Scaling up the Synthesis of a Hydroxyquinoline-Functionalized p-tert-Butylcalix[4]arene

Article abstract of DOI:10.1021/acs.oprd.9b00315

The optimization for an upscaled technical production of a lower rim-functionalized p-tert-butylcalix[4]arene, furnished with N₄O₄-donor ligands for superior solvent extraction separation between heavy and light lanthanides, is described. We demonstrate that reducing the polarity of the aprotic solvent in the (1,3)-distal esterification of p-tert-butylcalix[4]arene 1 does not compromise the quality or yield of product 2. It was possible to use the technical quality educt 1, that is, without prior crystallization in toluene, in conjunction with reductions in reaction time and solvent volume. The raw diester product 2 could be used, without prior recrystallization (which originally required 3 days) in the condensation reaction with hydrazine monohydrate to form hydrazide 3. Most importantly, the solvent volume required in the final condensation reaction of 3 with 8-hydroxyquinoline-2-carboxaldehyde could be reduced by an order of magnitude by using chloroform. Not only was the final disubstituted product yield improved but also the purity of the final product could be ensured by preventing the precipitation of the intermediate monosubstituted product during reaction. The filtration characteristics of the final product, as well as its solvation properties during solvent extraction of lanthanides were significantly improved.

Full text of DOI:10.1021/acs.oprd.9b00315

Subscriber access provided by - Access paid by the | UCSB Libraries
Full Paper
Scaling up the Synthesis of a Hydroxyquinoline-
Functionalized p-tert-Butylcalix[4]arene
Quirina Roode-Gutzmer, Lara N. Holderied, Florian Glasneck,
Berthold Kersting, Peter Fröhlich, and Martin Bertau
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.9b00315 • Publication Date (Web): 18 Sep 2019
Downloaded from pubs.acs.org on September 19, 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 50
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Scaling up the Synthesis of a Hydroxyquinoline-
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
Functionalized p-tert-Butylcalix[4]arene
†
†
‡
‡
Quirina I. Roode-Gutzmer, Lara N. Holderied, Florian Glasneck, Berthold Kersting,
†*
†*
Peter Fröhlich, and Martin Bertau
†
Institute of Chemical Technology, Freiberg University of Mining and Technology,
Leipziger Straße 29, 09599 Freiberg, Germany
‡
Institute for Inorganic Chemistry, University of Leipzig, Johannisallee 29,
4103 Leipzig, Germany
0
1
ACS Paragon Plus Environment

Organic Process Research & Development
Page 2 of 50
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
Table of Content Graphic
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
ACS Paragon Plus Environment

Page 3 of 50
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Abstract
The optimization for an up-scaled technical production of a lower rim functionalized p-tert-
butylcalix[4]arene, furnished with N O -donor ligands for superior solvent extraction
4
4
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
separation between heavy and light lanthanides, is described. We demonstrate that reducing
the polarity of the aprotic solvent in the (1,3)-distal esterification of p-tert-
butylcalix[4]arene 1 does not compromise the quality or yield of product 2. It was possible to
use technical quality educt 1, i.e. without prior crystallization in toluene, in conjunction with a
reduction in reaction time and solvent volume. The raw diester product 2 could be used,
without prior recrystallization (which originally required 3 days) in the condensation reaction
with hydrazine monohydrate to form the hydrazide 3. Most importantly, the solvent volume
required in the final condensation reaction of 3 with 8-hydroxyquinoline-2-carboxaldehyde
could be reduced by an order of magnitude by using chloroform. Not only was the final di-
substituted product yield improved, but also the purity of the final product could be ensured
by preventing the precipitation of the intermediate mono-substituted product during reaction.
The filtration characteristics of the final product, as well as its solvation properties during
solvent extraction of lanthanides was significantly improved.
KEYWORDS: rare earths, lanthanides, actinides, solvent extraction, calixarene synthesis,
scale up
3
ACS Paragon Plus Environment

Organic Process Research & Development
Page 4 of 50
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
INTRODUCTION
Calixarenes are cyclic oligomers, named as such in the early 1970s by Gutsche, by virtue of
1
the cup-like cavity contained within an aromatic scaffold. These macrocyclic compounds
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
therefore exhibit host-guest complex-forming capacities, where guests range from neutral to
charged metal or molecular species. The calixarene-host can essentially be designed to tune
the selectivity towards specific guest species. The appropriately functionalized calixarenes
can often be produced in appreciable yields via facile laboratory syntheses, which makes them
suitable for industrial applications. The first industrial application of calixarenes for the
2
removal of cesium from aqueous radioactive waste patented in 1984 remains industrially
3
4
relevant today. The first lanthanide calixarene complex described in the literature involved
an organic-solvent soluble complex of p-tert-butylcalix[8]arene with europium, which
demonstrated the possibility that calixarenes can be used for solvent extraction and
purification of lanthanides. The significance of calixarenes in the separation technology of
5
6
f-elements today is reflected in the research literature, as well as in patent literature.
Calixarene derivatives display an immense diversity in specialized applications. Their
stability, such as resistance to high temperatures and chemical attack, give calixarenes the
property of durability. Furthermore, the synthesis of calixarenes on an industrial scale is
7,8
becoming promising when one considers the patent literature, especially the most recent
patents.9,10 Our paper contributes towards the development of industrial calixarene
synthesis—a technology that is expected to emerge within the next few years. According to
the Global Calixarene Industry Market Research Report for 2017 global production of all
calixarenes amounted to ~ 5.5 metric tons, of which 2 tons reported to p-tert-
1
1
butylcalix[4]arene and 1.5 tons to p-tert-butylcalix[8]arene. The revenue obtained from the
latter two calixarenes amounted to ~520 EUR/kg. Only ~ 5 % of calixarenes produced
4
ACS Paragon Plus Environment

