摘要:

J. Chem. SOC., Faraday Trans. I, 1982, 78, 89-99 Electroactivity of Hydrogen-Nitrogen Compounds in Ionic Liquids Platinum Rotating-disc Electrode Voltammetric Behaviour of Ammonia in the (Na, K)NO, Equimolar Melt BY ELIO DESIMONI AND PIER G. ZAMBONIN* Istituto di Chimica Analitica, Universita degli Studi di Bari, Via Amendola 173, 70126 Bari, Italy Received 20th November, 1980 In the context of a systematic investigation on light-fuel electrodes in ionic solvents, the voltammetric behaviour of ammonia was studied in the (Na, K)NO, equimolar melt at 518 K. Solutions containing NH,, NH, + H20, NH, +NO; and NH; were tested by using platinum rotating-disc electrode systems. Mechanistic models for the electro-oxidation and electroreduction reactions of ammonia are suggested which fit the experimental electrochemical findings and, in particular, the constant value of the ratio between the electroreduction and the electro-oxidation limiting currents of ammonia.The diffusion coefficient of ammonia in the given experimental conditions was estimated to be 1.9 x cm2 s-l at the given temperature. The chemical behaviour of amide ions, which are involved in the ammonia electroreduction pathway, is described. ‘Concealed limiting-current ’ phenomena observed when ammonia is flowed in nitrite-containing melts are explained on the basis of acid-base reactions between each of these two species and the electro-oxidation product of the other. A systematic investigation is currently being performed in this laboratory into the behaviour of hydrogen in the (Na, K)NO, equimolar melt.In the field of hydrogen- fused-salt systems1 this is the only attempt to study quantitatively various aspects of the problem by combining different experimental techniques. Data collected to date concern the solubility of hydrogen,, its physicochemical interactions with the nitrate melt,, electrode mechanisms4-’ and the analytical detection of the chemical species involved therein.’-1° A recent voltammetric surveyll9 l2 showed that high hydrogen concentrations could possibly be obtained in the melt via the catalysed chemical decomposition of ammonia. Ammonia might thus be a useful hydrogenated fuel in a hydrogen l4 and direct ammonia fuel cells appear15-19 promising. However, no prototype for technological use has been developed and this can probably be attributed, at least in part, to insufficient information on the basic behaviour of ammonia in the given solvents.So far only the solubility of ammonia in nitrate2O9,l and perchlorate21 melts and the qualitative voltammetric behaviour22 of the NH,/NH;, H, system in fused hydroxides have been studied in ionic liquids. In this context a systematic investigation was initiated in this laboratory into the chemical and electrochemical behaviour of ammonia in the (Na, K)NO, equimolar melt. This paper refers the results of rotating-disc electrode (r.d.e.) voltammetric experiments performed on solutions of ammonia in anhydrous or slightly wet melts. 4 89 FAR 190 ELECT RO A C T I V I T Y 0 F HY D R 0 G EN-N I TROG E N COMPOUNDS EXPERIMENTAL Most of the information on chemicals and methodology, as well as on the cell, the voltammetric set-up and the thermostatic system has been recently reviewed.l The solvent, the (Na, K)NO, equimolar melt at 518 K, was purified from NO; and OH- traces by direct injection of concentrated nitric acid.7 Since Teflon-shielded platinum counter- electrodes and Teflon-lined cells were used, only the platinum surface of the working electrode was in contact with the melt during the experiments. When necessary the melt was dehydrated in bubbling ammonia (anhydrous grade, Matheson) and/or nitrogen (UPP grade, SIO Milan) subsequently dried by sodium amide and 5A molecular sieve traps, respectively.Variable NH, + N, mixtures were prepared by using a Matheson 7400 rotameter. A controlled partial pressure of water vapour could be maintained in the cell with the aid of thermostatted water saturators.’ Sodium amide (97%, Alfa) and NH,NO, (previously dried at 423 K) were directly added to the melt, while increasing concentrations of NO, ions were obtained by adding small drops of NaNO, solid solutions in the specified solvent.The rotating-disc electrode was a platinum wire (1.0 mm in diameter) coaxially sealed into a thick-walled soft-glass tubing and ground flat to form a centre disc with protecting rim.,, The Pt r.d.e. was cleaned by ‘mirror’ grinding on Al,O, (polishing powder grade B, Fisher) and then was washed with water and organic solvents. The normal rotation speed was 600 r.p.m. Graphical verification of the Levich equation was performed at eight different rotation speeds ranging from 250 to 2000 r.p.m.Potentials are referred to a AgNO, (0.07 mol per kg of the same nitrate solvent)/Ag reference half-cell connected to the solution via an asbestos wick.,, The ESCA investigation was performed on platinum-sheet samples electrochemicaIly treated at different, controlled potential values (vide infra). A Kratos ES 200 B spectrometer was used with X-ray excitation from aluminium K, at an energy of 1486.6 eV. RESULTS AND DISCUSSION SOLUTIONS OF NH, I N ANHYDROUS MELTS The Pt r.d.e. base voltammogram obtained in a (Na, K)NO, equimolar melt maintained under anhydrous nitrogen atmosphere at 518 K is given in fig. 1, curve (a). No wave is displayed in the solvent electroactivity range because the residual concentration of the usual impurities, water8* 24-26 and nitrite2’* 28 is lower than 1 x mol kg-l.Only the known passivation peak at ca. - 1.8 V29-32 and the fairly constant post-peak current, hp,30* 3 2 9 33 are recorded up to the alkali ions’ reduction Completely different current-potential profiles are obtained when anhydrous ammonia is bubbled in the melt. Wave (b) in fig. 1 is recorded on starting the scan from -0.6 V towards more positive potentials. It can be attributed to the ammonia electro-oxidation and its half-wave potential was found to be - 0.15 _+ 0.01 V in the ammonia concentration range 9 x The relevant limiting current, h, in fig. 1 , is linearly related to the ammonia concentration and to the square root of the rotation speed (correlation coefficients r = 0.998 and 0.997, respectively) and then can be described by the Levich equation35 339 34 < [NH,]/mol kg-l < 1.83 x I = 0.62 nFADf v-4 cui c where n is the number of electrons, A is the disc surface area (cm2), D is the diffusion coefficient (cm2 s-l), v is the kinematic viscosity (cm2 s-l), o is the angular velocity of the disc (rad s-l) and c is the concentration of the electroactive species (mol drn-,).By working under 1 atm anhydrous ammonia, e.g. in the presence ofE. DESIMONI AND P. G. ZAMBONIN 91 I Vi 4 I I 1:o 0 0 1 I + .I 1 , 4 0 -2.0 thp -3.0 a -300 I , po ten tial/V FIG. I.--(a) Pt r.d.e base voltammogram recorded in a purified (Na, K)NO, equimolar melt maintained at 518 K under flow of anhydrous nitrogen. (b) and (c) Pt r.d.e voltammograms recorded under the same experimental conditions but at 1 atm anhydrous ammonia.( d ) Example of Pt r.d.e. voltammogram obtained on reversing the scan direction immediately after recording curve (b). [NH,] = 1.83 x lo-, mol kg-1 [the solubility at 518 K was obtained by extrapolation of data given in ref. (20)] h, was found to be 135 f 10 pA (Pt r.d.e. area = 7.85 x lov3 cm2; rotation speed = 600 r.p.m.; T = 518 K). If, on the contrary, the scan is started from -0.6 V towards more negative potentials curve (c) in fig. 1 is obtained in which a cathodic wave is recorded just before the peak. At the same time the after-peak current appears markedly increased (compare h, and h, in fig. 1). The cathodic wave can be attributed to the ammonia electroreduction; its limiting current, h,, linearly related to the ammonia concentration and to the square root of the rotation speed (correlation coefficients 0.996 and 0.995, respectively) can be described by eqn (1).By working under 1 atm anhydrous ammonia h, was found to be 360f 15 pA under the given experimental conditions. The two ammonia waves are well-reproducible if one waits for the zeroing of possible residual currents at the initial potential (- 0.6 V) before starting the scan. Frequent cleanings of the Pt r.d.e. surface were necessary to minimize a rather evident current noise (see ammonia waves in fig. 1) probably due to electrode-surface poisoning and hydrodynamic turbulence effects. The h,/h, current ratio was 2.7k0.3 in the ammonia concentration range 9 x lo-* < [NHJmol kg-l < 1.83 x lop2.On reversing the scan direction immediately after recording the ammonia electro- oxidation wave (b), irreproducible voltammetric profiles such as curve ( d ) in fig. 1 were observed. Sometimes two consecutive cathodic waves were recorded before the peak, even if the total limiting reduction current, h,, did not change. A series of voltammograms obtained on starting the scan from -0.6 V, recording the NH, anodic wave and inverting the scan direction at different potential values, is shown in fig. 2. Since curves such as (el) in fig. 2 can be also obtained by applying -0.2 V to the Pt r.d.e. for some minutes before scanning towards more negative potentials, the shape of curves (b’)-(e’) depends on the adsorption on the electrode surface of the 4-292 ELECT R 0 ACT 1 V I TY OF HY D R 0 GEN-N I TRO GEN C 0 M P 0 UND S 0.5 0.0 -0.5 -1.0 -1.5 -2.0 potential/V FIG.2.-Voltammograms recorded in a melt maintained under flow of an anhydrous mixture of N, + NH, (partial pressure of N, = 0.9 atm; partial pressure of NH, = 0.1 atm). (a) reference voltammogram obtained on starting the scan from - 0.6 V after disappearance of any residual current. (b) + (b’), (c) + (c’), ( d ) + ( d ’ ) and (e)+(e’) Voltammograms obtained on inverting the scan direction at +0.03, +0.1, +0.3 and + 0.73 V, respectively, after recording the ammonia electro-oxidation wave. ammonia electro-oxidation products and not on the inversion-potential value. Chemical modifications of the actual status of the electrode surface were excluded because ESCA analysis did not show any significant difference between platinum samples polarized at + 0.4 and - 1.45 V for 2 min in an ammonia-saturated melt and the blank, i.e.an unpolarized platinum sheet dipped into the same melt under a nitrogen atmosphere. SOLUTIONS OF AMMONIA I N WET MELTS Curve ( a ) in fig. 3 is the base voltammogram which can be recorded in a nitrite-free melt maintained under atmosphere of slightly wet nitrogen (water partial pressure of ca. 6 Torr.). The cathodic wave starting at ca. - 1 .O V is the known ‘water which is due to the overall reaction H,O + NO; + 2e = NO, + 20H-. (2)E. DESIMONI AND P. G . ZAMBONIN --600 -300 93 1 1 .o I \ / \ / ' '-/-- u 0 0 1, ~ ~ ~~~~ potential/V FIG. 3.+a) Base voltammogram obtained under flow of wet nitrogen (water partial pressure x 6 Torr).(b) and ( c ) Voltammograms obtained in the same wet melt under flow of ammonia (partial pressures of water and ammonia = 6 and 754 Torr, respectively). potential/V FIG. 4.--(a) Pt r.d.e. voltammogram recorded in a melt containing [NO;] = 3 x mol kg-' and maintained under flow of anhydrous nitrogen. (6) and (c) Obtained in the same melt maintained under flow of anhydrous N,+NH, (partial pressure of ammonia = 0.1 atm, partial pressure of nitrogen = 0.9 atm). When ammonia is flowed through the melt in place of nitrogen curves (b) and (c) in fig. 3 are obtained. The presence of water does not seem to influence the ammonia waves [compare curves (b) and ( c ) in fig. 1 and 31. Again the h,/h, ratio is ca.2.7 if the water diffusion current, h,, is subtracted from the total cathodic limiting current, h,, ( i e . in this case h, = hT- h,).94 E LE C T RO A C TI V I TY OF HYDRO G E N-N I TR 0 GE N C 0 M POUND S SOLUTIONS OF AMMONIA I N NITRITE-CONTAINING ANHYDROUS MELTS The current-potential profile recorded in an anhydrous melt maintained under a nitrogen atmosphere and con.taining a nitrite concentration = 3 x mol kg-l is reported in fig. 4, curves (a) + (a’). The anodic wave (half-wave potential = + 0.48 V) is due to the k n ~ w n , ~ ~ 28 electro-oxidation of NO; to NO,. When anhydrous ammonia (partial pressure of ammonia = 0.1 atm) is flowed through the melt in place of nitrogen, curves (b) and ( c ) are recorded in which no NO, wave can be seen after the NH, oxidation wave. The disappearance of the nitrite is only apparent since curves (a) and (a’) in fig.4, showing the original nitrite content, can be newly recorded if pure nitrogen is flowed through the melt in place of ammonia. CHEMICAL BEHAVIOUR OF NH, IONS I N ANHYDROUS MELTS The chemical behaviour of amide ions in the (Na, K)NO, equimolar melt was qualitatively investigated since they can participate in electrodic mechanisms involving NH,. 0.5 0 A h 15 I 30 Icl potential/V FIG. 5 . 4 4 Pt r.d.e. voltammogram obtained in a (Na, K)NO, melt maintained under flow of anhydrous nitrogen. (b) and (c) Obtained after 2 and 6 min, respectively, from the addition of 0.072 g of sodium amide (theoretical analytical concentration = 9.9 x lo-, mol kg-l). Curve (a) in fig.5 is the base voltammogram recorded for a Pt r.d.e. dipped in a purified, anhydrous melt maintained under a flow of nitrogen. In a typical experiment 0.072 g of solid NaNH, (corresponding to FH;] = 9.9 x mol kg-l) were added to the melt immediately after recording curve (a): a violent reaction occurred at the surface of the melt, accompanied by light flashes and the eruption of a white vapour. At the same time the solution became pale yellow and, for a few seconds, ammonia could be detected in the gas flow at the cell output. Curves (b) and (c) in fig. 5 were recorded 2 and 6 min after the addition respectively. On the basis of the experimental half-wave potential (Ei x -0.2 V) the anodic wave in curves (b) and (c) can be attributed to the kn0wn~9~ electro-oxidation reaction of the OH- ion 20H- = H20+$0,+2e.(3) The relevant diffusion current, h, constant over several hours, could be increased by adding NaOH ions to the melt.E. DESIMONI A N D P. G. ZAMBONIN 95 The hydroxide concentration produced by decomposition of the amide, calculated on substituting in eqn (1) the known5 diffusion coefficient of the OH- ion [D(OH-) = 3.6 x lod6 cm2 s-l at 513 K] and the experimental value of the hydroxide limiting current ( h = 0.028 mA from fig. 5) was found to be [OH-] = 8.7 x lo-, mol kg-l, i.e. it was ca. 10% lower than the analytical concentration of the added amide. On considering that amide losses by evaporation during the initial violent reaction cannot be excluded, the result seems to indicate a nearly complete and fast one-to-one production of OH- from NH;, in the simplest manner according to the reaction NH;+NO; = NO;+OH-+NH. (4) The NH radicals can rapidly decompose according to reaction and/or NH = $N2 + $H2 2NH +NO, = N, + H,O +NO,.The initial production of traces of ammonia can be ascribed to the reaction of NH; with the water produced by reaction (6) NH, + H,O = NH, + OH-. (7) Hydrogen and/or ammonia dissolved in the fused solvent may explain the observed anodic maximum of curve (b) in fig. 5, which is characteristic of nitrate melts containing ‘electrodic’ excess of OH- with respect to hydrogen1, 5 9 or ammonia.12 Of course the maximum disappears as soon as hydrogen and/or ammonia are stripped from the melt [see curve (c) in fig. 51. Nitrite ions produced via reactions (4) and (6) give a characteristic yellow colour to the melt.However, a quantitative voltammetric determination of NO; ions in the presence of a comparable concentration of OH- is not possible36 due to partial ‘concealed limiting-current’ 37 In the simplest way these effects can be explained by a partial chemical consumption of the depolarizers (i.e. hydroxide and/or nitrite) on the electrode, in the potential range where both OH- and NO; can be electro-oxidized. The following parallel pathways can be suggested ( 2(OH- = OH* +e) 1 20H- +NO; = H20+NO; (solvent) (8) (9) (1) ( 2(NO; = NO,+e) (10) 1 2N0, + 20H- = H,O + NO; +NO; (solvent) (11) (1 1) The second anodic wave in curves (b) and ( c ) of fig. 5 can be easily explained by considering the electro-oxidation of NO; ions not consumed via mechanisms (I) and (W- DISCHARGE MECHANISMS OF AMMONIA I N NITRATE MELTS The experimental results described can be rationalized on the basis of the following The electro-oxidation wave (b) in fig.1, 3 and 4 can be ascribed to the overall (12) mechanistic model. reaction 4NH, = 3NHt + $N2 + 3e which was also in liquid ammonia. Several items of experimental evidence support the production of NH,+ ions: (a) Only NHf ions were voltammetrically detected39 after a controlled-potential, massive electrolysis performed by applying +0.4 V to a platinum sheet dipped in a96 E LEC TRO ACT I V I TY OF H Y D R OG E N-N I TR OG EN CO M POUNDS melt maintained under a flow of anhydrous ammonia. A few hours after the end of the electrolysis (the lapse of time dependent on stirring, gas flow, etc.) NH,+ ions disappeared completely from the melt and a white deposit, identified as NH4N0,, appeared on the ‘cold’ top of the Pyrex cell.Under our working conditions, i.e. under flow of anhydrous ammonia of the order of 100 cm3 min-l, the phenomenon is probably due to direct evaporation and/or stripping of NH4N0, from the 41 The thermal decomposition of ammonium nitrate into N20 + H20, which usually accounts for 98% of the decompo~ition,~~-~* is relatively slow at the working temperature and is inhibited4, by the presence of the ammonia atmosphere. (b) The ammonium ions produced by massive electrolysis completely disappear if nitrite ions are added to the melt. Under our experimental conditions, in fact, NH; ions react with nitrite according to the equation NHZ+NO; = N2+2H20. (13) The stoichiometry of this reaction was confirmed in a separate experiment by arnper~metric~~ titration of a known amount of NO; with solid, dehydrated NH4N0,.The time necessary for the titration was sufficiently short and the kinetics of reaction (13) sufficiently rapid to neglect the thermal loss of ammonium nitrate. (c) On substituting in eqn (1) the known values of c (1.83 x mol kg-l = 3.96 x mol dm-,, Z = 0.135 mA and n = i), a diffusion coefficient for ammonia was calculated at 5 18 K [D5ls(NH,) = 1.9 x cm2 s-l] which is practically coincident with the values obtaineds* lo for other uncharged, polar species under similar experimental conditions, i.e. water at 503 K and carbon dioxide at 510 K.On considering that hydrogen is produced during the Pt-catalysed decomposition of ammonia11$12 and that proton-acceptor species (such as OH- and CO:-) are necessary1? 5-7 for hydrogen electro-oxidation in nitrate melts, the most probable pathway for the overall reaction (1 2) is the following: NH, = 4N2 + iH2 $H2 = 3H++3e 3(H+ + NH, = NHZ) 4NH, = 3NH; + 4N2 + 3e. I (111) The simplest pathway for the ammonia electroreduction [see curves (c) in fig. 1,3 and 41 seems to be the following: NO; + 2e = NO; + 02- NH, +NO; + 2e = NO; + NH; + OH- (17) (18) (19) (IV) [ NH, + 02- = NH; +OH- Reaction (17) is the known electroreduction of the solvent while step (18), in which oxide ions are consumed by the acidic species NH,, was observed even at low working temperatures for the direct reaction of Cs20 with NH,.45 Mechanism (IV) is analogous to the one previously proposeds for the ‘water wave’ in nitrate melts: in that case water, instead of ammonia, was the acidic species which consumes oxide ions and permits the ‘anticipation’ of the solvent discharge.Mechanism (IV) does not take into account the chemical reaction of NH; ions with the melt. At present, however, it is not possible to define how fast and/or complete reactions (4)-(7) can be and, in particular, if amide ions chemically decompose within the diffusion layer.E. DESIMONI A N D P. G. ZAMBONIN 97 As further indirect support for the electrode reactions proposed for reduction and oxidation of NH, in the given solvent, the experimental ratio (2.7 f 0.3) of the relevant limiting currents h, and h, [see fig.1 and 3 and eqn (12) and (19)] matches the theoretical value = 2.66. h2 %ed - h, no, 314 -=--- DISAPPEARANCE OF NITRITE UNDER AMMONIA ATMOSPHERE Ammonia cannot react with NO; ions. However, both species can be electro-oxidized under the present experimental conditions (the relevant half-wave potentials are - 0.15 and + 0.48 V, respectively) and each of them can in theory react (via acid-base processes) with the electro-oxidation product of the other. The two alternative or parallel pathways can be written as ( NO,=NO,+e 1 N02+NH, = products and (12) (13) 4NH, = 3NHi + 4N2 + 3e (VI) 1 1 3(NH: + NO; = N, + 2H,O) which permits us to explain this second ‘concealed limiting-current ’ phenomenon [see ref. (37), p. 1961.It can be readily understood that when nitrogen is flowed through the melt to remove ammonia, the electro-oxidation wave of nitrite can be again recorded [see curves (a) and (b) in fig. 41. THE POST-PEAK CURRENT It is known29-32 that the cathodic peak recorded in a (Na, K)NO, melt (see for example fig. 1) is due to the passivation of the electrode surface by a solid film of Na,O formed during the electroreduction of the nitrate ion [see reaction (17)], while the post-peak current is kinetically controlled by the oxide-film dissolution rate. If a chemical reaction occurs between the oxide film and species present in solution ( e g ~ nitrate ions, water, ammonia, etc.) the dissolution rate (and the post-peak current also) becomes higher with respect to the value due to the simple ‘physical’ process.The situation can be represented as follows: physical dissolution h 2Na+ + 02- +NO; chemical dissolution + 2Na+ +NO; + 0;- (23) (24) * (25) b (26) + H,O chemical dissolution 2Na+ + 20H- + NH, chemical dissolution 2Na+ + OH- + NH, I N a 2 W where (s) denotes the solid phase on the electrode surface. In dry melts only reactions (23) and (24) (prevalent) are re~ponsible~~ for the redissolving of the oxide layer and their rates determine the value of current h, in fig. 1. In wet melts an additional important chemical contribution to the total dissolution rate from reaction (25). Similarly, reactions (26) or (25) and (26) increase the Na,O dissolution rate when anhydrous or wet ammonia is present in solution. From fig. 1 and 3 it can be inferred that the post-peak contributions from NH, and H,O are very close to the pre-peak diffusion currents of the two species.At potential values more negative then the peak potential, any water or ammonia molecule which98 EL E C T R 0 A C T I V I T Y 0 F H Y D R 0 G E N-N I T R 0 GE N C 0 M P 0 U N D S chemically reacts with a Na20 molecule [see reactions (25) and (26)] induces the formation of a new oxide molecule uia a global reaction involving two electrons 2Na+ + NO; + 2e = NO; + Na20J. On the other hand two electrons are also required for the electroreduction of one molecule of water8? 2 5 9 26 or ammonia [see reactions (2) and (19), respectively]. The fact that the post-peak current contributions from water and ammonia are still controlled (almost to a first approximation) by the diffusion of the two species, seems to indicate that reactions (25) and (26) are fast and shifted to the right.This work was carried out with the financial assistance of the Italian National Research Council (C.N.R. Rome). Some results were obtained with the help of G. Fuggiano and N. Cardellicchio during their undergraduate research training. P. G. Zambonin, E. Desimoni, F. Palmisano and L. Sabbatini, in Ionic Liquids, ed. D. Lovering and D. Inman (Plenum, London, 1981). E. Desimoni, F. Paniccia and P. G. Zambonin, J. Chem. Soc., Faraday Trans. I , 1973, 69, 2014. E. Desimoni, F. Paniccia and P. G. Zambonin, J. Phys. Chem., 1977, 81, 1985. E. Desimoni, F. Paniccia, L. Sabbatini and P. G. Zambonin, J. Appl. Electrochem., 1976, 6, 445.E. Desimoni, F. Palmisano and P. G. Zambonin, J. Electroanal. Chem., 1977,84, 323. E. Desimoni, B. Morelli, F. Palmisano and P. G. Zambonin, Ann. Chim. (Rome), 1977, 67, 541. E. Desimoni, F. Palmisano, L. Sabbatini and P. G. Zambonin, Anal. Chem., 1978, 50, 1895. P. G. Zambonin, V. L. Cardetta and G. Signorile, J. Electroanal. Chem., 1970, 28, 237. P. G. Zambonin, Anal. Chem., 1971, 43, 1571. E. Desimoni, F. Paniccia and P. G. Zambonin, Ann. Chim. (Rome), 1980, 68, 351. Society of Electrochemistry, Venice (Italy), 1980. lo P. G. Zambonin, Anal. Chem., 1972,44, 763. l2 E. Desimoni, F. Palmisano, L. Sabbatini and P. G. Zambonin, 31st Meeting of the International l3 D. P. Gregory, Sci. Am., 1973, 228, 13. l4 D. P. Gregory, CRY0 '78 Conference, Oak Brook, Illinois, (1978).l 5 E. J. Cairns, E. L. Simons and A. D. Tavebaugh, Nature (London), 1968, 217, 780. l6 E. L. Simons, E. J. Cairns and D. J. Surd, J. Electrochem. SOC., 1969, 116, 556. l7 D. W. Mckee, A. J. Scarpellino, I. F. Danzig and M. S. Pak, J. Electrochem. SOC., 1969, 116, 562. D. W. Mckee and A. J. Scarpellino, Fr. Patent 1571482 (Cl ; HOIm) 20 June 1969, U.S. Patent Appl. 1 June 1967. l9 J. Giner and J. R. Moser, U S . Patent 3650838 (C1 136/86, HOIm) 21 Mar. 1972, Appl. 491759, 30 Sept. 1965. 2o F. Paniccia and P. G. Zambonin, J. Chem. Soc., Faraday Trans. I , 1973, 69, 2019. 21 S. Allulli, J. Phys. Chem., 1969, 73, 1084. 22 J. Goret, Thesis (Universite de Paris, 1966). 23 P. G. Zambonin, Anal. Chem. 1969,41, 868. 24 M. Peleg, J. Phys. Chem., 1967, 71, 4553.25 D. G. Lovering, R. M. Oblath and A. K. Turner, J. Chem. SOC., Chem. Commun., 1976,673. 26 A. Espinola and J. Jordan, in Characterization of Solutes in Non-aqueous Solvents, ed. G. Mamantov 27 E. Desimoni, F. Palmisano and P. G. Zambonin, J. Electroanal. Chem., 1977, 84, 315. 28 F. Palmisano, L. Sabbatini, E. Desimoni and P. G. Zambonin, J. Elecrroanal. Chem., 1978, 89, 31 1. 29 G. J. Hills and K. E. Johnson, Advances in Polarography, Proceedings of the 2nd Znt. Congress (1959) 30 H. S. Swofford and H, A. Laitinen, J. Electrochem. Soc., 1963, 110, 814. 31 L. B. Topol, R. A. Osteryoung and J. M. Christie, J. Phys. Chem., 1966, 70, 2857. 32 P. G. Zambonin, J. Electroanal. Chem., 1970, 24, 365. 33 H. A. Laitinen, Talanta, 1965, 12, 1237. 34 H. E. Bartlett, K. E. Johnson, J. Electrochem. SOC., 1967, 114, 64. 35 V. Levich, Physicochemical Hydrodynamics (Prentice Hall, Englewood Cliffs, N.J., 1962). 36 P. G. Zambonin, unpublished data. 37 J. Heyrowski and J. Kuta, Principles of Polarography (Academic Press, New York, 1966). 38 J. J. Lagowski, Znt. Symp. on Non Aqueous Electrochemistry (Butterworths, London, 1970), p. 441. (Plenum, New York, 1978), p. 311. (Pergamon Press, Oxford, 1960), p. 974.E. DESIMONI A N D P. G. ZAMBONIN 99 39 E. Desimoni and P. G. Zambonin, paper in preparation on the voltammetric behaviour and detection of NH: ions in alkali nitrate melts. G. Feick and R. M. Hainer, J. Am. Chem. SOC., 1954, 76, 5860. 41 K. S . Barclay and J. M. Crewe, J. Appl. Chem., 1967, 17, 21. 42 A. G. Keenan and B. Dimitridates, Trans. Faraduy SOC., 1961, 57, 1019. 43 W. A. Rosser, S. H. Inami and H. Wise, J . Phys. Chem., 1963, 67, 1753. 44 H. L. Saunder, J. Chem. SOC., 1922, 121, 698. 45 E. Rengade, C.R. Acad. Sci., 1907, 143, 592. (PAPER O/ 1796)

