Arylazide cycloaddition to methyl propiolate: DFT-based quantitative prediction of regioselectivity

Article abstract of DOI:10.1002/chem.200204681

Several 1-(4-substituted)-phenyl-4- or 5-methoxycarbonyl-1,2,3-triazoles have been synthesized by 1,3-dipolar cycloaddition of the corresponding arylazides to methyl propiolate in carbon tetrachloride. The regioselectivity of these reactions cannot be rat

Full text of DOI:10.1002/chem.200204681

FULL PAPER
Arylazide Cycloaddition to Methyl Propiolate:
DFT-Based Quantitative Prediction of Regioselectivity
Giorgio Molteni[b] and Alessandro Ponti*[a]
Abstract: Several
1-(4-substituted)-
applied to this problem a quantitative
formulation of the HSAB principle to
this problem developed within density
functional theory. Global and local re-
activity indices were computed at
B3LYP/6-311 ‡ G(d,p) level both in va-
cuo and in carbon tetrachloride (by the
COSMO approach). The direction of
charge transfer upon reactive encounter
has been determined and the computed
regioselectivity has been shown to be in
good agreement with the experimental
results. The relationship between com-
puted and experimental data and how it
phenyl-4- or 5-methoxycarbonyl-1,2,3-
triazoles have been synthesized by 1,3-
dipolar cycloaddition of the correspond-
ingarylazides to methyl propiolate in
carbon tetrachloride. The regioselectiv-
ity of these reactions cannot be ration-
alized on the basis of the electronic
demands of the reactants or frontier
molecular-orbital theory. Therefore, we
Keywords: azides ¥ cycloaddition ¥
density functional calculations
¥
HSAB principle ¥ regioselectivity
ized within the framework of the density functional theory
(DFT).[11] Well-known examples[12] are the electron chemical
potential m, which represents the escapingtendency of
molecular electrons, and the molecular softness S, which is
the sensitivity of the total number of electrons to a change in
m. Within DFT, any reaction can be considered as split in two
steps:[13, 14] 1) as soon as reactant molecules approach each
other, they form a weakly interacting, promoted complex,[15]
whereby charge is transferred between the reactants in order
to equalize the electron chemical potential at constant
external potential; 2) a charge reshuffling at constant electron
chemical potential occurs by which the promoted complex
evolves toward the product(s) or back to the reactants (see
Figure 1). If one assumes that the second step can be
neglected, it can be shown that the most favorable situation
occurs when the reactants have equal softness. This is the DFT
formulation[13, 16] of Pearson×s hard soft acid base (HSAB)
principle.[17]
However, to study regioselection a local (atomic) reactivity
index is needed. The best suited is local softness s(r),[12] which
represents the sensitivity of the molecular electron density at
point r to a change in m. A local HSAB principle[18] can then be
devised: a regioisomer is favored when the new bonds form
between atoms with equal softness. The local HSAB principle
has provided many reliable qualitative prediction of regiose-
lectivity for 1,3-dipolar cycloadditions (1,3-DC).[19] Since in
1,3-DCs the relative energy of transition states is paralleled by
the relative energy of the weakly interacting complexes
formingin the early stage of the reaction [10] (cf. Figure 1), the
neglect of the charge reshuffling term is reasonable.[20]
Moreover, such successful predictions suggest that the early
interaction energy between the reactants should be closely
Introduction
The first example of arylazide cycloaddition to unsaturated
compounds appeared over a century ago, when Arthur
Michael reacted phenylazide with dimethylacetylene dicar-
boxylate.[1] Seventy years later, Huisgen fully investigated the
mechanism, scope and limitations of azide cycloadditions,[2]
which was recognized as the choice method for the direct
synthesis of 1,2,3-triazoles.[3] These and other azoles[4] were
the object of recent important investigation leading to the
™click∫ chemistry approach.[5] 1,2,3-Triazoles may display a
wide range of biological activity, such as anti-HIV[6] and anti-
microbial[7] agents, as well as selective b3 adrenergic receptor
agonists.[8] Due to these attractive activities, new insights
about the factors that influence the regiochemical output for
these compounds are of interest. FollowingHouk×s find-
ings,[9, 10] arylazide cycloaddition to methyl propiolate can be
controlled from both the HOMO and the LUMO of the 1,3-
dipole, and this implies that the observed regioselectivities are
difficult to rationalize on the grounds of simple FMO theory.
