FLOW CHEMISTRY
---'
-------------------------s�b�;;,- i�m A!
(°) N H moveable needle
�
HPLC pump
Waste
Back pressure valve outlet pressure sensor
Gas-liquid separator
Figure 3- Parallel sub-systems designed for rapid optimisation & scale-up
the system receives instant analytical feedback to intelligently guide the next parameter adjustments.
This self-optimising loop reduces the total number of required runs to find the absolute best reaction conditions, with each automated experiment taking only 15-20 minutes.
In the experiment depicted in Figures 1 and 2, our aim was to optimise the reduction of nitroquinoline 1 into the corresponding amine 2 and the tetrahidroquinoline analogue 3. Using a Bayesian algorithm on the data immediately available from the in-line IR spectrometer, we were able to determine the optimum conditions in nine steps.
By combining automation, in-line analytics, and intelligent experimental design, the platform reduced both development time and experimental effort while delivering reproducible optimisation results.
Case study 2
In traditional batch chemistry, scaling up a reaction from a round-bottom flask to a larger reactor is notoriously hazardous and time-consuming. Heat transfer, mass transfer and mixing dynamics change unpredictably at larger volumes, often leading to dangerous runaway exotherms or unexpected by-product profiles that
Heating block( 1 % Pt / C CatCart)
Spinsolve NMR spectrometer
Gas-liquid mixer
Phoenix · ™11. flow reactor
Inlet pressure S e
MeOH
0
Gas-liquid mixer
• Phoenix TM II. flow reactor
Washing solvent
Pressure module ™
, '----------- Fl�� Path
force chemists to re-optimise the reaction from scratch.
FlowChemFleet eliminates these scale-up bottlenecks by bridging the gap between discovery and production. By networking a smallscale and a larger-scale flow system within the same fleet, parallel optimisation can be performed swiftly. Because flow chemistry parameters, such as residence time and temperature, translate almost directly across scales, the small-scale system can act as a pilot to find ideal conditions that are instantly applied to the larger system without the typical safety risks.
While integrated, automated liquid handlers can precisely collect samples for high-resolution offline analysis, and incorporating in-line analytical tools provides instant, real-time feedback. This allows chemists to monitor reaction progression dynamically, ensuring safe, rapid and predictable scale-up.
Using sub-system A( small-scale, Figure 3), the automated injection of nitrobenzene was achieved by a Brooks liquid handler. Small portions of this solution were injected into an H-Cube Advance instrument, which performed the reduction of the nitromoiety to the primary amine.
The system was extended by a phase separator, which eliminated
I
A + i(( i I LY
- rI
,. '•
:::::_ S� /J ����fem
I----,
B.
I
Tempera ture: moniton ��-
_,'
N
C)
F
,• •
I I the excess hydrogen gas from the stream. The liquid line, containing the crude product, was fed into a Magritek benchtop NMR spectrometer, allowing live monitoring of the conversion via
19
F-NMR spectroscopy. After testing ten different parameter combinations, we achieved a full conversion and 98 % selectivity.
Data collected by sub-system A could immediately be used for the rapid optimisation of a similar synthesis on a larger-scale, in subsystem B. This setup featured the synthesis of the key intermediate for a Linezolid analogue. With the information constantly arriving from sub-system B, a rapid optimisation of the hydrogenation step could be carried out, using only six experimental data points within less than three hours of operation time.
Outlook
The future of flow chemistry extends beyond advances in individual reactor technologies. As research laboratories continue to adopt digital workflows, flexible platforms that can readily incorporate new analytical tools, automation technologies, and rapid reaction optimisation will play an important role in the future of chemical research.
By combining modular instrumentation and intelligent automation, platforms like FlowChemFleet demonstrate how the next generation of flow chemistry laboratories can become more autonomous, more reproducible and better suited to the growing demands of modern process development. ●
*- THS System Controller, THS ReAction, CatCart, H-Genie and H-Cube are registered trademarks of ThalesNano; FlowChemFleet and Phoenix are unregistered trademarks of ThalesNano
Dr Nándor Kánya
DIRECTOR OF CHEMISTRY
THALESNANO INC. k + 36 3053 83439 J nandor. kanya @ thalesnano. com j www. thalesnano. com
44 SPECIALITY CHEMICALS MAGAZINE ESTABLISHED 1981