Much improved conditions for the Negishi cross-coupling of iodoalanine derived zinc reagents with aryl halides

Article abstract of DOI:10.1021/jo902238n

(Chemical Equation Presented) A combination of Pd₂(dba) ₃ and SPhos (1:2 molar ratio) is an excellent precatalyst for the Negishi cross-coupling of the serine-derived organozinc reagent 2 with aryl halides, including previously difficult ortho-substituted examples. In the case of meta- and para-substituted aryl halides, Pd-loadings of 0.5 mol % give satisfactory results. Use of 2-iodoaniline as substrate gives the lactam 12 in good yield. 2009 American Chemical Society.

Full text of DOI:10.1021/jo902238n

pubs.acs.org/joc
Negishi reaction.) Use of a combination of Pd(OAc)2 (5 mol
Much Improved Conditions for the Negishi Cross-
Coupling of Iodoalanine Derived Zinc Reagents with
Aryl Halides
%) as palladium source with o-tol3P (10 mol %) allowed
cross-coupling of aryl bromides in moderate yields, provided
the reactions were conducted at 50 °C.4 However, the yields
remain moderate, and moreover the typical palladium load-
ings required [2.5 mol % of Pd2dba3, 5 mol % of Pd(OAc)2]
demonstrate that the number of catalytic turnovers prior to
organozinc reagent decomposition is modest; furthermore,
purification of the protected phenylalanine analogues is
complicated by the presence of catalyst residues.
Andrew J. Ross, Hannah L. Lang, and
Richard F. W. Jackson*
Department of Chemistry, The University of Sheffield,
Dainton Building, Sheffield S3 7HF, United Kingdom
r.f.w.jackson@shef.ac.uk
Received October 19, 2009
In recent work, the relative reactivity in Negishi cross-
coupling reactions of the β-aminoalkyl zinc reagents 7 and
8, used for the synthesis of protected β- and γ-amino acids,
has been studied.5 The conclusion was that the major
factor influencing reactivity is the position of the ester
group: the closer the ester function to the carbon-zinc
bond, the less reactive the reagent. In accordance with
this trend, the pseudo-second order rate constant deter-
mined for the reaction of iodobenzene with reagent 2 [with
o-tol3P and Pd2dba3 (4:1 ratio) as precatalyst] was
approximately five times less than the corresponding rate
constants measured for reagent 7a under the same condi-
tions;5,6 reagent 7a is in turn ten times less reactive than
reagent 8a.5 It is therefore apparent that the reason for the
modest yields of protected phenylalanine analogues 3 and
4 is the low reactivity of the serine-derived reagents 1 and
2, combined with their tendency to decompose by elimina-
tion and protonation. Furthermore, the only way to
increase the yields was either by using more catalyst or,
in cases where the aryl iodide was the more valuable
component, by use of an excess of reagent 1.7 Neither of
these solutions is ideal, and the need for a more active
catalyst is clear.
A combination of Pd2(dba)3 and SPhos (1:2 molar ratio)
is an excellent precatalyst for the Negishi cross-coupling
of the serine-derived organozinc reagent 2 with aryl
halides, including previously difficult ortho-substituted
examples. In the case of meta- and para-substituted aryl
halides, Pd-loadings of 0.5 mol % give satisfactory
results. Use of 2-iodoaniline as substrate gives the lactam
12 in good yield.
The Negishi cross-coupling of the serine-derived organo-
zinc reagents 1 and 2 with aryl halides under palladium
catalysis has proven to be a convenient method for the direct
preparation of protected phenylalanine analogues 3 and 4.1
Under the original conditions reported (Zn/Cu, ultrasonica-
tion, benzene/DMA, (o-tol3P)2PdCl2) yields of the products
3 of cross-coupling with aryl iodides based on iodoalanine 5
were highly variable (10-67%).2 A substantial increase in
yields, especially with ortho-substituted aryl iodides, was
achieved in the cross-coupling of zinc reagent 2 with aryl
iodides, using a combination of o-tol3P and Pd2dba3 (in a 4:1
ratio, thus providing a 2:1 ratio of ligand to Pd) as the
precatalyst, and using neat DMF as the solvent, together
with some modifications to the zinc activation protocol.3
Under these improved conditions, both yields (40-70%,
based on iodoalanine 6) and reproducibility increased. (An
excess of organozinc reagent is often used in Negishi reac-
tions, since the organozinc reagent is generally the less valua-
ble component. In all the work reported here, yields are
based on the amount of protected iodoalanine 6; the yield
therefore depends both on the efficiency of formation of the
organozinc reagent 2 and the efficiency of the subsequent
Evidence has been building that the most reactive type of
Pd catalyst in cross-coupling reactions is in fact monoligated
(4) Oswald, C. L.; Carrillo-Marquez, T.; Caggiano, L.; Jackson, R. F. W.
