020, 2970, 2900, 1600, 1465, 1385, 990,
855 cm–1.
[1-(4,4-Dimethyl-2,2-diphenyloxetan-3-yl)ethoxy](1,1-
dimethylethyl)dimethylsilane (5i )
Major Diastereomer (relative configuration not known):
Anal. Calcd for C21H26O2 (310.4): C, 81.25; H, 8.44. Found: C,
80.91; H, 8.29.
Irradiation of 2,2,5-Trimethyl-4-hexen-3-ol (4h) and Benzo-
phenone (1)
1H NMR (200 MHz, C6D6): = –0.09 (s, 3H), –0.08 (s, 3H), 0.90 (s,
9H), 1.03 (d, 3H, J = 5.8 Hz), 1.42 (s, 3H), 1.71 (s, 3H), 3.26 (d, 1H,
J = 10.8 Hz), 3.97 (dq, 1H, J = 10.8, 5.8 Hz), 6.94–7.23 (m, 6H),
7.51–7.65 (m, 4H).
13C NMR (50 MHz, CDCl3): = –4.5 (q), –2.6 (q), 17.9, 22.3, 24.9,
26.0 (q), 32.3, 61.1, 66.8, 81.2, 84.1, 125.9 (d), 126.6 (d), 127.0 (d),
127.6 (d), 127.7 (d), 127.9 (d), 143.0 (s), 148.5 (s).
A solution of 2,2,5-trimethyl-4-hexen-3-ol (4h, 517 mg, 3.63
mmol) and benzophenone (656 mg, 3.60 mmol) in benzene (15 mL)
was irradiated for 20 h according to the above general procedure.
Flash chromatography on silica gel (65 g; PE–Et2O, 6:1) yielded
905 mg (78%) of threo-5h as a colorless, sticky oil, which solidified
within 3 days (mp 71–72 °C) on standing at r.t. (approx. 20 °C).
Synthesis 2001, No. 8, 1203–1214 ISSN 0039-7881 © Thieme Stuttgart · New York

PAPER
Preparation of Oxetanes with up to Three Stereogenic Centers as Synthetic Building Blocks
1211
Determination of the Relative Configuration of threo-5i
A diastereomeric mixture (approx. 92:8) of threo,erythro-5e (113
mg, 400 mol) was dissolved in CH2Cl2 (1.75 mL), and 4-(N,N-dim-
ethylamino)pyridine (24.4 mg, 200 mol), Et3N (50.5 mg, 500 mol)
and tert-butyldimethylsilyl chloride (75.4 mg, 500 mol) were add-
ed. The reaction mixture was stirred for 22 h at r.t. and the solvent
was evaporated (20°C, 15 Torr). The residue was purified by flash
chromatography on silica gel (18 g; PE–Et2O, 50:1) to afford 130
mg (82%) of the diastereomerically pure (by NMR analysis) threo-
5i as a colorless oil.
(63%) of a mixture (dr 88:12) of the threo,cis- and erythro,cis-6f
diastereomers as a colorless oil. An enriched sample of the eryth-
ro,cis-6f diastereomer was obtained upon repeated column chroma-
tography.
[2 (R*),4 ]-1-(3,3,4-Trimethyl-4-phenyloxetan-2-yl)-propanol
(threo,cis-6f)
1H NMR (200 MHz, CDCl3): = 0.71 (s, 3H), 1.01 (dd, 3H, J = 7.6,
6.9 Hz), 1.24–1.36 (m, 2H), 1.27 (s, 3H), 1.70 (s, 3H), 2.39 (br s,
1H), 3.60–3.70 (m, 1H), 4.21 (d, 1H, J = 8.7 Hz), 7.18–7.37 (m,
5H).
Irradiation of 3-Methyl-2-buten-1-ol (4a) and Acetophenone (2)
A solution of 3-methyl-2-buten-1-ol (4a, 215 mg, 2.50 mmol) and
acetophenone (2, 300 mg, 2.50 mmol) in benzene (10 mL) was irra-
diated for 48 h according to the above general procedure. The con-
version of the allylic alcohol 4a (85%) was determined by 1H NMR
spectroscopy on the crude photolysate. Flash chromatography on
silica gel (60 g; PE–Et2O, 2:1) yielded 233 mg (53%) of cis-6a as a
colorless oil. The relative cis configuration of 6a was determined by
a combination of HMBC and NOE spectroscopy.
