摘要:

J.C.S. F ~ ~ u Y I. 1980, 76, 101-106.Partial Molar Volumes of Organic Compounds in WaterPart 7.-Sodium and Hydrochloride Salts of a,o-Aminocaboxylic AcidsBY FEREIDOQN SHAHIDITDepartment of Chemistry, University of Toronto,Toronto, Ontario M5S lA1, CanadaandDepartment of Chemistry, McGill University,Montreal, Quebec H3A 2K6, CanadaReceived 13th February, 1979The partial molar volumes of hydrochloride and sodium salts of a number of a,w-aminocarboxylicacids in water at 25°C have been measured and related to their van der Waals volumes. Volumes ofproton ionization have been determined and the electrostriction of the solvent calculated. It isconcluded that 5.85 k 0.45 water molecules hydrate an amino acid, of which 3.85 rt 0.35 molecules arebound to the dipolar ends of the molecule.In a previous paper in this series we examined the influence of the charges uponthe partial moalr volumes, P, of a variety of amino acids and showed that the ap-parent electrostriction, E, could be expressed as the sum of the true electrostriction,V", and hydrophobic hydration effects, VH.The apparent electrostriction forsufficiently long chain a,w-aminocarboxylic acids (n > 6, where n is the number ofintervening methylene groups) was essentially constant at = 16.2 cm3 mol-'. Thiswas obtained by the conventional method of comparing 5" values for a,w-amino-carboxylic acids with those for their unchanged isomers, hydroxyamides. Fromexamination of the packing density values, ref. (3), Vw/ 5", for a,w-amino acids and acomparison with those of n-alkylamine hydrochlorides and the sodium salts of n-alkane carboxylic acids, it was suggested that the solvent molecules along the hydro-carbon section of the a,w-amino acids are highly ordered and hence have largerVw/ Vo values.In this paper we report on the measurement of the partial molar volumes ofhydrochlorides and sodium salts of a,w-aminocarboxylic acids up to and includingthose of 11-aminoundecanoic acid in water at 25 "C.This serves as an alternativeroute for calculating electrostriction, as well as providing a means of evaluating thevolume changes upon proton ionization, A V;, for a,w-aminocarboxylic acids.EXPERIMENTALMATERIALSThe source of the amino acids and their purification have been described previously.*Solutions of the corresponding hydrochlorides and sodium salts were prepared immediatelyprior to use by titration with stoichiometric amounts of HCl or NaOH solution in doublydistilled, deionized water.A serial dilution method was used to obtain solutions at variousconcentrations in order to minimize experimental error.t Address correspondence to the author at the University of Toronto.10102 PARTIAL MOLAR VOLUMES OF AMINO ACIDSAPPARATUS A N D MEASUREMENTSAll measurements were carried out at 25.00+O.0l0C using a Paar ditigal density meter,(model DMA 02C), as described el~ewhere.~ Between 8 and 16 measurements were madewithin the concentration range 0.2-3.0 % for each compound. The observed apparentmolar volumes, @JS, were calculated from eqn (1)where the parameters have their usual meaning^.^TREATMENT OF DATABecause hydrochlorides of a,w-aminocarboxylic acids dissociate [process (a)] andtheir sodium salts hydrolyse [srocess (b)] in aqueous solution, corrections to theq5GbS values were applied (vide infra)+C1-H3N-(CHJn--COOH(A.MCl) s H3N-(CNJnCOO-(A) + HC1 (4H20 + H2N--(CH2),-C0O-Na(NaA) e H3N-(CH2),-C00-(A) +NaOH. (b)The procedure followed to correct 4v for process (a) is similar to the one used forhydrochlorides of a,cll-diamines. For process (b) corrected 4v values are obtainedfromwhere a is the degree of hydrolysis of the a,a-aminocarboxylic acid sodium salt, usingliterature pK v a l ~ e s . ~Total molar concentrations, Coy were deduced from weight fractions and densitydata. The molar concentration of each species in solution, C = Co(l -a), wasobtained from these data.The corrections required for hydrochloride salts were allsmall and always < 1.0 cm3 mol-l ; however, the corrections required for hydrolysisof sodium salts were significant, (up to 20 cm3 mol-l).Partial molar volumes, V O y were then obtained from a least squares fit of theresults to eqn (2), the Redlich-Meyer equationwhere Sv, the Debye-Hiickel limiting slope is 1.868 cm3 dm3i2 m ~ l - ~ ' ~ for mono-valent electrolytes in water at 25 "C? Plots of & against JC did not show anyrapid increase in @v over the concentration range studied, indicating that no m i d -lization occurred in either series.+ f$v(NaA) = (GbS(NaA) - a[&(A) + VO(Na0H) - VO(H20)])/(1 - a)&-&JC= P+b& (2)RESULTS AND DISCUSSIONExperimental partial molar volumes for hydrochlorides and sodium salts of a,w-aminocarboxylic acids, together with bv values of eqn (2) are reported in table 1.Also shown in this table are calculated values, based on eqn (3) developed by Tera-sawa et aL9where Vw is the van der Waals volumelo and a and b are empirical constants depend-ing on the type of compound.As can be seen from the data presented, the bv valuesfor the hydrochlorides axe generally small, as found for other series of mine hydro-ch1orides.l. 8 s 11* l2 The corresponding values for sodium salts are all negative andV'= aVw+b (3F. SHAHIDI 103significantly larger. These values vary approximately linearly with chain length foreach series.ELECTROSTRICTIONFrom the calculated values based on eqn (3) for a,o-aminocarboxylic acids atdifferent pH conditions, using a method described elsewhere,l average values of14.5 f 1.7 and 4.5 $0.6 cm3 mol-1 have been obtained for the apparent electrostric-tion, E, for an NH3 or a Coy group, respectively.These compare with a value ofw 16.2 cm3 mol-l obtained for the total electrostriction of dipolar end groups for+TABLE ~.-PARTTAL MOLAR VOLUMES, v", AND bv VALUES FOR ~,CO-A~BCARE+OXYLIC ACIDHYDROCHLORIDES C1-H3N--(CH&--COOH AND ~,CO-AMI"CARBOXYLIC ACID SODIUMSALTS H2N--(CH2)-COO-Na+ IN WATER AT 25°C+~~~VO/~rn~rn01-~ n expt. ca1c.a bv/cm3 dm3 mo1-2a,w-a.minocarboyxlic acid hydrochlorides1234567891067.22+ 0.05b83.02$0.0599.76$ 0.06116.16+0.04133.23 $ 0.12C149.38 & 0.06165.46$ 0.23181.35k0.12197.69 & 0.5021 3.65 & 0.4767.2283.5699.89116.23132.57148.92165.80181.60197.93214.27a,o-aminocarboxylic acid sodium salts1234567891042.81 $ 0.1 5d58.78 f0.1674.29k0.1690.32+0.21106.44+_0.40e121.93 -F.0.18137.97+ 0.48154.07$ 0.38170.02$-0.28185.95k0.4042.6958.6174.5290.43106.35122.26138.16154.10169.98185.90+0.72+0.17 + 0.41 & 0.14+ 0.19+ 0.19 + 0.50+ 1.93- 0.61 k 0.47- O.89$0.70- l.18tl.01-2.24& 1.53- 4.94+ 3.09- 5.09+6.OO-0.1 1 k0.37- 1.2050.52- 2.32+ 0.42- 5.58 + 0.70- 5.63+ 1.57- 6.61 & 0.88-9.73k4.80- 16.89+ 4.90- 17.01 4 5.30- 17.04k4.21~ ~~a Calculated using eqn (3) with a = 1.596, b = -23.27 cm3 mol-l for hydrochlorides and a =1.554, b = -10.68 cm3 mol-1 for sodium salts.The chloride or sodium ion has not been includedin these calculations of a and b values. b Ref. (25) reports 67.0 cm3 mol-I and ref. (26) reports 68.3cm3 mol-'. C Ref. (25) reports 135.4 cm3 mol-I. d Ref. (25) reports 43.4 cm3 mol-l and ref. (26)reports 44.4 cm3 mol-'. e Ref. (25) reports 106.0 cm3 mo1-l.a,co-aminocarboxylic acids by comparing their vaiues with those of their cor-responding isomers, hydroxyamides.2 Using previously established values for theinfluence of the charges upon hydrophobic hydration of the aliphatic portion of thesemolecules,2 values of l l .O + 1.7 and 2.0k0.6 cm3 mol-l were obtained for the tru104 PARTIAL MOLAR VOLUMES OF AMINO ACIDS+electrostriction, V", for NE.' and COT groups, respectively. These results were usedto calculate the number of water molecules, N", hydrating an amino acid, using therelationship E Y 3.0 N e developed by Millero et aZ.l4# l5 These workers, using ap-propriate data for naturally occurring amino acids, obtained a value of 4.4 k 0.3 for N".However, naturally occurring a-amino acids suffer from incomplete hydration due tothe overlap of the hydration speheres about their charged end groups and to thepresence of alkyl side chains.2* 16-1 The data for longer chain a,o-aminocarboxylicacids and their corresponding sodium and hydrochloride salts indicate that an averageof 5.8550.45 water molecules hydrate an amino acid, of which 3.85k0.35 moleculesare bound to the dipolar end groups of these molecules.It has been suggested that water molecules align themselves along the hydro-carbon portion of a,w-a.minocarboxylic acids.20 Plots of Vw/P against Vw3 fora,@-aminocarboxylic acids, sodium n-alkane carboxylates and n-alkylamine hydro-chlorides showed that a,m-amino acids had considerably larger Vw/ v" values, whichTABLE 2.-THERMODYNAMIC PARAMETERS OF PROTON IONIZATION FOR THE SERIES + +OF H3N-(CH2)n-COOH AND H3N-(CH2)n-C00- IN WATER AT 25°C an AP/cm3 mol-i AS'lcal mol-1 deg-l pK+HBN--(CH~)~-COOH1 - 5.92+ 0.34'2 - 6.72+ 0.44'3 - 8.46k0.354 - 10.86+0.335 - 10.8450.51e6 -11.5850.257 - 11.56k0.528 - 12.2550.519 - 12.95k0.7910 - 13.25k0.86H3N-(CH2),-COO-f1 0.52k0.44-f2 1.4950.543 2.00k0.454 3.23+0.535 2.95 0.796 3.1450.377 3.08 20.278 3.98k0.779 3.93k0.5710 4.563-0.79-7.1- 13.4- 17.7-20.1-23.1- 24.1 - 23.6-23.5 - 23.7- 24.6- 10.76- 9.75 - 7.86- 2.87- 3.52 - 1.93- 1.9- 0.5 - 0.43-22.343.554.044.234.374.454.524.544.584.639.8410.3610.6210.8510.9410.9710.971 1.0211.0211.070 Values of pK and ASo aretaken from ref.( 5 ) and quantities refer to processes HA+ + A+H+or A + A- + H+. b Values of for a, o-aminocarboxylic acids are taken from ref. (2) and P(Na+) =-6.61 cm3 mo1-1 [ref. (711, P(HC1) = 17.80&0.09 cm3 mol-l [ref.(l)] and p(H+) = -5.4 cm3mol-l [ref. (7), (2711. C Ref. (3) reports - 6.8 cm3 mol-l, ref. (7), (27) report - 5.9 cm3 mol-l andref. (28) reports -6.4 cm cm3 mol-l. Ref. (28) reports - 13.2 cm3 mol-l. fRef. (28) reports - 1.1 cm3 mol-'. g Ref. (28) reports - 1.6 cm3 mol-I.Ref. (28) reports -7.5 cm3 mol-lF. SHAHIDI 105was taken in support of alignment of water molecules along hydrocarbon middlesections of the a,u-amino acids.2 An examination of such plots for a,o-amino-carboxylic acids and their corresponding sodium and hydrochloride salts indicatesthat while these salts have limiting values of 0.69 & 0.01 for Vw/ P, amino acids them-selves have Vw/ Vo values levelling off at 0.73. This trend lends support to our pre-vious findings.VOLUMES OF PROTON IONIZATIONThe AV" values for the first and second steps in proton ionization of a,amt-n.ino-carboxylic acids calculated from the V o values of species involved are assembled intable 2 using a conventional method.'.21 Corresponding pK values along with ASodata for each step are also given in the table. From the data presented, values ofA Vo and AS" for the first step in the ionization of a,o-aminocarboxylic acids are allnegative and varying in the same direction, levelling off at a chain length of n > 5.The A P and AS" values for the second step, although they increase in the samedirection, have opposite signs and tend to level off at n 3 4. Furthermore, AS' andA v o limiting values for both steps of proton ionization parallel those of the corre-ponding n-alkylamine hydrochlorides and sodium n-alkane carboxylates. Thisobservation, in turn, may suggest that the electrostatic interation between a chargedamino or carboxyl group and the solvent is independent of the other polar substitu-ents present in the molecule, provided that they do not overlap with the hydrationshell around each charged group.This idea is further supported by a comparison of theexperimental partial molar volumes, v", and the calculated V* values using eqn (4)developed by Cabani et aL21V*p-(CH2),-YJ = T[X-(CH2),-H] + P[H-(CHJ,-Y] -P[H-(CH,),-H] (4)where X and Y denote the hydrophilic centres as usual.1*21 Using available literaturedata for n-alkane carboxylic acids22 n-alkylamines2 and their corresponding~ a l t ~ , ~ ~ - ~ ~ ~ ~ ~ Po values for ethane and propane24 and an increment of 15.9cm3 mol-l for the methylene group contribution for higher homologues, limitingvalues of 2 cm3 mol-l are obtained for V* - 7" in the series of apaminocarboxylicacids and their corresponding series of sodium or hydrochloride salts. This constantlimiting value implies that the slight residual interaction between hydrophilic centresmay not be electrostatic.T thank Prof.J. T. Edward and P. G. Farrell for laboratory accommodation atMcGill University, where this work was started. Also the suggestions of Dr. G. E.Langford and Prof. J. T. Edward are appreciated.Part 6, F. Shahidi and P. G. Farrell, J. SoZution Chern., 1978,7, 549.F. Shahidi and P. G.Farrell, J.C.S. Paraday I, 1978,74, 858.E. J. King, J. Phys. Chem., 1969, 73, 1220.J. T. Edward, P. G. Farrell and F. Shahidi, J.C.S. Furaday I, 1977, 73, 705.J. T. Edward, P. G. Farrell, J. L. Job and B-L Poh, Canad. J. Chem., 1978,56,1122 ; J. J. Christ-ensen, J. L. Oscarson and R. M. Izatt, J. Amer. Chem. SOC., 1968,90,5949.0. Redlich and D. M. Meyer, Chern. Rev., 1964,64,221.J. E. Desnoyers and M. Arel, Canad. J. Chem., 1967,45,359.S . Terasawa, H. Itsuki and S. Arakawa, J. Phys. Chem., 1975,79,2345.' F. J. Millero, Chem. Rev., 1971, 71, 147.l o J. T. Edward, J, Chem. Educ., 1970, 47,261106 PARTIAL MOLAR VOLUMES OF AMINO ACIDSl1 M. Sakurai, T. Komatsu and T. Nakagawa, Bull. Chem. SOC. Japan, 1975,48,3491 ; M. Sakurai,l2 P. A. Leduc, J. L. Fortier and J. E. Desnoyers, J. Phys. Chem., 1974,78,1217.l 3 F. Shahidi and P. G. Farrell, J.C.S. Faraday I, 1978,74,1268.l4 F. J. Millero, G. K. Ward, F. K. Kepple and E. V. Hoff, J. Phyr. Chem., 1974,78,1636.l6 R. Zana, J. Phys. Chem., 1977,81,1817.l7 J. C. Ahluwalia, C. Ostiguy, G. Perron and E. J. Desnoyers, Canad. J. Chem., 1977,55,3364.l9 L. G. Longsworth, in Electrochemistry in Biology and Medicine, ed. T. Shedlovsky (Wileym2o J. T. Edward, P. G. Farrell and J. L. Job, J. Amer. Chem. Soc., 1974,96,902.21 S. Cabani, V. Mollica, L. Lepori and S, T. Lobo, J. Phys. Chem., 1977, 81,987.22 H. Hailand, Acta Chem. Scad. A, 1974,28,699.23 S . Cabani, G. Conti and L. Lepori, J. Phys. Chem., 1974,78, 1030.24 W. L. Masterton, J. Chem. Phys., 1954,22,1830.25 J. Daniel and E. J. Cohn, J. Amer. Chem. SOC., 1936,58,415.26 E. J. Cohn, T. L. McMeekin, J. T. Edsall and M. H. Blanchard, J. Amer. Chem. Soc., 1934,27 R. Zana and E. Yeager, J. Phys. Chem., 1966,70,954.28 W. Kauzmann, A. Bodanszky and J. Rasper, J. Amer. Chem. SOC., 1962,84,1777.1973,46,1596.F. J. Millero, A. LoSurdo and C. Shin, J. Phys. Chem., 1978,82,784.J. T. Edsall and J. Wyman, Jr., J. Amer. Chem. Soc., 1935,57,1964.Interscience, N.Y., 1955), chap. 12.56,784.(PAPER 9/232

ISSN:0300-9599    

DOI:10.1039/F19807600101

出版商:RSC    

年代:1980

数据来源: RSC

摘要:

J.C.S. Faraday I, 1980, 76, 107-111Moderated CopolymerizationPart 4.-Penultimate Unit Effect in Chain-transfer :System Styrene/Methyl Methacrylate/l-Butane ThiolBY CLEMENT H. BAMFORD* AND SULAIMAN N. BASAHELDepartment of Inorganic, Physical and Industrial Chemistry,University of Liverpool, Liverpool L69 3BXReceived 22nd February, 1979The technique of moderated copolymerization has been applied in determinations of the transferconstant of styryl radicals towards 1-butane thiol in the system styrene/methyl methacrylate/l-butanethiol at 60°C. It has been found that the transfer constant does not vary significantly (outside thelimits of experimental error) with the ratio [styrene]/[methyl methacrylate] so that there is no evidencefor the existence of a penultimate unit effect as found in the system styrene/methyl methacrylate/carbon tetrabromide.The mean value 22.4 lies in the range of literature values.These results are consistent with the view that the penultimate unit effect in the system with CBr4arises from local steric interference between the a-methyl of the radical - MMA-St- and the in-coming bromine atom.Studies of the chain-transfer reaction between styryl radicals and carbon tetra-bromide, made with the aid of the moderated copolymerization technique,' havebeen rep0rted.l~ With methyl methacrylate as moderating comonomer theobserved transfer constant was shown to depend strongly on the ratio styrene :methyl methacrylate, increasing as the ratio increases. We concluded that thetransfer reaction is sensitive to the nature of the penultimate unit and we obtainedvalues of the transfer constants for styryl radicals with penultimate units of methylmethacrylate (MMA) and styrene (St) of 57 and 368, respectively.Retaining thenomenclature of the earlier publications we shall refer, in general, to the two mono-mers as A and B, A being the moderating comonomer, so that the transfer constantsof radicals (I) and (11) will be designated CB, and CBB; CB,obs will refer to theobserved transfer constant of styryl (or w Be-type) radicals containing contributionsfrom both (I) and (II).MMA-St* (I) CBA = 60WAV St-st* (11) CBB 337In evaluating CBA, CBB we assumed that no other units in the chains influencethe transfer reaction and that penultimate unit effects in propagation axe negligible.There are no reasons for believing that the latter are operative in the styrene/methylmethacrylate system and indeed the recent accurate data of BontA, Gallo and Russoargue strongly against their existence.Further, the findings of Part 2 and the presentpaper cannot be rationalized in terms of penultimate unit effects on propagation.Recently, the data of Part 2 have been analysed by the more accurate proceduredescribed in Part 3 which takes account of all concentration changes during polymer-ization and the resulting values of CBA and CBB are given in eqn (I).The difference between these transfer constants appeared to be much too large10108 MODERATED COPOLYMERIZATIONto be attributable to likely electronic effects, so we sought other explanations.Theend-portions of chains formed by transfer between radical (I) and carbon tetra-bromide have the structure (111). Models show that in (111) there is significantsteric interference between the a-methyl of the MMA unit and the terminal bromineatom. In (IV), the corresponding structure arising from (II), the a-methyl is replacedby a hydrogen atom and there is much less congestion, in spite of the presence of theadditional phenyl group. We proposed that these considerations also apply t othe transition states of the transfer reactions, the difference between the transferconstants in eqn (1) being attributable to local steric hindrance.CH3 BrI II I(MN CH2-C-CH2-C-HH BrI II I ww CH2-C-CH2-C-HThis hypothesis may be tested in two ways.First, steric hindrance of the typedescribed would not be encountered when the labile atom of the transfer agent issmall, so that with such a transfer agent no penultimate unit effect in the methylmethacrylatelstyrene system would be expected. Secondly, a moderating comonomermay be chosen which does not produce unfavourable contacts with carbon tetra-bromide; in this case also there should be no penultimate unit effect. These twoapproaches are explored in this and the following paper which describe studies ofthe systems (i) styrenelmethyl methacrylatell-butane thiol and (ii) styrenelmethylacrylate/carbon tetrabromide and styrene/methyl acrylate/l-butane thiol.EXPERIMENTAL1-Butanethiol (B.D.H.) was fractionated at atmospheric pressure with a 1-foot column ; the boilingpoint of the final product was 98.5"C (lit.98.4"C). Azo-bis-isobutyronitrile, used as initiator,was twice recrystallized from methanol before use.copolymerization of the degassed monomers was carried outat 60°C in the presence of the transfer agent with azo-bis-isobutyronitrile as initiator.Copolymers were precipitated into petroleum ether (6O-8O0C), dissolved in benzene, re-precipitated and dried to constant weight. Number average molecular weights weremeasured with the aid of a Hewlett-Packaxd membrane osmometer.Styrene and methyl methacrylate were purified by conventional methods.As in the earlier work,l'RESULTS AND DISCUSSIONFour different initial values of [St]/[MMA] were chosen and for each a set ofexperiments was performed with a range of 1-butane thiol concentrations. ResultC.H. BAMFORD AND S. N. BASAHEL 109TABLE 1 .-SYSTEM STYRENE(B)/METHYL METHACRLATE(A)/~ -BUTANE THIOL AT 60°C103[S10 102A(A+ B) 10-3Fn 103u Cl3,obs[A10 = 8.85 mol dm-3, [B], = 0.091 mol dmW3, uo = 10.28~03.626.538.169.9513.0514.6916.8220.4130.8129.9628.9228.2725.3925.7525.3925.4325.267.1432.0411.2661.0420.86960.71430.63290.55250.44649.9879.91 19.9219.9299.9659.9609.9659.9649.966mean :[A10 = 8.48 rnol dm-3, [Bl0 = 0.4358 mol dm-3, uo = 51.39~02.985.967.458.9411.9213.4114.9017.8820.3819.8919.1819.3618.4617.5617.5417.8616.257.1431.3700.78740.70420.56820.44250.40000.34970.300350.3050.3350.3750.3650.4150.4550.4550.4450.52mean :[A~o = 8.064 mol dm--3, [El], = 0.8308 mol ~€m-~, uo = 103.0~02.45.27.08.211.112.414.016.720.3418.2017.6617.6117.4917.4117.3016.9316.785.0011.1700.62130.51420.44-430.34250.30190.27320.2346101.2101.4101.4101.5101.5101.5101.5101.5101.5mean :[A10 = 7.34 mol dm-3, [Bl0 = 1.51 mol dm-3, uo = 205.7~02.44.96.17.29.810.912.414.815.7614.5214.1213.9113.1212.7712.6512.6212.394.0000.83330.50000.41680.34480.27030.25000.21740.1888203.6203.8203.8203.9204.0204.0204.0204.1204.1-21.524.024.825.721.322.723.726.823.8--26.024.521 .o22.821.921.523 .O22.022.8--24.423.320.822.521.822.321.220.522.1--23.620.320.921.820.119.421.220.0mean : 20.9~~~~~[azo-bis-isobutyronitrile] = 8.29 X lov4 mol dmM3. All concentrations are expressed inmol dm-3 at 60°C110 MODERATED COPOLYMERIZATIONare presented in table 1, in which A(A+B) repr ients the total conversion of themonomers and S is the transfer agent.Transfer constants of the polystyryl radicalsCB,obs were computed from eqn (2) and (3) below. The derivation of these equationsis given in an earlier paper;4 eqn (3) is a rearranged form of eqn (22) of ref. (4).Eqn (2) and (3) take account of changes in concentration of all the reactants duringpolymerization.whereandrA- 1 A(A+B) 1 In ---= - ~ACA { [S]* (Pn ;:))In eqn (2) and (3) rA, r, are reactivity ratios and CA is the transfer constant of poly-(methyl methacrylate) radicals towards l-BuSH ; u is the ratio [B]/[A], initiallyhaving the value uo, and Fa, P: have their customary significance.Since, for givenconversion, all the quantities in eqn (2) except u are known, the latter may be calcu-lated; insertion of the value of u into eqn (3) then permits estimation of CB,obs.We have taken r A = 0.47, rB = 0.50 and CA = 0.66.5a Newton’s method hasbeen used in computing u from eqn (2).The small trend towards decreasing CBSobs with increasing uo noticeable in table 1is hardly outside the range of experimental error.(Small errors may be introduced,for example, by loss of styrene-rich copolymers arising from incomplete precipitationof low polymers in petroleum ether.) The mean value, CB,obs = 22.4, lies withinthe range reported in the literature 5 b for transfer constants of 1-butane thiol derivedfrom observations on the homopolymerization of styrene, viz. 21-25.The ratio of radical concentrations [(II)]/[(I)] in these experiments covers the range(5 x 10-3)-0.1, approximately. Clearly the present results show that (11) does nothave a much higher reactivity than (I) towards 1-butane thiol, so that there is nomarked penultimate unit effect of the type found with carbon tetrabromide. Ifdifferences in the electronic properties of the radicals (I) and (11), arising from thedifferent penultimate groups, were responsible for the penultimate unit effectobserved in transfer to carbon tetrabromide shown in eqn (l), they should also beoperative in the same sense with 1-butane thiol, since the “ polarities ” of the transitionstates axe qualitatively (though not quantitatively) similar for the two transfer agents.6The present results therefore favour our views about the stereochemical origin ofthe influence of the penultimate unit in the St/MMA/CBr, system.C.H. Bamford, J.C.S. Faraday I, 1976,72,2805.C. H. Bamford and S. N. Basahel, J.C.S. Faraday I, 1978,74, 1020.G. Bonth, B. M. Gallo and S. RUSSO, J.C.S. Faraday I, 1973, 69, 328.C. H. Bamford and S. N. Basahel, Polymer, 1978,19,943C. H. BAMFORD AND S. N. BASAHEL 111G. C. Eastmond, in Comprehensive Chemical Kinetics, ed. C. H. Barnford and C. F. H. Tipper(Elsevier, Amsterdam, 1976), vol. 14A, (a) p. 208 ; (b) p. 172.C. H. Barnford, A. D. Jenkins and R. Johnston, Trans. Faraday SOC., 1958, 55, 418; C H.Bamford and A. D. Jenkins, J. Polymer Sci., 1961,53,149 ; Trans. Furaday SOC., 1963,59,530 ;N. C. Billingham and A. D. Jenkins, in Comprehensive Chemical Kinetics, ed. C . H. Bamfordand C. F. H. Tipper (Elsevier, Amsterdam, 1976), vol. 14A, chap. 5.(PAPER 9/284

ISSN:0300-9599    

DOI:10.1039/F19807600107

出版商:RSC    

年代:1980

数据来源: RSC

摘要:

J. C.S. FUYU~~Y I, 1980,76,112-117Moderated CopolymerizationPart 5.-Penultimate Unit Effect in Chain-transfer. Systems StyrenelMethylAcrylate/Carbon Tetrabromide and StyrenelMethyl Acrylatell -Butane Thioland Some General ConsiderationsBY CLEMENT H. BAMFORD" AND SULAIMAN N. BASAHELDepartment of Inorganic, Physical and Industrial Chemistry,University of Liverpool, Liverpool L69 3BXReceived 22nd February, 1979Investigation of the transfer constants of styryl radicals in copolymerizations of styrene (St) hasbeen continued by the technique of moderated copolymerization with methyl acrylate (MA) ascomonomer. Using a low value of [styrene]/[methyl acrylate] (8.75 x we fhd that the transferconstant towards carbon tetrabromide is 302 at 60°C; this value, which refers essentially toWMA-St.radicals, approximates to that found previously for WSt-St- radicals (337). Theresults therefore support the view that the low transfer constant found for WMMA-St radicals(MMA = methyl methacrylate) is attributable to steric interference between the a-methyl of thepenultimate MMA unit and the incoming bromine atom. With 1-butane thiol as transfer agent,the transfer constant found for -MA-St- (20.0) is close to that of WMMA-St. (22.4), a resultwhich also strongly supports the hypothesis of local steric interference mentioned above.The general case of copolymerization of monomers A and I3 in which both MA. and -B-radicals are subject to penultimate unit effects is considered and general equations are developed.Finally, the possibility of penultimate unit effects in chain-transfer in homopolymerizations ismentioned.In this paper we report studies of the transfer constant of styryl radicals towardscarbon tetrabromide and 1-butane thiol made with the aid of the moderated co-polymerization technique with methyl acrylate (MA) as moderating comonomer.The general background of this work has been described in Part 4;2 the presentpaper concludes the examination of our hypothesis, advanced in Part 2,3 that thedifference between the reactivities of the styryl radicals (I) and (11) (St = styrene) inchain-transfer to carbon tetrabromide arises predominantly from steric interferencebetween the a-methyl group of the penultimate methyl methacrylate (MMA) unitin (I) and the incoming bromine atom.wMM A-S t (I)-St-St* (11)Preliminary results on the styrenelmethyl acrylate/carbon tetrabromide systemhave already been reported briefly in Part 2.3EXPER1,MENTALMethyl acrylate, styrene, carbon tetrabromide and 1-butane thiol were purified asdescribed in the preceding paper.2Copolymerizations were carried out at 60°C with azo-bis-isobutyronitrile as initiatorand number-average molecular weights were determined with the aid of a Hewlett-Packardmembrane osmometer.Like other workers, we experienced difficulty in obtaining satisfactory precipitation of11C.H. BAMFORD A N D S . N. BASAHEL 113copolymers rich in methyl acrylate. Finally, we adopted the procedure of precipitatinginto petroleum ether (40-60°C) cooled to - 50°C, allowing to stand overnight then recoolingto -50°C before filtrati~n.~.Polymers were dried by prolonged evacuation at roomtemperature. Attempts were also made to remove unreacted monomers by evaporationin vacuum at room temperature, but this appeared to lead to some cross-linking, no doubton account of the presence of residual initiator.RESULTS AND DISCUSSIONSYSTEM STYRENE(B)/METHYL ACRYLATB(A)/CARBON TETRABROMIDE(S)In these experiments the value of u,( = [B],/[A],) was as low as convenientlypossible. Further, carbon tetrabromide concentrations were also kept low, tominimize problems (mentioned above) associated with isolation of the copolymersby precipitation. Values of CB,obs (the observed overall transfer constants for-Be radicals) were computed from the measured E a s described in Part 4,2 takingthe reactivity ratios rA = 0.18, rB = 0.75 and the transfer constant for methylacrylate towards carbon tetrabromide CA = 0.41.Results are presented in table I ,in which the total monomer conversions are denoted by A(A + B).TABLE 1 .-SYSTEM STYRENE~METHYL ACRYLATE~CARBON TETRABROMIDE AT 60'C104[S]o 102A(A+ B) ~ O - ~ K 10314 CBpbs0 39.56 62.47 7.428 -0.95 34.97 10.36 7.620 3151.42 33.61 7.434 7.663 3061.89 33.48 5.707 7.668 3102.84 33.24 4.022 7.675 2983.32 33.48 3.663 7.668 279mean : 302[styrene], = 0.091 mol dnr3, [methyl acrylate], = 10.35 mol dm-3, uo = 8 . 7 9 2 ~[azo-bis-isobutyronitrile] = 8.29 x mol dnr3. All concentrations are expressed inrnol dnr3 at 60°C.The mean value of CBYobs in table 1 is comparable with that for radical (11) (CBB),viz.337.2 As a consequence of the low value of uo in these experiments > 99 %of the radicals with terminal styryl units have structure (111)so that we may conclude that CBA = 302, approximately. Comparison with thevalue CBA = 60 (for I)3 (A = methyl methacrylate) reveals the large influence exertedby the a-methyl substituent in the penultimate unit on the transfer reaction. Theelectronic effect arising from this methyl must be small and in any case would tendto increase the transfer constant.' The present results therefore strongly favour theview that steric congestion involving the a-methyl group in (I) is responsible for therelatively low reactivity of this radical towards CBr4.gave CBA = 169.Webelieve that incomplete precipitation of the lower copolymers was responsible forthe smaller values of CBA.wMA--St* (111)Preliminary results for this system reported in Part 114 MODERATED COPOLYMERIZATIONAs mentioned in Part 2,3 transfer constants calculated by the moderated co-polymerization procedure are sensitive to the magnitude of rB in the styenelmethylacrylate system, for which a range of values extending from 0.60 to 1.1 exists in theliterature.8a Recalculation of the present results with V, = 0.60, rA = 0.18 yields amean value of CBA = 378. This figure is also reasonably close to the value CBB = 337for styryl radicals with penultimate styrene units.2SYSTEM STYRENE(B) /METHYL ACRYLATE(A) / I -BUTANE THIOL( S)In computing CB,-,bs the following values of the kinetic parameters were used:The mean value CB,obs = 20.0, which effectively refers to radical (In) (i.e., CBA),is close to the transfer constant previously determined for radical (I) (22.4).Thus,for chain-transfer to 1-butane thiol, the presence of an a-methyl group in thepenultimate unit has little effect. These conclusions are not affected by takingr A = 0.18, rB = 0.60;8a with these values CB,obs (mean) = 25.0.r A = 0.18, r~ = 0.75,6 CA = 1.69.8bTABLE 2.-sYSTEM STYRENEIMETHYL ACRYLATE/l-BUTANE THIOL AT 60°C104[Slo 102A(A+B) 10-3E 103u c30 39.56 62.47 7.428 I1.85 19.54 18.59 8.123 18.43.79 24.78 10.25 7.950 22.55.68 24.58 7.463 7.956 19.919.1 7.50 24.41 5.882 7.962mean : 20.0-[styrenelo = 0.091 mol dm-3, [methyl acrylatel0 = 10.35 mol dm-3, uo = 8.792~[azo-bis-isobutyronitrile] = 8.29 x mol dm-3.All concentrations are expressed inmol dm-3 at 60°C.CONCLUSIONSTable 3 presents transfer constants for styryl radicals with different penultimateunits. These data demonstrate that, in the systems studied, (a) the electronicproperties of the penultimate unit in a styryl radical do not exert a major influenceon the transfer constant and (b) the effect of a penultimate unit becomes very smallwhen either the unit or the transfer agent is such as to eliminate steric congestion ofthe type described. The conclusions evidently support our hypothesis of the stericorigin of the penultimate unit e f f e ~ t .~It is possible that smaller penultimate unit effects, perhaps of different origin,arise in these and other systems, but the experimental results at present available arenot sufficiently precise to establish their existence. We hope, however, that theymay be observable through the development of sufficiently refined procedures basedon moderated copolymerization.COPOLYMERIZATION I N WHICH BOTH A- AND B* RADICALS SHOWPENULTIMATE-UNIT INFLUENCES I N CHAIN TRANSFEREvaluation of CB,obs from the equations already presented [eqn (4) of Part 1,leqn (22), (24) and (27) of Part 3 and eqn (3) of Part 4 2] requires a knowledge oC. H. BAMFORD AND S . N. BASAHEL 115the transfer constant of the moderating comonomer A. Hitherto we have assumeda constant value CA for this quantity, but our results suggest that in some systemsMA. radicals may show penultimate unit effects so that CAA # CAB andCA = CA,obs (which contains contributions from wAA.and wBA.) becomes afunction of [B]/[A]. In the system styrene/methyl methacrylatelcarbon tetrabromideit seems probable that radical (IV) is more reactive than (V) towards CBr,WSt-MMA.wMMA-MMA-Our estimates of CB,obs and hence of CBB have neglected such differences; in thisparticular instance reasonable changes in CA arising in this way do not lead toappreciable errors in CBB, since the terms in C, are relatively small.TABLE 3.-TRANSFER CONSTANTS AT 60°Ctransferradical ‘F. CBr4 1 -BUSHwst-st. 337 22WMMA-St. 60 2220 MMA-S t 302In general, if penultimate unit effects are significant for both MA- and M B ~radicals, eqn (3) of Part 1,l applicable at infinitesimal conversions, must be replaced byIn the styrenelmethyl methacrylate system CA,obs and CB,obs include contributionsfrom radicals of types (IV), (V) and (I), (11).Both CA,obs and CB,obs may be functionsof [B]/[A] of the form shown in eqn (2) [cf. eqn (6) of Part 111CA,obs = I___ (rACAA/uO f CAB),$-rA/uOCombination of eqn (1) and (2) gives eqn (3) :In principle, eqn (3) permits calculation of CAA, CABy CBA, CBB from copolymerizationexperiments carried out over a range of [A] and [B]11 6 MODERATED COPOLYMERIZATIONWith finite (but small) conversions, it is appropriate to use eqn (14) and (22) ofPart 3.9 We write the latter relation in the formin whicha =P =( 5 )[It is assumed that CA,obs and CB,obs remain effectively constant in the compositionrange between uo and u.This is the reason for the restriction to small conversionsstated above; otherwise, eqn (14) and (22) of Part 3 are valid for all conversions.]Combination of eqn (Z), (4) and (5) yieldsCBBarA au0 PCAB+ l+r,uo CAA + -uO + rA uO + rAIn eqn (6), a and p axe calculable from eqn (5) for given values of uo and u, so thatby carrying out a series of (modified) copolymerization experiments over a range ofvalues of uo a set of simultaneous equations in CAA, CAB, CBA, CBB may be obtainedwith the aid of this equation.Determination of these four quantities would reveal penultimate unit influenceson both MA* and -Be radicals.Direct measurement of CAA by studies of thehomopolymerization of the moderating comonomer A would normally be possibleand so facilitate application of eqn (3) or (6). Clearly, great experimental precisionis essential if significant values are to be obtained in this way.PENULTIMATE UNIT INFLUENCES I N CHAIN-TRANSFER INHOMOPOLYMERIZATIONThere is no reason why penultimate unit effects of the type described should beconfined to copolymerizations; on the contray, our data imply that they shouldalso exist in some homopolymerizations. For example, we might expect that, inthe homopolymerization of methyl methacrylate, the rate of transfer with carbontetrabromide is not determined solely by the properties (reactivity) of the terminalunit, but is reduced by the presence of the a-methyl in the penultimate residue.Evidently this possibility must be borne in mind when assessing reactivities ofpropagating chains from chain-transfer data.'C. H. Bamford, J.C.S. Faraday I, 1976,72,2805.C. H. Bamford and S. N. Basahel, J.C.S. Faraday 4, 1979,75, 107.C. H. Bamford and S. N. Basahel, J.C.S. Faraday I, 1978,74, 1020.H. Fuhrman and P. B. Mesrobian, J. Amer. Chem. SOC., 1954,76, 3281.T. Huff and E. Perry, J. Polymer Sci. A, 1963, 1553.F. M. Lewis, C. Walling, W. Cummings, E. R. Briggs and F. R. Mayo, J. Amer. Chem. SoC.,1948,70,1519C . H. BAMFORD AND S . N. BASAHEL 117' C. H. Bamford, A. D. Jenkins and R. Johnston, Trans. Faraday SOC., 1958, 55, 418; C. H.Bamford and A. D. Jenkins, J. Polymer Sci., 1961, 53, 149 ; Trans. Favaday SOC., 1963, 59,530 ; N. C. Billingham and A. D. Jenkins, in Comprehensive ChemicaE Kinetics, ed. C . H. Bam-ford and C. F. H. Tipper (Elsevier, 1976), vol 14A, chap. 5. * G. C. Eastmond, in Comprehensive Chemical Kinetics, ed. C . H. Bamford and C. F. H. Tipper(Elsevier, Amsterdam, 1976), vol. 14A, (a) p. 379; (b) p. 226.C. H. Bamford and S . N. Basahel, Polymer, 1978, 19,943.(PAPER 9/285

