The continuous flow synthesis of butane-2,3-diacetal protected building blocks using microreactors

Article abstract of DOI:10.1039/b924309g

The continuous flow synthesis of butane-2,3-diacetal protected derivatives has been achieved using commercially available flow chemistry microreactors in concert with solid supported reagents and scavengers to provide in-line purification systems. The BDA protected products are all obtained in superior yield to the corresponding batch processes.

Full text of DOI:10.1039/b924309g

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
The continuous flow synthesis of butane-2,3-diacetal protected building blocks
using microreactors†
a
a
b
a
Catherine F. Carter, Ian R. Baxendale, John B. J. Pavey and Steven V. Ley*
Received 19th November 2009, Accepted 24th December 2009
First published as an Advance Article on the web 10th February 2010
DOI: 10.1039/b924309g
The continuous flow synthesis of butane-2,3-diacetal protected derivatives has been achieved using
commercially available flow chemistry microreactors in concert with solid supported reagents and
scavengers to provide in-line purification systems. The BDA protected products are all obtained in
superior yield to the corresponding batch processes.
Introduction
chromatography, extractions and aqueous washes. It is our view
that many of these labour intensive events are best relegated to
a machine assisted synthesis approach whereby skilled operator
time is released for improved molecular design and better synthetic
The synthesis of complex natural products has reached a high
level of sophistication; nevertheless, it is not without its problems,
not least of which is the necessary preparation of the required
2,3
planning.
1
coupling partners and building blocks on scale. More often than
4,5
In our work we have made extensive use of 1,2-diacetals,
not, the scaling process can be frustrated by lack of consistency and
consequently this leads to the need for constant route optimisation
and redevelopment. In addition, the scale-up procedure involves
many repetitive experiments where the knowledge gained is
minimal yet the skilled workforce commitment is substantial. This
arises due to the multiple problems of down-stream processing
and unit work-up operations such as solvent handling and
disposal, combined with losses that accrue via recrystallisation,
and in particular butane-2,3-diacetals (BDA), to effect selective
6–8
vicinal diol and a-hydroxyacid protection, diastereoselective
9–12
and facially selective processes, and to prepare a whole variety
of chiral building blocks for natural product synthesis programmes
13–16
(
Fig. 1).
Here, we show how key members of this series
can be assembled successfully with high reproducibility using
flow chemistry methods and associated continuous processing
techniques.
a
Department of Chemistry, Innovative Technnology Centre, University of
Results
Cambridge, Lensfield Road, Cambridge, UK CB2 1EW. E-mail: svl1000@
cam.ac.uk
BDA protected tartrates
b
AstraZeneca, Bakewell Road, Loughborough, Leics, LE11 5RH
We have recently reported the synthesis of butane-2,3-diacetal
tartrates 1–3 in flow, assisted by the use of an in-line IR flow
†
This paper is part of an Organic & Biomolecular Chemistry web theme
issue on enabling technologies for organic synthesis
Fig. 1 Butane-2,3-diacetals in synthesis.
1
588 | Org. Biomol. Chem., 2010, 8, 1588–1595
This journal is © The Royal Society of Chemistry 2010

View Article Online
Scheme 1 Flow syntheses of butane-2,3-diacetal tartrates using the Uniqsis FlowSyn.
17
cell for reaction monitoring. Three tartrate building blocks were
successfully synthesised and purified in-line using commercially
available flow chemistry equipment, resulting in higher yields than
the corresponding batch processes (Scheme 1).
The full details of these syntheses and additional new work is
further reported here. All experiments were performed using the
18
Uniqsis FlowSyn platform. For our purposes the system was
operated with two HPLC pumps, a 14 mL PTFE reactor coil
and OmniFitꢀR columns packed with solid-supported reagents
and scavengers (typically 10 ¥ 150 mm columns, see experimental
section for full details), with a 100 psi back pressure regulator
fitted at the output flow stream.
A particularly successful feature of the BDA tartrate procedures
19
(
Scheme 1) was the use of in-line solid-supported reagents
Scheme 2 In-line work-up and purification procedure for the synthesis of
BDA protected tartrate 1.
to purify the reaction stream. BDA protected tartrate 1 was
synthesised on a 38 g scale by passing stock solutions of the
◦
reagents in MeOH through a reactor coil held at 90 C. As
previously described, the in-line purification consisted of firstly
an OmniFitꢀR column filled with benzylamine resin (QP-BZA)
20
followed by a second column containing periodate resin 4. As
shown in Scheme 2, when the crude reaction mixture enters the
first column the excess butanedione becomes immobilised as the
imine species and the camphor sulfonic acid (CSA) catalyst is
sequestered as an ion pair salt. In order to remove unreacted
dimethyl-L-tartrate, the supported periodate performs a rapid
glycol cleavage leading to volatile byproducts. On exiting these
columns, the reaction stream contains pure product 1.
Scheme 3 A resin for scavenging iodine.
removes diisopropyl amine and the short plug of silica gel
effectively traps any inorganic lithium salts.
The corresponding meso-BDA-protected derivative 3 was then
synthesised as required on a 3 g scale batch using the H-Cube
For the synthesis of alkene 2, a polymer-supported reagent
was required to remove the residual iodine that was employed
in forming the alkene double bond. To facilitate this, thiosulfate
resin 5 was freshly prepared according to a procedure reported by
TM
Midi . The reaction was found to be ineffective in THF so it was
21
Parlow et al. (Scheme 3). The image shows the crude reaction
mixture from the synthesis of alkene 2 before and after being
passed through a column of the resin, visibly demonstrating its
effectiveness at removing the iodine from the reaction mixture. The
sulfonic acid column attached after the reactor coil (Scheme 1)
not possible to telescope the syntheses of 2 and 3 into a single
continuous process without needing a solvent switch. However,
recycling a solution of alkene 2 in MeOH through the hydro-
genation reactor affords BDA-protected tartrate 3 in quantitative
yield, with evaporation of solvent being the only manual operation
This journal is © The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 1588–1595 | 1589

View Article Online
required. This represents a considerable advantage over the batch
hydrogenation process, which requires 5 days of stirring in a vessel
pressurised with 80 bars of hydrogen gas.
silica gel chromatography we postulated that the use of 1.8 eq
of butanedione increases the yield as it reduces the propensity to
form tris-protected byproducts.
