Allylation of aldehyde and imine substrates with in situ generated allylboronates - A simple route to enatioenriched homoallyl alcohols

Article abstract of DOI:10.1002/ejoc.200500069

Allylation of aldehyde and imine substrates was achieved using easily available allylacetates and diboronate reagents in the presence of catalytic amounts of palladium. This operationally simple one-pot reaction has a broad synthetic scope, as many functionalities including, acetate, carbethoxy, amido and nitro groups are tolerated. The allylation reactions proceed with excellent regio- and stereoselectivity affording the branched allylic isomer. By employment of commercially available chiral diboronates enantioenriched homoallyl alcohols (up to 53 % ee) could be obtained. The mechanistic studies revealed that the in situ generated allylboronates react directly with the aldehyde substrates, however the allylation of the sulfonylimine substrate requires palladium catalysis. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2005.

Full text of DOI:10.1002/ejoc.200500069

FULL PAPER
Allylation of Aldehyde and Imine Substrates with In Situ Generated
Allylboronates – A Simple Route to Enatioenriched Homoallyl Alcohols
Sara Sebelius[a] and Kálmán J. Szabó*[a]
Keywords: Allylic compounds / Boron / Diboronates / Electrophilic addition / Homogenous catalysis / Palladium
Allylation of aldehyde and imine substrates was achieved
using easily available allylacetates and diboronate reagents
in the presence of catalytic amounts of palladium. This oper-
ationally simple one-pot reaction has a broad synthetic
scope, as many functionalities including, acetate, carbethoxy,
amido and nitro groups are tolerated. The allylation reactions
proceed with excellent regio- and stereoselectivity affording
the branched allylic isomer. By employment of commercially
available chiral diboronates enantioenriched homoallyl
alcohols (up to 53% ee) could be obtained. The mechanistic
studies revealed that the in situ generated allylboronates re-
act directly with the aldehyde substrates, however the al-
lylation of the sulfonylimine substrate requires palladium ca-
talysis.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2005)
regio- and stereodefined products.[1–5] Furthermore, tartrate
based chiral allylboronates react with aldehydes with good
Introduction
The reactions of allylboronates with aldehyde and other to high level of enantioselectivity.[4,6,7] Although, the parent
electrophiles offer an attractive approach for synthesis of allylboronate and its alkyl-substituted analogs are easily
Scheme 1. One-pot allylation of aldehyde 3 and imine 4 substrates through palladium-catalyzed formation of transient allylboronates.
Scheme 2. Various diboronates employed in this study.
available reagents, the limited access to properly function-
[a] Stockholm University, Arrhenius Laboratory, Department of
alized and/or chiral allylboronates often limit the synthetic
Organic Chemistry,
scope of the allylation reactions. Recently, we have commu-
nicated[8] an efficient one-pot allylation procedure based on
in situ generation of allylboronates (Scheme 1). In this pro-
cedure the transient allylboronates were prepared by a pal-
10691, Stockholm, Sweden
Fax: +46-8-15-49-08
E-mail: kalman@organ.su.se
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
Eur. J. Org. Chem. 2005, 2539–2547
DOI: 10.1002/ejoc.200500069
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2539

S. Sebelius, K. J. Szabó
FULL PAPER
ladium-catalyzed process from bis(pinacolato)diboron (1a) Table 1. (continued).
(Scheme 2) and easily available functionalized allylacetates
2a–g. Using this procedure the electrophiles nitrobenzalde-
hyde 3a and sulfonylimine 4 were allylated with a high re-
gio- and stereoselectivity affording the homoallyl alcohols
5a–g and amines 6a–d as products. In this paper we give a
full account of our results on these types of reactions. In
addition, we present our new results on some important
synthetic and mechanistic aspects of this reaction: (i) Pos-
sibilities to extend the synthetic scope of the reaction to
nonactivated aldehydes 3b and 3c; and (ii) application of
the chiral diboronate reagents 1b–f to prepare enantiomer-
ically enriched homoallyl alcohols.
Results and Discussion
Allylation of Aldehyde and Imine Electrophiles
As we communicated before,[8] the reaction conditions of
the palladium-catalyzed boronation of the allylacetates 2
are fully compatible with the electrophilic substitution of
Table 1. Allylation of aldehyde and imine electrophiles by catalyti-
cally generated allylboronates.
[a] The reactions with were conducted in DMSO (for 1a) or
DMSO/toluene, 1:1 (for 1b–f) using 6 mol-% Pd catalyst at given
the temperature/reaction time [hours]/[°C]. [b] Diastereomer ratio
(dr) (anti/syn); ao = anti isomer only; so = syn isomer only. [c]
Enantiomeric excess. The major enantiomer is depicted in the pro-
duct column. [d] Isolated yield. [e] In DMSO solvent.
