Diastereoselective chain-elongation reactions using microreactors for applications in complex molecule assembly

Article abstract of DOI:10.1002/chem.201003148

Diastereoselective chain-elongation reactions are important transformations for the assembly of complex molecular structures, such as those present in polyketide natural products. Here we report new methods for performing crotylation reactions and homopropargylation reactions by using newly developed low-temperature flow-chemistry technology. In-line purification protocols are described, as well as the application of the crotylation protocol in an automated multi-step sequence. Polyketides in flow: By using new low-temperature reactors, the scope of reactions conveniently performed in a flow process has been extended to encompass diastereoselective chain elongations. Various aldehydes undergo Roush crotylations and Marshall homopropargylation reactions smoothly in flow (see scheme), with incorporated in-line work-ups that drastically reduce the amount of time required for these transformations. Copyright

Full text of DOI:10.1002/chem.201003148

DOI: 10.1002/chem.201003148
Diastereoselective Chain-Elongation Reactions Using Microreactors for
Applications in Complex Molecule Assembly
Catherine F. Carter,[a] Heiko Lange,[a] Daiki Sakai,[b] Ian R. Baxendale,[a] and
Steven V. Ley*[a]
Abstract: Diastereoselective chain-elongation reactions are important transforma-
tions for the assembly of complex molecular structures, such as those present in
polyketide natural products. Here we report new methods for performing crotyla-
tion reactions and homopropargylation reactions by using newly developed low-
temperature flow-chemistry technology. In-line purification protocols are de-
scribed, as well as the application of the crotylation protocol in an automated
multi-step sequence.
Keywords: asymmetric synthesis ·
crotylation · flow chemistry · homo-
propargylation · microreactors
Introduction
ral products often display significant biological activity,
these compounds are required in significant quantities, ren-
dering stereoselective methods for their preparation an ap-
propriate target for flow-chemistry applications.
Over the last ten years, our group has made significant ad-
vances in the field of machine-assisted synthesis, in particu-
lar by using flow-chemistry platforms in combination with
polymer-supported reagents.[1] Ultimately, we are aiming to
achieve the multi-step synthesis of complex natural products
and other significantly functionalised molecules using these
methods. We have reported previously on many transforma-
tions that benefit from the elevated temperatures and pres-
sures that can be used safely in flow, particularly for the syn-
thesis of aromatic heterocyclic compounds.[2] As flow
chemistry and new reactor technology develops, the next
step is to adapt stereoselective reactions that require low
temperatures and rigorously dry conditions to flow proto-
cols. This would harness the advantage of superior control
over temperature, particularly for large-scale applications in
which the maintenance of low internal temperatures, that is,
below À208C, can be a significant issue. Recently, we have
reported the use of flow coils immersed in baths containing
dry ice to handle organometallics on scale.[3] Herein, we
report the development of a series of diastereoselective
chain-elongation methods. These reactions have been consis-
tently employed in the synthesis of polyketides. These natu-
Results and Discussion
Roush crotylation: The Roush crotylation uses a family of
enantiopure (Z)- or (E)-boronates, 1 or 2, respectively, to
give access to syn or anti propionate relationships
(Scheme 1).[4] They are particularly effective in reactions
with chiral aldehydes, furnishing complex motifs with high
diastereoselectivity through doubly matched asymmetric re-
actions. Our aim was to convert these reactions into a flow
process, whereby the desired diastereoisomer could be
formed, the alkene converted to the aldehyde and the chain
elongated further with a second crotylation in an iterative
sequence.
Technical details: The following reactions were performed
on a Vapourtec R2+/R4 flow-chemistry platform.[5] A relia-
ble procedure to establish anhydrous flow conditions was an
essential first step in the development of a suitable process.
It was found that flushing the pumps, valves and reaction
coils with isopropyl alcohol (IPA), followed by acetone and
then the dry solvent of choice under an inert atmosphere for
at least two hours was needed to eliminate all traces of
water from the machine (see general procedure). Currently,
there are a number of reactors available to perform low-
temperature reactions in flow (Figure 1). A self-made coil
can be simply immersed in a dry ice bath. Although this is
highly convenient, it can create high backpressures and
places a large degree of strain on the tubing and PEEK con-
nectors. Alternatively, a number of low-temperature reactors
are becoming available on the market. We made use of the
Vapourtec low-temperature coil, which uses a stream of ni-
[a] C. F. Carter, Dr. H. Lange, Dr. I. R. Baxendale, Prof. S. V. Ley
Innovative Technology Centre, Department of Chemistry
University of Cambridge, Lensfield Road
Cambridge, CB2 1EW (UK)
Fax : (+44)1223-336-442
E-mail: svl1000@cam.ac.uk
[b] Dr. D. Sakai
Mitsubishi Tanabe Pharma Corporation
1000, Kamoshida-cho, Aoba-ku, Yokohama 227-0033 (Japan)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201003148. This contains de-
tailed experimental procedures and full experimental data for novel
compounds.
