Enantiopure furo[3,4-c]pyrazole derivatives by intramolecular nitrilimine cycloaddition: a stereoselectivity rationale based upon MP2 calculations

Article abstract of DOI:10.1016/j.tetasy.2008.05.007

Silver carbonate treatment of hydrazonoyl chloride 4 promoted the in situ generation of the corresponding nitrilimine bearing a stereocentre at the α-position of the ethylenic dipolarophile. Intramolecular cycloaddition of the latter intermediate involves the formation of 4-(S)-methyl-6-oxo-3,3a,4,5-tetrahydro-furo[3,4-c]pyrazole derivatives with very good yield and diastereoselectivity. Full rationalization of the experimentally observed stereoselectivity has been pursued by means of MP2 calculations.

Full text of DOI:10.1016/j.tetasy.2008.05.007

Tetrahedron: Asymmetry 19 (2008) 1381–1384
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
Enantiopure furo[3,4-c]pyrazole derivatives by intramolecular nitrilimine
cycloaddition: a stereoselectivity rationale based upon MP2 calculations
b
Giorgio Molteni a, , Alessandro Ponti
*
a Università degli Studi di Milano, Dipartimento di Chimica Organica e Industriale, Via Golgi 19, 20133 Milano, Italy
b Consiglio Nazionale delle Ricerche, Istituto di Scienze e Tecnologie Molecolari, Via Golgi 19, 20133 Milano, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Silver carbonate treatment of hydrazonoyl chloride 4 promoted the in situ generation of the correspond-
Received 28 April 2008
Accepted 13 May 2008
Available online 10 June 2008
ing nitrilimine bearing a stereocentre at the
a-position of the ethylenic dipolarophile. Intramolecular
cycloaddition of the latter intermediate involves the formation of 4-(S)-methyl-6-oxo-3,3a,4,5-tetrahy-
dro-furo[3,4-c]pyrazole derivatives with very good yield and diastereoselectivity. Full rationalization of
the experimentally observed stereoselectivity has been pursued by means of MP2 calculations.
Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction
nale since it was based upon the examination of stereomodels,
which mimic the proposed transition structures. Thus, a validation
The synthesis of annulated pyrazoles based on the intramole-
cular cycloadditions of C-substituted nitrilimines was pioneered
by Garanti and Zecchi.1 This route disclosed the opportunity to
obtain annulated 4,5-dihydropyrazole derivatives of relevance in
organic synthesis,2 as well as some related products, which have
received interesting biological applications.3 In recent times, our
main interest in intramolecular 1,3-dipolar cycloadditions of C-
substituted nitrilimines focused on the behaviour of homochiral
substrates towards stereoselection.4 Thus, we were able to exploit
this strategy as the key step of synthetic sequences leading to var-
ious kinds of annulated 4,5-dihydropyrazoles in enantiopure form,
which are not easily accessible by other routes.5 As far as stereo-
selectivity is concerned, we first described the synthesis of the pyr-
rolo[3,4-c]pyrazole skeleton in the enantiopure form by means of
stereoselective cycloaddition of a homochiral nitrilimine bearing
of the stereochemical model, which works in the intramolecular
cycloadditions of C-substituted nitrilimines, needs to be formu-
lated. In order to achieve this goal, we herein report the behaviour
of the novel C-substituted-nitrilimine intermediate 5, which is
derived from (S)-(+)-but-3-en-2-ol and hence bears the stereocen-
tre in the
a-position of the ethylenic dipolarophile, leading to
enantiopure 4-(S)-methyl-6-oxo-3,3a,4,5-tetrahydro-furo[3,4-c]-
pyrazole derivatives 6. The cycloaddition stereoselectivity outcome
was fully investigated through a computational study of the diaste-
reoisomeric transition structures at the MP2 level.