13C NMR (50 MHz, CDCl3): = 9.7 (q), 20.7 (q), 23.3 (q), 25.3 (q),
25.7 (t), 42.6 (s), 72.8 (d), 87.2 (d), 88.3 (s), 124.0 (d), 126.5 (d),
127.9 (d), 145.3 (s).
[2 (S*),4 ]-1-(3,3,4-Trimethyl-4-phenyloxetan-2-yl)-propanol
(erythro,cis-6f)
1H NMR (200 MHz, CDCl3): = 0.80 (s, 3H), 1.03 (dd, 3H, J = 7.5,
7.3 Hz), 1.32 (s, 3H), 1.29–1.43 (m, 2H), 1.68 (s, 3H), 3.58–3.73
(m, 1H), 4.16 (d, 1H, J = 9.0 Hz), 7.17–7.37 (m, 5H).
13C NMR (50 MHz, CDCl3): = 9.3 (q), 20.7 (q), 23.7 (q), 25.3 (q),
25.9 (t), 43.0 (s), 72.6 (d), 86.7 (d), 88.1 (s), 124.0 (d), 126.3 (d),
127.8 (d), 145.6 (s).
cis-3,3,4-Trimethyl-4-phenyloxetan-2-ylmethanol (cis-6a)
IR (NaCl): = 3410, 3060, 2975, 1490, 1470, 1375, 1025, 890 cm–1.
1H NMR (200 MHz, CDCl3): = 0.71 (s, 3H), 1.31 (s, 3H), 1.70 (s,
3H), 2.23 (br s, 1H), 3.66 (dd, 1H, J = 11.8, 4.4 Hz), 3.82 (dd, 1H,
J = 11.8, 7.3 Hz), 4.58 (dd, 1H, J = 7.3, 4.4 Hz), 7.21–7.39 (m, 5H).
For the mixture of diastereomers:
IR (NaCl): = 3424, 3060, 2930, 1470, 1375, 1055, 970, 900 cm–1.
13C NMR (50 MHz, CDCl3): = 20.3 (q), 23.3 (q), 25.6 (q), 42.7 (s),
63.1 (t), 85.4 (d), 89.1 (s), 124.0 (d), 126.5 (d), 127.9 (d), 145.5 (s).
Irradiation of 2,5-Dimethyl-4-hexen-3-ol (4g) and Aceto-
phenone (2)
Anal. Calcd for C13H18O2 (206.3): C, 75.69; H, 8.79. Found: C,
75.43; H, 8.84.
A solution of 2,5-dimethyl-4-hexen-3-ol (4g, 641 mg, 5.00 mmol)
and acetophenone (2, 601 mg, 5.00 mmol) in benzene (18 mL) was
irradiated for 82 h according to the above general procedure. Flash
chromatography on silica gel (100 g; PE–CH2Cl2, 1:1) yielded first
203 mg of unreacted allylic alcohol 4g and then 485 mg (57%) of a
mixture (dr 92:08) of the threo,cis- and erythro,cis-6g diastere-
omers as a colorless oil. An enriched sample of the erythro,cis-6f
diastereomer was obtained upon repeated column chromatography.
Irradiation of 4-Methyl-3-penten-2-ol (4e) and Acetophenone
(2)
A solution of 4-methyl-3-penten-2-ol (4e, 376 mg, 3.75 mmol) and
acetophenone (2, 451 mg, 3.75 mmol) in benzene (15 mL) was irra-
diated for 28 h according to the above general procedure. Flash
chromatography on silica gel (80 g; PE–Et2O, 2:1) yielded 525 mg
(64%) of a mixture (dr 82:18) of the threo,cis- and erythro,cis-6e
diastereomers as a colorless oil. An enriched sample of the eryth-
ro,cis-6e diastereomer was obtained upon repeated column chroma-
tography.
[2 (R*),4 ]-1-(3,3,4-Trimethyl-4-phenyloxetan-2-yl)-2-methyl-
propanol (threo,cis-6g)
1H NMR (200 MHz, CDCl3): = 0.74 (s, 3H), 0.94 (d, 3H, J = 6.9
Hz), 0.98 (d, 3H, J = 7.0 Hz), 1.28 (s, 3H), 1.52–1.65 (m, 1H), 1.71
(s, 3H), 2.40 (br s, 1H), 3.56 (dd, 1H, J = 7.3, 3.4 Hz), 4.43 (d, 1H,
J = 7.3 Hz), 7.19–7.38 (m, 5H).
