[3+2] Cycloaddition reactions of arylacetylenes with carbonyl ylides derived from 1-aryl-1-diazohexane-2,5-diones

Article abstract of DOI:10.1016/S0040-4039(02)00630-5

The synthesis and 1,3-dipolar cycloaddition reactions of α-aryl-α-diazodiones with aryl acetylenes (up to 76% ee) are described.

Full text of DOI:10.1016/S0040-4039(02)00630-5

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 3927–3930
[3+2] Cycloaddition reactions of arylacetylenes with carbonyl
ylides derived from 1-aryl-1-diazohexane-2,5-diones
David M. Hodgson,a,* Rebecca Glena and Alison J. Redgraveb
aDyson Perrins Laboratory, Department of Chemistry, University of Oxford, South Parks Road, Oxford OX1 3QY, UK
bGlaxoSmithKline Medicines Research Centre, Gunnels Wood Road, Stevenage SG1 2NY, UK
Received 14 February 2002; accepted 28 March 2002
Abstract—The synthesis and 1,3-dipolar cycloaddition reactions of a-aryl-a-diazodiones with aryl acetylenes (up to 76% ee) are
described. © 2002 Elsevier Science Ltd. All rights reserved.
There are currently few methods to achieve catalytic
enantioselective 1,3-dipolar cycloadditions, despite the
potential utility of such asymmetric transformations.1
Following our report of the first examples of enantiose-
lective tandem carbonyl ylide formation–cycloaddition
in 1997,2 we have been interested in exploring its scope
and limitations.3 In order to begin to elucidate the
factors influencing enantioselectivity in this new asym-
metric process we focused initially on electronic effects
within the dipole. Work by Hashimoto et al. (Scheme 1,
X=H)4 and ourselves (Scheme 1, X=Me,5 CO2Et3)
illustrates that the nature of the dipole (be it steric
and/or electronic) has an important bearing upon the
level of asymmetry induced and we wished to investi-
gate this further. We selected a-aryl-a-diazodiones (e.g.
1, Scheme 2) as test substrates, since substituent varia-
tion within the aryl group (for example p-methoxy or
O
OH
O
a, b
83%
MeO OMe
MeO OMe
c
N
OMe
quant.
O
O
Rh
Rh
O
O
37%
O
N
f
d
72%
O
MeO OMe
MeO OMe
4
O
N2
Ph
Rh2(S-BPTV)4
MeO2C CO2Me
O
X
O
O
O
NO2
X
Rh2Ln
O
e
65%
g
64%
PhCF3
O
N2
O
N2
X = H
80% ee (50% y)4
51% ee (65% y)5
CO2Me
Ph
X = Me
X = CO2Et 33% ee (23% y)3
O
MeO2C
O
O
NO2
2
1
O
X
Scheme 2. (a) MeOH, (MeO)3CH, H2SO4, reflux, 18 h; (b)
Me(MeO)NH·HCl, i-PrMgCl, THF −10°C, 2.5 h; (c)
MeMgBr, THF, 0°C, 1.5 h; (d) BnMgCl, THF, 0°C, 2 h; (e)
4-(NHAc)C6H4SO2N3, DBU, MeCN, 0°C, 16 h, then 10%
citric acid; (f) t-BuOK, 4-(NO2)C6H4CHO, NH3, −70°C, 8 h;
(g) 4-(NHAc)C6H4SO2N3, Et3N, MeCN, 0°C, 2 h, then 10%
citric acid.
Scheme 1.
Keywords: asymmetric catalysis; carbonyl ylides; cycloadditions;
diazo compounds; rhodium.
* Corresponding author. E-mail: david.hodgson@chem.ox.ac.uk
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)00630-5

3928
D. M. Hodgson et al. / Tetrahedron Letters 43 (2002) 3927–3930
p-nitro groups) allows electronic perturbations on the
system to be analysed without concomitant alteration
in the steric environment of the dipole. Intermolecular
cycloadditions also provided opportunities for ready
variation of the dipolarophile electronics using, for
example, aryl acetylenes.
