Kinetics and mechanisms of the gas-phase elimination of 2-substituted primary, secondary and tertiary hydroxy groups in nitroalkanes

Article abstract of DOI:10.1002/poc.743

The kinetics of the gas-phase elimination of several 2-substituted primary, secondary and tertiary hydroxy groups in nitroalkanes were determined in a static reaction system over the temperature range 220-400°C and pressure range of 29-235 Torr. The reactions, in seasoned vessels, are homogeneous and unimolecular and obey a first-order rate law. The presence of secondary and tertiary hydroxy substituent at the 2-position of the nitro group in nitroalkanes leads to a retro-aldol type of decomposition. The mechanism may be rationalized in terms of a six-membered cyclic transition state to give the corresponding aldehyde or ketone and the nitroalkane, respectively. However, some of the primary 2-hydroxy groups in nitroalkanes undergo a dehydration process with very limited isomerization to the corresponding alkyl nitrate. The mechanism of dehydration is believed to proceed through a six-membered rather than the already reported four-membered cyclic transition state to give the nitroalkene and water. In the case of the primary hydroxy substituent in 2-methyl-2-nitro-1-pentanol, the products of elimination are HNO₂ gas and 3-hydroxy-2-methyl-1-propene. This reaction is rationalized in terms of a four-membered cyclic transition state type of mechanism. The kinetic and thermodynamic parameters of the hydroxynitroalkane substrates are presented and discussed. Copyright 2004 John Wiley & Sons, Ltd.

Full text of DOI:10.1002/poc.743

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY
J. Phys. Org. Chem. 2004; 17: 399–408
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/poc.743
Kinetics and mechanisms of the gas-phase elimination of
2
-substituted primary, secondary and tertiary hydroxy
groups in nitroalkanes
Rosa Mar ´ı a Dom ´ı nguez, Armando Herize, Alexandra Rotinov, Alvaro Alvarez-Aular, Gonzalo Visbal and
Gabriel Chuchani*
Centro de Qu ı´ mica, Instituto Venezolano de Investigaciones Cient ı´ ficas (IVIC), Apartado 21827, Caracas 1020-A, Venezuela
Received 27 October 2003; revised 10 December 2003; accepted 15 December 2003
ABSTRACT: The kinetics of the gas-phase elimination of several 2-substituted primary, secondary and tertiary
ꢀ
hydroxy groups in nitroalkanes were determined in a static reaction system over the temperature range 220–400 C
and pressure range of 29–235 Torr. The reactions, in seasoned vessels, are homogeneous and unimolecular and obey a
first-order rate law. The presence of secondary and tertiary hydroxy substituent at the 2-position of the nitro group in
nitroalkanes leads to a retro-aldol type of decomposition. The mechanism may be rationalized in terms of a six-
membered cyclic transition state to give the corresponding aldehyde or ketone and the nitroalkane, respectively.
However, some of the primary 2-hydroxy groups in nitroalkanes undergo a dehydration process with very limited
isomerization to the corresponding alkyl nitrate. The mechanism of dehydration is believed to proceed through a six-
membered rather than the already reported four-membered cyclic transition state to give the nitroalkene and water. In
the case of the primary hydroxy substituent in 2-methyl-2-nitro-1-pentanol, the products of elimination are HNO gas
2
and 3-hydroxy-2-methyl-1-propene. This reaction is rationalized in terms of a four-membered cyclic transition state
type of mechanism. The kinetic and thermodynamic parameters of the hydroxynitroalkane substrates are presented
and discussed. Copyright # 2004 John Wiley & Sons, Ltd.
KEYWORDS: gas-phase kinetics; eliminations; pyrolyses; 2-hydroxynitroalkanes
INTRODUCTION
ð3Þ
Previous investigations on the gas-phase elimination of dif-
ferent series of primary, secondary and tertiary 2-hydroxy
compounds have been reported as follows: 2-hydroxyalk-
1–6
enes,
7,8
2,9–13
2-hydroxyacetylenes,
2-hydroxy esters,
18,19
ð4Þ
ð5Þ
ð6Þ
4,9,14–17
-hydroxy ketones,
2
2-hydroxynitriles
and 2-
20
hydroxyalkylbenzenes [reactions (1)–(6)]. From experi-
mental kinetic data and some theoretical studies of these
substrates, a six-membered cyclic transition state type of
mechanism was suggested as depicted in reaction (7).
ð1Þ
ð2Þ
ð7Þ
*
Correspondence to: G. Chuchani, Centro de Qu ´ı mica, Instituto
Venezolano de Investigaciones Cient ´ı ficas (IVIC), Apartado 21827,
Caracas 1020-A, Venezuela. E-mail: chuchani@quimica.ivic.ve
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2004; 17: 399–408

´
R. M. DOMINGUEZ ET AL.
4
00
The C —C bond is believed to be polarized
ꢀ
The average experimental P /P at several different
f 0
ꢁ
ꢂꢁ
ꢂþ
C
position rates increase from primary to tertiary C con-
. . . C in the reaction trasition state. The decom-
temperatures and after 10 half-lives is 2.2 (Table 1). The
small departure from stoichiometry was due to the for-
mation of trace amounts of unidentified reaction pro-
ducts. HOCH CH NO also yields a small amount of the
ꢀ
ꢁ
ꢁ
sistent with stabilization of the ꢂþ charge.
4
August et al. were able to demonstrate, by comparing
2
2
2
different series of primary, secondary and tertiary 2-
hydroxy compounds, mainly 2-hydroxy ketones, 2-hy-
droxy esters and 2-hydroxyalkenes, that the higher the
nucleophilicity of the main group the faster is the rate of
isomer CH CH ONO . However, the stochiometry of
3 2 2
reaction (8), up to 50% reaction, was confirmed by
comparing the percentage decomposition of the substrate
from pressure measurements with that obtained by chro-
matographic analysis of nitroethene formation (Table 2).
