Synthesis of macrocyclic chromotropic acid-based sulfonamides; Their complexation properties and an unexpected photochemical reaction

Article abstract of DOI:10.1016/j.tetlet.2013.04.081

Chromotropic acid (CTA), a dihydroxynaphthalenedisulfonic acid is a reagent that is widely used for dyestuff manufacture, as a chromogen-forming reagent for the analysis of formaldehyde and as an intermediate for azo derivatives used as analytical reagents. The synthesis of two new CTA-based macrocyclic tetra-sulfonamides 7 and 7a and their corresponding bis-sulfonamides 10 and 10a and their complexation properties toward tetra-n-butylammonium halides were studied. The X-ray structure of 10a was determined and it provided key evidence for an unexpected photochemical reaction that the tetra-sulfonamides underwent.

Full text of DOI:10.1016/j.tetlet.2013.04.081

Tetrahedron Letters 54 (2013) 3444–3448
Contents lists available at SciVerse ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Synthesis of macrocyclic chromotropic acid-based sulfonamides; their
complexation properties and an unexpected photochemical reaction
⇑
Hisham Fathy Sleem, Louise Nicole Dawe, Paris Elias Georghiou
Department of Chemistry, Memorial University of Newfoundland, St. John’s, Newfoundland and Labrador, Canada A1B 3X7
a r t i c l e i n f o
a b s t r a c t
Article history:
Chromotropic acid (CTA), a dihydroxynaphthalenedisulfonic acid is a reagent that is widely used for dye-
stuff manufacture, as a chromogen-forming reagent for the analysis of formaldehyde and as an interme-
diate for azo derivatives used as analytical reagents. The synthesis of two new CTA-based macrocyclic
tetra-sulfonamides 7 and 7a and their corresponding bis-sulfonamides 10 and 10a and their complexa-
tion properties toward tetra-n-butylammonium halides were studied. The X-ray structure of 10a was
determined and it provided key evidence for an unexpected photochemical reaction that the tetra-sulfon-
amides underwent.
Received 10 April 2013
Revised 16 April 2013
Accepted 19 April 2013
Available online 27 April 2013
Keywords:
Chromotropic acid
Macrocycles
Ó 2013 Elsevier Ltd. All rights reserved.
Sulfonamides
Host–guest complexation
Tetrabutylammonium halides
We have been interested in the chemistry of chromotropic
acid 1 (4,5-dihydroxy-2,7-naphthalenedisulphonic acid, or ‘CTA’)
for some time now.1 CTA has been used as a key reagent for
the syntheses of various azo derivatives, which can be used not
only in the dye industry, but also in a variety of analytical chem-
istry applications.2 Eegriwe3 first described its use in 1937 as a
formaldehyde-specific reagent but it was only more recently in
1989 that the mechanism of the chromogen-forming reaction
was deciphered.1 In the same year, Poh et al. showed that the
reaction of 2, the disodium salt of CTA, with formaldehyde in
an aqueous solution, in the absence of concentrated sulfuric acid
produced a water-soluble cyclic tetramer which he called ‘cyclo-
tetrachromotropylene’ (3) (Scheme 1).4 This compound which can
be considered to be a calixarene analogue, could not, however, be
crystallized and was mainly characterized by its mass spectrum.
Nevertheless, Poh’s group published several papers5 which re-
vealed that 3 showed typical host-guest properties with various
guest species.
We were therefore interested in investigating the potential
host–guest properties of macrocyclic CTA-based sulfonamides
which could be formed using the diamino compounds, Jeffam-
inesÒ EDR-148 and EDR-176.6 The products formed from these
diamino compounds with isophthaloyl dichloride were previously
studied for their complexation properties with Groups 1 and 2
halides and also with tetrabutylammonium halides (TBAX, where
X = Cl, Br, I).7 As further impetus for our work, Lan et al.8 had re-
ported the synthesis (Scheme 2) of macrocyclic tetrasulfonamides
4 based upon 3,6-dihydroxy-2,7-naphthalene disulfonic acid, a
regioisomer of CTA, via the corresponding 3,6-dimethoxy-dis-
ulfonyldichloride (5). These authors reported that their com-
pounds were used as good hosts for small solvent molecules
9
´
such as DMF. Bochenska et al. reported some other macrocyclic
sulfonamides however which were ion-selective ionophores for
K+, Rb+, and Cs+.
