Solid-supported gallium triflate: An efficient catalyst for the three-component ketonic strecker reaction

Article abstract of DOI:10.1002/cssc.201100370

In light of the growing interest in the use of rare earth metal triflates as water-tolerant Lewis acid catalysts, we embarked upon the development of a solid-supported gallium triflate (PS-Ga(OTf)₂) derivative as a means of increasing the clean

Full text of DOI:10.1002/cssc.201100370

DOI: 10.1002/cssc.201100370
Solid-Supported Gallium Triflate: An Efficient Catalyst for
the Three-Component Ketonic Strecker Reaction
Charlotte Wiles*[a] and Paul Watts[b]
In light of the growing interest in the use of rare earth metal
triflates as water-tolerant Lewis acid catalysts, we embarked
upon the development of a solid-supported gallium triflate
(PS-Ga(OTf)2) derivative as a means of increasing the cleanli-
ness and cost effectiveness of using these increasingly expen-
sive catalytic materials in synthetic processes. Having previous-
ly highlighted the advantages associated with coupling solid-
supported catalysis and the emerging area of micro-reaction
technology, we screened PS-Ga(OTf)2 for activity towards the
ketonic Strecker reaction, in which the target a-aminonitriles
were obtained in higher yield and purity compared to reac-
tions reported in literature, in which the analogous homogene-
ous catalyst was used.
Introduction
Solid-supported catalysis under flow
Micro reaction technology (MRT) offers the synthetic chemist a
new method of executing chemical reactions, which in addi-
tion to obtaining enhanced reaction control, affords the user a
means of transferring methodology developed within the labo-
ratory to production without the need for lengthy re-optimiza-
tion.[1] This approach enables the conditions identified within
the R&D laboratory to be harnessed within a production envi-
ronment. This not only reduces the time taken to scale up a
process, but also removes the risks conventionally associated
with increasing production volume.
Compared to solution-phase catalysis, the immobilization of
catalysts onto inert supports can be advantageous, affording
ease of product isolation and recycling, which leads to in-
creased efficiency, cost effectiveness, and reduced waste gen-
eration. Immobilization of catalysts can, however, lead to ex-
tended reaction times compared to their homogeneous coun-
terparts, a feature that is frequently overcome by the use of
increased reactant stoichiometries, which drives the reactions
to completion. Stirring or shaking the catalytic material can
also lead to mechanical degradation of the support, which
makes recovery and re-use inefficient especially on small
scales.
Based on our experience of continuous-flow reactors[1,10] we
proposed that the disadvantages could be overcome by cou-
pling heterogeneous catalysis with micro-reaction technology
(Figure 1), whereby the catalytic material is retained within a
small packed bed (Labtrix device 3026). Using this approach,
the issue of mechanical degradation is removed, which leads
to increased catalyst lifetimes and reaction efficiency compared
to batch reactions, thus affording a more sustainable process.
In addition, recycling is easier as filtration of the reaction prod-
uct is no longer required and, therefore, no loss of material
arises between reactions; reduced leaching of the catalyst is
also observed, which can be problematic in batch reactions.[11]
Furthermore, reaction times are reduced owing to the high
surface-to-volume ratio obtained within such systems; a high
catalyst-to-substrate ratio is obtained whilst only milligram
Based on the findings of initial research programs, which
served to illustrate the practical advantages associated with
MRT, the field of continuous-flow synthesis has grown, with
many research groups now involved in the exploration of this
methodology. Recent examples include the synthesis of biolog-
ically relevant molecules, such as efaproxiral,[2] pristane,[3] and
ibuprofen,[4] with DSM demonstrating the use of glass reactors
for the large-scale production of pharmaceutically relevant ma-
terials, such as naproxcinod.[5]
Several interesting molecules have been prepared by using
this technology; however, to continue to increase product
complexity generated in this way, it is imperative that multiple
reaction steps are combined in single continuous processes,
rather than the recent trend, where the vast majority of reac-
tions are conducted via a series of single steps, with the reac-
tion products from each step purified off-line.[6] To enable reali-
zation of this, researchers have developed methodologies that
allow processes, such as in-line separations[7] and micro-distilla-
tions,[8] to be performed in series. The use of solid-supported
reagents, catalysts, and scavengers is another approach that af-
fords a facile method of increasing the number of transforma-
tions possible within microreactors and enables the synthesis
of analytically pure materials without the need to perform
formal purifications between reaction steps.[9]
[a] Dr. C. Wiles
Chemtrix BV, Burgemeester Lemmensstraat 358
6163JT Geleen (The Netherlands)
Fax: (+31)464759416
E-mail: c.wiles@chemtrix.com
[b] Dr. P. Watts
Department of Chemistry, University of Hull
Cottingham Road, Hull, HU6 7RX (United Kingdom)
332
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2012, 5, 332 – 338

Solid-Supported Gallium Triflate Catalyst for the Strecker Reaction
Scheme 1. Generalization of the multi-component Strecker reaction and the
synthetic versatility of the resulting a-aminonitrile.
We found the technique to be chemoselective, which ena-
bled the synthesis of aldehydic Strecker products in the pres-
ence of ketonic functionalities. The investigation culminated in
the chemoselective synthesis of 2-(4-acetylphenyl)-2-(phen-
ethylamino)acetonitrile (2) obtained in 99.8% yield and quanti-
tative purity (Scheme 2). During this investigation, we were
Figure 1. The Labtrix Start equipment (top) and the glass microreactor used
for the evaluation of PS-Ga(OTf)2 under continuous flow (bottom).
quantities of catalytic material are used. The inherently high
level of reaction control obtained also enables the user to
obtain previously inaccessible reaction selectivities, as demon-
strated in the oxidation of alcohols[12] and the thioacetalisation
of 4-acetylbenzaldehyde.[13]
Incorporating such materials into multi-step flow processes
has to-date been limited. This often consists of solution-phase
reactions that are performed in separate modules to polymer-
assisted reaction steps, and although techniques exist to per-
form reactions in series, the use of multiple interconnections
result in regions of large dead volume.[14]
Scheme 2. The sole reaction product obtained from the Strecker reaction of
4-acetylbenzaldehyde (3).
