333J. CHEM. SOC. FARADAY TRANS.. 1994. m21. 333-33s Photochromism, Thermochromism and Solvatochromism of Some Spiro[indolinoxazine]-photomerocyanine Systems:Effects of Structure and Solvent G. Favaro,* F. Masetti, U. Mazzucato and G. Ottavi Dipartimento di Chimica, Universita di Perugia . 06122 Perugia , Italy P. Allegrini and V. Malatesta" EniChem Synthesis, 20097 San Donato Milanese Itat./~ Three spiro[indoline-naphthoxazines] and a spiro[ind~iine-phenanthroxazine], which exhibit photochromic and thermochromic properties, have been investigated. So vent and structure effects on the absorption spectra of the merocyanines produced under UV irradiation and kine:ic parameters for the ring-closure and ring-opening reac- tions were studied. Positive solvatochromism was fcund, indicating that the opened form is a weakly polar species.Equilibrium constants and rate constants for the forward and back reactions spiroxazine emerocya-nine increase with increasing the solvent polarity and with electron-donating groups in the oxazine moiety. The reaction is endothermic by 10-20 kJ mol-' and almost isoentropic. The activation entropy is generally negative, while the activation Gibbs energy is approximately independent of solvent and structure. Photochromism involving changes in the visible absorption pi wnd. I ,3,3-trimethylspiro[indoline-2.3'[3H]naphth[2,1 -h] spectrum has attracted much attention in the last decades [I .J]oxazine] 1, the 6'-piperidine substituted compound 2. because of the variety of practical applications of photo- the 5-Br compound 3 and the unsubstituted spiro[indoline-2. chromic systems.' The photochromic properties of the spir o-2' [?H]phenanthr[9,10-b][ 1,4)oxazine) 4. A11 these molecules compounds are well known. Spiropyrans have been supplied by EniChem Research Sp.4. The solvents were wIw extensively studied2 and, more recently. spiroxazines hac t' re igent grade Carlo Erba products. been the subject of many investigation^.^ Is Interest in these compounds is justified by their high durability with respect to photoexci tation. The photochromism of these molecules is due to photo- cleavage of the spirobond under UV irradiation to give 'in open merocyanine structure (photomerocyanine) which absorbs in the visible region.'YGeneral agreement can be found in the literature about W some aspects of the mechanistic behaviour of these photo- 4chromic systems. It is well known, from experiments uirh 1 2 3 SH H Brpicosecond time resolution, that the C-0 bond breakage in Y H piperidine H the excited state occurs on the picosecond timescale.-' Experimental evidence has been reported for the production of several (at least two') merocyanine isomers in a transoid Equipment structure.' Thermal bleaching of the coloured form is knoun Almrption spectra were recorded on Perkin-Elmer Lambda to be a relatively slow process (rate constant: 0.01 1Cl 5 ind Lambda 16 spectrophotometers. For absorption mea-1 1.'L.5.13 sus'ements at varying temperatures. a cryostat (Oxford However, there are some doubts about the nature of the In~rument)was used, equipped with a temperature control- primary photoproduct 'X', whether it is a non-planar cisoid ler operating between 77 K (if liquid nitrogen was used for structure,-or a transoid form undergoing very fast thermd co ,ling) and 500 K.A 250 W medium-pressure mercury lamp equilibration.' Conflicting results have been reported on the fihered by a CS 7-54Corning filter.which transmits 240-400 quantum efficiency of the photocolouration reaction, which nn light. was used for producing the coloured form in some cases differs, for the same molecule, by more than 100°/0.4.5.'3Large discrepancies can also be found in the MIdar Absorption Coefficientsmolar absorption coefficients of the open forms determined by different method^.^.' 