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 infraredmicrospectroscopy

Samantha L. Jenkins, Matthew J. Almond∗, Samantha D.M. Atkinson,Peter Hollins�, 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 andtrans-cinnamic acids with 200 Torr pressure of ammonia gas in a sealed glass bulb at 20◦Cgenerates the corresponding ammonium salts; there is no sign of any 1:2 adduct as has been reported previously for related systems. Isotopics ha enea -position, i.e.t hei h NOgt gas–solidr©

K

1

iRptttorrt

fides a

mand to-thatpec-tantsead-eeny

ls arees torac-

lved

1d

ubstitution using ND3 has been used to aid identification of the products. Adipic acid likewise reacts with NH3 gas to form a product in whicmmonium salts are formed at both carboxylic acid groups. Reaction of 0.5 Torr pressure of NO2 gas with single crystals of 9-methylanthracnd 9-anthracenemethanol in a flow system generates nitrated products where the nitro group appears to be attached at the 10

he positiontrans to the methyl or methoxy substituent on the central ring. Isotopic substitution using15NO2 has been used to confirm tdentity of the bands arising from the coordinated NO2 group. The products formed when single crystals of hydantoin are reacted wit2as 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

o a product containing no nitro groups. The findings show the general applicability of infrared microspectroscopy to a study ofeactions of organic single crystals.

2005 Elsevier B.V. All rights reserved.

eywords: Gas–solid reactions; Carboxylic acid; Hydantoin; Solvent-free nitration; Infrared microspectroscopy; Raman microspectroscopy

. Introduction

In recent years there has been a strong renewal of interestn the reactions of solid crystalline organic substrates[1–5].eactions of this type are desirable from an environmentaloint of view in that they obviate the need for large quan-

ities of solvent as the reaction medium[4,6]. At the sameime it has become increasingly recognised that such reac-ions give considerable stereochemical control, as the formf the reaction product may be directed by the arrangement ofeactant molecules within the crystal[2,3,7–11]. Performingeactions in the solid-state has been found in many exampleso increase yields and improve specificity. The explosion of

∗ Corresponding author. Fax: +44 1 1893 16332.E-mail address: m.j.almond@reading.ac.uk (M.J. Almond).

� Deceased.

interest incombinatorial chemistry where a wide range oreagents are tested on solid beads of substrates provfurther impetus to research in this area.

We have utilised the techniques of infrared and Ramicrospectroscopy to monitor reactions of this type ancharacterise reaction products[1–3,12]. The principal advantages of vibrational microspectroscopy in this context aresingle crystals may be monitored in situ and that the stra give considerable structural information about reacand products. Such structural information is much less rily forthcoming from an alternative approach, which has bused elsewhere[4,6,13,14]in which atomic force microscopis employed. Here morphological changes to the crystaseen upon reaction, but it is difficult to relate these changthe chemical processes ongoing within the crystal. Difftion methods have, of course, also been used[8,9,15] butare still undergoing development for use in time-reso

386-1425/$ – see front matter © 2005 Elsevier B.V. All rights reserved.oi: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 investigationdegrade 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�′-polymorphof ortho-ethoxy-trans-cinnamic acid in the solid-state[18].Until now we have focused our attention upon photochemicaldimerisation reactions of single crystals oftrans-cinnamicacid and its derivatives[1–3,12]. An integrated approachusing infrared and Raman microspectroscopy clearly demon-strates thetopotactic nature of these reactions where theform 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 thetime-dependency of the dimerisation of molecules of�-2,4-dichloro-trans-cinnamic acid within a single crystal[1]. Thereaction is shown to follow strictly first-order kinetics, ratherthan the zeroth-order kinetics predicted by some theoreticalmodels[1].