[2 (R*),4 ]-1-(3,3,4-Trimethyl-4-phenyloxetan-2-yl)-ethanol
(threo,cis-6e)
13C NMR (50 MHz, CDCl3): = 16.0 (q), 19.6 (q), 20.9 (q), 23.1 (q),
25.4 (q), 30.7 (d), 43.0 (s), 74.4 (d), 85.2 (d), 88.4 (s), 123.9 (d),
126.4 (d), 127.9 (d), 145.2 (s).
1H NMR (200 MHz, CDCl3): = 0.70 (s, 3H), 1.07 (d, 3H, J = 6.2
Hz), 1.28 (s, 3H), 1.70 (s, 3H), 2.20 (br s, 1H), 3.90 (dq, 1H, J = 8.8,
6.2 Hz), 4.14 (d, 1H, J = 8.8 Hz), 7.22–7.36 (m, 5H).
13C NMR (50 MHz, CDCl3): = 18.0 (q), 20.5 (q), 23.3 (q), 25.3 (q),
42.4 (s), 67.6 (d), 88.3 (d), 88.8 (s), 124.0 (d), 126.5 (d), 127.9 (d),
145.2 (s).
[2 (S*),4 ]-1-(3,3,4-Trimethyl-4-phenyloxetan-2-yl)-2-methyl-
propanol (erythro,cis-6g)
1H NMR (200 MHz, CDCl3): = 0.80 (s, 3H), 0.97 (d, 3H, J = 5.6
Hz), 1.04 (d, 3H, J = 5.4 Hz), 1.33 (s, 3H), 1.69 (s, 3H), 1.97–2.09
(m, 1H), 2.35 (br s, 1H), 3.62 (dd, 1H, J = 7.3, 2.7 Hz), 4.30 (d, 1H,
J = 7.3 Hz), 7.19–7.37 (m, 5H).
[2 (S*),4 ]-1-(3,3,4-Trimethyl-4-phenyloxetan-2-yl)-ethanol
(erythro,cis-6e)
1H NMR (200 MHz, CDCl3): = 0.80 (s, 3H), 1.29 (d, 3H, J = 6.1
Hz), 1.33 (s, 3H), 1.69 (s, 3H), 2.42 (br s, 1H), 3.82–3.99 (m, 1H),
4.09 (d, 1H, J = 8.7 Hz), 7.18–7.37 (m, 5H).
13C NMR (50 MHz, CDCl3): = 15.2 (q), 19.4 (q), 20.7 (q), 23.8 (q),
24.8 (q), 28.6 (d), 43.0 (s), 75.0 (d), 84.5 (d), 88.2 (s), 145.6 (s); only
the well resolved signals are reported.
13C NMR (50 MHz, CDCl3): = 19.6 (q), 20.7 (q), 23.8 (q), 25.3 (q),
42.9 (s), 67.6 (d), 88.0 (d), 88.2 (s), 124.0 (d), 126.0 (d), 126.3 (d),
145.6 (s).
Irradiation of 2,2,5-Trimethyl-4-hexen-3-ol (4h) and Aceto-
phenone (2)
A solution of 2,2,5-trimethyl-4-hexen-3-ol (4h, 427 mg, 3.00
mmol) and acetophenone (2, 180 mg, 1.50 mmol) in benzene (15
mL) was irradiated for 72 h according to the above general proce-
dure. The excess of allylic alcohol 4h was removed by Kugelrohr
distillation (55°C, 0.2 Torr). To remove by-products of higher mo-
lecular weight, the residue was dissolved in Et2O (3 mL), then sand
Irradiation of 5-Methyl-4-hexen-3-ol (4f) and Acetophenone (2)
A solution of 5-methyl-4-hexen-3-ol (4f, 428 mg, 3.75 mmol) and
acetophenone (2, 451 mg, 3.75 mmol) in benzene (15 mL) was irra-
diated for 48 h according to the above general procedure. Flash
chromatography on silica gel (90 g; PE–Et2O, 1:1) yielded 553 mg
Synthesis 2001, No. 8, 1203–1214 ISSN 0039-7881 © Thieme Stuttgart · New York

1212
W. Adam, V. R. Stegmann
PAPER
[2 (S*),4 ]-1-(3,3-Dimethyl-4-phenyloxetan-2-yl)-ethanol
(approx. 2 g) was added, and the solvent was evaporated (20°C, 15
Torr). The residue was triturated with petroleum ether (3 5 mL)
and the solid matter was removed by filtration. Evaporation (20°C,
15 Torr) of the solvent yielded 194 mg (49%) of threo,cis-6h as a
pale yellow oil.