Recently, many important concepts and indices useful for
the understandingof chemical reactivity have been rational-
[a] Dr. A. Ponti
Consiglio Nazionale delle Ricerche
Istituto di Scienze e Tecnologie Molecolari
via Camillo Golgi 19, 20133 Milano (Italy)
Fax : (‡39)02-5031-4300
E-mail: a.ponti@istm.cnr.it
[b] Dr. G. Molteni
¡
Universita degli Studi di Milano
Dipartimento di Chimica Organica e Industriale
via Camillo Golgi 19, 20133 Milano (Italy)
2770
¹ 2003 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
DOI: 10.1002/chem.200204681
Chem. Eur. J. 2003, 9, 2770 2774

2770 2774
Figure 1. Energy profile of a prototypical reaction leading to several
products (e.g. regioisomers). Reactants R form weakly interacting pro-
moted complexes Ci which evolve to the correspondingtransition states T i
and then to the products (not shown). If the local HSAB principle applies,
the energy differences between the Ci×s are proportional to the corre-
spondingenergy differences between the T i×s.
Scheme 1.
related to the transition state energy. On these grounds, a
generalization of the local HSAB principle has been recently
introduced,[21] which enables one to compute, from m and s of
the reactants only, the grand potential variation DW[22] due to
the charge transfer occurring in the very first step of the bond-
forminginteraction between specific atoms of the reactants.
For the above reasons, DW is expected to be proportional to
the transition state energy and to provide a quantitative
prediction of regioselectivity without the need of locating the
transition state. This method has been successfully applied by
us to the 1,3-DC between nitrilimines and alkynyl- or alkenyl
dipolarophiles.[23] Continuingour investigations, we present
here the first quantitative prediction of the regioselectivity
involved in the cycloaddition between 1-(4-substituted)phe-
nylazides 2 and methyl propiolate 3, which is based upon the
DFT theory and the HSAB principle.
Table 1. Experimental yields and 4/5 yield ratios of the cycloaddition
between arylazides 2 and methyl propiolate 3 in refluxingCCl
.
4
R
Yields[a] [%]
Yield ratio[b]
4 ‡ 5
4/5
a
b
c
d
e
f
H
> 96
> 96
91
95
93
75:25
73:27
68:32
70:30
68:32
55:45
Me
MeO
F
Cl
NO2
> 96
1
[a] Isolation yields. [b] Deduced from H NMR of reaction crudes.
Table 2. Results of B3LYP/6-311 ‡ G(d,p) calculations either in vacuo or in CCl4
(COSMO model). Electron chemical potential difference between arylazides 2
and methyl propiolate 3 alongwith dDW difference[a] and predicted 4/5 yield ratio
for their mutual cycloaddition.
Vacuum
CCl4
R
m(2) À m(3) dDW
Predicted m(2) À m(3) dDW
Predicted
Ratio[b] 4/5
[eV]
[kJmolÀ1
]
Ratio[b] 4/5 [eV]
[kJmolÀ1
]
Results and Discussion
H
1.04
À 1.60
À 0.79
À 1.14
À 1.16
À 1.07
0.22
74:26
66:34
69:31
70:30
69:31
54:46
1.14
1.28
1.46
1.08
1.06
À 1.81
À 1.30
À 1.35
À 1.31
À 0.55
À 0.02
74:26
69:31
69:31
69:31
70:30
54:46
Me[c]
MeO
F
1.19
1.35
0.97
0.97
The 1,3-dipolar species 2 were prepared from the correspond-
inganilines by diazotization, followed by addition to sodium
azide (Scheme 1). The subsequent cycloadditions were per-
formed by refluxing 2 in dry CCl4 in the presence of an
equimolecular amount of 3. Reaction time, product yields,
and yield ratios are summarized in Table 1. The structures of
regioisomeric 1-(4-substituted)phenyl-4-methoxycarbonyl-
1,2,3-triazoles (4) and 1-(4-substituted)phenyl-5-methoxycar-
bonyl-1,2,3-triazoles (5) were unambiguously determined
through analytical and spectroscopic data. In particular, as
far as H NMR spectra are concerned, the protons in the 4-
and 5-positions of the 1,2,3-triazole ringshow resonances that
are in perfect agreement with literature data.[24
The main results of DFT calculations at the B3LYP/6-311 ‡
G(d,p) level both in vacuo and in CCl4 are reported in Table 2.