Tetrahedron 2008, 64, 681–687.
(1) Rilatt, I.; Caggiano, L.; Jackson, R. F. W. Synlett 2005, 2701–2719.
(2) Jackson, R. F. W.; Wishart, N.; Wood, A.; James, K.; Wythes, M. J.
J. Org. Chem. 1992, 57, 3397–3404.
(3) Jackson, R. F. W.; Moore, R. J.; Dexter, C. S.; Elliot, J.; Mowbray,
C. E. J. Org. Chem. 1998, 63, 7875–7884.
(5) Rilatt, I.; Jackson, R. F. W. J. Org. Chem. 2008, 73, 8694–8704.
(6) Rilatt, I. Ph.D. Thesis, The University of Sheffield, 2005.
(7) Dumez, E.; Snaith, J. S.; Jackson, R. F. W.; McElroy, A. B.; Overington,
J.; Wythes, M. J.; Withka, J. M.; McLellan, T. J. J. Org. Chem. 2002, 67, 4882–
4892.
DOI: 10.1021/jo902238n
r
Published on Web 11/25/2009
J. Org. Chem. 2010, 75, 245–248 245
2009 American Chemical Society

Ross et al.
JOCNote
Pd, rather than bisligated Pd.8-10 This has led to the design
and widespread applications of biarylphosphine ligands,11-14
which have been applied to great effect in Suzuki-Miyaura
cross-coupling reactions,15 as well as aryl amination reac-
tions.16,17 The most useful generally applicable ligand is
SPhos 9 and the closely related analogue RuPhos 10, which
each have the capacity to stabilize a monoligated Pd(0)
center by interaction between the ipso-carbon of the lower
ring and Pd, and also to stabilize a Pd(II) center by coordina-
tion of the alkoxy group to Pd. This combination results in
highly reactive catalysts that are also rather stable. In the
initial report,13 ligand to Pd ratios of 1:1 [with Pd(OAc)2 as
the Pd source] were shown to be most effective for Suzuki-
Miyaura reactions with aryl chlorides at room temperature.
In the subsequent full paper, ligand to Pd ratios of 2:1 were
generally used for reactions conducted at elevated tempera-
tures.14 The use of very low loadings of RuPhos and SPhos,
in combination with Pd, for the Negishi reaction to form
biaryls was reported by Buchwald in 2004;18 the active
catalysts were prepared by combining Pd2(dba)3 with 4 equiv
of the biarylphosphine ligand (resulting in a ligand to Pd
ratio of 2:1). Following this report, Bach reported the use of
RuPhos in combination with Pd2(dba)3 for the cross-cou-
pling of a range of functionalized alkylzinc halides to β-
bromo-R,β-unsaturated lactams.19 In a series of papers,
Knochel has shown that SPhos (and RuPhos), in combina-
tion with Pd(OAc)2, is a highly effective catalyst for the
Negishi cross-coupling of a wide range of organozinc re-
agents with a number of substrates, including those contain-
ing acidic protons.20-23 It is interesting that Knochel’s
procedures also employ a ligand to Pd ratio of 2:1, although
the use of a Pd(II) precursor does leave open the possibility
that 1 equiv of ligand might be consumed reducing Pd(II) to
Pd(0), rather than the reduction being effected either by zinc
or the alkylzinc halide. Very recently, Buchwald has intro-
duced a new ligand, CPhos 11, which, in combination with
Pd(OAc)2, is highly effective for the Negishi cross-coupling
of secondary organozinc reagents.24 Once more a ligand to
Pd ratio of 2:1 was employed. We now report our results on
TABLE 1. Influence of Catalysta on the Yield of Protected Phenylala-
nine 4a
entry
Pd precursor
mol %
ligand
mol %
yield (%)b
1
2
3
4
5
6
7
8
9
Pd(OAc)2
Pd2dba3
Pd2dba3
Pd(OAc)2
Pd(OAc)2
Pd2dba3
Pd2dba3
Pd2dba3
Pd2dba3
Pd2dba3
Pd2dba3
5
RuPhos
RuPhos
SPhos
RuPhos
SPhos
RuPhos
SPhos
SPhos
SPhos
SPhos
SPhos
10
10
10
2
2
2
2
1
0.6
0.5
0.2
70
85
80
25
63
24
73
72
59
51
48
2.5
2.5
1
1
0.5
0.5
0.25
0.15
0.125
0.05
10
11
aIn all cases, the Pd:L ratio is 2:1. bYields determined by 1H NMR.
