Gas-solid reactions of single crystals: A study of reactions of NH 3 and NO2 with single crystalline organic substrates by infrared microspectroscopy

Article abstract of DOI:10.1016/j.saa.2005.03.030

Reaction of single crystals of benzoic and trans-cinnamic acids with 200 Torr pressure of ammonia gas in a sealed glass bulb at 20 °C generates the corresponding ammonium salts; there is no sign of any 1:2 adduct as has been reported previously for related systems. Isotopic substitution using ND ₃ has been used to aid identification of the products. Adipic acid likewise reacts with NH₃ gas to form a product in which ammonium salts are formed at both carboxylic acid groups. Reaction of 0.5 Torr pressure of NO₂ gas with single crystals of 9-methylanthracene and 9-anthracenemethanol in a flow system generates nitrated products where the nitro group appears to be attached at the 10-position, i.e. the position trans to the methyl or methoxy substituent on the central ring. Isotopic substitution using ¹⁵NO₂ has been used to confirm the identity of the bands arising from the coordinated NO₂ group. The products formed when single crystals of hydantoin are reacted with NO₂ gas under similar conditions depend on the temperature of the reaction. At 20 °C, a nitrated product is formed, but at 65 °C this gives way to a product containing no nitro groups. The findings show the general applicability of infrared microspectroscopy to a study of gas-solid reactions of organic single crystals.

Full text of DOI:10.1016/j.saa.2005.03.030

Spectrochimica Acta Part A 62 (2005) 1131–1139
Gas–solid reactions of single crystals: A study of reactions of NH3
and NO2 with single crystalline organic substrates by infrared
microspectroscopy
Samantha L. Jenkins, Matthew J. Almond∗, Samantha D.M. Atkinson,
Peter Hollinsꢀ, John P. Knowles
School of Chemistry, University of Reading, Whiteknights, Reading RG6 6AD, UK
Received 26 January 2005; accepted 18 March 2005
Abstract
Reaction of single crystals of benzoic and trans-cinnamic acids with 200 Torr pressure of ammonia gas in a sealed glass bulb at 20 ◦C
generates the corresponding ammonium salts; there is no sign of any 1:2 adduct as has been reported previously for related systems. Isotopic
substitution using ND3 has been used to aid identification of the products. Adipic acid likewise reacts with NH3 gas to form a product in which
ammonium salts are formed at both carboxylic acid groups. Reaction of 0.5 Torr pressure of NO2 gas with single crystals of 9-methylanthracene
and 9-anthracenemethanol in a flow system generates nitrated products where the nitro group appears to be attached at the 10-position, i.e.
the position trans to the methyl or methoxy substituent on the central ring. Isotopic substitution using 15NO2 has been used to confirm the
identity of the bands arising from the coordinated NO2 group. The products formed when single crystals of hydantoin are reacted with NO2
gas under similar conditions depend on the temperature of the reaction. At 20 ◦C, a nitrated product is formed, but at 65 ◦C this gives way
to a product containing no nitro groups. The findings show the general applicability of infrared microspectroscopy to a study of gas–solid
reactions of organic single crystals.
© 2005 Elsevier B.V. All rights reserved.
Keywords: Gas–solid reactions; Carboxylic acid; Hydantoin; Solvent-free nitration; Infrared microspectroscopy; Raman microspectroscopy
1. Introduction
interest in combinatorial chemistry where a wide range of
reagents are tested on solid beads of substrates provides a
further impetus to research in this area.
In recent years there has been a strong renewal of interest
in the reactions of solid crystalline organic substrates [1–5].
Reactions of this type are desirable from an environmental
point of view in that they obviate the need for large quan-
tities of solvent as the reaction medium [4,6]. At the same
time it has become increasingly recognised that such reac-
tions give considerable stereochemical control, as the form
of the reaction product may be directed by the arrangement of
reactant molecules within the crystal [2,3,7–11]. Performing
reactions in the solid-state has been found in many examples
to increase yields and improve specificity. The explosion of
We have utilised the techniques of infrared and Raman
microspectroscopy to monitor reactions of this type and to
characterise reaction products [1–3,12]. The principal advan-
tages of vibrational microspectroscopy in this context are that
single crystals may be monitored in situ and that the spec-
tra give considerable structural information about reactants
and products. Such structural information is much less read-
ily forthcoming from an alternative approach, which has been
used elsewhere [4,6,13,14] in which atomic force microscopy
is employed. Here morphological changes to the crystals are
seen upon reaction, but it is difficult to relate these changes to
the chemical processes ongoing within the crystal. Diffrac-
tion methods have, of course, also been used [8,9,15] but
are still undergoing development for use in time-resolved
∗
Corresponding author. Fax: +44 1 1893 16332.