ISSN:0300-9599    

DOI:10.1039/F19827800089

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

J . G e m . Soc., Faraday Trans. I, 1982, 78, 101-107 Microscopic and Macroscopic Contact Angles BY M. AMARAL FORTES Departamento de Metalurgia, Instituto Superior Tecnico, Av. Rovisco Pais, 1096 Lisboa, Portugal Receiued 24th Nouember, I980 A simple relation is derived between the microscopic and macroscopic angles of contact of a fluid interface with a solid by taking into account the interaction of the interfaces in the regon of contact. It is also shown how the macroscopic angle can be related to the perturbed interfacial tensions. Both angles are a property of the three phases in contact, independent of geometry and weak gravitational fields. Finally, calculated profiles of axially symmetric fluid interfaces in the neighbourhood of a planar solid surface are compared with those that would occur in the absence of interaction between the interfaces. Young’s equation for the contact angle of a fluid interface with a solid is of great importance in the interpretation of many surface phenomena.In its original, simplest form the equation is written as1* YSZ - Y S l Yo cos 0, = - where 8, is the angle of contact of fluid 1 with the solid (in the presence of fluid 2), yo is the fluid surface tension and ysl, ys2 are the interfacial tensions of the two fluids and solid. What is important in Young’s equation, and is the origin of its wide application, is not so much the relation between the contact angle and the interfacial tensions (which are very difficult to measure, with the exception of yo), but the following facts: (i) it asserts that the contact angle is independent of the size and shape of the fluid phases and of the shape of the solid surface, i.e.the geometry of the system, and (ii) it is independent of ‘weak’ applied fields, such as gravitational and centrifugal fields, that have a negligible effect on the structure of the interfaces. It is implicitly assumed that the values of the y in eqn (1) are the same as those that would be measured in each separate interface, i.e. no account is taken of a possible interaction between the interfaces in the region of contact. The angle 8, calculated in this way will be termed the unperturbed angle of contact. If the interactions are taken into account the interfacial tensions will be perturbed in the region of contact and Young’s equation, at least in the form of eqn (l), is no longer valid.Both molecular and continuum approaches have been used by various a u t h o r ~ ~ - ~ to analyse the interactions between interfaces. The general outcome of these studies is the concept of a microscopic angZe of contact, Om, which is the actual angle between the fluid and solid-fluid interfaces. It is generally accepted that this angle is a property of the three phases in contact, independent of the macroscopic dimensions of the system and of the shape of the solid. The microscopic angle should be distinguished from the macroscopic angZe of’ contact, O M , 3 - 5 9 that would be measured macroscopically, e.g. from an optical photograph of the contact region. The two angles are indicated in fig. 1. To show the macroscopic angle in this figure, the fluid interface profile in the unperturbed region 101102 MICROSCOPIC A N D MACROSCOPIC CONTACT ANGLES FIG.1.-Macroscopic and microscopic angles of contact of fluid 1 with a solid, S, in presence of fluid 2. is extrapolated to the solid surface. Attempts have been made to relate 8, and OM. In these studies3* a cylindrically symmetric geometry was assumed. While this may be questionable,8 the fact is that no general conclusions can be drawn from these studies; in particular, it is not clear whether 8, is also a property of the three phases in contact, independent of the geometry of the system and gravitational fields. Owing to the interactions between the three interfaces, the interfacial tensions will no longer be independent of position in the interfaces, and the equilibrium shape of the fluid interface will be altered near the solid, relative to its shape under unperturbed conditions.Several authors [e.g. ref. (7) and (S)] following Gibbsg have used the concept of a line tension to characterize thermodynamically the contact region. However, it is not possible using the line tension to calculate the shape of the interface in the contact region; therefore the concept is not appropriate for relating the various contact angles. It thus seems preferable to retain the concept of interfacial tension, while allowing for the possibility that interfacial interactions change the values of the tensions. That is, the interfacial tensions should be regarded as a function of position in the interfaces; this is essentially the approach used by Feijter and Vrij7 in their study of soap films.A variational analysis of the equilibrium configuration of three phases in contact, with the interfacial tensions varying with position, has been undertaken recently.’, Equilibrium equations were derived simultaneously for the fluid interface profile and for the microscopic angle of contact, 8,. If the effect of inclination of the fluid interface on the interfacial tension is neglected, the following equation is obtained for 8,: where the j j are the actual perturbed interfacial tensions measured at the line of contact. This equation is formally analogous to Young’s equation. However, the relation between 8, and 8, can only be established from a detailed analysis of the contribution of molecular interactions to the interfacial tensions.The important angle in a large number of surface problems is the macroscopic angle of contact. It is this angle, and not Om, that is accessible to direct measurement and governs the mascrocopic shape of the fluid interface and possibly also the capillary forces due to the interface (since these forces are essentially determined by the shape of the fluid interface and the way it changes when the geometry of the system is changed). However, while 8, and 8, may be regarded as properties of the three phases in contact, it is not clear whether the macroscopic angle also has this property. If 8M were dependent on geometry or externally applied fields, a considerable amount of published work on surface science would be put in question.The purpose of this study is to find a relation between the three angles defined above and, as a consequence, to show that the macroscopic angle of contact is, in fact, a property of the three phases in contact.102 MICROSCOPIC A N D MACROSCOPIC CONTACT ANGLES FIG. 1.-Macroscopic and microscopic angles of contact of fluid 1 with a solid, S, in presence of fluid 2. is extrapolated to the solid surface. Attempts have been made to relate 8, and OM. In these studies3* a cylindrically symmetric geometry was assumed. While this may be questionable,8 the fact is that no general conclusions can be drawn from these studies; in particular, it is not clear whether 8, is also a property of the three phases in contact, independent of the geometry of the system and gravitational fields. Owing to the interactions between the three interfaces, the interfacial tensions will no longer be independent of position in the interfaces, and the equilibrium shape of the fluid interface will be altered near the solid, relative to its shape under unperturbed conditions.Several authors [e.g. ref. (7) and (S)] following Gibbsg have used the concept of a line tension to characterize thermodynamically the contact region. However, it is not possible using the line tension to calculate the shape of the interface in the contact region; therefore the concept is not appropriate for relating the various contact angles. It thus seems preferable to retain the concept of interfacial tension, while allowing for the possibility that interfacial interactions change the values of the tensions.That is, the interfacial tensions should be regarded as a function of position in the interfaces; this is essentially the approach used by Feijter and Vrij7 in their study of soap films. A variational analysis of the equilibrium configuration of three phases in contact, with the interfacial tensions varying with position, has been undertaken recently.’, Equilibrium equations were derived simultaneously for the fluid interface profile and for the microscopic angle of contact, 8,. If the effect of inclination of the fluid interface on the interfacial tension is neglected, the following equation is obtained for 8,: where the j j are the actual perturbed interfacial tensions measured at the line of contact.This equation is formally analogous to Young’s equation. However, the relation between 8, and 8, can only be established from a detailed analysis of the contribution of molecular interactions to the interfacial tensions. The important angle in a large number of surface problems is the macroscopic angle of contact. It is this angle, and not Om, that is accessible to direct measurement and governs the mascrocopic shape of the fluid interface and possibly also the capillary forces due to the interface (since these forces are essentially determined by the shape of the fluid interface and the way it changes when the geometry of the system is changed). However, while 8, and 8, may be regarded as properties of the three phases in contact, it is not clear whether the macroscopic angle also has this property.If 8M were dependent on geometry or externally applied fields, a considerable amount of published work on surface science would be put in question. The purpose of this study is to find a relation between the three angles defined above and, as a consequence, to show that the macroscopic angle of contact is, in fact, a property of the three phases in contact.104 MICROSCOPIC A N D MACROSCOPIC CONTACT ANGLES CYLINDRICALLY SYMMETRIC INTERFACES Integration of eqn (4a) between z = 0 (apex) and z = z, (solid surface) gives yo - y cos 8, = f ( a + Apg*z) dz, (6) since the interfacial tension at the apex has its unperturbed value yo (provided the drop or bubble is not microscopic). 7 is, as previously, the fluid interfacial tension at the line of contact.If we now assume that y = yo at any point of the interface we obtain Comparing eqn (6) and (7) we finally have 7 0 cos 6, = 7 cos 8,. (8) This equation shows that the macroscopic angle of contact is in general different from the microscopic angle, but is independent of the dimensions of the interface and of weak gravitational fields, because Om, yo and j j have these properties. Combining eqn (2) and (8) we also obtain 7s2 - 7Sl Y o cos = -, (9) which is the correct relation between the macroscopic angle of contact and the interfacial tensions. 6, only coincides with 8,, the unperturbed angle, if j7s277sl = ys2-ysl, i.e. if the interaction produces equal changes in the interfacial tensions with the solid. It is worth noting that the above relations for cylindrically symmetric interfaces are exact for drops or bubbles of any size (except when they are so small that the value of y at the apex is perturbed).AXIALLY SYMMETRIC INTERFACES In this case, the equation of the interface profile [eqn (4 b)] contains the term in l/x, so that the argument presented above for cylindrically symmetric interfaces cannot be applied straightforwardly. However, this term will be negligible provided x is large enough in the region where the interaction occurs, i.e. in the region where y # yo. Let the width of this region be denoted by 6 (fig. 2). Then, if the radius of the line of contact with the solid is much larger than 6, eqn (4b) reduces to eqn (4a) and the same procedure will lead to the same relations between the macroscopic and microscopic contact angles, and between and the interfacial tensions.Eqn (8) and (9) therefore also apply to axially symmetric interfaces, We shall now show that neglecting the term in l / x is a correct approximation for sufficiently large drops or bubbles with axial symmetry. To show that the approximation is valid we have determined the actual interface profile by numerical integration of eqn (4b) (with g* = 0) assuming the following specific law for the variation of y with the distance from the solid surface, located at z = z, (see fig. 2) : y = y o ; O d z d z , (10) y = y , [ l + A ( y y ] ; z 2 2,M. A. FORTES 105 where 6 is the range of the interaction (6 = z,--zJ; n > 1 and A are constants. The fluid interfacial tension at the solid surface is p = y o ( l + A ) .(1 1) This law [eqn (lo)] does not pretend to describe any particular physical situation, but it is qualitatively plausible. We introduce adimensional variables X and 2 defined by (fig. 2) (13) Y (4 and define r(z) = -. Yo Therefore Eqn (4b) takes the form (for g* = 0) where X, = X( 1) and 8, is the angle for 2 = 1 (2, = -cot OJ, see fig. 2. Note that the radius of curvature, R, at the apex (I' = 1 ; 2 = 0) is R/6 = X,/sin 8,. Eqn (1 5) was integrated between 2 = 1 and 2 = 0 by the method of Runge-Kutta (4th order) for various values of A , n and of the initial conditions X,, X,. The interface z= 1 Z= 0 1 x-x, 0 1 x -x, x-x, -2 - 1 0 x-x, - 1 0 FIG. 3.-Calculated profiles (full curves) of axially symmetric interfaces of two sizes, for A = 0.2, n = 2 [eqn (14)] and for two values of 8,.The dotted curves are the unperturbed, circular profiles of radius 2X,. The inserts show schematically the complete profiles. The values of 8, and 8, are indicated in table 1 . (a) el = 300, x, = 5 ; (b) 8, = 300, x, = 500; (c) el = 1500, x, = 5 ; ( d ) 8, = 1500, x, = 500.106 MICROSCOPIC AND MACROSCOPIC CONTACT ANGLES TABLE VA VALUES OF 8, AND 8, (IN ") 8, = 30" 6, = 150" cos 8, cos 8, cos 8, n x; 8, 6M a n X / om OM 2 5 20 50 100 500 6 5 100 500 49.266 45.145 44.337 44.07 1 43.859 49.902 44.116 43.868 40.002 32.75 1 31.127 30.568 30.114 40.002 30.568 30.1 14 1.1739 1.1924 1.1968 1.1984 1.1997 1.1893 1.1993 1.1999 2 5 20 50 100 500 6 5 100 500 140.530 137.484 136.7 18 136.457 136.247 141.919 136.504 136.256 165.022 1.2514 153.002 1.2106 151.167 1.2034 150.578 1.2017 150.1 15 1.2003 165.022 1.2273 150.578 1.2007 150.115 1.2001 a The radius of the interface profile in the unperturbed region is R = 2X,6.profile for 2 > 1 is, of course, a circle. The contact angle 8, was determined and compared with 8M, which is easily obtained by noting that the unperturbed profile is a circle down to the solid. In fig. 3 the calculated profiles (full lines) are compared with the unperturbed, circular profiles (dotted lines). Examples of results for A = 0.2 are shown in table 1. For A = 0.2 the ratio cos 8,/COS 8, is 1.2 according to eqn (80) and (1 1). The results of table 1 therefore show that eqn (8) is correct to within Z 0.15 % for drops of radius R > 100 6.That is, provided the drop is of macroscopic dimensions, the term in l / x can be neglected and eqn (8) and (9) apply equally to axially symmetric interfaces. The results in table 1 also show that the relation between the two angles is independent of the way y changes from yo to p, provided the ratio ? / y o is unchanged. DISCUSSION In the analysis undertaken on the contact-angle question, the details of the interaction at the triple junction have been avoided. Instead we have used the concept of perturbed interfacial tensions, variable with position in the interfaces, the perturbation being due to the molecular interactions in the contact region. There seems to be no formal thermodynamic difficulty with this concept, which has in fact been used previously by other a ~ t h o r s .~ ? 6 * The perturbed fluid interfacial tension can then be introduced in a modified, general form of the Laplace equation, with which the interface shape in the contact region is calculated. Simultaneously, the microscopic contact angle can be expressed in terms of these perturbed tensions. In this way the microscopic and macroscopic angles can be related to each other and to the interfacial tensions. The macroscopic angle of contact is distinct from the microscopic angle (except when their value is 90°), but there is a definite relation between the two, independent of geometry and of the intensity of gravitational fields. Both angles are therefore well-defined properties of the interfaces in contact. The actual relation between the two angles [eqn (S)] depends on the type of interaction between the fluid interface and the solid.For example, if the interaction increases the fluid interfacial tension (compared with its unperturbed value), the microscopic angle will be larger than the macroscopic angle for 8, > n/2 and smaller if 8, c n/2 (cf. table 1 and fig. 3). TheM. A . FORTES 107 form of eqn (8) is not altered even if the fluid interfacial tension changes with the inclination of the fluid interface relative to the solid. The relation between the macroscopic angle and the interfacial tensions [eqn (9)] is similar to Young’s equation, with the interfacial tensions with the solid (but not the fluid interfacial tension) replaced by their perturbed values at the line of contact.It would not be surprising if the macroscopic and unperturbed angles were nearly the same in most systems, considering the form of eqn (1) and (9). The equality of the two angles has been suggested by White5 and Pethica,* but only a detailed analysis of the molecular interactions can confirm this possibility. The fundamental equations derived in this paper [eqn (8) and (9)] apply to cylindrically symmetric interfaces down to dimensions of the order of the range 6 of the interaction, and to axially symmetric interfaces of dimensions greater than ca. 1006, in contact with a planar solid surface. However, considering the short-range nature of the interaction, it is expected that the same relations apply to general geometries of the solid and fluid interface. It was assumed that the ratio of the perturbed interfacial tensions in the second term of eqn (2) is in the interval [ - 1, I].For values outside this interval, eqn (2) is not applicable and situations of complete wetting (spreading) will occur. However, even if eqn (2) gives an acceptable value for cos Om, it may occur that cos 8, calculated from eqn (8) has a value outside the permissible interval. The actual macroscopic angle is probably 0 or n in this case. If this is correct, then a zero (or n) macroscopic angle of contact may not imply a zero (or n) microscopic angle. Finally, note that centrifugal fields probably give rise to a macroscopic angle of contact which depends on the intensity of the field (although 8, does not). This is because an additional term in x2 will appear in eqn (4a) and (4b). An analysis of the effect of a centrifugal field on the macroscopic angle of contact will be published el sew here. Computations were carried out by Mr Braz Fernandes of the Universidade Nova de Lisboa. Financial support by Centro de Meciinica e Materiais da Universidade Tecnica de Lisboa (CEMUL) is acknowledged. T. Young, Miscellaneous Works, ed. Peacock (Murray, London, 1 809), vol. 1. A. W. Adamson, Physical Chemistry of Surfuces (Wiley, New York, 1976). M. V. Berry, J. Phys. A , 1974, 7, 231. G. J. Jameson and M. C. G. del Cerro, J. Chem. Soc., Faraday Trans. I , 1976, 72, 883. L. R. White, J. Chem. Soc., Faraday Trans. 1, 1977, 73, 390. G. Saville, J . Chem. SOC., Faraday Trans. 2, 1977, 8, 1122. B. A. Pethica, J. Colloid Interface Sci., 1977, 62, 567. J. W. Gibbs, The ScientGc Papers (Dover, New York, 1961), vol. 1, p. 288. ? J. A. de Feijter and A. Vrij, Electroanal. Chem. Interfacial Electrochem., 1972, 37, 9. lo M. A. Fortes, Phys. Chem. Liquids, 1980, 9, 285. (PAPER O/ 18 18)