2540
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 2539–2547

Allylation of Aldehyde and Imine Substrates with Allylboronates
FULL PAPER
the transient allylboronates with aldehyde and imine elec- further studies. According to Roush and co-workers[4] the
trophiles. In a typical reaction the diboronate 1, the allyl- highest ee with chiral allylboronates can be achieved in tol-
acetate 2, the appropriate electrophile (3 or 4) and catalytic uene solvent. Therefore, we conducted the asymmetric al-
amounts of Pd2(dba)3 [dba = (dibenzylidene)acetone] were lylation reactions in the presence of toluene co-solvent (in
mixed in DMSO and after the allotted reaction time toluene/DMSO, 1:1). In fact, conducting the reactions in
(Table 1) the corresponding product (5 or 6) was isolated. toluene/DMSO mixture gives higher ee and isolated yield
Using 1a as diboronate reagent the typical reaction tem- of the product than the same process in neat DMSO (c.f.
perature was room temperature or 40 °C, however the crotyl Entries 12 and 13).
substrate (Entry 2) required a somewhat higher reaction
We have found that employment of chiral tartrate deriva-
temperature (60 °C). The broad synthetic scope is a particu- tive 1b in place of 1a leads to a higher reactivity, and there-
larly important feature of this reaction. Many functionali- fore to a faster reaction. Thus the allylation reactions could
ties, such as NO2, COOEt, CONH2 and OAc (Entries 4– also be performed under mild reaction conditions using
7), are tolerated under the applied reaction conditions. The benzaldehyde (3b) and aliphatic aldehyde 3c. The reaction
reactions involving substituted allylacetates provide the with benzaldehyde (3b) gave predominantly (R)-5h with
branched products with very high regioselectivity. The dia- lower enantioselectivity (Entry 18, 45% ee) than the corre-
stereoselectivity of the reaction is also very high. In many sponding reaction with nitrobenzaldehyde 3a (Entry 12,
cases a single diastereomer was obtained (Entries 2, 6, 7 53% ee). The reaction with cyclohexanecarbaldehyde (3c)
and 9), while in the rest of the reactions one of the dia- (Entry 25) requires an extended reaction time to give (R)-
stereomers dominated of the products. It was found that the 5l (48% ee). It is interesting to note that Roush and co-
aldehyde electrophile 3a reacted with anti diastereoselectiv- workers[4] obtained the same enantiopreference with chiral
ity (Entries 2–7), while the imine substrate 4 gave exclusively allylboronate (S,S)-7 and 3c (Scheme 3).
or predominantly syn product (Entries 9–11).
We have found that the best solvent for the reaction was
DMSO. In other solvents, such as THF, acetonitrile, ben-
zene or toluene, black palladium(0) was precipitated imme-
diately after addition of the allylacetate component. On the
other hand, we were not able to stabilize the palladium(0)
catalyst in these solvents by addition of phosphanes [e.g.
PPh3, P(OPh)3] or activated alkenes (e.g. maleic anhydride
and COD), as these additives strongly inhibited the cata-
lytic process. Other strongly coordinating ligands, such as
Scheme 3. Reaction of isolated chiral allylboronate (S,S)-7 with 3c
reported by Roush and co-workers.[4]
We have also studied the reactions of functionalized allyl-
halide salts also retarded or completely inhibited the cata- acetates 2b–f with chiral diboronate 1b and aldehydes 3a–c.
lytic formation of the transient allylboranes. Therefore, em- The high regio- and stereoselectivity of the allylation reac-
ployment of Pd2(dba)3 as catalyst source and DMSO as sol- tions is maintained for each allylacetates, and accordingly
vent or co-solvent (vide infra) seems to be indispensable for the allylation reactions provided the corresponding
the catalytic generation of allylboronates from diboronate branched product with anti diastereoselectivity (Entries 20–
precursors.
24 and 26). The increased reactivity provided by the tartrate
derivative 1b allowed lowering of the reaction temperature
to room temp. using crotylacetate as substrate (c.f Entries
2 and 20). Allylation with crotyl and cinnamylacetate (2b
and 2c, Entries 20–22) proceeds with lower enantio-
Application of Chiral Diboronates 1b–f
The high regio- and diastereoselectivity of the reaction selectivity (33–43% ee) than the corresponding process with
inspired us to conduct the reaction (Scheme 1) in the pres- the parent compound 2a (53% ee, Entry 12). Increase of
ence of the commercially available chiral diboronates 1b–f the polarity of the allylic substituent seems to improve the
in place of 1a. We have found that tartrate-based dibo- enantioselection. As one goes from crotylacetate (2b)
ronates 1b–d reacted with the allylacetate 2a and nitrobenz- through cinnamylacetate (2c) to the diacetate 2f the ee in-
aldehyde (3a) to give homoallyl alcohol 5a up to 53% ee creases from 34% to 50% (c.f. Entries 20–24). Thus, using
(Table 1). Using d-tartrate derivatives (1b or 1d, Entries 12, diacetate 2f and 3a the reaction provided allylacetate (S)-5f
13 and 15) with 3a the major product was the R-enantiomer with excellent regio- and stereochemistry and with 50% ee
[(R)-5a]. As expected, the enantioselectivity of the reaction (Entry 23). It was also found that cyclohexanecarbaldehyde
is reversed using l-tartrate derivative 1c with 3a providing (3c) is allylated with a higher enantioselectivity than benzal-
(S)-5a with 48% ee (Entry 14). Tartaramide derivative 1e dehyde (3a) (c.f. Entries 22 and 26).