E-mail address: m.j.almond@reading.ac.uk (M.J. Almond).
ꢀ
Deceased.
1386-1425/$ – see front matter © 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.saa.2005.03.030

1132
S.L. Jenkins et al. / Spectrochimica Acta Part A 62 (2005) 1131–1139
monitoring of solid-state reactions [16]. Single crystal meth-
ods may present difficulties if the crystals under investigation
degrade during the experiment [17]. In this context, however,
it is interesting to note a very recent single crystal diffrac-
tion study of the photodimerisation of the ␣ꢀ-polymorph
of ortho-ethoxy-trans-cinnamic acid in the solid-state [18].
Until now we have focused our attention upon photochemical
dimerisation reactions of single crystals of trans-cinnamic
acid and its derivatives [1–3,12]. An integrated approach
using infrared and Raman microspectroscopy clearly demon-
strates the topotactic nature of these reactions where the
form of the dimer obtained is dependent upon the orienta-
tion of the monomer molecules within the starting compound
[2,3]. We have also extended our studies to investigate the
time-dependency of the dimerisation of molecules of ␣-2,4-
dichloro-trans-cinnamic acid within a single crystal [1]. The
reaction is shown to follow strictly first-order kinetics, rather
than the zeroth-order kinetics predicted by some theoretical
models [1].
We have now turned our attention to the reactions of sin-
gle crystalline organic substrates with a range of gaseous
reagents. Such reactions are considered to be relatively envi-
ronmentally friendly and often they give yields at least as high
asthoseobtainedbyanalogoussolutionreactions[4–6]. Their
full exploitation, however, in synthesis is held back because
of uncertainty as to how controllable the reactions are and as
to how the products of the reactions relate to products formed
by analogous reactions in solution. We selected two areas for
study where we felt that there was uncertainty regarding the
nature of the reaction products and/or the reaction mechanism
which could readily be addressed by vibrational microscopy.
The two areas were reactions with (i) ammonia and (ii) NO2
gas.
Scheme 1. The reactions of hydantoin with NO2 gas.
system. Insufficiently activated aromatic rings do not, how-
ever, react readily with NO2. Thus, the non-activated arenes
toluene and chlorobenzene are better nitrated with the ternary
mixture NO–NO2–O2 [5]. It is thought that nitrogen triox-
ide is involved in this reaction. The second substrate whose
nitration reaction we wished to monitor was hydantoin. Reac-
tionsofhydantoin(C3H4N2O4)—aderivativeofurea—areof
interest as derivatives of this compound find a wide range of
uses from their pharmaceutical role as anticonvulsants in the
treatment of epilepsy to their use as preservatives in inhibiting
the growth of moulds in shampoos and cosmetics. Moreover,
with this substrate the reaction products are reported to vary
with changes in temperature [4]. At room temperature a nitra-
tion product is formed while as the temperature is raised this
gives way to an oxidation product containing three carbonyl
groups (see Scheme 1).
It is thus apparent that there remain uncertainties in the
interpretation of these reactions and that, with the exception
of a number of studies by single crystal X-ray diffraction
[8,9,15,16] and by atomic force microscopy [4,6,13,14], few
investigations of single crystals have been made. Vibrational
microspectroscopy has not, to our knowledge, previously
been employed to investigate the reaction of any organic sin-
gle crystal with a gaseous reagent. Given the current general
interest in such reactions and our previous success in utilising
vibrational microspectroscopy to monitor photodimerisation
reactions in single crystals we felt that the study reported
here was very timely. We felt, moreover, that our approach
could usefully address specific questions regarding these par-
ticular reactions such as whether 1:1 salts or 1:2 adducts are
formed from ammonia and the nature of the products formed
by nitration.
Gas–solid reactions of carboxylic acids with ammonia gas
were first observed in 1884 [19]; here interest lies in whether
1:1 or 1:2 products are formed. In this case it appears that
some acids, for example, p-nitrobenzoic acid, react with am-
monia gas to give a 1:1 salt RCOO−NH4+ which then reacts
+
with more ammonia to form a 1:2 adduct RCOO−NH4 ·NH3
[20]. Upon standing the 1:2 adduct reverts to the 1:1 salt. By
contrast p-chloro- and p-bromobenzoic acid, while giving the
1:1 salt, could not be induced to form a 1:2 adduct [21]. That
such a simple gas–solid reaction should in fact show subtle
complexities lends weight to the argument that it is difficult
to predict the products of such reactions with certainty and
provides a rationale for investigating reactions within single
crystals. Even more complex are the reactions of NO2 gas.
