Spontaneous dehydration mechanism of aromatic aldehyde reactions with hydroxyl and non-hydroxyl amines

Article abstract of DOI:10.1002/kin.10095

The plot of rate constants vs. pH for the dehydration step of the reaction between furfural and 5-nitrofurfural with hydroxylamine, N-methylhydroxylamine, and O-methylhydroxylamine, shows two regions corresponding to the oxonium ion-catalyzed and spontaneous dehydration. The oxonium ion-catalyzed dehydration region of the reaction of furfural with the above mentioned hydroxylamines exhibits general acid catalysis with excellent Bronsted correlation (Bronsted coefficients: 0.76 (r = 0.986), 0.68 (r = 0.987), and 0.67 (r = 0.993) respectively). However, the rate constants of the spontaneous dehydration of these hydroxylamines, where water is considered the general acid catalyst, exhibit a large positive deviation from the Bronsted line. This fact was not observed in the reaction of non-hydroxyl amines with different aromatic aldehydes by other authors, thus supporting that the spontaneous dehydration steps for these reactions proceed by intramolecular catalysis. The mechanism of intramolecular catalysis might be stepwise. First, a zwitterionic intermediate is formed. It can then evolve in the second step by loss of water, or follow a concerted pathway, with the transference of a proton through a five-membered ring (general intramolecular acid catalysis). In the case of non-hydroxyl amines, data suggested the possibility of a mechanism of intramolecular proton transfer through one or two water molecules, from the nitrogen of the amine to the leaving hydroxide ion.

Full text of DOI:10.1002/kin.10095

Spontaneous Dehydration
Mechanism of Aromatic
Aldehyde Reactions with
Hydroxyl and Non-Hydroxyl
Amines
ˆ
SILVANA AZZOLINI TRAVALON, INES MARIA COSTA BRIGHENTE,
ROSENDO AUGUSTO YUNES
Departamento de Quı´mica, Universidade Federal de Santa Catarina, Campus Universita´rio, Trindade, 88040-900
Floriano´polis, SC, Brazil
Received 5 March 2002; accepted 29 July 2002
DOI 10.1002/kin.10095
ABSTRACT: The plot of rate constants vs. pH for the dehydration step of the reaction bet-
ween furfural and 5-nitrofurfural with hydroxylamine, N-methylhydroxylamine, and O-methyl-
hydroxylamine, shows two regions corresponding to the oxonium ion-catalyzed and sponta-
neous dehydration. The oxonium ion-catalyzed dehydration region of the reaction of furfural
with the above mentioned hydroxylamines exhibits general acid catalysis with excellent
Brønsted correlation (Brønsted coefficients: 0.76 (r = 0.986), 0.68 (r = 0.987), and 0.67 (r =
0.993) respectively). However, the rate constants of the spontaneous dehydration of these
hydroxylamines, where water is considered the general acid catalyst, exhibit a large positive
deviation from the Brønsted line. This fact was not observed in the reaction of non-hydroxyl
amines with different aromatic aldehydes by other authors, thus supporting that the spon-
taneous dehydration steps for these reactions proceed by intramolecular catalysis. The
mechanism of intramolecular catalysis might be stepwise. First, a zwitterionic intermediate is
formed. It can then evolve in the second step by loss of water, or follow a concerted pathway,
with the transference of a proton through a five-membered ring (general intramolecular acid
catalysis). In the case of non-hydroxyl amines, data suggested the possibility of a mechanism
of intramolecular proton transfer through one or two water molecules, from the nitrogen of
ꢀ
C
the amine to the leaving hydroxide ion.
685–692, 2002
2002 Wiley Periodicals, Inc. Int J Chem Kinet 34:
INTRODUCTION
Correspondence to: Dr. Rosendo Augusto Yunes; e-mail:
ryunes@qmc.ufsc.br.
Some reactions of amines with carbonyl compounds
occur with rate-determining dehydration of the addi-
tion intermediate under acidic, neutral, or basic condi-
tions, Eq. (1) [1–4].
