Substituted heterocyclic naphthalene diimides with unexpected acidity. Synthesis, properties, and reactivity

Article abstract of DOI:10.1021/jo9017342

(Chemical Equation Presented) Naphthalene bisimides (NDIs) with a heterocyclic 1,4-dihydro-2,3-pyrazinedione moiety have been synthesized from both 2,6-dibromonaphthalene and 2,3,6,7-tetrabromonaphthalene bisanhydrides by means of a stepwise protocol incl

Full text of DOI:10.1021/jo9017342

pubs.acs.org/joc
Substituted Heterocyclic Naphthalene Diimides with Unexpected Acidity.
Synthesis, Properties, and Reactivity
Filippo Doria,† Marco di Antonio,‡ Michele Benotti,† Daniela Verga,† and Mauro Freccero*,†
†
ꢀ
Dipartimento di Chimica Organica, Universita di Pavia, V. le Taramelli 10, 27100 Pavia, Italy, and
‡Dipartimento di Scienze Farmaceutiche, Universitaꢀ di Padova,Via Marzolo, 5, 35131 Padova, Italy
mauro.freccero@unipv.it
Received August 10, 2009
Naphthalene bisimides (NDIs) with a heterocyclic 1,4-dihydro-2,3-pyrazinedione moiety have been
synthesized from both 2,6-dibromonaphthalene and 2,3,6,7-tetrabromonaphthalene bisanhydrides
by means of a stepwise protocol including imidization, nucleophilic displacement of the bromine
atoms by ethane-1,2-diamine, in situ reductive dehalogenation, and further oxidation. These
heterocycles (R = n-pentyl, cyclohexyl) are yellow dyes with green emission in organic solvent,
where the acid form dominates. The orange nonfluorescent conjugate base can be generated
quantitatively by CH3COONBu4 addition in DMSO, where it exhibits a pKa=7.63. The conjugate
base becomes the only detectable species (by UV-vis spectroscopy), in water solution, even under
acid conditions (pH 1). In aqueous DMSO the acid/base equilibrium is a function of the DMSO/
water ratio. The unexpected acidity of these heterocyclic NDIs, which justifies the reactivity with
CH2N2, has been rationalized by DFT computational means [PBE0/6-31+G(d,p)] in aqueous
solvent (PCM models) as a result of a strong specific solvation effect, modeled by the inclusion of
three water molecules.
Introduction
anions, under mild conditions,3 they have been exploited as
electron-acceptor moieties in artificial light-harvesting sys-
tems.4 In supramolecular chemistry, several catenanes, ro-
taxanes, and self-assembling systems have been designed and
synthesized around NDI cores.5
1,4,5,8-Naphthalenetetracarboxylic acid diimides (NDIs)
and their core-substituted derivatives have been a thor-
oughly investigated class of compounds in the last decades.1
Such an interest is motivated by their numerous applications
in material science due to their semiconducting properties.2
Since NDIs can be reversibly reduced to stable radical
NDIs also attracted a great deal of interest in biological
and medical areas as DNA intercalating agents.6 In addition,
(1) (a) Bhosale, S. V.; Jani, C. H.; Langford, S. J. Chem. Soc. Rev. 2008, 37,
331–342. (b) Grimshaw, J.; De Silva, A. P. Chem. Soc. Rev. 1981, 10, 181–203.
(2) (a) Kishore, R. S. K.; Kel, O.; Banerji, N.; Emery, D.; Bollot, G.;
Mareda, J.; Gomez-Casado, A; Jonkheijm, P.; Huskens, J.; Maroni, P.;
Borkovec, M.; Vauthey, E.; Sakai, N.; Matile, S. J. Am. Chem. Soc. 2009,
131, 11106–11116. (b) Katz, H. E.; Lovinger, A. J.; Johnson, J.; Kloc, C.;
Siegrist, T.; Li, W.; Lin, Y.-Y.; Dodabalapur, A. Nature 2000, 404, 478–481.
(3) (a) Di Antonio, M.; Doria, F.; Mella, M.; Merli, D.; Profumo, A.;
Freccero, M. J. Org. Chem. 2007, 72, 8354–8360. (b) Heywang, G.; Born, L.;
Fitzky, H.-G.; Hassel, T.; Hocker, J.; Muller, H.-K.; Pittel, B.; Roth, S.
Angew. Chem., Int. Ed. 2004, 43, 668–698. (c) Zhong, C. J.; Kwan, W. S. V.;
Miller, L. Chem. Mater. 1992, 4, 1423–1428.
(4) (a) El-Khouly, M. E.; Kim, J. H.; Kay, K.-Y. C.; Chan, S.; Ito, O.;
€
Fukuzumi, S. Chem.;Eur. J. 2009, 15, 5301–5310. (b) Roger, C.; Yuliya, M.;
Brunner, D.; Holzwarth, A. R.; Wurthner, F. J. Am. Chem. Soc. 2008, 130,
€
(c) Wurthner, F. Angew. Chem. 2001, 113, 1069–1071. (c) Takenada, S.; Uto,
Y.; Saita, M.; Kondo, H.; David, W. Chem. Comm. 1998, 1111–1112.
€
5929–5939.
8616 J. Org. Chem. 2009, 74, 8616–8625
Published on Web 10/22/2009
DOI: 10.1021/jo9017342
r
2009 American Chemical Society

Doria et al.
JOCArticle
the propensity of NDIs to form excimers propelled the
development of NDI-based fluorescent probes for sensitive
and selective DNA detection.7
Very recently, tetrasubstituted naphthalene diimides, con-
taining two polar or cationic substituents on the NDI core
have been developed as a new class of promising ligands
toward human telomeric quadruplex DNA (G-quadruplex).
They showed selectivity over duplex DNA, inducing sene-
scence in cancer cell lines typical of telomere-directed effects.8
Functionalization of NDIs by core substitution has been
In an attempt to design new extended aromatics exhibiting
a T-shaped planar structure, to be exploited as G-quadruplex
ligands, we have combined an efficient two steps synthetic
route to NDIs bearing cyclic substituents at the same side of
the naphthalene core, at the 2,3 positions, yielding 4a and 4b,
with an oxidative protocol generating tetraazabenzo-
[e]pyrene-1,3,6,8,10,11-hexaones (5a and 5b, in Chart 2),
bearing two different alkyl substituents at the imide moieties.
The heterocycles 5a and 5b exhibit an unexpected acidity
both in DMSO and in aqueous solution, which has been
investigated experimentally and modeled by DFT computa-
tional means coupled with PCM solvation models.
€
used by Wurthner’s group to induce a structural mediated
tuning in the optical and redox properties of this class of
dyes, extending the scope of their application.9
Synthetically di- and trisubstituted NDIs at the aromatic
core are accessible from 2,6-dichloronaphthalene dianhydride
by imidization with primary amines, followed by substitution
of the chlorine atoms at the naphthalene core.10 A great
variety of core 2,6-di- and 2,3,6-trisubstituted NDIs (see
Chart 1 for numbering) have been synthesized starting from
2,6-dichloro- or 2,6-dibromonaphthalene dianhydride.10c,11,12
Very recently, NDIs bearing four electron-donating substitu-
ents at the naphthalene core (Chart 1) have been reported
using 2,3,6,7-tetrabromonaphthalene dianhydride as a pre-
cursor.13,14
CHART 2
CHART 1. Core 2,6-Di- (1), 2,3,6-Tri- (2), and 2,3,6,7-Tetra-
substituted NDIs (3)
Results and Discussion
Synthesis. Two different synthetic protocols have been
used to achieve the syntheses of the heterocycle 4a. The first
one used 2,3,6,7-tetrabromonaphthalene dianhydride as
starting material; the second one used the 2,6-dibromo-
naphthalene dianhydride. 2,3,6,7-Tetrabromonaphthalene
dianhydride was prepared by bromination of naphthalene
dianhydride with 2.5 equiv of dibromoisocyanuric acid
(DBI) in oleum (20% SO3) according to the procedure
report by Wurthner.14 The imidization of the anhydride by
€
pentylamine and cyclohexylamine outlined in Scheme 1 was
carried out in acetic acid. The tetrabromonaphtalene diimide
8a was synthesized as a major product in low yield 40%,
along with byproducts arising from complete or partial
reductive dehalogenation 6a (35%) and 7a (26%). In order
to improve the reaction yield of the diimide 8a, another
preparation was run in a dedicated microwave reactor, under
atmospheric pressure, in an open reaction vessel. The reac-
tion was carried out at 150 °C for a period of 30 min. Under
these conditions, 100 W was the power required for 45 s to
warm the reaction mixture at the desired temperature,
followed by pulses of 45-50 W to maintain the reaction
mixture at 150 °C. The microwave-assisted synthesis gener-
ates the imide adducts 8a, in a much better yield (65%),
reducing the dehalogenation byproducts.