(erythro,cis-7e)
1H NMR (600 MHz, C6D6): = 0.78 (s, 3H), 1.22 (s, 3H), 4.06 (d,
1H, J = 9.1 Hz), 5.30 (s, 1H); only the well resolved signals are re-
ported.
13C NMR (150 MHz, C6D6): = 21.7 (q), 28.0 (q), 42.7 (s), 67.5 (d),
88.2 (d), 90.3 (d); only the well resolved signals are reported.
[2 (S*),4 ]-1-(3,3,4-Trimethyl-4-phenyloxetan-2-yl)-2,2-
dimethylpropanol (threo,cis-6h)
1-(2,2-Dimethyl-4-phenyloxetan-3-yl)-ethanol (7e’) (four diaste-
reomers)
IR (NaCl): = 3515, 3060, 2960, 2870, 1495, 1470, 1445, 1375,
1090, 1025 cm–1.
1H NMR (600 MHz, C6D6): = 2.23 (dd, 1H, J = 9.8, 7.4 Hz), 2.36
(dd, 1H, J = 10.2, 7.8 Hz), 2.48 (dd, 1H, J = 10.9, 8.7 Hz), 2.64 (dd,
1H, J = 10.7, 8.4 Hz), 4.96 (d, 1H, J = 7.8 Hz), 5.50 (d, 1H, J = 8.4
Hz), 5.51 (d, 1H, J = 7.4 Hz), 5.67 (d, 1H, J = 8.7 Hz); only the well
resolved signals are reported.
13C NMR (150 MHz, C6D6): = 55.7 (d), 56.0 (d), 60.4 (d), 60.8 (d),
77.4 (d), 78.3 (d), 78.5 (d), 79.8 (d); only the well resolved signals
are reported.
1H NMR (200 MHz, CDCl3): = 0.82 (s, 3H), 0.93 (s, 9H), 1.27 (s,
3H), 1.73 (s, 3H), 3.03 (d, 1H, J = 6.1 Hz), 3.18–3.22 (m, 1H), 4.60
(d, 1H, J = 4.6 Hz), 7.15–7.36 (m, 5H).
13C NMR (50 MHz, CDCl3): = 21.1 (q), 23.0 (q), 25.6 (q), 26.1 (q),
34.2 (s), 43.8 (s), 75.0 (d), 82.8 (d), 89.3 (s), 123.6 (d), 126.5 (d),
128.0 (d), 145.0 (s).
Irradiation of 3-Methyl-2-buten-1-ol (4a) and Benzaldehyde (3)
A solution of 3-methyl-2-buten-1-ol (4a, 517 mg, 6.00 mmol) and
benzaldehyde (3, 318 mg, 3.00 mmol) in benzene (30 mL) was irra-
diated for 20 h according to the above general procedure. The ex-
cess allylic alcohol 4a was removed by Kugelrohr distillation
(50°C, 0.1 Torr) and flash chromatography of the residue on silica
gel (60 g; PE–Et2O, 1:1) yielded 387 mg (67%) of a mixture of the
cis-7a and 7a’ regioisomers as a colorless oil. Enriched samples of
both regioisomers were obtained upon repeated column chromatog-
raphy.
Irradiation of 5-Methyl-4-hexen-3-ol (4f) and Benzaldehyde (3)
on the 1H NMR Analytical Scale
A solution of 5-methyl-4-hexen-3-ol (4f, 28.5 mg, 0.250 mmol) and
benzaldehyde (3, 26.5 mg, 0.250 mmol) in d6-benzene was placed
into a Quartz-NMR tube and irradiated for 3 h according to the
above general procedure. The 1H NMR analysis of the crude pho-
tolysate showed the formation of the threo,cis- and erythro,cis-7f
and four diastereomers of 7f’ (see Table 3), which were assigned by
1
their typical H and 13C resonances with the help of HMQC spec-
troscopy.
cis-3,3-Dimethyl-4-phenyl-2-oxetanemethanol (cis-7a)
IR (NaCl): = 3410, 3065, 2975, 1715, 1450, 1275, 1120,
1025 cm–1.
1H NMR (200 MHz, CDCl3): = 0.69 (s, 3H), 1.41 (s, 3H), 2.31 (br
s, 1H), 3.69 (dd, 1H, J = 12.0, 4.4 Hz), 3.88 (dd, 1H, J = 12.0, 7.3
Hz), 4.55 (dd, 1H, J = 7.3, 4.4 Hz), 5.49 (s, 1H), 7.24–7.40 (m, 5H).