Contract grant sponsor: CNPq, Brazil.
Contract grant sponsor: CAPES, Brazil.
Contract grant sponsor: FINEP, Brazil.
c
ꢀ
2002 Wiley Periodicals, Inc.

686
TRAVALON, BRIGHENTE, AND YUNES
(1)
The rate–pH profile of the dehydration reaction
might exhibit three regions: (1) a region catalyzed
by the oxonium ion at low pH, (2) a hydroxide ion-
catalyzed region at high pH, and (3) a region of spon-
taneous reaction around neutral pH.
The study of these reactions can provide insights
into the behavior of some dehydratases of the fam-
ily of mandelate racemase, which includes galactonate
dehydratase from Escherichia coli [5], glucarate dehy-
dratase from Pseudomonas putida [5], and information
about the importance of the ꢀ-protons of the amines in
the mechanism of ꢁ-elimination [6].
furfural and 5-nitrofurfural, respectively) was mon-
itored. The concentration of aldehyde did not ex-
ceed 1.0 × 10−4 M. An adequate excess of nitro-
gen base (between 1.0 × 10−1 and 1.0 × 10−3 M)
was employed to ensure pseudo-first-order kinetic
behavior. For the reaction between furfural and N-
cyclohexylhydroxylamine, an excess of aldehyde was
employed (2.0 × 10−2 M of aldehyde and 1.0 ×
10−3 M of amine). Rate constants were determined
using established procedures [7]. Reactions were fol-
lowed for three half-lives (t1/2) and first-order rate con-
stants (kobs) were calculated using a computer.
It has been suggested [3] that in the reaction be-
tween N-phenylhydroxylamine and furfural, the devi-
ation from the Brønsted line of the point of the rate con-
stant of water, as general acid catalyst, indicated that
the addition intermediate is catalyzed by intramolecu-
lar proton transfer in a five-membered ring, from the
hydroxyl group bonded to nitrogen.
This work was carried out in order to (i) characterize
the dehydration step of the addition intermediate of
the reaction of aromatic aldehydes with hydroxyl and
non-hydroxyl amines and (ii) observe the importance
of the acidity of the proton that is eliminated in the
dehydration step.
According to Eq. (1), as the rate-determining step
is the dehydration of the addition intermediate in equi-
librium with the reagents, the kinetic law is
rate = kdeh[I]
(2)
When excess of nitrogen base is used, considering the
equilibrium of the addition intermediate [I] formation
and the conservation equation for the total carbonyl
compound (CO)t
Kad = [I]/[CO]f[Nu]
(3)
where [CO]f is the concentration of the free carbonyl
compound:
EXPERIMENTAL
Material
[CO]t = [CO]f + [I]
obtaining [CO]f from (4) and substituting into (3):
[I] = Kad[CO]t[Nu]/1 + Kad[Nu]
then
(4)
(5)
(6)
Organic reagents employed were commercially avail-
able products either redistilled or recrystallized. In-
organic chemicals and carboxylic acids used as
buffers were reagent grade and used without further
purification.
rate = kdeh Kad[CO]t[Nu]/1 + Kad[Nu]
thus, the observed first-order rate constant is
Kinetic Measurements
Reactions of carbonyl compounds with nitrogen bases,
in water at 25.0◦C and ionic strength 1.0 M (KCl),
were followed on a Varian DMS 80 spectrophotometer
equipped with a thermostated cell holder. Till the end,
the formation of oxime (260 nm for the reaction be-
tween hydroxylamine with furfural and 5-nitrofurfural,
250 nm for O-methylhydroxylamine and furfural, and
348 and 330 nm for N-cyclohexylhydroxylamine with
5-nitrofurfural and furfural, respectively) and nitrone
(320 and 370 nm for N-methylhydroxylamine with
kobs = rate/[CO]t = kdeh Kad[Nu]/1 + Kad[Nu] (7)
The second rate constants were calculated by divid-
ing the first-order rate constant by the concentration of
the free nitrogen base.
k2obs = kdeh Kad = kobs/[Nu]f × fc
fc = 1/1 + Kad[Nu]f
(8)
(9)

MECHANISM OF AROMATIC ALDEHYDE REACTIONS WITH HYDROXYL AND NON-HYDROXYL AMINES
687
working at a pH corresponding to the pKaꢁ of the cat-
alyst (pKaꢁ = pKa previously determined [7,9] under
the experimental conditions of this study: water, ionic
strength 1.0 M (KCl), 25◦C). The pKaꢁ values are 1.70,
3.56, 4.65, 5.48, 6.55, 7.21, and7.99, respectively[7,9].