3398
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 3398 – 3405

FULL PAPER
Scheme 2. General set-up for performing Roush crotylation reactions in
flow.
be reacted at higher temperatures with no detriment to the
diastereomeric ratio.
Work-up procedure: There are three alternative work-up
procedures that can be employed for this flow-crotylation
process. For procedures with an “off-line” work-up, the
output stream was directed into a solution of NaOH to hy-
À
drolyse the B O bond present in immediate reaction prod-
uct 7 (Scheme 3), followed by a typical aqueous extraction.
However, in order for a fully continuous flow process to be
realised, an in-line work-up was necessary. This would
enable the reaction to be telescoped as part of a multi-step
sequence with no off-line purification processes being neces-
Scheme 1. Roush crotylation reactions and their stereochemical implica-
tions.
sary.
A number of polymer-supported reagents were
screened to effect the in-line clean up (Scheme 3). Solid-
supported (SS) diol resin and polyol resin IRA-743 were
found to effectively scavenge the boron residues from the
reagent stream in sub-stoichiometric amounts. However, the
resins bind so strongly to boron that (À)-diisopropyl d-tar-
trate (8) was released as a breakdown product, resulting in
product contamination. Nevertheless, this procedure would
be useful for the large-scale process, since ester 8 can be
readily crystallised at À208C and recovered. By comparison,
in the off-line work-up, the chiral ester is hydrolysed and re-
moved by filtration over Celite. As an alternative in-line
work-up, a column was packed with polymer-supported hy-
droxide resin followed by a plug of Celite to mimic the hy-
drolysis process. Taking advantage of a chromatographic
effect caused by the scavenging system, we used in-line
Figure 1. a) PTFE coil and T-piece immersed in a À788C bath. b) Va-
pourtec low-temperature coil. c) Close-up of the low-temperature coil.
trogen flowing over dry ice to establish and maintain low
temperatures.
infrared monitoring[6] to give the product cleanly. A plug of
ACHTUNGTRENNUNG
silica gel in place of Celite was also examined, but resulted
in a very poor dispersion of the product, whereby a reaction
that normally proceeds in 30 min took four hours due to the
additional time required for the product to pass through the
column. Overall, the work-up procedure chosen depends
upon the scale and overall application (i.e., single step
versus multi-step) and does not affect the diastereomeric
ratio or yield of the reaction under study.
To perform the crotylation reactions on a small scale, one
sample loop is loaded with a solution of aldehyde in toluene
and the other with a solution of the desired boronate re-
agent in toluene (Scheme 2). For larger scale reactions, the
reagents can be passed directly through the pumps without
any damage to the system. The two streams are mixed at
low temperature in a T-piece and passed through the reactor
coil and a back-pressure regulator, which holds the system
under pressure. By using this method, reactions commonly
performed in batch mode were easily reproduced under
flow conditions, both in terms of yield and diastereomeric
ratio (Table 1). We also found that by using the flow condi-
tions, simple substrates (Table 1, entries 1–3 and 11) could
Reduction in flow: Prior to the crotylation event the initial
aldehyde is often produced from the corresponding ester,
either by direct reduction or more commonly by full reduc-
tion to the alcohol, necessitating subsequent re-oxidation.
Issues with over-reduction have prevented this transforma-
Chem. Eur. J. 2011, 17, 3398 – 3405
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3399

S. V. Ley et al.
Table 1. Roush crotylation reactions performed in flow using the conditions detailed in Scheme 2.
sition, as is common with pre-
cursors for crotylation reac-
tions, these can be sensitive to
racemisation. To circumvent
these issues we have developed
a procedure for performing re-
duction reactions in flow direct-
ly from the ester to the corre-
sponding aldehyde, with the
aim of further reacting the
product in flow with a third
stream containing a crotylation
reagent, and thereby minimis-
ing handling of the sensitive in-
termediates. Following drying
of the pumping system, as de-
scribed earlier, a solution of
ester (S)-9a and a solution of
DIBAL-H in toluene were
loaded into sample loops and
passed through the low-temper-
ature coil (Scheme 4). For opti-
mum results, it is recommended
to work at dilutions of 0.5 m or
lower, to avoid inorganic pre-
cipitates blocking the channels
within the valves. The yields of
the resulting aldehydes were
consistently high (typically over
90%). Performing similar reac-
tions in batch led to significant
over-reduction and irreproduci-
bility. The greater reproducibili-
ty in the flow process may be a
result of the precise control
over stoichiometry and temper-
ature compared with adding
Aldehyde
Reagent Product
T [8C] Work-up Yield d.r.[a]
[%]
1
2
3a
3a
N
syn-5a
syn-5a
À78
À40
in-line
off-line
89
85
16:1
16:1
3
3a
syn-5a
À20
off-line
in-line
87
14:1
4
3a
syn-5a
À78
90
18:1
5
6
7
(S)-4a
(S)-4a
(R)-4a
G
anti,syn-6a À78
syn,anti-6a À78
anti,syn-6a À78
in-line
off-line
in-line
80
82
75
8:1
3.6:1
4:1
8
9
G
(S,S)-1
anti,syn-6b À78
in-line
83
70
11:1
5:1
anti,syn-6b À78
anti,syn-6c À78
off-line
10
11
off-line
in-line
86
94
16:1
20:1
syn-5b
À40
1
[a] Determined by GC-MS and H NMR spectroscopy.
