Iron-salen complexes as efficient catalysts in ring expansion reactions of epoxyalkenes

Article abstract of DOI:10.1002/adsc.200606107

The optimisation of the iron-catalysed ring expansion reaction of epoxyalkenes was considerably improved when the original phosphine ligand system [FeCl₂(dppe)] was altered to include nitrogen-containing ligand systems. Especially, the very pot

Full text of DOI:10.1002/adsc.200606107

FULL PAPERS
DOI: 10.1002/adsc.200606107
Iron-Salen Complexes as Efficient Catalysts in Ring Expansion
Reactions of Epoxyalkenes
Gerhard Hilt,a,* Christian Walter,a and Patrick Bolzea
a
Fachbereich Chemie, Philipps-Universität Marburg, Hans-Meerwein-Str., 35043 Marburg, Germany
Fax : (+49)-6421-282-5677; e-mail: hilt@chemie.uni-marburg.de
Received: March 15, 2006; Accepted: May 9, 2006
Dedicated to Prof. Ryoji Noyori on the occasion of his 68th birthday.
Abstract: The optimisation of the iron-catalysed ring preformed iron(II)-salen complex the yields and dia-
expansion reaction of epoxyalkenes was considerably stereoselectivities were greatly enhanced and the
improved when the original phosphine ligand system scope of the reaction could also be enlarged.
[FeCl2ACHTREUNG(dppe)] was altered to include nitrogen-con-
taining ligand systems. Especially, the very potent
class of salen ligands gave the best results in inter- Keywords: catalysis; epoxide; iron; ring expansion;
and intramolecular ring expansion reactions. With a salen ligands
Introduction
es with low-valent iron species, we found that a ring
expansion of epoxides 1 takes place when iron-phos-
Iron-catalysed reactions in organic synthesis have phine complexes are reduced with zinc powder in the
been a long-time goal for many organic chemists due presence of triethylamine and an alkene 2.[2] Thereby,
to the low toxicity and the abundance of many iron tetrahydrofuran derivatives 3 are generated in an in-
salts and compounds. Among these iron-catalysed re- termolecular approach from readily available starting
actions are carbon-carbon bond formation processes, materials (Scheme 1). However, polymerisation side
cycloadditions, cross-coupling reactions and aldol-type reactions could only be suppressed when intramolecu-
reactions, just to name a few of the most relevant lar reactions were investigated. Under these circum-
transformations.[1] It can be assumed that in some stances the epoxide ring expansion of 4 gives rise to
iron-catalysed reactions lower oxidation states are the the formation of bicyclic compounds such as 5a.[3]
active species in the catalytic cycle. Therefore, the
active species are often generated in situ upon the ad-
dition of an appropriate reducing agent. In an attempt Results and Discussion
to investigate carbon-carbon bond formation process-
The epoxyalkene 4 (Scheme 1) needed for the intra-
molecular epoxide ring expansion reaction is pre-
pared from 1-phenylcyclopent-1-ene (6) which is con-
verted into 7 in an ozonolysis followed by reductive
work-up in 82% yield (Scheme 2).[4] A chemoselec-
tive Wittig olefination converts the aldehyde into the
alkene 8 in 78% yield as an E/Z mixture (1.0:1.5).
The following epoxidation of the ketone 8 via sulfur
ylides produces the desired epoxyalkene 4 (79%) in a
straightforward reaction sequence in good overall
yield.
The iron-catalysed ring expansion reaction is be-
lieved to be initiated by a single electron transfer
(SET) from the low-valent iron species 9 to the epox-
ide 1 which is coordinated to the iron centre.[5] Ac-
cordingly, due to the reaction mechanism proposed
Scheme 1. Inter- and intramolecular iron-catalysed ring ex-
pansion reactions.
Adv. Synth. Catal. 2006, 348, 1241 – 1247
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1241

Gerhard Hilt et al.
FULL PAPERS
Ligand Screening
For testing the influence on the diastereoselectivity of
the iron-catalysed process, the catalysts were generat-
ed in situ prior to the reduction of the complex with
zinc powder in acetonitrile. As a starting point the
preliminary results using the FeCl2ACHTRE(UNG dppe) complex
yielding 5 in 83% and a 93:7 cis:trans diastereoselec-
tivity were chosen as comparison for the investiga-
tion.[2] Several different ligand types (Figure 1) were
tested in the reaction of 4 with respect to the diaste-
reoselectivity for the generation of 5a and 5b while
the yields were not optimised at this stage
(Scheme 4).