The homogeneity of this elimination was examined by
using a packed reaction vessel with a surface-to-volume
ratio six times greater than that of the unpacked vessel
(Table 3). The rates were unaffected by the packed and
unpacked seasoned vessels, whereas a significant hetero-
geneous effect was obtained with the packed and un-
packed clean Pyrex vessels. The effect of the free radical
inhibitor toluene is shown in Table 4. No induction period
was observed. The rates are reproducible with a standard
deviation not greater than 5% at a given temperature.
The first-order rate coefficient of nitroethanol calcu-
lated from k ¼ (2.303/t)log[P /(2P – P )] was indepen-
—
elimination. In other words, the C—O bond is a better
—
nucleophile than the C—CH bond, which accounts for
2
the difference in reactivity in the order keto-
nes > esters > alkenes. Among the types of compounds
containing 2-substituted hydroxy groups as described in
reactions (1)–(6), it seemed of interest to determine the
gas-phase elimination kinetics of several primary, sec-
ondary and tertiary 2-hydroxynitroalkanes and to con-
sider their mechanisms.
RESULTS AND DISCUSSION
Primary 2-hydroxy substituent
1
0
0
t
dent of their initial pressures (Table 5). A plot of log[P /
0
(
5
2P ꢁ P )] against time t gave a good straight line up to
0
t
2
of 2-nitroethanol described in reaction (8), the final
pressure, P , should be approximately twice the initial
f
-Nitroethanol. According to the elimination products
0% reaction. The variation of the rate coefficients with
temperature and the corresponding Arrhenius equation is
given in Table 6 (95% confidence limits from a least-
squares procedure).
pressure, P , i.e. P ¼ 2P .
0
f
0
2
-Nitro-1-propanol
ð8Þ
The elimination products of 2-nitro-1-propanol described
in reaction (9) require approximately a twofold increase
Table 1. Ratio of final (P ) to initial pressure (P )
f
0
ꢀ
a
a
Substrate
Temperature ( C)
P0 (Torr)
Pf (Torr)
Pf/P0
Average
2.2
2
2
-Nitroethanol
303.2
65
70
106
65
127
157
239.5
162
192.5
214
228.5
124
103
91
150
159.5
186
210.5
134
368
304
165
2.0
2.2
2.3
2.5
2.0
2.0
2.1
2.1
2.1
2.3
2.4
1.9
1.9
1.9
2.0
2.0
2.0
2.0
2.0
1.9
1.9
1.9
1.9
3
3
3
14.0
20.5
29.5
-Nitro-1-propanol
270.0
98
2.1
2
2
2
80.2
90.3
98.5
110
111.5
59.5
49.5
40
61.5
92.5
98
110
68
183
155
84
2
3
-Methyl-2-nitro-1-propanol
-Nitro-2-butanol
260.4
70.2
79.3
240.2
2.3
1.9
2
2
2
2
2
50.4
60.5
70.3
3
2
-Nitro-2-pentanol
360.5
2.0
1.9
3
3
4
70.8
80.7
00.8
198
55
70
82
386
-Hydroxy-2-methylnitropropane
245.6
101.5
133.5
155
2
2
2
55.5
64.8
70.0
72
138
a
1
Torr ¼ 133.3 Pa.
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2004; 17: 399–408

GAS-PHASE ELIMINATION OF HYDROXY GROUPS IN NITROALKANES
Table 2. Stoichiometry of the elimination reaction
401
ꢀ
Substrate
Temperature ( C) Parameter
Value
14
2
2
2
3
3
2
-Nitroethanol
303.2
270.0
235.7
260.5
380.7
235.7
Time (min)
5
10
18
Reaction (%) (pressure)
Nitroethene (%) (GLC)
Time (min)
23.7 30.2 41.5
26.3 30.1 38.7
10 15
16.9 30.6 43.1
19.8 34.0 47.0
10 15
20.1 37.0 51.5
-Hydroxy-2-methyl-1-propene (%) (GLC) 21.6 39.3 52.8
51.6
48.5
20
-Nitro-1-propanol
5
30
Reaction (%) (pressure)
52.8 66.3
53.9 63.7
18
57.1
60.1
2
-Nitropropene (%) (GLC)
Time (min)
Reaction (%) (pressure)
-Methyl-2-nitro-1-propanol
-Nitro-2-butanol
5
3
Time (min)
3
4.5
17.8 30.7 41.8
19.6 31.4 39.4
3.5
25.0 34.9 55.1
21.2 30.0 55.4
10 15 20
25.3 36.1 42.6
23.2 34.2 40.9
7
10
14
Reaction (%) (pressure)
Nitroethane (%) (GLC)
Time (min)
Reaction (%) (pressure)
3-Nitropropane (%) (GLC)
Time (min)
50.5 62.8
53.3 65.7
15
74.5
76.5
-Nitro-2-pentanol
1
7
-Hydroxy-2-methylnitropropane
25
51.4 56.8
49.8 53.8
30
Reaction (%) (pressure)
Acetone (%) (GLC)
in the final pressure, P , i.e. P /P ¼ 2.0. The actual
The effect of the surface area on the rate of elimination
was tested by carrying out several runs in a vessel with a
surface-to-volume ratio of 6.0 relative to that of the
normal vessel, which is equal to 1.0 (Table 3). Packed
and unpacked Pyrex vessels seasoned with allyl bromide
showed no effect on rates. However, the packed and
unpacked clean Pyrex vessels had a marked heteroge-
neous effect on the rate coefficients of 2-nitro-1-propanol.
The addition of different proportions of the free radical
inhibitor toluene (Table 4) showed no effect on rates. No
induction period was observed. The rate coefficients were
reproducible with a relative standard deviation of 5% at
any given temperature.
f
f
0
experimental results for P /P at four different tempera-
f
0
tures and 10 half-lives is 2.1 (Table 1). Although 2-
nitropropanol produces small amounts of the isomer
propyl nitrate, the stoichiometry of reaction (9), up to
7
0% decomposition, was confirmed by comparing the
pressure measurements with the quantitative chromato-
graphic analyses of 2-nitropropene (Table 2).
