Towards continuous flow, highly enantioselective allylic animation: ligand design, optimization and supporting

Article abstract of DOI:10.1002/adsc.200900163

A family of enantiopure diphenylphosphinooxazolines (PHOX) containing in their structures a sterically tunable alkoxymethyl group (-CH₂OR) has been optimized for the palladium-catalyzed asymmetric allylic amination. The optimal catalyst (R=CH₃), depicting very high catalytic activity and broad scope applicability, has been further modified to include an ω-alkynyloxy substituent of variable length for polymer supporting via click chemistry, and has been anchored onto slightly cross-linked azidomethyl poly(styrene). The length of a polymethylene chain connecting the PHOX unit with the 1,2,3-triazole linker has been optimized, and the first polymer-supported PHOX ligands for the highly enantioselective allylic amination have been prepared in this manner. Conditions for catalyst recovery and reuse in microwave-promoted amination reactions have been established, and the system has been finally adapted to continuous flow operation.

Full text of DOI:10.1002/adsc.200900163

FULL PAPERS
DOI: 10.1002/adsc.200900163
Towards Continuous Flow, Highly Enantioselective Allylic
Amination: Ligand Design, Optimization and Supporting
Dana Popa,a Rocꢀo Marcos,a Sonia Sayalero,a Anton Vidal-Ferran,a,b,*
and Miquel A. Pericꢁsa,c,*
a
Institute of Chemical Research of Catalonia (ICIQ), Av. Paꢂsos Catalans 16, 43007 Tarragona, Spain
Fax : (+34)-977-920-222; e-mail: avidal@iciq.es or mapericas@iciq.es
ICREA (Instituciꢃ Catalana de Recerca i Estudis AvanÅats), Pg. Lluꢀs Companys 23, 08010 Barcelona, Spain
Departament de Quꢀmica Orgꢄnica, Universitat de Barcelona, 08028 Barcelona, Spain
b
c
Received: March 13, 2009; Published online: June 30, 2009
This paper is dedicated to Professor Luis Castedo on the occasion of his 70th birthday.
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/adsc.200900163.
Abstract: A family of enantiopure diphenylphosphi- lene chain connecting the PHOX unit with the 1,2,3-
nooxazolines (PHOX) containing in their structures triazole linker has been optimized, and the first poly-
a sterically tunable alkoxymethyl group (-CH2OR) mer-supported PHOX ligands for the highly enantio-
has been optimized for the palladium-catalyzed selective allylic amination have been prepared in this
asymmetric allylic amination. The optimal catalyst manner. Conditions for catalyst recovery and reuse
(R=CH3), depicting very high catalytic activity and in microwave-promoted amination reactions have
broad scope applicability, has been further modified been established, and the system has been finally
to include an w-alkynyloxy substituent of variable adapted to continuous flow operation.
length for polymer supporting via click chemistry,
and has been anchored onto slightly cross-linked azi- Keywords: amination; catalyst recycling; flow pro-
domethyl polyACHTUNGTRENNUNG
Introduction
gen s-donor group.[2] In a pioneering study, Pfaltz and
Helmchen described moderate to high enantioselec-
The palladium-catalyzed asymmetric allylic substitu- tion in the allylic amination of symmetrical p-allyl
tion reaction is a versatile and widely employed meth- substrates with various nitrogen nucleophiles in the
odology for the enantioselective construction of presence of 2-(diphenylphosphinophenyl)ozaxoline li-
carbon-carbon and carbon-heteroatom bonds.[1] The gands (Scheme 1).[3] These initial results, and those
development of new chiral ligands for asymmetric al- obtained with analogous PHOX ligands[4] suggest that
lylations, mainly phosphorus-, nitrogen- or sulfur-con- the asymmetric allylic amination is quite sensitive to
taining bidentate species, has attracted a considerable modifications in the electronic characteristics of the
degree of attention over the last years.[1d] In particu- structural aryl group or in the environment of the ste-
lar, the efficiency of chiral phosphinooxazoline- and reogenic center adjacent to the nitrogen in the oxazo-
other related P,N-palladium complexes in the asym- line moiety (C-4). The modification of the spacer be-
metric allylic amination reaction has been well estab- tween the two coordinating heteroatoms in the ligand
lished.[1,2]
framework has provided a diversity of PHOX[5] and
Phosphinooxazolines (PHOX) belong to a non-C2 oxazoline-based P,N ligands[6] with useful characteris-
symmetrical class of bidentate ligands which present tics in this palladium mediated process. However, sub-
the advantage of producing electronic discrimination strate specificity and low reaction rates are still impor-
between the two terminal allylic carbons in the h3-al- tant limitations to overcome, commonly for unhin-
lylpalladium intermediates due to the combination of dered disubstituted substrates and especially in case
the characteristics of a “soft” phosphorus donor of the monosubstituted ones, which have proved to
group with p-acceptor properties and a “hard” nitro- require more active catalysts.[1] Although several ex-
Adv. Synth. Catal. 2009, 351, 1539 – 1556
ꢅ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1539

Dana Popa et al.
FULL PAPERS
Scheme 1. Enantioselective allylic amination of 1,3-disubsti-
tuted allyl substrates in the presence of a Pfaltz–Helmchen
chiral PHOX ligand.
Figure 1. Polymer-supportable, modular phosphinooxazoline
complexes 2a–d.
amples of solid-supported chiral phosphine ligands,
and bidentate N,N-, P,N- and P,S-palladium complexes
as catalysts for asymmetric allylic substitution reac-
tions can be found in the literature,[1d,7] the use in cat- new strategy for supporting catalysts onto Merrifield-
alysis of immobilized chiral phosphinooxazoline li- type resins through copper(I)-catalyzed Huisgen 1,3-
gands remains completely unexplored.[8]
dipolar cycloaddition (click chemistry)[12–14] and have
The preparation and use of solid-supported catalyt- shown that the resulting resins behave as highly
ic systems able to induce enantioselective transforma- active, enantioselective and diastereoselective, yet re-
tions is an area of increasing relevance in chemistry usable, catalysts.[11f,13]
due to the improved sustainability characteristics of
On the other hand, we have recently described a
this type of process. Thus, the suppression of complex new family of PHOX ligands derived from modular,
work-up operations for catalyst separation and the re- enantiopure b-amino alcohols (Figure 1). The pres-
moval of metal-containing by-products from reaction ence of an additional alkoxymethyl substituent at C-5
mixtures lead to cleaner alternatives for the produc- in the oxazoline ring of 2a–d leads to improved cata-
tion of metal-free, enantioenriched compounds. As an lytic characteristics with respect to analogous PHOX
additional bonus, the recovery and reuse of the cata- ligands also bearing a phenyl substituent in C-4, but
lytic system becomes possible.[9]
lacking the alkoxymethyl substituent at C-5. Thus,
The ultimate goal in this type of catalysis is per- palladium complexes 2a–d have proved to be highly
forming chemical transformations in a continuous efficient chiral mediators for the asymmetric allylic al-
mode in a flow reactor under single-pass conditions, kylation.[15] Besides its primary effect on catalytic ac-
where reaction and catalyst separation are carried out tivity and on enantioselectivity, the alkoxymethyl sus-
simultaneously.[10] In spite of these potential advantag- bstituents at C-5 offer the additional possibility of al-
es, the application of flow-through processes in cata- lowing the heterogenization of the PHOX ligands
lytic enantioselective reactions has received little at- onto insoluble supports. The remarkably good perfor-
tention, probably due to the apparently inherent low mance and robustness of complexes 2a–d make them
activity associated to supported systems and to the attractive for evaluation as catalysts in the more chal-
need of engineering effort for the optimization of the lenging asymmetric allylic amination.[1] Moreover, the
different parameters controlling conversion and enan- results of a benchmark-guided structure optimization
tioselectivity in this kind of process.
suggest that their immobilization onto polymers
In recent years, we have been involved in the should lead to minimal deterioration of catalytic
design and synthesis of functional amino alcohol li- properties, thus opening the way to continuous flow
gands that could be immobilized onto inorganic or operation.
polymeric supports without perturbation of the mo-
Herein we report the development of modular, po-
lecular regions where the catalytic activity resides.[11] lymer-supported chiral phosphinooxazoline ligands
As a fruit of this effort, we have recently succeeded and their use in the palladium-catalyzed enantioselec-
in developing the first polymer-supported ligand for tive amination of allyl acetates under batch and con-
the fast, continuous-flow, highly enantioselective alky- tinuous flow conditions.
lation of aldehydes.[11g] Taking into account the limita-
tions associated to ligand anchoring through nucleo-
philic substitution, we have thoroughly investigated a
1540
asc.wiley-vch.de
ꢅ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 1539 – 1556

Towards Continuous Flow, Highly Enantioselective Allylic Amination
FULL PAPERS
Scheme 2. Preparation of 2-fluorophenyloxazolines 6e–f.
Results and Discussion
ACHTUNGTRENNUNG
Ligand Synthesis
ACHTUNGTRENNUNG
The starting point of this research was the modular induced cyclization (5% KOH/MeOH) through an
phosphinooxazoline p-allylpalladium complexes 2a–d, SN2 mechanism. Due to the limited stability of the
readily available from Sharpless epoxy ethers through mesylates, the cyclization was carried out within a few
well
established
synthetic
transformations hours after their preparation. The activation-cycliza-
(Figure 1).[15] The preparation of oxazolines 6, specifi- tion sequence took place in 73–81% overall yield for
cally designed for polymer-supporting, was planned the two steps. The trans stereochemistry at C-4/C-5 in
through a similar, five-step sequence (Scheme 2). The the oxazoline ring was confirmed by the values of the
presence of alkynyloxy side chains in 6 is instrumental coupling constant (J=7.2–7.3 Hz) between C-4 ad C-
for the supporting of PHOX ligands onto slightly 5 protons, similar to those reported for analogous,
cross-linked polystyrene resins through
a
click trans-4,5-disubstituted oxazolines.[15] (See Supporting
chemistry strategy. Within this strategy, the variable Information for NOE experiments on 6f).
length spacers were designed to allow testing the
effect of the distance between the bulky polymeric
backbone and the catalytic site on the performance of Polymer-Supported Phosphinooxazolines
the catalytic process.
The alkynyl fragments were installed on enantio- Our strategy for the preparation of click-PHOX
pure phenylglycidol (>99% ee) by alkylation with the resins is outlined in Scheme 3. The complementary
corresponding halides in the first step of the se- azido group for the dipolar cycloaddition was incorpo-
quence.[16] The so-prepared epoxy ethers 3e–f were rated onto commercial Merrifield resin (1% DVB,
next submitted to aminolysis in 25% aqueous ammo- f0 =0.81 mmolClgÀ1 resin) by reaction with sodium
nia with thermal activation.[17] The epoxide ring-open- azide.[19] According to elemental analysis data,[11a] the
ing took place stereospecifically and with complete calculated degree of functionalization of the resulting
regioselectivity leading to amino alcohols 4e,f which azido resin was f=0.8 mmolgÀ1. Oxazolines 6e,f were
were not isolated. The crude amino alcohols were di- grafted onto that resin by a Cu-catalyzed Huisgen cy-
rectly N-acylated under standard conditions with 2- cloaddition to afford 2-fluorophenyl oxazoline-func-
fluorobenzoyl chloride in the presence of triethyl- tionalized resins 7e,f. As a general rule, stirring the re-
Adv. Synth. Catal. 2009, 351, 1539 – 1556
ꢅ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1541

Dana Popa et al.
FULL PAPERS
Scheme 3. Synthesis of click-supported PHOX and of their p-allyl Pd-complexes.
action mixture (shaker) in 1:1 DMF/THF at 458C for
16 h led to complete conversion. The progress of the
cycloaddition reaction could be easily followed by IR
spectroscopy, through the disappearance of the azide
band (ca. 2094 cmÀ1). Elemental analysis of the final
resins in conjunction with high-resolution magic angle
spinning (HRMAS) 13C NMR spectroscopy allowed
us to establish that the oxazoline anchors to the resin
in quantitative yield. Treatment of resins 7e,f with po-
tassium diphenylphosphide in THF at 08C provided
the desired click-PHOX resins 8e,f. The 19F NMR
spectra (gel phase) of the resulting resins provided
evidence for total conversion in the nucleophilic dis-
placement of the fluoride group. Finally, formation of
the p-allyl complexes 9 was easily performed by addi-
tion of p-allylpalladium chloride dimer [PdACTHNUTRGNEUNG
(h3-
C3H5)Cl]2 to the resins previously swollen in toluene.
The final resins 9e,f had functionalization degrees f=
0.49–0.53 mmolgÀ1, in agreement with those expected
for quantitative complexation. This was also indicated
by 31P NMR (See Supporting Information for
31P NMR spectra of 8e,f and 9e,f).
Figure 2. Preparation of reference, monomeric PHOX-Pd
complexes 11e–f.
the corresponding p-allylpalladium complexes 11e,f
For comparison purposes, compounds 11e,f by reaction with p-allylpalladium chloride dimer
(Figure 2) were prepared following a similar proce- [PdACTHNUTRGNEUNG
(h3-C3H5)Cl]2 in ethanol in the presence of
dure. The Cu(I)-catalyzed reaction between 6e,f and NH4PF6.[18] Complexes 11e,f were isolated in 80–82%
in situ generated benzyl azide (from benzyl bromide yield by precipitation from the reaction mixture at
and sodium azide), took place uneventfully in 1:1 t- 48C.
BuOH/water under microwave irradiation to afford
compounds 10e,f in 81–82% yield. The nucleophilic
displacement of fluoride from 10 with potassium di- Allylic Aminations with Pd-PHOX Complexes 2
phenylphosphide proceeded smoothly in THF at
À208C, the corresponding PHOX ligands being isolat- The modular p-allylpalladium complexes 2a–d were
ed in essentially pure form in 85–90% yield after fil- first evaluated in the amination reaction of rac-(E)-3-
tration through a short pad of deoxygenated SiO2. acetoxy-1,3-diphenyl-prop-1-ene (S1) with benzyl-
The crude PHOX ligands were directly converted to
amine (both widely used as a model substrates).[3–6]
ACHTUNGTRENNUNG
1542
asc.wiley-vch.de
ꢅ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 1539 – 1556