3*1J Thr determination of the molar absorption coefficients of the In this paper we report the effect of the solvent and strub- op:n forms was carried out at 223 K in order to minimize the ture on the thermal equilibrium between coloured and efihct of thermal bleaching. The room-temperature concentra- colourless forms, on the kinetics of thermal bleaching of the tio 1s of the solutions ( lo-' mol dm-3) were corrected photomerocyanines and on their absorption properties.for the volume contraction upon cooling. In toluene. owing These measurements. which give an overall view of the ener- to the sharp decrease of solubility of the spiro compounds getics of these systems, can also give mechanistic information with increasing temperature.a further correction was neces- about the thermal breaking and reforming of the spirobond. sar v to account for partial solute precipitation, even when the corcentration was kept below 5 x mol dm- '. The solu- Experimental tio IS were irradiated for 10 min. sufficient to obtain a con- Materials sta it absorption of the coloured form. In order to avoid The photochromic molecules under study were three diflusion effects. the whole surface of the sample cell was spiro[indoline-naphthoxazines] : the unsubstit uted com-hoiwgeneously irradiated. Equilibrium Constants The constants of the thermal equilibrium K = [merocy-anine]/[spiroxazine], were determined in ethanol and toluene by measuring the visible absorbance of the open form in cor- respondence to the maximum absorption (570-615 nm) where the spiro form does not absorb.The determinations were carried out in the temperature range 280-335 K, using sample concentrations on the order of lo-' mol dm-'. The accuracy can be considered within 20%. Kinetics of Thermal Bleaching The kinetics of ring-closure reaction were studied following the disappearance of the coloured form at the wavelength of maximum absorbance. The solutions (ca. lo-' mol dm-3) were previously irradiated with the filtered light of the mercury lamp for 5 min and analysed immediately. The mea- surements were performed in the temperature range 266-300 K, ca. 30 min after having set the temperature control in order to allow the solution to reach thermal equilibrium. First-order rate constants were obtained from linear log A us.time plots. Results and Discussion Absorption Spectra The absorption spectra of the colourless and coloured forms of the four molecules in ethanol are illustrated in Fig. 1. 1.2 1 0.6 ,,, , 0.0 0.2 Q) C $n 0.0 (D 1.O - , I ' I , , 0.5 0.0 4 0.2 0.0250 350 450 550 650 750 L/nm Fig. 1 Absorption spectra of the colourless (--) and coloured (---) forms of the 1,2,3and 4 in ethanol at 223 K J. CHEM. SOC. FARADAY TRANS., 1994, VOL. 90 Table 1 Absorption characteristics of the spiro[indolinoxazines] in ethanol solution indoline moiety oxazine moiety A,,/nm Emax /dm3 mol-' an-' Amax /nm Emax /dm3 mol-' cm-' 1 233 36000 318 4400 2 249 24000 363 loo00 3 231 50900 310 8OOO 4 252 43600 340 6OOo The absorption spectra of the colourless forms consist of localised n-transitions in the UV region, belonging to the two orthogonal halves of the molecule.The band maxima and molar absorption coefficients of the indoline moiety and those of the naphthoxazine or phenanthroxazine moieties are reported in Table 1. The effect of the electron-donating sub- stituent and electron-withdrawing substituent on the naphtho-derivatives is to shift the absorption band to lower and higher energies, respectively. For 4, the phenanthrenic structure was observed. The spectra of the photomerocyanines, which were produc- ed at 223 K in order to avoid thermal bleaching, are charac- terized by an intense band in the visible region (A,,, rz 600 nm); thus the solutions are deeply coloured.The spectral characteristics of the coloured forms in two solvents are given and compared with literature data in Table 2. The agreement is not completely satisfactory for both A,,, and molar absorption coefficients. This is particularly evident in the case of 1 for which E values in ethanol range from 51 x lo3 dm3 mol-' cm-' '' to 73 x lo3 dm3 mol-' ~m-'.~The ques- tion of the poor correspondence between molar absorption coefficients determined in different laboratories has been raised by Wilkinson et ~1.'