The electron chemical potential difference between 2 and 3
determines the direction of the overall charge flow upon
interaction of the reactants, as electrons flow towards regions
Cl
NO2 À 0.21
À 0.03
[a] Difference in grand potential variation for the pathways leading to triazoles 4
and 5. [b] From computed dDW and Equations (2) and (3) as appropriate;
uncertainty Æ1%. [c] Excluded from regression.
at low electron chemical potential m. It turns out that, both in
vacuo and in CCl4, 2a e act as nucleophiles, whereas 2 f acts
as an electrophile (note, however, that the chemical potentials
of 2 f and 3 are close to each other, especially in solution). The
ꢀ0.1 eV increase of the m(2) À m(3) difference in CCl4 in
mostly due to the decrease of m(2); m(2 f) À m(3) is slightly
larger because of a comparable decrease in m(2 f). Our
calculations thus show that in the reaction of 3 with 2a
charge flows from the arylazide to methyl propiolate, whereas
FMO theory is not able to provide a clear-cut prediction, since
1
26]
e
Chem. Eur. J. 2003, 9, 2770 2774
www.chemeurj.org
¹ 2003 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
2771

A. Ponti and G. Molteni
FULL PAPER
the two possible HOMO-LUMO interactions are comparable
in size.
We now turn to regioselectivity prediction. As selectivity
criterion, we used the grand potential change due to two
bond-forminginteractions between 2 and 3, because of the
general agreement about the concertedness of 1,3-DC reac-
tions. The grand potential change for the pathway leading to
4-methoxycarbonyl-1,2,3-triazole 4 is given in Equation (1):
2
À1
1
DW(4) ˆ À ³2[m(2) À m(3)] {s(N1)s(C1)[s(N1) ‡ s(C1)]
(1)
‡ s(N3)s(C2)[s(N3) ‡ s(C2)]À1
}
Here the atoms are numbered as in Scheme 1. The grand
potential variation for DW(5) can be obtained by exchanging
s(C1) and s(C2). The difference dDWˆ DW(4) À DW(5) is
reported in Table 2. The negative sign of dDW shows that
cycloadduct 4 is the major one, in line with experimental
results. Note that dDW of the reaction 2 f ‡ 3 is negative only
in the solution calculation, thus showingthe beneficial effect
of solvent inclusion. The low regioselectivity of the reaction of
2 f with 3 is due to two concurrent causes. First, m(2 f) and m(3)
are so close that DW (and hence dDW) is at least one order of
magnitude smaller than in the remaining cases. Besides, the
Figure 2. Linear relationship between the computed difference dDW in
grand potential variation for the pathways leading to cycloadducts 4 and 5
and the correspondingdifference in activation energy dDE=, computed
from the experimental 4/5 yield ratio. The error bars show the uncertainty
in dDE= due to the error in yield ratio, estimated at 1%. dDW computed in
vacuo: circles and dashed regression line; dDW computed in CCl4 by the
COSMO solvation model: squares and solid regression line. Open symbols
denote the 2b (R ˆ Me) ‡ 3 reaction which has been excluded from both
regressions.
À
grand potential change due to the formation of the N1 C1
and comes at the border of statistical significance; the
À
bond is very close to that for the N3 C1 bond, so that dDW is
À1
remaining0.3 kJmol
term might vanish on improving the
mostly due to the energy difference of a single bond, that is,
that formingabout C2 of 3. Amongthe currently studied
cycloadditions, this is the only case in which the difference in m
has a significant effect on the regioselectivity.