We have now turned our attention to the reactions of sin-gle crystalline organic substrates with a range of gaseousreagents. Such reactions are considered to be relatively envi-ronmentally friendly and often they give yields at least as highas those obtained by analogous solution reactions[4–6]. Theirfull exploitation, however, in synthesis is held back becauseof uncertainty as to how controllable the reactions are and ast rmedb s fors g then nismw opy.T Og

gasw er1 thats -m sw[ t. Byc e1s ubtlec cultt andp glec .W n1 hat 9-n od-u eds[ ud-i roducf rec yl orm atic

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 arenestoluene and chlorobenzene are better nitrated with the ternarymixture NO–NO2–O2 [5]. It is thought that nitrogen triox-ide is involved in this reaction. The second substrate whosenitration reaction we wished to monitor was hydantoin. Reac-tions of hydantoin (C3H4N2O4)—a derivative of urea—are ofinterest as derivatives of this compound find a wide range ofuses from their pharmaceutical role as anticonvulsants in thetreatment of epilepsy to their use as preservatives in inhibitingthe growth of moulds in shampoos and cosmetics. Moreover,with this substrate the reaction products are reported to varywith changes in temperature[4]. At room temperature a nitra-tion product is formed while as the temperature is raised thisgives way to an oxidation product containing three carbonylgroups (seeScheme 1).

It is thus apparent that there remain uncertainties in theinterpretation of these reactions and that, with the exceptionof a number of studies by single crystal X-ray diffraction[8,9,15,16]and by atomic force microscopy[4,6,13,14], fewinvestigations of single crystals have been made. Vibrationalmicrospectroscopy has not, to our knowledge, previouslybeen employed to investigate the reaction of any organic sin-gle crystal with a gaseous reagent. Given the current generalinterest in such reactions and our previous success in utilisingvibrational microspectroscopy to monitor photodimerisationr rtedh achc e par-t ts aref medb

2

d byA e as

o how the products of the reactions relate to products foy analogous reactions in solution. We selected two areatudy where we felt that there was uncertainty regardinature of the reaction products and/or the reaction mechahich could readily be addressed by vibrational microsche two areas were reactions with (i) ammonia and (ii) N2as.

Gas–solid reactions of carboxylic acids with ammoniaere first observed in 1884[19]; here interest lies in wheth:1 or 1:2 products are formed. In this case it appearsome acids, for example,p-nitrobenzoic acid, react with amonia gas to give a 1:1 salt RCOO−NH4

+ which then reactith more ammonia to form a 1:2 adduct RCOO−NH4

+·NH320]. Upon standing the 1:2 adduct reverts to the 1:1 salontrastp-chloro- andp-bromobenzoic acid, while giving th:1 salt, could not be induced to form a 1:2 adduct[21]. Thatuch a simple gas–solid reaction should in fact show somplexities lends weight to the argument that it is diffio predict the products of such reactions with certaintyrovides a rationale for investigating reactions within sinrystals. Even more complex are the reactions of NO2 gashen anthracene was first reacted with NO2 gas, more tha

00 years ago, Liebermann and Lindemann suggested titro-10-nitrito-9,10-dihydroanthracene is the primary prct [22] while almost contemporary experiments by Le

23] led to the isolation of only anthraquinone. Later stes have suggested that anthraquinone is a secondary pormed via nitrated intermediates[24]. It appears that a moontrolled reaction may be achieved if deactivating methethoxy groups are attached to the 9-position of the arom

t

eactions in single crystals we felt that the study repoere was very timely. We felt, moreover, that our approould usefully address specific questions regarding thesicular reactions such as whether 1:1 salts or 1:2 adducormed from ammonia and the nature of the products fory nitration.

. Experimental

Samples of all organic solids were used as supplieldrich. The stated purities of the reagents used wer

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 suppliedby Aldrich. Ammonia (99.96% purity) and nitrogen diox-ide (>99.5% purity) gases were supplied in lecture bottlesby Argo International and Aldrich, respectively. Isotopicallylabelled ND3 (99% isotopic purity) and15NO2 (99% isotopicpurity) were used as supplied by Aldrich and MSD, respec-tively.

Reactions of ammonia gas were performed in a sealedglass bulb in which the single crystal under investigation wasplaced. A pressure of 200 Torr of ammonia was added to thebulb on a vacuum line and the system was left for a period of 1month to react. Because of the reactive nature of NO2 a flow,rather than a static system was employed to limit the contacttime between the solid reagent and the NO2 gas. Crystals ofthe reagent under investigation were held in glass wool in aglass tube through which NO2 gas was flowed on a vacuumline. A pressure of 0.5 Torr of gas was maintained and wasmonitored using a Pirani gauge.