and pinane derivative 1f are ineffective in chiral induction
The enantioselectivity of the allylation of the aldehydes
giving a nearly racemic product (Entries 16 and 17). Al- 3a–c is apparently lower with in situ generated allylbo-
though ethyl ester 1b is expected[4] to give less stable allylbo- ronates than with isolated ones. Roush and co-workers[4,9]
ronates than the isopropyl ester 1d, we obtained somewhat have shown that allylation of 3b and 3c with (S,S)-7 (or its
higher ee with 1b than 1d; and moreover, 1b is less expensive enantiomer) can be achieved with 71% and 86% ee
than 1d. Therefore, we employed chiral diboronate 1b in the (Scheme 3) at –78 °C in toluene. An obvious drawback of
Eur. J. Org. Chem. 2005, 2539–2547
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2541

S. Sebelius, K. J. Szabó
FULL PAPER
the presented reactions is that they are conducted at room sion of 2a was observed. It should also be noted that forma-
temp. in presence of DMSO cosolvent. Probably the rela- tion of (S,S)-7 in the previous reaction (Scheme 5) could be
1
tively high reaction temperature (required for the catalytic observed by H NMR spectroscopy, however isolation of
generation of allylboronates) is responsible for the lowering this allylboronate was encumbered by its low stability.
of the enantioselectivity of the process. Conducting the al-
lylation of 3b with isolated chiral allylboronate at room
temp. (23 °C/CH2Cl2) the enantiomeric excess drops to
30%,[9] which is in the same range as the analog reaction
with in situ generated allylboronate (Entry 18) giving 45%
ee.
As it appears above, aldehydes 3a–c give about 35–53%
ee with 1b and various allylacetates (2a–c and 2f) in the
presence of catalytic amounts of palladium. However, the
reaction of the sulfonylimine 4 with 2a and 1b under the
same catalytic condition gave racemic product (Entry 27).
The lack of the enantioselectivity in this process indicates
different mechanistic features for allylation of aldehyde and
imine electrophiles.
Figure 1. Palladium-catalyzed reaction of 2a with 1a in the pres-
ence (closed circles) and in the absence (closed square) of electro-
phile 3a. The reactions were performed in [D6]DMSO using iden-
tical amounts of Pd2(dba)3 catalyst. The progress of the reaction
Mechanistic Considerations
1
was followed by H NMR spectroscopy.
Palladium-catalyzed coupling of allylacetates with 1a to
give allylboronates was first reported by Miyaura and co-
The product inhibition in the above catalytic processes
workers.[10] These authors also found that in these processes (Figure 1 and Scheme 5) can be explained by formation of
a considerable amount of 1,5-hexadiene derivatives were bis(allyl)palladium complex 10 (Scheme 6) from the allylbo-
also formed (Scheme 4).
ronate product and the mono-allylpalladium intermediate
Formation of the transient allylboronate is also the intro- of the process (9). Similar reactions for formation of bis(al-
ducing step of the one-pot allylation reaction described lyl)palladium complexes from (η3-allyl)palladium com-
above (Scheme 1). However, in the one-pot process we did plexes and allylstannanes are well documented in the litera-
not observe formation of hexadiene products (such as 8), ture.[11–19] Bis(allyl)palladium complexes such as 10b are
since the allylboronate formed in the palladium-catalyzed well known to undergo allyl–allyl coupling reactions to give
process immediately react with the added electrophile. hexadiene products,[11,12,20–22] which also explain the forma-
Interestingly, we have found that in the palladium-catalyzed tion of 8 in the palladium-catalyzed boronation reaction in
reaction of 1a and 2a the consumption of the allylacetate the absence of electrophiles (Scheme 4).
(2a) is faster in the presence of the electrophile 3a than in
On the other hand, bis(allyl)palladium complexes 10 are
the absence of it (Figure 1). The retardation effect induced also prone to react with electrophiles, such as aldehydes and
by the accumulation of the allylboronate product is even imines.[13–19,22–25] In order to study the involvement of the
more pronounced when chiral diboronate 1b is employed. palladium catalyses in the allylation step of the above reac-
When 2a was reacted with 1b in the presence of palladium tion (Scheme 1), we reacted allylboronate 11 with ni-
catalyst, only 30% of 2a was converted in 20 hours at room trobenzaldehyde 3a and imine 4 under the usual reaction
temp. to (S,S)-7 (Scheme 5). On the other hand, conducting conditions but in the absence of palladium catalyst
this reaction in the presence of nitrobenzaldehyde (3a) un- (Scheme 7). The reaction of 3a with 11 gave smoothly 5a,
der the same reaction conditions (Entry 12) a full conver- clearly showing that the allylation of the aldehyde compo-
Scheme 4. Palladium-catalyzed boronation of allylacetates[10] in the absence of electrophiles.
Scheme 5. Conducting this reaction in the absence of electrophile only 30% conversion of allylacetate 2a could be achieved (c.f. Entry
12).