(5) For selected examples, see: (a) Fallon, G. D.; Lee, M. A.-P.;
Langford, S. J.; Nichols, P. J. Org. Lett. 2004, 6, 655–658. (b) Wang, X.-
Z.; Li, X.-Q.; Shao, X.-B.; Zhao, X.; Deng, P.; Jiang, X.-K.; Li, Z.-T.; Chen,
Y.- Q. Chem.;Eur. J. 2003, 9, 2904–2913. (c) Li, X.-Q.; Feng, D.-J.; Jiang,
X.-K.; Li, Z.-T. Tetrahedron 2004, 60, 8275–8284. (d) Gunter, M. J.;
Farquhar, S. M. Org. Biomol. Chem. 2003, 1, 3450–3457. (e) Johnstone, K.
D.; Yamaguchi, K.; Gunter, M. J. Org. Biomol. Chem. 2005, 3, 3008–3017. (f)
Gunter, M. J. Eur. J. Org. Chem. 2004, 8, 1655–1673. (g) Vignon, S. A.;
Jarrosson, T.; Lijima, T.; Tseng, H.-R.; Sanders, J. K. M.; Stoddart, J. F. J.
Am. Chem. Soc. 2004, 126, 9884–9885. (h) Hansen, J. G.; Feeder, N.;
Hamilton, D. G.; Gunter, M. J.; Becher, J.; Sanders, J. K. M. Org. Lett.
2000, 2, 449–452. (i) Kieran, A. L.; Pascu, S. I.; Jarrosson, T.; Gunter, M. J.;
Sanders, J. K. M. Chem. Commun. 2005, 1842–1844. (j) Pantos, G. D.; Pengo,
P.; Sanders, J. K. M. Angew. Chem., Int. Ed. 2007, 46, 194–197.
(6) (a) Chu, Y.; Hoffman, D. W.; Iverson, B. L. J. Am. Chem. Soc. 2009,
131, 3499–3508. (b) Lee, J.; Guelev, V.; Sorey, S.; Hoffman, D. W.; Iverson,
B. L. J. Am. Chem. Soc. 2004, 126, 14036–14042. (c) Guelev, V.; Sorey, S.;
Hoffman, D. W.; Iverson, B. L. J. Am. Chem. Soc. 2002, 124, 2864–2865. (d)
Sato, S.; Takenaka, S. J. Organomet. Chem. 2008, 693, 1177–1185. (e)
Takenaka, S.; Uto, Y.; Saita, H.; Yokoyama, M.; Kondo, H.; Wilson, W.
D. Chem. Comm. 1998, 1111–1112.
The bromo NDIs 6a-8a were then isolated by column
chromatography and used as precursors for the next step of
(7) (a) Lee, H. N.; Xu, Z.; Kim, S. K.; Swamy, K. M. K.; Kim, Y.; Kim, S.-
J.; Yoon, J. J. Am. Chem. Soc. 2007, 129, 3828–3829. (b) Mokhir, A.;
€
€
(10) (a) Thalacker, C.; Roger, C.; Wurthner, F. J. Org. Chem. 2006, 71,
Kramer, R.; Wolf, H. J. Am. Chem. Soc. 2004, 126, 6208–6209.
€
8098–8105. (b) Wurthner, F.; Ahmed, S.; Thalacker, C.; Debaerdemaeker, T.
Chem.;Eur. J. 2002, 8, 4742–4750.
(11) Jones, B. A.; Facchetti, A.; Marks, T. J.; Wasielewski, M. R. Chem.
Mater. 2007, 19, 2703–2705.
(12) Chaignon, F.; Falkenstrom, M.; Karlsson, S.; Blart, E.; Odobel, F.;
Hammarstrom, L. Chem. Commun. 2007, 64–66.
(13) (a) Vollmann, H.; Becker, H.; Corell, M.; Streeck, H. Liebigs Ann.
1937, 531, 1–159. (b) Kishore, R. S. K.; Ravikumar, V.; Bernardinelli, G.;
Sakai, N.; Matile, S. J. Org. Chem. 2008, 73, 738–740.
(8) (a) Cuenca, F.; Greciano, O.; Gunaratnam, M.; Haider, S.; Munnur,
D.; Nanjunda, R.; Wilson, W. D.; Neidle, S. Bioorg. Med. Chem. Lett. 2008,
18, 1668–1673. (b) Gunaratnam, M.; Swank, S.; Haider, S. M.; Galesa, K.;
Reszka, A. P.; Beltran, M.; Cuenca, F.; Fletcher, J. A.; Neidle, S. J. Med.
Chem. 2009, 52, 3774–3783.
€
(9) (a) Suraru, S.-L.; Wurthner, F. Synthesis 2009, 1841–1845. (b) Roger,
C.; Miller, M. G.; Lysetska, M.; Miloslavina, Y.; Holzwarth, A. R.;
€
Wurthner, F. J. Am. Chem. Soc. 2006, 128, 6542–6543. (c) Banerji, N.;
Furstenberg, A.; Bhosale, S.; Sisson, A. L.; Sakai, N.; Matile, S.; Vauthey, E.
€
(14) Roger, C.; Wurthner, F. J. Org. Chem. 2007, 72, 8070–8075.
J. Phys. Chem. B 2008, 112, 8912–8922.
J. Org. Chem. Vol. 74, No. 22, 2009 8617

Doria et al.
JOCArticle
SCHEME 1. Imidization of the 2,3,6,7-Tetrabromonaphthalene
ring closing (step iii, in Scheme 3), leading to a mixture of
products 4a and 10a was quantitative after an additional 14 h
at rt. The cyclization reaction could be an intramolecular
conjugate addition of the NH2 group to the ortho position,
with the aromatic ring bearing the amide moiety acting as
Michael acceptor, followed by an oxidative aromatization.16
Since more than 50% of the adduct arising from the cycliza-
tion in the mixture is dehalogenated, the NDI core acts as the
main oxidizing agent. The final step (iv, in Scheme 3) was a
mild and quantitative reductive dehalogenation, of both the
monobromo derivatives 10a and 10b, in aqueous DMSO by
Na2S2O4 to afford 4a and 4b, which were used, without
purification, to prepare the diamide analogues 5a and 5b.
Scheme 4 shows the final oxidative step of the synthesis.
To our knowledge, this oxidation is a novelty in the current
literature. The closest reaction reported in the literature is the
oxidation of the 1H-quinoxalin-2-one to 1,4-dihydroqui-
noxaline-2,3-dione by permanganate.17 Several oxidants
have been used such as DDQ, Dess-Martin, and Jones, of
which only the latter successfully oxidized the reagents into
the dicarbonyl products 5a and 5b in fairly good yields
(75%).
