15N-CIDNP during Photonitration of Phenol with Tetranitromethane and Reactions with Nitric and Nitrous Acid

Article abstract of DOI:10.1002/(sici)1099-0690(199805)1998:5<913::aid-ejoc913>3.0.co;2-r

During the irradiation of phenol (1) with ¹⁵N-enriched tetranitromethane in acetonitrile, and during the reaction of 1 with ¹⁵N-enriched nitric acid and nitrous acid, emission due to the nitration products o- and p-nitrophenol (2a, 2b) is observed in the ¹⁵N-NMR spectra. The CIDNP effects are built up by radical pairs formed by the encounters of the radicals NO2^? and 1^+? or PhO^?. During the reaction of 1 with nitrous acid, 2b is formed, in part due to a non-radical reaction, via oxidation of p-nitrosophenol (3).

Full text of DOI:10.1002/(sici)1099-0690(199805)1998:5<913::aid-ejoc913>3.0.co;2-r

FULL PAPER
15N-CIDNP during Photonitration of Phenol with Tetranitromethane and
Reactions with Nitric and Nitrous Acid
Manfred Lehnig* and Klaus Schürmann
Fachbereich Chemie der Universität Dortmund,
Otto-Hahn-Straße 6, D-44221 Dortmund, Germany
Received September 29, 1997
Keywords: Radical reactions / Electron transfer / Nitration / Radical ions / Nitric oxides
During the irradiation of phenol (1) with 15N-enriched tetra-
nitromethane in acetonitrile, and during the reaction of 1 NO2 and 1+• or PhO•. During the reaction of 1 with nitrous
with 15N-enriched nitric acid and nitrous acid, emission due
acid, 2b is formed, in part due to a non-radical reaction, via
up by radical pairs formed by the encounters of the radicals
•
to the nitration products o- and p-nitrophenol (2a, 2b) is ob- oxidation of p-nitrosophenol (3).
served in the 15N-NMR spectra. The CIDNP effects are built
The nitration of aromatic compounds plays an important systems, catalysis by nitrous acid (NAC) plays an important
role in organic chemistry[1]. The mechanism for the ni- role[3][9]. Ridd and coworkers suggest that NAC involves
tration reaction depends on the nature of the nitrating sys- an electron transfer between nitrosonium cations NOϩ and
tem, the actual reaction conditions, and in many cases is arenes (Eq. 4a)[10]. NOϩ is regenerated by the reaction
•
still under discussion. In various nitration systems, ni- given in Eq. 4b to give NO2 .
ϩ
tronium ions NO2 are involved as reaction intermediates.
The discussion then focusses on the question as to whether
NAC: ArH ϩ NOϩ Ǟ ArHϩ• ϩ NO•
(4a)
(4b)
ϩ
•
(NO• ϩ NO2 Ǟ NOϩ ϩ NO2 )
the cationic intermediate (σ complex) is formed by addition
ϩ
of the nitronium ions NO2 to the arenes (Eq. 1a), or by
Eq. 4b is enclosed in parentheses as it shows only the
stoichiometry of the reaction. It does not involve free ni-
tronium cations NO2ϩ but a more complex and, as yet, un-
identified route. Under these conditions, the rate of forma-
•
recombination of the radicals NO2 with arene radical cat-
ions (Eq. 1b)[2][3]
.
ϩ
ArH ϩ NO2 Ǟ ArHϩϪNO2 Ǟ ArϪNO2 ϩ Hϩ
(1a)
(1b)
ϩ
•
ArHϩ• ϩ NO2 Ǟ ArHϩϪNO2 Ǟ ArϪNO2 ϩ Hϩ
tion of NO2 is too small to allow the occurrence of the
1
(σ complex)
2
reaction given in Eq. 1a. The formation of nitration prod-
ucts via the reaction given in Eq. 1b has been shown by
15N-CIDNP to occur during the nitration of p-substituted
phenols, polymethylbenzenes, arylamines and naphtha-
1: Phenol; 2a: o-Nitrophenol; 2b: p-Nitrophenol
The radical cations ArHϩ• might be formed via electron
ϩ
transfer (ET) between NO2 and, if they are more reactive
lenes[5]
.
than toluene, arenes[4]
.
If two different nitration products are formed during the
nitration of phenolic compounds with nitric acid, their ratio
depends on the concentration of sodium nitrite added to
the reaction mixture. This indicates the presence of a second
reaction catalysed by nitrous acid, which is a nitrosation
ϩ
ET: ArH ϩ NO2 Ǟ ArHϩ• ϩ NO2•
(2)
It has been shown by both CIDNP investigations and
theoretical considerations that ET plays, if any, only a
minor role[5][6]. As the addition of NO2ϩ to activated arenes
(Eq. 1a) is a diffusion-controlled reaction, the reactions
given in Eqs. 1b and 2 have different rates, whilst the reac-
tion shown in Eq. 2 has a high activation energy due to the
requirement that the linear nitronium ions NO2ϩ transform
followed by oxidation (NOX)[11]
.