Strecker reaction
surprised by the lack of reaction observed for ketonic sub-
strates, with the starting materials recovered quantitatively
even at elevated reaction temperatures (508C) and prolonged
reaction times (2 h).
Building on our experience within the field of MRT, we recently
developed and evaluated an integrated borosilicate-glass
micro-reactor in which liquid–liquid and liquid–solid reaction
steps were performed in series in a single integrated reactor,
which reduced the dead volume between reaction steps dra-
matically and increased reaction complexity.
After this work was performed, the microreactor was com-
mercialized, which enabled the evaluation of solid-supported
reagents and catalysts at high temperature (À15–1958C) and
pressure (2 MPa, Figure 1). With this equipment in hand, we
tried to identify a solid-supported Lewis acid catalyst capable
of performing the ketonic Strecker reaction under continuous
flow.
Using this approach, more than 50 analytically pure a-
aminonitriles were synthesized, using the aldehydic Strecker
reaction (Scheme 1), with isolated yields in excess of 99.6% ob-
tained by using polymer-supported scandium triflate
[PS-Sc(OTf)2] (1) as the catalyst. In addition, the unique reaction
environment obtained within such micro-channel networks en-
abled us to rapidly evaluate the reaction mechanism by con-
trolling the order of reactant addition, preventing the forma-
tion of byproducts commonly associated with the Strecker
reaction (typically cyanohydrins and tetramethylsilane cyano-
hydrins).[15]
Ketonic Strecker reaction
The Strecker reaction is often referred to as one of the most
important multi-component reactions[16,17] owing to the ease
with which the reaction products can be converted into syn-
thetically useful diamines, a-amino acids (Scheme 1),[18] den-
ChemSusChem 2012, 5, 332 – 338
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
333

C. Wiles and P. Watts
drimers,[19] and substituted hydantoins.[20] Utilizing convention-
al batch reaction methodology, the ketonic Strecker reaction
has been performed by using catalysts such as bis(dialkylami-
no)cyanoborane,[21] palladium(II) complexes,[22] Fe(Cp)2PF6,[23]
rare earth metal triflates (RE(OTf)3),[24] and zinc halides.[25] How-
ever, as reaction times are long (24–72 h), large quantities of
the cyanide source are used to drive the reaction, thus making
work-up and isolation of the a-aminonitrile tedious, with some
examples even found to require high pressures to complete
the transformation.[26] In all cases, efficient recovery and re-use
of the catalytic material is difficult and, therefore, this catalysts
are not suitable to be used on a large scale and, unlike its alde-
hydic counterpart, no examples of the aromatic ketonic Stre-
cker reaction have been reported to be promoted by a solid-
supported catalyst.
Results and Discussion
To investigate the activity of 4 towards the ketonic Strecker
reaction, a D263T glass microreactor (Chemtrix BV, The Nether-
lands) was fabricated (Footprint=22.5 mmꢁ45.0 mmꢁ
2.0 mm), which consisted of a T-intersection, where two solu-
tions could be mixed [300 mm(wide)ꢁ60 mm(deep); volume=
5 mL] prior to entering the packed bed [2.5 mm(wide)ꢁ
3.3 cm(long)ꢁ600 mm (deep)].
Dichloromethane (DCM) was selected as the reaction sol-
vent, due to increased substrate solubility, ease of solvent re-
moval upon completion of the reaction, and a literature prece-
dent with respect to rare-earth-metal-triflate catalysis.
By using the reaction manifold previously described, 4
(0.01 g, 1.1ꢁ10À2 mmol Ga) was packed into the reactor and a
pre-mixed solution of 4-methylacetophenone (7, 0.4m) and
aniline (8, 0.4m) in DCM was introduced into the reactor from
inlet A, and a solution of trimethylsilylcyanide (TMSCN, 9, 0.4m
in DCM) was introduced into the reactor from inlet B. The reac-
tants were mixed, but did not react, prior to entering the
packed bed, upon which the ketoimine intermediate formed
and subsequently underwent nucleophilic addition of the cya-
nide anion to afford the respective a-aminonitrile, 2-(phenyl-
amino)-2-p-tolylpropanenitrile (10). The reaction products were
collected over a period of 1 h prior to evaporation of the reac-
tion solvent and dissolved in CDCl3. The reaction products
were subsequently analyzed by using 1H and 13C NMR spec-
troscopy, with the product conversion determined by compari-
son of the integrals obtained for the methyl signals in the
As a means of improving catalyst recovery and reducing the
reaction time required, we investigated the preparation of a
solid-supported Lewis acid catalyst capable of performing the
ketonic Strecker reaction. Based on our previous success with
(1)[15b,c] and the homogeneous investigation of Olah and co-
workers,[24] we evaluated the preparation of polymer-support-
ed gallium(III) bis(trifluoromethanesulfonate) [PS-Ga(OTf)2, 4,
À1
Scheme 3], affording a loading of 1.10 mmol
g
[determined
Ga
starting material
7 (d=2.31 ppm) and product 10 (d=
1.95 ppm) respectively.
