J. MATER. CHEM., 1994, 4( l), 35-39 Dynamic Properties of the Urea Molecules in a,cu=Dibromoalkane/ Urea Inclusion Compounds Investigated by *H NMR Spectroscopy Abil E. AlievJ Sharon P. Smart and Kenneth D. M. Harris*t Department of Chemistry, University of St. Andrews, St. Andrews, Fife, UK KYI 6 9ST 2H NMR spectroscopy has been used to investigate the dynamic properties of the urea molecules in Br(CH,),Br/[2H,]- urea inclusion compounds (n=7-1 0) and in pure crystalline [*H,]urea. In the Br(CH,),Br/urea inclusion compounds, the urea molecules form an extensively hydrogen-bonded host structure, which contains linear, parallel tunnels within which the Br(CH,),Br guest molecules are located. The urea molecules undergo a two-site 180" jump motion about their C=O axes, on the 2H NMR timescale, at temperatures greater than ca.200 K for the Br(CH2),Br/[2H,]urea inclusion compounds and at temperatures greater than ca. 300 K for pure crystalline [2H,J~rea. Quantitative details relating to the dynamic properties of the urea molecules in these solids are presented. We are currently interested in the structural, dynamic and chemical properties of urea inclusion compounds containing a diverse range of organic 'guest' molecules. In these inclusion compounds,192 the urea molecules form an extensively hydro- gen-bonded 'host' structure containing parallel one-dimensional tunnels that are densely packed with guest mol- ecules. The internal diameter of the tunnels in the urea host structure is ca. 5.1-5.9 A, and 'guest' molecules based on a sufficiently long n-alkane chain can be accommodated within these tunnels, provided that the degree of substitution of this chain is small.Appropriate guest molecules include alkanes and derivatives such as a,o-dihalogenoalkanes, diacyl per- oxides, carboxylic acids and carboxylic acid anhydrides. We have shown by X-ray diffraction3y4 and other techniques that urea inclusion compounds containing functionalized alkane guest molecules exhibit interesting structural proper- ties, particularly concerning the three-dimensional packing arrangement of the guest molecules. In particular, it has been shown4 by single-crystal X-ray diffraction that, at room temperature, a,o-dibromoalkane guest molecules [Br( CH2),Br with n =7-10] exhibit a characteristic three- dimensionally ordered packing arrangement in which As = cg/3, where cg denotes the periodic repeat distance of the guest molecules along the tunnel and Ag denotes the offset, along the tunnel axis, between the positions of guest molecules in adjacent tunnels. This guest structure is rhombohedral, and a given single crystal of the inclusion compound usually contains two domains of this guest structure, differing in orientation with respect to the host structure.Furthermore, the Br (CH,),Br/urea inclusion compounds also contain regions in which the guest molecules are ordered only along the tunnel axis; the periodic repeat distance, cg, is the same (within experimental error) for the one-dimensionally ordered regions and the three-dimensionally ordered regions discussed above.In contrast to this situation for the Br(CH,),Br guests, the molecular packing arrangement in the three-dimensionally ordered regions of the guest structure in alkanelurea inclusion compounds has A, =0. It is interesting and important to speculate whether the presence of terminal bromine atoms as opposed to methyl groups on the guest molecule gives rise to a similarly marked difference in the dynamic properties of the guest molecules. Similarly, it is possible that the presence of different functional groups on the guest molecule could exert an important t Present address: Department of Chemistry, University College London, 20 Gordon Street, London, UK WClH OAJ.influence on the dynamic properties of the urea molecules within the host structure. It is known5 that alkane/urea inclusion compounds undergo a phase transition from a low-temperature phase in which the host tunnel structure is orthorhombic to a high-temperature phase in which the host structure is hexagonal. This phase transition is also believed to be associated with an abrupt change in the dynamic properties of the alkane molecules, and it has been shown that