Reactions of Nitrosonium Ethyl Sulfate with Olefins and Dienes: An Experimental and Theoretical Study

Article abstract of DOI:10.1021/jo990679t

Nitrosonium ethyl sulfate (1), which is generated in situ from ethyl nitrite and sulfur trioxide, is a convenient reagent for the one-pot transformation of olefins and dienes into substituted aldehydes and ketones. New experimental and theoretical aspects of this reaction are discussed. A DFT and ab initio computational study is undertaken to provide further insight into the mechanism of electrophilic nitrosation, including the initial π-complexes, transition states, and the intermediates involved in subsequent carbonyl formation.

Full text of DOI:10.1021/jo990679t

J . Org. Chem. 1999, 64, 7121-7128
7121
Rea ction s of Nitr oson iu m Eth yl Su lfa te w ith Olefin s a n d Dien es:
An Exp er im en ta l a n d Th eor etica l Stu d y
Nikolai V. Zyk, Evgueni E. Nesterov, Andrei N. Khlobystov, and Nikolai S. Zefirov*
Department of Chemistry, Moscow State University, 119899 Moscow, Russian Federation
Loren A. Barnhurst and Andrei G. Kutateladze*
Department of Chemistry and Biochemistry, University of Denver, Denver, Colorado 80208
Received April 22, 1999
Nitrosonium ethyl sulfate (1), which is generated in situ from ethyl nitrite and sulfur trioxide, is
a convenient reagent for the one-pot transformation of olefins and dienes into substituted aldehydes
and ketones. New experimental and theoretical aspects of this reaction are discussed. A DFT and
ab initio computational study is undertaken to provide further insight into the mechanism of
electrophilic nitrosation, including the initial π-complexes, transition states, and the intermediates
involved in subsequent carbonyl formation.
Electrophilic addition across olefinic double bonds is a
fundamental process in organic chemistry that has both
theoretical value and far-reaching synthetic applications.
Indisputably, the AdE reaction of olefins constitutes a
versatile and reliable method for their functionalization,
making the search for new electrophilic reagents, as well
as the development of methods to enhance the reactivity
of weak electrophiles, an important task.1-3
In a previous paper5 we reported the application of this
concept to the electrophilic addition of ethyl nitrite to
cycloalkenes. This paper contributes an important syn-
thetic development of this idea augmented with density
functional theory (DFT) and ab initio theoretical studies
to better understand the mechanism of nitrosation.
Resu lts a n d Discu ssion
Earlier we found5 that ethyl nitrite reacts instantly
with an equimolar amount of SO3 at -50 to -30 °C in
CH2Cl2 to give a highly reactive nitrosating reagent,
nitrosonium ethyl sulfate (1), which was used in situ
without purification or isolation. Reagent 1 reacts with
cyclic olefins to form R-ethylsulfato cycloalkanones 2 in
good yields.
In recent years, we have proposed a number of new
electrophilic reagents based on the novel concept involv-
ing activation of weak electrophiles with sulfur trioxide.4
(1) (a) Schmid, G. H. Electrophilic addition to carbon-carbon double
bonds. In Chemistry of Double-Bonded Functional Groups; Patai, S.,
Ed.; Wiley: Chichester, U.K., 1989; Vol. 2, Part 1, p 679. (b) Freeman,
F. Chem. Rev. 1975, 75, 439. (c) Fahey, R. C. Top. Stereochem. 1968,
3, 237.
(2) (a) Zefirov, N. S.; Zyk, N. V.; Kolbasenko, S. I.; Kutateladze, A.
G. J . Org. Chem. 1985, 50, 4539 and references therein. (b) Smit, V.
A.; Zefirov, N. S.; Bodrikov, I. V.; Krimer, M. Z. Acc. Chem. Res. 1979,
12, 282.
(3) For selected recent examples, see: (a) Stavber, S.; SotlerPecan,
T.; Zupan, M. Bull. Chem. Soc. J pn. 1996, 69, 169. (b) Pitre, S. V.;
Reddy, M. V. R.; Vankar, Y. D.; Madhusudanan, K. P. Synth. Commun.
1997, 27, 267. (c) Sanseverino, N. M.; de Mattos, M. C. S. Synthesis
1998, 1584. (d) Tiecco, M.; Testaferri, L.; Santi, C.; Marini, F.; Bagnoli,
L.; Temperini, A. Tetrahedron Lett. 1998, 39, 2809.
(4) (a) For
a review, see: Zefirov, N. S. In Organic Synthesis:
Modern Trends, Proceedings of the 6th IUPAC Symposium on Organic
Synthesis, Moscow, 1986; Chizhov, O. S., Ed.; Blackwell: Oxford, UK,
1987, p 122. (b) SO3 + RSCl: Zefirov, N. S.; Kos’min, A. S.; Sorokin,
V. D.; Shastin, A. V.; Balenkova, E. S. Dokl. Akad. Nauk SSSR, 1984,
276, 1139. (c) SO3 + Cl2: Zefirov, N. S.; Kos’min, A. S.; Sorokin, V. D.;
Zhdankin, V. V. J . Org. Chem. 1984, 49, 4086. (d) SO3 + AcF: Krespan,
C. G.; England, D. C. J . Org. Chem. 1975, 40, 2937. Shastin, A. V.;
Gavrishova, T. N.; Balenkova, E. S. Zh. Org. Khim. 1985, 21, 1862.
Gavrishova, T. N.; Shastin, A. V.; Balenkova, E. S. Zh. Org. Khim.