The effect of reactant flow rate was initially evaluated, with
a total flow rate of 20 mLminÀ1, affording 25.6% conversion to
the target a-aminonitrile, with the residual being unreacted
starting materials 7 and 8 (Table 1). Reducing the flow rate,
Scheme 3. Reaction protocol used to prepare polymer-supported gallium(III)
bis(trifluoromethanesulfonate) (4) through the reaction of Amberlyst-15 (5)
with gallium(III)chloride (6).
Table 1. Results obtained for the optimization of 2-(phenylamino)-2-p-tol-
ylpropane-nitrile (10) under continuous flow.
Flow rate
T
[8C]
Conversion
[%]
Throughput
Flow rate
[mLminÀ1
]
[a]
[mLminÀ1
]
[mghÀ1
]
by using acid digestion and inductively coupled plasma mass
spectrometry (ICP–MS) analysis of the filtrate]. The activity of
the materials towards the ketonic Strecker reaction was subse-
quently evaluated under continuous flow, utilizing the micro-
reactor depicted in Figure 1 and the model reaction illustrated
in Scheme 4.
20[b]
10
5
1
20[b]
20[b]
20[b]
RT
RT
RT
RT
30
40
50
25.6
40.8
52.3
89.1
76.2
13.6
10.8
7.0
20[b]
10
5
1
20[b]
20[b]
20[b]
2.4
43.1
56.6
56.6
100.0
100.0
[a] After product purification. [b] Residence time=1 min.
and hence increasing the reactant residence time, was found
to increase the conversion of 7 to 10 (89.1%); however, this re-
sulted in a reduction in reactor productivity, affording only
2.4 mghÀ1 after purification.
To increase conversion and productivity, the effect of reactor
temperature was investigated by placing the microreactor
onto the Labtrix Start thermal control unit (Figure 1). To keep
Scheme 4. Model reaction used to evaluate the activity of polymer-support-
ed gallium(III) bis(trifluoromethanesulfonate) (4) towards the ketonic Stre-
cker reaction.
334
www.chemsuschem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2012, 5, 332 – 338

Solid-Supported Gallium Triflate Catalyst for the Strecker Reaction
the reactants and solvent in the liquid phase, a backpressure
regulator (2 MPa) was fitted to the reactor holder outlet, which
enabled controlled liquid handling within the microreactor
above the boiling point of the solvent (388C). By using this ap-
proach, a reactor temperature of 308C and a flow rate of 20 mL
minÀ1 resulted in an increase of conversion to 76.2%, which
corresponds to a throughput of 43.1 mghÀ1. Increasing the re-
actor temperature (40–508C) afforded quantitative conversion
electron-withdrawing and donating aromatic ketones along
with aliphatic precursors. Interestingly, compared to the homo-
geneous-batch investigation performed by Olah and co-work-
ers,[24] we were able to obtain the target a-aminonitriles in
higher yield and increased purity by using this combined ap-
proach, whilst reducing the proportion of 9 used from 3 to
1 equivalents.
of
7 to 10 and a 23.6 fold increase in productivity
(56.6 mghÀ1) compared with reactions performed at room
temperature. In addition, complete conversion of starting ma-
terials 7 and 8 was observed, implying that upon evaporation
of the reaction solvent, 10 was obtained in analytical purity
with no need for subsequent purification.
Aldehydic Strecker reaction
In an extension to our initial investigation, we subsequently
evaluated the activity of 4 towards the aldehydic Strecker reac-
tion. DCM was used as the reaction solvent and in contrast to
previous investigations utilizing 1, the effect of reactor temper-
ature was investigated (25–408C). By using the same aliquot of
4 (0.01 g, 1.1ꢁ10À2 mmol Ga), a solution of 4-bromobenzalde-
hyde (11 0.2m) and 8 (0.2m) was introduced into the reactor
from inlet A, and a solution of 9 (0.2m) was then introduced
into the reactor from inlet B, where it was mixed with the ald-
imine prior to reaction in the presence of catalyst 4. Again, the
reaction progress was determined by removal of the reaction
solvent and dissolution of the reaction products in CDCl3 prior
Having confirmed the ability to efficiently transform an aro-
matic ketone into the respective a-aminonitrile 10 by using
the Lewis acid catalyst 4, the reaction products were assessed
by using ICP–MS to quantify the proportion of Ga within 10.
As with previous flow reactions, no Ga was detected above the
instruments detection limit [the limit of detection (LOD) was
determined to be 8.8 ppb for Ga], which confirmed the long-
term mechanical stability of catalyst 4 compared to the batch
reaction, for which 400 ppm of Ga was detected. With regard
to catalyst activity, the same aliquot of material 4 remained
active over a one-month period of use without showing any
signs of degradation.
1
to using H and 13C NMR spectroscopy.
By using this approach, quantitative conversion to 2-(4-bro-
mophenyl)-2-(phenylamino)acetonitrile was obtained at a total
flow rate of 20 mLminÀ1 (34.4 mghÀ1) when the reaction was
performed at room temperature. Increasing the reactor tem-
perature to 408C enabled the reaction time to be reduced to
30 s, affording an increase in throughput to 68.9 mghÀ1. By
using sequential reactant addition, cyanohydrin formation was
again suppressed, and 2-(4-bromophenyl)-2-(phenylamino)ace-
tonitrile was obtained in analytical purity (99.9% yield). Based
on this observation, a library of five compounds was prepared
to illustrate the increased throughput obtained by using 4.
As illustrated in Table 3, increased yields were obtained in all
cases as a result of using a solid-supported Lewis acid within a
microreactor compared to batch reactions, with throughputs
doubled for 4 compared to 1.