ð9Þ
The rate coefficient also calculated from k ¼ (2.303/
1
t)log[P /(2P – P )] was invariable with respect to of their
0
0
t
initial pressures (Table 5). On plotting log[P /(2P – P )]
0
0
t
Table 3. Homogeneity of the elimination reaction
against time t, a good straight line, up to 70% reaction,
was obtained. The temperature dependence of the rate
coefficients is given in Table 6. The data were fitted to the
Arrhenius equation shown where 90% confidence limits
from a least-squares procedure are given.
4
10 k
4
Substrate
S/V
cm
10 k
1
1
ꢁ
1 a
)
ꢁ1 b
ꢁ1 c
(s )
(
(s
)
ꢀ
2
2
2
-Nitroethanol at 303.2 C
1
21.61
21.20
22.41
26.16
14.53
5.96
6.07
6.15
6.31
7.79
6
1
6
1
ꢀ
d
d
-Nitro-1-propanol at 270.0 C
2-Methyl-2-nitro-1-propanol
-Methyl-2-nitro-1-propanol
ꢀ
The unimolecular elimination of this substrate, in a
seasoned reaction vessel and in the presence of the free
radical suppressor toluene, proceeds according to reac-
tion (10). The stoichiometry demands P /P ¼ 2.0. The
at 270.0 C
6
1
6
1
6
1
23.13
15.02
41.13
6.03
11.92
18.64
8.25
12.85
13.23
5.60
5.87
4.26
ꢀ
-Nitro-2-butanol at 260.5 C
3
3
2
f
0
ꢀ
average P /P at three different temperatures and 10 half-
f 0
-Nitro-2-pentanol at 381.0 C
lives was 2.3. The departure from the theoretical stoi-
chiometry increases with increase in temperature. The
variation was due to the formation of trace amounts of
nitropopane, isobutyl nitrite, isobutyl nitrate, 1-hydroxy-
2-methyl-1-propene, 1-hydroxy-2-methyl-2-propene and
isobutyraldehyde. However, it was possible to check the
stoichiometry of reaction (10), up to 50% reaction, by
comparing the pressure measurements with quantitative
-Hydroxy-2-methylnitropropane
ꢀ
at 235.7 C
6
19.26
4.10
a
S ¼ surface area; V ¼ volume.
b
Clean Pyrex vessel.
Vessel seasoned with the products decomposition of allyl bromide (420–
c
ꢀ
4
50 C).
Average k values.
d
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2004; 17: 399–408

´
R. M. DOMINGUEZ ET AL.
4
02
Table 4. Effect of the free radical chain inhibitor on rates
ꢀ
a
Ps (Torr)
b
Pi (Torr)
4
10 k (s
ꢁ1
Substrate
Temperature ( C)
Pi/Ps
)
1
c
2
-Nitroethanol
314.0
88.5
—
53.5
87
244
—
—
0.6
1.2
3.4
—
0.4
0.9
1.7
2.5
—
1.2
2.5
3.0
3.5
—
0.5
0.9
1.5
2.4
—
6.25
9.64
9
7
7
6
0
1.5
9.61
9.36
2
-Nitro-1-propanol
290.3
270.2
260.5
380.5
85
95
26
11.5
73
72
24.10
23.75
23.55
23.73
23.52
7.12
7.62
7.79
7.72
7.80
12.55
12.68
12.78
12.47
12.90
6.10
5.51
5.08
5.25
7.47
1
1
1
84
123.5
185
199.5
—
64.5
113
121
133
—
62.5
82.5
138.5
163.5
—
134
168
182
—
2
3
-Methyl-2-nitro-1-propanol
-Nitro-2-butanol
5
4
4
3
3
5.5
0
8
103
28
1
9
9
7
0.5
3.5
1.5
3
2
-Nitro-2-pentanol
235
84
60
0.6
1.1
2.9
—
0.6
1.3
1.7
1
1
6
3
-Hydroxy-2-methylnitropropane
245.6
94
01
59
79.5
112
7.56
7.67
6
6
1
4.5
7.51
a
b
c
P
P
s
¼ pressure of substrate.
i
¼ pressure of free radical inhibitor.
Average k value.
Table 5. Invariability of the rate coefficients from initial pressure
ꢀ
Substrate
Temperature ( C)
Parameter
Value
92
14.32
111
23.41
54.5
4.16
103
12.55
182
5.07
89.5
31.40
2
2
2
3
3
2
-Nitroethanol
320.5
P (Torr)
0
0 k (s
40
14.00
73
23.52
29
4.34
44.5
12.15
84
5.51
66.5
31.81
66
14.06
87
23.23
49
4.38
71.5
12.90
157
5.15
74
31.73
106
14.36
126
23.55
66
4.17
128
12.68
210
5.70
105
31.65
4
ꢁ1
1
P (Torr)
)
)
)
)
)
)
1
-Nitro-1-propanol
-Methyl-2-nitro-1-propanol
-Nitro-2-butanol
290.3
260.4
260.5
380.5
195
0
0 k (s
4
ꢁ1
1
P (Torr)
23.75
97
4.16
150
12.56
235
5.60
1
0
0 k (s
4
ꢁ1
1
1
P (Torr)
0
4
ꢁ1
0 k (s
1
1
-Nitro-2-pentanol
P (Torr)
1
0
0 k (s
4
ꢁ1
1
-Hydroxy-2-methylnitropropane
270.0
P (Torr)
0
4
ꢁ1
0 k (s
1
1
chromatographic analysis of the main product 3-hydroxy-
-methyl-1-propene (Table 2). The homogeneity of this
affected by the presence of different proportions of the
free radical inhibitor toluene (Table 4). No induction
period was observed. The k values are reproducible
with a relative standard deviation not greater than 5%
at a given temperature.