ISSN:0300-9599    

DOI:10.1039/F19807600112

出版商:RSC    

年代:1980

数据来源: RSC

摘要:

J.C.S. Faraday I, 1980,76, 118-125Use of the Gibbs Equation to Calculate Adsorptioninto Monolayer-covered SurfacesBY DAVID M. ALEXANDER* AND GEOFFREY T. BARNESDepartment of Chemistry, University of Queensland, Brisbane, AustraliaReceived 5th December, 1977A formal equation has been derived to describe the change in equilibrium surface pressure whenthe concentration of a soluble surfactant is varied in a subphase covered by a constant amount ofinsoluble monolayer on a fixed area of surface. Approximations are suggested which enable theequation to be applied to experimental data for the calculation of the surface excess concentration ofsurfact ant.An example is given and the surface penetration results discussed.The adsorption of a solute at an interface can, in principle, be calculated fromsurface tension data by means of the Gibbs equation.However, in penetrationexperiments, where adsorption is occurring at a surface that carries an insolublemonolayer, there axe difficulties with such a calculation because the activity of themonolayer component in a bulk phase can be neither controlled nor measured.has attempted a solution to this problem, but his treatment containsassumptions which will be shown to limit the validity of the final equation to restrictedregions of high surfactant concentration and low monolayer coverage. In the presentpaper we have derived a more general expression which illuminates the assumptionsof the Pethica equation, defining the limits of its applicability, and which, usingcertain assumptions, may be applied to certain systems to give results for conditionswhere Pethica's equation is not applicable.PethicaTHEORETICALFor simplicity we will consider the case where both monolayer? and surfactantare uncharged.We will adopt the Gibbs convention €or defining the position of thesurface. For this situation the Gibbs equation can be written :lwhere rM is the adsorption of the insoluble monolayer substance and rgl) the adsorp-tion of the surfactant relative to water. AM is the absolute activity of the insolublemonolayer in the surface phase and As the absolute activity of the surfactant in boththe surface phase and the bulk phase at equilibrium.By definition, the axea per molecule of M isdn[ = RTTL d In AM + R T r p d In A:, (1)(2)(3)h AM = A/nh = 1 / r M .Furthermore, if the surfactant concentration is low we can write :d In As = d ln ms,? The term " monolayer " is used here to refer to the spread monolayer substance, M.11D.M. ALEXANDER AND G. T. BARNES I19where m, is the molality of surfactant. Inserting eqn (2) and (3) into eqn (1) gives(4)RTdIT = - d In AG+RTal) d In mS.A MIn an equilibrium penetration experiment, surface pressure isotherms are deter-mined on surfactant solutions of various concentrations. A plot of surface pre2sureagainst the logarithm of the surfactant concentration at a constant value of AM isthen constructed and the slope determined. Eqn (4) gives an expression for the slope :(E) = RT -(-) a h a , +RTr-I)a In mS A,ng AM a In mS A,ngRT a h & - - - ( - allI ) (s) +RTTh1),JM 2x mS A,ngwhich on rearrangement yieldsPARTIAL MOLAR AREAIt can be shown that Pethica's eqn 6(a), 9(u) and (11) are only self-consistent andconsistent with our eqn (5) ifwhere & is the partial molar area of M.However, the constant conditions ofeqn (6) do not correspond to those required for a rigorous definition of the partialmolar mea, as we will show.For a Gibbs model for the surfacedGz = -S" d T - A dy+x py dnt. (7)iAt constant T (a constraint not explicitly shown in the following expressions),andTherefore,The latter equality is in accord with the usual definition of 3,. Thus €or the penetra-tion experimentd In 1,AM = RT(T5r)ng,ng*It follows that eqn (6) is incorrect and cannot be substituted into eqn (5).Anotherapproach based on eqn (11) must be used120 ADSORPTION INTO MONOLAYERSRIGOROUS EXPRESSION FOR THE SLOPE, (anla In ms)A,"gAt constant T and n&, the activity of M will depend on the surface pressure andon the amount of adsorbed surfactant :For eqn (5) we require the differential with respect to II at constant area and n&.Thus, with the substitution of eqn (ll)y we obtainIf the last term in the denominator is zero the expression reduces to Pethica'sequation :SATURATION ADSORPTIONThe concept of saturation adsorption has proved useful in treating adsorptionat monolayer-free surfaces and it can also be applied to monolayer-covered surfaces.A suitable definition of saturation adsorption at a monolayer-covered surface wouldbe :Application of this condition to eqn (14) not only leads to the Pethica equation[eqn (15)], but also gives a constant value for the slope, (allla In ms)As ag.If, to thegeneral relation,A = n$&+n",A,,we apply the constant conditions of eqn (16) and the reasonable assumption thatAM and As do not vary with II in opposite directions, we see that AM, & and 2,must all be constant at saturation adsorption and that consequently= constant.Linear portions often observed 1* in plots of n against In m, are thus likelyto be a necessary condition for saturation adsorption, but because of experimentalinaccuracies not a sufficient condition. Within such a region the Pethica equationis valid and can be used to calculate the adsorption.For this calculation a value of AM is required.Pethica suggested that a reason-able estimate could be obtained by equating AM to the area per molecule in a mono-layer of pure M at the surface pressure selected for measuring the slope (alI/a In ms)A, nsD . M. ALEXANDER AND G . T . BARNES 121This assumption appears acceptable if the excess surface region is dilute with respectto surfactant. Fortunately the c?lculated adsorption values do not depend stronglyon the value used for AM unless AM is small.The appearance of linearity in the plot of Ii against lnm, may be deceptive,particularly at low areas where there is often appreciable scatter in the data. Insuch doubtful cases it is better to use the more general equations developed below.GENERAL CASEFor a more general treatment it is convenient to modify eqn (14). By definitionA = n&& = n;/r$'),&ril) = ngln;.whenceCombination with eqn (17) yieldsand eqn (14) can then'be written(20)Approximations are necessary if this equation is to be used to evaluate and &.(i) APPROXIMATION FOR (a In LM/an& ng.We assume that at constant surface pressure and temperature :This postulates ideal dilute solution behaviour for the excess surface components,the monolayer acting as solvent and the surfactant as solute, and can be assumedvalid only where ng 4 n;.It is assumed that under these c nditions the surfactantmolecules are molecularly dispersed in the excess phase. (The formation of twophases in the surface region would result in the loss of a degree of freedom whichwould be experimentally observable.)This equation applies when the surfactant is unionized.In the case where thesurfactant is ionized and surface hydrolysis does not occur the expression is morecomplex but tends to a value twice that given by eqn (22) as n:ln& + 0.{ii) APPROXIMATION FOR (6nz/aII)As ng.Differentiating eqn (17) with respect to II, keeping n& and A constant, giveswhenc122At constant nkADSORPTION INTO MONOLAYERSthereforeSimilarlyThe Gibbs-Duhem equation applied to the areas isSubstituting the above three equations into eqn (23) ;Provided that n: is appreciably less than n s an approximation for the last term ineqn (25) is acceptable.We shall assume that the partial compressibilities of thetwo components are equal, i.e.,Substitution into eqn (25) givesCALCULATION OF ;IsFor n$ 4 n&, which is a condition most frequently encountered when AM is small,we can combine eqn (22), (26) and (20) and using eqn (17) obtain :AM and (azM/aII) may be evaiuated by assuming, with Pethica,l that their valuesin the penetrated monolayer are the same as in a monolayer of pure M at the samesurface pressure. Eqn (27) can then be solved for As and eqn (19) gives the adsorp-tion, rp.The assumption of the constancy of & at constant I3 implies by the Gibbs-Duhemeqn (24) that As is also constant. Any treatment postulating a varying AM woulduse the Gibbs-Duhem equation, but experimental accuracy has not been high enoughto justify any such treatment.In the general case when AM is not lage it is apparent that a value of As can becalculated only if a very precise value of d?fM in the pure monolayer has been deterD.M. ALEXANDER AND G . T. BARNES 123mined, However, if the data were accurate enough AM and As could each beevaluated as constant parameters giving the best fit to the experimental data.In the case of a highly incompressible monolayer the situation is more favourablesince, because (i3&/i31X)ng, nE is small, the quotient in the brackets of eqn (27) is smallat all but very low surface concentrations of surfactant. Thus the approximationsmade become less critical if the monolayer is highly incompressible.There is another set af conditions which allow a solution for eqn (20). This iswhere eqn (20) reduces to eqn (19, Pethica’s equation, so that As can be calculatedusing eqn (19).The conditions are a Lasge value of AM [so that AM-& is large ineqn (20)] and bulk surfactant concentrations which are not too low [so that (i3nXldII), n%is small in eqn (2011.DISCUSSIONThe data of Pethica at 298 K on the penetration of cholesterol monolayers bydodecylsulphate can be treated with some confidence over an experimentally accessibleconcentration range because the cholesterol layer is so incompressible. We assumethat surface hydrolysis of the surfactant does not occur, though this has been verifiedfor the pure aqueous surfactant system only down to a concentration of 4 x moldnn3, well above the concentrations used in Pethica’s penetration experiments.I I I ‘ 0 tQ.36 0.38 0.40 0.42 0.44area, A M l n m z mohuje-1FIG.1.-Experimental slopes for the cholesterol + dodecylsulphate system from Pethica at28 mN m-’ (0) ; slopes calculated according to eqn (28) (solid curves) for As/run2 molecule-l of0.50 (upper curve), 0.55 (middle curve) and 0.60 (lower curve); experimental slope for themonolayex-free system (broken line).A surface pressure of 28 mN m-l was chosen as being common to the greagstnumber of Pethica’s isotherms (n, In m,) each of which applies to a particular AM.The slope of each isotherm ( X I / a In ms)A,ng was determined at this surface pressur124 ADSORPTION INTO MONOLAYERSand plotted against &.From Pethica's isotherm for the pure monolayer, ai at28 mN m-1 was 0.369 x 10-1 m2 molecule-1 and= - 1.62 x m4 molecule"2.The appropriate form of eqn (27) for the case of an electrolyte solute undergoing nosurface hydrolysis and penetrating a layer with which it has no common ion is I-'. (28) (E) = ? F [ l - 2RT;Z,(a;l,/aJJ),,,,,ms A,nS (AIM - A M ) A M ( A M A M + As)A further approximation is introduced in the factor of two in the last quotient. Thisis discussed above following eqn (22). This approximation is again valid at lowsurface concentrations of surfactant.Experimental and theoretical values are compared in fig. 1 which includes thepoint ( X I / a In ms)A, ng = 0 where A, = &. Plots are included for As = 0.50,0.55 and 0.60 nm2 molecule-'.The fit is best at the lowest adsorption (high mono-layer coverage), where the approximations are most valid, and for & = 0.55 nm2molecule-l.For the pure surfactant on water at 298 K the value of A, at surface pressuresabove 40 mN m-1 is in the range 0.41-0.58 nm2 molecule-l, indicating that thearea change on mixing monolayer and surfactant in the surface is not very large.Some estimate can be made of the error involved in this case when the assumptionfollowing eqn (25) breaks down. If, in the condensed S state the surfactant is assumedto be five times as compressible as the cholesterol instead of being of equal com-pressibility, then the maximum error resulting in the slopes calculated and plotted onfig.1 is 5 %, falling to 2 % at the lowest AM values used. A compressibility differenceof more than a factor of five would be unusual for pure monolayers in the S state.ACONCLUSIONIt has been shown that it is sometimes possible in practice to interpret penetrationisotherms by a thermodynamic treatment in terms of partial molar areas in regionswhere the surface is dilute in surfactant or in regions where the monolayer is of higharea per molecule and the surfactant is of relatively high concentration. In othercases other methods such as the use of radioactive tracers must be used to calculatethe surface excess of surfactant.The data of Pethica fit our equations as well as can be expected and lead to acalculated value of A, and the surface excess of surfactant at low surface areas ofmonolayers.The assumption is made that no surface hydrolysis takes place.Pethica's measurements of surface pressures over a subphase of 0.145 mol dm-3sodium chloride do not clarify this assumption, partly because of experimental un-certainties and partly because some measurements were made in more dilute solutionsmol d.~n-~) than used in the other penetration measurements, where the situationwith respect to surface hydrolysis may be different.Fowkes has discussed the problem of equilibrium penetration using a monolayermodel. His theory is based on a fundamental equation consistent with the treatmentabove but his assumptions regarding the ideality of the mixtures of the surface mono-layer are extended to high surface concentrations of surfactantD. M. ALEXANDER A N D G . T . BARNES 125NOTATIONarea of surface= A/$', area per surface excess mole of ipartial molar area of i defined by eqn (10)partial molar area of monolayer without surfactantGibbs energymolality of i in bulk phasesurface excess amount, Gibbs adsorption of component iAvogadro constantgas constantentropytemperature= np/Cnpsurface tensionsurface excess concentration (= nP/A)relative adsorption of component i with respect to component 3absolute activity of ichemical potential of isurface pressureisubscriptsM monolayer substances surfactantsuperscripts0 excess quantity referring to the Gibbs surfaceB. A. Pethica, Trans. Faraday SOC., 1955, 51, 1402.R. Aveyard and D. A. Haydon, An Introduction to the Principles of Surface Chemistry (Cam-bridge, London, 1973), pp. 12, 13.R. Defay, I. Prigogine, A. Bellemans and D. H. Everett, Surface Tension and Adsorption(Longmans, Green and CQ, London, 1966), p. 160.F. van Voorst Vader, Trans. Faraday SOC., 1960,56,1067.M. A. McGregor and G. T. Barnes, J. Colloid Interface Sci., 1976,54,439.J. E. Bujake and E. D. Goddard, Trans. Faraday SOC., 1965, 61, 19(F. M. Fowkes, J. Phys. Chem., 1961, 65, 355.(PAPER 712131

ISSN:0300-9599    

DOI:10.1039/F19807600118

出版商:RSC    

年代:1980

数据来源: RSC

摘要:

J.C.S. Faraday I, 1980,76,126-134Ultraviolet Photoelectron Spectroscopic and ThermalDesorption Studies of the Chemisorption and Decompositionof Benzene, Cylcohexadiene, Cyclohexene andCyclohexane on W( 100)BY ASHOK K. BHATTACHARYADepartment of Physical Chemistry, University of Cambridge,Lensfield Road, Cambridge CB2 1EPReceived 24th August, 1978The chemisorption and decomposition of the cyclic hydrocarbons, viz. benzene, 1,4 (and 1,3)cyclohexadiene, cyclohexene and cyclohexane on a W(100) surface at 300 K have been investigatedusing ultraviolet photoelectron spectroscopy and thermal desorption studies. The hydrocarbonsall decompose on the surface to give an adsorbed layer of hydrogen and the same molecular species.Increase in exposure to the hydrocarbon results in a monotonic increase in the surface concentrationof the molecular fragment, while the surface concentration of the adsorbed hydrogen, in the caseof the unsaturated hydrocarbon, first increases and then decreases so that at saturation coverageonly the molecular species remains on the surface. On heating the adsorbate covered surface themolecular species decomposes and hydrogen is found to be the only observable desorption productwhile the surface is left with an adsorbed layer of carbon.The nature and bonding of the molecularfragment are also discussed.The elucidation of the nature and bonding of the species formed when a gasadsorbs on well characterized surfaces of different metals is of vital importance to abetter understanding of heterogeneous catalysis.With this in mind we have studiedthe chemisorption, decomposition and interaction with hydrogen of benzene, 1 , 4 (and1, 3) cyclohexadiene, cyclohexene, and cyclohexane on a W(100) surface usingultraviolet photoelectron spectroscopy (u.P.s.) and thermal desorption techniques.The chemisorption of benzene has been previously studied on Ni(lll),l Ir(100)2and Pt(100)3 surfaces by U.P.S. and on Ni(l.11) and faces by high-resolutionelectron energy loss spectroscopy (h.r.e.1.s.). These experiments indicated thatbenzene adsorbs on these surfaces with relatively little distortion. The plane of themolecule was thought to lie parallel to the surface and the interaction to take placevia the n electrons of the ring. Similar conclusions were also drawn by Somorjaiand Gland 5 s on the basis of their investigations of the adsorption of benzene onPt(ll1) and (100) surfaces using 1.e.e.d.and work-function measurements. U.P.S.experiments have further indicated that, like benzene, cyclohexane adsorbs onNi(ll1) with relatively little distortion.EXPERIMENTALThe apparatus used in the present investigation, the geometry employed to record thephotoelectron (p.e.) spectra with HeQ) (21.2 eV) radiation and the method of cleaning thecrystal have been described elsewhere?For thermal desorption studies the crystal was heated radiatively by a hlament at aconstant rate m 12 K s-l.12A . K . BHATTACHARYA 127The hydrocarbons (reagent grade) were all dried and purified by multiple freeze-thaw-vaporisation cycles in the gas handling line of the apparatus.The purity of the hydrocarbonvapours was checked mass spectrometrically.RESULTSThe p.e. spectra of a W(100) surface on sequential exposure to benzene at 300 Kare shown in fig. 1. The corresponding adsorbate-induced emissions are shown by thedifference curves in fig. 2. Each of the difference curves has been obtained byplotting the difference between the p.e. spectrum of an adsorbate-covered surfaceand that of the clean surface. The difference curves can be best described by consider-ing two regions: one between 0 and 4eV and the other beyond 4eV below theFermi level (EF). In the first region an exposure of 1 L (= 1.33 x Pa s) producesa peak at - 3 eV with two shoulders at - 1.3 and -2.2 eV (the peak positions inthis paper axe given with respect to EF).Increasing the exposure causes a rapidincrease in the intensity of photoemission in this region and for an exposure ofx 2 L there is a sharp and intense peak at - 1.9 eV and a shoulder at = -3.2 eV.Further exposure causes the intensity of photoemission to decrease. Saturation isobtained at an exposure of = 6 L when two peaks with maxima -2.2 and -3.2 eVare obtained. In the second region, for an exposure of 1 L peaks appear at -5.0,- 7.0, - 9.0, - 10.4 and - 12.4 eV. Increasing the exposure causes the intensity ofphotoemission in this region to increase monotonically until saturation. However,with increasing coverage the position of the second peak shifts towards higherbinding energy, while the positions of the other peaks remain unaltered.At satura-tion coverage the second peak has its maxima at -7.9 eV.hVI I I I I I I I 1 IL18 -16 -14 -12 -1'0 -8 -6 -4 -2 0E- EFleVFIG, 1.-He(1) photoelectron spectra of a W(100) surface on sequential exposure to benzene at300K: (a)(--- ) clean surface ; (b) (-) 1.0 L of benzene ; (c) (- - -) 2.0 L of benzene and(d) (-) 6.0 L of benzene128 ADSORPTION OF CYCLIC HYDROCARBONS ON TUNGSTENrn+d .*3ti;iii!i21 19 17 15 13 11 9i.p./eV Jw I 1 I 1 I I-14 -12 -10 -8. -6 -4 -2 0E- &levFIG. 2 . 4 0 ) Difference curves for varying exposures of benzene at 300 K : (i) (-) 1.0 L ; (ii) (- - -)2.0 L ; (iii) (- - -) 3.0 L and (iv) (-) 6.0 L ; (b) gas-phase photoelectron spectrum of benzene.-0.41 2 3 4 5 6 7 8exposure/LFIG.3.-Change in A 4 as a function of exposure at 300 K: (a) (-) benzene; (b) (. . .) cyclo-hexadiene ; (c) (- - 9 -) cyclohexane and (d) (- - -) cyclohexeneA . K . BHATTACHARYA 129Each of the other hydrocarbons studied in the present investigation upon adsorp-tion on a W(100) surface gives rise to similar p.e. spectra with the same number ofpeaks at the same positions as are obtained on exposing the W(100) surface to benzene.However, two observations can be made: (a) In the case of cyclohexane the p.e.spectrum saturates at an exposure of FZ 2 L when a sharp intense peak is obtained inthe first region. Further exposure to cyclohexane, unlike that of the other unsaturatedhydrocarbons, does not result in a decrease in intensity of the sharp peak.(b) Fora given low exposure (c 2 L) the intensities of photoemission in the first region forthe different adsorbates have the following order of magnitudes :(both 1, 4 and 1, 3 cyclohexadiene behave identically in all the measurements thatwe have carried out).= 4.6 ev)on sequential exposure to each of the hydrocarbons are shown in fig. 3. Theaccuracy of A 4 is k0.05 eV. The values of A+ at the maxima decrease in the order :Hydrogen was the- only observable desorption product when the W(100) surfaceexposed to any of the hydrocarbons was heated. Fig. 4 shows a series of H2 thermaldesorption spectra taken after varying exposure to benzene at 300 K. For comparisonthe H2 thermal desorgtion spectrum obtained from a W(100) surface saturated withpure hydrogen at 300 K is also shown in fig.4. The H2 thermal desorption charac-teristics of a W(100) surface after varying exposures to each of the other hydro-carbons is similar to that of a W(100) surface exposed to benzene. In table 1 we show< C6HS < C6HlO < C6H12The changes in A+ of a W(100) surface at 300 K (A4 = + - &eaa ;C6Hlo > C6Hg > C6H6.A4 00 500 600 700TIKFIG. 4.-Hz flash desorption spectra from a W(100) surface after';varying exposure to benzene at300 K. (a) 0.5, (b) 1.0, (c) 2.0, (d) 6.0 L. The H2 flash desorption_spectrum from a W(100) surfacesaturated with pure Hz is shown by the dashed curve.1-130 ADSORPTION OF CYCLIC HYDROCARBONS O N TUNGSTENthe ratios of the normalised peak(s) areas for different exposures to the unsaturatedhydrocarbons [a normalised peak(s) area is obtained by dividing the total area underthe desorption peak(s) for a given coverage by the total area under the desorptionpeaks obtained when the surface is saturated with pure hydrogen].TABLE RATIOS OF THE NORMALISED PEAK AREAS FOR VARIOUS EXPOSURESratio of the normalisedexposure/L peak areasCsH6 :CGHs :CGH100.5 1 : 1.6 : 1.91 .o 1 : 1.2 : 1.32.0 1 : 0.9 : 16.0 1 : 0.9 : 1(saturated)Fig.5 shows the room temperature pe. spectra obtained after sequentially flashingthe W(100) surface saturated with benzene at 300 K to M 450 K (which is greater thankhe decomposition temperature of the molecular fragment at saturation coverage.See fig.4.) and to = 1100 K. The difference spectrum plotted in fig. 6 shows that-16 -I4 -12 -10 -8 -6 --A -2 0E- EF/evFIG. 5.-He@) photoelectron spectra of a W(100) surface saturated with benzene after flashing todifferent temperatures : (a) (-) clean surface ; (b) (- - -) 450 K and (c) (-) 1100 K.the fragment left after desorption of hydrogen gives rise to emissions at - 1.6, -4.05,- 5 , -6 and -9.7 eV. The peak at -4.05 is very sharp and has a width at halfheight of M 0.55 eV. Heating the crystal from m 450 to = 1100 K results in adecrease in intensity of the p.e. spectrum below -4.5 eV while the peaks at -1.6and -4.05 eV become sharper. The value of A 4 increases from -0.15 to +0.10 eVA . K. BHATTACHARYA 131-!2 -10 -8 -6 -L -2 0E-EF/eVFIG.6.-Difference curve obtained for the carbon overlayer by plotting the difference fig. 5(c)-(a).The p.e. spectra obtained after exposing a W(100) surface preadsorbed withhydrogen to benzene at 300 K axe shown in fig. 7. By cornparing fig. 1 and 7 it canbe concluded that benzene displaces adsorbed hydrogen. Both cyclohexadiene andcyclohexene behave similarly to benzene. Cyclohexane, however, does not displacehydrogen adsorbed on W(100).E-EFleVFIG. 7.--He(I) photoelectron spectra of a W(100) surface at 300 K under different conditions :(a) (. . .) clean surface ; (b) (-) from (a) after saturation with H2 ; (c) (- - -) from (b) after anexposure of 2 L of benzene and (d) (-) from (c) after an exposure of 24 L of benzene4 32 ADSORPTION OF CYCLIC HYDROCARBONS ON TUNGSTENDISCUSSIONThe p.e.spectra shown in fig. 1 can be readily understood in terms of a de-composition reaction giving rise to an adsorbed layer of a molecular fragment whichgives rise to, among others, multiple photoemission peaks in the region below -4 eVand an atomic species which gives rise to intense photoemission in the region between- 1 and -4 eV. Of the two pussible atoms here, C and H, only the latter is known[ref. (8) and (9), as well as from the present investigation] to give rise to intense emissionin the region between - 1 and -4 eV. Therefore, it is concluded that all the cyclichydrocarbons studied in the present investigation decompose upon adsorption on aW(100) surface to give an adsorbed layer of the same molecular fragment and hydro-gen.This continues up to an exposure of = 2 L. On further exposure to the hydro-carbons (excepting cyclohexane) an increase in the surface concentration of the molecu-lar fragment causes a displacement of adsorbed hydrogen. Indeed the results of theU.P.S. experiments (fig. 7) also show that the unsaturated cyclic hydrocarbons alldisplace adsorbed hydrogen while there is simultaneous chemisorption of the molecu-lar fragment, The inability of cyclohexane to displace adsorbed hydrogen is alsoreflected in the failure of the intense photoemission peals at - 1.9 eV to decrease inintensity for exposures of > 2 L of cyclohexane. If we assume that the stickingcoefficients of all the unsaturated hydrocarbons under investigation here are the samethen it is expected that at low coverages the amount of chemisorbed hydrogen, for agiven exposure of the adsorbate, would decrease with increasing unsaturation in thehydrocarbons.The observed intensities of photoemission in the region between- 1 and -4 eV for exposures of < 2 L would seem to support the above-mentionedtrend.Chemisorption of hydrogen is known to increase the work-function of a W(lO0)surface. If we consider the molecular fragment to be an electron donor then thework-function behaviour of a W(100) surface on sequential exposure to each of thehydrocarbons (fig. 3) would also support the above-mentioned mechanism of chemi-sorption of the cyclic hydrocarbons on W(lO0).It is not clear, however, whythe value of A 4 for an exposure of w 1 L of cyclohexane is not more than that for asimilar exposure of cyclohexene, although pee. spectra suggest that there is morechemisorbed hydrogen in the former case than the latter.From the results of the thermal desorption studies we can arrive at the followingconclusions : (a) it is the strongly bound hydrogen that is displaced initially, followedby the displacement of the weakly bound hydrogen, and (b) the chemisorbed hydrogenformed during the initial adsorption of the hydrocarbons and the hydrogen bound tothe molecular fragment both give rise to the desorption peak around z 450 K. Sincehydrogen may be retained in the chemisorbed state on the surface after dissociation,this desorption temperature may, or may not, be the temperature at which the chemi-sorbed molecular fragment dehydrogenates.It does, however, set a maximumlimit to the temperature of the decomposition reaction. In accord with the above-mentioned mechanism of chemisorption it is seen from table 1 that at initial coveragesthe amounts of hydrogen desorbing on heating the crystal exposed to the differentunsaturated hydrocarbons are almost in the ratio of the numbers of hydrogen permolecule of the hydrocarbons. However, as expected, the ratio of the amounts ofdesorbing hydrogen with increasing coverage deviates from that required by themolecular stoichiometry and for exposures of > 2 L the amounts of desorbinghydrogen are the same (within experimental errors) for 411 the unsaturated hydro-carbon adsorbatesA .K . BHATTACHARYA 133From fig. 1 and 2 (and also from the p.e. spectra recorded for other exposuresbut not shown in these figures for clarity), it can be seen that the photoemission fromadsorbed benzene in the region between - 1 and - 4 eV consists of two maxima at-2.2 and -3.2 eV whose positions do not alter with varying coverage. The othermaxima in this region, on the other hand, shift and at first increase and then decreasein intensity. Having regard to the photoemission from hydrogen chemisorbed onW(100) surface it can be concluded that the maxima at -2.2 and -3.2 eV (whichare also the only ones present at saturation coverage in this binding energy region)are due to the molecular fragment.However, from the evidence that is availableit is not possible to say if the molecular fragment is the only species present on thesurface at saturation coverage. A small amount of chemisorbed hydrogen may bepresent which, although not distinctly identifiable by an inspection of the p.e.spectrum, could be responsible for some of the intensity in the photoemission atsaturation coverage. In any case, the amount of such chemisorbed hydrogen canonly be very small. The peak at 12.4 eV (fig. 2) is probably associated with a changein the electronic structure of tungsten following adsorption, as a peak is found toexist in this region in the p.e. spectra of W(100) surfaces having widely differentadsorbates.Unambiguous identification of the molecular fragment is not possible on thebasis of the data available to us.It is clear, however, that the fragment is obtainedby partial dehydrogenation of benzene.A mechanism for the adsorption of benzene on a W(lO0) surface involving asequence of fragmentation of benzene upon adsorption to form acetylene, dissociativeadsorption of the acetylene formed as HC=C--- and H- and finally displacementof H- by HC=C- at higher coverages (> 2 L) would be consistent with some ofthe above-mentioned observations. This mechanism is, however, not consistentwith the mechanism of chemisorption of acetylene. It is believed lo that acetylene,upon adsorption on a W(100) surface, initially decomposes completely to give anadsorbed layer of atomic C and H. Further exposure of the surface to acetyleneHresults in the formation of an ethylenic species, \C_dH, and a simultaneous / \displacement of the adsorbed H.Ethylene and butadiene have also been found loto behave like acetylene on W(100).A possible structure of the molecular fragment in question here could be oneinvolving di-o-bonded benzene shown below : 0 / \Indeed, examples of organo-metallic clusters with benzene bonded at the 1 , 2 positionsto two metal atoms are kn0wn.l The gas phase p.e. spectrum of a benzene moleculea-bonded to two W atoms at the ortho-positions is not accessible. However, on thebasis of the known effects of substitution l2 on the p.e. spectrum of benzene, we expectthat the highest occupied double degenerate n-orbitals of benzene would split by = 1 eV on forming the above-mentioned structure. Moreover, all the three benzenen-orbitals, which extend over a large region, would interact strongly with the metallic&orbitals, which also extend over a large region at the surface.Such n-d interactionswould increase the binding energies of all the n-orbitals, though by differing amounts.The energies of the a-orbitals, on the other hand, are likely to alter only a little134 ADSORPTION OF CYCLIC HYDROCARBONS O N TUNGSTENFig. 2(b) shows the gas phase p.e. spectrum of benzene.12 The vertical ionizationpotentials of the various 71- and a-orbitals are given below :n : 9.4 (doubly degenerate) and 11.5 eVa : 11.8, 12.2, 13.9, 14.2, 14.8, 15.4, 16.9 and 19.2eVOn comparing fig.2(b) and (aiv), it can be seen that the molecular fragment could,indeed, have the di-a-bonded structure. For the peaks at -2.2 and -3.2 eV inthe difference spectra could be due to the doubly degenerate n-orbitals followingsplitting due to ortho-substitution and further stabilization through n--d interaction.The peak at -5 eV could then correspond to the second band in the p.e. spectrumof benzene after a slight shift in the energy of the a-orbitals and a stabilization throughn-dinteraction of the third n-orbital of benzene. The peak at - 7 eV for low coverage[fig. 2(a i)] agrees well with the third band in the p,e. spectrum of benzene. The shiftof this peak to higher binding energy with increasing exposure could be due to achange in the C-W bond energy with increasing coverage.Further, after a slightshift to lower binding energy of the S t h band, the fourth and fifth bands of the p.e.spectrum of benzene could be made to coincide with the peaks at -9 and - 10.4 eVof the difference spectra. It must be emphasized at this point that, in spite of thereasonable agreement between the expected p.e. spectrum for di-a-bonded benzeneand the observed emissions induced by the molecular fragment in question here,one cannot rule out the possibilities of other structures for the fragment. Comple-mentary studies by X.P.S., l.e.e.d., i.r. and h.r.e.1.s. would, therefore, be useful inresolving the problem.The emissions induced by the carbon overlayer (fig. 6) left after complete de-hydrogenation of the molecular fragment formed upon adsorption of the cyclichydrocarbons on a W(100) surface do not agree with the emission of carbided W(100)surface.This is indicative of the overlayer being chemisorbed carbon. Moreover,the sharpness of the emission at -4.05 eV (fig. 6) and the fact that the overlayerremains stable until x 1600 K seems to suggest that the carbon atoms on the surfaceare perhaps strongly bonded to each other and may exist as a graphitic overlayer.L.e.e.d. studies would, of course, be useful in this respect. However, on heating thissurface to a temperature == 1600 K, its p.e. spectrum changes from that shown infig. 6 to the p.e. spectrum of a carbided W(lO0) surface.8The author thanks the Royal Commission for the Exhibition of 1851 for theaward of a Research Fellowship which made this work possible. Useful conversa-tions with Prof. J. Lewis and R. Hoffman and Dr. J. Q. Broughton are gratefullyacknowledged.J. E. Demuth and D. E. Eastman, Phys. Rev. Letters, 1974,32, 1123.G. Broden, T. Rhodin and W. Capehart, Surface Sci., 1976, 61, 143.T. E. Fischer, S. R. Kelemen and H. P. Bonzel, Surface Sci., 1977, 64, 157.J. C. Bertolini, G. Dalmai-Imelik and J. Rosseau, Surface Sci., 1977, 67,478.J. L. Gland and G. A. Somorjai, Surface Sci., 1973, 38, 157.J. L. Gland and G. H. Somorjai, Surface Sci., 1974, 41, 387.W. F. Egelhoff, J. W. Linnett and D. L. Perry, Farday Disc. Chem. SOC., 1974,58, 35.E. W. Plummer, Topics in AppZ. Phys., ed. R. Gomer (Springer-Verlag, Berlin, Heidelberg,New York, 1975), vol. 4, p. 144.' A. J. Bhattacharya, J. Q. Broughton and I). L. Perry, J.C.S. Faraday I, 1979,75, 850.lo A. K. Bhattacharya, J.C.S. Faraday I, 1979,75, 863.l1 J. Lewis and B. F. G. Johnson, Pure AppZ. Chem., 1975,44,43.l2 D. W. Turner, G. Baker, A. D. Baker and C. R. Brundle, Molecular PhotueZectron Spectroscopy(Wiley-Interscience, New York, 1970), p. 264.(PAPER 8/1554

ISSN:0300-9599    

DOI:10.1039/F19807600126

出版商:RSC    

年代:1980

数据来源: RSC

摘要:

J.C.S. Faraday I, 1980,76,135-151Improved Representation of Velocity Correlations inAqueous Electrolyte Solutions3~ ALFONS GEIGER AND H. GERHARD HERTZ*Institut fur Physikalische Chemie und Elektrochemie der Universitat Karlsruhe,Kaiserstr. 12, 7500 Karlsruhe, West GermanyReceived 18th December, 1978A general dehition of the salt molecule for any asymmetric polyvalent electrolyte in solution isgiven which fulfils the requirement that the solute mass flux in the presence of an electric fieldstrength vanishes when referred to the proper representative point of the molecule. Effectively thisis the definition of the mass flux in the microscopic representation of an electrolyte solution. Theelectric current density arising from a gradient in the chemical potential also vanishes wheu referredto the same representative point.Time integrals over the velocity cross-correlation functions havebeen calculated for aqueous NaCl, LiCI, CaCI2 and 33aClz solutions using conductivity data, transportnumbers and mutual diffusion coefficients fulfilling the above requirements. A first approximationis also given for the calculation of velocity correlation coefficients in the case where transport numbersare not available.In a number of previous papers lm3 we have reported time integrals over velocitycross-correlation functionsViC,NYf i i = - (vy)(O) vg)(t)) dt i = a, c3 0 s v,c,NV *<v(1‘)(0) vy’(t)> dt i , j = c, a i # j3 0 s fij = -for aqueous electrolyte solutions in the concentration range 0 < &/moldm-3 < 4.The salt may be any electrolyte which ionizes in a solvent w (water), to give v, anions(a) of valence z, and v, cations (c) of valence 2,. c, is the salt concentration inmol ~ r n - ~ , N is Avogadro’s number, V is the solution volume and vy)(t) (i = a, c)denotes the velocity of particle n of constituent i at time t.The velocities uti)(t) aremeasured relative to the laboratory (or cell) coordinate system; they are vectorquantities, but for simplicity we omit vector notation. As usual the pointedbrackets represent the ensemble average.The three independent quantities given by eqn (1) and (2) were calculated fromexperimental values of the equivalent conductance, A, the transport number of thecation, t,, and the mutual diffusion cmEcient of the salt against water, D,.Thequantity D,, describes the flux of a neutral substance, the dissolved salt, whereas thetransport number is a quantity which describes the motion of a single ionic speciesunder the action of the electric field; the microscopic representation of the electriccurrent consists of the separate motion of ionic species. In order to incorporate themutual diffusion data in the treatment it was necessary to generalize the concept of amolecule to a solution of a strong electrolyte, i.e., to a situation where the intra-molecular distances of the anions and cations are not constant. For the transportproperties considered here, the velocity of the salt molecule is of particular interest ;this velocity will be given in terms of the velocities of the cations and anions belonging13136 VELOCITY CORRELATIONS IN AQUEOUS ELECTROLYTE SOLUTIONSto the molecule.In our previous work we used the following expression connectingthe instantaneous ionic velocities with the instantaneous molecular velocitywhere D, and Da are the self-diffusion coefficients of the cation and anion, respec-tively.D~ = 3 J: <u(1')(0) vy)(t)> dt i = c, a. (4)Eqn (3) was chosen in such a form that in the limit of high salt dilution the correctlimiting value for D,, results :which represents the well-known Nernst limiting formula when the self-diffusioncoefficients are replaced by the ionic equivalence conductances Li using the relationwhere P is the Faraday constant.however, it is not yet clear which conditionDi = RPRT/(zilP2 (i = c, a)Thus there is no doubt that eqn (3) is correct at very low salt concentrations;has to apply at high salt concentrations. In the meantime this condition has beenworked out for 1-1 electrolyte^.^ We require that the quantitymust vanish at all concentrations, i.e.,A = 0 0 < C, < csatwhere csat is the saturation concentration of the salt. The validity of eqn (7) is aconsequence of the requirement that the salt molecule is electrically neutral ; thusin a homogeneous solution (c, = const.) there cannot be any mass transport underthe action of electric field strength. It appears that the application of the molecularvelocity definition eqn (3) in general does not satisfy eqn (7).Only in the case thatthe cations and anions involved are not strongly hydrated was eqn ('7) found to beapproximately fulfilled.These are the ions K+, Cs+, Cl- and I-, typically those ionswhich are usually characterized as being structure-breaking. The chlorides of Na+,Li+, Ca2+ and Ba2+ showed distinct deviations from eqn (7).that for symmetrical electrolytes the quantitiesDc and Da in the numerator of eqn (3) have to be replaced by*It has been shown elsewhereD,+P i = c, awherein order to guarantee the validity of eqn (7) for all concentrations.* In ref. (5) the quantity S* is denoted by 6A . GEIGER AND H. G . HERTZ 137The present paper has three purposes : (i) To generalize the correction given ineqn (8) in such a way that we write :In particular we shall investigate the choice6, = 66, = -6.The use of eqn (Sa) and (8b) rather than eqn (8) has only practical reasons; in thisway singularities of the function ( O f + P ) - l are avoided.(ii) To work out the correctformulae for non-symmetric electrolytes. (iii) To recalculate the velocity correlationcoefficients faa, f c 0 fac as defined by eqn (1) and (2) using the new definition of themolecular velocity for the salts NaCl, EiCl, CaCl, and BaCl,. The behaviour ofthe newLj remains practically the same as previously reported.2*THEORETICALMASS FLUX AND ELECTRIC CURRENT DENSITY: GENERALWe consider a binary electrolyte solution consisting of any polyvalent electrolyteand water. In the microscopic representation of this system the instantaneousvelocities of the cations areand the instantaneous velocities of the anions areu p , u p y u p y .. . a,& (a)where N , = VcsN.in the system; j: is defined by the relationThere may be a inacroscopic (i.e., observable) mass flux of the solute,jf = j f (Y, t ) ,(10) - - a’s - -div j zattogether with j : = 0 at those positions where grad ps = 0 and ps is the partial massdensity of the solute. According to eqn (10) j $ is defined in the laboratory (or cell)coordinate system. Having j $ at each point, it is easy to transform jz to a centreof volume fixed referenceThere may also exist an electric current density jq at each point of the system.jq is connected with the rate of increase of the local internal energy; it may bemeasured via, e.g., its magnetic field.The microscopic representation of the instan-taneous electric current density of the system isthe solute mass flux is then denoted by j,.1Jq = ,(q,Vp+q,up+q,vy’+ . . . + q a u ~ ’ + q a u ~ ) + q a u p + . . .). (1 1)g, and q a are the electric charges of the cation and anion, respectively. We havewhere --e is the charge on the electron. z, and 2, are related to one anotherz,v, = zavaso that the salt molecule is electrically neutral. The velocities in eqn (11) are no138 VELOCITY CORRELATIONS I N AQUEOUS ELECTROLYTE SOLUTIONSentirely independent : a volume element which is macroscopically small has to remainelectrically neutralZ,CL(t) = z,cL(t)where c;(t) ; i = c, a, are the instantaneous number densities of cations and anions.The instantaneous mass flow density in the microscopic representation is given bythe relationwhere mc and ma are the masses of the cation and anion, respectively.v$) is theinstantaneous velocity of the kth molecule. Now, as has already been describedin the introduction [see eqn (3), (8), (8a) and (8b)], we write $1 in terms of the instan-taneous velocities of the v, cations and v, anions as the following relation :l*This then implies that the rnicroscapic representation of the instantaneous solutemass flux is defined as the quantity :In eqn (14a) and (14b) 6 -+ 0 as c, +. 0 ; then these expressions lead to the correctlimiting behaviour for c, 4 0 as given in eqn (5).l'According to the results of the linear response theory' the observable meanelectric current density isand likewise the observable mean mass flow density isjm = -3kTwhere A is the time derivative of the effective perturbation energy connected with theflux in question, the time dependence being given by the microscopic motion of thesystem in the equilibrium state.We consider here two kinds of perturbation energy : (1) the electrostatic oneA,, = (q,rr)+ qcrp)+ .. . + q,ry)+ q , r f ) + . . .)& (17)where rii) (i = c, a) is the position of the Zth cation or anion, 8 is the gradient of themacroscopic electric potential (i.e., the electric field strength) ; (2) the thermo-dynamic oneA m = -(vcmc+vama)(yl+~2 +Y, + * -1 grad PZ (18)where the meaning of the yk is that the mass of the kth molecule is to be placed atyk, k = 1,2, .. . N,. p,* is the specific chemical potential of the solute. Since thegradient of the latter quantity is always connected with the gradient of the specifiA . GEIGER AND H . G . HERTZ 139chemical potential of the solvent, A , contains a factor 1 +ps/pw which we omit forsimplicity.6 pw is the partial mass density of the solvent water. NQW we mayinsert both the quantities eqn (17) and (18) in eqn (15). This then gives two kinds ofelectric current density ; one connected with 8 and another with grad ,us*. Likewise,substitution of eqn (17) and (18) in eqn (16) yields the corresponding two kinds ofmass flux. We begin with eqn (16).MASS FLUXESFirst we must find Ael. It follows from eqn(l3) and (14a) that the position of thekth salt molecule isy k = [ ( Y o f Va)(VcD, k = 1 , 2 .. . N , (19)where r?’ = r$), r la) = r g ) are the positions of the Zth cation and anion, respectively ;then t$) = ~$1, j k = uf), where vector notation has again been dispensed with forsimplicity. However, according to eqn (17) the contribution to the potential energyof a salt molecule due to the electric field strength 8 isThe essential feature of eqn (14a) is the redistribution of weights of the cationic andanionic velocities. Usually the cation is strongly hydrated and as a consequencethe thermal motion is more strongly damped. To compensate partly for this effect,in the expression for the total solute mass flux, the instantaneous cation velocitycontribution is increased by a factorrelative to that of the anions.Thus, the dynamical variable “ mass flux of the salt ”,apart from a multiplication by the mass, consists of a redistribution of weights to theinstantaneous ionic velocities. However, if the velocities have different weights,then the forces must also be modified in the corresponding way. Thus, the effectivepotential energies of the cations and anions with respect to the dymmical variablemass flux areand, in order to obtain the total effective potential energy of the molecule, one mustreplace the field strength in eqn (20) by the effective field strengths which occur ineqn (21) and (22) as the factors of z,ergf and -z,er$). So we arrive at the effectiveperturbation energy of the system of N, points described by eqn (19) 140We now form the time derivative of eqn (17) in which 8 = const.the result is :VELOCITY CORRELATIONS IN AQUEOUS ELECTROLYTE SOLUTIONSWith eqn (23)and we can introduce eqn (13), (14b) and (24) into eqn (16) to obtain :(u\,ui,> = Jb (v!?(O) ug)(t)) dtfor the intramolecular velocity correlations andi , j = c, a I = 1 , 2(vfu{) = (u(l')(O) v i j ) ( t ) ) dt i , j = c, a I = 1,2Consider the case that all v, + v, ions axe tightly bound so as to form a moleculeThen we havel: for the intermoleculm velocity correlations.in the conventional sense.D, = Da = ~ ( U ~ ~ U ( I ) i , j = c,a I = 1 , 2and<vC,v;> = (uC,v;) = (u;u;) = (u;u;).We set 6 = 0 and introduction of these relations into eqn (25) yieldsIf there is an electrical field strength acting on a system containing dissolved neutralmolecules, then no mass flux of the solute is observed ; this is a well-known result.When we insert the velocity correlation coefficients fee, faa and fa,, defined in theintroduction, in eqn (25) and neglect the terms involving products of N, and intra-molecular cross-correlation coefficients, i.e., those containing (I${ vyj} (i = c, a),the result is :j,, = 0.(26A . GEIGER AND H . G. HERTZ 141We now choose 6 such that we obtain j,, = 0. If zc = Za and if zc # z,butfac = 0, then we find from the requirementj,, = 0 according to eqn (27)In the general case we proceed as shown below.the specific chemical potential. With eqn (18) and (19) we have :Next we give the expression for the mass flux which arises from the gradient ofThis expression we insert in eqn (16) together with eqn (13) and (14a) which gives theresult for the mass flux (taking into account the factor 1 + p , / ~ , ~ )ELECTRIC CURRENT DENSITYSo far, we have given expressions for the mass fluxes.Next we turn to thecorresponding formulae for the electric current density, see eqn (1 5).The instantaneous electric current density is given by eqn (11). If we considerthe mean electric current connected with the gradient of the electric potential, thenthe potential energy involves direct coupling with each individual ion and not with apoint representing a part of the molecule. Thus, we can apply eqn (17) as it standsand we haveA,, = [ezc(vF) + vp) + .. .) - ez,(vl;') + u p ) + . . ,118. (3 1)Then combination of eqn (ll), (15) and (31) yields= K B .From eqn (32a) and (33) we obtain the expression for the reducedductivi t y :ART Z A * = - - - - Dc + f c c - 2fac + ?(faa + Da) * ZcF2 ZCWe now consider j,,:, the electric current density arising from thespecific chemical potential.J(32b)(33)equivalent con-(34)gradient of th142for the velocities of the single ions :VELOCITY CORRELATIONS I N AQUEOUS ELECTROLYTE SOLUTIONSA,.,., is given by eqn (29) and needs no further comment. In eqn (11) we write~2:) = jk+V(k(i') a = c, a (35)where 9, is the time derivative of the position given in eqn (19).u p =is the velocity of the ion ki relative to the velocity of the point $.'a.Since all micro-scopic velocities occurring in the treatment are taken at equilibrium, the system isfully isotropic and thus we have< u p ' ) = 0.As a consequence, the mean microscopic electric current density carried by the vcNsrepresentative points $1 and the vaNs representative points xia) isCombination of eqn (15), (29) and (36) then gives the result :Apart from the gradient of the chemical potential this is the s m expression as eqn(27). Thus, if the quantity 6 is chosen such that the mass flux connected with thegradient of the electric potential vanishes, then, at the same time, the electric currentconnected with the gradient of the chemical potential vanishes.TRANSPORT NUMBERSFor the total analysis we need the transport numbers which we introduce in thefollowing way.If the boundary arrangements of the electrolyte solution are con-structed suitably (i.e.? if there axe electrodes), then in the presence of the electriccurrent densityj, and in the non-uniform parts of the system, a variation of the solutepartial density with time, dp,/dt, may be observed. We write this quantity in the form :The two quantities tc and ta are the transport numbers for the cation and anion,respectively. We consider them to be defined by a pair of excess constituent massfluxes, j : and j A . GEIGER AND H. G . HERTZ 143We emphasize that our choice of a microscopic representation of the mass flux j zis such that j$ = 0 when grad j ~ : = 0 even if G$ # 0. In contrast to this, the excessconstituent mass fluxes j [ (i = c, a) do not have this property; they axe proportionalto jq.Thus j ; # 0 if 8 # 0. This is the justification of the designation " excessconstituent mass flu ". The strict coupling between the two excess constituentfluxes is given by the relationta+tc = 1 (41)which is a consequence of the requirement of electrical neutrality. With eqn(39)-(41), eqn (38) becomes :2' = - div j,* - div j b - div j iatj*Fzcvc= - div j:- -(vcMc + v,M,) grad t,.Integration of eqn (42) with respect to x givesxo and xl are two coordinates given by the apparatus. This equation allows themeasurement of t,. One possibility is to measure the left-hand side and to arrangethe set-up by adding another cationic species such thatdivj: dx = 0 1::in the range of interest x1 < x c x2.Now in eqn (43) ti = t,. This is the movingboundary method. Or we may havewhere the integration range 6x is a very thin layer at the electrode and j : may bemeasured. Now in eqn (43) ti = t , = (1 - t,). This is the Hittorf method. Havingthus established the procedure to measure the constituent excess mass fluxes, weneed a microscopic representation of these fluxes. In order to find the correctanswer we return to eqn (11) and write this expression in the following form :In the same way as each ion carries a fixed charge, so each ion also carries a fixedmicroscopic mass. Thus, the instantaneous electric current density is rigidly coupledwith an instantaneous mass flow density. As a consequence, the microscopicrepresentation of the electric current density is at the same time a microscopic re-presentation of some kind of mass flux.In particular, let us consider the twocontributions J g ) and J$') as defined in eqn (44) separately. We can then say:the .Tt) (i = c, a), converted from a charge-characterized to a mass-chaxacterizedform, are the microscopic representations of the excess constituent mass fluxes 144 VELOCITY CORRELATIONS I N AQUEOUS ELECTROLYTE SOLUTIONSOf course, in the equilibrium state we have SE) = J k ) = 0. The potential energywhich acts as perturbation in order to cause J:) f 0 is again for each particle theproduct of the gradient of the electric potential with the corresponding couplingparameter, i.e., the electric charge.Thus, in order to calculate the mean value ofJ:)? we have to apply eqn (16) with the perturbation eqn (17) and the dymamicalvariable eqn (45a). The result iswhen the definitions of the velocity correlation coefficients are introduced. Equatingthis with the macroscopic definition of the excess constituent mass flux [eqn (39)], weobtainFinally, the combination of eqn (47) with eqn (326) yieldsNote that the mean values of the excess constituent fluxes axe not themselves directlyobservable quantities. Only the fact that the divergence of j : is strictly coupledwith the divergence of j :2, 2, - div jk = - div j :Mc Mawhich givesgrad tc = -grad t ,leads to dp,/dt or j s [uia eqn (43)] as quantities concerning the neutral solute, thusallowing the determination of t,.MUTUAL DIFFUSION COEFFICIENTFinally, the mass fluxj, given by eqn (30) has to be connected with the experimentalquantity dp,/a't introduced in eqn (10).The mass flux j: measured in the cell-coordinate system is- j z = D,* grad ps (49)where Df is the mutual diffusion coefficient in the cell-coordinate system.other hand, j s in eqn (30) may be written asOn thA . GEIGER AND H . G . HERTZ 145where QSs is the phenomenological coefficient and D, is the diffusion coeficient of thesolute in the local centre-of-mass fixed reference frame. The mutual diffusioncoefficient in the volume fixed reference frame isandD,, = pT$D,0: = D,,- if volume changes during diffusion are absent.the solvent water.arrangement : 'V$ is the partial specific volume ofCombination of eqn (50), (51a) and (30) yields, after some re-' (52) Dc + f c c + 2vafacvc ~ { Dc+6 [(DE+S)(Da-6)lffYaand considering eqn (51b), (49) and (10) one sees that D,, is a measurable quantity.c$ and cs are the solute concentrations given in mol g-1 and mol ~ m - ~ , respectively,p is the total density of the solution (g c ~ n - ~ ) and y* is the mean activity coefficientof the solute.EVALUATIONNext we introduce the following abbreviations :u = Da-tfaav = Dc+fcc.Then combination of eqn (34) and (48) yields :(53)(54)fac = v-tCA." (55)andwithFurthermore, in eqn (52) we set :Dsw(vaDc + VcDaXDa +Dc)cslcZ L =p(1 +c,*Ms)DaDC(l +c,* d In y*/dc,*)' (57)With these abbreviations and with eqn (53) and (54) eqn (52) can be written in theform :[v + A*( 1 - 2tc) Jc-Da-6 (58)L 2, v - tcA* --[(D, + 6)(Da - a)]+ + va- V C ' j + 2 %Da + D146and the requirementVELOCITY CORRELATIONS IN AQUEOUS ELECTROLYTE SOLUTIONS=jm, = 0according to eqn (27) takes the form :L, Da, D,, A* and t , are given as a function of the concentration; they areexperimental quantities or directly obtainable from other experimental quantities.The two unknowns u and 6 then have to be determined such that they fulfil botheqn (58) and (59).For symmetrical electrolytes the situation simplifies. Now theright-hand side of eqn (59) equals zero and S = 6" [see eqn (28)]. In this situation6 = 6" according to eqn (28) can directly be introduced into eqn (58). The result is :or with eqn (56)(61)L ~ v , ( v - t,A*)v(g-' + 1 ) + A*@- 2t,)c-' -(" 4- = ( V [ V +A*( 1 -2tc)]c- '}faFrom this equation v has to be determined as a function of the salt concentration.When v is available, then fa,, fc, and fac may be calculated using the set of eqnThe numerical solution to eqn (61) was achieved by a standard Newton-Raphsoniteration.To solve the system of the two non-linear eqn (58) and (59) for v and 6,a subroutine (library IMSL, computer system UNIVAC 1108) based on Brown'smethod was used. Tests insured that the stopping criterion for the iteration processgives a numerical accuracy in the results of at least six digits.(53)-( 5 6).FIRST APPROXIMATION FOR THE VELOCITYCORRELATION COEFFICIENTSSo far we have used expression (14b) as the definition of the instantaneoussolute mass flux in the molecular picture.The quantity 6 occurring in this formulawas determined in such a way that the instantaneous electric current density and theinstantaneous excess constituent mass fluxes are represented by eqn (11) and (451,respectively, and that at the s m e time the solute mass fluxj, vanishes if grad p,* = 0,whatever the electrical potential gradient may be. The electric current density andthe solute mass flux are well-defined quantities even if the excess constituent massfluxes do not exist. This situation occurs in all those arrangements where the massflux j , vanishes at the boundaries of the system. For an ordinary mutual diffusionexperiment it is clear that, for the electric current density to be non-zero, one has toplace the electrolyte solution in a capacitor with insulated plates and to charge thecapacitor. During the charging process jq # 0 and, of course, if grad ,uz = 0 wealso have j, = j,, = 0 whilst 8 # 0.In this case the excess constituent mass fluxesare undetermined and the mean displacements of a cation and of an anion are both zero.In contrast to this, when we have mass fluxes at the boundaries, excess constituentfluxes also exist and the mean displacement of a given ion does not vanish ; for instance,a cation undergoes a displacement in the direction of the cathode.Having no observation on which to fix the values of the excess constituent mass fluxeA.GEIGER AND H . G. HERTZ 147the parameter 6 in eqn (14) and (14a) remains undetermined. We may then chooseany &value with the property 6 -+ 0 as c, -+ 0 and, since the results for the velocitycorrelation coefficientsfcc,faa andfac depend on this choice, they axe no longer uniquelydefined. The simplest approximation is to set 6 = 0 over the entire concentrationrange. Then the three unknowns fco faa and fac axe to be determined from theequivalent conductance, eqn (33), from the mutual diffusion coefficient according toeqn (52) (with 6 = 0) and from the condition j,, = 0 according to eqn (27) (againwith 6 = 0). Since we do not need the transport numbers (for which in many casesexperimental data for high concentrations axe not available) we call the V.C.C.obtainedin this way " first approximation velocity correlation coefficients ". After somealgebraic operations one obtains from eqn (27) with j,, = 0, 6 = 0, eqn (33) andeqn (52) :A* - D,, z -"(Fv)- '[( 1 -:$la - 2(DaDC)*]f c c - ZC l + y -- /- \ 2withU CF =v =f a c =(?f+D?)Va+Vcfaa is then given by eqn (32b).For the salts containing only structure-breaking or structurally indifferent ionsthe first approximation to the velocity Correlation coefficients provides the correctset off,,,faa andf,,.l* 2*RESULTS AND DISCUSSIONIn fig. 1 and 2 the results of the computation Offaa,fcc,fac and 6, using eqn (58) and(59) for the unsymmetrical electrolytes, and eqn (61) for the 1-1 electrolyte, togetherwith eqn (53)-(56), are shown.The experimental data are the same as used in ourearlier papers,,. where literature references are given. It is not the purpose of thepresent paper to discuss the physical meaning of the shapes of the curves shown in thefigures. This has been amply done in the preceding papers of this ~eries.l-~g Theresults shown in fig. 1 and 2 are practically identical with those previously reported.2*The dashed curves in these figures represent the previous values in those cases wherethe deviation allows separate drawing. In all other cases the difference is smaller.So, obviously the fulfilment of the requirement j,, = 0 [eqn (231 is not of greatimportance for the final numerical results. We also applied eqn (61), which is onlyvalid for symmetrical electrolytes, to the system CaCl, and BaCl,.The numericalresults for faa, fac and fCc again are practically indistinguishable from the exact com-putations, only 6 is different. The 6 values of this approach are indicated as dottedcurves in fig. 2248 VELOCITY CORRELATIONS I N AQUEOUS ELECTROLYTE SOLUTIONSIn a further computation we examined the influence of the choice of 6 on thefinal results for the velocity correlation coefficients. Whereas the fully drawn curvesin fig. 1 and 2 correspond to the choice 6, = 6, 6, = -6 and the dashed curvesrepresent the results for 6 = 0 (using the t, values) we show in fig. 3 the resultingvelocity correlation coefficients when one sets 6, = 6, = 6* [see eqn (8) and (9)].-0.4cI I II L 1 2 3 F+0.2 --0.2-0.3--0.4 -facdfaafccI I I2 3 *1FIG.1.-Velocity correlation coefficients fa,, fac and fcc for aqueous NaCl (a) and LiCl (b) solutionsat 25°C. The salt molecule location parameter 6 is also given. The dashed curves are our previousresults where the condition of vanishing mass fluxj,~ has not yet been fulfilled. Zs is the salt con-centration in mol dm-3.The qualitative behaviour is similar to that shown in fig. 1 and 2, but all curves areshifted in the direction of more negative values. However, the -6 values are muchcloser to the D,, D, values; in the CaC1, system at E =:4 mol dm-3 we have D, = - 6,which causes a singularity of (D, + 6)-l. For this reason the choice 6, = 6,6, = - 6is preferred.Finally, in fig. 4 we give the first approximation to the velocity Correlation co-efficients as defined in the previous section.It may be seen from fig. 1, 2 and 4that the fac are almost identical in both representations, whereas the behaviourof fc, and faa is only qualitatively similar. In particular, the order of fc, and faa haschanged, in fig. 4 Ifaa] > lfccl in all cases. This is due to the fact that we have D, > D ,and according to eqn (6) and ('7) (which approximately hold also for the 1-2 electro-lytes) faa/Da = f,,/Dc when the excess constituent mass fluxes a.re undetermined.In contrast to this, when the excess constituent mass fluxes are fixed, expressed bya knowledge of the transport numbers, then the relative weight of the cations to forA. GEIGER AND H. G. HERTZ 149rl L+0.10- 0.1-0.2- 0.3- 0.4+ 0.2i i3t1 2 4 ' F+ 0.10- 0.1- 0.2- 0.3-0.4 1 , I I1 (6) 1 2 3 4 '25FIG.2.-Velocity correlation coefficients fa,, faC andf, for aqueous BaCl, (a) and CaCI, (b) solutionstogether with molecule location parameter 6 (T = 25°C). The dashed curves are our previous resultswhere the condition of vanishing mass fluxj,s has not yet been fulfilled. c, is the salt concentrationin mol dm-3. For other details see text.i 1 1 t1 2 3FIG. 3.-Velocity correlation coefficients for aqueous solutions of NaCl (l), LiCl(2), CaCl, (3) andBaCl, (4) when computed with the location parameter S* (dot-dashed curves) according to eqn (8)and (9). Where curves are broken a* leads to a singularity (2' = 25°C)150the salt molecule is increased compared with that given by the friction property alone.This then also increases the magnitude of fCc relative to faa.In summary, the velocity coefficients are not very sensitive to changes in theparameters defining the salt molecule.One can demonstrate clearly the hierarchyof definability of a salt molecule. In equilibrium the definition of a salt molecule isentirely arbitrary, the ions themselves may be considered to be the independentparticles. Of course, this holds as long as no equilibrium properties lead to theconclusion that ion pairs should exist.VELOCITY CORRELATIONS IN AQUEOUS ELECTROLYTE SOLUTIONS+ 0.2 c- - 0 6 L 0.7 1 2 3 42sFIG. 4.-First approximation to velocity correlation coefficients for aqueous solutions of NaCl (l),LiCl(2), CaC12 (3) and BaC12 (4).In the presence of an electric current, when the system remains uniform, the samestatements hold true.If there is a non-uniformity in the system, mass fluxes occur in the analyticalfield q(r, t ) (i = 1,2) represented by the solution.The analytical operation of takinga sample always yields the neutral salt, in spite of the fact that the ionic mobilitiesdiffer. This fundamental result requires the definition of the salt molecule. How-ever, the location of the salt molecule is not uniquely specified; in eqn (19) thedefinition of the position vectors yk may involve any &value, provided 6 + 0 asc, + 0. We have, arbitrarily, chosen the example 6 = 0 for all concentrations(fig.4).The next requirement is that the excess constituent mass fluxes in the presence ofelectrodes ( j , = 0) are also represented by the concept of the salt molecule. Themass fluxes of the solute towards or from the electrodes are correctly described bythe first approximation, but the “rotation” of the salt molecule representing theexcess constituent mass fluxes, eqn (39) and (40), has also to be described correctly.In fact, the definition of the molecule does not a priori fix the location of the ionsbelonging to the molecule and we have also to account for the “ salt molecule ” whoseparts are extending over the entire electrolytic cell from the cathode to the anode.At the cathode the anion is formed and at the anode the cation is formed; thiA .GEIGER AND H. G . HERTZ 151molecule then ‘‘ rotates ” about the point y k as given by eqn (19) with the 6 valuesshow in fig. 1 and 2. This is the second aspect of the concept of a salt molecule : thecoherent effects at the cathode and anode, which have the form uf two diffusion fluxesof solute towards and away from the electrodes (which in general are asymmetric atthe boundaries). Were they symmetric then we would always have t, = t , = 1 2‘Woolf and Harris have also computed velocity correlation coefficients using theapproach based on two ionic fluxes in the solvent at rest as developed by Miller, losee also: ref. (11) and (12). Miller’s phenomenological coefficients Zij are related tothe velocity correlation coefficients via phenomenological coefficients in the localcentre-of-mass fixed coordinate system. Although this treatment in its physicalessence is different from ours because mass flux and electric current density are con-sidered to be directly coupled, the final results for the velocity correlation coefficientsaxe very similar to our data presented in fig. 1 and 2. To give a rough estimate, thedeviation is of the order of 10 %. Thus even here we see that the velocity cor-relation coefficients are fairly insensitive to the path on which they were derived.H. G. Hertz, Ber. Bunsenges. phys. Chem., 1977,81, 656.H. G. Hertz, K. R. Harris, R. Mills and L. A. Woolf, Ber. Bunsenges.phys. Chem., 1977,81,664.H . G. Hertz and R. Mills, J. Phys. Chem., 1978,82,952.See, e.g., R. A. Robinson and R. H. Stokes, Electrolyte Solutions (Butterworths, London, 1970).H. G. Hertz, in Protons and Ions Involved in Fast Dynamic Phenomena, ed. P. Laszlo (Elsevier,Amsterdam, 1978), p. 1.See, e.g., D. D. Fitts, Non-equilibrium Thermodynamics (McGraw-Hill, New York, 1962).New York, 1969).K. M. Brown, SIAM J. Numerical Analysis, 1969, 6, 560.L. A. Woolf and K. R. Harris, J.C.S. Faraday I, 1978,74, 933.D. 6. Miller, Faraday Disc. Chem. SOC., 1977, 64, 295.’ See, e.g., E. W. Steele, in Transport Phenomena in Fluids, ed. H. J. M. Hanley (Marcel Dekker,lo D. G. Miller, J. Phys. Chem., 1966, 70,2639.l2 R. Paterson, Furaday Disc. Chem. SOC., 1977, 64,304 and discussion on p. 349.(PAPER 812172