Reactivity of 5a toward CH2N2. It has been previously
shown that heterocycles such as 2,3-dihydroxyquinoxalines
and 2-pyridones can slowly react with diazomethane to
afford in low yield both O- and N-methylation, only in the
presence of acid silica gel catalysis.18 In contrast, the hetero-
cycle 5a reacts instantaneously with diazomethane in organic
solvents (Et2O and CH2Cl2), with N2 generation, exhibiting a
behavior very similar to that of a carboxylic acid. By con-
trolling the reaction time it has been possible to selectively
generate the monomethylated and the bis-methylated ad-
ducts 15a and 16a (Scheme 5). The O-methylated adducts
have been used as standards to investigate the tautomeric
and acid-base equilibriums of 5a and 5b in organic solvents
and in aqueous media by UV-vis spectroscopy.
Bisanhydride by Pentylaminea
aReagents and conditions: (a) dibromoisocyanuric acid (DBI), oleum
(20% SO3), 25 °C, 3 h, yield 93%; (b) pentylamine 150 °C, 30 min,
microwave assisted (8a, 65%, yield).
synthesis: a nucleophilic substitution of the halogen atoms
with ethane-1,2-diamine (EDA).14 The nucleophilic aro-
matic substitution (SNAr) on both 6a and 7a gave 4a in good
yields (g85%), in DMF at 130 °C. The SNAr reaction on 8a
was carried out using three different conditions: (a) in DMF,
(b) in neat EDA, both at 130 °C with an oil bath, for 30 min,
and (c) in DMF microwave-assisted (MW) at 150 °C, 250 psi,
200 W, 10 min, as outlined in Table 1. The ring-closure by
two sequential SNAr, and the further reductive dehalogena-
tion yielding 4a, occurred in one pot in DMF in the presence
of EDA at 130 °C for 30 min (Scheme 2). The product 11a
resulting from a complete SNAr, which is always a bypro-
duct, was generated in higher yield in the microwave-assisted
protocol. To improve the synthesis of 4a minimizing the
side-reaction products, we followed a more efficient syn-
thetic route starting from 2,6-dibromonaphthalene dianhy-
dride as depicted in Scheme 3. Since this second synthetic
protocol is much more efficient then the first one, particu-
larly for bulky R substituents, we used it also to synthesize
4b and 5b.
Spectroscopic Properties in Organic and Aqueous Solvents.
Absorbance and Emission Spectra. In order evaluate the
possibility of exploiting the new NDIs as a fluorescent
chemosensor, and more specifically as probes for G-quad-
ruplex sensing, we decided to measure the absorbance and
emission spectra of the NDIs 5a and 5b and their precursors
4a and 4b. The disubstituted NDI 4a is a green fluorescence
compound (Figure 1a, λmax = 498 nm) with an absorption
spectra characterized by an electronic transition in the
400-550 nm range, with λmax = 488 nm (Figure 1a). Com-
TABLE 1. Reactants and Conditions for the Nucleophilic Aromatic
Substitution on Polybrominated NDIs 6a-8a and 12a
reactant
conditionsa
products (%, yield)
4a (92)
4a (85), 10a (<5%)
6a
7a
8a
8a
8a
a
a
b
a
c
DMF, Ar, 130 °C, 30 min
DMF, Ar, 130 °C, 30 min
neat EDA, Ar, 130 °C, 30 min 9a (24), 4a (27), 11a (25)
DMF, Ar, 130 °C, 30 min
DMF, MW, 150 °C, 250 psi,
200 W, 10 min
9a (3), 4a (86)
9a (35), 4a (16), 11a (7)
8a
12a
12a
d
d
d
neat EDA, Ar, rt, 16 h
neat EDA, Ar, rt, 16 h
neat EDA, Ar, rt, 2 h
4a (5), 10a (72)
4a (52), 10a (40)
4a (21), 14a (50), 10a(8)
€
pared to 2,6-dialkylamino NDIs synthesized by Wurthner
(λmax = 620 nm),10a the absorption maximum is blue-shifted
by 132 nm. The UV-vis property of 4a is much more similar
to that of 2,6-dialkoxy NDIs (λmax = 470 nm).10a This
evidence suggests that cyclic amine substituents at 2,3 posi-
tions on the NDI core are less efficient electron donor than
the conformational flexible acyclic amine moieties at 2,6
positions. After the oxidation, the UV-vis spectra of the
resulting 5a (Figure 1b) is 34 nm blue-shifted in comparison
to the spectra of 4a (Figure 1a). Compound 5a exhibits an
aIn neat ethane-1,2-diamine (EDA) or in DMF.
The 2,6-dibromo-substituted NDIs 12a and 12b were
synthesized by a classical method of imidation of the 2,6-
dibromo-1,4,5,8-naphthalenetetracarboxylic acid dianhy-
dride. The following SNAr yielding 14a and subsequent
ring-closure was carried out in neat EDA, at room tempera-
ture (condition d, in Table 1), following a similar pro-
€
cedure used by Wurthner for the functionalization of
2,6-dichloro NDIs.15 The reaction was monitored by
reversed-phase HPLC. The reactant 12a was converted into
the monosubstituted adduct 14a after 2 h. The following
(16) An intermediate resulting from the cyclization step (iii) in Scheme 3
was detected by 1H NMR in CDCl3. The structural characterization and its
decay are currently under investigation.
(17) Obafemi, C. A.; Pfleiderer, W. Helv. Chim. Acta 1994, 77, 1549–56.
(18) (a) Nishiyama, H.; Nagase, H.; Ohno, K. Tetrahedron Lett. 1979, 48,
4671–4674. (b) Cheeseman, G. W. H. J. Chem. Soc. 1955, 1804–1809.
€
(15) Roger, C.; Ahmed, S.; Wurthner, F. Synthesis 2007, 12, 1872–1876.
8618 J. Org. Chem. Vol. 74, No. 22, 2009

Doria et al.
JOCArticle
SCHEME 2. Nucleophilic Aromatic Substitution on Polybromo-NDIsa
aReagents and conditions are shown in Table 1.
SCHEME 3. Synthesis of 4a and 4b according to the Second
emission spectra of the NDI 5b are almost superimposable
on that of 5a (Supporting Information).
Synthetic Protocola
The UV-vis spectra of both NDI 5a and 5b were inves-
tigated in several organic solvents: benzene, CHCl3, metha-
nol, acetonitrile (ACN) (Figure 2a), DMSO, and aqueous
DMSO (Figure 2b).
The absorbance spectra in organic solvents is not substan-
tially affected by the polar orprotic nature of the solvent, since
in the above solvents including DMSO the spectra are very
similar (for 5a see Figures 2aand 2b, redline). The appearance
of a new absorbance in water (Figure 2b, blue line) reveals
the presence of another additional chemical species, which is
in equilibrium with 5a, as suggested by the two isosbestic
points at 366 and 469 nm in the spectra recorded in aqueous
DMSO (Figure 2c). The equilibrium is shifted progressively
toward the new species increasing the water ratio, which
becomes the most populated in water containing a small
amount of DMSO (2%) (Figure 2b). The comparison of the
UV-vis spectra of the new species in water to the spectra of
both the mono- and the dimethoxy-substituted heterocycles
15a and 16a (Figure 3a), suggests that the new species is not a
tautomer of the diamide 5a. On the other hand, the addition of
a weakbase, such astriethylamine(TEA, 0.3 equivalents), to a
solution of 5a in CHCl3 changes the spectra generating a
new band with two absorption maxima (λmax = 487, 509 nm,
Figure 3b), which are fairly similar to those of the new species
in water solution (λmax = 485, 495 nm). The effect of triethyl-
amine (TEA, 1 equiv) or CH3COONBu4 on the spectra,
generating a new red-shifted absorption with two maxima in
DMSO (λmax=521, 549 nm), and in ACN (λmax=521, 549
nm), suggests that the heterocycles 5a in organic solvents is
quantitatively deprotonated by a weak base, and the new
bands have to be assigned to the anion 5aA (Scheme 6) free or
as ion pair with the triethylammonium cation.
aReagents and Conditions: (i) pentylamine, CH3COOH, 130 °C, 30 min;
(ii) EDA, 25 °C, 2 h, Ar; (iii) EDA, 25 °C, 14 h, Ar; (iv) Na2S2O4, DMSO/
H2O = 20:80, 40 °C, 2 h, Ar.