ISSN:0300-9599    

DOI:10.1039/F19827800101

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

J. Chem. SOC., Faraday Trans. I, 1982, 78, 109-118 Effects of Preferential Solvation and of Solvent-Solvent Interaction on the Rates of Nucleophilic Substitution involving Anions in Binary Mixed Solvents Theoretical Approach BY YASUHIKO KONDO* AND SHIGEKAZU KUSABAYASHI Department of Applied Chemistry, Faculty of Engineering, Osaka University, Suita, Osaka 565, Japan Received 8th December, 1980 Theoretical procedures for investigating rate constants and activation parameters measured in binary mixed solvents have been presented on the basis of the concept of ideal associated mixtures. In methanol + acetonitrile mixtures the behaviour of the rate constant and of activation parameters for the ethyl iodide plus bromide ion reaction were interpreted as resulting from the specific interaction of bromide ion with methanol.In methanol + NN-dimethylacetamide mixtures association complex formation between methanol and NN-dimethylacetamide makes a significant contribution to the activation parameters, and this factor must be taken into account in interpreting the observed rate behaviour. Recently, the importance of a chemical interaction between an electrolyte and solvents has been stressed in interpreting the solution thermodynamic properties of electr~lytes.l-~ In favourable cases various kinds of solvates have been detected by n.m.r. techniques, which provides supporting evidence for the ~oncept.~ In the previous work9 and in the experimental part of this work,l0 various aspects of anion solvation and effects on the rates of nucleophilic substitution reactions have been investigated.One of the salient features of these studies is the fact that the transfer enthalpy of bromide ion and the rates of the nucleophilic substitution reaction respond characteristically to the thermodynamic properties of the solvent mixtures, depending on whether the pair of solvents shows an exothermic or an endothermic heat of rni~ing.~9 lo In this work we will be concerned with the theoretical approach to the effects of preferential solvation of anions and of solvent-solvent interactions on the rates of a nucleophilic substitution reaction in binary mixed solvents. The rates of the bromide ion plus ethyl iodide reaction in methanol + acetonitrile and in methanol + NN- dimethylacetamide mixtures will be analysed using this approach. Also, we will discuss how the information obtained through experimental observations is considered in the model treatment.THEORETICAL For the reaction type considered, A+ B -+ M# -+ products, 109110 NUCLEOPHILIC SUBSTITUTION IN BINARY SOLVENTS the rate constant in solvent j , ki, is related to the rate constant in the gas phase, k,, by the equation1’ where Pir is the activity coefficient of the solute i in solvent j , and is defined by eqn (2)11 mgi is the concentration of solute i in the vapour phase above the solution. m,, is the concentration of solute i in the solution. Dij = (mgi/mli)j- (2) The Henry’s constant of solute i in solvent j , Hii, is defined by eqn (3)12 Hii = lim X i 4 0 (3) wheref, and xi are the fugacity and the mole fraction of solute i, respectively.At extremely low concentrations of solute i the ideal gas law, eqn (4), holds and the solute concentration is related to the mole fraction of the solute and the molar volume of the solvent, 5, by eqn ( 5 ) : (4) fi = pi = (ngi/V) RT = mgiRT mli = n l i / ( x nli K) z x i / 5. ( 5 ) i Thus the activity coefficient defined by eqn ( 2 ) can be found from the Henry’s constant and the molar volume of the solvent /Iij = (Hii Y / R T ) . (6) By following this procedure the original problem of solution kinetics can be In ideal associated mixtures the fugacity of solute 2, fi, is related to that of pure transformed into that of non-electrolyte solution thermodynamics. solute,fz, by eqn (7)12 f i =f20y2 (7) where y2 is the mole fraction of solute defined using the number of moles actually existing in solution.On the other hand, Henry’s constant is defined by using the stoichiometric mole fraction, xi, by eqn ( 3 ) . The combination of eqn (3), (6) and (7) reduces to the required equation for the activity coefficient of solute 2 in ideal associated mixtures 82, mix = The next step is to calculate the for a relevant model. MODEL In the following treatment the subscripts 1 and 4 stand for solvent components and (1) Solvent 4 forms various association complexes with solvent 1, each of which the subscript 2 for a solute. We consider the following ideal associated mixtures.Y. KONDO A N D S. KUSABAYASHI 111 is composed of one molecule of solvent 4 and k molecules of solvent 1, i.e. s4 * (Sl)k, and the equilibrium constants for the processes are given by the relations S4+S1 = S4*S1; Kc,l = K S4’(Sl)k-1+S1 = S4’(sl)k; Kc,k = (l/k)K* (2) A solute, M, is in equilibrium with clusters that contain (z-j) molecules of solvent 4, and j molecules of solvent 1, i.e.M-(S4)z-j-(Sl)j, and the equilibrium constants of the process are given as follows: M + z S ~ + M * ( S ~ ) ~ ; KO M * (S4)z-j+l* (Sl)j-l+ S1 =t M * (S4)z-j (S1)j + S4; Kj. (3) All the equilibrium constants are expressed using the mole fraction defined by the number of moles in solution as follows: &, k = Y c , k / b 1 Yc, k-1) Kj = ~ 2 , j ~ 4 / c Y 2 , j - 1 YJ where c refers to the complex. The following i before mixing, material balances hold between the number of moles of component Ni, and the number of moles in solution, Mi, Mc, k and M2, j, 00 Z 00 2 where and M2,j stand, respectively, for the number of moles of the associated complex containing k molecules of solvent 1, and of the solute cluster composed of (z -5) molecules of solvent 4 and of j molecules of solvent 1.The stoichiometric mole fraction, xi, and the mole fraction counted by the number of moles in solution, yi and Yc,k, are defined by the equations xi = NJN, (i = 1,2,4) (12) where Nt = Ni+N2+N4 (1 3) yi = MJM, (i = 1,2,4) (14) Yc,k = (k = l-O0) (15) m z From the basic assumptions (1) and (3) for the system, and making use of the rela tion exp(x) = 1 +x+(x2/2!)+(x3/3!)+. . .t 12 NUCLEOPHILIC SUBSTITUTION IN BINARY SOLVENTS In the case of no added solute, letting M, = M,,j = 0 (for all values ofj), eqn (16) reduces to eqn (20), making use of eqn (19): Similarly the stoichiometric mole fraction can be related to the mole fraction y1 by combining eqn (9)-(16), (18) and (20): On rearranging, the final formula for y1 as a function of solvent composition x, is (22) derived, KX4y21-(Kxq+ 1)y1+ 1-x4 = 0.Limiting values of (y,/x,) can be expressed as a function of mole fraction yf by combining eqn (9)-(16) and (18), which finally reduces to eqn (23) (23) 1 +KYl(l -u1) - lim (y,/x,) = - 1 + ky1(1 -Yl) 22'0 j-0 making use of assumption (3). Alternatively, for the system in which a solvent exchange process can be defined on each solvation site around the ion, the term which expresses the distribution of clusters, i.e. is rewritten as z n CV4+Kse,rn~1), m-1 making use of the equilibrium constants defined for the solvent exchange process on the respective solvation site rn around the ion, Kse,m.g?13 Finally the activity coefficient of solute 2 in the mixed solvent is derived as m-1 / For an electrolyte the assumption KO >> 1 would not be unreasonable; thus the activity coefficient in a mixed solvent as a relative value to that in the reference solvent 1 is given by eqn (25): logp,,mix = logP2,1 +log [1 +KY~(' -~1)1 2 z +log II K,e,m-log n ( Y 4 + K s e , m ~ l ) + l o g ( ~ m i x / ~ ) .(25) m-1 m-i For the system in which there is no association between components 4 and 1, substituting the relations y , = x, and K = 0 into eqn (25) gives eqn (26): z zY. KONDO A N D S. KUSABAYASHI 113 In this treatment the transfer free energy of a solute from solvent 1 to a mixed solvent is given by the relation AG: = RT In Thus the transfer free energy derived in this work has essentially the same functional dependence on the solvent composition as derived on a different except for the extra term, log( Vmix/ K) contained in eqn (26) because of the different standard state adopted in previous treatrnent~~g - RT In &.APPLICATION TO REACTION RATES As was discussed previouslyg and will be discussed in the experimental part of this work,l0 for the bromide ion plus ethyl iodide reaction the electrostatic contribution to the transfer activation parameters are largely cancelled out between the initial and the transition state, and as a result rates are mostly controlled by the specific solvation of bromide ion by methanol.Also, it was suggested that in amide +methanol mixtures only free methanol participates in specific solvation of bromide ion, and associated methanol acts as only neutral solvent; thus y , in eqn (25) should be replaced by (1 --y1).99 lo Considering these suggestions and substituting eqn (25) into one of the activity coefficients of the reactants in the fundamental equation (l), the rate constant in mixed solvents is derived as eqn (27): z z Eqn (27) contains many parameters to be adjusted. To understand the general trends implied by the equation it is appropriate to start the analysis with a simplified formula. For the reaction in non-associated media, on letting K = 0, eqn (27) reduces to eqn (28), and on letting x, = 1 in eqn (28), eqn (29) is also derived: z z This equation, eqn (29), serves as the boundary condition for the value of Kse,m.by the equation Performing the differentiations and making use of the relation The activation enthalpy is defined as the temperature derivative of the rate constant (a1nkmix/aT) = (AHgiX+RT)/RT2. (a In Kse , m /a T = AH;, m /RT2 the following equations are derived. [Throughout this work, of the term ( Vmix/ K) will be neglected.] temperature derivatives z AHf-AH,Z = AHze,,. m-1114 NUCLEOPHILIC SUBSTITUTION I N BINARY SOLVENTS The reaction proceeds much faster in acetonitrile than in methanol. If we take acetonitrile as solvent 4 and methanol as solvent 1, this leads to the relation 2 n Kse,m 1 m-1 on the basis of eqn (29).Substituting the activation enthalpies in acetonitrile and in methanol into eqn (31) the relation is also derived. These relations confirm the view of specific solvation of bromide ion by methanol reached from transfer-enthalpy analy~is.~, lo Calculations of the rate constants have been repeated assuming various sets of Kse,m values under the condition of eqn (29), and the optimum agreement was reached assuming three steps of solvent exchange processes.* Considering the success of the model calculation on the transfer enthalpy of bromide ion in mixed solvent^,^ and partly because of the brevity of the treatment, all three AHZe,, values were assumed to be equal. Under this condition, theoretical values of the activation enthalpy can be calculated from eqn (30) without recourse to any adjustable parameters, since the value AH& can be determined uniquely by eqn (31), and Kse,m values have already been determined in the rate-constant analysis.From the theoretical values of log kmix and of AH&, activation entropies have also been calculated. They are compared with the experimental values in fig. 1-3. For the sake of the consistency of the treatment in the various series and of the brevity of the analysis, the rate behaviour in NN-dimethylacetamide + methanol mixtures has been analysed under the presumption of a binominal distribution of clusters with three steps of solvent exchange equilibria. Under this condition, eqn (27) reduces to eqn (32) and (33), and eqn (33) then gives the boundary condition for Kse,m : Again, differentiating eqn (32), (33) and (22) with respect to temperature and making (a lnK/aT) = AHo/RT2, use of the relation theoretical equations for the activation enthalpy were derived as follows : AH? -AH? = 3AHie,, (35) * Throughout the work the density of the mixed solvent has been calculated from those of pure solvents from the equation dmix = xld1+x4d4.Y.KONDO AND S. KUSABAYASHI 115 t (x 1) FIG. 1 .-Plots of log kmix against mole fraction of methanol for the reaction of ethyl iodide with bromide ion (30 OC): 0, experimental results in methanol + acetonitrile mixtures; 0, experimental results in methanol + NN-dimethylacetamide mixtures; (-) calculated by eqn (28) with K,,,, = 75.0, = 7.16 It is clear from eqn (33) that the value is uniquely determined from the rate constants in pure solvents.Thus, in rate-constant analysis only one parameter, K, remains to be adjusted. For an assumed value of K, y1 values have been calculated from eqn (22), and then theoretical values of logkmi, have been calculated from eqn (32) as a function of solvent composition. Similarly, the AHZe,, value was uniquely determined from eqn (35). The values of K, Kse,m and of y , have already been determined. For an assumed value of AH', the values of RT2 (3yl/aT) have been calculated from eqn (36). On substituting all these values into eqn (34), theoretical values of AH&x have been obtained. Activation entropies have also been calculated from these theoretical values. In these analyses NN-dimethylacetamide was taken as solvent 4 and methanol as solvent 1.The relations, Kse,m > 1 and AHZe,m -= 0, derived from the substitution of the experimental results into eqn (33) and (35) under the above notation, also agree with the views attained from the transfer-enthalpy analysi~.~~ lo All the optimum values are given in fig. 1-3.116 NUCLEOPHILIC SUBSTITUTION IN BINARY SOLVENTS 9 5.0 90.0 - I - i? 3 $ 80.0 \ 7 0.0 I .o 0.5 0 (x 1) FIG. 2.-Plots of AH,$ix against mole fraction of methanol: 0, experimental results in methanol + aceto- nitrile mixtures; 0, experimental results in methanol + NN-dimethylacetamide mixtures; (-) calculated by eqn (30) with AH,",,,, = -5.9 kJ mol-l and the set of K,,,, values determined above; (-) calculated by eqn (34) with = -8.2 kJ mol-l, AHo = -8.8 kJ mol-I, and the set of K,,,, and of K values determined above.DISCUSSION In the analysis of the rate behaviour in methanol + acetonitrile mixtures, three steps of solvent exchange processes and also strong preferential solvation of bromide ion by methanol have been suggested. Transfer-enthalpy analysis also supports the view of preferential solvation of bromide ion by methan01.~ However, as is partly inferred from measurements of the partial molal heat of mixing,g, lo the activity coefficient of methanol is significant in the region of low methanol content, and this might possibly bring about phenomena which would be taken as originating from the preferential solvation of bromide ion. A detailed analysis of the preferential solvation will have to be postponed until exact activity coefficients are available.Recently Cox and Waghorne noticed the sharp minimum which appeared in the transfer entropy against composition profile for alkali metal ions, and ascribed the phenomenon to the preferential solvation of these ions by one of the component s01vents.l~ The activation entropy maximum observed in methanol + acetonitrile mixtures in this work corresponds to their observations and would be interpreted on the basis of their theory as preferential solvation of bromide ion by methanol. In our treatment, however, the main features in the three activation parameters could be-20.0 - c1 I 2 - I !& +tE r, 1 .?I - 40.0 2 - 6 0.C 1.0 Y. KONDO AND S. KUSABAYASHI 0.5 (x 1) 0 117 FIG. 3.-Plots of AS& against mole fraction of methanol: 0, experimental results in methanol + aceto- nitrile mixtures; , experimental results in methanol + NN-dimethylacetamide mixtures; (-) calculated in methanol + acetonitnle mixtures; (-) calculated in methanol + NN-dimethylacetamide mixtures.reproduced consistently without recourse to an ad hoc expression for the respective activation parameters. Comparative studies in two thermodynamically contrasted systems clearly indicate the contrasting behaviour in the activation parameters; especially, maximum and minimum values in the activation entropy against composition profiles are very pronounced. Solvent-solvent interactions seem to play a significant role in associated mixtures. An amide molecule could interact with methanol at least at three sites, i.e. the carbonyl z-electron, the oxygen lone pair and the nitrogen lone pair.In addition, on dissolving into methanol there would be formed new structures around each amide molecule. This phenomenon would be non-stoichiometric in origin and could not be simulated by any stoichiometric means. In this work the interaction has been taken into account as infinite-chain association with attenuation in calculating equilibrium constants. According to the model, the interaction manifests itself in the activation parameters through various forms, i.e. y l , K, AH', and (2yJi37'). The calculation clearly simulates the interactions in the region of high methanol content, as partly inferred from the minimum in the activation entropy against composition profile. Grunwald and Effio warned against the conventional use of the constant- composition condition in analysing the solution phenomena observed in mixed118 NUCLEOPHILIC SUBSTITUTION IN BINARY SOLVENTS They put forward a method of avoiding the complexity originating from the non-ideal nature of thermodynamic quantites of the solvent mixtures, although their procedures were criticized later? In contrast, the model approach as performed in this work can provide a way of investigating the origin of such complexities in solution kinetics and thermodynamics, and in favourable cases can give information that is not otherwise obtainable. This suggests that the experimental analysis should proceed in conjunction with theoretical approaches.We thank Dr M. H. Abraham, University of Surrey, for helpful discussions on the original version of this manuscript. E. Grunwald, G. Baugham and G. Kohnstam, J . Am. Chem. SOC., 1960,82, 5801. C. V. Krishnan and H. L. Friedman, J . Phys. Chem., 1969, 73, 1572. Y. Kondo and N. Tokura, Bull. Chem. SOC. Jpn, 1972, 45, 818. A. K. Covington, T. H. Lilley, K. E. Newman and G. A. Porthouse, J. Chem. SOC., Faraday Trans. I , 1973, 69, 963, 973. A. K. Covington and K. E. Newman, Pure Appl. Chem., 1979,51,2041, and references cited therein. B. G. Cox, A. J. Parker and W. E. Waghorne, J. Phys. Chem., 1974, 78, 1731. 'I G. Clune, W. E. Waghorne and B. G. Cox, J. Chem. SOC., Faraday Trans. I , 1976, 72, 1294. * E. M. Amett, B. Chawla, L. Bell, M. Taagepera, W. J. Hehre and R. W. Taft, J. Am. Chem. SOC., 13 Y. Kondo, K. Yuki, T. Yoshida and N. Tokura, J . Chem. SOC., Faraday Trans. I , 1980, 76, 812. lo Y. Kondo, M. Itto and S. Kusabayashi, to be published. l1 S. Glasstone, H. Eyring and K. J. Laidler, The Theory of Rate Processes (McGraw Hill, New York, l2 I. Prigogine and R. Defay, Chemical Thermodynamics, translated by D. H. Everett (Longmans, l3 Y. Kondo, T. Kato and N. Tokura, Bull. Chem. SOC. Jpn, 1975, 48, 285. l4 B. G. Cox, W. E. Waghorne and C. K. Pigott, J. Chem. SOC., Faraday Trans. I , 1979, 75, 227. l5 E. Grunwald and A. Effio, J. Am. Chem. SOC., 1974, 96, 423. l6 G. L. Bertrand and T. F. Fagley, J . Am. Chem. SOC., 1976,98, 7944. 1977, 99, 5729. 1941). London, 1954). (PAPER O/ 1897)