Towards Continuous Flow, Highly Enantioselective Allylic Amination
FULL PAPERS
Table 2. Asymmetric allylic amination of S1 with different
N-nucleophiles catalyzed by Pd/PHOX complex 2a.[a]
Entry Nucleophile
Prod- Time Yield ee
uct
[h]
[%][b] [%][c]
1
2
3
4
5
6
p-Methoxybenzylamine 13
4
2
4
24
24
48
94
99
99
94
99
65
94
97
99
96
94
92
Scheme 4. Asymmetric allylic amination of S1 with benzyl-
ACHTUNGTRENNUNGamine catalyzed by 2a.
Propargylamine
Diallylamine
Benzhydrylamine
Benzoylhydrazine[d]
Phthalimide[e]
14
15
16
17
18
Since 2a (R=Me) had proven to be the optimal cata-
lyst for the allylic alkylation of S1 with dimethyl mal-
onate,[15] experiments aimed at optimizing reaction
conditions were performed with this catalyst. To our
delight, 98% yield and 95% ee were achieved in 2 h
in the presence of 2.5 mol% of 2a at room tempera-
ture under solvent-free conditions, by using an excess
of benzylamine, N,O-bis(trimethylsilyl)acetamide
(BSA) and potassium acetate as additives (Scheme 4).
It is interesting to note that this result is among the
best ones reported in the literature for this particular
reaction using PHOX-like ligands.[3–4] In addition, the
reaction was remarkably fast and nearly total conver-
[a]
All reactions were run at room temperature with
2.5 mol% catalyst, 3 equiv. of amine nucleophile, 3 equiv.
of BSA and 2.5 mol% KOAc.
Yield of isolated product after purification by flash chro-
matography.
Enantiomeric excesses were measured by chiral HPLC.
6 equiv. of BSA were needed.
Phthalimide potassium salt was used. The reaction was
carried out at 508C.
[b]
[c]
[d]
[e]
sion was achieved after 2 h at room temperature. On potassium salt, which required higher temperatures
the basis of previously reported data, the absolute and longer reaction times (65% yield, entry 6). In any
configuration of 12 formed in the reaction was deter- case, this result is comparable with those observed for
mined to be S.[3]
this nucleophile with other PHOX-based ligands.[4a]
Encouraged by these results we next examined the Under the optimized reaction conditions, p-methoxy-
activity of complexes 2b–d under our optimized ex- benzylamine, propargylamine and diallylamine react-
perimental conditions (Table 1). While complexes 2b– ed in 2–4 h to give the expected products with excel-
d (entries 2–4) provided full conversion after 2 h, the lent enantiomeric excess and in essentially quantita-
corresponding enantioselectivities were significantly tive yield (entries 1–3). Interestingly, neither propar-
lower than those with 2a, ranging from 83 to 88% ee. gylamine nor diallylamine have been previously stud-
Thus, the presence of a non-bulky alkoxymethyl sub- ied as nitrogen nucleophiles in asymmetric allylic
stituent at C-5 in the oxazoline ring is the key to high amination processes, despite the potential application
enantioselectivity in the asymmetric allylic amination of the resulting products 14 and 15 in Pauson–Khand,
reaction (entry 1).
enyne cyclization, or metathesis-type reactions. With
To extend the scope of the asymmetric allylic ami- bulkier nucleophiles such as benzhydrylamine, ben-
nation, the reaction of S1 was evaluated using a wide zoylhydrazine and phthalimide, longer reaction times
range of nitrogen nucleophilic compounds (Table 2). were required for complete conversion. Although the
Excellent results were obtained in terms of yield and reaction with benzoylhydrazine required a large
enantioselectivity except in the case of phthalimide excess of BSA, the corresponding product 17 was iso-
lated in quantitative yield and 94% ee (entry 5). In
any case, enantioselectivities, which range from 94 to
99% ee, are among the highest recorded with this
Table 1. Asymmetric allylic amination of S1 with benzyl-
amine catalyzed by Pd/PHOX complexes 2a–d.[a]
type of nucleophiles.[1]
ACHTUNGTRENNUNG
The configurations of (S)-17[3] and (S)-18[4b,20] were
established by comparison of the signs of their optical
rotations or the elution order in HPLC with those re-
ported in the literature. The absolute configurations
of 13–16 were assigned as S by analogy.[1]
Entry
Catalyst
Product
Yield [%][b]
ee [%][c]
1
2
3
4
2a
2b
2c
2d
12
12
12
12
98
99
99
99
95
84
83
88
[a]
All reactions were run at room temperature for 2 h with
2.5 mol% catalyst, 3 equiv. of benzylamine, 3 equiv. of
BSA and 2.5 mol% KOAc.
Yield of isolated product after purification by flash chro-
matography.
Allylic Aminations with Polymer-Supported Pd-
PHOX Complexes 9
[b]
[c]
The results obtained with complexes 2a–d clearly indi-
cated that a non-bulky substituent at C-5 in the oxa-
Enantiomeric excesses were measured by chiral HPLC.
Adv. Synth. Catal. 2009, 351, 1539 – 1556
ꢅ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1543

Dana Popa et al.
FULL PAPERS
Table 3. Asymmetric allylic amination of S1 with different N-nucleophiles catalyzed by Pd/PHOX complexes 11e,[a] 9e[a] and
9f.[b]
Entry
Catalyst
Nucleophile
Product
Time [h]
Yield [%][c]
ee [%][d]
1
2
3
4
5
6
7
8
11e
11e
11e
11e
11e
9e
9e
9e
9e
9e
9e
9f
9f
9f
9f
9f
Benzylamine
12
13
14
15
16
12
12
13
14
15
16
12
13
14
15
16
24
24
30
48
42
48
2.5
240
72
72
240
2
95
98
97
99
99
99
99
85[h]
91
99
89[j]
99
99
95
98
99
91
91
93
99
89
86
81
85
84
88
82
91
85
93
87
83
p-Methoxybenzylamine[e]
Propargylamine
Diallylamine
Benzhydrylamine[e]
Benzylamine[f]
Benzylamine[g]
p-Methoxybenzylamine[e]
Propargylamine[f,i]
Diallylamine[e]
9
10
11
12
13
14
15
16
Benzhydrylamine[e]
Benzylamine
p-Methoxybenzylamine[k]
Propargylamine
Diallylamine
2
4
12
4
Benzhydrylamine[k]
[a]
Unless otherwise specified, the reactions were run at room temperature with 2.5 mol% catalyst, 3 equiv. of amine nucleo-
phile, 3 equiv. of BSA and 2.5 mol% KOAc.
All reactions were run at 408C (setting temperature) under microwave irradiation with 7 mol% catalyst, 3 equiv. of
amine nucleophile and 3 equiv. of BSA.
Yield of isolated product after purification by flash chromatography.
Enantiomeric excesses were measured by chiral HPLC.
9 mol% of catalyst was used.
[b]
[c]
[d]
[e]
[f]
No KOAc was used
[g]
[h]
[i]
Reaction was run at 408C (setting temperature) under microwave irradiation.
13% of S1 was recovered with 21% ee determined by HPLC.
4 mol% of catalyst was used.
10% of S1 was recovered with 27% ee determined by HPLC.
9 mol% of catalyst and 2.5 mol% KOAc were used.
[j]
[k]
zoline ring was optimal for high enantioselectivity in linker as a part of the alkoxy substituent at C-5 in the
the amination reaction. This observation guided the oxazoline ring.
design of the clickable precursors of the e and f fami-
Reactions with the resin-supported Pd complex 9e
lies, where one or four methylene groups are interca- were run under the same reaction conditions as with
lated between the oxygen atom in the C-5 substituent 11e, but with smooth stirring in an orbital shaker to
and the carbon-carbon triple bond on the same avoid mechanical deterioration of the polymer beads.
moiety.
When the reactions were performed in this manner,
For reference purposes, we first examined the be- the amination products could be obtained in high to
haviour of the monomeric model for click-supported excellent yields, but after long reaction times (en-
PHOX/Pd complexes 11e (containing the short meth- tries 6 and 8–11 in Table 3). Except for the highly re-
ylene spacer) in the allylic amination of S1. To this active benzylamine (entry 6), more than 2.5 mol% of
end, S1 was reacted with several N-nucleophiles at 9e was required in order to complete the reaction in
room temperature in the presence of 11e, BSA and 48 h. Taking into account that no work-up is required
potassium acetate. As can be seen in Table 3 (en- under these conditions, and that the catalyst can be
tries 1–5) the reaction was slower in all cases than separated by simple filtration, we decided to investi-
with 2a, 24–48 h being needed to achieve complete gate the possibility of performing the reaction without
conversion. Nevertheless, the results obtained with the presence of the potassium acetate additive. This
this model catalyst are noteworthy since they are would avoid the presence of insoluble solids in the re-
comparable with those obtained with 2a (see Table 2). action media and would facilitate catalyst recovery.
Accordingly, we anticipated that the enantioselectivity Interestingly, with highly reactive substrates such as
of the amination reaction would remain essentially benzylamine (entry 6) and propargylamine (entry 9)
unaffected by the immobilization of the phosphinoox- the reactions take place to completion in reasonable
azoline-palladium complexes through a 1,2,3-triazole periods of time under these conditions. Microwave ir-
1544
asc.wiley-vch.de
ꢅ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 1539 – 1556

Towards Continuous Flow, Highly Enantioselective Allylic Amination
FULL PAPERS
radiation[15] was also studied as an alternative activa-
tion method. Thus, benzylamine reacted with S1 at
408C (temperature setting) providing the amination
product in quantitative yield in only 2.5 h, with only
slightly lower enantioselectivity than in the reaction
at room temperature (compare entries 6 and 7). In
summary, 9e showed the viability of performing asym-
metric allylic aminations with a click-supported phos-
phinooxazoline palladium complex, but its overall cat-
alytic activity and enantioselectivity still required fur-
ther refinement.
Since the reaction environment with the polymer-
supported catalyst is quite different from that of the
homogeneous reaction, both the catalytic activity and
the enantioselectivity could be affected by the close
vicinity of the bulky polystyrene backbone. To miti-
gate this problem, catalyst 9f containing a longer tet-
ramethylene spacer between the PHOX moiety and
the 1,2,3-triazole linker was synthesized. Gratifyingly
enough, resin 9f exhibited an optimal catalytic perfor-
mance under microwave irradiation, with improved
Scheme 5. Asymmetric allylic amination of allyl acetates S2–
S5 with benzylamine.
enantioselectivity over 9e (entries 12–16 in Table 3). sterÀic hindrance around the metal, as in the case of
Furthermore, microwave-assisted allylic amination PF6 anion.[21]
turned out to be more effective for all the nucleo-
In summary, high activities and good to excellent
philes tested and, in general, complete conversions enantioselectivities (in general, higher than 85% ee)
were achieved at 408C in only 2–4 h. As an additional can be achieved in the amination of S1 with different
bonus, it was in general possible to perform the ami- N-nucleophiles under microwave irradiation using po-
nations catalyzed by 9f without the presence of the lymer-supported PHOX/Pd complex 9f as the cata-
potassium acetate additive; only with the less reactive lyst. Notably, these results are similar to those record-
amines, like p-methoxybenzylamine and benzhydryl- ed with the monomeric analogue 11e (89–99% ee)
ACHTUNGTRENNUNGamine, was the presence of this additive required to and, therefore, comparable with those obtained with
ensure complete conversion in short reaction times. 2a (94–99% ee).
This characteristic is relevant in connection with our
ultimate goal of developing a flow system for continu-
ous catalytic enantioselective amination.
Amination of a Family of Allylic Substrates with
In a further attempt to improve the characteristics Benzylamine
of the supported catalysts 9e,f, we studied the effect
of the counterion on their performance. Displacement Regarding the allylic acetate counterpart, a represen-
of chloride anion by the poorly coordinating hexa- tative set of substrates was reacted with benzylamine
fluorophosphate anion in resins 9e,f was easily per- in the presence of the catalysts 2a and 9f (Scheme 5).
formed in the presence of NH4PF6 (See Supporting For comparison purposes, the model complex 11e was
Information). In particular, these anion-exchanged also tested in the reactions. The results of this study
complexes were tested in the amination of S1 with are collected in Table 4. Products 19–22 are known
benzylamine in the reaction conditions optimized for compounds, and their absolute configuration was con-
9e,f. Somewhat dissappointingly, while the yield re- firmed by comparison of the signs of their optical ro-
mained quantitative, the enantioselectivity was rather tation with those reported in the literature.
low in comparison with that recorded with 9e,f (69–
While the remarkable catalytic activity depicted by
71% ee, see with entries 6 and 12 in Table 3 for the the new family of PHOX ligands is rather independ-
results with 9e,f). As the exo-endo ratio of the two ent on the nature of the allylic substrate, the enantio-
isomeric allylpalladium species formed in the catalytic selectivity of the process is strongly substrate-depen-
cycle determines the stereochemical outcome of the dent, as is generally observed in this reaction. In par-
reaction, a fast equilibrium between the two isomers ticular, the amination of S2 and S3 with 2a (entries 1
is essential for elevated enantioselectivities. The equi- and 4) leads to ee values among the highest recorded
librium rate is affected by the anion either by ion with PHOX ligands in the considered reaction.[4b] Fol-
pairing or by coordination to palladium. Anions that lowing the same trend previously observed for S1,
stick to the Pd(II) complexes (ion pairing) modify the resin 9f provided slightly lower enantiomeric excesses
than 2a (entries 2 and 5). In any case, the results are
Adv. Synth. Catal. 2009, 351, 1539 – 1556
ꢅ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1545