The next stage of the sequence was to effect the glycol
cleavage of the central diol bond. The transformation to the
BDA-protected aldehyde 7 was likely to be facile, particularly as
BDA-protected glyceraldehyde
Following the preparation of the family of BDA-protected tar-
trates, we continued to investigate other BDA-protected substrates,
namely aldehyde 7 and ester 8. In batch mode, the synthesis of
the BDA-protected mannitol 6 proceeds via condensation with
butanedione, using trimethyl orthoformate as a dehydrating agent
NaIO supported on silica has previously been used to cleave
4
25
mannitol protected as the bis-acetonide. The original batch
procedure for this transformation involves overnight stirring in
CH
of MgSO
2
Cl
2
and aqueous NaHCO
3
, and then stoichiometric addition
24
4
and Na
2
S
2
3
O . It was found that the reaction could be
22–24
and BF
Sodium periodate in CH
3
·THF complex as the activating Lewis acid (Scheme 4).
performed rapidly, in flow, by recirculating a solution of the bis-
protected BDA-protected mannitol 6 through a column packed
with periodate resin 4 (Scheme 6). A small amount of water is
required for the success of this reaction; therefore, on completion,
the solution was finally directed through a drying column of
2
Cl is then used to cleave the central diol
2
bond to yield aldehyde 7. When this transformation is carried out
in methanol, the hemiacetal is initially obtained, which is then
oxidised in situ with bromine to afford the ester derivative 8.
MgSO . This procedure furnishes the aldehyde 7 in superior yield
4
to the batch process, mainly due to the purity of the starting
bis-protected material, providing the aldehyde in high purity and
hence avoiding the need for aqueous work-up and reduced pressure
distillation.
Scheme 4 Batch synthesis of BDA-protected mannitol derivatives.
A significant drawback with this procedure is the purity of
the initially formed adduct 6, since there are seven possible
alternatives that could arise from the first step. Consequently,
following the glycol cleavage the aldehyde and ester require
extensive purification by a non-trivial reduced pressure distillation,
resulting in the relatively low yields.
Scheme 6 Continuous flow synthesis of BDA-protected aldehyde 7.
The synthesis of the BDA-protected ester 8 proceeded in a
similar fashion, with oxidation of the intermediate hemiacetal
being carried out in-line with a polymer-supported perbromide
The first challenge in order to achieve a flow synthesis protocol
was the insolubility of the D-mannitol itself, since in the batch
process it slowly dissolves as the reaction proceeds. Insolubility
can be a major issue for flow-based approaches because solid
deposits may cause unreliable flow rates and blockages, resulting
in overpressure and automatic system shutdown. This problem
was solved by premixing D-mannitol with CSA and trimethyl
orthoformate. Interestingly, this was the only combination of
reagents in which the D-mannitol could be fully dissolved and
remain solublised. This observation could be further evidence
that the BDA protection mechanism proceeds via pre-activation
of the diol through an orthoester intermediate, as described in our
26
resin (Scheme 7).
Scheme 7 Continuous flow synthesis of BDA-protected ester 8.
17
The simplicity of the above processes makes them more
amenable and reproducible in scale up experiments than the
corresponding batch protocols.
previous paper.
A series of rapid optimisation reactions conducted in flow found
◦
that a slight warming of the reaction mixture to 40 C, and the use
of precisely 1.8 eq of butanedione afforded the best yield (55%)
of the desired BDA-protected mannitol product 6 (Scheme 5).
Although the product still required a simple purification using
BDA-protected glycolate
The final BDA-protected building block investigated was pro-
tected glycolate 12. In the batch preparation of this compound,
the standard BDA protection conditions of refluxing in methanol
with CSA, butanedione and trimethyl orthoformate gave 9 in
excellent yield (Scheme 8). Elimination of HCl using KOtBu
then yields alkene 10. A problem, however, with this step is that
the exo-alkene readily isomerises to the more stable endo-alkene
27
11. Finally, ozonolysis of 10 under reductive work-up conditions
Scheme 5 Continuous flow synthesis of BDA protected D-mannitol 6.
furnishes the BDA-protected glycolate 12.
1
590 | Org. Biomol. Chem., 2010, 8, 1588–1595
This journal is © The Royal Society of Chemistry 2010

View Article Online
In order to try and avoid the use of ozone, a new approach was
required for the flow synthesis of lactone 12. Various polymer-
supported osmium species have been reported as alternatives
28
to in solution OsO for the dihydroxylation of double bonds.
4
It has also been shown in our laboratories that combining
TM
OsEnCat (osmium tetroxide encapulsated in a polyurea matrix)
with sodium periodate in solution achieves the cleavage of the
29
diol to the corresponding carbonyl group in high yields. Using
this procedure, alkene 10 was converted to lactone 12, although a
second byproduct was also observed (1 : 1) which was identified to
be the enantioenriched diketone 13 (Scheme 11).
Scheme 8 Batch synthesis of BDA glycolate 12.
As the batch synthesis of BDA derivative 9 requires similar
conditions to the preparation of BDA-protected tartrate 1,
the optimum conditions for the flow synthesis of 1 were first
investigated (Scheme 9). The procedure proved highly successful,
providing 9 in 95% yield with no temperature-induced racemisa-
tion of the starting material and no manual purification required.
Additionally, due to the higher conversion achieved, there was no
need to remove any unreacted diol.
Scheme 11 Formation of byproduct 13.
To investigate this side product further, the reaction was
followed using a ReactIR 45m with a fibre optic cable and diamond
probe to monitor the peak intensity of the alkene C=C and ketone
30
C=O stretches over time. After a series of control experiments,
it was determined that the formation of ketone 13 was dependant
on the periodate, as none of the byproduct was observed by IR
Scheme 9 Continuous flow synthesis of BDA-protected chloropropane-
diol 9.
TM
when the alkene was simply stirred in the presence of OsEnCat
.
However, when the alkene was stirred together with sodium
periodate in THF–H O (2 : 1), it decayed linearly into exclusively
2
Ideally, for flow applications, a polymer-supported base would
have been used to deprotonate 9 and generate the alkene 10.
