Reconfigurable Flow Platform for Automated Reagent Screening and Autonomous Optimization for Bioinspired Lignans Synthesis

Article abstract of DOI:10.1021/acs.joc.9b02263

Naturally occurring benzoxanthenones, which belong to the vast family of lignans, are promising biologically relevant targets. They are biosynthetically produced by the oxidative dimerization of 2-propenyl phenols. In this manuscript, we disclose a powerf

Full text of DOI:10.1021/acs.joc.9b02263

Subscriber access provided by - Access paid by the | UCSF Library
Article
Reconfigurable Flow Platform for Automated Reagent Screening
and Autonomous Optimization for Bio-inspired Lignans Synthesis
Ehu Camille Aka, Eric Wimmer, Elvina Barre, Natarajan Vasudevan, Daniel Cortés-
Borda, Tchirioua Ekou, Lynda Ekou, Mireia Rodriguez-Zubiri, and François-Xavier Felpin
J. Org. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.joc.9b02263 • Publication Date (Web): 30 Sep 2019
Downloaded from pubs.acs.org on October 5, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the
course of their duties.

Page 1 of 32
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Reconfigurable Flow Platform for Automated Reagent Screening and
Autonomous Optimization for Bio-inspired Lignans Synthesis
‡
‡
‡
‡
Ehu Camille Aka, Eric Wimmer, Elvina Barré, Natarajan Vasudevan, Daniel Cortés-
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
‡
†
†
‡
Borda, Tchirioua Ekou , Lynda Ekou, Mireia Rodriguez-Zubiri, and François-Xavier
Felpin*‡
‡
Université de Nantes, CEISAM, CNRS UMR 6230, 2 rue de la Houssinière, 44322 Nantes
Cedex 3, France
†
Université Nangui Abrogoua, Laboratoire de Thermodynamique et de Physico-Chimie du
Milieu, 02 BP 801 Abidjan 02, Côte d’Ivoire
■
ABSTRACT GRAPHIC
■
ABSTRACT
Naturally occurring benzoxanthenones which belongs to the vast family of lignans, are
promising biologically relevant targets. They are biosynthetically produced by the oxidative
dimerization of 2-propenyl phenols. In this manuscript, we disclose a powerful automated flow-
based strategy for identifying and optimizing a cobalt-catalyzed oxidizing system for the bio-
inspired dimerization of 2-propenyl phenols. We designed a reconfigurable flow reactor
associating on-line monitoring and process-control instrumentation. Our machine was first
configured as an automated screening platform to evaluate a matrix of 4 catalysts (plus the
blank) and 5 oxidant (plus the blank) at two different temperatures, resulting in an array of 50
reactions. The automated screening was conducted on micromole scale at a rate of one fully
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 2 of 32
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
characterized reaction every 26 minutes. After having identified the most promising cobalt-
catalyzed oxidizing system, the automated screening platform was straightforwardly
reconfigured to an autonomous self-optimizing flow reactor by implementation of an
optimization algorithm in the closed-loop system. The optimization campaign allowed the
determination of very effective experimental conditions in a limited number of experiments
which allowed to prepare natural products carpanone and polemannone B as well as synthetic
analogues.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
■
INTRODUCTION
Naturally occurring benzoxanthenones constitute a small group of natural products
1
isolated as racemates and belonging to the vast family of lignans. The most
representative members of benzoxanthenone-based natural products include carpanone
1
, polemannone A-C 2-4 and sauchinone 5 and feature a highly oxygenated polycyclic
structure with five contiguous stereogenic centers (Figure 1). They have been identified
as biologically relevant targets with anti-inflammatory,2-3 hepatoprotective,4-5 and
6
antitumor properties. Synthetic analogues have also been reported to be potent
inhibitors of exocytosis and might be useful for the study of vesicular traffic.7-8
Therefore, the significant complexity of the benzoxanthenone skeleton associated to the
potential of these structures to interfere with several biological mechanisms, make these
structures appealing targets for synthetic chemists.
It has been postulated that the biosynthesis of the benzoxanthenone skeleton occurs through
1
the oxidative dimerization of the corresponding 2-propenyl phenol. This assumption was latter
supported by synthetic studies showing that the bio-inspired oxidative dimerization of 2-
propenyl phenols using a variety of oxidants such as PdCl2,9-10 [Co]/O2,11-12 CuCl /()-
2
sparteine/O2,13-14 PhI(OAc)27-8 and laccase, indeed furnishes the benzoxanthenone skeleton.
We recently described the multi-step synthesis of carpanone 1 in flow using a modular
autonomous flow reactor.16 Advantages of conducting reactions in flow with respect to
traditional batch reactors include higher mass and heat transfers, better reproducibility,
improved kinetics, safer experimental conditions and easier automation.17-24
15
In our approach, all discrete steps of the flow synthetic route were optimized with the use of
a black-box optimization algorithm working without a priori reaction and gradient information.
In particular, the oxidative dimerization furnishing carpanone 1 was optimized in 1,2-
ACS Paragon Plus Environment

Page 3 of 32
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
II
dichloroethane using molecular oxygen as the oxidant associated to a Co (salen) complex as
the catalyst.
H
H
O
MeO
MeO
O
O
O
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
H
H
O
OMe
O
O
O
OMe
O
Carpanone - 1
Polemannone A - 2
H
H
O
MeO
O
O
H
H
O
O
OMe
MeO
O
MeO
OMe
O
OMe
O
OMe
OMe
Polemannone B - 3
Polemannone C - 4
Oxidative dimerisation
H
O
O
H
O
H
OH
O
R
O
H
O
O
R
Sauchinone - 5
FIGURE 1. Naturally occurring benzoxanthenones lignans. Oxidative dimerization of 2-
propenyl phenols as inset.
While these experimental conditions furnished carpanone 1 in 82% isolated yield, we later
faced difficulties to prepare multi-gram amounts of carpanone 1 for two reasons. First, under
the optimal conditions, carpanone 1 was prepared with an excellent isolated yield (82%) but
with a modest productivity (ca. 17 mg/h) due to the high dilute experimental conditions (ca.
0.015 M) and the modest kinetic (ca. 40 min residence time). Second, the use of a stock solution
of 1,2-dichloroethane saturated with pure oxygen posed severe safety concerns for our
25
industrial partner in charge of the scaling-up. While the first issue could be addressed with an
increase of the initial concentration of starting materials, the second issue required us to
reinvestigate the oxidative conditions through the screening of alternative oxidants and catalysts
compatible with a flow regime. The experimental flow setup we developed for the synthesis of
carpanone in our previous investigations was not adapted for reagents screening as it would
require the use of three streams for the starting material, oxidant and catalyst, respectively. This
also implies that an automated screening of oxidants and catalysts was not possible as it would
require a manual change of the stock solutions. In this manuscript, we report the development
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 4 of 32
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
of a reconfigurable and intelligent synthesis platform that allows both automated micromole-
scale reagent screening and autonomous reaction optimization in flow. The automated platform
includes a number of benefits as it allows very reproducible experiments on a micromole scale
due to the use of a liquid handling robot and minimizes the consumption of solvent due to the
use of a single stream. Additionally, the use of a liquid handling robot in charge of the reaction
mixture preparation prevents issues associated with the reproducibility of the mixing process.
This machine led to the discovery of uncovered oxidative conditions for the construction of
the benzoxanthenone skeleton and allowed the synthesis of naturally occurring capanone 1 and
polemannone B 3 as well as a synthetic analogue.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
■
RESULTS AND DISCUSSION
The bio-inspired oxidative dimerization of 2-propenyl phenols has been reported in
batch under a variety of experimental conditions which are, for most of them, unsuitable
for flow conditions due to either solubility issues (PdCl laccase) or long reaction times
2
,
II
(
CuCl /()-sparteine/O , 24 h). We demonstrated that the use of Co (salen)/O as the
2
2
2
oxidizing system could be efficiently transferred from batch to continuous flow
1
6
conditions; however, the use of molecular oxygen was associated with safety concerns,
especially in the industrial environment. Considering that the reasonable price of cobalt-
based catalysts was an important asset, we decided to explore the combination of a
selection of commercially available Co catalysts with oxidants compatible with flow
conditions to properly address safety concerns for scaling-up experiments; keeping in
mind that oxidants are, by nature, hazardous compounds requiring careful handling,
especially on large scale.
In our initial experimental procedure, a stock solution of cobalt catalyst in 1,2-
1
6
dichloroethane was saturated with oxygen gas (O ). While this protocol features an
2
26
obvious simplicity reaching the main principles of sustainable chemistry, it suffers
from safety concerns upon scaling-up, as the stock solution could represent a highly
2
7
flammable volume of several liters. The other issue associated with this protocol is
related to the difficulty in measuring the exact volume of O dissolved in 1,2-
2
dichloroethane so that the amount of oxygen available for the oxidative dimerization
could not be precisely determined. This point is particularly relevant upon increasing the
concentration of starting material as the amount of oxygen required by unit of volume
would substantially increase until reaching a value where oxygen becomes the limiting
ACS Paragon Plus Environment