Page 5 of 50
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
globally represented more functionalized calixarenes and are considered as speciality
chemicals.
Currently, important applications for calixarenes, many of them patented, include:
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
sequestration of metals, selective separation of neutral organic molecules,13–15 stabilizers in
12
1
6
17,18
organic polymers, phase transfer agents, catalysis,
chiral calixarenes in asymmetric
1
9
20
21
catalysis, ion scavengers for electronic devices, ion and molecular recognition, and even
2
2
in biological, as well as medical applications. Particularly interesting are the recent
23
patents, including the use of calixarenes to chelate leaked electrode cations in the solid
electrolyte in lithium ion batteries,24,25 and even as cathode material for lithium ion
batteries.26,27 Of further interest is the employment of calixarenes in the selective solvent
28
extraction of lithium. Furthermore, calixarene complexes also demonstrate special optical
properties.29,30 Liquid crystal (LC) calixarenes31,32 are useful in the preparation, particularly in
terms of self-assembly in the production of functional materials deemed to be promising in
applications of display- and membrane-technologies.
The recovery of rare earth elements (which includes all the lanthanides) is complex,
3
3
laborious and cost-intensive. The separation of individual rare earth elements, as well as the
necessary removal of radioactive uranium and thorium arising from the ores, poses a great
technological challenge. The demand for rare earths in the production of permanent magnets
for wind turbine and electric vehicle technology in developed countries is expected to increase
significantly in the near future. Even though the rare-earth-containing ore reserves worldwide
are sufficient to cover this demand, the security of rare-earth supply to countries outside
3
4
China is not guaranteed. According to the most recent USGS report (2019), China currently
produces 70-80 % of the world’s supply of rare earths. Artificial supply shortages in recent
years have occurred due to increased environmental protection regulations and prevention of
illegal production in China. Resilience to supply bottlenecks for lanthanides used in
5
ACS Paragon Plus Environment

Organic Process Research & Development
Page 6 of 50
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
permanent magnets (i.e. neodymium, praseodymium and dysprosium) is low in the EU.35
Increasing global demand for rare earths concomitant with supply bottlenecks for the
production of permanent magnets, catalysts and luminescent materials necessitates a recycling
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
strategy in the near future.33
The scope of our work in a joint research and development project involved the design
of functionalized calix[4]arenes to enable selective solvent extraction of f-elements from ore
leachates. Further objectives included the scale-up synthesis of the most promising
calix[4]arene candidates with subsequent application in a pilot-scale mixer-settler solvent-
extraction operation.
To this end, modifications to a p-tert-butylcalix[4]arene scaffold were undertaken with
the objective of (i) separating f-elements from other elements in aqueous solutions arising
from ore leachates, and (ii) separating f-elements from one another, particularly actinides
from lanthanides, thereby achieving radioactive decontamination. The most suitable
molecules arising from our research that fulfill both these objectives are respectively:
(i) a tetra-phosphonate-functionalized p-tert-butylcalix[4]arene based on work by Glasneck
3
6
et al. and (ii) the hydroxyquinoline-functionalized p-tert-butylcalix[4]arene 4 described by
3
7
Jäschke et al. The solvent extraction performance of both these molecules is described by
Schreiter et al.38
Bidentate ligands containing hard O-donors are very suitable for efficient extraction of
3
9
trivalent actinides and lanthanides from solutions. In combination with a soft N-donor, the
ligand exhibits selectivity for actinides over the lanthanides due to differences in the way 5f-
and 4f-cations are complexed.39–41 Excellent selectivity for heavier lanthanides, with respect
to lighter lanthanides, was achieved by appending two 8-hydroxyquinoline-2-carboxaldehyde
units via hydrazone-carbonylmethoxy linkages in a 1,3 arrangement to the lower rim of p-tert-
3
7
butylcalix[4]arene 4. The synthesis (Figure 1), coordination properties, as well as molecular
6
ACS Paragon Plus Environment

Page 7 of 50
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
37
and crystal structures of this calixarene is described by Jäschke et al. Of all the calixarenes
we investigated, this particular hydroxyquinoline-functionalized p-tert-butylcalix[4]arene 4
demonstrated the (i) highest selectivity between light and heavy lanthanides,
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
38
i.e. β (Er/Dy) = 4.4 and β (Dy/Tb) = 2.5, in a solvent extraction process, and (ii) enabled
4
2
radioactive decontamination. To investigate the solvent extraction performance of this
functionalized calixarene 4 more extensively, it was necessary to scale up its synthesis in the
most optimal way possible. This paper describes the results of this investigation.
4
3
Figure 1: i) Synthesis of 1 according to Gutsche and Iqbal. Synthesis of 4 from 1 according to
Jäschke et al. with ii) ethyl bromoacetate, K CO , acetonitrile, reflux, overnight iii) hydrazine
monohydrate, ethanol, reflux overnight, and iv) 8-hydroxyquinoline-2-carboxaldehyde, ethanol, reflux
overnight.
3
7
2
3
RESULTS AND DISCUSSION
The laboratory synthesis for our hydroxyquinoline-functionalized p-tert-butylcalix[4]arene 4,
4
4
originally described in the PhD thesis of Jäschke, differ slightly in reaction conditions
3
7
described in the published synthesis. The original reaction conditions involved refluxing the
reactions in each reaction step at the boiling point of the respective solvents overnight
4
4
(16 hours). We investigated the minimum reaction time required to obtain maximum yield
in each reaction step. Initially, solvents in analytical reagent purity were employed. This was
7
ACS Paragon Plus Environment