This then would have allowed facile telescoping of these two
steps, giving alkene 10 as a continuous procedure from the
starting chloropropanediol. Unfortunately, none of the bases
screened achieved this transformation cleanly. The best conditions,
therefore, involve the use of a solution of KOtBu in THF, combined
at a T-piece with a solution of the BDA-protected chloropropane
diol also in THF. The reaction mixture was then passed through
product 13 over 2 h (Scheme 12). A plausible mechanism for this
transformation is shown below, resulting from the formation of
periodic acid, which could add to the enol ether to form species
14, which then eliminates to give the observed product.
◦
a 14 mL reactor coil, maintained at 70 C with a residence time of
7
0 min, affording alkene 10 in 81% isolated yield (Scheme 10).
Scheme 10 Flow synthesis of BDA protected alkene 9.
Due to the known problems of isomerisation of the double
bond in 10, it was not possible to add a further column of solid
supported acid in-line to scavenge excess KOtBu. Instead, the
reaction stream was collected directly into water and extracted in a
typical batch manner. However, flow chemistry provides one major
advantage for this reaction, which is the precise control over the
ratio of products. The reaction was found to be very reproducible,
with a ratio of 24 : 1 for exo : endo alkene observed consistently.
In contrast, the batch process is reported to result in ratios
varying from 15 : 1 to 5 : 1 in the worst case. This improvement is
probably a factor of the precise temperature control in the reactor
compared with global temperature measurement recorded in a
round-bottomed flask.
Scheme 12 Synthesis of diketone 13 monitored using an FTIR diamond
probe.
Despite varying the reaction conditions, the formation of
this byproduct could not be entirely suppressed and a modified
This journal is © The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 1588–1595 | 1591

View Article Online
Table 1 A comparison of the batch and flow syntheses of BDA-protected derivatives
Product
Synthetic method Yield (%) Purification technique
Product
Synthetic method Yield (%) Purification technique
b
Batch
70
Aqueous extraction,
recrystallisation
Batch
42
Aqueous extraction,
reduced pressure
distillation
a
b
Flow
Batch
Flow
Batch
Flow
Batch
Flow
75
60
65
In-line
Flow
Batch
Flow
Batch
Flow
Batch
Flow
49
90
95
80
81
61
85
In-line
Aqueous extraction,
recrystallisation
In-line
Aqueous extraction
In-line
a
100
100
Filtration, column
chromatography
In-line
Aqueous extraction
Aqueous extraction
Recrystallisation
In-line
a
b
41
Aqueous extraction, reduced
pressure distillation
In-line
b
47
a
b
See ref. 17. Yield determined over two steps from D-mannitol.
approach was required. We initially evaluated using a supported
readily than the manually operated batch method. We believe
that these new technologies have real value for the on demand
generation of starting materials for natural product synthesis
programmes.
periodate resin 4 to exclude the formation of periodic acid in
TM
solution. In this experiment, in order to use 5 mol% of OsEnCat
,
a stoichiometric amount of N-methylmorpholine N-oxide (NMO)
was added as a reoxidant because the supported periodate was
unable to effectively turnover the catalyst. Pleasingly, using this
protocol, the BDA-protected alkene was converted in 85% yield
to the BDA-protected lactone 12 with none of the undesired
byproduct 13 (Scheme 13). In practise, the reaction mixture
was recycled though an OmniFitꢀR column containing a mixed
Experimental
General considerations
Unless specified, reagents were obtained from commercial sources
and used without further purification. Solvents were obtained
from Fischer Scientific and distilled before use. The petroleum
ether used was the 40–60 boiling fraction. THF was dried over
TM
bed of OsEnCat and the supported periodate resin 4. When
the reaction was judged to have reached completion by LCMS,
the flow stream was redirected through two scavenger columns;
firstly, a supported acid to remove stoichiometric quantities
of N-methyl morpholine and then Quadrapure thiourea (QP-
TU), a metal scavenging resin, to remove any leached osmium
contaminants.
calcium hydride and LiAlH
and distilled under an atmosphere of dry argon. CH
4
with triphenylmethane as an indicator
Cl and
2
2
MeCN were dried over calcium hydride and distilled under an
atmosphere of dry argon. Quadrapure sulfonic acid resin (QP-SA),
Quadrapure benzylamine resin (QP-BZA), Quadraupre thiourea
TM
(
QP-TU) and OsEnCat were obtained from Reaxa Ltd and used
without further purification.
1
H NMR spectra were recorded in CDCl on a Bruker
3
Avance DPX-400 (400 MHz) spectrometer with residual CHCl
3
1
3
(
7.26 ppm) as the internal reference. C NMR spectra were
recorded using a Bruker Avance DPX-400 (100 MHz) spec-
trometer using the central resonance of CDCl as the internal
reference (77.0 ppm). Coupling constants are quoted to the nearest
.1 Hz. Infrared spectra were recorded neat on a Perkin-Elmer
3
Scheme 13 Continuous flow synthesis of BDA protected lactone 12.
0
Spectrum One FT-IR spectrometer using Universal ATR sampling
accessories. Letters in parentheses refer to the relative absorbency
of the peak: w–weak (< 40% of the most intense peak), m–medium
Conclusions
(
40-70% of the most intense peak), s–strong (>70% of the most
In conclusion, the syntheses of the family of BDA-protected
chiral building blocks have been achieved using flow chemistry
in concert with supported reagents and scavengers. As shown
in Table 1, the yields are superior to the batch processes and
no manual purification is required, with the exception of BDA
alkene 10. In this case, the machine-assisted approach has the
added advantage that precise control can be achieved more
intense peak). Optical rotations were measured on a Perkin-Elmer
Model 343 digital polarimeter. High resolution mass spectrometry
was carried out on a Waters Micromass LCT Premier spectrometer
using time of flight with positive electrospray ionisation. Elemental
analyses were determined in the microanalytical laboratories at the
Departmental of Chemistry, University of Cambridge. Column
chromatography was carried out using a Biotage SP1 Flash
1
592 | Org. Biomol. Chem., 2010, 8, 1588–1595
This journal is © The Royal Society of Chemistry 2010

View Article Online
TM
Purification system with prepacked SNAP cartridges containing
0 g of silica gel.