1991, 27, 673. (e) SO3 + R2NCl: Zefirov, N. S.; Zyk, N. V.; Kolbasenko,
S. I.; Kutateladze, A. G. Sulfur Lett. 1984, 2, 95. Zefirov, N. S.; Zyk,
N. V.; Kolbasenko, S. I.; Kutateladze, A. G. J . Org. Chem. 1985, 50,
4539. (f) SO3 + XeF2: Brel’, V. K.; Gakh, A. A.; Zhdankin, V. V.; Zefirov,
N. S.; Koz’min, A. S.; Korkin, A. A.; Kutateladze T. G.; Caple, R.;
Lermontov, S. A.; Plokhikh, I. G.; Safronov, S. O.; Stang, P. J .;
Chovnikova, N. G. Dokl. Akad. Nauk SSSR 1990, 313, 1131. (g) SO3
+ RSSR′: Kutateladze, A. G.; Zefirov, N. S.; Zyk, N. V. Sulfur Rep.
1992, 11, 233. (h) SO3 + RSNR2′: Zefirov, N. S.; Zyk, N. V.; Kutate-
ladze, A. G.; Lapin, Yu. A. Zh. Org. Khim. 1987, 23, 392. Zyk, N. V.;
Kolbasenko, S. I.; Kutateladze A. G.; Lapin, Yu. A. Zh. Org. Khim.
1988, 24, 1209. (i) SO3 + RSOR′: Zefirov, N. S.; Zyk, N. V.; Lapin, Yu.
A.; Kutateladze, A. G.; Ugrak, B. I. Zh. Org. Khim. 1992, 28, 1226. (j)
SO3 + R2N-S-NR2: Zefirov, N. S.; Zyk, N. V.; Kutateladze, A. G.;
Lapin, Yu. A. Zh. Org. Khim. 1987, 23, 229. Zefirov, N. S.; Zyk, N. V.;
Kutateladze, A. G.; Potekhin, K. A.; Struchkov, Yu. T. Sulfur Lett.
1987, 6, 139. (k) SO3 + RONO2: Zefirov, N. S.; Zyk, N. V.; Lapin, Yu.
A.; Kutateladze, A. G. Sulfur Lett. 1988, 8, 143. Zyk, N. V.; Lapin, Yu.
A.; Kutateladze, A. G.; Zefirov, N. S. Zh. Org. Khim. 1988, 24, 889. (l)
SO3 + R2NSCl: Kutateladze, A. G.; Zyk, N. V.; Denisko, O. V.; Zefirov,
N. S. Zh. Org. Khim. 1991, 27, 659.
As a development of this idea we now report the results
of our experimental and theoretical studies on the
reactions of 1 with terminal olefins and dienes. In the
case of 1-hexene (3a ), 1-heptene (3b), and 1-octene (3c)
the reaction furnished the corresponding R-ethylsulfato-
substituted aldehydes 4a -c in good yields (70-75%).
The reactions proceed regiospecifically in accordance
with the Markovnikov rule. It should be emphasized that
the above reaction of 1 is a rare case of direct conversion
of the terminal olefinic carbon into the aldehyde group.
(5) Zefirov, N. S.; Zyk, N. V.; Lapin, Yu. A.; Nesterov, E. E.; Ugrak,
B. I. J . Org. Chem. 1995, 60, 6771.
10.1021/jo990679t CCC: $18.00 © 1999 American Chemical Society
Published on Web 08/31/1999

7122 J . Org. Chem., Vol. 64, No. 19, 1999
Zyk et al.
Parent ethylene (5) also reacted with reagent 1; pass-
ing gaseous ethylene through freshly prepared 1 in CH2-
Cl2 at -40 °C yielded the ethyl sulfate of glycolaldehyde
(6) (40%).
dihydrooxazinium cation IIIa (a boat conformation) lies
only 4.2 kcal/mol (B3LYP/6-31G(d) or 11.8 at MP2/6-
311G(d)) above the s-cis π-complex Ia , which makes it
easily accessible at the reaction temperatures.
Summarizing our experimental and theoretical evi-
dence, we can therefore offer the following mechanistic
rationale for the process. Despite somewhat contradicting
views on the mechanism of olefin nitrosation,9 there is
little doubt that the reaction with isoprene begins with
an electrophilic attack of the nitrosonium cation (NO+).
Our computational data seem to suggest that the s-cis
π-complex Ia (which is in equilibrium with dihydroox-
azinium cation IIIa ) is formed as an initial product of
such electrophilic attack, followed by deprotonation to
form oxazine 11 (or its salt 10). In the case when an
equimolar amount of the nitrosating reagent is used, a
basic workup of the reaction mixture then furnishes
oxazine as the only product.10
An interesting result was obtained for a simple con-
jugated diene, isoprene (7), which is one of the most
important industrial dienes, as well as the main building
block of many naturally occurring molecules. Reaction
of 1 with isoprene at -55 °C followed by a slow warming
to room temperature and a subsequent aqueous workup
resulted in a single product, 4-hydroxyaldehyde 8 (78%),
with E-configuration of the CdC double bond.6
If a 2-fold excess of the nitrosating reagent is used,
N-nitrosooxazinium is formed, which in turn can be
quenched at low temperatures by a water/alcohol mixture
and hydrolyzed into the Z-hydroxyaldehyde 9 (Path A,
Scheme 1). Alternatively, in the absence of water it can
ring-open to form N-nitrosoaldoxime (B-Z), which upon
warming to room temperature (prior to aqueous workup)
has a chance to equilibrate with (B-E).