2
pyrolytic elimination was examined using a reaction
vessel with different surface-to-volume ratios, i.e. the
packed vessel had a 6.0 times greater surface-to-volume
ratio than the unpacked vessel (Table 3). The packed and
unpacked clean Pyrex vessels resulted in a marked
heterogeneous effect. The packed seasoned vessel
showed a very small surface effect. This substrate was
always pyrolyzed in the unpacked seasoned Pyrex vessel.
The elimination of this nitro alcohol substrate was not
ð10Þ
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2004; 17: 399–408

GAS-PHASE ELIMINATION OF HYDROXY GROUPS IN NITROALKANES
Table 6. Variation of the rate coefficients with temperature
403
Substrate
Parameter
Temperature ( C)
Value
314.0
ꢀ
2
-Nitroethanol
291.2
296.2
3.37
log k (s ) ¼ (11.14 ꢂ 0.48) – (158.9 ꢂ 5.3) kJ mol (2.303 RT) , r ¼ 0.9978
303.2
5.96
320.5
14.28
329.5
22.91
4
0 k (s
ꢁ1
1
Rate equation
)
2.68
9.56
1
ꢁ
1
ꢁ1
ꢁ1
1
ꢀ
2
-Nitro-1-propanol Temperature ( C)
248.2
259.8
3.16
log k (s ) ¼ (12.40 ꢂ 0.47) – (162.1 ꢂ 5.0) kJ mol (2.303 RT) , r ¼ 0.9995
270.0
280.2
290.3
298.7
4
0 k (s
ꢁ1
1
Rate equation
)
1.44
6.15
13.24
23.41
37.75
1
ꢁ
1
ꢁ1
ꢁ1
1
ꢀ
2-Methyl-2-nitro-1- Temperature ( C)
propanol
239.4
249.5
2.04
log k (s ) ¼ (11.64 ꢂ 0.06) – (153.4 ꢂ 0.6) kJ mol (2.303 RT) , r ¼ 0.9998
260.4
270.2
279.3
4
10 k (s
ꢁ1
)
Rate equation
1.01
4.24
7.79
13.61
1
ꢁ
1
ꢁ1
ꢁ1
1
ꢀ
3
3
2
-Nitro-2-butanol
Temperature ( C)
220.4
230.9
1.70
log k (s ) ¼ (11.84 ꢂ 0.21) – (150.4 ꢂ 2.1) kJ mol (2.303 RT) , r ¼ 0.9996
400.8
240.2
250.4
260.5
270.3
4
0 k (s
ꢁ1
1
Rate equation
)
0.86
3.55
7.00
12.85
24.52
1
ꢁ
1
ꢁ1
ꢁ1
1
ꢀ
-Nitro-2-pentanol Temperature ( C)
340.7
351.1
0.90
log k (s ) ¼ (12.93 ꢂ 0.24) – (202.8 ꢂ 3.0) kJ mol (2.303 RT) , r ¼ 0.9995
360.7
370.8
380.8
391.0
4
0 k (s
ꢁ1
1
Rate equation
)
0.50
1.62
2.91
5.24
9.44
17.37
1
ꢁ
1
ꢁ1
ꢁ1
1
ꢀ
-Hydroxy-2-
Temperature ( C)
225.7
230.4
2.43
log k (s ) ¼ (11.66 ꢂ 0.40) – (146.9 ꢂ 4.0) kJ mol (2.303 RT) , r ¼ 0.9981
235.7
245.6
255.5
264.8
270.0
31.81
4
ꢁ1
methylnitropropane 10 k (s
)
Rate equation
1.79
4.26
7.55
14.25
24.92
1
ꢁ
1
ꢁ1
ꢁ1
1
The rate coefficients, in seasoned vessel, from
k ¼ (2.303/t)log[P /(2P ꢁ P )] were found to be inde-
period was detected. The rate coefficients are reproduci-
ble with a relative standard deviation not greater than 5%
at a given temperature.
1
0
0
t
pendent of the initial pressure of the substrate (Table 5),
and the first-order plots of log[P /(2P ꢁ P )] versus time t
The rate coefficients of decomposition of 3-nitro-
1-butanol were calculated from k ¼ (2.303/t)log[P /
0
0
t
are satisfactorily linear up to 50% reaction (Table 5). The
temperature dependence of the rate coefficients and the
corresponding Arrhenius equation is described in Table 6
1
0
(2P ꢁ P )] and were found to be independent of the
0
t
initial pressure (Table 5). A plot of log[P /(2P ꢁ P )]
0
0
t
(
90% confidence levels with a least-squares method).
against time t gave a good straight line up to 70%
reaction. The variation of the rate coefficients with
temperature and the corresponding Arrhenius equation
are given in Table 6 (90% confidence limits from a least-
squares procedure).
Secondary 2-hydroxy substituent
3
the elimination of 3-nitro-2-butanol in the gas phase
-Nitro-2-butanol. The experimental stoichiometry for
3-Nitro-2-pentanol. The elimination reaction of this
substrate, in a seasoned vessel and in the presence of a
free radical inhibitor [reaction (12)] implies P /P ¼ 2.0.
[
reaction (11)] is P ¼ 2P . The average P /P at four
f
0
f
0
different temperatures and 10 half-lives was 1.9. The
small departure from the theoretical stoichiometry was
due to the very slight polymerization of acetaldehyde.
Further verification of the stoichiometry of reaction (11)
was possible by comparing, up to 70% decomposition,
the measurements of pressure increase with the quantita-
tive analysis of nitroethane formation (Table 2).
f
0
The average experimental P /P value measured after 10
f
0
half-lives and at four different temperatures was 2.0
(Table 1). A further verification of the stoichiometry of
reaction (12), up to 75% decomposition, is that the
quantitative chromatographic analysis of nitropropane
formation corresponds to the pressure increase (Table 2).