Statistical corrections in the Brønsted plot were applied
to the catalytic (kc) constants according to the method
proposed by Bell and Evans [10], where p and q are
the numbers of acidic and basic sites, respectively.
Corrections were made according to Laviron et al. [8]
for the reaction between 5-nitrofurfural and nitrogen
bases, considering the constant Kꢁ of hydration equilib-
rium of this carbonyl compound, according to Eq. (10),
where Kꢁ = 1.
k2obs = kobs(1 + Kꢁ)/[Nu]f × fc
(10)
When excess of carbonyl compound is used, a sim-
ilar equation can be deduced:
Equilibrium Constants
rate = kdeh Kad[Nu]f[CO]t/1 + Kad[CO]t
(11)
The equilibrium constants Kad for formation of the
addition intermediate from several amines with fur-
fural and 5-nitrofurfural in aqueous solution and ionic
strength 1.0 M (KCl) were spectrophotometrically de-
termined [11] mixing an aldehyde solution with sev-
eral concentrations of amines, of pH between 7.5 and
8.5 (0.01 M of phosphate buffer), measuring the ini-
tial decrease in the absorbance of the carbonyl group at
270 and 308 nm (for the reaction between amines with
furfural and 5-nitrofurfural, respectively) and extrapo-
lating to time equals to zero.
but as in all cases Kad [CO]t
simplified as
1 Eq. (11) can be
rate = kdeh Kad [Nu]f [CO]t
(12)
The rate constant should be expressed as a function
of the total concentration of nitrogen base. In addition,
considering its pKa, it is possible to obtain the follow-
ing relation:
The values obtained represent the average of 15 de-
terminations and were calculated from Eq. (18) [9]:
[Nu]f = [Nu]t Ka/Ka + [H+]
(13)
thus
1/ꢀA = 1/[C O]t + 1/Kad[C O]t × 1/[Nu]f (18)
rate = kdeh Kad[Nu]t[CO]t Ka/Ka + [H+]
(14)
where [C O]t is the total concentration of the carbonyl
compound and [Nu]f is the free-amine concentration.
The equilibrium constant was determined from the neg-
ative abscissa intercept of the double reciprocal plot of
the difference of absorbance as a function of the amine
concentration.
then, the first-order (kobs) and second order (k2obs) ob-
served rate constants are
kobs = rate/[Nu]t = k2 Kad[CO]t Ka/Ka + [H+]
(15)
Calculation of the pKa
of ꢀ-N-Hydroxy Group
k2obs = k2 Kad = rate/[CO]t × fc
fc = Ka/Ka + [H+]
(16)
(17)
The pKa for the ionization of the hydroxy group,
bonded to the nitrogen atom of the addition in-
termediate formed from the reaction of furfural
with hydroxylamine and N-methylhydroxylamine,
were calculated from the pKa of the addition in-
termediate for the reaction between furfural and
N-phenylhydroxylamine [3], considering the change
caused by the replacement of the phenyl group for a hy-
drogen atom or a methyl group. The acid-strengthening
effects of these species can be calculated from the
σ* values (0.75 for Ph [12] 0.49 for H [12], and
0.00 for CH3 [12]), using ꢀpKa = 0.06 + 0.63σ*
[12]. As these species are bonded to ꢀ-C, considering
that their behavior is normal in this position, the
values of ꢀpKa obtained from the equation above
should be divided by 0.4 [12]. Thus, the pKa values
+
Third-order rate constants (kaH ) and rate constants
for general acid catalysis (kc) were obtained from the
slopes of plots of second-order constants as a function
of the concentration of oxonium and catalytic species,
respectively. The rate constant of spontaneous reaction
(k0) was directly obtained from the plot of log k2obs vs.
pH. kH O was obtained from k0 (spontaneous dehydra-
2
tion rate constant) by dividing it by 55.5 M.
