Rapid and scalable synthesis of chiral porphyrin cage compounds

Article abstract of DOI:10.1016/j.tet.2019.07.009

An improved and scalable synthetic route to chiral porphyrin cage compounds, which will be used as catalytic machines for the encoding of information into polymers, has been developed. The porphyrin cage was made chiral by introducing one or two nitro gro

Full text of DOI:10.1016/j.tet.2019.07.009

Accepted Manuscript
Rapid and scalable synthesis of chiral porphyrin cage compounds
Pieter J. Gilissen, Anne Swartjes, Bram Spierenburg, Jeroen P.J. Bruekers, Paul
Tinnemans, Paul B. White, Floris P.J.T. Rutjes, Roeland J.M. Nolte, Johannes A.A.W.
Elemans
PII:
S0040-4020(19)30751-3
https://doi.org/10.1016/j.tet.2019.07.009
TET 30451
DOI:
Reference:
To appear in: Tetrahedron
Received Date: 18 May 2019
Revised Date: 1 July 2019
Accepted Date: 5 July 2019
Please cite this article as: Gilissen PJ, Swartjes A, Spierenburg B, Bruekers JPJ, Tinnemans P, White
PB, Rutjes FPJT, Nolte RJM, Elemans JAAW, Rapid and scalable synthesis of chiral porphyrin cage
compounds, Tetrahedron (2019), doi: https://doi.org/10.1016/j.tet.2019.07.009.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Graphical Abstract
To create your abstract, type over the instructions in the template box below.
Fonts or abstract di
Rapid and scalable synthesis of chiral
porphyrin cage compounds
Leave this area blank for abstract info.
Pieter J. Gilissen, Anne Swartjes, Bram Spierenburg, Jeroen P. J. Bruekers, Paul Tinnemans, Paul B. White,
Floris P.J.T. Rutjes, Roeland J.M. Nolte, Johannes A.A.W. Elemans
Radboud University, Institute for Molecules and Materials, Heyendaalseweg 135, 6525 AJ, Nijmegen, The
Netherlands. E-mail J.Elemans@science.ru.nl, R.Nolte@science.ru.nl
An improved and scalable synthetic route to molecular catalytic machines, to encode information onto
polymers, has been developed. Regioselective nitration reactions have been performed to introduce chirality on
the catalyst. The mechanism behind the regioselectivity is herein investigated.

ACCEPTED MANUSCRIPT
Rapid and scalable synthesis of chiral porphyrin cage compounds
Pieter J. Gilissen#, Anne Swartjes#, Bram Spierenburg, Jeroen P.J. Bruekers, Paul Tinnemans,
Paul B. White, Floris P.J.T. Rutjes, Roeland J.M. Nolte*, Johannes A.A.W. Elemans*
Radboud University, Institute for Molecules and Materials, Heyendaalseweg 135, 6525 AJ, Nijmegen, The Netherlands. E-mail
J.Elemans@science.ru.nl, R.Nolte@science.ru.nl
# These authors contributed equally to this research
Abstract
An improved and scalable synthetic route to chiral porphyrin cage compounds, which will be
used as catalytic machines for the encoding of information into polymers, has been developed.
The porphyrin cage was made chiral by introducing one or two nitro groups on its xylylene
sidewalls. This nitration was performed with fuming nitric acid at low temperature and
occurred in a highly regioselective fashion. The latter was thought to be the result of the
binding of a nitronium cation inside the cavity of the cage compound, directing the reaction to
the sidewalls. However, 1H NMR titrations of the porphyrin cage compound with either nitric
acid or the nitronium salt [NO2][BF4] revealed that the effect of the host-guest binding of the
nitronium ion on the selectivity of the reaction is negligible. Instead, protonation of the
porphyrin plays an essential role as it prevents the ring from being oxidized, allowing the
nitration to be directed to the sidewalls.
Keywords: Glycoluril; Porphyrin; Host-guest chemistry; Cage molecule; Chirality; Nitration.
1. Introduction
The amount of information that is produced nowadays is increasing exponentially and will
lead to storage problems in the near future1,2. As a result, new approaches to store data are
currently being developed and an interesting option is to use biological (e.g. DNA) and
synthetic polymers for this purpose3–7. We have started a program to write and store digital
information onto polymers in the form of chemical functions with the help of catalytic
machines that thread onto a polymer chain, and while moving along it, write (R,R)- and (S,S)
epoxides representing the digits 0 and 18,9. The aforementioned catalytic machines are based
1

ACCEPTED MANUSCRIPT
on a chiral manganese(III) porphyrin cage compound derived from diphenylglycoluril. The
synthesis of the parent achiral porphyrin cage compound H21 (Figure 1) was first reported in
1999, and required 9 steps10. Later in 2007 the first improvements on the laborious synthesis
of the porphyrin cage were made11. With the challenging goals of writing and storing
information on polymers with catalytic machines in mind, our aim is to further reduce the
number of synthetic steps and to increase the overall yield, such that also substantial amounts
of chiral derivatives of H21, i.e. compounds H22 and H23, are easily accessible. These
compounds are the starting materials for the machines that can write chiral epoxides.
The introduction of chirality into the porphyrin cage compound (H22 and H23) was reported
in an earlier study12, i.e. by treating cage compound H21 with fuming nitric acid, which lead
to functionalization of exclusively the xylylene sidewalls with one or two nitro-groups.
In.addition to conferring chirality, the nitro groups also serve as excellent handles to further
functionalize the cage13. Remarkably, the nitration occurred in a highly regioselective fashion,
which was initially thought to be a consequence of a nitronium cation being bound in the
cavity of the cage compound12. Because of its proximity it would then only nitrate the
aromatic rings of the sidewalls. Interestingly, it has been reported that a simple nitration of
meso-tetraphenylporphyrin (TPP) with HNO3 can nitrate the β-pyrroles, as well as the phenyl
rings14–17. Other work on the nitration of of meso-aryl porphyrins using a radicalar18 (NO2•) or
−
anionic19 (NO2 ) nitro-source revealed selectivity of nitration for the β-pyrrolic positions of
the porphyrin ring. These previous findings suggest that vastly different mechanisms are at
play during the nitration of H21. In the present study, we address this issue further by
performing additional studies that aim to shine light on the mechanism behind the observed
regioselectivity.
Figure 1. Porphyrin cage compounds
2