[2 (R*),4 ]-1-(3,3-Dimethyl-4-phenyloxetan-2-yl)-propanol
(threo,cis-7f)
1H NMR (600 MHz, C6D6): = 0.51 (s, 3H), 0.97 (s, 3H), 3.64 (ddd,
1H, J = 8.8, 8.7, 2.9 Hz), 4.18 (d, 1H, J = 8.7 Hz), 5.26 (s, 1H); only
the well resolved signals are reported.
13C NMR (50 MHz, CDCl3): = 17.0 (q), 27.7 (q), 42.1 (s), 63.3 (t),
87.6 (d), 89.3 (d), 124.9 (d), 127.4 (d), 128.1 (d), 139.5 (s).
13C NMR (150 MHz, C6D6): = 17.4 (q), 27.2 (q), 31.0 (t), 42.2 (s),
72.7 (d), 88.2 (d), 89.8 (d); only the well resolved signals are report-
ed.
+
HRMS (CI): Calcd for C12H16O2NH4+ (M+NH4 ) 210.1494. Found:
210.1493.
[2 (S*),4 ]-1-(3,3-Dimethyl-4-phenyloxetan-2-yl)-propanol
(erythro,cis-7f)
2,2-Dimethyl-4-phenyl-3-oxetanemethanol (7a’)
1H NMR (600 MHz, C6D6): = 5.31 (s, 1H); only the well resolved
signals are reported.
1H NMR (200 MHz, CDCl3): = 1.39 (s, 3H), 1.61 (s, 3H), 3.02
(ddd, 1H, J = 8.2, 8.1, 7.5 Hz), 3.37–3.54 (m, 2H), 5.82 (d, 1H,
J = 8.2 Hz), 7.25–7.38 (m, 5H).
13C NMR (150 MHz, C6D6): = 42.8 (s), 88.2 (d), 88.8 (d); only the
well resolved signals are reported.
13C NMR (50 MHz, CDCl3): = 24.2 (q), 30.5 (q), 48.9 (d), 59.6 (t),
77.0 (d), 82.6 (s), 125.4 (d), 127.4 (d), 128.3 (d), 139.5 (s).
1-(2,2-Dimethyl-4-phenyloxetan-3-yl)-propanol (7f’; four diastere-
omers)
Irradiation of 4-Methyl-3-penten-2-ol (4e) and Benzaldehyde
(3) on the 1H NMR Analytical Scale
1H NMR (600 MHz, C6D6): = 2.32 (dd, 1H, J = 9.8, 7.4 Hz), 2.47
(dd, 1H, J = 10.2, 7.7 Hz), 2.59 (dd, 1H, J = 10.8, 8.6 Hz), 2.74 (dd,
1H, J = 10.9, 8.5 Hz), 5.00 (d, 1H, J = 7.7 Hz), 5.51 (d, 1H, J = 8.6
Hz), 5.53 (d, 1H, J = 7.7 Hz), 5.68 (d, 1H, J = 8.5 Hz); only the well
resolved signals are reported.
A solution of 4-methyl-3-penten-2-ol (4e, 25.0 mg, 0.250 mmol)
and benzaldehyde (3, 26.5 mg, 0.250 mmol) in d6-benzene was
placed into a Quartz-NMR tube and irradiated for 3 h according to
1
the above general procedure. The H NMR analysis of the crude
13C NMR (150 MHz, C6D6): = 53.6 (d), 54.1 (d), 58.8 (d), 59.1 (d),
77.5 (d), 78.3 (d), 78.6 (d), 79.5 (d); only the well resolved signals
are reported.
photolysate showed the formation of the threo,cis- and erythro,cis-
7e and four diastereomers of 7e’ (see Table 3), which were assigned
by their typical 1H and 13C resonances with the help of HMQC spec-
troscopy.
Irradiation of 2,5-Dimethyl-4-hexen-3-ol (4g) and Benzalde-
hyde (3) on the 1H NMR Analytical Scale
[2 (R*),4 ]-1-(3,3-Dimethyl-4-phenyloxetan-2-yl)-ethanol (threo,
cis-7e)
A solution of 2,5-dimethyl-4-hexen-3-ol (4g, 32.1 mg, 0.250 mmol)
and benzaldehyde (3, 26.5 mg, 0.250 mmol) in d6-benzene was
placed in a Quartz-NMR tube and irradiated for 3 h according to the
above general procedure. The 1H NMR analysis of the crude pho-
tolysate showed the formation of the threo,cis- and erythro,cis-7g
and four diastereomers of 7g’(see Table 3), which were assigned by
1H NMR (600 MHz, C6D6): = 0.46 (s, 3H), 0.92 (d, 3H, J = 6.2
Hz), 0.96 (s, 3H), 3.85 (dq, 1H, J = 8.8, 6.2 Hz), 4.10 (d, 1H, J = 8.8
Hz), 5.25 (s, 1H); only the well resolved signals are reported.