ISSN:0300-9599    

DOI:10.1039/F19827800109

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

J. Chem. SOC., Faraday Trans. 1, 1982,78, 119-125 New Approach to the Evaluation of Single-ion Conductances in Pure and Mixed Non-aqueous Solvents [Part 21 BY DIP SINGH GILL* AND MADHU BALA SEKHRI Department of Chemistry, Himachal Pradesh University, Simla- 17 1005, India Received 9th December, 1980 Equivalent conductances of tetrabutylammonium tetraphenylborate (Bu,NBPh,) in acetonitrile (AN) and in several acetonitrile+ benzene (AN + BN) mixtures covering the dielectric constant range 12.3 < E < 36.0 have been measured at 25 OC. Limiting ion conductances ,I: for Bu,N+ and Ph,B- have been determined by the method of Fuoss and coworkers; the ion conductances thus obtained are in quantitative agreement (within an average of +2%) with the values predicted by an empirical equation already proposed from our laboratory.These investigations thus provide further positive support for our new method for the evaluation of limiting ion conductances. In a previous paper, we have proposed an equation for the evaluation of limiting ion conductances dp in pure and mixed non-aqueous so1vents.l This equation predicts limiting ion conductances of Bu,N+, Pr,N+ and Et,N+ in pure and mixed non-aqueous solvents in quantitative agreement with the experimentally determined values. In the present paper we report some conductance measurements of Bu,NBPh, in pure acetonitrile (AN) and in several acetonitrile + benzene (AN + BN) mixtures covering the dielectric constant range 12.3 < E < 36.0 at 25 O C to provide further positive support for the usefulness of our equation for evaluating limiting ion conductances in pure and mixed non-aqueous solvents.EXPERIMENTAL Acetonitrile (E. Merck, 99 % purity) was purified by fractional distillation (two times) over P,O, through a long vertical column. The purified solvent with density 0.7766 g ~ m - ~ , viscosity 0.341 x kg m-' s-l (cP), dielectric constant 36.0 and specific conductance (1-4) x lo-* S2-l cm-l (all at 25 "C) was used. These physical constants agree well with the literature2 and also with our previously reported value^.^ The water content of the solvent was found to be < 50 ppm. The purified solvent was used immediately after distillation. Benzene (B.D.H. AnalaR of 99.5% purity) was refluxed over sodium metal for 8-10 h and was then slowly fractionated through a long vertical column.The purified solvent had a density of 0.8735 g C M - ~ and viscosity of 0.608 kg m-ls-l (cP), values which agree well with the literature values., Bu,NBPh, was prepared by the method of Accascina et aL5 Conductances were measured with a calibrated Toshniwal conductance bridge model CL10/02A at a frequency of 3000 Hz. A conductance cell similar in design to that reported by Shedlovsky6 with bright platinum electrodes was used. The cell constant was determined following the method of Fuoss and coworkers7 using aqueous potassium chloride solutions in the concentration range (3-70) x lo-, mol dm-3. All measurements were carried out at 119120 SING LE-ION CONDUCT A N C ES TABLE 1 .-EQUIVALENT CONDUCTANCES, A, AND MOLAR CONCENTRATIONS, c, OF Bu,NBPh4 IN ACETONITRILE AND ACETONITRILE + BENZENE MIXTURES AT 25 O C c/ 10-4 A c/ 10-4 A c/ 10-4 A 100% AN 88% AN 56% AN 9.744 110.1 1 21.02 104.39 28.30 102.02 35.40 99.50 42.32 97.47 49.08 97.04 55.67 94.99 62.10 93.60 94% AN 9.299 16.38 23.33 30.09 36.67 43.07 49.3 1 61.29 106.00 103.02 99.66 97.70 95.5 1 94.03 92.78 91.29 22.77 94.52 28.69 93.05 34.49 90.22 42.06 88.27 47,60 86.65 54.83 85.49 61.87 84.24 74% AN 5.610 96.26 17.76 90.9 1 28.01 87.57 32.97 86.44 37.78 84.37 44.82 82.83 66% AN 5.150 80.10 8.930 76.85 23.42 69.45 29.3 1 65.25 33.01 64.53 37.85 62.42 51% AN 82.16 1.442 4.350 78.43 8.67 1 74.14 13.47 71.87 15.06 69.72 19.41 67.62 21.63 65.88 45% AN 5.400 88.80 10.60 84.92 14.53 82.63 19.70 79.90 23.30 78.31 26.90 76.79 3.856 71.97 13.98 63.32 16.03 62.25 18.69 59.38 21.30 58.1 1 23.83 56.65 TABLE DERIVED CONDUCTANCE PARAMETERS FOR Bu4NBPh, IN ACETONITRILE AND ACETONITRILE + BENZENE MIXTURES AT 25 *C BY SHEDLOVSKY'S METHOD 100 94 88 74 66 56 51 45 36.0" 33.3b 29.7b 22.9b 19.4b 1 5.7b 14.0b 12.3b 0.341" 0.355b 0.367b 0.395b 0.414b 0.43gb 0.450b 0.466b 1 19.96 1 15.69 11 1.40 105.74 99.28 91.96 89.25 84.32 11 9 13 13 21 31 25 52 a Ref.(3); ref. (11).D. S. GILL AND M. B. SEKHRI 121 25.00+0.01 O C . AN+BN mixtures were prepared by weight and in each case a range of concentrations of the salt was produced by adding stock solutions of appropriate concentrations from a weight burette to a known quantity of the solvent mixture taken in the conductance cell. In all cases the measurements were repeated with different stock solutions to obtain reproducible results.The reproducibility of the conductance measurements is ca. 0.2%. RESULTS AND DISCUSSION The measured equivalent conductances, A, and the corresponding molar concen- trations, c, of Bu,NBPh, in AN and in AN+BN mixtures at 25 O C are reported in table 1. The association constant, KA and the limiting equivalent conductance (A,) in all cases have been iteratively calculated by a least-squares treatment with an IBM 1620 computer using Shedlovsky’s rnethods-lo which involves the following set of equations - 1 SA - S = D = and a = 8.204 x 105A, 82.5 +--- (&T)$ 7](ET)* 1.8246 x lo6 ( c a ) i / ( ~ T ) i 1 + 50.29 x los R (ca)l/(&T)t - For the analysis of conductance data, values of the dielectric constant ( E ) and viscosity (r) for AN and AN + BN mixtures were taken from the literature3.l1 and are also reported in table 2. Justice12 has suggested that the Bjerrum critical distance, e2 q = - 2 ~ k T should be used for the calculation of mean-ion activity coefficients from eqn (4). Our derived conductance parameters reported in table 2 have also been obtained after setting R = q in eqn (4). The standard deviations in A, and KA values of table 2 obtained by applying standard’statistical equations13 were found to be always less than +0.2% and lo%, respectively. The root mean-square deviations oA calculated from the standard deviations of the individual points in no case exceeded the experimental uncertainty of the present conductance measurements, i.e.0.2 %. This shows the good applicability of Shedlovsky’s equation to our conductance data. The conductance measurements in the present work have been made over a wide range of dielectric constant (36.0 2 e 2 12.3) and the precision of our conductance data is ca. +0.2%. The use of all other conductance equations which demand a 5 FAR 1122 SING LE-ION COND U C T A N CES precision in the conductance data much better than +0.1% were not thought appropriate for the analysis of the present data. Fuoss and c o - w ~ r k e r s ~ ~ ~ ~ ~ also preferred to analyse their conductance data for a number of solvent mixtures where the measurements were made over a wide range of dielectric constant by the Fuoss meth0d,~7 l6 which is closely similar to Shedlovsky's method.Fuoss and Shedlovsky9 TABLE 3.-EXPERIMENTAL AND THEORETICALLY CALCULATED LIMITING ION CONDUCTANCES FOR Bu,N+ AND Ph,B- IN ACETONITRILE AND ACETONITRILE + BENZENE MIXTURES AT 25 "C mol % RP(Bu,N+) % AN & 1.4% AZ(BU,N+)~ difference 100 94 88 74 66 56 51 45 62.26 60.04 57.82 54.88 51.53 47.73 46.32 43.76 63.62 60.63 58.12 53.04 49.76 46.94 45.49 43.74 - 2.2 - 1.0 - 0.6 3.4 3.5 1.7 1.8 0.1 mol % RP(Ph,B -) % AN & I % Az(Ph,B-)b difference 100 94 88 74 66 56 51 45 57.70 55.65 53.58 50.86 47.75 44.23 42.93 40.56 58.24 - 55.57 53.27 48.68 46.06 43.16 41.84 40.24 .1.0 0.2 0.6 4.3 3.6 2.5 2.6 0.8 a For the calculation of these values from eqn (6), ri was taken as 5.00 A and r y as 0.85 A For the calculation of these values from eqn (6), ri was taken as 5.35 A and ; have shown that the two treatments yield the same values of A.and slightly different values of association constant; in the range lo3 >, KA >, 1 Shedlovsky's treatment should be preferred for extrapolation. As our KA values for Bu,NBPh, lie in this range, we have thus preferred to analyse our conductance data by Shedlovsky's method and not by the Fuoss method. To give an indication of the precision of the present conductance measurements, our A, value, 119.96 i2-l cm2 molt1 for Bu4NBPh, in pure AN from table 2, can well be compared with the A. value of 119.7 C2-I cm2 mol-l reported by Coetzee and Cunningham.17 These values are in agreement with each other within +0.2%, i.e. the claimed accuracy of the present measurements. LIMITING ION CONDUCTANCES Using the assumption of Fuoss and c o ~ o r k e r s ~ ~ ~ l5 (which is claimed to be valid within 1 %) that the limiting transference number of Bu,N+ in Bu,NBPh, is 0.519 and is independent of the so1vent,l49 l5 limiting ion conductances 2; for Bu,N+ andD.S. GILL AND M. B. SEKHRI 123 Ph,B- in pure AN and in AN + BN mixtures have been calculated from the A,, values of Bu,NBPh, from table 2. These values are reported as A:(Bu,N+) and Ay(Ph,B-) in table 3. The validity of the assumption of Fuoss and coworkers can be checked from a comparison of the limiting ion conductances of Bu,N+ and Ph,B- in pure AN from table 3 with such values obtained from direct transference-number measurements in AN. The limiting ion conductances of 62.26 and 57.70 0-l cm2 mol-1 for Bu,N+ and Ph,B- in AN in table 3 are within 1.4 and 1 .Ox, respectively, in agreemenf with the values 61.4 and 58.3 0-l cm2 mol-l reported by Springer et al.lS from direct transference-number measurements in AN.PREDICTION OF LIMITING I O N CONDUCTANCE OF Bu,N+ I N ACETONITRILE In a previous paper1 we have proposed an equation which theoretically predicts limiting ion conductances of Bu,N+, Pr,N+ and Et4N+ in pure and mixed non-aqueous solvents within an average uncertainty of +2%. This equation can be written in the form AND ACETONITRILE+BENZENE MIXTURES 27 = IZIF2/(6~N~)[ri-(0.0103 &+ry)] (6) TABLE 4.-LIMITING ION CONDUCTANCES AP(Ph,B-) AND ri VALUES FOR Ph,B- IN SOME NON- AQUEOUS SOLVENTS AT 25 "C solvent & VlCP AP(Ph,B-) riIA acetoni trile i-butyronitrile n-butyronitrile nitromethane nitrobenzene acetone NN-dimethylformamide ethyl methyl ketone dimethylsulphoxide propylene carbonate methanol ethanol 36.0a 23.8lC 24.26d 36.7a 34.3h 20.7j 37.6a 18.014" 46.6a 65.0a 32.6a 24.3a 0.341" 0.48Y 0.553d 0.61 18f 1 .839i 0.304 0. 796a 0.3774" 1 .990a 2.48a 0.545" 1.084" 58.3b 39.3OC 34. 50d 31.57f.g 10.79f. 62.65k 24.5l 50.1 6"3 10.6lng 8.26* 37.054 19.34' 5.34 5.40 5.40 5.48 5.33 5.37 5.44 5.37 5.48 5.52 5.53 5.29 a Ref. (3); ref. (18); ref. (14); A. D'Aprano and R. M. Fuoss, J. Solution Chem., 1974, 3,45 ; calculated from (i-Am),BuNBPh, assumption; f ref. (1 7) ; based on (i-Am), NB(i-Am), as a reference electrolyte: E. Hirsch and R. M. Fuoss, J . Am. Chem. SOC., 1960,82,1018;i R. A. RobinsonandR. H. Stokes, Electrolyte Solutions (Butterworths, London, 1959), p.458; D. F. Evans, J. Thomas, J. A. Nadas and M. A. Matesich, J. Phys. Chem., 1971, 75, 1714; ' V . M. Tsentovskii, V. P. Barabanov, N. K. Mochalov and N. A. Tumasheva, Zh. Obshsch. Khim., 1974,44,1938; " S . R. C. Hughes and D. H. Price, J. Chem. SOC. A , 1967, 1093; D. E. Arrington and E. Griswold, J . Phys. Chem., 1970,74, 123; P A,, value for Bu,NBPh, reported by R. M. Fuoss and E. Hirsch, J. Am. Chem. Soc., 1960, 82, 1013 was corrected for a viscosity ratio of 2.55312.48 and from the resulting A. value RP(Ph,B-) was calculated by combining this A. value with the ionic conductance of Bu,N+ from L. M. Mukherjee, D. P. Boden and R. Lindauer, J. Phys. Chem., 1970, 74, 1942; Q calculated by combining A?(Bu,N+) = 38.94 from the results of R.L. Kay, C. Zawoyski and D. F. Evans, J. Phys. Chem., 1965, 69, 4208 with the A. value 75.99 for Bu,NBPh, reported by M. A. Coplan and R. M. Fuoss, J. Phys. Chem., 1964, 68, 1177; ' S. Schiavo and G. Marrosu, Z . Phys. Chem. (N.F.), 1977, 105, 157. R. H. Boyd, J . Chem. Phys., 1961, 35, 1281; 5-2124 SINGLE-ION CONDUCTANCES where ri is the crystallographic radius of the ion and ry is the empirical constant 0.85 A for dipolar non-associated s01vents.l~~ 2o The validity of eqn (6) for Bu,N+ in AN and AN + BN mixtures has been tested by calculating A: values theoretically, using eqn (6) and substituting ri and r y values equal to 5.00 and 0.85 A, respectively.' The theoretically calculated A: values for Bu4N+, which are recorded as AE(Bu,N+) in table 3, are in excellent agreement (within an average uncertainty of Ifr 1.8%) with the AP(Bu,N+) values in the same table.TABLE 5.-LIMITING ION CONDUCTANCES A:(Ph,B-) AND ri VALUES FOR Ph,B- IN SOME MIXED NON-AQUEOUS SOLVENTS AT 25 "C solvent mixture & q/cP Ay(Ph4B-)d ri/A AN + carbon tetrachloride" 19.87 17.18 13.92 11.16 1-BN + chlorobenzeneb 20.98 16.26 11.99 9.90 i-BN + o-dichlorobenzeneb 20.86 15.85 11.76 9.99 i-BN + 1,2-di~hloroethane~ 18.46 14.63 12.61 10.34 i-BN + benzeneC 19.04 14.79 10.38 i-BN + dioxanc 18.98 13.95 10.31 0.4676 0.5042 0.5566 0.5992 0.527 0.596 0.661 0.696 0.638 0.916 1.184 1.317 0.659 0.735 0.768 0.795 0.477 0.482 0.503 0.541 0.630 0.720 40.90 37.70 33.69 30.87 36.60 32.27 29.00 27.06 30.79 21.09 15.86 14.56 29.40 26.17 25.04 24.12 39.24 38.24 36.34 34.94 30.01 26.17 5.34 5.34 5.36 5.39 5.32 5.28 5.25 5.30 5.23 5.25 5.34 5.23 5.27 5.26 5.24 5.24 5.43 5.44 5.45 5.38 5.34 5.30 a D.S. Berns and R. M. Fuoss, J. Am. Chern. Soc., 1960, 82, 5585; ref. (15); ref. (14); A:(Ph,B-) in all mixed solvents were obtained from the A,, values of Bu,NBPh, using the assumption of Fuoss and coworkers, ref. (14) and (1 5). TEST OF THE VALIDITY OF EQN (6) FOR Ph,B- I N PURE AND MIXED NON-AQUEOUS SOL VENTS Eqn (6) contains the ri parameter which is equal to the crystallographic radius of the ion.' The crystallographic radius of Ph,B- is not accurately known in literature, therefore, the validity of eqn (6) for A:(Ph,B-) can not be tested as such. AP(Ph,B-), however, can be accurately determined both in pure and mixed non-aqueous solvents from the conductance data available in the literature.Therefore, for a test of the validity of eqn (6) for Ph,B- it would be meaningful to evaluate first the ri values by using the literature value of iZP(Ph,B-) and examining whether ri in pure and mixed non-aqueous solvents remains constant. If it remains constant within experimental uncertainty this constant value of ri can then be used to calculate AP(Ph,B-)D. S. GILL AND M. B. SEKHRI 125 theoretically in AN and AN+ BN mixtures; these values can then be compared with our experimental values of table 3. Taking accurate values of Ap(Ph,B-) in various pure and mixed non-aqueous solvents from the literature (as reported in tables 4 and 5 ) and the corresponding E and values for the pure solvents or the solvent mixtures, ri values have been calculated using eqn (6) and substituting ry = 0.85 Al. These ri values are reported in tables 4 and 5.A careful examination of tables 4 and 5 shows that ri values for Ph,B- in pure non-aqueous solvents (table 4) and in mixed non-aqueous solvents (table 5 ) calculated using eqn (6) remain constant and equal to 5.35 A within the average uncertainty of ca. 2%. PREDICTION OF LIMITING I O N CONDUCTANCE OF Ph,B- I N ACETONITRILE AND ACETONITRILE+BENZENE MIXTURES Using ri = 5.35 A for Ph,B- in eqn (6), limiting ion conductances of Ph,B- in pure AN and in AN + BN mixtures have been theoretically calculated. These values have been reported as A:(Ph,B-) in table 3 for comparison with our experimental values.The agreement between the AE(Ph,B-) and AP(Ph,B-) values of table 3 is very good (within an average uncertainty of *2%), thereby showing that eqn (6) predicts limiting ion conductances for Ph,B- also in quantitative agreement with the experi- mental values. M. B. S. thanks the C.S.I.R., New Delhi for a Junior Research Fellowship. D. S. Gill, J. Chem. Soc., Faraday Trans. I, 1981, 77, 751. J. A. Riddick and W. B. Bunger, Organic Solvents (Wiley-Interscience, New York, 1970). D. S. Gill, J. Solution Chem., 1979, 8, 691. I. N. V’yunnik, A. M. Zholnovach and A. M. Shkodin, Zh. Obshch. Khim., 1977, 47, 1681 F. Accascina, S. Petrucci and R. M. Fuoss, J. Am. Chem. Soc., 1959, 81, 1301. ti T. Shedlovsky, J . Am. Chem. Soc., 1932, 54, 141 1. J. E. Lind, Jr., J. J. Zwolenik and R. M. Fuoss, J. Am. Chem. Soc., 1959, 81, 1557. T. Shedlovsky, J . Franklin Inst., 1938, 255, 739. R. M. Fuoss and T. Shedlovsky, J . Am. Chem. Soc., 1949, 71, 1496. lo R. M. Fuoss and F. Accascina, Electrolytic Conductance, (Interscience, New York, 1959). l 1 I. N. V’yunnik, A. M. Zholnovach and A. M. Shkodin, Elektrokhimiya, 1976, 12, 1334. l 2 J-C. Justice, Electrochim. Acta, 1971, 16, 701. l3 W. J. Youden, Statistical Methods for Chemists (John Wiley, New York, 1951), p. 42. l4 C. J. James and R. M. Fuoss, J . Solution Chem., 1975, 4, 91. l 5 A. D’Aprano and R. M. Fuoss, J . Solution Chem., 1975, 4, 175. l6 R. M. Fuoss, J . Am. Chem. Soc., 1935, 57,488. l 7 J. F. Coetzee and G. P. Cunningham, J. Am. Chem. Soc., 1965,87, 2529. l9 D. S. Gill, Electrochim. Acta, 1977, 22, 491. 2o D. S. Gill, Electrochim. Acta, 1979, 24, 701. C. H. Springer, J. F. Coetzee and R. L. Kay, J. Phys. Chem., 1969, 73, 471. (PAPER O/ 1905)