General acid catalysis for the reaction of furfural
with hydroxylamine, N-methylhydroxylamine, and N-
cyclohexylhydroxylamine were determined using the
following acids: phosphoric acid, formic acid, acetic
acid, succinatemonoanion, phosphatemonoanion, imi-
dazolium ion, and tris(2-hydroxyethyl)ammonium ion,

688
TRAVALON, BRIGHENTE, AND YUNES
calculated are 12.50 and 13.12 for the N OH of the
addition intermediate derived from hydroxylamine
and N-methylhydroxylamine, respectively.
The pH dependence of the second-order rate con-
stant (k2obs) extrapolated to zero buffer concentration
for the reactions of hydroxylamines with furfural and
5-nitrofurfural is shown in Fig. 1. Equation (20) is con-
sistent with the experimental results and
It is similarly possible to consider that the pKa of the
hydroxy group C OH as a base of the addition inter-
mediate of the reaction between furfural and hydroxy-
lamine must be similarto thatof asubstituted methanol.
It is then possible to calculate the effects of the sub-
stituents considering both the pKa of the conjugate acid
of methanol −2.05 [13], and the behavior of the basic-
ity of alcohols to be similar to that of amines. The
base-weakening effect of the NHOH group was cal-
culated using the equation −ꢀpKa = 0.28 + 0.83ꢂ*
[12]. As the σ* of the NHOH group is 0.30 [12],
ꢀpKa is −0.53. This value corresponds to the sub-
stituent attached to ꢁ-carbon, but in this case it is at-
tached to ꢀ-carbon, assuming its normal behavior in
this position, ꢀpKa = −0.54/0.4 = −1.32. In a simi-
lar way, the base-weakening effect of the furfuryl group
(C4H3O ) was calculated as −1.22 from σ∗ = 0.25
[12]. Thus, the pKa of the C OH was calculated
as −2.05 − 1.32 − 1.22 = −4.59. In the case of the
hydroxy group C OH of the addition intermediate
of the 5-nitrofurfural and hydroxylamine, considering
that the effect of the nitro group should be similar to
that of p-nitrophenyl with respect to the phenyl group
[14], (the σ∗ of the 5-NO2C4H2O can be calculated
as 0.76, and its ꢀpKa as −2.28) the pKa calculated
using the previous method is −5.65.
k2obs = Kadkdeh = KadkaH + [H+] + Kadk0 (20)
includes the rate constants for the oxonium ion-
+
catalyzed dehydration (kaH ) of the addition intermedi-
ate, previously formed, and the corresponding sponta-
neous dehydration (k0). Kad is the equilibrium constant
for formation of the addition intermediate. The lines in
Fig. 1 were calculated using Eq. (20).
The values of the equilibrium constant for the for-
mation of the neutral addition intermediate Kad from
amines and furfural, and 5-nitrofurfural are indicated in
Table I. The accuracy of these values are very important
in order to determine the true value of the dehydration
rate constant (kdeh = k2obs/Kad).
The values of the oxonium ion-catalyzed dehydra-
tion rate constants and of the spontaneous dehydration
rate constants for the reactions of hydroxylamine, N-
methylhydroxylamine, O-methylhydroxylamine, and
N-cyclohexylhydroxylamine with furfural and 5-
nitrofurfural are indicated in Table II.
Thereactionexhibitscatalysisbydifferentbuffersin
the pH region of oxonium ion-catalyzed dehydration.
The third-order rate constants, determined in the pres-
ence of acetic acid/acetate buffers, were plotted against
the molar fraction of the acid for several concentrations
of the buffer, and a least-square fit was made showing
that the acidic component is the only catalytic species.