When anthracene was first reacted with NO2 gas, more than
100 years ago, Liebermann and Lindemann suggested that 9-
nitro-10-nitrito-9,10-dihydroanthracene is the primary prod-
uct [22] while almost contemporary experiments by Leeds
[23] led to the isolation of only anthraquinone. Later stud-
ies have suggested that anthraquinone is a secondary product
formed via nitrated intermediates [24]. It appears that a more
controlled reaction may be achieved if deactivating methyl or
methoxy groups are attached to the 9-position of the aromatic
2. Experimental
Samples of all organic solids were used as supplied by
Aldrich. The stated purities of the reagents used were as

S.L. Jenkins et al. / Spectrochimica Acta Part A 62 (2005) 1131–1139
1133
follows: benzoic acid, trans-cinnamic acid and adipic acid
(all 99% purity) and hydantoin (99%) were used as supplied
by Aldrich. Ammonia (99.96% purity) and nitrogen diox-
ide (>99.5% purity) gases were supplied in lecture bottles
by Argo International and Aldrich, respectively. Isotopically
labelled ND3 (99% isotopic purity) and 15NO2 (99% isotopic
purity) were used as supplied by Aldrich and MSD, respec-
tively.
Reactions of ammonia gas were performed in a sealed
glass bulb in which the single crystal under investigation was
placed. A pressure of 200 Torr of ammonia was added to the
bulb on a vacuum line and the system was left for a period of 1
month to react. Because of the reactive nature of NO2 a flow,
rather than a static system was employed to limit the contact
time between the solid reagent and the NO2 gas. Crystals of
the reagent under investigation were held in glass wool in a
glass tube through which NO2 gas was flowed on a vacuum
line. A pressure of 0.5 Torr of gas was maintained and was
monitored using a Pirani gauge.
The infrared (Brucker IRscope II attached to Brucker
Equinox 55 FT interferometer) microscope which was used
to record vibrational spectra of single crystals has been
described elsewhere [2,3,25]. Infrared spectra of bulk sam-
pleswererecordedusingaPerkin-Elmer1720-XFTIR, NMR
spectra were measured on a Bruker DPX 250 MHz spectrom-
eter at 250 Hz in DMSO-d6 and mass spectra were obtained
using a VG Autospec mass spectrometer.
spectrum. The latter feature overlapped with νsym.(OCO) of
the carboxylate ion. In order to confirm which bands arise
from the ammonium cation and, in particular, to observe the
ν4 band free from overlap with other features, the reaction
was repeated using ND3 in place of NH3. A broad band is
now seen centred at 2340 cm−1 together with a much sharper
band at 1070 cm−1. These findings demonstrate that forma-
tion of the ammonium salt occurs upon exposure of benzoic
acid to ammonia gas. It appears that the reaction has gone
to completion as the ν(C O) carboxyl peak has decayed
to extinction. The bands seen before and after exposure of
a single crystal of benzoic acid to NH3 gas are listed in
Table 1, and the observed infrared spectra are illustrated in
Fig. 1.
We then turned our attention to the reaction of trans-
cinnamic acid with ammonia. We wished to explore the
possibility that the presence of a conjugated unsaturated
C C bond might influence the reaction and, moreover, we
wished to utilise the position of the ν(C C) band to follow
the reaction. A single crystal of trans-cinnamic acid was
exposed to 200 Torr pressure of ammonia gas for a period of
1 month. The observed infrared spectra of the crystal before
and after exposure are illustrated in Fig. 2. The positions of
bands are listed in Table 2. The principal changes observed
are as follows. First, ν(C O) at 1682 cm−1 decays in inten-
sity, but does not decay to extinction. Secondly, ν(C C),
aliphatic at 1632 cm−1 suffers a blue shift to 1640 cm−1
.
Thirdly, bands arising from νasym. and νsym.(OCO) of the
carboxylate group (CO2−) may be identified in the product
at ca. 1540 and 1400 cm−1. Fourthly, rather broad bands of
the ammonium cation from the ν3 and ν4 vibrations [26]
may be identified at ca. 3100 and 1400 cm−1. The bands
from νasym. and νsym.(OCO) and from ν4(NH4+) appear as a
3. Reactions of organic substrates with ammonia gas
The first reagent selected for study was benzoic acid.
A single crystal of benzoic acid was exposed to 200 Torr
pressure of NH3 gas for 1 month. During this time the fol-
lowing spectral changes were observed. The band arising
from ν(C O), initially at 1694 cm−1, decayed to extinction,
as would be expected as the carboxyl (CO2) group is con-
verted to a carboxylate (CO2−) group. Bands arising from
νasym. and νsym.(OCO) of the carboxylate group were identi-
fied in the product at 1580 and 1420 cm−1. Also, rather broad
bands of the ammonium cation from the ν3 and ν4 vibrations
[26] were seen around 3100 and 1400 cm−1 in the product
rather broad overlapping feature centred around 1450 cm−1
.