We ran DFT calculations (B3LYP/6-31G*) for isomers
8 and 9 and found the E-isomer 8 to be approximately
2.4 kcal/mol more stable than the Z-isomer 9. Self-
consistent reaction field computations to account for
solvent effects (self-consistent isodensity polarized con-
tinuum model, SCIPCM) showed the same energy dif-
ference for the two isomers.
The transformation of the nitroso group in the initial
products of electrophilic addition into a carbonyl is known
to occur via nitrosooxime tautomerization with subse-
quent hydrolysis of the oxime into the free carbonyl. Such
hydrolyses are facilitated by NO+ when excess nitrosating
species are present. This reaction, however, has been a
point of controversy in the literature. In a mechanistic
study of a ketoxime reacting with 18O-enriched nitrous
acid, Wieland and Grimm considered two possible mech-
anisms:
However, a small variation in the reaction conditions
led to a dramatically different result. When aqueous
ethanol was added to the reaction mixture at -55 °C
prior to increasing the temperature, a mixture of two
stereoisomeric 4-hydroxyaldehydes 8 and 9 is obtained
in the ratio of 1:6.5,7 (42%). That is, the Z-isomer 9 is
now the major product.
In both cases we used a 2:1 ratio of reagent 1 to
isoprene in accordance with our previous observations.5
However, in the case of equimolar amounts of 1 and
isoprene (the conditions of “the primary adduct stage”,
see ref 5), 4-methyl-1,2,6H-oxazinium ethyl sulfate (10)
was obtained as the only product.
The subsequent treatment of the salt 10 with an
aqueous solution of sodium carbonate furnished oxazine
11 as a free base (44%).
To gain mechanistic insight we conducted a DFT study
on the species potentially involved in this reaction
sequence. Not unexpectedly, we were unable to locate a
σ-complex on the energy hypersurface. It was found that
the most stable initial species of the electrophilic attack
is a s-cis π-complex Ia , which is about 1 kcal/mol more
stable than s-trans π-complex IIa (Table 1). The fact that
a “classical” σ-complex is much less stable than the
π-complex is certainly not unprecedented in olefin nit-
rosation. Raghavachari et al.8 showed that, for example,
in ethylene nitrosation the π-complex (CH2CH2/NO+) is
about 14 kcal/mol more stable than the σ-complex (ON-
CH2CH2+) at CCDST(4)/6-31G(d) and MP4/6-31G(d) lev-
els of theory. At the same time we find that the
On the basis of a 89% enrichment of nitrous oxide (N2-
18O) they ruled out the four-membered intermediate,
concluding that the reaction involves an intermediate
formation of N-nitrosooxaziridine.11 Kliegman and Bar-
(6) E-configuration of the double bond in 8 was determined by a
NOE experiment.
(7) Determined from the integral intensities of corresponding peaks
(9) (a) Kadzjauskas, P. P.; Zefirov, N. S. Usp. Khim. 1968, 37, 1243.
(b) Ingold, C. K. Structure and mechanism in organic chemistry; Cornell
University Press: Ithaca, NY, 1953. (c) Meinwald, J .; Meinwald, Y.
C.; Baker, T. N. J . Am. Chem. Soc. 1963, 85, 2513.
in the 1H NMR spectrum.
(8) Raghavachari, K.; Reents, W. D., J r.; Haddon, R. C. J . Comput.
Chem. 1986, 7, 265-73.

Nitrosation of Olefins and Dienes
J . Org. Chem., Vol. 64, No. 19, 1999 7123
Ta ble 1. NO+ Ad d ition to Isop r en e, Dim eth ylbu ta d ien e, a n d Bu ta d ien e; DF T a n d a b In itio Resu ltsa ,b
nes later revised Wieland’s results, contending that there
is no need to invoke the three-membered ring intermedi-
ate as the mechanism can simply be envisioned as a
hydrolysis of the N-nitrosooxime.12 They, however, re-
vived the oxadiazacyclobutene intermediate to explain
partial N2 formation in this reaction.
We explored this transformation theoretically and
carried out both DFT and high level ab initio studies of
the species potentially involved in the rearrangement.
First, we optimized molecular geometries of syn- and
anti-N-nitrosoformaldoximes (deprotonated in Table 2
and protonated in Table 3) at B3LYP/6-31G(d) and MP2/
6-311G(d) levels of theory. We then optimized the geom-
etries of the three- and four-membered ring intermedi-
ates and located the respective transition states leading
to these intermediates. We also located the transition
states for the N2O extrusion regenerating the free car-
bonyl. All of these computations were accompanied by
vibrational analyses. No imaginary frequencies were
observed for the intermediates, and all of the calculated
transition states had one imaginary frequency corre-
sponding to the expected normal mode associated with
the reaction coordinate. We also ran higher level single
point MP4/6-311+G(d) computations for the MP2/6-311G-
(d) geometries. Additionally, we estimated the impact of
the solvent by employing the self-consistent reaction field
computations at the MP4/6-311+G(d) level in conjunction
with SCIPCM for all of the species involved, including
the transition states in the deprotonated series. Our
findings seem to indicate that the barriers for the
formation of the three- and four-membered intermedi-
ates, about 50 kcal/mol, are too high to be accessible for
the molecules even at room temperature (see Tables 2
(10) Such process is certainly not unprecedented in the literature;
as early as three decades ago oxazine 11 was synthesized by the
reaction of nitrosonium hydrosulfate NOHSO4 with isoprene: Kla-
mann, D.; Fligge, M.; Weyerstahl, P.; Kratzer, J . Chem. Ber. 1966, 99,
556.