ð11Þ
ð12Þ
The homogeneity of this elimination was investigated
by using a packed reaction vessel with a surface-to-
volume ratio six times greater than that of the unpacked
reaction vessel (Table 3). The rate coefficients are very
similar in the packed and unpacked vessels seasoned with
allyl bromide. However, the packed and unpacked clean
Pyrex vessels showed a marked effect on the rates.
The process of elimination of this substrate was not
affected by the presence of different proportions of a free
radical inhibitor such as toluene (Table 4). No induction
To examine the effect of surface area on the rates,
several runs in the presence of toluene inhibitor were
performed in a vessel with a surface-to-volume ratio
six times greater than that of the normal vessel
(Table 3). The packed clean Pyrex vessel resulted in a
significant heterogeneous effect. This nitropentanol sub-
strate was always pyrolyzed in seasoned vessel and in the
presence of at least a twofold excess of the free radical
inhibitor toluene (Table 4). No induction period was
observed and the rates were reproducible with a relative
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2004; 17: 399–408

´
R. M. DOMINGUEZ ET AL.
4
04
standard deviation not greater than 5% at a given tem-
perature.
versus time t, a good straight line, up to 50% decomposi-
tion, was obtained. The variation of the rate coefficients
with temperatures is shown in Table 6. The experimental
data were fitted to the Arrhenius equation (Table 6),
where 90% confidence limits from a least-squares proce-
dure are given.
The first-order rate coefficient of this substrate calcu-
lated from k ¼ (2.303/t)log[P /2(2P ꢁ P )] was inde-
1
0
0
t
pendent of the initial pressure (Table 5). A plot of
log[P /(2P ꢁ P )] against time t gave a good straight
0
0
t
line up to 75% decomposition. The temperature depen-
dence of the rate coefficients together with the corre-
sponding Arrhenius equation are given in Table 6 (90%
confidence limits from a least-squares method).
Tertiary 2-hydroxy substituent
2
-Hydroxy-2-methylnitropropane. The product for-
mation of this nitro alcohol is described in reaction (13).
The stoichiometry requires that, for long reaction time, Pf/
P ¼ 2.0. The average experimental P /P values at four
0
f
0
different temperatures and 10 half-lives was 1.9 (Table 1).
Further verification of the above stoichiometry, up to 50%
reaction, was possible by comparing the percentage
decomposition of the substrate from pressure measure-
ments with that obtained from the quantitative chromato-
graphic analyses of the product acetone (Table 2).
ð14Þ
ð13Þ
The experimental results and the data specified in
Table 7 require careful analyses for mechanistic interpret-
ations. In the case of the substrates with primary
2-substituted hydroxy groups, as in 2-nitroethanol and
2-nitro-1-propanol, the likely process for nitrate forma-
tion is an abstraction of the H of the OH by the oxygen of
The homogeneity of this pyrolytic elimination was
examined in a vessel with a surface-to-volume ratio six
times greater than that of the normal vessel (Table 3). The
rates of formation of acetone and/or nitromethane were
unaffected in seasoned packed and unpacked Pyrex
vessels. However, clean packed and unpacked Pyrex
vessels showed a significant effect on the rates. The effect
of different proportions of the free radical inhibitor
toluene on the elimination reaction is given in Table 4.
No induction period was observed. The rate coefficients
were reproducible with a standard deviation not greater
than 5% at a given temperature.
the NO group as shown in reaction (14A). This process
2
of isomerization was found to be extremely rapid and a
kinetic determination was not possible. However, the
main elimination product is the dehydration reactions
(8) and (9). Because of the low-temperature pyrolysis
work on primary nitro alcohols and the values of log
A ¼ 11.28 and 12.40 (Table 7) of these substrates, the
water elimination through a four-membered cyclic state
transition state type of mechanism [reaction (14C)]
appears to be unlikely. Moreover, primary aliphatic
The rate coefficients estimated from k ¼ (2.303/
1
t)log[P /(2P ꢁ P )] were invariant with respect to their
0
0
t
initial pressure (Table 5). On plotting log[P /(2P ꢁ P )]
alcohols with a C —H bond are difficult to dehydrate
ꢁ
0
0
t
ꢀ
Table 7. Kinetic and thermodynamic parameters for pyrolysis of the 2-hydroxynitroalkanes at 280 C
4
6
6
6¼
ꢀG
Z
k ꢃ 10
Ea
(kJ mol
Log A
)
ꢀS
(J mol
ꢀH
(kJ mol
1
ꢁ
1
ꢁ1
ꢁ1
ꢁ1 ꢁ1
ꢁ1
ꢁ1
(kJ mol )
(
s
)
)
(s
)
)
HOCH CH NO
2
1.35
12.30
14.13
42.66
0.006
60.26
158.9 ꢂ 5.3
162.1 ꢂ 5.0
153.4 ꢂ 0.6
150.4 ꢂ 2.1
202.8 ꢂ 3.0
146.9 ꢂ 4.0
11.14 ꢂ 0.48
12.40 ꢂ 0.47
11.64 ꢂ 0.06
11.84 ꢂ 0.21
12.93 ꢂ 0.24
11.66 ꢂ 0.40
ꢁ45.0
154.3
157.5
148.8
145.8
198.2
142.3
179.2
169.1
168.4
163.3
204.2
161.7
2
2
HOCH CH(CH )NO
2
ꢁ20.9
ꢁ35.5
ꢁ31.6
ꢁ10.8
ꢁ35.1
2
3
HOCH C(CH ) NO
2
2
3 2
CH CH(OH)CH(NO )CH
3
3
2
CH CH(OH)CH(NO )CH CH
3
3
CH ) C(OH)CH NO
2
2
(
3 2 2 2
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2004; 17: 399–408

GAS-PHASE ELIMINATION OF HYDROXY GROUPS IN NITROALKANES
405
ꢀ
Table 8. Kinetic and thermodynamic parameters for pyrolysis of HOCH CH Z at 280 C
2
2
4
ꢁ1
ꢁ1
ꢁ1
6
ꢁ1 ꢁ1
6
ꢁ1
6
ꢁ1
Z
k ꢃ 10 (s
)
Ea (kJ mol
)
Log A (s
)
ꢀS (J mol
K
)
ꢀH (kJ mol
)
ꢀG (kJ mol
)
1
COCH3
CN
C H
43.9
155.7
227.0
229.2
158.9
12.35
13.12
12.40
11.14
ꢁ21.9
ꢁ7.1
151.1
222.4
224.6
154.3
163.2
226.3
236.2
179.2
0.0000478
0.00000563
1.35
ꢁ20.9
ꢁ45.0
6
5
NO2
ꢀ
even at temperatures higher than 500 C, for a four-
membered cyclic transition state elimination in the gas
parallel reactions have only been reported for hydroxy
18
hydroxynitriles and hydro-
4
esters, hydroxy ketones,
4,16
20
19
xylalkylbenzenes [reactions (15)–(18)].
