The heterocyclic ring of 4-substituted-1,8-naphthalimides is NOT inert to nucleophilic attack, contrary to earlier reports

Article abstract of DOI:10.1039/c3ob40639c

The heterocyclic ring of N-aryl-4-chloro-1,8-naphthalimides, reported to be resistant to nucleophilic attack, reacts with primary amine nucleophiles at room temperature to give 4-chloro-N-alkyl-1,8-naphthalimides. The reaction is first order in the naphth

Full text of DOI:10.1039/c3ob40639c

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
The heterocyclic ring of 4-substituted-1,8-
naphthalimides is NOT inert to nucleophilic attack,
contrary to earlier reports†
Cite this: Org. Biomol. Chem., 2013, 11,
4390
Ryan K. McKenney, Leah L. Groess, Kyle M. Kopidlansky, Kelsey L. Dunkle and
David E. Lewis*
Received 30th March 2013,
Accepted 20th May 2013
The heterocyclic ring of N-aryl-4-chloro-1,8-naphthalimides, reported to be resistant to nucleophilic attack,
reacts with primary amine nucleophiles at room temperature to give 4-chloro-N-alkyl-1,8-naphthalimides.
The reaction is first order in the naphthalimide. The Hammett plot is linear (R2 0.996) with a large positive
slope (+3.0), consistent with substantial negative charge development at nitrogen in the activated complex.
DOI: 10.1039/c3ob40639c
www.rsc.org/obc
become accepted dogma that the heterocyclic ring of the
4-substituted-1,8-naphthalimides, once formed, is unreactive
Introduction
The 4-amino-1,8-naphthalimide fluorophore has been widely towards nucleophilic and electrophilic reagents alike. We are
used in a variety of applications, including fluorescent sensing aware of no reports of nucleophilic displacement in the hetero-
of cations1 and anions,2 fluorescent logic gates and switches,3 cyclic ring of these compounds.
selective staining of live cells,4 and photochemical welding of
In contrast to the lack of reactivity of the heterocyclic ring, a
tissues with visible light.5 Part of the importance of this fluoro- halogen or nitro group at the 4-position of the N-alkyl-1,8-
phore arises from its resistance to photochemical bleaching, naphthalimide nucleus has been found to be readily displaced
and part to its chemical stability. In fact, the resistance of the by amine nucleophiles, providing a convenient entry into the
ring system towards nucleophilic attack has become a defining highly fluorescent 4-amino-1,8-naphthalimide system.7,8
characteristic of this important fluorophore—in his synthesis Similar copper-catalyzed displacement of the halogen by meth-
of Lucifer Yellow anhydride from Brilliant Sulfoflavin (1), oxide anion in a series of N-aryl-4-bromo-1,8-naphthalimides
Stewart6 established the conventional wisdom that the hetero- has given the corresponding 4-methoxy compounds, without
cyclic ring of N-arylnaphthalimides is inert to nucleophilic loss of the N-aryl group.9 Thus, confirming Stewart’s impli-
attack except under forcing conditions, or under conditions of cations, it has been general experience that the imide ring,
unusual electron deficiency, with the following statement: “A once formed, is resistant to attack by heteroatom nucleophiles:
striking property of 1 is its extreme resistance to hydrolysis. N-alkyl-4-chloro-1,8-naphthalimides have been found to react
Prolonged reflux of 1 with aqueous NaOH or H2SO4 failed to with amines in large excess to give products only of displace-
give the diacid or the corresponding anhydride”.
ment of the halogen. To our knowledge, Stewart’s synthesis of
Lucifer Yellow anhydride represents the only published
example of successful nucleophilic attack on the heterocyclic
ring of the 4-amino-1,8-naphthalimide nucleus, and in the
reactive compound the naphthalimide ring and the N-aryl
group each carried a pair of strongly electron-withdrawing sul-
fonyl groups. Because of the apparent high stability of the
heterocyclic ring system and its resistance towards nucleophi-
lic attack, reinforcing Stewart’s observations, there has been
little incentive to study its reactions with nucleophiles, with
the result that there has been no systematic study, to date, of
these reactions.
As a consequence of Stewart’s observation, and the experience
of three decades of synthetic endeavors in this area,7 it has
In the course of other work, we needed a sulfonylaziridine,
which we proposed to prepare by standard methods from the
corresponding sulfonamide (4) by the approach summarized
Department of Chemistry, University of Wisconsin-Eau Claire, Eau Claire, WI, USA.