these molecules undergo considerable motion in the high-temperature phase. Differential scanning calorimetry has shown that the Br(CH,),Br/urea inclusion compounds undergo a phase trans- ition similar to the well established phase transition for the alkane/urea inclusion compounds.Powder X-ray diffraction' has shown that this transition is associated with the same distortion of the host tunnel structure as that established previo~sly~,~for the alkane/urea inclusion compounds, Although many studies, by NMR and other techniques, have been carried out to probe the motion of the guest molecules in urea inclusion compounds, little attention has been devoted to studies of the dynamic properties of the urea molecules. Recently, Heaton et al. have studied the dynamic properties of the urea molecules in the nonadecane/C2H4]- urea inclusion compound by powder' and single-crystal" ,H NMR spectroscopy. This work led to the proposal that the urea molecules undergo 180"jumps about their C=O axes, with no evidence (on the 2H NMR timescale) for rotation of the NH, groups about the C-N bonds (although rapid librational motion about the C-N bonds cannot be ruled out).It is possible that the exact nature of the guest molecules (and particularly the presence of different types of functional group on the guest molecules) could have a significant bearing upon the urea jump motion. In pure crystalline urea, 180" jumps of the urea molecule about its C=O axis are also believed to above ambient temperature, and it has been proposed that simultaneous rotation about the C-N bond may also occur.11,12,15 In this paper we report ,H NMR investigations of the dynamic properties of the urea molecules in Br(CH,),Br/[2H4] urea inclusion compounds (n =7-10) and in pure crystalline C2H4]urea.Experimental Inclusion compounds containing Br (CH,),Br guest molecules (n= 7-10) in L2H4]urea were prepared by slowly cooling warm solutions of Br(CH,),Br and C2H4]urea in CH,OD. The degree of deuteriation of the urea molecules was shown by infrared spectroscopy to be greater than ca. 98%. Powder X-ray diffraction confirmed that all the crystals had the characteristic tunnel host structure of the conventional urea inclusion compounds. The phase-transition temperatures for these inclusion compounds, established by differential scan- ning calorimetry, are in the range ca. 145-170 K. 2H NMR spectra were recorded at 76.78 MHz on a Bruker MSLSOO spectrometer, using a standard Bruker 5 rnm high- power probe.The stability and accuracy of the temperature controller (Bruker B-VT1000) were ca. & 2 K. Temperature calibration of the probe was established via 2H NMR studies of the melting transitions in CD,OD (175 K) and D20 (277K). 2H NMR spectra were recorded using the conven- tional quadrupole echo [( 90"),-z-( 90"),-z-acquire-recycle] pulse sequence,16 with 2H 90" pulse duration=2.4 ys, echo delay z =13 ps, and recycle delay =60 s. Typically, 2048 data points were accumulated, with a dwell time of 0.5 ys. The 2H NMR spectra have not been artificially symmetrized. 2H NMR spectra were recorded for all Br(CH2),Br/[2H,]- urea inclusion compounds (n=7-10) at 293 K, and for Br(CH2),Br/C2H,] urea at several temperatures between 140 K and 293 K.2H NMR spectra were recorded for pure crystalline ['H,]urea at 293, 333 and 358 K. Simulations of 2H quadrupole echo NMR spectra were obtained using the program MXQET.17 The spectral simu- lations were obtained by Fourier transformation of calculated echo decays, with both Gaussian and Lorentzian apodization applied before Fourier transformation. The calculations using the MXQET program include effects arising from the virtual free induction decay and from the echo, and corrections for imperfect spectral coverage (due to finite pulse power) are also considered. The program also takes into consideration effects on the spectrum which can ariseI8 when exchange processes occur on a timescale comparable to z. Results and Discussion In ['H,] urea" and in r2H4] urea inclusion compounds'q2 there are two crystallographically inequivalent deuteron environments which are denoted here as 'axial' and 'equatorial' deuterons; the N-D bond for the axial deuterons forms an angle P~l80" with the axis of