Organic Process Research & Development
Page 8 of 50
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
important to observe impurities arising from other factors. Using technical grade solvents,
1
however, did not compromise the purity of the products (mp, TLC, H-NMR) obtained in our
syntheses.
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
Two constraints were identified at the outset for a commercial technical production
of 4: the first refers to a production capacity limitation in that per 100 g of final product 4, a
4
4
solvent volume of ethanol amounting to ~ 36 L (based on Jäschke) is required in the final
condensation reaction step; the second refers to an economic constraint in the production of
the N O -donor ligand, the 8-hydroxyquinoline-2-carboxaldehyde (commercially obtainable
4
4
at a price of 37 EUR/g), which typically requires the highly toxic SeO as the selective
2
oxidation agent in its synthesis.45–47 By virtue of the extreme toxicity of SeO , an up-scaled
2
in-house synthesis of 8-hydroxyquinoline-2-carboxaldehyde was not undertaken. To
circumvent this cost factor, attempts were made to attach a similar ligand moiety onto the
p-tert-butylcalix[4]arene framework, by using 6-hydroxymethylpyridine-2-carboxaldehyde in
48
the final reaction to produce the functionalized calix[4]arene 5 (Figure 2). This calix[4]arene
product 5 yielded complexation constants of similar orders of magnitude, but did not vary
sufficiently between the lanthanides to enable selectivity. This reveals that the benzene ring
attached to the hydroxyl group of the hydroxyquinoline plays a very significant role in the
selective complexation of lanthanides. Cation-π interaction with aromatic rings is well
4
9
known, the electrostatic interaction contributing significantly towards cation binding at long
range. Whether the selectivity for specific lanthanide ions of the additional aromatic ring is
driven solely by electrostatic effects, or also by steric effects would need to be established by
DFT calculations.
8
ACS Paragon Plus Environment

Page 9 of 50
Organic Process Research & Development
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 2: Ligand 5.
To minimize toxicity, the scale-up experiments reported here involved the use of
ethanol as opposed to methanol. An industrial production of 4 would typically employ the use
of methanol by virtue of the lower cost of its production. Ethanol is produced industrially by
fermentation and cannot compete commercially with the cheaper production of methanol from
50
51,52
synthesis gas. Calixarenes are known to form complexes with solvents,
which may
change the subsequent complexation behavior to cations, as well as solvation behavior in
solvent extraction processes. The differences in using methanol instead of ethanol in the
various process steps was not investigated in this work.
The various sets of experiments are summarized in Table 1.
The products of the four sets of experiments (listed in Table 1) were all used to
conduct a very detailed investigation into the solvent extraction- and re-extraction-
performance of lanthanides in the presence of other competing metal ions in aqueous
solutions. These results are described by Schreiter et al.38
9
ACS Paragon Plus Environment

Organic Process Research & Development
Page 10 of 50
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
Table 1: Summarized results of experiments performed.
Synthesis of Frame of reference
a) Batch A
b) Batch B
c) Optimum
synthesis
parameters
4
4
Jäschke
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
Commercially
Acquired
Commercially
acquired
Synthesized in-house after
Gutsche and Iqbal43
(without recrystallization)
Yield
ACROS Organics
(Purity 98-99 %)
(
w. r. t. p-tert-Butylphenol): max 62 %
mp. 338-343°C
3
Batches: ~250 g
7 h
2
Overnight (16 h)
16 h
16 h
Solvent-volume
reduced 3 x
Product
Product
Product
Raw product
No recrystallization
Yield
crystallization (1 x)
Yield
crystallization (1 x)
Yield
crystallization (2 x)
Yield
(
w. r. t. 1): 75 %
(w. r. t. 1): 78 %
mp: 182-184°C
(w. r. t. 1): 79 %
mp: 179 °C
(w.r.t. 1): 81 %
mp: 177-179 °C
-
3
4
Overnight (16 h)
12 h
12 h
10 h
Yield
Yield
Yield
Yield
(w. r. t. 2): 84 %
(w. r. t. 2): 87 %
mp: 242 °C
(w. r. t 2): 87 %
mp: 242 °C
(w. r. t. 2): 88 %
mp: 241-243 °C
-
Overnight (16 h)
12 h§
12 h
7 h
Ethanol (75 mL)ǂ
3:1 EtOH:CHCl
3
Chloroform
Solvent-volume
reduced by 95 %
Yield
Chloroform
Solvent-volume
reduced by 97 %
Yield
0
.2 g of 3
Solvent-volume
reduced by 80 %
Yield
Yield
w. r. t. 3):
(
(w. r. t. 3):
(w. r. t. 3):
(w. r. t. 3):
7
5 % (no trap)
90 % (no trap)
70 % (Dean-Stark trap) 80 % (no trap)
6
5 % (Zeozorb 3A)†
Filtration tedious
4 g 4
using 33 g of 1)
*
Improved filtration.
67 g 4
3
(
(using 82 g of 1)
ǂ The volume of ethanol is reported as being 35 mL in Jäschke et al.,37 and as 75 mL in the PhD thesis of Jäschke.44
*
§
Filtration problems cannot be realistically assessed on such small scales.
After 7 h, a sample of the reaction solution was dried and dissolved in CDCl
. 1H-NMR reveals the presence of appreciable amounts of
3
intermediate product 6.
† Small laboratory scale batch performed using a molecular sieve, Zeozorb 3A, to absorb the water of condensation (using Batch A synthesis
conditions).
1
0
ACS Paragon Plus Environment