We now proceed one step further by demonstratingthat
dDW is a quantitative regioselectivity index for 1,3-DC
reactions. The difference in activation energy dDE= of the
two reaction paths can be obtained as dDE=ˆ ÀRTlog(Y),
whereby Tˆ 350 K is the reaction temperature, and Y is the
experimental 4/5 ratio. Estimatingthe error in Y at Æ1% and
considering 2b as an outlier (vide infra), weighted least-
squares linear regression results in Equations (2) and (3):
solvation model. The slopes in Equations (2) and (3) are
almost independent of solvation. Their values mean that the
difference in transition state energy is about 30% larger than
the energy difference between the promoted complexes.
Although this is in line with theory, we were prompted to
check the main approximation used, namely, the neglect of
the constant electron potential term in Equation (1), since we
previously obtained a nearly equality of dDW and dDE= in the
case of 1,3-DCs involvingnitrilimines. To this end, we
computed the contribution of the step at constant electron
chemical potential [Eq. (4)]:
À1
DWm(4) ˆ À ³2 l{[s(N1) ‡ s(C1)] ‡ [s(N3) ‡ s(C2)]À1
}
(4)
1
dDWˆ (0.73 Æ 0.07)dDE=‡ (0.6 Æ 0.1) kJmolÀ1
dDWˆ (0.71 Æ 0.07)dDE= ‡ (0.3 Æ 0.1) kJmolÀ1
1 ˆ 0.99 (in vacuo) (2)
1 ˆ 0.98 (in CCl4) (3)
Here DWm(5) can be obtained by exchanging s(C1) and
s(C2), and l is a positive parameter related to an effective
number of valence electrons. As the value of l is not precisely
set by theory, we carried out a bilinear regression of dDE=
with dDW and dDWm, whereby l was considered a parameter
to be optimized. Such a regression did not show any improve-
ment over the linear one in Equation (2). This confirms that in
early transition state reactions, such as the 1,3-DCs, the
relative transition-state energy depends only on the relative
energy of the promoted complex formed by charge transfer in
the chemical-potential equalization step at the beginning of
the reactive encounter.
The experimental selectivity towards 4b is higher than that
predicted by computation. The reason for this discrepancy is a
too negative value of DW(5). Dissectingthe latter into bond
contributions, it turns out that the too large stabilization of the
5b is caused by the particularly favorable interaction between
N3 and C1. This in turn arises from a large computed softness
at N3 of 2b. Therefore, the wrongprediction of the selectivity
towards 4b is mostly due to an incorrect evaluation of the
Here 1 is the linear correlation coefficient (see Figure 2).
The predicted 4/5 ratios (Table 2), obtained from computed
dDW values [Eqs (2) and (3)], are in very good agreement
with the experimental values (Table 1), except for 2b. Figure 2
and Table 2 also show that regioselectivity data cannot be
rationalized only on the basis of the electron demand of R of
2, for example, by usingHammett s parameters. This is best
illustrated by the fact that maximum regioselectivity is
observed when R ˆ H, which is the zero of the usual
electron-demand scale. A satisfactory description of regiose-
lectivity must take into account local variations of charge
density due to the interaction between reactants.
The positive intercept in the above equations implies that
there is a slight preference towards the major product 4, since
dDE=< 0 when dDWˆ 0. Such preference is independent of
the specific interaction between the reactants. However, this
constant term halves when the solvent is taken into account
2772
¹ 2003 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2003, 9, 2770 2774

Arylazide Cycloaddition
2770 2774
‡
nÄ ˆ 1740 cmÀ1; MS: m/z: 217 [M ]; elemental analysis calcd (%) for
charge on the terminal nitrogen of 2b. Figure 2, however,
shows that the situation is somewhat improved when solvation
is taken into account thanks to a better evaluation of the
charge rearrangement about N3 of 2b upon electrophilic
attack.
C11H11N3O2: C 60.82, H 5.10, N 19.34; found: C 60.88, H 5.14, N 19.28.