Scheme 2. Reaction sequence for the synthesis of the sub-
strate 4.
for the ring expansion reaction, as outlined in
Scheme 3, it can be assumed that the carbon-carbon
(reaction from 10 to 11) and the carbon-oxygen bond
formation processes (reaction from 11 to 3) proceed
within the ligand sphere of the iron complex. A back
Scheme 3. Proposed mechanism for the intermolecular iron-
catalysed ring expansion reaction.
Figure 1. Amine-type ligands applied in the iron-catalysed
ring expansion reaction of 4.
electron transfer (BET) generates the stabilised car-
bocation 12 which cyclises to the product 3. At this
stage the diastereoselectivity is controlled by the li-
gands of the iron catalyst. Therefore, modifications in
The ring expansion reaction did not proceed at all
the ligand sphere are supposed to have a profound in- when simple amine-type ligands such as the bidentate
fluence on the reactivity as well as on the diastereose- ligand 13, the tridentate amine ligand 14, bipyridine
lectivity of the reaction.
(15) or phenanthroline (16) were used (Table 1). To-
Scheme 4. Intramolecular iron-catalysed ring expansion reaction of 4.
1242
www.asc.wiley-vch.de
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 1241 – 1247

Ring Expansion Reactions of Epoxyalkenes
FULL PAPERS
Table 1. The iron-catalysed ring expansion reaction of 4 with
ligands 13–24.
good diastereoselectivity of 93:7. With triphenylphos-
phine (entry 9) or the P,N ligand (21) the reactivity of
the catalysts was further restored. Compared to the
benchmark reaction performed with the dppe ligand
(Scheme 1), however, side products were detected by
GC-MS while the starting material was not complete-
ly consumed. The yields of the mixture of products 5a
and 5b were considerably lower with 20% (for PPh3
as ligand) and 27% (for 21 as ligand), respectively.
The reactivities of the diimine-type ligands such as 22
and 23 were on similar levels but also side products
could be detected while the conversion was not com-
plete after 20 h. However, the desired products were
obtained in good diastereoselectivity, 5a:5b=94:6
(Table 1). The pyridine-diimine ligand 24 gave accept-
able yields of 30% and the diastereomers were
formed in a good selectivity of 95:5. Although the
yields were considerably lower than with the dppe
ligand (83%), a noticeable effect on the cis-trans dia-
stereoselectivity of the reaction was observed (see
Table 1, entries 9 and 12/13) . Accordingly, it must be
concluded that the assumption that the bond forma-
tion processes take place in the ligand sphere of the
iron catalysts was correct and that further investiga-
Entry[a]
Ligand
Yield
5a:5b (cis:trans)
1
13
0%
-
2
14
0%
-
3
15
0%
-
4
16
0%
-
5
17
0%
-
6
18
0%
-
7
8
9
10
11
12
13
19
20
0%
-
14%[b,c]
20%[b,c]
27%[b,c]
5%[b,c]
28%[b,c]
30%
93:7
80:20
91:9
94:6
94:6
95:5
PPh3 (4 equivs.)
21
22
23
24
[a]
Reaction conditions: iron complex (30 mol %), zinc (2.0
equivs.), NEt3 (45 mol %), CH3CN (1.0 mL), 4 (50 mg,
189 mmol), 20 h, 608C.
[b] Incomplete conversion.
[c]
(Unidentified) side products (>5%) were detected by
GC-MS.
sylated amine ligands such as 17 and 18 as well as the tion regarding other ligands are reasonable.
disulfide ligand 19 also gave no conversion of the The salen-type ligands and structurally related li-
starting material, whereas the oxalyl diamide ligand gands (Figure 2) were then tested in the intramolecu-
20 yielded the desired compound in 14% yield with a lar iron-catalysed ring expansion reaction of 4. The
Figure 2. Salen-type ligands applied in the iron-catalysed ring expansion reaction of 4.
Adv. Synth. Catal. 2006, 348, 1241 – 1247
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
1243

Gerhard Hilt et al.
FULL PAPERS
Table 2. The iron-catalysed ring expansion reaction of 4 with
ligands 25–36.
ratio while the yields of the desired product were
again in an acceptable range. Almost exclusive diaste-
reoselectivities (99:1) were obtained with ligand 36.