Page 5 of 32
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
reagent. The use of a diluted gas mixture (5-10% O in N ) is an alternative strategy
2
2
which allows users to keep the O concentration in the organic solvent below the
2
explosive limit. However, the low amount of O which can be solubilized with such a
2
strategy could preclude the reaction to reach complete conversion.28
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
With the aim of addressing these limitations we embarked on the search for a new oxidizing
system avoiding the use of O . In this frame, we designed a new reconfigurable and fully
2
automated screening platform capable of preparing reaction mixture from specified reagents on
the μmol scale (Figure 2, see also Figure S2 in SI for a photograph of the platform). Our one-
stream system includes a liquid handling robot in charge of the reaction mixture preparation, a
pump, a thermostated reactor coil and a switch valve which samples a small fraction of the
crude mixture for the on-line HPLC analysis.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 6 of 32
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Automated screening platform
H
MeO
O
Reactant 6
H
O
Internal standard 7
Oxidant 8a-e
Catalyst 9a-b
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
0
OMe
Fraction collector
Switch
valve
Heater
Sample
handler
UV
Pump
Reactor
Column
Pump
C H Cl
2
2
4
MeOH/H O
2
Matlab control
Reagents screened
Reactant
Internal standard
, [0.2 M]
7, [0.5 M]
Oxidant
Catalyst
[2.5 x 10-3 M]
6
[0.4 M]
[Co]-
9a
[Co]-9b
[Co]-9c
8
8
a - t-BuOOH,
b - Cumene-OOH
8c - PhI(OAc)2
8d - Oxygen
8e - Air
OH
MeO
[
Co]-9d
Reaction mixture : [0.021 M]
Catalysts structure
O
O
Co
N
O
[
O
N
O
N
O
N
N
N
Co
9b
Co]-
N
N
Co
N
N
t-Bu
t-Bu
N
t-Bu
t-Bu
N
O
N
O
[
Co]-9a
[
Co]-9d
Co
[Co]-9c
FIGURE 2. Evaluation of oxidizing systems through a mol scale automated screening platform
ACS Paragon Plus Environment

Page 7 of 32
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
An automatic data processing integrates the chromatogram and calculates the yield using an
internal standard. All modules from the screening platform are connected to a computer through
RS-232 ports and controlled with an ad-hoc process-control software.
To start a screening campaign, an operator populates the 56-well plate with vials containing
the required reagents (starting material, catalyst, base…). After having fed the process-control
software with the concentration of the stock solution-containing vials, the composition of the
reaction mixture, the stoichiometry chosen for the screening and the volume injected in the coil
reactor, the sampling robot automatically determines the volume of suction in each vial to
prepare the reacting mixture injected in the coil reactor. Two different setups were designed
depending on the nature of the oxidant. Non-gaseous oxidants were sampled by the liquid
handling robot from the stock solutions and the reaction occurred in a PEEK reactor. Gaseous
oxidants such as air and oxygen were supplied to the reaction mixture through the use of a gas-
permeable coil reactor made from the AF-2400 polymer.29-30 The search for a new oxidizing
system using the screening platform was carried out with the benchmark oxidative dimerization
of 2-propenyl phenol 6 to benzoxanthenone 10. We prepared a 40-member 3-dimensional array
from a selection of 3 non-gaseous oxidants 8a-c (plus one blank) and 4 cobalt catalysts 9a-d
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(plus one blank). The array was split in two runs of 20 experiments carried out at 25 and 50 °C.
An additional 10-member 2D array from a selection of catalysts 9a-d (plus one blank) and 2
gaseous oxidants 8d-e were conducted at 25 °C. For safety reasons we did not studied gaseous
oxidant, and especially oxygen, at 50 °C. In this screening campaign the use of oxygen served
as a standard to assess the efficiency of other oxidants screened. To our knowledge, the cobalt-
catalyzed oxidative dimerization of 2-propenyl phenols had only been investigated with oxygen
as oxidant and the potential of alternative oxidants in promoting the expected dimerization
remained to be demonstrated.
Reactions explored with the screening platform were arbitrarily conducted using 4
μmol of 6 with 1.5 equivalents of non-gaseous oxidant and 10 mol% of cobalt catalyst
at either 25 or 50 °C. By using a high loading of cobalt catalysts and an excess of oxidant,
we minimize the risk of missing an effective Co/oxidant combination. 2-Propenyl phenol
6
, non-gaseous oxidants and cobalt catalysts were prepared as stock solutions in 1,2-
dichloroethane at 0.2, 0.4, and 0.0025 M, respectively and placed in the 56-well plate of
the liquid handling robot. For the use of gaseous oxidants, the experimental setup was
slightly modified by switching the PEEK reactor for a gas permeable tubing reactor
made from AF-2400 polymer that passed through a chamber supplied with a flow of
29
either air or oxygen. This procedure differs from the one we previously published and
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 8 of 32
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
addresses the risk of lacking oxidant when using concentrated solution of 2-propenyl
phenols. In order to allow the screening at an elevated throughput, all reactions were
conducted at a single residence time of 10 minutes. Each fully automated experiment,
including the robot-mediated reaction mixture preparation, the reaction time, the on-line
HPLC analysis and the data processing (integration and yield calculation) was completed
in ca. 26 minutes and the screening campaign was achieved at a rate of 27 experiments
every 12 hours; for safety reason the machine was not used overnight. Both oxidants and
catalysts were selected for their (i) potential to promote the oxidative dimerization, (ii)
solubility in 1,2-dichloroethane, (iii) ease of handling under standard conditions and (iv)
commercial availability. Several oxidants were discarded as they were either insoluble
in DCE (DDQ and TEMPO) or required aqueous biphasic conditions (H O and
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
2
Oxone®).
The analysis of the 50-reaction screen revealed several interesting features (Figure 3). First,
reaction conducted at 50 °C consistently gave lower HPLC yields of the targeted
benzoxanthenone 10, suggesting a pronounced thermal sensitivity of either starting materials
or benzoxanthenone 10. Second, [Co]-9a-b gave significantly higher yields than [Co]-9c-d
whatever the oxidant used. Regarding the effect of oxidants it was observed that all 3 non-
gaseous oxidants 8a-c performed similarly while air and oxygen gave lower yields under such
short residence time. These results validated our strategy to switch oxygen for a non-gaseous
oxidant. A closer look on background reactions, revealed that PhI(OAc) 8c was the only
4
oxidant promoting the dimerization in a significant extent in the absence of cobalt catalyst.
While the ability of PhI(OAc) 8c to promote the dimerization of 2-propenyl phenols to
4
benzoxanthenone skeletons was already known,7-8 our results show an uncovered exalted
reactivity in the presence of cobalt catalysts. Despite the slightly better results obtained with
PhI(OAc) 8c, we selected t-BuOOH 8a as it is less expensive and only produces t-BuOH as
4
by-product. Regarding the catalyst selection, both [Co]-9a and [Co]-9b gave promising results.
-3
Despite its much lower price, [Co]-9b proved to be particularly insoluble above 2.5 × 10 M,
precluding any scale-up experiments under more concentrated conditions. We therefore
preferred the use of [Co]-9a for our further studies. The process for discovering a new cobalt-
catalyzed oxidative system for the dimerization of 2-propenyl phenols to benzoxanthenone
skeletons required 50 reactions at a throughput of one reaction every 26 minutes. Each reaction
used 4 µmol of starting material 6 and 0.4 µmol of cobalt catalysts 9a-d.
ACS Paragon Plus Environment

Page 9 of 32
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
FIGURE 3. Heatmap visualization of the automated micromole-scale screen across a 50-member
array.
Having identified [Co]-9a/t-BuOOH 8a as the most promising oxidative system from
the screening of 50 reactions, we further optimized experimental conditions using an
autonomous self-optimizing flow reactor. This emerging technology combines the
power of automated flow devices with an assistive optimization algorithm and integrates
a quasi-real time in-line/online monitoring.31-35
In most chemical laboratories, optimizations are traditionally conducted by modifying
one factor at a time until reaching an optimum. While the simplicity of execution is the
major asset explaining the popularity of this approach, the data obtained fail to explain
interactions between variables, precluding any efficient automation. The automated
screening of a set of predetermined experimental conditions generated from design-of-
experiment (DoE) methods partially addresses the issues of one-factor-at-a-time
optimizations. However, multi-variable optimizations using DoE methods usually
require a high number of experiments to locate a satisfactory optimum.
The use of feedback algorithms working in black box-type systems is a powerful
alternative to traditional one-factor-at-a-time and DoE methods for optimizing chemical
reactions.36-37 Several black box algorithms of different numerical complexity have been
reported in the literature for integration in self-optimizing flow devices.38-49 From these
5
0
algorithms, the Nelder-Mead method, also known as the simplex method, has been
reported to be one of the simplest for implementation in automated flow reactors.51-60
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 10 of 32
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Numerically, it is one of the simplest optimization algorithms and does not require
gradient approximation which is a strong asset when working with expensive
experiments. However, the Nelder-Mead method suffers from two important drawbacks:
(i) it does not converge rapidly when optimizing large multi-variable problems and (ii)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
it only converges to a local optimum which quality highly depends on the initialization
6
1
and the studied system. We addressed these issues through the development of a
profoundly modified version of the Nelder-Mead method that includes (i) the possibility
to temporarily modify the dimensionality of the search to explore a subspace, (ii)
multiple stopping criteria to lower the number of experiments, (iii) diversification and
intensification mechanisms to escape from an unsatisfactory local optimum, (iv) the
golden search method for 1-D optimizations which cannot be addressed by the Nelder–
62
Mead method. Moreover, in addition to the traditional linear constraints imposed by
boundary constraints, our algorithm contains an optional mechanism to restrict the
search space through linear constraints in order to exclude experimental conditions
considered as unsound (lack of reactivity, poor stability of reagents/products…).
The transformation of our automated screening platform to an autonomous self-
optimizing flow reactor was straightforward and convenient as the only modification
consisted of the integration of our additional optimization algorithm to the process-
control software. Our approach allows to either screen discrete variables for reaction
development or continuous variables for a fine-tuning of experimental conditions (self-
optimization). While the use of automated screening platforms have been described in
the literature,63-68 the use of a reconfigurable platform for either screening or self-
optimization offers a much higher flexibility. The optimization was conducted at higher
concentrations with stock solutions of 2-propenyl phenol 6, [Co]-9a and t-BuOOH 8a at
concentration of 1 M, 0.03 M and 2 M, respectively, in order to assure an elevated
throughput.
The reaction yield was optimized in a 4-dimension space where the residence time,
temperature, equivalent of t-BuOOH 8a and loading of [Co]-9a were selected as input variables
in the range of 5-60 minutes, 25-50 °C, 1-2 equivalents, 1-10 mol%, respectively. In addition
to the bounded search space, we also fixed the initial point X at 5 minutes of residence time,
0
25 °C, 1 equivalent of t-BuOOH 8a and 5 mol% of [Co]-9a with d values of 10 min, 5 °C, 0.2
equivalent and 2 mol%, respectively. While we preferred to focus on the reaction yield
optimization, we were also interested in developing experimental conditions allowing decent
ACS Paragon Plus Environment