Compound 5c: Pale yellow solid, m.p. 98 1008C; 1H NMR (300 MHz,
CDCl3): d ˆ 3.88 (3H, s), 3.99 (3H, s), 6.90 7.40 (4H, m), 8.24 ppm (1H, s);
13C NMR (75 MHz, CDCl3): d ˆ 52.40 (q), 55.60 (q), 113.71 (d), 122.37 (d),
127.13 (d), 129.16 (s), 157.81 (s), 160.32 ppm (s); IR (Nujol): nÄ ˆ 1735 cmÀ1
;
‡
We have thus shown that the combined use of DFT
reactivity indices of the reactants with the local HSAB
principle provides quantitative rationalization of regioselec-
tivity for a series of 1,3-DCs not amenable to FMO and
electron-demand theory. Indeed, the electron demand of the
substituent R only affects the electron chemical potential m,
which cannot fully account for regioselectivity. This can be
adequately rationalized only when the substituent effect on
the sensitivity of the electron density at the reactive atoms is
taken into account. Other advantages of the DFT-HSAB
approach are that transition states need not to be located and
that insight into the details of atom-atom interactions can be
easily obtained. The inclusion of solvation effects leads to a
tiny worseningof the correlation of dDE= with dDW, but
significantly improves the description of individual cases, such
as 2b ‡ 3 and 2 f ‡ 3, which is less satisfactory when the
systems are treated in vacuo.
MS: m/z: 233 [M ]; elemental analysis calcd (%) for C11H12N3O3: C 56.65,
H 4.75, N 18.02; found: C 56.68, H 4.78, N 17.92.
1
Compound 5d: White solid, m.p. 79 808C; H NMR (300 MHz, CDCl3):
d ˆ 3.86 (3H, s), 7.20 7.70 (4H, m), 8.26 ppm (1H, s); 13C NMR (75 MHz,
CDCl3): d ˆ 51.22 (q), 115.60 (d), 121.48 (d), 125.41 (d), 127.31 (s), 159.45
(s), 162.76 ppm (s); 19F NMR (282 MHz, CDCl3): d ˆ À110.51 ppm; IR
‡
(Nujol): nÄ ˆ 1740 cmÀ1; MS: m/z: 221 [M ]; elemental analysis calcd (%)
for C10H8FN3O2: C 54.30, H 3.65, N 19.00; found: C 54.36, H 3.67, N 19.07.
Compound 5e: Pale yellow solid, m.p. 91 938C; 1H NMR (300 MHz,
CDCl3): d ˆ 3.87 (3H, s), 7.50 7.70 (4H, m), 8.27 ppm (1H, s); 13C NMR
(75 MHz, CDCl3): d ˆ 52.07 (q), 121.96 (d), 125.46 (d), 129.92 (d), 132.21
(s), 140.17 (s), 157.68 (s), 161.28 ppm (s); IR (Nujol): nÄ ˆ 1735 cmÀ1; MS: m/
‡
z: 249 [M ]; elemental analysis calcd (%) for C10H8ClN3O2: C 52.92, H
3.23, N 16.83; found: C 52.88, H 3.19, N 16.87.
Computational methods: DFT calculations were performed with the
Gaussian 98[27] program suite by means of a Beowulf PC cluster. The
hybrid B3LYP functional was employed with the standard 6-311 ‡ G(d,p)
basis set. The geometry of 2a f and 3 in vacuo was fully optimized and
characterized with vibrational analysis at the same level of theory. The
anion and cation of 2a f and 3 were treated at the UB3LYP/6-311 ‡
G(d,p) level by usingthe geometry of the neutral systems. Calculations
of the solvated systems were carried out by the COSMO model[28] with
dielectric constant e ˆ 2.228 at the in vacuo geometry. The COSMO
approach describes the solvent reaction field by means of apparent
polarization charges distributed on the surface of the cavity in which the
solute molecule is embedded. Atomic electron populations were evaluated
followingthe Merz Kollman scheme [29] (includingfittingof atom-centered
dipoles). This scheme, which already proved to be reliable,[30] has been used
in most DFT calculations of regiochemistry of 1,3-DCs, so that our results
can be directly compared with existingliterature. It has also been recently
considered as an appropriate local descriptor of charge.[31] Reactivity
indices were computed within the finite difference approximation:[11] m ˆ
À(IP ‡ EA)/2 and S ˆ (IP-EA)À1, whereby IP and EA are the (vertical)
ionization potential and electron affinity, respectively. The local softness s
Experimental Section
General methods. Meltingpoints were determined in open tubes and are
uncorrected. IR spectra were recorded with a FTIR spectrophotometer.