The experiments to determine the catalytic constant
of different acids for the reaction of the amines with fur-
fural were carried out using six different acids, taking
an average of their catalytic constant from three deter-
minations for each one, to obtain the Brønsted plots
RESULTS AND DISCUSSION
The reaction of furfurals with nitrogen nucleophiles
follows a two-step process: attack of the nitrogen nu-
cleophile on the carbonyl group to form an addition
intermediate kad, and dehydration of this intermediate
to form the product kdeh (Eq. (19)).
(19)

MECHANISM OF AROMATIC ALDEHYDE REACTIONS WITH HYDROXYL AND NON-HYDROXYL AMINES
689
Table I Equilibrium Constants for the Formation of
Addition Intermediates from Amines and Carbonyl
Compoundsa,b
Carbonyl
Compoundc
Aminesd
pKaꢁ Amine Kad (M−1
)
Furfural
Furfural
Furfural
Furfural
NH2OH
6.17
6.15
4.73
6.00
4.63 0.23
CH3NHOH
NH2OCH3
C6H11NHOH
0.84 0.03
1.99 0.08
0.17 0.005
700 35
5-Nitrofurfural NH2OH
5-Nitrofurfural CH3NHOH
5-Nitrofurfural C6H11NHOH
107 5.3
17.89 0.71
a Water at 25◦C, ionic strength 1.0 M (KCl).
b Standard deviation of Kad was about 5%.
c [Carbonyl compound] = 4.0–8.0 × 10−5 M.
d [amine] = 0.25–0.03 M.
Figure 1 pH dependence of the logarithms of second-order
rate constants for oxime formation from hydroxylamine with
furfural ( ) and 5-nitrofurfural ( ) in water at 25◦C and ionic
Different authors have suggested that this reaction
is general acid catalyzed, including the oxonium ion
[3,9,15,16]. The values of the spontaneous reaction rate
constants (k0), as divided by 55.5 M [17–19], are sig-
nificantly higher than expected from the Brønsted line,
for the reactions of furfural with hydroxylamine, N-
methylhydroxylamine, and O-methylhydroxylamine.
If the reaction with water as catalyst followed the
general acid-catalyzed mechanism, the rate constant
should fit the Brønsted line [17] as it happens with the
rate constant of the oxinium ion.
ꢀ
ꢁ
strength 1.0 M.
with minimum error (Fig. 2). The Brønsted coefficients
obtained were 0.76 (r = 0.986), 0.68 (r = 0.987),
and 0.67 (r = 0.993) for the reactions between fur-
fural with hydroxylamine, N-methylhydroxylamine,
and O-methylhydroxylamine, respectively. Reimann
and Jencks [9] obtained α = 0.77 for the reactions of
p-chlorobenzaldehyde with N-methylhydroxylamine
and hydroxylamine.
Intramolecular Proton Transfer
in the Transition State
of the Spontaneous Dehydration
McClelland and Coe [18] pointed out that in the decom-
position of benzaldehyde hydrates the rate constant of
the water-catalyzed reaction is considerably above the
line of the Brønsted correlation with acids and very
near the line of the Brønsted correlation with oxyan-
ion bases. In some cases, as in the decomposition of
formaldehydehydratesandhemiacetals[19], thewater-
catalyzed rate constants were found to fit the Brønsted
line for carboxylate bases within a factor of 2. On this
basis, Funderburck et al. [19] suggested that in these
cases the water-catalyzed reaction is mechanistically
similar to the general base reaction.
The reaction of p-chlorobenzaldehyde with aceto-
hydrazide, semicarbazide, phenylhydrazine-p-sulfo-
nic acid, hydrazine, and thiosemicarbazide, studied by
Sayer et al. [20] shows that for water the points are
39, 24, 7, 15, and 100 times above the Brønsted line
of the acid catalysis. It is important to note that the
value corresponding to the oxonium ion, which was
not included in the correlation, fits near the Brønsted
Figure 2 Brønsted plot for general acid catalysis of the
dehydration step of the reaction between hydroxylamine
and furfural at 25◦C and ionic strength 1.0 M (KCl); α =
0.76, r = 0.986. H3O+ and H2O not considered.