The blue shift in ν(C C) aliphatic may be understood in
terms of loss of conjugation to the carboxyl (C O) group
upon formation of the ammonium salt. Taken together these
findings demonstrate that formation of the ammonium salt
occurs upon exposure of trans-cinnamic acid to ammonia
gas, but that the reaction does not go entirely to completion
even after 1 month’s reaction time. In order to confirm the
assignments of bands of the product of trans-cinnamic acid
Table 1
Positions of infrared absorptions (cm−1) of a single crystal of benzoic acid before and after exposure to 200 Torr pressure of ammonia gas for 1 month
Assignment
Benzoic acid (cm−1
)
Benzoic acid after exposure to
Benzoic acid after exposure to
ND3 gas (cm−1
)
ammonia gas (cm−1
)
+
ν(N H(D)) (NH(D)4
)
∼3100 br
∼3227 br
–
2340 br
3060 br
–
ν(C H) unsaturated (aromatic)
ν(C O) (carboxyl)
∼3072 br
1694 vs
νasym.(OCO) carboxylate
νsym.(OCO) carboxylate
1580
1590 s
1410 s
1070
1420 sa
∼1400 sha
+
a
δ(N H(D)) (N(D)4
)
δ(C H), δ(C O) and ring vibrationsb
1494, 1454 vs, 1326 s, 1291 s, 1181 s,
1129 s, 1073, 934 vs, 809 s, 711 vs
1181, 1134, 1103 s, 1073 s,
944, 814 s, 715 s, 666 s
1174 w, 1100, 960 w, 834 s,
720 s, 680
a
It is difficult to identify the position of this band with certainly because of the presence of broad overlapping bands in this region.
Bands in this spectral region are difficult to assign with certainty because of coupling of vibrations and overlap of bands.
b

1134
S.L. Jenkins et al. / Spectrochimica Acta Part A 62 (2005) 1131–1139
Fig. 1. Infrared spectra of a single crystal of benzoic acid (bottom), after exposure to 200 Torr pressure of ammonia gas for 1 month (centre) and after exposure
to 200 Torr pressure of ND3 for 1 month (top).
exposed to ammonia the experiment was repeated using ND3
The third reagent that we selected for study in its reac-
tion with ammonia was adipic acid (HOOCH2CH2CH2
CH2COOH), which contains two carboxylic acid groups. The
interest in studying this reaction is to determine whether one
or both acid groups react and whether these groups react
simultaneously or sequentially.
+
in place of NH3. The bands assigned to ν3 (ν(N H))(NH4
)
and ν4 (δ(N H))(NH4+) at 3100 and 1400 cm−1 shift to ca.
2350 and 1065 cm−1 in the deuterated ammonia product.
+
Such shifts are in line with the formation of an ND4 cation
+
in place of NH4 [26d].
Fig. 2. Infrared spectra of a single crystal of trans-cinnamic acid before (below) and after (above) exposure to 200 Torr pressure of ammonia gas for 1 month.

S.L. Jenkins et al. / Spectrochimica Acta Part A 62 (2005) 1131–1139
1135
Table 2
Positions of infrared absorptions (cm−1) of a single crystal of trans-cinnamic acid before and after exposure to 200 Torr pressure of ammonia gas for 1 month
Assignment
Trans-cinnamic acid (cm−1
)
Trans-cinnamic acid after
exposure to NH3 gas (cm−1
Trans-cinnamic acid after
exposure to ND3 gas (cm−1
)
)
+
ν(N H(D)) (NH(D)4
)
–
∼3100 br
∼3000 br
2800
1681 w
1640
∼2350 br
∼3000 br
2810
1685 w
1640
ν(C H) unsaturated (aromatic)
ν(C H) unsaturated (aliphatic)
ν(C O) (carboxyl)
∼3025 br
–
1682
1632
ν(C C) (aliphatic)
νasym.(OCO) carboxylate
νsym.(OCO) carboxylate
–
–
–
∼1540 br
∼1400 bra
∼1400 bra
∼1540 br
∼1400 br
1065
∼1450 br, 1290, 1250 s, 1080 s,
1025, 978 vs, 920, 880 s, 846,
771 s, 715 s.
+
a
δ(N H(D)) (NH(D)4
)
δ(C H), δ(C O) and ring vibrationsb
1494, 1450 s, 1422 s, 1316, 1209,
1205, 1177, 1068, 1027, 935, 872 s.