H-Cube Midi flow hydrogenation reactor. The solution was
◦
1
recycled for 2 h at 60 bar of H and 40 C. After this time
2
the product was pumped out and the reactor was flushed with
(
2R,3R,5R,6R)-dimethyl 5,6-dimethoxy-5,6-dimethyl-1,4-
MeOH (30 mL). Evaporation of the solvent in vacuo yielded BDA
17
25
dioxane-2,3-dicarboxylate 1
protected meso tartrate 3 as a white solid (3.80 g, 100%). [a]
D
-
1
-
(
95.5 (c 0.7 in CHCl
3
); n˜ max/cm (neat) 2954w (C–H), 1770m
Solutions of dimethyl-L-tartrate (30 g, 170 mmol) and trimethyl
orthoformate (55.8 mL, 510 mmol) dissolved in MeOH (30 mL)
and butadione (16.7 mL, 190 mmol) and CSA (3.95 g, 17 mmol)
dissolved in MeOH (60 mL) were pumped at 0.3 mL min
(
1
C=O), 1736m (C=O), 1140s (C–O), 1034s (C–O); H NMR
(400 MHz, CDCl
3
): d
H
= 4.69 (1H, d, J = 3.8 Hz, CH), 4.50 (1H,
d, J = 3.8 Hz, CH), 3.77 (3H, s, CO
2
CH
3
), 3.75 (3H, s, CO
), 1.38 (3H, s, CCH
); C NMR (100 MHz, CDCl ): d
2
CH
3
),
),
-
1
3
1
.30 (3H, s, COCH
.33 (3H, s, CCH
3
), 3.21 (3H, s, COCH
3
3
0.15 mL min- per channel) into a T-piece and through a 14 mL
1
13
3
3
C
= 169.6
◦
CFC reactor at 90 C. The crude solution was passed through a
(
(
(
(
CO), 168.8 (CO), 100.4 (C), 99.3 (C), 69.1 (CH), 66.1 (CH), 52.1
CO CH ), 51.9 (CO CH ), 50.1 (COCH ), 48.4 (COCH ), 17.8
CCH ), 17.3 (CCH ), HRMS (+ES): m/z Calcd. for C12
1
5 ¥ 150 mm Omnifit column filled with QP-BZA resin (25 g) and
then through a second 15 ¥ 150 mm Omnifit column containing
PS-NMe IO 4 (17 g). The output of the reactor was collected
2
3
2
3
3
3
3
+
3
H
20NaO
8
3
4
MNa ): 315.1056. Found: 315.1074.
until no further product was eluted and the solvent removed
in vacuo to yield the product as a crystalline off-white solid (38 g,
17
Polymer-supported periodate 4
2
5
7
C
(
5%). [a]D -138.3 (c 1.0 in CHCl
requires C, 49.31; H, 6.90%); n˜ max/cm (neat) 2991w
C–H), 2949w (C–H), 1736s (C=O), 1110s (C–O), 1023s (C–O);
3
); (Found: C, 49.35; H, 6.90.
-
1
Amberlite IRA 900, chloride form (25 g, 100 mmol, 4 mmol g-1
loading) was added to a solution of sodium metaperiodate (20 g,
94 mmol) in H O (200 mL). The mixture was agitated at 250 rpm
for 18 h using an orbital shaker and then filtered. The resin was
then added to a further solution of sodium metaperiodate (20 g,
12
H
20
O
8
1
H NMR (400 MHz, CDCl
6H, s, 2 ¥ CO CH ), 3.32 (6H, s, 2 ¥ OCH
); C NMR (100 MHz, CDCl ): d
= 168.4 (2 ¥ CO), 99.2
), 48.5 (2 ¥ OCH ), 17.3
); HRMS (+ES): m/z Calcd. for C12 (MNa ):
15.1056. Found: 315.1035.
3
): d
H
= 4.53 (2H, s, 2 ¥ CH), 3.77
2
(
2
3
3
), 1.35 (6H, s, 2 ¥
1
3
CH
(
(
3
3
C
2 ¥ C), 68.8 (2 ¥ CH), 52.5 (2 ¥ OCH
2 ¥ CH
3
3
94 mmol) in H O (200 mL) and shaken for a further 24 h. The
2
2
2
+
3
H
20NaO
8
reaction was filtered and the resin washed with H O (600 mL),
3
THF (200 mL) and Et O (200 mL) and dried in vacuo to yield
-
1
-
the desired product 4 (28 g, 2.0 mmol g ). (Found: Cl , 0.00.
-
(
1
5R,6R)-dimethyl 5,6-dimethoxy-5,6-dimethyl-5,6-dihydro-
,4-dioxine-2,3-dicarboxylate 2
PS–NMe
3
IO
4
requires Cl , 0.00).
1
7
Polymer-supported thiosulfate 5
A solution of n-BuLi (1.6 M in hexanes, 1.38 mL, 2.2 mmol) was
added dropwise to a stirred solution of diisopropylamine (0.35 mL,
Amberlyst A900, chloride form (11 g, 46.4 mmol) was added to a
◦
2
.5 mmol) in THF (3 mL) at -78 C. A solution of BDA tartrate
solution of sodium thiosulfate (36 g, 230 mmol) in H O (200 mL)
and the mixture agitated at 250 rpm for 16 h using an orbital
shaker. The reaction was filtered, the resin washed with H
(100 mL), THF (200 mL) and Et O (200 mL) and dried in vacuo to
give the desired product 5 (8.3 g, 3.4 mmol g ). (Found: N, 4.81.
PS–NMe requires N, 4.76).