the C=O bond, whereas the N-D bond for the equatorial deuterons forms an angle P~z60' with the axis of the C=O bond (Fig.1). In principle, the axial and equatorial deuterons could have different values of the static quadrupole coupling constant and the static asymmetry parameter (and perhaps also different values of isotropic chemical shift). Before undertaking studies of the dynamic properties of the urea molecules in the B~(CH,),BI-/[~H,] urea inclusion com- pounds, 2H NMR studies of pure crystalline C2H,]urea were carried out in order to corroborate the known dynamic properties of this material.2H NMR spectra of pure crystalline C2H,]urea recorded at 293, 333 and 358 K are shown in Fig. 2(a), (b) and (c), respectively. The spectrum recorded at 293 K was fitted well by a spectrum [Fig. 2(d)] simulated assuming no motion of the deuterons, and with static quadru- 0 Fig. 1 Molecular structure of the ['HJ urea molecule, indicating the designation of the two different types of deuteron as 'axial' and 'equatorial' J. MATER. CHEM., 1994, VOL. 4 pole coupling constant x =212 kHz and static asymmetry parameter q=0.15. We conclude from this that the static quadrupole interaction parameters do not differ significantly for the two crystallographically distinguishable deuteron environments. Thus, in the spectral simulations carried out in this work, it was assumed that all deuterons in the [2H4]urea molecule have the same values of the static quadrupole interaction parameters.These values of x and q are in close agreement with those determined previously12 from single-crystal 2H NMR studies of C2H4]urea at room temperature (for the two crystallograph- ically distinguishable deuterons, the following values were obtained: x=210.8_+ 1.0 kHz and q=0.139_+0.010; x= 210.7_+ 1.0 kHz and q =0.146 +O.OlO). Similar values were also obtained previously' from 2H NMR lineshape analysis for a polycrystalline sample of C2H4]urea at 303 K (x= 212 f2 kHz and y =0.145 k0.005). 2H NMR spectra of C2H4]urea recorded at 333 and 358 K [Fig.2(b) and (c)] are clearly not static powder patterns, although the total widths of these spectra are only slightly less (ca. 2%) than that of the spectrum recorded at 293 K. The 'inner' powder pattern with intensity maxima at f. 14 kHz and the shoulders at k 99 kHz confirm the existence of mobile deuterons at 333 and 358 K. The existence of two crystallo- graphically distinguishable deuteron environments, and the spectral features observed at 333 and 358 K, suggest that the lineshapes measured at higher temperatures can be interpreted as a superposition of two powder patterns (one of which is similar in appearance to the powder pattern of a static system). It should be noted that at all temperatures, a single peak at zero frequency is also present, and its intensity increases gradually as temperature is increased.This is probably due to ['H,] urea molecules undergoing isotropic motion (e.g. in the gas phase). The changes in lineshape with temperature for C2H4] urea are characteristic of the changes in lineshape for a two-site 180" jump motion about an axis forming an angle of 60" with respect to the direction of the z axis (principal component) of the electric-field gradient tensor of the deuteron. Analogous lineshapes (with ca. 15-30 kHz separation between the 'inner' intensity maxima) have been reported for such motions in nonadecane/C2H,] urea,9 1,4-disubstituted benzenes'* and other systems.20 Based on this interpretation, the 2H NMR lineshapes at 333 and 358 K have been simulated successfully on the basis of a dynamic model consisting of a two-site 180' jump motion about the C=O axis of the urea molecule.Although both types of deuteron necessarily undergo jumps at the same rate, the N-D,, vector is almost parallel to the C=O vector (the jump axis), and the D,, deuterons do not significantly change their orientation with respect to the applied magnetic field during the jump motion. Thus, their contribution to the spectrum is almost indistinguishable from that of static deuterons. In contrast, the N-D,, bond forms an angle of ca. 60" with the C=O vector, and the orientation of the D,, deuterons relative to the applied magnetic field changes appreciably