13C NMR (150 MHz, C6D6): = 17.2 (q), 17.9 (q), 27.2 (q), 41.9 (s),
67.7 (d), 88.3 (d), 91.3 (d); only the well resolved signals are report-
ed.
Synthesis 2001, No. 8, 1203–1214 ISSN 0039-7881 © Thieme Stuttgart · New York

PAPER
Preparation of Oxetanes with up to Three Stereogenic Centers as Synthetic Building Blocks
1213
the typical 1H and 13C resonances with the help of HMQC spectros-
copy.
1H NMR (200 MHz, CDCl3): = 0.76 (s, 9H), 1.63 (s, 3H), 1.71 (s,
3H), 3.25 (dd, 1H, J = 9.2, 5.1 Hz), 3.35–3.40 (m, 1H), 5.67 (d, 1H,
J = 9.2 Hz), 7.26–7.50 (m, 5H).
[2 (R*),4 ]-1-(3,3-Dimethyl-4-phenyloxetan-2-yl)-2-methyl-
propanol (threo,cis-7g)
13C NMR (50 MHz, CDCl3): = 26.6 (q), 27.2 (q), 33.0 (q), 35.0 (s),
48.7 (d), 77.2 (d), 78.3 (d), 85.8 (s), 126.8 (d), 127.7 (d), 128.3 (d),
140.6 (s).
1H NMR (600 MHz, C6D6): = 0.58 (s, 3H), 3.58 (dd, 1H, J = 7.7,
3.4 Hz), 4.38 (d, 1H, J = 7.7 Hz), 5.28 (s, 1H); only the well resolved
signals are reported.
Third Diastereomer (in the order of elution, relative configuration
not known):
1H NMR (200 MHz, CDCl3): = 0.80 (s, 9H), 1.52 (s, 3H), 1.56 (s,
3H), 2.81 (dd, 1H, J = 8.1, 6.9 Hz), 3.61 (d, 1H, J = 6.9 Hz), 5.63 (d,
1H, J = 8.1 Hz), 7.23–7.53 (m, 5H).
13C NMR (50 MHz, CDCl3): = 24.0 (q), 25.6 (q), 30.8 (q), 34.7 (s),
54.4 (d), 77.1 (d), 77.9 (d), 81.7 (s), 126.4 (d), 127.6 (d), 128.3 (d),
143.0 (s).
13C NMR (150 MHz, C6D6): = 17.6 (q), 42.7 (s), 74.6 (d), 88.0 (d),
88.4 (d); only the well resolved signals are reported.
[2 (S*),4 ]-1-(3,3-Dimethyl-4-phenyloxetan-2-yl)-2-methyl-
propanol (erythro,cis-7g)
1H NMR (600 MHz, C6D6): = 4.35 (d, 1H, J = 7.7 Hz), 5.32 (s,
1H); only the well resolved signals are reported.
13C NMR (150 MHz, C6D6): = 43.1 (s), 86.9 (d), 88.3 (d); only the
well resolved signals are reported.
Fourth Diastereomer (in the order of elution, relative configuration
not known):
1H NMR (200 MHz, CDCl3): = 0.72 (s, 9H), 1.56 (s, 3H), 1.63 (s,
3H), 2.98 (dd, 1H, J = 10.5, 8.2 Hz), 3.69 (d, 1H, J = 10.5 Hz), 5.25
(d, 1H, J = 8.2 Hz); only the well resolved signals are reported.
13C NMR (50 MHz, CDCl3): = 23.1 (q), 25.7 (q), 31.6 (q), 34.9 (s),
53.9 (d), 79.1 (d), 79.9 (d), 82.5 (s), 140.3 (s); only the well resolved
signals are reported.