DIBAL-H dropwise into
a
batch vessel. There are also op-
tions for the in-line work-up of
these reactions, which simplify
and eliminate manual down-
stream processing. For example,
a column packed with sodium
sulfate decahydrate and MgSO4
removes aluminium salts in-
line. However, due to the void
volume in a column containing
sodium sulfate decahydrate, a
large dispersion of the product
results, which is insignificant for
single step operations but prob-
lematic for multi-step sequen-
Scheme 3. Different possible polymer-supported reagents for in-line work-up of the Roush crotylation reac-
tions.
ces. Consequently,
a column
packed with IRA-743 and silica
tion from being used to its full effect in industrial process-
gel gives the same clean product but a much better disper-
sion profile. Although the main advantage of this method is
es.[7] When the aldehyde contains a chiral centre in the a-po-
3400
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 3398 – 3405

Chain-Elongation Reactions
FULL PAPER
Applications in consecutive multistep sequences: Having es-
tablished a convenient in-line work-up for the reduction of
esters to aldehydes, we are now in a position to perform the
reduction–crotylation joint reaction sequence as a single
flow process (Scheme 5). This was achieved by timing the
addition of a third stream of the crotylation reagent through
infra-red monitoring.[8] This whole sequence takes around
4 h to complete, which should be compared to at least one
day of work using conventional batch techniques and purifi-
cations.
Marshall homopropargylation: A disadvantage of the Roush
crotylation procedures adapted for the flow transformation
is the fact that a syn,syn relationship is not easy to establish.
Furthermore, depending on the desired application, a termi-
nal triple bond in the product might be more useful for di-
versification than the alkene produced in the crotylation re-
action. One way to address these issues would be to use the
Marshall procedure, whereby a chain elongation is achieved
through the addition of a chiral allene to an aldehyde.[9,10] In
order to compliment the crotylation protocol, we have
therefore developed a flow process for the boron trifluoride
etherate mediated homopropargylation under Marshall con-
ditions.[10]
In a series of optimisation reactions, we found that the
flow process allowed for a significant reduction in the
amount of toxic stannane needed to obtain comparable
yields to the corresponding batch processes. To perform the
reaction under flow conditions, allenyltin compound 12[11]
and the aldehyde are mixed as a 1m solution in CH2Cl2 and
BF3·Et2O is dissolved in CH2Cl2 as a 2m solution. These two
standard solutions are then injected into the sample loops of
a conventional flow-chemistry platform (Uniqsis FlowSyn[12]
Scheme 4. DIBAL-H reduction reactions in flow.
the controlled formation of aldehydes from esters, it can
also be used to great effect for reduction to the alcohol oxi-
dation level. Starting from ester 10, para-bromobenzyl alco-
hol (11) was produced in 98% yield in 30 min for the com-
plete reduction process, while in batch-mode a lengthy Ro-
chelleꢁs salt work-up is required.
Scheme 5. Reduction and crotylation multi-step flow sequence.
Chem. Eur. J. 2011, 17, 3398 – 3405
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3401

S. V. Ley et al.
or the Vapourtec R2+/R4 unit[5]), followed by mixing in a
cooled T-piece. The reaction mixture is then passed through
a cooled coil of appropriate length with respect to the flow-
rate employed. The desired stereotriads were obtained in
yields and selectivities comparable to the previously pub-
flow process was generally superior to the batch process em-
ploying the same conditions. Further application of the Mar-
shall protocols using other Lewis acids, such as magnesium
bromide,[14] indium triiodide[15] or indium iodide,[16] to effect
syn,anti or anti,anti stereochemical relationships within the
stereotriad are inherently more complicated due to solubili-
ty issues encountered in flow. These extensions are currently
being investigated in more detail.[17]
lished batch reactions, in which less favourACTHNUGRTENUNGable stoichiome-
tries of toxic reagents were used (Scheme 6, Table 2).[13] The
Work-up procedure and in-line purification: The develop-
ment of an in-line scavenging system for the Marshall reac-
tion was very challenging since a successful chain elongation
will result initially in a stannylated alcohol functionality.