ISSN:0300-9599    

DOI:10.1039/F19827800119

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

J . Chem. SOC., Furuduy Trans. 1 , 1982, 78, 127-136 Pulsed Irradiation of Water-soluble Porphyrins BY RAYMOND BONNET AND R. J. RIDGE Department of Chemistry, Queen Mary College, Mile End Road, London El 4NS A N D EDWARD J. LAND Paterson Laboratories, Christie Hospital and Holt Radium Institute, Manchester M20 9BX AND ROY s. SINCLAIR, DAVID TAIT AND T. GEORGE TRUSCOTT* Chemistry Department, Paisley College, High Street, Paisley PA 1 2BE Received 1 1 th December, 1980 Nanosecond laser flash photolysis and pulse radiolysis have been used to characterise the excited states and radicals of meso-tetraphenylporphyrin tetrasulphonic acid as the tetrasodium salt (Na,TPPS) and of meso-tetra-N-methylpyridylporphyrin as the tetra-p-toluenesulphonate (TMPyrP). After 347 nm irradia- tion, aqueous Na,TPPS forms triplet states (dT = 0.76) and photoionises monophotonically = 0.029), while TMPyrP forms only triplets (aT = 0.92) and does not photoionise.The difference and corrected spectra of the triplet states and the semi-oxidised and semi-reduced radicals of these molecules have been obtained. It has been known for over 70 years that porphyrins such as haematoporphyrin cause photodynamic activity in man.l In recent years2 it has been established that certain porphyrins, including haematoporphyrin, are useful in the phototherapy of malignant tumours. In such treatment the tumour accumulates porphyrin and subsequent irradiation of the tumour can lead to marked regression. Currently the most used porphyrin is the so-called haematoporphyrin derivative containing, among other constituents, haematoporphyrin mono- and di-a~etates.~ Several other por- phyrins are under clinical investigation4 as possible alternatives to the haematopor- phyrin derivative for this important therapeutic use.These include Na,TPPS and TMPyrP, which are the subject of the present paper. We have attempted to generate and to characterise the excited state and radical species of such porphyrins; clearly a complete understanding of the phototherapeutic process must depend on a thorough understanding of the role of such species. EXPERIMENTAL meso-Tetraphenylporphyrin tetrasulphonic acid [5,10,15,20-tetra(p-hydroxysulphonylphen- yl)porphyrin] was prepared as its tetrasodium salt (25% yield) by the direct sulphonation of meso-tetraphenylp~rphyrin~ and was recrystallised from water + ethanol.meso-Tetra(N-methyl- 4-pyridy1)porphyrin tetra-p-toluenesulphonate was obtained by the alkylation of meso-tetra(4- pyridy1)porphyrin with methyl-p-toluenesulphonate (reflux in dimethylformamide overnight).6 The product crystallised on cooling: it was removed by filtration, washed with acetone and dried in zmcuo (36%). These two porphyrins are both soluble in water because of the peripheral charged substituents: in one case the substituents are negatively charged, in the other they are positively charged. The porphyrin radical ions were studied by pulse radiolysis using a 9-12 MeV Vickers linear accelerator, as previously described by Keene,7 with 1 cm or 2.5 cm quartz cells. 127128 PULSED IRRADIATION 0 F W A T ER-S 0 LU B LE PO R PH Y R I N S One-electron oxidations were carried out on aqueous 1 x mol dm-, porphyrin solutions containing 5 x lod2 mol dm-, NaN, or KCNS in lop2 mol dm-, phosphate buffer, pH 7.0.At this pH the porphyrins exist as the free base (H,P). Pulse radiolysis of aqueous NaN, or KSCN generates, respectively, Nj and (SCN);- radicals via electron transfer from the primary OH' (1) radical : From earlier studies,8 it might be expected that N, or (SCN);- would oxidise porphyrins (2) according to : Since OH' itself tends to form adducts with organic solutes, it is not an ideal oxidant. The N; (or 2 SCN-)+OH' -+ OH-+Nj [or (!EN);-1. H,P+Nj [or (SCN),-] -+ H2P'+ +N; (or 2 SCN-). solutions were saturated with N,O to remove hydrated electrons (e&) by: eLq+N,O -+ N,+OH-+OH'. (3) Under such conditions G(0H') is currently believed to be 6.1.9 The concentration of radical cations produced was determined using air-saturated aqueous KSCN as the dosimeter G[(SCN);-] = 2.9, E ( ~ ~ ~ , ; - taken to be 7100 dm3 mol-' cm-l at 500 nm.l0 The difference extinction coefficient of the porphyrin radical cation (A&Hzp.+) can be found from the expression : 2.9 AODHzp*~+ AEHzP'f = '(SCN); - __ 6.1 O D ( s c N ) ; - where AOD,,, * + is the observed difference (radical - ground state) optical density and O D ( s , N ) ; - denotes the thiocyanate radical optical density obtained in air-saturated solution. The corrected extinction coefficient ( E ~ ~ ~ .+ ) is obtained by adding on the appropriate ground-state extinction ( E ~ ) .Due to possible errors in the G value estimates used, the extinctions obtained are considered to be correct to & 15%. One-electron reduced porphyrin radicals were generated in 1 OW3 mol dmW3 phosphate buffer (PH 7.0) and in 10-1 rnol dm-, NaOH containing lop4 mol dm-, porphyrin and 10-l mol dm-, propan-2-01. Radicals of similar porphyrins have been shown previously to have a pK value around 9.7." Thus, at neutral pH, the radical H,P' is observed, while in 10-l mol dm-, NaOH the radical H,P'- is observed. The porphyrin may be reduced directly by primary eiq as in: H,P+eiq -+ H,P'-. (4) ( 5 ) (6) The interfering primary radicals OH' and H' are removed by adding propan-2-01, as in: (CH,),CHOH + OH' (or H') -+ (CH,),cOH+ H,O (or H,). H,P + (CH,),cOH -+ H2P'- + (CH,),CO + H+.The propan-2-01 radical produced also reduces porphyrins by : Since this reaction takes place 10 times more slowly than does reaction (4), eiq was removed by saturating with N,O so that eiq was ultimately replaced by an equivalent amount of (CH,),eOH. This simplified the growth kinetics of the porphyrin radical. In 10-l mol dm-, NaOH, the propan-2-01 radical is deprotonatedl, to the anion (CH,),CO-, which reacts with H,P or various deprotonated forms of H2P as in: H,P+ (CH,),CO- --* H,P'-+(CH,),CO. (7) The concentration of porphyrin radicals produced was estimated by monitoring the eiq absorption at 700 nm (E = 18500 dm3 mol-1 cm-')13 in N,-saturated solution containing no porphyrin. The difference extinction coefficients (AE) of the radicals were obtained in N,O saturated solutions using the expression:BONNETT, RIDGE, LAND, SINCLAIR, TAIT AND TRUSCOTT 129 where G(0H') = 6.1 in N,O-saturated solution, G(H') = 0.6 in N,O-saturated solution and G(eiq) = 2.7 in N,-saturated solution.A& was corrected for ground-state absorption by adding on the appropriate ground-state extinction coefficient ( E ~ ) to give cHsP- -(oT FrP.). The extinctions in alkali may be slightly in error since the yield of OH' under such conditions could be higher since OH- will scavenge H+ from the spurs, thus enhancing the yield of eiq reacting with H,O. Laser flash photolysis was carried out on aqueous (5-10) x 1 0-6 mol dm-3 porphyrin solutions in lop2 mol dm-3 phosphate buffer, pH 7.4, using the 347 nm line of the frequency-doubled ruby laser previously described by McVie et al.14 Triplet-state extinction coefficients (cT) were measured by the complete conversion method described earlier for pr~toporphyrin.'~ For Na,TPPS, the technique had to be modified to allow for interfering photoionisation, in the presence of which the observed difference optical density in argon-saturated solution consists of both triplet-state and the semi-oxidised radical absorption. Since the semi-oxidised radical forms by: hv H,P + H,P' + + eiq (8) the concentration of H,P'+ and hence its AOD can be deduced by monitoring eiq. When all the ground-state molecules have been converted to triplet or H,P' +, the triplet concentration and AOD are readily derived so that AE, is given by the expression: AoDobserved - ([e~qlAEH2P' ' r) = ([ground state] - [e;J) I where [eiq] = [AODeG(720 nm)/A~~;~(720 nm)]l and 1 is the path length of material excited by the laser beam.AODobserved is the maximum AOD observed in argon-saturated solution at complete conversion of the ground state to triplet or H,P'+. cT was obtained by adding on cG. Quantum yields of triplet formation (a,) and of photoionisation (OJ were obtained by the comparative techniquels using anthracene in cyclohexane (cT = 64700 dm3 mol cm-l at 414 nm," OT = 0.71) as actinometer. This technique compares the concentration of porphyrin transient with the concentration of anthracene triplets formed by the same number of photons. At laser intensities low enough to ensure < 10% depopulation of the ground state,l* the quantum yield of porphyrin transient is given by: AOD, O, (or a,) = AE, AODgnth where anth denotes anthracene and X denotes porphyrin triplet, porphyrin radical cation or The second-order rates for triplet quenching by oxygen kt2, were calculated from the eiq- equation : k, -k, kt2 = [o,l where k, is the first-order rate constant for triplet decay in air-saturated solution, k, is the first-order rate constant for triplet decay in argon-saturated solution and [O,] is the oxygen concentration in air-saturated water = 2.65 x TheO,, N,, argon and N,O gases used in these experiments were all high purity from the British Oxygen Company.mol dm-3 at 25 OC.19 RESULTS PULSE RADIOLYSIS OF Na,TPPS The semi-oxidised radical (H,P' +) difference and corrected spectra in the 440-900 nm region shown in fig.1 were generated with N3 as oxidant (k = 5 x lo9 dm9 mol-1 s-l). Using KSCN dosimetry the corrected extinction coefficient ( E ) was found to be 25 150130 PULSED IRRADIATION OF WATER-SOLUBLE PORPHYRINS FIG. 1.-Semi-oxidised radical (H,P'+) of Na,TPPS in water at pH 7.4. (a) Laser flash photolysis of 0,-saturated lop5 mol dm-3 solutions: (0) difference spectrum + - E ~ ) ; (0) corrected for ground-state absorption ( E ~ , ~ - + ) . (b) Pulse radiolysis of N,O-saturated lo-, mol dm-3 solutions containing 10+ mol dm-3 NaN,: (0) difference spectrum ( E ~ , ~ . + - E ~ ) ; (A) corrected spectrum (eHPp.+). dm3 mol-l cm-l at 450 nm. The spectrum derived from OH' oxidation differed slightly from the ones above, probably indicating the presence of OH '-H,P adducts.No transient attributable to H,P'+ was observed when (SCN);- was used as the oxidant. In argon-saturated solution the initial fast growth of the semi-reduced radical H,P' , presumably generated by reaction (4), corresponds to a second-order quenching of eCq by H,P of 8.9 x lo9 dm3 mol-l s-l. This was followed by a slower growth, probably resulting from reaction (6), the corresponding rate constant being at least one order of magnitude less than that for reaction (4). The growth kinetics of H3P' were much simpler in N,O-saturated solution, where only reaction (6) is observed. By assuming that the rate of decay of (CH,),cOH in the absence of Na,TPPS is negligible, the second-order rate constant was estimated as 2 x lo9 dm3 mol-l s-l for this reaction.H3P' decayed by mixed-order kinetics with a first half-life of 7 ms to a relatively stable product. Using e,,c, as dosimeter, E for H,P' was found to be 45 380 dm3 mol-1 cm-l at 460 nm. This value was used to obtain the difference and corrected spectra given in fig. 2. The semi-reduced radical at high pH (H,P'-) was generated in N,O-saturated 10-l mol dm-3 NaOH by reaction (7). The second-order rate constant for growth of thisBONNETT, RIDGE, LAND, SINCLAIR, TAIT A N D TRUSCOTT 131 radical corresponded to a value of 1 x lo9 dm3 mol-1 s-l for reaction (7). H,P'- decayed with a first half-life of 640 ps to a permanent product absorbing in the visible region. The value for E ~ ~ ~ - - of 22000 dm3 mol-1 cm-l at 460 nm was used to derive the difference and corrected spectra shown in fig.3. PULSE RADIOLYSIS OF TMPyrP No semi-oxidised radical absorption was observed for this molecule with Nj as the oxidant. Because of the results described previously for Na,TPPS, experiments using OH' or (SCN);I- as oxidants were not attempted. 1 X/nm -10 FIG. 2.-Semi-reduced radical (H,P') of Na,TPPS in water at pH 7.0. Pulse radiolysis of N,O-saturated lo-, mol dm-3 solutions containing lo-' mol dm-, propan-2-01; (0) difference spectrum ( E ~ , ~ * - c ~ ) ; (0) corrected for ground-state absorption (cHSp.). -1oL X/nm FIG. 3.-Semi-reduced radical (H,P'-) of Na,TPPS in lo-' mol dm-3 NaOH solution. Pulse radiolysis of N,O-saturated mol dm-, solutions containing lo-' mol dm-, propan-2-01: (0) difference spectrum (cHZp.- -cG) ; (0) corrected for ground-state absorption ( E ~ , ~ - - ) .132 PULSED IRRADIATION OF WATER-SOLUBLE PORPHYRINS The semi-reduced radical was studied using 7 x mol dm-3 TMPyrP with 10-l mol dmh3 propan-2-01.In N,O-saturated solutions the major process detected corresponded to a growth (l; z 2 ps) to a permanent product. This is interpreted in terms of reaction (6), with a corresponding second-order rate constant of 5 x lo9 dm3 mot1 s-l. However, these observations were complicated by slower growth kinetics I I I L 500 600 700 800 h/nm FIG. 4.-Semi-reduced radical (H,P') of TMPyrP in water at pH 7.0. Pulse radiolysis of N,O saturated 3.5 x lop5 mol dmd3 solutions containing lo-' mol dmm3 propan-2-01: (0) difference spectrum (eH,.,. - E ~ ) ; (0) corrected for ground-state absorption (eHsP-).at certain values of 3, (such as 500, 525, 575 and 600 nm). These may arise from the small yields of radicals other than (CH,),COH in the system20 reacting with this porphyrin. Thus, in order to obtain the spectrum of H,P' with least contamination from the slow growing product(s) the difference spectrum given in fig. 4 was measured soon after the pulse (2 ps). The difference extinction ( E ~ , ~ . ) was determined at 480 nm where the slowly growing transient appeared not to absorb. The ground-state absorption spectrum of TMPyrP in 10-1 mol dme3 NaOH indicated that the porphyrin was a mixture of HP- and P2-. The semi-reduced radical grew in with an apparent second-order rate constant of 6:3 x lo9 dm3 mol-1 s-l [reaction (7)].The radical decayed (ti = 55 ps) to a relatively stable product. The apparent semi-reduced radical extinction coefficient was 42 300 dm3 mol-1 cm-l at 480 nm and fig. 5 gives the difference and corrected spectra.BONNETT, RIDGE, LAND, SINCLAIR, TAIT AND TRUSCOTT 133 50 40 7 30- 5 2 - I - E 2 -u 2 0 - --. ru 10 - - - 0 500 600 7 00 8 00 X/nm FIG. 5.-Semi-reduced radical (H,P'-) of TMPyrP in lo-' mol dm-3 NaOH. Pulse radiolysis of N,O-saturated 7 x mol dm-3 solutions containing lo-' mol dmP3 propan-2-01 : (0) difference spectrum ( E ~ , ~ * - - e G ) ; (0) corrected for ground-state absorption (cHSp. ). LASER FLASH PHOTOLYSIS OF Na,TPPS Following 347 nm excitation of argon-saturated Na,TPPS solutions [(0.5-10) x lop5 mol dm-3] a short-lived (ti = 2.0 ps)? species (A) was observed superimposed on a much longer-lived (ti = 400 ps)t species (B).The absorption of (A) increased linearly with laser intensity implying that it was formed by a monophotonic process. The short-lived, broad absorption spectrum peaking at 720 nm was quenched by N,O suggesting that it was due to eLq. Only part of (B), presumably the triplet, was efficiently quenched by 0, (k$ = 1.8 x lo9 dm3 mol-1 s-l), to leave a residual absorption decaying by first-order kinetics (k = 2.7 x lo3 s-l). The difference spectrum of this residual absorption measured in 0,-saturated solution 20 p s after the laser pulse closely resembled that of the semi-oxidised radical generated by pulse radiolysis. Because much lower porphyrin concentrations could be used in laser flash photolysis, it was possible to obtain the H,P'+ difference and corrected spectra in the Soret region and these are shown in fig.1. t At Na,TPPS concentration of c a 10-5 mol dm-3.134 PULSED IRRADIATION OF WATER-SOLUBLE PORPHYRINS The modified complete conversion technique described earlier gave a value for E, of 44250 dm3 mol-1 cm-l at 440 nm. The AOD of triplet used in the CD, calculation was measured by subtracting the small H,P' + absorption observable in 0,-saturated solution together with a contribution from the H3P' absorption formed in argon- saturated solution from the overall AOD observed in argon-saturated solution. Using the ACT derived from the E, given above, CD, was found to be 0.76 ( + O . O S ) . The quantum yield of eiq formation monitored at 720 nm in argon-saturated solution was (&TI.FIG. 6.-Triplet-triplet spectrum of Na,TPPS in water, pH 7.4. Laser flash photolysis of argon-saturated lov5 mol dm-3 solutions: (0) difference spectrum (ET-EG); (0) corrected for ground-state absorption 6o 40 t - ' 20 5 E mE -20 p -40 - - I X/nm -3 1 -60 -80 FIG. 7.-Triplet-triplet spectrum of TMPyrP in water, pH 7.4. Laser flash photolysis of argon-saturated mol dm-3 solutions: (0) difference spectrum ( E ~ - E ~ ) ; (0) corrected for ground-state absorption ( E dBONNETT, RIDGE, LAND, SINCLAIR, TAIT AND TRUSCOTT 135 0.023 ( f 0.005). This value corresponds well with the semi-oxidised radical yield of 0.029 ( f 0.005) observed in 0, saturated solution. The full triplet-triplet difference and corrected spectra are shown in fig.6. LASER FLASH PHOTOLYSIS OF TMPyrP Excitation of TMPyrP with 347 nm radiation at pH 7.4 in argon- or N,O-saturated solution produced only one strong, long-lived ( 4 = 120 p s ) transient. This species was quenched by 0, with a diffusion-controlled second-order rate constant of 1.5 x lo9 dm3 mo1-1 s-l, implying that it was the triplet. Straightforward application of the complete conversion and the comparative methods yielded, respectively, cT = 28 600 dm3 mol-1 cm-l at 450 nm and CD, = 0.92 ( 0.06). The difference and corrected triplet spectra are given in fig. 7. DISCUSSION The very efficient formation of long-lived triplet states by Na,TPPS (CDT = 0.76; ti = 280 p s ) and TMPyrP (CD, = 0.92; ti = 120 p s ) suggests that both of these compounds are good photosensitisers.While Na,TPPS does have a lower CD, than TMPyrP, it also photoionises monophotonically (CDI = 0.029) yielding the very reactive eiq. Subsequent reactions of e.& may occur, e.g. with 0, to form the superoxide (0; -) radical. We have observed that uro- and cupro-porphyrins also photoionise in aqueous The fact that TMPyrP is not oxidised by N j emphasises the reluctance of this molecule to ionise. This difference in behaviour is not surprising when the polycationic nature of TMPyrP is compared with the polyanionic nature of Na,TPPS. This implies that the different meso-substituents strongly influence the reduction potential of the porphyrin macrocycle. Thus in TMPyrP, electron density may to some extent be drawn away from the porphyrin ring towards the electron deficient N-methylpyridyl part, while for Na,TPPS, the reverse is the case.This is not inconsistent with the reported reduction potentials of these porphyrins.22 While we have not studied the fluorescence of these porphyrins in detail, that of TMPyrP is clearly much lower than that of Na,TPPS. This may reflect either aggregation in aqueous solution for TMPyrP or a more rapid intersystem crossing from S, to T, which is consistent with the higher value of CD, obtained for this molecule. The radical and triplet spectra are similar in the region up to 650 nm. Where measurements have been extended to longer wavelengths the semi-oxidised and semi-reduced radicals appear to have another absorption band at ca. 700 nm. The rate constants for reaction of TMPyrP and Na,TPPS with eiq, (CH,)$OH and (CH,),CO-, along with the spectra of the semi-reduced radicals produced, compare well with information reported for other porphyrins.23 This work was supported by grants from the S.R.C., the M.R.C. and the Cancer Research Campaign. D. T. acknowledges an S.R.C. research studentship. F. Meyer-Betz, Dtsch. Arch. Klin. Med., 1913, 112, 476. T. J. Dougherty, D. Boyle, K. Weishaupt, C. Gomer, D. Borcicky, J. Kaufman, A. Goldfarb and G. Grindey, in Research in Photobiology, ed. A. Castellani (Plenum Press, New York, 1977), p. 435. R. Bonnett, A. A. Charalambides, K. Jones, I. A. Magnus and R. J. Ridge, Biochern. J., 1978, 173, 693. D. A. Musser, J. M. Wagner and N. Datta-Gupta, J . Natl. Cancer Znst., 1978, 61, 1397.136 PULSED IRRADIATION OF W ATER-SO L U B LE PO R PHY R I NS T.S. Srivastava and M. Tsutsui, J. Org. Chem., 1973, 38, 2103. R. F. Pasternack, P. R. Huber, P. Boyd, G. Engasser, L. Francesconi, E. Gibbs, P. Fasella, G. C. Venturo and L. de C. Hinds, J. Am. Chem. Soc., 1972, 94, 451 1. J. P. Keene, J. Sci. Instrum., 1964, 41, 493. J. P. Chauvet, R. Viovy, E. J. Land and R. Santus, C.R. Acad. Sci., 1979, 288, 1423. R. H. Schuler, L. K. Patterson and E. Janata, J. Phys. Chem., 1980,84, 2088. Keene, A. J. Swallow and J. H. Baxendale (Academic Press, London, 1965), p. 117. lo G. E. Adams, J. W. Boag, J. Currant and B. D. Michael, in Pulse Radiofysis, ed. M. Ebert, J. P. l1 P. Neta, A. Scherz and H. Levanon, J. Am. Chem. SOC., 1979, 101, 3624. l2 K. D. Asmus, A. Henglein, A. Wigger and G. Beck, Ber. Bunsenges. Phys. Chem., 1966, 70, 756. l 3 E. M. Fielden and E. J. Hart, Trans. Faraday SOC., 1967, 63, 2975. l4 J. McVie, R. S. Sinclair and T. G. Truscott, J. Chem. Soc., Faraday Trans. 2, 1978, 74, 1870. l5 R. S. Sinclair, D. Tait and T. G. Truscott, J. Chem. Soc., Faraday Trans. I , 1980, 76, 417. l7 R. Bensasson and E. J. Land, Trans. Faraday Soc., 1971, 67, 1904. I8 R. Bensasson, C. R. Goldschmidt, E. J. Land and T. G. Truscott, Photochem. Photobiof., 1978, 28, l9 S. L. Murov, in Handbook of Photochemistry (Marcel Dekker, New York, 1973), p. 89. zo K. D. Asmus, K. Mockel and A. Henglein, J. Phys. Chem., 1973, 77, 1218. 21 Unpublished work by the present authors. z2 R. F. X. Williams and P. Hambright, Bioinorg. Chem., 1978, 9, 537. 23 Y. Hare1 and D. Meyerstein, J. Am. Chem. Soc., 1974, 96, 2720. B. Amand and R. Bensasson, Chem. Phys. Lett., 1975, 34, 44. 277. (PAPER 0/1919)

ISSN:0300-9599    

DOI:10.1039/F19827800127

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

J. Chem. SOC., Faraday Trans. I , 1982, 78, 137-141 Salt-induced Metachromatic Behaviour of an Azo Dye in Aqueous Solution BY MICHEL DE VYLDER Laboratorium voor Fysische Scheikunde, Rijksuniversiteit Gent, Krijgslaan 27 1, B-9000 Gent, Belgium Received 16th January, 198 1 Addition of substantial amounts of simple electrolytes into basic aqueous methyl orange solutions causes a marked decrease of the intensity of the light absorption of the dye in the visible range and the appearance of a total new absorption band at 358 nm. Addition of a non-electrolyte such as urea shows no such effect. The phenomenon is attributed to interactions between the added ions and the electrons of the chromophoric system. The formation of a new absorption band in the visible range on the short-wavelength side and at the expense of the original absorption band, which can be observed on increasing the concentration of some water-soluble dyes, is known as metachromatic behaviour.Where this phenomenon is not attributable to the presence of aggregated species, interactions between dye-ions and counter-ions are taken into account. Similar spectral changes have been observed on adding foreign ions in large excess into dye solutions of a defined Concentration' but hitherto azo dyes have apparently been devoid of such behaviour. Only the qualitative effect of KC1 and CaC1, on the spectrum of methyl orange (I) has been mentioned, albeit without explanation.2 We report here investigations of the influence of several electrolytes [NaCl, KCl, LiC1, CaCl,, MgCl,, NaNO,, LiNO, and Ca(NO,),] and of a non-electrolyte (urea) on the spectrum of methyl orange in aqueous solution at different temperatures and pH values, we also report our attempts to interpret the results.We observed that in basic solutions the addition of increasing amounts of each electrolyte used causes a proportionally marked decrease in the dye absorbance at 465 nm, while urea does not influence the spectrum at all, even at concentrations as high as 7.5 mol drn-,. However, the concentration required to induce the phenomenon depends on the nature of both the anion and the cation (fig. 1). Upon further salt addition, we noticed the formation of a new absorption band at 358 nm. Fig. 2 shows the typical development of the absorption spectrum as a function ofelectrolyteconcentration. The presence of an isosbestic point proves the simultaneous existence of only two absorbing species in solution.The original spectrum is gradually restored on heating and becomes fully recovered to its state before salt addition at ca. 75 O C . On recooling, the spectrum no longer reverts to its initial state, however (fig- 3)- 137138 METACHROMATIC BEHAVIOUR OF A N AZO DYE 2*5 t 2.5 5 7.5 c/mol d n ~ - ~ FIG. 1.-Decrease in the molar absorptivity ( E ) of the methyl orange anion in basic aqueous solution as a function of the addition of different substances (A = 465 nm, 4.28 x mol dm-, methyl orange, 8 < pH < 10, temp. 20 f 0.5 OC, corresponding salt solutions as reference). 0, MgCl,; 0, CaC1,; + , NaCl; A, Ca(NO,),; 0, KCl; @, NaNO,; x , LiCl; A, LiNO,; a, urea.DISCUSSION Water solubility is conferred to the methyl orange ions by the presence of the lyophilic SO; group and by the hydration of the lyophobic group via hydrogen bonds at the p-azo nitr~gen.~ Interactions resulting in spectral changes are thus only likely after the protecting water shields around the dye ions are broken up. The order of influence of the univalent cations which we found experimentally can be related to the series of their specific powers of breaking the water structure, as determined by Frank and R~binson.~ The greater effect of Mg2+ and Ca2+ follows from their larger size and subsequently greater breaking power. The slighter effectiveness of nitrates with respect to chlorides can be attributed to their having a larger extent of ass~ciation.~ The formation of a new absorption band can be justified in different ways.Thus, shifts in the absorption spectra of various ions have been ascribed to the broadening of the ionic atmosphere resulting from the release of surrounding water molecules.s Such an explanation is not applicable here since thermal agitation should cause at leastM. DE VYLDER 139 500 450 400 350 X/nm FIG. 2.-Changes in the visible absorption spectrum of the methyl orange anion in basic aqueous solution caused by the addition of CaC1, (4.28 x mol dm-3 methyl orange, pH z 10, temp. 20k0.5 OC, salt solution as reference). Concentration CaCI, (in mol dm-3): (1) 1.2, (2) 1.6, (3) 2.0, (4) 2.4, (5) 2.8. qualitatively similar effects unlike our experimental finding that increasing the temperature causes the spectrum to revert to its shape before the addition of the salt.Ascribing the phenomenon to aggregated species is also unrealistic because of the lack of unequivocal reversibility of the spectrum on varying the temperature (fig. 3). This conviction is further supported by earlier statements revealing that methyl orange fails to show a distinct dimer or polymer band, even at concentrations near saturation. Although similar marked shifts of the absorption band have been observed in other solvents whose dielectric properties are comparable with those prevailing in aqueous electrolyte solutions, and related to the ionizing power of the solvent,8 an analogous interpretation of our results is inadequate in view of the evident differences in the solvation state.Photochromism of methyl orange, resulting in a decrease in the main absorption band, has been demonstrated on irradiating alkaline solutions of the dyeg but, since only a partial conversion could be achieved in this way, a new band due to the cis isomers was not seen. The application by ReeveslO of the theoretical approach elaborated by Fischer displays this eventual absorption band near 360 nm, in140 METACHROMATIC BEHAVIOUR OF AN AZO DYE 1 . 1 500 450 LOO 350 h/nm FIG. 3.-Influence of the temperature on the spectral changes of the methyl orange anion in basic aqueous solution caused by the addition of 2.8 mol dm-3 CaC1, (4.28 x moi dm- dye, pHoz 8.5, CaC1, solution as reference). Curves 1-4 were recorded with increasing temperature (4,20,40 and 55 C) and curves 5 and 6 with decreasing temperature (20 and 4 "C).accordance with the Amax value of the band we report. Nevertheless, the ratio between the absorption intensities of the presumed cis and trans forms calculated at 100% conversion, differs slightly from the ratio we obtain experimentally. The plausibility of trans-cis isomerism of methyl orange in electrolyte solutions should be confirmed by a suitable direct technique. Recent interpretations of laser-excited resonance spectra attribute the bands observed at 1400 and 1300 cm-l to trans- and cis-azo groups, respectively.ll However, we did not observe a similar shift on adding electrolytes but were confronted with a systematic lowering of all but one of the bands, the latter being assigned to the C-N stretching of the aromatic amine.Therefore, a likely explanation should involve strong interactions between the electrons of the chromophoric part of the dye and the added counter-ions. The electron distribution is influenced by the strong repelling effect of the dimethylamino group. Charges induced in this way are of course neutralized by adjacent water dipoles. Added salt ions will take up the positions of the water dipoles after they have broken up the protecting water shields. Related effects caused by alkali metal ions, observed in the absorption band of crown-ether dyes, have been explained in the sameM. DE VYLDER 141 way.I2 The appearance of a new band at 357 nm has only been reported on the addition of Ba2+ salts, however.The involvement of charges is further evidenced on considering the absence of the effect for urea. EXPERIMENTAL The spectrophotometric experiments in the visible range were performed with a Varian Techtron 635 spectrophotometer and carried out on fresh solutions. Identical spectra could usually be obtained within several hours, except in the presence of the highest electrolyte concentrations where turbidity occurred within a few minutes. Resonance Raman spectra were recorded on a Coderg spectrometer. The laser excitation was supplied by a He-Ne laser and the frequency used was 632.8 nm. We thank Mr Tony Haemers (Labo Algemene en Anorganische Scheikunde) for recording the Raman spectra. R. B. McKay and P. J. Hillson, Trans. Faraday Soc., 1965, 61, 1800. V. V. Palchevskii, M. C. Zakharevskii and T. M. Kalvarskaya, Vestn. Leningr. Univ., Ser. Fiz. Khim., 1962, 17(3), 125. R. L. Reeves, R. S. Kaiser, M. S. Maggio, E. A. Sylvestre and W. H. Lawton, Can. J. Chem., 1973, 51, 628. H. Frank and R. Robinson, J . Chem. Phys., 1940, 8, 933. J. McKenzie and R. Fuoss, J . Phys. Chem., 1969, 73, 1501. M. Smyth and M. C. Symons, Discuss. Faraday SOC., 1957, 24, 206. ’ F. Quadrifoglio and V. Crescenzi, J. Colloid Interface Sci., 1971, 35, 447. Ch. Williamson and A. Corwin, J. Colloid Interface Sci., 1972, 38, 567. R. Lovrien, P. Pesheck and W. Tisel, J . Am. Chem. Soc., 1974, 96, 244. lo R. L. Reeves and Sh. A. Harkaway, in Micellization, Solubilization and Microemulsions, ed. K. L. Mittal (Plenum, New York, 1977). l 1 P. R. Carey, H. Schneider and H. J. Bernstein, Biochem. Biophys. Res. Commun., 1972, 47, 588; K. Machida, B. Kim, Y. Saito, K. Igarashi and T. Uno, Bull. Chem. SOC. Jpn, 1974, 47, 78. l 2 J. P. Dix and Fr. Vogtle, Angew. Chem. Int. Ed. Engl., 1978, 17, 857. (PAPER 1 /065)