ACCEPTED MANUSCRIPT
2. Results and discussion
Our improved and shorter synthetic route towards the parent porphyrin cage compound H21
is shown in Scheme 1. The synthetic strategy encompasses two parts: the synthesis of the
tetratosyl clip compound (7) and the synthesis of the porphyrin roof (8). The clip molecule
was constructed starting from diphenylglycoluril, which was prepared in up to quantitative
yield (70–99%) by the acid-catalyzed condensation of benzil and urea in toluene using a
Dean–Stark apparatus. Because diphenylglycoluril is hardly soluble in any common organic
solvent, it was further functionalized to cyclic ether 5 in reasonable yield (50–70%) by a
reaction with formaldehyde and a subsequent acid-catalyzed ring closing reaction. Cyclic
ether 5 is a weak electrophile in Friedel–Crafts alkylation reactions and, hence, in the past
more reactive intermediates were synthesized from this compound for the attachment of the
aromatic sidewalls (see Supporting Information, Scheme S1). These sidewalls are derived
from pyrocatechol. Previously, the direct bis-alkylation of this compound with 2-
chloroethanol turned out to be cumbersome, as the product was formed in low yield (<30%).
We found that under the reaction conditions, 2-chloroethanol is transformed into oxirane
(detected by 1H NMR), which either reacts with deprotonated catechol, or leaves the reaction
mixture as a gaseous byproduct. By keeping the pH of the reaction mixture between 9 and 11,
catechol was always present in its deprotonated form. In this way, a multigram scale
alkylation of catechol with 2-chloroethanol in water gave the diol intermediate in quantitative
yield, without the need for further purification. Subsequent tosylation of the hydroxy groups
afforded the sidewall molecule 6 in good yield after a single recrystallization step.
3

ACCEPTED MANUSCRIPT
Scheme 1. Synthesis of cage compound H21. Reagents and conditions: a) CO(NH2)2, TFA,
PhMe, reflux; b) HCHO, NaOH, DMSO/H2O, 20 °C; then HCl, reflux; c) 2-chloroethanol,
KOH, H2O, 60 °C; d) TsCl, pyridine, DCM, 0 → 20 °C; e) SOCl2, ZnCl2, DCM, 20 °C; f)
SOCl2, Sn(OTf)2 (cat.), DCM, 20 °C; g) pyrrole, EtCO2H, reflux; then DDQ, CHCl3,
reflux; h) BBr3, DCM, −25 → 20 °C; i) K2CO3, MeCN, reflux.
The Friedel–Crafts alkylation reaction between 6 and the tetrachloromethyl derivative (see
Supporting Information, Scheme S1) of 5 had always been performed under reflux conditions
in 1,2-dichloroethane, moderated by a stoichiometric amount of tin(IV) chloride as the Lewis
acid, to furnish clip molecule 7 in moderate yield (typically 30–60%). It is assumed that the
Lewis acid activates the carbon-chlorine bond to form an N-acyliminium intermediate, which
is prone to nucleophilic attack from the electron-rich arene 6. We reasoned that the same N-
acyliminium intermediate will be present in any conversion of 5 to a more reactive N,O-acetal
derivative. Hence, a one-pot activation and Friedel–Crafts alkylation reaction should work,
starting from cyclic ether 5 or any similar N,O-acetal (Supporting Information, Table S1). We
found that cyclic ether 5 could be activated in situ with thionyl chloride and a stoichiometric
amount of a Lewis acid (e.g. zinc chloride) to furnish the clip molecule in reasonable yield
after a single trituration of the crude product as the sole purification step. The reaction also
worked with a catalytic amount of Lewis acid (e.g. tin(II) trifluoromethanesulfonate), again
providing clip molecule 7 in reasonable yield. The same conditions were used in multigram
scale experiments (up to 21 g) without significant yield losses. Now, tosyl clip 7 can be
obtained in fewer steps and with an increased overall yield (31%) as compared to earlier work
(11%)10,11
.
The porphyrin roof (8) was synthesized in two steps from 2-methoxybenzaldehyde and
pyrrole in a reproducible yield of 6-7%, which could not be improved further. The methyl
ether was chosen as a protecting group to simplify the workup during the porphyrin synthesis.
The protecting group was easily removed in quantitative yield via a reaction with boron
tribromide.
Finally, a highly diluted 1:1 mixture of tosyl clip 7 and porphyrin roof 8 was refluxed in
acetonitrile, with potassium carbonate as a base, to yield free base porphyrin cage compound
H21 in acceptable yield (typically 25–58%) after purification by a single chromatographic
event. The yield of the cage synthesis was highly dependent on the duration of the reaction.
As described previously10,11, a yield of 25–30% was obtained after just 1 day, while a yield of
up to an impressive 58% can be reached reproducibly by refluxing the mixture for
approximately 10 days, as we show here. As such, up to 450 mg of porphyrin cage compound
4