To confirm that the transformation could be attributed to 4,
the flow reaction was performed in the presence of Amberlyst-
15, (5), the material used as the solid support, and subsequent-
ly 2% cross-linked polystyrene beads; in all cases, no reaction
was observed, and recovery of the starting materials 7 and 8
1
monitored by using H NMR spectroscopy. In addition, the sta-
bility of the pre-mixed ketone 7 and amine 8 stock solution
1
was validated by using H NMR spectroscopy after 1 and 24 h,
whereby no ketoimine formation was observed.
Having demonstrated the ability to synthesize 10 within a
packed-bed reactor, affording increased product purity com-
pared to batch, we evaluated the reactivity of a series of ke-
tones with 8. As is illustrated in Table 2, high yields were ob-
tained under the optimized reaction conditions for a range of
Table 2. Results obtained for the evaluation of PS-Ga(OTf)2 (4) as a Lewis
acid catalyst for the ketonic Strecker reaction performed under flow
conditions.
Table 3. Results obtained for the evaluation of PS-Ga(OTf)2 (4) as a Lewis-
acid catalyst for the aldehydic Strecker reaction performed under contin-
uous flow conditions.
[a]
[a]
Ketone
Yield [%]
Throughput [mghÀ1
]
Aldehyde[a]
Yield [%]
Throughput [mghÀ1
]
4-methylacetophenone (7)
acetophenone (12)
4-bromoacetophenone (13)
4-nitroacetophenone (14)
cyclohexanone (15)
99.6 (78)[b]
99.8 (98)[b]
99.5 (95)[b]
99.7
56.6
53.3
72.3
67.4
48.0
benzaldehyde (16)
99.7 (90)[b]
99.9
99.8 (90)[b]
99.8 (88)[b]
99.9
49.9 (24.9)[c]
68.9 (34.4)[c]
58.3 (28.9)[c]
54.3
4-bromobenzaldehyde (11)
4-chlorobenzaldehyde (17)
4-fluorobenzaldehyde (18)
methyl-4-formylbenzoate (19)
99.8 (85)[b]
63.9 (31.8)[c]
[a] Flow Rate=20 mLminÀ1 and temperature=408C. [b] The number in
parentheses represents the isolated yield obtained by Olah et al.[24] in
4–6 h.
[a] Flow Rate=20 mL minÀ1 and temperature=408C. [b] The number in
parentheses represents the isolated yield obtained by Olah et al.[24] in 4–
6 h. [c] Throughput obtained by using PS-Sc(OTf)2 (1).
ChemSusChem 2012, 5, 332 – 338
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
335

C. Wiles and P. Watts
outlet, secured with 6–32 PEEK nuts (Upchurch Scientific, USA);
previously, epoxy resin was used for prototyping purposes.[15a,b]
Reagents were delivered to the reactor by using displacement
pumps (F-200, Chemyx, USA), capable of delivering three solutions
at flow rates between 0.1–100 mL minÀ1. A maximum flow rate of
25 mLminÀ1 per feed is recommended to keep the pressure drop
generated in the system low. Inlets A and B were fed by using
1000 mL gas-tight syringes (SGE, USA), and the reaction products
collected at the outlet of the ultra-low, dead-volume, back pressure
regulator (Upchurch Scientific, USA), which was set to 2 MPa by
using a hand-held pressure gauge (Omega, DPG 120). Prior to use,
the catalyst was graded by using metal sieves to remove particu-
lates that were <45 mm and >75 mm to enable efficient and con-
trolled packing of the microreactor.
Conclusions
Through the preparation of a novel, non-hydrolysable solid-
supported analogue of Ga(OTf)3, we were able to develop a
versatile and efficient continuous-flow technique for the syn-
thesis of a-aminonitriles by utilizing stoichiometric quantities
of the cyanide source. Incorporation of catalyst 4 into a flow
reactor enabled a dramatic reduction in reaction time from 8 h
to 1 min for ketones and 30 sec for aldehydes, with yields in
excess of 99% for all substrates evaluated (0.24 mmolgÀ1 hÀ1).
As the catalytic material 4 is confined within a micro channel,
no loss of catalyst occurs between reactions, and ICP–MS anal-
ysis of the reaction products confirmed no detectable leaching
of Ga from catalyst 4 over the course of the investigation. This
represents a significant advantage over conventional homoge-
neous systems (400 ppm in a standard batch reaction).
Preparation of polymer-supported gallium(III) bis(trifluoro-
methanesulfonate) (4)
We have, therefore, developed a continuous-flow method
suitable for the preparation of a-aminonitriles, derived from
both aromatic and aliphatic aldehydes and ketones, by utiliz-
ing mild reaction conditions.
By using Amberlyst-15 (5) as the solid-support (Scheme 3) was pre-
pared by using the following procedure. A-15 (1.00 g, 4.2 mmol
SO3H gÀ1) was added to a stirred solution of GaCl3 (1.85 g,
10.5 mmol) in DCM (10 mL). After 24 h at room temperature, the
solid-supported material was isolated under vacuum filtration and
was washed with DCM (3ꢁ50 mL), EtOH (1ꢁ50 mL), deionized H2O
(1ꢁ50 mL), EtOH (1ꢁ50 mL), and DCM (1ꢁ50 mL). The solid-sup-
ported gallium(III) chloride (6, 1.00 g) was subsequently wetted
with DCM (10 mL) prior to the addition of trifluoromethanesulfonic
acid (6 equiv) under N2. The reaction mixture was stirred at room
temperature for 48 h prior to isolation of 4 under filtration. The
material 4 was washed with DCM (1ꢁ50 mL), EtOH (1ꢁ50 mL), de-
ionized H2O (1ꢁ50 mL), EtOH (1ꢁ50 mL), and DCM (1ꢁ50 mL),
which was followed by oven drying at 908C. By using a pestle and
mortar, the catalyst was ground into a fine powder and sieved
(Endcotts, UK) to afford a particle size distribution of 45–75 mm.