In this Letter we report the synthesis of four new CTA-based
macrocyclic sulfonamides which were conducted in order to eval-
uate their potential host–guest properties. Some of these proper-
ties were evaluated in complexation studies with Groups 1 and
2, and tetrabutylammonium (TBA) halides. As well, the single-crys-
tal X-ray structure of one of these compounds was determined and
is reported. During the course of these studies, an unexpected pho-
tochemical reaction was observed which is also described herein.
The original macrocyclic sulfonamides which were targeted as
shown in Scheme 3, were 6 and 6a in which the naphthalene rings
bear free hydroxyl groups, as in CTA itself. The reaction of the dis-
ulfonyldichloride of CTA with 1,8-diamino-3,6-dioxaoctane (9,
‘JeffamineÒ EDR-148’)6 or with 1,10-diamino-4,7-dioxadecane
(9a, ‘JeffamineÒ EDR-176’) was envisioned as a potential route to
6 and 6a, respectively. Paruch et al.10 in 2000 showed that an effi-
cient way to functionalize CTA to its disulfonyldichloride 8, was to
first convert it, via its sodium salt 2, to the corresponding ditosy-
late which could then form 8 in good yields. Therefore, this ap-
proach was taken by us and after considerable experimentation,
conditions were found from which the syntheses of the tosylated
macrocycles 7 and 7a were achieved.
⇑
Corresponding author.
E-mail address: parisg@mun.ca (P.E. Georghiou).
0040-4039/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tetlet.2013.04.081

H. F. Sleem et al. / Tetrahedron Letters 54 (2013) 3444–3448
3445
NaO3S
SO3Na
SO3Na
NaO3S
OH
OH
OH
OH
HO
HO
OH
OH
One week
(HCHO)n
+
SO3R
RO3S
OH
OH
NaO3S
SO3Na
1:
R = H
2: R = Na+
NaO3S
SO3Na
3
Scheme 1. Poh’s synthesis of cyclotetrachromotropylene (3).4
H3CO
O2S
OCH3
SO2
HN
R
R
R
HN
H3CO
OCH3
SO2Cl
R
R
Et3N/CH2Cl2
12 h
+
ClO2S
H2N
NH2
NH
SO2
R
NH
O2S
5
H3CO
OCH3
R= n-C7H15O-; i-C5H11O-; or H
4
Scheme 2. Lan’s synthesis of macrocyclic tetrasulfonamides.8
In our initial experiments reacting 8 with 9 with triethylamine,
in acetonitrile under high dilution conditions, at either 0 °C or
À78 °C afforded only the [1+1] product 10 in 65% yield. However,
when dichloromethane was used instead as the solvent, at
À100 °C, the desired [2+2] product 7 was formed in 47% yields in
a mixture with 10 which was formed in 16% yields. Similarly, reac-
tion of 8 with 9a in acetonitrile only afforded the corresponding
[1+1] product 10a, in 76% yields. Using dichloromethane instead
as the solvent at À100 °C however, formed the desired [2+2] prod-
uct 7a in 45% yields in a mixture together with 10a in 13% yields.
Although Paruch et al.10 reported being able to de-tosylate their
CTA-derived product, all attempts to de-tosylate our compounds
only afforded intractable products which could not be character-
ized. Therefore since it was not possible to produce the free hydro-
complexation studies were conducted on all four of the tosylated
products instead.
Each of the macrocyles 7, 7a, 10, and 10a were tested using 1H
NMR titration experiments with Groups 1 and 2 chlorides. An anal-
ogous study with isophthaloyl macrocyclic amides was previously
reported by us.7 In all cases, in contrast to the findings with the
macrocyclic amides, no chemically-induced shifts (CIS) could be
observed when solutions of the CTA macrocycles in a 9:1 CDCl3;
DMSO-d6 solvent mixture were saturated with the metal salts.