ISSN:0300-9599    

DOI:10.1039/F19807600135

出版商:RSC    

年代:1980

数据来源: RSC

摘要:

J.C S. Faraahy I, 1980,76,152-161Thermal Decomposition of Solid Chromium(II1)Tris-N-benzo yl-N-phenylhy dr oxylamineBY JOS6 C. MACHADO," MAURO M. BRAGA, ANA M. P. R. DA LUZ ANDGILLES DUPLATRE~Departamento de Quimica, Universidade Federal de Minas Gerais,30.000 Belo-Horizonte, BrazilReceived 6th February, 1979The thermal decomposition of the CrIII tris-N-benzoyl-N-phenylhydroxylamine chelate,Cr(BPHA)3, has been investigated using radiometric methods by labelling the complex with 'Cr.The decomposition is found to occur in the temperature range 16O-18O0C, according to an auto-catalytic process. Kinetic models are discussed which should account for the experimental evidence.Although the coordination chemistry of a large number of metal complexes iswell documented, most thermal studies on coordination compounds have beenof a rather qualitative nature and very little is yet known about the precise kineticsand mechanism of thermal decomposition of chelates in the solid state.The presentpaper is the first of a series that has the purpose of studying the thermal decompositionof some metal complexes with the N-benzoyl-N-phenylhydroxylamine (BPHA)ligand. This ligand forms complexes with many metals and has been widely usedas a versatile organic analytical reagent.lm4 With C P , the Cr(BPHA)3 chelate isobtained, in which the central metal atom is hexacoordinated by the six oxygenatoms of the l i g a ~ ~ d . ~ The thermal stability of some BPHA metal chelates has beenstudied previously.6 In the present paper, the kinetics of thermal decomposition ofCr(BPHA)3 are reported and kinetic models are discussed in order to elucidate someaspects of the decomposition mechanism.EXPERIMENTALP R E PA R AT1 0 N 0 F cr-L A B E LLE D Cr(BPHA)3CrC13 .6Hz0 from Merck was irradiated with thermal neutrons in the Triga reactor of theNuclebras at Belo-Horizonte to obtain the 51Cr radioisotope (26 d). The irradiated chloridewas dissolved in deionised water and chromium hydroxide precipitated before centrifugingand washing with water. After the hydroxide was dissolved in dilute HCl aqueous solution,the pH was adjusted to 4.5k0.2 with ammonium acetate. A 10 % excess BPHA alcoholicsolution was then added under continuous stirring for 2 h. The finely-divided green precipi-tate was subsequently filtered, washed with hot water and dried.The compound wasfurther purified by chromatography through a cellulose column, with benzene as eluent.After the solvent was evaporated, the complex was vacuum-dried overnight and ground topass 300 mesh.The elemental analysis on an inactive sample prepared under the same conditions asgiven above led to the following results : calculated content per molecule of Cr(C13H1002N)3in wt %: Cr, 7.55; C, 68.02; H, 4.39; N, 6.10; found: Cry 8.10; C, 66.32; H, 4.55;N, 5.88.7 Permanent address : Laboratoire de Chimie NuclCaire, Centre des Recherches NuclCaires,B.P.20,67037 Strasbourg, Cedex, France.15MACHADO, BRAGA, D A L U Z A N D DUPLATRE 153THERMAL TREATMENTS, SEPARATION AND COUNTINGThe thermal decomposition experiments were carried out using an oil-bath with a thermalstability within +0.2"C.The samples ( M 15 mg) were introduced into 10 cm long, 0.5 cmdiameter thin-walled Pyrex tubes which were immersed in the oil-bath. After heating, thetubes were rapidly cooled in water, at room temperature, before analysis.The non-gaseous stable products of the thermal decomposition of Cr(BPHA)3 have beenseparated in a chromatographic column with cellulose as the stationary phase and mixturesof solvents of various polarities (cycloliexane, benzene, chloroform, acetone, methanol andacetic acid) as eluents. The stable products which have been characterized by infraredspectroscopy after the chromatographic separation were : benzanilide, chromium(1n)benzoate and a further unidentified chromium product, in agreement with previous resultse6For the chromium containing fractions, in the systematic experiments, thin layerchromatography was employed as the analytical technique, using a Merck microcrystallinecellulose layer E 1 mm thick applied on a 10 x 20 cm2 glass plate.After heating, the sampleswere dissolved in a small volume of chloroform, then deposited onto the plate. Purebenzene served as solvent and the material was eluted twice. The decomposed fractioncontaining the metal ion did not move significantly, while the fraction related to the un-decomposed chelate was eluted with Rf M 1. After elution, the plate was air dried and thecellulose film was cut into ten 1.5 cm wide strips, perpendicularly to the direction of elution.Each of the trips was taken off for radioactivity measurements.The countings were made on the 320 keV photopeak of 51Cr, using a NaI(T1) well-typescintillator coupled to a multichannel analyser.The experimental procedures were systematically carried out in the dark becauseCr(BPHA)3 is photosensitive, especially when in solution.The average error for a givenexperiment is estimated as k0.02 in terms of the decomposed fraction percentage, a. Theresults presented arise from a single batch of the compound, but preliminary experimentsshowed no significant change, viz. within experimental error, to appear in the a against Tcurves when using samples from various preparations.RESULTS AND DISCUSSIONPreliminary isochronal experiments with heating times of 20 and 40min havebeen conducted, in order to determine the temperature interval for Cr(BPHA),thermal unstability.The results are shown in fig. 1 in terms of the decomposedfraction, a. The decomposed fraction a at time t and for temperature Tis definedas the ratio of the radioactivity associated with the Cr-containing decomposedproducts to the total radioactivity of the sample. Examination of the curves infig. 1 led us to choose the 160-180°C interval as most suitable for the study of thethermal decomposition.A series of isothermal experiments in this temperature range was hence run andthe resulting curves are shown in fig. 2. The a against t plots display the well knownsigmoid-shape characteristic of many solid state thermal decompo~itions.~The isothermal curves were first analysed on the basis of a unimolecular law :-ln(l-a-) = kt+c.(1)A typical plot of -In (1 -a) against t is shown in fig. 3(a). Obviously, a singlefirst-order process cannot account for the decomposition process. Rather, twosubsequent first-order processes or some more complex reaction involving an inductionperiod are to be considered.Several reviews have been concerned with solid state de~omposition.~-~ O Theseshow that for almost all of the solids, nucleation and propagation of the nuclei havebeen invoked for the mechanism of decomposition.ll Owing to the presence of a1 54 THERMAL DECOMPOSITION OF CHROMIUM(III) COMPLEX2 JT("CRG. 1.-Isochronal variations of the decomposed fraction, a, as a function of temperature, T, forheating times of: 0 , 2 0 min ; ., 40 min.I0 20 40 60 80t/minFIG.2.4sothermal variations of the decomposed fraction, a, as a function of time, t, for tempera-turesof: +,161.5; *, 167.4; 0,170.5; A, 171.4; V,173.5; v, 175.0; X , 177.0; A, 179.0"CMACHADO, BRAGA, DA LUZ AND DUBLATRE 155induction time in the experimental isothermal curves, a process involving the growthof nuclei that would favour the decomposition of the chelate molecules is an a prioripossibility. Simple laws have been suggested for the early stages of decompositionon this basis,' sueh as :a = Ct", (2)where C is a constant at a given temperature and n an integer.t/rninFIG. 3.-Variations as a function of time, t, of (a) -In (1-a) at 175.OoC, (6) a* at: e, 167.4;0,170.5; v, 175.0; A, 179OC.Fig.3(b) displays some typical plots of af against t showing that eqn (2) wouldhold in the present case, with n = 3 and for values of a such as 0.03 < a < 0.5.The physical meaning of eqn (2) with n = 3 would be either a nucleation proceedingat a rather low rate with a two-dimensional growth of the nuclei or a rapid increaseof the nuclei number associated with a three-dimensional growth. Obviously,eqn (2) is only meaningful for small values of a. A further step was to use the Avramiequation based on random nucleation :a = 1-exp(-Kt)". (3)This equation was fitted to the experimental data using a least-square fitting procedurewith a resulting rz value of 3.9 kO.1 for all isothermal curves, except for the two lowertemperatures : n = 4.8 at 167.4"C and n = 2.1 at 161.5"C.However, for the lattertemperature the U-values are all below 0.12 and hence have a large relative error. Avalue of n = 4 is expected on the basis of the Avrami equation supposing a constantrate, k l , for spontaneous nucleation per unit transformed volume leading to a linea156 THERMAL DECOMPOSITION OF CHROMIUM(III) COMPLEXincrease with time of the number of nuclei and a cubic increase with time of the volumeof the growth nuclei at a linear rate k2. Fixing the n parameter value to 4 led to theK values given in table 1 and the resulting fitting curves were not significantly differentfrom the solid lines in fig. 2. The K parameter should be expressed as :K E klk$ (4)where both k , and k2 should obey the Arrhenius law.An Arrhenius plot of K isshown in fig. 4(a) and a least-square fitting led to :K(s-l) = (9.8 x 1015) exp (-165 kJ rnol-l/RT).-10.0,-.y* -1.1.0- -d --1 2.02.20 2.24 2.2 8 2.32FIG. 4.-Arrhenius plots for rate constants : (a) K; (b) ks ; (c) kl.The treatment of the experimental data in terms of the Avrami equation thuspoints to a random nucleation followed by a three-dimensional growth of the nucleias representative of the thermal decomposition process of Cr(BPHA)3. But thenature of the nuclei is not disclosed by the kinetics. Isotropic growth of the nucleishould allow visual observation of the nuclei, at least in samples where the decom-position is well advanced.The observation of samples using a platinum heateMACHADO, BRAGA, D A LUZ A N D DUPLATRE 157coupled to a microscope showed the formation of small liquid regions, which diffuseslowly until the whole system is in the liquid phase. The formation of these smallliquid regions is initiated at E 161°C. Very strikingly, 161°C is the melting point ofbenzanilide,13 the major stable product of the composition of Cr(BPHA),. Thisexperimental observation suggests that the nuclei may consist in a second phasebuilt up by the molecules of benzanilide. The autocatalytic process of decompositionof Cr(BPHA), could hence be interpreted in terms of the formation of a liquid phase,in proportion as the chelate molecules are being decomposed in the initial solid phase,that would dissolve the complex thus giving way to a faster rate of decomposition.TABLE VALUES OF KINETIC PARAMETER K FROM THE AVRAMI EQUATION, FOR n = 4temperature K/lO-$ s-l temperature K/lO-$ s-l/"C /"C161 .-5 1.30 173.5 4.40167.4 2.33 175.0 5.43170.5 3.02 177.0 6.12171.4 3.45 179.0 7.37A simple reaction scheme would be :ksA, + B+C .. .kiA~ --, B ~ C . . .where A, and Al represent the original molecules A in the solid and in the liquidphase, respectively, and B and C are the products.Supposing product B is responsible for the creation of the liquid phase, the varia-tions of A with time may be expressed, as suggested by Bawn,' by making use of themolecular solubility s :where A and B are the numbers of molecules of A and B, respectively.symbols A(t) etc., have been avoided for the sake of clarity] :Thus [theNow : Al + A , + B = A,, the initial number of undecomposed molecules, so :dA- = - k,[A, - B(s + l)] - k1sB.dtDividing by A , and noting that a = B/Ao = 1 - (A/&) :(7)dadt - = k,+&[k,s- k,(s+ l)]158Integrating eqn (8) leads to :THERMAL DECOMPOSITION OF CHROMIWM(III) COMPLEXwith D = k,s-k,(s+ 1).(10)Obviously, eqn (9) is only valid for A, > 0 ; as soon as A, = 0, a11 molecules of Aare dissolved :4 A s = - = -B B'The limiting value of a for eqn (9) is thus :and the time needed for the complete dissolution of the chelate is given byFor a > alim, only reaction (11) is occurring and :a = [-k1(t-tdI} +slim* (13)Eqn (9) and (13) have been least-square fitted to the data and led to a mean value ofs = 1.05+0.1 for all curves.Fixing the value of s as 1 led to the parameters dis-played in table 2 and the corresponding calculated curves axe drawn as solid lines infig. 2 for each temperature. Although the chi-square value was lower using eqn (9)and (13) than using Avrami quation, the resulting calculated curves are not signifi-cantly different.TABLE 2.-mETIC PARAMETERS DEDUCED USING EQN (9) AND (13) FOR S = 1161.5167.4170.5171.4173.5175.0177.0179.00.631.071.331.752.022.443.263.860.491.131.331.451.932.252.563.09The least-square fitting of the Arrhenius plots of the k, and kl parameters givenin fig.4(b) and (c), respectively, led to :k,(s-l) = (4.4 x 1015) exp (- 174 kJ mol-l/RT)k,(s-l) = (9.6 x 10l6) exp (- 169 kJ rnol-l/RT)MACHADO, BRAGA, DA LUZ AND DUPLATRE 159Although not common behaviour for inorganic or metal chelate compounds,thermal decomposition accompanied by the formation of a liquid phase and aconsequent increase in the reaction rate is not unusual in organic solids, as discussedby Bawn in ref. (7). Preliminary experiments show that if a small amount ofbenzanilide is mixed with the Cr(BPHA), crystallites prior to heating, a pronouncedacceleratory effect of the thermal decomposition is promoted. Furthermore, iftwo samples consisting of the chromium chelate and of pure benzanilide, respectively,are deposited on the microscope plate close to each other, it is observed that themelting of the benzanilide is followed by the immediate dissolution of the Cr(BPHA)3sample into the liquid phase.These facts lend support to the mechanism suggestedabove. However, the frequency factor related to kl, although certainly tainted with arather large error of very possibly more than an order of magnitude, would appear toohigh for a process in a liquid phase [but see Bawn in ref. (7)]. In addition, preliminaryresults on Fe(BPHA), indicate the formation of one molecule of benzanilide perdecomposed molecule of the chelate. If this were the case for Cr(BPHA)3, thevalue s = 1 found when fitting eqn (9) and (13) to the data would imply that a unitcell of the " liquid " phase would be defined by a single molecule of benzanilide.Inthis case, the definition of a " liquid " phase would hardly hold on a microscopicscale, at least in the initial stage of the decomposition. The physical situation mightrather be described as the molecule of benzanilide influencing the immediate sur-rounding undamaged molecules in such a way that one of the latter could decomposeat a higher rate than in a normal lattice site of the original solid.The important finding remains that some molecule(A)-molecule(B) interactionseems to be responsible for the acceleration of the decomposition as time proceeds.Supposjng the benmnilide molecules possess an important characteristic of liquidbehavisur, namely a high mobility, so that they can move freely in the lattice, theautocatalytic process may be described by the reaction scheme :Then :k3 A + B+C...k4 A+B 3 2BfC.. ..Dividing by A*, the initial number of undecomposed molecules, and noting thata = (B/A) = 1 -(A/&) leads toda - = k3(l -a)+k4Aoa(l -a). dtIntegrating eqn (15) gives :with CT = k3 +k4Ao.Fitting eqn (16) to the data led to the parameters recorded in table 3. Thecalculated curves were not very different from the solid lines in fig. 2, but the cor-responding chi-square values were higher than those found when using eqn (9) and(13). Further, the values at 167.4 and 1615°C for k3 and at 161.5"C for k4Ao ha160to be discarded in the Arrhenius plots (these are not reported). The least-squarefitting of these plots gave :k3(s-l) = (2.1 x 1O2O) exp (-200 kJ mol-l/BT)ksAo(s-l) = (8.4 x 1014) exp (- 149 kJ mol-lIRT).THBRM A L DE c o M P o SI TI o N o F c HR o MI UM ( I 11) c o MP LE xIn spite of the rather rough approximation made when supposing the benzanilidemolecules are free to move in the lattice even at the early stage of decomposition,the preceding treatment would tend to confirm the important role these moleculesplay in the acceleration of the decomposition rate.Owing to the vicinity of thebenzanilide fusion point in the temperature range where the decomposition occurs,the niolecules of benzanilide produced at the early stage of the decomposition mayprovide a larger free volume at the lattice site where they occur, thus favouring thedecomposition of neighbouring Cr(BPHA), molecules.TABLE 3.-KINETIC PARAMETERS DEDUCED USING EQN (16)ternperaturel'c k3/10-6 s-l k4Ao/10-3 s-l161.5167.4170.5171.4173.5175.0177.0179.05.922.055.137.857.909.9714.217.20.531.721.982.1 12.853.503.774.53Visual observatioii indicates that a genuine liquid phase is being formed.Sincethe latter two models used strongly suggest that the autocatalytic reaction proceedsvia a single benzanilide molecule, it is not excluded that this molecule does provide asufficient free volume for one neighbouring Cr(BPHA), molecule to decomposepractically at the same rate as it would in the genuine liquid phase. On this assump-tion, no change would be expected in the kinetics even if a liquid phase were indeedbeing built up by the benzanilide molecules.After all, this seems a likely possibilityin a system for which the molecular solubility would effectively be unity.Regarding the kinetics, the use of the concept of nucleation via the Avramiequation is undoubtedly quite attractive, owing to its large flexibility to account forthe variable kinetics that may be encountered with the various solids. However, inthe present case it seems that the nuclei effectively consist of molecules of one of theproducts of decomposition, the benzanilide.This investigation was supported by grants from Conselho de DesenvolvimentoCientifico e Tecnol6gico (CNPq) and Financiadora de Estudos e Projetos (FINEP).We thank both referees for their helpful remarks.S . C. Shome, Analyst, 1950,75,27.0. A. Vita, W. A. Levier and E. Litteral, Analyt. Chim. Ada, 1968, 42, 87.I. P. Alimarin, F. P. Sudakov and B. G. Golovkin, Russ. Chem. Reo., 1962, 31,466MACHADO, BRAGA, D A LUZ AND DUPLATRE 161A. D. Shendrikar, Talantu, 1969,16, 51.A. Syamal, J. prakt. Chem., 1970,312,954.R. A. Meyer, J. D. Hazel and Y. M. McNabb, Anulyt. Chim. Acta, 1964, 31, 419.Chemistry of the Solid State, ed. W . E. Garner (Butterworths, London, 1955).E. K. Gill and J. A. Morrison, in Annual Review of Physical Chemistry, ed. H. Eyring (AnnualReviews Inc., Palo Alto, 1963), vol. 14.D. A. Young, in Decomposition of Solids (Pergamon Press, Oxford, 1966).lo S. R. Yoganarasimhan, in Modern Aspects of Solid State Chemistry, ed. C . N. R. Rao (PlenumPress, New York, 1970).l1 F. C. Tompkins, Pure Appl. Chem., 1964,9,387.l2 M. Avrami, J. Chem. Phys., 1940,8,212.l3 Handbook of Chemistry and Physics, ed. C. D. Hodgman (Chemical Rubber Publishing Co.,Cleveland, Ohio, 43rd edn, 1961).(PAPER 9/184