À
This tin oxygen bond is difficult to cleave and, following
the release of the free alcohol group, the toxic tin waste has
to be separately scavenged.[18,19] Extensive screening of poly-
mer supported reagents showed that numerous systems
À
could affect the cleavage of the Sn O bond, but that the
scavenging of the resulting tin waste was more problematic.
Some of the standard scavenging systems resulted in the for-
mation of by-products (elimination products) or further de-
composition. The scavenging protocol displayed in
Scheme 6b, employing QuadraSil TA (QS-TA; a silica-sup-
ported trisamine) and conventional silica gel, proved to be
the most reliable combination, yielding the standard testing
product 14a in >95% purity and 77% yield (Scheme 6,
Table 2, entry 1). Although this in-line work-up and purifica-
tion protocol has an effect on the overall yield, the approach
is valuable as handling of the toxic tin waste can be circum-
vented and the overall amount of time required to produce
clean material was significantly reduced compared with run-
ning the full batch sequence with its associated work up dif-
ficulties.
Reduction of the Marshall homopropargylation products in
flow: In the challenging case in which both a syn,syn rela-
tionship of stereocentres and a terminal double bond are re-
quired for the synthesis programme, the product of the Mar-
shall flow process can be immediately injected into the H-
Cube to effect a partial reduction of the triple bond
(Scheme 7).[20] The isolated immediate product syn-14a was
re-dissolved in ethyl acetate (0.1 m), and the solution sub-
jected to hydrogenation conditions (258C, 10 bar) employing
the Lindlar catalyst (Pd/BaCO3/PbO), yielding syn-5a in
95% yield without compromising the diastereomeric ratio
(determined by 1H NMR spectroscopy). In order to tele-
scope steps in future applications,[21] we were able to per-
form the hydrogenation in CH2Cl2, a solvent which is scarce-
ly used in this kind of transformation, by simply increasing
the pressure to 30 bar (Scheme 7).
Scheme 6. Marshall homopropargylation reactions in flow.
method was found to be applicable to standard aliphatic al-
dehydes such as cyclohexyl carbaldehyde (3a; Table 2, en-
tries 1–5 and 13) and to a range of substrates frequently
used in polyketide synthesis (Table 2, entries 6–9). Unfortu-
nately, chiral aldehydes 4b and (R)-4c did not react smooth-
ly with the allenyltin species (Table 2, entries 8 and 9). In
both cases, the diastereoselectivities dropped significantly.
For aromatic substrates, the results show that the substitu-
tion pattern and electronics have a substantial influence on
the homopropargylation with respect to both yields and se-
lectivities (Table 2, entries 10–12). In the case of a-un-
branched substrates, the selectivity was less impressive
(Table 2, entry 13), although this result is comparable to re-
sults obtained in conventional batch processes.
Conclusion
With respect to larger scale applications, we were interest-
ed in the overall performance of the flow Marshall reaction
in comparison with a corresponding batch reaction at higher
temperatures. As can be seen in Table 2, entries 1–5, the
In summary, we have extended the applications of flow-
chemistry technology to stereoselective organic transforma-
tions that require inert conditions and low temperatures. By
3402
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 3398 – 3405

Chain-Elongation Reactions
FULL PAPER
Table 2. Marshall homopropargylation reactions performed in flow using the conditions detailed in Scheme 6.
forming these reactions as a
single process together with re-
ductions or subsequent hydro-
Aldehyde
Reagent Product
T [8C] Work-up Yield d.r.[a]
[%]
1
2
3a
3a
3a
3a
3a
rac-12
rac-12
rac-12
rac-12
rac-12
syn-14a
syn-14a
syn-14a
syn-14a
syn-14a
À78
À78
À20
0
in-line
77
94
77
78
76
3.6:1 genations, this machine-assisted
off-line
off-line
off-line
off-line
4.3:1
3.1:1
3.2:1
3.0:1
approach provides the common
starting materials used in com-
plex molecule assembly in a
much shorter time period than
the corresponding batch reac-
tions.
