Energetic and structural characterization of 2-R-3-methylquinoxaline-1,4- dioxides (R = benzoyl or tert-butoxycarbonyl): Experimental and computational studies

Article abstract of DOI:10.1002/poc.1184

The gaseous standard molar enthalpies of formation of two 2-R-3-methylquinoxaline-1,4-dioxides (R = benzoyl or tert-butoxycarbonyl), at T = 298.15 K, were derived using the values for the enthalpies of formation of the compounds in the condensed phase, me

Full text of DOI:10.1002/poc.1184

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY
J. Phys. Org. Chem. 2007; 20: 491–498
Published online 9 May 2007 in Wiley InterScience
(www.interscience.wiley.com) DOI: 10.1002/poc.1184
Energetic and structural characterization
of 2-R-3-methylquinoxaline-1,4-dioxides
(R ¼ benzoyl or tert-butoxycarbonyl):
experimental and computational studies
1
1
1
2
2
1
´
´
Jose R. B. Gomes, Emanuel A. Sousa, Jorge M. Gonc¸alves, Luıs Gales, Ana M. Damas, Paula Gomes,
3
4
1
*
Siddharth Pandey, William E. Acree Jr and Maria D. M. C. Ribeiro da Silva
1
´
´
ˆ
Centro de Investigac¸a˜o em Quımica (CIQ), Departamento de Quımica, Faculdade de Ciencias, Universidade do Porto (UP), Rua do Campo
Alegre 687, P-4169-007 Porto, Portugal
2
ˆ
´
Instituto de Biologia Molecular e Celular (IBMC) and Instituto de Ciencias Biomedicas de Abel Salazar (ICBAS), Universidade do Porto,
Portugal
3Department of Chemistry, Indian Institute of Technology Delhi (IIT-Delhi), Hauz Khas, New Delhi 110016, India
4Department of Chemistry, University of North Texas (UNT), Denton, TX 76203-5070, USA
Received 26 January 2007; revised 7 March 2007; accepted 13 March 2007
ABSTRACT: The gaseous standard molar enthalpies of formation of two 2-R-3-methylquinoxaline-1,4-dioxides
(R ¼ benzoyl or tert-butoxycarbonyl), at T ¼ 298.15 K, were derived using the values for the enthalpies of formation
of the compounds in the condensed phase, measured by static bomb combustion calorimetry, and for the enthalpies of
sublimation, measured by Knudsen effusion, using a quartz crystal oscillator. The three dimensional structure of
2-tert-butoxycarbonyl-3-methylquinoxaline-1,4-dioxide has been obtained by X-ray crystallography showing that the
two N—O bond lengths in this compound are identical. The experimentally determined geometry in the crystal is
similar to that obtained in the gas-phase after computations performed at the B3LYP/6-311 þ G(2d,2p) level of theory.
The experimental and computational results reported allow to extend the discussion about the influence of the
molecular structure on the dissociation enthalpy of the N—O bonds for quinoxaline 1,4-dioxide derivatives. As found
previously, similar N—O bond lengths in quinoxaline-1,4-dioxide compounds are not linked with N—O bonds having
the same strength. Copyright # 2007 John Wiley & Sons, Ltd.
KEYWORDS: thermochemistry, bond dissociation energies, density functional theory calculations, nitrogen—oxygen bonds
INTRODUCTION
from quinoxaline 1,4-dioxide, which were activated
under hypoxic conditions, are at different stages of
development to be used as drugs. These compounds are
thought to be active due to the creation of cytotoxic
radicals formed after N—O bond cleavage taking place in
these N-dioxide compounds. Accordingly, the knowledge
of the energetics of the N—O bonds in this class of
compounds has a fundamental importance for the
characterization of their behavior. Therefore, the thermo-
chemical study of heterobicyclic molecules derived from
quinoxaline N,N0-dioxide has been receiving special
attention,12–16 aimed at construction of relationships
between the energetic and structural properties with the
reactivities of such compounds,15,17 in order to clarify the
chemical behavior of this class of molecules.
Quinoxaline 1,4-di-N-oxide derivatives are key structures
in a large variety of biochemical processes and this
justifies the biological and industrial significance of that
group of compounds. In the last two decades, they have
been synthesized and evaluated with the objective of
determining the influence of different substituents in
positions 2 and 3 of the quinoxaline ring on their
biological activity.1–6
In fact, the quinoxaline 1,4-dioxides are a group of
synthetic antibacterial agents largely used as medicinal
feed additives7,8 and they are also used as bioreductive
cytotoxic agents/species.9–11 Several compounds derived
*Correspondence to: M. D. M. C. Ribeiro da Silva, Centro de Inves-
The present work reports the standard (po ¼ 0.1 MPa)
molar enthalpies of formation, in the gaseous state, at
T ¼ 298.15 K, for two quinoxaline 1,4-dioxides deriva-
tives, namely, 2-benzoyl-3-methylquinoxaline-1,4-dioxide
´
tigac¸a˜o em Quımica (CIQ), Departamento de Quımica, Faculdade de
Ciencias, Universidade do Porto (UP), Rua do Campo Alegre 687,
´
ˆ
P-4169-007 Porto, Portugal.