Page 11 of 32
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
productivity. Therefore, for experimental conditions displaying the highest yields, the
associated productivity have been considered as well.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
FIGURE 4. (a) Maximization of the yield of benzoxanthenone 10. (b) Representation of the four-
dimensional experimental conditions for the maximization of the yield for benzoxanthenone
10.
The first 5 experiments corresponding to the initial 4-D simplex, gave promising yields
ranging from 44 to 62% (Figure 4a-b, See Table S1 in SI for details). In the subsequent
experiments, the algorithm rapidly decreased the loading of [Co]-9a until reaching the lower
bound at 1 mol%, while in the same time the residence time progressively increased to ca. 30
minutes. On the other hand, both the equivalents of t-BuOOH 8a and the temperature were set
to ca. 1.00-1.10 equivalents and 27-33 °C, respectively. The algorithm reduced the dimension
of the search from experiment 17 by fixing the catalyst loading to its lower bound, reaching an
optimum corresponding to 95% yield. After 5 consecutive simplexes which did not improve
the reaction yield, the algorithm reached a stopping criterion and the optimization stopped after
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 12 of 32
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
experiment 24. From the 24 experiments conducted through this optimization campaign, we
observed two appealing solutions corresponding to experiments 16 and 17 which led to
benzoxanthenone 10 as a single diastereoisomer with 92 and 95% yield, respectively. A closer
look on the experimental conditions revealed that experiment 16 (27 minutes, 32 °C, 1.09 equiv
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
8
a and 1 mol% 9a) provided a better productivity than experiment 17 (31 minutes, 29 °C, 1.08
-1
equiv 8a, and 1 mol% 9a) with values of 887 and 800 mg.h , respectively. The calculated
productivity for each experiment can be found in Table S1. Unfortunately, upon purification by
flash chromatography, benzoxanthenone 10 turned to be extremely instable in our hand, leading
to a modest isolated yield (55%).
We further pursued our efforts toward the synthesis of natural lignans carpanone 1 and
polemannone B 3 using the optimized conditions from experiment 16.
Desmethoxycarpacine 12, obtained in three steps from sesamol 11 following our
previously published flow procedure, was dimerized to carpanone 1 in 85% isolated
1
6
yield (Scheme 1). A minor isomer, identified on the crude mixture (ca. 3%), was lost
during the purification step and carpanone 1 was isolated as a single isomer. In such
experimental conditions the productivity spectacularly increased from ca. 17 mg/h using
our previous setup to ca. 263 mg/h in this work. The reproducibility of our automated
screening platform machine was also assessed with the oxidative dimerization of
desmethoxycarpacine 12 to carpanone 1 on 6 consecutive runs in the optimal conditions
of experiment 16. The reaction yields calculated by the online HPLC system were 85,
8
6, 86, 86, 86 and 87%, respectively, corresponding to an average HPLC yield of 86%
(vs 85% for the isolated yield) and a standard deviation of only 0.63.
ACS Paragon Plus Environment

Page 13 of 32
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
OH
OH
O
O
O
O
Sesamol 11
3 steps, 49% overall
Reference 16
12
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
8
[
1, [0.25 M]
a, [0.5 M]
Heater
Sample
handler
Co]-9a, [0.03 M]
H
O
O
O
0.19 mL/min
H
Pump
Reactor
O
O
V = 5 mL
T = 32 °C
Carpanone - 1, 85% yield
C H Cl
2
2
4
Matlab control
SCHEME 1. Improved flow synthesis of carpanone 1 in flow.
Regarding the total synthesis of polemannone B 3, we prepared the required 2-propenyl
phenol 17 in 7 steps from commercially available sesamol (Scheme 2). A mixed batch-flow
synthesis was developed for the synthesis of 17. The allylation of sesamol 11 with allyl iodide
was conducted in flow providing the corresponding allyl ether 13 in 76% yield. The required
methoxy group was installed through a 4-step batch sequence. The formylation of 13 with
DMF/n-BuLi furnished the expected aldehyde 14 with a modest yield (51%). A Baeyer-Villiger
oxidation of aldehyde 14 furnished the formate intermediate which was immediately
hydrolyzed with aqueous NaOH to reveal the phenol function of 15 (48%, two steps). The latter
was etherified with methyl iodide to give 16 in 84% yield. The thermal [3-3]-Claisen
rearrangement was conducted under flow-mode as it provided an improved reaction rate and
much safer experimental conditions. The rearrangement occurred in acetone at 250 °C and 70
bar with a residence time of only 2.5 minutes to provide phenol 17 in 89% yield. Under such
high-temperature/high-pressure conditions acetone reached its supercritical state.69-71 Lastly,
allylphenol 17 was isomerized to the corresponding 2-propenyl phenol 18 by treatment with t-
BuOK in 49% yield. The 8-step total synthesis of polemannone B 3 was smoothly achieved
through the oxidative dimerization of 2-propenyl phenol 18 using the automated platform under
optimal conditions developed for the synthesis of 10, providing polemannone B 3 in 84%
isolated yield as a single isomer. This synthesis complements the still scarce examples of total
synthesis of natural products using a mixed batch/flow approach.72
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 14 of 32
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
OH
O
O
Pump
Heater
BPR
O
O
O
Sesamol - 11
32 L/min
MeOH/H O
[
0.4 M]
+
15 bar
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Reactor
2
KOH [0.6 M]
13, 76%
V = 2 mL
T = 75 °C
Pump
I
2
9 L/min
[
0.8 M]
MeOH/H O
2
1
. m-CPBA
CHCl3, 0 °C
4 h
n-BuLi, THF
78 °C, 45 min
MeI, K CO3
2
O
O
O
-
O
O
acetone
2
2
. NaOH
THF/H2O
5 °C, 1 h
O
then, DMF
78 °C, 30 min
25 °C, 72 h
84%
-
1
4
CHO
15 OH
2
51%
4
8%, two steps
O
O
O
2 mL
Heater
BPR
S
OH
t-BuOK, THF
O
O
loop
OMe
16
[1.5 M]
Pump
70 °C, 90 min
70 bar
Reactor
49%
OMe
7, 89%
V = 5 ml
T = 250 °C
1
2
mL/min
Acetone
OH
O
O
8
a, [0.5 M]
Heater
Sample
handler
[
Co]-9a, [0.03 M]
H
O
O
O
0
.19 mL/min
OMe
8, [0.25 M]
H
1
Pump
O
OM
e
Reactor
O
OMe
O
V = 5 ml
T = 32 °C
Polemannone B - 3, 84%
yield
C H Cl
2
2
4
Matlab control
SCHEME 2. Total synthesis of Polemannone B 3 under mixed batch/flow modes.
The mechanism of the oxidative dimerization is still a subject of debate which has not
led to a unanimous consensus at present; some authors also propose the possibility of
10
several mechanisms depending on the experimental conditions. Though, the most
ACS Paragon Plus Environment