Page 11 of 50
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Synthesis of 1
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
Scheme 1: Reaction of p-tert-butylphenol with formaldehyde under alkaline conditions to produce 1.
In-house production of educts is often associated with a cost-reduction. The industrial
7
synthesis of p-tert-butylcalix[4]arene is well-established. The essential starting material for
synthesizing our functionalized calixarene is p-tert-butylcalix[4]arene 1, which we produced
4
3
in bulk according to Gutsche and Iqbal. This one-pot base-catalyzed condensation reaction
of p-tert-butylphenol with formaldehyde is an electrophilic aromatic substitution reaction, in
which the phenol is deprotonated by the base. The respective weak C-nucleophile then attacks
the positive carbonyl carbon to yield dimers at first. Eventually linear oligomers are formed,
which in boiling diphenyl ether undergo a cyclisation through the formation of intra-
53
molecular hydrogen bonding to p-tert-butylcalix[8]arene. Upon heating a molecular mitosis
occurs in which the intra-molecular hydrogen bonding increases in strength, thereby
transforming the octamer into the thermodynamically more stable tetramer in twice the molar
quantity.54
The molecular mitosis is required to take place at higher temperatures, so that a high
boiling point (bp) organic liquid is required as reaction medium. Diphenyl ether with a bp of
258 °C is very suitable, but in industrial productions of 1, cheaper alternatives can be found
7
with petrol fractions with appropriate boiling point ranges. In order to attain the temperature
required for the condensation of the linear oligomer (the precursor), the water of condensation
4
3
44
needs to be removed. This reaction as performed by Gutsche and Iqbal and Jäschke utilizes
nitrogen to purge the solvent of water vapor, which results in the undesirable formation of
1
1
ACS Paragon Plus Environment

Organic Process Research & Development
Page 12 of 50
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
foam. Ideally, a Dean-Stark trap in conjunction with a condenser should be employed to
remove the water of condensation.7,55
The melting point of in-house produced 1 (Batch B, Table 1) was measured to be 338-
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
43
3
43 °C, which is comparable to 342-344 °C obtained in the literature. The melting point
measurement suggests the absence of p-tert-butylcalix[8]arene, which possesses a
5
6
considerably higher mp at 411-412 °C. By virtue of p-tert-butylcalix[8]arene being
4
3
insoluble in toluene, the raw product can be recrystallized in boiling toluene. In our case,
this step was not necessary, and would result in a yield loss (w. r. t. p-tert-butylphenol) from
6
1 % to 49 %.43
The IR-spectrum revealed the absence of p-tert-butylcalix[8]arene, which would be
-1
57
apparent by a shift of the O–H stretching vibration from 3165 cm for the tetramer to a
-1
higher wavenumber of 3225 cm for the octamer (Figure S1, supporting information). The
O-H stretching frequencies of p-tert-butylcalix[n]arenes (n = 4-20) are listed by Stewart and
Gutsche.58 Spectroscopic analysis, IR (Figure S2, supporting information) and H-NMR
1
(Figure S3, supporting information), of in-house produced 1 reveals traces of acetic acid and
ethyl acetate arising from the washing step.
Heating 1 at 120 °C in air did not achieve a removal of the solvent residues.
Furthermore, the acquisition of product 2 in the next synthesis step was deleteriously affected
by prior thermal treatment of 1 in air. A calixarene can easily exchange a complexed solvent
molecule for another kind of solvent molecule, but there is no energetic advantage for a
calixarene to give up its solvent guest molecule entirely. In an NMR crystallographic study of
p-tert-butylcalix[4]arene complexes with solvents, it is very apparent that a considerable
5
9
structural change occurs when the calixarene molecule is guest-free. At raised temperatures,
especially in air, calixarene tends to decompose. In general, TGA-DSC analysis reveals that
the removal of solvent molecule(s) from a calixarene to be an endothermic process, whereas
the subsequent molecular rearrangement and eventual decomposition of the calixarene to be
1
2
ACS Paragon Plus Environment

Page 13 of 50
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
60
an exothermic process (e.g. in work by Morales et al.) Commercially, 1 is purified by
recrystallization in boiling toluene, a time-consuming and cost- and energy-intensive process
step. The toluene of crystallization can then be removed under high vacuum > 140 °C.43
Removing solvents of crystallization from 1 is generally unnecessary prior to employment in
the synthesis of 2. It is expected, however, that the type of solvent complexed within the
calixarene may influence solvation characteristics.
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
Since the employment of the in-house produced 1, containing the solvent residues, did
not compromise the yield or quality of the diester product 2, or any of the subsequent products
3
and 4, a purification step was unnecessary.
Synthesis of 2
Scheme 2: Reaction of 1 to 2.
The (1,3)-distal esterification of 1 to form 2 involves a nucleophilic substitution reaction, bi-
molecularly, i.e. via an S 2 mechanism.44,61 This reaction has been fairly well investigated
N
and optimized.61–64 The selectivity of the functionalization is sensitive to the basicity and type
of solvent. Calixarenes are weak acids, retaining similar properties of phenolic hydroxyl
groups, each hydroxyl group having a different pKa value. Distal di-alkylation of p-tert-
butylcalix[4]arene is favored using a weak base such as K CO with a limiting amount of
2
3
6
5
alkylating agent. For the di-protonation of the calixarene, spatial separation of the anion is
1
3
ACS Paragon Plus Environment