ACCEPTED MANUSCRIPT
H21 could be obtained per batch. The physical properties of this compound were in line with
our previous reports. An X-ray structure of H21 could be determined after crystallization of
this compound from CDCl3 at −60 °C (Figure 2). This chloroform solvate is a pseudo-
polymorph of an X-ray structure of the same cage compound, reported by us before, which
was a chloroform-acetonitrile solvate20. In the current structure, H21 contains two chloroform
molecules in the asymmetric unit, both disordered over two positions (Figure 2b).
Interestingly, one of these solvent molecules is hosted by the cavity of H21.
Figure 2. X-ray structure of compound H21 (a) excluding and (b) including co-crystallized
chloroform solvent molecules.
Functional handles can be added selectively to the xylylene side walls of H21 in the form of
either one nitro-group, creating the chiral mono-nitrated cage H22, or two nitro-groups,
leading to the chiral anti-di-nitro cage H23 and the achiral syn-di-nitro cage H24 (Scheme 2).
This is achieved by reacting the cage with fuming nitric acid at low temperature, as we
reported earlier12.
5

ACCEPTED MANUSCRIPT
Scheme 2. Synthesis of cage compounds H22, H23 and H24.
We noticed that the formation of either the mono- or the di-nitro products is very sensitive
to changes in the concentration of the reactants and the temperature. When the temperature of
the reaction mixture in chloroform is strictly kept between −50 and −40 °C, the major product
is the mono-nitro-compound H22, which can be isolated in a yield of up to 75% (but typically
50–60%). The di-nitro products H23 and H24 already quickly form when the temperature is
slightly increased. In an attempt to elucidate the kinetic factors of the nitration reaction, we
monitored its progress over time with the help of 1H NMR spectroscopy (Figure 3). We used a
ten times more dilute concentration of H21 (0.743 mM) than is normally used during the
reaction, because in that case it is possible to perform the NMR experiments at a milder
temperature (0 °C). This temperature grants us a longer timespan to perform the
measurements before completion of the reaction, and although the yield of H22 is somewhat
lower, the reaction is easier to handle and to reproduce. Over time, we observed the
consumption of H21 and the formation of H22 and H24; the formation of H23 was very slow
and was therefore not considered in the kinetics plot. All changes in concentration are fitted
6

ACCEPTED MANUSCRIPT
using first order kinetics. Therefore, it appears that the formation of H22 follows a first order
kinetic process. However, the conversion rate could not be accurately determined since it
became clear that the concentration of H21 was decreasing with a much faster rate than H22
and H24 were formed. Hence, it became clear that the mass balance did not remain constant,
indicating decomposition. The longer the reaction ran, the more decomposition occurred. TLC
analysis also confirmed the formation of degradation products. It was unclear whether either
the starting material or the nitrated products decomposed, or both of these. However, the
apparent decomposition implied that a short reaction time would be optimal, which could
indeed be achieved by using a higher concentration of H21 in combination with lower
temperatures.
Figure 3. Progress of the nitration of H21 (initial concentration 0.743 mM) as a function of
time. T = 0 °C in CDCl3, concentrations are based on integration of relevant signals in the
1H NMR spectra. For the calculation of the concentrations, dibromomethane was used as
internal standard.
Because of the observed sensitivity of the reaction to the nitration conditions, we attempted
to introduce the nitro-groups in an earlier stage. When we treated clip molecule 7 with fuming
nitric acid under the same conditions, the material decomposed and no nitrated product could
be isolated or starting material recovered. This might indicate that the cage plays a role in
facilitating the nitration reaction, as we reported previously12. Another possibility was to
nitrate compound 6, but this reaction led to nitration at the meta-position of the
7

ACCEPTED MANUSCRIPT
dialkoxybenzene ring, which would make it impossible to obtain the nitrated version of 7 in
the subsequent step. Alternatively, H21 was subjected to a mild nitration method with
ammonium nitrate and trifluoroacetic acid anhydride in chloroform at room temperature, to
investigate if a higher selectivity for the mono-nitrated product and a lower amount of
decomposition could be achieved. Instead, the previously observed regioselectivity for the
sidewalls, was lost, and an inseparable mixture of many different nitrated products was
generated. The observation that the nitration occurred regioselectively with nitric acid and
non-selectively with ammonium nitrate could indicate that the cationic nature of the nitronium
ion plays an important role, since this cation is not formed under the mild nitration conditions
of the latter reagent.
To further investigate whether the observed sidewall selectivity was caused by the inclusion
of the nitronium cation in the cavity of the cage compound during the nitration reaction,
1
several H NMR titrations were performed. When a solution of cage molecule H21 was
titrated with HNO3, the NMR signals corresponding to the β-pyrrole protons (protons a in
Figure 4) and those belonging to the meso-phenyl rings (b) attached to the porphyrin roof
showed broadening. This is indicative of proton-exchange after the porphyrin is mono-
protonated. These signals became well-defined and sharp again after the addition of ~2
equivalents of fuming nitric acid, which suggests bis-protonation of the porphyrin roof
(Figure 4). Furthermore, the pyrrole-NH signal (g, −2.70 ppm) was no longer observed after
the addition of HNO3, due to broadening of the signal as a result of exchange with water.
8