The resulting 4 was characterized by using ICP–MS; by using acid
digestion of the solid-supported Àc1atalyst 4 and analysis of the fil-
trate, a loading of 1.10 mmolGa g was confirmed. Catalyst 4 was
stored in a screw-top amber vessel and found to remain active for
3 years (to-date).
In addition to improvements in catalyst recyclability and
stability, the use of a micro-flow reactor is advantageous, as it
enables the detailed screening of small quantities of catalytic
material when compared to conventional macro packed-bed
reactors. It is, however, acknowledged that should novel reac-
tions be identified by using a microreactor, efficient and pre-
dictable scale-up techniques are required to realize the synthe-
sis of larger quantities of products. Future work will focus on
the development of meso-reactors suitable for the perfor-
mance of heterogeneously-catalyzed reactions, initially at the
gram to kilogram scale and subsequently aimed at industrial
production volumes.
Experimental Section
Materials
All solvents were purchased as puriss grade (>99.5%) over molec-
ular sieves (H2O<0.005%) from Fluka (Gillingham, UK) and, unless
otherwise stated, chemicals were purchased from Sigma–Aldrich
(Gillingham, UK) and used as-received. Prior to use, Amberlyst-15
(5) was ground and sieved to afford a particle size distribution of
45–75 mm (Endcotts, UK).
Filling the microreactor
Prior to filling, the device was cleaned with acetone and oven
dried (908C) to remove any residual solvent. Catalyst 4 was sieved,
and the graded material packed into the reactor through the filling
hole (Figure 1, bottom). The reactor was inverted and tapped to
settle the material and to ensure no voids were present. The reac-
tor was then laid flat on a bench and tapped to settle the catalyst
around the filling hole. Before placing the plug into the filling hole,
the surface of the reactor was wiped with a piece of tissue wetted
with DCM to remove any particulates. As the outlet channel is shal-
low (60 mm) compared to the packed-bed (600 mm) the catalyst is
physically retained in the device once the plug is installed. Care
must be taken not to overfill the hole, leaving space for the plug
to be fitted. The cover plate was then fitted, and the thumbscrews
tightened evenly to afford a liquid tight seal.
Instrumentation
1H and 13C NMR spectra were obtained at room temperature as so-
lutions in deuteriochloroform (CDCl3), using tetramethylsilane
(TMS) as an internal standard, or deuterated acetonitrile (CD3CN).
The spectra were recorded by using a Jeol GX400 spectrometer,
and all spectral data of previously reported compounds were con-
sistent with the literature. Inductively coupled plasma mass spec-
trometry (ICP–MS) measurements were performed by using a
Perkin–Elmer Optima 5300DV instrument.
Micro reactions were performed by using the Labtrix Start micro-
reactor development apparatus (Chemtrix BV, The Netherlands) il-
lustrated in Figure 1. Fluidic connections to the microreactor (3026,
Figure 1, bottom) were made by using polyether ether ketone
(PEEK) tubing (1/32“ o.d.ꢁ90 mm i.d.ꢁ10 cm) for the inlets and
Emptying the microreactor
To remove catalyst 4 from the reactor, the system was flushed with
DCM (2 mL). The reactor was removed from the holder, and the
solvent was allowed to evaporate. Once the catalyst was dry, the
336
www.chemsuschem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2012, 5, 332 – 338

Solid-Supported Gallium Triflate Catalyst for the Strecker Reaction
reactor was inverted, and the catalyst tapped gently out of the fill-
ing hole. The reactor was then returned to the reactor holder, with-
out the plug and DCM pumped through the packed-bed to
remove any remaining particulates. The reactor was removed from
the holder, the solvent allowed to evaporate in a fume cupboard,
and then the reactor was stored for subsequent investigations.
408C, the compound was obtained as a colorless crystalline solid
(72.3 mg, 99.5%). 1H NMR (400 MHz, CDCl3/TMS): d=1.93 (s, 3H;
CH3), 4.27 (s, 1H; NH), 6.52 (d, J=7.8 Hz, 2H; ArH), 6.83 (t, J=
7.8 Hz, 1H; ArH), 7.13 (d, J=7.8 Hz, 2H; ArH), 7.51–7.55 ppm (m,
4H; ArH); 13C NMR (100 MHz, CDCl3/TMS): d=33.4 (CH3), 56.8 (C0),
115.9 (2CH), 120.3 (CN), 120.4 (CH), 122.7 (C0), 126.8 (2CH), 129.2
(2CH), 132.5 (2CH), 139.2 (C0), 143.2 ppm (C0).
2-(4-Nitrophenyl)-2-(phenylamino)propanenitrile (Entry 4, Table 2):
By using 4-nitroacetophenone (14) and 8 as reactants, along with
a total flow rate of 20 mLminÀ1 and a reactor temperature of 408C,
the compound was obtained as a pale yellow solid (67.4 mg,
General ketonic Strecker flow procedure
Gallium(III) bis(trifluoromethanesulfonate) (PS-Ga(OTf)2, 4) was dry
packed into the packed-bed of a 3026 microreactor (0.01 g, 1.1ꢁ
10À2 mmol Ga, Figure 2) and a solution of ketone (0.4m); amine
1
99.7%). H NMR (400 MHz, CDCl3/TMS): d=1.98 (s, 3H; CH3), 4.38
(s, 1H; NH), 6.49 (d, J=7.8 Hz, 2H; ArH), 6.83–6.87 (m, 1H; ArH),
7.14 (t, J=7.8 Hz, 2H; ArH), 7.83 (d, J=8.8 Hz, 2H; ArH), 8.27 ppm
(d, J=8.8 Hz, 2H; ArH); 13C NMR (100 MHz, CDCl3/TMS): d=33.2
(CH3), 56.8 (C0), 115.8 (2CH), 119.7 (CN), 120.8 (CH), 124.7 (2CH),
126.3 (2CH), 129.3 (2CH), 142.7 (C0), 147.1 (C0), 148.2 ppm (C0NO2).