NOX: ArH ϩ HNO2 Ǟ ArNO ϩ H2O
ArNO ϩ HNO3 Ǟ ArNO2 ϩ HNO2
(5a)
(5b)
•
Aromatic C-nitrosation (Eq. 5a) is often described as an
electrophilic substitution reaction with nitrosonium ions
NOϩ as the electrophile (Eq. 6a)[11]. It has been suggested
that the reaction might have a radical mechanism (Eq. 6b),
analogous to the nitration reaction (Eq. 1b).
into the bent structure of the NO2 radicals during the elec-
tron transfer process. Radical reactions only occur if the
radicals are generated in a different way. This is the case
during irradiation of arenes with tetranitromethane; ET
then takes place in excited charge-transfer complexes[7][8]
.
•
[C(NO2)4, ArH ]* Ǟ ArHϩ• ϩ NO2
ϩ
ϪC(NO2)3
(3) ArH ϩ NOϩ Ǟ ArHϩϪNO
ArHϩ• ϩ NO• Ǟ ArHϩϪNO
Ǟ ArϪNO ϩ Hϩ
Ǟ ArϪNO ϩ Hϩ
3
(6a)
(6b)
•
Arϩ• and NO2 might also be formed during nitration of
reactive aromatic systems with dilute nitric acid. In such 3: p-Nitrosophenol.
Eur. J. Org. Chem. 1998, 913Ϫ918
WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
1434Ϫ193X/98/0505Ϫ0913 $ 17.50ϩ.50/0
913

M. Lehnig, K. Schürmann
FULL PAPER
The purpose of this work is to investigate the reaction been measured after completion of the reaction are 180 s
mechanism of the nitration reaction of 1 using different ni- and 120 s. For a quantitative analysis the reaction is too
tration agents. First, 15N-CIDNP results obtained during fast. In further experiments lower concentrations of the
the photonitration of 1 with 15N-labelled tetranitromethane starting compounds have been used in order to retard the
will be presented. The appearance of free radicals during reaction rate (see Table 1). The time dependence of the
the photoreaction of tetranitromethane with activated ar- emission signals during the reaction of 1 (0.2 ) with 15N-
enes has already been shown using time-resolved spec- enriched nitric acid (0.5 ) is shown in Table 2.
troscopy and ESR[7][12], and has been postulated to occur
The nature of the nitration products, the yields and the
during the photoreaction with 1[13] The second reaction is 15N-CIDNP effects observed, are the same in both systems,
the nitration of 1 with dilute nitric acid, which will be stud- indicating an identical formation path for the products. It
ied by 15N-CIDNP in water and different organic solvents is concluded from the 15N-CIDNP effects that they are built
under the conditions customary in synthetic work. The rad- up in radical pairs which are formed by encounters of the
•
ical character of the reaction (NAC, Eqs. 4) has been postu- independently generated radicals 1ϩ• and 15NO2 (ЉFЉ- case,
lated using kinetic arguments[14][15] but has not been proven Eq. 7).
by spectroscopic techniques. There is no doubt, however,
•
• F
1ϩ•
ϩ
15NO2 Ǟ [1ϩ•
,
15NO2 ] Ǟ 2a,b ϩ Hϩ
(7)
that an electrophilic substitution reaction takes place dur-
ing the nitration of 1 with a mixture of concentrated nitric
and sulfuric acid (Eq. 1a) in the presence of effective scaven-
The radicals are formed by the reactions shown in Eqs. 3
gers for nitrous acid[16]. During the reaction of 1 with nitric and 4. Similar CIDNP effects have been observed during
acid in the presence of sodium nitrite, 2b is formed in high the nitration of a number of activated aromatic compounds
yield, which has been explained by NOX (Eqs. 5a,b)[14]. To with nitric acid, and these have been explained in an anal-
provide evidence for this 15N-CIDNP studies have also been ogous fashion[5][18]. Eq. 7 describes only one possibility for
performed during the reaction of 1 with sodium nitrite.
product formation during the photoreaction of tetranitro-
The magnitude of the 15N-CIDNP effects, taking into ac- methane with phenolic compounds. The photo-15N-CIDNP
count enhancement factors which are determined exper- effects might therefore be generated by, at least in part, a
imentally, will be discussed and compared with calculations different radical reaction. They might be built up in gemi-
basing on PedersenЈs formulation of the radical-pair the- nate pairs formed by the decomposition of excited charge-
ory[17]. This procedure, which allows the importance of the transfer complexes from triplet states (ЉTЉ case). Free rad-
radical reactions to be determined, has already been applied icals might also be formed during the decomposition of un-
with success to the nitration of 1,2-dimethoxybenzene, ani- stable intermediates such as the nitro-trinitromethyl adducts
sole and mesitylene[18]
.