E-mail: mdsilva@fc.up.pt
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 491–498

492
J. R. B. GOMES ET AL.
333 K in a water-bath. Calcium hydroxide was added
portion-wise to the warm mixture, which immediately
turned into an orange solution. A condenser was adapted
to the flask and the reaction proceeded for 2.5 h, with
monitoring by thin-layer chromatography. The crude
product precipitated as a yellow solid directly from the
reaction mixture upon cooling, being collected by suction
filtration, rinsed with cold isopropanol, and recrystallized
from methanol. Interestingly, large and well-defined
crystals of pure 2 could be additionally isolated upon slow
(over 1 week) evaporation of the remnant methanolic
mother-liquor. Pure compound 2 was isolated as a bright
yellow solid with correct spectroscopic and analytical
data, as follows: dH (CDCl3, 300 MHz): 8.55 (dd,
J ¼ 7.16 Hz; J ¼ 2.13 Hz, 1H), 8.51 (dd, J ¼ 7.13 Hz,
J ¼ 2.23 Hz, 1H), 7.81 (td, J ¼ 7.03 Hz, J ¼ 1.77 Hz, 1H),
7.76 (td, J ¼ 7.03 Hz, J ¼ 1.77 Hz, 1H); 2.53 (s, 3H), 1.62
(s, 9H); dC (CDCl3, 75 MHz): 158.68, 138.63, 137.76,
136.89, 136.14, 132.29, 131.36, 120.25, 120.10, 86.26,
29.67, 27.98, 27.29; 14.17; mass fraction for
C14H16N2O4: found C, 0.6089; H, 0.0595; N, 0.1011;
calculated: C, 0.6086, H, 0.0584, N, 0.1014.
Figure 1. Structural formulas for 1, 2-benzoyl-3-methyl-
quinoxaline-1,4-dioxide, 2, 2-tert-butoxycarbonyl-3-methyl-
quinoxaline-1,4-dioxide, 3, 2-benzoyl-3-methylquinoxaline
and, 4, 2-tert-butoxycarbonyl-3-methylquinoxaline
and 2-tert-butoxycarbonyl-3-methylquinoxaline-1,4-
dioxide, represented in Figure 1. Those values were
calculated from the experimental values of the enthalpies
of combustion of the crystalline compounds and their
enthalpies of sublimation. The experimental measure-
ments were performed using static bomb calorimetry and
Knudsen effusion techniques.
Density functional theory (DFT) calculations have been
performed using the same approach used in other
works13,14,16 regarding the computation of the gas-phase
N—O bond dissociation enthalpies (BDEs). From the
consideration of the N—O (BDEs) for 1 and 2 and also of
their standard molar gas-phase enthalpies of formation,
DfHmo (g), it was also possible to estimate the DfHmo (g) for
compounds 3 and 4.
The crystal structure of compound 2 has been obtained
by X-ray crystallography and the structural data, in clear
agreement with the parameters coming from the DFT
computations, are interpreted jointly with the mean N—O
BDEs.
It should be emphasized that the control of reaction time
is quite important for the successful synthesis of 2, as
maximum yield is achieved within 2.5 h; in our hands,
longer reaction times led to loss of the tert-butyl ester
group, that is, conversion of 2 into 2-methyl-quinox-
aline-1,4-dioxide to a significant extent, as confirmed by
1H-NMR, 13C-NMR and elemental analysis (data not
shown) of reaction products. It is also noteworthy that, in
our hands, using Ca(OH)2 as the base catalyst2 for the
synthesis reaction led to significantly improved yield and
purity, as compared to syntheses carried out using
morpholine3–5 or triethylamine4–6 as base catalysts. The
synthesis of 2 has been recently described by Jaso et al.6
who also used benzofuroxane and tert-butyl acetoacetate
as starting materials; their procedure was based in the
EXPERIMENTAL
˚
addition of 4 Amolecular sieves to a methanolic solution of
Synthesis and characterization of compounds
those reactants, after which the methanol was evaporated at
2938C and the reaction allowed to proceed at 3638C on the
surface of the molecular sieves for about 1 h, without
drying. However, this synthetic method gave the desired
compound 2 with only a 3% yield, after purification.
The average ratio of the mass of carbon dioxide
recovered after combustion experiments to that calculated
from the mass of samples (1: 0.9996 ꢀ 0.0008, and 2:
0.9997 ꢀ 0.0004) confirmed the composition of the
samples. The thermal behavior of the compounds were
studied by differential scanning calorimetry. The thermo-
grams show that there is no transition phase before the
melting temperatures: 488 K (with decomposition) and
418 K, respectively for 1 and 2.
Compound 1 was prepared by slow addition of
(benzoylacetone þ triethylamine) solution to a warm
triethylamine solution containing benzofuroxan accord-
ing to the published method of Haddadin et al.18 The
resulting solution was stirred for 24 h at ambient room
temperature. The crude product was collected by vacuum
filtration and washed with chilled triethlamine. The
compound was further purified by three crystallizations
from anhydrous methanol. Elemental analyses were in
excellent agreement with calculated values: mass fraction
for C16H12N2O3: found C, 0.6866, H, 0.0428, N, 0.1011;
calculated: C, 0.6856, H, 0.0432, N, 0.0999.