Organic Process Research & Development
Page 14 of 50
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
66
necessary, a factor that makes the choice of solvent important. The resulting phenoxide is a
good nucleophile, which is sterically hindered by the intra-molecular hydrogen-bonded
calix[4]arene scaffold. The halide in ethyl bromoacetate is attached to a primary carbon and is
therefore a good leaving group. It is important to emphasize that ethyl bromoacetate is
classified as highly toxic and may be fatal upon inhalation or upon contact with skin.
Replacing the alkylating agent with ethyl chloroacetate, however, should not be considered as
a cheaper alternative, because it is even more hazardous than ethyl bromoacetate.
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
Using a polar aprotic solvent, thereby not participating in hydrogen bonding with the
nucleophile itself, renders the nucleophile naked. Ideally, the polar aprotic solvent should also
solvate the leaving group. A protic solvent, on the other hand, would stabilize the nucleophile
by hydrogen bonding and promote the reverse reaction towards the educts. The weak base
K CO (with pKa = 10.25), therefore, ensures a 1,3-distal deprotonation of 1. The ethyl
2
3
bromoacetate is employed in a molar quantity in slight excess twice that of the calixarene.
The synthesis of 2 was first performed with commercially obtained calixarene 1
3
7
(
Batch A, Table 1). The laboratory scale synthesis of Jäschke et al. employed the use of
acetonitrile to produce 2. The cheapest alternative polar aprotic solvent is acetone, which has
a slightly lower dipole moment than acetonitrile, i.e. 2.9 D compared to 3.9 D, respectively.
For a technical production of our calixarene, it makes sense to employ acetone as solvent, as it
is produced in greater supply inexpensively. This is because acetone is formed in significant
quantities as a by-product from the Hock process in the industrial production of phenol.
Acetonitrile, on the other hand, is produced in much smaller quantities in the industrial
production of acrylonitrile. Acetone is also a relatively harmless substance, whereas
acetonitrile is metabolized to cyanide when absorbed through the skin or lungs.
The synthesis of 2 has already been reported using acetone as a solvent.61–64 The
dissolved product was filtered to remove the K CO . After evaporation the product was
2
3
1
4
ACS Paragon Plus Environment

Page 15 of 50
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
dissolved in CH Cl , to which ethanol was added in a ratio of 5:1 as was done for the
2
2
3
7
laboratory scale synthesis. Crystallization of the product required at least 3 days and was
1
accelerated at -30 °C. An impurity-free diester product 2 (mp, TLC, H-NMR) could be
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
obtained in a yield of 78 % (Batch A) comparable to that found in the literature.37,62–64
Replacing the commercially acquired educt 1 with in-house produced 1 (containing solvent
impurities acetic acid and ethyl acetate) did not compromise the quality (mp, TLC, IR,
1H-NMR) or yield (with respect to 1) of the diester product 2.
61
Contrary to the results of Wall, we observed a higher temperature-dependent
solubility of the educt in acetone compared to acetonitrile (Figure 3). Despite the lower bp of
acetone (56 °C) relative to acetronitrile (82 °C), it was possible to reduce the solvent volume
37
three-fold relative to the laboratory scale of Jäschke et al. Reducing the reaction temperature
down to 45 °C, i.e. below the bp of acetone, resulted in a decrease in yield of 2 from 75 % to
57 %. This is in accordance with the van’t Hoff rule, that a reduction in reaction temperature
of 10 K decreases the reaction rate of the reaction two- to three-fold, which necessitates
longer reaction times to achieve the maximum yield.
The crystallization step is time- and energy-consuming, which when implemented on a
technical scale, would require recycling a two-solvent crystallization-solution. In Batch B,
product 2 was re-crystallized twice in order to ensure the removal of possible impurities
arising from the in-house produced educt. Subsequent experiments (c in Table 1)
demonstrated that a recrystallization of product 2 did not contribute to an observable
1
5
ACS Paragon Plus Environment

Organic Process Research & Development
Page 16 of 50
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: Solubility of 1 in acetone and acetonitrile (500 mg solid in 15 mL solvent for 4h) as a
function of temperature.
improvement in purity (IR, NMR). The raw diester product 2 (without re-crystallization) was
1
investigated by H-NMR (Figure S3, supporting information). The main impurity was shown
to be unreacted ethyl bromoacetate, which is added in slight excess in the synthesis. As ethyl
bromoactetae is soluble in ethanol, the raw diester product 2 can simply be recrystallized in
ethanol, which required only 30 minutes to achieve a product in a yield of 77 % free of ethyl
bromoacetate.
The re-crystallization process adapted from the laboratory synthesis of Jäschke et al.37
required 3-5 days to harvest the product from a two-stage crystallization process. Subsequent
experiments (c in Table 1) showed that the recrystallization step was not necessary. As the
reaction of 2 to 3 takes place in ethanol, residues of ethyl bromoacetate would be diluted in
the reaction solution and mostly remain in solution when product 3 is crystallized out. The
quality and yield of product 3 was not at all compromised by omitting the recrystallization of
the diester product 2. It was, in fact, found that it is also irrelevant whether educt 1 was
commercially acquired or in-house produced (containing traces of acetic acid and ethyl
acetate solvent residues).
Yield of 2 (with respect to 1) was investigated as a function of reaction time, and it
was found that a reaction time of 7 hours at the refluxing temperature was necessary to reach
1
6
ACS Paragon Plus Environment