during the motion.Thus, in the rapid motion regime, the spectrum due to the D,, deuterons is very similar to a static 2H NMR powder pattern, whereas the spectrum due to the D,, deuterons is substantially motionally averaged. The best-fit spectral simulation at each temperature is shown in Fig. 2(e) and (f)together with the value of the jump frequency (K). In order to achieve good agreement between simulated spectra and the experimental spectra recorded at 333 and 358 K, it was necessary to use slightly smaller values of the static quadrupole coupling constant x and the static asymmetry parameter q than those determined from the 'static' spectrum recorded at 293 K. The best-fit values of x and q J.MATER. CHEM., 1994, VOL. 4 I"""""""""'I''""'"'"""'''~ 200 0 -200 kHz Fig. 2 2H NMR spectra recorded for pure crystalline C2H4]urea at: (a) 293; (b) 333; (c) 358 K. The best-fit simulated 2H NMR spectra corresponding to the experimental spectra (a)-(c) are shown in (d)-(f)respectively. The simulated spectrum (d)was calculated assuming no motion of the deuterons, whereas (e)and (f)were calculated assuming the two-site 180" jump motion discussed in the text. The following parameters were used in these spectral simulations: (d) static quadrupole coupling constant x =212 kHz, static asymmetry parameter rj =0.15; (e)x=209 kHz, q=O.13, jump frequency IC=1.5 x lo5s-'; (f)x=208 kHz, q=O.13, IC=~x lo6 s-'. The agreement between the simulated and experimental spectra is noticeably degraded if x is changed by more than k1 kHz or if q is changed by more than kO.01 were found to be x=209 kHz, q=O.13 at 333 K and x= 208 kHz, q=O.13 at 358 K.An analogous decrease in the value of 2was found in the case of spectral simulations of a two-site 180" jump of the ND2 deuterons of p-nitroaniline,20 and it was proposed that this behaviour arises from an increase in the amplitude of restricted, rapid libration of the N-D bonds about their equilibrium orientations. It is also possible that the exact nature of the hydrogen bonding in these systems may have a critical influence on the temperature- dependence of x and q, particularly when the motion is a large-angle jump process involving breakage and formation of hydrogen bonds.The frequency separation between the 'inner' intensity maxima in the spectral simulations depends critically upon the angle p for the equatorial deuterons, and best-fit simu- lations for the experimental spectra recorded at 333 and 358 K were obtained with /3 =60.5f0.3". The corresponding angle determined from neutron diffraction studies" of crystalline C2H4]urea is 61.2'. In our spectral simulations, the angle /3 for the axial deuterons was fixed at 177', based on the value ( 176.6') obtained from these neutron diffraction studies. Additional spectral simulations showed that variation of /3 for the axial deuterons by as much as f5' gives no perceptible I change in the simulated 2H NMR lineshape. 2H NMR spectra recorded at 293K for the Br(CH,),Br/ C2H4] urea inclusion compounds with n=7-10 are shown in Fig.3. The temperature-dependence of the 2H NMR spectrum for Br(CH2)7Br/[2H4]urea is shown in Fig. 4. As for pure crystalline C2H4]urea, the 2H NMR spectra of the Br(CH,),Br/C2H4] urea inclusion compounds are consistent with the existence of two crystallographically distinguishable deuterons D,, and D,, in the host structure in the inclusion compound. In contrast to pure crystalline C2H4] urea (Fig. 2), however, the 'inner' powder patterns in the spectra of the Br(CH2),Br/C2H4] urea inclusion compounds recorded at 293 K (Fig. 3) have intensity maxima at f8 kHz and should- ers at +95 kHz. Fig. 3 'H NMR spectra recorded for different Br(CH,),Br/[2H4]urea To prove the existence of the two crystallographically inclusion compounds at 293 K: (a)n=7; (b)n=8; (c)n=9; (d)]'I=10 A I""""'""~""'~"""'""'"""'~ 200 0 -200 kHz Fig.4 'H NMR spectra recorded for the Br(CH,),Br/['H,]urea inclusion compound at: (a) 140; (b) 160; (c) 200; (d)240; (e)293 K distinguishable deuteron environments, 2H NMR spectra were recorded for the Br(CH,),Br/C2H4] urea inclusion compound with very short recycle delay (10ms) at 240 and 293 K (Fig. 5). With this very short recycle delay, there is a substantial