(11) Wieland, T.; Grimm, D. Chem. Ber. 1963, 96, 275.
(12) Kliegman, J . M.; Barnes, R. K. J . Org. Chem. 1972, 37, 4223.

7124 J . Org. Chem., Vol. 64, No. 19, 1999
Zyk et al.
Sch em e 1
and 3). We therefore contend that a simple hydrolysis of
N-nitrosooximes is the most likely channel for NO+-
assisted transformation of oximes. Subsequent degrada-
tion of the initial inorganic product of such hydrolysis,
hyponitrous acid, could conceivably account not only for
the formation of N2O but also for the formation of the
other two components of the gaseous mixture, N2 and NO:
The DFT study of the 2,3-dimethylbutadiene nitrosa-
tion again showed that the most stable initial species in
this reaction is the s-cis π-complex (Ib) (Table 1), just as
in the reaction with isoprene. It is 2.47 kcal/mol below
its s-trans isomer (IIb). The π-complexes are in equilib-
rium with the oxazinium (IIIb), which lies only 2.7 (10.4
MP2) kcal/mol above the s-cis isomer. Conceivably, the
introduction of the second methyl group simply prevents
the oxazinium ring from opening under a nucleophilic
attack by either water or ethyl sulfate anion. As a result
of such steric hindrance the oxazinium cation (IIIb) then
has only one channel for stabilization, deprotonation,
which furnishes the oxazine 13.
Finally, the reaction of 1 with 1,3-butadiene (14) did
not produce any monomeric product but rather yielded
a mixture of oligomeric products, which showed the
presence of formyl and hydroxy groups by IR spectro-
scopy. Computationally, one difference was that both
B3LYP/6-31G(d) and MP2/6-311G(d) computations in this
case showed a slight energy preference for the s-trans
π-complex IIc (Table 1). It is plausible that such an
acyclic intermediate would be much more prone to
polymerization. The absence of methyl substitution may
also allow for secondary nitrosation of the double bond
being formed between atoms 2 and 3 in the initial 1,4-
adduct.
In summary, with minor variations in the reaction
conditions, nitrosation of isoprene 7 with nitrosonium
ethyl sulfate 1 can be directed to afford at least three
products with surprisingly good selectivity. One of them,
E-aldehyde 8, is an important starting material for
syntheses of various isoprenoid compounds.13 It should
also be noted that the regiochemistry of this isoprene
nitrosation is exclusively Markovnikov, in keeping with
the results of 1 reacting with olefins 3a -c.
Despite our multiple efforts to obtain 4-hydroxyalde-
hydes (of type 8 and 9) only 4,5-dimethyl-1,2,6H-oxazine
(13) was obtained in a quantitative yield (97%) as a result
of the nitrosation of 2,3-dimethyl-1,3-butadiene (12).
Cyclopentadiene (15), a widely used 1,3-diene with
rigidly fixed s-cis arrangement of double bonds, yields
only a single product, 2-hydroxycyclopent-3-enone (16)14
(55%), when reacted with 1.
Ketone 16 is the result of the Markovnikov 1,2-addition
of 1 to one of the double bonds. Thus, the reaction
proceeds differently in comparison to acyclic dienes where
(13) Pattenden, G.; Way, G. E.; Weedon, B. C. L. J . Chem. Soc., C
1970, 235. Pommer, H.; Nu¨rrenbach, A. Pure Appl. Chem. 1975, 43,
527; Wehrli, P. A., Schaer, B. Synthesis 1977, 649.

Nitrosation of Olefins and Dienes
J . Org. Chem., Vol. 64, No. 19, 1999 7125
Ta ble 2. Rela tive En er gies,a Ch a r ges,b a n d Bon d Len gth s in IVa -XIa
the products of 1,4-addition were exclusively formed. Our
DFT study revealed that the π-complex XII is again by
far the most stable initial species. In fact, we were unable
to optimize the geometry of an oxazinium-like intermedi-
ate without imposing certain geometry constraints. Such
unconstrained computations, starting with a bicyclo-
[2.2.1] structure similar to XIII, would inevitably revert
to the π-complex XII. Only when we constrained the C-N
and C-O bond lengths to that of the respective bond
length values in oxazinium IIIa were we able to optimize
the geometry at B3LYP/6-31G(d) level. Surprisingly, the
vibrational analysis showed no imaginary frequencies,
which would imply that the bicyclic structure XIII is a
very shallow minimum with an insignificantly low barrier
leading to the π-complex XII. The energy of XIII is still
about 20.3 kcal/mol above the π-complex (see Table 4,
compare with 2.7-4.2 kcal/mol at this level of theory for
the oxaziniums IIIa -c, Table 1). We therefore conclude
that the absence of the isomeric product, 3-hydroxycy-
clopent-4-enone, is due to high skeletal strain in XIII
compared with acyclic dienes.
(14) The structural assignment of 24 was based on the presence of
the intramolecular H-bond in the IR experiment and the absence of
spin-spin coupling constant 3J for H2C and HCO protons in the 1H
NMR experiment.