3[b]
4
5[b]
RT
6
7
(S)-4a
(R)-4a
(P)-12
syn,syn-13a À78
syn,syn-13a À78
off-line
in-line
75
54
10:1
(M)-12
>10:1
Experimental Section
8
9
G
syn,syn-13b À78
syn,syn-13c À78
in-line
53
2:1
General Remarks: Toluene was dried
and freshly distilled from sodium wire
before use. Dichloromethane was dis-
tilled from calcium hydride. All reac-
tions were performed under an atmos-
phere of argon and carried out using
flow-chemistry platforms the pumps,
tubing, and so forth of which have
been dried by using the protocol out-
lined below. Unless otherwise speci-
fied, reactants and reagents, as well as
polymer-supported reagents and scav-
engers were used as purchased, with-
out further purification. Infrared spec-
tra were recorded as neat samples on
a Perkin–Elmer Spectrum One FTIR
spectrometer fitted with an ATR sam-
pling accessory. 1H NMR spectra were
recorded at 278C on a Bruker DPX-
400. Residual protic solvent was used
as the internal reference (CHCl3 dH =
7.27 ppm). 13C NMR spectra were re-
corded at 125 MHz on a Bruker DPX-
400. The resonance of CDCl3 (dC =
in-line
64
75
2.5:1
10:1
10
syn-14b
syn-14c
À78
À78
off-line
11
off-line
49
3.9:1
12
13
syn-14d
syn-14e
À78
À78
off-line
in-line
22
43
3.6:1
1:1
[a] Determined by H NMR spectroscopy. [b] Only 1.2 equivalents of Lewis acid were used.
77.0 ppm, t) was used as the internal reference. GC-MS was performed
on a Perkin–Elmer Turbomass Autosystem XL with a Supelco SLB-5 ms
column (30 mꢂ0.25 mmꢂ0.25 mm film thickness) and positive electron
ionisation (EI+). Optical rotation was measured by using a Perkin–
Elmer Polarimeter 343 with the sample temperature maintained at 258C.
[a]2D5 is reported in units of 10À1 deggÀ1 cm2. Concentration is quoted in
units of 0.01 gcmÀ3. HRMS were recorded on a Waters Micromass LCT
Premier Q-TOF spectrometer by electrospray ionisation (ESI) or an
ABI/MDS Sciex Q-STAR Pulsar. Unless otherwise stated, the mass re-
ported contains the most abundant isotopes. Limit: Æ5 ppm. Flash
column chromatography for obtaining analytically pure material was car-
ried out by using silica gel [Merck or Breckland (230–400 mesh)] under
pressure with DE/PE mixtures (DE=diethyl ether, PE=petroleum
ether, b.p. 40–608C).
Procedure for the Roush crotylation reaction under flow conditions, em-
ploying an in-line work-up: The system manifold [pumps, valves, tubing
(0.5 mm id, perfluoroalkoxy (PFA)) and reactor coil] of a Vapourtec
R2+/R4 unit were dried with IPA (2 mLminÀ1
, 15 min), acetone
Scheme 7. Marshall homopropargylation reaction and subsequent reduc-
tion in flow.
(1 mLminÀ1, 1 h) and anhydrous toluene (0.5 mLminÀ1, 2 h). A solution
of aldehyde in toluene (0.5 m) was then loaded into a PEEK sample loop
(2 mL), and the crotylation reagent (1 mmol) in toluene (2 mL) was
loaded into the other PEEK sample loop (2 mL). The two solutions were
pumped through the apparatus at a combined rate of 0.3 mLminÀ1 and
mixed in a Vapourtec low-temperature coil (10 mL plus two premixing
coils) held at À788C. The output of the coil was directed into an Omnitfit
column (10ꢂ100 mm) containing Amberlyst IRA-700 hydroxide resin
(2 g) and a plug of Celite (2 cm). The solution was collected and evapo-
rated in vacuo. By using this procedure, (S)-3-tert-butyldimethylsilyloxy-
developing reliable flow procedures for both the Roush cro-
tylation reaction and the Marshall homopropargylation reac-
tion, we have significantly increased the repertoire of organ-
ic transformations possible in flow, particularly for future
polyketide natural-product synthesis. Furthermore, by per-
Chem. Eur. J. 2011, 17, 3398 – 3405
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3403