~in a recent paper which appeared when this work was already in progress.In that paper a method is described for measuring the molar absorption coef- ficient of photochromic compounds which seems completely convincing since it requires a minimum of assumptions. However, the E values determined are generally smaller than those previously reported (as well as those obtained in this work). These discrepancies are underlined by Wilkinson but no reason is proposed for them. Our determinations were carried out under conditions similar to those used by Chu4 and Kh~lmanskii,'~ that is at a temperature low enough to inhibit thermal bleaching. Thus, possible sources of error could arise from photochemical ring-closure or spectral changes which occur upon cooling. The first source can be excluded for several reasons.First, the quantum efficiency of the back reaction, Ob,was found to be Table 2 Absorption characteristics of the photomerocyanines (223K) ethanol toluene Amax Emax Amax Lax /nm /dm3 mol-' cm-' /nm /dm3 mol- ' cm-1 614 73600 601 58700 (612)" (73000)" (594)b (31000)c (607)b (51000)' (592)' (52000)' (594)e 2 593 77000 574 62400 (5W (32000)'(565)'3 615 68100 600 57000 4600 63300 588 68300 (592)J From ref. 4; from ref. 5; from ref. 13; 'from ref. 14; from ref. 15; from ref. 6. J. CHEM. SOC. FARADAY TRANS.. 1994. VOL. 90 very small compared with that of the forward reaction. a+. (U+ + Ob2 Of);'3 secondly, the absorbance of the coloured form in the exciting wavelength range is also very small: finally, not taking photobleaching into account should led to smaller, rather than higher, E values.Thus, we believe that the main reason for the discrepancies lies in some spectral evolution occurring upon cooling. The best agreement. in fact, is found with Chu's data,4 not only in the E value. but also in the spectral position of the absorption maximum of 1. which is shifted to lower energies with respect to that mea-sured at room temperature. It is possible that the tern-perature difference (ca. 70 K) between different kinds of measurements is critical to establish equilibrium among dif- ferent merocyanine isomers. To verify this hypothesis, absorption spectra of merocya- nines (complete conversion) were recorded over a range of temperatures (165-225 K) low enough to ensure exclusion of thermal bleaching.As an example, the behaviour of 2 in ethanol is shown in Fig. 2. It can be observed that lowering the temperature enhances the visible absorption intensit) without appreciably affecting the UV region. A slight, but sig- nificant, red shift of the coloured band was also observed. This behaviour is qualitatively similar for all the molecules under study. It is worthwhile noting that the absorbance rs temperature plot is linear (Fig. 2, inset). at least in the tem- perature range explored. The increase in E with decreasing temperature is ca. 0.3-0.4°/0 dm3 mol-' cm-' K-' in ethanol. Toluene solutions could not be investigated due to the poor solubility of the substrates in this solvent at 10% temperature.However, similar spectral modifications were also found in a non-polar solvent (methylcyclohexane). Assuming that linearity also holds approaching room teni-perature. E values at 298 K could be extrapolated. Solvatochromism The absorption band of the open forms exhibited a batho- chromic shift with increasing solvent polarity. An example is given in Fig. 3. To correlate the transition energies in differ-ent solvents. the empirical solvent parameter ET(30) of DimrothI6 was used, which accounts for hydrogen-bond interactions as well as dipolarity effects. Moreover, it is based on a reference compound (a pyridinium-N-phenolate-betaine) which is similar in structure to the merocyanines investigated here. The following correlations were obtained: 1, F/cm-' =: 17880 -27.1 ET(30); 2, F/cm-' = 20 390 -63.8 €,(30): 4.f;cm -= 18 600 -34.1 ET(30); and are illustrated in Fig. 4. 250 350 450 550 650 753 i.nm Fig. 2 Spectra of the colourless (---I and coloured (--) forms of 2 in ethanol at 215, 205. 