Reaction of 1 with 1,5-cyclooctadiene (17) afforded

7126 J . Org. Chem., Vol. 64, No. 19, 1999
Ta ble 3. Rela tive En er gies,a Ch a r ges,b a n d Bon d Len gth s in IVb-VIIIb a n d Xb
Zyk et al.
Ta ble 4. B3LYP /6-31G(d ) Rela tive En er gies,a Ch a r ges,b
Predictably, 1 reacted with the more active (cy-
clobutene) double bond of diene 19, yielding the product
of 1,2-addition (20, 76%). As in the previous case, we did
not observe the formation of the transannular cross-
cyclization products, which are typical for electrophilic
additions to this diene.17
a n d Bon d Len gth s in XII a n d XIII
exclusively the ethylsulfato ketone 18 (74%).
Dicyclopentadiene (21) adds the reagent 1 exclusively
to the less strained double bond to yield ethylsulfato
ketone 22 (82%).
(15) (a) See ref 2a. (b) Zefirov, N. S.; Bodrikov, I. V. Zh. Org. Khim.
1983, 19, 225. (c) See also ref 15 in: Zefirov, N. S.; Zyk, N. V.;
Borisenko, A. A.; Krysin, M. Yu.; Schestakova, T. G. Tetrahedron 1983,
39, 3145.
(16) Zefirov, N. S.; Sadovaja, N. K.; Novgorodtseva, L. A.; Bodrikov,
I. V. Zh. Org. Khim. 1978, 14, 1806.
(17) Kondo, A.; Yamane, T.; Ashida, T.; Sasaki, T.; Kanematsu, K.
J . Org. Chem. 1978, 43, 1180. Zefirov, N. S.; Koz’min, A. S.; Kirin, V.
N.; Zhdankin, V. V.; Caple, R. J . Org. Chem. 1981, 46, 5264.
Transannular ring closure to give bicyclo[3.3.0]octane
derivatives or hydride shifts, which commonly accompany
the additions of “effectively strong electrophiles” (for
definition see ref 15) to this diene,16 are not observed in
this case.

Nitrosation of Olefins and Dienes
J . Org. Chem., Vol. 64, No. 19, 1999 7127
mL of CH2Cl2, and 0.4 g (4.8 mmol) of 3a in 20 mL of CH2Cl2
was obtained 0.75 g (70%) of ethyl sulfate of 2-hydroxyhexanal
(4a ) (eluent hexane/AcOEt/CHCl3 ) 1:3:3; colorless oil, Rf
0.16): IR 1760, 1400, 1200 cm-1 1H NMR (CDCl3) δ 0.91
;
(narrow t, 3H), 1.20-1.55 (m, 4H), 1.43 (t, 3H, J ) 7.1 Hz),
1.87 (m, 2H), 4.45 (m, 2H), 4.83 (dd, 1H, J ) 7.0, 5.5 Hz), 9.64
(s, 1H). Anal. Calcd for C8H16O5S: C, 42.84; H, 7.19. Found:
C, 42.75; H, 7.19.
In conclusion, we have developed a new one-pot
transformation for terminal olefins and dienes into
R-substituted aldehydes and internal olefins into the
corresponding ketones. The studied reaction can be
considered a versatile technique for olefin functionaliza-
tion, affording valuable synthetic intermediates. The
common feature of the mechanism is that, unlike a
typical electrophilic addition, the nitrosation reaction
seems to proceed via a π-complex, not a σ-intermediate.
Insignificant development of carbocationic character in
such π-complexes could then explain the known “reluc-
tance” of nitrosation reactions to undergo skeletal rear-
rangements.
Rea ction of 1-Hep ten e (3b). From 1.28 g (16.0 mmol) of
SO3 in 25 mL of CH2Cl2, 1.64 g (22.0 mmol) of EtONO in 20
mL of CH2Cl2, and 0.71 g (7.2 mmol) of 3b in 20 mL of CH2Cl2
was obtained 1.29 g (75%) of ethyl sulfate of 2-hydroxyheptanal
(4b) (eluent hexane/AcOEt ) 3:1; colorless oil, Rf 0.23): IR
1740, 1400, 1200 cm-1; 1H NMR (CDCl3) δ 0.9 (narrow t, 3H),
1.20-1.50 (m, 6H), 1.4 (t, 3H, J ) 7.1 Hz), 1.7-2.0 (m, 2H),
4.4 (m, 2H), 4.78 (dd, 1H, J ) 7.7, 5.6 Hz), 9.6 (s, 1H); 13C
NMR δ 13.8, 14.4, 22.1, 23.8, 29.5, 31.0, 70.1, 85.5, 196.1. Anal.
Calcd for C9H18O5S: C, 45.36; H, 7.61. Found: C, 45.54; H,
7.63.
Rea ction of 1-Octen e (3c). From 1.67 g (21.0 mmol) of
SO3 in 30 mL of CH2Cl2, 2.13 g (28.4 mmol) of EtONO in 25
mL of CH2Cl2, and 1.06 g (9.5 mmol) of 3c in 25 mL of CH2Cl2
was obtained, 1.67 g (70%) of ethyl sulfate of 2-hydroxyoctanal
(4c) (eluent heptane/AcOEt ) 3:1; colorless oil, Rf 0.24): IR
1740, 1400, 1200 cm-1; 1H NMR (CDCl3) δ 0.9 (narrow t, 3H),
1.1-2.1 (group of m, 13H), 4.3 (m, 2H), 4.7 (m, 1H), 9.5 (s,
1H). Anal. Calcd for C10H20O5S: C, 47.60; H, 7.99. Found: C,
47.48; H, 7.89.