E-mail: lewisd@uwec.edu; Fax: +1-715-836-4979; Tel: +1-715-836-4744
†Electronic supplementary information (ESI) available: 1H NMR and 13C NMR
spectra of imides 3, 7a–d, and 11; data from kinetic runs; Cartesian coordinates
of the complex 10. See DOI: 10.1039/c3ob40639c
in Fig. 1. We prepared the N-aryl-4-chloroimide
3 by
heating sulfanilamide and 4-chloro-1,8-naphthalic anhydride
4390 | Org. Biomol. Chem., 2013, 11, 4390–4396
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
Fig. 2 Synthesis of 4-chloro-N-aryl-1,8-naphthalimides and their reactions with
butylamine.
we found that every 4-chloroimide underwent some degree of
ring cleavage with butylamine at 4 °C to give N-butyl-4-chloro-
1,8-naphthalimide, although the reaction rate did vary widely
with the phenyl substituent; displacement of the chloride
became the dominant reaction when the substituent on the
N-phenyl ring became electron-releasing. Thus, we suggest
that in the presence of even a mildly electron-withdrawing sub-
stituent at the 4-position (the Hammett σp constant for chlo-
rine is 0.23, compared to 0.35 for the sulfonate group and 0.78
for the nitro group10), an N-aryl group is sufficient to render
attack of an amine nucleophile at the imide carbonyl group
preferred.
Fig. 1 Attempted synthesis of the 4-alkylamino-N-aryl-1,8-naphthalimide gives
unexpected products of substitution in the heterocyclic ring.
(2) in o-dichlorobenzene or benzonitrile, and subjected the
product to heating in hexylamine in an attempt to effect its
transformation into the aziridine precursor 4. However, the
only product isolated from this reaction was N-hexyl-4-hexyl-
amino-1,8-naphthalimide, 5. Even at room temperature, the
reaction gave 4-chloro-N-hexyl-1,8-naphthalimide (6), which
showed that the substitution of the aromatic amine is faster
than the displacement of the halogen.
In order to quantify the effects of the phenyl substituent,
we determined the kinetics of the displacement of the aro-
matic amine by butylamine using the nucleophile as solvent;
under these conditions, all reactions exhibited excellent
pseudo-first order kinetics. Solutions that varied in concen-
tration from 0.06 to 0.6 mM were kept at 25.00 0.02 °C, and
aliquots were removed from the reaction flask and quenched
These results were entirely unexpected, given three decades
of experience that the heterocyclic ring is inert to attack by
1
with 8 M hydrochloric acid prior to extractive work-up and H
nucleophiles. Thus—unexpectedly—the combination of
a
NMR analysis. The work-up removes both the butylamine
solvent and the aniline produced, leaving a mixture of imides
for 1H NMR analysis.
4-chloro substituent on the naphthalimide ring and an
electron-withdrawing substituent on the N-aryl ring is clearly
sufficient for heterocycle cleavage to occur with primary amine
nucleophiles. This then raised the question of just what level
of electron deficiency is required for the displacement of the
aniline from a 4-substituted-N-aryl-1,4-naphthalimide, and just
how general this reaction is.
The 1H NMR spectra of these compounds (Fig. 3) are
especially suited for kinetic studies because the resonances of
the naphthalimide ring system are relatively insensitive to
changes in substituent at the heterocyclic nitrogen. The proton
at position 2 of the naphthalimide ring in chloroimides 7 and
chloroimide 3, regardless of the N-aryl substituent, resonates
at a chemical shift close to δ 8.5, well separated from the rest
of the spectrum, and sufficiently far from the resonance of the
same proton in chloroimide 8 (always slightly upfield from the
starting imide) to permit its use in quantifying the amount of
unreacted starting material in most samples. At the same
time, the protons at positions 5 and 7 resonate as a complex
multiplet or pair of doublets in the region δ 8.6–8.8 in the
spectra of all 4-chloro-1,8-naphthalimides (both N-aryl and
N-butyl) that we have prepared; although this multiplet is
Results and discussion
To assess the requirement for electron-withdrawing groups on
the N-phenyl substituent, we prepared a series of N-aryl-4-
chloro-1,8-naphthalimides carrying both electron-withdrawing
and electron-releasing substituents on the phenyl ring by stan-
dard methods from 4-chloro-1,8-naphthalic anhydride (Fig. 2).
The resultant 4-chloro-N-aryl-1,8-naphthalimides were then
treated with butylamine at 4 °C. Contrary to our expectations,
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 4390–4396 | 4391

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 3 1H NMR resonances used in monitoring kinetics (taken from compound
7d).
Fig. 4 Hammett plot for the loss of the substituted N-phenyl group of the N-
aryl-4-chloro-1,8-naphthalimides.
sometimes difficult to deconvolute in the mixture of reactant
and product, it can be used as an internal integration
standard.