690
TRAVALON, BRIGHENTE, AND YUNES
Table II Rate Constants of the Dehydration Step for Nitrone and Oxime Formation from Furfural and 5-Nitrofurfural
with Aminesa,b
kaH (10−3 M−2 s−1
)
k0 (10−3 M−1 s−1
)
+
Amines
Furfural
5-NO2-Furfural
3.16 0.09
Furfural
5-NO2-Furfural
1.12 0.03
NH2OH
56.2 1.40
3.9 0.10
56.2 1.69
31.6 0.10
3.16 0.08
0.20 0.01
5.01 0.15
10.00 0.02
NH2OCH3
CH3NHOH
C6H11NHOH
5.62 0.14
1.78 0.05
1.41 0.04
1.74 0.04
a Water at 25◦C, ionic strength 1.0 M (KCl).
b Standard deviation of kaH and k0 was about 3%.
+
line, which gives confidence to the value of the angular
coefficient.
stable electronic interactions or some kind of special
catalysis.
The values of the water catalytic rate constant for the
reaction of p-chlorobenzaldehyde with hydroxylamine
and N-methylhydroxylamine obtained by Reimann
and Jencks [9], are similar to those determined in
this work. Those values gave a positive deviation of
38,000 and 30,000 fold respectively with respect to the
Brønsted line for those reactions.
The electronic interaction in the TS, with differ-
ent degrees of stabilization, has been suggested [21]
to be the cause of the different Brønsted relations ob-
served for different acids types [22]. The same factor
has been suggested to explain the enhanced reactivity
of phosphonate for the acid-catalyzed reaction of the
hydration-dehydration of benzaldehydes [18].
In the dehydration reaction of the addition inter-
mediate formed from furfural and N-phenylhydro-
xylamine, the spontaneous dehydration rate constant
considered as water-catalyzed reaction is approxi-
mately 400 times larger than that predicted by the
Brønsted line [3].
In the present work it was observed that the sponta-
neous dehydration rate constants for the reaction of fur-
fural with hydroxylamine, N-methylhydroxylamine,
and O-methylhydroxylamine are 7500, 2000, and 100
times, respectively above the corresponding point on
the Brønsted line.
Taking this data into consideration, three important
facts should be noted: (i) There is a clear behavioral
difference between hydroxyl and non-hydroxyl
amines. Thus, the water-catalyzed rate constant for the
reactions of hydroxyl amines with furfural and with
p-chlorobenzaldehyde exhibits a notably positive de-
viation from the Brønsted line, 3–4 log units, while the
non-hydroxyl amines exhibit a maximum deviation of
2 log units. (ii) This difference does not depend on the
pKa of the corresponding nitrogen bases. (iii) These
significant deviations of the spontaneous dehydration
rate constant from the Brønsted line for general acid
catalysis, suggest that the spontaneous dehydration oc-
curs through different transition states.
However, in the case of the reactions of hy-
droxyl and non-hydroxyl amines with furfural and p-
chlorobenzaldehyde, several pieces of evidence sug-
gest that the spontaneous dehydration of the addition
intermediate follows a mechanism of intramolecu-
lar catalysis by proton transfer from the hydroxy
group bonded to the nitrogen to the leaving hydroxyl
group. They can be summarized as follows: (i) The
spontaneous rate constants (k0) of hydroxyl amines
that are more than 10 times higher than the non-
hydroxyl ones; (ii) The high deviation of the consid-
ered water-catalyzed rate constants of hydroxyl amines
from the Brønsted line for acids (the electronic inter-
action could explain only small deviations, between
2 and 10 times, in reactivity) and the lack of this
high deviation with non-hydroxyl amines such as O-
methylhydroxylamine, hydrazine, etc.; (iii) The pKa
values of the hydroxy groups bonded to the nitrogen
atomoftheadditionintermediatesformedfromfurfural
with hydroxylamine and N-methylhydroxylamine
were calculated as 12.50 and 13.12, respectively. Con-
sequently, these hydroxy groups are stronger acids than
water (pKa = 15.7) [23] and for this reason are able to
act as general acid catalyst.