195, 185, 175 and 165 K (in direction of the arrow). Inset : plot of absorbance rs. temperature. i. nm Fig. 3 Solvatochromic effect on the absorption spectrum of 4 at 298 K in (u)methylcyclohexane, (b)ethyl acetate, (c)acetonitrile and (d) etl anol l9 30 35 40 45 50 55 60 ET(30) Fig 4 Correlation of the transition energy of the photomerocya- nin :s with the solvent polarity parameter €,(30): 1 (0); 4 (A)2 (0); 'The bathochromic shift observed (positive solvato-chi omism)' ' is indicative of an increased dipole moment up+In electronic excitation.Therefore, the ground-state we tkly polar molecule should approach the configuration of the quinoid form. This hypothesis was previously put forward for 1 by Kellmann et a!. using the Brooker's empirical param- ete.s for the correlation,5 as well as by Lenoble and Becker for the photomerocyanine of an indolinespirobenzopyran. In contrast. in the case of some merocyanines derived from the spironitropyrans. which are also closely related in struc-tur.: to the molecules investigated here, a negative solvatoch- roniism was found.lg Consequently, the substituent effect on sol.:atochromism was opposite to that found here, that is, the sen iitivity to solvent polarity increased with electron-dorlating groups.The results from the solvatochromic study alsc, seem to conflict with recent theoretical calculations on the..e molecules20 which predict a prevalent zwitterionic stri cture for the ground state. However, since calculations consider the system in the absence of solute-solvent inter-actions. the disagreement is not surprising. Tht,rmosquilibriumand Thermochrornism Concentrated solutions of 1, 2 and 4 (c 4c lop3mol dm-3) showed a low intensity absorption band in the visible region at 'oom temperature, denoting that a thermal equilibrium was established between the open and closed forms. The abs xption intensity. which was weaker in a non-polar J.CHEM. SOC. FARADAY TRANS., 1994, VOL. 90 Table 3 Thermodynamic parameters of the reaction spiroxazine +photomerocyanine (298 K) 1 2 3 4 ethanol toluene ethanol 1.1 0.12 58K 298/1o4 AH"/kJ mol-' 20.9 21.7 15.0 (1 7.9)" AG"/kJ mol -' 22.6 28.0 12.8 ASo/J mol -' K --5.6 -21 7.4 a From ref. 4. solvent than in a polar one, increased as the temperature increased, that is, these molecules exhibited thermochromic properties. For 3, the coloured band was hardly detectable, even in a saturated solution of a polar solvent at 320 K (absorbance at 600 nm less than 0.005). In principle, the absorbance in the visible region of non- irradiated concentrated solutions can be used to calculate the equilibrium constant of the spiroxazine merocyanine system, if the molar absorption coefficient of the open form is known.To our knowledge, such a calculation has never been reported before, probably because of the difficulty of obtain- ing reliable E values for the merocyanines. The equilibrium constants determined in ethanol and toluene are reported in Table 3. It can be observed that the K29, values are spread over a large interval ranging from to lo-'. The equi- librium constant increases with the solvent polarity and with electron-donating groups, while it decreases to an unmeasur- able value (K29, < in the presence of the electron- withdrawing Br substituent. Variations introduced in K values by both solvent and structure are so large that the uncertainty of the E values used for their calculations should not invalidate the relative comparison.The enthalpy of reaction could be evaluated by measuring the absorbance of the coloured form at several temperatures, according to the van't Hoff equation, d In K/d(l/ T)= d In A/d(l/T) = -AH/R. Plots of In A us. 1/T are shown in Fig. 5. AHo values are reported in Table 3 along with AGO and AF calculated by means of the thermodynamic relationships. Variations in AHo from ethanol to toluene for the same molecule are within the experimental error, while the decrease of AHo with electron-donating groups and the increase with the Br electron-withdrawing substituent are sig- nificant. The reaction entropy is generally small. Considering that is was estimated by calculating a small difference between two large and close numbers, the estimated values are almost of the same order of magnitude of the experimen- -0.5 I I I I -1.5 -C -2.5 -3.5 2.9 3.1 3.3 3.5 3.7 103 KIT Fig.5 Data for thermal equilibrium treated according to the van't Hoff equation in ethanol (empty symbols) and in toluene (filled symbols): l(0);2 (A);4 (0) toluene ethanol/toluene ethanol toluene 6.0 < 110 51 18.0 > 34 10.9 9.6 18.4 >34 11.2 13.1 -1.3 - -1.0 -11.7 tal error. Negligible ASo values indicate that the positive con- tribution