The infrared (Brucker IRscope II attached to BruckerEquinox 55 FT interferometer) microscope which was usedto record vibrational spectra of single crystals has beendescribed elsewhere[2,3,25]. Infrared spectra of bulk sam-ples were recorded using a Perkin-Elmer 1720-X FTIR, NMRspectra were measured on a Bruker DPX 250 MHz spectrom-eter at 250 Hz in DMSO-dand mass spectra were obtainedu

3

acid.A Torrp l-l risingf ,a -v mν ti-fi db[ t

spectrum. The latter feature overlapped withνsym.(OCO) ofthe carboxylate ion. In order to confirm which bands arisefrom the ammonium cation and, in particular, to observe theν4 band free from overlap with other features, the reactionwas repeated using ND3 in place of NH3. A broad band isnow seen centred at 2340 cm−1 together with a much sharperband at 1070 cm−1. These findings demonstrate that forma-tion of the ammonium salt occurs upon exposure of benzoicacid to ammonia gas. It appears that the reaction has goneto completion as theν(C O) carboxyl peak has decayedto extinction. The bands seen before and after exposure ofa single crystal of benzoic acid to NH3 gas are listed inTable 1, and the observed infrared spectra are illustrated inFig. 1.

We then turned our attention to the reaction oftrans-cinnamic acid with ammonia. We wished to explore thepossibility that the presence of a conjugated unsaturatedC C bond might influence the reaction and, moreover, wewished to utilise the position of theν(C C) band to followthe reaction. A single crystal oftrans-cinnamic acid wasexposed to 200 Torr pressure of ammonia gas for a period of1 month. The observed infrared spectra of the crystal beforeand after exposure are illustrated inFig. 2. The positions ofbands are listed inTable 2. The principal changes observedare as follows. First,ν(C O) at 1682 cm−1 decays in inten-sity, but does not decay to extinction. Secondly,ν(C C),aTc cta oftm sf ar mT intu hesefi salto iag etione thea

TP id befo th

A to

ν

ν

ν

ν

ν

δ

δ s, 1181711 vs

se of tecause

6sing a VG Autospec mass spectrometer.

. Reactions of organic substrates with ammonia gas

The first reagent selected for study was benzoicsingle crystal of benzoic acid was exposed to 200

ressure of NH3 gas for 1 month. During this time the foowing spectral changes were observed. The band arom ν(C O), initially at 1694 cm−1, decayed to extinctions would be expected as the carboxyl (CO2) group is conerted to a carboxylate (CO2−) group. Bands arising froasym.andνsym.(OCO) of the carboxylate group were idened in the product at 1580 and 1420 cm−1. Also, rather broaands of the ammonium cation from theν3 andν4 vibrations

26] were seen around 3100 and 1400 cm−1 in the produc

able 1ositions of infrared absorptions (cm−1) of a single crystal of benzoic ac

ssignment Benzoic acid (cm−1)

(N H(D)) (NH(D)4+)

(C H) unsaturated (aromatic) ∼3072 br(C O) (carboxyl) 1694 vs

asym.(OCO) carboxylate

sym.(OCO) carboxylate(N H(D)) (N(D)4

+)a

(C H), δ(C O) and ring vibrationsb 1494, 1454 vs, 1326 s, 12911129 s, 1073, 934 vs, 809 s,

a It is difficult to identify the position of this band with certainly becaub Bands in this spectral region are difficult to assign with certainty b

liphatic at 1632 cm−1 suffers a blue shift to 1640 cm−1.hirdly, bands arising fromνasym. and νsym.(OCO) of thearboxylate group (CO2−) may be identified in the produt ca. 1540 and 1400 cm−1. Fourthly, rather broad bands

he ammonium cation from theν3 and ν4 vibrations [26]ay be identified at ca. 3100 and 1400 cm−1. The band

rom νasym.andνsym.(OCO) and fromν4(NH4+) appear as

ather broad overlapping feature centred around 1450 c−1.he blue shift inν(C C) aliphatic may be understood

erms of loss of conjugation to the carboxyl (CO) grouppon formation of the ammonium salt. Taken together tndings demonstrate that formation of the ammoniumccurs upon exposure oftrans-cinnamic acid to ammonas, but that the reaction does not go entirely to complven after 1 month’s reaction time. In order to confirmssignments of bands of the product oftrans-cinnamic acid

re and after exposure to 200 Torr pressure of ammonia gas for 1 mon

Benzoic acid after exposure toammonia gas (cm−1)