Dana Popa et al.
FULL PAPERS
Table 4. Asymmetric allylic amination of allyl acetates S2–
S5 with benzylamine catalyzed by Pd/PHOX complex 2a,[a]
9f[b] and 11e.[a]
Recovery and Reuse of Polymer-Supported Catalysts
9
As a further step towards our ultimate goal of devel-
oping a continuous amination process, the possibility
of recycling and reusing the polymer-supported
PHOX catalysts 9e,f in the allylic amination of S1
with benzylamine was studied. All the reactions were
carried out under the conditions previously optimized
for each catalyst, without the addition of potassium
acetate. After each run, the reaction mixture was sep-
arated by decantation, and the resin was simply
rinsed with deoxygenated dichloromethane and dried
under Ar before the next use. In the case of 9e, the
reactions were performed at room temperature with
orbital shaking. Although, as already discussed, reac-
tions mediated by this catalyst take a long time for
completion, it is worth mentioning that both catalytic
activity and enantioselectivity remain essentially unaf-
fected after three consecutive runs (entries 1–5 in
Entry Catalyst Substrate Product Time Yield
ee
[h]
[%][c]
[%][d]
1
2
3
4
5
6
7
8
2a
9f
11e
2a
9f
11e
2a
9f
11e
2a
9f
S2
S2
S2
S3
S3
S3
S4
S4
S4
S5
S5
S5
19
19
19
20
20
20
21
21
21
22
22
22
1[e]
6[f]
72
15[g]
4
99
96
98
99
70[h]
85
99
99
82
99
99
96
94
82
86
89
82
83
63
62
43
37
29
39
7[i]
4
4
72
4
3
72
9
10
11
12
11e
[a]
[b]
[c]
All reactions were run at room temperature with
2.5 mol% catalyst, 3 equiv. of benzylamine, 3 equiv. of
BSA and 2.5 mol% KOAc.
Reactions were run at 408C (setting temperature) under Table 5).
microwave irradiation with 5 mol% catalyst, 3 equiv. of
benzylamine, 3 equiv. of BSA and 2.5 mol% KOAc.
Yield of isolated product after purification by flash chro-
matography.
Enantiomeric excesses were measured by chiral HPLC.
6 equiv. of benzylamine and 6 equiv. of BSA were used
KOAc was not added.
Reaction was run at 508C. 5 Equiv, of benzylamine were
required.
28% of S3 was recovered. Kinetic resolution was not ob-
served.
On the other hand, a significant decrease in catalyt-
ic activity was observed in the recycling of 9f under
microwave irradiation with a temperature setting of
408C (entries 9–13). In this case, reaction times had
to be increased up to four times to achieve the con-
version of the preceding cycle (compare entries 9, 11,
and 13). Less satisfactory results were achieved when
the reactions were performed at 558C (entries 6–8).
Under these conditions, the decrease in the activity of
the catalyst was more evident after the first cycle
(compare entries 7 and 10). Since the resin turned
black as the recycling progressed, we interpreted that
thermally-induced precipitation of Pd(0) was taking
place, and that this fact was responsible for the activi-
ty decrease. In an attempt to minimize this deactiva-
[d]
[e]
[f]
[g]
[h]
[i]
Reaction was heated at 458C (setting temperature) in a
microwave reactor. 13% of S3 was recovered. Kinetic
resolution was not observed.
comparable with those obtained in solution with the tion mechanism, a new set of recycling experiments
monomeric analogue 11e as the catalyst (entries 3 and (entries 14–17) was performed at low temperature
6). The bulkiness of the ipso substituent on the ace- (278C) with the following experimental modifications.
tate substrate has a remarkable influence on enantio- First, the resin was pre-swollen with dichloromethane
selectivity. Thus, replacement of a rather bulky aryl in order to improve the flow of the reactants into the
substituent (like in S1 or S2) by a less bulky methyl polymer beads. Second, to avoid temperature peaks
group at the reaction site (as in S3 or S4) is accompa- that could be deleterious for the stability of the palla-
nied by a decrease in enantioselectivity, which is nota- dium complex, the reactions were performed in
ble in case of S4 (entries 7 and 8). Nevertheless, power control mode (the reaction time was fixed at
either 2a or 9f afforded product 21 in higher enantio- 35 min and the microwave power at 1 W). Interesting-
meric excess than related PHOX ligands reported in ly, this importantly mitigated the loss of catalytic ac-
the literature.[4b] Notably, no decrease in enantioselec- tivity, while enantioselectivity reached 87% and re-
tivity is observed for this substrate between 2a and 9f, mained essentially constant over four cycles. Conver-
while 11e is much less effective (entry 9). This pro- sion and enantioselectivity were determined after
vides a clear indication on the importance of the each cycle.
spacer when phosphinooxazolines are supported onto
Taken together, these results clearly show that
polymers via click chemistry. Finally, the cyclic sub- while microwave activation is the key to high catalytic
strate S5 was also explored. Although enantioselectiv- activity when the polymer-supported PHOX catalyst
ities are only moderate in this case (entries 10–12), 9f is used, temperature control during these experi-
these are the first results on the amination of cyclic ments is of paramount importance to avoid catalyst
substrates reported with PHOX ligands.
deactivation.
1546
asc.wiley-vch.de
ꢅ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 1539 – 1556

Towards Continuous Flow, Highly Enantioselective Allylic Amination
FULL PAPERS
Table 5. Asymmetric allylic amination of S1 with benzylamine catalyzed by polymer-supported Pd/PHOX complexes 9e[a]
and 9f.[b,c]
Entry
Cycle
Catalyst
Temp. [8C]
Time [h]
Yield [%][d] (Conv. [%])[e]
ee [%][f]
1
2
3
4
5
6
7
8
1
2
9e
25
25
48
48
57
48
58
2
2
8
2
2
99
(90)
99
(89)
99
99
(20)
86
99
(47)
97
(10)
95
86
nd
86
nd
84
88
nd
82
91
nd
87
nd
86
87
87
86
86
3
25
1
2
9f[b]
9f[b]
55[g]
55[g]
9
1
2
40
40
10
11
12
13
14
15
16
17
8
7
3
40
35
0.58
0.58
0.58
0.58
1
2
3
4
9f[c]
27[h]
27[h]
27[h]
27[h]
(99)
(91)
(82)
(76)
[a]
All reactions were run with 2.5 mol% catalyst, 3 equiv. of benzylamine and 3 equiv. of BSA.
[b]
[c]
Reactions were run under microwave irradiation with 7 mol% catalyst, 3 equiv. of benzylamine and 3 equiv. of BSA.
Reactions were run under microwave irradiation with 10 mol% catalyst, 4 equiv. of benzylamine, 4 equiv. of BSA and
7 equiv. of CH2Cl2.
[d]
[e]
[f]
Yield of isolated product after purification by flash chromatography.
1
Conversion determined by H NMR of an aliquot.
Enantiomeric excesses were measured by chiral HPLC.
4 mol% of catalyst was used.
Reactions were performed in power control mode (1 W). Temperature of the reaction mixture was measured with a
teflon-coated Pt-100 probe.
[g]
[h]
Continuous Flow System for the Single-Pass Allylic
Amination of S1
series of infinitely thin coherent plugs, each of them
characterized by a uniform composition which varies
along the axial direction of the reactor as the reaction
With the optimal conditions for recycling the poly- proceeds. For this model to be applicable, the fluid
mer-supported Pd catalyst in hand, efforts were then traveling through the reactor must be perfectly mixed
directed towards the implementation of 9f in a con- in the radial direction but not in the axial one. In the
tinuous flow system. Thus, with the experience gained flow system under consideration, where the tubular
in the design of a system for the enantioselective eth- reactor is completely filled with the swollen functional
ylation of aromatic aldehydes through a single-pass, resin, isothermal piston flow operates.[23] Under these
continuous flow process,[11g] we planned a similar set- conditions, conversion achieved in a given reactor (of
up suited to the new purposes.
given geometry and containing a given amount of cat-
The simplicity of the system used for the study of alyst) is only determined by the feed composition, re-
the allyllic amination of S1 with benzylamine can be action temperature, and mean residence time of the
appreciated in Figure 3. The continuous flow system reactants in the reactor.
consists of a vertical, 1/4 inch internal diameter teflon
In the continuous flow allylic amination experi-
tube containing the supported catalyst, fitted with two ments, the resin was first swollen by pumping anhy-
adapters which allows to connect it with 1/16 inch in- drous dichloromethane at a 0.12 mLminÀ1 flow rate
ternal diameter tubes to pump in the reagents and to during 30 min. Then, microwave irradiation (1 W) was
collect the products (Figure 4). The microwave-assist- started and a solution of BSA, S1 and benzylamine in
ed flow experiment was performed in an open-vessel the right proportions in dichloromethane, placed in a
manner at 1 W irradiation power. During operation, flask under argon, was pumped through the system
the reagents were pumped in through the bottom end for 180 min at the same flow rate (0.12 mLminÀ1).
of the tube using a piston pump.
Under these conditions, temperature of the reaction
Chemical reactions in continuous, flowing systems mixture (measured at the inlet and outlet of the tubu-
are best described by the plug flow reactor model.[22] lar reactor with a Pt-100 probe) kept constant within
In this model, the tubular reactor is considered as a 18C (22Æ18C). The system was operated in single-
Adv. Synth. Catal. 2009, 351, 1539 – 1556
ꢅ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1547

Dana Popa et al.
FULL PAPERS
Figure 3. Continuous flow system used in the allylic amination of S1 with benzylamine.
Table 6. Continuous flow asymmetric allylic amination of S1
with benzylamine catalyzed by polymer-supported Pd/
PHOX complex 9f under single-pass conditions.[a]
Entry
Time [min]
Conv.[%][b]
ee [%][c]
[d]
1
15
–
–
2
3
4
5
6
7
8
9
30
45
60
75
85
86
80
75
70
66
63
60
55
55
54
81
83
82
83
83
83
85
84
84
84
86
90
105
120
135
150
165
180
10
11
12
Figure 4. Detail of the teflon tube containing the polymer-
supported catalyst in the microwave cavity.
[a]
Reaction was run under microwave irradiation (1 W). A
loading of 240 mg of resin 9f (12 mol%) was used. The
employed dichloromethane solution of reagents was pre-
pared by mixing S1 (9.6 mmol) in CH2Cl2 (4.3 mL) with
benzylamine (48.1 mmol) and BSA (38.5 mmol).
Conversions were determined by H NMR.
Enantiomeric excesses were measured by chiral HPLC.
Only CH2Cl2 was collected.
pass manner, so that the output flow was directly col-
lected. Samples of the output solution were taken
every 15 min for analysis purposes. At the end of the
experiment, dichloromethane was pumped for a fur-
ther 30 min in order to wash the resin and the me-
chanical parts of the reactor. The conversion and ee
of each aliquot are plotted in Table 6 and Figure 5. As
[b]
[c]
[d]
1
shown, while the conversion slowly decays from 85% enantioselectivity achieved (86% ee) is comparable
in the first 30 min (entry 2) to 54% after 3 h with those obtained in batch processes.
(entry 12), the ee increases in a slight but regular
The flow rate used in this experiment involves a
manner from 81% to 86% ee. Gratifyingly enough, residence time of the reagents within the catalyst bed
conversion remains near 55% after 3 h of continuous of only 8.5 min.[24] The output flow corresponding to
flow allylic amination of S1 (entry 12) and the level of the whole operation period (3 h) was concentrated in
1548
asc.wiley-vch.de
ꢅ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 1539 – 1556