Mass spectra were determined with a 70 eV EI apparatus. 1H NMR
(300 MHz), 13C NMR (75 MHz), and 19F NMR (282 MHz) spectra were
taken in CDCl3 at 297 K. Chemical shifts are given as ppm from
tetramethylsilane (hexafluorobenzene for 19F NMR), J values are given
in Hz.
Compounds 4a,[24] 4b,[25] 4 f,[26] 5a,[24] and 5 f[26] are known in the literature.
(condensed to each individual atom[32]) was computed as s ˆ S [p(N0 ‡
‡
Cycloaddition between arylazides 2 and methyl propiolate 3: General
procedure: A solution of 2 (5.0 mmol) and 3 (0.43 g, 5.0 mmol) in dry
carbon tetrachloride (25 mL) was refluxed for the time indicated in Table 1.
Evaporation of the solvent in vacuo gave a residue which was separated by
chromatography on a silica gel column with ethyl acetate/hexane 1:2. Major
4-methoxycarbonyl-1,2,3-triazole 4 was eluted first, followed by minor
5-methoxycarbonyl-1,2,3-triazole 5. Crystallization from diisopropyl ether
gave analytically pure 4 and 5.
1) À p(N0)] for electrophiles and as sÀ ˆ S [p(N0) À p(N0 À 1)] for nucleo-
philes, whereby p(N), N ˆ N0 À 1, N0, N0 ‡ 1, was the atomic electron
population of the cationic, neutral, and anionic system, respectively.
Acknowledgement
Compound 4c: Pale yellow solid, m.p. 93 958C; 1H NMR (300 MHz,
CDCl3): d ˆ 3.84 (3H, s), 3.88 (3H, s), 7.00 7.60 (4H, m), 8.42 ppm (1H, s);
13C NMR (75 MHz, CDCl3): d ˆ 52.20 (q), 55.51 (q), 114.23 (d), 126.86 (d),
A.P. is grateful to G. Mezza (CNR ISTM) for his expert assistance with
Beowulf cluster hardware and software.
138.02 (d), 127.13 (s), 158.22 (s), 160.67 ppm (s); IR (Nujol): nÄ ˆ 1730 cmÀ1
;
‡
MS: m/z: 233 [M ]; elemental analysis calcd (%) for C11H12N3O3: C 56.65,
H 4.75, N 18.02; found: C 56.70, H 4.77, N 17.94.
[1] A. J. Michael, Prakt. Chem. 1893, 48, 94.
[2] R. Huisgen, in 1,3-Dipolar Cycloaddition Chemistry, Vol. 1, Wiley-
Interscience, New York, 1984, pp. 1 176.
1
Compound 4d: White solid, m.p. 79 808C; H NMR (300 MHz, CDCl3):
d ˆ 3.99 (3H, s), 7.20 7.60 (4H, m), 8.49 ppm (1H, s); 13C NMR (75 MHz,
CDCl3): d ˆ 50.71 (q), 114.41 (d), 126.64 (d), 136.63 (d), 127.36 (s), 160.18
(s), 163.32 ppm (s); 19F NMR (282 MHz, CDCl3): d ˆ À111.11 ppm; IR
[3] E. F. V. Scriven, K. Turnbull, Chem. Rev. 1988, 88, 297; W. Lwowsky, in
1,3-Dipolar Cycloaddition Chemistry, Vol. 1, Wiley-Interscience, New
York, 1984, pp. 621 627; The Chemistry of the Azido Group (Ed.: S.
Patai), Wiley, London, 1971.
[4] Z. P. Demko, K. B. Sharpless, Org. Lett. 2001, 3, 4091; Z. P. Demko,
K. B. Sharpless, Org. Lett. 2002, 4, 2525; Z. P. Demko, K. B. Sharpless,
Angew. Chem. 2002, 114, 2214; Angew. Chem. Int. Ed. 2002, 41, 2110;
Z. P. Demko, K. B. Sharpless, Angew. Chem. 2002, 114, 2217; Angew.
Chem. Int. Ed. 2002, 41, 2113; F. Himo, Z. P. Demko, L. Noodleman,
K. B. Sharpless, J. Am. Chem. Soc. 2002, 124, 12210.
‡
(Nujol): nÄ ˆ 1730 cmÀ1; MS: m/z: 221 [M ]; elemental analysis calcd (%)
for C10H8FN3O2: C 54.30, H 3.65, N 19.00; found: C 54.34, H 3.69, N 19.06.