∼1450 br, 1289, 1246 s, 1201,
1163, 1072 s, 972 s, 915, 877 s,
847, 711 s, 712 s, 687 s.
a
It is difficult to identify the position of this band with certainly because of the presence of broad overlapping bands in this region.
Bands in this spectral region are difficult to assign with certainty because of coupling of vibrations and overlap of bands.
b
Infrared spectra show that adipic acid undergoes a reaction
with ammonia (200 Torr pressure of NH3; 1 month reaction
time). The band arising from ν(C O), at around 1700 cm−1
0.5 Torr) in a flow system on a vacuum line for ca. 2 h. At the
end of this time crystals were removed and were studied by
infrared microspectroscopy.
,
decays almost to extinction over the course of the reaction.
A sharp, weak feature at 1716 cm−1 seen in the product may
be due to a small amount of unreacted starting material or it
may arise from an overtone or combination band of the prod-
uct. It is thus seen that the reaction occurs at both carboxyl
groups. No spectra, obtained on crystals after shorter reaction
times, were observed where just one group has reacted, so it
would appear that the reaction at both groups is more-or-less
simultaneous. Broad bands of the ammonium cation from the
ν3 and ν4 vibrations [26] may be identified at ca. 3100 and
1400 cm−1. Bands seen before and after exposure of a single
crystal of adipic acid to NH3 gas are listed in Table 3.
Infrared spectra of a crystal of 9-methylanthracene before
and after exposure to NO2 gas are shown in Fig. 3 and the
positions of infrared bands are listed in Table 4. Striking
changes to the spectrum are observed. Water is clearly a
by-product of the reaction, as evidenced by the growth of
intense broad bands centred around 3464 and 1663 cm−1
.
The presence of a nitro group in the product is shown by
intense broad bands centred around 1550 and 1300 cm−1
which may be assigned to νasym. and νsym.(NO2) [26]. A
strong band at 828 cm−1 and a weaker feature at 625 cm−1
,
which are not seen for the starting material probably arise
from δ(ONO) and ν(C N) modes of the nitro group [26].
To back up this assignment a spectrum of a single crystal
of an authentic sample of 9-nitroanthracene was recorded.
This shows strong bands centred around 1550, 1340, 840 and
630 cm−1 corresponding to the bands seen above. The simi-
larity in position between the bands of the nitration product
and of 9-nitroanthracene suggest that the nitration product
may be 9-methyl-10-nitroanthracene. In order to confirm
the assignments of bands of nitrated 9-methylanthracene the
experiment was repeated using 15NO2 in place of 14NO2. The
wavenumber shifts of bands upon isotopic substitution are
given in Table 5. The calculated isotopic shifts assume har-
monic diatomic N O or N C oscillators and are based upon
4. Reactions with NO2 gas
Two crystalline substrates were selected for study in their
reactions with NO2 gas. These were: (i) 9-methylanthracene
and (ii) hydantoin, and they were chosen because a rich chem-
istry with NO2 gas had been suggested in previous work
[4,24]. As NO2 is highly reactive towards these substrates,
and as we wished to study primary reaction products, we
modified our experimental procedure such that the crystals,
supported in glass wool, were exposed to low pressures (ca.
Table 3
Positions of infrared absorptions (cm−1) of a single crystal of adipic acid before and after exposure to 200 Torr pressure of ammonia gas for 1 month
Assignment
Adipic acid (cm−1
)
Adipic acid after exposure to ammonia gas (cm−1
)
+
ν(N H) (NH4
)
∼3100 br
∼2917 br,
1716
ν(C H) unsaturated (aromatic)
ν(C O) (carboxyl)
∼2966 br,
1700 br,
1573
νasym.(OCO) carboxylate
+
a
δ(N H) (NH4
δ(C H)b
)
∼1400 bra
1478, 1300 br, 1196, 1045, 939,
904, 815 w, 732, 674
1311 br, 1208, 1154 s, 1103, 1094, 1079, 1028, 909 s,
784 s, 648 s
a
It is difficult to identify the position of this band with certainly because of the presence of broad overlapping bands in this region.
Bands in this spectral region are difficult to assign with certainty because of coupling of vibrations and overlap of bands.
b

1136
S.L. Jenkins et al. / Spectrochimica Acta Part A 62 (2005) 1131–1139
Fig. 3. Infrared spectra of a single crystal of 9-methylanthracene before (below) and after (above) exposure to 0.5 Torr pressure of NO2 gas in a flow system
for 2 h.
reduced mass calculations (Table 5). Given the broadness
of the observed bands a highly satisfactory correspondence
between observed and calculated values is found. Taken
together these results suggest that the product of this reaction
is 9-methyl-10-nitro-anthracene in line with previous find-
ings.
