Palladacyclic and platinacyclic catalysts for the allylation of aldehydes

Article abstract of DOI:10.1016/j.tetlet.2008.04.157

A range of palladacyclic and platinacyclic catalysts have been tested for activity in the allylation of aldehydes with allyl tributyltin. The bulky, π-acidic palladacycle [{Pd(μ-Cl){κ²-P,C-P(OC₆H₂-2,4-^tBu_2<

Full text of DOI:10.1016/j.tetlet.2008.04.157

Tetrahedron Letters 49 (2008) 4216–4219
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Palladacyclic and platinacyclic catalysts for the allylation of aldehydes
*
Robin B. Bedford , Lukasz T. Pilarski
School of Chemistry, University of Bristol, Cantock’s Close, Bristol BS8 1TS, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
A range of palladacyclic and platinacyclic catalysts have been tested for activity in the allylation of
Received 29 January 2008
Revised 18 April 2008
Accepted 29 April 2008
Available online 2 May 2008
2
t
aldehydes with allyl tributyltin. The bulky, p-acidic palladacycle [{Pd(l-Cl){j -P,C-P(OC
6
H
2
-2,4- Bu
2
)-
t
(OC
6
H
3
-2,4- Bu
2 2
) }}
2
] shows particularly good activity at room temperature with a variety of unsaturated
substrates.
Ó 2008 Elsevier Ltd. All rights reserved.
Catalytic allyl palladium chemistry is a vast field dominated by
the reactions of nucleophiles with allyl complexes, by contrast the
reactions with electrophilic substrates such as aldehydes and
syntheses of bis(phosphite) pincer complexes can be hampered
by the challenging C–H activation of the parent ligand or alterna-
tively the need to introduce a halogen onto the resorcinol
backbone of the ligand to facilitate complex formation via C–X
oxidative addition. By contrast, the synthesis of closely related
orthopalladated triarylphosphite complexes of the type 4 and their
adducts 5 is facile and these species have been shown to be active
catalysts for the conjugate addition reactions of both arylboronic
1
imines (Scheme 1) are less well explored.
Whilst tin reagents are toxic, their ease of handling and the fact
that they typically require only mild reaction conditions make
2
them particularly useful allylating reagents for such reactions.
Recent work by the groups of Szabó and le Floch showed that
palladium ‘ECE’-pincer complexes, such as 1 and 2 are very useful
catalysts for the reactions of allyl tin reagents with aldehydes and
6
acids and arylsiloxanes. We were therefore interested to see if
these systems would act as good catalysts for the allylation of
aldehydes with allyl tin reagents; this indeed turns out to be the
case and the preliminary findings of this study are presented
below.
In the first instance we investigated the allylation of benzalde-
hyde with allyltributyl tin using a range of palladacyclic catalysts
and the results from this study are summarised in Table 1. As
can be seen, the bulky triarylphosphite complex 4a gave good
activity at room temperature, giving a 93% spectroscopic yield of
phenylbut-3-ene-1-ol within 24 h at 5 mol % Pd. When the reac-
tion was repeated under air a significant loss in activity resulted
3
,4
imines.
O
O
PPh
Pd TFA
PPh
2
Ph
2
P
S
Pd Cl
2
Ph P
2
S
1
2
R
R
O
P(OAr)*
Pd
P(OAr)*
2
2
R
O
P(OAr)
R
O
P(OAr)
2
X
Pd Cl
2
Pd Cl
L
(
entry 2). In general the reaction times were not optimised, but
when the reaction was run for a shorter period (8 h) then the alco-
hol was produced in only 63% spectroscopic yield (entry 3). When
the reaction was performed at 50 °C then a 75% product yield was
obtained after 4 h (entry 4). Dichloromethane proved to be the
solvent of choice for the reactions; all other solvents tried (entries
O
R
R
3
4
5
The chiral phosphite-based pincer complexes 3 have been found
to show both good activity and modest to good enantioselectivity
in the allylation of aldehydes and imines.5 Unfortunately, the
5
–9) led to reduced or no reaction.
Reducing the size of the orthometallated ring (entry 11) is delete-
rious to catalyst performance. By contrast, the more electron-rich
phosphinite complex showed only slight reduction in
5
a
E
EH
[
cat]
performance (entry 12). The nitrogen based-palladacycles 6 and 7
performed poorly, but surprisingly showed some activity, in contrast
with the arylation of enones by arylboronic acids where no activity is
M
+
R
H
R
E = O; NR
6
a,7
Scheme 1. Nucleophilic allylation of aldehydes and imines.
observed with these or related complexes.
The platinacyclic ana-
logue of 4a, complex 8, showed reduced activity compared with the
palladium complex as has been noted previously for Lewis acid
catalysis using either metallacycles or ‘PCP’-pincer complexes.6
*
Corresponding author. Tel.: +44 0 117 3317538.
E-mail address: r.bedford@bristol.ac.uk (R. B. Bedford).
a,8
0
040-4039/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2008.04.157