Page 17 of 50
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
67
the maximum yield (Figure 4). A microwave synthesis of 2 has been shown to reduce the
reaction time to 30 minutes when microwave radiation at 300 W was employed, and even
increased the yield of 2 (with respect to 1) to 85 %. There are no known industrial processes,
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
however, that employ microwave-assisted synthesis of organic compounds.68
Figure 4: Yield of products 2, 3 and 4 (w. r. t. 1, 2, 3), respectively synthesized in optimum minimum
volumes of acetone for 2, ethanol for 3 and chloroform for 4.
Synthesis of 3
Scheme 3: Reaction of 2 to 3.
1
7
ACS Paragon Plus Environment

Organic Process Research & Development
Page 18 of 50
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 the condensation reaction of diester 2 to produce the hydrazine-functionalized
calix[4]arene 3, the carbonyl of the ester functional groups undergoes nucleophilic attack via
a nitrogen atom arising from the hydrazine monohydrate. The ethanolate molecule leaves the
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
44
calixarene molecule by means of an S 2 reaction mechanism. Jäschke found that the
N
product of this reaction could be produced in high purity and yield (84 %). Our work
1
confirmed this observation (TLC, H-NMR), which essentially renders the solvent volume as
the only parameter requiring optimization. The solvent choice of ethanol is in accordance with
using alcohols as the typical solvent for reactions of esters with hydrazine to form hydrazides.
In addition, the hydrazine monohydrate, which is employed in an excess molar quantity in this
reaction, is soluble in ethanol. No evidence of hydrazine could be observed in the final
1
product ( H-NMR). It was also found that washing the filtered product 3 with water, as done
44
by Jäschke, was not necessary. It was clearly observed in Batch A and B that washing the
filter cake with water resulted in the filtrate becoming cloudy with an associated final yield
loss of 1 %. This product displayed optimal filtration characteristics and, therefore, should not
pose problems in a scaled up technical production.
The solubility of the diester 2 in ethanol dictates the minimum reaction temperature
and minimum solvent volume of 40-45 °C (Figure 5).
Figure 5: Solubility of 2 in ethanol (500 mg solid in 15 mL solvent for 4h) as a function of
temperature.
1
8
ACS Paragon Plus Environment

Page 19 of 50
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Investigating the yield of the hydrazide product 3 (with respect to 2) as a function of
reaction time, it can be established that a reaction time of 10 hours is necessary at the
refluxing temperature in ethanol (78 °C) (Figure 4). The minimum solvent volume is dictated
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
44
by the solubility of 2. The solvent volume relative to that of Jäschke’s, could only be
decreased negligibly, i.e. by 3 %.
Synthesis of 4
Scheme 4: Reaction of 3 to 4.
The synthesis of the final product 4 involves an S 2t condensation reaction (where ‘t’ refers
N
3
to the tetragonal sp intermediate step) between 8-hydroxyquinoline-2-carboxaldehyde
(a Schiff base) and the hydrazide 3: the electrophilic carbonyl atom from the Schiff base is
attacked by the nucleophilic nitrogen atom of 3. After a proton rearrangement, water leaves as
a condensation product.
The synthesis of our hydroxyquinoline-functionalized calix[4]arene 4 as described by
37
Jäschke et al. requires 35 mL ethanol in relation to 0.5 g educt 3 in the final condensation
reaction. The educts were also separately dissolved in ethanol, and the solution containing
8
-hydroxyquinoline-2-carboxaldehyde was added drop-wise during the reaction. We found
that the amount and type of solvent was insufficient to ensure complete reaction to 4, by
1
9
ACS Paragon Plus Environment

Organic Process Research & Development
Page 20 of 50
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
virtue of the formation of the insoluble mono-substituted intermediate 6 (Figure 6) (seen by
1
additional signals in the H-NMR spectrum arising from phenolic protons on the quinoline
substituents, and the nitrogen-bound protons) (Figure 7).
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
6
Figure 6: Mono-substituted intermediate product 6.
4
4
In the original description of the synthesis, 75 mL of ethanol was used per 0.2 g
educt 3. In this educt to solvent ratio, we found that despite long reaction times, that
appreciable amounts of the mono-substituted intermediate 6 were found in our final product
(1H-NMR) (Figure 7).
2
0
ACS Paragon Plus Environment

Page 21 of 50
Organic Process Research & Development
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
1
Figure 7: H-NMR spectrum (400 MHz, CDCl ) of 4 produced in (3:1) CH OH:HCCl (right).
3
3
3
Signals δ (ppm): 11.82 (s, 2H, OH)(designated as 9 for the di-substituted product 4) and 8.52
(s, 2H, N=CH)(designated as 7 for the di-substituted product 4). When the mono-substituted product 6
is present in the raw product (left), additional signals are observed at 11.88 ppm (1H, quinoline-OH),
and 8.49 ppm and 9.85 ppm (2H,–NH).
In order to free up reactor capacity it was necessary to adjust the solvent composition
in such a way that the mono-substituted intermediate 6 remained in solution enabling it to
react further with 8-hydroxyquinoline-2-carboxaldehyde to form the final di-substituted
product 4. It was found that the solvent volume could be reduced by 80 % by increasing the
chloroform content in ethanol from 0 to 25 %. This also resulted in a final product yield of 4
37
(
with respect to 3) from 75 % (laboratory scale according to Jäschke et al.) to 90 %. At the
start of the reaction, all educts are solvated. The product precipitates during the course of the
3
8
reaction. No improvement in the filtration could be obtained. Subsequent TGA-IR results of
product 4 synthesized in ethanol:chloroform (3:1) revealed that the solvent molecules are
fairly strongly bound in the calixarene molecule: EtOH Δm = -2.5 % at 156 °C (bp 78 °C),
CHCl Δm = 1.1 % at 230 °C (bp 61 °C). Considering the densities of ethanol and
3
chloroform (0.789 and 1.49 g/mL respectively), ethanol is included in a higher weight ratio to
2
1
ACS Paragon Plus Environment