In order to confirm this supposition a bulk sample
of the product was analysed by mass spectrometry. The
Table 4
Positions of infrared absorptions (cm−1) of a single crystal of 9-methylanthracene before and after exposure to 0.5 Torr of NO2 gas in a flow system for 2 h
Assignment
9-Methylanthracene ν (cm−1
)
9-Methylanthracene exposed to NO2
(cm−1
ν
)
ν(O H) (H2O)
3464 br, s
3053
2918
1663
1550 s, br
1297 br
1601 sh, 1493 w, 1448, 1178, 1069 w,
993 s, 894, 768, 746, 697, 648.
828
625
ν(C H) aromatic
ν(C H) aliphatic
δ(H OH) (H2O)
νasym.(NO2)
νsym.(NO2)
δ(O H), δ(C H), ν(C C), ν(C O)a
3084, 3046
2934 s
1622 s, 1561 s, 1445, 1409 s, 1348 s, 1257, 1183 s,
1160 s, 998 s, 968 vs, 948, 885 s, 862 s, 838 s, 777 s.
δ(NO2)b
ν(C N)c
a
Bands in this spectral region are difficult to assign with certainty because of coupling of vibrations and overlap of bands.
Assigned to δ(NO2) on basis of wavenumber shift upon substitution with 15NO2.
b
c
Assigned to ν(C N) on basis of wavenumber shift upon substitution with 15NO2.
Table 5
Observed and calculated isotopic shifts for bands of coordinated NO2 group in 9-methyl-10-nitroanthracene formed by reaction of 9-methylanthracene with
(i) 14NO2 gas and (ii) 15NO2 gas
14NO2 observed
15NO2 observed
∆
∆
Robserved
(ν 14NO2/ν 15NO2)
Rcalculated
(ν 14NO2/ν 15NO2)a
Assignment
observed
calculated
(cm−1
)
(cm−1
)
(ν 14NO2 − ν 15NO2)
(ν 14NO2 − ν15NO2)
(cm−1
)
(cm−1
)
a
1550
1300
828
625
a
1526
1285
819
24
15
9
28
24
15
10
1.016
1.012
1.011
1.015
1.018
1.019
1.018
1.016
νasym.(NO2)
νsym.(NO2)
δ(NO2)
616
9
ν(C N)
Calculated values are maximum isotopic shifts based upon harmonic diatomic N O or C N oscillators.

S.L. Jenkins et al. / Spectrochimica Acta Part A 62 (2005) 1131–1139
1137
Fig. 4. Infrared spectra of a single crystal of hydantoin before (below) and after exposure to 0.5 Torr pressure of NO2 gas in a flow system at 20 ◦C (middle)
and 65 ◦C (top) for a period of 2 h in each case. The band marked (+) in the middle spectrum is assigned to νasym.(NO2) and the band marked (*) in the same
spectrum is assigned to νsym.(NO2).
observed principal peaks are m/z (EI): 236 (C15H10NO2),
222 (C14H8NO2), 209 (C14H11NO), 195 (C14H11O). It was
expected that 9-methylanthracene would be likely to nitrate
at the 10-position. The methyl group has activating +I and
+M effects; this makes the ring more susceptible to attack by
the electrophilic N atom of the NO2 molecule and the +M
effect makes the 10-position the most readily attacked. We,
therefore, repeated the experiment using single crystals of
the reagent 9-anthracene methanol. Here the methoxy group
at the 9-position has a −I, but a +M effect. As such the
ring should be a little less susceptible to nitration, but again
attack at the 10-position should be favoured. After exposure
to NO2 gas a single crystal of the product shows new infrared
absorptions at 1553, 1300, 817 and 625 cm−1 which may be
assigned to νasym.(NO2), νsym.(NO2), δ(NO2) and ν(C N),
respectively. Upon reaction with 15NO2 these bands shift to
1524, 1275, 810 and 615 cm−1, confirming the above assign-
ments.
The interest in studying hydantoin reactions came, in part,
from the fact that two quite different products have been
reported depending on the reaction temperature. Single crys-
tals of hydantoin were exposed to a flow of NO2 gas at a
pressure of ca. 0.5 Torr for ca. 2 h. The first reaction was car-
ried out at room temperature (20 ◦C), the second at 65 ◦C.