ACCEPTED MANUSCRIPT
1
1
Figure 4. H NMR titration of H21 with HNO3. Each H NMR spectrum was recorded
after the addition of the indicated amount of HNO3. The signals are assigned to different
parts of the cage compound, see structure on the right. Solvent CDCl3.
After the addition of 3.0 equivalents of HNO3, the formation of H22 could be observed
through the upcoming three singlets of the xylylene sidewall protons (d). This indicates that 2
equivalents of nitric acid are needed to protonate the porphyrin roof prior to nitration. When
less than 2 equivalents of nitric acid were added, the 14N NMR spectrum of the mixture only
showed the signal of nitrate anions, which suggests that HNO3 protonates the porphyrin roof,
−
leaving NO3 behind. After the addition of 2 equivalents, the signal corresponding to nitric
acid appeared in the 14N NMR spectrum as well (Supporting Information, Figure S1). If a
nitronium cation had been bound inside the cavity, 14N NMR should have shown an additional
signal belonging to the shielded cationic nitronium guest. As expected, 14N signals that would
14
arise from the porphyrin cages are not observed, since N signals in larger molecules and in
an asymmetric environment are broadened because of quadrupolar interactions. In another 1H
NMR titration, nitronium tetrafluoroborate was used as a potential nitronium source (Figure
5).
9

ACCEPTED MANUSCRIPT
1
1
Figure 5. H NMR titration of H21 (in CDCl3) with [NO2][BF4] (in CD3CN). Each H
NMR spectrum was recorded after the addition of the indicated amount of [NO2][BF4]. The
signals are assigned to different parts of the cage compound, see structure on the right.
Similar to the titration with HNO3, broadening of the signals corresponding to the
porphyrin roof (a, b, g) occurred after adding the salt. After the addition of ~1 equivalent of
[NO2][BF4], a well-defined spectrum was obtained again, indicating either bis-protonation of
the porphyrin, or the presence of a 1:1 host-guest complex of the porphyrin cage with either a
nitronium or a nitrate ion. After the addition of 10 equivalents of [NO2][BF4], nitration of the
sidewalls could be observed through the three growing singlets at between 6.50 and 6.25 ppm.
These peaks continued to grow upon the addition of more of the nitronium salt. In the
presence of less than 1 equivalent of the salt, the 14N NMR spectrum again revealed the
presence of nitrate as the only symmetric nitrogen-containing species. Upon the addition of
more of the salt, the signal corresponding to nitric acid appeared in the 14N NMR spectrum as
well (Supporting Information, Figure S2). Combined with trace amounts of water, each
nitronium tetrafluoroborate ion pair can act as a diprotic acid. To further support these
1
findings, the H NMR spectrum of H21 in the presence of 1.0 equiv. of [NO2][BF4] was
compared to a spectrum of H21 in the presence of TFA (5.0 equiv, to ensure bis-protonation
of the porphyrin) (Figure 6). It is obvious that the two spectra are basically the same, apart
10

ACCEPTED MANUSCRIPT
from small deviations that are probably caused by the different anions. Hence, it is very likely
that protonation of the porphyrin is the initial event during the nitration reaction, which is in
line with the observed color change from purple to green when the fuming HNO3 is added to
the solution of H21.
1
1
Figure 6. (Top) H NMR spectrum of H21. (Middle) H NMR spectrum of H21 in the
1
presence of 1 equivalent of [NO2][BF4]. (Bottom) H NMR spectrum of H21 in the
presence of 5 equivalents of TFA. Solvent CDCl3-CD3CN (1:1, v/v).
In a final attempt to probe the presence of a nitronium cation in the cavity of the cage
molecule during the nitration reaction, a nickel(II) ion was inserted in the porphyrin ring to
afford a stable 16 electron square planar diamagnetic complex Ni1, which also resists
1
protonation of the porphyrin (Figure 7, top H NMR spectrum). When a solution of Ni1 in
chloroform-acetonitrile (1:1, v/v) was treated with 1.0 equiv. of [NO2][BF4], a new symmetric
1
1
compound was formed (Figure 7, bottom H NMR spectrum). Its H NMR spectrum showed
the absence (or extreme broadening) of all the β-pyrrole resonances of the porphyrin (a).
Moreover, the signals of the meso-phenyl substituents (b) were significantly broadened as
compared to those in the spectrum of Ni1. The signals arising from the diphenylglycoluril clip
part (c, e) remained basically unaffected. We propose that in the case of Ni1 the nitronium
cation serves as a one-electron oxidant, which oxidizes the porphyrin to a radical cation which
11

ACCEPTED MANUSCRIPT
is delocalized over the entire porphyrin ring system. Hence, the porphyrin resonances in the
1H NMR spectrum become extremely broadened, and protons in the vicinity of the porphyrin
experience a similar but smaller effect. These paramagnetic effects are highly dependent on
the distance to the paramagnetic center, and as a result proton signals further away from it are
less affected.
1
1
Figure 7. (Top) H NMR spectrum of Ni1. (Bottom) H NMR spectrum of Ni1 in the
presence of 1 equivalent of [NO2][BF4]. Solvent CDCl3-CD3CN (1:1, v/v).
Alternative to the formation of a radical cation, the nickel(II) center can be oxidized to a
nickel(III) species, which also implies the formation of a paramagnetic entity. One-electron
oxidations of nickel(II) porphyrins to either the radical cation or the nickel(III) species have
been reported before21–23, but not in combination with nitronium tetrafluoroborate as the
oxidant. As an additional experiment, we also performed the nitration reaction under the
standard conditions on Ni1 to investigate the outcome of the regioselectivity. Interestingly, all
1
selectivity for nitration on the xylylene sidewalls appeared to be lost. H NMR (Supporting
Information, Figure S3) and MALDI-TOF (Supporting Information, Figure S4) analysis
showed the formation of a mixture of many different nitrated species. This confirms our
findings that protonation of the porphyrin ring plays an essential role in the regioselectivity of
12