Compound 2 was prepared from benzofuroxan and
tert-butyl acetoacetate, following the method described
by Robertson and Kasubick.2 Benzofuroxan was sus-
pended in isopropanol in a round-bottom flask, tert-butyl
acetoacetate was added and the suspension was heated to
Structural determination of 2
Single crystals of C14H16N2O4 were obtained during the
synthesis of the compound, mounted on top of a
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 491–498
DOI: 10.1002/poc

STUDY OF 2-R-3-METHYLQUINOXALINE-1,4-DIOXIDES
493
fiberglass, and transferred to the diffractometer. Diffrac-
tion data were collected at 293 K with a Stoe IPDS plate
4-hydroxy-2-methylquinoline), and there was an excel-
lent agreement among the results obtained with those
reported in the literature. The vapor effusing from the
Knudsen cell was allowed to condense on a quartz crystal
positioned above the effusion hole; changes in the
frequency of oscillation of the quartz crystal, Df, were
proportional to the mass condensed in its surface,28
Df ¼ CfDm, where Cf is a proportionally constant. From
the Knudsen equation, the vapor pressure is given by Eqn
(1), where (Dm/Dt) is the rate of mass loss, a is the
effective area of the effusion hole, and M is the molar
mass of the effusing vapor.
˚
equipped with MoKa radiation (l ¼ 0.71073 A).
Calorimetric measurements
The energies of combustion of 1 and 2 were measured
with a static bomb calorimeter, with a twin valve bomb
and an internal volume of 0.290 dm3, assembled originally
in Teddington,19,20 and now installed in Thermochemistry
Laboratory, at the University of Porto.21 The energy
equivalent of the calorimeter was determined from the
combustion of benzoic acid (BDH Thermochemical
Standard, batch 693976/01) having a massic energy of
combustion, under standard bomb conditions, of
ꢁ(26435.1 ꢀ 3.5) J ꢂ gꢁ1. The calibration results were
corrected to an energy equivalent e(calor) corresponding
to an average mass of water added to the calorimeter of
2900.0 g. Seven calibration experiments were made in
oxygen atmosphere at p ¼ 3.04MPa, with 1.00cm3 of
water added to the bomb, leading to an energy equivalent
1=2
p ¼ ðDm=DtÞ ꢂ aꢁ1 ꢂ ð2pRT=MÞ
(1)
The measured rate of change of frequency of
oscillation with time, n ¼ Df/Dt, is directly proportional
to the rate of sublimed mass of the crystalline
sample,28n ¼ Cf Dm/Dt, so it is possible to establish the
Eqn (2), relating directly the vapor pressure with n.
1=2
n ꢂ T1=2 ꢂ ð2pR=MÞ
p ¼
(2)
a ꢂ Cf
Using the integrated form of the Clausius–Clapeyron
of the calorimeter e(calor) ¼ (15551.6 ꢀ 2.6) J ꢂ Kꢁ1
,
where the uncertainty quoted is the standard deviation
of the mean.
equation, the enthalpy of sublimation, DgcrHmo , is derived
½
The samples were ignited, in pellet form, in oxygen
atmosphere (p ¼ 3.04 MPa) at T ¼ 298.15 K, with
1.00 cm3 of water inside the bomb. The small amount
available for 2 led us to use n-hexadecane as an auxiliary
combustion material (standard massic energy of combus-
tion, Dc uo ¼ ꢁ(47160.8 ꢀ 4.1) J ꢂ gꢁ1) in the combustion
experiments, in order to produce an appropriate increase
of temperature. For the cotton thread fuse of empirical
from the slope of ln(n T ) against Tꢁ1. From at least five
independent sets of experimental measurements of the
frequency of the quartz oscillator for each compound, at
convenient temperature intervals, it was possible to obtain
independent results for the enthalpy of sublimation of
each compound, referenced to the mean temperature of
the experimental range.
formula CH1.686O0.843, Dc uo ¼ ꢁ16250 J ꢂ g
.
ꢁ1 22 Correc-
tions for nitric acid formation were based on
ꢁ59.7 kJ ꢂ gꢁ1 for the molar energy of formation of 0.1
mol ꢂ dm3 HNO3(aq) from N2(g), O2(g), and H2O(l).23 At
T ¼ 298.15 K, (@u/@p)T for the solid was assumed to be
Theoretical details
The Gaussian 03 suite of program29 has been used to
compute, at the B3LYP level of theory,30 structural and
energetic details for all compounds considered here. The
density functional theory based B3LYP method uses the
exchange functional introduced by Becke in 1988
together with the LYP correlation functional of Lee
et al.31,32
The atomic electron density of hydrogen, carbon,
nitrogen, and oxygen atoms has been described with two
different basis sets. The 6-31Gꢃ(d) basis set has been used
to fully-optimize the geometry of all compounds and to
compute the vibrational frequencies. The absence of
negative frequencies ensured that all structures were
minima on the potential energy surface. These calcu-
lations permitted us to extract the thermal corrections for
T ¼ 298.15 K as well as to obtain structural guesses for the
calculations performed with a larger basis set. The larger
basis set was the 6-311 þ G(2d,2p) basis set, which was
used only for the full-optimization of the structures of all
compounds. Finally, the enthalpies of all compounds
were obtained by combining the energy coming from the
ꢁ1
ꢁ0.2 J ꢂ g ꢂ MPaꢁ1, a typical value for organic solids.