Table 6
Positions of infrared absorption (cm−1) of a single crystal of hydantoin before and after exposure to 0.5 Torr pressure of NO2 gas in a flow system at 20 ◦C and
at 65 ◦C
Assignment
Hydantoin ν (cm−1
)
Hydantoin exposed to NO2 at
Hydantoin exposed to NO2 at
65 ◦C ␯ (cm−1
)
20 ◦C ␯ (cm−1
)
ν(O H), free
3554
3457
ν(N H), ν(O H) H bonded
ν(C H)
3100 br
2760
3100 br
3004, 2763 s
3100 br
2873
Overtones/combinations
2612, 2473, 2388 s, 2292, 2078, 1981 s
2488 s, 2350, 2273, 2099,
2370, 2355s
2065, 2011, 1974
ν(C O)
1700 br
1700 br
1766, 1711 s
νasym.(NO2)
νsym.(NO2)
ν(C C), δ(C H)a
1503 s
1238 s
1434, 1297, 1204 s, 1078 s, 992 s
1350, 1195 w, 988 vs, 926,
1581, 1415, 1355, 1200, 1142 s,
882 s
758
646
1064 s, 937 s, 850, 720, 680
–
–
δ(NO2)
ν(C N)
a
–
–
Bands in this spectral region are difficult to assign with certainty because of coupling of vibrations and overlap of bands.

1138
S.L. Jenkins et al. / Spectrochimica Acta Part A 62 (2005) 1131–1139
The observed spectra before and after exposure to NO2 gas
are illustrated in Fig. 4 and the observed infrared bands are
listed in Table 6. It may be seen that two quite different reac-
tion products have been formed under the different reaction
conditions.
microspectroscopy is an excellent way to identify such prod-
ucts. We have extended the work to a preliminary study of the
reaction of hydantoin with NO2 gas where different products
are formed depending on the temperature of the reaction. We
have shown that our approach allows these products to be
readily distinguished and allows structural information to be
obtained. The spectra show quite conclusively the presence
of an NO2 group in the product of reaction at 20 ◦C and the
absence of such a group when the reaction is performed at
65 ◦C.
An additional importance is given to the findings from
this work by the known use of these reactions in organic
synthesis. Nitrated hydantoins are useful as their reduction
products—the aminohydantoins—are important intermedi-
ates in the preparation of several hydantoin pharmaceuticals
[4].
Having carried out this preliminary survey of reactions to
establish the general applicability of our technique we will
now extend the work to look at specific aspects regarding
the time-dependency and temperature-dependency of prod-
uct formation of specific reactions. This will involve in situ
kinetic monitoring of the reactions of single crystals using
the approach that we have developed elsewhere for following
photochemical reactions of single crystals of trans-cinnamic
acid derivatives [1].
Upon reaction at 20 ◦C bands which may be assigned
to a coordinated nitro group appear in the product. These
bands are listed in Table 6; they occur at 1503, 1358, 756 and
646 cm−1. Assuming that the identity of the product shown
in Scheme 1 is correct, these bands may be identified with
νasym.(NO2), νsym.(NO2), δ(NO2) and ν(C N) vibrations of
the coordinated NO2 moiety. When the reaction was repeated
at 65 ◦C none of the above bands were seen in the product.
Rather the appearance of the spectrum in the ν(C O) region
changed considerably in comparison to the starting material
spectrum. New bands are seen at 1766 and 1711 cm−1. The
spectral changes are in keeping with the product shown in
Scheme 1.
5. Discussion
Gas–solid reactions of organic single crystals have been
followed for the first time by the technique of infrared
microspectroscopy. Our experiments have shown the general
applicability of this approach to a study of such reactions.
The relative simplicity of these techniques makes them very
attractive for monitoring such processes; moreover, structural
information regarding the products is readily obtained. At
the same time it is straightforward to relate the vibrational
spectra obtained from single crystals to vibrational spectra
obtained from bulk samples. Thus, it may be confirmed that
the product observed in a single crystal is the same as the bulk
product of the reaction. This point is important when reac-
tions where many different products may be generated, e.g.
nitration reactions, are considered. Taken together these con-
siderations give an advantage to our method over alternative
approaches where diffraction methods [8,9,15,16] or atomic
force microscopy [4,6,13,14] are utilised to follow reactions
of this type.
Acknowledgement
We are grateful to the University of Reading for providing
a studentship for SLJ and to EPSRC for providing a stu-
dentship for SDMA.
References
[1] S.D.M. Atkinson, M.J. Almond, S.J. Hibble, P. Hollins, S.L. Jenkins,
M.J. Tobin, K.S. Wiltshire, Phys. Chem. Chem. Phys. 6 (2004) 4.
[2] S.D.M. Atkinson, M.J. Almond, G.A. Bowmaker, M.G.B. Drew, E.J.
Feltham, P. Hollins, S.L. Jenkins, K.S. Wiltshire, J. Chem. Soc.,
Perkin Trans. 2 (2002) 1533.