TABLE 2. Influence of L:Pd Ratio on the Yield of 4a, Using Pd2dba3
entry
Pd2dba3(mol. %)
ligand
mol %
L:Pd
yield (%)a
1
2
3
4
5
6
7
8
9
10
0.5
0.5
0.5
0.25
0.25
0.25
0.5
0.5
0.5
SPhos
SPhos
SPhos
SPhos
SPhos
SPhos
RuPhos
RuPhos
RuPhos
2
1
2:1
1:1
0.5:1
2:1
1:1
1:1
2:1
1:1
0.5:1
0:1
73
77
75
0.5
1
0.5
0.5
2
72
80b
73c
24
1
61
47
17
0.5
none
0.5
aYields determined by 1H NMR. bPd2dba3 and SPhos added as
separate solutions. cA solution of Pd2dba3 and SPhos used.
the application of biarylphosphine ligands to the Negishi
cross-coupling of serine-derived organozinc reagent 2 with
aromatic halides.
As a model reaction, we selected the Negishi cross-
coupling of organozinc reagent 2 with iodobenzene at room
temperature. Initial studies employed a ligand to Pd ratio of
2:1, and the results from screening combinations of Pd
precursor and biarylphosphine ligand at varying concentra-
tions are included in Table 1. At high Pd loadings (5 mol %)
use of either RuPhos or SPhos provided excellent yields of
the product 4a. It was striking that at reduced Pd loadings
(1 mol %) SPhos was clearly superior (entries 5 and 7
compared with entries 4 and 6). In all cases, use of Pd2dba3
gave improved results to those obtained with Pd(OAc)2.
Even with substantially lower catalyst loadings (down to
0.1 mol % Pd), respectable yields of product 4a were
obtained, although use of less than 0.5 mol % Pd (i.e., less
than 0.25 mol % Pd2dba3, entries 9-11) did result in a signi-
ficant reduction in yield. These very encouraging results
testify to the suitability of SPhos as a ligand for Negishi
coupling reactions of unreactive organozinc reagents.
The next stage of the optimization process focused on
determining the influence of the ligand to Pd ratio. As is
evident from Table 2, the optimum results arise from a ligand
to Pd ratio of 1:1, which appears to be entirely consistent
with the proposals by Buchwald on the identity of the active
catalytic species.12,14 The optimum catalytic system is a 1:1
ratio of SPhos to Pd (entries 5 and 6), at a Pd loading of
0.5 mol %. It is striking that an excess of RuPhos is actually
worse than a substoichiometric amount (entries 7 and 9),
which implies that the formation of L2Pd complexes results
in a smaller concentration of the catalytically active species.
The control experiment (entry 10) established that the
(8) Paul, F.; Patt, J.; Hartwig, J. F. J. Am. Chem. Soc. 1994, 116, 5969–
5970.
(9) Hartwig, J. F.; Paul, F. J. Am. Chem. Soc. 1995, 117, 5373–5374.
(10) Galardon, E.; Ramdeehul, S.; Brown, J. M.; Cowley, A.; Hii, K. K.;
Jutand, A. Angew. Chem., Int. Ed. 2002, 41, 1760–1763.
(11) Strieter, E. R.; Blackmond, D. G.; Buchwald, S. L. J. Am. Chem. Soc.
2003, 125, 13978–13980.