Compound 4e: Pale yellow solid, m.p. 98 998C; 1H NMR (300 MHz,
CDCl3): d ˆ 4.00 (3H, s), 7.00 7.20 (4H, m), 8.49 ppm (1H, s); 13C NMR
(75 MHz, CDCl3): d ˆ 52.37 (q), 114.09 (d), 126.14 (d), 128.12 (s), 138.39
(d), 159.36 (s), 162.2 ppm (s); IR (Nujol): nÄ ˆ 1730 cmÀ1; MS: m/z: 249
‡
[M ]; elemental analysis calcd (%) for C10H8ClN3O2: C 52.92, H 3.23, N
16.83; found: C 52.97, H 3.26, N 16.78.
1
Compound 5b: White solid, m.p. 85 878C; H NMR (300 MHz, CDCl3):
[5] H. C. Kolb, M. G. Finn, K. B. Sharpless, Angew. Chem. Int. Ed. 2001,
113, 2056; Angew. Chem. Int. Ed. 2001, 40, 2004.
[6] S. Velazquez, R. Alvarez, C. Perez, F. Gago, C. De, J. Balzarini, M. J.
Camarasa, Antiviral Chem. Chemother. 1998, 9, 481; R. Alvarez, S.
d ˆ 2.41 (3H, s), 3.84 (3H, s), 7.30 7.70 (4H, m), 8.24 ppm (1H, s);
13C NMR (75 MHz, CDCl3): d ˆ 20.82 (q), 52.09 (q), 120.58 (d), 125.33 (d),
129.93 (d), 133.76 (s), 139.46 (d), 158.01 (s), 160.71 ppm (s); IR (Nujol):
Chem. Eur. J. 2003, 9, 2770 2774
www.chemeurj.org
¹ 2003 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
2773

A. Ponti and G. Molteni
FULL PAPER
Velazquez, F. San, S. Aquaro, C. De, C. Perno, A. Karlsson, J.
Balzarini, M. J. Camarasa, J. Med. Chem. 1994, 37, 4185.
[7] M. J. Genin, D. A. Allwine, D. J. Anderson, M. R. Barbachyn, D. E.
Emmert, S. A. Garmon, D. R. Graber, K. C. Grega, J. B. Hester, D. K.
Hutchinson, J. Morris, R. J. Reischer, C. W. Ford, G. E. Zurenko, J. C.
Hamel, R. D. Schaadt, D. Stapert, Y. B. H., J. Med. Chem. 2000, 43,
953.
[21] A. Ponti, J. Phys. Chem. A 2000, 104, 8843.
[22] The grand potential is the natural thermodynamic quantity to describe
the behavior of the reactants× atoms, which are open subsystems freely
exchanging energy and electrons.
[23] A. Ponti, G. Molteni, J. Org. Chem. 2001, 66, 5252; G. Molteni, A.
Ponti, M. Orlandi, New J. Chem. 2002, 26, 1340; A. Ponti, G. Molteni,
New J. Chem. 2002, 26, 1346.
[8] L. L. Brockunier, E. R. Parmee, H. O. Ok, M. R. Candelore, M. A.
Cascieri, L. F. Colwell, L. Deng, W. P. Feeney, M. J. Forrest, G. Hom,
J., D. E. MacIntyre, L. Tota, M. J. Wywratt, M. H. Fisher, A. E. Weber,
Bioorg. Med. Chem. Lett. 2000, 10, 2111.
[9] K. N. Houk, J. Sims, R. E. Duke, R. W. Strozier, J. K. George, J. Am.
Chem. Soc. 1973, 95, 7287.
[10] K. N. Houk, J. Sims, C. R. Watts, L. J. Luskus, J. Am. Chem. Soc. 1973,
95, 7301.
[11] R. G. Parr, W. Yang, Density Functional Theory of Atoms and
Molecules, Oxford University Press, Oxford, 1989.
[24] R. K. Huisgen, L. Mˆbius, G. Szeimies, Chem. Ber. 1965, 98, 4014.
[25] C. Wentrup, V. V. Ramana Rao, W. Frank, B. E. Fulloon, D. W.