It was found that, pleasingly, such substrates under-
went the tandem catalytic ylide generation–intermolecu-
lar cycloaddition process9 with Rh2(OAc)4 to deliver the
cycloadducts 6 from a range of (commercially avail-
able) aryl acetylenes 3–5 in good to excellent yields
(Scheme 3, Table 1).10 Only one of the possible
regioisomers was observed in each crude reaction mix-
ture and, once isolated, the cycloaddition regiochem-
istry was determined by NOE experiments to be the
same for all cycloadducts (Scheme 3).11 In the absence
of catalyst and using heat to decompose the diazo-
diones, the cycloadducts were formed with the same
regioselectivity, but in much reduced yields (2–19%).
The previously unknown a-aryl-a-diazo-containing
diones were synthesised from levulinic acid via the
sequence in Scheme 2 (attempted preparation of the
p-methoxy derivative failed due to the instability of the
diazodione). For the phenyl diazodione 1, DBU instead
of Et3N was necessary to effect diazo transfer6 and the
presence of the ketal at the diazo transfer step was
crucial to prevent cyclopentenone formation via an
intramolecular aldol reaction.7 For nitrophenyl diazodi-
In order to examine any effect a catalyst might have on
relative rates of the cycloadditions each diazodione was
first individually heated in the presence of equal quanti-
ties (5 equiv.) of all three acetylenes and the product
ratio determined. Similar experiments were then carried
out (at room temperature) with Rh2(OAc)4 and two of
the chiral catalysts which are currently best able to
deliver asymmetric induction in carbonyl ylide cycload-
ditions: carboxylate catalyst Rh2(S-BPTV)4 (shown in
Scheme 1) and phosphate catalyst Rh2(R-BNP)4 (Fig.
1, Table 2).
one
2 an unusual formyl displacement from p-
nitrobenzaldehyde8 was used to generate the a-aryl
ketone.
Z
H
O
N2
Z
O
O
O
Y
Z = OMe 3
For phenyl diazodione 1, in the absence of a catalyst,
reaction with nitrophenylacetylene dominated (Table 2,
entry 1), whereas for nitrophenyl diazodione 2 reaction
with methoxy- and nitrophenylacetylenes were slightly
favoured over phenylacetylene (Table 2, entry 5).
Cycloaddition competition experiments with phenyl
diazodione 1 under dirhodium catalysis gave a signifi-
cantly greater proportion of cycloaddition from nitro-
phenylacetylene, with all three catalysts showing a
Z = H
Z = NO2
4
5
Y = H
Y = NO2
1
2
Y
6
Scheme 3.
Table 1. Cycloadditions using Rh2(OAc)4
a
Entry
Diazodione
Dipolarophile
Yieldb (%)
1
2
3
4
5
6
1
1
1
2
2
2
3
4
5
3
4
5
76
60
82
89
82
58
R
R = H
R = C12H25
O
O
O Rh
O Rh
Rh2(R-BNP)4
Rh2(R-DDBNP)4
P
R
a All reactions carried out in toluene (0.05 M) at 25°C with 5 equiv.
acetylene and 0.5–1 mol% Rh2(OAc)4.
b Isolated yields.
4
Figure 1.
Table 2. Cycloaddition competition reactions
Entrya
Diazodione
Catalyst
Solvent
Cycloadduct ratiob from dipolarophile
3
4
5
1
2
3
4
5
6
7
8
1
1
1
1
2
2
2
2
None
Toluene
Toluene
Toluene
PhCF3
Toluene
Toluene
Toluene
PhCF3
1
1
1.5
1
2.5
1.5
2
1
1
1
1
1
1
1
1
4
9
8
11
2
1
1
2.5
Rh2(OAc)4
Rh2(R-BNP)4
Rh2(S-BPTV)4
None
Rh2(OAc)4
Rh2(R-BNP)4
Rh2(S-BPTV)4
2.5
a Carried out at room temperature, apart from entries 1 and 5 at reflux.