Page 15 of 32
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
common mechanism proposed in the literature involves an initial anti-β,β-radical
dimerization followed by an endo-[4+2] cycloaddition (Scheme 3).
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
OH
O
-H, -e
anti-,-coupling
RO
RO
RO
A
B
endo
H
O
RO
O
H
O
[
4+2]
O
RO
C
D
OR
SCHEME 3. Mechanistic proposal involving sequential anti-β,β-radical dimerization/endo-
[4+2] cycloaddition.
However, this mechanistic proposal fails to properly explain the perfect anti-selectivity
observed at the methyl groups from the β,β-radical dimerization. Moreover, radical B would be
reactive and could abstract, at least partially, a hydrogen atom from the solvent or the starting
phenol, leading to the ortho-quinone methide E which is known to rapidly dimerize into the
corresponding benzopyrane F (Scheme 4).15
RO
RO
H
+
O
OR
O
B
E
F
HO
A
RO
[
4+2]-cycloaddition
OR
O
HO
SCHEME 4. Dimerization of ortho-quinone methides E to benzopyrane F.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 16 of 32
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
In all our tests, regardless of the oxidant used, we never observed the formation of
benzopyrane-type structures. In a previous report we proposed an initial para-quinol-type
1
6
coupling followed by an endo-[4+2] cycloaddition (Scheme 5, top). While this mechanism
properly explains the anti-selectivity at the methyl group, we did not have any experimental
evidence at that time to support this proposal. We have reinvestigated this mechanism, and we
are now able to confirm our proposal when using [Co]-9a/t-BuOOH 8a as oxidizing system.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
III
The oxidation of [Co]-9a by t-BuOOH 8a produces the corresponding Co (salen)(OH)
73
complex 19 along with the tert-butoxy radical (Scheme 5, bottom). Ligand exchange on
III
III
Co (salen)(OH) complex 19 furnishes Co (salen)(OOt-Bu) complex 20 which restored
•
starting cobalt complex 9a along with t-BuOO . The latter abstracts the phenolic hydrogen from
A to give the corresponding phenoxy radical H which by delocalization of the unpaired electron
furnishes aryl radical I. Dimerization of H with I furnishes the para-quinol-type intermediate
G which further cyclizes through an endo-[4+2] cycloaddition to give the benzoxanthenone
skeleton D.
ACS Paragon Plus Environment

Page 17 of 32
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Formation of benzoxanthenones
H
[
Ox]
RO
O
endo-[4+2]
RO
O
OH
H
O
para-quinol
cycloaddition
O
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
RO
coupling
OR
A
G
D
OR
Mechanistic insights
RO
II
Co (salen)
t-BuOOH
RO
9a
8
a
A
OH
O
t-BuOO
O
t-BuO
I
H
RO
III
III
Co (salen)(OH)
Co (salen)(OOt-Bu)
19
20
t-BuOOH H O
2
8a
H
endo-[4+2]
RO
O
RO
O
dimerization
H
O
cycloaddition
O
D
OR
OR
G
#
O
RO
OR
O
SCHEME 5. Mechanistic proposal for the bio-inspired formation of benzoxanthenones lignans
through Co(II)/t-BuOOH oxidative dimerization of 2-propenyl phenols.
■
CONCLUSION
In this work we designed a reconfigurable flow reactor which could be used either as an
automated screening platform or as an autonomous self-optimizing reactor for natural
product synthesis. The capability of our automated screening platform to screen discrete
variables on the micromole scale and analyze the crude content through an online
monitoring in a short time frame is a strong asset in reaction discovery. By associating
process control instrumentation and online analysis to an optimization algorithm in a
closed-loop system, the screening platform can be easily transformed to a powerful
autonomous self-optimizing flow reactor for the optimization of continuous variables.
The automated screening platform identified a new cobalt-catalyzed oxidative system
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 18 of 32
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
for the dimerization of 2-propenyl phenols to benzoxanthenone skeletons from an array
of 50 reactions which represent a total of only 200 μmol (0.66 mg per reaction) of starting
material. The cobalt-catalyzed oxidative system identified with the screening platform
was further optimized in a 4-dimension space using an autonomous closed-loop system.
A very satisfactory optimum was located after only 16 experiments and the system
reached a stopping criterion after only 24 experiments. This cobalt-catalyzed oxidizing
proved to be particularly efficient to prepare natural products carpanone and
polemannone B as well as a synthetic analogue.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Through this work, we demonstrated that while essentially used for preparative
procedures, flow reactors are also particularly well suited for automated reagent
screening and autonomous reaction optimization. While the development of automation
in chemistry is a modern preoccupation,74-79 the results described in this work shows that
process control instrumentation associated to menial and assistive algorithms assist
chemists and improve their intellectual productivity.
■
EXPERIMENTAL SECTION
General information. All commercially available chemicals were used as received unless
1
13
otherwise noted. H and C NMR spectra were recorded at 300 or 400 and 75 MHz or 100
1
13
MHz, respectively. H and C NMR spectra were referenced to the internal deuterated solvent
(CDCl ) at 7.26 and 77.16 ppm, respectively. FT-IR spectra were recorded in the ATR mode.
3
-1
Wavelengths of maximum absorbance (νmax) are quoted in wave numbers (cm ). High
resolution mass spectrometry (HRMS) was recorded on a microTOF spectrometer equipped
with orthogonal electrospray interface (ESI). Analytical thin-layer chromatography (TLC) was
carried out on silica gel 60 F254 plates and visualized with a UV lamp at 254 nm or stained with
a basic potassium permanganate solution. Flash column chromatography was performed using
silica gel 60 (40−63 µm).
Details of the experimental setup. HPLC pumps (JASCO PU2080) equipped with a RS-
2
32 port were employed to flow the solution through the system. A sampler handler (JASCO
AS 2055) equipped with a RS-232 port was used to inject the reagents in the line. The reactor
coil was heated with a heating plate (Heidolph, MR Hei-Connect) equipped with a RS-232 port.
A 2-way 6-port valve (VICI, Cheminert C2-3006D) equipped with a RS-232 port was used to
inject an aliquot of the crude mixture within the on-line HPLC unit. The HPLC column outlet
was connected to a UV detector (JASCO, UV 2075) equipped with a RS-232 port. The flow
ACS Paragon Plus Environment

Page 19 of 32
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
outlet was connected to a programmable fraction collector (Advantec, CHF 1225C). All units
®
equipped with a RS-232 port were autonomously controlled with MATLAB through the use
of communication protocols provided by the manufacturers.
1
3
Synthesis of (E)-4-methoxy-2-(prop-1-en-1-yl)phenol (6). To a solution n-BuLi (13.85 mL,
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
6.3 mmol, 1.9 M in hexane) in distilled THF (65 mL) was added ethyltriphenylphosphonium
bromide (9.76 g, 26.3 mmol) at 25 °C. The resulting mixture was stirred for 45 min and then 2-
hydroxy-5-methoxybenzaldehyde (2 g, 13.2 mmol) was added in one portion. After being
stirred for 3 h, the reaction was quenched with 0.5 M HCl (30 mL) and the aqueous phase was
extracted three time with EtOAc (3 × 30 mL). The collected organic extracts were dried over
anhydrous MgSO , filtered and concentrated under reduced pressure. Purification by flash
4
chromatography (20% EtOAc-petroleum ether) furnished 6 (1.60 g, 74%) as pale yellow liquid.
IR (ATR) ν 3372, 3035, 2937, 2912, 2834, 1652, 1607, 1497, 1428, 1280, 1201, 1160, 1037,
-1 1
968 cm . H NMR (300 MHz, CDCl ) δ 6.86 (d, 1H, J = 2.9 Hz), 6.72 (d, 1H, J = 8.6 Hz), 6.66
3
(dd, 1H, J = 2.9, 8.7 Hz), 6.57 (dq, 1H, J = 1.7, 15.8 Hz), 6.14-6.26 (dq, 1H, J = 6.6, 15.8 Hz),
1
3
1
4
.73 (br d, 1H, J = 2.8 Hz), 3.77 (s, 3H), 1.91 (dd, 3H, J = 1.7, 6.6 Hz). C{ H} NMR (100
MHz, CDCl ) δ 153.9, 146.7, 128.4, 126.0, 125.5, 116.6, 113.7, 112.3, 55.9, 19.0. HRMS
3
+
+
(ASAP ) m/z [M + H] Calcd for C H O 165.0916; Found 165.0920
1
0
13
2
Experimental setup of the automated screening platform. An automatic sample handler
prepared 200 µL of the reaction mixture from stock solutions of 2-propenyl phenol 6 (0.2 M