Moloney, T. Mosandl, J. Org. Chem. 1999, 64, 3608.
[26] R. Fusco, G. Bianchetti, D. Pocar, R. Ugo, Gazz. Chim. Ital. 1962, 92,
1040.
[27] Gaussian 98 (Revision A.11.3), M. J. Frisch, G. W. Trucks, H. B.
Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, V. G. Zakr-
zewski, J. J. A. Montgomery, R. E. Stratmann, J. C. Burant, S.
Dapprich, J. M. Millam, A. D. Daniels, K. N. Kudin, M. C. Strain, O.
Farkas, J. Tomasi, V. Barone, M. Cossi, R. Cammi, B. Mennucci, C.
Pomelli, C. Adamo, S. Clifford, J. Ochterski, G. A. Petersson, P. Y.
Ayala, Q. Cui, K. Morokuma, N. Rega, P. Salvador, J. J. Dannenberg,
D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman, J.
Cioslowski, J. V. Ortiz, A. G. Baboul, B. B. Stefanov, G. Liu, A.
Liashenko, P. Piskorz, I. Komaromi, R. Gomperts, R. L. Martin, D. J.
Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, M.
Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong, J. L.
Andres, C. Gonzalez, M. Head-Gordon, E. S. Replogle, J. A. Pople,
Gaussian, Inc., Pittsburgh, 2002.
[28] V. Barone, M. Cossi, J. Phys. Chem. A 1998, 102, 1995.
[29] U. C. Singh, P. A. Kollman, J. Comput. Chem. 1984, 5, 129; B. H.
Besler, K. M. Merz, Jr., P. A. Kollman, J. Comput. Chem. 1990, 11,
431.
[30] F. De Proft, J. M. L. Martin, P. Geerlings, Chem. Phys. Lett. 1996, 256,
400.
[12] H. Chermette, J. Comput. Chem. 1999, 20, 129.
[13] P. K. Chattaraj, H. Lee, R. G. Parr, J. Am. Chem. Soc. 1991, 113, 1855.
¬
[14] A. Toro-Labbe, J. Phys. Chem. A 1999, 103, 4398.
[15] J. L. Gazquez, J. Phys. Chem. A 1997, 101, 9464.
[16] A. Cedillo, P. K. Chattaraj, R. G. Parr, Int. J. Quant. Chem. 2000, 77,
403.
[17] R. G. Pearson, J. Am. Chem. Soc. 1963, 85, 3533.
¬
¬
[18] J. L. Gazquez, F. Mendez, J. Phys. Chem. 1994, 98, 4591.
[19] A. K. Chandra, M. T. Nguyen, J. Comput. Chem. 1998, 19, 195; A. K.
Chandra, M. T. Nguyen, J. Phys. Chem. A 1998, 102, 6181; F. Mendez,
J. Tamariz, P. Geerlings, J. Phys. Chem. A 1998, 102, 6292; T. N. Le,
L. T. Nguyen, A. K. Chandra, F. De Proft, P. Geerlings, M. T. Nguyen,
J. Chem. Soc. Perkin Trans. 2 1999, 1249; A. K. Chandra, T. Uchimaru,
M. T. Nguyen, J. Chem. Soc. Perkin Trans. 2 1999, 2117; P. Geerlings, F.
De Proft, Int. J. Quant. Chem. 2000, 80, 227; L. T. Nguyen, F. De Proft,
A. K. Chandra, T. Uchimaru, M. T. Nguyen, P. Geerlings, J. Org.
Chem. 2001, 66, 6096.
[31] P. K. Chattaraj, J. Phys. Chem. A 2001, 105, 511.
[32] W. Yang, W. J. Mortier, J. Am. Chem. Soc. 1986, 108, 5708.
[20] Of course, the reshufflingterm may be needed in other reactions
where the transition state is not early. For instance see: S. Damoun, G.
Van de Woude, K. Choho, P. Geerlings, J. Phys. Chem. 1999, 103, 7861;
S. Pal, K. R. S. Chandrakumar, J. Am. Chem. Soc. 2000, 122, 4145.
Received: December 18, 2002 [F4681]
2774
¹ 2003 Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2003, 9, 2770 2774