SCHEME 4. Oxidation of 4 to 5a
aConditions: (a) K2Cr2O7, H2SO4, H2O/acetone, 100 °C, 2 h.
SCHEME 5a
It is worth noting that the efficient solvation of the trietylam-
monium cation leaves the anion 5aA free in DMSO, resulting
in a UV-vis spectrum red-shifted by 70 nm in comparison to
the spectrum in chloroform where the ion pair 5aA/trietyl-
ammonium cation is present. Efficient ion-pair separation
results in a progressive red shift of the UV-vis absorbance
of the anion 5aA (passing from chloroform, ACN, and DMSO
in the presence of TEA). Due to the similarity of the spec-
tra in water to that in chloroform in the presence of TEA,
we suggests that the new absorbance in water solution has to be
ascribed to the anion 5aA.
aReagent and conditions: (a) CH2N2, CHCl3, rt, 30 s; (b) CH2N2,
CHCl3, rt, 10 min.
The NDI 5b, with the more bulky cyclohexyl substituents
replacing the n-pentyl groups at the imide nitrogen atoms,
mirrors the behavior of 5a, with the generation of an anion
intense cyan (blue-green) bright fluorescence emission
(Figure 1b), more intense than 4a. The absorption and
J. Org. Chem. Vol. 74, No. 22, 2009 8619

Doria et al.
JOCArticle
FIGURE 1. (a) UV-vis absorption (black lines) and fluorescence spectra (colored lines) of (a) 4a (10-5 M) and (b) 5a (2 ꢀ 10-5 M).
FIGURE 2. (a) UV-vis absorption spectra of 5a in ACN (6 ꢀ 10-5M), benzene, CH3OH and CHCl3. (b) UV-vis absorption spectra of 5a in
DMSO (5 ꢀ 10-5M, red line) and in water/DMSO = 98:2 (blue line). (c) UV-vis absorption spectra of 5a in aqueous DMSO (5 ꢀ 10-5M) from
77:23 to 52:48 DMSO/water volume ratio.
FIGURE 3. (a) UV-vis spectra of 5a in water (3 ꢀ 10-5 M), 15a and 16a in chloroform (3 ꢀ 10-5 M). (b) UV-vis spectra of 5a in water (3 ꢀ
10-5 M, fuchsia), in neat chloroform (5 ꢀ 10-5 M, brown), in chloroform with TEA (0.3 equiv, orange), and in DMSO with TEA (1 equiv,
cyan).
SCHEME 6. Acid-Base Equilibrium Involving 5a,b and Their Conjugate Bases 5aA and 5bA
5bA in water solution. For a comparison of the UV-vis
absorbance of the two anions 5aA and 5bA, see Figure 4.
Although the more bulky cyclohexyl substituent slightly
reduces the tailing of the UV-vis spectra over 500 nm
(probably caused by clustering of the conjugate base with
the acid), the shape and the maxima of the two spectra are
8620 J. Org. Chem. Vol. 74, No. 22, 2009

Doria et al.
JOCArticle
are evaluated instead of absolute pKa. This requires the choice
of a similar reference molecule, in our case, 1,4-dihydro-2,3-
quinoxalinedione (QXH). QXH is a cyclic bis-amide (Scheme 7)
exhibiting structural and electronic features quite similar to
those of the model NDI, for which the experimental pKa has
been experimentally measured. The experimental pKa value
measured in water [pKa(exp)(QXH)] is 9.27, in good agreement
with previously published data (9.52, 9.75).22 The pKa of the
model NDIH has been calculated from the computed ΔGaq
value for the proton-transfer reactions in Scheme 7 and the
experimental pKa of QXH, according to eq I.
pKaðNDIHÞ ¼ ΔGaq=RT ln 10þpKaðexpÞðQXHÞ ðIÞ
Two different proton-transfer reaction models (a) and (b)
have been considered (Scheme 7). The most simple one (a)
does not include any water molecules. The second proton-
transfer reaction model (b) includes three water molecules H-
bonded to each acid/base couple. The choice of the “three
water molecules model” is the compromise between the needs
tocontainthe molecularcomplexityofthe water clusterandto
describe the specific solvation involving the most negatively
charged atoms of the anionNDI-. The second model has been
suggested by the fact that error of the PCM method in the
predictions of acidities in water may became as large as 7 pKa
units, if the solvent-solute interactions (mainly due to hydro-
FIGURE 4. UV-vis spectra of the anions 5aA (3 ꢀ 10-5 M) (blue
line) and 5bA (red line) in water solution.
very similar. The maxima in the UV-vis spectra are un-
affected by concentration in the 10-4-10-5 M range.
NDI Acidity Measurements. The UV-vis spectra of 5a
change for the addition of an equimolar amount of TEA in
DMSO as a consequence of the equilibrium depicted in Scheme 6.
The titration of 5a, using CH3COONBu4 as a base in DMSO
(Figure5),gaveapKa 7.63, which is 10 orders of magnitude lower
than the pKa of 2-pyridone in DMSO (pKa 17.0).19
gen bonding) in the first solvation shell are neglected.23
A
remedy for this problem was to use a cluster-continuum
model.24 This cluster-continuum model is a hybrid approach
that combines gas-phase clustering by explicit solvent mole-
cules and solvation of the cluster by the dielectric continuum.
Using the cluster-continuum model, pKa between -10 and 50
for 17 acids in aqueous solution were calculated with an error
of 2.2 pKa units.23 The above results have clearly demon-
strated that, by using the PCM method, one is able to
predict the pKa in aqueous solution with a precision of about
0.5-2.2 pKa units. Therefore, the cluster-continuum model
should allow a better evaluation of the solvation effects
of water on the NDI acidity. Bulk solvation effects have
been computed for both models (a) and (b) optimizing
the acid and conjugate base structures in the solvent bulk by
PCM solvation model at PBE0/6-31þG(d,p) level of theory,25
using UAHF-radii.26
Geometries in Gas Phase and in Water Bulk. In the model
(a), all the structures, with the exception of the anion NDI-,
are planar. The nonplanarity of the anion NDI- is due to the
strong electrostatic repulsion between the nitrogen and
oxygen atoms, both negatively charged. The effect of the
solvent bulk reduces significantly the out of plane bending
of the carbonyl moiety relative to the dihydroxypyrazine
ring. Model (b) appeared to be computationally more de-
manding since the three water molecules are loosely bound
to the heterocycles by a H-bonding network with great
FIGURE 5. Absorption spectra of a DMSO solution of 5a (1 ꢀ
10-4M) measured during a titration with CH3COONBu4.
The UV-vis absorbance of 5aA does not change in water
solution in the pH range 1-8. Therefore, it has not been
possible to measure the pKa of 5a in water solution since it
should be lower than 1.
Computational Acidity. In order to evaluate the acidity of
the NDIs 5a and to describe the effect of water as solvent on the
acidity, we have computed the pKa of a prototype NDI model
(NDIH, exhibiting methyl substituents, on both the imide
moieties, see Scheme 7) by a semiempirical computational stra-
tegy, which has been used successfully to evaluate the pKa for
twisted amide20 and unstable cyclopropylpyrroleindoles.21
Considerable cancellation of errors is expected if relative pKa
(22) Albert, A.; Phillips, J. N. J. Chem. Soc. 1956, 1294–1304.