2. Results and discussion
Our synthetic sequence starts from (S)-(+)-but-3-en-2-ol 1,
which was devised as a suitable starting chiral building block for
our synthesis. Following literature procedures,8 enantiopure (S)-1
was prepared by yeast reduction of 3-chlorobutan-2-one
the L-alanine benzylester pendant as a chiral auxiliary at the c-
position with respect to the nitrilimine moiety.6 However, since
control of the absolute stereochemistry was really modest, we
reasoned that the large distance between the starting stereocentre
and the new one could be responsible for the observed and disap-
pointing outcome. To circumvent this drawback, we planned an-
(½a 2D5
¼ þ25:9, lit. +25.4). Hydrazonoyl chloride 4, which consti-
ꢀ
tutes as the precursor of nitrile imine 5, was synthesized from
(S)-1 following a well-established procedure previously elaborated
by Garanti and Zecchi9 (Scheme 1). The in situ generation of the la-
bile intermediate 5 was accomplished by treating the correspond-
ing hydrazonoyl chloride 4 with a twofold molar excess of silver
carbonate in dry dioxane at room temperature.10 Diastereoiso-
meric 4-(S)-methyl-6-oxo-3,3a,4,5-tetrahydro-furo[3,4-c]pyrazoles
6 were formed after 96 h with 90% overall yield and 91:9
diastereoisomeric ratio. These cycloadducts were obtained as ana-
lytically (and enantiopure) solids after silica gel column chroma-
tography followed by crystallization. Spectral data of products 6
were in good agreement with those reported for similar furo[3,4-
c]pyrazoles.6,11 The absolute configuration of the newly formed
other
a-aminoester-based synthesis of C-substituted nitrilimines,
in which the stereocentre was placed at the
a-position with re-
spect to the dipolar function. As a result, much better cycloaddition
stereoselectivities were observed, which are superimposable to
those obtained from analogous C-substituted nitrilimines synthe-
sized from enantiopure
a
-hydroxyesters.7 However, in all these
cases, the observed diastereoselectivity found a very rough ratio-
* Corresponding author. Tel.: +39 02 50314139; fax: +39 02 50314141.
E-mail address: giorgio.molteni@unimi.it (G. Molteni).
0957-4166/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2008.05.007

1382
G. Molteni, A. Ponti / 19 (2008) 1381–1384
O
O
O
O
O
O
O
SO2Cl2
0-5°C
OH
O
O
xylene, Δ
Cl
3
1
2
_
N
O
O
O
+
_
Ph
N
O
+
N
N
Ag2CO3
Ph
Cl
PhNH
N
O
0-5°C
dioxane
Cl
4
5
O
O
1
N
N
2
(90%)
6
5
O
+
N
N
Ph
Ph
4
3
H
H
91
:
9
(3aS,4S)-6
(3aR,4S)-6
Scheme 1.
stereocentre of minor (3aR,4S)-6 was determined by the mutual
NOE enhancements between the methyl hydrogens and the hydro-
gen in the 3a- position of the furo[3,4-c]pyrazole skeleton (see
Fig. 1d). It is apparent that the observed NOE effect can be opera-
tive only in the case of the stereochemical arrangement of minor
(3aR,4S)-6. As far as the diastereoselection is concerned, the exper-
imental ratio (3aS,4S)-6: (3aR,4S)-6 = 91:9 was quite satisfactory in
the light of the average stereoselectivities reported for the intra-
molecular cycloadditions of C-substituted nitrilimines.4,12
To gain deeper insights about this point, we undertook a
computational study principally aimed at locating the relevant
transition states. Despite the investigated molecules being
medium-sized (28 atoms, 114 electrons), we were able to carry
out correlated calculations (comprising full geometry optimiza-
tion and harmonic analysis) using second-order Møller-Plesset
(MP2) perturbation theory and Dunning’s correlation consistent
basis sets. Nitrilimine 5, cycloadducts (3aS,4S)-6 and (3aR,4S)-6,
and the corresponding transition states TS[(3aS,4S)-6] and
TS[(3aR,4S)-6] were fully optimized at the MP2/cc-pVDZ level. Har-
monic analysis was performed to ensure that they actually were
minimum energy and transition state structures, as appropriate.
The results of second-order Møller-Plesset (MP2) calculations with
three different basis sets are reported in Table 1, and the structure
of the products and of the transition states are shown in Figure 1.