Unfortunately, the yields were rather low and the
conversion was incomplete after 20 h reaction time.
Accordingly, the ligand design would favour sterically
bulky substituents on the phenyl ring for a better dia-
stereoselectivity and a possible chiral modification in
the carbon backbone as realised in the Jacobsen
ligand for possible enantioselective versions of the re-
action which have not been the subject of this investi-
gation so far.
Entry[a]
Ligand
Yield
5a:5b (cis:trans)
1
2
3
4
5
6
7
8
25
26
27
28
29
30
31
32
33
34
35
36
65%
40%
58%
61%
22%
35%
40%
44%
61%
61%
56%
29%[b]
83:17
96:4
94:6
87:13
91:9
89:11
89:11
87:13
80:20
83:17
83:17
99:1
9
Even though good candidates in the salen-type
ligand series have already been identified for the
iron-catalysed ring expansion reaction. Nevertheless,
also the salicylenimine-type ligands shown in Figure 3
10
11
12
[a]
Reaction conditions: iron complex (30 mol %), zinc (2.0
equivs.), NEt3 (45 mol %), CH3CN (1.0 mL), 4 (50 mg,
189 mmol), 20 h, 608C.
[b] Incomplete conversion.
results summarised in Table 2 show that the salen-
type ligands are better suited for the reaction than
the previously discussed ligands. While the product
mixtures of 5a and 5b are generated in acceptable
yields of up to 65% (entry 1) acceptable to very good
cis:trans selectivities of up to 96:4 (entry 2) can be ob-
served without optimisation at this stage. Interesting-
ly, the amounts of side products detectable by GC-MS
or TLC were considerably reduced and the starting
materials were completely consumed. Nevertheless,
the mass balance suggests that some polymerisation
side products could have been formed. On the other
hand, small losses during the work-up on the small
Figure 3. Salicylenimine-type ligands applied in the iron-cat-
alysed ring expansion reaction of 4.
scale applied (189 mmol) easily result in>10% loss of were investigated in the iron-catalysed reaction. Al-
material and yield. When the carbon backbone of the though the iron to ligand ratio was altered (1.0 and
salen ligand was stepwise elongated by one carbon at 2.0 equivs. of ligands were used) almost identical
a time (25, 28, 29 and 30) it could be seen that the re- yields and diastereoselectivities were observed for the
activities for the ligands 25 and 28 are comparable to salicylenimine-type ligands 37, 38 and 39 (Table 3, en-
the benchmark dppe ligand and drop when longer
carbon backbones are involved. On the other hand,
the selectivity is slightly enhanced and reaches 91:9
for 29. The introduction of bulky substituents on
Table 3. The iron catalysed ring expansion reaction of 4 with
ligands 37–40.
Entry[a]
Ligand
Yield
5a:5b (cis:trans)
either the phenyl substituent of the salen-type ligands
(26, 27) or in the backbone of the salen-type ligands
(31, 32 and 33) has an effect both on the diastereose-
lectivities as well as on the yields of the reaction.
While in the case of the modifications on the phenyl
ring (26 and 27), inspired by the work of Jacobsen,[6]
the introduction of tert-butyl substituents increased
the diastereoselectivity to excellent levels while the
yield was somewhat diminished for 26. On the other
hand, electronic modifications of the salen-type li-
gands on the phenyl rings (34 and 35) gave almost
identical results with around 60% yields and identical
diastereoselectivities of 83:17. However, substituents
in the backbone (31, 32, 33) decreased the cis:trans
1
2
3
4
5
6
7
37
16%[b,c]
20%[b,c]
22%[b,c]
20%[b,c]
20%[b,c]
18%[b,c]
44%[b,c]
90:10
86:14
90:10
89:11
95:5
37 (2 equivs.)
38
38 (2 equivs.)
39
39 (2 equivs.)
40
95:5
90:10
[a]
Reaction conditions: iron complex (30 mol %), zinc (2.0
equivs.), NEt3 (45 mol %), CH3CN (1.0 mL), 4 (50 mg,
189 mmol), 20 h, 608C.
[b] Incomplete conversion.
[c]
(Unidentified) side products (>5%) were detected by
GC-MS.