ACCEPTED MANUSCRIPT
the nitration reaction on H21. From the above experiments we cannot completely exclude that
the nitronium ion is first bound in the cavity of H21 before it nitrates the xylylene sidewalls of
this compound in a regioselective fashion. What is evident is that the first step of the reaction
is the protonation of the porphyrin ring, which protects it from being oxidized or nitrated.
These claims are confirmed by the loss of regioselectivity that takes place during nitration
after nickel is inserted in the porphyrin ring.
3. Conclusions
We have optimized the synthesis of porphyrin cage compound H21 by reducing the number
of steps by two, and alongside increased the yields of the other steps. As such, the overall
yield of the bottom part of the cage compound has been increased from 11 to 31%. Moreover,
the yield of the final step, in which the bottom part is connected to the porphyrin, could be
increased to 58%. We have gained more insight into the possible mechanisms involving the
regioselective nitration of cage compound H21. The high regioselectivity is most probably not
caused by the location of the nitronium cation in the cavity of the cage molecule, but
controlled by an initial double protonation of the porphyrin ring, which deactivates it for
electrophilic aromatic substitution.
4. Experimental
4.1. General Information
Catechol was recrystallized successively from toluene and chloroform and stored in the
freezer under an argon atmosphere. Pyrrole was either filtered through a pad of Alumina I just
before use or it was distilled under reduced pressure and stored in the freezer under an argon
atmosphere. Dichloromethane was distilled from calcium hydride under a nitrogen
atmosphere. Other solvents and reagents were obtained from commercial suppliers and used
without further purification. Reactions were followed using thin-layer chromatography (TLC)
on silica gel-coated plates (Merck 60 F254). Detection was performed with UV light at 254
nm and/or by charring at 150 °C after dipping in a solution of KMnO4. Column
chromatography was performed manually using Acros silica gel, 0.035−0.070 mm, 60A, and
Acros aluminium oxide, 0.050–0.200 mm, 60A. NMR spectra were recorded at 298 K on on
an Agilent Inova 400 spectrometer (400 MHz) equipped with a dual-channel inverse probe,
on a Bruker Avance III 500 spectrometer (500 MHz) equipped with a Prodigy BB cryoprobe
1
and on a Bruker Avance III 400 spectrometer (400 MHz) equipped with a BBFO probe. H
13

ACCEPTED MANUSCRIPT
NMR chemical shifts (δ) are given in parts per million (ppm) and were referenced to
tetramethylsilane (0.00 ppm). Coupling constants are reported as J values in Hertz (Hz). Data
1
for H NMR spectra are reported as follows: chemical shift (multiplicity, coupling constant,
integration). Multiplicities are abbreviated as s (singlet), d (doublet), t (triplet), m (multiplet),
b (broad). Mass spectra were recorded on a Thermo Finnigan LCQ Advantage Max mass
spectrometer (MS), an JEOL AccuTOF CS JMS-T100CS mass spectrometer or on a Bruker
Microflex LRF MALDI-TOF system in reflective mode employing dithranol as a matrix.
4.2. Typical synthesis procedures
4.2.1.
1,6:3,4-Bis(2-oxapropylene)tetrahydro-3a,6a-diphenylimi-dazo[4,5-d]imidazole-
2,5(1H,3H)-dione (5)
TFA (24 mL, 0.31 mol, 1.25 equiv) was added to a solution of urea (31 g, 0.51 mol, 2.05
equiv) and benzil (53 g, 0.25 mol, 1.0 equiv) in toluene (500 mL). The resulting mixture was
refluxed for 26 hours in a Dean–Stark apparatus under an argon atmosphere. Upon
completion of the reaction, indicated by the disappearance of the yellow color, the mixture
was allowed to cool to 20 °C. The resulting precipitate was successively filtered off, washed
with ethanol (100 mL) and dried under high vacuum to afford diphenylglycoluril (74 g, 99%)
as a white solid. Paraformaldehyde (23 g, 0.78 mol, 5.0 equiv) was added to a suspension of
diphenylglycoluril (46 g, 0.16 mol, 1.0 equiv) in DMSO (250 mL). The pH of the reaction
mixture was adjusted to 10 by the dropwise addition of aqueous 1M NaOH. The resulting pale
yellow solution was stirred at 20 °C for 16 hours. Then, the pH of the reaction mixture was
adjusted to 1 by the addition of aqueous 37% HCl. The resulting mixture was refluxed for 2
hours. The suspension was allowed to cool to 20 °C and water (100 mL) was carefully added
while stirring. The resulting precipitate was successively filtered off, washed with water (100
mL) and cold ethanol (100 mL), and dried under high vacuum to afford cyclic ether 5 (37 g,
1
61%) as a white solid. H NMR (400 MHz, CDCl3) δ 7.20–7.10 (m, 10H), 5.66 (d, J = 11.2
Hz, 4H), 4.57 (d, J = 11.2 Hz, 4H). Spectral data were in agreement with literature values24.
4.2.2. 1,2-Di-(4-methylbenzene-sulfonoxy-ethoxy)benzene (6)
A solution of KOH (33 g, 0.59 mol, 2.6 equiv) in water (800 mL) was purged with argon
for 60 minutes. Then, catechol (25 g, 0.23 mol, 1.0 equiv) was added and the reaction mixture
was heated to 60 °C. 2-Chloroethanol (50 mL, 0.75 mol, 3.3 equiv) was added dropwise over
15 minutes and the solution was stirred at 60 °C for 99 hours under an argon atmosphere.
Additional 2-chloroethanol was added after 24 hours (50 mL, 0.75 mol, 3.3 equiv), 48 hours
14