(23) Pliego, J. R., Jr.; Riveros, J. M. J. Phys. Chem. A 2002, 106, 7434–
7439.
(24) Pliego, J. R., Jr.; Riveros, J. M. J. Phys. Chem. A 2001, 105, 7241–
7247.
(25) (a) Saracino, G. A. A.; Improta, R.; Barone, V. Chem. Phys. Lett.
2003, 411–415. (b) Adamo, C.; Barone, V. J. Chem. Phys. 1999, 110, 6158–
6170. (c) Cimino, P.; Gomez-Paloma, L.; Barone, V. J. Org. Chem. 2004, 69,
7414–7422.
(26) Barone, V.; Cossi, M.; Tomasi, J. J. Chem. Phys. 1997, 108, 3210–
3221.
(19) Bordwell, F. G. Acc. Chem. Res. 1988, 21, 456–463.
(20) Mujika, J. I.; Mercero, J. M.; Lopez, X. J. Phys. Chem. A 2003, 107,
6099–6107.
(21) Freccero, M.; Gandolfi, R. J. Org. Chem. 2005, 70, 7098–7106.
J. Org. Chem. Vol. 74, No. 22, 2009 8621

Doria et al.
JOCArticle
FIGURE 6. Geometries in the gas phase and in water bulk (in parentheses, PCM model, UAHF radii) for the model anion NDI- 3H2O in the
presence of three water molecules at the PBE0/6-31þG(d,p) level of theory.
SCHEME 7. Proton-Transfer Reaction without (a) and with Three Water Molecules H-Bonded to Each Acid/Base Couple (b) Used as
Models for the Computational Evaluation of the pKa of 5a
conformational mobility. Actually, we were able to locate
only one conformational minima both in the gas phase and in
solvent bulk for the three stationary points QXH 3H2O, QX-
3H2O, and NDIH 3H2O (Scheme 7). On the other hand we
were able to locate four relative minima for the conjugate base
NDI- 3H2O in gas phase (A-D in Figure 6), all exhibiting a
planar structure. Optimization in water bulk was successful
only for the two most stable water complex A and C. In the
two most stable NDI- 3H2O complexes, including three water
molecules, the “anionic cavity” in the heterocyclic structure
(in blue, Scheme 7) tightly binds a coplanar water molecule,
with three H-bonding interactions (Figure 6).
8622 J. Org. Chem. Vol. 74, No. 22, 2009

Doria et al.
JOCArticle
TABLE 2. Electronic Energies (Hartree), in the Gas Phase (Egas), Thermal Correction to Gibbs Free Energy in the Gas Phase (δGgas), Gibbs Free Energy
in the Gas Phase (Ggas), and Aqueous Solution (Gaq) at the PBE0/6-31þG(d,p) Level of Theory with PCM Solvation Models
compd
QX-
QXH
NDI-
NDIH
Egas
δGgas
Ggas
Gaq
-567.300039
-567.8494478
-1360.4827393
-1361.0147537
0.086272
0.100720
0.205033
0.220398
-567.213767
-567.748728
-1360.277706
-1360.794356
-567.398792
-567.875130
-1360.558389
-1361.019470
including specific interactions with 3 H2O
QX- 3H2O
QXH 3H2O
NDIH 3H2O
NDI- 3H2O
-796.4192348
-796.9458311
-1590.092377
-1589.5912688a
-1589.5911693b
-1589.5950532c
-1589.5872889d
0.148900
0.162589
0.278762
0.266906
-796.270335
-796.783242
-1589.813615
-1589.324363a
-796.496536
-796.968522
-1590.115295
-1589.660287a
0.270646
-1589.324407c
-1589.663297c
aConformer A. bConformer B. cConformer C. dConformer D, in Figure 6.
From a geometrical point of view, the H-bonding network
becomes more tight passing from gas phase to water bulk,
with the H-bonding between water and the anion NDI-
shorter in condensed phase than in gas phase. This aspect
suggests that the specific interactions between the water
molecules and the anions 5aA and 5bA are stronger than in
QX- 3H2O, due to an “anionic cavity”, which for its size and
charge binds a water molecule.
the electron-withdrawing nature of the NDI core and the
strong specific solvation by water. Therefore, both the new
NDIs 5a and 5b exist mainly as non fluorescent conjugate
bases in water solution at pH g 1, which exhibit and “anionic
cavity” on the heterocyclic structures. Along this line,
we are currently exploring other NDI analogues exhibiting
two cationic arms to be developed as new G-quadruplex
ligands.
Energy in Gas Phase and in Water Bulk. The energy data in
Table 2 have been used to calculate the free energy for the
proton exchange reactions (a) and (b) depicted in the
Scheme 7, both in the gas phase and in the solvent bulk
(Table 3). These data suggest that the reactions (a) and (b) are
very exoergonic by more than 10 kcal/mol. Using the eq I and
the experimental pKa value (9.27) measured for the reference
acid/base couple QXH/QX-, in water, we have been able to
estimate the pKa of 5a in water solution. The model (a) which
takes into account only the bulkeffect of the solvent suggests a
pKa 1.8. The more refined model (b), which uses proton
exchange between water clusters, including specific interac-
tions, suggests a lower value (pKa 0.1, Table 3). These data
unequivocally explainwhy the anions5aAand 5bAare stillthe
main species in water solution at pH g 1.
Experimental Section
2,3,6,7-Tetrabromo- and 2,6-dibromonaphthalene dianhy-
dride were synthesized via standard published procedures.14,10c
Procedure for the Synthesis of 6a, 7a, and 8a. To a stirred
suspension of 2,3,6,7-tetrabromodianhydride (600 mg, 0.001
mol) in glacial acetic acid (15 mL) was added pentylamine (1.6
mL). After being stirred for 30 min at 130 °C, the reaction
mixture was cooled to room temperature and put into ice to
induce precipitation. The crude orange solid was filtered on a
Hirsch filter and purified by column chromatography (CH2Cl2),
yielding 6a (35%), 7a (26%), and 8a (21%).
Microwave-Assisted Synthesis of 7a and 8a. The reaction was
runina dedicatedmicrowave reactor (CEMDiscovermodel, with
PC control) under atmospheric pressure in an open reaction
vessel. The microwave method was power-controlled (100 W,
power input) to maintain the desired temperature (150 °C). An IR
noncontact sensor was used for temperature measurement of
vessel contents. To a stirred suspension of 2,3,6,7-tetrabromo-
dianhydride (600 mg, 0.001 mol) in glacial acetic acid (25 mL) was
added pentylamine (1.6 mL). After being stirred for 30 min, the
reaction mixture was cooled to room temperature and quenched
in ice to induce a precipitation. The crude yellow solid was filtered
on a Hirsch filter and purified by column chromatography
(CH2Cl2), yielding 7a (14%) and 8a (65%).
TABLE 3. Free Energy for the Proton Exchange Reaction in Scheme 7
in the Gas Phase and in Aqueous Solution with and without Three Water
Molecules [at PBE0/6-31þG(d,p) Level of Theory with PCM Solvation
Models]a
gas phase gas phase þ 3H2O water bulk water bulk þ 3H2O
ΔG
pKa
-11.5
-14.8
-1.6
-10.2
1.8
-12.5
0.1
aThe pKa of the prototype NDIH was computed according to eq I.
N,N0-Dipenthyl-2,3-dibromonaphthalene-1,4,5,8-tetracarboxy-
lic Acid Bisimide (6a). Orange solid. 1H NMR (300 MHz,
CDCl3, 25 °C, TMS): δ 8.98 (s, 2H), 4,20 (t, J = 7.4 Hz, 4H),
1.73 (broad s, 4H), 1.41 (broad s, 8H), 0.92 (t, J = 6.0 Hz, 6H).