Benzoic acid after exposureND3 gas (cm−1)

∼3100 br 2340 br∼3227 br 3060 br

– –1580 1590 s1420 sa 1410 s

∼1400 sha 1070s, 1181, 1134, 1103 s, 1073 s,

944, 814 s, 715 s, 666 s1174 w, 1100, 960 w, 834 s,720 s, 680

he presence of broad overlapping bands in this region.of coupling of vibrations and overlap of bands.

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 exposureto 200 Torr pressure of ND3 for 1 month (top).

exposed to ammonia the experiment was repeated using ND3in place of NH3. The bands assigned toν3 (ν(N H))(NH4

+)andν4 (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

+ cationin place of NH4

+ [26d].

The third reagent that we selected for study in its reac-tion with ammonia was adipic acid (HOOCH2CH2CH2CH2COOH), which contains two carboxylic acid groups. Theinterest in studying this reaction is to determine whether oneor both acid groups react and whether these groups reactsimultaneously or sequentially.

F below) 1 month.

ig. 2. Infrared spectra of a single crystal oftrans-cinnamic acid before ( and after (above) exposure to 200 Torr pressure of ammonia gas for

S.L. Jenkins et al. / Spectrochimica Acta Part A 62 (2005) 1131–1139 1135

Table 2Positions of infrared absorptions (cm−1) of a single crystal oftrans-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 afterexposure to NH3 gas (cm−1)

Trans-cinnamic acid afterexposure to ND3 gas (cm−1)

ν(N H(D)) (NH(D)4+) – ∼3100 br ∼2350 br

ν(C H) unsaturated (aromatic) ∼3025 br ∼3000 br ∼3000 brν(C H) unsaturated (aliphatic) – 2800 2810ν(C O) (carboxyl) 1682 1681 w 1685 wν(C C) (aliphatic) 1632 1640 1640νasym.(OCO) carboxylate – ∼1540 br ∼1540 brνsym.(OCO) carboxylate – ∼1400 bra ∼1400 brδ(N H(D)) (NH(D)4

+)a – ∼1400 bra 1065δ(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.

∼1450 br, 1290, 1250 s, 1080 s,1025, 978 vs, 920, 880 s, 846,771 s, 715 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.b Bands in this spectral region are difficult to assign with certainty because of coupling of vibrations and overlap of bands.

Infrared spectra show that adipic acid undergoes a reactionwith ammonia (200 Torr pressure of NH3; 1 month reactiontime). The band arising fromν(C O), at around 1700 cm−1,decays almost to extinction over the course of the reaction.A sharp, weak feature at 1716 cm−1 seen in the product maybe due to a small amount of unreacted starting material or itmay arise from an overtone or combination band of the prod-uct. It is thus seen that the reaction occurs at both carboxylgroups. No spectra, obtained on crystals after shorter reactiontimes, were observed where just one group has reacted, so itwould appear that the reaction at both groups is more-or-lesssimultaneous. Broad bands of the ammonium cation from theν3 andν4 vibrations[26] may be identified at ca. 3100 and1400 cm−1. Bands seen before and after exposure of a singlecrystal of adipic acid to NH3 gas are listed inTable 3.

4. Reactions with NO2 gas

Two crystalline substrates were selected for study in theirreactions with NO2 gas. These were: (i) 9-methylanthraceneand (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, wemodified our experimental procedure such that the crystals,s s (ca.

TP before

A

ν

ν

ν

ν

δ

δ 1045, 9 09 s,

se of tecause

0.5 Torr) in a flow system on a vacuum line for ca. 2 h. At theend of this time crystals were removed and were studied byinfrared microspectroscopy.