decrease in the intensity of the powder pattern assigned to the axial deuterons (as a consequence of their longer spin- lattice relaxation times) relative to the intensity of the powder pattern assigned to the equatorial deuterons.Qualitative features of the 2H NMR spectra recorded at 293 K are identical for all of the Br(CH,),Br/[2H,]urea inclusion compounds studied, and we concentrate our dis- cussion on the spectra for Br(CH2),Br/['H4] urea. The spectra recorded (Fig. 4) between 140 and 200 K are characteristic of 'static' ,H NMR powder patterns; there is no discernible change in lineshape upon increasing the temperature in this range, even though the phase transition occurs within this range. The spectra recorded at 240 and 293 K can be con- sidered in terms of a superposition of two powder patterns: an 'inner' powder pattern (assigned to the D,, deuterons) and an 'outer' powder pattern (assigned to the D,, deuterons). A single peak at zero frequency is also present, and is probably due to [2H4] urea molecules undergoing isotropic motion (e.g.in the gas phase). The spectra recorded at 140 and 160 K are fitted well by a spectrum [Fig. 6(a)] simulated assuming no motion of the deuterons, and with static quadrupole coupling constant x =212 kHz and static asymmetry parameter q = J. MATER. CHEM., 1994, VOL. 4 ,I,,, I",'/ " I 1 200 100 0 -100 -200 kHz Fig. 5 'H NMR spectra recorded for the Br(CH,),Br/['H,]urea inclusion compound with a short recycle delay (10 ms) at: (a) 240; (b)293 K Fig. 6 Simulated 'H NMR spectra calculated (a) assuming no motion of the deuterons, and (b)-(d)assuming the two-site 180" jump motion discussed in the text. The following parameters were used in these spectral simulations: (a) static quadrupole coupling constant x = 212 kHz, static asymmetry parameter q =0.18 [compare with Fig.4(a)]; (b) x=207 kHz, q=0.17; jump frequency ~=2 xlo5s-' [compare with Fig. 4(d)]; (c) x=207 kHz, q=0.17; K= 1.5 xlo6s-' [compare with Fig. 3(a)]; (d) x=207 kHz; q=O.17; ~=4xlo6s-l [compare with Figs. 3(b), (c) and (d)].The agreement between the simulated and experimental spectra is noticeably degraded if x is changed by more than k1 kHz or if q is changed by more than +O.Ol J. MATER. CHEM., 1994, VOL. 4 0.18. These values of x and y are in close agreement with those (3: =208 k3 kHz and rj =0.155k0.005) determined pre- viouslyg from lineshape analysis of motionally averaged 2H NMR powder patterns recorded for the nonadecane/C2H4]- urea inclusion compound at 258-338 K. As in the case of pure crystalline C2H4]urea, the 2H NMR spectra recorded at 293 K (Fig.3) for the Br(CH,),Br/ C2H4] urea inclusion compounds have been simulated success- fully (Fig. 6) on the basis of a dynamic model consisting of a two-site 180" jump motion about the C=O axis of the urea molecule. The best-fit values of x and rj at 293 K were found to be x=207 kHz and y=O.l7. The best-fit of the frequency separation for the 'inner' powder pattern in the spectral simulations was obtained with the angle p for the equatorial deuterons in the range 59.5f0.3". As in our studies of pure crystalline ['H,]urea, the angle fl for the axial deuterons was fixed at 177".The spectra recorded at 293 K for the C2H4]urea inclusion compounds with Br(CH,),Br, Br(CH2)9Br and Br(CH,),,Br guest molecules were fitted well by a spectrum [Fig. 6(d)] simulated using jump frequency ic =4 x lo6s-'. For the Br( CH,),BI-/[~H~] urea inclusion compound, the best-fit value of K is 1.5x lo6 s-' at 293 K [Fig. 6(c)] and 2 x lo5 s-' at 240 K [Fig. 6(b)]. Concluding Remarks If the comparatively bulky bromine atoms on the Br(CH,),Br guest molecules hinder the jump motion of the urea molecules, the dynamics of the urea molecules should be expected to depend on the length (n) of the guest molecule. For shorter Br(CH,),Br guests, there is a higher 'density' of bromine atoms per unit length of tunnel, and the motion of a higher proportion of the urea molecules (i.e.those in the vicinity of bromine atoms) might be expected to be hindered. It is interesting that the jump frequency (K) is lower (at the same temperature) for Br(CH,),Br/['H,] urea than for the Br(CH2),Br/['H4] urea inclusion compounds with longer guest molecules. In the