The conversion from 4 to 6 is exothermic and the most stable
product is minor (3aR,4S)-6. Conversely, the activation energy is
lower for the transition state leading to the major diastereoisomer
(3aS,4S)-6 by 1.2–1.4 kcal molꢁ1 depending on the basis set em-
ployed. The computed diastereoisomeric ratio is in excellent agree-
ment with the experimental stereochemical outcome. The energy
difference between the two transition states can be correlated to
their structures. The distances between the atoms involved in the
formation of the new bonds are as follows: C–C = 2.325 for
(3aS,4S)-6 and 2.234 for (3aR,4S)-6; C–N = 2.754 and (3aS,4S)-6
and 2.938 for (3aR,4S)-6. Hence, the C–C bonds have an equal
length in the two transition states whereas the C–N bond is signif-
icantly less developed in the higher-energy TS[(3aR,4S)-6]. The
activation energy difference seems related to an unequal degree
of formation of the C–N bond due to the structural constraints
inherent to this intramolecular cycloaddition (see Fig. 1a and b).
It is also interesting to analyze the dependence of the activation
energy on the basis set. First, it should be noted that the variation
Figure 1. Structures of the transition states TS[(3aS,4S)-6] (a) and TS[(3aR,4S)-6] (b) leading to the major (3aS,4S)-6 (c) and minor (3aR,4S)-6 (d) diastereomeric products,
respectively, optimized at the MP2/cc-pVDZ level. The double arrow marks the hydrogens giving rise to the NOE effect used to assign the absolute configuration (see text).

G. Molteni, A. Ponti / 19 (2008) 1381–1384
1383
Table 1
Reaction energy
DEr, activation energy D
Eà and diastereoisomeric ratio of the intramolecular 1,3-dipolar cycloaddition of 4 to (3aS,4S)-6 and (3aR,4S)-6 calculated at the MP2 level
with three basis sets using the MP2/cc-pVDZ optimized geometry
Basis set
D
Er (kcal molꢁ1
)
D
Eà (kcal molꢁ1
)
Diastereoisomeric ratioa
(3aS,4S)-6
(3aR,4S)-6
(3aS,4S)-6
(3aR,4S)-6
(3aS,4S)-6:(3aR,4S)-6
cc-pVDZ
aug-cc-pVDZ
cc-pVTZ
ꢁ50.9
ꢁ51.9
ꢁ50.5
ꢁ51.5
ꢁ52.3
ꢁ51.0
4.6
2.7
4.0
5.9
3.9
5.4
89:11
88:12
91:9
a
The predicted diastereoisomeric ratio has been computed assuming that the reaction occurs under purely kinetic control and with an exponential activation law.
in molecular energy, when different basis sets are used, amounts to
ꢂ400 kcal molꢁ1 (data not shown), the variation in activation
energy to 2 kcal molꢁ1, and the variation in the activation energy
difference between the two transition states amounts to just
0.2 kcal molꢁ1. This shows that the computed stereochemical out-
come is very robust with respect to the choice of the basis set. Sec-
ond, the augmentation of the cc-pVDZ set with diffuse functions,
leading to the aug-cc-PVDZ set, improves the description of the
long, incipient C–C and C–N bonds (as shown by the lower activa-
tion energy) but not the computed stereochemistry. The latter is
improved when the double-f cc-pVDZ set is replaced by the
triple-f cc-pVTZ set.
Finally, as further proof of the above assignment of the absolute
configuration, the average distance between the methyl hydrogens
and the hydrogen in the 3a-position of the furo[3,4-c]pyrazole
skeleton is 3.06 Å for minor (3aR,4S)-6 and 3.98 Å for major
(3aS,4S)-6, which translate into average NOE distances (i.e., aver-
age of the inverse sixth power of the distance) of 2.3 ꢃ 10ꢁ3 Åꢁ6
for minor (3aR,4S)-6 and 0.3 ꢃ 10ꢁ3 Åꢁ6 for major (3aS,4S)-6. These
values substantiate the observed mutual NOE enhancement.
residue gave the analytically pure 2 (1.81 g, 93% yield). Colourless
oil; bp 71 °C (0.8 mmHg); ½a D25
¼ þ37:3 (c 0.88, CHCl3); IR (neat):
ꢀ
1740, 1720 cmꢁ1 1H NMR d 1.31 (3H, d, J = 7.6, CH3–CH—), 2.25
;
(3H, s, CH3–), 3.42 (2H, s, –CH2–), 4.89 (1H, dd, J = 7.6, 5.8, CH3–
CH—), 5.15–5.30 (2H, m, CH2@), 5.75–5.90 (1H, m, –CH@). MS:
m/z 156 (M+). Anal. Calcd for C8H12O3: C, 61.51; H, 7.75. Found:
C, 61.43, H, 7.71.