3
plus xylene as the internal standard at 0.5 M), oxidants (0.4 M) and catalysts (2.5 × 10 M).
The reaction mixture was injected in a stream of C H Cl pumped at 0.5 mL/min as depicted in
2
4
2
Figure 2. The oxidative dimerization occurred either in a PEEK reactor coil (5 mL, 0.75 mm
id) heated at the required temperature (25 or 50 °C) for oxidants 8a-c or in a AF-2400 reactor
coil that passed through a glass bottle closed with a GL-45 cap and supplied with a flow of
2
9
either air or oxygen. The reactor outlet was connected to an automatic 2-way 6-port switch
valve which injected 0.2 µL of the crude mixture in the HPLC unit while the remaining stream
was collected in a fraction collector. A mixture of MeOH/H O (70/30, v/v) was used as mobile
2
phase for the HPLC analysis at a flowrate of 0.4 mL/min. A UV detector was connected to the
outlet of the HPLC column (Agela Promosil C18, 3.5 mm × 150 mm, 5 µm) to follow the
absorbance at a wavelength of 270 nm. Peak integration and yield calculation were under full
MATLAB automation.
Experimental setup of the autonomous self-optimizing flow reactor. Synthesis of
benzoxanthenone (10). An automatic sample handler prepared 200 µL of the reaction mixture
from stock solutions of 2-propenyl phenol 6 (1 M plus xylene as the internal standard at 2.5 M),
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 20 of 32
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
t-BuOOH 8a (2 M) and [Co]-9a (0.03 M). The reaction mixture was injected in a stream of
C H Cl pumped at the required flow. The oxidative dimerization occurred in a PEEK reactor
2
4
2
coil (5 mL, 0.75 mm id) heated at the required temperature. The reactor outlet was connected
to an automatic 2-way 6-port switch valve which injected 0.2 µL of the crude mixture in the
HPLC unit while the remaining stream was collected in a fraction collector. A mixture of
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
MeOH/H O (70/30, v/v) was used as mobile phase for the HPLC analysis at a flowrate of 0.4
2
mL/min. A UV detector was connected to the outlet of the HPLC column (Agela Promosil C18,
3
.5 mm × 150 mm, 5 µm) to follow the absorbance at a wavelength of 270 nm. Peak integration
and yield calculation were under full MATLAB automation. The calculated yield was
automatically sent to the algorithm which set new experimental conditions to the units via RS-
2
32 ports. A 4-D optimization of the reaction yield was conducted using the temperature,
residence time, loading of [Co]-9a and equivalents of t-BuOOH 8a as the input variables. The
initial experiment of the simplex was: 25 °C, 5 min of residence time, 5 mol% of [Co]-9a and
1
equivalent of t-BuOOH 8a with d values of 5 °C, 10 min, 2 mol% and 0.2 equivalent,
respectively. The lower and upper boundaries of the research space were the following: 25-50
C, 5-60 min, 1-10 mol% and 1-2 equiv. for the temperature, residence time, catalyst loading
°
and equivalents of oxidant, respectively. An optimum giving 92% HPLC yield was found in
experiment 16 at 32 °C, 27 min of residence time, 1 mol% of [Co]-9a and 1.09 equivalents of
t-BuOOH 8a (see Table S1 for details). An analytical sample of benzoxanthenone 10 was
obtained by flash chromatography (10% Et O-cyclohexane) as a white solid (23.8 mg, 55%).
2
mp 88.5-89.5 °C. IR (ATR) ν 2958, 2929, 2873, 2835, 1678, 1617, 1492, 1268, 1199, 1115,
-
1 1
1
6
035 cm . H NMR (300 MHz, CDCl ) δ 7.24 (d, 1H, J = 10.2 Hz), 7.01-7.03 (m, 1H), 6.89-
3
.90 (m, 1H), 6.75 (d, 1H, J = 8.7 Hz), 6.66 (dd, 1H, J = 2.8, 8.8 Hz), 6.31 (d, 1H, J = 10.2 Hz),
3.76 (s, 3H), 3.31-3.33 (m, 1H), 3.31 (s, 3H), 3.12-3.16 (m, 1H), 2.59-2.66 (m, 1H), 2.17-2.23
1
3
1
(
m, 1H), 1.15 (d, 3H, J = 7.2 Hz), 0.67 (d, 3H, J = 7.5 Hz). C{ H} NMR (75 MHz, CDCl ) δ
3
1
86.8, 154.4, 145.1, 143.7, 142.7, 131.8, 128.1, 125.6, 118.0, 113.6, 113.3, 55.8, 49.3, 37.1,
+
3
6.8, 35.2, 34.0, 21.8, 21.4. HRMS (ESI) m/z [M + Na] Calcd for C H NaO 349.1416;
2
0
22
4
Found 349.1407.
Synthesis of Carpanone (1). An automatic sample handler prepared 200 µL of the reaction
mixture from stock solutions of (E)-6-(prop-1-enyl)-1,3-benzodioxol-5-ol 1216 (0.25 M, 1
equiv), t-BuOOH 8a (0.5 M, 1.09 equiv) and [Co]-9a (0.03 M, 1 mol%). The reaction mixture
was injected in a stream of C H Cl pumped at 0.19 mL/min as depicted in Scheme 1. The
2
4
2
oxidative dimerization occurred in a PEEK reactor coil (5 mL, 0.75 mm id) heated at 32 °C.
The reactor outlet was connected to an automatic 2-way 6-port switch valve which injected 0.2
ACS Paragon Plus Environment

Page 21 of 32
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
µL of the crude mixture in the HPLC unit while the remaining stream was collected in a fraction
collector. A mixture of MeOH/H O (70/30, v/v) was used as mobile phase for the HPLC
2
analysis at a flowrate of 0.4 mL/min. A UV detector was connected to the outlet of the HPLC
column (Agela Promosil C18, 3.5 mm × 150 mm, 5 µm) to follow the absorbance at a
wavelength of 270 nm. Peak integration and yield calculation were under full MATLAB
automation. An analytical sample of carpanone 1 was obtained by flash chromatography (10%
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
5
Et O-cyclohexane) as a white solid (34 mg, 85%). mp 192-193 °C [Lit. 189-190 °C]. IR
2
-1
1
(
ATR) ν 2872, 1674, 1622, 1478, 1158, 1032, 910, 840 cm . H NMR (400 MHz, CDCl ) δ
3
7.00-7.02 (m, 1H), 6.80 (s, 1H), 6.33 (s, 1H), 5.90 (d, 1H, J = 1.4 Hz), 5.87 (d, 1H, J = 1.4 Hz),
.69 (s, 1H), 5.66 (s, 1H), 5.64 (s, 1H), 3.27 (dd, 1H, J = 7.4, 2.6 Hz), 3.19 (dt, 1H, J = 7.5, 2.5
Hz), 2.52 (br qd, 1H, J = 6.9, 2.1 Hz), 2.18-2.24 (m, 1H), 1.15 (d, 3H, J = 7.2 Hz), 0.71 (d, 3H,
5
13
1
J = 7.6 Hz). C{ H} NMR (100 MHz, CDCl ) δ 187.0, 168.5, 146.8, 145.3, 143.3, 142.7, 126.5,
3
115.4, 107.3, 101.4, 100.6, 100.3, 99.4, 98.9, 36.4, 35.6, 35.4, 33.7, 21.6, 21.3. HRMS (ESI)
+
m/z [M + H] calcd for C H O , 355.1182, found 355.1180.
20 19
6
5
-Allyloxy-1,3-benzodioxole (13). The experimental setup consisted in two streams as
depicted in Scheme 2. The first stream contained a solution of sesamol 11 (0.4 M) and KOH
(0.6 M) in MeOH/H O (95/5, v/v) while the second stream contained a solution of alkyl iodide
2
(
0.8 M) in MeOH/H O (95/5, v/v). Each solution was continuously pumped with two
2
independent pumps at 32 µL/min for the first stream and 29 µL/min for the second stream. The
two streams met in a stainless steel T-shaped piece (internal volume: 0.57 μL). The resulting
mixture was introduced in a stainless steel reactor coil (2 mL, 0.75 mm id) heated at 75 °C. The
reactor outlet was connected to a back pressure regulator (BPR) to maintain the internal pressure
at ca. 15 bar. An analytical sample of 5-allyloxy-1,3-benzodioxole 7 was obtained by flash
chromatography (10% AcOEt-petroleum ether) as a colourless oil (1.04 g, 76%). IR (ATR) ν
-
1 1
2
8
1
283, 1629, 1483, 1178, 1035, 923, 780 cm . H NMR (300 MHz, CDCl ,) δ 6.70 (d, 1H, J =
3
.5 Hz), 6.52 (d, 1H, J = 2.5 Hz), 6.34 (dd, 1H, J = 2.5, 8.5 Hz), 6.03 (ddt, 1H, J = 5.3, 10.6,
7.2 Hz), 5.91 (s, 2H), 5.39 (dq, 1H, J = 1.6, 17.3 Hz), 5.28 (dq, 1H, J = 1.4, 10.5 Hz), 4.46 (dt,
1
3
1
2H, J = 1.5, 5.3 Hz). C{ H} NMR (100 MHz, CDCl ): δ 154.2, 148.3, 141.8, 133.5, 117.7,
3
+
1
08.0, 106.1, 101.2, 98.4, 69.9. HRMS (ASAP+) m/z [M] Calcd for C H O 178.0630; Found
1
0
10
3
1
78.0631.
5-Allyloxy-4-formyl-1,3-benzodioxole (14). A solution of n-BuLi (6 mL, 1.6 M) was added
8
0
dropwise to a solution of 5-allyloxy-1,3-benzodioxole 13 (1.41 g, 7.91 mmol) in dry THF (20
mL) at 78 °C. The resulting yellow solution was stirred for 45 min and then dry DMF (1.25
mL, 15.8 mmol) was added. After 30 min at 78 °C, the reaction mixture was hydrolyzed with
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 22 of 32
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
saturated NH Cl solution and extracted with Et O (3 × 20 mL). The collected organic fractions
4
2
were dried over anhydrous MgSO , filtered and concentrated under reduced pressure.
4
Purification by flash chromatography (60% CH Cl -petroleum ether) furnished the
2
2
8
0
corresponding aldehyde 14 (823 mg, 51%) as a yellow solid. mp 98-99 °C [Lit. 98-100 °C].
-1 1
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
IR (ATR) ν 2881, 1675, 1630, 1455, 1234, 1075, 913, 745 cm . H NMR (300 MHz, CDCl )
3
δ 10.42 (s, 1H), 6.89 (d, 1H, J = 8.6 Hz), 6.33 (d, 1H, J = 8.6Hz), 6.09 (s, 2H), 5.98-6.12 (m,
1
H), 5.42 (dq, 1H, J = 1.6, 17.2 Hz), 5.31 (dq, 1H, J = 1.4, 10.5 Hz), 4.56 (dt, 2H, J = 1.5, 5.2
13
1
Hz). C{ H} NMR (75 MHz, CDCl ) δ 188.3, 155.3, 148.5, 142.8, 132.6, 118.1, 113.1, 111.3,
3
+
103.8, 103.1, 70.2. HRMS (ESI) m/z [M + Na] Calcd for C H O Na 229.0477; Found :
1
1
10
4
2
29.0485.
-Allyloxy-4-formate-1,3-benzodioxole. m-CPBA (5.62 g, 19.5 mmol, 60% in water) was
5
added to a solution of 4-formyl-5-allyloxy-1,3-benzodioxole (2.12 g, 10.3 mmol) in CHCl (34
3
mL) at 0 °C and the resulting mixture was stirred at 0 °C for 24 h. After completion, the mixture
was diluted with chloroform (50 mL) and was successively washed with saturated aqueous
Na SO solution (50 mL), saturated aqueous Na CO solution (50 mL), brine (50 mL) and water
2
3
2
3
(
50 mL). The organic phase was dried over anhydrous MgSO and concentrated under reduced
4
pressure. The crude formate was obtained as a yellow oil (1.82 g, 80%) and directly used in the
next step without further purification.
5
-Allyloxy-4-hydroxy-1,3-benzodioxole (15).80 5-Allyloxy-4-formate-1,3-benzodioxole
(1.82 g, 8.2 mmol) in solution in THF (30 mL) was hydrolyzed by 3N NaOH (9 mL). After
being stirred for 1 h at 25 °C, the reaction mixture was diluted with CH Cl (50 mL). The
2
2
organic phase was extracted with water (4 × 30 mL). The aqueous phase was acidified with 2
N HCl and extracted with CH Cl (3 × 50 mL). The organic phases were dried over anhydrous
2
2
MgSO and concentrated under reduced pressure. Purification by flash chromatography (30%
4
CH Cl -petroleum ether) furnished 5-allyloxy-4-hydroxy-1,3-benzodioxole 15 (964 mg, 61%)
2
2
as a white solid. mp 51-52 °C. IR (ATR) ν 3301, 2902, 1656, 1471, 1244, 1053, 913, 767 cm-
1. 1
H NMR (300 MHz, CDCl ) δ 6.33 (s, 1H), 6.33 (s, 1H), 5.98-6.11 (m, 1H), 5.94 (s, 2H),
5.47 (s, 1H), 5.39 (dq, 1H, J = 1.5, 17.2 Hz), 5.31 (dq, 1H, J = 1.3, 10.4 Hz), 4.53 (dt, 2H, J =
3
1
3
1
1
.4, 5.5 Hz). C{ H} NMR (75 MHz, CDCl ) δ 143.6, 142.4, 134.1, 133.1, 131.1, 118.6, 104.8,
3
+
1
01.8, 98.9, 71.2. HRMS (ESI) m/z [M + Na] Calcd for C H O Na 217.0477; Found :
1
0
10
4
217.0470.
5
-Allyloxy-4-methoxy-1,3-benzodioxole (16). To a solution of 5-allyloxy-4-hydroxy-1,3-
benzodioxole (1.69 g, 8.70 mmol) in acetone (9 mL) was added sequentially K CO (2.40 g,
2
3
1
7.39 mmol) and CH I (2.70 ml, 43.48 mmol, 5 equiv). The resulting reaction mixture was
3
ACS Paragon Plus Environment