1244
www.asc.wiley-vch.de
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 1241 – 1247

Ring Expansion Reactions of Epoxyalkenes
FULL PAPERS
tries 1–6). Furthermore, an interesting ligand motif
derived from an a-amino acid was tested (40), which
would allow a simple access to chiral iron complexes.
However, in all cases the conversion was not com-
plete and varying amounts of side products were de-
tected. Therefore, the salen-type ligands, which can
easily be accessed and sterically as well as electroni-
cally modified, seem to be the most promising candi-
dates for further investigations.
The oxidation state of the iron precursor did not
have a large effect on the reaction of 4. When iron(II)
chloride was used as precursor for the in situ generat-
ed salen complex with ligand 25, a 65% yield and a
diastereoselectivity of 83:17 were found. Similar re-
Scheme 6. Intermolecular iron catalysed ring expansion re-
action with the 2,3-dimethyl-1,3-diene.
sults were found when ironCAHTRE(UNG III) chloride was used as starting materials is enhanced considerably, 43 was
iron source with ligand 25 (59% yield, 80:20). There- obtained in 83% as a 4.2:1.0 mixture of 43a:43b as
fore, we assume that the reducing agent (zinc single regioisomers.
powder) converts the two complexes to identical low-
valent iron species.
Therefore, the preformed iron-salen complexes
seem to be a very useful catalyst motif for further
The good results for the iron-salen complex (ligand inter- and intramolecular epoxide ring expansion re-
28) were then further optimised on a larger scale and actions for the synthesis of regiochemically, diastereo-
in these reactions (1.0 mmol) easily up to 88% yields and (hopefully) enantiomerically pure tetrahydrofur-
of the isolated bicyclic product 5 were achieved.[7]
A
an and bicyclic heterocyclic ring systems.
further application and a considerable enlargement of
the substrate scope were realised in the iron-salen-
catalysed intermolecular reaction of styrene oxide (1) Conclusions
with 41 (Scheme 5).
In the iron-catalysed ring expansion reaction of epoxy-
alkenes considerable improvements are realised when
the original phosphine ligand system [FeCl2A(dppe)] is
CTHREUNG
altered to include nitrogen-containing ligand systems.
Thereby, the very potent class of salen ligands (26,
27) gave the best results concerning the yields and the
diastereoselectivity in an intramolecular test reaction.
When the reaction was performed with a preformed
iron-salen complex the yields of intra- and intermo-
lecular epoxide ring expansion reactions were greatly
enhanced. Also, for the first time, internal alkenes
Scheme 5. Intermolecular iron-catalysed ring expansion re-
action with an internal double bond.
In this reaction, for the first time, an internal ac- could be used as substrates in intermolecular ring ex-
ceptor-substituted double bond was brought to reac- pansion reactions so that the scope of the iron-cata-
tion and the yield of 69% is excellent compared to lysed process was enlarged during this study.
earlier results in intermolecular reactions.[2] Further-
more, the product 42 was isolated as a single diaste-
reomer with a trans correlation of the phenyl substitu-
Experimental Section
General Remarks
ent and the imide moiety as was concluded from the
coupling constants of the corresponding protons of
7.4 and 3.0 Hz, nOe experiments and an X-ray struc-
ture of 42.
Another most considerable improvement in the
iron-catalysed ring expansion reaction was achieved
when the preformed Fe(II)-(salen) complex was used
in the intermolecular reaction of styrene oxide (1)
with 2,3-dimethyl-1,3-butadiene to yield the tetrahy-
drofuran derivatives 43a and b (Scheme 6). In this re-
action with the preformed iron-salen complex the
regio- and diastereoselectivities are also very good
NMR spectra were recorded on a Bruker Avance 300 or
DRX 500 (1H: 300 MHz or 500 MHz, 13C: 75 MHz or
125 MHz) spectrometer using TMS as internal standard (d=
0) unless otherwise noted. Mass and GC-mass spectra were
measured on a Hewlett Packard 6890 GC-System including
a Hewlett Packard 5973 Mass Selective Detector. For (high
resolution) mass spectra a Finnigan MAT 95S and a Finni-
gan LTQ (ESI, HRMS) spectrometer were used. Analytical
but the yield of these quite polymerisation-sensitive thin layer chromatography was performed on Merck silica
Adv. Synth. Catal. 2006, 348, 1241 – 1247
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
1245

Gerhard Hilt et al.