2542
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 2539–2547

Allylation of Aldehyde and Imine Substrates with Allylboronates
FULL PAPER
Scheme 6. Possible explanation of the product inhibition in the palladium-catalyzed formation of allylboronates.
nent does not require palladium catalysis. On the contrary, form (η3-allyl)palladium complex 9b.[26,27] The next step is
reaction of 4 with 11 does not provide the expected product addition of the diboron reagent to the allyl moiety of 9b to
6a unless catalytic amount of palladium is added to the form the transient allylboronate 11. Considering these type
reaction mixture. This latter experiment indicates that the of reactions with analog dimetal reagents,[10,17–19,28–30] such
allylation reaction of imine 4 requires palladium catalysis, as hexaalkyl/aryldisilanes (R3Si–SiR3) and distannanes
and that this reaction proceeds via bis(allyl)palladium com- (R3Sn–SnR3) the boron–carbon bond formation probably
plex 10a.
takes place by a inner-sphere nucleophilic attack. Thus the
Considering the above findings two different catalytic cy- nucleophilic attack is preceded by formation of complex 13,
cles can be envisaged for the allylation of the aldehyde 3 and in which the boronate group is coordinated to palladium.
the imine substrate 4. The introducing step for allylation Coordination of the boronate group to palladium involves
of aldehydes 3a–c (Scheme 8) is oxidative addition of the ligand exchange, which is certainly hindered in the presence
palladium(0) catalyst to the corresponding allylacetate to of strongly coordinating ligands, such as phosphanes, acti-
Scheme 7. Allylation of aldehyde 3a and imine 4 in the presence and absence of palladium catalyst.
Scheme 8. Catalytic cycle for the allylation of aldehydes 3a–c.
www.eurjoc.org
Eur. J. Org. Chem. 2005, 2539–2547
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2543

S. Sebelius, K. J. Szabó
FULL PAPER
Scheme 9. Catalytic cycle for the allylation of sulfonylimine 4.
vated alkens or halogenides. This would explain our find- Development of the Selectivity
ings that the addition of strongly coordinating species (vide
supra) inhibit the catalytic formation of the allylboronates
Probably the most important step of the catalytic process
11. On the other hand, in the absence of phosphane and is the electrophilic attack on the allylmetal moiety, since this
other ligands DMSO is the only solvent, which is able to step determines the selectivity of the allylation process. The
stabilize the palladium(0) catalyst source. The final step in stereo- and enantioselectivity of the electrophilic attack on
the catalytic cycle is the direct attack of the aldehyde elec- the aldehyde 3 and the imine 4 substrates is markedly dif-
trophile on the boronate 11 (Scheme 7) to give 14, which ferent, which can be ascribed to the different reaction
subsequently hydrolyses to the homoallyl alcohol product.
mechanism.
The allylation of imine 4 substrates also starts with palla-
The direct attack on the aldehyde substrate 3 is known
dium-catalyzed formation of the allylboronate 11. In con- to proceed with a high anti diastereoselectivity.[1] The expla-
trast to aldehyde 3, imine 4 is not able to undergo direct nation of this diastereoselectivity is based on the assump-
electrophilic substitution with 11 (Scheme 7), and therefore tion that the reaction takes place via six-membered cyclic
the allylation step requires palladium catalysis. The second TS (Scheme 10), in which substituents R and RЈ occupy
catalytic cycle (Scheme 9) is assumed to start with trans- equatorial positions. This arrangement of the substituents
metallation of 11 to 9b providing the bis(allyl)palladium in the TS leads to formations of the anti product, which
complex 10a. This complex undergoes electrophilic attack explains the anti diastereoselectivity of the allylation reac-
with imine 4 to give compound 15. Similar reactions have tions with aldehyde electrophiles.
been reported for palladium-catalyzed allylation of 4 with
The predominant formation of the R enantiomer from
the transient (S,S)-7 and the corresponding aldehydes 3a–c
allyl stannane substrates.[17–19]
Scheme 10. Assumed TS structures in the selectivity determining step of the allylation of aldehydes.
2544 www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2005, 2539–2547

Allylation of Aldehyde and Imine Substrates with Allylboronates
FULL PAPER
Scheme 11. Evolution of the steroechemistry in the allylation of sulfonylimine 4.
can be explained by the selectivity model presented by amounts of palladium. Using this operationally simple one-
Roush and co-workers.[9,31,32] According to this model pot synthesis the tedious isolation of unstable allylbo-
(Scheme 10) TS structure 16a is stabilized by attractive in- ronates can be avoided. The reactions proceed with an ex-
teractions between the aldehyde carbonyl carbon (δ+) and cellent regio- and stereoselectivity providing the branched
the ester carbonyl oxygen (δ–).[31,32] On the other hand, 16b allylic isomer. The allylation reaction with aldehydes 3 pro-
is destabilized by four-electron interactions induced by the ceeds with anti diastereoselectivity, while the corresponding
close proximity of the lone-pair electrons in the aldehyde reaction with sulfonylimine substrate 4 takes place with syn
carbonyl and in one of the ester carbonyls of the tartrate diastereoselectivity. By employment of chiral diboronates
functionality. Considering these interactions TS 16a is fa- 1b–d enatiomerically enriched homoallyl alcohols (up to
vored, and therefore the allylation reaction gives predomi- 53% ee) can be obtained using aldehyde electrophiles. Al-
nantly to the R enantiomer.
