Synthesis of α-iminoimidates by palladium catalysed double isonitrile insertion

Article abstract of DOI:10.1039/b408673m

The synthesis of α-iminoimidates by palladium catalyzed double isonitrile insertion was described. The influence of the quantity of isonitrile was also studied. It was observed that the bidentate ligands dppe and dppp gave poor conversions with extensive

Full text of DOI:10.1039/b408673m

View Article Online / Journal Homepage / Table of Contents for this issue
Synthesis of ꢀ-iminoimidates by palladium catalysed double
isonitrile insertion
Richard J. Whitby,*a C. Gustaf Salustea and Mark Furberb
a School of Chemistry, University of Southampton, Southampton, HANTS, UK SO17 1BJ.
E-mail: rjw1@soton.ac.uk; Tel: 44 23 80592777
b AstraZeneca Charnwood, Department of Medicinal Chemistry, Bakewell Road,
Loughborough, Leics, UK LE11 5RH
Received 8th June 2004, Accepted 9th June 2004
First published as an Advance Article on the web 18th June 2004
Palladium catalysed selective double insertion of isonitriles
into aryl bromides with trapping by sodium alkoxides
provides an efficient 4-component synthesis of unusual
ꢀ-iminoimidates.
The influence of the quantity of isonitrile was next studied,
and as expected increased amounts gave a much higher ratio of
4 : 3 (Table 1, entries 1–6). With 5 equiv of cyclohexylisocyanide
with respect to bromobenzene greater than 90% selectivity for
the bis-insertion product 4 is obtained (entry 6) with only trace
amounts (detected by electrospray MS) of tris-inserted product.
It is significant that even at very low isonitrile concentrations
substantial amounts of the bis-inserted product 4 are formed
(Table 1, entry 2). The reaction was also found to be faster
at lower isonitrile concentrations, possibly due to partial
deactivation of the palladium catalyst by the isonitrile at higher
concentrations.† Lower temperatures strongly favoured
formation of the bis-inserted product 4 (Table 1 entries 1, 7–9),
but at the expense of long reaction times and poorer yields.
We then turned our attention to the effect of the amount
of sodium ethoxide used. We changed to dioxane as solvent,
preliminary experiments demonstrating that it gave similar 3 : 4
ratios as toluene and had the advantage that mixtures remained
homogeneous. Using a constant amount of ethanol, to reduce
solvent effects, we varied the amount of sodium ethoxide
and found that lower concentrations strongly favoured the
formation of the bis-inserted product 4 (Table 2 entries 1–4).
For preparative reactions when variation in the alkoxide com-
ponent is required it would be more convenient to use a base
separate from the alcohol component of the α-iminoimidate.
Use of Cs2CO2 and K2CO3 gave poor conversions, but NaOtBu
worked well. To determine the optimum amount of alcohol we
carried out a series of experiments in which the amount of
ethanol was varied and were interested to find a dramatic effect
on the ratio of 3 : 4 (Table 2, entries 5–8). Use of solid NaOtBu
to further reduce the amount of hydroxylic solvent present,
with 5 equivalents of ethanol to avoid formation of products
from tBuOϪ addition, gave optimum conditions for both yield,
and selectivity for formation of the α-iminoimidate 4 (Table 2
entry 9). We also examined how the ethoxide metal counterion
affected the 3 : 4 ratio. LiOtBu gave a very slow and messy
reaction, but NaOtBu, KOtBu and CsOtBu gave good yields of
the desired products, and with a dramatic preference for mono-
insertion of isonitrile as the metal ion gets larger (i.e. as the
ethoxide anion gets more nucleophilic) (Table 2 entries 10–12).
A related observation was that addition of 15-crown-5 (2 equiv)
to reactions using NaOEt under the conditions of Table 1 entry
1 increased the ratio of 4 : 3 from 1.3 : 1 to 4.7 : 1.
We recently reported the palladium catalysed synthesis of
amidines, imidates, and thioimidates from aryl halides, iso-
nitriles, and amines, alcohols, and thiols respectively.1,2 In the
case of the formation of imidates 1 from aliphatic alcohols and
electron rich aryl bromides, yields under our initial conditions
were poor and the problem was traced to the formation of
substantial amounts of the product 2 of bis-insertion of the
isonitrile [eqn. (1)]. Selective mono-insertion was obtained by
slow addition of the isonitrile, but we now report optimisation
to afford the α-iminoimidates 2.
(1)
Palladium-catalysed double carbonylation reactions are
well-established for the synthesis of α-keto-esters and -amides.3
To our knowledge, metal catalysed double isonitrile insertion is
not known. The stoichiometric double and triple insertion of
isonitriles into palladium–carbon bonds is known4,5 as well
as palladium catalysed isonitrile polymerisation.6 Although
α-iminoimidates are virtually unknown,7 1,2-diimines have
found applications as ligands.8 A recent route is the palladium
catalysed reductive dimerisation of imidoyl chlorides.9
We first noted formation of α-iminoimidates in a system con-
sisting of bromobenzene (1 eq.), cyclohexylisonitrile (1.5 eq.),
NaOEt (5 eq., as 2 M soln, in EtOH), palladium dichloride
(5 mol%), 1,1Ј-bis(diphenylphosphinyl)ferrocene (dppf )
(10 mol%) in toluene at 98 ЊC which gave around a 1 : 1.3 ratio
of mono- : bis-inserted products 3 and 4 [eqn. (2)]. Use of
iodobenzene gave the same ratio of 3 : 4, but in a slow and
low yielding reaction. A range of alternative ligands were
investigated [1,3-bis(diphenylphosphinyl)propane (dppp), 1,2-
bis(diphenylphosphinyl)ethane (dppe), P(o-Tol)3, PPh3,
PPh2Me, P(Furyl)3, P(Cy)3 and P(tBu)3] but had little effect on
the bis : mono ratio (1.3–2). For PPh3 the ratio was 2.8, but
conversion was poor, and for P(o-tol)3 it was 1.1 and a very fast
reaction was noticed. The bidentate ligands dppe and dppp
gave poor conversions, with extensive formation of unidentified
side products.
The above experiments, and others, lead to conditions
optimised for the formation of bis-inserted products: 1.2 eq.
NaOtBu, 5 eq. R2OH, 3 eq. R1NC in dioxane at 98 ЊC. Under
these conditions, a range of α-iminoimidates 5 were obtained
in high yield (Table 3). Generally less than 5% of the
mono-inserted product was formed. The successful double
insertion using electron poor aromatic systems is notable since
under our original conditions for imidate formation,2 double
insertion was not observed. No double insertion was observed
using phenol as the nucleophile, a result which may be related to
(2)
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 1 9 7 4 – 1 9 7 6
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4
1974