The kinetics of the reaction were followed by measuring the
disappearance of the resonance of the signal of H-5 of imide 7
(or imide 3). All reactions were run in triplicate. The reproduci-
bility of the peak integrations was tested by measuring the
spectrum of a standard set of samples in quintuplicate. These
samples showed less than 0.1% variation in the value of the
rate constant and the intercept for the same samples, indicat-
ing that there is no substantial error introduced by this
method of measurement. The results of the kinetic measure-
ments are gathered in Table 1 (errors are from weighted least
squares at the 95% confidence level).
The half-lives of the reactions increased from ca. 5 hours
when the substituent was m-Cl, to more than two days when
the phenyl ring was unsubstituted. The chloroimide 3 exhibi-
ted a strong propensity to form a solution from which the
starting imide would precipitate at different times during the
reaction, making any kinetic measurements for this com-
pound unreliable; this compound was therefore omitted from
the reaction rate analysis. The reactions of compounds carry-
ing groups less strongly electron-withdrawing than m-methoxy
(σp = 0.12) become very difficult to follow, because displace-
ment of the chlorine becomes the dominant reaction. This
problem required that we use a different resonance to monitor
the reaction of the parent compound, but fortunately, the res-
onance of the C-6 proton of the product is well separated from
the rest of the resonances. For the same reason, we have
restricted this study to substituents with 0.37 > σ > 0.
Fig. 5 Rate-determining step of the substitution of N-aryl-4-chloro-1,8-
naphthalimides by butylamine.
more scatter in the plots for the p-bromo and m-methoxy com-
pounds, which we attribute to the much lower solubility of
these two compounds in the reaction medium. The Hammett
plot is linear (R2 = 0.996), with a slope of +3.0 1.2, which is
comparable to the Hammett ρ values for Hofmann elimin-
ations of quaternary ammonium ions, where computations
suggest the development of a formal negative charge of 0.3e−
at the β carbon.5 This is consistent with an increase in the elec-
tron density at the heterocyclic nitrogen as the reaction pro-
ceeds, with the rate determining step of the reaction being the
opening of the heterocyclic ring of 9, formed in a pre-equili-
brium involving addition of butylamine to the imide carbonyl
group (Fig. 5).
The large slope of the Hammett plot may also reflect the
ability of the lone pair on nitrogen to conjugate with the aro-
matic ring as the ring opens; it is well known that primary
alkyl groups bonded to the heterocyclic nitrogen adopt a con-
formation where the group is orthogonal to the naphthalimide
ring system,7 and substituents on an N-phenyl substituent do
not interact with the fluorophore.11 An orthogonal N-aryl
group would be incapable of conjugating with the lone pair on
the imide nitrogen, but rotation about the C–N bond as the
ring opens (as in 10) would allow improved conjugation of
the electron density on nitrogen with the phenyl ring. Thus,
the large positive ρ value may, in part, reflect this confor-
mational change in the transition state.
The pseudo-first order rate constants for reactants with sub-
stituents having 0.37 > σ > 0 give the linear Hammett plot in
Fig. 4. It is worthwhile noting that there was considerably
Table 1 Pseudo-first order rate constants for the reaction of imides 7 with
n-butylamine at 25.00 °C
R
Rate constant (min−1
)
R2
m-Cl
p-Br
m-OMe
H
2.02 0.04 × 10−3
7.9 1.2 × 10−4
3.21 0.6 × 10−4
1.64 0.07 × 10−4
0.995
0.994
0.956
In an effort to quantify some of the structural changes
0.999 occurring during this rate-limiting step, we have carried out
4392 | Org. Biomol. Chem., 2013, 11, 4390–4396
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
computations, at the MP3/6-31G* level, of the energy profile of
intermediate 10 as a function of the C–N bond distance. The
geometry of the activated complex was then optimized at the
MP3/6-311+G** level.‡ These computations reveal that
the energy of the system passes through a maximum when the
C–N distance is close to 2.02 Å to give the activated complex
model shown in Fig. 6. The major changes in bond lengths of
the groups around the heterocyclic ring are also summarized
in Fig. 6 for the optimized reactant and product for the ring
cleavage. These bond changes are as expected for the for-
mation of the two amide groups: the C–OH bond shortens sig-
nificantly, from 1.424 to 1.313 Å, as does the C–NHBu bond
(from 1.448 to 1.400 Å). The bond between the heterocyclic
nitrogen and the phenyl ring shortens by 0.04 Å, and, at the
same time, the dihedral angle between the N-phenyl ring and
the carbonyl group of the incipient amide anion decreases
from 92° to 43°. These changes are consistent with 10 as a
reasonable model for the activated complex for this reaction.