Rea ction of Eth ylen e (5). To a solution of 1.27 g (16.0
mmol) of SO3 in 25 mL of CH2Cl2 was added the solution of
2.38 g (32.0 mmol) of EtONO in 20 mL of CH2Cl2 dropwise
while the temperature was maintained at -55 °C. After 0.5 h
of stirring at this temperature, gaseous ethylene was passed
through the reaction mixture for 2 h. Further treatment was
performed according to the general procedure and gave 0.45 g
(40%, based on SO3) of ethyl sulfate of glycolaldehyde (6)
(eluent hexane/AcOEt ) 2:1; yellowish oil, Rf 0.15): IR 1720,
1380, 1200 cm-1; 1H NMR (CDCl3) δ 1.4 (t, 3H, J ) 7.1), 4.0-
4.9 (m, 4H), 9.3 (s, 1H).
Exp er im en ta l Section
Gen er a l. 1H and 13C NMR spectra were recorded in CDCl3.
All reactions were followed by TLC with precoated aluminum
TLC plates (silica gel, Silufol, Czech Republic). Preparative
column chromatography involved silica gel (Silpearl) and ethyl
acetate-hexane mixtures as eluent. All solvents and reagents
were additionally purified and dried by standard techniques.
Ethyl nitrite was synthesized as previously reported.18 SO3 was
obtained from 60% oleum. Freshly distilled SO3 was used,
which was weighed and dissolved in CH2Cl2.
Com p u ta tion s. Ab initio and DFT computations were
performed on a dual mips R10000 processor SGI Octane
workstation equipped with 1GB memory using Gaussian 94
Revision E.2 computational package.19 The input geometries
were created and preoptimized using a force field geometry
optimization as implemented in Chem3D (Cambridgesoft). Full
geometry optimizations were then performed at B3LYP/6-31G-
(d) and MP2/6-311G(d) levels of theory. Single point computa-
tions at the MP4/6-311G(d) level of theory were performed for
the series of nitrosoformaldoximes (Table 2). Self-consistent
reaction field (SCRF) computations were then performed to
account for solvent polarity effects utilizing SCIPCM. These
also were run as single point calculations without further
geometry optimization, MP4-SCIPCM/6-311G(d)//MP2/6-311G-
(d) (Table 2).
Gen er a l P r oced u r e for th e Ad d ition Rea ction . A three-
neck flask fitted with an addition funnel, stirrer, and argon
inlet was charged with a solution of SO3 in CH2Cl2. The
mixture was cooled to -50 °C, and a solution of EtONO in
CH2Cl2 was added dropwise while the temperature was
maintained at -50 to -30 °C. The resulting mixture was
stirred for 0.5 h at this temperature, and a solution of olefin
in CH2Cl2 was added dropwise at -50 to -30 °C. Further
workup involved 1 h of stirring at -50 °C, after which the
temperature was allowed to rise slowly to room temperature.
After treatment with cold water, extraction with CHCl3, and
drying over MgSO4, the solvent was evaporated in vacuo.
Crude product was chromatographed.
Rea ction of Isop r en e (7). (a ) In Con d ition s of Kin etic
Con tr ol. To a solution of 1.3 g (16.3 mmol) of SO3 in 30 mL of
CH2Cl2 was added the solution of 2.65 g (35.3 mmol) of EtONO
in 25 mL of CH2Cl2 dropwise while the temperature was
maintained at -55 °C. After 0.5 h of stirring at this temper-
ature, the solution of 0.48 g (7.1 mmol) of 7 in 18 mL of CH2-
Cl2 was added dropwise. After 1 h of stirring at -55 °C, the
1:1 water-ethanol mixture (20 mL) was added as quickly as
possible. Further treatment was performed according to the
general procedure. As a result, 0.3 g (42%) of a mixture of Z-
and E-isomers of 4-hydroxy-2-methylbut-2-enal (9 and 8)20 was
obtained (eluent hexane/AcOEt ) 3:1; colorless oil, Rf 0.32):
1
IR 3600-3200, 1690, 1650 cm-1; H NMR (CDCl3) δ 1.85 (q,
3H of 8, J ) 7.0 Hz), 1.92 (q, 3H of 9, J ) 7.0 Hz), 5.27 (dq,
2H of 8, J ) 7.0, 1.7 Hz), 5.41 (dq, 2H of 9, J ) 7.0, 1.7 Hz),
6.37 (tq, 1H of 9, J ) 7.0, 1.7 Hz), 6.46 (tq, 1H of 8, J ) 7.0,
1.7 Hz), 9.48 (s, 1H of 8), 10.07 (s, 1H of 9); 13C NMR δ 16.81
(1C of 9), 19.30 (1C of 8), 67.21 (1C of 9), 68.45 (1C of 8), 134.82
(1C of 9), 140.37 (1H of 8), 140.60 (1C of 9), 147.20 (1C of 8),
190.33 (1C of 9), 193.53 (1C of 8). Anal. Calcd for C5H8O2: C,
59.98; H, 8.05. Found: C, 59.85; H, 7.93.