2
1
(292 mg, 1.0 mmol) in THF (3 mL) was then added dropwise
over 10 min. A second solution of iodine (254 mg, 1 mmol) in
2
O
◦
THF (8 mL) was prepared and also held at -78 C. The two stock
2
-
1
-1
solutions were each pumped at 0.5 mL min (combined flow rate
-
1
of 1.0 mL min ), mixed in a T-piece and allowed to warm to
RT through a 14 mL PTFE coil. The crude solution was passed
through a 10 ¥ 100 mm Omnifit column packed with QP-SA resin
3
S
2
O
3
(1S,2S)-1,2-bis((2R,5R,6R)-5,6-dimethoxy-5,6-dimethyl-1,4-
22
(
1 g) and then through a second 10 ¥ 100 mm Omnifit column
containing PS-NMe 5 (2 g) and a short plug of silica gel
0.5 g). The output of the reactor was collected until no further
product was observed and the solvent removed in vacuo to yield
dioxan-2-yl)ethane-1,2-diol 6
3
S
2
O
3
D-mannitol (1 g, 5.5 mmol), CSA (127 mg, 0.6 mmol) and trimethyl
orthoformate (2.41 mL, 22 mmol) were dissolved in MeOH (2 mL)
(
◦
2
5
and heated to 60 C for 1 min by microwave irradiation using
the product as pale yellow needles (188 mg, 65%). [a] -293 (c
D
a Biotage Initiator before being transferred to a 5 mL sample
loop. Butanedione (0.92 mL, 10.5 mmol) in MeOH (4 mL) was
also loaded into a second 5 mL sample loop. The two solutions
1
4
1
.0 in CHCl
3
); (Found: C, 49.94; H, 6.27. C12
H
18
O requires C,
8
-
1
9.65; H, 6.25%); n˜ max/cm (neat) 2960w (C–H), 1733m (C=O),
1
648m (C=C), 1133s (C–O), 1090s (C–O); H NMR (400 MHz,
-
1
were pumped at 0.07 mL min , mixed at a T-piece and passed
CDCl
3
): d
H
= 3.76 (6H, s, 2 ¥ CO
2
CH
3
), 3.31 (6H, s, 2 ¥ OCH
3
),
= 162.2
), 48.8 (2 ¥
); HRMS (+ES): m/z Calcd. for C12
MNa ): 313.0899. Found: 313.0922.
◦
1
3
through a 14 mL CFC reactor held at 40 C. The output was
1
(
.49 (6H, s, 2 ¥ CH
2 ¥ CO), 130.3 (2 ¥ C), 98.2 (2 ¥ CH), 51.8 (2 ¥ OCH
OCH ), 16.1 (2 ¥ CH
3
); C NMR (100 MHz, CDCl
3
): d
C
then passed through a 10 ¥ 100 mm Omnifit column filled with
QP-BZA resin (2 g). The solvent was evaporated in vacuo and the
residue purified by column chromatography (SP1, 6 : 4 Pet ether:
3
3
3
H
18NaO
8
+
(
-
1
EtOAc, 25 column volumes on 10 g silica, 15 mL min , R
f
0.15)
2
D
5
to yield the product 6 as a white foam (1.25 g, 55%). [a] -78.8 (c
(
2R,3S,5R,6R)-dimethyl 5,6-dimethoxy-5,6-dimethyl-1,4-
-
1
17
0.85, CHCl
(
3
); n˜ max/cm (neat) 3482br (OH), 2994w (C–H), 1114s
dioxane-2,3-dicarboxylate 3
1
C–O), 1033s (C–O); H NMR (400 MHz, CDCl
3
): d
H
= 4.10 (2H,
BDA-protected oxidised tartrate 2 (3.78 g, 13.0 mmol) was
ddd, J = 10.9, 6.2, 3.6 Hz, 2 ¥ CHH), 3.80-3.71 (4H, m), 3.65 (2H,
dd, J = 11.4, 3.5 Hz, 2 ¥ CHH), 3.28 (6H, s, 2 ¥ OCH ), 3.27
(6H, s, 2 ¥ OCH ), 2.75 (2H, br, 2 ¥ OH), 1.29 (6H, s, 2 ¥ CCH ),
dissolved in MeOH (30 mL) and pumped at a flow rate of
3
1
3
mL min- through a 5% Rh/Al
2
O
3
cartridge installed in the
3
3
This journal is © The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 1588–1595 | 1593

View Article Online
1
3
1
(
4
.28 (6H, s, 2 ¥ CCH
2 ¥ C), 98.1(2 ¥ C), 69.5 (2 ¥ CH), 68.5 (2 ¥ CH), 60.9 (2 ¥ CH
8.1 (2 ¥ OCH ), 48.1 (2 ¥ OCH ), 17.8 (2 ¥ CH ), 17.5 (2 ¥ CH
3
); C NMR (100 MHz, CDCl
3
): d
C
= 99.3
was of sufficient purity to be used for further synthesis. A small
amount was isolated by column chromatography for analytical
2
),
);
3
3
3
3
purposes (SP1, Pet ether–Et
2
O, 25 column volumes on 10 g silica,
+
-1
25
-1
HRMS (+ES): m/z Calcd. for C18
H
34NaO10 (MNa ): 433.2050.
15 mL min , R
f
0.3). [a]D -101.3 (c 0.8, CHCl
3
); n˜ max/cm (neat)
1
Found: 433.2030.
2948w (C–H), 1374, 1114s (C–O), 1035s (C–O), 878m (C–Cl); H
NMR (400 MHz, CDCl
6.2 Hz, CH), 3.62-3.58 (2H, m, OCH
6.2 Hz, CHHCl), 3.38 (1H, dd, J = 11.3, 6.2 Hz, CHHCl), 3.29
3H, s, OCH ), 3.26 (3H, s, OCH ), 1.30 (3H, s, CH ), 1.28 (3H, s,
); C NMR (100 MHz, CDCl ): d
= 99.2 (C), 97.9 (C), 67.3
CH), 61.3 (CH Cl), 47.8 (OCH ), 47.7 (OCH ), 17.4 (CCH ), 17.2
CCH ); HRMS (+ES): m/z Calcd for C (MNa ):