The mechanism of the intramolecular general acid
catalysis might take place in a stepwise fashion. First,
a zwitterionic intermediate (Structure 1) is formed and
can evolve, in the second step, by loss of water fol-
lowed by N-deprotonation or could directly follow by
a, concerted pathway, the transition state indicated in
Structure 2. These mechanisms are consistent with the
Following this line of reasoning, it is clear that
there exists some effect that makes the transition state
(TS) of the spontaneous reaction more stable than
the hypothetical mechanism of general acid catalysis
by water. This effect can be a consequence of more

MECHANISM OF AROMATIC ALDEHYDE REACTIONS WITH HYDROXYL AND NON-HYDROXYL AMINES
691
In the dehydration reaction of the tetrahedral ad-
dition intermediate of p-chlorobenzaldehyde with hy-
drazines [20], it is important to note that acids as weak
as the protonated quinuclidinone (pKa = 11.55) ex-
hibit general acid catalysis in the dehydration process.
This is not observed in the case of the reaction of fur-
fural with O-alkylhydroxylamine. This fact associated
to the weak positive deviation of the Brønsted line, al-
ready indicated for the water catalysis, lead to the sug-
gestion that in this last reaction the dehydration cannot
be formulated merely as an spontaneous expulsion of
the hydroxide ion from the neutral addition intermedi-
ate. A viable possibility accounts for all these facts and
should consider that there exists intramolecular catal-
ysis by proton transfer from the nitrogen to the leaving
hydroxide ion through one or two water molecules in
a six- or eight-membered ring (Structure 3).
Structure 1
lower destabilizing effect of the nitro group in the spon-
taneous dehydration, because the central carbon atom
of the reaction is not charged.
The TS of the spontaneous dehydration of the addi-
tion intermediates of the non-hydroxyl amines should
be analyzed at this point. According to Sayer et al.
[20], the relationship between the rate constant for
oxonium ion-catalyzed and spontaneous dehydration
and the basicity of the hydrazines, previously indicated,
suggest that there is electron donation from the nitro-
gen atom to the carbon atom in the transition state
of these reactions. However, the correlations of the
plot of the logarithms of the catalytic constants ver-
sus the pKa of the hydrazines studied by Sayer et al.
show that the Brønsted coefficients are β = 0.34 (r =
0.895) and β = 0.31 (r = 0.624) for the oxonium ion-
catalyzed and spontaneous reaction, respectively. In
the case of the oxonium ion-catalyzed dehydration,
the coefficient of correlation is not completely satis-
factory. It can, however, be accepted that this correla-
tion indicates that the contribution of the basicity of
hydrazines is important in the transition state of the
oxonium ion-catalyzed dehydration. This is in accor-
dance with the requirement of stabilization of the pos-
itive charge developed on the central carbon atom, but
the unsatisfactory value of the correlation coefficient of
the spontaneous process indicates that, even taking into
account the contribution of the basicity of the amines
in the transition state, there must exist another predom-
inant factor. In the case of the reaction of furfural with
amines, it is evident that the basicity is not a decisive
factor, since hydroxylamine, N-methylhydroxylamine,
and N-cyclohexylhydroxylamine have similar basicity
but very different values of the rate constant in the ox-
onium ion-catalyzed and spontaneous dehydration.
The Importance of the pKa
of the Leaving Proton in ꢀ-Position
A comparison between the rate of spontaneous dehy-
dration of the addition intermediate of furfural with
hydroxylamine and that of O-methylhydroxylamine
shows that the oxonium ion-catalyzed dehydration
rate constant, the same as the spontaneous dehydra-
tion rate constant, is approximately 15 times faster
in the first case. However, in the intermediate of O-
methylhydroxylamine, the N H bond should be more
acidic than that of hydroxylamine, considering their
pKa values (6 for hydroxylammonium and 4.75 for O-
methylhydroxylammonium) and also calculating the
pKa of the N H of the intermediates as acid, using
the equation [12] ꢀpKa = 0.06 + 0.63σ∗. Thus, it can
be stated that the pKa of the ꢀ-proton is not significant
in the kinetics of these reactions.