We recognize that these computations omit the hydrogen
bonding between 10 and the solvent, and that care must there-
fore be exercised not to over-interpret the results. Nevertheless,
it is gratifying to see that the results of the computations are
consistent with the effects implied by the Hammett plot.
The fact that the reaction occurs for N-aryl compounds, but
not for N-(primary)alkyl compounds reaction may be due to
either steric or electronic factors. The question of whether or
not the aryl substituent is actually necessary for the reaction to
occur, or whether any sterically bulky alkyl substituent bonded
to nitrogen would suffice, was settled by preparing the N-cyclo-
Fig. 6 Optimized structure (MP3/6-311+G**) of the activated complex for
heterocyclic ring cleavage in the butylamine adduct of N-phenyl-4-chloro-1,8-
naphthalimide.
hexyl analogue (11) and treating it with butylamine. Only one lone pair and capable of conjugating with the imide carbonyl
reaction was observed: displacement of the halogen—there groups suppresses substitution of the heterocyclic nitrogen;
was no evidence for substitution of the heterocyclic nitrogen. this would account for Stewart’s failure to observe hydrolysis
Thus, we conclude that the N-aryl group is essential for substi- in N-aryl-4-amino-1,8-naphthalimides.
tution of the heterocyclic nitrogen. We suggest that this may
Our efforts to determine the kinetic order of the reaction
be due to the additional stabilization of the incipient amide with respect to butylamine have been complicated by the
anion of 10 by the aromatic ring; similar stabilization is not effects of solvent on the rate of the reaction, which are un-
available to saturated alkyl analogues.
expectedly complex. Thus, while the reactions of 7a with butyl-
In similar vein, we sought to define the required properties amine using dichloromethane as co-solvent (where hydrogen
of the substituent at the 4-position by preparing 4-dimethyl- bonding to the co-solvent is not possible), the observed rate
amino-N-(4-aminosulfonylphenyl)-1,8-naphthalimide (12), and constant decreases with decreasing concentration of the butyl-
treating it with butylamine. The synthesis of the imide, which amine, but the large amount of scatter in the values obtained
could not be prepared by direct displacement, was effected does not allow the kinetic order with respect to the amine to
from the 4-chloronaphthalimide by a modification of Qian’s be determined. In methanol (which is a strongly hydrogen-
procedure12 for the synthesis of 4-dimethylamino-1,8-naphtha- bonding cosolvent) on the other hand, changing from butyl-
lic anhydride. The reaction provided the 4-dimethylamino-N- amine (concentration 10 M) to 5 M butylamine in methanol
aryl-1,8-naphthalimide in modest yield. The reaction of this leads to an increase in the rate constant of the reaction despite
compound with butylamine was extremely slow, and it pro- the decreased nucleophile concentration. We are exploring
vided only the 4-butylamino derivative 13. Thus, we conclude these solvent effects, and the effects of the structure of the
that an electron-releasing group at the 4-position bearing a primary amine nucleophile, further.
‡The calculations were originally carried out using an unrestricted (open shell)
Conclusions
calculation. A reviewer pointed out that this system should be a closed-shell
system, so the computations were repeated as the closed-shell calculations speci-
In conclusion, we report an unprecedented, facile substitution
fied in the body of the text. The changes did not materially affect the outcome of
in the heterocyclic ring of N-aryl-4-chloro-1,8-naphthalimides
the calculations, with minor differences in equilibrium geometry and energy
being noted.
by primary amines, in contrast to the conventional wisdom in
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 4390–4396 | 4393

View Article Online
Paper
Organic & Biomolecular Chemistry
this system. The aromatic ring on the heterocyclic nitrogen and recrystallized from acetic acid. The purity of each com-
appears to be essential for this reaction, and the effects are not pound used in the kinetic runs was analyzed by HPLC.
simply steric in origin. Electron-donating substituents carrying
a lone pair at the 4-position completely suppress the reaction: (7d). 4-Chloro-1,8-naphthalic
6-Chloro-2-phenyl-1H-benz[de]isoquinolin-1,3(2H)-dione
anhydride (3.0147 g,
4-dimethylamino-N-aryl-1,8-naphthalimides undergo only slow 12.96 mmol) was dissolved in aniline (15 mL, 15.0262 g), and
substitution at the 4-position, and no substitution in the the solution was stirred 24 h under reflux. The solution was
heterocyclic ring is observed. The Hammett plot for the reac- poured into 10% HCl (≈ 200 mL), and the solid that precipi-
tion (σ > 0) is linear with a large positive slope, consistent with tated was collected by vacuum filtration and recrystallized
the generation of a substantial partial negative charge on nitro- from acetic acid, using decolorizing carbon, to give pure
gen, or much improved interaction between the phenyl ring 6-chloro-2-phenyl-1H-benz[de]isoquinolin-1,3(2H)-dione as
a
and the lone pair on nitrogen, in the rate-determining step, a pale grey solid (1.8078 g, 45%). HPLC analysis of this material
conclusion supported by computational results at the B3LYP/ showed it to be >98% pure. Material for use in the kinetics
6-31G* level of theory.
experiments was obtained by a second recrystallization of this
imide (0.5045 g) from ethyl acetate–hexane, which gave chro-
matographically homogeneous imide (0.3902 g, 77% recovery),
m.p. 248–249 °C.