4.2. Synthesis of (S)-but-3-en-2-yl 2-chloroacetoacetate 3
A solution of sulfuryl chloride (1.68 g, 12.5 mmol) in dry chloro-
form (15 mL) was added over 1 h to a solution of 2 (10 mmol) in
dry chloroform (8 mL), on keeping the temperature in the range
0–5 °C. After 2 h at room temperature, chloroform (25 mL) was
added and the organic solution was washed with 5% aqueous
sodium hydrogencarbonate (20 mL). Then, the organic layer was
washed with water (2 ꢃ 30 mL) and dried over sodium sulfate.
The solvent was removed to give chloroacetoacetate 3 as an undis-
tillable oil that was not analytically pure (2.19 g, 92% yield). Pale
yellow oil; IR (neat): 1740, 1730 cmꢁ1 1H NMR d 1.30 (3H, d,
;
J = 7.1, CH3–CH—), 2.25 (3H, s, CH3–), 4.70 (1H, s, –CHCl–), 5.11
(1H, dd, J = 7.1, 5.7, CH3–CH—), 5.20–5.40 (2H, m, CH2@), 5.75–
5.90 (1H, m, –CH@). MS: m/z 190 (M+).
3. Conclusions
The dipolarophilic behaviour of the intramolecular cycloaddi-
tion between the ethylenic double bond bearing a stereocentre at
the allylic position and the nitrilimine moiety occurred with good
yield and stereoselectivity. This latter aspect of the intramolecular
cycloaddition has found a full rationale on the basis of MP2
calculations. It is worth noting that the preferential formation of
enantiopure furo[3,4-c] pyrazole (3aS,4S)-6 can be predicted
quantitatively when using the cc-pVTZ basis set.
4.3. Synthesis of hydrazonoyl chloride 4
A
cold aqueous solution of benzenediazonium chloride
(5.0 mmol) was slowly added to a solution of 3 (5.0 mmol) in
80% aqueous methanol (20 mL) under vigorous stirring and ice-
cooling. During the addition, the pH was adjusted to 5 by adding
sodium acetate. After 6 h at room temperature, the solvent was
partly removed under reduced pressure and the resulting mixture
was extracted with diethylether (2 ꢃ 30 mL). The organic layer was
washed with 5% sodium hydrogencarbonate (25 mL), then with
water (50 mL), and dried over sodium sulfate. Evaporation of the
solvent gave a solid and subsequent recrystallization from diiso-
propylether gave the hydrazonoyl chloride 4 in a pure state
4. Experimental
Melting points were determined with a Büchi apparatus in open
tubes and are uncorrected. IR spectra were recorded with a Perkin–
Elmer 1725 X spectrophotometer. Mass spectra were determined
with a VG-70EQ apparatus. 1H NMR (300 MHz) and 13C NMR
(75 MHz) spectra were taken with a Bruker AMX 300 instrument
(in CDCl3 solutions at room temperature). Chemical shifts are given
as ppm from tetramethylsilane and J values are given in Hz. NOE
experiments were performed by setting the following parameters:
relaxation delay (d1) 2 s, irradiation power (dl2) 74 dB, and total
irradiation time (for each signal) 1.8 s. Optical rotations were re-
corded on a Perkin–Elmer 241 polarimeter at the sodium D-line
at 25 °C. (S)-(+)-But-3-en-2-ol 1 was prepared as described in the
literature.8
(0.71 g, 56% yield). Pale yellow powder; mp 90 °C; ½a D25
¼ þ59:7
ꢀ
(c 0.52, CHCl3); IR (Nujol): 3270, 1710 cmꢁ1 1H NMR d 1.40 (3H,
;
d, J = 7.0, CH3–CH—), 4.92 (1H, dd, J = 7.0, 5.1, CH3–CH—), 5.20–
5.40 (2H, m, CH2@), 5.75–5.90 (1H, m, –CH@), 7.05–7.20 (5H, m,
–Ph), 8.30 (1H, br s, –NH–). MS: m/z 252 (M+). Anal. Calcd for
C12H13ClN2O2: C, 57.02; H, 5.19; N, 11.09. Found: C, 56.97, H,
5.22; N, 11.12.