ISSN:0300-9599    

DOI:10.1039/F19807600152

出版商:RSC    

年代:1980

数据来源: RSC

摘要:

J. C.S. Faraday I, 1980, 76,162-1 73Inorganic Photophysics in SolutionPart 4.-Deactivation Mechanisms of the 2Ea State of Crxsl Complexes from LifetimeStudies1BY STEPHEN R. ALLSOPP, ALAN Cox, TERENCE J. Urn* AND W. JOHN REEDDepartment of Chemistry and Molecular Sciences, University of Warwick,Coventry CV4 7AL, West MidlandsANDSILVANA SOSTERO AND OUZIO TRAVERSO1st it u to Chimico dell’ Univer siti di Fer rar a,Via L. Borsari 46, 44100 Ferrara, ItalyReceived 26th February, 1979The temperature dependences of the luminescence lifetimes, 7lWn, of several CrIII complexes ofparticular photochemical interest have been determined over temperature ranges 77-370 K in avariety of media, especially 9 mol dmm3 LiCl+ HzO and cellulose acetate film. In most cases, kl,,fits an expression k ~ 1 + A m exp( -AEzR/RT) where the subscripts refer to temperature-independentand (temperature-dependent) non-radiative terms. The strong solvent-dependence of AE&, whichis always only a fraction of the “TzB.;Eg splitting, suggests a chemical pathway for deactivation ofthe 2E’ state rather than a purely physical route such as back-intersystem crossing. For complexeswhich have been examined in a wide variety of media, values of In ANR correlate fairly well withthose of AEzR, implying a common overall mechanism.~One of the persisting themes of inorganic photochemistry for over a decade hasbeen the identification of the photoreactive state of octahedral CrlI1 complexes.2 Byanalogy with the situation established for polyacenes, initial discussion centred 2 *around the weakly absorbing (and in many instances luminiscent) 2Eg state which wasknown (i) to be of longer lifetime than the lowest quartet state, 4T2g, and (ii) to bepopulated by intersystem crossing from the latter (fig.1). Later approaches 2 *noted that additives capable of quenching phosphorescence from the 2Eg level ofcertain CrII’ complexes did not always quench their photochemical activity, whilstdirect population of 2Eg states either by energy transfer 2 b s or by laser photolysis6did not result in the same quantitative photochemistry realised by direct irradiation intothe quartet states and opinion moved generally in favour of the 4T2g state as the levelresponsible for Cr”’ photochemistry.2 More recent work has re-established aphotochemical role for 2Eg states, although it remains to be resolved whether this ismerely as a source, through back-intersystem crossing, of (reactive) 4T2g states or as agenuine photoreactive state per se.Thus Maestri et al. 8a suggest that 97 % of thephotoaquation of [Cr(bi~y)~]~+ (bipy = 2,2’-bipyridine) proceeds directly through the2Eg state, whilst Sandrini et aLSb propose that the photoaquation of trans-[Cr(en),(NCS),]+ involves two routes, i.e. 20% directly through 4T2g and 80% byback-intersystem crossing into 4T2g from the 2Eg level. Another recent view is thatof regarding the 4T2g and 2Eg states as separate minima on a continuous potentialsurface. b* 8 b The Jablonski diagram for CrIII is further complicated by the existenceof several cases of CrlllJluorescence, e.g.in [Cr(urea),13+ ion, which exhibit very largeStokes shifts, typically of the order of several thousand cm-l, with emission maxima16ALLSOPB, COX, KEMP, REED, SOSTERO, TRAVERSO 163of considerably longer wavelengths than the phosphorescence, indicating the emittingstate of the 4T2g level to be greatly distorted from its initial Frank4ondon geometryand to be at a lower energy than the 2Eg level. (The appearance of fluorescence wasassociated later with a small 4T2e-2EcI splitting).9d This situation prompted thedesignation of " thermally relaxed excited " or " thexi " state for the fluorescent4T2g level.2e-photoreactionFIG. l.-Jablonski dia,gram for a CrIn complex of octahedral symmetry.ISC refers fo irrtprsysterncrossing ; BISC to back-intersystem crossing.Photophysical investigations of CrlI1 complexes have reflected the principalphotochemical problem, i.e. is luminescence from the 2Eg state quenched by (i) aphotochemical mechanism, (ii) by back-intersystem crossing, O (iii) by direct energytransfer into vibrational modes of the ligands or the solvent l1 or (iv) by some com-bination of (i)-(iii) which varies from complex to complex ? Methods of investigationof this problem have included measurements both of 4p and 7p and of their temperatureactivation (particularly of crystalline,1° but also of glassy and fluid sarnples),l2 ofjudicious changes in the ligand and solvent environments of the complex,ll* ofintersystem crossing rates and efficiencies1 and of lifetime dependence on excitation~avelength.'~ The general picture appears to be one where 2Eg states lose theirenergy by (a) a temperature-independent radiative pathway (possibly coupled with atemperature-independent non-radiative pathway) and (b) a temperatursdependentnon-radiative pathway .In this account we present data for 2Eg state lifetimes for several key CrlI1 com-plexes in (i) fluid and glassy aqueous solutions (9 mol dm-3 LiCl) over a wider tem-perature range than investigated hitherto,l (ii) organic solvents covering a wide rangeof polarity, (iii) polymer film.We have also analysed the data of Pfeil l6 into tern-perature-independent (TI) and temperature-dependent non-radiative (NR) compon-ents and have compared the results for AE& (in the total expression164 INORGANIC PHOTOPHYSICS I N SOLUTIONwith those given for CrXI1 complexes in crystalline and glassy alkanolic xnedia,loa*enabling a discussion of the character of the non-radiative processes.EXPERIMENTALLIFETIME MEASUREMENTSThese were determined by monitoring emission from solutions or glasses of the Cr"'complexes following delivery of a 25 ns pulse of 347 nm radiation.The means of tempera-ture control, measurement and processing of individual kinetic runs and of iterated fittingto eqn (1) have been given bef0re.l' A new frequency doubling crystal of rubidium dihydro-gen arsenate was employed.MATERIALSWater was multiply distilled from alkaline K[Mn04]. Organic solvents were of spectro-scopic grade.CrlI1 complexes were prepared by conventional routes. In the case of[Cr(phen),][ClO,J,, the very long room-temperature lifetime in deaerated aqueous solution,2 7 0 ~ s ~ which compares with a figure of 2 7 0 ~ s due to Serpone and Bolletta,12g could beachieved only by repeated recrystallisation from water combined with chloroform extractionof organic impurities.ABBREVIATIONSThese are as follows : bipy = 2,2'-bipyridine, phen = 1,lO-phenanthroline, acac = acetyl-acetone, en = 1 ,Zdiaminoethane, glycol = 1 ,2-dihydroxyethaneY terpy = 2,2',2"-ter-pyridine, DMF = dimethylformamide, tn = 1 ,=l-diaminopropane (often called trismethylene-diamine), CA = cellulose acetate.RESULTS AND DISCUSSIONFORM OF TEMPERATURE DEPENDENCEFor all complexes we have examined, irrespective of the solvent medium, plots ofIn klum (the luminescence decay constant) against T-l display two distinct regions,illustrated by fig.2(a)-(c). klum barely increases in the temperature range 77-200 K,4 6 8 10 12103 KITFIG. ALLSOPP, COX, KEMP, REED, SOSTERO, TRAVERSO 165t-121806lo3 KITlo3 KITFIG. 2.-Temperature dependence of luminescence lifetime of CrIII complexes. x , experimentalpoints ; full line, computer fit to eqn (1) (for individual parameters see table 1). (a) [Cr(en)J3+ in9 mol dm-3 LiCl + H20 ; (b) [Cr(phen)J3+ in cellulose acetate film ; (c) [Cr(ter~y)~]~+ in 9 mol dm-3LiCl+ H20.but at higher temperatures normal Arrhenius behaviour is found leading to an activa-tion energy, AE&, and a corresponding frequency factor, ANR.In some cases alldata points could be iterated by computer to an expression of the form described byeqn (l), yielding values for kTI, AE& and AN& In a few cases the iteration was un-successful : in some of these this is probably due to the presence of a phase transitionwithin the .matrix, but it was nonetheless possible to discern two principal terms of thetype described in eqn (1). Magnitudes of the three parameters are collated in table 1,together with those given elsewhere in the literature for the complexes we havTABLE 1 .-ACTIVATION PARAMETBRS FOR THERMAL DEACTIVATION OF LUMINESCENCE OF C r I n COMPLEXESLINE HOSTScomplex mediuM1 o-2kn/s-' 10-lodpq&-(1.09&0.88)x lo2- --(1.1 10 & 0.002) x lo2(9.18k0.09)~ 10'i1.87+ 0.04--1.05 x2 .9 ~ 103(1.3$0.7JX ---1.05 X(1.41+1.1)X2.1 x 1-0.36O.lS+0.05-ALLSOPP, COX, KEMP, REED, SOSTERO, TRAVERSO 167Y". wv,t 2Xn0.4O N rl- 0 0Xn I I3> 3 2 e B en3 UL4TABLE 1 .-continuedcomplex10-lOAm/~-medium 10-2k+-1MeOH+ H20+ glycol 2.62+ 0.10[Al(acac)31 (2.30k0.01) x lo1crystalline (7.82k 0.76) x lo150 % in [Al(a~ac)~] (4.27k0.40) x lo110 % in [Al(acac)J (2.2320.06)~ lo10.1 % in [Al(a~ac)~] (2.34k0.05) x lo1EtOH (2.1740.09) x lo1n-PrOH (2.04+ 0.24) x lo1n-BuOH (2.1040.08) x lo1n-C5H90H (2.07k0.13) x lo1n-C6H1 1OH (2.04k0.15)~ lo1n-C7H130H (2.28k0.06) x lo1n-C8H150H (2.1720.09)~ lo1isopentane + 3-methylpentane (2.31 k0.05) x lo1poly(methy1 methacrylate) (2.17k0.02) x 10'MeOH + H20 + glycolformamidef MeOH[for many other solvents, see ref.12(e)][Cr(terpy> 213+ HzO+ LiCl (1.7 +O.l)x lo1CA film (2.0520.09) x 10'5 % in A1Cl3.6D20 (2.36k0.62)~ lo1Cr3+ 2 % in NaMgAI(C20&.9H20 (1.16_+0.04)~ lo15 % in K3[Co(CN)6] (8.2010.35)~ether+isopentane+ alcohol -glycerin+ H2O - 2.88k0.8- [Cr(CN) 61 3-(8.26+0.90)(2.54k0.47)(1.54+ 3.33)(1.39k0.99)~(1.36k0.97)(7.70+ 1.43)(4.1 k 4 . 8(8.1 k 6.8)(8.2 & 13.9)(4.2 4 8 . 6(9.2 k 4 . 2(4.8 k 0 . 9(3.16+ 0.76)14.1 f1.4(2.1 + 1 . 1(7.8 2 8 . 8(3.0 k0.8)(1.7650.97)~(1.69k0.63)~(8.88k2.82)(4.03 k 4.74)(7.372 1.99)Adapted from ref.12(c) except for [Cr(CN)$'-, which was calculated from data of ref 12(e). b GraphicaALLSOPP, COX, KEMP, REED, SOSTERO, TRAVERSO 169examined (or their close analogues). We have also extracted activation parametersfrom the luminiscence lifetime data of Pfeil16 on a number of CrlI1 complexes insolution over fairly extended temperature ranges, which also reveal good fitting toeqn (1). Finally, we have fitted the data of Forster et aLIOa* on [Cr(acac),] lumines-cence in crystalline hosts, alkanolic media and in polymer film to eqn (1) to yieldvalues for kTI, ANR and AE& (table 1). Forster et preferred a more complexexpression to fit the temperature dependence of their data, i.e.but we have found that these data can be reasonably well-fitted by iteration to eqn (1),although the comparatively few data points associated with each temperature study(typically 12 to 15) naturally introduces a considerable standard deviation, particularlyin ANR of eqn (1).The lifetime of [Cr(ter~y)~]~+ is remarkably short at 300 K(z = 180 ns) compared with those of [Cr(bipy),I3+ (87.8 ps) and [Cr(phen)J2+(270 ps). This influence of the terpy ligand parallels that found for the analogousRu" complexes, thus 7298K for [Ru(bipy),12+ and [Ru(phen),12+ is 612 and 1280 ns,respectively, whilst [R~(terpy)~]~+ is essentially non-luminescent * (all data refer toaqueous 9 mol dm-, LiCl solution).MAGNITUDES OF ANR A N D AE&For a given CrIX1 complex, the values of Am and AE& can cover a wide range. Forexample, AE& (in kJ mol-l) for [Cr(phen),13+ varies from 17.962 1.2 in H20lZc and12.4kl.O in MeOH to 68.8f2.1 in cellulose acetate film.However, the latter verylarge energy requirement is compensated by NN 1O6-fold larger A factor and the variousdata for [Cr(phen),I3+ fit a Barclay-Butler plot [fig. 3(a)] reasonably well with a cor-relation coefficient of 0.994. A fair correlation (of Coefficient 0.927) between In ANRand AE$ is also exhibited by the activation data for [Cr(acac),] in 16 environments ofwidely differing character [fig. 3(b)]. Finally, the data of Wasgestian and coworkers1 2efor [Cr(CN)6]3- luminescence in a variety of organic solvents [fig. 3(c)] also show thissame correlation, but less exactly (correlation coefficient = 0.722). This type ofapproximate fit has been adduced l5 as evidence of a common mechanism for a seriesof reactions displaying a range of AE* values and has been used by Adamson 5a in adiscussion of the phosphorescence of [Cr(NH,), (NCS),]- in 12 solvents : in this casethe slope is 0.41 mol kJ-l, in good agreement with those of fig.3 (see legend).MECHANISM OF NON-RADIATIVE ENERGY-LOSS FROM 2Eg LEVELThe principal findings of this study are : (i) the (temperature-dependent) non-radiative energy-loss mechanism for each complex is adequately described by a singleenergy term and therefore may well refer to a single process ; (ii) the activation energyterm for this single process is extremely sensitive to environment for certain complexes,e.g. those of phen, acac and terpy ; (iii) despite (ij), good Barclay-Butler plots (for datacovering all media) are given in these particular cases, indicating the probability of acommon mechanism; (iv) (as noted before 12ai c* the magnitude of AE,f, is invari-ably only a fraction of the spectroscopic splitting of the 4T2g and 2Eg levels (table 1).Since the locations of the spectroscopic 4T2g and 2Eg levels for a given C Pcomplex are not noticeably solvent-sensitive the splitting of these levels is similarlyindependent of environment.Consequently it is to be expected that a mechanism of2Eg quenching by back-intersystem crossing, even though this may involve an activa-tion barrier (to a cross-over point) greater than the spectroscopic splitting energy(typically of magnitude 12c 5000-10 008 cm-1 or 50-100 kJ mol-I), should not sho170 INORGANIC PHOTOPHYSICS I N SOLUTIONCA filmL i C I - H0 20 40 60 80AE&/kJ mol-l7 "/013I I 1 I I I I I5 10 15 20 25 30 35 40A%*R/kJ mol-1FIG.ALLSOPP, COX, KEMP, REED, SOSTERO, TRAVERSO 1711012I I I 1 I20 25 30 35 40AE&/kJ mol-lFIG. 3.-Barclay-Butler plots for the Arrhenius parameter for the non-radiative decay term for CrIII(a) Data for [Cr(phen),]’* (slope = 0.32 k 0.02 mof kJ-’).(b) Data for [Cr(acac)J (slope = 0.48 k0.05 mol kJ-l). Key : I, 100 % crystal ; 2, 50 % infAl(acac)3]; 3, 10 % in [Al(acac)J; 4, 0.1 % in [AI(acac),]; 5, EtOH; 6, n-PrOH; 7, n-BuOH;8, n-pentanol ; 9, n-hexanor ; 10, n-heptanol ; 11, n-octanol ; 12, ether+isopentane+alcohol ; 13,isopentane+ 3-methylpentane ; 14, poly(methy1 methcrylate) ; 15, MeOH + H20 +glycol ; 16,[Al(aca~).~].(c) Data for [Cr(CN),I3- (slope = 0.52 0.2 mol kJ-l).Key : 1, DMF ; 2, MeCN ; 3, i-PrCN ;4, PhCN ; 5, i-PrOH ; 6, n-PrOH ; 7, EtOH ; 8, EtOD ; 9, MeOH ; 10, MeOD ; 11, methyrformamide ;12, fomamide.complexes in various media taken from table 1.much solvent dependence on AE*. This situation should also apply even if back-intersystem crossing occurs to a cross-over point 12a between the ”Eg level and a“ thermally relaxed ” 4T2g level, which would account for the s d l vdues of AE&.The large variation with solvent of AE& found in this work, therefore, points to theoperation of some other mechanism which embraces cmsjtderable partieiptkm bythe solvent atmosphere of the complex. One eminently suitable candidate must bea mechanism of ligand dissociation-return of the typ172 INORGANIC PHOTOPHYSICS I N SOLUTIONThis would accommodate those extremely large values for AE& sometimes obtainedwith cellulose acetate film as host, in addition to the low photodecomposition quantumyields found with some CrIrl complexes under conditions where they luminescestrongly. An alternative mechanism which would reflect solvent character is that ofcoupling of the electronic excitation energy to excited vibrational modes of the solvent :however, the magnitudes of solvent isotope effects (klum(NzO) /klUmcD, em too smallfor this to be a dominant pathway and it is not clear why such a 1 should haveactivation energies of the magnitude found.We thank the S.R.C.for support of S. R. A. throughapost-doctoralresearchassist-antship, W. J. R. through a post-graduate studentship and for a grant to purchase thelaser system. We acknowledge support from NATO for the Warwick-Ferraraexchange and from the Royal Society for a grant to purchase the RDA crystal.Finally, we thank Mr. R. H. Frowen for help with preliminary studies of the[Cr(bi~y)~]~+ system and Prof. V. Carassiti for valuable comments.Part 3. S . R.-Allsopp, A. Cox, T. J. Kemp, W. J. Reed, V. Carassiti and 0. Traverso, J.C.S.Faraday I, 1979,75,353.(a) G. B. Porter, in Concepts of Inorganic Photochemistry, ed. A. W. Adamson and P. D,Fleischauer (Wiley, New York, 1975), chap. 2 ; (6) E.Zinato, ref. (a), chap. 4 ; (c) A. W.Adamson, Pure Appl. Chem., 1970, 24, 451 ; ( d ) A. D. Kirk, Mol. Photochem., 1973, 5, 127 ;(e) P. D. Fleischauer, A. W. Adamson and G. Sartori, Prog. Inorg. Chem., 1972, 17, 1 ;H. Schlafer, 2. Chem., 1970, 10, 9 ; cf) V. Balzani and V. Carassiti, Photochemistry of Co-ordination Compounds (Academic Press, New York, 1970), chap. 7.(a) H. L. Schafer, J. Phys. Chem., 1965,69,2201; (b) R. A. Plane and J. P. Hunt, J. Amer. Chem.SOC., 1957, 79, 3343.(a) H. F. Wasgestian, J. Phys. Chem., 1972,76,1947; (6) S. N. Chen and G. B. Porter, Chem.Phys. Letters, 1970, 6, 41 ; (c) C. H. Langford and L. Tipping, Canad. J. Chem., 1972,50, 887.(a) V. Balzani, R. Ballardini, M. T. Gandolfi and L. Moggi, J. Amer. Chern. SOC., 1971, 93,339; (b) A.W. Adamson, J. E. Martin and F. D. Camassei, J. Amer. Chem. Suc., 1969, 91,7530 ; (c) J. E. Martin and A. W . Adamson, Theor. Chim. Acta, 1971,20,119.C. H. Langford and C. P. J. Vuik, J. Amer. Chem. SOC., 1976,98, 5409. ' (a) M. Maestri, F. Bolletta, L. Moggi, V. Balzani, M. S . Henry and M. Z. Hoffman, J.C.S.Chem. Comm., 1977, 491 ; (b) N. A. P. Kane-Maguire, D. E. Richardson and C. G. Toney,J. Amer. Chem. Soc., 1976,98, 3996; (c) R. Ballardini, G. Varani, H. F. Wasgestian, L. Moggiand V. Balzani, J. Phys. Chem., 1973,77,2947 ; ( d ) N. A. P. Kane-Maguire and C . H. Langford,J. Amer. Chem. Sac., 1972,94, 2125. * (a) M. Maestri, F. Bolletta, L. Moggi, V. Balzani, M. S. Henry and M. Z . Hoffman, J. Amer.Chem. Soc., 1978,100,2694 ; (b) D.Sandrini, M. Gandolfi, L. Moggi and V. Balzani, J. Amer.Chem. Soc., 1978,100,1463.(a) G. B. Porter and H. L. Schlafer, Zphys. Chem. (Frankfurt), 1963,37,109 and 1964,40,280 ;(b) D. M. Klassen and H. L. Schlafer, Ber. Bunsenges. Phys. Chem., 1968,72, 663 ; (c) W. M.Watson, Y. Wang, J. T. Yardley and G. A. Stucky, Inorg. Chem., 1975, 14, 2374; ( d ) H. L.Schlafer, H. Gausmann and H. Witzke, J. Chem. Phys., 1967, 46,1423.lo (a) W. Targos and L. S . Forster, J. Chem. Phys., 1966,44,4342 ; (b) F. D. Camassei and L. S .Forster, J. Chem. Phys., 1969,50,2603 ; (c) F. Castelli and L. S . Forster, J. Amer. Chem. SOC.,1975,97,6306 ; ( d ) F. Castelli and L. S . Forster, 9. Phys. Chem., 1977,81,403 ; (e) L. S . Forster,in Inorganic Compounds with Unusual Properties, Adv. Chem. Ser., 1976,150, 172.l1 (a) M. S . Henry, J. Amer. Chem. Soc., 1977, 99, 6138; (6) N. Serpone, M. A. Jamieson, M. S .Henry, M. 2. Hoffman, F. Bolletta and M. Maestri, J. Amer. Chem. SOC,. 1979, 101, 2907,l2 (a) H. E. Schlafer, H. Gausmann and H. Witzke, 2. phys. Chem. (Frankfirt), 1967, 56, 55 ;(b) J. T. Yardley and J. K. Beattie, J. Amer. Chem. Soc., 1972, 94, 8925 ; (c) N. A. P. Kane-Maguire and C. H. Langford, J.C.S. Chem. Cumm., 1971, 895 ; (d) T. Ohno and S . Kato,Bull. Chem. SOC. Japan, 1970, 43, 8 ; (e) R. Dannohl-Fickler, H. Kelm and F. Wasgestian,J. Luminescence, 1975,10, 103 ; (f) A. W. Adamson, C. Geosling, R. Pribush and R. Wright,Inorg. Chim. Acta, 1976, 16, L5 ; (9) N. Serpone and F. Bolletta, quoted by V. Balzani, F.Bolletta, M. T. Gandolfi and M. Maestri, Topics in Current Chemistry, 1978, 75, 1ALLSOPP, COX, KEMP, REED, SOSTERO, TRAVERSO 173l3 (a) F. Bolletta, M. Maestri and V. Balzani, J. Phys. Chem., 1976, 80, 2499; (b) A. D. Kirk,P. E. Hoggard, G. B. Porter, M. G. Rockley and M. W. Windsor, Chem. Phys. Letters, 1976,37, 199.(a) A. R. Gutierrez and A. W. Adamson, J. Phys. Chem., 1978,82,902 ; (6) R. T. Walters andA. W. Adamson, Acta Chem. Scmd. Ser. A, 1979 A33,53; (c) A. W. Adamson, Pure. Appl.Chem., 1979, 51, 313.R. C. Young, J. K. Nagle, T. J. Meyer and D. G. Whitten, J. Amer. Chem. SOC., 1978,100,4773.l4 C. Conti and L. S. Forster, J. Amer. Chem. SOC., 1977, 99, 613.l6 A. Pfeil, J. Phys. Chem., 1971, 93, 5395.l7 S. R. Allsopp, A. Cox, T. J. Kemp and W. J. Reed, J.C.S. Faraday I, 1978,74, 1275.(PAPER 9 131 3