1H NMR (δ, CD3COCD3): 8.72 (1H, d J = 8.5), 8.67 (1H, d J =
7.3), 8.53 (1H, d J = 7.9), 8.07 (1H, t J = 7.9), 8.05 (1H, d J = 7.9),
7.56 (2H, t J = 7.6), 7.49 (1H, t J = 6.9), 7.46 (2H, d J = 6.9) ppm.
13C NMR (δ, CD3SOCD3): 163.3, 163.0, 137.5, 135.8, 131.6,
130.8, 130.1, 129.0, 128.9, 128.64, 128.57, 128.3, 127.7, 123.4,
122.1 ppm.
νmax (KBr): 3070, 1709, 1665 cm−1
.
HR-APCI-MS (M + H, m/z): Found: 308.0462, 310.0416;
C18H11ClNO2+ requires 308.0478, 310.0449.
6-Chloro-2-(3-methoxyphenyl)-1H-benz[de]isoquinolin-1,3-
(2H)-dione (7c). This imide was prepared by the standard pro-
cedure from 4-chloro-1,8-naphthalic anhydride (2.9535 g,
12.70 mmol) and m-anisidine (2.4366 g, 19.78 mmol) in acetic
acid (31 mL). The reaction was heated 16 h under reflux. The
imide obtained (3.3367 g, 78%), m.p. 273–274 °C, was pure
enough for use without a second recrystallization. HPLC analy-
sis showed the material to be >98% pure.
Experimental
Melting points were measured using a hot-stage melting point
apparatus, and are uncorrected. Butylamine was dried over
KOH pellets for at least 24 h prior to use. 4-Chloro-1,8-
naphthalic anhydride and the aromatic amines were used as
supplied. 1H NMR spectra were recorded at 400 MHz; peaks
are reported in ppm (δ) downfield from Me4Si, and are
reported as singlet (s), doublet (d), triplet (t), quartet (q), and
multiplet (m). 13C NMR spectra were recorded at 100 MHz;
peaks are reported in ppm (δ) downfield from Me4Si. Analytical
HPLC was carried out using a C-18 reversed-phase column
with methanol or 10% water in methanol as the eluting
solvent. A blank was run to ensure that artifacts from solvent
change were eliminated. High resolution mass spectra were
recorded using a dual mode (APCI/ESI) ion source; values of
m/z were calibrated using the isotope cluster of the M or M + H
ion from 2,4,6-tribromoaniline (C6H4Br3N+) as internal
standards.
1H NMR (δ, CDCl3): 8.74 (1H, dd J = 0.9, 7.3), 8.69 (1H, dd
J = 0.9, 8.6), 8.58 (1H, d J = 7.8), 7.92 (1H, dd J = 7.4, 8.4), 7.89
(1H, d J = 7.8), 7.49 (1H, t J = 8.1), 7.06 (1H, ddd J = 0.7, 2.4,
8.3), 6.93 (1H, ddd J = 0.8, 2.4, 7.7), 6.88 (1H, t J = 2.3), 3.86
(3H, s) ppm.
13C NMR (δ, CDCl3): 163.8, 163.6, 160.5, 139.4, 136.2, 132.4,
131.5, 131.0, 130.1, 129.5, 128.0, 127.5, 123.2, 121.7, 120.7,
114.9, 114.2, 55.4 ppm.
νmax (KBr): 3074, 1709, 1661 cm−1
.
HR-APCI-MS (M + H, m/z): Found: 338.0594, 340.0568;
C19H13ClNO3+ requires 338.0584, 340.0554.
6-Chloro-2-cyclohexyl-1H-benz[de]isoquinolin-1,3(2H)-dione
(11). This imide was prepared by the standard procedure from
4-chloro-1,8-naphthalic anhydride (3.1344 g, 13.47 mmol) and
cyclohexylamine (20.6550 g, 208.26 mmol) in acetic acid
(31 mL). The mixture was stirred 46 h under reflux. The imide
was obtained as an off-white solid (3.5927 g, 90%), m.