lylation of sulfonylimine 4 with allylacetates in the presence
The reaction of 2a, chiral diboronate 1b and sulfonylim- of chiral diboronates leads to racemic products. The mech-
ine 4 leads to the racemic homoallylamine 6a (Entry 27). anistic studies have shown that under the applied reaction
Formation of the racemic product is the consequence of conditions the aldehydes 3 react directly with the transient
the palladium-catalyzed allylation of imine 4 with transient allylboronates, while the allylation of sulfonylimine 4 re-
allylboronate (S,S)-7. This reaction proceeds via the bis(al- quires palladium catalysis. This latter reaction is assumed
lyl)palladium complex 10a (Scheme 9) with a complete loss to proceed via the bis(allyl)palladium intermediates 10. The
of the chiral information. Another interesting feature of the stereo- and enantioselectivity of the reaction can be ex-
allylation of 4 is that this reaction takes place with syn dia- plained invoking six-membered cyclic TS structures
stereoselectivity, while the analog reaction with aldehydes (Scheme 10 and Scheme 11). Since the presented operation-
3a–c proceeds with anti diastereoselectivity. Our recent ally simple one-pot reaction has a broad synthetic scope,
studies[19] on the electrophilic substitution reactions via bis- high regio- and stereoselectivity and a promising enantio-
(allyl)palladium complexes have shown that the electro- selectivity, it can be employed for selective synthesis of func-
philic attack in these processes also proceed through six- tionalized homoallyl alcohols.
membered cyclic TS. Accordingly, the development of the
stereoselectivity in the reaction of sulfonylimine 4 with allyl-
substituted substrates 2c–e (Entries 9–11) can be discussed
Experimental Section
on the basis of a six-membered TS structure model 17
All reactions were conducted under argon employing standard
(Scheme 11). The geometry of the TS structure 17 is biased manifold techniques. The 1H NMR and 13C NMR spectra were
recorded in CDCl3 (internal standard: 7.26 ppm, 1H NMR;
by the trans geometry of the phenylsulphonyl and phenyl
groups across the carbon–nitrogen double bond in 4. Be-
cause of this trans geometry, the lone-pair on nitrogen (in-
teracting with palladium) and the phenyl group are in a cis
77.36 ppm,13C NMR) or [D6]DMSO (internal standard: 2.54 ppm,
1H NMR; 40.45 ppm, 13C NMR) solutions at room temperature
with Varian 400 spectrometer. Merck silica gel 60 (230–400 mesh)
was used for the chromatography. The enantiomeric excess of the
products was determined by HPLC using chiracel OD-H and OJ
arrangement. Furthermore, it is reasonable to assume that
the steric interactions between the metal atom and the
or chiralpak AD columns with hexane/isopropyl alcohol as eluent.
bulky phenylsulfonyl group will be avoided. As a conse-
General Procedure A for the Allylation Reactions: The correspond-
ing electrophile 3–4 (0.3 mmol) and Pd2(dba)3 (0.009 mmol) were
dissolved in DMSO (3 mL) followed by addition of the allylic sub-
strate 2a–g (0.36 mmol). This reaction mixture was stirred for
quence, the preferred orientation of 4 in TS structure 17
renders the phenyl group to an axial position. Since the
allylic substituent (R) is in an equatorial position, this
model predicts formation of the syn diastereomer
(Scheme 11). It is interesting to note that a similar syn dia-
stereoselectivity was reported by Chan[33] and Lu for forma-
tion of 6b in the indium- and zinc-mediated coupling of 4
with cinnamyl bromide.
10 min at room temperature under Ar. After addition of diboron
reagent 1a (0.36 mmol) the reaction mixture was stirred for the al-
lotted temperatures and times listed in Table 1. Thereafter, the reac-
tion mixture was diluted with water (3 mL) and stirred for one hour
at room temperature. This solution was extracted with diethyl ether
(4×6 mL), and the combined ether phases was dried with MgSO4
and the solvents evaporated. The products 5–6 were isolated by
column chromatography using a pentane/EtOAc eluent.
Conclusions
Aldehyde 3 and imine 4 substrates can be allylated with
General Procedure B for the Allylation Reactions: The correspond-
allylacetates 2 in the presence of diboronates and catalytic ing electrophile 3 (0.1 mmol) and Pd2(dba)3 (0.003 mmol) were dis-
Eur. J. Org. Chem. 2005, 2539–2547
www.eurjoc.org © 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 2545

S. Sebelius, K. J. Szabó
FULL PAPER
solved in DMSO (0.7 mL). Thereafter, the allylic substrate 2a–f
(0.12 mmol) was added. This reaction mixture was stirred for
10 min at room temperature under Ar. After addition of diboron
reagent 1b (0.12 mmol) the reaction mixture was stirred until a
clear yellow solution was obtained. Thereafter, toluene (0.7 mL)
was added and the reaction mixture was stirred for the allotted
temperatures and times listed in Table 1. Subsequently, the reaction
mixture was diluted with water (1 mL) and stirred for one hour at
room temperature, then the solution was extracted with diethyl
ether (4×2 mL), and the combined ether phases was dried with
MgSO4 and the solvents evaporated. The products 5a–m were iso-
lated by column chromatography using a pentane/EtOAc eluent.
(d, J = 17.3 Hz, 1 H, trans), 4.90 (d, J = 3.7 Hz, 1 H), 4.32 (dd, J
= 11.2 Hz, 7.7 Hz, 1 H), 4.02 (dd, J = 11.2 Hz, 6.1 Hz, 1 H), 2.7
(m, 1 H), 2.66 (s, 1 H, OH), 2.08 (s, 3 H) ppm. 13C NMR (CDCl3):
δ = 171.7, 149.8, 147.7, 132.8, 127.4, 123.8, 120.9, 72.2, 64.6, 51.1,
21.2 ppm.