R. B. Bedford, L. T. Pilarski / Tetrahedron Letters 49 (2008) 4216–4219
4217
Yieldb (%)
41
Table 1
Table 1 (continued)
Catalyst screeninga
Entry
Catalyst
Solvent
O
H
OH
tBu
O PPh
2
[cat]
SnBu
19
3
+
Pd PPh
Cl
3
10
tBu
Entry
1
Catalyst
Solvent
Yieldb (%)
93
tBu
t
20
[PdCl {P(OC H -2,4-
t
Bu ) } ], 11
14
O
P(OC6H3-2,4- Bu2)2
2
6
3
2 3 2
a
Dichloromethane
Conditions: PhCHO (0.5 mmol), C
3
H
5
SnBu
3
(0.64 mmol), solvent (2 ml), rt 24 h.
Spectroscopic yield determined by H NMR (CDCl , 300 MHz) spectroscopy
b
1
Pd Cl
3
4a
(1,3,5-trimethoxybenzene internal standard).
2
c
tBu
Under air
8 h reaction time.
50 °C (sealed tube), 4 h.
d
2
3
4
5
6
7
8
9
58c
e
d
63
f
e
4 h
75
Toluene
0
1,4-Dioxane
Diethyl ether
Acetonitrile
Tetrahydrofuran
Dichloromethane
33
48
42
35
The phosphine, phosphite and carbene adducts 9a–c and 10
showed poor to modest activity (entries 16–19), depending on
the nature of the co-ligand; in all cases they performed less well
than the dimeric complex 4a. The importance of the orthopallada-
tion of the phosphite ligand is demonstrated by the low activity
observed with the non-orthometallated complex 11 (entry 20).
Having established that complex 4a shows the best activity, this
catalyst was then used in the reaction of allyltributyltin with a
range of aldehydes and the results are summarised in Table 2.9
Good to excellent activity was observed with all the aryl aldehydes
used. The introduction of methoxy groups on the aromatic ring led
to a lowering of yields (entries 5 and 6) as did the introduction of
ortho-substituents. The catalyst is tolerant of aryl halide, hydro-
f
1
0
28
O
P(OPh)
2
2
1
1
60
85
39
Pd Cl
2
4b
tBu
tBu
O
P Pr
i
1
1
2
3
Pd Cl
5
2
NMe2
1
0
xyl and ferrocenyl functions. The tolerance to bromide is partic-
ularly noteworthy when it is considered that the palladacyclic
catalyst 4a is an excellent pre-catalyst for Suzuki, Stille and Heck
Pd TFA
2
6
7
reactions of aryl bromides at elevated reaction temperatures.11
A
control experiment performed with the electronically activated
substrate 4-chlorobenzaldehyde (entry 3) showed no reaction in
N
1
4
36
12
the absence of catalyst.
Pd TFA
2
In addition to the aryl aldehydes tested, alkyl aldehydes can
be used as substrates (entries 15 and 16) albeit with lower
yields obtained. The reaction with N-benzylidene tosylamine
was attempted, but the result obtained was disappointing
tBu
O
t
2 2
P(OC H -2,4- Bu )
6
3
(
entry 17).
1
1
5
6
Pt Cl
45
25
With regards to the mechanism, when pincer complexes are
2
8
used as catalysts Szabó postulates the formation of an intermediate
g -allyl complex which subsequently reacts with the electrophilic
partner, whilst Le Floch favours a Lewis acidic pathway in which a
cationic intermediate with a coordinated aldehyde reacts with the
allyl tin reagent. By contrast, we find that complex 4a readily
reacts with allyltributyl tin to give the g -allyl complex 12
tBu
tBu
1
t
3
O
P(OC
6
H
3
-2,4- Bu
2
)
2
4
Pd PCy3
Cl
3
9a
tBu
tBu
(Scheme 2) and suspect that this is a probable catalytic intermedi-
1
3
31
ate. The P NMR spectrum of the reaction mixture shows the
presence of a new peak at 151.9 ppm, this low field chemical shift
is consistent with an orthopalladated triarylphosphite complex.
The H NMR spectrum of 12 is consistent with an g -allyl com-
plex. Subsequent addition of benzaldehyde shows a reduction
in the amount of 12 and the concomitant reformation of 4a
(determined by P NMR spectroscopy). This is presumably due
to the reaction of the putative intermediate 13 with the ClSnBu
produced during the formation of 12 (Scheme 2).
t
O
O
P(OC
6
H
3
-2,4- Bu
2
)
2
1
4
1
7
17
Pd P(OEt)
Cl
3
1
3
9b
15
tBu
tBu
3
1
t
P(OC
6
H
3
-2,4- Bu
2
)
2
3
Pd Cl
N
The reaction sequence outlined in Scheme 2 is consistent with
1
8
32
3
tBu
Ar
Ar
the formation of an g -allyl intermediate being the first step in
the catalytic cycle, followed by reaction with the aldehyde. It is
not possible at this stage to comment on whether this second step
N
9c
i
6 3 2
Ar = C H -2,6- Pr
3
proceeds via a five-coordinate transition state with an g -allyl