phase. Ethanol was difficult to pyrolyze, but under
maximum inhibition and at a single temperature of
ꢀ
ꢁ5
5
25 C gave a k value of 1.1 ꢃ 10 . However, propanol
ð15Þ
ð16Þ
showed a vatiation of the rate coefficient with tempera-
ture according to the following Arrhenius expression:
ꢁ
1
ꢁ1
ꢁ1
log k ðs Þ ¼ 13:64 ꢁ 279:4 kJ mol (2.303RT) .Com-
1
parison of the rate coefficents of ethanol, propanol
and 2-nitroethanol at 525 C are given in Related Com-
parison 1.
ꢀ
ð17Þ
ð18Þ
ꢁ
1
ꢁ1
Propanol: log k (s ) ¼ 13.64–279.4 kJ mol
(2.303
1
ꢁ1
ꢁ1
ꢁ1
RT) 2-Nitroethanol: log k (s ) ¼ 11.28–160.6kJ mol
1
ꢁ
1
(
2.303 RT)
Etanol
Propanol 2-Nitroethanol
ꢁ
1
ꢀ
1.1 ꢃ 10ꢁ5 2.3 ꢃ 10ꢁ5
Related Comparison 1
According to the data in Table 8, where the range of
log A is 11.28–12.40, with the apparent exception of 2-
hydroxypropionitrile (13.12), a six-membered rather than
a four-membered cyclic transition state appears to be the
most probable mechanism for water elimination. This
assumption appears to be supported by the very negative
entropy of activation. In addition to this consideration,
when estimating the rate of elimination of water molecule
k1 (s ) at 525 C
5.85
The replacement of the H of ethanol by an NO group
2
cannot be a factor in increaseing the dehydration process
of the alcohol through the same mechanistic process of a
four-membered cyclic transition state with a rate as fast
as 500 000 s . This dramatic difference is hardly ob-
tained in gas-phase elimination reactions in the same
series of compounds. Consequently, the high dipole of the
ꢁ
1
ꢀ
of the hydroxy compounds listed in Table 8 at 525 C, the
difference is found to be extremely large (Related Com-
parison 2). With the same arguments for the transition
state of dehydration of nitroethanol, the substrates de-
scribed in Table 8 may well be undergo a similar six-
membered cyclic transition state as pictured in Scheme 1.
þ
ꢁ
—
NO group, as O—N —O and the proximity of the
2
adjacent C —H for charge stabilization through isomer-
ꢀ
ization must not be overlooked. Therefore, it is not
unreasonable to believe that the mechanism of dehydra-
tion may well go through a six-membered cyclic transi-
tion state as described in reaction (14B).
The rate of dehydration of 2-nitro-1-propanol was
found to be faster than that of 2-nitroethanol (Table 7).
This marked difference apparently arises from the ab-
straction of the acidic H of the —ONOH group by OH to
form a water molecule [reaction (14B)], thus leaving the
adjacent C atom partially positive. This positive C atom
Relative
H
CH3 C H CN
6
NO2
CH CO
3
5
ꢀ
3
3
rate a 525 C 1
2.1
166 226 532 ꢃ 10 13100 ꢃ 10
Related Comparison 2
ꢁ
ꢁ
may then be stabilized through the electronic release of
the CH in the nitropropanol for augmentation in elim-
ination.
3
Scheme 1
The dehydration process of the primary 2-hydroxy
compounds has also been observed as a parallel reaction
described in reactions (1)–(6). The authors concerned ass-
umed a four-membered cyclic transition state mechanisms
for this side-reaction. Complete kinetic studies of these
With regard to the other primary 2-hydroxynitroalkane,
i.e. 2-methyl-2-nitro-1-propanol, there is no H attached to
C ; instead, there are two CH group, i.e.—C (CH ) .
ꢀ
3
ꢀ
3 2
Consequently, no isomerization of the H for dehydation
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2004; 17: 399–408

´
R. M. DOMINGUEZ ET AL.
4
06
may take place. This appears to support the mechanism in
reaction (14B), based on the importance of the H at the
C atom. In this case, the NO group is found to be
described in reactions (1)–(6). Therefore, secondary and
tertiary hydroxy groups at the 2-position of nitroalkanes
appear to proceed through a six-membered cyclic transi-
tion state type of mechanism as described in reaction (7).
Therefore, the process of decomposition may be depicted
as in reactions (19) and (20).
ꢀ
2
attached to a tertiary carbon atom. Consequently, the
nitro group is displaced by the C —H bond of the CH3
ꢁ
group through a four-membered cyclic transition state
type of mechanism (Scheme 2). This type of elimination
is similar to the known and well-established mechanism
for tertiary butyl compounds containing a leaving group
Z attached to the tertiary carbon atom (Scheme 2).