S. V. Ley et al.
2-methylpropanal ((S)-4a) was reacted with crotylation reagent (S,S)-1 to
yield olefin anti,syn-6a as a colourless oil (80%, d.r.=8:1). 1H NMR
(500 MHz, CDCl3): d=5.84 (ddd, J=7.3, 10.4, 17.5 Hz, 1H), 5.05–5.02
(m, 2H), 3.82 (dd, J=3.9, 9.9 Hz, 1H), 3.61–3.55 (m, 2H), 3.38 (dt, J=
4.6, 6.7 Hz, 1H), 2.35–2.27 (m, 1H), 1.82–1.73 (m, 1H), 1.03 (d, J=
6.8 Hz, 3H), 0.89 (d, J=7.1 Hz, 3H), 0.87 (s, 9H), 0.05 ppm (d, J=
2.1 Hz, 6H); 13C NMR (126 MHz, CDCl3): d=142.47, 113.88, 79.60,
67.98, 41.26, 36.62, 25.85, 18.14, 14.10, 13.40, À5.61, À5.64 ppm; IR
(ATR): n˜ =3510, 2929, 2858, 1463, 1253, 1075 cmÀ1; [a]D25 =+6.4 (c=2.00,
CHCl3). The spectroscopic data correspond to the literature.[4a]
Ley, I. F. McConvey, B. Al-Duri, G. A. Leeke, R. C. D. Santos,
J. P. K. Seville, Chem. Commun. 2005, 2175–2177; m) G. N. Doku,
W. Verboom, D. N. Reinhoudt, A. van den Berg, Tetrahedron 2005,
61, 2733–2742; n) G. Jas, A. Kirschning, Chem. Eur. J. 2003, 9,
5708–5723; o) A. M. Hafez, A. E. Taggi, T. Lectka, Chem. Eur. J.
2002, 8, 4114–4119; p) W. Ehrfield, V. Hessel, H. Lçwe, Microreac-
tors: New Technology for Modern Chemistry, Wiley-VCH, Wein-
heim, 2000; q) A. M. Hafez, A. E. Taggi, H. Wack, W. J. Drury, T.
Lectka, Org. Lett. 2000, 2, 3963–3965.
[2] Recent applications of flow chemistry from our group: a) J. Sedel-
meier, S. V. Ley, I. R. Baxendale, M. Baumann, Org. Lett. 2010, 12,
3618–3621; b) C. H. Hornung, B. Hallmark, M. R. Mackley, I. R.
Baxendale, S. V. Ley, Adv. Synth. Catal. 2010, 352, 1736–1745;
c) C. H. Hornung, B. Hallmark, M. Baumann, I. R. Baxendale, S. V.
Ley, P. Hester, P. Clayton, M. R. Mackley, Ind. Eng. Chem. Res.
2010, 49, 4576–4582; d) F. Venturoni, N. Nikbin, S. V. Ley, I. R. Bax-
endale, Org. Biomol. Chem. 2010, 8, 1798–1806; e) M. OꢁBrien,
I. R. Baxendale, S. V. Ley, Org. Lett. 2010, 12, 1596–1598; f) M. D.
Hopkin, I. R. Baxendale, S. V. Ley, Chem. Commun. 2010, 46, 2450–
2452; g) C. F. Carter, I. R. Baxendale, J. B. J. Pavey, S. V. Ley, Org.
Biomol. Chem. 2010, 8, 1588–1595; h) M. Baumann, I. R. Baxen-
dale, S. V. Ley, Synlett 2010, 749–752; i) Z. Qian, I. R. Baxendale,
S. V. Ley, Synlett 2010, 505–508; j) I. R. Baxendale, S. C. Schou, J.
Sedelmeier, S. V. Ley, Chem. Eur. J. 2010, 16, 89–94; k) C. F. Carter,
I. R. Baxendale, M. OꢁBrien, J. B. J. Pavey, S. V. Ley, Org. Biomol.
Chem. 2009, 7, 4594–4597; l) J. Sedelmeier, S. V. Ley, H. Lange,
I. R. Baxendale, Eur. J. Org. Chem. 2009, 4412–4420; m) M. Bau-
mann, I. R. Baxendale, L. J. Martin, S. V. Ley, Tetrahedron 2009, 65,
6611–6625; n) A. Palmieri, S. V. Ley, A. Polyzos, M. Ladlow, I. R.
Baxendale, Beilstein J. Org. Chem. 2009, 5, No. 23; o) A. Palmieri,
S. V. Ley, K. Hammond, A. Polyzos, I. R. Baxendale, Tetrahedron
Lett. 2009, 50, 3287–3289; p) I. R. Baxendale, S. V. Ley, A. C. Mans-
field, C. D. Smith, Angew. Chem. 2009, 121, 4077–4081; Angew.
Chem. Int. Ed. 2009, 48, 4017–4021; q) J. Sedelmeier, I. R. Baxen-
dale, S. V. Ley, Green Chem. 2009, 11, 683–685.