TBAX halides were also tested with each of macrocyles 7, 7a, 10,
and 10a, again, in analogous experiments to those conducted with
the isophthaloyl macrocyclic amides7 described above. In these
tests, in CDCl3 solutions, no 1H NMR CIS were observed with 10
or 10a.
xy group-containing sulfonamides
6
or 6a, host–guest
OR
OR
H2N
O
O
NH2
+
n
n
X
X
9:
2: R = H; X = SO3Na
8: R = Ts; X = SO2Cl
n = 1 ("Jeffamine 148")
9a: n = 2 ("Jeffamine 176")
n
n
O
O
SO2NH
NHSO2
OTs OTs
RO
RO
OR
SO2
O2S
HN
OR
NH
SO2NH
NHSO2
O
O
O
O
n
n
n
n
6: R
n
6a: R
n
10:
10a:
= H;
= 1
= H; =2
n = 1
n = 2
7: R = Ts; n = 1 7a: R = Ts; n =2
Figure 1. 1:1 Binding plots for the titration of 7 (0.62 Â 10À3 M in CDCl3) with
Scheme 3. Synthetic routes to macrocyclic CTA-based bis- and tetra-sulfonamides.
TBACl using 1H NMR (300 MHz) CIS for the N–H protons).

3446
H. F. Sleem et al. / Tetrahedron Letters 54 (2013) 3444–3448
Figure 4. Expanded sections of 1H NMR (300 MHz) spectra of the NH signals of 7
and 10 from the stepwise timed photo-chemical reaction of 7.
be more significant is the difference in the ring sizes leading to
more conformational possibilities and a larger entropic effect
therefore which needs to be overcome in the case of 7a.
Figure 2. 1:1 Binding plots for the titration of 7 (0.62 Â 10À3 M in CDCl3) with
TBACl using 1H NMR (300 MHz) CIS for the intra-annular naphthyl protons.
ESI-mass spectral data were also determined qualitatively using
ESI-mass spectra in both positive and negative modes and a sum-
mary table is given in the Supplementary data. In the negative-ion
mode, for the TBACl:7 solutions a m/z signal at 1515.8 correspond-
ing to [7+ClÀ], and in the positive-ion mode, m/z signals at 1722
and 1781 corresponding to [7+TBA+ ] and [7+TBACl+Na+], respec-
tively were present. For the TBABr:7 solutions, in the negative-
ion mode, signals at m/z 1561.8 corresponding to [7+BrÀ], and in
the positive-ion mode signals at m/z 1722 and 1803 corresponding
to [7+TBA+] and [7+TBABr+H+], respectively were present. Using
the same conditions and parameters for the ESI-MS spectra used
in the previous cases for TBAI did not reveal signals in the posi-
tive-ion mode indicating any complexation. However, when the
solvent was changed from chloroform to methanol, and the dry
temperature was decreased on the chromatograph to 200 °C, the
ESI-MS spectra in the negative-ion mode showed a m/z signal at
1607.9 corresponding to [7+IÀ], and in the positive-ion mode, sig-
nals at m/z 1799 and 1849 corresponding to [7+TBA+3H2O+Na+]
and [7+TBAI+H+], respectively.
Figure 3. Expanded sections of 1H NMR (300 MHz) spectra of the NH signals of 7a
and 10a from the stepwise timed photo-chemical reaction of 7a.