to entropy due to increased torsional freedom in the open structure is approximately compensated by the solvent reorganization around the more polar merocyanine form.However, less negative (positive for 2) ASo values in ethanol than in toluene are responsible for the increased K in the polar solvent. Kinetics of Thermal Bleaching The kinetics of thermal bleaching of the photomerocyanines were investigated in ethanol and toluene following the fading of colour at the maximum absorption. The ring-closure reac- tion was found to be strictly first order in the temperature range explored for all molecules in both solvents. Typical first-order kinetic plots (p > 0.99) are shown in Fig. 6. The rate constants are given in Table 4, which also includes liter- ature data. The agreement with the latter can be considered satisfactory. However, we did not find any sign of the bi- exponential decay reported by Kellmann et d.for 1 in a non- polar solvent.' This is probably due to the different time-resolutions of the detection systems used. An increase in the reaction rate with increased solvent polarity is evident from the data reported in Table 4.Such a trend was also noted previously.' In addition, interesting structure effects are observed, such as the low rate for 2 in toluene and the high one for 4 in ethanol, while a close simi- larity is observed between 1 and 3. As can be seen from Table 4, the activation energy is more structure dependent in ethanol (41-81 kJ mol-') than in toluene (47-66 kJ mol- '). The corresponding Arrhenius plots are shown in Fig. 7. The thermodynamic activation parameters, calculated from the kinetic data, are also reported in Table 4.It can be seen that AG' is scarcely influenced by the solvent and structure (AG' = 70-80 kJ mol-'). This is due to a compensation 0 -1 -2 -? 5 -3 -4 (e )I,IIIIII 0 25 50 75 100 125 150 time/s Fig. 6 First-order kinetics for the thermal bleaching of merocyanine 4 in toluene. (a) 281, (6)286, (c) 291, (d)296 and (e) 301 K. J. CHEM. SOC. FARADAY TRANS., 1994, VOL. 90 Table 4 Kinetic parameters for the reaction of thermal bleaching (298 K) 1 2 3 4 ethanol toluene ethanol toluene ethanol toluene ethanol toluene ils -0.23 0.23 0.58 0.035 0.55 0.18 3.2 0.14 (0.21)o (0.27)* (0.044)b (5.0)c (0.l)d(0.54)e (0.04)d (0.07)' (0.4)d (0.03)J (0.28) E,/kJ mol-l 81.1 47.3 41.1 64.4 73.5 58.0 65.6 66.5 (82.4)" (72.0)cAH#/kJ mol-' 78.6 44.8 38.5 AG#/kJ mol-I 76.5 76.7 74.2 ASf/J mol-' K-' 6.7 -107 -120 From ref.4; from ref. 13; from ref. 6; from ref. 15; from ref. 5; 1992,89, 897. between energetic (AH') and entropic (AS') factors. The activation entropies, in fact, range from values close to zero when AHf is large (1 and 3 in ethanol) to fairly negative values in the cases where AHf is small (2 in ethanol). Combining the kinetic parameters of the ring-closure reac- tion and the thermodynamic data obtained from the study of thermal equilibrium, the rate constants and kinetic activation parameters for the thermal breakage of the spiro bond were calculated and are reported in Table 5. Despite the slowness of the forward r_eaction (k = 10-6-10-2 s-'), the large differ- ence between k and & allows thermal equilibration to be attained in a few seconds.It can be observed that the rate parameters of the forward reaction vary over a much larger range than those of the back reaction, thus determining the large differences in the equilibrium constant and thermochro- mic behaviour. Concluding Remarks These results point to a similar reaction mechanism in toluene for all the molecules investigated. This mechanism passes through an activated state where entropy is lost, prob- ably because of the partial charge separation (negative on the oxygen and positive on the indoline nitrogen). The dipolar transition state interacts with the solvent much more than the closed and open forms, both being weakly polar molecules, as confirmed for the last species by the positive solvatochromic effect.In ethanol, the kinetic behaviour is more structure depen- dent. Even though AG' is similar for all the molecules (AG' = 70-76 kJ mol-' for the back reaction and AG' = 81-99 kJ mol- for the forward reaction), marked differences