Towards Continuous Flow, Highly Enantioselective Allylic Amination
FULL PAPERS
careful optimization of reaction conditions for catalyst
recovery and recycling has ultimately served to fix the
technological aspects allowing the conversion of the
initial, batch amination process into the first, single-
pass, continuous flow process of this class. During this
optimization, an important acceleration of the hetero-
geneously-catalyzed reaction by microwave irradia-
tion that does not involving macroscopic heating of
the reaction mixture has been observed. Work aimed
at improving the thermal stability of the polymer-sup-
ported palladium complex for an extended life of the
catalytic system and at applying highly selective mi-
crowave activation to other processes where polymer-
Figure 5. Results of the continuous flow allylic amination of
S1 with benzylamine.
vacuum, diluted with diethyl ether and washed with supported reactants or catalysts are involved is being
water. After evaporation, flash chromatography of actively pursued in our laboratories and will be re-
the reaction crude furnished the desired product 12 in ported in due course.
68% yield and 83% ee. Using a flow rate of
0.12 mLminÀ1 the system has
a production of
9.18 mmolhÀ1 per gram of resin 9f, allowing the prep- Experimental Section
aration of 1.98 g of pure (+)-(S,E)-N-benzyl-(1,3-di-
General aspects of the experimental section can be found in
the Supporting Information.
phenyl-2-propenyl)amine (12) with high enantioselec-
tivity in a 3 h run.
The role exerted by microwave irradiation on the
acceleration of these reactions is rather intriguing
since, as discussed above, the very low microwave
power used is not enough to provoke any significant
increase in the macroscopic temperature of the flow-
ing reaction mixture. This is in agreement with the
low loss tangent values (tan d) of dichloromethane
and polystyrene,[25a] which should make very ineffi-
cient the conversion of electromagnetic energy into
heat in the employed reaction system. Although the
highly selective microwave heating of the Pd(I) com-
plexes grafted onto the polymer chains could be the
primary origin of the observed rate enhancement,[25b]
the dissipation of the energy harvested by the metal
Typical Procedure for the O-Alkylation of (2S,3S)-3-
Phenylglycidol; Synthesis of Alkynyloxy Epoxides
3e,f
A solution of enantiomerically pure (2S,3S)-3-phenylglycidol
(10 mmol) in anhydrous DMF (10 mL) was added under
argon to a suspension of sodium hydride (60% in mineral
oil, 11.7 mmol) in anhydrous DMF (15 mL) previously
cooled at À208C. The mixture was stirred for 20 min and
then the corresponding propargyl halide (13 mmol) was
added dropwise. The reaction mixture was stirred at À208C
for 3 h, then allowed to reach room temperature and further
stirred for 24 h. The reaction was quenched with MeOH
(10 mL) and brine (20 mL) and extracted with CH2Cl2 (3ꢆ
complexes through the polymer chains where they are 20 mL). The organic phase was dried (MgSO4) and the sol-
vent removed under reduced pressure. The purification of
the products 3e,f was carried by flash chromatography (hex-
anes/EtOAc from 100:0 to 80:20).
imbedded could also play a role in this behaviour.
Thus, an increased mobility of the rather flexible po-
lymer matrix would greatly facilitate the contact be-
tween the polymer-supported catalyst and the liquid
phase containing the reactants by opening all the pos-
sible channels in the polymer structure. In this way,
mass transfer limitations to reaction rate could be ef-
ficiently overcome.
3e: Following the general method from (2S,3S)-3-phenyl-
glycidol (1.4 g, 9.3 mmol), sodium hydride (0.45 g,
11.2 mmol) and propargyl bromide (80% in toluene, 1.3 mL,
12 mmol), compound 3e was obtained as a pale yellow oil;
1
yield: 1.32 g (75%); [a]2D3: À49.0 (c 1.1 in CHCl3); H NMR
(CDCl3): d=7.36–7.25 (m, 5H), 4.29–4.20 (m, 2H), 3.90 (dd,
J=11.5, 3.0 Hz, 1H), 3.81 (d, J=2.3 Hz, 1H), 3.68 (dd, J=
11.5, 5.0 Hz, 1H), 3.24–3.21 (m, 1H), 2.47–2.46 (m, 1H);
13C NMR (CDCl3): d=136.7 (C), 128.5 (2CH), 128.3 (CH),
125.7 (2CH), 79.2 (C), 75.0 (CH), 69.4 (CH2), 60.7 (CH2),
58.6 (CH), 55.9 (CH); IR (film): n=3285, 2854, 1461, 1100,
879, 699 cmÀ1; HR-MS (ESI+): m/z=211.0721, calcd. for
C12H12NaO2 [M+Na]+: 211.0735.
Conclusions
In summary, we have described a family of highly
modular PHOX ligands (2) for the Pd-catalyzed
asymmetric allylic amination, and have used this
result as a guide for the design of an analogue that
could be supported onto polystyrene via click chemis-
try. The corresponding polymer-supported catalysts
(9e,f), efficiently induce allylic amination, especially
3f: Following the general method from (2S,3S)-3-phenyl-
glycidol (1.5 g, 10 mmol), sodium hydride (0.47 g,
11.7 mmol) and 6-chloro-1-hexyne (1.6 mL, 13 mmol), com-
pound 3f was obtained as a pale yellow oil; yield: 1.5 g
1
(65%); [a]2D3:À148.6 (c 1.3 in CHCl3); H NMR (CDCl3): d=
when the reactions are promoted by microvawes. The 7.35–7.25 (m, 5H), 3.81–3.76 (m, 2H), 3.59–3.50 (m, 3H),
Adv. Synth. Catal. 2009, 351, 1539 – 1556
ꢅ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1549

Dana Popa et al.
FULL PAPERS
3.2–3.17 (m, 1H), 2.22 (td, J=7.1, 2.7 Hz, 2H), 1.95 (t, J=
2.7 Hz, 1H), 1.76–1.67 (m, 2H), 1.65–1.58 (m, 2H);
13C NMR (CDCl3): d=136.9 (C), 128.5 (2CH), 128.2 (CH),
125.7 (2CH), 84.2 (C), 71.1 (CH2), 70.6 (CH2), 68.5 (CH),
61.2 (CH), 55.9 (CH), 28.7 (CH2), 25.1 (CH2), 18.2 (CH2);
J=9.8, 5.1 Hz, 1H), 2.67 (d, J=6.35 Hz, 1H), 2.43–2.41 (m,
1H); 13C NMR (CDCl3): d=162.9 (d, JC,F =3.5 Hz, C=O),
160.9 (d, JC,F =248.0 Hz, CF), 138.3 (C), 133.2 (d, JC,F
=
9.3 Hz, CH), 132.2 (d, JC,F =2.1 Hz, CH), 128.6 (2CH), 127.7
(CH), 127.3 (2CH), 124.7 (d, JC,F =3.2 Hz, CH), 120.9 (d,
JC,F =11.0 Hz, CH), 115.9 (d, JC,F =24.8 Hz, CH), 78.8 (C),
IR (film): n=3293, 2921, 1459, 1259, 1112, 880, 750 cmÀ1
;
HR.MS (ESI+): m/z=253.1198, calcd. for C15H18NaO2
75.1 (CH), 72.0 (CH), 70.8 (CH2), 58.6 (CH2), 56.8 (CH); IR
(film): n=3316, 3266, 3032, 2941, 1636, 1613, 1527, 1481,
1452, 1099, 752, 702 cmÀ1; HRMS (ESI+): m/z=350.1153,
calcd. for C19H18FNNaO3 [M+Na]+: 350.1168, found.
[M+Na]+: 253.1204.
Preparation of Amino Alcohols 4e,f
5f: Yield: 1.32 g (70%); white foam; [a]2D3: +32.8 (c 1.1 in
1
The corresponding epoxy ether (1.0 mmol), 30% aqueous
NH3 (50 mmol), LiClO4 (1 mmol) and isopropyl alcohol
(5 mL), were heated in a sealed pressure tube at 1008C for
6–7 h. The mixture was left to reach room temperature and
the solvent was removed under reduced pressure. The crude
reaction product was dissolved in CH2Cl2 (5 mL) and
washed with water (3ꢆ5 mL). The organic phase was dried
(MgSO4) and the solvent was removed under reduced pres-
sure. The resulting amino alcohols 4e,f were used in the next
step without further purification.
CHCl3); H NMR (CDCl3): d=8.35–8.30 (m, 1H), 7.95 (dt,
J=7.9, 1.9 Hz, 1H), 7.39–7.02 (m, 8H), 5.49–5.48 (m, 1H),
4.09–4.03 (m, 1H), 3.38–3.22 (m, 4H) 2.17 (td, J=7.1,
2.7 Hz, 2H), 1.95 (t, J=2.7 Hz, 1H) 1.69–1.64 (m, 2H),
1.58–1.52 (m, 2H); 13C NMR (CDCl3): d=163.0 (d, JC,F
=
3.0 Hz, C=O), 160.4 (d, JC,F =248.2 Hz, CF), 138.8 (C), 133.2
(d, JC,F =9.3 Hz, CH), 131.8 (d, JC,F =2.0 Hz, CH), 128.4
(2CH), 127.4 (CH), 127.2 (2CH), 124.6 (d, JC,F =3.3 Hz,
CH), 121.2 (d, JC,F =12.0 Hz, CH), 115.9 (d, JC,F =24.3 Hz,
CH), 84.1 (C), 71.7 (CH), 71.6 (CH2), 71.0 (CH2), 68.7
(CH), 57.1 (CH), 28.4 (CH2), 25 (CH2), 18.2 (CH2); IR
(film): n=3316, 3266, 3032, 2941, 1636, 1613, 1527, 1481,
1452, 1099, 752, 702 cmÀ1; HR-MS (ESI+): m/z=392.1638,
calcd. for C22H24FNNaO3 [M+Na]+: 392.1650.
4e: Obtained as a pale yellow oil after flash chromatogra-
phy (hexanes/AcOEt from 100:0 to 50:50); yield: 99%; [a]2D3:
1
À33.3 (c 0.9 in CHCl3); H NMR (CDCl3): d=7.36–7.20 (m,
5H), 4.12–4.10 (m, 3H), 3.95 (ddd, J=5.1, 5.1, 5.1 Hz, 1H),
3.48–3.46 (m, 2H), 2.44 (bs, 2H), 2.41–2.40 (m, 1H);
13C NMR (CDCl3): d=141.7 (C), 128.4 (2CH), 127.5 (CH),
127.2 (2CH), 79.4 (C), 74.8 (CH), 73.5 (CH), 71.1 (CH2),
58.6 (CH2), 57.8 (CH); IR (film): n=3337, 3277, 2852, 1601,
1494, 1453, 1071, 763, 701 cmÀ1; HR-MS (ESI+): m/z=
206.1178, calcd. for C12H16NO2 [M+H]+: 206.1181.
Preparation of (2-Fluoro)phenyloxazolines 6e,f
Methanesulfonyl chloride (1.1 mmol) was slowly added to a
solution of the corresponding compound 5e,f (1.0 mmol)
and Et3N (2.2 mmol) in CH2Cl2 (8.2 mL) under an argon at-
mosphere at 08C. The mixture was left to reach room tem-
perature and stirred for an additional 2 h period. This mix-
ture was added to a saturated aqueous solution of NH4Cl
(9 mL), the two phases were separated and the aqueous
phase was extracted with CH2Cl2 (3ꢆ6 mL). The organic
phases were washed with brine (5 mL) and dried (Na2SO4).
The drying agent was filtered off and the solvent was re-
moved under reduced pressure. The residue was added to a
5% solution of KOH in MeOH (9 mL, 2.0 mmol), and the
mixture was stirred for 15 h at room temperature. The reac-
tion was quenched with water (5 mL) and the mixture ex-
tracted with CH2Cl2 (3ꢆ5 mL). The organic phases were
dried (Na2SO4) and solvent removed under reduced pres-
sure. The final product was purified by flash chromatogra-
phy (hexane/EtOAc from 100:0 to 90:10).
4f: Obtained as a pale yellow oil after flash chromatogra-
phy (hexanes/AcOEt from 100:0 to 50:50); yield: 90%; [a]2D3:
1
À68.8 (c 1.0 in CHCl3); H NMR (CDCl3): d=7.36–7.20 (m,
5H), 4.13 (d, J=5.0 Hz, 1H), 3.95 (ddd, J=5.0, 5.0, 5.0 Hz,
1H), 3.40–3.28 (m, 4H), 2.19 (td, J=7.1, 2.6 Hz, 2H), 1.94
(t, J=2.7 Hz, 1H), 1.69–1.62 (m, 2H), 1.69–1.52 (m, 2H);
13C NMR (CDCl3): d=141.0 (C), 128.5 (CH), 127.6 (CH),
127.2 (3CH), 84.2 (C), 73.0 (CH), 71.7 (CH2), 70.9 (CH2),
68.6 (CH), 57.8 (CH), 28.5 (CH2), 25.1 (CH2), 18.2 (CH2);
IR (film): n=3287, 2865, 1668, 1454, 1114, 704, 632 cmÀ1
;
HR-MS (ESI+): m/z=248.1658, calcd. for C15H22NO2 [M+
H]+: 248.1651.
Preparation of Hydroxy Amides 5e,f
Et3N (1.95 mmol) was added under argon to a solution of
the corresponding amino alcohol 4e,f (1.0 mmol) in anhy-
drous THF (1.5 mL) and the mixture was cooled at 08C. A
solution of 2-fluorobenzoyl chloride (0.95 mmol) in anhy-
drous THF (1.3 mL) was slowly added to the previous solu-
tion. The mixture was left to reach room temperature and
stirred for an additional 2 h period. The solvents were re-
moved under reduced pressure. The crude reaction product
was diluted with CH2Cl2 (5 mL) and washed with brine (3ꢆ
5 mL). The organic phase was dried (Na2SO4) and the sol-
vent removed under reduced pressure. The residue was puri-
fied by flash chromatography (hexanes/EtOAc from 100:0
to 60:40).
6e: Yield: 0.85 g (73%); colourless oil; [a]2D3: +52.0 (c 1.2
in CHCl3); 1H NMR (CDCl3): d=7.98 (ddd, J=7.5, 1.8,
1.8 Hz, 1H), 7.5–7.14 (m, 8H), 5.18 (d, J=7.3 Hz, 1H), 4.64
(ddd, J=7.3, 4.7, 4.7 Hz, 1H), 4.32–4.23 (m, 2H), 3.89–3.83
(m, 2H), 2.47–2.45 (m, 1H); 13C NMR (CDCl3): d=161.1
(d, JC,F =258.5 Hz, CF), 160.8 (d, JC,F =5.4 Hz, C=N), 141.7
(C), 133.0 (d, JC,F =8.4 Hz, CH), 131.3 (d, JC,F =1.4 Hz, CH),
128.7 (2CH), 127.6 (CH), 126.6 (2CH), 123.9 (d, JC,F
3.8 Hz, CH), 116.6 (d, JC,F =21.0 Hz, CH), 115.8 (d, JC,F
=
=
10.2 Hz, C), 85.4 (CH), 79.1 (C), 75.1 (CH), 72.4 (CH), 70.3
(CH2), 58.7 (CH2); 19F NMR (CDCl3): d=À108.8 (s, F); IR
(film): n=290, 2857, 1648, 1495, 1457, 1053, 766, 701 cmÀ1
;
5e: Yield: 1.32 g (69%); white foam; [a]2D3: +46.1 (c=0.8
HR-MS (ESI+): m/z=332.1049, calcd. for C19H16FNNaO2
1
in CHCl3); H NMR (CDCl3): d=8.11 (bs, 1H), 8.06 (ddd,
[M+Na]+: 332.1063.
J=7.8, 1.9 Hz, 1H), 7.49–7.10 (m, 8H), 5.48–5.44 (m, 1H),
4.19–4.14 (m, 3H), 3.59 (dd, J=9.6, 3.6 Hz, 1H), 3.41 (dd,
6f: Yield: 0.7 g (81%); colourless oil; [a]2D3: +47.2 (c 1.1 in
CHCl3); 1H NMR (CDCl3): d=8.00 (ddd, J=7.5, 1.8,
1550
asc.wiley-vch.de
ꢅ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 1539 – 1556