In the cases of N-methylhydroxylamine and N-
cyclohexylhydroxylamine, the mechanism is not of ꢁ-
elimination. Even then, it is possible to observe that the
addition intermediate of N-cyclohexylhydroxylamine
Structure 2
Structure 3

692
TRAVALON, BRIGHENTE, AND YUNES
exhibits a slower acid-catalyzed dehydration rate con-
stant and a faster spontaneous dehydration rate constant
than that corresponding to the addition intermediate of
the N-methylhydroxylamine. In this case, the steric ef-
fects should be as important as in enzymatic catalysis.
10. Bell, R. P.; Evans, P. G. Proc Roy Soc (London), Ser A
1966, 291, 297.
11. Jencks, W. P. J Am Chem Soc 1959, 81, 475.
12. Perrin, D. D.; Dempsey, B.; Serjeant, E. P. pKa Predic-
tion for Organic Acids and Bases; Chapman & Hall:
London, 1981.
13. Perdocin, G.; Scorrano, G. S. J Am Chem Soc 1977, 99,
6983.
14. Exner, O. In Advances in Linear Free Energy Relation-
ships Chapman, N. B.; Shorter, J. (Eds.); Plenum: New
York, 1972; p. 43.
15. Jencks, W. P. Catalysis in Chemistry and Enzymology;
McGraw-Hill: New York, 1969.
16. Palmer, J. L.; Jencks, W. P. J Am Chem Soc 1980, 102,
6466.
BIBLIOGRAPHY
1. Calzadilla, M.; Malpica, A.; Diaz, P. M. Int J Chem Kinet
1996, 28, 687.
2. Malpica, A.; Calzadilla, M.; Baumrucker, J.; Jimenez,
J.; Lopez, L.; Escobar, G.; Montes, C. J Org Chem 1994,
59, 3398.
3. Fett, R.; Simionato, E. L.; Yunes, R. A. J Phys Org Chem
1990, 3, 620.
17. Dal Magro, J.; Yunes, R. A. Bull Chem Soc Jpn 1998,
71, 2381.
4. Brighente, I. M. C.; Budal, R. M.; Yunes, R. A. J Chem
Soc, Perkin Trans 2 1991, 11, 1861.
18. McClelland, R. A.; Coe, M. J Am Chem Soc 1983, 105,
2718.
5. Palmer, D. R. J.; Wieczorek, S. J.; Hubbard, B. K.;
Mrachko, G. T.; Gerlt, J. A. J Chem Am Soc 1997, 119,
9580.
6. Hanson, K. R.; Rose, I. A. Acc Chem Res 1975, 8, 1.
7. Rosemberg, S.; Silver, S. M.; Sayer, J. M.; Jencks, W. P.
J Am Chem Soc 1974, 96, 7986.
19. Funderburk, I. H.; Aldwin, L.; Jencks, W. P. J Am Chem
Soc 1978, 100, 5444.
20. Sayer, J. M.; Peskin, M.; Jencks, W. P. J Am Chem Soc
1973, 95, 4277.
21. Kresge, A. J.; Chiang, Y. J. J Am Chem Soc 1973, 95,
803.
8. Laviron, E.; Troncin, H.; Tirouflet, J. Bull Soc Chem Fr
1962, 524.
22. Harron, J.; McClelland, R. A.; Thankadran, C.; Tidwell,
T. T. J Org Chem 1981, 46, 903.
9. Reimann, J. E.; Jencks, W. P. J Am Chem Soc 1966, 88,
3973.
23. Hammett, L. P. Physical Organic Chemistry; McGraw-
Hill: New York, 1970; p. 23.