1-(Phenylamino)cyclohexanecarbonitrile (Entry 5, Table 2): By using
cyclohexanone (15) and 8 as reactants, along with a total flow rate
of 20 mLminÀ1 and a reactor temperature of 408C, the compound
was obtained as
a
colorless oil (48.0 mg, 99.8%). 1H NMR
(400 MHz, CDCl3/TMS): d=1.27–1.38 (m, 3H; CH2 and CH), 1.59–
1.83 (m, 7H; 3CH2 and CH), 4.01 (s, 1H; NH), 6.91–7.10 (m, 3H;
ArH), 7.21–7.40 ppm (m, 2H; CH); 13C NMR (100 MHz, CDCl3/TMS):
d=22.1 (2CH2), 24.9 (CH2), 35.9 (2CH2), 53.1 (C0), 113.7 (2CH), 118.2
(CN), 120.2 (CH), 129.8 (2CH), 147.3 ppm (C0N).
Figure 2. The reaction manifold (Labtrix 3026) used to investigate the keton-
ic Strecker reaction under continuous flow.
(0.4m in DCM) was pumped into the reactor from inlet A and a so-
lution of trimethylsilylcyanide (9, 0.4m in DCM) from inlet B to
afford a 1:1:1 ratio and a final concentration of 0.2m. The micro-
reactor was placed onto the Labtrix Start thermal control unit,
where it was heated to 408C by using a Peltier element and the re-
action products collected into a pre-weighed sample tube for a
period of 1 h. The reaction products were concentrated in vacuo
General aldehydic Strecker flow procedure
Catalyst 4 was dry-packed into a packed-bed reactor (0.01 g, 1.1ꢁ
10À2 mmol Ga, (Figure 2); a solution of aldehyde and amine (0.2m)
was introduced from inlet A and a solution of 9 (0.2m in DCM)
from inlet B to afford a 1:1:1 ratio and a final concentration of
0.1m. The microreactor was placed within the Labtrix Start unit
and heated to 408C. The reaction products were collected into a
pre-weighed sample tube for a period of 1 h prior to concentrating
in vacuo. Dissolution of the product in CDCl3 enabled analysis of
1
prior to analysis as solutions in CDCl3 or CD3CN by H and 13C NMR
spectroscopy. The synthesis of the following compounds was de-
scribed in Ref. [24].
2-(Phenylamino)-2-p-tolylpropanenitrile (Entry 1, Table 2): By using
4-methylacetophenone (7) and aniline (8) as reactants, along with
a total flow rate of 20 mLminÀ1 and a reactor temperature of 408C,
the compound was obtained as a colorless, crystalline solid
1
the crude reaction product by using H and 13C NMR spectroscopy.
The synthesis of the following compounds (unless marked other-
wise) was described in Ref. [15b].
1
(56.6 mg, 99.6%). H NMR (400 MHz, CD3CN): d=1.84 (3H, s, CH3),
2-Phenyl-2-(phenylamino)acetonitrile (Entry 1, Table 3): By using
benzaldehyde (16) and 8 as reactants, along with a total flow rate
of 40 mLminÀ1 and a reactor temperature of 408C, the compound
was obtained as a colorless crystalline solid (49.9 mg, 99.7%).
1H NMR (400 MHz, CDCl3/TMS): d=4.38 (s, 1H; NH), 5.36 (s, 1H;
CH), 6.73 (dd, J=7.6 and 1.2 Hz, 2H; ArH), 6.86 (t, J=7.6 Hz, 1H;
ArH), 7.23 (dd, J=7.6 and 1.2 Hz, 2H; ArH), 7.40–7.52 (m, 3H; ArH),
7.54–7.58 ppm (m, 2H; ArH); 13C NMR (100 MHz, CDCl3/TMS): d=
50.0 (CH), 114.1 (2CH), 115.2 (CN), 120.1 (CH), 127.1 (2CH), 128.3
(CH), 129.2 (2CH), 129.5 (2CH), 133.8 (C0) and 144.6 ppm (C0N).
1.91 (s, 3H; CH3), 4.22 (s, 1H; NH), 6.46 (d, J=7.8 Hz, 2H; 2ArH),
6.70 (t, J=7.8 Hz, 1H; ArH), 7.01–7.13 (m, 4H; ArH), 7.42 ppm (d,
J=7.8 Hz, 2H; ArH); 13C NMR (100 MHz, CDCl3/TMS): d=21.1 (CH3),
33.0 (CH3), 56.8 (C0), 115.4 (2CH), 119.6 (CN), 119.7 (CH), 124.7
(2CH), 128.9 (2CH), 129.8 (2CH), 136.9 (C0), 138.2 (C0), 146.9 ppm
(C0N).