4
218
R. B. Bedford, L. T. Pilarski / Tetrahedron Letters 49 (2008) 4216–4219
Table 2 (continued)
Table 2
Yieldb (%)
62 (51)d
a
Entry
Product
Allylation of aldehydes and imines
E
EH
OH
OH
4a
SnBu
3
E = O; NR'
Yieldb (%)
tBu
+
R
H
R
1
3
4
Entry
Product
OH
tBu
1
93
OH
1
Fe
34
OH
2
3
95
0
c
OH
Cl
1
1
5
6
59
33
OH
4
5
95
73
OH
OH
NHTosyl
1
2
7
27
MeO
OH
a
Conditions: aldehyde (0.5 mmol), C
4 h.
3
H
5
SnBu
3
(0.64 mmol), solvent (2 ml), rt
MeO
MeO
6
50
b
Spectroscopic yield determined by 1H NMR (CDCl
3
, 300 MHz) spectroscopy
(
1,3,5-trimethoxybenzene internal standard).
c
No catalyst.
Isolated yield.
OMe
OH
d
tBu
tBu
O
P(OAr)
Pd
SnBu
3
2
7
8
85
88
2 2
CD Cl
4a
-
ClSnBu
3
OH
12
PhCHO
4
a
tBu
tBu
OH
O
P(OAr)
Pd
2
Ph
ClSnBu
3
9
77
68
+
O
OSnBu
3
2
Ph
OH
1
3, not observed
10
11
12
Scheme 2.
Cl
1
OH
ligand, or a four-coordinate species with an g -allyl ligand. Further
studies are in progress to resolve this issue.
71 (63)d
16
In summary, the simple, commercially available palladacyclic
complex 4a is an active catalyst for the allylation of aldehydes with
Br
allyltributyl tin under mild conditions.
OH
Acknowledgements
31
We thank the EPSRC (Advanced Research Fellowship for RBB,
DTA studentship for LTP) for funding, Johnson Matthey for funding