View Article Online
Table 1 Effect of [CyNC] and temperature on formation of 3 and 4a
Entry
CyNC/mmol
Temp/ЊC
Time/h
3/%b
4/%b
4 : 3
1.30
0.52
0.66
1.22
4.73
11.69
5.71
5.76
2.02
1
2
3
4
5
6
7
8
9
0.6
0.1
0.2
0.4
0.8
2.0
0.6
0.6
0.6
98
98
98
98
98
98
50
70
90
2
2
2
2
2
36.7
10.0
15.3
21.8
7.0
5.2
0.7
47.8
5.2
10.1
26.7
33.1
60.8
4.0
2
60
60
12
6.6
20.0
38.0
40.3
a 0.4 mmol PhBr, 0.02 mmol PdCl2, 0.04 mmol dppf, 2 mmol EtONa in 1 mL EtOH, 3 mL toluene. b Yields by GC vs tridecane as internal standard.
Table 2 Effect of base and [ethanol] on formation of 3 and 4a
Entry
ROM/mmol
EtOH
Time/h
3/%b
4/%b
4 : 3
1
2
3
4
5
6
7
8
9
0.12 NaOEtc
0.24 NaOEtc
0.48 NaOEtc
1.0 NaOEtc
0.44 mL
0.38 mL
0.26 mL
0
10 mmol
1 mmol
0.5 mmol
0.25 mmol
1 mmol
1 mmol
1 mmol
1 mmol
15
15
15
15
15
15
15
15
15
4
2.0
2.9
6.1
13.4
10.9
7.9
1.8
0.7
5.3
5.3
27.8
41.0
48.5
49.4
43.9
53.3
62.0
56.4
92.3
51.0
25.3
10.2
13.9
14.1
7.9
3.7
4.0
0.24 NaOtBud
0.24 NaOtBud
0.24 NaOtBud
0.24 NaOtBud
0.24 NaOtBu
0.24 NaOtBu
0.24 KOtBu
6.7
34.4
80.6
17.4
9.6
0.44
0.16
10e
11e
12e
4
4
57.0
65.3
0.24 CsOtBu
a 0.2 mmol PhBr, 0.6 mmol CyNC, 0.01 mmol PdCl2, 0.02 mmol dppf, 2 mL dioxane, 98 ЊC. b By GC vs. tridecane as internal standard. c As a 2 M
soln. in EtOH. d As a 1M solution in tBuOH. e 0.3 mmol CyNC used.
Table 3 Preparation of α-iminoimidatesa
Entry
Ar
R1
R2
5
Yield (%)b
1
2
3
4
5
6
7
Ph
Ph
Ph
Cy
Bu
Cy
Cy
Bu
Cy
Cy
Et
Et
iPr
Et
Et
iPr
Et
a
b
c
d
e
f
79
70
88
72
74
71
73
p-MeOC6H4
p-MeOC6H4
p-MeOC6H4
p-NCC6H4
Fig. 1 Possible catalytic cycles.
g
a ArBr (1 equiv), tBuONa (1.2 equiv), R2OH (5 equiv), R1NC (3 equiv),
PdCl2 (0.05 eqiv), dppf (0.1 equiv), dioxane, 98 ЊC, 4h. b Isolated yield
of distilled product.
Pd–iminoacyl bond preceding alkoxide attack in formation of
the α-iminoimidate is one of many possible variations.
Overall we have described an efficient palladium catalysed
4-component synthesis of α-iminoimidates.
the reported failure of palladium catalysed bis-carbonylations
when aniline was the nucleophile.3
A crystal structure of 5c was obtained confirming the 1E,2Z
Acknowledgements
We thank AstraZeneca (Charnwood) for funding this work and
Professor M. B. Hursthouse and Dr M. E. Light of the EPSRC
national X-ray service for the crystal structure of 5c.
stereochemistry indicated and showing a 90Њ dihedral angle for
the N᎐CC᎐N group.‡ The stereochemistry of the products
᎐
᎐
probably tells us little about the mechanism of the reaction
since E/Z isomerisation of the C᎐N bonds is likely to occur
᎐
under the reaction conditions.10 The observed 1E,2Z isomer of
5 formed is probably the thermodynamically most stable of the
4 possibilities.§
The work described above was aimed at optimisation, rather
than mechanistic study, but the strong effects on the ratio of
mono- to bis-inserted products formed of both the con-
centration of isonitrile and the ethoxide concentration and
nucleophilicity, suggest a direct competition between the two at
the point where the catalytic cycles diverge. The catalytic cycle
shown in Fig. 1 is one possibility. Insertion of isonitrile into the
Notes and references
† Tris-insertion of isonitrile may give a very stable chelated palladium
complex—ref 5.
‡ Crystal data for 5c: C23H34N2O, Mr = 354.52, monoclinic, space group
P21, a = 9.1902(5), b = 14.4440(7), c = 15.8727(10) Å, β = 94.398(2)Њ, U =
2100.8(2) Å3, Z = 4, µ(Mo-Kα) = 0.068 mmϪ1, T = 120(2) K, 6538
independent reflections (Rint = 0.0447), final R1 = 0.0753, wR2 = 0.1018
for all data. CCDC reference number 232994. See http://www.rsc.org/
suppdata/ob/b4/b408673m/ for crystallographic data in .cif or other
electronic format.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 1 9 7 4 – 1 9 7 6
1975