FULL PAPERS
gel 60 F254. For column chromatography Merck silica gel 60
(230–400 mesh ASTM) was used. All reactions were carried
out under inert atmosphere (nitrogen or argon) using stan-
dard Schlenk techniques. Dichloromethane and acetonitrile
were dried over phosphorus pentoxide, tetrahydrofuran and
diethyl ether over sodium. 5-Oxo-5-phenylpentanal (7) was
synthesised according to the literature method of Hsu.[3]
The ligands were prepared according to or by adapting liter-
ature methods.[8] For the larger scale reaction the Fe(II)-
(salen) complex was synthesised by an adapted method.[9]
bined organic phases were washed with water (50 mL),
brine (50 mL), dried over sodium sulfate and concentrated
under reduced pressure. The residue was purified by column
chromatography (triethylamine:ethyl acetate:pentane=
1:5:100) to yield 4 as a colourless oil as a mixture of diaste-
reomers (E:Z=1.0:1.5); yield: 420 mg (1.6 mmol, 80%).
1H NMR (300 MHz, CDCl3; main diastereomer, Z-isomer):
d=7.66–7.44 (m, 10H), 6.67 (d, 1H, J=11.6 Hz), 5.85 (dt,
1H, J=11.7, 7.0 Hz), 3.19 (d, 1H, J=5.3 Hz), 2.97 (d, 1H,
J=5.3 Hz), 2.66–2.57 (m, 2H), 2.06–1.90 (m, 2H), 1.90–1.69
(m, 2H). Further resolved signals of the minor diastereomer
(E-isomer): d=6.41 (dt, 1H, J=15.9, 6.9 Hz), 3.22 (d, 1H,
J=5.3 Hz), 2.53–2.43 (m, 2H). 13C NMR (75 MHz, CDCl3;
main diastereomer, Z-isomer): d=139.8, 137.5, 132.2, 129.1,
128.6, 128.3, 128.2, 128.0, 127.3, 126.4, 125.8, 55.4, 34.9, 28.3,
25.1. Further resolved signals of the minor diastereomer (E-
isomer): d=139.9, 137.6, 130.2, 130.1, 128.8, 128.3, 60.1, 34.8,
32.7, 24.5. MS (EI): m/z (%)=264 (M+, 1), 250 (3), 144 (3),
130 (100), 115 (27), 105 (16), 91 (21), 77 (27), 65 (4), 51 (7).
HM-RS (EI): m/z=264.1519, calcd. for C19H20O: 264.1514.
1,6-Diphenylhex-5-en-1-one (8)
Potassium tert-butoxide (15.0 mmol, 1.68 g) was suspended
in 70 mL of dichloromethane and cooled to 08C. Benzyltri-
phenylphosphonium bromide was added and the solution
turned red. After stirring for 10 min, the solution was cooled
to À788C and 7 (12.5 mmol, 2.21 g) dissolved in 20 mL of di-
chloromethane was slowly added by syringe. The solution
was kept at À788C for 1 h and stirred at room temperature
for 20 h. After addition of water (100 mL) the phases were
separated and the water phase was extracted with dichloro-
General Procedure for Iron-Catalysed Intramolecular
methane (3”30 mL). The combined organic phases were Ring Expansion of Epoxides (Ligand Screening)
washed with water (50 mL) and brine (50 mL), dried over
MgSO4 and concentrated under reduced pressure. The resi-
due was purified by column chromatography (ethyl acetate/
pentane, 1:10) to afford 8 as a colourless oil as a mixture of
diastereomers (E:Z=1.0:1.4); yield: 2.47 g (9.8 mmol,
78%). 1H NMR (300 MHz, CDCl3; main diastereomer, Z-
isomer): d=7.99–7.91 (m, 2H). 7.60–7.16 (m, 8H), 6.47 (d,
1H, J=11.8 Hz), 5.69 (dt, 1H, J=11.6, 7.2 Hz), 2.99 (t, 2H,
J=7.3 Hz), 2.44 (qd 2H, J=7.5, 1.8 Hz), 1.89–1.86 (m, 2H).