2-Phenyl-2-(phenylamino)propanenitrile (Entry 2, Table 2): By using
acetophenone (12) and 8 as reactants, along with a total flow rate
of 20 mLminÀ1 and a reactor temperature of 408C, the compound
was obtained as
a
colorless oil (53.3 mg, 99.8%). 1H NMR
2-(4-Bromophenyl)-2-(phenylamino)acetonitrile (Entry 2, Table 3): By
using 4-bromobenzaldehyde (11) and 8 as precursors, the microre-
action was performed at a total flow rate of 40 mLminÀ1 and a re-
action temperature of 408C to afford the compound as a colorless
crystalline solid (68.9 mg, 99.9% yield). 1H NMR (400 MHz, CDCl3/
TMS): d=3.65 (s, 1H; NH), 5.32 (d, J=3.7 Hz, 1H; CH), 6.66 (d, J=
8.4 Hz, 2H; ArH), 6.85 (t, J=7.7 Hz, 1H; ArH), 7.20 (t, J=7.7 Hz, 2H;
ArH), 7.36 (d, J=7.7 Hz, 2H; ArH), 7.47 ppm (d, J=8.4 Hz, 2H;
ArH); 13C NMR (100 MHz, CDCl3/TMS): d=49.8 (CH), 112.3 (2CH),
114.9 (CN), 116.9 (CH), 122.6 (C0Br), 128.3 (2CH), 129.1 (C0), 131.3
(2CH), 131.4 (2CH), 143.7 ppm (C0).
(400 MHz, CDCl3/TMS): d=1.95 (s, 3H; CH3), 6.49–6.57 (m, 2H;
ArH), 6.71–6.89 (m, 1H; ArH), 7.05–7.18 (m, 2H; ArH), 7.35–7.45 (m,
3H; ArH), 7.61 ppm (d, J=7.9 Hz, 2H; ArH), NH not observed;
13C NMR (100 MHz, CDCl3/TMS): d=33.0 (CH3), 57.1 (C0), 105.7
(2CH), 120.2 (CH), 121.1 (CN), 125.1 (2CH), 128.3 (CH), 128.7 (2CH),
128.9 (2CH), 130.1 (C0), 147.3 ppm (C0N).
2-(4-Bromophenyl)-2-(phenylamino)propanenitrile (Entry 3, Table 2):
By using 4-bromoacetophenone (13) and 8 as reactants, along
with a total flow rate of 20 mLminÀ1 and a reactor temperature of
ChemSusChem 2012, 5, 332 – 338
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
337

C. Wiles and P. Watts
[3] K. Tanaka, S. Motomatsu, K. Koyama, S. Tanaka, K. Fukase, Org. Lett.
2007, 9, 299–302.
[4] A. R. Bogdan, S. L. Poe, D. C. Kubis, S. J. Broadwater, D. T. McQuade,
Angew. Chem. 2009, 121, 8699–8702; Angew. Chem. Int. Ed. 2009, 48,
8547–8550.
[5] S. Braune, P. Pochlauer, R. Reintjens, S. Steinhofer, M. Winter, O. Lobet,
R. Guidat, P. Woehl, C. Guermeur, Chim. Oggi 2009, 27, 26–29.
[6] L. F. Tietze, D. Liu, Arkivoc 2008, 193–210.
[7] a) H. R. Sahoo, J. G. Kralj, K. F. Jensen, Angew. Chem. 2007, 119, 5806–
5810; Angew. Chem. Int. Ed. 2007, 46, 5704–5708; b) H. Lange, M. J. Car-
penter, A. X. Jones, C. J. Smith, N. Nikbin, I. R. Baxendale, S. V. Ley, Syn-
lett 2011, 869–873; c) L. J. Martin, A. L. Marzinzik, S. V. Ley, I. R. Baxen-
dale, Org. Lett. 2011, 13, 320–323.
2-(4-Chlorophenyl)-2-(phenylamino)acetonitrile (Entry 3, Table 3): By
using 4-chlorobenzaldehyde (17) and 8 as precursors, the microre-
action was conducted at a total flow rate of 40 mLminÀ1 and a re-
actor temperature of 408C to afford the compound as a colorless
crystalline solid (58.3 mg, 99.8%). 1H NMR (400 MHz, CDCl3/TMS):
d=4.04 (s, 1H; NH), 5.42 (s, 1H; CH), 6.75 (dd, J=7.6 and 1.1 Hz,
2H; ArH), 6.92 (dd, J=7.6 and 1.1 Hz, 1H; ArH), 7.27 (dt, J=7.6
and 1.1 Hz, 2H; ArH), 7.42 (d, J=7.2 Hz, 2H; ArH), 7.54 ppm (d, J=
7.2 Hz, 2H; ArH); 13C NMR (100 MHz, CDCl3/TMS): d=49.6 (CH),
114.3 (2CH), 117.8 (CN), 120.6 (CH), 128.5 (4CH), 129.6 (2CH), 132.4
(C0), 135.6 (C0Cl), 144.3 ppm (C0N).
2-(4-Fluorophenyl)-2-(phenylamino)acetonitrile (Entry 4, Table 3):[23]
By using 4-fluorobenzaldeyde (18) and 8 as reactants, the microre-
action was conducted at a total flow rate of 40 mLminÀ1 and a re-
actor temperature of 408C to afford the compound as a colorless
oil (54.3 mg, 99.9%). 1H NMR (400 MHz, CDCl3/TMS): d=4.03 (d,
J=8.4 Hz, 1H; NH), 5.40 (d, J=8.4 Hz, 1H; CH), 6.77–6.82 (m, 2H;
ArH), 6.91–7.14 (m, 1H; ArH), 7.12–7.18 (m, 2H; ArH), 7.24–7.31 (m,
2H; ArH), 7.52–7.63 ppm (m, 2H; ArH); 13C NMR (100 MHz, CDCl3/
TMS): d=49.7 (CH), 114.3 (2CH), 116.3 (d, J 21.4, 2CH), 118.6 (CN),
120.4 (CH), 126.7 (CH), 126.9 (d, J=3.1 Hz, CH), 129.1 (d, J=8.4 Hz,
2CH), 144.5 (C0), 163.4 ppm (d, J=248.5 Hz, C0F).