47.0713. Found: 247.0733.
3
): d
H
= 4.08 (1H, ddd, J = 11.3, 9.2,
(
2R,5R,6R)-5,6-dimethoxy-5,6-dimethyl-1,4-dioxane-2-
2
), 3.43 (1H, dd, J = 11.3,
22
carbaldehyde 7
(
3
3
3
BDA-protected D-mannitol (350 mg, 1.3 mmol) was dissolved in
CH Cl –H O (9 : 1, 4 mL) and recycled through a 10 ¥ 100 mm
Omnifit column of PS-periodate 4 (1 g, 2 mmol) at 0.5 mL min
for 14 h. After this time the flow stream was directed through a
0 ¥ 100 mm Omnifit containing MgSO (500 mg) and the solvent
13
CH
(
(
3
3
C
2
2
2
2
3
3
3
-
1
+
3
9
H
17ClNaO
4
2
1
4
removed in vacuo to give aldehyde 7 as a colourless liquid (477 mg,
(
1
2R,3R)-2,3-dimethoxy-2,3-dimethyl-5-methylene-1,4-dioxane
0
2
5
-1
9
1
0%). [a]D -101.3 (c 0.8, CHCl
3
); n˜ max/cm (neat): 2949w (CH),
27
1
733w (C=O), 1110s (C–O), 1033s (C–O); H-NMR (400 MHz,
CDCl
CH), 3.73-3.65 (2H, m, CH
OCH ), 1.36 (3H, s, CCH
100 MHz, CDCl ): d
= 200.2 (CHO), 99.8 (C), 98.3 (C), 72.4
CH), 58.2 (CH ), 48.4 (OCH ), 48.2 (OCH ), 17.6 (CCH ), 17.5
CCH ).
3
): d
H
= 9.63 (1H, s, CHO), 4.32 (1H, dd, J = 10.1, 4.9 Hz,
), 3.30 (3H, s, OCH ), 3.24 (3H, s,
), 1.28 (3H, s, CCH ); C NMR
A 1 mL PEEK sample loop was loaded with a solution of BDA-
protected chloropropanediol 9 (224 mg, 1 mmol) in THF (1 mL)
and a second 1 mL loop loaded with a solution of KOtBu (224 mg,
2 mmol) in THF (1 mL). The two solutions were mixed at a T-
piece at a combined flow rate of 0.2 mL min (0.1 mL min
per pump) and flowed through a 14 mL CFC reactor heated to
2
3
1
3
3
3
3
(
(
(
3
C
-
1
-1
2
3
3
3
3
◦
7
0 C. The output stream was collected in H
2
O (2 mL), diluted with
(
2R,5R,6R)-methyl 5,6-dimethoxy-5,6-dimethyl-1,4-dioxane-2-
Et O (10 mL), separated, and the organic phase dried over sodium
sulfate to yield the title compound 10 as a colourless oil (152 mg,
2
22
carboxylate 8
2
5
-1
8
1
1%). [a]D -166.9 (c 1.2, CHCl
3
); n˜ max/cm (neat) 2950w (C–H),
BDA-protected D-mannitol (350 mg, 1.3 mmol) was dissolved in
MeOH (4 mL) and recycled through a 10 ¥ 100 mm Omnifit
column of PS-periodate 4 (1 g, 2 mmol) at 0.5 mL min for 14 h.
After all the starting material had been consumed (monitored
by TLC, Pet ether: EtOAc (6 : 4), R
was directed into a column of PS-pyridinium perbromide (1 g,
2
1
659m (C=C), 1106s (C–O), 1039s (C–O); H NMR (400 MHz,
CDCl
4
OCH
CCH
3
): d = 4.45 (1H, s, CHH), 4.27 (1H, d, J = 13.1 Hz, OCHH),
-
1
.26 (1H, s, CHH), 3.97 (1H, d, J = 13.3 Hz, OCHH), 3.34 (3H, s,
3
), 3.29 (3H, s, OCH
3
), 1.37 (3H, s, CCH
); C NMR (100 MHz, CDCl
), 98.1 (CCH ), 93.2 (OCH ), 59.728 (CCH
8.3 (OCH ), 17.6 (CCH ), 17.4 (CCH ).
3
), 1.30 (3H, s,
), 101.1
f
0.4) the output stream
13
3
3
): d = 152.6 (CCH
2
(
CCH
3
3
2
2
), 48.5 (OCH ),
3
-
1
.7 mmol) at a flow rate of 0.05 mL min . Excess solvent was
4
3
3
3
removed in vacuo to yield ester 8 as a colourless oil (489 mg,
8
7
6%). (Found: C, 51.53; H, 7.70. C10
H
18
O
6
requires C, 51.27; H,
27
(
5R,6R)-5,6-dimethoxy-5,6-dimethyl-1,4-dioxan-2-one 12
2
5
); n˜ max/cm-1 (neat) 2952w (C–
.75%); [a]D -119.6 (c 1.22, CHCl
3
H), 1764m (C=O), 1734m (C=O), 1140s (C–O), 1034s (C–O);