(12) Barder, T. E.; Biscoe, M. R.; Buchwald, S. L. Organometallics 2007,
26, 2183–2192.
(13) Walker, S. D.; Barder, T. E.; Martinelli, J. R.; Buchwald, S. L.
Angew. Chem., Int. Ed. 2004, 43, 1871–1876.
(14) Barder, T. E.; Walker, S. D.; Martinelli, J. R.; Buchwald, S. L. J. Am.
Chem. Soc. 2005, 127, 4685–4696.
(15) Martin, R.; Buchwald, S. L. Acc. Chem. Res. 2008, 41, 1461–1473.
(16) Biscoe, M. R.; Barder, T. E.; Buchwald, S. L. Angew. Chem., Int. Ed.
2007, 46, 7232–7235.
(17) Surry, D. S.; Buchwald, S. L. Angew. Chem., Int. Ed. 2008, 47, 6338–
6361.
(18) Milne, J. E.; Buchwald, S. L. J. Am. Chem. Soc. 2004, 126, 13028–
13032.
(19) Albrecht, D.; Bach, T. Synlett 2007, 1557–1560.
(20) Manolikakes, G.; Schade, M. A.; Hernandez, C. M.; Mayr, H.;
Knochel, P. Org. Lett. 2008, 10, 2765–2768.
(21) Manolikakes, G.; Hernandez, C. M.; Schade, M. A.; Metzger, A.;
Knochel, P. J. Org. Chem. 2008, 73, 8422–8436.
(22) Manolikakes, G.; Dong, Z. B.; Mayr, H.; Li, J. S.; Knochel, P.
Chem.;Eur. J. 2009, 15, 1324–1328.
(23) Dong, Z. B.; Manolikakes, G.; Li, J. S.; Knochel, P. Synthesis 2009,
681–686.
(24) Han, C.; Buchwald, S. L. J. Am. Chem. Soc. 2009, 131, 7532–7533.
246 J. Org. Chem. Vol. 75, No. 1, 2010

Ross et al.
JOCNote
SCHEME 1. Improved Synthesis of Phenylalanine Derivatives
4a-t
TABLE 3. Synthesis of Phenylalanine Derivatives 4a-t
ArI
ArBr
method Aa method Bb method Cc
Ar
product
yield (%)d
yield (%)d
yield (%)d
C6H5
4a
4b
4c
4d
4e
4f
4g
4h
4i
80
29
79
84
77
1-naphthyl
2-naphthyl
9-anthracenyl
2-MeC6H4
2-HOC6H4
2-MeOC6H4
2-FC6H4
79
40
67
23
50
8
66
75
66
70
76
66
88
71
53
83
82
76
80
69
79
51
FIGURE 1. Correlation between the chemical shift difference
of the two diastereotopic benzylic methylene protons and the
Hammett σ-parameter of protected phenylalanines 4a and 4n-t.
58
65
21
29
49
70
67
34
53
77
65
77
58
49
38
2-ClC6H4
2-O2NC6H4
3-HOC6H4
3-MeOC6H4
3-O2NC6H4
4-MeC6H4
4-HOC6H4
4-MeOC6H4
4-FC6H4
4j
4k
4l
substrates containing a nitro group, the organozinc reagent
was transferred from excess zinc (in order to avoid reduction
of the nitro group) prior to the cross-coupling, and this will
contribute to the lower yields obtained. Simple modification
of the reaction conditions by conducting the reaction for
3 h at 50 °C (method C), rather than room temperature,
allowed good yields to be obtained in the coupling reaction
with aromatic bromides, a very substantial improvement
on our previous efforts.4 It is striking that the yields of
product derived from ortho-substituted iodo- and bromo-
benzene derivatives are very substantially improved over all
our previous results.3,4 This outcome appears entirely
consistent with the presence of a sterically unencumbered
monoligated-Pd catalytic species, compared with the
presumed Pd(o-tol3P)2 that was employed previously.
The availability of such a variety of identically protected
phenylalanine derivatives 4a-t allowed the recording of
specific rotations of the compounds under identical condi-
tions. The data are included in the Supporting Information.