The amount of compound burnt in each experiment was
determined from the total mass of carbon dioxide
produced after allowance for that resulted from the
cotton thread fuse and n-hexadecane. For each exper-
iment, the value of Dcuo was calculated by using the
procedure given by Hubbard et al.24 The relative atomic
masses used throughout this paper were those recom-
mended by the IUPAC Commission in 2001.25
Knudsen effusion technique
For both 2-R-3-methyl-quinoxaline-1,4-dioxide deriva-
tives, the respective standard molar enthalpies of
sublimation were determined using a quartz crystal
microbalance,26,27 by the Knudsen effusion method. The
effusion system was previously tested with three
compounds (benzanthrone, squaric
acid, and
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 491–498
DOI: 10.1002/poc

494
J. R. B. GOMES ET AL.
calculations completed at the highest level of theory and
the thermal corrections retrieved from the scaled
vibrational frequency (factor ¼ 0.98) calculations per-
formed at the B3LYP/6-31G(d) level. For the open-shell
species, namely the gas-phase oxygen atom, the unrest-
ricted UB3LYP approach has been employed both for the
calculation of the total energy and corresponding thermal
corrections.
agreement with those of other quinoxaline-1,4-dioxide
derivatives;36–42 a total of 10 crystal structures are
available in the Cambridge Crystallographic Data Base
˚
and the N—O distances range between 1.28–1.31 A. In
three of these compounds,36–38 the crystallographic
packing leads to hydrogen bonds involving the oxygen
atom of the N—O group, which may contribute to the
elongation of this bond. This is the case for
2-(N-(2-hydroxyethyl)carboxamide)-3-methylquinoxali-
ne 1,4-dioxide,36 very similar to 2, that displays N—O
˚
distances of 1.286 and 1.304 A. However, the N—O group
RESULTS AND DISCUSSION
˚
with a bond length of 1.304 A is involved in an
intermolecular hydrogen bond with the N—H bond of
the N-(2-hydroxyethyl)carboxamide. In compound 2,
there is no evidence for intermolecular hydrogen bonding
involving the oxygen atoms. Another synthesized
heteroaromatic N-oxide with high similarity to 2 is
2-methyl-3-carboethoxyquinoxaline.39 The crystal struc-
ture of this compound shows that there are no hydrogen
bonds in the crystal packing and that the N—O bond
Crystal structure determination of C14H16N2O4
The crystal structure of C14H16N2O4 was solved using
SHELXS33 and refined using SHELXL34 program. C5 to
C8-bound hydrogen atoms were refined freely with
isotropic displacement parameters while all of the
hydrogen’s from the methyl groups were positioned with
idealized geometry and refined riding on their parent C
˚
˚
lengths are slightly longer (1.292 and 1.303 A) than the
N—O bond lengths in 2.
atoms at distances of 0.93 A, with Uiso (H) ¼ 1.2 Ueq (C).
Tables containing the final fractional coordinates,
temperature parameters, bond distances, and bond angles
are deposited with the Cambridge Crystallographic Data
Centre (CCDC reference number 633371). Briefly, the
most important details for this crystallographic structure
Calorimetric studies
˚
are: M ¼ 276.29, monoclinic, a ¼ 5.9167(18) A, b ¼
The standard enthalpies of formation in condensed phase,
at T ¼ 298.15 K, for the two quinoxaline 1,4-dioxide
derivatives were calculated from the respective massic
energies of combustion, Dc uo, determined by static bomb
calorimetry. Typical results for one combustion exper-
iment of each compound are presented in Table 1. The
energy involved in the isothermal bomb process, DU(IBP),
after ignition of the samples at T ¼ 298.15 K, is derived
from Eqn (3), where DTad is the calorimeter temperature
change corrected for heat exchange and the work of
stirring, Dm(H2O) is deviation of the mass of water added
to the calorimeter from 2900.0 g (the mass assigned for
3
˚
˚
˚
9.708(5) A, c ¼ 12.051(4) A, U ¼ 690.7 A , T ¼ 290 K,
space group P21 (no. 4), Z ¼ 2, m(Mo-Ka) ¼
0.099 mmꢁ1, 8018 reflections measured, 1737 unique
(Rint ¼ 9.84) which were used in all calculations. The final
R(F2) was 0.056 and the final wR(F2) was 0.152 (all data).
A perspective view of the compound, obtained using
ORTEP35 and showing the atomic numbering scheme is
presented in Figure 2. The N1—O1 and the N2—O2 bond
lengths are 1.286(2) Aand 1.288(2) A, respectively, which
are in the upper limit of the data retrieved for 158 N—O
groups, with N belonging to a ring system (the N—O
distance ranges between 1.22–1.31 A) from the crystal
structures deposited in the Cambridge Crystallographic
Data Base. Moreover, these N—O distances are in good
˚
˚
˚
Table 1. Typical combustion experiments for 1 and 2, at
T ¼ 298.15 K
1
2
m(CO2, total)/g
m(cpd)/g
m(hexadecane)/g
1.17997
0.46797
—
0.00265
0.86723
13.71
ꢁ0.10
13497.77
31.04
0.61
9.65
1.61510
0.44734
0.19711
0.00282
1.37144
14.94
m0(fuse)/g
DTad/K
ef/(J Kꢁ1
)
Dm(H2O)/g
ꢁDU(IBP)/Ja
DU(HNO3)/J
DU(ign)/J
1.20
21354.93
32.24
0.60
DUS/J
10.57
DU(hexadecane)/J
DU(fuse)/ꢁJ 1
ꢁDcuo/J g
—
43.10
28664.19
9296.06
45.80
26758.75
Figure 2. View of the molecular structure of C14H16N2O4,
showing the atom-labeling scheme. Displacement ellipsoids
are drawn at the 50% probability level. Hydrogen atoms are
represented by circles of arbitrary size
a DU(IBP) already includes the DU(ign).