ISSN:0300-9599    

DOI:10.1039/F19827800137

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

J. Chem. SOC., Faraday Trans. I, 1982, 78, 143-146 Isotopic Study of the Temperature-programmed Desorption of Oxygen from Silver BY SHUICHI KAGAWA,* MASAKAZU IWAMOTO AND SHUJI MORITA Department of Industrial Chemistry, Faculty of Engineering, Nagasaki University, Nagasaki 852, Japan AND TETSURO SEIYAMA Department of Materials Science and Technology, Graduate School of Engineering Sciences, Kyushu University, Fukuoka 8 12, Japan Received 16th January, 198 1 The properties of oxygen adsorbed on silver were studied by a temperature-programmed desorption technique using an isotopic-tracer method. Oxygen adsorbates desorbed to form a well-defined single peak with a maximum at ca. 573 K. On the basis of the isotopic distribution during desorption and the surface coverage of the adsorbed oxygen, molecular and atomic oxygen adsorbates and embedded oxygen species are suggested to exist on or in the silver.The latter species can easily exchange its position with the atomic adsorbates. The behaviour of oxygen over silver has been extensively investigated in connection with the mechanism of the catalytic oxidation of ethy1ene.l Many studies on the chemisorption of oxygen have suggested that molecular and atomic oxygen species coexist on the surface of the silver catalyst; however, there is considerable disagreement concerning the adsorption mode of oxygen and the oxygen species active in catalytic oxidation.2 The thermal desorption of oxygen from silver has also been studied by several group^.^-^ Their observations agree in terms of the appearance of only one desorption peak with a maximum between 473 and 573 K upon adsorption of oxygen at 373-473 K, but the desorption kinetics were reported to be first-5 or ~econd-order.~~ These discrepancies may result from the cooperative desorption of molecular and atomic oxygen species and/or from residual oxygen' existing in the sample.The present study is an attempt to resolve some of the above-mentioned ambiguities in the oxygen-silver system by an isotopic-tracer method combined with a temperature- programmed desorption (t.p.d.) study. EXPERIMENTAL The apparatus used in the present work was essentially the same as that previously reported except that the desorption behaviour of oxygen was monitored with a quadrupole mass spectrometer (Nichiden Varian TE-600) instead of a thermal-conductivity detector.8 The silver powder was prepared by a method described el~ewhere.~ Its surface area was 0.20 m2 g-l.Pretreatment at 623 K was as follows: the sample was evacuated for 1 h in a sample tube, exposed to 300 Torr of hydrogen for 1 h and re-evacuated for an additional 1 h. Oxygen was adsorbed at 453 K for a prescribed period and removed by evacuation at the same temperature for 1 h. The sample was then cooled to 273 K in a static vacuum. After these processes the carrier gas (He) was diverted to flow through the sample at a rate of 30 cm3 min-l and then the programmed heating was started at a uniform heating rate of 5 K min-l. 143144 T.P.D. OF 0, FROM Ag RESULTS AND DISCUSSION The t.p.d. chromatogram of oxygen from silver showed a well-defined single peak which began at ca.443 K and passed through a maximum at ca. 573 K, in good agreement with thermal-desorption behaviour previously r e p ~ r t e d . ~ - ~ The amounts of desorbed oxygen were well reproduced within an experimental error of ca. lo%, whereas the temperature of the peak maximum was found to vary between 553 and 603 K in each experiment. This variation probably came from a slight change in the oxidation state of the silver surface owing to a slight variation in the pretreatment; however, in this paper no further reference will be made to this phenomenon. r 0 0 1 I I I . WY h Y ." c T/K FIG. 1 .-Distribution of oxygen isotopes in the t.p.d. chromatogram after adsorption of lsO2 at 453 K and 113 Torr for 30 min on silver.Upon adsorption of 1802 (l80 atomic fraction = 99.1 %) at 113 Torr and 453 K for 30 min, large amounts of l60l8O and 1602 as well as 1802 were desorbed in the t.p.d. chromatogram as shown in fig. 1. The number of desorbed oxygen-16 atoms reached as much as 30-40% of the total amount of desorbed oxygen. This is due to residual oxygen which was not removed even after reduction at 623 K for 1 h. In an attempt to minimize the influence of such oxygen, the period of hydrogen pretreatment was extended to 8 h; nevertheless, the result was almost the same as above. The surface coverage, i.e. the ratio of the number of desorbed oxygen atoms to the number of surface silver atoms,* was in the range 0.9-1.1 under the experimental conditions of fig. 1. On the basis of these results and other work,' a fair amount of oxygen is concluded to exist on or in the silver which cannot be removed from the sample upon hydrogen pretreatment at temperatures as high as 623 K.In spite of its firm binding, this oxygen can easily exchange positions with the oxygen adsorbed later. In order to investigate whether the isotopic species in the desorbed gas are in statistical equilibrium, we introduced a factor, R, given by R = [16018O]2/[16O2] "",] * The number of surface silver atoms has been reported to be 1.31 x lO1O atom m-2.10S. KAGAWA, M. IWAMOTO, S. MORITA A N D T. SEIYAMA 145 where the square brackets designate the concentration of each molecule. Values of R were calculated from mass spectra. As mentioned already, the surface coverage was almost one.Therefore, if only atomic adsorbates existed on the surface and the isotopic effect can be assumed to be negligible, then R should be equal to 4, since two oxygen atoms would combine at random to desorb as oxygen gas. The value of R shown in fig. 1 , however, was nearly one at the initial stage of desorption and increased gradually as desorption progressed to reach the value of 4. When adsorption was carried out at 453 K by using an equimolar mixture of lS02 and lS02, the variation of R in the desorbed gas was almost the same as above. These results suggest that the oxygen species adsorbing on the silver consist not only of atomic oxygen but also of molecular oxygen and that this molecular oxygen desorbs in the first half of the desorption peak.This conclusion is further supported by the following facts. The surface coverage of oxygen exceeded unity under the adsorption conditions at oxygen pressures > 120 Torr. The oxygen adsorbates detected in the present study were confirmed in separate experiments to react with ethylene yielding ethylene oxide as well as carbon dioxide.s desorb as O7 in early stage of t.p.d. to give low R desorb as 0, in l a t e stage of t.p.d. to give R = 4 P type 1 type I 1 type 111 FIG. 2.-Model of the behaviour of oxygen species on or in the silver surface. The desorption peak was always single and the shape and range of the desorption curves were roughly the same, regardless of the adsorption conditions and the amount of desorbed oxygen. Taking the above into consideration, it is proposed that the coupling reaction of the atomic adsorbates to form molecular adsorbates on the silver surface is very fast at the desorption temperatures of the molecular species.This would be the reason why there is only one desorption peak from silver and why the adsorption modes reported by several ~ o r k e r s ~ - ~ are in disagreement. Next, the silver was pretreated several times at 773 K with an equimolar mixture of lS02 and lS02. The t.p.d. chromatogram was measured following the adsorption of a mixture of 1 6 0 2 and 1 8 0 2 (1 : 1) at 453 K and 100 Torr. This was done in an attempt to reproduce earlier work7 in which it was reported that an isotopic equilibrium was achieved in the desorbed gas upon exposing the sample in advance to a mixture of lS02 and lS02 (1 : 1) at 773 K.The change in R, however, was very similar to that in fig. 1. Backx et aZ.ll found by electron energy-loss spectroscopy that upon adsorption of oxygen at 473 K and pressures greater than a few Torr diatomic oxygen adsorbates begin to be observed on the silver. The pressures of oxygen adopted in the earlier experiments,' 1.7-2.3 Torr, seem so low that most of the adsorbed oxygen should be atomic. In conclusion, the chemisorbed oxygen exists in two states, molecular (type I) and atomic (type 11), at relatively high adsorption pressures of oxygen. In addition, there146 T.P.D. OF 0, FROM Ag exists a firmly bound oxygen (type 111) which can easily exchange its position with the type I1 oxygen. This model is depicted in fig.2. The existence of type I11 is supported by the observations of Czanderna et al. that oxygen was found to penetrate into the bulk at 373-473 K from the change of work function and FEM pattern^.^? l 2 It remains uncertain whether the superstructures of (n x 1) oxygen on silver, detected by Rovida et aL3 correspond to our type I1 or I11 oxygen, since these phases become mobile above 423 K. The molecular oxygen found in this work might consist of 0,- ions or of diamagnetic species, as found to exist on silver by Clarkson et al.13 and Abou-Kais et al.14 This work was supported by a Grant-in-Aid for Scientific Research from the Ministry of Education of Japan under Grant no. 575535. P. A. Kilty and W. M. H. Sachtler, Catal. Rev. Sci. Eng., 1974, 10, 1. X. E. Verykios, F.P. Stein and R. W. Coughlin, Catal. Rev. Sci. Eng., 1980, 22, 197. G. Rovida, F. Pratesi, M. Maglietta and E. Ferroni, Surf. Sci., 1974, 43, 230; G. Rovida, F. Pratesi and E. Ferroni, J. Catal., 1976, 41, 140; G. Rovida, J. Phys. Chem., 1976, 80, 522. W. Kollen and A. W. Czanderna, J. Colloid Interface Sci., 1972, 38, 152; R. J. Ekern and A. W. Czanderna, J. Catal., 1977, 46, 109. Y. Tokoro, T. Uchijima and Y. Yoneda, J. Catal., 1979, 56, 110. S. Kagawa, M. Iwamoto, H. Mori and T. Seiyama, J. Phys. Chem., 1981, 85, 434. Amsterdam, 1965), p. 227; Y. L. Sandler and D. D. Durigon, J. Phys. Chem., 1965, 69, 4201. 'I Y. L. Sandler and W. M. Hickam, Proc. ThirdZnt. Congr. Catal., Amsterdam, 1964 (North Holland, * M. Iwamoto, Y. Yoda, E. Egashira and T. Seiyama, J. Phys. Chem., 1976, 80, 1989. * G. Brauer, Handbuch der Praparativen Anorganischen Chemie (Ferdinand Enke Verlag, Stuttgart, lo B. E. Sandquist, Acta Metall., 1964, 12, 67. l1 C. Backx, C. P. M. de Groot and P. Biloen, Appl. Surf. Sci., in press. l2 A. W. Czanderna, 0. Frank and W. A. Schmidt, Surf. Sci., 1973,38, 129. l3 R. B. Clarkson and A. C. Cirillo Jr, J. Catal., 1974,33,392; R. B. Clarkson and S . McClellan, J. Phys. l4 A. Abou-Kais, M. Jarjoui, J. C. Vedrine and P. C. Gravelle, J. Catal., 1977, 47, 399. 1954), p. 771. Chem., 1978, 82, 294. (PAPER 1 /067)

ISSN:0300-9599    

DOI:10.1039/F19827800143

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

J. Chem. SOC., Faraday Trans. 1, 1982, 78, 147-151 Structure of the ‘K’ Phase in Ternary Mixtures of Potassium Decanoate/Octanol/Water, as Determined by X-ray Diffraction, Polarising Microscopy and Nuclear Magnetic Resonance Spectroscopy BY PAUL G. NEE SON^ AND GORDON J. T. TIDDY* Unilever Research, Port Sunlight Laboratory, Quarry Road East, Bebington, Wirral, Merseyside L63 3JW Received 19th January, 1 98 1 The ‘ K’ mesophase region in the system potassium decanoate/octanol/water has been investigated using optical microscopy, low-angle X-ray diffraction and nuclear magnetic resonance spectroscopy. The phases formed below 300 K are lamellar and crystalline surfactant, while above 300 K an additional reversed hexagonal phase occurs. There is no evidence for the occurrence of a novel ‘K’ structure.The structures of the lamellar, hexagonal and reversed hexagonal lyotropic liquid crystals are generally agreed, and cubic phases are commonly accepted, although their structures have not yet been fully e1~cidated.l.~ In addition, there exist reports of other mesophases which are not universally accepted. These include the deformed hexagonal, complex hexagonal, tetragonal and rectangular It is important to establish whether or not these phases occur, not only for their intrinsic interest, but also because any theoretical description of lyotropic phases must be able to account for the additional structures that are proven to exist. A rationalisation of the known phase structures in terms of intra- and inter-micellar forces has been propo~ed.~ It is hard to envisage how the molecular arrangements of complex hexagonal or tetragonal phases could fit into the sequence of str~ctures.~ Previous investigations of the Ekwall ‘C’ phase,, which was assigned a tetragonal structure, have shown that it is in fact an emulsion of lamellar phase and micellar solution.6 An investigation of the complex hexagonal phase reported for the sodium dodecylsulphate/water system shows that the fundamental unit is a rod micelle, and the structure (deformed hexagonal) is very similar to that of the normal hexagonal p h a ~ e .~ The very extensive studies of mixed amphiphile mesophases by Ekwall and his collaborators2$ 5 9 8-10 provide a catalogue of information against which any theory can be checked. Among these exists a report of a reversed tetragonal phase labelled ‘ K ’, for the potassium decanoate/octanol/water system;8y9 this study is concerned with an attempt to establish its structure using low-angle X-ray diffraction, polarising microscopy and nuclear magnetic resonance (n.m.r.) spectroscopy, techniques which were successful in the previous The ‘K’ phase is reported to occur at high potassium decanoate concentrations, and has tie-lines to the lamellar phase (La), the liquid octanol phase (L,) and solid potassium decanoate (KC,,).An illustration of the relevant part of the phase diagram is given in fig. 1. The reversed tetragonal structure originally proposed for ‘ K’ phase8* 7 Permanent address : Department of Physics, Brunel University, Uxbridge, Middlesex UB8 3PH. 147148 STRUCTURE OF LYOTROPIC MESOPHASE ‘ K ’ FIG.1 .-Partial phase diagram of potassium decanoate (KC,,)/octanol (C,OH)/water system showing region of ‘K’ phase at 293 K, taken from P. Ekwall, L. Mandell and K. Fontell, J. Colloid Interface Sci., 1969, 31, 508. Triangular three-phase regions not labelled. Symbols are: L,, lamellar phase; L,, octanol solutions; S, crystalline KC,,, K, ‘K’ region. Points indicate compositions of samples studied. Note that measurements listed in table 1 refer to 31 1 K (X-rays) and 303 K (n.m.r.). consists of long water rods arranged on a square lattice in a surfactant-chain continuum. Arguments against this structure4 include the fact that the low-angle X-ray diffraction pattern does not include the expected d,,/2/2 line. Very recently, in a review of lyotropic phases Stenius and Ekwall list the occurrence of the ‘K’ phase without suggesting a structure.lO Mesophase ‘K’ is ‘a stiff, fairly clear, gel’ with a distinctive microscope texture;*-1° it can be separated by centrifugation from lamellar phase and L, phase.We have examined compositions in and around the ‘K’ phase using the three techniques mentioned above and have included measurements over a range of temperatures. EXPERIMENTAL Potassium decanoate was prepared by neutralisation of decanoic acid with standard techniques. Water was deionised and distilled, while octanol (B.D.H., specially pure) and deuterium oxide (99.7 %) were commercially available. Samples were mixed by heating and centrifugation in sealed tubes. Low-angle X-ray diffractograms were obtained using a Philips X-ray generator, a Kratky camera and a Philips proportional counter plus associated equipment.Typical sample scans were carried out in ca. 15 min. N.m.r. measurements were made using a Bruker BKr-322s pulsed spectrometer and variable-temperature probe operating at 12.0 MHz for 2H resonance with a 7r/2 pulse length of ca. 10 ps. Quadrupole splittings were obtained from the free-induction decay following a 71/2 pulse, and where necessary signal averaging was employed. The accuracy is estimated to be f 5%. Temperatures were measured by inserting a thermometer into the probe before and after each measurement and were constant within & 1 K. A Reichert Thermopan microscope with attached Koffler hot stage and camera were used to obtain photomicrographs.J. Chem.SOC., Faraday Trans. 1, Vol. 78, part 1 Plate 1 oc tanol liquid (L,) reversed hexagonal lamellar (L a ) PLATE 1 .-Photomicrograph of octanol penetration into a ternary mixture of KC,,/C,0H/2H,0 (43.8”/,/26.40/,/29.8%) at 302 K. (With crossed polars, magnification x 100.) NEESON AND TIDDY (Facing p . 149)P. G. NEESON A N D G. J. T. TIDDY 149 RESULTS MACROSCOPIC APPEARANCE A N D OPTICAL MICROSCOPY Samples with compositions corresponding to ‘K’ phase formed a stiff white gel at ambient temperatures. Repeated mixing and further storage at ca. 313 K (where a translucent material with no ‘whitish’ appearance did form) did not alter its state to the eye. Examination of these samples and those in neighbouring regions of the lamellar phase showed that they contained needle-like crystals corresponding to undissolved KC,,.(The amount of solid material increased markedly with KC,, concentration.) The disappearance of these crystals at higher temperatures (up to ca. 308 K) corresponded to the vanishing of the ‘whitish’ appearance. At the higher temperatures we were able to observe the previously reported* microscope textures of ‘K’ phase. We attempted to establish the occurrence of the ‘ K’ phase as a separate region using a ‘penetration experiment ’,11 where a lamellar phase sample containing crystals was surrounded by octanol and the temperature varied on the microscope stage. The presence of solid surfactant ensured that the chemical potential of lamellar soap was close to that of the pure material.Addition of liquid octanol ensures that the mesophase forming at the octanol (L,)/lamellar boundary also has an high octanol chemical potential. Thus if ‘K’ phase occurs as an independent entity with tie-lines to L, and solid KC,, it should be observed as a separate band under crossed polars. We were unable to observe such a band at 293 K. However, at 300 K a new band with a distinctive texture did occur (see plate 1). The texture corresponds to one commonly observed for normal or reversed hexagonal phases, with the latter possibility being judged the most likely. TABLE 1 .-X-RAY AND N.M.R. RESULTS FOR POTASSIUM DECANOATE/OCTANOL/WATER MIXTURES (FOR LOCATIONS ON PHASE DIAGRAM SEE FIG. 1) water A valuesc/kHz C,OHa 2H ,O( w t . %) X-ray spacingsb/A KC10 0.603 0.603 0.603 0.603 0.603 0.603 0.603 0.693 0.693 0.693 0.693 38.80 34.61 29.80 19.92 18.42 17.88 17.8 1 38.78 21.82 18.70 16.18 - - - 2.26 31 27d 2.80 30 27d 2.84 - - x 2.7 w 1.6 30 26 x 3.0 x 1.6 - x 2.4 x 1.6 - x 3.0 x 1.7 - - 2.50 - 30 26 w 3.0 w 1.7 2.04 30 25 - 1.86 - - - - - 25 - a Weight ratio; 31 1 K; 303 K; very low intensity.LOW-ANGLE X-RAY DIFFRACTION At room temperature (296 K) we obtained one diffraction peak at ca. 31 A for the ‘ K’ phase composition. This value is similar to the results of Ekwall et aL8 who report do values in the range 29.3-3 1.09 A. Additionally, they observed the d,/2 and d,/3 lines, which we were unable to obtain. A value of 30 A was also measured for samples in150 STRUCTURE OF LYOTROPIC MESOPHASE ‘ K ’ the neighbouring region of the lamellar phase, again similar to the previous results.8 At higher temperatures (31 1 K) we frequently observed a second diffraction peak at 25 A, in addition to the 30 A peak (see table 1). The second peak is consistent with the occurrence of a reversed hexagonal phase, which would be expected to have a lower do value than that of the lamellar phase (because it consists of a two-dimensional array and the surfactant aggregate is smaller).Similar do values have been reported for reversed hexagonal phase in the sodium octanoate/decanol/water system.12 We were unable to observe the d 0 / d 3 line required to confirm this structure by the X-ray technique. N . M . R . RESULTS The n.m.r. samples were prepared with deuterium oxide, because the 2H nucleus ( I = 1) gives rise to quadrupole splittings (A) in mesophases,13 and the relative magnitudes of these splittings provide an additional way of characterising phase structure.Other things being equal, normal or reversed hexagonal phases will have A values one half that of a lamellar phase with the same composition. Samples containing two liquid-crystal phases will give two superimposed spectra, with intensities corresponding to the amounts of water in each phase. At 296 K it was impossible to measure A values from the ‘ whitish ’ gel samples because the free-induction decay was too fast. This phenomenon is often observed with lyotropic liquid-crystal samples that contain dispersed solid phase. The n.m.r. results from various samples are listed in table 1 for measurements at 303 K.A single A value in the range 2.2-3 kHz is observed for lamellar samples, while ‘K’ phase samples show a second A value of ca. 2 kHz (where two overlapping spectra are present the accuracy of the measured A values is reduced). For a mesophase containing surfactants with labile hydrogens, the A values observed are a weighted average of the contributions from water and head-group deuterons if exchange is fast on the n.m.r. time-scale. This is the case with other soap/alcohol systemss*14 and is likely to occur here. In addition, the A contribution from water hydrogens arises mainly from the fraction of water ‘bound’ to head groups; the number of ‘ bound’ water molecules per surfactant is frequently observed to be invariant over a wide range of surfactant concentrations.Thus for a given decanoate/octanol molar ratio we expect the A values to be given by: where C is a constant, and na and nw refer to the concentration of amphiphile (soap plus alcohol) and water, respectively. From the A values of the lamellar phase we calculate A w 5 kHz for a lamellar composition in the ‘K’ region. The A values for the second phase in the ‘K’ region are less than half this; thus a lamellar structure can be excluded. The reduced A values do indicate the presence of a normal or reversed hexagonal structure because the A values expected for these phases are lower than those of a lamellar phase with the same composition by a factor of i. An exact agreement cannot be expected because the compositions of the two coexisting phases have not been determined.The fact that the spectrum of the first phase was not observed at the higher C,OH/KC,, ratio does not mean that no lamellar phase is present, because the technique is insensitive to small quantities of an additional phase. It does mean that the fraction of lamellar phase is reduced. Again, from the compositions of the second phase, we infer a reversed hexagonal structure.P. G. NEESON A N D G. J. T. TIDDY 151 DISCUSSION With all these techniques we have evidence for a difference in behaviour of ‘ K’ phase samples at ambient temperature and above 300 K. At room temperature the structure is lamellar plus crystalline KC,,, possibly with a small fraction of L, phase, although none was detected even with 2H n.m.r.measurements. (Water in the L, phase would give a sharp resonance that should be observed separately from the broad mesophase signal.) The ‘whitish’ appearance probably arises from light scattered by small crystals, while the presence of the needle-shaped crystals increases the apparent viscosity. Above 300 K an additional reversed hexagonal phase occurs. This is similar to the behaviour reported for many other systems by Ekwall and co-workers2~ 9 9 12. The volume fraction of reversed hexagonal phase increases with C,OH concentration, and samples close to the high octanol boundary of the ‘ K’ region are likely to form only the reversed hexagonal structure. From the compositions (50% KC,,/35% C,,OH), X-ray spacing (25 A), and assuming relative densities of 1.0 and 0.9 for aqueous and hydrocarbon regions, we calculate3 a water + head-group cylinder diameter of ca.18 A and a minimum alkyl-chain thickness of 10.8 A. The maximum extension of alkyl chains calculated for this structure is ca. 7.6 A. Thus both chains can easily pack into a reversed hexagonal structure. To summarise, all our evidence is consistent with the ‘K’ region consisting of lamellar and solid below 300 K, and the occurrence of an additional reversed hexagonal phase above 300 K. There is no requirement to invoke a novel ‘K’ structure. We thank Dr C. D. Adams and Mr C. King for technical assistance with the X-ray measurements, and the S.R.C. for the award of a CASE studentship to P.G.N. P. A. Winsor, Chem. Rev., 1968, 68, 1. P. Ekwall, in Advances in Liquid Crystals, ed.G. H. Brown (Academic Press, New York, 1971), vol. 1, chap. 1, p. 1. V. Luzzati, in Biological Membranes, ed. D. Chapman (Academic Press, New York, 1968), chap. 3, p. 71. G. J. T. Tiddy, Phys. Rep., 1980, 57, 1 . K. Fontell, paper presented at the 8th International Liquid Crystals Conference, Japan, July 1980. G. J. T. Tiddy, J . Chem. SOC., Faraday Trans. I , 1972,68,369; N. 0. Persson, K. Fontell, B. Lindman and G. J. T. Tiddy, J . Colloid Interface Sci., 1975, 53, 461 ; G. Lindblom, B. Lindman and G. J. T. Tiddy, Acta Chem. Scand., 1975,29,867; A. Forge, J. E. Lydon and G. J. T. Tiddy, J. Colloid Interface Sci., 1977, 59, 186. ’ I. D. Leigh, M. P. McDonald, R. Woods, G. J. T. Tiddy and M. A. Trevethan, J. Chem. SOC., Faraday Trans. I , 1981, 77, 2867. * P. Ekwall, L. Mandell and K. Fontell, J . Colloid Interface Sci., 1969, 31, 508. P. Ekwall, L. Mandell and K. Fontell, J . Colloid Interface Sci., 1969, 31, 530. lo P. Stenius and P. Ekwall, to be published. l1 A. S. C. Lawrence, in Liquid Crystals 2, ed. G. H. Brown (Gordon & Breach, New York, 1969). l 2 K. Fontell, L. Mandell, H. Lehtinen and P. Ekwall, Acta Polytechnica Scand., 1968, chap. 74, 111. l 3 A. Johansson and B. Lindman, in Liquid Crystals and Plastic Crystals, ed. G. W. Gray and P. A. Winsor (Ellis Horwood, Chichester, 1974), vol. 2, p. 192. l4 N. 0. Persson, H. Wennerstrom and B. Lindman, Acta Chem. Scand., 1973, 27, 1667; G. Lindblom, N. 0. Persson and B. Lindman, Proc. VI Int. Con$ Surface Active Agents (Carl Hanser Verlag, Miinchen, 1973), p. 939. vol. 1 , p. 1. (PAPER 1 /086)