Towards Continuous Flow, Highly Enantioselective Allylic Amination
FULL PAPERS
1.8 Hz, 1H), 7.50–7.15 (m, 8H), 5.19 (d, J=7.2 Hz, 1H),
4.64 (ddd, J=7.4, 5.0, 5.0 Hz, 1H), 3.79–3.74 (m, 2H), 3.64–
3.54 (m, 2H), 2.27 (td J=7.1, 2.7 Hz, 2H), 1.95 (t, J=
2.7 Hz, 1H), 1.78–1.71 (m, 2H), 1.68–1.59 (m, 2H);
13C NMR (CDCl3): d=161.5 (d, JC,F =260.0 Hz, CF), 160.9
(d, JC,F =5.4 Hz, C=N), 141.9 (C), 133.0 (d, JC,F =8.9 Hz,
CH), 131.3 (d, JC,F =1.5 Hz, CH), 128.7 (2CH), 127.6 (CH),
Preparation of the Polymer-Supported Phosphino-
oxazolines 8e,f
A solution of KPPh2 (1.37 mmol, 2.74 mL of 0.5M solution
in THF) was added dropwise under argon at 08C to an
oven-dried Schlenk flask which contained the corresponding
resin 7e,f (0.98 mmol) previously swollen with anhydrous
and degassed THF (10 mL). The reaction mixture was
shaken at 08C for 2 h, then allowed to reach room tempera-
ture and further shaken for 12 h at this temperature. The so-
lution was removed under argon via cannula and the resin
was washed with anhydrous and degassed CH2Cl2 (7ꢆ
10 mL) and dried under vacuum for 10 h. Resins 8e,f were
characterized by gel-phase 31P NMR and were immediately
transformed into the corresponding palladium complexes
9e,f to minimize oxidative deterioration.
126.7 (2CH), 123.9 (d, JC,F =4.0 Hz, CH), 116.7 (d, JC,F
=
22.0 Hz, CH), 116.0 (d, JC,F =11 Hz, C), 85.7 (CH), 84.2 (C),
72.5 (CH), 71.6 (CH), 71.2 (CH2), 68.5 (CH2), 28.6 (CH2),
25.1 (CH2), 18.1 (CH2); 19F NMR (CDCl3): d=À108.9 (s, F);
IR (film): n=3300, 2863, 1648, 1613, 1495, 1457, 1111,
760 cmÀ1
; HR-MS (ESI+): m/z=352.1720, calcd. for
C22H23FNO2 [M+H]+: 352.1713.
Resin 8e: 31P NMR (CDCl3): d=À3.2 (s, PPh2).
Resin 8f: 31P NMR (CDCl3): d=À3.0 (s, PPh2).
Preparation of the Click Resins 7e,f
The N3-functionalized resin[19] (1.54 g, f=0.98 mmolgÀ1) was
reacted with the corresponding alkynyloxymethyl oxazoline
6e,f (1.62 mmol), CuI (2 mg, 0.01 mmol) and DIPEA
(0.17 mL, 0.99 mmol) in a 1:1 mixture of DMF and THF
(10 mL) at 458C. The progression of the reaction was moni-
tored by IR spectroscopy. After disappearance of the azide
signal (40 h) the resin was collected by filtration and sequen-
tially washed with water (250 mL), DMF (250 mL), THF
(250 mL), THF-MeOH 1:1 (250 mL), MeOH (250 mL) and
THF (250 mL). The solid was dried under vacuum overnight
at 408C.
Preparation of the Polymer-Supported Phosphino-
oxazoline p-Allylpalladium Complexes 9e,f
A solution of [PdACTHNUTRGNE(UNG C3H5)Cl]2 (0.018 mmol, 67 mg) in anhy-
drous and deoxygenated toluene (1 mL) was added to an
oven-dried Schlenk flask which contained the corresponding
resin 8e,f (0.5 g) previously swollen with anhydrous and de-
gassed toluene (5 mL). The reaction mixture was shaken for
1 h. The resin was filtered, rinsed with toluene (10 mL) and
CH2Cl2 (200 mL) and dried in vacuo for 12 h. Spherical
beads with a 0.102 mm diameter (120 mesh) were obtained
in this way.
1
Resin 7e: H NMR (HRMAS, CDCl3): d=7.99–7.91 (m,
1
Resin 9e: H NMR (HRMAS, CDCl3): d=8.32–8.20 (m,
1H), 7.45–6.27 (m, polymer), 5.40–5.10 (m, polymer), 5.12
(m, 1H), 4.73–4.66 (m, 2H), 4.65–4.60 (m, 1H), 3.85–3.81
(m, 2H), 2.12–0.85 (m, polymer); 13C NMR (HRMAS,
CDCl3): d=161.1 (d, JC,F =258.5 Hz, CF), 160.0 (C=N),
146.3–144.6 (m, polymer), 141.8 (C), 135.3 (CH), 133.2 (d,
1H), 7.65–6.25 (m, polymer), 5.78–5.41 (m, polymer), 5.30
(m, 1H), 4.96–4.34 (m, 2H), 4.61 (m, 1H), 4.07–3.69 (m,
4H), 3.34–3.08 (m, 2H), 2.99–2.76 (m, 2H), 2.33 (d, J=
16.5 Hz, 2H), 2.14 (d, J=23.6 Hz, 2H), 1.99–0.90 (m, poly-
mer); 13C NMR (HRMAS, CDCl3): d=146.3–144.3 (m, po-
lymer), 139.6 (CH), 135.4–130.9 (m, CH), 130.7–124.8 (m,
CH), 114.0–108.5 (m, polymer), 87.2 (CH), 75.2 (CH), 69.5
(CH2), 64.5 (CH2), 55.3–53.4 (m, CH2), 47.1–41.5 (m, poly-
mer), 41.3–38.5 (m, polymer), 29.7 (CH2), 25.6 (CH2);
31P NMR (CDCl3): d=25.5 (s, PPh2, exo), 24.8 (s, PPh2,
endo); IR (ATR): n=3057, 3024, 2919, 1625, 1600, 1542,
1492, 1451, 1435, 1350, 1116, 1098, 728, 696, 538 cmÀ1. A
100% yield of functionalization was calculated on the basis
of nitrogen elemental analysis calcd. (%): N 2.95; found: N
2.93; f=0.53 mmolgÀ1.
J
C,F =8.4 Hz, CH), 131.4 (CH), 128.8 (CH), 128.0 (CH),
127.7 (CH), 126.7 (CH), 125.7 (CH), 124.0 (d, JC,F =3.7 Hz,
CH), 116.6 (d, JC,F =22.0 Hz, CH), 115.8 (C), 112.3–109.5
(m, polymer), 85.7 (CH), 72.3 (CH), 71.2 (CH2), 68.0 (CH2),
40.7–40.1 (m, polymer), 29.7 (CH2), 25.6 (CH2); 19F NMR
(CDCl3): d=À109.6 (s, F); IR (ATR): n=3058, 3024, 2919,
1647, 1600, 1492, 1451, 1308, 1221, 1180, 1066, 1027, 753,
696, 556 cmÀ1. A 100% yield of functionalization was calcu-
lated on the basis of nitrogen elemental analysis calcd. (%):
N 3.61; found: N 3.91; f=0.71 mmolgÀ1.
1
1
Resin 7f: H NMR (HRMAS, CDCl3): d=7.99–7.91 (m,
Resin 9f: H NMR (HRMAS, CDCl3): d=8.33–8.24 (m,
1H), 7.42–6.20 (m, polymer), 5.38–4.96 (m, polymer), 5.15
(m, 1H), 4.62–4.34 (m, 1H), 3.72–3.66 (m, 2H), 3.60–3.47
(m, 2H), 2.74–2.55 (m, 2H), 2.12–0.85 (m, polymer);
13C NMR (HRMAS, CDCl3): d=161.5 (d, JC,F =263.9 Hz,
CF), 160.0 (C=N), 146.3–144.6 (m, polymer), 142.0 (C),
133.2 (CH), 131.4 (CH), 128.8 (CH), 128.0 (CH), 127.7
1H), 7.71–6.18 (m, polymer), 5.65–5.08 (m, polymer), 5.48
(m, 1H), 4.89–4.46 (m, 2H), 4.61 (m, 1H), 3.93–3.64 (m,
4H), 3.62–3.46 (m, 2H), 3.32–3.05 (m, 2H), 3.03–2.80 (m,
2H), 2.79–2.54 (m, 2H), 2.34 (d, J=18.0 Hz, 2H), 2.14 (d,
J=24.3 Hz, 2H), 2.03–1.21 (m, polymer); 13C NMR
(HRMAS, CDCl3): d=146.3–144.3 (m, polymer), 139.5
(CH), 134.9–130.8 (m, CH), 129.9–124.6 (m, CH), 114.1–
105.6 (m, polymer), 87.2 (CH), 75.6 (CH), 71.5 (CH2), 70.4
(CH2), 54.1 (CH2), 46.4–41.8 (m, polymer), 41.6–39.4 (m, po-
lymer), 29.6 (CH2), 26.0 (CH2), 25.5 (CH2); 31P NMR
(CDCl3): d=25.5 (s, PPh2); IR (ATR): n=3024, 2919, 1624,
1600, 1491, 1451, 1435, 1307, 1180, 1118, 749, 695, 540 cmÀ1.
A 100% yield of functionalization was calculated on the
basis of nitrogen elemental analysis calcd. (%): N 2.88;
found: N 2.99; f=0.68 mmolgÀ1.
(CH), 126.8 (CH), 125.7 (CH), 124.0 (CH), 116.7 (d, JC,F
=
23.0 Hz, CH), 116.1 (C), 112.3–109.5 (m, polymer), 85.9
(CH), 72.5 (CH), 71.8 (CH2), 71.6 (CH2), 68.0 (CH2), 40.7–
40.1 (m, polymer), 29.3 (CH2), 26.1 (CH2), 25.7 (CH2), 25.5
(CH2); 19F NMR (CDCl3): d=À109.6 (s, F); IR (ATR): n=
3058, 3024, 2919, 1647, 1600, 1492, 1451, 1308, 1255, 1180,
1066, 1027, 753, 696, 556 cmÀ1. A 100% yield of functionali-
zation was calculated on the basis of nitrogen elemental
analysis calcd. (%): N 3.52; found: N 3.82; f=0.68 mmolgÀ1.
Adv. Synth. Catal. 2009, 351, 1539 – 1556
ꢅ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1551