1-Phenylbut-3-en-1-ol [(R)-5h]: This product was obtained by gene-
ral procedure B. The NMR spectroscopic data are in agreement
with the literature values.[35] Optical rotation: [α]2D0 = +24.0 (c =
0.67, benzene) for 45% ee; ref.[35] [α]2D0 = +51.2 (benzene) for 97%
ee.
2-Methyl-1-phenylbut-3-en-1-ol [(R)-5i]: General procedure B was
employed to obtain the enantiomerically enriched product (R)-5i.
The NMR spectrum is identical with that reported in the litera-
ture.[38] Optical rotation: [α]2D0 = +57.0 (c = 0.46, CHCl3) for 33%
ee; ref.[39] [α]2D0 = +92.0 (CHCl3) for 97% ee.
1-(4-Nitrophenyl)but-3-en-1-ol [5a, (R)-5a and (S)-5a]: General pro-
cedure A was employed for synthesis of the racemic product 5a.
The NMR spectroscopic data obtained of this product is in agree-
ment with the literature values.[34,35] The enantiomerically enriched
product (R)-5a was obtained using general procedure B. Optical
1,2-Diphenylbut-3-en-1-ol [(R)-5j]: General procedure B was em-
ployed to obtain this product. The NMR spectrum is identical with
that reported in the literature.[40] Optical rotation: [α]2D0 = +5.5 (c
= 0.74, CHCl3) for 43% ee; ref.[40] given for the enantiomer form,
(1S,2R)-1,2-diphenylbut-3-en-1-ol, [α]2D0 = –12.5 (CHCl3) for 97.4%
ee.
rotation: [α]2D0 = +14.1 (c = 0.92, C6H6) for 53% ee; ref.[35] [α]2D0
=
+21.5 (C6H6) for 81% ee. The same reaction (according to general
procedure B) was performed with l-tartrate 1c to afford the enan-
tiomerically enriched product (S)-5a. Optical rotation: [α]2D0 = –28.0
(c = 0.67, CHCl3) for 49% ee; ref.[36] [α]2D0 = –33.2 (CHCl3) for 65%
ee.
1-[Hydroxy(phenyl)methyl]allyl Acetate [(S)-5k]: This product was
2-Methyl-1-(4-nitrophenyl)but-3-en-1-ol [(R)-5b and (R)-5b]: Gene-
ral procedure A was employed for synthesis of the racemic product.
The NMR spectroscopic data obtained of this product is in agree-
ment with the literature values.[37] General procedure B was em-
ployed to obtain the enantiomerically enriched product (R)-5b. Op-
tical rotation: [α]2D0 = +23.0 (c = 0.54, CHCl3) for 34% ee. The
configuration was assigned on the basis of the optical rotation data
reported for the parent compound (R)-5i (see below).
obtained by general procedure B. The NMR spectrum is identical
with that reported in the literature.[41] Optical rotation: [α]2D0
=
+19.7 (c = 0.23, CHCl3) for 45% ee. Determination of the configu-
ration is based on the optical rotation data published[42] for
(1S,2R)-(+)-1-phenylbut-3-ene-1,2-diol which was obtained by hy-
drolysis of (S)-5k.
1-(Cyclohexyl)but-3-en-1-ol [(R)-5l]: This product was obtained by
general procedure B. The NMR spectrum is identical with that re-
ported in the literature.[43,44] Optical rotation: [α]2D0 = +3.0 (c =
0.68, EtOH) for 48% ee; ref.[44] [α]2D0 = +9.7 (EtOH) for 98% ee.
1-(4-Nitrophenyl)-2-phenylbut-3-en-1-ol (5c): General procedure A
was employed for synthesis of the racemic product. The NMR
spectrum is in agreement with the literature data.[18]
1-Cyclohexyl-2-phenylbut-3-en-1-ol [(S)-5m]: This product was ob-
tained by general procedure B. The NMR spectrum is identical
with that reported in the literature.[45] [α]2D0 = +42.2 (c = 0.36,
CHCl3) for 53% ee. The configuration was assigned on the basis
of the optical rotation data reported for the phenyl analog (R)-5j
(see above).