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 491–498
DOI: 10.1002/poc

STUDY OF 2-R-3-METHYLQUINOXALINE-1,4-DIOXIDES
495
e(calor)), DUS is the correction to the standard state and
Knudsen effusion measurements
the remaining terms were previously described.24
The results for the determination of the standard molar
enthalpies of sublimation of 1 and 2, obtained by the
Knudsen method, are summarized in Tables 3 and 4,
respectively, together with the mean temperatures of the
experimental ranges and the standard molar enthalpies of
sublimation at these mean temperatures, DgcrHmo (hTi); the
parameter of the Clausius–Clapeyron equation corre-
sponding to the slope was obtained using a least square
fitting of the experimental data. The value of DgcrHmo (hTi)
È
É
(3)
DUðIBPÞ ¼ ꢁ "ðcalorÞ þ cpðH2O; lÞ ꢂ DmðH2OÞ þ "f
DTad þ DUðignÞ
The corresponding values of the standard molar energy,
of combustion, DcUm0 (cr), and the standard molar
enthalpy of combustion, DcHm0 (cr) were calculated using
the mean values of massic energy of combustion, hDcuoi,
of the crystalline compounds, at T ¼ 298.15 K. The value
of the standard molar enthalpy of formation in the
crystalline state for each of the compounds, DfHmo (cr),
was calculated using the corresponding value for
DcHmo (cr) and the values for the standard molar
enthalpies of formation for H2O(l) and for CO2(g) taken
from the literature,43 ꢁ(285.83 ꢀ 0.04) kJ ꢂ molꢁ1 and
ꢁ(393.51 ꢀ 0.13) kJ ꢂ molꢁ1, respectively. Table 2 pre-
sents a resume of the final results of these parameters
derived for the two compounds.
was corrected to T ¼ 298.15 K assuming DgcrCpo;m
¼
ꢁ50 J ꢂ Kꢁ1 ꢂ molꢁ1 26
,
yielding the final DgcrHmo . The
uncertainties assigned are twice the overall standard
deviations of the mean.
Experimental gaseous standard molar
enthalpies of formation
The experimental results for the standard molar
enthalpies of formation and sublimation of crystalline
Table 2. Derived standard (po ¼ 0.1 MPa) molar energies of combustion, DcUmo , standard molar enthalpies of combustion,
DcHmo , and standard molar enthalpies of formation, DfHmo , for crystalline 1 and 2 compounds at T ¼ 298.15 K
ꢁDcuo
ꢁDcUmo ðcrÞ
ꢁDcHmo ðcrÞ
DfHmo ðcrÞ
kJ ꢂ molꢁ1
kJ ꢂ molꢁ1
kJ ꢂ molꢁ1
J ꢂ gꢁ1
Compound
1
2
28666.5 ꢀ 5.6
26758.5 ꢀ 8.2
8034.7 ꢀ 4.3
7393.1 ꢀ 5.3
8035.9 ꢀ 4.3
7395.6 ꢀ 5.3
24.8 ꢀ 4.3
ꢁ 400.2 ꢀ 5.6
Table 3. Parameters of Clausius–Clapeyron equation and standard enthalpies of sublimation for 1: DgcrHmo (298.15 K) ¼
153.8 ꢀ 1.8 kJ ꢂ molꢁ1
DgcrHmo ðhTiÞ
DgcrHmo ð298:15 KÞ
kJ ꢂ molꢁ1
kJ ꢂ molꢁ1
Exp.
T/K
hTi/K
a
ꢁb/Kꢁ1
r
1
2
3
4
5
6
393.1–408.1
394.6–403.5
383.1–398.1
385.7–400.6
381.6–396.8
386.5–401.7
400.7
399.2
390.7
393.2
389.2
394.2
46.57 ꢀ 0.29
46.47 ꢀ 0.55
45.75 ꢀ 0.93
46.19 ꢀ 0.16
45.14 ꢀ 0.52
46.96 ꢀ 0.43
18039 ꢀ 115
17993 ꢀ 220
17754 ꢀ 363
17945 ꢀ 67
17539 ꢀ 203
18252 ꢀ 168
0.9999
0.9998
0.9992
0.9999
0.9997
0.9998
150.0
149.6
147.6
149.2
145.8
151.7
155.1
154.6
152.2
153.9
150.4
156.5
Table 4. Parameters of Clausius–Clapeyron equation and standard enthalpies of sublimation for 2: DgcrHmo (298.15 K) ¼
164.1 ꢀ 1.8 kJ molꢁ1
DgcrHmo ðhTiÞ
DgcrHmo ð298:15KÞ
kJ ꢂ molꢁ1
kJ ꢂ molꢁ1
Exp.
T/K
hTi/K
a
ꢁb/Kꢁ1
r
1
2
3
4
5
363.2–372.3
359.2–371.1
362.2–370.1
357.3–369.7
356.7–369.2
367.8
365.3
366.2
363.4
363.3
53.94 ꢀ 1.82
54.20 ꢀ 1.45
55.63 ꢀ 1.08
54.05 ꢀ 1.09
54.97 ꢀ 2.66
19252 ꢀ 667
19146 ꢀ 533
19701 ꢀ 393
19110 ꢀ 397
19444 ꢀ 968
0.9988
0.9992
0.9994
0.9991
0.9963
160.1
159.2
163.8
158.9
161.7
163.6
162.5
167.2
162.1
164.9
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 491–498
DOI: 10.1002/poc