Dana Popa et al.
FULL PAPERS
538 cmÀ1
;
HR-MS (ESI+): m/z=755.1763, calcd. for
Preparation of Model Fluorooxazolines 10e,f
C41H38N4O2PPd [MÀPF6]+: 755.1767.
Benzyl bromide (0.13 mL, 1.07 mmol) was added to a mix-
ture of the corresponding alkynyloxymethyloxazoline 6e,f
(1 mmol), sodium azide (0.148 g, 2.13 mmol), CuSO4·5H2O
(5 mg, 0.002 mmol) and sodium l-ascorbate (43 mg,
0.21 mmol) in tert-butyl alcohol:water 1:1 (3 mL). The mix-
ture was submitted to microwave irradiation (150 W, 1008C,
2 min ramp and 40 min hold time). The reaction mixture
was extracted with ethyl acetate (3ꢆ25 mL) and the com-
bined organic phases were dried (Na2SO4) and concentrated
under reduced pressure. The residue was purified by flash
chromatography (hexanes/EtOAc from 80:20 to 20:80). See
Supporting Information for the physical and spectroscopic
data of compounds 10e,f.
General Procedure for the Palladium-Catalyzed
Allylic Amination of S1 with Different N-Nucleo-
philes in the Presence of 2a–d
To an oven-dried Schlenk flask containing the corresponding
phosphinooxazoline palladium complex 2a–d (0.025 mmol)
under argon were successively added (E)-3-acetoxy-1,3-di-
phenyl-1-propene (S1) (0.25 g, 1 mmol), the N-nucleophile
(3 mmol), BSA (0.74 mL, 3 mmol) and KOAc (2.5 mg,
0.025 mmol). The mixture was stirred at room temperature
for 2–24 h (unless otherwise stated, see Table 1) and then di-
luted with diethyl ether and washed with water. The organic
phase was dried (MgSO4) and concentrated under reduced
pressure. The residue was purified by flash chromatography
(hexanes/EtOAc from 100:0 to 80:20).
Preparation of Model Phosphinooxazoline p-Allyl-
palladium Complexes 11e,f
A solution of KPPh2 (0.77 mmol, 1.55 mL of 0.5M solution
in THF) was added dropwise under Ar at À788C to an
oven-dried Schlenk flask which contained the corresponding
fluorooxazoline 10e,f (0.55 mmol) in THF (2 mL). The tem-
perature was allowed to reach À208C. The reaction mixture
was stirred for an additional 2 h period at this temperature,
then allowed to reach room temperature, further stirred for
12 h at this temperature, quenched with Na2SO4·10H2O in
order to hydrolyze the excess of diphosphine, and filtered
through a short SiO2 pad eluting with CH2Cl2. Solvents were
removed under vacuum, and the residue was purified by
flash chromatography over SiO2 under argon using deoxy-
genated solvents. The resulting phosphinooxazolines (see
(+)-(S,E)-N-Benzyl-(1,3-diphenyl-2-propenyl)amine (12)[3]:
From S1 (0.25 g, 1.0 mmol), 2a (18.7 mg, 0.025 mmol), ben-
zylamine (0.33 mL, 3.0 mmol), BSA (0.74 mL, 3.0 mmol)
and KOAc (2.5 mg, 0.025 mmol). Yield: 0.29 g (98%); col-
ourless oil; [a]23: +16.4 (c 0.85 in CHCl3), 95% ee. The en-
antiomeric excDess was determined by HPLC on an OD-H
column (0.6 mLminÀ1 n-hexane/isopropyl alcohol 99:1,
254 nm): (R)-12 Rt=19.5 min, (S)-12 Rt=20.8 min.
(+)-(S,E)-N-(p-Methoxybenzyl)-(1,3-diphenyl-2-propenACHTUNGTRENNUNGyl)-
amine (13): From S1 (0.25 g, 1.0 mmol), 2a (18.7 mg,
0.025 mmol), p-methoxybenzylamine (0.396 mL, 3.0 mmol),
BSA (0.74 mL, 3.0 mmol) and KOAc (2.5 mg, 0.025 mmol).
Yield: 0.31 g (94%) yellow oil; [a]2D3: +28.2 (c 0.82 in
CHCl3), 94% ee. The enantiomeric excess was determined
by HPLC on a AD-H column (0.7 mLminÀ1 n-hexane/iso-
propyl alcohol 94:6, 254 nm): (S)-13 Rt=18.0 min, (R)-13
Rt=20.9 min; 1H NMR (CDCl3): d=7.29–7.17 (m, 12H),
6.88–6.84 (m, 2H), 6.56 (d, J=15.8 Hz, 1H), 6.30 (dd, J=
15.8, 7.5 Hz, 1H), 4.37 (d, J=7.5 Hz, 1H), 3.79 (s, 3H),
3.75–3.68 (m, 2H); 13C NMR (CDCl3): d=158.7 (C), 143.0
(C), 137.0 (C), 132.7 (CH), 132.5 (C), 130.3 (CH), 129.4
(2CH), 128.6 (2CH), 128.5 (2CH), 127.5 (CH), 127.4 (2CH),
127.3 (2CH), 126.4 (CH), 113.8 (2CH), 64.5 (CH), 55.3
(CH3), 50.8 (CH2); IR (film): n=3024, 2832, 1610, 1510,
1492, 1245, 1174, 966, 914, 745 cmÀ1; HR-MS (ESI+): m/z=
352.1666, calcd. for C23H23NONa [M+Na]+: 352.1677.
1
Supporting Information for H- and 31P NMR spectra) were
immediately transformed into the corresponding palladium
complexes 11.
A
solution of the corresponding phosphinooxazoline
(0.38 mmol) in deoxygenated EtOH (4 mL) was via cannula
to
a
Schlenk flask which contained
a
solution of
[Pd
A
mixture was stirred 1 h under argon at room temperature;
NH4PF6 (0.41 mmol, 0.068 g) was then added, and the result-
ing solution was further stirred for 14 h. Then, the mixture
was cooled to À208C and allowed to stand at this tempera-
ture for several hours to induce crystallization of the p-allyl-
palladium complex. The crystallized phosphinooxazoline p-
allylpalladium complex was filtered off, washed with cold
EtOH (3 mL) and dried under vacuum. See Supporting In-
formation for the physical and spectroscopic data of com-
pound 11f.
(+)-(S,E)-N-Propargyl-(1,3-diphenyl-2-propenyl)amine
(14): From S1 (0.25 g, 1.0 mmol), 2a (18.7 mg, 0.025 mmol),
propargylamine (0.19 mL, 3.0 mmol), BSA (0.74 mL,
3.0 mmol) and KOAc (2.5 mg, 0.025 mmol). Yield: 0.247 g
(99%); yellow oil; [a]D23: +26.6 (c 1.15 in CHCl3), 97% ee.
The enantiomeric excess was determined by HPLC on a
AD-H column (0.5 mLminÀ1 n-hexane/isopropyl alcohol
90:10, 254 nm): (R)-14 Rt=17.0 min, (S)-14 Rt=18.4 min;
1H NMR (CDCl3): d=7.43–7.12 (m, 10H), 6.63 (d, J=
15.9 Hz, 1H), 6.26 (dd, J=15.9, 7.7 Hz, 1H), 4.60 (d, J=
7.7 Hz, 1H), 3.43 (dd, J=17.3, 2.6 Hz, 1H), 3.34 (dd, J=
17.3, 2.6 Hz, 1H), 2.24–2.25 (m, 1H); 13C NMR (CDCl3):
d=142.0 (C), 136.8 (C), 131.6 (CH), 130.8 (CH), 128.6
(2CH), 128.5 (2CH), 127.5 (2CH), 127.4 (2CH), 126.4
(2CH), 82.1 (C), 71.5 (CH), 63.6 (CH), 35.7 (CH2); IR
(film): n=3290, 3025, 2832, 1491, 1449, 1330, 1110, 968, 913,
Model compound 11e: Yield: 82%; orange solid; [a]2D7:
À27.9 (c 1.1 in CHCl3); 1H NMR (CD3CN): d=8.28–8.22
(m, 2H), 7.83–6.97 (m, 42H), 6.85 (d, J=7.4 Hz, 4H), 5.56
(m, 4H), 5.30 (d, J=7.4 Hz, 2H), 4.74–4.61 (m, 6H), 3.95
(d, J=12.0 Hz, 2H), 3.78 (d, J=12.0 Hz, 2H); 13C NMR
(CD3CN): d=165.2 (CN), 139.5 (C), 134.7 (CH), 133.9
(CH), 133.8 (CH), 133.5 (CH), 133.3 (CH), 132.7 (CH),
132.3 (CH), 132.1 (CH), 131.9 (CH), 129.8 (CH), 129.7
(CH), 129.5 (CH), 129.3 (CH), 129.2 (CH), 129.0 (CH),
128.7 (CH), 128.6 (CH), 128.1 (CH), 127.1 (CH), 87.0 (CH),
76.3 (CH), 69.3 (CH2), 64.1 (CH2), 56.0 (CH2); 31P NMR
(CDCl3): d=24.9 (s, PPh2, exo), 24.4 (s, PPh2, endo),
À141.05 (h, J=712.5 Hz, PF6); IR (film): n=3060, 2920,
1622, 1480, 1456,1436, 1451, 1307, 1118, 831, 746, 694, 556,
749 cmÀ1
C18H17NNa [M+Na]+: 270.1268.
; HR-MS (ESI+): m/z=270.1259, calcd. for
1552
asc.wiley-vch.de
ꢅ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 1539 – 1556