Infrared spectra of a crystal of 9-methylanthracene beforeand after exposure to NO2 gas are shown inFig. 3 and thepositions of infrared bands are listed inTable 4. Strikingchanges to the spectrum are observed. Water is clearly aby-product of the reaction, as evidenced by the growth ofintense broad bands centred around 3464 and 1663 cm−1.The presence of a nitro group in the product is shown byintense broad bands centred around 1550 and 1300 cm−1

which may be assigned toνasym. and νsym.(NO2) [26]. Astrong band at 828 cm−1 and a weaker feature at 625 cm−1,which are not seen for the starting material probably arisefrom δ(ONO) andν(C N) modes of the nitro group[26].To back up this assignment a spectrum of a single crystalof an authentic sample of 9-nitroanthracene was recorded.This shows strong bands centred around 1550, 1340, 840 and630 cm−1 corresponding to the bands seen above. The simi-larity in position between the bands of the nitration productand of 9-nitroanthracene suggest that the nitration productmay be 9-methyl-10-nitroanthracene. In order to confirmthe assignments of bands of nitrated 9-methylanthracene theexperiment was repeated using15NO2 in place of14NO2. Thewavenumber shifts of bands upon isotopic substitution aregiven inTable 5. The calculated isotopic shifts assume har-m on

upported in glass wool, were exposed to low pressure

able 3ositions of infrared absorptions (cm−1) of a single crystal of adipic acid

ssignment Adipic acid (cm−1)

(N H) (NH4+)

(C H) unsaturated (aromatic) ∼2966 br,(C O) (carboxyl) 1700 br,

asym.(OCO) carboxylate 1573(N H) (NH4

+)a

(C H)b 1478, 1300 br, 1196,904, 815 w, 732, 674

a It is difficult to identify the position of this band with certainly becaub Bands in this spectral region are difficult to assign with certainty b

and after exposure to 200 Torr pressure of ammonia gas for 1 month

Adipic acid after exposure to ammonia gas (cm−1)

∼3100 br∼2917 br,

1716

∼1400 bra

39, 1311 br, 1208, 1154 s, 1103, 1094, 1079, 1028, 9784 s, 648 s

he presence of broad overlapping bands in this region.of coupling of vibrations and overlap of bands.

onic diatomic N O or N C oscillators and are based up

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 systemfor 2 h.

reduced mass calculations (Table 5). Given the broadnessof the observed bands a highly satisfactory correspondencebetween observed and calculated values is found. Takentogether 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 sampleof the product was analysed by mass spectrometry. The

Table 4Positions 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-Methylanthraceneν (cm−1) 9-Methylanthracene exposed to NO2 ν

(cm−1)

ν(O H) (H2O) 3464 br, sν(C H) aromatic 3084, 3046 3053ν(C H) aliphatic 2934 s 2918δ(H OH) (H2O) 1663νasym.(NO2) 1550 s, brνsym.(NO2) 1297 brδ(O H), δ(C H), ν(C C), ν(C O)a 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.1601 sh, 1493 w, 1448, 1178, 1069 w,993 s, 894, 768, 746, 697, 648.

δ(NO2)b 828ν(C N)c 625

a Bands in this spectral region are difficult to assign with certainty because of coupling of vibrations and overlap of bands.b Assigned toδ(NO2) on basis of wavenumber shift upon substitution with15NO2.c Assigned toν(C N) on basis of wavenumber shift upon substitution with15NO2.

Table 5Observed 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 gas14NO2 observed(cm−1)

15NO2 observed(cm−1)

∆observed

(ν 14NO2 − ν 15NO2)∆calculated

(ν 14NO2 − ν15NO2)a

Robserved

(ν 14NO2/ν 15NO2)Rcalculated

(ν 14NO2/ν 15NO2)aAssignment

11

(cm−1) (cm−1)

550 1526 24 28300 1285 15 24828 819 9 15

625 616 9 10a Calculated values aremaximum isotopic shifts based upon harmonic diatom

1.016 1.018 νasym.(NO2)1.012 1.019 νsym.(NO2)1.011 1.018 δ(NO2)

1.015 1.016 ν(C N)

ic NO 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 toνasym.(NO2) and the band marked (*) in the samespectrum is assigned toνsym.(NO2).