knowledge that the host structure does not differ significantly4 (within experimental error) between the different Br(CH,),Br/['H,] urea inclusion com- pounds investigated here, the fact that there is a measurable difference in the jump frequency of the urea molecules in Br(CH2),Br/[2H4]urea suggests that the nature of the guest molecules (e.g. the 'density' of bromine atoms along the tunnel) does indeed exert some influence upon the dynamic properties of the host.However, in view of the fact that the motions of the Br(CH,),Br guest molecules [for r1=8-10]~'*~~occur on a timescale that is several orders of magnitude shorter than the jump motion of the urea molecules established here, it is clear that there is no direct correlation between the dynamic properties of the host and guest molecules. Note that the dynamic properties of the guest molecules in Br(CH2),Br/[2H,] urea have not yet been investigated. Finally, it is interesting to recall that there is an incommen- surate structural relation~hip'?~ between the host and guest substructures in the Br(CH,),Br/urea inclusion compounds with n=7-10, and this may be expected to give rise to a distribution of K values for the jump motion of the urea molecules (a lower value of K may be expected for urea molecules in the vicinity of a bromine atom of the guest molecule).However, interpretation of the 2H NMR spectra reported in this paper did not require a distribution of ic values to be invoked. It is known21*22 that translational motions of the guest molecules along the tunnel occur on a timescale that is several orders of magnitude shorter than the timescale for the motion of the urea molecules, and it is possible that this translational motion of the guest molecules leads to each urea molecule experiencing the same average jump frequency over the timescale of the 2H NMR measure- ment. This provides a possible explanation for our ability to interpret the 2H NMR spectra of the Br(CH2),Br/C2H4] urea inclusion compounds on the basis of a single value of ic, even although each urea molecule has a different instantaneous environment as a consequence of the incommensurate relationship between host and guest substructures.We are grateful to the SERC (studentship to S.P.S.and general support to K.D.M.H.) and the Royal Society (postdoctoral fellowship to A.E.A.) for financial support. References 1 A. E. Smith, Acta Crystallogr., 1952,5,224. 2 K. D. M. Harris and J. M. Thomas, J. Chem. SOC.,Faradaj Trans., 1990,86,2985. 3 K. D. M. Harris and M. D. Hollingsworth, Proc. Roy. SOC. London, Ser. A, 1990,431,245. 4 K. D. M. Harris, S. P. Smart and M. D. Hollingsworth, J. Chem.SOC.,Faraday Trans., 1991,87,3423. 5 N. G. Parsonage and R. C. Pemberton, Trans. Faraday So,.., 1967, 63,311, and earlier references cited therein. 6 Y. Chatani, H. Anraku, and Y. Taki, Mol. Cryst. Liq. Cryst., 1978, 48,219. 7 K. D. M. Harris, I. Gameson and J. M. Thomas, J. Chern. SOC., Furaduy Trans., 1990,86,3135. 8 I. J. Shannon and K. D. M. Harris, in preparation. 9 N. J. Heaton, R. L. Vold and R. R. Vold, J. Am. Chern. Sol ., 1989, 111,3211. 10 N. J. Heaton, R. L. Vold and R. R. Vold, J. Mugn. Reson., 1989, 84, 333. 11 J. W. Emsley and J. A. S. Smith, Trans. Faraday SOC.,1961, 57, 1233. 12 T. Chiba, Bull. Chem. SOC.Jpn., 1965,38,259. 13 A. Zussman, J. Chem. Phys., 1973,58,1514. 14 H. H. Mantsch, H. Saito and I. C. P. Smith, Prog. NMR Spectrosc., 1977, 11,211. 15 T. P. Das, J. Chem. Phys., 1957, 27, 763; see also J. Chem. Phys., 1961,35,1897. 16 J. H. Davis, K. R. Jeffrey, M. Bloom, M. I. Valic and T. P. Higgs, Chem. Phys. Lett., 1976,42, 390. 17 M. S. Greenfield, A. D. Ronemus, R. L. Vold, R. R Vold, P. D. Ellis and T. R. Raidy, J. Magn. Reson., 1987,72, 89. 18 T. M. Barbara, M. S. Greenfield, R. L. Vold and R. R. Vold, J.Mugn. Reson., 1986,69, 311. 19 J. E. Worsham, H. A. Levy and S. W. Peterson, Acta Crysfullogr., 1957, 10, 319. 20 M. A. Kennedy, R. R. Vold and R. L. Vold, J. Magn. Reson, 1991, 91, 301. 21 S. P. Smart, F. Guillaume, K. D. M. Harris, C. Sourisseau and A. J. Dianoux, Physica B, 1992,180& 181,687. 22 F. Guillaume, S. P. Smart, K. D. M. Harris and A. J. Dianoux, J. Phys., Condens. Matter, in the press. Paper 3/04273A; Received 20th Julj, 1993