496
J. R. B. GOMES ET AL.
Table 5. Derived standard (po ¼ 0.1 MPa) molar enthalpies
of formation, DfHmo , and standard molar enthalpies of sub-
limation, DgcrHmo , for compounds 1 and 2 at T ¼ 298.15 K
DfHmo ðcrÞ
DgcrHmo
kJ ꢂ molꢁ1
DfHmo ðgÞ
kJ ꢂ molꢁ1
kJ ꢂ molꢁ1
Compound
1
2
24.8 ꢀ 4.3
153.8 ꢀ 1.8
178.6 ꢀ 4.7
ꢁ 400.2 ꢀ 5.6
164.1 ꢀ 1.8
ꢁ 236.1 ꢀ 5.9
1 and 2 allow us to calculate the gaseous enthalpies of
formation, at T ¼ 298.15 K, for these quinoxaline 1,4-
dioxides derivatives. The final standard molar enthalpies
of formation, DfHmo , in both crystalline and gaseous states,
as well as the standard molar enthalpies of sublimation,
DgcrHmo , at T ¼ 298.15 K, for the two compounds are
summarized in Table 5.
Figure 4. First, second, and total (N–O) bond dissociation
enthalpies for (a) compound 1 and (b) compound 2. All
values are in kJ ꢂ molꢁ1
Computed geometries and N—O bond
dissociation enthalpies
probably the cause for a COR moiety that is less normal
with respect to the quinoxaline ring. Interestingly, the
N—O bonds lengths seem to be unaffected if the bond is
adjacent to a —CH3 or to a —COR group.
The present DFT approach was previously found to be
suitable for the optimization of the structures of
these kinds of compounds.13 Views of the B3LYP/
6-311 þ G(2d,2p) fully-optimized geometries of 1 and 2
are depicted in Figure 3 where selected bond lengths and
angles are also included. The bond lengths and angles
are very similar for both compounds suggesting that the
influence of the R group (see Figure 1) is almost the same
in both cases. The largest differences are found for
The B3LYP/6-311 þ G(2d,2p)//B3LYP/6-31G(d) values
for the first, second, total, and mean N—O bond
dissociation enthalpies for compounds 1 and 2, and
corresponding derivatives, are given in Figure 4. The
gas-phase calculations suggest that, in principle, both
compounds will have a similar behavior regarding their
promptness to form active cytotoxic radicals. In fact,
these two bulky R groups have been chosen in order to see
if at least one of the N—O bonds was weakened, lowering
the energy required to cleave the bond and, in principle,
leading to easy formation of active cytotoxic radicals.
However, the enthalpies of dissociation are nearly the
same as those computed previously for 2-ethoxy-
carbonyl-3-methylquinoxaline-1,4-dioxide (R group in
Figure 1 is —OCH2CH3); first is 242.9 kJ ꢂ molꢁ1, second
—
—
the C O and C( O)—C bond lengths but, even in
—
—
˚
these two cases, they are lower than 0.01 A. The
comparison between the experimental and computational
geometries for 2 reveals that the DFT method yields bond
lengths that almost match the experimental results. Again,
—
the largest difference in bond lengths is found for the C
—
O group. The X-ray result is 1.191 A while the B3LYP/
˚
˚
6-311 þ G(2d,2p) value is 1.204 A. The experimental
is 267.0 kJ ꢂ molꢁ1 and total is 510.0 kJ ꢂ molꢁ1 13
.
There-
NCCO dihedral angle, white dotted line in Figure 3, is
109.38 while the computed value is only 101.48 which is a
consequence of the different physical state of the
molecule in experimental and computational investi-
gations. In the X-ray experiments, the lateral intermo-
lecular interactions in the packed crystalline state are
fore, it is concluded that attaching these three different R
groups to the carbonyl substituent in position 2 of the
quinoxaline ring has little influence on the strength of
the N—O bonds. A possible explanation for this may be
due to the ability of the carbonyl group to attract and
Figure 3. B3LYP/6-311þG(2d,2p) optimized structures of compounds 1 (left) and 2 (right). Selected bond lengths and angles
˚
are in A and in degrees, respectively
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 491–498
DOI: 10.1002/poc

STUDY OF 2-R-3-METHYLQUINOXALINE-1,4-DIOXIDES
497
approach in this case. The DfHmo values computed for 3
and 4 are 192.9 kJ ꢂ molꢁ1 and ꢁ221.7 kJ ꢂ molꢁ1, respec-
tively. The reader should be aware of the uncertainties
associated with both the experimental DfHmo (g) for 1 and
2 and the computed N—O BDEs. Therefore, the
estimated DfHmo values for the compounds 3 and 4 are
O
QNX.ring
O
C
QNX.ring
C
CH3
O
C
CH3
H3C
associated with an uncertainty of ꢄ10 kJ ꢂ molꢁ1
.
Scheme 1
CONCLUSIONS
attenuate the electron density coming from the different R
groups due to mesomeric or inductive effects. Con-
sequently, the carbonyl ‘absorbs’ the electronic character
of the different R groups, blocking any effect that these
might exert on the quinoxaline ring, cf. Scheme 1.