For 7a the ESI-MS of the TBACl:7a solution in chloroform in the
negative-ion mode showed a mass signal at m/z 1573 correspond-
ing to [7a+ClÀ], and in the positive-ion mode mass signals at m/z
1779 and 1838 corresponding to [7a+TBA+] and [7a+TBACl+Na+],
respectively. For the TBABr:7a solution the ESI-MS spectra in the
negative-ion mode showed a mass signal at m/z 1617 correspond-
ing to [7a+BrÀ ], and in the positive-ion mode, mass signals at m/z
1779 and 1882 corresponding to [7a+TBA+] and [7a+TBABr+Na+],
respectively. But, as before, in the case of TBAI, using the same con-
ditions and parameters of the ESI-MS spectra used for the chloride
and bromide salts, no mass signals corresponding to any complex-
ation in the positive-ion mode could be detected. However, when
the solvent was changed from chloroform to methanol instead,
and the dry temperature to 200 °C on the instrument, as was done
with 7, the ESI-MS spectra in the negative-ion mode did reveal a
signal at m/z = 1663.6 corresponding to [7a+IÀ] and in the posi-
tive-ion mode, signals at m/z 1779, 1798, and 1929 corresponding
to [7a+TBA+], [7a+TBA++H2O], and [7a+TBAI+Na+], respectively.
All attempts to form suitable crystals for single-crystal X-ray
analysis of each of the four products obtained, or from the com-
plexation studies as their complexes, failed with the exception of
10a itself which crystallized by the slow evaporation at room tem-
perature of an aqueous methanol/ethyl acetate solution (see pho-
tochemistry section below however!). 10a crystallized in the
non-centrosymmetric orthorhombic space group P212121, with
intermolecular hydrogen bonding, leading to the formation of a
However, with 7 and 7a, significant complexation could be
determined: For
7 ;
average Kassoc values of 808 18 MÀ1
229 16 MÀ1, and 222 21 MÀ1, respectively, and for 7a average
Kassoc values of 252 8 MÀ1; 87 6 MÀ1, and 52 5 MÀ1, respec-
tively, were determined for the chloride, bromide, and iodo salts.
These values were determined using non-linear 1:1 binding iso-
therms11 for the NH and aromatic intra-annular proton singlet
CIS changes (see Figs. 1 and 2) which were averaged. The results
confirmed the stronger relative affinities for the chloride anion as
seen previously also with the isophthaloyl macrocyclic amides.
The other ‘non-complexing’ anions tested, namely the tetrafluoro-
borate and phosphorus hexafluoride salts showed negligible chem-
ical shift changes for which association constants could not be
determined.
The association constants for 7 and 7a are nearly 3-, 2.5-, and 4-
fold higher for the ClÀ, BrÀ and IÀ salts, respectively. There are sev-
eral possible factors which could account for these differences.
Firstly, the mode of complexation of the halide ion is presumed
to be via hydrogen-bonding between the halide ions and the pro-
tons of the NH groups. As found by others in similar studies, the
trend is: ClÀ > BrÀ > IÀ. Secondly, the tetrabutylammonium ion is
non-coordinating so it is presumed that the ether oxygen atoms
are not as significant in the case of the TBAX salts compared with
the metal halides. Since presumably the distances between the NH
groups in the two macrocycles are not different, a factor that could

H. F. Sleem et al. / Tetrahedron Letters 54 (2013) 3444–3448
3447
H2N
CHO
CHO
O
N
N
N
N
N
N
O
+
O
O
H2N
12
11
14
13
Scheme 4. Synthesis and equilibrium process of macrocycles 11 and 12 reported by Jurczak and co-workers.18
two-dimensional infinite polymer in the ab-plane (Figures S1–3
and Supplementary Table S1. Crystallographic data for 10a (CCDC
924367) has been deposited with the Cambridge Crystallographic
Data Centre).
O2
S
O2
S
O2
S
O2
S
HN
HN
NH2
NH2
HN
HN
H
N
7
or
7a
hν
When a chloroform solution of the [2+2] macrocyclic sulfon-
amide 7a was set aside to crystallize from various solvents or sol-
vent mixtures for three to four weeks in a closed vial on the bench-
top, it was noticed that the color of the solutions gradually changed
from colorless to light yellow. It was suspected that a photochem-
ical reaction might have occurred, as similar solutions of the same
compound which were protected from ambient light with alumi-
num foil did not change in the same way. By way of contrast, solu-
tions of the [1+1] macrocyclic sulfonamide 10a did not change in
the same way, and afforded crystals whose X-ray structure was de-
scribed above. When crystals did form from the methanol: chloro-
form solution of 7a, it was anticipated that these would
complement the X-ray structure for 10a. Surprisingly, the X-ray
structure of the putative 7a however turned out to be identical
to that obtained with 10a itself. This observation indicated that a
photochemical transformation of the [2+2] macrocycle formed
the corresponding [1+1] macrocycle.