1
b Ratios of the three cycloadducts produced could be determined from the H NMR spectrum (C6D6) of the crude reaction mixture by integration
of the distinct olefinic H singlets.

D. M. Hodgson et al. / Tetrahedron Letters 43 (2002) 3927–3930
3929
similar product profile (Table 2, entries 2–4). These
results indicate that the catalyst may be influencing the
energies of the molecular orbitals of the ylide. However,
with nitrophenyl diazodione 2 there was no significant
change in cycloadduct profile under dirhodium cataly-
sis, suggesting that during cycloaddition the catalysts
do not strongly perturb the molecular orbital energies
of the comparatively more electron deficient ylide
derived from 2.
BNP)4 and Rh2(S-BPTV)4 were the same as when each
acetylene was used in isolation.] Thus, a simple reaction
scheme whereby a reduction in ee is due to loss of
catalyst from the ylide, followed by cycloaddition on
the achiral ylide is shown not to be valid,1c at least in
these experiments.
Whilst only modest levels of asymmetric induction are
recorded in Table 3, these are the first examples of
phenyl-substituted dipoles and of acetylenes (other than
diester-substituted systems) in such cycloadditions and
provide encouragement to investigate further substrate
and catalyst combinations in this newly emerging enan-
tioselective process. Indeed, the following experiments
show that improvement in asymmetric induction with
these types of substrates is certainly possible. From
Table 3, it is evident that the highest ees were generally
obtained using phenylacetylene as dipolarophile and
further experimentation established that using phenyl-
acetylene as both the dipolarophile and reaction sol-
vent proved advantageous, with 66% ee (yield 73%)
being obtained at room temperature from the nitro-
phenyl-substituted diazodione 2 and Rh2(R-BNP)4.
Changing to the more hydrocarbon-soluble Rh2(R-
Knowing the relative rates of cycloaddition, a series of
experiments were then carried out to determine the
influence of the dipole and dipolarophile on asymmetric
induction. Cycloadditions were carried out with each
diazodione and dipolarophile combination and two sol-
vent–catalyst combinations (Table 3). Yields of
cycloadducts are generally lower than those obtained
using Rh2(OAc)4; Rh2(S-BPTV)4 is slightly more
efficient than Rh2(R-BNP)4 in all cases [except with
nitrophenyl dione 2 and nitrophenylacetylene 5 when
Rh2(S-BPTV)4 gives a significant improvement in yield
(Table 3, entries 9 and 12)]. Each of the four possible
diazo substrate–catalyst interactions will initially gener-
ate a catalyst-associated ylide with a certain level of
diastereomeric excess; the results indicate that each of
these catalyst-associated ylides do not lead to equally
enantioenriched cycloadducts with the three acetylenic
dipolarophiles [except for the case of the nitrophenyl-
substituted dipole and Rh2(S-BPTV)4 (Table 3, entries
10–12)]. The absolute values of ee are modest, but the
differences in ee between cycloadducts arising from the
same catalyst-associated ylide and the three different
dipolarophiles provide the first unambiguous demon-
stration that electronic effects play a role in determining
the level of asymmetric induction in such cycloaddition
processes. Moreover, for the phenyl-substituted dipole,
where the competition reactions conclusively ascer-
tained that nitrophenylacetylene was the most reactive
dipolarophile (Table 2, entries 3 and 4), it is shown that
higher rates of cycloaddition do not lead to better
asymmetric induction (Table 3, entries 1–6). [The ees
obtained in the competition reactions with Rh2(R-
3
DDBNP)4 improved the yield to 80%, whilst the ee
remained the same (66%). At 0°C the ee increased to
76% (yield 83%); however, further lowering of the
temperature totally inhibited diazo decomposition.