Furthermore, a systematic study of the dependence of NMR
spectroscopic parameters on the substituent, in particular
the anisochrony of the two diastereotopic benzylic methy-
lene protons,26 was possible. In the series of 4-substituted
phenylalanines 4n-t, including phenylalanine itself, 4a, a
reasonable correlation between the chemical shift difference
of the two diastereotopic benzylic methylene protons and the
Hammett σ-parameter was noted (Figure 1). The correlation
is entirely consistent with an early proposal that the magnetic
anisotropy of a phenyl group strongly influences such non-
equivalence.27
4m
4n
4o
4p
4q
4r
4s
4t
80
51
69
75
4-ClC6H4
4-H2NC6H4
4-O2NC6H4
59
aMethod A: 0.25 mol % of Pd2dba3, 0.5 mol % of SPhos, room
temperature, overnight. bMethod B: 2.5 mol % of Pd2dba3, 5 mol % of
SPhos, room temperature, overnight. cMethod C: 2.5 mol % of Pd2dba3,
5 mol % of SPhos, 50 °C, 3 h. dIsolated yields.
presence of phosphine ligand is helpful but not essential for
cross-coupling.
Having established that the optimum catalyst composition
for the coupling of serine-derived organozinc reagent 2 with
iodobenzene is Pd2dba3 (0.25 mol %) in combination with
SPhos (0.5 mol %), the scope of this catalytic system (method
A) for the preparation of a range of phenylalanine deriva-
tives containing substituents in ortho-, meta-, and para-posi-
tions was explored (Scheme 1, Table 3). While the use of a Pd
loading of 0.5 mol % proved entirely satisfactory for most of
the meta- and para-substituted iodobenzene derivatives in-
vestigated, the yields with all of the ortho-substituted iodo-
benzene derivatives were poor (with the exception of 2-iodo-
aniline, see below). Very substantial increases in yield could
be obtained by increasing the catalyst loading by a factor of
10 (method B); these conditions provide good to excellent
yields in most cases. The excellent yields obtained with
coupling to all three iodophenols (with the yields based on
protected iodoalanine 6) are striking, showing an improve-
ment on our previous results,25 and confirming clearly the
benefit of the highly reactive catalysts derived from SPhos
for Negishi couplings with substrates containing acidic pro-
tons.20,21 The only substrates where moderate yields were
obtained were 1-iodo-3-nitrobenzene and 1-iodo-4-nitro-
benzene, for which use of the original catalytic system
employing o-tol3P as ligand is superior.3 For all the
Use of 2-iodoaniline as a substrate in the cross-coupling
reaction with organozinc reagent 2 led to the formation of
the lactam 12 in excellent yield, rather than the expected
2-aminophenylalanine derivative. The low yield of 2-amino-
phenylalanine derivatives that had previously been obtai-
ned3 suggests that the lactam 12 may also have been formed,
but was not isolated. Lactam 12 has been used as a building
block for the synthesis of glycogen phosphorylase inhibitors,
and has been prepared previously by multistep sequences
(26) Jennings, W. B. Chem. Rev. 1975, 75, 307–322.
(27) Whitesides, G. M.; Holtz, D.; Roberts, J. D. J. Am. Chem. Soc. 1964,
86, 2628–2634.
(25) Jackson, R. F. W.; Rilatt, I.; Murray, P. J. Org. Biomol. Chem. 2004,
2, 110–113.
J. Org. Chem. Vol. 75, No. 1, 2010 247

Ross et al.
JOCNote
SCHEME 2. Preparation of Lactam 12
Experimental Section
Preparation of Organozinc Reagent 2. Zinc dust (190 mg,
3 mmol) was added to a flame-dried, nitrogen-purged side arm
round-bottomed flask. Dry DMF (1 mL) was added via syringe,
followed by a catalytic amount of iodine (40 mg, 0.15 mmol). A
color change of the DMF was observed from colorless to yellow
and back again. Protected iodoalanine 6 (329 mg, 1 mmol) was
added immediately, followed by a catalytic amount of iodine
(40 mg, 0.15 mmol). The solution was stirred at room tempera-
ture; successful zinc insertion is accompanied by a noticeable
exotherm. The solution of organozinc reagent 2 was allowed to
cool to room temperature before use.