Organic Process Research & Development
Page 22 of 50
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
chloroform than expected by the volume ratio of 3:1. The high temperature at which
chloroform is removed from the calixarene host is indicative of favorable binding energetics
of chloroform as guest in the calixarene host.
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
The improvement that resulted from the addition of chloroform to the ethanol solvent,
begged the question whether the synthesis could be conducted solely in chloroform, i.e.
without any ethanol. This proved to be the case, and the results exceeded our expectations.
Using chloroform in only 5 % of the solvent volume used in the original laboratory synthesis
44
1
of Jäschke, it was possible to produce a pure product 4 (TLC, H-NMR, IR) free from the
mono-substituted intermediate with a yield of 70 % (with respect to 3) (Batch B). All
substances (educts, products, intermediate products) remained in solution throughout the
entire reaction time. The solution at the outset is a golden yellow (as in Batch A and
Jäschke’s37,44 experiments) arising from the 8-hydroxyquinoline-2-carboxaldehyde. As the
product started forming, the color of the solution turned a deep olive green (Figure 8). This
color was not observed when the solvent contained ethanol either completely or in various
ratios with chloroform. The extended π-conjugated calixarene scaffold that results from
solvation of 4 in chloroform therefore acts as a chromophore, thereby changing the visible
spectral properties from yellow to green.
2
2
ACS Paragon Plus Environment

Page 23 of 50
Organic Process Research & Development
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 8: Yield of 4 (w. r. t. 3) as a function of solvent used and whether condensation water was
removed from reaction, either by using a a) molecular sieve in the reaction solution (Zeozorb 3A) or a
b) Dean-Stark trap for removing water from denser solutions. Small round-bottom flasks indicate
laboratory scale synthesis, whereas large round bottom flasks indicate up-scaled synthesis.
The dissolved product 4 was allowed to cool to room temperature and then
precipitated by adding ethanol to the reaction mixture in a ratio of 3:1 (ethanol:chloroform)
(Figure 9). Upon adding ethanol, the color of the solution changed instantaneously from green
to yellow, while the solid product that crystallized out is colorless (white, tainted slightly
yellow). This solution contains a significant amount of product particles in suspension. This
mixture was allowed to settle (ripen) overnight. The filtration process was improved
significantly (compared to Batch A), despite the average particle-size remaining in the order
ACS Paragon Plus Environment
2
3

Organic Process Research & Development
Page 24 of 50
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 5 µm. The filtration properties in this case, therefore, cannot be attributed per se to the
particle size of the product. It is, however, important to be cognizant of the fact that samples
analyzed for particle-size were subjected to prior ultrasonic treatment.
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 9: Batch B: The reaction solution immediately turns yellow upon addition of ethanol. The
white product 4 precipitates slowly (crystalizes) out of solution.
The effect we are observing regarding filtration characteristics is likely to be partly
ascribable to non-covalent interactions between calixarene molecules (π-π interactions
between aromatic rings of separate calixarene particles) which will be influenced by the type
of solvent complexed within the calixarene molecules and the solvent solvating the calixarene
complex. Ultrasonic treatment is known to break π-π interactions between aromatic rings
(observed, for example, for the separation of graphene flakes from one another). The particle-
size analysis, therefore, does not necessarily reconcile with the real particle-size during
filtration.
The differences in filtration characteristics that have been observed between the final
product synthesized solely in chloroform relative to the product synthesized in solvents
containing ethanol is ascribed to structural changes in the host-guest complex. The crystal
modification (i.e. the packing structure) that the precipitated calixarene adopts, is influenced
ACS Paragon Plus Environment
2
4

Page 25 of 50
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
13
by the kind of solvent molecule that is complexed within the calixarene host. This
phenomenon can be explained by the CH-π interactions, which plays an important role in
6
9
maintaining the structure of host-guest compounds in solution, as well as the crystal packing
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
70
of organic compounds. The type of solvent guest in the calixarene host will influence the
degree of CH-π interactions, which again influences solvation enthalpy as a function of
solvent polarity (or permittivity).71
What is interesting to note that even though the formed product remained solvated in
chloroform, once it was precipitated / crystallized out with ethanol, it was not possible to re-
dissolve the product in chloroform. It can, therefore, be assumed that the product is present as
solvated monomers and dimers prior to precipitation / crystallization.
It was possible to produce 67 g of the hydroxyquinoline-functionalized p-tert-
butylcalix[4]arene 4, which was subsequently employed in solvent-extraction investigations
3
8
by Schreiter et al. The solvent-extraction results using this product, revealed that a complete
removal of the lanthanides in the re-extraction step was possible, which with previous
products synthesized in the presence of ethanol (synthesis batch A), either on its own or in
various ratios with chloroform, yielded compromised re-extraction performance in this regard.
Furthermore, the solubility characteristics of product 4 synthesized solely in chloroform were
advantageous in the solvent extraction procedure.38
In Batch B the amount of solvent used to produce 4 from 3 was reduced to 5 % of the
4
4
volume used by Jäschke. In subsequent optimization experiments (c, Table 1) the solvent
volume could be reduced to a minimum of 3 % of the volume used by Jäschke.44
In terms of solvent volume when using chloroform, we found that the lower the
solvent volume employed in the reaction, the faster the product precipitated when ethanol was
added. When 10 mL of chloroform was used (for every 0.2 g hydrazide 3 employed),
2
5
ACS Paragon Plus Environment