of 1 or trinitromethyl-substituted phenols, which are
formed as primary products during the irradiation of tetra-
nitromethane with phenolic derivatives[6]. Under our con-
Results and Discussion
Photonitration of 1 with Tetranitromethane and Nitration ditions, however, we did not detect any 15N-NMR or 15N-
with Nitric Acid in Acetonitrile: 15N-NMR spectra taken CIDNP signals which might be due to such compounds.
before, during and after irradiation of 1 in 15N-labelled
The importance of non-radical nitration reactions will be
tetranitromethane and acetonitrile are given in Figures discussed by comparing experimentally determined en-
1aϪc. Irradiation was carried out with the visible light (λ ϭ hancement factors (E) with calculated ones (Ecalc), since
488, 514 nm) of an argon ion laser in the modified probe only the radical reactions lead to 15N-CIDNP. E values,
head of a 15N-NMR spectrometer. The 15N-NMR signals which have been determined from the data given in the
observed after irradiation are due to HC(15NO2)3 and the Tables 1 and 2 by applying Eqs. 9 and 10, Ecalc values with
nitration products 2a and 2b (see Eqs. 1b, 3), which are g(1ϩ•) ϭ 2.00291[19], and the parameters used for calculat-
formed with a ratio of 1.1. This is similar to the value of 1.3 ing enhancement factors of 15N-CIDNP effects in the ni-
found by Seltzer and coworkers for cyclohexane[13]. During tration products of 1,2-dimethoxybenzene and anisole[18]
irradiation the 15N-NMR signals due to 2a and 2b show are listed in Table 3. The radical cation 1ϩ• might exist in its
emission, and the signals of HC(15NO2)3 and C(15NO2)4 do unprotonated form PhO•, and so further calculations have
not show CIDNP. The assignment of the signals and therefore been performed using g(PhO•) ϭ 2.00461[19]. E
further details are given in Table 1.
values lie between Ϫ500 and Ϫ1000 and Ecalc values be-
After treatment of 1 (0.5 ) with 15N-labelled nitric acid tween Ϫ1000 and Ϫ1100. Within the limits of the exper-
(1.25 ) in acetonitrile at room temp., 2a and 2b are again imental and theoretical accuracy the values are comparable,
formed with a ratio of 1.1. If the reaction is carried out in showing that there is no reason to assume that non-radical
the probe head of the 15N-NMR spectrometer, the signals reactions are of major importance during the photochem-
show emission for about 1/2 h (see Figure 2a). After 2 h, ical nitration of 1 with tetranitromethane, or during the ni-
the reaction is completed, and the intensities of the 15N- tration of 1 with nitric acid. It cannot be determined from
NMR signals of 2a and 2b correspond to the yields of the the 15N-CIDNP experiments whether radical cations 1ϩ• or
products (see Figure 2b); the assignments and the intensities phenoxyl radicals PhO• are involved in the nitration of 1 .
of the signals are given in Table 1. The longitudinal relax-
Reactions of 1 with Nitric Acid in Water and Acetic Acid:
ation times T1 of the 15N nuclei in 2a and 2b which have The nitration of 1 with diluted nitric acid was first de-
914
Eur. J. Org. Chem. 1998, 913Ϫ918

15N-CIDNP during Photonitration of Phenol with Tetranitromethane and Reactions
Table 1. 15N-CIDNP during nitration of 1 with 15N-enriched systems (ACN: acetonitrile)
FULL PAPER
[b]
[c]
System
Solvent
15N Signals[a]
Imax
Imax
Io
tE
C(NO2)4 (0.18 )
ACN/[D3]ACN
Ϫ38.6 [C(15NO2)4]
Ϫ24.8 [HC(15NO2)3]
Ϫ0.2 (2b)
1.8 (2a)
(Figure 1)[d]
ϭ 50:1
[e]
[e]
E
E
Ϫ19.0