13C NMR (300 MHz, CDCl3, 25 °C, TMS): δ 160.61; 138.94;
128.21; 127.61; 125.24; 123.98; 41.46; 29.05; 27.46; 22.25; 13.85.
MS (EI): m/z calcd for C24H24Br2N2O4 562.01, found 564.27.
Anal. Calcd: C, 51.09; H, 4.29; Br, 28.32; N, 4.96; O, 11.34.
Found: C, 51.20; H, 4.22; Br, 28.15; N, 5.00.
N,N0-Dipentyl-2,3,6-tribromonaphthalene-1,4,5,8-tetracarboxy-
lic Acid Bisimide (7a). Pink solid. 1H NMR (300 MHz, CDCl3,
25 °C, TMS): δ 9.01 (s, 1H), 4,20 (t, J = 7.4 Hz, 4H), 1.73 (broad
s, 4H), 1.41 (broad s, 8H), 0.92 (t, J = 6.0Hz, 6H). 13C NMR (300
MHz, CDCl3, 25 °C, TMS): δ 160.64; 151.73; 138.94; 138.19;
128.22; 127.61; 125.23; 123.96; 120.43; 41.48; 29.06; 27.46; 22.27;
13.87. MS (EI): m/z calcd for C24H23Br3N2O4 643.16, found
Conclusion
In summary, this investigation describes the synthesis of
new heterocyclic dihydropyrazinediones starting from both
2,6-dibromonaphthalene and 2,3,6,7-tetrabromonaphtha-
lene bisanhydride by a stepwise protocol including imidiza-
tion, aromatic nucleophilic substitution by ethane-1,2-
diamine, in situ reductive dehalogenation, and oxidation.
These new heteroccyles exhibit a strong green florescence in
organic solvents including DMSO, revealing an unexpected
acidity both in DMSO (pKa 7.6) and in water (pKa < 1).
A computational investigation at PBE0/6-31þG(d,p) level
of theory, using PCM solvation models, predicts for a
model NDI a pKa value close to zero, as a result of both
J. Org. Chem. Vol. 74, No. 22, 2009 8623

Doria et al.
JOCArticle
644.27. Anal. Calcd: C, 44.82; H, 3.60; Br, 37.27; N, 4.36; O, 9.95.
Found: C, 44.80; H, 3.65; Br, 37.32; N, 4.41.
orange solid was filtered on a Hirsch filter and purified by
column chromatography (CH2Cl2), yielding 12a (45%), 13a
(47%), and 12b (11%).
N,N0-Dipentyl-2,3,6,7-tetrabromonaphthalene-1,4,5,8-tetra-
carboxylic Acid Bisimide (8a). Yellow solid. 1H NMR (300 MHz,
DMSO-d6): δ 4,13 (m, 4H), 1.73 (broads, 4H), 1.41 (broad s, 8H),
0.92 (t, J = 6.0 Hz, 6H). MS (EI): m/z calcd for C24H22Br4N2O4
722.06, found 723.05. Anal. Calcd: C, 39.92; H, 3.07; Br, 44.26; N,
3.88; O, 8.86. Found: C, 40.00; H, 3.15; Br, 44.28; N, 3.91.
Thermal Procedure for a Ring-Closure Nucleophilic Substitu-
tion. Reaction Conditions (a). The reactant 8a (150 mg, 0.0002 mol)
was heated under argon in DMF with ethane-1,2-diamine (48 mg,
0.0008 mol) at 135 °C for 30 min. After this period, the solution
became brownish and then was quenched in ice to induce precipi-
tation of the product. The crude was purified by column chroma-
tography (Cy/AcOEt = 8:2) to yield 9a (3%) and 4a (86%).
Reaction Conditions (b). The reactant 8a (150 mg, 0.0002 mol)
was heated under argon in ethane-1,2-diamine (15 mL) at 135 °C
for 30 min. After this period, the solution become brownish and
then was cooled to room temperature, water was added (50 mL),
and the solution was extracted with CH2Cl2 (3 ꢀ 50 mL). The
combined organic phases were washed with 1 N HCl. The
solvent was then removed under vacuum and purified by
column chromatography (CHCl3/MeOH 9:1), yielding 9a
(24%), 4a (27%), and 11a (25%).
N,N0-Dipentyl-2,6-dibromonaphthalene-1,4,5,8-tetracarboxy-
lic Acid Bisimide (12a). Yellow solid. 1H NMR (300 MHz,
CDCl3, 25 °C, TMS): δ 9.02 (s, 2H), 4,21 (t, J = 7.6 Hz, 4H),
1.76 (broad s, 4H), 1.42 (broad s, 8H), 0.94 (t, J = 6.8 Hz, 6H).
13C NMR (300 MHz, CDCl3, 25 °C, TMS): δ 160.63; 138.93;
128.19; 127.61; 125.25; 124.00; 41.45; 29.04; 27.45; 22.23; 13.81.
MS (EI): m/z calcd for C24H24Br2N2O4 562.01, found 564.27.
Anal. Calcd: C, 51.09; H, 4.29; Br, 28.32; N, 4.96; O, 11.34.
Found: C, 51.01; H, 4.31; Br, 28.41; N, 5.02.
N,N0-Dicyclohexyl-2,6-dibromonaphthalene-1,4,5,8-tetracar-
boxylic Acid Bisimide (12b). Yellow solid. 1H NMR (300 MHz,
CDCl3, 25 °C, TMS): δ 8.7 (s, 2H), 5.09 (m, 2H), 2.63-2.48 (m,
4H), 1.93 (m, 4H), 1.75 (m, 6H), 1.45-1.27 (m, 6H). 13C NMR
(300 MHz, CDCl3, 25 °C, TMS): δ 163.19; 130.70; 126.79; 54.34;
29.59; 28.99; 26.37; 25.22. MS (EI): m/z calcd for C26H24-
Br2N2O4 588.22, found 588.31. Anal. Calcd: C, 53.08; H, 4.11;
Br, 27.16; N, 4.76; O, 10.88. Found: C, 53.15; H, 4.10; Br, 27.21;
N, 4.74.
N,N0-Dipentyl-2-bromonaphthalene-1,4,5,8-tetracarboxylic
Acid Bisimide (13a). White solid. 1H NMR (300 MHz, CDCl3,
25 °C, TMS): δ 8.96 (s, 1H), 8.81 (AB system, 2H), 4,20 (m, 4H),
1.76 (broad s, 4H), 1.43 (broad s, 8H), 0.94 (m, 6H). MS (EI): m/
z calcd for C24H25BrN2O4 485.37, found 486.45. Anal. Calcd: C,
59.39; H, 5.19; Br, 16.46; N, 5.77; O, 13.19. Found: C, 59.28; H,
5.15; Br, 16.42; N, 5.82.
Reaction Conditions (c), Microwave-Assisted in a Closed
Vessel. Compound 8a (150 mg, 0.0002 mol) was heated in a
closed vessel in DMF with ethane-1,2-diamine (48 mg, 0.0008
mol) at 170 °C for 10 min, 200 psi, with a power of 200 W. After
this period, the brown solution was cooled to room temperature,
and water was added (50 mL) to induce precipitation. The crude
product was filtered and purified by column chromatography
(CHCl3/MeOH 9:1) to yield 9a (35%), 4a (16%), and 11a (7%).
Reaction Conditions (d). The title compound was placed in a
flask containing ethane-1,2-diamine (15 mL), and the mixture was
stirred at rt for 16 h under argon. The resulting red mixture was
poured in HCl (1 N, 100 mL). The solid orange solid was filtered and
washed with water. Further purification by column chromatogra-
phy (CHCl3/Cy 6:4) gave pure 4a (5% yield) and 10a (72%yield).