To confirm this hypothesis, a sample of 7a was prepared by dis-
solving a small amount in deuterated chloroform in an NMR tube
and then exposing the sample to a single low-energy 350-nm lamp
in a photochemical reactor, for short periods and monitoring the
solution directly by 1H NMR. Preliminary experiments led to the
optimal conditions which are summarized below. As can be seen
in Fig. 3 increasing the light exposure time resulted in a gradual in-
crease in the formation of the [1+1] macrocycle 10a. This can be
followed by comparing the triplet NH signals of the different spec-
tra with the triplet NH signals of the reference macrocycles 7a and
10a. Thus, the [2+2] macrocycle can be seen to undergo a gradual
conversion into the corresponding [1+1] macrocycle.
Additional support for this process was obtained from TLC mon-
itoring and from the positive-mode APCI-mass spectra of the solu-
tion which had been irradiated for only 8 min which, in the
positive mode, showed m/z signals at 769 and 1538 corresponding
to the masses of [10a+H+] and [7a+H+], respectively. Sulfonamide 7
showed a similar photochemical behavior and was monitored by
the 1H NMR spectra from a similar timed 350-nm irradiation
experiment (Fig. 4).
Photochemical reactions of sulfonamides in general have prece-
dents in the literature and many are particularly of interest due to
the fact that pharmaceutical aryl sulfonamide and sultam com-
pounds may be photolabile.12–14 Döpp, for example, conducted
photochemical studies on sultams related to saccharin and pro-
posed several mechanisms to account for the resulting photoprod-
ucts obtained.15 In 1980 Lancaster and Smith reported the
photochemical extrusion of SO2 from 1,3-dihyrobenz[c]-2,1-iso-
thiazoles to form a bicyclic benzazetine.16 Weiss et al. also con-
ducted a series of photochemical reactions on sulfonamides and
included a photo-Fries rearrangement among several different
pathways proposed to account for their observed products.17
NH
S
S
S
S
O2
O2
O2
O2
A
B
Decomposition
product(s)
H
N
O2
S
O2
H
N
h
ν
S
+
10 10a
or
N
H
S
N
H
S
O2
O2
C
10 10a
or
Scheme 5. A proposed mechanism for the photochemical reactions.
The mechanism of the photochemical transformations which
we observed in our examples can only be speculated upon. An
analogous but non-photochemical disproportionation equilibrium
process however was reported recently by Jurczak and co-work-
ers18 involving the macrocyclic [1+1] bis- and [2+2] tetra-imines
11 and 12 formed, respectively, by the condensation of dialdehyde
13 with the diamine 14 (Scheme 4).
In our instances however and under the condition studied, there
was no evidence of an equilibrium process occurring, and a possi-
ble rationale for the observed transformation(s) which occur can
be visualized in Scheme 5. It is presumed that a first step involves
the hemolytic cleavage of a sulfur-nitrogen bond in 7 (or 7a) as
shown in A to form the diradical species B which can then undergo
an intramolecular homolytric bond cleavage and concomitant for-
mation of 10 from 7 (or 10a from 7a) and the diradical species C
which can either form 10 (or 10a from 7a) or directly undergo for-
mation of decomposition products.
In conclusion, two new bis-sulfonamide macrocycles 10 and
10a and two tetra-sulfonamide macrocycles 7 and 7a have been
synthesized by reacting the ditosyl-protected CTA disulfonyl
dichloride derivative 8 with the diamines 9 and 9a, respectively,
and a single-crystal X-ray crystal structure was determined for
10a. Complexation studies of 7 and 7a were conducted with both
Group 1 and 2 chlorides and TBA salts using both 1H NMR and
ESI MS indicating that these new macrocycles were moderately
good hosts for 1:1 complexation with TBAX halides. Finally, the
[2+2] macrocycles 7 and 7a were shown to be photolabile and
unexpectedly formed the corresponding [1+1] macrocyclic com-
pounds 10 and 10a, respectively.