Ϫ27.7
13.5
14.5
76
7
HNO3 (1.25 )
ACN/H2O ϩ D2O
ϭ 9:1
Ϫ5.6 (H15NO3)
0.8 (2b)
(Figure 2)[f]
E
E
E
E
E
E
Ϫ1200
6
6
13.5
15.0
4.5
5.0
Ϫ
30
2.3 (2a)
Ϫ1800
Ϫ330
Ϫ390
Ϫ35
30
HNO3 (0.5 )
(Table 2)[g]
ACN/H2OϩD2O
ϭ 9:1
H2O/D2O
ϭ 20:1
0.8 (2b)
25
120
2.3 (2a)
25
120
120
120
HNO3 (1.7 )
0.7 (2b)
240[h]
240[h]
(Figure 3a)[f]
1.8 (2a)
Ϫ50
Ϫ
4.3 (H15NO3)
Ϫ0.6 (2b)
0.9 (2a)
HNO3 (1.25 )[f]
HNO3 (0.5 )[g]
AcOH/D2O
ϭ 9:1
AcOH/D2O
ϭ 9:1
H2OϩD2O/H2SO4
ϭ 9:1
A
5
6
6
5
3
[i]
[i]
A
3
Ϫ0.6 (2b)
0.9 (2a)
0.8 (2b)
E
Ϫ677
Ϫ1000
Ϫ17
Ϫ25
22
2
2
3
3
3
3
3
3
15
12
Ϫ
5
5
E
NaNO2 (0.3 )
E
6[j]
6[j]
10[j]
90[l]
90[l]
40[l]
(Figure 3b)[g]
1.7 (2a)
E
Ϫ
Ϫ
1.2
0.7
5.0
Ϫ
3.2 (15NO3
Ϫ0.6 (2b)
0.9 (2a)
)
N[k]
E
NaNO2 (0.3 )
AcOH/D2O
ϭ 9:1
Ϫ10
(Figure 4, Table 2)[g]
E
Ϫ15
Ϫ
3.2 (15NO3
60 (3)
)
N[k]
N
20
[a]
[b]
δ values against Ph15NO2, positive δ values downfield; E: emission, A: enhanced absorption, N: no 15N-CIDNP. Ϫ Imax: maximal
intensity of E or A (signal-to-noise ratio). tmax: time of Imax after addition of 15N-enriched HNO3 or NaNO2 to the solutions (min). Ϫ
[c]
[d]
Io: signal intensity after reaction (signal-to-noise ratio), tE: duration of the 15N-CIDNP effects (min). Ϫ [PhOH] ϭ 0.64 ; T ϭ 298 K.
Irradiation with the unfiltered light of a LEXEL 95 argon ion laser within the probe head of the 15N-NMR spectrometer, 30 pulses with
[e]
[f]
13°, delay time 3 s, after irradiation 501 pulses. Ϫ
15N-CIDNP intensities are constant during the irradiation. Ϫ
T ϭ 300 K,
single 90° pulses, [PhOH] ϭ 0.5 . Ϫ T ϭ 300 K, single 90° pulses, [PhOH] ϭ 0.2 . Ϫ Besides H15NO3, 15N-NMR signals could not
[g]
[h]
be detected after reaction (2a and 2b are not soluble in diluted HNO3). Ϫ [i] The reaction was completed before taking the first spectrum.
[k]
Ϫ
[j] 15N-NMR signals of 2a and 2b could not be detected after reaction (2a and 2b are not soluble in diluted H2SO4). Ϫ
The 15N-
NMR signal might exhibit a small A-type CIDNP effect. Ϫ [l] Io values taken from spectra taken after reaction, 64 pulses with 90°, delay
time 10 min.
Figure 1. 15N-NMR spectra of 1 and 15N-enriched C(NO2)4 in effects observed in acetonitrile (Figure 2a) shows that the
ACN/[D3]ACN (ACN: acetonitrile), (a) before (30 scans), (b) dur-
signals are due to 2a and 2b. As 2a and 2b are not soluble
ing (30 scans) and (c) after irradiation (501 scans) with the unfil-
in diluted nitric acid, 15N-NMR signals could not be ob-
tered light of an argon ion laser (pulse angle 13°, delay time 3 s)
served after the reaction, and a quantitative analysis of the
15N-CIDNP effects was not possible. Nevertheless, the rad-
ical character of the nitration reaction which was first pos-
tulated by Ross and coworkers[14] is evident even in water.
The formation of 2a and 2b is also described with acetic
acid as the solvent[21], and 15N-CIDNP effects have also
been observed. During the reaction of 15N-enriched nitric
acid (1.25 ) with 1 (0.5 ) , the reaction is completed be-
fore taking the first spectrum, and the signals of 2a and 2b
show enhanced absorption. Using a solution of 15N-en-
riched nitric acid (0.5 ) and of 1 (0.2 ), the reaction is
completed within 5 min, and the 15N-NMR signals of 2a
and 2b appear in emission, as was the case with acetonitrile
and water. The results are compiled in Table 1. The reac-
tions are too fast to allow a quantitative analysis. 15N-
NMR signals of 2a and 2b can be detected after the reac-
tion which allows the determination of the T1 values shown
in Table 3.