observed principal peaks arem/z (EI): 236 (C15H10NO2),222 (C14H8NO2), 209 (C14H11NO), 195 (C14H11O). It wasexpected that 9-methylanthracene would be likely to nitrateat the 10-position. The methyl group has activating +I and+M effects; this makes the ring more susceptible to attack bythe electrophilic N atom of the NO2 molecule and the +Meffect makes the 10-position the most readily attacked. We,therefore, repeated the experiment using single crystals ofthe reagent 9-anthracene methanol. Here the methoxy groupat the 9-position has a−I, but a +M effect. As such thering should be a little less susceptible to nitration, but againattack at the 10-position should be favoured. After exposure

to NO2 gas a single crystal of the product shows new infraredabsorptions at 1553, 1300, 817 and 625 cm−1 which may beassigned toνasym.(NO2), νsym.(NO2), δ(NO2) andν(C N),respectively. Upon reaction with15NO2 these bands shift to1524, 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 beenreported depending on the reaction temperature. Single crys-tals of hydantoin were exposed to a flow of NO2 gas at apressure 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 6Positions 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 andat 65◦C

Assignment Hydantoinν (cm−1) Hydantoin exposed to NO2 at20◦C � (cm−1)

Hydantoin exposed to NO2 at65◦C � (cm−1)

ν(O H), free 3554 3457ν(N H), ν(O H) H bonded 3100 br 3100 br 3100 brν(C H) 2760 3004, 2763 s 2873Overtones/combinations 2612, 2473, 2388 s, 2292, 2078, 1981 s 2488 s, 2350, 2273, 2099,

2065, 2011, 19742370, 2355s

ν(C O) 1700 br 1700 br 1766, 1711 sνasym.(NO2) 1503 sνsym.(NO2) 1238 sν a s,

δ

ν

ecause

(C C), δ(C H) 1434, 1297, 1204 s, 1078 s, 992 s

(NO2) –(C N) –a Bands in this spectral region are difficult to assign with certainty b

1350, 1195 w, 988 vs, 926,882 s

1581, 1415, 1355, 1200, 11421064 s, 937 s, 850, 720, 680

758 –646 –

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 gasare illustrated inFig. 4 and the observed infrared bands arelisted inTable 6. It may be seen that two quite different reac-tion products have been formed under the different reactionconditions.

Upon reaction at 20◦C bands which may be assignedto a coordinated nitro group appear in the product. Thesebands are listed inTable 6; they occur at 1503, 1358, 756 and646 cm−1. Assuming that the identity of the product shownin Scheme 1is correct, these bands may be identified withνasym.(NO2), νsym.(NO2), δ(NO2) andν(C N) vibrations ofthe coordinated NO2 moiety. When the reaction was repeatedat 65◦C none of the above bands were seen in the product.Rather the appearance of the spectrum in theν(C O) regionchanged considerably in comparison to the starting materialspectrum. New bands are seen at 1766 and 1711 cm−1. Thespectral changes are in keeping with the product shown inScheme 1.

5. Discussion

Gas–solid reactions of organic single crystals have beenfollowed for the first time by the technique of infraredmicrospectroscopy. Our experiments have shown the generalapplicability of this approach to a study of such reactions.T verya turali . Att nals ctrao thatt e bulkp ac-t , e.g.n con-s ativeaf so

tw alts.I givea Hs eo doesn actsw ac-t itha ctiont lyh hichh ndly( lds.O sin-g onal

microspectroscopy is an excellent way to identify such prod-ucts. We have extended the work to a preliminary study of thereaction of hydantoin with NO2 gas where different productsare formed depending on the temperature of the reaction. Wehave shown that our approach allows these products to bereadily distinguished and allows structural information to beobtained. The spectra show quite conclusively the presenceof an NO2 group in the product of reaction at 20◦C and theabsence of such a group when the reaction is performed at65◦C.

An additional importance is given to the findings fromthis work by the known use of these reactions in organicsynthesis. Nitrated hydantoins are useful as their reductionproducts—the aminohydantoins—are important intermedi-ates in the preparation of several hydantoin pharmaceuticals[4].