Procedure for the Marshall homopropargylation reaction under flow con-
ditions, employing an in-line work-up: A pre-dried (see above) Vapour-
tec R2+/R4 unit or Uniqsis FlowSyn was used as the flow-chemistry
platform, equipped with two sample loops (2 mL), a T-piece and a self-
made PTFE reactor coil (30 mL, inner diameter 0.5 mm). This reactor
coil was attached to an Omnifit glass column (6.6ꢂ10 mm), which was
filled with Quadrasil-TA (300 mg) and silica gel (300 mg), followed by a
back pressure regulator (100 psi). The T-piece and reaction coil were im-
mersed in a dry ice/acetone bath and the setup was purged with anhy-
drous CH2Cl2. A solution of aldehyde (2 mL, 0.1 m) and (tri-n-butylstann-
yl)-1,2-butadiene (12) in CH2Cl2 and a solution of BF3·OEt2 in CH2Cl2
(2 mL, 0.2 m) were injected into independent sample loops and pumped
at a combined rate of 0.5 mLminÀ1. The solution exiting the system was
collected for 3 h and concentrated in vacuo to give the desired homopro-
pargylic alcohol. By using this procedure, (R)-3-tert-butyldimethylsily-
loxy-2-methylpropanal ((R)-4a) was reacted with (M)-(tri-n-butylstanyl)-
1,2-butadiene ((M)-12) in the presence of BF3·OEt2 to yield diastereo-
merically pure (2R,3R,4S)-1-tert-butyldimethylsilyloxy-2,4-dimethyl-5-
butyn-3-ol (syn,syn-13a, d.r.>10:1, 54%, 28 mg, 0.11 mmol) as a colour-
less oil (purity >95% by 1H NMR spectroscopy). 1H NMR (400 MHz,
CDCl3): d=3.87 (dd, J=3.0, 9.8 Hz, 1H), 3.72 (dd, J=3.9, 9.8 Hz, 1H),
3.70–3.72 (m, 1H), 3.42 (d, J=1.5 Hz, 1H), 2.48–2.56 (m, 1H), 2.08–2.18
(m, 1H), 2.05 (d, J=2.0 Hz, 1H), 1.30 (d, J=6.8 Hz, 3H), 1.00 (d, J=
7.1 Hz, 3H), 0.92 (s, 9H), 0.08 ppm (s, 6H); 13C NMR (100 MHz,
CDCl3): d=86.4, 78.2, 69.9, 69.4, 36.3, 30.2, 25.9, 18.2, 17.7, 9.4, À5.6,
À5.7 ppm; IR (ATR): n˜ =3495, 3313, 2956, 2930, 2908, 2884, 2858, 1472,
1463, 1391, 1374, 1362, 1254, 1158, 1139, 1092, 1059, 1018, 1006, 990, 966,
939, 908, 834, 815, 776, 681, 666 cmÀ1; [a]D25 =À9.3 (c=0.74, CDCl3). The
spectroscopic data correspond to the literature.[13b]
[3] For an example of handling highly water-sensitive compounds in
flow, namely n-butyl lithium, see ref. [2d].
[4] a) W. R. Roush, K. Ando, D. B. Powers, A. D. Palkowitz, R. L. Hal-
terman, J. Am. Chem. Soc. 1990, 112, 6339–6348; b) W. R. Roush,
A. D. Palkowitz, K. Ando, J. Am. Chem. Soc. 1990, 112, 6348–6359;
c) W. R. Roush, A. D. Palkowitz, M. J. Palmer, J. Org. Chem. 1987,
52, 316–318.
[5] Vapourtec R2+ and R4 units are available from Vapourtec, Place
Farm, Ingham, Suffolk, IP31 1NQ, UK. Website: www.vapourtec.
co.uk.
Acknowledgements
[6] C. F. Carter, H. Lange, I. R. Baxendale, S. V. Ley, J. Goode, N.
Gaunt, B. Wittkamp, Org. Process Res. Dev. 2010, 14, 393–404.
[7] L. Ducry, D. M. Roberge, Org. Process Res. Dev. 2008, 12, 163–167.
[8] H. Lange, C. F. Carter, M. Hopkin, A. Burke, J. G. Goode, I. R.
Baxendale, S. V. Ley, Chem. Sci. 2011, DOI: 10.1039/C0SC00603C.
[9] Recent reviews: a) J. A. Marshall, J. Org. Chem. 2007, 72, 8153–
8166; b) J. A. Marshall, Chem. Rev. 2000, 100, 3163–3185.
[10] J. A. Marshall, X.-J. Wang, J. Org. Chem. 1990, 55, 6246–6248.
[11] Chiral allenyltin species 12 was prepared according to J. A. Mar-
shall, H. Chobanian, Org. Synth. 2005, 82, 43–54.
[12] The FlowSyn unit is available from Uniqsis, 29 Station Road, She-
preth, Cambridgeshire, SG8 6GB, UK. Website: www.uniqsis.com.
[13] The reaction displayed in Scheme 6a yielding stereotriad 13a has
been reported by Marshall and co-workers twice using different con-
ditions, see: a) J. A. Marshall, K. C. Ellis, Org. Lett. 2003, 5, 1729–
1732: (S)-4a (1.38 mmol, 1 equiv), (P)-12 (1.6 equiv) and BF3·OEt2
(3.1 equiv); b) J. A. Marshall, G. M. Schaaf, J. Org. Chem. 2001, 66,
7825–7831: (R)-4a (19.8 mmol, 1 equiv), (M)-12 (1.15 equiv) and
BF3·OEt2 (3.0 equiv).