N,N0-Dipentylnaphthalene-1,4,5,8-tetracarboxylic Acid Bis-
imide (9a). Gray solid. 1H NMR (300 MHz, CDCl3, 25 °C,
TMS): δ 8.75 (s, 1H), 4,19 (t, J = 7.5 Hz, 4H), 1.75 (broad s, 4H),
1.42 (broad s, 8H), 0.96 (t, J = 6.0 Hz, 6H). 13C NMR (300
MHz, CDCl3, 25 °C, TMS): δ 162.70; 133.01; 131.07; 130.79;
126.50; 40.83; 29.05; 27.62; 22.27; 13.84. MS (EI): m/z calcd. for
C24H26N2O4 406.47, found 406.50. Anal. Calcd: C, 70.92; H,
6.45; N, 6.89; O, 15.74. Found: C, 70.95; H, 6.51; N, 6.82.
N,N0-Dipentyl-5,8,9,10-(1,4-diaza-1,2,3,4-tetrahydroanthra-
N,N0-Dipentyl-2-(20-aminoethylamino)-6-bromo-1,4,5,8-naph-
thalenetetracarboxylic Acid Bisimide (14a). Compound 12a (200
mg, 0.35 mmol) was placed in a flask containing EDA (15 mL),
and the mixture was stirred at rt for 2 h under argon. The
resulting red mixture was washed with a solution of NaHCO3,
and extracted, with CHCl3 (3 ꢀ 100 mL). The organic phases
were combined, washed with water, and purified by column
chromatography (CHCl3) yielding 14a as a red solid (50%,
1
yield). H NMR (300 MHz, CDCl3, 25 °C, TMS): δ 10.31 (s,
1H), 8.82 (s, 1H), 8.26 (s, 1H), 4.19 (m, 4H), 3.66 (t, J = 5.8 Hz,
2H), 3.19 (t, J = 5.8 Hz, 2H), 1.74 (broad s, 4H), 1.42 (broad s,
8H), 0.94 (m, 6H). MS (EI): m/z calcd for C26H31BrN4O4
543.45, found 543.52. Anal. Calcd: C, 57.46; H, 5.75; Br,
14.70; N, 10.31; O, 11.78. Found: C, 57.41; H, 5.76; Br, 14.71;
N, 10.36.
Synthesis of 10a and 10b according synthetic procedure (d):
The procedure (d), descried above, was applied to compound
12a, 12b (0.35 mmol) yielding: 10a (40%), 4a (52%) and 10b
(33%), 4b (60%), respectively.
1
cene)tetracarboxylic Acid Bisimide (4a). Yellow-gold solid. H
N,N0-Dipentyl-6-bromo-5,8,9,10-(1,4-diaza-1,2,3,4-tetrahy-
droanthracene)tetracarboxylic Acid Bisimide (10a). 1H NMR
(300 MHz, CDCl3, 25 °C, TMS): δ 11.13 (s, 1H), 10.73 (s, 1H),
8.68 (s, 1H), 4.27 (m, 4H), 3.91 (s, 4H), 1.70 (broad s, 4H), 1.45
(broad s,8H), 0.95 (m, 6H). 13C NMR (300 MHz, CDCl3, 25 °C,
TMS): δ 166.11; 165.55; 162.22; 161.30; 144.33; 143.36; 131.56;
125.04; 122.16; 121.41; 119.93; 40.69; 40.28; 38.16; 37.96; 29.26;
29.14; 27.57; 27.49; 22.36; 13.93; 13.89. MS (EI): m/z calcd for
C26H29BrN4O4 541.44, found 542.50. Anal. Calcd: C, 57.68; H,
5.40; Br, 14.76; N, 10.35; O, 11.82. Found: C, 57.42; H, 5.51; Br,
14.72; N, 10.28.
NMR (300 MHz, CDCl3, 25 °C, TMS): δ 10.60 (s, 2H), 8.25 (s,
2H), 4.15 (m, 4H), 3.80 (s, 4H), 1.70 (broad s, 4H), 1.45 (broad s,
8H), 0.95 (m, 6H). 13C NMR (300 MHz, CDCl3, 25 °C, TMS): δ
166.27; 163.28; 143.72; 130.75; 126.25; 124.48; 122.2; 40.18;
38.07; 29.20; 27.63; 22.36; 13.89. MS (EI): m/z calcd for
C26H30N4O4 462.23, found 462.54. Anal. Calcd: C, 67.51; H,
6.54; N, 12.11; O, 13.84. Found: C, 67.42; H, 6.55; N, 12.32.
N,N0-Dipentyl-5,6,11,12-(1,4,7,10-tetraaza-1,2,3,4,7,8,9,10-
octahydroanthracene)tetracarboxylic Acid Bisimide (11a). Violet
solid. 1H NMR (300 MHz, CDCl3, 25 °C, TMS): δ 10.79 (s, 4H),
4,15 (broad s, 4H), 3.73 (m, 8H), 1.73 (broad s, 4H), 1.43 (broad
s, 8H), 0.94 (broad s, 6H). MS (EI): m/z calcd for C28H34N6O4
518.61, found 519.70. Anal. Calcd: C, 64.85; H, 6.61; N, 16.20;
O, 12.34. Found: C, 65.02; H, 6.65; N, 16.12.
General Thermal Procedure for the Synthesis of 12a and 13a
According the Second Synthetic Protocol (See Scheme 3). To a
stirred suspension of dibromodianhydride (600 mg, 0.001 mol)
in glacial acetic acid (15 mL) was added pentylamine (1.6 mL).
After being stirred for 30 min at 130 °C, the reaction mixture was
cooled to room temperature and quenched into ice. The crude
N,N0-Dicyclohexyl-6-bromo-5,8,9,10-(1,4-diaza-1,2,3,4-tetra-
hydroanthracene)tetracarboxylic Acid Bisimide (10b). Orange
solid. 1H NMR (300 MHz, CDCl3, 25 °C, TMS): δ 11.0 (s,
1H), 10.60 (s, 1H), 8.60 (s, 1H), 5.09 (m, 2H), 3.79 (s, 4H),
2.63-2.51 (m, 4H), 1.91 (m, 4H), 1.75 (m, 6H), 1.45-1.27 (m,
6H). MS (EI): m/z calcd for C28H29BrN4O4 465.46, found
466.54. Anal. Calcd: C, 59.47; H, 5.17; Br, 14.13; N, 9.91; O,
11.32. Found: C, 59.41; H, 5.20; Br, 14.11; N, 10.01.
Mild Reductive Dehalogenation Yielding 4a (See Scheme 3).
The reactant 10a (0.07 mmol) was suspended in a solution of
8624 J. Org. Chem. Vol. 74, No. 22, 2009

Doria et al.
JOCArticle
34 mL of DMSO and 6 mL of water, degassed under nitrogen
atmosphere, and stirred at room temperature. Sodium dithio-
nite (0.6 mmol) was added, and the new mixture was stirred for 2
h at 40 °C. After this period, the reaction the excess of sodium
dithionite was quenched by oxygen bubbling. After few minutes
the precipitation of the product occurred. The resulting suspen-
sion was filtered and purified by column chromatography
(CHCl3/Cy 6:4) giving 4a as a yellow solid (92%, yield).