ISSN:0300-9599    

DOI:10.1039/F19827800147

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

J. Chem. SOC., Faraday Trans. 1, 1982, 78, 153-164 Adsorption of Hydrogen on Zinc Oxide Energy of Interaction and Related Mechanisms BY BICE FUBINI,* ELIO GIAMELLO, GIUSEPPE DELLA GATTA AND GIOVANNI VENTURELLO Istituto di Chimica Generale ed Inorganica, Universita di Torino, Facolta di Farmacia, Via Pietro Giuria, 9, 10125 Torino, Italy Received 26th January, 198 1 The heat of adsorption of hydrogen on zinc oxide (Kadox 25) has been measured at room temperature using a Calvet microcalorimeter. Two main forms of adsorbed hydrogen have been found, coinciding with the literature types I and 11. Both always occur simultaneously (although with different kinetics) and a clear-cut separation between the two forms is not achieved on the basis of their reversibility. Differential heats of adsorption decrease with coverage from 60 to 14 kJ mol-*.Apart from the lowest coverages, where a real heterogeneity in site energies is present, the decrease is basically due to the simultaneous adsorption of type I (ca. 40 kJ mol-I) and type I1 (ca. 14 kJ mol-l), the latter prevailing at increasing coverage. The surprising feature that the ‘irreversible’ type I1 has a lower adsorption heat than the reversible type I is interpreted as evidence for diffusion of the adsorbate into the bulk, in agreement with kinetic observations. The interaction of hydrogen with ZnO has been the subject of many investigations because of the role adsorbed hydrogen plays in both methanol synthesis and the hydrogenation of unsaturated hydrocarbons. Various adsorption processes are known to occur, characterized by different mechanisms, kinetics and stability of adsorbed species, as shown in a recent review.I At room temperature two main processes have been identified, yielding two species called type I and type I1 by Dent and Kokes,2 but more than two peaks have been found by thermal desorption a n a l ~ s i s .~ - ~ Type I has been defined as rapid, dissociative, reversible and i.r. active; type I1 as slow and irreversible.,T Type I participates directly in catalytic reactions such as methanol synthesis, hydrogenation and H,-D, exchange, whereas it is not clear whether type I1 plays any role at all, as its effect on the rate of hydrogenation is still the subject of The two forms of adsorption have been assigned to the interaction with the two different crystal faces known to occur predominantly on ZnO Kadox,@ namely the polar ones (0001) for type I and the apolar ones (0101) for type 11.The model adopted for type I adsorption on the basis of i.r. investigations identifies a Zn-0 pair as active site on the (00Oi)-type faces., Boccuzzi et a1.l0 hypothesize a site originating from a triplet of Zn2+ exposed by an oxygen vacancy and at least one reactive 0,- ion. Hydrogen undergoes heterolytic dissociation thus yielding surface hydroxyls and hydrides. Taking into account that only 10% of the actual surface is involved in this type of adsorption, lateral interaction between adsorbed species should be ruled out and a substantial homogeneity in this adsorption process is expected, unless adsorption occurs in very well defined patches of the surface.T.p.d. peaks corresponding to type I are indeed unaffected by coverage;5 however, accurate kinetic analysis by Aharoni and Tompkinsll? l2 and the continuous shift of the i.r. bands with coverage found by Boccuzzi et al.l0 indicate the existence of slight heterogeneity or mutual interaction. Moreover, two different mechanisms giving rise to these species, 6 153 FAR 1154 ENERGY OF INTERACTION OF H, ON ZnO one involving charge transfer and one not, have recently been proposed by Murphy et a1.13 which can easily account for slight differences in the final surface arrangement. Type I1 hydrogen is less well defined. It is desorbed below 373 K but the t.p.d. peak temperature varies irregularly with c~verage.~ It has long been considered an i.r.inactive species21 although recently a broad i.r. band in the low-frequency region has been assigned to hydrogen bridge-bonded to couples of Zn2+ or 0,- on the apolar faces.l0 In spite of the many investigations of H, adsorption and catalytic mechanism on ZnO, energy data related to these processes are almost non-existent, the only calorimetric data available being those reported by Garner and Veal14 many years ago. We have therefore measured the heat of interaction of H, with ZnO as a function of coverage in order to characterize the different processes, assign energy values to the various species, detect possible surface heterogeneity and, finally, find out whether different binding energy can account for the difference in reversibility and catalytic activity between hydrogen adsorbed as type I and as type 11.EXPERIMENTAL The heat of adsorption and related values were measured in a Tian-Calvet microcalorimeter connected to a volumetric apparatu~.'~~ l6 ZnO samples were slowly heated under vacuum up to 673 K and 50 kPa 0, were admitted and removed several times; the sample was then cooled under 0, and transferred into the calorimeter. Complete evacuation was monitored through the endothermic signal arising from desorption of 0,. Adsorption was obtained by successive admission of increasing doses of the gas, up to ca. 3 kPa. Each dose was admitted only when equilibrium conditions were attained (constant pressure, no deviation from the calorimetric base-line). Stepwise desorption was obtained by expansion of the gas phase into a defined volume.Eventually the sample was pumped off by direct connection to high vacuum, in order to remove all the adsorbate that was reversible at the operating temperature. These procedures were performed several times on the same sample and will be hereafter indicated as runs I, I1 and 111. In one case the sample was contacted with an excess of H, (50 kPa) in order to saturate all sites and to define the maximum adsorption capacity. The calorimeter was kept at a constant temperature of 298 K. Zinc oxide (Kadox 25, New Jersey Zinc Co.) with 10 m2 g-l B.E.T. surface area was used. RESULTS The adopted pretreatment procedure leads to a surface composition which is as near as possible stoichiometric and fairly reproducible as far as activity towards simple gases such as CO is ~0ncerned.l~ Loss of surface area, which.is known to occur easily on Zn0,3911 was observed only in particular cases when the same sample was reactivated many times.A test of the adsorption capacity of CO ensured that in all cases the surface area was the same. However, as has been described elsewhere,17 slight differences in activity towards H, can be found from one sample to the other, due to the extreme sensitivity of the ~ystem.~ Results reported in the present paper have all been obtained on samples exhibiting similar activity towards H,, the activity being close to the average value obtained using a larger set of specimens. Three successive adsorption-desorption runs are reported in fig. 1 as Q int against p .In runs I1 and 111 adsorption values are reported from the last desorption point, obtained by complete evacuation of the adsorbate. As a corresponding volumetric point is lacking, similar n, against p curves cannot be drawn. However, they all follow the same trend for the first run (fig. 2) as the calorimetric ones.B. FUBINI, E. GIAMELLO, G. DELLA GATTA AND G. VENTURELLO 155 6 L 1 2 3 p/kPa FIG. 1 .-Heat related to hydrogen interaction in three successive adsorption4esorption runs (Qint against p). Insert: saturation with an excess of H,. Open symbols: adsorption; filled symbols: desorption. 1 2 3 PlkPa FIG. 2.-Adsorption and desorption in the first run (n, against p ) . Open symbols: adsorption; filled symbols: desorption. The main features of these curves are: (i) Adsorption is pressure dependent in the whole range examined.Even a very small adsorbed amount, such as the first point in fig. 1 and 2 (0.09 pmol m-2), yields a pressure of 20 Pa over the sample. (ii) The slopes of the adsorption-desorption curves show that the adsorption process is far from finished at the end of each adsorption run. Higher pressure ranges have not been examined in detail; however, ‘saturation’ of a sample by exposure to a pressure of ca. 50 Pa for a period of four days involved a heat of adsorption (from 2.8 kPa to total saturation) of 4.35 x low2 J m-2, which corresponds to ca. 50% of the total heat evolved (insert of fig. 1). (iii) Adsorption and desorption follow different paths in the whole range examined, not only in the first run but in all the successive runs as well.6-2156 ENERGY OF INTERACTION OF H, ON ZnO This behaviour appears even in successive adsorption4esorption runs performed after ‘saturation’ of the ZnO sample with H,. (iv) In each adsorption run some reactions occur yielding adsorbed species which are not removable by evacuation at room temperature. This fraction of adsorbate is measured on the ordinate axis in fig. 1 by the last desorption point, which represents the amount of heat evolved upon adsorption and not recovered upon desorption. Although a slight decrease in the amount of this fraction in successive runs can be detected, the formation of ‘irreversible adsorbate’ is far from being exhausted in the third run. This fraction seems to be related to the time of contact of H,.Points in fig. 1 and 2 have therefore to be considered pseudo-equilibrium ones, although taken when the system appeared to be in equilibrium as far as was detectable by our instruments. Integral heats of adsorption on various samples (run I) are reported in fig. 3 as a function of adsorption. The integral heat of desorption obtained by successive expansions in all runs (I, 11, 111) and on various samples are reported in fig. 4, as a function of desorption. I 1 I 0.4 0.8 1.2 n,/pmol md2 FIG. 3.-Integral heat of adsorption against amount adsorbed (run I) on various samples. Different symbols indicate different samples. Fig. 3 shows a smooth curve, whose slope decreases with coverage, indicating a heterogeneity in site energy distribution. On the other hand, all the points in fig.4, although scattered, lie on a straight line passing through the origin. The differential heat of desorption calculated from the slope (37 kJ mol-l) is thus fairly constant over the coverage range examined by stepwise desorption. The variation of differential heat of adsorption with coverage has been obtained by differentiation of the curve in fig. 3 and it is reported in fig. 5 as a solid line. The same data for run I1 are also reported; run I11 coincides with run 11. The heat of desorption is reported in fig. 5 as a broken line at the appropriate coverage value. Three main features emerge from the figure: (i) Differential heats of adsorption strongly decrease with the coverage from an initial value of 60 kJ mol-l down to ca. 14 kJ mol-l.(ii) The heat of adsorption is lower than the heat of desorption, in the stepwise desorption range. (iii) The initial heat of adsorption is higher in run I1 (69 kJ mol-l) than in run I (60 kJ mol-l). Such an increase of ca. 15% has been observed on various samples.B. FUBINI, E. GIAMELLO, G. DELLA GATTA A N D G. VENTURELLO 157 2 N E h a2 0.4 ndlpmol m-2 FIG. 4.-Integral heat of desorption against amount desorbed (runs I, I1 and 111) on various samples. Different symbols indicate different samples. Dashed symbols: run I1 ; crossed symbols : run 111. 80 20 1 I I 1 I I I 0.2 0.4 0.6 0.8 1.0 1 2 n,/pmol m-2 FIG. 5.-Differential heat of adsorption against amounts adsorbed (runs I and 11). Broken line: differential heat of desorption. The heat of emission peaks corresponding to the adsorption of the first dose (runs I and 11) are reported in fig.6(a). They both exhibit the characteristic thermokinetic behaviour of activated chemisorption,16 i.e. the curve is less sharp than the standard one for instantaneous heat evolution and the calorimetric zero is attained after a much longer time than the zero due to instrumental inertia. Note that in our case this time was always ca. 6-7 h for the first peak of the first run. Subsequent adsorption either in the same or in successive runs was accomplished within 24 h. Something peculiar is thus occurring with the first contact of H, with ZnO samples. Although after the attainement of the calorimetric zero no further appreciable development of heat was observed, in agreement with previous results by Dent and Kokes,2f we noticed that, after several days of contact, the H, pressure over the ZnO samples decreased slightly.Adsorption corresponding to this decrease is such that the158 ENERGY OF INTERACTION OF H, ON ZnO corresponding heat, slowly evolved over a period of days, could not be detected in our apparatus. Desorption obtained by evacuation of the sample is monitored on the calorimeter as a broad endothermic peak [fig. 6(b)]. Several such peaks obtained by evacuation at different pseudo-equilibrium pressures are reported. The general features of these curves are similar, except for the time taken for the attainment of the calorimetric zero and for a marked variation in the position of the maximum, which is displaced on the time axis as the coverage decreases.Fig. 6 also compares desorption heat peaks from both stepwise expansions [fig. 6(c)] and direct evacuation [fig. 6(d)] with the similar peaks obtained by desorption of CO from ZnO (broken line). The activation energy for desorption of hydrogen is clearly higher than the activation energy for desorption of carbon monoxide. Jx3.3 I I time/min 0 120 380 time/min time/min time/min 20 40 20 40 .- I zi ‘c) FIG. 6.-Themokinetics. (a) Heat emission peak corresponding to adsorption of the first dose in run I (solid line) and run I1 (dashed line). (b) Desorption by evacuation of H, from ZnO at different coverages : curve 1, p = 2 kPa; curve 2, p = 0.6 kPa; curve 3, p = 0.133 kPa; curve 4, p = 0.066 kPa. (c) Stepwise desorption peaks of H, (solid line) and CO (broken line).(d) Desorption by evacuation of H, (solid line) and CO (broken line).B. FUBINI, E. GIAMELLO, G. DELLA GATTA A N D G. VENTURELLO 159 DISCUSSION DIFFERENT FORMS OF ADSORBED HYDROGEN At room temperature hydrogen interacts with ZnO in at least two different forms, which we will identify to a first approximation with the ‘reversible’ and ‘irreversible’ species type I and type 11, hereinafter referred to as H(1) and H(I1). These two processes take place simultaneously and both depend on H, pressure so that they cannot be separated at different coverages along the same isotherm, as is possible in other cases.l* Adsorption and desorption do not coincide in the whole range examined. Moreover, adsorption in the so-called irreversible form occurs to a considerable extent after successive adsorption-desorption cycles, so that we cannot draw conclusions as to the contributions from the two forms by a simple comparison of the first and second runs.Adsorption values fall within the range of those obtained by other authors2$ 5 9 l9 at the same pressure. Dent and Kokes29’ did separate H(1) from H(I1) by simple evacuation (30 min) of a sample exposed to high H, pressure (17 kPa): all H, successively adsorbed on this sample was then assigned to the H(1) form. In our case, adsorption of H(1) is accompanied in all successive runs by adsorption of H(I1). It could be inferred that, as H(I1) adsorption is a very slow and pressure- dependent process, the adsorption capacity up to 3 kPa is not exhausted in the first run.In that case the ‘irreversible’ adsorption seen in later runs could be interpreted as a mere continuation of the slow process interrupted in the preceeding run. Although this can account for some irreversible adsorption taking place after many adsorption- desorption runs, we discard this as a unique interpretation for two reasons: (i) At the end of each run, the adsorption rate is very slow so we can confidently consider the isotherms in fig. 2 as pseudo-equilibrium isotherms. If adsorption of H(I1) in the next run were just a continuation of it, it should occur at an even lower rate, being at a lower pressure than the final one in the first run. This is not the case. Adsorption in form I1 occurs in each successive run in the same way, with the same kinetics and in considerable amounts. Only a slight decrease in the total irreversible adsorption (fig.1) occurs run after run. (ii) Adsorption and desorption after evacuation of the H,-saturated sample still show the same features, i.e. different adsorption and desorption paths and new sites for ‘irreversible ’ adsorption. Note also that the time lag of desorption in our system (4 h directly pumping-off) is much longer than the one adopted by Dent and Kokes (30 min); even when the pressure over the adsorbent is very low, the calorimeter monitors the presence of endothermic processes still occurring, ascribable at a first approximation to the desorption of the more strongly held absorbate. Therefore some of the so-called ‘irreversible ’ adsorption becomes ‘ reversible ’ in our apparatus.Nevertheless, some indications of the adsorption energies related to the two forms can be drawn by molar heats, obtained by partial separation of the two processes. Final total values in run I of both n, and Qint (fig. I and 2) are slightly higher than the corresponding values in run 11, if the run I1 isotherms are computed from a zero initial point. This difference has to be ascribed to irreversible chemisorption, which occurs to a higher extent in run I. The partial molar heat of this fraction measured at 3 kPa is 16 kJ mol-l, much lower than the overall molar heat at the same pressure (33 kJ mol-l). Accordingly the initial differential heat is higher in run I1 than in run I (fig. 5). Moreover the heat of desorption in the range of stepwise desorption is much higher (37 kJ mol-l) than the heat of adsorption in the same coverage range (20 kJ mol-l).160 ENERGY OF INTERACTION OF H, ON ZnO These findings clearly indicate that the heat of adsorption related to the formation of H(1) (fast and reversible) would be higher than the heat of adsorption related to H(I1) (considered as slow and irreversible).This fact is very unusual. As a general rule an irreversible adsorption has always been considered to have a higher heat of adsorption than a reversible adsorption. This would account for the higher energy of desorption and hence the different possibilities of removing the adsorbate by pumping-off. In some cases, where both molecular and dissociative adsorption were present, an inversion of these characteristics has been observed, caused by a significant difference in the entropic content of the two adsorbates.20 In the present case we are dealing with two dissociative chemisorptions on two different crystal lo It seems very difficult to envisage such a great difference in adsorption mechanisms, able to account for such a different kinetic behaviour, when heat of adsorption values indicate the opposite.We have to discard any hypothesis concerning desorption of gases other than H,, in order to justify a heat of desorption higher than the heat of adsorption in the same coverage range: in that case the heat of desorption would not be constant in such a wide coverage range and could by no means be the same in different runs (fig. 4).Moreover none of the authors who investigated the same system by t . ~ . d . ~ * ~ mentioned the possibility of desorption peaks not arising from H, desorption. We have therefore to look at the complexity of the adsorption features of H(II) (p and t dependence, occurrence at each run) for the cause of its peculiarly low heat of adsorption. On the basis of one adsorption-desorption cycle (run I) we have attempted a breakdown of calorimetric and volumetric isotherms, in the range where data are available for both (0.133-3 kPa). The decomposition scheme is illustrated in fig. 7(a). At any pressure value (p,) the difference between the adsorption and the desorption branches (a,) represents the amount adsorbed (or heat evolved) related to irreversible processes occurring from that point to the end of the run.Considering a set of regular pressure intervals, the difference a, - a,,, measures irreversible adsorption in the pressure interval p , - ~ ~ + ~ . A set of data related only to the irreversible processes can thus be obtained from a limiting value (133 Pa) up to the end of the run and is reported in fig. 7(b) and (b’) as calorimetric and volumetric isotherms. The corre- sponding ‘ reversible ’ isotherms, obtained by subtraction, are also reported (broken line). Adsorption below 133 Pa is here incorrectly reported as wholly reversible, even if a small fraction (0.2 x lo-, J m-,) is irreversible, the corresponding adsorption value being unknown. Relevant Qint against n, plots are reported in fig, 7(c). The main results obtained by this decomposition are: (i) The ‘reversible’ adsorption has an asymptotic trend and most adsorption occurs below 0.7 kPa, whereas the ‘ irreversible ’ adsorption increases almost linearly with pressure and is not exhausted in the pressure range examined.This is nearly the opposite of what one could expect in classical reversible and irreversible adsorption. (ii) Both processes proceed in the pressure range examined with constant differential heat of adsorption ; namely qdiff (reversible) = 40 kJ mol-1 and qdiff (irreversible) = 14 kJ mol-l. These values are fairly consistent with the heat of desorption (37 kJ mol-l) and confirm a very low heat of adsorption in the ‘irreversible’ form. The reversible process, having a differential heat of 37-40 kJ mo1-1 above 133 Pa, can be easily ascribed to the species denoted as H(I).On the other hand, the assignment of the irreversible processes with a low heat of adsorption to a well defined model is more difficult. Assignment of the whole irreversible adsorption to the species assigned by Boccuzzi et aZ.1° to H atoms bridged on apolar faces has to be disregarded, the heat of adsorption being far too low [by comparison with that for H(I)] for an irreversible and dissociative chemisorption.B. FUBINI, E. GIAMELLO, G. DELLA GATTA AND G. VENTURELLO 161 Some characteristics of the process would instead indicate that possible diffusion in the bulk onto sub-surface layers has also to be taken into account. This would be in agreement with : (i) The continuous irreversible uptake in subsequent adsorption- desorption runs.(ii) The correlation between time of contact and irreversibly adsorbed amounts. (iii) The linear isotherm trend, characteristic of dissolution and not of irreversible chemisorption. (iv) The very low heat related to this process. ct 0.8 E - 0 --. f ,m 0.4: t P - _--- /- /- +.+' I /+ ( b ) /+' I I I , 1' / I +,+'+ 4 1 h 7 2 2 '54 . I c I 1' / / / / I I 4 I 1 - 0.4 0.8 n,/pmol m-* FIG. 7.-Decomposition of experimental pint and n, against p curves in their reversible and irreversible fraction. (a) Decomposition scheme. (b) pt against p (0.133 < p/kPa < 2.8). (b') n, against p (0.133 < p/kPa < 2.8). (c) Integral heat curves (Qint against n,). It is well-known that in the case of absorption, i.e. any sort of penetration into the bulk from the surface, the heat evolution can be very low and in some casesz1 endothermic, as the process is mainly guided by entropic effects.We recall here that the ZnO lattice has a very open structure with straight channels located on the apolar faces (0701) of 0.2 nm diameter separated by a trigonal squeeze point of 0.12 nm through which a hydrogen atom or even a molecule can easily penetrate into the bulk.zz, 23 The squeeze point can account for the extreme slowness of the process and its substantial irreversibility. A schematic diagram for energy relationships in such a system is reported in fig. 8. The possibility of some penetration into the bulk had been also invoked by Dent and Kokes to account for the non-detectability by i.r. of type I1 and by Baransky and Galuzska to account for the irregular t.p.d.peak temperat~re.~ Interstitial Zn+ and oxygenz4 in ZnO have been identified. Mechanisms for diffusion of hydrogen in other oxides, e.g. TiOZz5 and have been described recently. On the ammonia synthesis catalyst, hydrogen is reversibly adsorbed but irreversibly dissolved in the subsurface The irreversible uptakes therefore would be comprised of at least two phenomena: the dissociative adsorption, yielding a bridged hydrogen as found by Boccuzzi et a1.l0 and occurring at a first stage, followed by diffusion of the adsorbate in the bulk, through the wide channels present on the apolar faces. z8162 ENERGY OF INTERACTION OF H, ON ZnO t FIG. 8.-Schematic diagram for energy relationship in the case of hydrogen adsorption-absorption on ZnO.Activation energy for adsorption, E, ; for desorption, E,; for penetration into subsurface layers, E3; for diffusion in the bulk, E4. -AHl, enthalpy change upon adsorption; +AH,, enthalpy change from adsorption to absorption; -AH,, enthalpy change from gaseous to absorbed state. KINETICS As to the kinetic features of H, adsorption a possible interpretation is the following: (1) The first adsorbed dose of the first run is comprised of H(1) and H(I1) in the i.r. active form. This latter process, being dissociative and irreversible, will probably be slower than the former, thus explaining the slow thermokinetics in this first step. The slow step could also be caused by generation of active centres upon chemisorption, as pointed out by Naito et al.19 (2) Diffusion into the bulk, through channels present at the apolar faces, occurring all along the adsorption process, has a very low overall heat of reaction, probably due to some activated endothermic steps (fig.8). As a consequence, although very slow, this process is not detectable as slow thermokinetics, but is mainly shown by the continuous decrease of the pressure over the adsorbent for a period of several days. We now suggest that the characterization of the various processes as 'reversible' and ' irreversible ' is misleading because we are dealing with pseudo-equilibrium data. H(1) and H(I1) at the surface and H(I1) diffusion into the bulk should probably be regarded as processes characterized by different kinetic behaviour accounting for their apparent reversibility and irreversibility. Note that previously some authors2y ' 9 29 already distinguished a fast part and a slow part in the adsorption of the H(I1) form, probably corresponding to our surface and bulk processes.Although adsorption of the H(1) form is the fastest process, thermokinetics indicate that this process is also activated. By comparing the peak for desorption of H, with those of CO we compare the kinetics of H(1) with those of a fast, instantaneous and reversible coordination of CO on similar sites30 with a slightly higher heat of adsorption (44 kJ m~l-').~' As the activation energy for desorption is higher for H, than for CO [fig. 6(c) and (43 it must be that Ea(des) H(1) = Ea(ads) H(1) + 35 kJ mol-l > Ea(des) CO = 44 kJ mol-l; thus &(ads) H(1) > 9 kJ.Accordingly values of E, = 17 kJ have been reported by Baransky et al.5 All forms of dissociative adsorption of H, on ZnO at room temperature haveB. FUBINI, E. GIAMELLO, G. DELLA GATTA A N D G. VENTURELLO 163 therefore to be regarded as kinetically activated processes. The possibility for one species [H(II)] to diffuse into the bulk accounts for the apparent irreversibility and slowness of the process. On the other hand, the complete reversibility of adsorption of H(I), in spite of its dissociative and activated features, is probably connected with the characteristics of adsorption sites on polar faces, where 0-H and Zn-H are located at such a short a distance that a low-frequency factor for desorption as H, can be invoked.SURFACE HETEROGENEITY The variation of the differential heat of adsorption against coverage shown in fig. 5 is characteristic of interaction of gases on heterogeneous surfaces and similar to data found by Garner and Veal.14 ZnO heterogeneity towards hydrogen has been already pointed out and considered both as structural and induced on the basis of kinetic The limited shift of Zn-H and Zn-OH stretching frequencies with coverage in i.r. spectralo seems to confirm this hypothesis. On the other hand, t.p.d. data3-6 indicated that adsorption at room temperature leads to three peaks, namely IA, IB and 11, whose peak temperature did not vary with coverage as expected in the case of heterogeneous adsorption. Our data indicate an intermediate situation; in a wide pressure range (p > 133.3 Pa), adsorption in forms H(1) and H(I1) occurs with fairly constant heat (fig.4 and 7) for each mechanism and the overall decrease in the heat of adsorption against coverage plot (fig. 5) is due only to the superposition of the two processes and the prevalence of H(I1) compared with H(1) at increasing pressure (fig. 7). This accounts for the constancy of the peak temperature against coverage in t.p.d. data, although the presence of two different peaks (IA and IB) would suggest a different mechanism for thermal desorption. Below 133.3 Pa, 0.4 mol rn-, are adsorbed mainly as H(1) or H(I1) in the i.r. active, surface form. These processes occur with a higher heat of adsorption (partial molar heat of adsorption = 55 kJ mol-l) which decreases markedly with coverage from a value of 60 kJ mol-l (zero coverage) to a value of 40 kJ mol-l (133.3 Pa).Most probably in this first range, not only do the two processes occur together but some heterogeneity in H(1) adsorption sites is present, in accordance with kinetic and i.r. data and with the possible existence of more than one mechanism yielding H(I).13 A confirmation of the substantial heterogeneity in adsorption of H, in the ' reversible form ' at low coverage is given by fig. 6 (b). The desorption curves obtained by pumping-off the sample are a representation of desorption rates, although deformed by the calorimeter inertia. Their general smoothed shape is obviously due to a decrease of the desorption rate along the desorption process itself.In the case of the desorption of a species characterized by a constant heat of adsorption, the desorption curve presents the same shape starting from any coverage, the only difference being in the overall heat measured, as in curves 1 and 2 (high H, coverage). On the other hand, significant variation occurs at low coverage (curves 3 and 4) indicating a variable energy of activation for desorption. CONCLUSIONS On the basis of calorimetric and thermokinetic results, the two main species arising from hydrogen adsorption on ZnO, denoted as H(1) and H(II), have to be distinguished more on a kinetic basis (slow and fast) than on their reversibility or irreversibility. The slow and irreversible process [H(II)] probably comprises two different steps : an adsorption at the apolar faces forming a stable bridged H complex, i.r.active, and a very slow diffusion into the bulk, which escapes detection by surface techniques but164 ENERGY OF INTERACTION OF H, ON ZnO is probably the prevailing form from a quantitative point of view, at the highest coverage. Diffusion process can account for the very low heat values (14 kJ mol-l) in spite of the ‘irreversibility’ of the process. Adsorption as H(1) on the polar faces occurs to some extent with a constant heat of adsorption, on isolated and similar sites; nevertheless at low coverage some heterogeneity is found, which indicates the presence of a fraction of high-energy sites. Although such a process has been regarded as fast, it has a definite activation energy in accordance with the proposed dissociative mechanism. Part of the experimental work was performed by Dr.G. Trucco while preparing her thesis. This research was supported by the Italian Consiglio Nazionale delle Ricerche. C. S . John, Catalysis by Zinc Oxide, in Catalysis, ed. C. Kemball and D. A. Dowden (Specialist Periodical Report, The Chemical Society, London, 1980), vol. 3, p. 169. A. L. Dent and R. J. Kokes, J. Phys. Chem., 1969, 73, 3781. A. Baranski and R. J. Cvetanovic, J. Phys. Chem., 1971, 75, 208. A. Baranski, R. J. Cvetanovic, T. Dal and J. Galuszka, Proc. IInd Int. Symp. Heterogeneous Catalysis, Varna, November 1971, Commun. Dept. Chem., Bulg. Acad. Sci., 1973, 6, 135. A. Baranski and J. Galuszka, J. Catal., 1976, 44, 259. M. Watanabe, J . Res. Inst. Catal., Hokkaido Univ., 1978, 26, 63.R. J. Kokes and A. L. Dent, Adv. Catal., 1972, 22, 1. D. Narayana, J. La1 and V. Kesavulu, J. Phys. Chem., 1970, 74, 4150. T. Morimoto and L. Moriskige, J. Phys. Chem., 1975, 79, 1573. lo F. Boccuzzi, E. Borello, A. Zecchina, A. Bossi and M. Camia, J. Catal., 1978, 51, 150. l1 C. Aharoni and F. C. Tompkins, Trans, Faraday SOC., 1970, 66, 434. l2 C. Aharoni and F. C. Tompkins, Adv. Catal., 1970, 21, 1. l3 W. R. Murphy, T. F. Veerkamp and T. W. Leland, J. Catal., 1976, 43, 304. l4 W. E. Garner and F. J. Veal, J. Chem. Soc., 1935, 1436 and 1487. l5 G. Della Gatta, B. Fubini and G. Venturello, J. Chim. Phys., 1973, 70, 64. l6 B. Fubini, Rev. Gen. Therm., 1979, 18, 297. l7 E. Giamello and B. Fubini, React. Kinet. Catal. Lett., in press. lS S . Naito, H. Shimizu, E. Hagiwara, T. Onishi and K. Tamaru, Trans. Faraday Soc., 1971, 67, 1519. 2o B. Fubini, G. Della Gatta and G. Venturello, J. Colloid Interface Sci., 1978, 64, 470. 21 T. B. Flanagan, Hydrides Energy Storage, ed. A. F. Andresen (Pergamon Press, Oxford, 1977), 22 A. L. Dent and R. J. Kokes, J. Phys. Chem., 1969, 73, 3781. 23 W. Gopel, Ber. Bunsenges. Phys. Chem., 1978, 82, 744. 24 A. Hausmann and B. Schallenberger, Z . Phys. B, 1978, 31, 269. 25 J. B. Bates, J. C. Wang and R. A. Perkins, Phys. Rev. B, 1979, 19, 4130. 26 J. Vitko Jr, C. M. Hartwig and P. L. Mattern, Proc. Int. Con$, New York, 1978, ed. R. Pantelides 27 V. E. Ostrovskii and E. G. Ingranova, Kinet. Catal. (USSR), 1978, 19, 68 1. 28 E. G. Igranova, V. E. Ostrovskii and M. I. Temkin, Kinet. Catal. (USSR), 1976, 17, 1257. 2s R. P. Eischens, W. A. Pliskin and M. J. Low, J. Catal., 1962, 1, 180. 30 F. Boccuzzi, E. Garrone, A. Zecchina, A. Bossi and M. Camia, J . Catal., 1978, 51, 160. 31 E. Giamello and B. Fubini, to be published. 32 R. J. Kokes, C . C. Chang, L. T. Dixon and A. L. Dent, J. Am. Chem. Soc., 1972, 94, 4429. G. Della Gatta, B. Fubini and L. Stradella, J. Chem. Soc., Faraday Trans. 2, 1977, 73, 1040. p. 135. (Pergamon Press, Oxford, 1978), p. 215. (PAPER 1 / 1 15)