Towards Continuous Flow, Highly Enantioselective Allylic Amination
FULL PAPERS
(+)-(S,E)-N,N-Diallyl-(1,3-diphenyl-2-propenyl)amine
(15): From S1 (0.25 g, 1.0 mmol), 2a (18.7 mg, 0.025 mmol),
diallylamine (0.37 mL, 3.0 mmol), BSA (0.74 mL, 3.0 mmol)
and KOAc (2.5 mg, 0.025 mmol). Yield: 0.287 g (99%);
yellow oil; [a]2D3: +27.3 (c 2.2 in CHCl3), 99% ee. The enan-
tiomeric excess was determined by HPLC on a AD-H
acetate S2–S5 as starting material and benzylamine as N-nu-
cleophile at room temperature (unless stated otherwise, see
Table 2).
(+)-(S,E)-N-Benzyl-[1,3-bis(2-chlorophenyl)-2-propenyl]-
amine (19)[4b]: [a]D23: +3.8 (c 0.93 in CHCl3), 94% ee. The en-
antiomeric excess was determined by HPLC on an OD-H
column (0.5 mLminÀ1 n-hexane/isopropyl alcohol 99:1,
254 nm): (R)-19 Rt=13.1 min, (S)-19 Rt=15.6 min.
(+)-(R)-N-Benzyl-[1-methyl-3,3-diphenyl-2-propenyl]am-
column (0.4 mLminÀ1 n-hexane from
0
to 5 min,
0.2 mLminÀ1 n-hexane from 5 to 40 min, 254 nm): (R)-15
Rt=19.0 min, (S)-15 Rt=20.0 min; 1H NMR (CDCl3): d=
7.39–7.20 (m, 10H), 6.54 (d, J=15.7 Hz, 1H), 6.33 (dd, J=
15.7, 9.1 Hz, 1H), 5.92–5.82 (m, 2H), 5.19–5.12 (m, 4H),
4.43 (d, J=9.1 Hz, 1H), 3.23–3.10 (m, 4H); 13C NMR
(CDCl3): d=142.2 (C), 137.0 (C), 136.0 (CH), 132.4 (CH),
129.8 (CH), 128.6 (CH), 128.5 (2CH), 128.3 (2CH), 127.9
(CH), 127.5 (CH), 127.0 (CH), 126.7 (CH), 126.4 (2CH),
117.1 (2CH2), 67.1 (CH), 52.6 (2CH2); IR (film): n=3059,
3025, 2920, 2813, 1739, 1641, 1599, 1492, 1448, 1417, 1229,
1028, 968, 917, 743 cmÀ1; HR-MS (ESI+): m/z=290.1917,
calcd. for C21H24N [M+H]+: 290.1909.
AHCTUNGTRENNUNG
ine (20)[4b,26]: [a]28: +89.0 (c 1.28 in CHCl3), 89% ee. The en-
antiomeric exceDss was determined by HPLC on an OJ-H
column (0.6 mLminÀ1 n-hexane/isopropyl alcohol 90:10,
254 nm): (S)-20 Rt=12.0 min, (R)-20 Rt=18.2 min.
(+)-(R,E)-N-Benzyl-(1-methyl-2-butenyl)amine (21)[3,26]
:
[a]2D8: +16.0 (c 1.76 in CHCl3), 63% ee. The enantiomeric
excess was determined by HPLC on an OD-H column
(0.5 mLminÀ1 n-hexane, 230 nm): (R)-21 Rt=30.7 min, (S)-
21 Rt=36.4 min.
(R)-N-Benzyl-(cyclohex-2-enyl)amine (22)[27,28]: The enan-
tiomeric excess was determined by HPLC on an OB-H
column (0.5 mLminÀ1 n-hexane/isopropyl alcohol 95:5,
230 nm): (R)-22 Rt=10.7 min, (S)-22 Rt=12.0 min.
(+)-(S,E)-N-Benzhydryl-(1,3-diphenyl-2-propenyl)amine
(16): From S1 (0.25 g, 1.0 mmol), 2a (18.7 mg, 0.025 mmol),
benzhydrlylamine (0.53 mL, 3.0 mmol), BSA (0.74 mL,
3.0 mmol) and KOAc (2.5 mg, 0.025 mmol). Yield: 0.287 g
(94%); colourless oil; [a]23: +30.6 (c 1.2 in CHCl3), 96% ee.
The enantiomeric excessDwas determined by HPLC on a
AD-H column (0.6 mLminÀ1 n-hexane/isopropyl alcohol
General Procedure for the Palladium-Catalyzed
Allylic Amination of Substrates S1–S5 in the
Presence of 11e
1
98:2, 254 nm): (S)-16 Rt=8.0, (R)-16 Rt=9.2 min; H NMR
(CDCl3): d=7.75–7.05 (m, 20H), 6.49(d, J=15.8 Hz, 1H),
6.29 (dd, J=15.8, 7.5 Hz, 1H), 4.87 (s, 1H), 4.28 (d, J=
7.2 Hz, 1H); 13C NMR (CDCl3): d=144.0 (C), 143.9 (C),
143.0 (C), 137.0 (C), 132.4 (CH), 130.6 (CH), 128.4 (2CH),
128.6 (2CH), 128.5 (2CH), 128.3 (2CH), 127.6 (2CH), 127.4
(2CH), 127.3 (2CH), 127.0 (2CH), 126.9 (2CH), 126.4
(2CH), 63.5 (CH), 61.9 (CH); IR (film): n=3025, 2925,
2850, 1949, 1810, 1598, 1492, 1180, 966, 914, 745 cmÀ1; HR-
MS (ESI+): m/z=376.2056, calcd. for C28H25N [M+H]+:
376.2065.
To an oven-dried, 5-mL conical flask equipped with a
septum, containing 11e (22.5 mg, 0.025 mmol) under argon
were successively added the corresponding allylic acetate
S1–S5 (1 mmol), the corresponding N-nucleophile (3 mmol),
BSA (0.74 mL, 3 mmol) and KOAc (2.5 mg, 0.025 mmol).
The mixture was stirred at room temperature for 24 À72 h
(unless stated otherwise, see Table 3 and Table 4) and then
concentrated under reduced pressure. The residue was puri-
fied by flash chromatography (hexanes/EtOAc from 100:0
to 80:20).
(+)-(S,E)-N-(1,3-Diphenyl-2-propenyl)-N’-benzoylhydra-
zine (17)[3]: From S1 (0.25 g, 1.0 mmol), 2a (18.7 mg,
0.025 mmol), benzoylhydrazine (0.42 g, 3.0 mmol), BSA
(1.48 mL, 6.0 mmol) and KOAc (2.5 mg, 0.025 mmol). Yield:
0.325 g (99%); [a]28: +36.4 (c 0.73 in CHCl3), 94% ee. The
enantiomeric exceDss was determined by HPLC on an OJ
column (0.5 mLminÀ1 n-hexane/isopropyl alcohol 85:15,
254 nm): (S)-17 Rt=27.7 min, (R)-17 Rt=31.4 min; (AD-H
column, 0.5 mLminÀ1 n-hexane/isopropyl alcohol 85:15,
254 nm): (R)-17 Rt=32.1, (S)-17 Rt=36.9 min.
General Procedure for the Palladium-Catalyzed
Allylic Amination of S1 in the Presence of 9e
S1 (0.05 g, 0.2 mmol), the corresponding N-nucleophile
(0.6 mmol) and BSA (0.148 mL, 0.6 mmol) were syringed
into an oven-dried, 5-mL conical flask equipped with a
septum, containing 9e (9.5 mg, 0.005 mmol) and KOAc
(2.5 mg, 0.025 mmol) under argon. The reaction mixture was
smoothly shaken (orbital shaker) at room temperature for
the time indicated in Table 3. Then the resin was filtered off
and rinsed with anhydrous CH2Cl2 (3ꢆ3 mL). The combined
filtrates were concentrated under reduced pressure and the
residue was purified by flash chromatography (hexanes/
EtOAc from 100:0 to 80:20).
(+)-(S,E)-N-(1,3-Diphenyl-2-propenyl)phthalimide (18)[4b,20]
:
From S1 (0.25 g, 1.0 mmol), 2a (18.7 mg, 0.025 mmol), potassi-
um phthalimide (0.573 g, 3.0 mmol), BSA (0.74 mL, 3.0 mmol)
and KOAc (2.5 mg, 0.025 mmol). Yield: 0.22 g (65%); [a]2D9:
+21.5 (c 0.87 in CHCl3), 92% ee. The enantiomeric excess was
determined by HPLC on an OD-H column (0.5 mLminÀ1 n-
hexane/isopropyl alcohol 98:2, 254 nm): (S)-18 Rt=25.6 min,
(R)-18 Rt=31.6 min.
General Procedure for the Palladium-Catalyzed
Allylic Amination of S1 in the Presence of 9f
Reactions were performed as described for the allylic ami-
nation of S1 in the presence of 9e, but in a vial for micro-
wave reactor and without the presence of the KOAc addi-
tive, using 9f (26 mg, 0.014 mmol) as the catalyst and heat-
ing the reaction mixture in a microwave reactor at 408C
(setting temperature) for 2–12 h (unless stated otherwise,
General Procedure for the Palladium-Catalyzed
Allylic Amination of Substrates S2–S5 in the
Presence of 2a
The procedure was analogous to the one described above
for S1 but using 2a as the catalyst, the corresponding allylic
Adv. Synth. Catal. 2009, 351, 1539 – 1556
ꢅ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1553

Dana Popa et al.
FULL PAPERS
see Table 3). Isolation of compounds 12–16 was performed
as stated before (vide supra).
the resin and the mechanical parts of the reactor. The col-
lected output flow was concentrated, diluted with Et2O
(5 mL), washed with water (3ꢆ20 mL), dried (MgSO4) and
concentrated under vacuum. Flash chromatography of the
reaction crude (hexanes/EtOAc from 100: 0 to 90:10) af-
forded 12 in 68% yield and 83% ee.
General Procedure for the Palladium-Catalyzed
Allylic Amination of Substrates S2–S5 in the
Presence of 9f
The procedure was analogous to the one described for allyl-
ic amination of S1 under microwave-assisted conditions but
using 5 mol% of 9f (18.6 mg, 0.01 mmol) and KOAc
(2.5 mg, 0.025 mmol) as co-additive (unless stated otherwise,
see Table 4). Isolation of compounds 19–22 was performed
as stated before (vide supra).
Acknowledgements
We thank MICINN (Grant Nos. CTQ2005–02193/BQU,
CTQ2008–00947/BQU, CTQ2008–00950/BQU), DURSI
(Grant No. 2005GR225), Consolider Ingenio 2010
(CSD2006–0003) and ICIQ and Ramꢀn Areces Foundations
for financial support. R. M. is grateful to the Spanish Minis-
terio de Ciencia e Innovaciꢀn for a predoctoral fellowship.
We also thank Dr. G. Gonzꢁlez and Dr. K. Gꢀmez (ICIQ
support unit) for their help with HRMAS 13C NMR measure-
ments and Dr. C. Rodrꢂguez-Escrich for his assistance in the
design and performance of the continuous flow system.
Recycling Experiments
As a representative example: S1 (0.062 g, 0.25 mmol), ben-
zylamine (0.107 mL, 0.98 mmol) and BSA (0.24 mL,
0.98 mmol) were added via syringe to an oven-dried vial for
microwave reactor containing 9f (51 mg, 0.025 mmol), previ-
ously swollen with anhydrous and degassed CH2Cl2
(0.11 mL, 1.72 mmol), under argon. The reaction mixture
was heated in a microwave reactor in power control mode
(1 W) for 35 min. The temperature of the reaction mixture,
measured with an internal, teflon-coated Pt-100 probe, was
26–278C. Then, the solution was removed under argon via
cannula and the resin was rinsed with CH2Cl2 (3ꢆ1 mL) and
dried under argon for 10 min. The resin was swollen again
with CH2Cl2 (0.11 mL, 1.72 mmol), the reactants were added
and the mixture was reacted as indicated before. The same
resin was used for each cycle and no further Pd source was
added. Isolation of compound 12 was performed as stated
before (vide supra).
References
[1] For reviews, see: a) B. M. Trost, D. L. van Vranken,
Chem. Rev. 1996, 96, 395; b) A. Pfaltz, M. Lautens, in:
Comprehensive Asymmetric Catalysis, Vol. 2, (Eds.:
E. N. Jacobsen, A. Pfaltz, H. Yamamoto), Springer,
Berlin, 1999, Chap. 24; c) B. M. Trost, M. L. Crawley,
Chem. Rev. 2003, 103, 2921; d) Z. Lu, S. Ma, Angew.
Chem. 2008, 120, 264; Angew. Chem. Int. Ed. 2008, 47,
258, and references cited therein; e) P. Butti, R.
Rochat, A. D. Sadow, A. Togni, Angew. Chem. 2008,
120, 4956; Angew. Chem. Int. Ed. 2008, 47, 4878.
[2] a) C. G. Frost, J. Howart, J. M. J. Williams, Tetrahedron:
Asymmetry 1992, 3, 1089; b) J. M. J. Williams, Synlett
1996, 705; c) B. Bartels, G. Helmchen, Chem. Commun.
1999, 741; d) G. Helmchen, A. Pfaltz, Acc. Chem. Res.
2000, 33, 336; e) H. A. McManus, P. J. Guiry, Chem.
Rev. 2004, 104, 4151; f) A. Pfaltz, W. J. Drury III, Proc.
Natl. Acad. Sci. USA 2004, 101, 5723.
Continuous Flow System Description and Details of
the Reaction Procedure
The continuous flow system (see Figure 4) was set up
around a vertically mounted 1/4 inch teflon tube, which was
loaded with resin 9f (240 mg). Teflon connectors allowed us
to connect it through 1/16 inch teflon tubes to an Ismatec
piston pump (reagents input) and to a collector vessel. A
schematic drawing of all the pieces and connections appears
in the Supporting Information. A solution of S1 (9.63 mmol,
2.43 g), BSA (38.52 mmol, 9.56 mL), and benzylamine
(48.15 mmol, 5.27 mL) in dry CH2Cl2 (67.42 mmol, 4.32 mL)
was prepared in a flask under argon. Dry CH2Cl2 was kept
in an independent bottle with a connector cap with valves
for an easy and inert connection to the tubing. The resin
was swollen by pumping CH2Cl2 at a 0.12 mLminÀ1 flow
during 30 min. Care was taken that all the resin beads
packed in the tube were inside the microwave cavity (see
Figure 5). Then, the microwave apparatus was connected
(1 W, open-vessel) and the solution of the reagents was
pumped through the system at a 0.12 mLminÀ1 flow for 3 h.
The efluent reaction mixture was collected through the
upper end of the reactor, taking aliquots every 15 min to
measure 1H NMR spectra and to determine the ee of the
amination product by HPLC. Temperature of the efluent
(22–238C) was monitored with a Pt-100 probe, and turned
out to be identical (Æ18C) to that of the reagents mixture.
When all the reacting solution had been pumped, CH2Cl2
was passed for an additional 30 min period in order to wash
[3] P. von Matt, O. Loiseleur, G. Koch, A. Pfaltz, C. Lefeb-
er, T. Feucht, G. Helmchen, Tetrahedron: Asymmetry
1994, 5, 573.
[4] a) R. Jumnah, A. C. Williams, J. M. J. Williams, Synlett
1995, 821; b) A. Sudo, K. Saigo, J. Org. Chem. 1997, 62,
5508; c) R. N. Constantine, N. Kim, R. C. Bunt, Org.
Lett. 2003, 5, 2279; d) P. B. Armstrong, L. M. Bennett,
R. N. Constantine, J. L. Fields, J. P. Jasinski, R. J. Sta-
ples, R. C. Bunt, Tetrahedron Lett. 2005, 46, 1441.
[5] a) K. Selvakumar, M. Valentini, M. Woerle, P. S. Prego-
sin, A. Albinati, Organometallics 1999, 18, 1207;
b) Y. M. Malone, P. J. Guiry, J. Organomet. Chem. 2000,
603, 110; c) W.-P. Deng, S.-L. You, X.-L. Hou, L.-X.
Dai, Y.-H. Yu, W. Xia, J. Sun, J. Am. Chem. Soc. 2001,
123, 6508; d) S.-L. You, X.-L. Hou, L.-X. Dai, Y.-H.
Yu, W. Xia, J. Org. Chem. 2002, 67, 4684; e) G. P. Xu,
S. R. Gilbertson, Tetrahedron Lett. 2002, 43, 2811; f) C.
Bolm, L. Xiao, M. Kesselgruber, Org. Biomol. Chem.
2003, 1, 145.
1554
asc.wiley-vch.de
ꢅ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 1539 – 1556