General Cross-Coupling: Method A. Solutions of Pd2dba3
(11 mg, 0.0125 mmol) in dry DMF (0.5 mL) and SPhos 9
(11 mg, 0.025 mmol) were prepared in dry flasks under nitrogen.
Aliquots of the Pd2dba3 solution (100 μL, 0.0025 mmol) and
SPhos solution (100 μL, 0.005 mmol) were added via syringe to
the solution of organozinc reagent 2, followed by the aryl iodide
(1.3 mmol). The reaction mixture was stirred at room tempera-
ture overnight under positive pressure of nitrogen. The crude
reaction mixture was applied directly to a silica gel column to
afford the purified cross-coupled product.
General Cross-Coupling: Method B. Pd2dba3 (22 mg, 0.025
mmol), SPhos (21 mg, 0.05 mmol), and the aryl iodide (1.3 mmol)
were added to the solution of organozinc reagent 2 and the mixture
was stirred at room temperature overnight, under a positive pressure
of nitrogen. The crude reaction mixture was applied directly to a
silica gel column to afford the purified cross-coupled product.
General Cross-Coupling: Method C. Pd2dba3 (22 mg, 0.025
mmol), SPhos (21 mg, 0.05 mmol), and the aryl bromide (1.3
mmol) were added to the solution of organozinc reagent 2 and
the mixture was heated at 50 °C for 3 h, under a positive pressure
of nitrogen. The reaction mixture was allowed to cool to room
temperature and applied directly to a silica gel column to afford
the purified cross-coupled product.
involving either a resolution,28 or employing phenylalanine
as a chiral pool starting material.29 The one-step synthesis
reported in Scheme 2 represents a convenient alternative.
The isolation of such a good yield from an ortho-substituted
aryl iodide with only 0.5 mol % Pd was somewhat unex-
pected given the results obtained with the other ortho-
substituted substrates, and suggests a rate enhancement
due to the o-amino group. It is tempting to suggest that an
interaction of the amino group with the coordinatively
unsaturated SPhos Pd complex prior to oxidative addition
3
may be responsible. It is possible that a similar effect is
responsible for the much higher yield obtained when using
only 0.5 mol % Pd in the reaction of organozinc reagent 2
with 2-iodophenol (4f), compared with 2-iodoanisole (4g),
although in this case it is not possible to exclude a steric
argument.
In conclusion, it has been established that the combination
of Pd2dba3 and SPhos in a 1:2 molar ratio (giving a Pd:SPhos
ratio of 1:1) is the most efficient precatalyst so far employed
for the Negishi coupling of serine-derived organozinc re-
agents 2 with aryl halides to give protected phenylalanine
derivatives 4. In the case of meta- and para-substituted
aryl iodides, Pd loadings of 0.5 mol % give satisfactory
results, but higher yields are obtained with aryl bromides
and ortho-substituted aryl iodides provided Pd loadings of 5
mol % (2.5 mol % Pd2dba3) are used; these conditions are
therefore recommended as a starting point for the synthesis
of protected phenylalanine derivatives from organozinc
reagent 2.
Acknowledgment. We thank Dr. Craig S. Harris (Astra-
Zeneca, Reims) for helpful suggestions and discussion, and
the Department of Chemistry, The University of Sheffield,
for the award of an EPSRC DTA studentship (A.R.).
Supporting Information Available: Comparison of NMR
chemical shift for the diastereotopic protons of substituted
phenylalanine derivatives (Table 4), tabulated specific rotations
for all phenylalanine derivatives in CHCl3 (Table 5), characteri-
zation data for all compounds, and 1H and 13C NMR spectra for
all compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
(28) Rosauer, K. G.; Ogawa, A. K.; Willoughby, C. A.; Ellsworth, K. P.;
Geissler, W. M.; Myers, R. W.; Deng, Q.; Chapman, K. T.; Harris, G.;
Moller, D. E. Bioorg. Med. Chem. Lett. 2003, 13, 4385–4388.
(29) Birch, A. M.; Kenny, P. W.; Oikonomakos, N. G.; Otterbein, L.;
Schofield, P.; Whittamore, P. R. O.; Whalley, D. P. Bioorg. Med. Chem. Lett.
2007, 17, 394–399.
248 J. Org. Chem. Vol. 75, No. 1, 2010