R. B. Bedford, L. T. Pilarski / Tetrahedron Letters 49 (2008) 4216–4219
4219
10. NMR data for the product (Table 2, entry 13): 1H NMR (300 MHz, CDCl
and the loan of precious metal salts and Dr. Chris Barnard (Johnson
Matthey) for useful discussions.
3
): d 1.28
t
t
(
s, 9H, Bu), 1.42 (s, 9H, Bu), 2.52–2.82 (m, 2H, CH
2
), 4.84 (ddd, 1H, J = 9.22,
4.46, 2.49, ArCH(OH)CH
2
), 5.20–5.30 (m, 2H, CH@CH ), 5.80–5.96 (m, 1H,
2
CH@CH ), 6.83 (d, 1H, J = 2.49, OH), 7.23 (s, 1H, ArH), 7.24 (s, 1H, ArH), 8.19 (s,
2
13
References and notes
1H, OH). C NMR (75.5 MHz, CDCl
22.0, 123.6, 125.5, 134.5, 137.0, 141.3, 152.6.
1. (a) Albisson, D. A.; Bedford, R. B.; Lawrence, S. E.; Scully, P. N. Chem. Commun.
3
): d 29.8, 31.7, 34.3, 35.2, 41.8, 76.0, 119.4,
1
1
1
1
2
.
.
For a recent discussion see: Szabó, K. J. Chem. Eur. J. 2004, 10, 5268.
1
1
998, 2095; (b) Albisson, D. A.; Bedford, R. B.; Scully, P. N. Tetrahedron Lett.
998, 39, 9793; (c) Bedford, R. B.; Hazelwood, S. L.; Limmert, M. E.; Albisson, D.
For recent examples: (a) Zhang, T.; Shi, M.; Zhao, M. Tetrahedron 2008.
doi:10.1016/j.tet.2008.01.017; (b) Barczak, N. T.; Grote, R. E.; Jarvo, E. R.
Organometallics 2007, 26, 4863; (c) Yao, Q.; Sheets, M. J. Org. Chem. 2006, 71,
A.; Draper, S. M.; Scully, P. N.; Coles, S. J.; Hursthouse, M. B. Chem. Eur. J. 2003,
, 3216.
9
5
384.
2. Le Floch and co-workers showed low levels of background reaction in the
absence of catalyst with p-nitrobenzaldehyde as substrate at room
temperature, see Ref. 4.
3
.
(a) Wallner, O. A.; Szabó, K. J. Chem. Eur. J. 2006, 12, 6976; (b) Solin, N.;
Kjellgren, J.; Szabó, K. J. J. Am. Chem. Soc. 2004, 126, 7026; (c) Wallner, O. A.;
Szabó, K. J. Org. Lett. 2004, 6, 1829.
1
3. Complex 12 can also be produced by the reaction of 4a with allyl magnesium
bromide.
4
.
.
Piechaczyk, O.; Cantat, T.; Mézailles, N.; Le Floch, P. J. Org. Chem. 2007, 72,
4
228.
(a) Aydin, J.; Kumar, K. S.; Sayah, M. J.; Wallner, O. A.; Szabó, K. J. J. Org. Chem.
007, 72, 4689; (b) Baber, R. A.; Bedford, R. B.; Betham, M.; Blake, M. E.; Coles, S.
1
4. Samples of 12 are moisture sensitive as demonstrated by 31P NMR
spectroscopy which shows small amounts of two isomers of a hydrolysis
product at d 120.3 and 119.8 ppm, minor and major isomers, respectively.
Addition of water to complex 12 leads to rapid formation of this product. The
similarity of the chemical shifts compared with 4a suggests that this may be
5
2
J.; Haddow, M. F.; Hursthouse, M. B.; Orpen, A. G.; Pilarski, L. T.; Pringle, P. G.;
Wingad, R. L. Chem. Commun. 2006, 3880; (c) Wallner, O. A.; Olsson, V. J.;
Eriksson, L.; Szabó, K. J. Inorg. Chim. Acta 2006, 359, 1767.
the hydroxyl bridged analogue.
6
.
(a) Bedford, R. B.; Betham, M.; Charmant, J. P. H.; Haddow, M. F.; Orpen, A. G.;
Pilarski, L. T.; Coles, S. J.; Hursthouse, M. B. Organometallics 2007, 26, 6346; (b)
He, P.; Lu, Y.; Hu, Q.-S. Tetrahedron Lett. 2007, 48, 5283.
1
15.
H NMR (300 MHz, CD Cl ): d 7.70 (dd, J = 2.4 and 1.5 Hz, 1H, H6
2 2
4
5
orthometallated ring), 7.37 (dd,
metallated ring), 7.35 (dd,
J
HH = 2.6 Hz,
J
PH = 1 Hz, 2H, H3 non-
4
5
0
J
HH = 2.6 Hz,
J
PH = 1 Hz, 2H, H3 non-metallated
7
8
9
.
.
.
He, P.; Lu, Y.; Dong, C.-G.; Hu, Q.-S. Org. Lett. 2007, 9, 343.
Longmire, J. M.; Zhang, X.; Shang, M. Organometallics 1998, 17, 4374.
General method for catalysis: A solution of aldehyde or imine (0.5 mmol),
3
4
ring), 7.23 (dd,
JHH = 8.4 Hz, JPH = 1.7 Hz, 2H, H6 non-metallated ring), 7.20
dd, 3
J
HH = 8.5 Hz, 4JPH = 2.1 Hz, 2H, H6 non-metallated ring), 7.08 (partially
0
(
3
HH = 8.7 Hz, 4
obscured, 1H, H4 orthometallated ring), 7.06 (dd,
2
J
J
HH = 2.7 Hz,
C
3
H
5
SnBu
stirred at room temperature for 24 h. The reaction mixture was quenched with
water (10 ml), the product extracted with CH Cl
(3 ꢀ 10 ml), the combined
organic phase dried (MgSO ) and the solvent removed under reduced pressure.
,3,5-Trimethoxybenzene (internal standard, 28 mg) was added and the yield
3 2 2
(0.20 ml, 0.64 mmol) and catalyst (5 mol %), in CH Cl (2 ml) was
H, H5 non-metallated ring), 7.02 (dd, 3
J
HH = 8.5 Hz, 4
J
HH = 2.5 Hz, 2H, H5 non-
0
metallated ring), 5.27 (m, 1H, allyl central H), 3.71 (m, 2H allyl syn Hs), 3.15
2
2
apparent t, J = 14.4 Hz, 1H, allyl anti H), 2.19 (dd, 3
H, allyl anti H), 1.41, 1.39, 1.31, 1.25, 1.24, 1.20 (s, 9H, Bu).
16. Sigma–Aldrich and Johnson Matthey Catalysts.
2
(
1
JHH = 14.1 Hz, JPH = 2.9 Hz,
t
4
1
1
3
was determined by H NMR spectroscopy (CDCl , 300 MHz).