ACCEPTED MANUSCRIPT
(50 mL, 0.75 mol, 3.3 equiv), and 72 hours (25 mL, 0.38 mol, 1.7 equiv). Meanwhile, KOH
pellets were added to keep the pH of the reaction mixture between 9 and 11. Upon
completion, the mixture was cooled to 20 °C and the product was extracted with DCM (4 ×
400 mL). The combined organic extracts were dried over sodium sulfate and the solvent was
removed in vacuo. The residue was dissolved in THF (150 mL) at 40 °C and precipitated by
the addition of n-heptane (50 mL). The precipitate was successively filtered off, washed with
n-pentane (200 mL) and dried under high vacuum to afford the diol intermediate (44 g, 98%)
as a white solid. Pyridine (110 mL, 1.3 mol, 6.0 equiv) was added to a 0 °C solution of the
diol intermediate (44 g, 0.22 mol, 1.0 equiv) in dry DCM (500 mL). Then, TsCl (130 g, 0.67
mol, 3.0 equiv) was added portion-wise over 15 minutes, the ice bath was removed and the
resulting mixture was stirred at 20 °C for 23 hours. The mixture was then diluted with DCM
(500 mL). The organic layer was successively washed with water (800 mL), aqueous 6M HCl
(600 mL), water (600 mL), and brine (600 mL); then dried over sodium sulfate, after which
the solvent was removed in vacuo. The residue was recrystallized from toluene (300 mL) to
afford ditosylate 6 (84 g, 78%) as a white solid. 1H NMR (400 MHz, CDCl3) δ 7.83–7.78 (m,
4H), 7.36–7.30 (m, 4H), 6.93–6.87 (m, 2H), 6.84–6.78 (m, 2H), 4.35–4.29 (m, 4H), 4.19–4.13
(m, 4H), 2.44 (s, 6H). Spectral data were in agreement with literature values10.
4.2.3. Tetratosyl clip (7)
Method A: Thionyl chloride (7.7 mL, 100 mmol, 20 equiv) was added to a suspension of
cyclic ether 5 (1.9 g, 5.0 mmol, 1.0 equiv), ditosylate 6 (6.3 g, 2.5 mmol, 2.5 equiv) and zinc
chloride (6.1 g, 45 mmol, 9.0 equiv) in dry DCM (50 mL). The resulting dark blue mixture
was stirred at 20 °C for 24 hours under an argon atmosphere. The mixture was diluted with
DCM (100 mL) and washed with aqueous 1M HCl (2 × 100 mL) and brine (100 mL). The
yellow-orange organic layer was dried over sodium sulfate and the solvent was removed in
vacuo. The residue was triturated with acetone (40 mL) and kept at 4 °C for 30 minutes. The
precipitate was successively filtered off, washed with acetone (40 mL) and dried under high
vacuum to afford clip molecule 7 (3.3 g, 50%) as a white solid. Method B: Thionyl chloride
(42 mL, 0.58 mol, 15 equiv) was added to a suspension of cyclic ether 5 (14.9 g, 39.5 mmol,
1.0 equiv), ditosylate 6 (43.1 g, 85.1 mmol, 2.1 equiv) and tin(II) trifluoromethanesulfonate
(3.36 g, 8.1 mmol, 0.2 equiv) in dry DCM (500 mL). The resulting dark green mixture was
stirred at 20 °C for 120 hours under an argon atmosphere. Upon completion, the mixture was
washed with water (2 × 500 mL) and brine (500 mL). The yellow-orange organic layer was
dried over sodium sulfate and the solvent was removed in vacuo. The residue was triturated
15