Page 23 of 32
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
stirred at 25 °C for 72 h and then evaporated to dryness under reduced pressure. Inorganic
residues were dissolved in water (15 mL) and the aqueous phase was extracted with Et O (3 ×
2
1
5 mL). The collected organic extracts were dried over anhydrous MgSO and concentrated
4
under reduced pressure. Purification by flash chromatography (30% CH Cl -petroleum ether)
2
2
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
furnished 5-Allyloxy-4-methoxy-1,3-benzodioxole 16 as a colorless oil (1.52 g, 84%). IR
-1 1
(ATR) ν 2886, 1630, 1459, 1231, 1049, 923, 778 cm . H NMR (300 MHz, CDCl ) δ 6.42 (d,
3
1
1
H, J = 8.5 Hz), 6.35 (d, 1H, J = 8.5 Hz), 5.99-6.12 (m, 1H), 5.90 (s, 2H), 5.37 (dq, 1H, J = 1.5,
13
1
7.2 Hz), 5.25 (dq, 1H, J = 1.3, 10.5 Hz), 4.50 (dt, 2H, J = 1.4, 5.4 Hz), 4.00 (s, 3H). C{ H}
NMR (75 MHz, CDCl ) δ 146.3, 143.5, 138.2, 135.0, 133.7, 117.7, 107.3, 101.4, 101.3, 71.4,
3
+
6
0.4. HRMS (ESI) m/z [M + H] Calcd for C H O 209.0814; Found : 209.0806.
11 13
4
5
-hydroxy-4-methoxy-6-(2-propenyl)-1,3-benzodioxole (17). A solution of 5-Allyloxy-4-
methoxy-1,3-benzodioxole 16 in acetone (1.5 M) was loaded in a PEEK injection loop (2 mL)
connected to a stream of acetone pumped at a flow rate of 2 mL/min as depicted in Scheme 2.
The [3,3]-Claisen rearrangement occurred in a stainless steel reactor coil (5 mL, 0.75 mm id)
heated at 250 °C. The line was constantly pressurized at 70 bar with an electronic back pressure
regulator (BPR) to prevent the vaporization of acetone. The product collected in a flask was
concentrated under reduced pressure and the resulting residue was purified by flash
chromatography (from 30% CH Cl -petroleum ether to 50% CH Cl -petroleum ether) to give
2
2
2
2
1
7 as a white solid (791 mg, 89%). mp 31.5-32 °C. IR (ATR) ν 3503, 2885, 1637, 1432, 1176,
-1 1
1045, 981, 915, 645 cm . H NMR (300 MHz, CDCl ) δ 6.33 (s, 1H), 5.88-6.02 (m, 1H), 5.85
3
1
3
1
(
s, 2H), 5.40 (br s, 1H), 5.02-5.09 (m, 2H), 4.04 (s, 3H), 3.31 (dt, 2H, J = 1.2, 6.5 Hz). C{ H}
NMR (75 MHz, CDCl ) δ 141.8, 140.0, 136.9, 134.4, 131.2, 117.7, 115.5, 103.0, 101.0, 60.0,
3
+
3
3.9. HRMS (ASAP+) m/z [M] Calcd for C H O 208.0736; Found : 208.0736.
1
1
12
4
(E)- 5-hydroxy-4-methoxy-6-(prop-1-enyl)-1,3-benzodioxole (18). A solution of t-BuOK in
THF (3.5 mL, 3.47 mmol, 1 M, 1.2 equiv) was added to a solution of 5-hydroxy-4-methoxy-6-
2-propenyl)-1,3-benzodioxole 17 (603 mg, 2.90 mmol, 1 equiv) in THF (3 mL) under argon.
(
The resulting mixture was stirred at 70 °C for 90 min and then quenched with water (50 mL)
and 5% HCl (10 mL). The aqueous layer was extracted with EtOAc (3 × 30 mL). The collected
organic extracts were dried over anhydrous MgSO and concentrated under reduced pressure.
4
Purification by flash chromatography (30% CH Cl -petroleum ether) furnished 18 as a white
2
2
solid (295 mg, 49%). mp 66-67 °C. IR (ATR) ν 3382, 2886, 1622, 1422, 1178, 1046, 977, 927,
-1 1
8
03 cm . H NMR (400 MHz, CDCl ) δ 6.61 (dq, 1H, J = 1.5, 15.8 Hz), 6.58 (s, 1H), 6.08 (dq,
3
1
H, J = 6.6, 15.8 Hz), 5.56 (s, 2H), 5.54 (s, 1H), 4.04 (s, 3H), 1.87 (dd, 3H, J = 1.7, 6.6 Hz).
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 24 of 32
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
13
1
C{ H} NMR (100 MHz, CDCl ) δ 142.3, 139.7, 134.9, 131.2, 124.9, 124.8, 116.9, 101.1, 99.0,
3
+
60.1, 18.8. HRMS (ESI) m/z [M + H] Calcd for C H O 209.0814; Found 209.0806.
1
1
13
4
Synthesis of Polemannone B (3). An automatic sample handler prepared 200 µL of the
reaction mixture from stock solutions of (E)-5-methoxy-6-(prop-1-enyl)-1,3-benzodioxol-5-ol
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
8 (0.25 M, 1 equiv), t-BuOOH 8a (0.5 M, 1.09 equiv) and [Co]-9a (0.03 M, 1 mol%). The
reaction mixture was injected in a stream of C H Cl pumped at 0.19 mL/min as depicted in
2
4
2
Scheme 2. The oxidative dimerization occurred in a PEEK reactor coil (5 mL, 0.75 mm id)
heated at 32 °C. The reactor outlet was connected to an automatic 2-way 6-port switch valve
which injected 0.2 µL of the crude mixture in the HPLC unit while the remaining stream was
collected in a fraction collector. A mixture of MeOH/H O (70/30, v/v) was used as mobile
2
phase for the HPLC analysis at a flowrate of 0.4 mL/min. A UV detector was connected to the
outlet of the HPLC column (Agela Promosil C18, 3.5 mm × 150 mm, 5 µm) to follow the
absorbance at a wavelength of 270 nm. Peak integration and yield calculation were under full
MATLAB automation. An analytical sample of polemannone B 3 was obtained by flash
chromatography (20% Et O-cyclohexane) as a white solid (25.2 mg, 84%). mp 160-161°C
2
81
-1 1
[
Lit. 170 °C]. IR (ATR) ν 2899, 1666, 1618, 1429, 1284, 1097, 1044, 888, 758 cm . H NMR
(400 MHz, CDCl ) δ 7.03-7.08 (m, 1H), 6.52 (s, 1H), 5.87 (dd, 2H, J = 1.3, 14.9 Hz), 5.65 (d,
3
2
7
H, J = 8.5 Hz), 3.96 (s, 3H), 3.93(s, 3H), 3.26 (dd, 1H, J = 2.6, 7.6 Hz), 3.17 (dt, 1H, J = 2.3,
.5 Hz) 2.45-2.52 (m, 1H), 2.17-2.26 (m, 1H), 1.13 (d, 3H, J = 7.1 Hz), 0.70 (d, 3H, J = 7.6
1
3
1
Hz). C{ H} NMR (100 MHz, CDCl ) δ 183.3, 151.7, 143.9, 143.2, 138.1, 136.0, 133.8, 131.7,
3
1
26.6, 117.0, 101.5, 101.3, 100.8, 98.9, 60.4, 59.9, 36.5, 35.8, 35.3, 34.3, 21.6, 21.2. HRMS
+
(ESI) m/z [M + Na] Calcd for C H O Na 437.1212; Found : 437.1205.
2
2
22
8
■
ASSOCIATED CONTENT
The supporting information is available free of charge on the ACS Publication website at DOI:
0.1021/acs.joc.xxxxxxx
1
Table S1 and NMR spectra of purified products: 1, 3, 6, 10, 12, 13, 14, 15, 16, 17, 18 (PDF)
■
AUTHOR INFORMATION
Corresponding author
E-mail: fx.felpin@univ-nantes.fr
ORCID
*
ACS Paragon Plus Environment