(b) In Con d ition s of Th er m od yn a m ic Con tr ol. From
1.21 g (15.0 mmol) of SO3 in 70 mL of CH2Cl2, 2.27 g (30.0
mmol) of EtONO in 20 mL of CH2Cl2, and 0.51 g (7.5 mmol) of
7 in 15 mL of CH2Cl2 was obtained 0.59 g (78%) of 4-hydroxy-
2-methylbut-2(E)-enal (8) (eluent hexane/AcOEt ) 3:1; color-
Rea ction of 1-Hexen e (3a ). From 0.95 g (11.9 mmol) of
SO3 in 20 mL of CH2Cl2, 1.78 g (23.7 mmol) of EtONO in 16
1
less oil, Rf 0.32): IR 3600-3200, 1690, 1650 cm-1; H NMR
(18) Blatt, A., Ed. Organic Syntheses; Wiley: New York, 1946;
Collect. Vol. 2, p 204.
(CDCl3) δ 1.85 (q, 3H, J ) 7.0 Hz), 5.27 (dq, 2H, J ) 7.0, 1.7
Hz), 6.46 (tq, 1H, J ) 7.0, 1.7 Hz), 9.48 (s, 1H); 13C NMR δ
19.30, 68.45, 140.37, 147.20, 193.53. Anal. Calcd for C5H8O2:
C, 59.98; H, 8.05. Found: C, 59.90; H, 7.97.
(19) Frisch, M. J .; Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.;
J ohnson, B. G.; Robb, M. A.; Cheeseman, J . R.; Keith, T.; Petersson,
G. A.; Montgomery, J . A.; Raghavachari, K.; Al-Laham, M. A.;
Zakrzewski, V. G.; Ortiz, J . V.; Foresman, J . B.; Cioslowski, J .;
Stefanov, B. B.; Nanayakkara, A.; Challacombe, M.; Peng, C. Y.; Ayala,
P. Y.; Chen, W.; Wong, M. W.; Andres, J . L.; Replogle, E. S.; Gomperts,
R.; Martin, R. L.; Fox, D. J .; Binkley, J . S.; Defrees, D. J .; Baker, J .;
Stewart, J . P.; Head-Gordon, M.; Gonzalez, C.; Pople, J . A. Gaussian
94, Revision E.2; Gaussian, Inc.: Pittsburgh, PA, 1995.
(c) In Con d ition s of Kin etic Con tr ol a t 1:1 Stoich iom -
etr y of 1 to 7. Reaction was performed according to the
(20) Ranu, B. C.; Sarkar, D. C. Synth. Commun. 1987, 17, 155.

7128 J . Org. Chem., Vol. 64, No. 19, 1999
Zyk et al.
procedure given in (a ). From 0.55 g (7.0 mmol) of SO3 in 30
mL of CH2Cl2, 0.59 g (7.9 mmol) of EtONO in 15 mL of CH2-
Cl2, and 0.47 g (7.0 mmol) of 7 in 15 mL of CH2Cl2 was obtained
crude product. This was dissolved in 50 mL of water and
washed with Na2CO3 solution until pH 8-9, followed by
extraction with CHCl3 (2 × 30 mL). Solvent removal and
chromatography purification gave 0.3 g (44%) of 4-methyl-
1,2,6H-oxazine (11) (eluent hexane/AcOEt ) 3:1; yellowish oil,
Rf 0.38): IR 1667, 1555 cm-1; 1H NMR (CDCl3) δ 1.75 (s, 3H),
4.10 (d, 2H, J ) 4.0 Hz), 5.52 (t, 1H, J ) 4.0 Hz), 7.18 (s, 1H);
13C NMR δ 18.1, 63.4, 123.5, 151.0. Anal. Calcd for C5H7NO:
C, 61.82; H, 7.27; N, 14.43. Found: C, 61.95; H, 7.31; N, 14.52.
Rea ction of 2,3-Dim eth yl-1,3-bu ta d ien e (12). Reaction
was performed analogously to the isoprene procedure. From
0.39 g (4.9 mmol) of SO3 in 25 mL of CH2Cl2, 0.75 g (10.0 mmol)
of EtONO in 25 mL of CH2Cl2, and 0.4 g (4.9 mmol) of 12 in
15 mL of CH2Cl2 was obtained 0.52 g (97%) of 4,5-dimethyl-
1,2,6H-oxazine (13) (eluent hexane/AcOEt ) 3:1; yellowish oil,
Rf 0.40): IR 1675, 1570 cm-1; 1H NMR (CDCl3) δ 1.78 (s, 3H),
1.87 (s, 3H), 4.31 (s, 2H), 7.62 (s, 1H); 13C NMR δ 17.1, 19.6,
66.3, 145.4. Anal. Calcd for C6H9NO: C, 64.82; H, 8.17; N,
12.61. Found: C, 64.72; H, 8.05; N, 12.61.
in 30 mL of CH2Cl2, 2.32 g (31.0 mmol) of EtONO in 20 mL of
CH2Cl2, and 1.53 g (6.2 mmol) of 1921 in 20 mL of CH2Cl2 was
obtained 2.2 g of crude product. Chromatography of 1.15 g of
this residue (eluent hexane/AcOEt ) 1:1) gave 0.95 g (76%) of
ethyl sulfate of exo-4-hydroxy-cis-endo-9,10-di(methoxycarbo-
nyl)-tricyclo[4.2.2.02,5]dec-7-en-3-one (20) (white crystalline
powder, mp 123-129 °C with partial decomp, Rf 0.23): IR
1
1810, 1750, 1730, 1390, 1200 cm-1; H NMR (CDCl3) δ 1.4 (t,
3H, J ) 7.0 Hz), 2.84 (dt, 1H, J ) 10.0, 3.9 Hz), 2.94 (dd, 1H,
J ) 11.0, 1.7 Hz), 3.02 (dd, 1H, J ) 11.0, 1.7 Hz), 3.34 (m,
1H), 3.45-3.56 (m, 2H), 3.57 (s, 3H), 3.59 (s, 3H), 4.35 (q, 2H,
J ) 7.0 Hz), 4.81 (t, 1H, J ) 3.9 Hz), 6.38 (t, 1H, J ) 6.9 Hz),
6.44 (t, 1H, J ) 6.9 Hz); 13C NMR δ 14.34, 33.38, 33.44, 34.93,
45.32, 45.61, 51.84, 61.38, 70.86, 88.54, 131.01, 132.87, 171.77,
171.92, 199.27. Anal. Calcd for C16H20O9S: C, 49.47; H, 5.19.