1-(2,2-Dimethyl-4-phenyloxetan-3-yl)-2-methylpropanol (7g’; four
diastereomers)
1H NMR (600 MHz, C6D6): = 2.56 (dd, 1H, J = 9.8, 7.4 Hz), 2.69
(dd, 1H, J = 10.6, 8.0 Hz), 2.80 (dd, 1H, J = 11.0, 8.7 Hz), 2.97 (dd,
1H, J = 11.3, 8.8 Hz), 5.03 (d, 1H, J = 7.4 Hz), 5.52 (d, 1H, J = 8.7
Hz), 5.54 (d, 1H, J = 8.0 Hz), 5.70 (d, 1H, J = 8.8 Hz); only the well
resolved signals are reported.
13C NMR (150 MHz, C6D6): = 51.2 (d), 51.5 (d), 56.5 (d), 57.0 (d),
77.4 (d), 78.4 (d), 78.6 (d), 79.6 (d); only the well resolved signals
are reported.
Acknowledgement
The authors thank the Volkswagenstiftung, the Deutsche For-
schungsgemeinschaft (DFG), and the Fonds der Chemischen Indu-
strie for financial support.
Irradiation of 2,2,5-Trimethyl-4-hexen-3-ol (4h) and Benz-
aldehyde (3)
A solution of 2,2,5-trimethyl-4-hexen-3-ol (4h, 427 mg, 3.00
mmol) and benzaldehyde (3, 318 mg, 3.00 mmol) in benzene (30
mL) was irradiated for 7 h according to the above general proce-
dure. The excess allylic alcohol 4h was removed by Kugelrohr dis-
tillation (50 °C, 0.1 Torr) and the residual product mixture was
purified by flash chromatography on silica gel (33 g; PE–Et2O, 10:1
to 3:1 gradient) to yield 107 mg (29%) of the threo,cis-7h oxetane
and a total of 95.3 mg (25%) of the four diastereomers of the 7h’
regioisomer as colorless oils.
References
(1) (a) Paternò, E.; Chieffi, G. Gazz. Chim. Ital. 1909, 39-1,
341. (b) Büchi, G.; Inman, C. G.; Lipinsky, E. S. J. Am.
Chem. Soc. 1954, 76, 4327.
(2) (a) Griesbeck, A. G.; Fiege, M. In Molecular and
Supramolecular Photochemistry, Vol. 6; Ramamurthy, V.;
Schanze, K. S., Eds.; Marcel Dekker: New York, 2000, 33.
(b) Bach, T. Synthesis 1998, 683. (c) Fleming, S. A.;
Bradford, C. L.; Gao, J. J. In Molecular and Supramolecular
Photochemistry, Vol. 1; Ramamurthy, V.; Schanze, K. S.,
Eds.; Marcel Dekker: New York, 1997, 187. (d) Mattay, J.;
Conrads, R.; Hoffmann, R. In Houben-Weyl, 4th ed., Vol.
E21c; Helmchen, G.; Hoffmann, R. W.; Mulzer, J.;
Schaumann, E. E., Eds.; Thieme: Stuttgart, 1995, 3133.
(e) Griesbeck, A. G. In Handbook of Photochemistry and
Photobiology; Horspool, W.; Song, P.-S., Eds.; CRC Press:
Boca Raton, 1995, 550 and 755. (f) Porco, J. A.; Schreiber,
S. L. In Comprehensive Organic Synthesis, Vol. 5; Trost, B.
M.; Paquette, L. A., Eds.; Pergamon Press: Oxford, 1991,
151.
(3) (a) Adam, W.; Peters, K.; Peters, E.-M.; Stegmann, V. R. J.
Am. Chem. Soc. 2000, 122, 2958. (b) Adam, W.; Saha-
Möller, C. R.; Schambony, S. B. J. Am. Chem. Soc. 1999,
121, 1834. (c) Yokoyama, A.; Mizuno, K. Org. Lett. 2000,
2, 3457. (d) D'Auria, M.; Racioppi, R.; Romaniello, G. Eur.
J. Org. Chem. 2000, 3265. (e) Bach, T.; Bergmann, H.;
Harms, K. J. Am. Chem. Soc. 1999, 121, 10650. (f) Sydnes,
L. K.; Hansen, K. I.; Oldroyd, D. L.; Weedon, A. C.;
Jørgensen, E. Acta Chem. Scand. 1993, 47, 916.
[2 (R*),4 ]-1-(3,3-Dimethyl-4-phenyloxetan-2-yl)-2,2-dimethyl-
propanol (threo,cis-7h)
1H NMR (200 MHz, CDCl3): = 0.82 (s, 3H), 0.94 (s, 9H), 1.36 (s,
3H), 3.07 (d, J = 6.0 Hz, OH), 3.27 (dd, 1H, J = 6.0, 3.7 Hz), 4.66 (d,
1H, J = 3.7 Hz), 5.54 (s, 1H), 7.22–7.42 (m, 5H).