N,N0-Dicyclohexyl-5,8,9,10-(1,4-diaza-1,2,3,4-tetrahydro-
anthracene)tetracarboxylic Acid Bisimide (4b). Yellow-gold so-
lid. 1H NMR (300 MHz, CDCl3, 25 °C, TMS): δ 10.68 (s, 2H),
8.33 (s, 2H), 5.09 (m, 2H), 3.79 (s, 4H), 2.63-2.51 (m, 4H), 1.93
(m, 4H), 1.75 (m, 6H), 1.45-1.27 (m, 6H). 13C NMR (300 MHz,
CDCl3, 25 °C, TMS): δ 167.18; 163.95; 143.92; 124.71; 122.76;
122.40; 53.57; 38.14; 29.00; 26.51; 25.41. MS (EI): m/z calcd
for C28H30N4O4 486.56, found 486.23. Anal. Calcd: C, 69.12;
H, 6.21; N, 11.51; O, 13.15. Found: C, 69.18; H, 6.26; N,
11.42.
minutes, an orange solid began to precipitate. After 10 min, the
solid was filtered on a Hirsh filter (yield: 98%). 1H NMR (300
MHz, CDCl3, 25 °C, TMS): δ 8.87 (s, 2H), 4.46 (s, 6H), 4.27 (m,
4H), 1.78 (broad s, 4H), 1.44 (broad s,8H), 0.95 (m, 6H). MS
(EI): m/z calcd for C28H30N4O6 518.56, found 519.7 (64), 504
(100), 490 (10). Anal. Calcd: C, 64.85; H, 5.83; N, 10.80; O,
18.51. Found: C, 64.94; H, 5.72; N, 10.72.
Computational Details. All calculations were carried out using
the Gaussian 0327 program package for gas phase and solvent
optimization. All the geometric structures of NDIH, QXH,
NDI-, QX-, and their water clusters NDIH 3H2O, QXH
3H2O, NDI- 3H2O, QX- 3H2O, including the four isomeric
clusters A-D, were fully optimized in the gas phase and in water
solution using the hybrid density functional method PBE0 with
the 6-31þG(d,p) basis set. PBE0 functional (also referred to as
PBE1PBE) is a combination of the exact exchange (25%) with
the Perdew-Burke-Ernzerhof exchange (PBE1) and correla-
tion functionals (PBE). It has been extensively used by Barone
for predicting reactivity in aqueous solution using PCM solva-
tion models.25 The bulk solvent effects on the geometries and
energies of the reactants were calculated via the self-consistent
reaction field (SCRF) method using the PCM as implemented in
the B.05 version of Gaussian 03. The cavity is composed by
interlocking spheres centered on non-hydrogen atoms with radii
obtained by the HF parametrization of Barone known as the
united atom topological model (UAHF).26 Such a model in-
cludes the nonelectrostatic terms (cavitation, dispersion, and
repulsion energy) in addition to the classical electrostatic con-
tribution.
2,7-N,N0-Dipentyltetraazabenzo[e]pyrene-1,3,6,8,10,11-hexa-
one (5a). The reactant 4a (150 mg, 0.0003 mol) was dissolved in
30 mL of H2O, 5 mL of concd H2SO4, and 5 mL of acetone. Then
K2Cr2O7 (200 mg, 0.0007 mol) was added to the solution. The
brown mixture was heated at 100 °C for 1 h. After this period,
the solution became clear, and then an other equal amount of
K2Cr2O7 was added and the solution was refluxed for other 2 h.
The solution was cooled to room temperature, quenched in ice,
neutralized with solid Na2CO3, and extracted with CHCl3 (3 ꢀ
50 mL). The combined organic phases were washed with brine,
and the solvent was removed under vacuum. The crude product
collect was purified by column chromatography (CHCl3) as a
yellow solid or by preparative HPLC using CH3CN/H2 = 1:1 as
eluent (yield: 75%). 1H NMR (300 MHz, CDCl3, 25 °C, TMS): δ
13.31 (s, 2H), 8.78 (s, 2H), 4.24 (m, 4H), 1.79 (broad s, 4H), 1.43
(broad s,8H), 0.95 (m, 6H). 13C NMR (300 MHz, CDCl3, 25 °C,
TMS): δ 165.11; 161.80; 152.31; 132.82; 130.30; 125.50; 122.81;
105.94; 41.00; 29.00; 27.43; 22.22; 13.82. MS (EI): m/z calcd for
C26H26N4O6 490.51, found 491.6 (100.0), 463 (55.1), 393 (48.8),
323 (60.7). Anal. Calcd: C, 63.66; H, 5.34; N, 11.42; O, 19.57.
Found: C, 63.62; H, 5.38; N, 11.58.
Acknowledgment. Financial support from the Italian
Ministry of University and Research (MIUR, FIRB-Ideas
Project RBID082ATK_003) and the Italian Association for
Cancer Research (Associazione Italiana per la Ricerca sul
Cancro, or AIRC Grant No. 5826) is gratefully acknowl-
edged. We are indebted to Dr. Lorenzo Mosca for helpful
support with the pKa measurements.
2,7-N,N0-Dicyclohexyltetraazabenzo[e]pyrene-1,3,6,8,10,11-
Supporting Information Available: UV/vis spectra of 5b. 1H
NMR and 13C NMR for the NDIs: 4a,b, 5a,b, 6a-9a, 10a,b,
11a, 12a,b, and -16a. Cartesian coordinates, energies (in
hartrees) and dipole moments (in debyes) for the structures
NDIH, QXH, NDI-, QX-, and the water clusters NDIH 3H2O,
QXH 3H2O, NDI- 3H2O, QX- 3H2O in the gas phase, and in
water solution, at the PBE0/6-31þG(d,p) and PBE0-PCM/6-
31þG(d,p) levels of theory. This material is available free of
charge via the Internet at http://pubs.acs.org.
1
hexaone (5b). Yellow solid. Yield: 64%. H NMR (300 MHz,
CDCl3, 25 °C, TMS): δ 13.48 (s, 2H), 8.85 (s, 2H), 5.17 (m, 2H),
2.67-2.60 (m, 4H), 2.08-2.04 (m, 4H), 1.90-1.86 (m, 6H),
1.45-1.27 (m, 6H). 13C NMR (300 MHz, CDCl3, 25 °C, TMS):
δ 164.67; 164.14; 152.49; 132.82; 130.44; 125.50; 122.81; 105.94;
55.06; 29.71; 28.98; 26.45; 25.30. MS (EI): m/z calcd for
C28H26N4O6 514.53, found 515.54. Anal. Calcd: C, 65.36; H,
5.09; N, 10.89; O, 18.66. Found: C, 65.52; H, 5.02; N, 10.93.
Synthesis of 15a. Compound 5a (0.030 g, 0.1 mmol) was
dissolved in CHCl3 (15 mL), and then a portion of CH2N2
(0.05 mmol; Et2O 1 mL) was added to the stirred solution. The
solution become yellow. Five milliliters MeOH was added after 5
min, and the solvent was evaporated under vacuum. The crude
product was purified by reversed-phase column chromatography
(27) Gaussian 03, revision D.02: Frisch, M. J.; Trucks, G. W.; Schlegel, H.
B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.;
Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.;
Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.;
Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda,
R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.;
Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.;
Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski,
V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.;
Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.;
Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith,
T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill,
P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A.
Gaussian, Inc.: Wallingford, CT, 2004.
1
(CH3CN/H2O = 1.1, pH 4) as a yellow solid (yield: 69%). H
NMR (300 MHz, CDCl3, 25 °C, TMS): δ 13.47 (s, 1H), 8.86 (d,
J = 7.6; 2H), 8.80(d, J = 7.6;2H), 4.42 (s, 3H), 4.25 (m, 4H), 1.78
(broad s, 4H), 1.44 (broads,8H), 0.95 (m, 6H). MS (EI): m/z calcd
for C27H28N4O6 504.53, found 505.45. Anal. Calcd: C, 64.27; H,
5.59; N, 11.10; O, 19.03. Found: C, 64.55; H, 5.42; N, 11.18.
Synthesis of 16a. Compound 15a (0.030 g, 0.1 mmol) was
dissolved in CHCl3 (15 mL) and then to the stirring solution was
added a portion of CH2N2 (0.2 mmol; Et2O 7 mL). After a few
J. Org. Chem. Vol. 74, No. 22, 2009 8625