BDA-protected alkene 10 (188 mg, 1.0 mmol) and NMO (175 mg,
1.5 mmol) were dissolved in THF–H O (2 : 1, 4.5 mL) and recycled
1
H NMR (400 MHz, CDCl
3
): d
H
= 4.56 (1H, dd, J = 11.2,
2
3
.7 Hz, CH), 3.85 (1H, t, J = 11.3 Hz, CHH), 3.76-3.71 (4H,
m, CO CH , CHH), 3.31 (3H, s, OCH ), 3.26 (3H, s, OCH ),
.37 (3H, s, CCH ), 1.29 (3H, s, CCH ); C NMR (100 MHz,
through a 10 ¥ 100 mm OmniFitꢀR column packed with a mixed
TM
2
3
3
3
bed of OsEnCat (250 mg, 0.05 mmol) and PS-periodate resin 4
1
3
-1
1
3
3
(1.5 g, 3 mmol) at 0.3 mL min for 7 h at room temperature. After
CDCl
3
): d
C
= 169.2 (CO), 99.7 (C), 98.0 (C), 67.1 (CH), 59.9
this time the flow stream was directed through a 10 ¥ 150 mm
OmniFitꢀR column containing QP-SA (1 g, 3 mmol) and QP-
TU (1 g, 3 mmol). Excess solvent was removed in vacuo to yield
(
1
2
CH
2
), 52.2 (CO
7.4 (CCH ); HRMS (+ES): m/z Calcd. for C10
57.1001. Found: 257.1007.
2
CH
3
), 48.4 (OCH
3
), 48.1 (OCH
3
), 17.6 (CCH ),
3
+
3
H
18NaO (MNa ):
6
2
5
glycolate 12 as a pale yellow oil (146 mg, 85%). [a] -241 (c 0.83,
D
-
1
CHCl
3
); n˜ max/cm (neat): 2952w (C–H), 1752m (C=O), 1107s (C–
1
(
2R,3R)-5-(chloromethyl)-2,3-dimethoxy-2,3-dimethyl-1,4-
O), 1032s (C–O); H NMR (400 MHz, CDCl
3
): d
H
= 4.31 (1H,
27
dioxane 9
d, J = 17.6 Hz, CHH), 4.15 (1H, d, J = 17.6 Hz, CHH), 3.44
(
(
1
(
(
3H, s, OCH
3H, s, CCH
05.0 (C), 97.8 (C), 60.3 (CH
CCH ), 16.9 (CCH ); HRMS (+ES): m/z Calcd. for C
3
), 3.31 (3H, s, OCH
3
), 1.50 (3H, s, CCH
); C NMR (100 MHz, CDCl ): d
= 167.5 (CO),
), 50.3 (OCH ), 49.1 (OCH ), 17.8
3
), 1.39
(
R)-Chloropropanediol (2 g, 18 mmol) and CSA (417 mg,
13
3
3
C
1
.8 mmol) were dissolved in MeOH (6 mL). A second solution
2
3
3
was prepared containing butanedione (1.93 mL, 22 mmol) and
trimethyl orthoformate (4.38 mL, 40 mmol) in MeOH (3 mL).
3
+
3
8
H
14NaO
5
MNa ): 213.0739. Found: 213.0722.
The two solutions were mixed at a T-piece at a flow rate of
.14 mL min- (0.07 mL per channel) and directed into a CFC
1
0
(
R)-3-methoxy-3-(2-oxopropoxy)butan-2-one 13
◦
reactor heated at 90 C. The solution was then passed through a
1
0 ¥ 100 mm Omnifit column filled with QP-BZA resin (2 g) and
BDA-protected alkene 10 (188 mg, 1.0 mmol) was dissolved in
THF–H O (2 : 1, 4.5 mL) and sodium periodate (642 mg, 3 mmol)
the output collected. Evaporation of the solvent in vacuo yielded
2
the desiredproduct 9 as a colourless oil (426 mg, 95%). The product
was added. The reaction was stirred at room temperature for 2 h,
1
594 | Org. Biomol. Chem., 2010, 8, 1588–1595
This journal is © The Royal Society of Chemistry 2010

View Article Online
monitored with the ReactIR 45 m fiber optic probe. After this time
the reaction mixture was filtered, diluted with CH Cl (10 mL) and
washed with sat. aq. sodium metabisulfite (10 mL). The organic
I. R. Baxendale and S. V. Ley, Org. Process Res. Dev., 2007, 11, 399–405;
I. R. Baxendale, C. M. Griffiths-Jones, S. V. Ley and G. K. Tranmer,
Synlett, 2006, 3, 427–430; I. R. Baxendale, J. Deeley, C. M. Griffiths-
Jones, S. V. Ley, S. Saaby and G. K. Tranmer, Chem. Commun., 2006,
2566–2568; M. Baumann, I. R. Baxendale, S. V. Ley, C. D. Smith and
G. K. Tranmer, Org. Lett., 2006, 8, 5231–5234.
S. V. Ley, D. K. Baeschlin, D. J. Dixon, A. C. Foster, S. J. Ince, H. W.
Priepke and D. J. Reynolds, Chem. Rev., 2001, 101, 53–80.
S. V. Ley and A. Polara, J. Org. Chem., 2007, 72, 5943–5959.
6 M. K. Cheung, N. L. Douglas, B. Hinzen, S. V. Ley and X. Pannecoucke,
Synlett, 1997, 257–260.
N. L. Douglas, S. V. Ley, U. Lucking and S. L. Warriner, J. Chem. Soc.,
Perkin Trans. 1, 1998, 51–65.
L. Green, B. Hinzen, S. J. Ince, P. Langer, S. V. Ley and S. L. Warriner,
Synlett, 1998, 440–442.
S. V. Ley, D. J. Dixon, R. T. Guy, F. Rodriguez and T. D. Sheppard,
Org. Biomol. Chem., 2005, 3, 4095–4107.
2
2
phase was separated and dried over MgSO
a yellow oil (162 mg, 93%). [a]D -9.5 (c 1.4, CHCl
); n˜ max/cm
4
to yield ketone 13 as
2
5
-1
3
4
5
(
neat) 2928m (C–H), 1730s (C=O), 1388w, 1119s (C–O), 1044s
1
(
C–O); H NMR (400 MHz, CDCl
3H, s, OCH ), 2.26 (3H, s, CH ), 2.21 (3H, s, CH
). C NMR (100 MHz, CDCl ): d
= 206.0 (CO), 205.6
), 50.0 (OCH ), 26.5 (CH ), 25.9 (CH ),
3
): d
H
= 4.02 (2H, s, CH
2
), 3.27
(
3
3
3
), 1.43 (3H, s,
1
3
CCH
3
3
C
7
8
9
(CO), 105.0 (C), 68.1 (CH
2
3
3
3
1
9.9 (CCH ).
3
Acknowledgements
1
0 K. R. Knudsen, A. F. Stepan, P. Michel and S. V. Ley, Org. Biomol.
Chem., 2006, 4, 1471–1473.
We gratefully acknowledge AstraZeneca (CFC), the Royal Society
IRB) and the BP 1702 Fellowship (SVL) for funding.
(
11 D. J. Dixon, S. V. Ley, A. Polara and T. Sheppard, Org. Lett., 2001, 3,
3
749–3752.
1
1
2 M. T. Barros, A. J. Burke, J. D. Lou, C. D. Maycock and J. R. Wahnon,
J. Org. Chem., 2004, 69, 7847–7850.
Notes and references
3 D. J. Dixon, A. C. Foster and S. V. Ley, Org. Lett., 2000, 2, 123–125.
1
C. J. Li and B. M. Trost, Proc. Natl. Acad. Sci. U. S. A., 2008, 105,
3197–13202; P. A. Wender, V. A. Verma, T. J. Paxton and T. H. Pillow,
14 D. J. Dixon, S. V. Ley, S. Lohmann and T. D. Sheppard, Synlett, 2005,
1
481–484.
Acc. Chem. Res., 2008, 41, 40–49.
