The Journal of Organic Chemistry
Article
with 4 g cartridges packed with monodisperse 15 μm spherical C18-
functionalized silica particles (PF-15C18XS-F0004). All organic
solvents used to compose eluents were purchased from Fisher
Scientific at SLR grade, except for acetonitrile (MeCN), isopropanol
(IPA), and n-hexane, which were HPLC grade. Deionized water used
for reversed-phase chromatography is produced on-site. Up to four
2.5 L Winchester flasks of solvent could be connected to the
instrument using color-coded tubing. Waste was collected to an empty
2.5 L Winchester flask. The volume of solvent reservoirs and waste
was automatically monitored, triggering an error that paused the
current chromatography method if the solvent level was too low or if
the waste volume was too high. The instrument was equipped with a
UV absorbance- (200−400 nm) and an ELS detector that could be
used to trigger fraction collection when the signal intensity surpassed
a user-defined threshold. Fractions were collected to one of three
racks of 48 test tubes (150 × 18 mm) with RFID sensors.
Off-line method development and optimization were performed by
manually injecting the crude reaction mixture through the system’s
injection port using a 1, 5, or 10 mL plastic syringe. Continuous
purification procedures were performed by connecting the output of
the flow reactor system directly to the injection port using a Luer lock
adaptor (male 1/4−28 UNF to male Luer, PEEK). The feed flow rate
of the flow reactor was synchronized to the chromatography method
duration to fill the sample loop with the desired volume of the crude
reaction mixture, typically 1−6 mL depending on the system.
Continuous purification sequences were generally initiated with a
″blank″ run, which ensured that the position of the valve system was
correct and prevented loading the same sample loop twice by mistake.
The chromatogram raw data could be obtained from the system
data files as a Microsoft Excel CSV file, which was imported and
processed in the OriginLab (2018) software to produce the stacked
plots presented in the manuscript. Processed chromatographs in the
absorbance in arbitrary units (a.u.).
port of the chromatograph. The chromatograph system was
programmed to run the same optimized chromatography method
50 times in a continuous sequence, alternating between the two
sample-loop/column systems. The systems were paused approx-
imately every 20 runs to refill solvent reservoirs, stock solution, and
empty waste. After 50 runs, the two silica cartridges were flushed with
IPA until the UV absorbance baseline achieved a steady state. The
instrument was switched off and left overnight with the silica
cartridges still connected and filled with IPA. The following day, the
chromatograph system was switched on and the silica cartridges were
re-equilibrated with ethyl acetate and n-hexane. The 50-run sequence
was repeated as before.
The fractions from every 10th run were collected and combined,
dried under vacuum, and analyzed for mass recovery. The 10th and
100th runs were analyzed by LC−MS, both displaying >99% purity.
1
The H, 13C{1H} NMR spectra and LC−MS traces were compared
against the commercial reagents and literature to confirm identity.57,58
Both compounds were recovered as white crystalline solids. The
overall recovery of 1 was 427 mg (3.44 mmol, 73%), and that of 2 was
767 mg (2.91 mmol, 86%). Characterization of the recovered
materials is presented in the Supporting Information (see Sections 7
Continuous Flow Photosensitized Oxidation of Fmoc-Met-
OH (3) to Fmoc-Met(O)-OH (4). Fmoc-L-Met-OH (3, 371 mg, 1.00
mmol) and 4,7-diphenyl-2,1,3-benzothiadiazole (5, 28 mg, 0.10
mmol) were added to a round-bottom flask, wrapped in tin foil to
exclude ambient light, and dissolved in MeCN (20 mL). A magnetic
stirrer bar was added to the flask before sealing with a septum. The
solution was stirred in the dark for 10 min before connecting to the
flow system illustrated in Scheme 2 using a stainless-steel needle to
pierce the septum and a second needle to balance pressure. The
reaction solution was pumped at 0.1 mL/min to a T-junction mixer
that was intercepted at 90° by a 0.9 mL/min flow of air. The resulting
gas−liquid slug flow passes through the UV-150 photoreactor
equipped with a 10 mL reactor coil and LED module (radiant
power: ∼18 W @ 420 nm, 3.37 mmol/min, lamp power: 60 W) with
a net velocity of 1 mL/min to achieve a residence time of 10 min.
After exiting the reactor, the flow stream was pumped into the center
of a test tube to collect the solution phase and vent the remnant gas
phase. After a sufficient volume of solution had collected (∼1 mL), a
third pump was used to flow the crude reaction mixture at 0.1 mL/
min from the base of the test tube to the injection port of the
chromatograph. The instrument was automated to run a sequence of
the optimized method 11 times continuously, the first of which was a
blank, followed by 10 continuous purifications of the crude reaction
mixture feed. The method duration was 13.5 min, purifying 1.35 mL
of the crude material each run for a total of 13.5 mL over 2 h and 20
min of continuous operation. Fmoc-L-methionine-D,L-sulfoxide (4)
was recovered from the collected fractions and dried under vacuum to
yield an amorphous off-white solid. NMR spectra were compared with
literature to confirm identity.39 5 was recovered separately (10 mg,
53%) as a crystalline yellow solid with identical LC chromatograms,
NMR, and mass spectra to the original synthesis procedure (vide
infra).