ACCEPTED MANUSCRIPT
with acetone (400 mL) and kept at 4 °C for 30 minutes. The precipitate was successively
filtered off, washed with acetone (200 mL) and dried under high vacuum to afford clip
molecule 7 (20.7 g, 39%) as a white solid. 1H NMR (400 MHz, CDCl3) δ 7.72–7.66 (m, 8H),
7.21–7.15 (m, 8H), 7.15–7.06 (m, 10H), 6.67 (s, 4H), 4.65 (d, J = 15.9 Hz, 4H), 4.21 (t, J =
4.7 Hz, 8H), 4.09 (d, J = 15.8 Hz, 4H), 3.96 (dt, J = 11.5, 4.5 Hz, 4H), 3.91 (dt, J = 11.4, 4.8
Hz, 4H), 2.38 (s, 12H). Spectral data were in agreement with literature values10.
4.2.4. 5,10,15,20-Tetrakis(2-hydroxyphenyl)porphyrin (8)
A mixture of 2-methoxybenzaldehyde (14 g, 0.10 mol, 4.0 equiv) and zinc acetate
dihydrate (5.5 g, 25 mmol, 1.0 equiv) in propionic acid (500 mL) was heated to reflux. Then,
pyrrole (6.9 mL, 0.10 mol, 4.0 equiv) was added dropwise over 10 minutes and refluxing was
continued for 2 hours. The resulting black mixture was slowly cooled to 20 °C. The resulting
precipitate was successively filtered off, washed with propionic acid (100 mL) and methanol
(4 × 50 mL), and dried under high vacuum to afford a dark purple solid. The solid material
was dissolved in chloroform (100 mL), DDQ (2.7 g, 13 mmol, 0.5 equiv) was added and the
resulting mixture was refluxed for 2 hours. Upon cooling, the mixture was purified by
Alumina III column chromatography (eluent CHCl3). The purified material was dissolved in
chloroform (100 mL) and successively washed with aqueous 37% HCl (100 mL), saturated
aqueous NaHCO3 (250 mL); then dried over sodium sulfate, after which the solvent was
removed in vacuo. The residue was dissolved in minimal amount of DCM and precipitated by
the addition of heptane. Most DCM was removed under reduced pressure and the resulting
suspension was centrifuged. The supernatant was removed and the precipitate was washed
with pentane and dried under high vacuum to afford H2TMPP (1.24 g, 7%) as a purple solid.
Boron tribromide (3.2 mL, 34 mmol, 20 equiv) was added dropwise over 1 minute to a −25
°C solution of H2TMPP (1.24 g, 1.7 mmol, 1.0 equiv) in dry DCM (25 mL) under an argon
atmosphere. The mixture was slowly warmed to 20 °C and stirred for 16 hours. Upon
completion, the mixture was poured onto ice water (300 mL). Then, ethyl acetate (250 mL)
was added and the biphasic mixture was stirred vigorously. Solid NaHCO3 was added
carefully until the organic phase turned purple. The organic layer was separated and
successively washed with saturated aqueous NaHCO3 (300 mL), dried over sodium sulfate,
after which the solvent was removed in vacuo. The residue was purified by silica gel column
chromatography (eluent DCM/MeOH, 19:1, v/v). The purified material was dissolved in
minimal amount of DCM and the product precipitated by the addition of heptane. Most DCM
was removed under reduced pressure and the resulting suspension was centrifuged. The
16

ACCEPTED MANUSCRIPT
supernatant was removed and the precipitate was washed with pentane and dried under high
vacuum to afford porphyrin 8 (1.16 g, 99%) as a purple solid. 1H NMR (400 MHz, CDCl3) δ
8.91 (s, 8H), 8.00–7.92 (m, 4H), 7.77–7.67 (m, 4H), 7.38–7.28 (m, 8H), 4.92 (bs, 4H), −2.79
(bs, 2H). Spectral data were in agreement with literature values11.
4.2.5. Porphyrin cage compound (H21)
Potassium carbonate (0.88 g, 6.4 mmol, 11 equiv) was added to a solution of tetratosyl clip
7 (0.78 g, 0.57 mmol, 1.0 equiv) and porphyrin 8 (0.43 g, 0.64 mmol, 1.1 equiv) in
acetonitrile (700 mL) and the resulting mixture was refluxed for 10 days. After cooling, the
mixture was filtered through a pad of Celite and the residue was washed with chloroform until
colorless washings were obtained. The combined filtrate was evaporated to dryness and then
purified by Alumina III column chromatography (eluent CHCl3). The purified material was
dissolved in a minimal amount of DCM and precipitated by the addition of n-heptane. Most
DCM was removed under reduced pressure and the resulting suspension was centrifuged. The
supernatant was removed and the precipitate was washed with n-pentane and dried under high
1
vacuum to afford porphyrin cage compound H21 (0.45 g, 58%) as a purple solid. H NMR
(400 MHz, CDCl3) δ 8.76 (s, 4H), 8.67 (s, 4H), 8.07 (d, J = 7.3 Hz, 4H), 7.76 (t, J = 8.0 Hz,
4H), 7.37 (t, J = 7.4 Hz, 4H), 7.34 (d, J = 8.4 Hz, 4H), 6.97–6.91 (m, 6H), 6.83–6.78 (m, 4H),
6.20 (s, 4H), 4.30–4.20 (m, 4H), 4.24 (d, J = 15.8 Hz, 4H), 4.09–4.01 (m, 4H), 3.74 (d, J =
15.8 Hz, 4H), 3.53–3.43 (m, 4H), 3.37–3.30 (m, 4H), −2.73 (bs, 2H). Spectral data were in
agreement with literature values10. A single crystal of this compound, suitable for X-ray
analysis, was grown from a CDCl3 solution at −60 °C. Crystallographic data have been
deposited at the Cambridge Crystallographic Data Centre under number CCDC 1913851.
4.2.6. Nickel(II) porphyrin cage compound (Ni1)
A solution of H21 (10 mg, 7.4 ꢀmol, 1.0 equiv) and nickel(II) acetate tetrahydrate (37 mg,
0.15 mmol, 20 equiv) in DMF (2 mL) was refluxed for 20 hours. Upon completion of the
reaction (confirmed by MALDI-TOF analysis) and after cooling, the mixture was poured into
water (40 mL). The formed precipitate was collected by centrifugation and was successively
washed with water (30 mL) and methanol (30 mL), and dried under high vacuum. The residue
was purified by silica gel column chromatography (eluent CHCl3/MeCN, 9:1, v/v). The
purified material was dissolved in a minimal amount of DCM and precipitated by the addition
of n-heptane. Most DCM was removed under reduced pressure and the resulting suspension
was centrifuged. The supernatant was removed and the precipitate was washed with n-pentane
17