[8] A. Hibara, K. Toshin, T. Tsukahara, K. Mwawtari, T. Kitamori, Chem. Lett.
2008, 37, 1064–1065.
[9] C. Wiles, P. Watts, S. J. Haswell, Lab Chip 2007, 7, 322–330.
[10] C. Wiles, P. Watts, Eur. J. Org. Chem. 2008, 1655–1671.
[11] I. R. Baxendale, S. V. Ley, Ernst Schering Found. Symp. Proc. 2007, 3, 151–
185.
[12] C. Wiles, P. Watts, S. J. Haswell, Tetrahedron Lett. 2006, 47, 5261–5264.
[13] C. Wiles, P. Watts, S. J. Haswell, Tetrahedron Lett. 2007, 48, 7362–7365.
[14] I. R. Baxendale, J. Deeley, C. M. Griffiths-Jones, S. V. Ley, S. Saaby, G. K.
Tranmer, Chem. Commun. 2006, 2566–2568.
[15] a) C. Wiles, P. Watts, Org. Process Res. Dev. 2008, 12, 1001–1006; b) C.
Wiles, P. Watts, Eur. J. Org. Chem. 2008, 5597–5613; c) C. Wiles, Spec.
Chem. Mag. 2009, 40–41.
Methyl-4-[cyano(phenylamino)methyl]benzoate (Entry 5, Table 3):
By using methyl-4-formyl benzoate (19) and 8 as reactants, the mi-
croreaction was conducted at a total flow rate of 40 mLminÀ1 and a
reactor temperature of 408C to afford the compound as a colorless
oil (63.9 mg, 99.9%). 1H NMR (400 MHz, CDCl3/TMS): d=3.95 (s,
3H; OCH3), 4.13 (d, J=8.4 Hz, 1H; NH), 5.52 (d, J=8.4 Hz, 1H; CH),
6.71–7.12 (m, 3H; ArH), 7.26–7.30 (m, 2H; ArH), 7.71 (d, J=8.3 Hz,
2H; ArH), 8.11 ppm (d, J=8.3 Hz, 2H; ArH); 13C NMR (100 MHz,
CDCl3/TMS): d=50.0 (CH), 52.4 (OCH3), 114.3 (2CH), 118.2 (CN),
120.6 (CH), 127.2 (2CH), 129.6 (2CH), 130.5 (2CH), 139.2 (C0), 144.3
(C0), 150.1 (C0CO2CH3), 168.3 ppm (CO).
[16] A. Strecker, Justus Liebigs Ann. Chem. 1850, 75, 46–51.
[17] D. J. Ramꢂn, M. Yus, Angew. Chem. 2005, 117, 1628–1661; Angew.
Chem. Int. Ed. 2005, 44, 1602–1634.
[18] a) D. Mendel, J. A. Ellman, P. G. Shultz, J. Am. Chem. Soc. 1991, 113,
2758; b) Chemistry and Biochemistry of the Amino Acids (Ed.: G. C. Bar-
rett), Chapman and Hall, London, 1985; c) M. S. Iyer, K. M. Gigstad, N. D.
Namdev, M. Lipton, Amino Acids 1996, 11, 259–268.
[19] M. A. Subhani, K. Mꢃller, F. KoÅ, P. Eilbracht, Org. Biomol. Chem. 2009, 7,
4000–4008.
[20] R. G. Murray, D. M. Whitehead, F. Le Strat, S. J. Conway, Org. Biomol.
Chem. 2008, 6, 988–991.
[21] M. Suginome, A. Yamamoto, Y. Ito, Chem. Commun. 2002, 1392–1393.
[22] J. Jarusiewicz, Y. Choe, K. S. Yoo, C. P. Park, K. W. Jung, J. Org. Chem.
2009, 74, 2873–2876.
[23] N. H. Khan, S. Agrawal, R. I. Kureshy, S. H. R. Abdi, S. Singh, E. Suresh,
R. V. Jasra, Tetrahedron Lett. 2008, 49, 640–644.
Keywords: gallium · heterogeneous catalysis · lewis acids ·
microreactors · strecker reaction
[24] G. K. S. Prakash, T. Mathew, C. Panja, S. Alconcel, H. Vaghoo, C. Do, G. A.
Olah, Proc. Natl. Acad. Sci. USA 2007, 104, 3703–3706.
[25] A. Kazemeini, N. Azizi, M. R. Saidi, Russ. J. Org. Chem. 2006, 42, 48–51.
[26] K. Kumamoto, H. Iida, H. Hamana, H. Kotsuki, K. Matsumoto, Heterocy-
cles 2005, 66, 675–681.
[1] a) C. Wiles, P. Watts, Chem. Commun. 2011, 47, 6512–6535; b) C. Wiles,
P. Watts in Micro reaction technology in organic synthesis, CRC Press,
2011; c) B. P. Mason, K. E. Price, J. L. Steinbacher, A. R. Bogdan, Chem.
Rev. 2007, 107, 2300–2318; d) C. Wiles, P. Watts, Chim. Oggi 2009, 27,
34–36; e) T. Razzaq, C. O. Kappe, Chem. Asian J. 2010, 5, 1274–1289;
f) J. Yoshida, H. Kim, A. Nagaki, ChemSusChem 2011, 4, 331–340.
[2] T. Gustafsson, F. Ponten, P. H. Seeberger, Chem. Commun. 2008, 1100–
1102.
Received: July 14, 2011
Revised: October 26, 2011
Published online on December 8, 2011
338
www.chemsuschem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2012, 5, 332 – 338