ð19Þ
Scheme 2
ð20Þ
2-Methyl-2-nitropropane has been reported to produce
isobutene and HNO in the gas phase with the following
21
2
ꢁ
1
ꢁ1
Arrhenius expresion: log k (s ) ¼ 13.63–179.3 kJ mol
1
ꢁ
1
2.303 RT) . The reaction rate of this compound is
(
expected to be faster than that of 2-methyl-2-nitro-1-
The comparative kinetic and thermodynamic parameters
for the gas-phase elimination of different series of 2-substi-
tuted secondary and tertiary hydroxy compounds are given in
Tables 9 and 10. These results seems to provide an additional
ꢀ
propanol (Table 7) at 280 C. However, the result is the
contrary (Related Comparison 3). The effect of the
HOCH group attached to the tertiary C atom should
2
destabilize the polarization of the C—Z bond, which is
rate determining. Therefore, the reaction rate of 2-methyl-
4
support for August et al.’s idea about the fact that the
faster elimination rate of the substrate is due to the higher
nucleophilic nature of the main group. This view accounts
for the difference in reactivity in the order hydroxy
ketone>hydroxynitroalkane>hydroxyester>hydroxyalkyne
2
-nitro-1-propanol should be slower than that of 2-methyl-
-nitropropane. Some additional work is needed to ac-
2
count for such a difference.
>
hydroxyalkane> hydroxynitrile > hydroxyalkylbenzene.
(
CH ) CNO
3
(HOCH )C(CH ) NO
2 3 2 2
3
2
ꢀ
4.90 ꢃ 10ꢁ4
14.13 ꢃ 10ꢁ4
At 280 C
Related Comparison 3
EXPERIMENTAL
According to the product formation from the secondary
reactions (11) and (12)] and tertiary hydroxynitroalkane
reaction (13)], a retro-aldol type of decomposition takes
2-Hydroxy-2-methyl-1-nitropropane. This substrate was
prepared by treating nitromethane (1.03 mol) with
LiAlH (0.0014 mol) in 15 ml of dry tetrahydrofuran.
4
[
[
place, which is similar to the elimination process
The reaction mixture was stirred under an argon
ꢀ
Table 9. Comparative kinetic and thermodynamic parameters for pyrolysis of secondary hydroxy substituent at 360 C
4
6
6
6¼
ꢀG
k ꢃ 10
Ea
(kJ mol
Log A
)
ꢀS
(J mol
ꢀH
(kJ mol
1
ꢁ
1
ꢁ1
ꢁ1
ꢁ1 ꢁ1
ꢁ1
ꢁ1
(kJ mol )
Z
(s
)
)
(s
)
)
Ref.
—
CH CH(OH)CH CH—CH
2
5.56
162.2
162.2
164.8
142.1
139.2
164.4
163.1
222.6
240.1
150.4
10.13
10.83
11.30
11.64
11.10
10.90
11.25
13.76
13.04
11.84
ꢁ65.5
156.9
156.9
159.5
136.8
133.9
159.1
157.8
217.3
234.8
145.1
198.4
189.9
186.8
160.0
163.6
191.3
185.7
214.8
241.0
165.9
4
5
6
16
4
4
11
18
20
3
2
2
7.86
50.16
8196.8
ꢁ52.1
ꢁ43.1
ꢁ36.6
ꢁ46.9
ꢁ50.8
ꢁ44.1
–
2
CH CH(OH)CH C–CH
3
CH CH(OH)CH COCH
3
2
3
7
253.1
21.55
61.76
0.25
0.0017
2683.6
CH CH(OH)CH COOCH
3
3
CH CH(OH)CH COOCH CH
2
3
2
2
3
–
CH CH(OH)CH C–N
3.97
3
CH CH(OH)CH C H
2
ꢁ9.81
3
2 6 5
CH CH(OH)CH(CH )NO
ꢁ32.8
This work
3
3
2
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2004; 17: 399–408

GAS-PHASE ELIMINATION OF HYDROXY GROUPS IN NITROALKANES
Table 10. Comparative kinetic and thermodynamic parameters for pyrolisis of tertiary Hydroxy substituent at 300 C
407
ꢀ
ꢁ
4
6
6
6¼
ꢀG
k ꢃ 10
Ea
(kJ mol
Log A
)
ꢀS
(J mol
ꢀH
(kJ mol
1
ꢁ
1
ꢁ1
ꢁ1
ꢁ1 ꢁ1
ꢁ1
ꢁ1
(kJ mol )
Z
(s
)
)
(s
)
)
Ref.
—
CH ) C(OH)CH CH—CH
(
1.94
160.3
160.3
166.9
133.4
133.9
159.9
174.5
212.5
222.9
146.9
10.90
10.90
11.60
11.36
11.13
11.18
12.39
13.68
12.65
11.66
ꢁ50.0
155.5
155.5
162.1
128.6
129.1
155.1
169.7
207.7
218.1
142.1
184.2
184.2
183.1
152.2
155.2
180.7
182.2
205.9
227.6
162.1
4
5
6
16
4
4
13
18
20
3
2
2
2
1
.94
2.43
1583.3
39.4
4.02
3.04
0.020
0.00021
185.8
ꢁ50.0
ꢁ36.6
ꢁ41.2
ꢁ45.6
ꢁ44.6
ꢁ21.5
–
2
(
(
CH ) C(OH)CH C–CH
3
2
CH ) C(OH)CH COCH
3
2
2
3
8
(
(
(
(
(
CH ) C(OH)CH COOCH
3
2
2
3
CH ) C(OH)CH COOCH CH
3
3
2
2
2
–
CH ) C(OH)CH C–N
3.21
3
2
CH ) C(OH)CH C H
2
ꢁ16.51
3
2
2 6 5
CH ) C(CH OH)NO
2
ꢁ35.46
This work
3
2
2
ꢀ
atmosphere at 0 C for 30 min. Anhydrous acetone
2.05 mol) was then slowly added (5 h) while stirring at
0
2-Nitroethanol (Aldrich), 2-nitro-1-propanol (K&K
(
Labs), 2-methyl-2-nitro-1-propanol (Aldrich), 3-nitro-2-
butanol (Aldrich) and 3-nitro-2-pentanol (Aldrich) were
shown to be better than 98.5% pure when analyzed by
GC–MS (Saturn 2000, Varian) with a DB-5MS capillary
column (30 m ꢃ 0.25 mm i.d., 0.25 mm film thickness).