Towards Continuous Flow, Highly Enantioselective Allylic Amination
FULL PAPERS
[6] a) K. Yonehara, T. Hashizume, K. Mori, K. Ohe; S.
Uemura, J. Org. Chem. 1999, 64, 9374; b) O. Pꢁmies,
M. Diꢇguez, C. Claver, J. Am. Chem. Soc. 2005, 127,
3646; c) Y. Mata, M. Diꢇguez, O. Pꢁmies, C. Claver,
Adv. Synth. Catal. 2005, 347, 1943; d) M. Diꢇguez, O.
Pꢁmies, Chem. Eur. J. 2008, 14, 3653.
[12] a) C. W. Tornøe, M. Meldal, in: Proceedings of the
Second International and Seventeenth American Peptide
Society Symposium, (Eds.: M. Lebl, R. A. Houghten),
American Peptide Society and Kluwer Academia Press,
San Diego, 2001, p 263; b) C. W. Tornøe, C. Christen-
sen, M. Meldal, J. Org. Chem. 2002, 67, 3057; c) V. V.
Rostovtsev, L. G. Green, V. V. Fokin, K. B. Sharpless,
Angew. Chem. 2002, 114, 2708; Angew. Chem. Int. Ed.
2002, 41, 2596. For a review on the concept of click
chemistry, see: d) H. C. Kolb, M. G. Finn, K. B. Sharp-
less, Angew. Chem. 2001, 113, 2056; Angew. Chem. Int.
Ed. 2001, 40, 2004.
[13] a) D. Font, C. Jimeno, M. A. Pericꢁs, Org. Lett. 2006, 8,
4653; b) D. Font, A. Bastero, S. Sayalero, C. Jimeno,
M. A. Pericꢁs, Org. Lett. 2007, 9, 1943; c) E. Alza,
X. C. Cambeiro, C. Jimeno, M. A. Pericꢁs, Org. Lett.
2007, 9, 3717; d) D. Font, S. Sayalero, A. Bastero, C.
Jimeno, M. A. Pericꢁs, Org. Lett. 2008, 10, 337.
[14] For additional examples of ligand supporting using
click chemistry, see: a) A. Gheorghe, A. Matsuno, O.
Reiser, Adv. Synth. Catal. 2006, 348, 1016; b) R. J.
Detz, S. A. Heras, R. de Gelder, P. W. N. M. van Leeu-
wen, H. Hiemstra, J. N. H. Reek, J. H. van Maarseveen,
Org. Lett. 2006, 8, 3227; c) A. Gissibl, M. G. Finn, O.
Reiser, Org. Lett. 2005, 7, 2325; d) K. M. Kacprzak,
N. M. Maier, W. Linder, Tetrahedron Lett. 2006, 47,
8721; e) M. Tilliet, S. Lundgren, C. Moberg, V. Levach-
er, Adv. Synth. Catal. 2007, 349, 2079; f) A. Gissibl, C.
Padie, M. Hager, F. Jaroschik, R. Rasappan, E. Cuevas-
Yanez, C.-O. Turrin, A.-M. Caminade, J.-P. Majoral, O.
Reiser, Org. Lett. 2007, 9, 2895; g) A. Schaetz, R. N.
Grass, W. J. Stark, O. Reiser, Chem. Eur. J. 2008, 14,
8262.
[7] a) For a review of polymeric palladium catalysts, see:
Y. Uozumi, Top. Curr. Chem. 2004, 242, 77. For some
recent examples, see: monodentate P ligands: b) Y. Ko-
bayashi, D. Tanaka, H. Danjo, Y. Uozumi, Adv. Synth.
Catal. 2006, 348, 1561; bidentate N,N ligands: c) M.
Bandini, M. Benaglia, T. Quinto, S. Tommasi, A.
Umani-Ronchi, Adv. Synth. Catal. 2006, 348, 1521; bi-
dentate P,N ligands: d) Y. Uozumi, H. Tanaka, K. Shi-
batomi, Org. Lett. 2004, 6, 281; e) H. Nakano, K. Taka-
hashi, R. Fujita, Tetrahedron: Asymmetry 2005, 16,
2133; f) Y. Uozumi, Pure. Appl. Chem. 2007, 79, 1481;
bidentate P,S ligands: g) H. Nakano, K. Takahashi, Y.
Suzuki, R. Fujita, Tetrahedron: Asymmetry 2005, 16,
609; h) B. Martꢀn-Matute, S. I. Pereira, E. PeÇa-Cab-
rera, J. Adrio, J. C. Carretero, Adv. Synth. Catal. 2007,
349, 1714.
[8] For the preparation of a polystyrene-supported phos-
phinooxazoline, see: R.-J. Chen, J.-M. Fan, J. Chin.
Chem. Soc. 2005, 52, 819.
[9] Reviews on immobilized catalysts include the follow-
ing: a) Chiral Catalyst Immobilization and Recycling,
(Eds.: D. E. de Vos, I. F. Vankelecom, P. A. Jacobs),
Wiley-VCH, Weinheim, 2000; b) Immobilized Cata-
lysts: Solid Phases, Immobilization and Applications,
in: Topics in Current Chemistry, (Ed.: A. Kirschning),
Springer, Berlin, 2004, 242, 1; c) Polymeric Materials in
Organic Synthesis and Catalysis, (Ed.: D. R., Buchmeis-
er), Wiley-VCH, Weinheim, 2006; d) M. Heitbaum, F.
Glorius, I. Escher, Angew. Chem. 2006, 118, 4850;
Angew. Chem. Int. Ed. 2006, 45, 4732; e) Handbook of
Asymmetric Heterogeneous Catalysis, (Eds.: K. Ding, Y.
Uozumi), Wiley-VCH, Weinheim, 2008.
[15] D. Popa, C. Puigjaner, M. Gꢃmez, J. Benet-Buchholz,
A. Vidal-Ferran, M. A. Pericꢁs, Adv. Synth. Catal. 2007,
349, 2265.
[16] A. Vidal-Ferran, A. Moyano, M. A. Pericꢁs, A. Riera,
J. Org. Chem. 1997, 62, 4970.
[10] For some general references on flow-through processes,
see: a) G. Jas, A. Kirschning, Chem. Eur. J. 2003, 9,
5708; b) P. Hodge, Ind. Eng. Chem. Res. 2005, 44, 8542;
c) F. Cozzi, Adv. Synth. Catal. 2006, 348, 1367; d) A.
Kirschning, W. Solodenko, K. Mennecke, Chem. Eur. J.
2006, 12, 5972; e) B. P. Mason, K. E. Price, J. L. Stein-
bacher, A. R. Bogdan, D. T. McQuade, Chem. Rev.
2007, 107, 2300; f) A. Corma, H. Garcꢀa, Top. Catal.
2008, 48, 8.
[11] a) A. Vidal-Ferran, N. Bampos, A. Moyano, M. A. Peri-
cꢁs, A. Riera, J. K. M. Sanders, J. Org. Chem. 1998, 63,
6309; b) M. A. Pericꢁs, D. Castellnou, I. Rodrꢀguez, A.
Riera, L. Solꢁ, Adv. Synth. Catal. 2003, 345, 1305;
c) J. M. Fraile, J. A. Mayoral, J. Serrano, M. A. Pericꢁs,
L. Solꢁ, D. Castellnou, Org. Lett. 2003, 5, 4333; d) D.
Castellnou, M. Fontes, C. Jimeno, D. Font, L. Solꢁ, X.
Verdaguer, M. A. Pericꢁs, Tetrahedron 2005, 61, 12111;
e) D. Castellnou, L. Solꢁ, C. Jimeno, J. M. Fraile, J. A.
Mayoral, A. Riera, M. A. Pericꢁs, J. Org. Chem. 2005,
70, 433; f) A. Bastero, D. Font, M. A. Pericꢁs, J. Org.
Chem. 2007, 72, 2460; g) M. A. Pericꢁs, C. I. Herrerꢀas,
L. Solꢁ, Adv. Synth. Catal. 2008, 350, 927; h) F. Micha-
lek, A. Lagunas, C. Jimeno, M. A. Pericꢁs, J. Mater.
Chem. 2008, 18, 4692.
[17] M. Pastꢃ, B. Rodrꢀguez, A. Riera, M. A. Pericꢁs, Tetra-
hedron Lett. 2003, 44, 8369.
[18] M. A. Pericꢁs, C. Puigjaner, A. Riera, A. Vidal-Ferran,
M. Gꢃmez, F. Jimꢇnez, G. Muller, M. Rocamora,
Chem. Eur. J. 2002, 8, 4164.
[19] a) K. Harju, M. Vahermo, I. Mutikainen, J. Yli-Kauha-
luoma, J. Comb. Chem. 2003, 5, 826; b) S. Lçber, P.
Gmeiner, Tetrahedron 2004, 60, 8699.
[20] T. Ohta, H. Sasayama, O. Nakajima, N. Kurahashi, T.
Fujii, I. Furukawa, Tetrahedron: Asymmetry 2003, 14,
537.
[21] a) U. Burckhardt, M. Baumann, A. Togni, Tetrahedron:
Asymmetry 1997, 8, 155; b) L. Hintermann, F. Lang, P.
Maire, A. Togni, Eur. J. Inorg. Chem. 2006, 1397.
[22] a) L. D. Schmidt, The Engineering of Chemical Reac-
tions, Oxford University Press, New York, 1998; b) O.
Levenspiel, Chemical Reaction Engineering, Wiley,
New York, 1998.
[23] Visual inspection of the system when a dichlorome-
thane solution of highly fluorescent coumarin 6 [3-(2-
benzothiazolyl)-7-(diethylamino)coumarin; CAS num-
AHCTUNGTREGbNNNU er 38215–36–0] is circulated through the tubular reac-
tion in the dark under ultraviolet irradiation (l=
365 nm) shows a flat and uniform liquid front.
Adv. Synth. Catal. 2009, 351, 1539 – 1556
ꢅ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1555

Dana Popa et al.
FULL PAPERS
[24] Residence time of the reaction mixture in the resin-
filled reactor was estimated by circulating a dichloro-
methane solution of coumarin 6 through the column
loaded with previously swollen resin 9f in the dark
under ultraviolet irradiation (l=365 nm), and measur-
ing the times when fluorescence is first observed in the
section of the reactor occupied by the resin and when
fluorescence first appears after this section of the reac-
tor.
[25] a) C. O. Kappe, A. Stadler, Microwaves in Organic and
Medicinal Chemistry, Wiley-VCH, Weinheim, 2005;
b) T. Razzaq, J. M. Kremsner, C. O. Kappe, J. Org.
Chem. 2008, 73, 6321.
[26] P. A. Auburn, P. B. Mackenzie, B. Bosnich, J. Am.
Chem. Soc. 1985, 107, 2033.
[27] D. A. Evans, K. R. Campos, J. S. Tedrow, F. E. Michael,
M. R. Gagnꢇ, J. Am. Chem. Soc. 2000, 122, 7905.
[28] D. Zhao, K. Ding, Org. Lett. 2003, 5, 1349.
1556
asc.wiley-vch.de
ꢅ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 1539 – 1556