Further resolved signals of the minor diastereomer (E-
isomer): d=6.42 (d, 1H, J=16.0 Hz), 6.23 (dt, 1H, J=15.8,
6.8 Hz), 3.03 (t, 2H, J=7.3 Hz), 2.33 (q, 2H, J=7.2 Hz).
13C NMR (75 MHz, CDCl3; main diastereomer, Z-isomer):
d=199.9, 137.4, 136.8, 132.8, 131.8, 129.5, 128.6, 128.4, 128.0,
127.8, 126.4, 37.8, 27.9, 24.2. Further resolved signals of the
minor diastereomer (E-isomer): d=200.0, 137.5, 136.9,
130.5, 129.8, 128.3, 126.8, 125.8, 37.6, 32.3, 23.6. MS (EI): m/
z (%)=250 (M+, 3), 130 (100), 115 (12), 105 (10), 91 (3), 77
(16), 51 (2); HM-RS (EI): m/z=250.1353, calcd. for
C18H18O: 250.1358.
Iron dichloride (7 mg, 57 mmol), the ligand (57 mmol or 114
mmol when 2.0 equivs. of ligand were applied) and zinc
(25 mg, 378 mmol) were suspended in 1.0 mL of acetonitrile
and triethylamine (9 mg, 85 mmol, 10 mL) was added. Then
the solution was heated until boiling. Afterwards, the epox-
yalkene 4 (50 mg, 189 mmol) was added by syringe and the
solution was stirred at 608C for 20 h. Then the mixture was
filtered over a pad of silica (eluent: diethyl ether) and con-
centrated under reduced pressure. The residue was purified
by column chromatography (ethyl acetate:pentane, 1:50)
1,3a-Diphenylcyclopenta[c]tetrahydrofuran (5a)
According to the general procedure the product 5 was ob-
tained as a colourless oil. Yields and diastereoselectivities,
see Tables 1–3. 1H NMR (300 MHz, CDCl3): d=7.40–7.08
(m, 10H), 4.45 (d, 1H, J=7.6 Hz), 4.20 (d, 1H, J=9.0 Hz),
3.75 (d, 1H, J=9.0 Hz), 2.79–2.68 (m, 1H), 2.09–1.73 (m,
6H); 13C NMR (125 MHz, CDCl3): d=148.8, 142.0, 128.4,
128.3, 127.3, 125.9, 125.8, 125.7, 88.8, 80.8, 60.7, 60.2, 37.7,
30.9, 25.0; MS (EI): m/z (%)=264 (M+, 25), 234 (37), 191
(17), 173 (17), 158 (100), 143 (80), 129 (72), 115 (67), 105
(35), 91 (95), 77 (41), 67 (18), 51 (15); HR-MS (EI): m/z=
264.1508, calcd. for C19H20O: 264.1514.
2-Phenyl-2-(5-phenyl-pent-4-en-1-yl)oxirane (4)
Sodium hydride (162 mg, 60% in mineral oil, 6.6 mmol) was
suspended in 5 mL of dimethyl sulfoxide. The suspension
was carefully heated until gas evolution started. Then the
solution was stirred for 20 min, 40 mL of tetrahydrofuran
were added and the mixture was cooled to 08C. After the
addition of trimethylsulfonium iodide (1.10 g, 6.0 mmol) the
solution was stirred for 1 h at 08C, compound 8 (500 mg,
2.0 mmol) was added by syringe and the mixture stirred for
24 h at room temperature. Then the solution was diluted
with water (50 mL) and methyl tert-butyl ether (50 mL) was
added. The phases were separated and the water phase was
extracted with methyl tert-butyl ether (3”50 mL). The com-
3,5-Diphenyltetrahydrofuro
(42)
ACHTRE[UNG 2,3-c]pyrrole-4,6-dione
Fe(II)-(salen) (64 mg, 0.20 mmol) and zinc powder (91 mg,
1.40 mmol) were suspended in 1.0 mL of acetonitrile and
triethylamine (30 mg, 0.30 mmol, 42 mL) was added. Then
the mixture was heated until boiling. After 5 min, N-phenyl-
1246
www.asc.wiley-vch.de
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 1241 – 1247

Ring Expansion Reactions of Epoxyalkenes
FULL PAPERS
maleimide (173 mg, 1.00 mmol) and styrene oxide (360 mg,
3.00 mmol, 0.34 mL) were added. After stirring for 14 h at
608C the mixture was filtered through a pad of silica
(eluent: diethyl ether) and concentrated under reduced
pressure. The residue was purified by column chromatogra-
phy (ethyl acetate:pentane, 1:4) to furnish the diastereo-
merically pure product 42 as a white solid; yield: 202 mg
(0.69 mmol, 69%); mp 166–1688C; 1H NMR (500 MHz,
CDCl3): d=7.53–7.48 (m, 2H). 7.42–7.29 (m, 8H), 5.08 (d,
1H, J=7.4 Hz), 4.23–4.18 (m, 2H), 3.88–3.84 (m, 1H), 3.60
(dd, 1H, J=7.4, 3.0 Hz); 13C NMR (75 MHz, CDCl3): 175.0,
173.1, 140.9, 131.3, 129.2, 129.1, 128.9, 127.5, 126.8, 126.2,
77.9, 74.5, 53.2, 49.0; MS (ESI): m/z (%)=316 (M+ +Na,
100), 294 (10), 277 (45), 198 (5); HRMS (ESI): m/z=
316.0944, calcd. for C18H15NNaO3: 316.0950.