Page 25 of 32
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
François-Xavier Felpin: 0000-0002-8851-246X
Notes
The authors declare no competing financial interest.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
■
ACKNOWLEDGEMENTS
This work was supported by the “Région des Pays de la Loire” in the framework of the “Pari
Scientifique Régional SmartCat” and the INCa “Institut National du Cancer”. E. C. Aka and M.
C. Nongbe thank the "Ministère de l’Enseignement Supérieur et de la Recherche Scientifique
de Côte d’Ivoire" for a grant. E.W. gratefully acknowledges the INCa for a grant. D.C.B.
gratefully acknowledges the “Région des Pays de la Loire” for a grant. We acknowledge Julie
Hémez (CEISAM, University of Nantes) for HRMS analyses.
■
REFERENCES
(
1) Brophy, G. C.; Mohandas, J.; Slaytor, M.; Sternhell, S.; Watson, T. R.; Wilson, L. A. Novel
lignans from a cinnamomum sp. from bougainville. Tetrahedron Lett. 1969, 10, 5159-5162.
2) Bae, H.-B.; Li, M.; Son, J.-K.; Seo, C.-S.; Chung, S.-H.; Kim, S.-J.; Jeong, C.-W.; Lee, H.-
(
G.; Kim, W.; Park, H.-C.; Kwak, S.-H. Sauchinone, a lignan from Saururus chinensis, reduces
tumor necrosis factor-α production through the inhibition of c-raf/MEK1/2/ERK 1/2 pathway
activation. Int. Immunopharmacol. 2010, 10, 1022-1028.
(
3) Gao, Y.; Zhao, H.; Li, Y. Sauchinone prevents IL-1β-induced inflammatory response in
human chondrocytes. J. Biochem. Mol. Toxicol. 2018, 32, e22033.
4) Sung, S. H.; Lee, E. J.; Cho, J. H. E. E.; Kim, H. S.; Kim, Y. C. Sauchinone, a Lignan from
(
Saururus chinensis, Attenuates CCl4-Induced Toxicity in Primary Cultures of Rat Hepatocytes.
Biol. Pharm. Bull. 2000, 23, 666-668.
(5) Kim, Y. W.; Lee, S. M.; Shin, S. M.; Hwang, S. J.; Brooks, J. S.; Kang, H. E.; Lee, M. G.;
Kim, S. C.; Kim, S. G. Efficacy of sauchinone as a novel AMPK-activating lignan for
preventing iron-induced oxidative stress and liver injury. Free Radical Biol. Med. 2009, 47,
1
082-1092.
6) Kim, Y. W.; Jang, E. J.; Kim, C.-H.; Lee, J.-H. Sauchinone exerts anticancer effects by
targeting AMPK signaling in hepatocellular carcinoma cells. Chem. Biol. Interact. 2017, 261,
08-117.
(
1
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 26 of 32
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
7) Goess, B. C.; Hannoush, R. N.; Chan, L. K.; Kirchhausen, T.; Shair, M. D. Synthesis of a
10,000-Membered Library of Molecules Resembling Carpanone and Discovery of Vesicular
Traffic Inhibitors. J. Am. Chem. Soc. 2006, 128, 5391-5403.
(8) Lindsley, C. W.; Chan, L. K.; Goess, B. C.; Joseph, R.; Shair, M. D. Solid-Phase Biomimetic
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Synthesis of Carpanone-like Molecules. J. Am. Chem. Soc. 1999, 122, 422-423.
(9) Chapman, O. L.; Engel, M. R.; Springer, J. P.; Clardy, J. C. Total Synthesis of Carpanone.
J. Am. Chem. Soc. 1971, 93, 6696-6698.
(
10) Liron, F.; Fontana, F.; Zirimwabagabo, J.-O.; Prestat, G.; Rajabi, J.; Rosa, C. L.; Poli, G.
A New Cross-Coupling-Based Synthesis of Carpanone. Org. Lett. 2009, 11, 4378-4381.
11) Matsumoto, M.; Kuroda, K. Transition Metal(II) Schiff's Base Complexes Catalyzed
(
Oxidation of trans-2-(1-propenyl)-4,5-methylenedioxyphenol to Carpanone by Molecular
Oxygen. Tetrahedron Lett. 1981, 22, 4437-4440.
(12) Baxendale, I. R.; Lee, A.-L.; Ley, S. V. A Concise Synthesis of Carpanone Using Solid-
Supported Reagents and Scavengers. J. Chem. Soc., Perkin Trans. 1 2002, 1850-1857.
(13) Daniels, R. N.; Fadeyi, O. O.; Lindsley, C. W. A New Catalytic Cu(II)/Sparteine Oxidant
System for β,β-Phenolic Couplings of Styrenyl Phenols: Synthesis of Carpanone and Unnatural
Analogs. Org. Lett. 2008, 10, 4097-4100.
(14) Fadeyi, O. O.; Nathan Daniels, R.; DeGuire, S. M.; Lindsley, C. W. Total Synthesis of
Polemannones B and C. Tetrahedron Lett. 2009, 50, 3084-3087.
(15) Constantin, M.-A.; Conrad, J.; Merişor, E.; Koschorreck, K.; Urlacher, V. B.; Beifuss, U.
Oxidative Dimerization of (E)- and (Z)-2-Propenylsesamol with O2 in the Presence and
Absence of Laccases and Other Catalysts: Selective Formation of Carpanones and Benzopyrans
under Different Reaction Conditions. J. Org. Chem. 2012, 77, 4528-4543.
(16) Cortés-Borda, D.; Wimmer, E.; Gouilleux, B.; Barré, E.; Oger, N.; Goulamaly, L.; Peault,
L.; Charrier, B.; Truchet, C.; Giraudeau, P.; Rodriguez-Zubiri, M.; Le Grognec, E.; Felpin, F.-
X. An Autonomous Self-Optimizing Flow Reactor for the Synthesis of Natural Product
Carpanone. J. Org. Chem. 2018, 83, 14286-14299.
(
17) Wegner, J.; Ceylan, S.; Kirschning, A. Ten Key Issues in Modern Flow Chemistry. Chem.
Commun. 2011, 47, 4583-4592.
18) Ley, S. V. On Being Green: Can Flow Chemistry Help? Chem. Rec. 2012, 12, 378-390.
(
(19) Vaccaro, L.; Lanari, D.; Marrocchi, A.; Strappaveccia, G. Flow Approaches Towards
Sustainability. Green Chem. 2014, 16, 3680-3704.
ACS Paragon Plus Environment