LC−MS Instrumentation and Analysis. LC−MS analysis was
performed with a Shimadzu Nexera-i LC-2040C 3D Plus liquid
chromatograph equipped with a Shim-pack Velox SP-C18 column
(2.7 μm, 4.6 × 150 mm), autosampler, and photodiode array detector.
The chromatograph was coupled to a Shimadzu LCMS-2020 mass
spectrometer. The stationary phase was equilibrated with LC−MS-
grade water (solvent A) and acetonitrile (solvent B) with 0.1% LC−
MS-grade formic acid additive. Eluent was pumped at 1 mL/min, and
the column was heated to 40 °C. Samples were dissolved in a suitable
solvent (MeCN, IPA, MeOH) at <1 mg/mL concentration. The same
HPLC chromatography method was used to analyze all samples
reported. The gradient method program was as follows:
0:00−1:00 min:s, A/B (95:5);
1:00−5:00 min:s, A/B (95:5 → 5:95);
5:00−7:00 min:s, A/B (5:95);
7:00−7:10 min:s, A/B (5:95 → 95:5);
7:10−10:00 min:s, A/B (95:5); method end.
The eluate was analyzed by dual electrospray ionization-
atmospheric pressure chemical ionization (DUI-ESI/APCI). Mass
spectra were collected by detecting ions over a range of 100−500 m/z
in both positive and negative phases simultaneously. Compounds that
did not ionize by ESI were subsequently analyzed by atmospheric
solid analysis probe mass spectrometry (ASAP-MS), detecting ions
over a range of 0−1000 m/z.
Fmoc-L-methionine-D,L-sulfoxide (4).39 Yield: 215 mg (0.55
mmol, 82%). Amorphous, off-white solid. 1H NMR (400 MHz,
DMSO-d6) δ ppm 7.89 (d, J = 7.5 Hz, 4H), 7.82−7.58 (m, 6H), 7.42
(td, J = 7.5, 1.2 Hz, 4H), 7.33 (tt, J = 7.4, 1.3 Hz, 4H), 4.31 (d, J = 7.1
Hz, 4H), 4.23 (t, J = 6.9 Hz, 2H), 4.14−4.03 (m, 2H), 2.91−2.60 (m,
4H), 2.55 (s, 3H), 2.53 (s, 3H), 2.18−2.07 (m, 2H), 2.05−1.88 (m,
2H). 13C{1H} NMR (101 MHz, DMSO-d6) δ ppm 173.0, 172.9,
156.10, 156.05, 143.8, 143.7, 140.7 (x2), 127.6 (x2), 127.0 (x2), 125.2
(x2), 120.1 (x2), 65.6 (x2), 53.1, 52.8, 49.8, 49.5, 46.7, 38.1, 37.8,
24.2, 23.7. LC−MS: tR = 5.020 min, peak area = 97.2%; MS (DUI-
ESI/APCI) calcd for [C20H22NO5S]+ ([M + H]+) = 388.12, (ESI+)
found: 388.15.
Proof-of-Principle Study. 4-Methoxyphenol (1, 4.966 g, 40.00
mmol) and 2,5-dibromo-p-xylene (2, 10.559 g, 40.00 mmol) were
added to a dry, 500 mL flask and dissolved in a mixture of n-hexane
and ethyl acetate (50/50 (v/v) 400 mL) to produce a 0.1 M stock
solution w.r.t. each of the reagents. An additional stock solution was
prepared, as necessary, following the same ratios. The flask was loaded
with a magnetic stirrer bar and sealed with a septum before stirring
until a homogeneous solution was obtained. The flask was connected
to the flow reactor via stainless-steel needles that pierced the septum
and connected to the PFA tubing (1 mm I.D.), as illustrated in
Scheme 1 of the manuscript. The solution was pumped via one the
flow reactor’s V-3 peristaltic pumps at 0.5 mL/min to the injection
Synthesis of the 4,7-Diphenyl-2,1,3-benzothiadiazole (5)
Photocatalyst. 4,7-Dibromo-2,1,3-benzothiadiazole (294 mg, 1.00
mmol), phenylboronic acid (244 mg, 2.00 mmol), and Pd(PPh3)4 (23
mg, 0.02 mmol) were added to a two-necked round-bottom flask
L
J. Org. Chem. XXXX, XXX, XXX−XXX