Quantitative chromatographic analyses and identifica-
tions of products were also performed by GC–MS (Saturn
2000, Varian) with a DB-5MS capillary column
(30 m ꢃ 0.53 mm. i.d., 0.53 mm film thickness).
ꢀ
C. The reaction mixture was kept overnight to reach
room temperature, then 30 ml of 1M HCl and 25 ml of
distilled water were added to stop the reaction. The
mixture was extracted with three portions of CH Cl .
2
2
The organic layer was dried over anhydrous Na SO ,
2
4
ꢀ
filtered and evaporated at 55 C. The excess of nitro-
ꢀ
methane was removed at 75 C under vacuum. The liquid
product (12% yield) was found to be of 99.1% pure as
determined by GC–MS (Saturn 2000, Varian) with a DB-
5
thickness. H NMR: ꢂ 1.3 (s, 6H), 3.3 (s, 1H), 4.4 (s, 2H).
MS capillary column, 30 m ꢃ 0.53 mm i.d., 0.53 mm film
Kinetics. Kinetic experiments were carried out in a static
reaction system as depicted in Fig. 1, as reported,
with an Omega DP41-TC/DP41-RTD high-performance
digital temperature indicator. The rate coefficients were
1
22,23
þ
þ
MS: m/z 119 (M ), 104 [CH C(OH)CH NO ], 59
3
2
2
þ
þ
–
(CH ) COH ], 43 (CH C–O ).
3 2 3
[
Figure 1. Static reaction system
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2004; 17: 399–408

´
R. M. DOMINGUEZ ET AL.
4
08
6
. Quijano J, David J, Sanchez C, Rincon E, Guerra D, Leon LA,
determined manometrically or by the quantitative
analyses of the starting material and/or by the corre-
sponding product. The temperature was controlled by a
Shinko DC-PS resistance thermometer controller and an
Notario R, Abboud JL. J. Mol. Struct. (Theochem) 2002; 580:
01–2057.
7. Viola A, MacMilan JH, Proverb RJ, Yates BL. J. Am. Chem. Soc.
971; 93: 6967–6974.
2
1
8
. Quijano J, Sanchez C, David J, Leon LA, Notario R, Abboud JL.
J. Mol. Struct. (Theochem) 2001; 572: 135–139.
Omega SSR280A45 solid-state relay, maintainedꢂ
ꢀ
0
.2 C, and measured with a calibrated platinum–plati-
9. Quijano J, Rodriguez M, Yepes M del S, Gallego LH. Tetrahedron
Lett. 1987; 28: 3555–3558.
0. Yates BL, Quijano J. J. Org. Chem. 1970; 35: 1239–1240.
1. Yates BL, Ramirez A, Velazquez O. J. Org. Chem. 1961; 36:
3579–3582.
2. Quijano J, Restrepo I, Gallego LH, Yepes M del S. Tetrahedron
Lett. 1994; 35: 4735–4738.
3. Dominguez RM, Chuchani G, Quijano J, Orozco LJ, Restrepo I.
React. Kinet. Catal. Lett. 1996; 57: 191–196.
4. Smith GG, Yates BL. J. Org. Chem. 1965; 30: 2067.
5. Yates BL, Quijano J. J. Org. Chem. 1969; 34: 2506–2508.
6. Rotinov A, Chuchani G, Machado R, Rivas C, Quijano J, Yepes M
del S, Restrepo I. Int. J. Chem. Kinet. 1992; 24: 909–915.
7. Feng W, Wang Y, Zhang S. Int. J. Quantum Chem. 1997; 62:
num–13% rhodium thermocouple. No temperature gra-
dient was observed along the reaction vessel. The starting
materials were all injected directly into the reaction
vessel with a syringe through a silicone-rubber septum.
The amount of substrate used for each reaction was
ꢄ0.05–0.2 ml.
1
1
1
1
1
1
1
REFERENCES
1
1
1
2
97–3027.
8. Chuchani G, Dominguez RM, Rotinov A, Quijano J, Valencia C,
Vicente B, Franco D. J. Phys. Org. Chem. 1999; 12: 19–23.
9. Chuchani G, Rotinov A, Dom ´ı nguez RM. Int. J. Chem. Kinet.
1
2
3
4
. Smith GG, Yates BL. J. Chem. Soc. 1965; 7242–7246.
. Arnold RT, Smolinsky G. J. Am. Chem. Soc. 1959; 81: 6443–6445.
. Smith GG, Taylor R. Chem. Ind. (London) 1961; 949–950.
. August R, McEwen I, Taylor R. J. Chem. Soc., Perkin Trans. 2
1
999; 31: 401–407.
2
2
2
2
0. Maccoll A, Thomas PJ. Prog. React. Kinet. 1967; 4: 119–148.
1. Nazin GM, Manelis GB. Russ. Chem. Rev. 1994; 63: 313–322.
2. Maccoll A. J. Chem. Soc. 1955; 965–973.
1
987; 1683–1689.
. Failes RL, Shapiro JS, Stimson VR. In The Chemistry of
the Hydroxy Group, Patai S (ed). Wiley: Chichester, 1980;
49–468.
5
3. Swinbourne ES. Aust. J. Chem. 1958; 11: 314–330.
4
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2004; 17: 399–408