13C NMR (50 MHz, CDCl3): = 17.7 (q), 25.6 (q), 27.2 (q), 34.2 (s),
43.6 (s), 75.5 (d), 84.9 (d), 89.0 (d), 124.9 (d), 127.4 (d), 128.1 (d),
138.9 (s).
1-(2,2-Dimethyl-4-phenyloxetan-3yl)-2,2-dimethylpropanol (7h’)
First Diastereomer (in the order of elution, relative configuration
not known):
1H NMR (200 MHz, CDCl3): = 0.82 (s, 9H), 1.07 (d, J = 5.0 Hz,
OH), 1.60 (s, 3H), 1.61 (s, 3H), 3.30 (dd, 1H, J = 9.3, 9.1 Hz), 3.50
(dd, 1H, J = 9.3, 5.0 Hz), 5.75 (d, 1H, J = 9.1 Hz), 7.25–7.63 (m,
5H).
13C NMR (50 MHz, CDCl3): = 25.8 (q), 26.2 (q), 31.4 (q), 34.7 (s),
50.2 (d), 75.7 (d), 78.6 (d), 83.4 (s), 127.4 (d), 128.2 (d), 128.6 (d),
140.7 (s).
Second Diastereomer (in the order of elution, relative configuration
not known):
(4) Adam, W.; Wirth, T. Acc. Chem. Res. 1999, 32, 703.
(5) Turro, N. J. Modern Molecular Photochemistry; University
Science Books: Sausalito, 1991, Chap. 11.3.
(6) Hoffmann, R. W. Chem. Rev. 1989, 89, 1841.
Synthesis 2001, No. 8, 1203–1214 ISSN 0039-7881 © Thieme Stuttgart · New York

1214
W. Adam, V. R. Stegmann
PAPER
(7) Caldwell, R. A.; Sovocool, G. W.; Gajewski, R. P. J. Am.
Chem. Soc. 1973, 95, 2549.
(13) Salem, L.; Rowland, C. Angew. Chem., Int. Ed. Engl. 1972,
11, 92.
(8) Helmchen, G. In Houben-Weyl, 4th ed., Vol. E21a;
Helmchen, G.; Hoffmann, R. W.; Mulzer, J.; Schaumann, E.
E., Eds.; Thieme: Stuttgart, 1995, 53.
(9) Paleta, O.; Svoboda, J.; Dedek, V. Collect. Czech. Chem.
Commun. 1978, 43, 2932.
(10) Griesbeck, A. G.; Mauder, H.; Stadtmüller, S. Acc. Chem.
Res. 1994, 27, 70.
(11) (a) Freilich, S. C.; Peters, K. S. J. Am. Chem. Soc. 1985, 107,
3819. (b) Freilich, S. C.; Peters, K. S. J. Am. Chem. Soc.
1981, 103, 6255. (c) Hu, S.; Neckers, D. C. J. Org. Chem.
1997, 62, 755.
(14) Griesbeck, A. G.; Buhr, S.; Fiege, M.; Schmickler, H.; Lex,
J. J. Org. Chem. 1998, 63, 3847.
(15) Dang, H.; Davies, A. G.; Davison, I. G. E.; Schiesser, C. H.
J. Org. Chem. 1990, 55, 1432.
(16) Gau, A.-H.; Lin, G.-L.; Uang, B.-J.; Liao, F.-L.; Wang, S.-L.
J. Org. Chem. 1999, 64, 2194.
(17) Bilas, W.; Hoebold, W.; Pritzkow, W. J. Prakt. Chem. 1982,
324, 125.
(18) Majewski, M.; Snieckus, V. J. Org. Chem. 1984, 49, 2682.
(19) Heathcock, C. H.; Young, S. D.; Hagen, J. P.; Pirrung, M. C.;
White, C. T.; VanDerveer, D. J. Org. Chem. 1980, 45, 3846.
(20) Peters, K.; Peters, E.-M.; Stegmann, V. R.; Adam, W. Z.
Kristallogr. NCS 2000, 215.
(12) Caldwell, R. A. In Kinetics and Spectroscopy of Carbenes
and Biradicals; Platz, M. S., Ed.; Plenum: New York, 1990,
77.
Synthesis 2001, No. 8, 1203–1214 ISSN 0039-7881 © Thieme Stuttgart · New York