In summary, a route to a-aryl-a-diazodiones has been
established; such substrates have been shown to
undergo tandem ylide formation–cycloaddition pro-
cesses. The first asymmetric induction in intermolecular
cycloadditions of carbonyl ylides with aryl acetylenes
has been observed, with the binaphthyl catalysts giving
the best enantioselectivities (up to 76% ee) and extend-
ing their usefulness.3,12 These results suggest that a
complex blend of electronic effects from the dipole and
dipolarophile, together with the nature of the catalyst
contribute to the origin of asymmetric induction. Fur-
ther studies are underway to investigate enantioselectiv-
ity in such cycloaddition processes.
Table 3. Enantioselective cycloadditions
Entry
Diazodione
Dipolarophile
Catalyst
Solvent
Yielda (%)
Eeb (%)
[h]TDc
1
2
3
4
5
6
7
8
1
1
1
1
1
1
2
2
2
2
2
2
3
4
5
3
4
5
3
4
5
3
4
5
Rh2(R-BNP)4
Rh2(R-BNP)4
Rh2(R-BNP)4
Rh2(S-BPTV)4
Rh2(S-BPTV)4
Rh2(S-BPTV)4
Rh2(R-BNP)4
Rh2(R-BNP)4
Rh2(R-BNP)4
Rh2(S-BPTV)4
Rh2(S-BPTV)4
Rh2(S-BPTV)4
Toluene
Toluene
Toluene
PhCF3
PhCF3
PhCF3
Toluene
Toluene
Toluene
PhCF3
PhCF3
PhCF3
41
37
50
43
47
73
48
54
33
65
69
72
10
38
31
3
28
1
18
48
31
40
43
38
+12.2
−63.1
+27.6
+2.7
+48.7
+2.0
−43.8
−105.5
+49.0
+70.7
+97.0
+54.4
9
10
11
12
a Isolated yields.
b Ees determined by chiral HPLC using Daicel Chiralpak AD (Chiralcel OD for entries 2 and 5) column with heptane/EtOH eluant.
c Entries 4–6 and 10–12: c=1, T=25°C; remaining c=0.88–1.12, T=20–24°C. All in CHCl3.

3930
D. M. Hodgson et al. / Tetrahedron Letters 43 (2002) 3927–3930
9. After commencement of our work the use of a-aryl-a-
Acknowledgements
diazo ketones in Rh(II)-catalysed oxidopyrylium cycload-
ditions was reported: Padwa, A.; Precedo, L.; Semones,
M. A. J. Org. Chem. 1999, 64, 4079–4088.
We thank the EPSRC and GlaxoSmithKline for a
CASE award (to R.G.), Dr B. Odell for NMR experi-
ments and the EPSRC National Mass Spectrometry
Service Centre for mass spectra.
10. Example procedure: 1-diazo-1-phenylhexane-2,5-dione (1)
(35 mg, 0.16 mmol) was dissolved in toluene (3 cm3) and
phenylacetylene (88 mL, 0.80 mmol) added. Rh2(Oac)4
(0.7 mg, 1 mol%) was added and the mixture stirred for
30 min at room temperature. The solvent was evaporated
under reduced pressure. Purification of the residue by
column chromatography [SiO2, 20:1 Et2O:light petroleum
(bp 40–60°C)] gave a white solid, 5-methyl-1,7-diphenyl-
8-oxabicyclo[3.2.1]oct-6-en-2-one 6 (Y, Z=H) (28 mg,
60%): Rf=0.65 [1:1 light petroleum (bp 40–60°C):Et2O];
mp 122–123°C [light petroleum (bp 40–60°C):EtOAc];
IR: wmax (KBr disc)/cm−1=2921 (w), 2854 (w), 1716 (s),
References
1. (a) Gothelf, K. V.; Jørgensen, K. A. Chem. Rev. 1998, 98,
863–909; (b) Karlsson, S.; Ho¨gberg, H.-E. Org. Prep.
Proced. Int. 2001, 33, 105–172; (c) Hodgson, D. M.;
Pierard, F. Y. T. M.; Stupple, P. A. Chem. Soc. Rev.