Synthesis of chloronaphthalimides
Standard procedure. 4-Chloro-1,8-naphthalic anhydride p. 239–240 °C. HPLC analysis showed the material to be ≥95%
(1.00 eq.) was mixed with the aromatic amine (1.1–2.0 eq.) and pure.
the mixture was dissolved in acetic acid (30–40 mL). The reac-
1H NMR (d, CDCl3): 8.65 (1H, d J = 7.3), 8.59 (1H, d J = 8.4),
tion mixture was stirred under reflux for 16–24 hours and 8.49 (1H, d J = 7.9), 7.86 (1H, dd J = 7.5, 8.2), 7.83 (1H, d J =
allowed to cool to room temperature. The resultant precipitate 8.1), 5.03 (1H, tt J = 3.7, 12.2), 2.56 (2H, d J = 3.4 of q J = 12.4),
was collected by vacuum filtration, washed well with water, 1.91 (2H, br. d J = 13.2), 1.0–1.8 (6H, complex) ppm.
4394 | Org. Biomol. Chem., 2013, 11, 4390–4396
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
13C NMR (δ, CDCl3): 164.1, 163.9, 138.6, 131.8, 131.0, 130.2,
13C NMR (δ, CD3SOCD3): 163.8, 163.1, 156.8, 143.7, 139.3,
129.12, 129.07, 127.8, 127.3, 123.7, 122.2, 53.9, 20.1, 26.5, 132.4, 131.9, 130.7, 130.2, 129.9, 126.3, 125.0, 124.3, 122.6,
25.4 ppm.
νmax (KBr): 3096, 1704, 1661 cm−1
HR-APCI-MS (M + H, m/z): Found: 314.0935, 316.0911;
C18H17ClNO2+ requires 314.0948, 316.0918.
113.3, 112.9, 44.4 ppm.
.
νmax (KBr): 1691, 1635 cm−1
.
HR-APCI-MS (M + H, m/z): Found: 396.1040; C20H18N3O4S+
requires 396.1013.
6-Chloro-2-(3-chlorophenyl)-1H-benz[de]isoquinolin-1,3(2H)-
2-(4-Aminosulfonylphenyl)-6-chloro-1H-benz[de]isoquinolin-
dione (7a). This imide was prepared by the standard pro- 1,3(2H)-dione (3). A mixture of 4-chloro-1,8-naphthalic anhy-
cedure from 4-chloro-1,8-naphthalic anhydride (3.0458 g, dride (3.1076 g, 13.36 mmol) and sulfanilimide (4.6162 g,
13.09 mmol) and m-chloroaniline (2.5843 g, 20.26 mmol) in 26.81 mmol) in benzonitrile (30 mL) was stirred 26 h under
acetic acid (30 mL). The reaction was heated 20 h under reflux. reflux. The reaction mixture was allowed to cool, and the crude
The imide obtained (4.1885 g, 93%), m.p. 266–267 °C, was product was isolated by vacuum filtration. Recrystallization of
pure enough for use without further purification. HPLC analy- the crude product from acetic acid gave the imide 3 as an iri-
sis showed the material to be >98% pure.
descent gold solid (2.9800 g, 53%), exhibiting a sharp m.p.
1H NMR (δ, CDCl3): 8.73 (1H, d J = 7.6), 8.71 (1H, d J = 8.4), >300 °C (approximately 305–306 °C).
8.57 (1H, d J = 7.8), 7.93 (1H, t J = 7.9), 7.90 (1H, d J = 7.9),
1H NMR (δ, CD3SOCD3): 8.70 (1H, dd J = 0.6, 7.9), 8.62
7.45–7.55 (2H, complex), 7.36 (1H, br. s), 7.20–7.25 (1H, br. m). (1H, d J = 7.2), 8.47 (1H, d J = 7.9), 8.10 (1H, d J = 7.9), 8.07
13C NMR (δ, CDCl3): 163.7, 163.4, 139.7, 136.2, 135.0, 132.6, (1H, dd J = 7.6, 7.9), 7.98 (2H, d J = 8.5), 7.63 (2H, d J = 8.5),
131.6, 131.2, 130.3, 129.5, 129.4, 129.19, 129.17, 128.0, 127.6, 7.52 (2H, s) ppm.
127.0, 123.0, 121.5 ppm.
13C NMR (δ, CD3SOCD3): 172.0, 163.2, 162.9, 144.0, 138.8,
131.6, 130.9, 130.3, 129.8, 128.8, 128.6, 128.5, 127.7, 126.4,
νmax (KBr): 3061, 1709, 1665 cm−1
.
HR-APCI-MS (M + H, m/z): Found: 342.0103, 344.0077, 123.2, 121.9 ppm.
345.9876; C18H10Cl2NO2+ requires 342.0089, 344.0059.