Acknowledgments
We thank the research support from Memorial University of
Newfoundland and from the Ministry of Higher Education, Egypt

3448
H. F. Sleem et al. / Tetrahedron Letters 54 (2013) 3444–3448
6. Jeffamines EDR-176 and EDR-148 are the trade names for diamines 9 and 9a as
supplied by The Huntsman Corporation, Texas.
7. Sleem, H. F.; Dawe, L. N.; Georghiou, P. E. New J. Chem. 2012, 36, 2451–2455.
8. He, L.; An, Y.; Feng, W.; Li, M.; Zhang, D.; Yamato, K.; Zheng, C.; Zheng, X. C.;
Gong, B. PNAS 2006, 103, 10850–10855.
for the Scholarship to H.F.S. The Huntsman Corporation, Texas, USA
is thanked for the generous gift of the Jeffamines used in this study.
Supplementary data
9. Bochen´ ska, M.; Biernat, J.; Bradshaw, J. S.; Koyama, H.; Izatt, R. M. J. Inclu.
Phenom. 1988, 6, 593–597.
10. Paruch, K.; Vyklicky, L.; Katz, T. J.; Incarvito, C. D.; Rheingold, A. L. J. Org. Chem.
2000, 65, 8774–8782.
11. Connors, K. A., Binding Constants, Wiley, New York, 1987: Kassoc values were
calculated using non-linear curve fitting using 1:1 and 2:1 binding constant
equations as derived after Connors and programmed into ORIGINPro 7.5 from
OriginLab Corporation. The titration spectral data and binding curves described
in this paper are presented in the SI.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.tetlet.2013.
04.081.
References and notes
12. Drugs: Photochemistry and Photostability; Albini, A., Fasani, E., Eds.; The Royal
Society of Chemistry: London, 1998.
13. Papageorgiou, G.; Corrie, J. E. T. Tetrahedron 1999, 55, 237–254.
14. Hill, R. R.; Jeffs, G. E.; Roberts, D. R.; Wood, S. A. Chem. Commun. 1999, 1735–
1736.
1. (a) Georghiou, P. E.; Ho, C. K. Can. J. Chem. 1989, 67, 871–876; (b) Georghiou, P.
E.; Ho, C. K.; Jablonski, C. Can. J. Chem. 1991, 69, 1207–1211.
2. (a) Budesinsky, B. In Chelates in Analytical Chemistry; Flaschka, H. A., Barnard, A.
J., Eds.; Marcel Dekker Inc.: New York, 1969; Vol. 2, (b) Prand, C.; Venturini, T. J.
Chromatogr. Sci. 1981, 19, 308–314; (c) Jandera, P.; Churacek, J. J. Chromatogr.
1980, 197, 181–187.
15. Döpp, D. Int. J. Photoenergy 2001, 3, 41–48.
16. Lancaster, M.; Smith, D. J. H. J. Chem. Soc., Chem. Commun. 1980, 471–472.
17. Weiss, B.; Dürr, H.; Haas, H. J. Angew. Chem., Int. Ed. Engl. 1980, 19, 648–650.
18. Ceborska, M.; Tarnowska, A.; Jurczak, J. Tetrahedron 2010, 66, 9532–9537.
3. Eegriwe, E. Anal. Chem. 1937, 110, 22–25.
4. Poh, B. L.; Lim, C. S.; Khoo, K. S. Tetrahedron Lett. 1989, 30, 1005–1008.
5. (a) Poh, B. L.; Tan, C. M. Tetrahedron 1993, 49, 9581–9592; (b) Poh, B. L.; Tan, C.
M.; Lim, C. H.; Tan, C. M.; Wong, W. M. Tetrahedron 1993, 49, 7259–7266; (c)
Poh, B. L.; Lim, C. S. Tetrahedron 1990, 46, 3651–3658.