2001, 30, 50–61; (d) Kitagaki, S.; Hashimoto, S. J. Synth.
Org. Chem. Jpn 2001, 59, 1157–1169; (e) Hodgson, D.
M.; Stupple, P. A.; Forbes, D. C. In Rodd’s Chemistry of
Carbon Compounds, Topical Volume, Asymmetric Cataly-
sis; Sainsbury, M., Ed.; Elsevier: Oxford, 2001; pp. 65–
99; (f) Forbes, D. C.; McMills, M. C. Curr. Org. Chem.
2001, 5, 1091–1105.
1
1448 (m), 1373 (w) and 1187 (m); H NMR (400 MHz;
CDCl3): l=7.42−7.40 (2H, m, 2×aromatic H), 7.31–7.20
(8H, m, 8×aromatic H), 6.54 (1H, s, ꢀCH), 2.92–2.74
(2H, m, COCH2), 2.43 (1H, ddd, J=13.5, 8.5 and 8.5,
COCH2CHH), 2.07 (1H, ddd, J=13.5, 8.5 and 2,
COCH2CHH) and 1.63 (3H, s, Me); 13C NMR (100
MHz; CDCl3): l=203.5 (CꢀO), 144.0 (aromatic C), 135.4
(ꢀCH), 135.0 (aromatic C), 132.1 (Cꢀ), 129.2 (2×aromatic
CH), 128.5 (aromatic CH), 128.3 (2×aromatic CH), 128.1
(2×aromatic CH), 128.0 (aromatic CH), 126.7 (2×aro-
matic CH), 95.1 (PhC-O), 84.5 (MeC-O), 35.0 (CH2),
34.7 (CH2) and 24.4 (Me); anal. calcd for C20H18O2: C,
82.7; H, 6.25%. Found: C, 82.6; H, 6.25; MS (CI, NH3):
m/z (%)=308 (100) [M+NH4+] and 291 (35) [M+H+]
(found: [M+NH+4], 308.1649. C20H22NO2 requires
308.1651).
2. Hodgson, D. M.; Stupple, P. A.; Johnstone, C. Tetra-
hedron Lett. 1997, 38, 6471–6472.
3. (a) Hodgson, D. M.; Stupple, P. A.; Johnstone, C. Chem.
Commun. 1999, 2185–2186; (b) Hodgson, D. M.; Stupple,
P. A.; Pierard, F. Y. T. M.; Labande, A. H.; Johnstone,
C. Chem. Eur. J. 2001, 7, 4465–4476.
4. Kitagaki, S.; Anada, M.; Kataoka, O.; Matsuno, K.;
Umeda, C.; Watanabe, N.; Hashimoto, S. J. Am. Chem.
Soc. 1999, 121, 1417–1418.
5. Donohue, A. C. University of Oxford, unpublished
11. Regiochemistry was confirmed by 2D NOE NMR experi-
ments which showed a strong correlation between the
methyl group and the olefinic H in each case. Aromatic
p–p interactions may contribute to the origin of the
regiochemistry, see: Ibata, T.; Motoyama, T.;
Hamaguchi, M. Bull. Chem. Soc. Jpn. 1976, 49, 2298–
2301.
results.
6. Mateos, A. F.; Coca, G. P.; Alonso, J. J. P.; Gonza´lez, R.
R.; Herna´ndez, C. T. Synlett 1996, 1134–1136.
7. Treatment of the diketone 1 (N2ꢀH2) with base (in our
case DBU) leads to 3-methyl-2-phenylcyclopent-2-enone,
see: Pecunioso, A.; Menicagli, R. J. Org. Chem. 1988, 53,
2614–2617.
12. Hodgson, D. M.; Petroliagi, M. Tetrahedron: Asymmetry
2001, 12, 877–881.
8. Iwasaki, G.; Saeki, S.; Hamana, M. Chem. Lett. 1986,
173–176.