νmax (KBr): 3352, 3252, 1713, 1670 cm−1
.
2-(4-Bromophenyl)-6-chloro-1H-benz[de]isoquinolin-1,3(2H)-
HR-APCI-MS (M + H, m/z): Found: 387.0212, 389.0184;
dione (7b). This imide was prepared by the standard pro- C18H12ClN2O4S+ requires 387.0206, 389.0177.
cedure from 4-chloro-1,8-naphthalic anhydride (3.0009 g,
12.90 mmol) and p-bromoaniline (3.4528 g, 20.07 mmol) in
Treatment of imide 2 with hexylamine
acetic acid (31 mL). The reaction was heated 20 h under reflux. The imide 2 (1.00 g, 2.59 mmol) was added to hexylamine
The imide obtained (3.9567 g, 90%), m.p. 296–297 °C, was (75 mL), and the mixture was stirred 18 h under reflux. The
pure enough for use without further purification. HPLC analy- reaction mixture was poured into dilute HCl (10%, 200 mL)
sis showed the material to be >98.5% pure.
and the resultant mixture was extracted with CH2Cl2 (3 ×
1H NMR (δ, CDCl3): 8.73 (1H, dd J = 1.1, 7.3), 8.71 (1H, dd 20 mL). The combined organic layers were dried and evapor-
J = 1.1, 8.5), 8.57 (1H, d J = 7.9), 7.93 (1H, dd J = 7.3, 8.5), 7.89 ated under reduced pressure to give a yellow solid that was
(1H, d J = 7.8), 7.70 (2H, AA′ of AA′BB′ system J = 8.7), 7.23 (2H, recrystallized from methanol to provide a bright yellow solid
AA′ of AA′BB′ system J = 8.7) ppm.
(0.70 g, 68%), m.p. 47–50 °C, shown to be identical with
13C NMR (δ, CDCl3): 163.9, 163.7, 139.7, 137.1, 132.7, 132.6, 2-hexyl-6-hexylamino-1H-benz[de]isoquinolin-1,3(2H)-dione (5)
131.7, 131.2, 130.4, 129.5, 129.4, 128.0, 127.6, 123.0, 122.9, by comparison with an authentic sample.
121.5 ppm.
Kinetics
νmax (KBr): 3100, 1704, 1670 cm−1
.
HR-APCI-MS (M + H, m/z): Found: 385.9588, 387.9567,
1. Standardization. All kinetics experiments were run in
389.9549; C18H10BrClNO2 requires 385.9583, 387.9554, triplicate in a 30 L water bath thermostated at 25.00 0.03 °C.
387.9563, 389.9534. Temperatures were monitored using two mercury-in-glass
+
2-(4-Aminosulfonylphenyl)-6-dimethylamino-1H-benz[de]iso- thermometers simultaneously immersed in the bath for the
quinolin-1,3(2H)-dione (12). A solution of imide 3 (1.38 g, duration of the experiment; temperature readings of the two
3.73 mmol) and copper(II) sulfate pentahydrate (1.69 g, thermometers did not differ by more than 0.01 °C at any time
10.59 mmol) in DMF (25 mL) was stirred 24 h under reflux. during the runs. Monitoring of the temperature during the run
The mixture was cooled, and then poured into water (75 mL) showed that the temperature remained constant to within
to precipitate the crude product. The crude product was col-
0.01 °C throughout the run. The variability of the 1H NMR
lected by vacuum filtration and recrystallized from acetic acid integrations was determined by running the same sample five
to give the imide 12 (0.46 g, 33%) as a yellow solid, m.p. times, and recording the integrations of the peaks used. The
>300 °C (dec.). HPLC analysis showed the material to be results showed that the ratios of the integrated areas of the
>99.5% pure.
critical peaks varied by less than 0.1%.
2. Solution preparation and kinetic data collection. The
1H NMR (δ, CD3COCD3): 8.65 (1H, d J = 8.6), 8.55 (1H, d J =
7.1), 8.45 (1H, d J = 8.4), 8.06 (2H, d J = 8.0), 7.82 (1H, dd J = samples were prepared by adding dry 1-butylamine (≈12 mL)
8.0, 7.8), 7.62 (2H, d J = 8.0), 7.32 (1H, d J = 7.8), 6.71 (2H, to an accurately weighed quantity of the chloronaphthalimide
br. s), 3.19 (6H, s) ppm.
(≈12 mg), and stirring briefly to dissolve as much of the solid
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 4390–4396 | 4395

View Article Online
Paper
Organic & Biomolecular Chemistry
as possible. The resultant homogeneous solution was trans-
ferred to a volumetric flask (10.00 mL), and the unused solu-
tion was evaporated under reduced pressure to provide the
mass of undissolved imide. The flask was placed in the ther-
mostated water bath, and aliquots (≈0.5 mL) were withdrawn
periodically and quenched with 20% aqueous HCl (≈10 mL).