Ethyl 2-[Hydroxy(4-nitrophenyl)methyl]but-3-enoate (5d): General
procedure A was employed for synthesis of this product. The NMR
spectrum is identical with that reported in the literature.[18]
2-[Hydroxy(4-nitrophenyl)methyl]but-3-enamide (5e): General pro-
cedure A was employed for synthesis of the racemic product. The
NMR spectrum is identical with that reported in the literature.[18]
N-(1-Phenylbut-3-enyl)benzenesulfonamide (6a): Both procedures
(A and B) gave racemic products. The NMR spectrum is identical
with that reported in the literature.[33]
1-[Hydroxy(4-nitrophenyl)methyl]allyl Acetate [5f[8] and (S)-5f]: Ge-
neral procedure A was employed for synthesis of the racemic pro-
duct. The crude product was purified by silica gel chromatography
N-(1,2-Diphenylbut-3-enyl)benzenesulfonamide (6b): General pro-
cedure A was employed for synthesis of this product. The NMR
spectrum is identical with that reported in the literature.[18]
1
using pentane/ethyl acetate (4:1) as eluent. H NMR (CDCl3): δ =
8.22 (d, J = 8.8 Hz, 2 H), 7.56 (d, J = 8.8 Hz, 2 H), 5.76 (ddd, J =
6.6 Hz, 10.4 Hz, 17.2 Hz, 1 H), 5.45 (dd, J = 6.6 Hz, 3.9 Hz, 1 H),
5.31 (d, J = 10.4 Hz, 1 H, cis), 5.26 (d, J = 17.2 Hz, 1 H, trans),
5.01 (t, J = 3.9 Hz, 1 H), 2.54 (d, J = 3.7 Hz, 1 H, OH) 2.09 (s, 3
H) ppm. 13C NMR (CDCl3): δ = 170.3, 146.8, 131.1, 128.8, 127.8,
123.8, 121.0, 78.3, 74.6, 21.4 ppm. General procedure B was em-
ployed to obtain the enantiomerically enriched product (S)-5f. Op-
tical rotation: [α]2D0 = +27.8 (c = 0.39, CHCl3) for 50% ee. The
configuration was assigned on the basis of the optical rotation data
reported for the hydrolysis product of the parent compound (S)-5k
(see below).
Ethyl 2-[Phenylsulfonylamino(phenyl)methyl]but-3-enoate (6c): Ge-
neral procedure A was employed for synthesis of this product. The
NMR spectrum is identical with that reported in the literature.[18]
2-[Phenylsulfonylamino(phenyl)methyl]but-3-enamide (6d):[8] This
product was obtained by general procedure B. The crude product
was purified by silica gel chromatography using dichloromethane/
methanol (20:1) as eluent. 1H NMR ([D6]DMSO): δ = 8.28 (d, J =
9.4 Hz, 1 H, NH), 7.48 (d, J = 7.4 Hz, 2 H), 7.39 (t, J = 7.4 Hz, 1
H), 7.26 (t, J = 7.4 Hz, 2 H), 7.04 (m, 5 H), 5.69 (ddd, J = 10.2 Hz,
10.0 Hz, 17.1 Hz, 1 H, anti), 5.51 (ddd, J = 9.4 Hz, 10.2 Hz,
17.2 Hz, 1 H, syn), 5.22 (d, J = 17.1 Hz, 1 H, trans, anti), 5.08 (d,
J = 10.0 Hz, 1 H, cis, anti), 4.84 (d, J = 17.2 Hz, 1 H, trans, syn),
4.80 (d, J = 10.2 Hz, 1 H, cis, syn), 4.68 (t, J = 9.4 Hz, 1 H, syn),
4.55 (t, J = 10.2 Hz, 1 H, anti), 3.28 (t, J = 10.2 Hz, 1 H, anti),
3.17 (t, J = 9.4 Hz, 1 H, syn) ppm. 13C NMR ([D6]DMSO): δ =
2-[Hydroxy(4-nitrophenyl)methyl]but-3-enyl Acetate (5g):[8] General
procedure A was employed for preparation of the racemic product.
The crude product was purified by silica gel chromatography using
1
pentane/diethyl ether (2:1) as eluent. H NMR (CDCl3): δ = 8.20
(d, J = 8.8 Hz, 2 H), 7.50 (d, J = 8.8 Hz, 2 H), 5.70 (ddd, J =
8.5 Hz, 10.4 Hz, 17.3 Hz, 1 H), 5.19 (d, J = 10.4 Hz, 1 H, cis), 5.04
2546
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 2539–2547

Allylation of Aldehyde and Imine Substrates with Allylboronates
FULL PAPER
[21] M. Méndez, J. M. Cuerva, E. Gómez-Bengoa, D. J. Cárdenas,
A. M. Echavarren, Chem. Eur. J. 2002, 8, 3620–3628.
[22] H. Nakamura, M. Bao, Y. Yamamoto, Angew. Chem. Int. Ed.
2001, 40, 3208–3210.
[23] H. Nakamura, H. Iwama, Y. Yamamoto, Chem. Commun.
1996, 1459–1460.
173.4, 142.4, 139.9, 135.7, 132.4, 129.2, 128.6, 128.2, 127.7, 127.2,
118.8, 60.1, 58.1 ppm.
Supporting Information (see also footnote on the first page of this
article): It contains the 13C NMR spectra of synthetized com-
pounds 5a–m and 6a–d.
[24] H. Nakamura, K. Nakamura, Y. Yamamoto, J. Am. Chem.
Soc. 1998, 120, 4242–4243.
[25] R. A. Fernandes, A. Stimac, Y. Yamamoto, J. Am. Chem. Soc.
2003, 125, 14133–14139.
Acknowledgments
[26] J. Tsuji, Palladium Reagents and Catalysis: Innovations, in: Or-
ganic Synthesis, Wiley, Chichester, 1995.
[27] J. Tsuji, Perspectives, in: Organopalladium Chemistry for the
21st Century, Elsevier, 1999.
This work was supported by the Swedish Science Research Council
(VR).
[28] Y. Tsuji, M. Funato, M. Ozawa, H. Ogiyama, S. Kajita, T. Ka-
wamura, J. Org. Chem. 1996, 61, 5779–5787.