were observed in the contributions of AHf and AS'. The entropic factor dominates for 2, which was indicated by the solvatochromic effect to be the least polar in the ground state. The activation energy of the thermal reaction is lower owing to solvent stabilization of the dipolar transition state, while entropy is lost because of the solvent reorganization around it.For 1 and 3 (only back reaction explored), very 61.9 71.0 55.5 63.1 64.0 81.1 74.5 77.2 69.9 77.7 -64.8 -11.6 -72.9 -23.0 -46.4 from D. Eloy, P. Escaffre, R. Gautron and P. Jardon, J. Chim.Phys., -0.5 -1.5 -2.5 -3.5 4.5 -5.5 3.4 3.6 3.8 4.0 103 KIT -1.5 -2.5 -3.5 4.5 3.3 3.5 3.7 3.9 103 KIT Fig. 7 Arrhenius plots for the thermal-bleaching reaction (a) in ethanol and (b) in toluene of 1 (0,M), 2 (A, A),3 (a,*)and 4 (0, small AS' values point to a transition state quite similar to the starting molecule, while 4 represents an intermediate situ- ation. From this study, even if the number of molecules investi- gated is limited, some aspects emerge which can be gener- alized. Increasing the solvent polarity always increases both the equilibrium constant and the rate constants.The relative .) Table 5 Kinetic parameters of the forward reaction (298 K) 1 2 3 4 ethanol toluene ethanol toluene ethanolltoluene ethanol toluene 1;/104 S-EfiJ mol-' 0.25 102 0.028 69.0 33.6 56.0 0.2 1 82.4 <0.005 > 100 352 76.5 7.1 76.1 AH#/~Jmol-' AGf/kJ mol-' AS#/J mol-' K-' 99.5 99.2 1 66.5 104.7 -128 53.5 86.9 -112 79.9 99.6 -66 >loo 74.0 81.2 -24 73.6 90.9 -58 338 J. CHEM. SOC. FARADAY TRANS., 1994, VOL. 90 contributions of enthalpy and entropy are mainly determined by the dipolar nature of the normal and activated state. An electron-donating substituent such as piperidine decreases the electron-accepting capacity of the oxygen atom, thus leading to a lower rate parameter for the ring-closure reaction.The rate parameters of the forward reaction are dramatically affected by both solvent and structure effects, thus leading to 5 6 7 8 9 A. Kellman, F. Tfibel, R. Dubest, P. Levoir, J. Aubard, E. Pottier and R. Guglielmetti, J. Photochem. Photobiol., A, 1989,49, 63. U. W. Grummt, M. Reichenbacher and R. Paetzold, Tetra-hedron Lett., 1981,22, 3945. S. Schneider, 2. Phys. Chem., Neue Folge, 1987,154,91. S. Schneider, A. Mindl, G. Elfinger and M. Melzig, Ber. Bun-senges. Phys. Chem., 1987,91, 1222. S. Aramaki and G. H. Atkinson, Chem. Phys. Lett., 1990, 170, marked changes in the thermochromic behaviour. 10 181. E. Pottier, A. Samat, R. Guglielmetti, D. Siri and G.Pepe, Bull. This work was undertaken under the National Programme of Research for Chemistry sponsored by the Italian ‘Minister0 per 1’Universita e la Ricerca Scientifica e Tecnologica’ (Tema lO-Consorzio R.C.E.). Financial contributions by the Italian 11 12 SOC. Chim. Belg., 1992, 101,207. C. Bohne, M. G. Fan, Z-J. Li, J. Lusztyk and J. C. Scaiano, J. Chem. SOC., Chem. Commun., 1990, 571. C. Bohne, M. G. Fan, Z-J. Li, Y. C. Liang, J. Lusztyk and J. C. Scaiano, J. Photochem. Photobiol., A, 1992,66, 79. Consiglio Nazionale delle Richerche are also gratefully acknowledged. 13 14 F. Wilkinson, J. Hobley and M. Naftaly, J. Chem. SOC., Faraday Trans., 1992,88, 151 1. A. S. Kholmanskii and K. M. Dyrmaev, Dokl. Akad. Nauk SSSR, 1988,303,1189 References 15 D. Eloy, P. Escaffre, R.Gautron and P. Jardon, Bull. SOC. Chim. Belg., 1992, 101, 779. 1 R. Guglielmetti, in Photochromism, ed. H. Durr and H. Bouas- Laurent, Elsevier, Amsterdam, 1990, p. 855; N. Y. C. Chu, p. 879; K. Ichimura, p. 903. 2 See e.g. R. Guglielmetti, in Photochromism, ed. H. Durr and H. Bouas-Laurent, Elsevier, Amsterdam, 1990, p. 3 14. 16 17 18 19 K. Dimroth, C. Reichardt, T. Siepmann and F. Bohlmann, Liebigs Ann. Chem., 1963,661, 1. P. Jacques, J. Phys. Chem., 1986,90,5535. C. Lenoble and R.S. Becker, J. Photochem., 1986,34,83. S. R. Keum, M. S. Hur, P. M. Kazmaier and E. Buncel, Can. J. 3 N. Y. C. Chu, in Photochromism, ed. H. Durr and H. Bouas- Chem., 1991,69, 1940. Laurent, Elsevier, Amsterdam, 1990, p. 493. 20 V. Malatesta, G. Ranghino, U. Romano and P. Allegrini, Int. J. 4 N. Y. C. Chu, Can. J. Chem., 1983,61,300. Quantum Chem., 1992,42,879. Paper 3/03115B; Receiued 1st June, 1993