15 J. A. Bull, E. P. Balskus, R. A. J. Horan, M. Langner and S. V. Ley,
Angew. Chem., Int. Ed., 2006, 45, 6714–6718.
16 J. A. Bull, E. P. Balskus, R. A. J. Horan, M. Langner and S. V. Ley,
Chem.–Eur. J., 2007, 13, 5515–5538.
2
For selected reviews in the area of flow chemistry and microreactors
see: X. Y. Mak, P. Laurino and P. H. Seeberger, Beilstein J. Org. Chem.,
2
009, 5, DOI: 10.3762/bjoc.5.19; R. L. Hartman and K. F. Jensen,
Lab Chip, 2009, 9, 2495–2507; K. Geyer, T. Gustafsson and P. H.
Seeberger, Synlett, 2009, 2382–2391; I. R. Baxendale, J. J. Hayward,
S. Lanners, S. V. Ley and C. D. Smith, in Microreactors in Organic
Synthesis and Catalysis, ed. T. Wirth, Wiley-VCH, Weinheim, 2008,
ch. 4.2, pp. 84–122; F. Benito-Lopez, R. J. Egberink, D. N. Reinhoudt
and W. Verboom, Tetrahedron, 2008, 64, 10023–10040; C. Wiles and P.
Watts, Eur. J. Org. Chem., 2008, 1655–1671; J.-I. Yoshida, A. Nagaki
and T. Yamada, Chem.–Eur. J., 2008, 14, 7450–7459; B. Ahmed-Omer,
J. C. Brandt and T. Wirth, Org. Biomol. Chem., 2007, 5, 733–740; I. R.
Baxendale and S. V. Ley, in New Avenues to Efficient Chemical Synthesis
17 C. F. Carter, I. R. Baxendale, M. O’Brien, J. B. J. Pavey and S. V. Ley,
Org. Biomol. Chem., 2009, 7, 4594–4597.
18 www.uniqsis.com.
19 For a comprehensive review of solid supported reagents see: S. V. Ley,
I. R. Baxendale, R. N. Bream, P. S. Jackson, A. G. Leach, D. A.
Longbottom, M. Nesi, J. S. Scott, R. I. Storer and S. J. Taylor, J. Chem.
Soc., Perkin Trans. 1, 2000, 3815–4195.
20 C. R. Harrison and P. Hodge, J. Chem. Soc., Perkin Trans. 1, 1982,
509–511.
21 J. J. Parlow, B. L. Case and M. S. South, Tetrahedron, 1999, 55, 6785–
-
Emerging Technologies, ed. P. H. Seeberger and T. Blume, Springer,
6796.
Berlin Heidelberg, 2007, pp. 151–185; T. L. LaPorte and C. Wang,
Curr. Opin. Drug Discov. Dev., 2007, 10, 738–745; I. R. Baxendale, J. J.
Hayward, S. V. Ley and G. K. Tranmer, Chem. Med. Chem., 2007, 2,
22 S. V. Ley and P. Michel, Synthesis, 2004, 147–150.
23 P. Michel and S. V. Ley, Angew. Chem., Int. Ed., 2002, 41, 3898–
3901.
7
68–788; B. P. Mason, K. E. Price, J. L. Steinbacher, A. R. Bogdan and
24 P. Michel and S. V. Ley, Synthesis, 2003, 1598–1602.
25 N. Hebert, A. Beck, R. B. Lennox and G. Just, J. Org. Chem., 1992, 57,
1777–1783.
D. T. McQuade, Chem. Rev., 2007, 107, 2300–2318; A. Kirschning, W.
Solodenko and K. Mennecke, Chem.–Eur. J., 2006, 12, 5972–5990; P.
Hodge, Curr. Opin. Chem. Biol., 2003, 7, 362–373.
-1
26 Pyridinium tribromide polymer bound, 2.7 mmol g , available from
Sigma Aldrich, product number 82795.
3
Selected examples of flow chemistry from our laboratories: J.
Sedelmeier, S. C. Schou, S. V. Ley and I. R. Baxendale, Chem. Eur. J.,
27 E. Diez, D. J. Dixon and S. V. Ley, Angew. Chem., Int. Ed., 2001, 40,
2
009, 16, 89–94; M. Baumann, I. R. Baxendale, L. J. Martin and
2906–2909.
S. V. Ley, Tetrahedron, 2009, 65, 6611–6625; A. Palmieri, S. V. Ley,
K. Hammond, A. Polyzos and I. R. Baxendale, Tetrahedron Lett.,
28 G. Cainelli, M. Contento, F. Manescalchi and L. Plessi, Synthesis,
1989, 45–47; G. Cainelli, M. Contento, F. Manescalchi and L. Plessi,
Synthesis, 1989, 47–48.
2
009, 50, 3287–3289; J. Sedelmeier, S. V. Ley, H. Lange and I. R.
Baxendale, Eur. J. Org. Chem., 2009, 4412–4420; I. R. Baxendale, S. V.
Ley, A. C. Mansfield and C. D. Smith, Angew. Chem., Int. Ed., 2009, 48,
29 S. V. Ley, C. Ramarao, A. L. Lee, N. Ostergaard, S. C. Smith and I. M.
Shirley, Org. Lett., 2003, 5, 185–187.
30 (a) ReactIR 45m with fibre optic cable and diamond probe available
from Mettler Toledo; www.mt.com; (b) The ReactIR 45m can now be
used in-line with a newly developed flow cell; C. F. Carter, H. Lange,
S. V. Ley, I. R. Baxendale, B. Wittkamp, N. L. Gaunt and J. G. Goode,
Org. Process Res. Dev., DOI: 10.1021/op900305v.
4
017–4021; M. Baumann, I. R. Baxendale, S. V. Ley, N. Nikbin, C. D.
Smith and J. P. Tierney, Org. Biomol. Chem., 2008, 6, 1577–1586; M.
Baumann, I. R. Baxendale and S. V. Ley, Synlett, 2008, 14, 2111–2114;
C. J. Smith, F. J. Iglesias-Sig u¨ enza, I. R. Baxendale and S. V. Ley, Org.
Biomol. Chem., 2007, 5, 2758–2761; C. H. Hornung, M. R. Mackley,
This journal is © The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 1588–1595 | 1595