4.4. Intramolecular cycloaddition of hydrazonoyl chloride 4
A solution of the hydrazonoyl chloride 4 (3.0 mmol) in dry diox-
ane (150 mL) was treated with silver carbonate (1.66 g, 6.0 mmol),
and stirred in the dark at room temperature for 96 h. The undis-
solved material was filtered off, the solvent evaporated and then
the residue was chromatographed on a silica gel column with
diethylether. Products and isolation yields are depicted in the
Scheme 1. Both major (3aS,4S)-6 and minor (3aR,4S)-6 were
4.1. Synthesis of (S)-but-3-en-2-yl acetoacetate 2
A solution of 1 (12.5 mmol) in xylene (5.0 mL) was treated with
2,2,6-trimethyl-4H-1,3-dioxin-4-one (1.78 g, 12.5 mmol). The
mixture was refluxed for 1.5 h. Evaporation of the solvent under
reduced pressure and subsequent in vacuo distillation of the

1384
G. Molteni, A. Ponti / 19 (2008) 1381–1384
obtained as analytically pure samples by recrystallization from
diisopropylether.
law, so that the diastereomeric excess is proportional to
exp(ꢁd
D D
Eà/kT), where d Eà is the activation energy difference.
(3aS,4S)-6: 0.53 g, 82%. White powder, mp 154 °C; ½a D25
ꢀ
¼ þ66:1
All computations were carried out by the GAUSSIAN03 suite.13
(c 0.77, CHCl3); IR (Nujol): 1750 cmꢁ1; 1H NMR d 1.60 (3H, d, J = 6.5,
CH3–CH—), 3.58–3.80 (2H, m, –CH2–), 4.38–4.59 (2H, m, CH3–CH—
and –CH—), 7.00–7.20 (5H, m, –Ph). After irradiation at 4.50 d:
3.64 (1H, d, J = 11.4, –HCH–), 3.71 (1H, d, J = 11.4, –HCH–). After
irradiation at 1.60 d: 4.42 (1H, ddd, J = 10.0, 8.1, 6.5, –CH—), 4.51
(1H, d, J = 8.1, CH3–CH—). 13C NMR d 23.8 (q, CH3–), 40.2 (d,
–CH—), 47.4 (t, –CH2N—), 58.8 (d, ˜CH–O–), 112.0–119.0 (aromatics),
120.8 (s, aromatic ˜C–N), 141.2 (s, ˜C@N–), 170.8 (s, –C@O). MS: m/
z 216 (M+). Anal. Calcd for C12H12N2O2: C, 66.64; H, 5.60; N, 12.96.
Found: C, 66.70, H, 5.57; N, 13.01.
Acknowledgements
Thanks are due to MURST (PRIN funding) and CNR for financial
support. We thank the NMR technician Dr. Lara De Benassuti,
University of Milan, for NOE experiments.
References
1. Garanti, L.; Sala, A.; Zecchi, G. Synth. Commun. 1976, 6, 269.
2. (a) Wade, P. A. In Comprehensive Organic Synthesis; Trost, B., Ed.; Pergamon
Press: New York, 1992; pp 1111–1168; (b) Padwa, A. In 1,3-Dipolar
Cycloaddition Chemistry; Padwa, A., Ed.; Wiley-Interscience: New York, 1984;
Vol. 2, p 277.
3. (a) Guerra, F. M.; Mish, M. R.; Carreira, E. M. Org. Lett. 2000, 2, 4265; (b) Conti,
P.; Roda, G.; Barberis Negra, F. F. Tetrahedron: Asymmetry 1999, 10, 487.