scribed by A. W. Hofmann in 1850[20]. If the reaction of 1
The emission, as well as the enhanced absorption, ob-
(0.5 ) with 15N-enriched nitric acid (1.7 ) is performed served in 2a and 2b can be explained as above (Eq. 7). The
within the probe head of the 15N-NMR spectrometer two reaction of independently generated radicals 1ϩ• and
•
emission signals are observed (see Figure 3a), which disap- 15NO2 only leads to emission if the reaction is run in mag-
•
pear after about 4 h. The comparison with the 15N-CIDNP netic fields which are large compared with aN(NO2 ). At
Eur. J. Org. Chem. 1998, 913Ϫ918
915

M. Lehnig, K. Schürmann
FULL PAPER
Ϫ
Table 2. 15N-NMR intensities I[a] of 2a,b and 15NO3 (a) during reaction of 1 (0.2 ) with 15N-enriched HNO3 (0.5 ) in ACN at 300
K, (b) during reaction of 1 (0.2 ) with 15N-enriched NaNO2 (0.3 ) in AcOH at 300 K
(a)
t[b]
5
Ϫ25
Ϫ24
15
Ϫ280
Ϫ190
25
Ϫ390
Ϫ330
35
Ϫ157
Ϫ127
45
Ϫ125
Ϫ100
55
Ϫ63
Ϫ45
65
Ϫ33
Ϫ25
75
Ϫ24
Ϫ19
85
Ϫ16
Ϫ11
95
Ϫ10
Ϫ6
105
Ϫ4
Ϫ2
115
Ϫ2
Ϫ1
125
Ϫ
Ϫ
600
5.0
4.5
I(2a)
I(2b)
(b)
t[b]
3
Ϫ15
Ϫ10
20
6
Ϫ58
Ϫ27
52
9
Ϫ42
Ϫ22
27
12
Ϫ24
Ϫ17
12
15
Ϫ24
Ϫ14
11
18
Ϫ18
Ϫ11
9
24
Ϫ17
Ϫ10
6
36
Ϫ12
Ϫ7
6
48
Ϫ12
Ϫ5
5
60
Ϫ6
Ϫ4
5
72
Ϫ5
Ϫ3
5
84
Ϫ2
Ϫ1
5
96
Ϫ
Ϫ
5
300
I(2a)
I(2b)
I(NO3
0.7[c]
1.2[c]
5
Ϫ
)
[a]
[b]
Determined from the signal-to-noise ratio using single 90° pulses. Ϫ t: time after mixing the reactants and putting the tube into the
[c]
probe head of the spectrometer (min). Ϫ Values from spectra taken after the reaction, 64 pulses with 90°, delay time 10 min.
low magnetic fields the recombination of free radicals
Figure 2. 15N-NMR spectra during reaction of 1 with 15N-enriched
HNO3 in ACN at 300 K (a) 6 min, (b) 30 min after mixing the should give enhanced absorption, as is observed[18]
.
reactants (single 90° pulses)
Figure 3. 15N-NMR spectra during reaction of 1 (a) with 15N-en-
riched HNO3 in H2O/D2O 120 min after mixing the reactants, (b)
with 15N-enriched NaNO2 in diluted H2SO4 3 min after mixing the
reactants (single 90° pulses)
Reactions of 1 with Nitrous Acid in Water and Acetic
Acid: The nitrosation of 1 was first studied by A. Baeyer
and H. Caro in 1874[23]. By treating an aqueous solution of
1 and NaNO2 or KNO2 with sulphuric acid or acetic acid,
p-nitrosophenol (3) is formed as the main product[22][23]
(Eq. 5a). During the reaction of 1 with 15N-enriched
NaNO2 in diluted sulfuric acid and acetic acid in the probe
head of the NMR spectrometer, two emission signals and
one absorption signal are observed (see Figures 3b, 4a and
Table 1). After the reaction in dilute sulfuric acid, 15N-
NMR signals of the reaction products could not be ob-
served because of their low solubility in water. This was not
the case with acetic acid as the solvent; Figure 4b shows a
15N-NMR spectrum of the products taken in acetic acid
120 min after mixing the reactants. It follows that the emis-
sion signals are due to 2a and 2b which are formed as side
products of the reaction. The effects are identical with those
Table 3. Experimental enhancement factors E of 15N-CIDNP ef-
[a]
fects in 2a and 2b and calculated ones Ecalc
.
Compound,
solvent
ESR[19] and
NMR parameters
Enhancement
factors
2a in ACN
2b in ACN
2a,b in ACN
T1 ϭ 180 s
T1 ϭ 120 s
E ϭ Ϫ500[b]
E ϭ Ϫ750[c]
E ϭ Ϫ550[b]
E ϭ Ϫ980[c]
Ecalc ϭ Ϫ1025
Ecalc ϭ Ϫ1092
E ϭ Ϫ1180[d]
E ϭ Ϫ650[d]
Ecalc ϭ Ϫ1260
Ecalc ϭ Ϫ1256
g(1ϩ•) ϭ 2.00291
g(PhO•) ϭ 2.00461
T1 ϭ 90 s
2a in AcOH
2b in AcOH
2a,b in AcOH
T1 ϭ 54 s
g(1ϩ•) ϭ 2.00291
g(PhO•) ϭ 2.00461
[a]
Calculated with PedersenЈs treatment of the radical-pair the-
ory[17][18]. Ϫ
During irradiation of 15N-enriched C(NO2)4 with
[b]
1. Ϫ [c] During reaction of 15N-enriched HNO3 with 1. Ϫ [d] During
reaction of 15N-enriched NaNO2 with 1.