Page 27 of 32
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
(
20) Gutmann, B.; Cantillo, D.; Kappe, C. O. Continuous-Flow Technology—A Tool for the
Safe Manufacturing of Active Pharmaceutical Ingredients. Angew. Chem. Int. Ed. 2015, 54,
688-6728.
21) Movsisyan, M.; Delbeke, E. I. P.; Berton, J. K. E. T.; Battilocchio, C.; Ley, S. V.; Stevens,
6
(
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
C. V. Taming Hazardous Chemistry by Continuous Flow Technology. Chem. Soc. Rev. 2016,
45, 4892-4928.
(22) Plutschack, M. B.; Pieber, B.; Gilmore, K.; Seeberger, P. H. The Hitchhiker’s Guide to
Flow Chemistry. Chem. Rev. 2017, 117, 11796-11893.
(23) Oger, N.; Le Grognec, E.; Felpin, F.-X. Handling Diazonium Salts in Flow for Organic
and Material Chemistry. Org. Chem. Front. 2015, 2, 590-614.
(24) Oger, N.; d’Halluin, M.; Le Grognec, E.; Felpin, F.-X. Using Aryl Diazonium Salts in
Palladium-Catalyzed Reactions under Safer Conditions. Org. Process Res. Dev. 2014, 18,
1786-1801.
(25) Gavriilidis, A.; Constantinou, A.; Hellgardt, K.; Hii, K. K.; Hutchings, G. J.; Brett, G. L.;
Kuhn, S.; Marsden, S. P. Aerobic oxidations in flow: opportunities for the fine chemicals and
pharmaceuticals industries. React. Chem. Eng. 2016, 1, 595-612.
(
26) Hone, C. A.; Kappe, C. O. The Use of Molecular Oxygen for Liquid Phase Aerobic
Oxidations in Continuous Flow. Top. Curr. Chem. 2018, 377, 2.
27) Osterberg, P. M.; Niemeier, J. K.; Welch, C. J.; Hawkins, J. M.; Martinelli, J. R.; Johnson,
(
T. E.; Root, T. W.; Stahl, S. S. Experimental Limiting Oxygen Concentrations for Nine Organic
Solvents at Temperatures and Pressures Relevant to Aerobic Oxidations in the Pharmaceutical
Industry. Org. Process Res. Dev. 2015, 19, 1537-1543.
(28) Ye, X.; Johnson, M. D.; Diao, T.; Yates, M. H.; Stahl, S. S. Development of safe and
scalable continuous-flow methods for palladium-catalyzed aerobic oxidation reactions. Green
Chem. 2010, 12, 1180-1186.
(
29) O’Brien, M.; Baxendale, I. R.; Ley, S. V. Flow Ozonolysis Using a Semipermeable Teflon
AF-2400 Membrane To Effect Gas−Liquid Contact. Org. Lett. 2010, 12, 1596-1598.
30) Brzozowski, M.; O’Brien, M.; Ley, S. V.; Polyzos, A. Flow Chemistry: Intelligent
Processing of Gas–Liquid Transformations Using a Tube-in-Tube Reactor. Acc. Chem. Res.
015, 48, 349-362.
(
2
(31) Rasheed, M.; Wirth, T. Intelligent Microflow: Development of Self-Optimizing Reaction
Systems. Angew. Chem. Int. Ed. 2011, 50, 357-358.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 28 of 32
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
32) Fabry, D. C.; Sugiono, E.; Rueping, M. Self-Optimizing Reactor Systems: Algorithms,
On-line Analytics, Setups, and Strategies for Accelerating Continuous Flow Process
Optimization. Isr. J. Chem. 2014, 54, 341-350.
(33) Fabry, D. C.; Sugiono, E.; Rueping, M. Online Monitoring and Analysis for Autonomous
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Continuous Flow Self-Optimizing Reactor Systems. React. Chem. Eng. 2016, 1, 129-133.
(34) Sans, V.; Cronin, L. Towards Dial-a-Molecule by Integrating Continuous Flow, Analytics
and Self-Optimisation. Chem. Soc. Rev. 2016, 45, 2032-2043.
(
35) Reizman, B. J.; Jensen, K. F. Feedback in Flow for Accelerated Reaction Development.
Acc. Chem. Res. 2016, 49, 1786-1796.
36) Henson, A. B.; Gromski, P. S.; Cronin, L. Designing Algorithms To Aid Discovery by
Chemical Robots. ACS Cent. Sci. 2018, 4, 793-804.
37) Gromski, P. S.; Henson, A. B.; Granda, J. M.; Cronin, L. How to explore chemical space
using algorithms and automation. Nat. Rev. Chem. 2019, 3, 119-128.
38) Krishnadasan, S.; Brown, R. J. C.; deMello, A. J.; deMello, J. C. Intelligent Routes to The
Controlled Synthesis of Nanoparticles. Lab Chip 2007, 7, 1434-1441.
39) McMullen, J. P.; Jensen, K. F. An Automated Microfluidic System for Online
Optimization in Chemical Synthesis. Org. Process Res. Dev. 2010, 14, 1169-1176.
40) Parrott, A. J.; Bourne, R. A.; Akien, G. R.; Irvine, D. J.; Poliakoff, M. Self-Optimizing
(
(
(
(
(
Continuous Reactions in Supercritical Carbon Dioxide. Angew. Chem. Int. Ed. 2011, 50, 3788-
3792.
(
41) Moore, J. S.; Jensen, K. F. Automated Multitrajectory Method for Reaction Optimization
in a Microfluidic System using Online IR Analysis. Org. Process Res. Dev. 2012, 16, 1409-
415.
1
(42) Skilton, R. A.; Parrott, A. J.; George, M. W.; Poliakoff, M.; Bourne, R. A. Real-Time
Feedback Control Using Online Attenuated Total Reflection Fourier Transform Infrared (ATR
FT-IR) Spectroscopy for Continuous Flow Optimization and Process Knowledge. Appl.
Spectrosc. 2013, 67, 1127-1131.
(43) Houben, C.; Peremezhney, N.; Zubov, A.; Kosek, J.; Lapkin, A. A. Closed-Loop
Multitarget Optimization for Discovery of New Emulsion Polymerization Recipes. Org.
Process Res. Dev. 2015, 19, 1049-1053.
(44) Holmes, N.; Akien, G. R.; Savage, R. J. D.; Stanetty, C.; Baxendale, I. R.; Blacker, A. J.;
Taylor, B. A.; Woodward, R. L.; Meadows, R. E.; Bourne, R. A. Online Quantitative Mass
Spectrometry for the Rapid Adaptive Optimisation of Automated Flow Reactors. React. Chem.
Eng. 2016, 1, 96-100.
ACS Paragon Plus Environment

Page 29 of 32
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
(
45) Holmes, N.; Akien, G. R.; Blacker, A. J.; Woodward, R. L.; Meadows, R. E.; Bourne, R.
A. Self-Optimisation of the Final Stage in the Synthesis of EGFR Kinase Inhibitor AZD9291
Using an Automated Flow Reactor. React. Chem. Eng. 2016, 1, 366-371.
(46) Walker, B. E.; Bannock, J. H.; Nightingale, A. M.; deMello, J. C. Tuning Reaction
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Products by Constrained Optimisation. React. Chem. Eng. 2017, 2, 785-798.
(47) Echtermeyer, A.; Amar, Y.; Zakrzewski, J.; Lapkin, A. Self-Optimisation and Model-
Based Design of Experiments for Developing a C–H Activation Flow Process. Beilstein J. Org.
Chem. 2017, 13, 150-163.
(48) Jeraal, M. I.; Holmes, N.; Akien, G. R.; Bourne, R. A. Enhanced Process Development
Using Automated Continuous Reactors by Self-Optimisation Algorithms and Statistical
Empirical Modelling. Tetrahedron 2018, 74, 3158-3164.
(49) Cherkasov, N.; Bai, Y.; Expósito, A. J.; Rebrov, E. V. OpenFlowChem – a platform for
quick, robust and flexible automation and self-optimisation of flow chemistry. React. Chem.
Eng. 2018, 3, 769-780.
(
50) Nelder, J. A.; Mead, R. A Simplex-Method for Function Minimization. Comput. J. 1965,
, 308-313.
51) McMullen, J. P.; Stone, M. T.; Buchwald, S. L.; Jensen, K. F. An Integrated Microreactor
7
(
System for Self-Optimization of a Heck Reaction: From Micro- to Mesoscale Flow Systems.
Angew. Chem. Int. Ed. 2010, 49, 7076-7080.
(52) Bourne, R. A.; Skilton, R. A.; Parrott, A. J.; Irvine, D. J.; Poliakoff, M. Adaptive Process
Optimization for Continuous Methylation of Alcohols in Supercritical Carbon Dioxide. Org.
Process Res. Dev. 2011, 15, 932-938.
(53) Jumbam, D. N.; Skilton, R. A.; Parrott, A. J.; Bourne, R. A.; Poliakoff, M. The Effect of
Self-Optimisation Targets on the Methylation of Alcohols Using Dimethyl Carbonate in
Supercritical CO2. J. Flow Chem. 2012, 2, 24-27.
(
54) Amara, Z.; Streng, E. S.; Skilton, R. A.; Jin, J.; George, M. W.; Poliakoff, M. Automated
Serendipity with Self-Optimizing Continuous-Flow Reactors. Eur. J. Org. Chem. 2015, 6141-
145.
55) Sans, V.; Porwol, L.; Dragone, V.; Cronin, L. A Self Optimizing Synthetic Organic Reactor
6
(
System Using Real-Time In-Line NMR Spectroscopy. Chem. Sci. 2015, 6, 1258-1264.
(56) Fitzpatrick, D. E.; Battilocchio, C.; Ley, S. V. A Novel Internet-Based Reaction
Monitoring, Control and Autonomous Self-Optimization Platform for Chemical Synthesis.
Org. Process Res. Dev. 2016, 20, 386-394.
ACS Paragon Plus Environment