Having carried out this preliminary survey of reactions toestablish the general applicability of our technique we willnow extend the work to look at specific aspects regardingthe time-dependency and temperature-dependency of prod-uct formation of specific reactions. This will involve in situkinetic monitoring of the reactions of single crystals usingthe approach that we have developed elsewhere for followingphotochemical reactions of single crystals oftrans-cinnamicacid derivatives[1].

A

inga stu-d

R

ins,4.

.J.oc.,

ec-

.2249.436.439.em.

[ . 36

[ . 36

[ J.

[ 993)

[

he relative simplicity of these techniques makes themttractive for monitoring such processes; moreover, struc

nformation regarding the products is readily obtainedhe same time it is straightforward to relate the vibratiopectra obtained from single crystals to vibrational spebtained from bulk samples. Thus, it may be confirmed

he product observed in a single crystal is the same as throduct of the reaction. This point is important when re

ions where many different products may be generateditration reactions, are considered. Taken together theseiderations give an advantage to our method over alternpproaches where diffraction methods[8,9,15,16]or atomic

orce microscopy[4,6,13,14]are utilised to follow reactionf this type.

Single crystals of benzoic andtrans-cinnamic acid reacith ammonia gas to give the corresponding ammonium s

n these cases it is shown that the reaction proceeds to1:1 salt and there is no sign of any 1:2 adduct of N3

uch has been seen for related systems[20]. The presencf a conjugated double bond system in cinnamic acidot appear to influence greatly the reaction. Adipic acid reith ammonia at both carboxylic acid groups. Nitration re

ions using NO2 gas are much more complex in nature wrange of products being observed depending on rea

imes and temperature[4,5,9,22,23]. This has unfortunateeld back the use of gas–solid reactions for nitration, wave advantages of being relatively environmentally friethey do not require solvents) and often giving high yieur experiments show that under controlled conditionsle isolable products may be formed and that vibrati

cknowledgement

We are grateful to the University of Reading for providstudentship for SLJ and to EPSRC for providing a

entship for SDMA.

eferences

[1] S.D.M. Atkinson, M.J. Almond, S.J. Hibble, P. Hollins, S.L. JenkM.J. Tobin, K.S. Wiltshire, Phys. Chem. Chem. Phys. 6 (2004)

[2] S.D.M. Atkinson, M.J. Almond, G.A. Bowmaker, M.G.B. Drew, EFeltham, P. Hollins, S.L. Jenkins, K.S. Wiltshire, J. Chem. SPerkin Trans. 2 (2002) 1533.

[3] S.D.M. Atkinson, M.J. Almond, P. Hollins, S.L. Jenkins, Sptrochim. 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)[7] S. Kanao, S. Kashino, M. Haisa, Acta Crystallogr. C46 (1990) 2[8] S. Kanao, S. Kashino, M. Haisa, Acta Crystallogr. C46 (1990) 2[9] V. Enkelmann, G. Wegner, K. Novak, K.B. Wagener, J. Am. Ch

Soc. 115 (1993) 10390.10] Y. Ito, B. Borecka, J. Trotter, J.R. Scheffer, Tetrahedron Lett

(1995) 6083.11] Y. Ito, B. Borecka, J. Trotter, J.R. Scheffer, Tetrahedron Lett

(1995) 6087.12] S.D.M. Allen, M.J. Almond, J.-L. Bruneel, A. Gilbert, P. Hollins,

Mascetti, Spectrochim. Acta Part A 56 (2000) 2423.13] G. Kaupp, J. Schmeyers, Angew. Chem. Int. Ed. Engl. 32 (1

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;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.

[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.[21] A. Korczynski, Chem. Zentralbl. 1908, 2009.

[22] C. Liebermann, L. Lindemann, Ber. Dtsch. Chem. Ges. 13 (1880)1584.

[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;(b) D.H. Williams, I. Fleming, Spectroscopic Methods in OrganicChemistry, fifth ed., McGraw Hill, Maidenhead, 1996;(c) N.B. Colthrup, L.H. Daly, R.B. Wiberly, Introduction to Infraredand Raman Spectroscopy, third ed., Academic Press, London, 1990;(d) K. Nakamoto, Infrared and Raman Spectra of Inorganic andCoordination Compounds, fourth ed., Wiley, New York, 1986.