916
Eur. J. Org. Chem. 1998, 913Ϫ918

15N-CIDNP during Photonitration of Phenol with Tetranitromethane and Reactions
FULL PAPER
observed during the reaction of 1 with nitric acid and ex- that NOϩ is less reactive than NO2ϩ by a factor of 1014 [28]
,
•
Ϫ
plained in the same manner. NO2 and NO3 are formed which makes an electrophilic substitution (Eq. 6a) unlikely
by disproportionation[24]
.
if there is any other reaction possibility.
Figure 4. 15N-NMR spectra (a) during reaction of 1 with 15N-enri-
ched NaNO2 in AcOH with a single 90° puls (b) 120 min after
completion of the reaction (64 90° pulses , delay time 10 min)
Conclusion
15N-CIDNP effects have been observed in the nitration
products of 1 during three different nitration procedures, a
photonitration with tetranitromethane and the reactions
with nitric and nitrous acid. The effects are identical indi-
cating a common reaction step, which is the recombination
•
of the radicals NO2 with radical cations 1ϩ• or phenoxyl
radicals PhO• (Eq. 7), for the product formation in all the
three systems, one of the rare examples of a radical cation-
radical reaction[27]. In the case of the photochemical reac-
tion, the radicals are generated by electron transfer in the
photoactivated state of a charge-transfer complex of tetra-
nitromethane (Eq. 3). They are also formed by nitrous acid
catalysis and decomposition during the reaction with nitric
and nitrous acid (NAC, Eqs. 4a,b, 8b).
Applying the nitrous acid system, 2b is formed in part
via a non-radical reaction, which is a nitrosation followed
by oxidation (NOX, Eqs. 5a,b). The presence of the radicals
NO• and 1ϩ• make it probable that the formation of 3 oc-
curs via recombination and deprotonation (Eq. 6b). It
might be concluded that not only the nitration, but also the
C-nitrosation of aromatic compounds with distinct elec-
tron-donor properties, is a radical reaction in general.
•
2 HNO2 v NO• ϩ NO2 ϩ H2O
(8a)
(8b)
•
Ϫ
2 NO2 ϩ H2O v HNO2 ϩ Hϩ ϩ NO3
The reaction is slow enough to allow the determination
of 15N-CIDNP intensities and enhancement factors E and
Ecalc (see Tables 2 and 3). In the case of 2a , the results
agree well with those found during the reaction of 1 with
nitric acid in acetonitrile. The E values of 2b are somewhat
lower than those observed for 2a, which is explained by a
non-radical formation of 2b in part via NOX (Eqs. 5a,b),
and this has been postulated to occur during the nitration
This work was supported by the Fonds der Chemischen Industrie.
We thank Radiant Dyes, Wermelskirchen, Germany, for giving us
the possibility for using the argon ion laser.
Experimental Section
of 1 with nitric acid in the presence of nitrous acid[14]
The absorption signal observed at δ ϭ 3.2 is assigned to
.
Photochemical 15N-CIDNP Experiments: The photochemical
experiments were performed within a modified 15N-NMR probe
head of a Bruker DPX-300 NMR spectrometer equipped with a
light conductor which was connected to a LEXEL 95 Ar-Ion Laser
(Polytec Inc.) The degassed reaction mixtures were irradiated at
room temp. with the unfiltered light of the laser [λ ϭ 514 nm (2.9
Ϫ
[25]
15NO3 or H15NO3
(Eq. 8b). The intensity is higher as
expected, indicating a slightly enhanced absorption. This
•
might be a polarisation of ЉeЉ-type built up in 15NO2 dur-
ing the reaction shown in Eq. 7[26], which is transferred to
Ϫ
15NO3 by Eq. 8b. The 15N-NMR signal of the nitrosation W), λ ϭ 488 nm (2.4 W)]. The spectra were taken with 30 scans of
130 pulses with a delay time of 3 s. After an irradiation time tE,
product (3), which is the main product, does not show a
spectra were taken with 501 scans. 15N-chemical shifts are reported
15N-CIDNP effect which might be formed during the re-
combination reaction of 15NO• with 1ϩ• (Eq. 6b).
1
in δ values downfield relative to the standard Ph-15NO2, H- and
13C-NMR data relative to TMS. Raman spectra were recorded on
a T 64000 (ISA) with an argon laser at λ ϭ 514.4 nm (Spectra
Physics). Enhancement factors E were determined by using Eq.
Both radicals are present as is indicated by the appear-
ance of 15N-CIDNP in 2a and 2b. The absense of 15N-
CIDNP effects does not prove the non-radical mechanism
of the nitrosation reaction. The low g value of NO•
(1.95[19]) makes it difficult or impossible to observe CIDNP.