[14] For use of a chiral allenyltin species in combination with MgBr2 as
the Lewis acid, see: a) ref. [13b]; b) J. A. Marshall, J. F. Perkins,
M. A. Wolf, J. Org. Chem. 1995, 60, 5556–5559; c) J. A. Marshall,
J. F. Perkins, J. Org. Chem. 1994, 59, 3509–3511; d) J. A. Marshall,
X.-J. Wang, J. Org. Chem. 1992, 57, 1242–1252.
The authors thank Dr. J. B. J. Pavey and AstraZeneca (C.F.C.), the
Alexander von Humboldt Foundation and the EPSRC (H.L.), Mitsubishi
Tanabe Pharma (D.S.), the Royal Society (I.R.B.) and the BP Endow-
ment (S.V.L.) for funding.
[1] Selected reviews on the concepts, benefits, applications and tools of
flow chemistry: a) D. Webb, T. F. Jamison, Chem. Sci. 2010, 1, 675–
680; b) X. Y. Mak, P. Laurino, P. H. Seeberger, Beilstein J. Org.
Chem. 2009, 5, No. 19; c) S. Y. Teh, R. Lin, L. H. Hung, A. P. Lee,
Lab Chip 2008, 8, 198–230; d) D. Wilms, J. Klos, H. Frey, Macro-
mol. Chem. Phys. 2008, 209, 343–356; e) B. K. Singh, N. Kaval, S.
Tomar, E. Van der Eycken, V. S. Parmar, Org. Process Res. Dev.
2008, 12, 468–474; f) C. Wiles, P. Watts, Eur. J. Org. Chem. 2008,
1655–1671; g) P. B. Mason, K. E. Proce, J. L. Steinbacher, A. R.
Bogdan, D. T. McQuade, Chem. Rev. 2007, 107, 2300–2318; h) I. R.
Baxendale, S. V. Ley in New Avenues to Efficient Chemical Synthe-
sis: Emerging Technologies (Ernst Schering Foundation Symposium
Proceedings) (Eds: P. H. Seeberger, T. Blume), Springer, Berlin,
2007, 151–185; i) A. Kirschning, W. Solodenko, K. Mennecke,
Chem. Eur. J. 2006, 12, 5972–5990; j) V. Hessel, H. Lçwe, A.
Mꢃller, G. Kolb, Chemical Micro Process Engineering: Processing
and Plants, Wiley-VCH, Weinheim, 2005; k) P. Hodge, Ind. Eng.
Chem. Res. 2005, 44, 8542–8553; l) C. K. Y. Lee, A. B. Holmes, S. V.
3404
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 3398 – 3405

Chain-Elongation Reactions
FULL PAPER
[15] For use of indium tribromide to generate a chiral allenylindium bro-
mide in situ, see: a) J. A. Marshall, B. A. Johns, J. Org. Chem. 2000,
65, 1501–1510; b) J. A. Marshall, B. A. Johns, J. Org. Chem. 1998,
63, 7885–7892.
not to be feasible in flow as the solid phosphorous-based scavenger
slowly dissolved in CH2Cl2 under different flow conditions: J. Srogl,
G. D. Allred, L. S. Liebeskind, J. Am. Chem. Soc. 1997, 119, 12376–
12377.
[16] For use of indium iodide to generate a chiral allenylindium iodide in
situ, see: a) J. A. Marshall, H. R. Chobanian, M.-M. Yanik, Org.
Lett. 2001, 3, 3369–3372; b) ref. [13a]; c) J. A. Marshall, C. M.
Grant, J. Org. Chem. 1999, 64, 8214–8219.
[20] The H-Cube and the catalyst-loaded cartridges are available from
ThalesNano, Gꢄzgyar u. 1–3, Budapest, Hungry H-1031. Website:
www.thalesnano.com.
[21] For a recent report on a multi-step machine-assisted synthesis in
which the H-Cube was incorporated, see ref. [2i].
[17] S. Newton, S. V. Ley, unpublished results.
[18] For a convenient way to remove tin waste in a solution-phase pro-
cess, see: D. P. Curran, C.-T. Chang, J. Org. Chem. 1989, 54, 3140–
3157, and literature cited therein.
[19] A protocol known to be very successful in the scavenging of tin
waste in Stille coupling reactions was tried as well, but was found
Received: November 2, 2010
Published online: February 23, 2011
Chem. Eur. J. 2011, 17, 3398 – 3405
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3405