ACCEPTED MANUSCRIPT
and dried under high vacuum to afford Ni1 (8.6 mg, 83%) as a dark red solid. 1H NMR (500
MHz, CDCl3) δ 8.69 (s, 4H), 8.56 (s, 4H), 7.96 (dd, J = 7.3, 1.8 Hz, 4H), 7.73–7.65 (m, 4H),
7.32 (td, J = 7.5, 1.1 Hz, 4H), 7.26 (d, J = 8.0 Hz, 4H), 7.01–6.92 (m, 6H), 6.86–6.79 (m, 4H),
6.26 (s, 4H), 4.30 (d, J = 15.9 Hz, 4H), 4.24 (ddd, J = 11.4, 7.8, 4.0 Hz, 4H), 4.02 (dt, J =
10.6, 4.2 Hz, 4H), 3.80 (d, J = 15.8 Hz, 4H), 3.52 (dt, J = 8.6, 4.5 Hz, 4H), 3.44 (ddd, J =
10.8, 7.8, 3.9 Hz, 4H). 13C NMR (126 MHz, CDCl3) δ 158.44, 157.16, 146.65, 143.67,
143.35, 134.81, 133.44, 131.34, 131.07, 130.58, 129.94, 129.54, 128.56, 128.48, 128.09,
119.78, 115.61, 115.05, 111.72, 84.89, 67.23, 66.50, 44.53. M.p. >300 °C (decomposition).
HRMS (ESI) calcd. for [C84H62N8NiO10 + Na]+ 1423.38401, found 1423.38381.
Acknowledgements
R.J.M.N. acknowledges support from the European Research Council (ERC Advanced Grant
ENCOPOL-74092) and from the Dutch National Science Organization NWO (Gravitation
program 024.001.035).
Appendix A. Supplementary data
Supplementary data to this article can be found online at …
References
1. Trends and Analysis Cisco Company. White Pap. June 2017.
2. J. Gantz, D. Reinzel, in Proceedings of IDC iView, IDC Anal. Future 2012.
3. A. Extance, Nature 2016, 537, 22–24.
4. P.Y. De Silva, G.U. Ganegoda, BioMed. Res. Int. 2016, 2016, 14.
5. M. Castillo, Am. J. Neuroradiol. 2014, 35, 1–2.
6. J.-F. Lutz, Macromolecules 2015, 48, 4759–4767.
7. G. Cavallo, A. Al Ouahabi, L. Oswald, L. Charles, J.-F. Lutz, J. Am. Chem. Soc. 2016,
138, 9417–9420.
8. S. Varghese, J.A.A.W. Elemans, A.E. Rowan, R.J.M. Nolte, Chem. Sci. 2015, 6, 6050–
6058.
18

ACCEPTED MANUSCRIPT
9. M.G.T.A. Rutten, F.W. Vaandrager, J.A.A.W. Elemans, R.J.M. Nolte, Nat. Rev. Chem.
2018, 2, 365–381.
10. J.A.A.W. Elemans, M.B. Claase, P.P.M. Aarts, A.E. Rowan, A.P.H.J. Schenning, R.J.M.
Nolte, J. Org. Chem. 1999, 64, 7009–7016.
11. J.A.A.W. Elemans, E.J.A. Bijsterveld, A.E. Rowan, R.J.M. Nolte, Eur. J. Org. Chem.
2007, 751–757.
12. S. Varghese, B. Spierenburg, A. Swartjes, P.B. White, P. Tinnemans, J.A.A.W. Elemans,
R.J.M. Nolte, Org. Lett. 2018, 20, 3719–3722.
13. S. Varghese, B. Spierenburg, J.P.J. Bruekers, A. Swartjes, P.B. White, J.A.A.W. Elemans,
R.J.M. Nolte, Eur. J. Org. Chem. 2019, 3525–3533.
14. S. Ostrowski, D. Szerszén, M. Ryszczuk, Synthesis 2005, 819–823.
15. S. Ostrowski, S. Grzyb, Tetrahedron Lett. 2012, 53, 6355–6357.
16. F. Ji, L. Zhu, X. Ma, Q. Wang, H. Tian, Tetrahedron Lett. 2009, 50, 597–600.
17. W.J. Kruper, T.A. Chamberlin, M. Kochanny, J. Org. Chem. 1989, 54, 2753–2756.
18. M. M. Catalano, M. J. Crossley, M. M. Harding, L. G. King, J. Chem. Soc., Chem.
Commun. 1984, 1535–1536.
19. J. Baldwin, M. J. Crossley, J. DeBernardis, Tetrahedron 1982, 38, 685–692.
20. P. Hidalgo Ramos, R. Coumans, A. B. C. Deutman, R. de Gelder, J. M. M. Smits, J. A. A.
W. Elemans, R. J. M. Nolte, A. E. Rowan, J. Am. Chem. Soc., 2007, 129, 5699–5702.
21. D. Chang, T. Malinski, A. Ulman, K.M. Kadish, Inorg. Chem. 1984, 23, 817–824.
22. G.S. Nahor, P. Neta, P. Hambright, L.R. Robinson, J. Phys. Chem. 1991, 95, 4415–4418.
23. J. Seth, V. Palaniappan, D.F. Bocian, Inorg. Chem. 1995, 34, 2201–2206.
24. F.G.M. Niele, R.J.M. Nolte, J. Am. Chem. Soc. 1988, 110, 172–177.
19

ACCEPTED MANUSCRIPT
Highlights:
•
Porphyrin cage compounds based on diphenylglycoluril can be synthesized in 8 steps
in batches up to 450 mg
•
•
Porphyrin cage compounds can be selectively nitrated at their cavity side-walls
Protonation of the porphyrin plane causes nitration by electrophilic nitronium ions to
be directed to the cavity side-walls