The transversal relaxation time T2 in solution is not known.
A low value of T2 has to be expected because of the or-
bitally degenerated ground state of the radicals NO•. This
would also diminish the magnitude of the effect. There are
further chemical reasons which make it probable that a rad-
ical C-nitrosation is the main reaction, and that an electro-
philic aromatic substitution reaction (Eq. 6a) only plays a
minor role, if any. C-nitrosation only happens with aromatic
compounds that are more reactive than toluene[11]. These
are the same systems which are nitrated via the radical
9[18][28]
.
E ϭ ImaxtE/IoT1
(9)
Imax, the intensity of the CIDNP signal, and Io, the intensity
after irradiation, were taken from the spectra under consideration
of the different scan numbers. An estimated filling factor of 2 was
taken into account. 15N-nuclear longitudinal relaxation times T1
were determined by applying πϪπ/2 pulse sequences. For getting
appropriate signal-to-noise ratios, delay times have been applied
which are much shorter than the T1 values[29]. 2a, 2b and
1
HC(NO2)3 were identified by using H-, 13C-, 14N- and 15N-NMR
data. The product ratio of 2a and 2b (1.1) was taken from 1H-
NMR spectra and proved by GC analysis [GC DANI Mod. 8521-
pathway NAC (Eqs. 4). Furthermore, it has been estimated A(25 m CP-SIL-5CB)].
Eur. J. Org. Chem. 1998, 913Ϫ918
917

M. Lehnig, K. Schürmann
FULL PAPER
[5]
[6]
Preparation of Tetranitromethane-15N : The compound was syn-
thesized as described[30]. NaNO3 with 10 atom% 15N was used (Iso-
tec Inc.). Ϫ 15N NMR (CD3CN): δ ϭ Ϫ38.6 (s). Ϫ 13C NMR
(CD3CN): δ ϭ 115.4 (q). Ϫ Raman: ν˜ ϭ 1615 cmϪ1, 1649 (νa NO2);
1341, 1247 (νs NO2); 360, 408, 417, 608 ( δ NO2).
Thermal 15N-CIDNP Experiments: After putting the reactants
into the 10-mm NMR tubes, they were quickly transferred into the
probe head of the 15N-NMR spectrometer and locked within 2 min
(internal lock: D2O). 15N-NMR spectra were then taken every
3Ϫ12 min using single pulses with a pulse angle of 90° until the
reactions were completed. 15N-NMR relaxation times T1 were then
determined applying πϪπ/2 pulse sequences. The repetition times
were long enough to avoid saturation. Shorter pulses have also been
applied. They give identical results, but have not been used for the
quantitative measurements, as the signal-to-noise ratio of Imax and
J. H. Ridd, Chem. Soc. Rev. 1991, 20, 149Ϫ165.
L. Eberson, F. Radner, Acta Chem. Scand. B, 1984, 38,
861Ϫ870; T. Lund, L. Eberson, J. Chem. Soc., Perkin Trans. 2,
1997, 1435Ϫ1443.
[7]
[8]
S. Sankararaman, W. A. Haney, J. K. Kochi, J. Am. Chem. Soc.
1987, 109, 5235Ϫ5249.
L. Eberson, M. P. Hartshorn, F. Radner, J. O. Svensson, Acta
Chem. Scand. 1996, 50, 885Ϫ898; C. P. Butts, L. Eberson, M.
Hartshorn, O. Persson, Acta Chem. Scand. 1997, 51, 718Ϫ732;
C. P. Butts, L. Eberson, M. P. Hartshorn, O. Persson, R. S.
Thompson, W. T. Robinson, Acta Chem. Scand. 1997, 51,
1066Ϫ1077.
[9]
H. Martinsen, Z. Phys. Chem. 1904, 50, 385Ϫ435.
J. C. Giffney, J. H. Ridd, J. Chem. Soc., Perkin Trans. 2, 1979,
618Ϫ623.
[10]
[11]
[12]
D. L. H. Williams, Nitrosation, Cambridge University Press,
Cambridge 1988 ; B. C. Challis, A. J. Lawson, J. Chem. Soc.
(B) 1971, 770Ϫ775; B. C. Challis, R. J. Higgins, A. J. Lawson,
J. Chem. Soc., Perkin Trans. 2, 1972, 1831Ϫ1836.
L. Eberson, M. P. Hartshorn, O. Persson, Angew. Chem. Int.
Ed. Engl. 1995, 34, 2268Ϫ2269; L. Eberson, O. Persson, F.
Radner, M. P. Hartshorn, Res. Chem. Intermed. 1996, 22,
799Ϫ820.