4. (a) Molteni, G. Heterocycles 2005, 65, 2513; (b) Molteni, G. Arkivoc 2007, 18,
224.
5. (a) Broggini, G.; Molteni, G.; Zecchi, G. Heterocycles 1998, 47, 541; (b)
Cycloaddition Reactions in Organic Synthesis; Kobayashi, S., Jørgensen, K. A.,
Eds.; Wiley-WCH: Weinheim, 2002; (c) Synthetic Applications of 1,3-Dipolar
Cycloaddition Chemistry Toward Heterocycles and Natural Products; Padwa, A.,
Pearson, W. H., Eds.; Wiley: New York, 2002.
6. Broggini, G.; Molteni, G.; Pilati, T.; Zecchi, G. Synth. Commun. 2001, 31, 3799.
7. De Benassuti, L.; Garanti, L.; Molteni, G. Tetrahedron: Asymmetry 2004, 15, 1127.
8. Ibrahim, T.; Grattan, T. J.; Whitehurst, J. S. J. Chem. Soc., Perkin Trans. 1 1990,
3317.
(3aR,4S)-6: 52 mg, 8%. White powder, mp 137 °C; ½a D25
¼ ꢁ21:4
ꢀ
(c 0.59, CHCl3); IR (Nujol): 1750 cmꢁ1; 1H NMR d 1.55 (3H, d, J = 6.6,
CH3–CH—), 3.60–3.90 (2H, m, –CH2–), 4.40–4.60 (2H, m, CH3–CH—
and –CH—), 7.00–7.20 (5H, m, –Ph). After irradiation at 4.50 d:
3.68 (1H, d, J = 11.0, –HCH–), 3.76 (1H, d, J = 11.0, –HCH–). After
irradiation at 1.55 d: 4.47 (1H, ddd, J = 10.8, 8.0, 6.6, –CH—), 4.56
(1H, d, J = 8.0, CH3–CH—). 13C NMR d 21.3 (q, CH3–), 40.9 (d,
–CH—), 45.1 (t, –CH2N—), 58.8 (d, ˜CH–O–), 110–120 (aromatics),
122.3 (s, aromatic ˜C–N), 140.8 (s, ˜C@N–), 187.6 (s, –C@O). MS:
m/z 216 (M+). Anal. Calcd for C12H12N2O2: C, 66.64; H, 5.60; N,
12.96. Found: C, 66.73, H, 5.66; N, 13.06.
4.5. Computational details
9. Garanti, L.; Sala, A.; Zecchi, G. J. Org. Chem. 1977, 42, 1389.
10. Broggini, G.; Bruché, L.; Garanti, L.; Zecchi, G. J. Chem. Soc., Perkin Trans 1 1994,
433.
11. De Benassuti, L.; Del Buttero, P.; Molteni, G. Tetrahedron: Asymmetry 2006, 17,
842.
The structure of nitrilimine 5, diastereomeric cycloadducts
(3aS,4S)-6 and (3aR,4S)-6, and the corresponding transition states
TS[(3aS,4S)-6] and TS[(3aR,4S)-6] were fully optimized at the
MP2//cc-pVDZ level in the frozen-core approximation. Harmonic
analysis confirmed that reactant and product structures were min-
imum energy structures and that TS[(3aS,4S)-6] and TS[(3aR,4S)-6]
structures were transition states with a single imaginary fre-
quency. To check the reliability of the computed energy differ-
ences, calculations were also performed with an augmented
double-f basis set (aug-cc-pVDZ) and with a triple-f basis set (cc-
pVTZ). The latter calculation, involving 114 electrons (98 active)
and 648 basis functions, is the largest calculation affordable with
our present computing facilities. The stereochemical outcome
was estimated by assuming that the reaction is first order in the
reactant and that the reaction rates are given by an activation
12. Pellissier, H. Tetrahedron 2007, 63, 3235.
13. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Montgomery, Jr. J. A.; Vreven, T.; Kudin, K. N.; Burant, J. C.;
Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.;
Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.;
Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao,
O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Adamo,
C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi,
R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.;
Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.;
Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.;
Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.;
Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe,
M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A.
GAUSSIAN 03, Revision C.02; Gaussian: Wallingford, CT, 2004.