Found: C, 49.28; H, 5.14.
Rea ction of Dicyclop en ta d ien e (21). From 0.5 g (6.2
mmol) of SO3 in 30 mL of CH2Cl2, 0.7 g (9.3 mmol) of EtONO
in 30 mL of CH2Cl2, and 0.41 g (3.1 mmol) of 21 in 20 mL of
CH2Cl2 was obtained 0.69 g (82%) of ethyl sulfate of exo-3-
hydroxytricyclo[5.2.1.02,6]dec-8-en-4-one (22) (eluent hexane/
1
AcOEt ) 3:1; yellowish oil, Rf 0.41): IR 1773, 1670 cm-1; H
Rea ction of Cyclop en ta d ien e (15). From 1.52 g (19.0
mmol) of SO3 in 50 mL of CH2Cl2, 2.12 g (28.2 mmol) of EtONO
in 35 mL of CH2Cl2, and 0.62 g (9.4 mmol) of 15 in 20 mL of
CH2Cl2 was obtained 0.50 g (55%) of 2-hydroxycyclopent-3-
enone (16) (eluent hexane/AcOEt ) 3:1; colorless oil, Rf 0.56):
IR 3000-2870, 1755, 1647 cm-1; 1H NMR (CDCl3) δ 2.84 (dd,
2H, J ) 2.0, 2.0 Hz), 3.25 (m, 1H), 5.94 (ddt, 1H, J ) 8.0, 2.0,
2.0 Hz), 6.14 (ddt, 1H, J ) 8.0, 2.0, 2.0 Hz); 13C NMR δ 43.7,
53.2, 127.1, 127.3, 202.0. Anal. Calcd for C5H6O2: C, 61.20;
H, 6.17. Found: C, 61.22; H, 6.12.
Rea ction of 1,5-Cycloocta d ien e (17). From 1.0 g (12.5
mmol) of SO3 in 35 mL of CH2Cl2, 1.42 g (18.9 mmol) of EtONO
in 30 mL of CH2Cl2, and 0.58 g (5.4 mmol) of 17 in 15 mL of
CH2Cl2 was obtained 0.98 g (74%) of ethyl sulfate of 2-hy-
droxycyclooct-5-enone (18) (eluent hexane/AcOEt ) 12:5; color-
less oil, Rf 0.43): IR 1723, 1644 cm-1; 1H NMR (CDCl3) δ 1.68
(t, 3H, J ) 7.0 Hz), 1.80-3.00 (m, 8H), 4.39 (q, 2H, J ) 7.0
Hz), 5.18 (dd, 1H, J ) 7.9, 3.9 Hz), 5.72 (dm, 1H, J ) 9.0 Hz),
5,76 (dm, 1H, J ) 9.0 Hz); 13C NMR δ 14.1, 20.4, 22.3, 30.0,
33.5, 70.1, 81.3, 130.7, 131.8, 176.5. Anal. Calcd for
NMR (CDCl3) δ 1.27 (dm, 1H, J ) -10.4 Hz), 1.32 (dm, 1H, J
) -10.4 Hz), 1.49 (t, 3H, J ) 7.0 Hz), 2.87 (m, 1H), 2.91-3.00
(m, 2H), 3.25 (d, 1H, J ) 4.9 Hz), 3.41-3.52 (m, 2H), 4.40 (q,
2H, J ) 7.0 Hz), 4.93 (d, 1H, J ) 2.8 Hz), 5.61 (dt, 1H, J )
5.8, <1 Hz), 5.76 (dt, 1H, J ) 5.8, <1 Hz); 13C NMR δ 15.1,
33.2, 38.0, 42.2, 48.7, 50.3, 53.3, 70.8, 81.0, 128.0, 133.3, 179.8.
Anal. Calcd for C12H16O5S: C, 52.93; H, 5.93. Found: C, 52.72;
H, 5.71.
Ack n ow led gm en t. This work was supported in part
by a grant from the Russian Foundation of Fundamen-
tal Investigations (99-03-33093). We thank Dr. A. A.
Borisenko for NMR experiments.
Su p p or tin g In for m a tion Ava ila ble: Optimized geom-
etries of intermediates and transition states and the respective
energies of I-XIII. This material is available free of charge
via the Internet at http://pubs.acs.org.
J O990679T
C
10H16O5S: C, 48.37; H, 6.50. Found: C, 48.11; H, 6.45.
Reaction of cis-en do-9,10-Di(m eth oxycar bon yl)-tr icyclo-
(21) Reppe, W.; Schlicting, O.; Klager, K.; Toepel, T. Liebigs Ann.
Chem. 1948, 560, 1.
[4.2.2.02,5]d ec-3,7-d ien e (19). From 1.24 g (15.5 mmol) of SO3