Ϫ
Io is somewhat smaller. The identification of 2a, 2b, 3 and NO3
was performed by comparing 13C-, 14N- and 15N-NMR data with
those of authentic material and the literature[25]. Product ratios
were taken from 15N-NMR and 13C-NMR spectra. The nitric acid
used was 9.4  in H2O and labelled with 60.3 atom% 15N (Isotec
Inc.). NaNO2 was labelled with 99.3 atom% 15N (Isotec Inc.). All
the compounds and solvents were commercial and used without
[13]
[14]
[15]
[16]
S. Seltzer, E. Lam, L. Packer, J. Am. Chem. Soc. 1982, 104,
6470Ϫ6471.
D. S. Ross, G. P. Hum , W. G. Blucher, J. Chem. Soc., Chem.
Comm. 1980, 532Ϫ534.
F. Radner, A. Wall, M. Loncar, Acta Chem. Scand. 1990, 44,
152Ϫ157.
further purification. E was determined using Eq. 10[18]
.
R. G. Coombes, J. G. Golding , P. Hadjigeourgiou, J. Chem.
Soc., Perkin Trans. 2, 1979, 1451Ϫ1459.
E ϭ Σ Ii∆ t(i,iϩ1)/IoT1
(10)
[17]
[18]
J. B. Pedersen, J. Chem. Phys. 1977, 67, 4079Ϫ4102.
M. Lehnig, J. Chem. Soc., Perkin Trans. 2, 1996, 1943Ϫ1948;
M. Lehnig, Acta Chem. Scand. 1997, 51, 211Ϫ213; M. Lehnig,
Chem. Phys. 1975, 8, 419Ϫ424; 1981, 54, 323Ϫ329.
Landolt-Börnstein, Neue Serie Bd. 9, 17, Magnetische Eigen-
schaften freier Radikale, Ed. H. Fischer, K.-H. Hellwege,
Springer-Verlag Berlin, Heidelberg, New York, 1987.
A. W. Hofmann, Ann. Chem. Pharm. 1850, LXXV, 356Ϫ368.
F. Arnall, J. Chem. Soc. 1924, 811Ϫ816.
Ii is the intensity of the ith measurement, ∆t(i,iϩ1) the time inter-
val between the ith and the (iϩ1)th measurement. The summation
was carried out as long as the CIDNP effects could be observed.
For improving the signal-to-noise ratio of Io, several 90° pulses with
a delay time of 10 min were used in some cases.
[19]
[20]
[21]
[22]
[23]
[24]
J. L. Bridge, Liebigs Ann. Chem. 1893, 277, 79Ϫ104.
A. Baeyer, H. Caro, Ber. Dtsch. Chem. Ges. 1874, 7, 963Ϫ968.
B. D. Beake, J. Constantine, R. B. Moodie, J. Chem. Soc., Perkin
Trans. 2 1992, 1653Ϫ1654.
[1]
G. A. Olah, R. Malhotra, S. C. Narang, Nitration, Verlag
Chemie, New York 1989; K. Schofield, Aromatic Nitration,
Cambridge University Press, Cambridge 1980.
[2]
[25]
T. M. Bockman, J. K. Kochi, J. Phys. Org. Chem. 1994, 7,
M. Witanowski, L. Stefaniak, G. A. Webb, Ann. Rep. NMR
Spectrosc. 1993, 25, 1, and references cited therein.
R. Kaptein, Chem. Comm. 1971, 732Ϫ733.
325Ϫ351; L. Eberson, M. P. Hartshorn, F. Radner, Advances in
Carbocation Chemistry, Ed. J. M. Coxon, JAI Press, London
1995, vol. 2, pp 207Ϫ263.
[26]
[27]
M. Schmittel, A.Burghart, Angew. Chem. 1997, 109,
2658Ϫ2699.
[3]
C. A. Bunton, E. D. Hughes, G. J. Minkoff , R. I. Reed, Nature
[28]
(London) 1946, 158, 514Ϫ515; E. D. Hughes, C. K. Ingold, R.
I. Reed, J. Chem. Soc. 1950, 2400Ϫ2440; C. A. Bunton, E. D.
Hughes, C. K. Ingold, D. I. H. Jacobs, M. H. Jones, G. J. Min-
koff, R. I. Reed, J. Chem. Soc. 1950, 2628Ϫ2656.
M. Lehnig, H. Fischer, Z. Naturforsch., Teil A, 1970, 25,
1963Ϫ1969; 1972, 27, 1300Ϫ1307.
[29]
[30]
R. R. Ernst, W. A. Anderson, Rev. Sci. Instrum. 1966, 37, 93.
W. Skawinski, J. Flisak, A. C. Chung, F. Jordan, R. Mendel-
sohn, J. Labelled Compds. 1990, 28, 1179Ϫ1183.
[97295]
[4]
J. Kenner, Nature (London) 1945, 156, 369Ϫ370; C. L. Perrin,
J. Am. Chem. Soc. 1977, 99, 5516Ϫ5518.
918
Eur. J. Org. Chem. 1998, 913Ϫ918