View Article Online
§ DFT calculations using B3LYP method and 6.31G* basis set gave
the relative energies of (1E,2Z)-, (1E,2E)-, (1Z,2Z)-, and (1Z,2E)-
PhC(᎐NiPr)C(᎐NiPr)OEt as 0, 17.6, 19.6, and 31.8 kJ molϪ1
and H. Yamazaki, J. Chem. Soc., Dalton Trans., 1991, 1531;
(c) G. R. Owen, R. Vilar, A. J. P. White and D. J. Williams,
Organometallics, 2003, 22, 3025; (d ) J. Vicente, J. A. Abad,
A. D. Frankland, J. Lopez-Serrano, M. C. Ramirez de Arellano and
P. G. Jones, Organometallics, 2002, 21, 272.
᎐
᎐
respectively.
1 C. G. Saluste, R. J. Whitby and M. Furber, Angew. Chem., Int. Ed.,
2000, 39, 4156.
2 C. G. Saluste, R. J. Whitby and M. Furber, Tetrahedron Lett., 2001,
42, 6191.
3 F. Ozawa, H. Soyama, H. Yanigihara, I. Aoyama, H. Takino,
K. Izawa, T. Yamamoto and A. Yamamoto, J. Am. Chem. Soc.,
1985, 107, 3235.
4 (a) J. Vicente, J. A. Abad, K. F. Shaw, J. Gil-Rubio, M. C. Ramirez
de Arellano and P. G. Jones, Organometallics, 1997, 16, 4557;
(b) K. Onitsuka, K. Ogawa, T. Joh, S. Takahashi, Y. Yamamoto
5 Y. Yamamoto, T. Tanase, T. Yanai, T. Asano and K. Kobayashi,
J. Organomet. Chem., 1993, 456, 287.
6 F. Takei, K. Yanai, K. Onitsuka and S. Takahashi, Chem. Eur. J.,
2000, 6, 983.
7 L. De Vries, J. Org. Chem., 1971, 36, 3442.
8 G. van Koten and K. Vrieze, Advances in Organometallic Chemistry,
Academic Press, New York, 1982, vol. 21, pp. 151–239.
9 H. Amii, M. Kohda, M. Seo and K. Uneyama, Chem. Commun.,
2003, 1752.
10 W. Walter and C. O. Meese, Chem. Ber., 1977, 110, 2463.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 1 9 7 4 – 1 9 7 6
1976