ISSN:0300-9599    

DOI:10.1039/F19827800153

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

J. Chem. SOC., Faraday Trans. I, 1982, 78, 165-169 Ultrasonic Behaviour of the System Water + Acetonitrile BY KOPPARAPU SUBBARANGAIAH, NERIYANURI M. MURTHY AND SARASWATULA V. SUBRAHMANYAM* Department of Physics, SVU Autonomous Post Graduate Centre, Anantapur 5 15 003, India Received 2nd February, 198 1 The effect of acetonitrile on the temperature of the adiabatic compressibility minimum (t.a.c.m.) and temperature of the sound velocity maximum (t.s.v.m.) of water has been studied. The structural contribution to the shift in t.a.c.m. and t.s.v.m., (ATg,str.)exp, and (ATU,str,)exp., has been found to be positive at low concentrations, becoming maximum around A', 2 0.0 18 and thereafter decreasing to become negative around A', 1.0.045. This behaviour is explained in terms of the effects of added acetonitrile on the structure of water.The behaviour of the water + acetonitrile system has attracted the attention of many investigators since it exhibits many unusual features in the variation of excess functions of mixing with comp~sition.l-~ The variation of excess dielectric constant with concentration for this system at different temperatures3 reflects the existence of many structural discontinuities and acetonitrile is not considered to be a strong structure breaker. A plot of the heat of solution against the structural contribution to the shift in the temperature of maximum density of water at infinite dilution for various solutess indicates that no decisive conclusions can be drawn regarding the structural propensity of acetonitrile.While most of the experimental data available in the literature indicate that acetonitrile is a weakly hydrophilic solute, no direct evidence exists clearly supporting the structure-promoting nature of acetonitrile. However, in very dilute solutions one cannot rule out the existence of a hydrophobic interaction between acetonitrile and water molec~les.~~ 5 7 In view of this peculiar behaviour of acetonitrile, we have studied the effect of acetonitrile on the temperature of the adiabatic compressibility minimum (t.a.c.m.) and the temperature of the sound velocity maximum (t.s.v.m.) of water in order to gain a better understanding of the structural interactions in dilute solutions. The results are reported in this paper. EXPERIMENTAL A.R. grade acetonitrile was used after purification following the procedure cited in ref.(10). The density of purified acetonitrile determined at 25 "C using a bicapillary-type pyknometer with an accuracy of 2 parts in lo5 was found to be 0.77804 kg M - ~ and this value is in agreement with the data of ref. (9). Ultrasonic velocities in pure liquids and solutions were determined using a single-crystal variable-path interferometer working at 3 MHz with an accuracy of + 0.003 %. The details of the experimental technique were reported in a previous paper." The densities of and ultrasonic velocities in aqueous acetonitrile solutions of different concentrations (prepared by weight using triply distilled degassed water) have been determined at intervals of ca. 2 "C over a range of 5 "C on either side of the t.a.c.m. and t.s.v.m.165ULTRASONIC BEHAVIOUR OF H,OfCH,CN TI"C I .-Adiabatic compressibility, B, plotted against temperature of solutions of aqueous acetonitrile. A, X, = O.oo00; B, X, = 0.0088; Ci X2= 0.0126; D, X2 = 0.0216; E, X, = 0.0292; F, X, = 0.0381; G, X, = 0.0491. A, D, F, right-hand scale; B, C, E, G, left-hand scale. RESULTS Ultrasonic velocity data have been corrected for diffraction effects.12 The results for the measurement of adiabatic compressibility (evaluated using the formula = u - ~ p-l with an accuracy of f 5 x m2 N-l) of and ultrasonic velocity in aqueous acetonitrile solutions as a function of temperature are shown graphically in fig. 1 and 2, respectively. Even though the water + acetonitrile system exhibits ultrasonic relaxation,' we do not expect relaxation at 3 MHz in aqueous acetonitrile solutions (< 0.05 mole fraction at 2 25 "C).The template method as described in our earlier comm~nication~~ has been employed to fix the t.a.c.m. and t.s.v.m. with an accuracy of f0.4 and f0.2 OC, respectively.K. SUBBARANGAIAH, N. M. MURTHY A N D S. V. SUBRAHMANYAM 167 3.0 2.5 2.0 - I E . J 46 48 50 52 54 56 5 8 60 6 2 64 66 ' 4.0 F G t 476 X t 4 1.0 t - - 3.0 G - / \ I I I I I I I 1 1 FIG. 2.-Ultrasonic velocity, u, plotted against temperature of solutions of aqueous acetonitrile. A, x2 = 0.0491. The values given on the ordinate have to be added to 1550 m s-l. A, B, D, F, right-hand scale; C, E, G, left-hand scale. X, = 0.0000; B, X, = 0.0088; C, X, = 0.0126; D, X, = 0.0216; E, X, = 0.0292; F, X, = 0.0381; G, DISCUSSION The structural contributions to the shift in t.a.c.m., ( A q , str.)exp., and t.s.v.m., (AG, str.)exp., have been evaluated using the following expressions, as discussed in a previous communication.l3 and %,id. = [74-(2) 2 K (&&9)(3]* (4) In the above expressions 41, b2 and W,, W, represent the volume fractions and weight fractions of pure water and the organic solute, respectively, in solution. ag and aj are168 ULTRASONIC BEHAVIOUR OF H,O+CH,CN the coefficients involved in the relation for adiabatic compressibility of organic solute as a function of temperature ( T ) , namely ( 5 ) and a, represents the coefficient involved in the expression for the temperature dependence of velocity in the organic solute, namely p2 = E+ap T+ag TL u, = ui - a, T.(6) u, and u2 are velocities in pure water and organic solute, respectively. The values of E, aP, ag, ui and a, for acetonitrile are evaluated using the experimentally determined ultrasonic velocity and density data at different temperatures and the values, respectively, are 65.629 x m2 N-l, 0.405 x m2 N-l O C - - l , 3.880 x m2 N-l O C 2 , 1381.38 m s-l and 3.9937 ms-l OC-l. Since the temperature dependence of 4, and d2 is small, $bl and $b2 used to determine q, id. G, id. were those calculated at q and of the solution found experimentally. A successive approximation method was used for evaluation of G, id.. The values of ( A q , str.)exp. and (AG, str.)exp. at different mole fractions (X2) of acetonitrile are given in table 1 and are shown graphically in fig.3. 5, id. 5, exp. q, id. c, exp. (Aq, str.)exp. str.)exp. x2 1°C 1°C 1°C 1°C /"c 1°C 0.0088 56.9 58.5 66.0 68.3 1.6 f 0.4 2.3 & 0.2 0.0126 54.0 56.6 63.0 66.0 2.6 f 0.4 3.0 f 0.2 0.0216 47.9 49.6 56.5 60.0 1.7f0.4 3.5 f 0.2 0.0292 43.2 44.3 51.8 53.8 1.1 k0.4 2.0 & 0.2 0.038 1 38.3 38.5 46.9 47.6 0.2 f 0.4 0.7 0.2 0.049 1 32.7 32.2 41.4 41.0 -0.5k0.4 - 0.4 f 0.2 -21 I 102 x, FIG. 3.--Structural shifts against mole fraction, X,, of acetonitrile. 0, (ATg,str.)exp. ; 0, (AT,, str.)exp.. Both (AZ), str.)exp. and (AG, str.)exp. are positive at low mole fractions of acetonitrile becoming maximum around X, 11: 0.0 15 and 0.01 8, respectively. Beyond these optimum concentrations both the structural shifts decrease becoming negative around X2 11 0.04 and 0.045.The negative values of the structural shifts clearly demonstrate the structure-disrupting nature of acetonitrile beyond X2 N 0.045. However, the positive values of the structural shifts as observed in this study indicate the structure-stabilizing tendency of acetonitrile at low concentrations. This behaviour of acetonitrile isK. SUBBARANGAIAH, N. M. MURTHY AND S. V. SUBRAHMANYAM 169 contrary to the density maximum studies of Wada and Umeda14 wherein acetonitrile has been classified as a structure breaker even though infrared absorption studies indicate that acetonitrile is not a strong structure breaker7. l5 when X2 < 0.04. The positive sign of the excess heat capacity of the water + acetonitrile system and the sigp and magnitude of the initial slope of the plot of apparent molar heat capacity against mole fraction of acetonitrile4 and slightly exothermic nature of enthalpy of mixing5 in the range 0 < Xz < 0.04 at 25 O C indicate, at least to some degree, the tendency of acetonitrile to promote the hydrogen-bonded structure of water; these support the results of the present study.The geometry of acetonitrile is such that it cannot fit exactly into the aqueous lattice and hence the added acetonitrile molecules occupy the void space of the hydrogen- bonded clusters leading to the formation of aggregates where cavities are still available. This may be considered as a stabilization of the hydrogen-bonded clusters against thermal collapse owing to the formation of hydrogen bonds between acetonitrile and water and it is perhaps this behaviour that is reflected in the positive values of (AG, str.)exp.and (AC, str.)exp.. These cavities are filled until a certain percentage which corresponds to the maximum in ( A q , str.)exp. and (AC, str.)exp. against X2, i.e. X2 N 0.01 8. The saturation of the cavities by acetonitrile is also reflected in the extrema of excess volume and dielectric polarization,l? enthalpy and entropic solvation parameters,16 ultrasonic absorption' and the inflection points of several excess functions2* 4-6 occurring Ot definite concentrations of acetonitrile. When X2 > 0.018 the structural shifts decrease with increasing X2 and this behaviour is due to the progressive collapsaof the larger aggregates into smaller ones as the concentration of acetonitrile increases.The complexity in the behaviour of the water + acetonitrile system as revealed from different studies may be attributed to the nature of measured parameter and the effect of temperature since different properties respond to different phenomena. l7 C . Moreau and G. Douheret, Thermochim. Acta, 1975, 13, 385. F. Mato and J. L. Hernandez, An. Quim., 1969, 65, 9. C. Moreau and G. Douheret, J. Chem. Thermodyn., 1976, 8, 403. C. DeVisser, W. J. M. Heuvelsland, L. A. Dunn and G. Somsen, J. Chem. SOC., Faraday Trans. I , 1978, 74, 1159. A. L. Vierk, 2. Anorg. Chem., 1950, 261, 283. K. W. Morcom and R. W. Smith, J. Chem. Thermodyn., 1969, 1, 503. D. A. Armitage, M. J. Blandamer, M. J. Foster, N. J. Hidden, K. W. Morcom, M. C . R. Symons and M. J. Wootten, Trans. Faraday Soc., 1968, 64, 1193. D. D. Macdonald, M. E. Estep, M. D. Smith and J. B. Hyne, J. Solution Chem., 1974, 3, 713. D. F. Grant-Taylor and D. D. Macdonald, Can. J. Chem., 1976, 54, 2813. vol. 7. lo A. Weissberger, Physical Method of Organic Chemistry (Interscience, New York, 2nd edn, 1955), l1 S. V. Subrahmanyam and N. Manohara Murthy, J. Solution Chem., 1975, 4, 347. l2 S. V. Subrahmanyam, V. Hyderkhan and C . V. Raghavan, J. Acoust. SOC. Am., 1969,46, 272. l3 N. Manohara Murthy and S. V. Subrahmanyam, Can. J. Chem., 1978, 56, 2412. l4 G. Wada and S. Umeda, Bull. Chem. Soc. Jpn, 1962,35, 646. l5 J. D. Worley and I. M. Klotz, J. Chem. Phys., 1966, 45, 2868. l6 C. Moreau and G. Douheret, J. Chim. Phys., 1974, 71, 1313. l7 D. D. Macdonald, M. D. Smith and J. B. Hyne, Can. J. Chem., 1971, 49, 2817. (PAPER 1 / 152)

ISSN:0300-9599    

DOI:10.1039/F19827800165

出版商:RSC    

年代:1982

数据来源: RSC