Organic Process Research & Development
Page 26 of 50
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
product 4 could be obtained after 24 hours, whereas when 50 mL was used, 3 days were
required to precipitate the product. The final yield of 4, however, remained constant
regardless of the volume of chloroform used.
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
Reaction time was also optimized, where 9 hours was sufficient to secure a maximum
yield of 4 (80 % with respect to 3) in chloroform (Figure 4).
An important aspect to consider is whether the water of condensation enhances the
intra-molecular hydrogen bonding in an intermediate stage in the synthesis of 3 to 4. That
water is included as a guest molecule in the host-guest structure of product 4 is already
3
7
crystallographically confirmed in the study by Jäschke et al. We observe reduced yields
when water of condensation of the reaction is removed (Figure 8). In a small laboratory scale
experiment using the same solvent conditions (type and volume ratio to solid) as in Batch A,
we added a molecular sieve (Zeozorb 3A) to the reaction solution in order to achieve in situ
removal of condensation water. We observed a lower yield of 4 (w. r. t. 3), i.e. 65 %, as
opposed to 70 % obtained in Batch A without using a Dean-Stark trap. In the optimized
synthesis we achieve a yield of 80 % of 4 (w. r. t. 3) when no trap is used, compared to 70 %
for Batch B using a Dean-stark trap (Table 1). A systematic study would need to be conducted
to investigate this phenomenon quantitatively.
Synthesis optimum for 4 from 1.
Optimization in this work involved reducing solvent volume capacity, minimizing the number
of process steps, reducing total process time and chemical costs, at maximum yield (target
value). A final synthesis from 1 to 4 was performed on laboratory scale with all the optimized
parameters. A maximum yield of 4 (w. r. t. 1) of 57 % was obtained, which represents an
4
4
increase of 8 % relative to the original synthesis of Jäschke, which served as frame of
reference for our work. Product loss occurs in each process step. The process step involving
2
6
ACS Paragon Plus Environment

Page 27 of 50
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
the recrystallization of 2 represents the largest loss. With the implementation of all the
optimum parameters, it was possible to produce 4 (with no signals for impurities in the NMR
spectra) within 36 hours (as opposed to ~ 100 hours). Most importantly, it was possible to
reduce the solvent volume required in the last reaction step, thereby creating volume capacity
by an order of magnitude (Table 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
Table 2: Solvent volume (L) used to produce 100 g of 4 from 3.
Batch A
Batch B
CHCl3
Jäschke44
After optimization
1.0
3
:1 (EtOH:CHCl3)
3
6
7.5
1.9
Chemical consumption and costs.
The consumption and cost of chemicals required to produce 100 g of 4 relative to that used in
the original synthesis is compiled in Table S1 (supporting information). The costs of solvents
in various grades and the chemicals required to synthesize 4 are contained respectively in
Tables S2 and S3 (supporting information).
In the calculation of costs in Table S1 diphenyl ether of analytical reagent grade is
used to produce half the required amount of 1 (with a yield of 57 %). This diphenyl ether is
re-used to produce the second half of the required amount of 1 (with a yield of 62 %). The
chemical costs excluding value added tax (VAT) to produce the required amount of 1 came to
3
2.82 EUR, of which 19 EUR reports to the diphenyl ether (which we used in analytical grade
quality). Due to diphenyl ether not being in continuous supply at our Institute, we have no
bulk price available. Furthermore, it is not possible to purchase technical grade diphenyl ether
in small quantities. The cost of producing 1 in-house amounts to ~ 315 EUR/kg (excluding
energy, equipment and other operational costs) compares competitively to the revenue that
2
7
ACS Paragon Plus Environment

Organic Process Research & Development
Page 28 of 50
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
11
can be gained at 520 EUR/kg (selling price). Considering, however, that ~ 58 % of the in-
house production cost of 1 reports to analytical grade diphenyl ether, there is a significant cost
reduction to be expected when the diphenyl ether is replaced with an inexpensive petroleum
fraction.
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
The purchasing cost of 1 to synthesize 100 g of 4 amounts to 589 EUR. The in-house
production of 1, therefore, reduces chemical costs by 94 %. Costs for energy, equipment and
miscellaneous are not considered.
The total chemical cost to produce 100 g of 4 under optimal conditions amounts to
1
567 EUR, of which 1 517 EUR reports to the expensive 8-hydroxyquinoline-
2-carboxaldehyde. All other chemicals combined contribute 50 EUR to the total chemical
costs. If ethanol were to be replaced by methanol, the cost reduction only amounts to 3 EUR
(per 100 g of 4)
The optimization of the original synthetic process resulted in a total chemical cost
reduction of 48 %. The most cost-intensive chemicals (in EUR) required to synthesize 100 g
of 4 under optimal synthesis conditions are shown in Table 3.
7
Diphenyl ether can of course be replaced by inexpensive crude oil fractions, which
would contribute significantly to a reduction in the cost of producing 1. When the in-house
production of 1 already reduces chemical costs by an order of magnitude, it is expected that a
cost reduction of similar magnitude is possible when 8-hydroxyquinoline-2-carboxaldehyde is
produced in-house.
Material flows for a possible industrial production of our calixarene is shown in a
schematic diagram in Figure 10.
Table 3: The most cost-intensive chemicals required to produce 100 g of 4 under optimal conditions.
2
8
ACS Paragon Plus Environment

Page 29 of 50
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Chemical compound
-Hydroxyquinoline-2-carboxaldehyde
Price in EUR (excl. VAT)
8
1 517
19
Diphenyl ether
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
Hydrazine monohydrate
Ethyl bromoacetate
18
12
2
9
ACS Paragon Plus Environment