In order to test this hypothesis, the dissociation
enthalpy of the two N—O bonds of 2-p-nitrobenzoyl-
3-methylquinoxaline-1,4-dioxide have also been calcu-
lated at the same B3LYP/6311 þ G(2d,2p)//B3LYP/
6-31G(d) level of theory. This compound was considered
since it is expected that the nitro group would also attract
electron charge from the aromatic ring and, therefore, it is
Complementary experimental and computational studies
on both structural and energetic aspects of two
quinoxaline-1,4-dioxide derivatives have been per-
formed. The two compounds studied differ from other
studied previously by the presence of two bulky
substituents attached to the quinoxalinic ring at position
2. Experimental and computational techniques yield the
same answer about the N—O bond lengths, that is, they
have almost the same size. However, previous work on an
analogue of Tirapazamine, 2-amino-3-quinoxalinecar-
bonitrile-1,4-dioxide,14 the similar bond lengths are not
necessarily connected with bonds having the same
strength and, thus, require different energy for their
cleavage. Also interesting is the fact that the N—O BDEs
are affected only by the type of atoms attached directly to
the quinoxalinic ring. As shown above, the difference
between the N—O BDEs calculated for several 2-R-
3-methylquinoxaline-1,4-dioxides, with R ¼ ꢁCOR0, is
almost negligible, even if R0 changes from an ethyl group
to a tert-butyl or phenyl group.
—
supposed to lead to a different electronic —C( O)R
—
environment that would affect the computed N—O bond
dissociation enthalpy. The computed total enthalpy for
the cleavage of the two bonds is 510.2 kJ ꢂ molꢁ1, which
yields a mean bond dissociation enthalpy for the two
N—O bonds of 255.1 kJ ꢂ molꢁ1. This value is identical to
those reported above, though somewhat smaller (5 kJ ꢂ molꢁ1)
than that reported for 2-benzyl-3-methylquinoxaline-
1,4-dioxide.13 Thus, it seems that all —C( O)R subs-
—
—
tituents bonded directly to the quinoxaline-1,4-dioxide
ring in positions 2 or 3 will have a similar effect on the
strength of the N—O bonds leading to identical enthalpies
to cleave these bonds in this class of compounds.
Finally, the similar lengths calculated for the N—O
bonds in the two title compounds, and also for other
differently substituted quinoxalines-1,4-dioxides,13,14 do
not necessarily imply that all N—O dative bonds will
have the same energy.
The combination of experimental and computational
techniques allowed also to determine the standard molar
gas-phase enthalpies of formation of 2-benzoyl-3-
methylquinoxaline and 2-tert-butoxycarbonyl-3-methyl-
quinoxaline, and of their corresponding 1,4-dioxide
derivatives.
Acknowledgements
ˆ
Thanks are due to Fundac¸a˜o para a Ciencia e a Tecno-
Estimated enthalpies of formation
logia, FCT, Lisbon, Portugal, and to FEDER for financial
support to the research project (POCTI/44471/QUI/
2002). EAS and JRBG thank FCT and the European
Social Fund (ESF) under the 3rd Community Support
Framework (CSF) for the award of the research grants
with references SFRH/BD/5355/2001 and SFRH/BPD/
24676/2005.
The enthalpies of formation of compounds 3 and 4 have
been estimated after the consideration of the enthalpy of
the following reaction:
1 or 2 ðgÞ ! 3 or 4 ðgÞ þ 2 3O ðgÞ
(4)
which is the same as after consideration of the total N—O
BDEs shown in Figure 4, for the loss of both oxygen
atoms in 1 or 2, and the DfHmo listed in Table 5 for
compounds 1 and 2 and for atomic oxygen available in the
literature.43 This strategy was found to be an adequate
approach for the estimation of the enthalpy of formation
of a very similar compound, namely, 2,3-dimethyl-
quinoxaline,13 and so, supposedly it should be also a good
REFERENCES
1. Amin KM, Ismail MMF, Noaman E, Soliman DH, Ammar YA.
Bioorg. Med. Chem. 2006; 14: 6917–6923.
2. Robertson RL, Kasubick RV. US Patent 1973; 3767657.
´
´
3. Monge Vega A, Gil MJ, Fernandez Alvarez E. J. Heterocylic
Chem. 1984; 21: 1271–1276.
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 491–498
DOI: 10.1002/poc

498
J. R. B. GOMES ET AL.
4. Jaso A, Zarranz B, Aldana I, Monge A. Eur. J. Med. Chem. 2003;
38: 791–800.
5. Zarranz B, Jaso A, Aldana I, Monge A. Bioorg. Med. Chem. 2003;
11: 2149–2156.
6. Jaso A, Zarranz B, Aldana I, Monge A. J. Med. Chem. 2005; 48:
2019–2025.
7. Carta A, Loriga M, Paglietti G, Mattana A, Fiori PL, Mollicotti P,
Sechi L, Zanetti S. Eur. J. Med. Chem. 2004; 39: 195–203.
8. Wu Y, Wang Y, Huang L, Tao Y, Yuan Z, Chen D. Anal. Chim. Acta
2006; 569: 97–102.
24. Hubbard WN, Scott DW, Waddington G. In Experimental Thermo-
chemistry, Vol. 1, Rossini FD (ed). Interscience: New York, 1956;
25. Loss RD. Pure Appl. Chem. 2003; 75: 1107–1122.
26. Burkinshaw PM, Mortiner CT. J. Chem. Soc. Dalton Trans. 1984;
75–77.
27. Ribeiro da Silva MAV, Gonc¸alves JM. J. Chem. Thermodynam.
1998; 30: 1465–1481.
28. Sauerbrey G. Z. Phys. 1959; 155: 206–222.
29. Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA,
´
Cheeseman JR, Montgomery JA, Jr, Vreven T, Kudin KN, Burant
9. Zarranz B, Jaso A, Aldana I, Monge A. Bioorg. Med. Chem. 2004;
12: 3711–3721.