The acidified solution was extracted with dichloromethane (2 ×
20 mL), and the combined organic layers were washed with
water (20 mL), dried (Na2SO4), and evaporated. The compo-
Sci., 2012, 11, 1675–1681; E. B. Veale and T. Gunnlaugsson,
Annu. Rep. Prog. Chem., Sect. B: Org. Chem., 2010, 106, 376–
406.
4 S. C. Hartsel and D. E. Lewis, U.S. Patent, 7,528,144, 2009;
X. Qian, Yi. Xiao, Yu. Xu, X. Guo, J. Qian and W. Zhu,
Chem. Commun., 2010, 46, 6418–6436.
5 B. A. Ristow, PhD Diss., South Dakota State University,
Brookings, SD, 2008.
6 W. W. Stewart, J. Am. Chem. Soc., 1981, 103, 7615–7620.
7 S.-C. Chang, R. E. Utecht and D. E. Lewis, Dyes Pigm.,
1999, 43, 83–94; S.-C. Chang, M.S. Thesis, South Dakota
State University, 1997; R. G. Baughman, S.-C. Chang,
R. E. Utecht and D. E. Lewis, Acta Crystallogr., Sect. C:
1
sition of the sample was analyzed by H NMR spectroscopy in
CDCl3 solution.
Cryst.
Struct.
Commun.,
1995,
51,
1189–1193;
Acknowledgements
M. S. Alexiou, V. Tychopoulos, S. Ghorbanian,
J. H. P. Tyman, R. G. Brown and P. I. Brittain, J. Chem.
Soc., Perkin Trans. 2, 1990, 837–842; J. E. Elbert,
S. Paulsen, L. Robinson, S. Elzey and K. Klein, J. Photo-
chem. Photobiol., A, 2005, 169, 9–19; B. Liu and H. Tian,
Acknowledgment is made to the Donors of the American
Chemical Society Petroleum Research Fund for support of this
research. The support of the UW-Eau Claire Office of Research
and Sponsored Programs is also gratefully acknowledged.
Chem.
Commun.,
2006,
3156–3158;
H.
He,
M. A. Mortellaro, M. J. P. Leiner, S. T. Young, R. J. Fraatz
and J. K. Tusa, Anal. Chem., 2003, 75, 549–555; W. Zhu,
N. Minami, S. Kazaoui and Y. Kim, J. Mater. Sci., 2003,
13, 2196–2201; G. R. Bardajee, A. Y. Li, J. C. Haley and
M. A. Winnik, Dyes Pigm., 2008, 79, 24–32.
8 V. Bojinov, G. Ivanova, J.-M. Chovelon and I. Grabchev,
Dyes Pigm., 2003, 58, 65–71.
Notes and references
1 K. A. Mitchell, R. G. Brown, D. Yuan, S.-C. Chang,
R. E. Utecht and D. E. Lewis, J. Photochem. Photobiol., A,
1998, 115, 157–161; F. Lupo, S. Gentile, F. P. Ballistreri,
G. A. Tomaselli, M. E. Fragalà and A. Gulino, Analyst, 2010,
135, 2273–2279; H. Wang, H. Wu, L. Xue, Y. Shi and X. Li,
Org. Biomol. Chem., 2011, 9, 5436–5444.
9 H. Cao, V. Chang, R. Hernandez and M. D. Heagy, J. Org.
Chem., 2005, 70, 4929–4934.
10 C. Hansch, A. Leo and R. W. Taft, Chem. Rev., 1991, 91,
165–195.
11 I. Grabchev and D. Staneva, Z. Naturforsch., 2003, 58a, 558–
562.
12 X. Qian, Z. Zhu and K. Chen, J. Prakt. Chem., 1992, 334,
161–164.
2 R. M. Duke, E. B. Veale, F. M. Pfeffer, P. E. Kruger and
T. Gunnlaugsson, Chem. Soc. Rev., 2010, 39, 3936–3953;
E. B. Veale, G. M. Tocci, F. M. Pfeffer, P. E. Kruger and
T. Gunnlaugsson, Org. Biomol. Chem., 2009, 7, 3447–3454.
3 S. Zheng, P. L. M. Lynch, T. E. Rice, T. S. Moody,
H. Q. N. Gunaratne and A. P. de Silva, Photochem. Photobiol.
4396 | Org. Biomol. Chem., 2013, 11, 4390–4396
This journal is © The Royal Society of Chemistry 2013