[29] I. Macsári, E. Hupe, K. J. Szabó, J. Org. Chem. 1999, 64, 9547–
9556.
[30] N. A. Bumagin, A. N. Kasatkin, I. P. Beletskaya, Izv. Akad.
Nauk SSSR, Ser. Khim. (Engl. Transl.) 1984, 636–642.
[31] B. W. Gung, X. Xue, W. R. Roush, J. Am. Chem. Soc. 2002,
124, 10692–10697.
[1] R. W. Hoffmann, Angew. Chem. Int. Ed. Engl. 1982, 21, 555–
642.
[2] Y. Yamamoto, N. Asao, Chem. Rev. 1993, 93, 2207–2293.
[3] H. C. Brown, U. S. Racherla, P. J. Pellechia, J. Org. Chem. 1990,
55, 1868–1874.
[4] W. R. Roush, L. K. Hoong, M. A. J. Palmer, J. C. Park, J. Org.
Chem. 1990, 55, 4109–4117.
[5] J. W. J. Kennedy, D. G. Hall, Angew. Chem. Int. Ed. 2003, 42,
[32] W. R. Roush, A. M. Ratz, J. A. Jablonowski, J. Org. Chem.
1992, 57, 2047–2052.
4732–4739.
[6] W. R. Roush, L. K. Hoong, M. A. J. Palmer, J. A. Straub, A. D.
Palkowitz, J. Org. Chem. 1990, 55, 4117–4126.
[7] W. R. Roush, K. Ando, D. B. Powers, A. D. Palkowitz, R. L.
Halterman, J. Am. Chem. Soc. 1990, 112, 6339–6348.
[8] S. Sebelius, O. A. Wallner, K. J. Szabó, Org. Lett. 2003, 5,
3065–3068.
[33] W. Lu, T. H. Chan, J. Org. Chem. 2000, 65, 8589–8594.
[34] T. Akiyama, J. Iwai, M. Sugano, Tetrahedron 1999, 55, 7499–
7508.
[35] S. E. Denmark, T. Wynn, J. Am. Chem. Soc. 2001, 123, 6199–
6200.
[36] A. V. Malkov, M. Orsini, D. Pernazza, K. W. Muir, V. Langer,
P. Meghani, P. Kocovský, Org. Lett. 2002, 4, 1047–1049.
[37] S. Jiang, G. E. Agoston, T. Chen, M.-P. Cabal, E. Turos, Orga-
nometallics 1995, 14, 4697–4709.
[38] T.-S. Jang, G. Keum, S. B. Kang, B. Y. Chung, Y. Kim, Synthe-
sis 2003, 775–779.
[9] W. R. Roush, A. E. Walts, L. K. Hoong, J. Am. Chem. Soc.
1985, 107, 8186–8190.
[10] T. Ishiyama, T.-A. Ahiko, N. Miyaura, Tetrahedron Lett. 1996,
37, 6889–6892.
[11] A. Goliaszewski, J. Schwartz, J. Am. Chem. Soc. 1984, 106,
5028–5031.
[12] A. Goliaszewski, J. Schwartz, Tetrahedron 1985, 41, 5779–5789.
[39] H. Lachance, X. Lu, M. Gravel, D. G. Hall, J. Am. Chem. Soc.
2003, 125, 10160–10161.
[13] H. Nakamura, H. Iwama, Y. Yamamoto, J. Am. Chem. Soc. [40] A. Hafner, R. O. Duthaler, R. Marti, G. Rihs, P. Roethe-Streit,
1996, 118, 6641–6647.
F. Schwarzenbach, J. Am. Chem. Soc. 1992, 114, 2321–2336.
[41] M. Lombardo, S. Morganti, C. Trombini, J. Org. Chem. 2003,
68, 997–1006.
[42] A. G. M. Barrett, J. W. Malecha, J. Chem. Soc., Perkin Trans.
1 1994, 14, 1901–1905.
[14] H. Nakamura, K. Aoyagi, J.-G. Shim, Y. Yamamoto, J. Am.
Chem. Soc. 2001, 123, 372–377.
[15] N. Solin, S. Narayan, K. J. Szabó, J. Org. Chem. 2001, 66,
1686–1693.
[16] N. Solin, S. Narayan, K. J. Szabó, Org. Lett. 2001, 3, 909–912.
[17] O. A. Wallner, K. J. Szabó, Org. Lett. 2002, 4, 1563–1566.
[18] O. A. Wallner, K. J. Szabó, J. Org. Chem. 2003, 68, 2934–2943.
[19] O. A. Wallner, K. J. Szabó, Chem. Eur. J. 2003, 9, 4025–4030.
[20] J. Krause, W. Bonrath, K. R. Pörschke, Organometallics 1992,
11, 1158–1167.
[43] A. Fürstner, D. Voigtländer, Synthesis 2000, 7, 959–969.
[44] K. Iseki, S. Mizuno, Y. Kuroki, Y. Kobayashi, Tetrahedron
Lett. 1998, 39, 2767–2770.
[45] M. Lombardo, S. Morganti, M. Tozzi, C. Trombini, Eur. J.
Org. Chem. 2002, 2823–2830.
Received: January 28, 2005
Eur. J. Org. Chem. 2005, 2539–2547
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2547