Ndene-Schiffer, M. Pierobon, S. Grimme, M. Geren-
kamp, C. Mꢁck-Lichtenfeld, Eur. J. Org. Chem. 2004,
2337; c) A. Gansäuer, S. Narayan, Adv. Synth. Catal.
2002, 344, 465; d) A. Gansäuer, H. Bluhm, T. Lauter-
bach, Adv. Synth. Catal. 2001, 343, 785; e) A. Gansäuer,
M. Pierobon, H. Bluhm, Synthesis 2001, 2500; f) A. Gan-
säuer, M. Pierobon, H. Bluhm, Angew. Chem. 1998, 110,
107; Angew. Chem. Int. Ed. 1998, 37, 101; g) S. C. Roy, J.
Indian Inst. Sci. 2001, 81, 477; h) K. Daasbjerg, H. Svith,
S. Grimme, M. Gerenkamp, C. Mꢁck-Lichtenfeld, A.
Gansäuer, A. Barchuk, F. Keller, Angew. Chem. 2006,
118, 2095; Angew. Chem. Int. Ed. 2006, 41, 2045.
[6] E. N. Jacobsen, M. H. Wu, in: Comprehensive Asymmet-
ric Catalysis, Vols. I–III, (Eds.: E. N. Jacobsen, A. Pfaltz,
H. Yamamoto), Springer Verlag, Berlin 1999,
Chapt.18.2, p 649.
[7] A detailed study on intramolecular ring expansion reac-
tion will be published elsewhere.
References
[8] a) A. Simion, C. Simion, T. Kanda, S. Nagashima, Y.
Mitoma, T. Yamada, K. Mimura, M. Tashiro, J. Chem.
Soc., Perkin Trans. 1 2001, 2071; b) K. Oyaizu, E. Tsu-
chida, Inorg. Chim. Acta 2003, 355, 414; c) M. Cheng,
D. R. Moore, J. J. Reczek, B. M. Chamberlain, E. B.
Lobkovsky, G. W. Coates, J. Am. Chem. Soc. 2001, 123,
8738; d) M. Stender, R. J. Wright, B. E. Eichler, J. Prust,
M. M. Olmstead, H. W. Roesky, P. P. Power, J. Chem.
Soc., Dalton Trans. 2001, 3465; e) K. J. Miller, T. T. Kita-
gawa, M. M. Abu-Omar, Organometallics 2001, 20, 4403.
[9] C. S. Marvel, S. A. Aspey, E. A. Dudley, J. Am. Chem.
Soc. 1956, 78, 4905.
[1] C. Bolm, J. Legros, J. Le Paih, L. Zani, Chem. Rev. 2004,
104, 6217 and references cited therein.
[2] G. Hilt, P. Bolze, I. Kieltsch, Chem. Commun. 2005,
1996.
[3] An X-ray analysis of a corresponding crystalline product
revealed that the relative stereochemistry is opposite to
the prior published stereochemistry (see ref.[2]). The
major diastereomer is the cis-product 5a.
[4] J.-L. Hsu, J.-M. Fang, J. Org. Chem. 2001, 66, 8573.
[5] For the mechanistically related Cp2TiACHTRE(UNG III)Cl chemistry,
see: a) T. V. RajanBabu, W. A. Nugent, J. Am. Chem.
Soc. 1994, 116, 986; b) A. Gansäuer, B. Rinker, N.
Adv. Synth. Catal. 2006, 348, 1241 – 1247
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
1247