Single crystals of benzoic and trans-cinnamic acid react
withammoniagastogivethecorrespondingammoniumsalts.
In these cases it is shown that the reaction proceeds to give
a 1:1 salt and there is no sign of any 1:2 adduct of NH3
such has been seen for related systems [20]. The presence
of a conjugated double bond system in cinnamic acid does
not appear to influence greatly the reaction. Adipic acid reacts
with ammonia at both carboxylic acid groups. Nitration reac-
tions using NO2 gas are much more complex in nature with
a range of products being observed depending on reaction
times and temperature [4,5,9,22,23]. This has unfortunately
held back the use of gas–solid reactions for nitration, which
have advantages of being relatively environmentally friendly
(they do not require solvents) and often giving high yields.
Our experiments show that under controlled conditions sin-
gle isolable products may be formed and that vibrational
[3] S.D.M. Atkinson, M.J. Almond, P. Hollins, S.L. Jenkins, Spec-
trochim. Acta Part A 59 (2003) 629.
[4] G. Kaupp, J. Schmeyers, J. Org. Chem. 60 (1995) 5494.
[5] H. Suzuki, T. Mori, J. Chem. Soc., Perkin Trans. 1 (1995) 291.
[6] G. Kaupp, U. Pogodda, J. Schmeyers, Chem. Ber. 127 (1994) 2249.
[7] S. Kanao, S. Kashino, M. Haisa, Acta Crystallogr. C46 (1990) 2436.
[8] S. Kanao, S. Kashino, M. Haisa, Acta Crystallogr. C46 (1990) 2439.
[9] V. Enkelmann, G. Wegner, K. Novak, K.B. Wagener, J. Am. Chem.
Soc. 115 (1993) 10390.
[10] Y. Ito, B. Borecka, J. Trotter, J.R. Scheffer, Tetrahedron Lett. 36
(1995) 6083.
[11] Y. Ito, B. Borecka, J. Trotter, J.R. Scheffer, Tetrahedron Lett. 36
(1995) 6087.
[12] S.D.M. Allen, M.J. Almond, J.-L. Bruneel, A. Gilbert, P. Hollins, J.
Mascetti, Spectrochim. Acta Part A 56 (2000) 2423.
[13] G. Kaupp, J. Schmeyers, Angew. Chem. Int. Ed. Engl. 32 (1993)
1587.
[14] G. Kaupp, Mol. Cryst. Liq. Cryst. 211 (1992) 1.

S.L. Jenkins et al. / Spectrochimica Acta Part A 62 (2005) 1131–1139
1139
[15] R.S. Miller, I.C. Paul, D.Y. Curtin, J. Am. Chem. Soc. 96 (1974)
6334;
[22] C. Liebermann, L. Lindemann, Ber. Dtsch. Chem. Ges. 13 (1880)
1584.
R.S. Miller, D.Y. Curtin, I.C. Paul, J. Am. Chem. Soc. 96 (1974)
6340.
[16] J.M. Cole, P.R. Raithby, M. Wulff, F. Schoffe, A. Plech, S.J. Teat,
G. Bushell-Wiye, Faraday Discuss. 122 (2003) 119.
[17] B.M. Kariuki, D.M.S. Zin, M. Tremayne, K.D.M. Harris, Chem.
Mater. 8 (1996) 565.
[23] A. Leeds, Ber. Dtsch. Chem. Ges. 14 (1881) 482.
[24] G.L. Squaddrito, F.R. Fronczek, S.F. Watkins, D.F. Church, W.A.
Pryor, J. Org. Chem. 55 (1990) 4322.
[25] S.D.M. Atkinson, Ph.D. Thesis, University of Reading, 2001.
[26] (a) W. Kemp, Organic Spectroscopy, third ed., Macmillan, Bas-
ingstoke, 1991;
[18] M.A. Fernandes, D.C. Levendis, Acta. Crystallogr. B B60 (2004)
315.
[19] G. Pellizzari, Gazz. Chim. Ital. 14 (1884) 362.
[20] R.S. Miller, D.Y. Curtin, I.C. Paul, J. Am. Chem. Soc. 96 (1908)
2009.
(b) D.H. Williams, I. Fleming, Spectroscopic Methods in Organic
Chemistry, fifth ed., McGraw Hill, Maidenhead, 1996;
(c) N.B. Colthrup, L.H. Daly, R.B. Wiberly, Introduction to Infrared
and Raman Spectroscopy, third ed., Academic Press, London, 1990;
(d) K. Nakamoto, Infrared and Raman Spectra of Inorganic and
Coordination Compounds, fourth ed., Wiley, New York, 1986.
[21] A. Korczynski, Chem. Zentralbl. 1908, 2009.