JC, Millam JM, Iyengar SS, Tomasi J, Barone V, Mennucci B,
Cossi M, Scalmani G, Rega N, Petersson GA, 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
JE, Hratchian HP, Cross JB, Bakken V, Adamo C, Jaramillo J,
Gomperts R, Stratmann RE, Yazyev O, Austin AJ, Cammi R,
Pomelli C, Ochterski JW, Ayala PY, Morokuma K, Voth GA,
Salvador P, Dannenberg JJ, Zakrzewski VG, Dapprich S, Daniels
AD, Strain MC, Farkas O, Malick DK, Rabuck AD, Raghavachari
K, Foresman JB, Ortiz JV, Cui Q, Baboul AG, Clifford S, Cio-
slowski J, Stefanov BB, Liu G, Liashenko A, Piskorz P, Komaromi
I, Martin RL, Fox DJ, Keith T, Al-Laham MA, Peng CY, Nanayak-
kara A, Challacombe M, Gill PMW, Johnson B, Chen W, Wong
MW, Gonzalez C, Pople JA. Gaussian 03, Revision C.01. Gaus-
sian, Inc., Wallingford CT, 2004.
´
10. Monge A, Palop JA, Cerain AL, Denador V, Martinez-Crespo FJ,
Sainz Y, Narro S, Garcia S, Miguel C, Gonc¸alez M, Hamilton E,
´
Barker AJ, Clarke ED, Greenhow DT. J. Med. Chem. 1995; 38:
1786–1792.
11. Ganley B, Chwdhury G, Bhansali J, Daniels JS, Gates KS. Bioorg.
Med. Chem. 2001; 9: 2395–2401.
12. Acree WE, Jr, Powell JR, Tucker SA, Ribeiro da Silva MDMC,
Matos MAR, Gonc¸alves JM, Santos LMNBF, Morais VMF, Pilcher
G. J. Org. Chem. 1997; 62: 3722–3726.
13. Ribeiro da Silva MDMC, Gomes JRB, Gonc¸alves JM, Sousa EA,
Pandey S, Acree WE, Jr. J. Org. Chem. 2004; 69: 2785–
2792.
14. Ribeiro da Silva MDMC, Gomes JRB, Gonc¸alves JM, Sousa EA,
Pandey S, Acree WE, Jr. Org. Biomol. Chem. 2004; 2: 2507–
2512.
30. Becke AD. J. Chem. Phys. 1993; 98: 5648–5652.
31. Becke AD. Phys. Rev. A 1988; 38: 3098–3100.
15. Gomes JRB, Ribeiro da Silva MDMC, Ribeiro da Silva MAV.
Chem. Phys. Lett. 2006; 429: 18–22.
32. Lee C, Yang W, Parr RG. Phys. Rev. B 1988; 37: 785–789.
33. Sheldrick GM. Acta Cryst. 1990; A46: 467–473.
¨
34. Sheldrick GM. SHELXL97. University of Gottingen: Germany,
1997.
35. Farrugia LJ. J. Appl. Cryst. 1997; 30: 565–565.
36. Bartczak TJ, Galdecki Z, Wolf W, Mak TCW. J. Crystallogr.
Spectrosc. Res. 1988; 18: 165–174.
16. Gomes JRB, Sousa EA, Gomes P, Vale N, Gonc¸alves JM, Pandey
S, Acree WE, Jr, Ribeiro da Silva MDMC. J. Phys. Chem. B 2007;
111: 2075–2080.
17. Acree WE, Jr, Ribeiro da Silva MDMC, Pilcher G. J. Phys. Chem.
Ref. Data 2005; 34: 553–572.
18. Haddadin MH, Taha MU, Jarrar AA, Issidorides CH. Tetrahedron
1976; 32: 719–724.
19. Gundry HA, Harrop D, Head AJ, Lewis GB. J. Chem. Thermo-
dynam. 1969; 1: 321–332.
37. Johnson PL. Acta Cryst. 1974; B30: 540–542.
38. Dvoryantseva GG, Lindeman SV, Aleksanyan MS, Struchkov YT,
Teten’chuk KP, Khabarova LS, Elina AS. Khim. Farm. Zh. 1990;
24: 80–84.
20. Bickerton J, Pilcher G, Al-Taklin G. J. Chem. Thermodynam. 1984;
16: 373–378.
39. Lin SK, Cong QZ. J. Mol. Struct. 1987; 159: 279–286.
40. Wang Y, Wang H, Wang Q. Acta Cryst. 1991; C47: 845–848.
41. Greer ML, Duncan JR, Duff JL, Blackstock SC. Tetrahedron Lett.
1997; 38: 7665–7668.
42. Cong QZ, Lin S-K, Wang HQ. Chinese J. Struct. Chem. 1989; 8:
31–36.
21. Ribeiro da Silva MDMC, Santos LMNBF, Silva ALR, Fernandes
O, Acree WE, Jr. J. Chem. Thermodynam. 2003; 35: 1093–1100.
22. Coops J, Jessup RS, Van Nes K. In Experimental Thermochemistry,
Vol. 1, Chapter 3, Rossini FD (ed). Interscience: New York, 1956.
23. The NBS Tables of Chemical Thermodynamic Properties. J. Phys.
Chem. Ref. Data 1982; 11: 1–392 (Supplement No. 2).
43. Cox JD, Wagman DD, Medvedev VA. CODATA Key Values for
Thermodynamics. Hemisphere: New York, 1989.
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 491–498
DOI: 10.1002/poc