ISSN:0300-9599    

DOI:10.1039/F19807600162

出版商:RSC    

年代:1980

数据来源: RSC

摘要:

J. C.S. Faraday I, 1980,76, 174-1 79Enthalpy of Mixing of Liquid Hydrogen Chloride andLiquid XenonComparison of Experiment and TheoryBY L~LIO Q. Loao AND LIONEL A. K. STAVELEY"Inorganic Chemistry Laboratory, Oxford University, Oxford OX1 3QRANDPAULETTE CLANCY AND KEITH E. GUBBINSSchool of Chemical Engineering, Cornell University, Ithaca, New York 14853,U.S.A.Received 27th February, 1979Calorimetric measurements are reported for the enthalpy of mixing HE of liquid hydrogenchloride and liquid xenon at 182.32K. These results are compared with values calculated fromperturbation theory, using two slightly different sets of intermolecular potential models for theXe+HCI system. Although these models give good results for the excess Gibbs energy GE andexcess volume VE, they are found to give values that are too high for HE.We have recently compared the measured values of the excess Gibbs energyGE and the volume of mixing VE for the liquid systems Xe + HCl and Xe+ HBr withthe values calculated by the perturbation theory developed by Gubbins and his co-w0rkers.l Each of these two systems is, of course, a mixture of non-polar moleculesand small, polar molecules and they were selected for experimental study of theirexcess functions as being suitable model systems on which to test theories of mixtureswhich take account of polarity.Since reporting this work, we have measuredcalorimetrically the enthalpy of mixing HE for the Xe + HC1 system. It is now recog-nized that successful prediction of this excess function is a more exacting test of atheory than the prediction of GE.We record here our experimental results for HEand compare them with the values estimated on the basis of the perturbation theoryused in the previous calculations of GE and VE.EXPERIMENTALThe measurements of HE were made with the calorimeter described by Lewis andStaveley.2 The results are presented in table 1 in the form used in earlier papers,2s to makeapparent the relative importance of the correction for the evaporation of the liquids into thegaseous phase in the mixing vessel. (There are other minor corrections, not recorded intable 1, which have to be considered, such as that for the non-ideality of the two componentsin the vapour phase.)The thermodynamic study of mixtures involving hydrogen chloride as one component ishandicapped by our present rather imperfect knowledge of the second virial coefficient B ofthis compound at low temperatures and by the almost complete lack of any experimentalinforniation on the second virial coefficient of binary mixtures of hydrogen chloride andother gases.In the measurement of HE it is necessary to know these quantities to makecorrect estimates of the amount of material in the vapour phase in the mixing vessel, as well17LOBO, STAVELEY, CLANCY AND GUBBINS 175as to allow for the pressure dependence of the internal energy of this gaseous material.Fortunately, for the particular system under discussion the uncertainty arising from this causeis not large, as the vapour pressures at which HE were determined were comparatively low.A reasonably self-consistent set of experimental values of B for hydrogen chloride is providedby the results of Prof.B. Schramm at 200, 250 and 295 K and the values at 328.7 and368.7 K of Glockler et aL5 By extrapolation, we estimated B(HC1) = -490 cm3 mol-l atour working temperature of 182.26 K.B(Xe) was taken to be - 334 cm3 mol-1 at this temperature from the work of Brewer.6We assumed the cross virial coefficient B12 to be the arithmetic mean of the values for thetwo pure components. Using these estimates of second virial coefficients, we obtained thevalues of HE given in table 1. Calculation of B(HC1) using the intermolecular potential forthis substance on which the calculation of GE and VE was based had given the numericallylower value at 182.26 K of -293 cm3 mol-l.l Use of this value of B(HC1) in the processingof our raw calorimetric results gave slightly higher values for HE, but for no mixture did thisincrease exceed 4 3 mol-l.The sources and purities of the two gases were the same as those of the samples used inthe earlier work on GE and VE.7TABLE 1 .-EXPERIMENTAL EXCESS ENTHALPIES FOR Xe+ HCl AT (182.262 0.05)K.Q DENOTESVAPORIZATION CORRECTION, (nl+n2) THE AMOUNT OF SUBSTANCE OF THE LIQUID MIXTURETHE ELECTRICAL ENERGY REQUIRED TO COMPENSATE FOR THE MIXING PROCESS, Av,,, THEAND X TEE3 MOLE FRACTION OF HC1. @(ps) IS TEIE MOLAR EXCESS ENTHALPY AT THE SATUR-ATION VAPOUR PRESWRE ps OF TIIE MIXTURE AND HE(p = 0) T’HE MOLAR EXCESS ENTHALPY ATZERO PRESSURE.A u,, n1-I- a2 HE@& HE@= 0)/J IJ /mol X /J mol-l /J mo1-lQ53.397 - 5.772 0.06559 0.3201 730.0 729.872.264 - 4.824 0.08361 0.4351 809.9 809.755.328 - 5.108 0.061 70 0.4971 822.2 822.058.241 - 4.938 0.06601 0.5081 811.5 811.362.758 - 3.994 0.07832 0.5902 753.7 753.640.579 - 3.522 0 .a 5 870 0.6%8 635.3 635-1RESULTSThe experimental values of HE ( p = 0) were fitted to a Redkkh-Kiskr quationHE/RT = ~ ( 1 - X) [ A + B(2x - 1) + C(2x - l)”] (1)with A = 2.1465, B = -0.3182 and C = -0.3595; the stamlard deviation beinga(HE) = k6.9 J mol-I.By combining this equation with that in our previous paper7 for GE, we obtain forthe excess entropy SEIn both equations x is the mole fraction of HCl in the liquid phase.The experimental results are plotted against composition in fig.1 , which shows therelation between the excess functions HE, GE and TSE. Too much reliance should notbe placed on the shape of the TSE curve at IQW and high mole fractions of HC1, sincethis curve is sensitive to the precise shape of the HE and GE curves in these regions,Nevertheless, there is no doubt that for this system TSE is relatively small in cornpari-son with HE and GE and over most of the composition range, at least, it is positive.SEJR = ~ ( 1 - X) [0.3368 - 0.2089 ( 2 ~ - 1) - 0.5575 ( 2 ~ - I)’]. (2176 HE FOR HCl(1) AND Xe(1)8000I I I0 '3.2 0.4 0.6 0.8 LOXHClFIG. 1.-Dependence on the hydrogen chloride mole fraction x of the molar excess Gibbs energy(GE) at 182.32 K, of the molar enthalpy of mixing (HE) at zero pressure at 182.26 K [the circles are theexperimental points and the curve is given by eqn (l)] and of TSE where SE is the molar excessentropy.Our value of HE for the equirnolar mixture is 813.216.9 J mol-l. Caladoet a1.' measured GE at 159.07, 182.26 and 195.42 K.Their results give a mem HEfor the equimolar mixture over this temperature range of 880 80 IT mol-l, in reason-able agreement with the more accurate calorimetric estimate.COMPARISON WITH THEORYThe theoretical approach 1* 8* is based on perturbation theory and involves anexpansion of the Helmholtz free energy A for the real system, in which the pairpotential is uaB(y0,w2), about the free energy of a reference system of sphericalmolecules, in which the pair potential is u&(r).Here Y is the vector from molecule 1(of component a) to molecule 2 (component p) and cot for linear molecules)represents the orientation of molecule i. The reference potential uZB is defined to bean unweighted average over the orientations of the full potential uas. With thischoice of reference the first-order term Al vanishes and the series to: third-order is usedas the basis for a simple Pad6 approximant,The detailed expressions for the second- and third-order terms A2 and A3 are givenin the previous papers.l* 8* Eqn (3) has been compared with computer simulationresults for dipole-dipole, quadrupole-quadrupole and anisotropic overlap potentialsof the type used here and found to give good agreement (within ~4 % for both thefree energy and configurational internal energy).Comparisons of theory and experi-ment should therefore provide a test of the intermolecular potential models used.of the system Xe + HC1 we have used two slightly differentmodel potentials, which we here refer to as models A and B. In both models theIn previous studies lpotentials are :Model AXe-Xe :HCl-HCl :Xe-HCl :Model B9Xe-Xe :HCl-HC1 :LOBO, STAVELEY, CLANCY AND GUBBINS 177central term is an (n, 6) potential, while the HC1-HCl interaction includes terms due todipole-dipole, dipole-quadrupole and quadrupole-quadrupole interactions and theXe-HC1 interaction includes two-body isotropic induction terms. Model B differsfrom A in the addition of anisotropic dispersion and overlap terms.These modelXe-HCl :Pu these equations u("s6) is the (n, 6) potential, u P p . . . uQQ are the dipole-dipole, . . ,quadrupole-quadrupole potentials, uo ind is the isotropic part of the induction poten-tial (dipole-induced dipole-dipole and quadrupole-induced dipole-quadrupole two-body terms), udis (101) and udis (011) are the leading anisotropic dispersion terms in anexpansion in spherical harmonics and uov (101) and uov (01 1) are the leading harmonicterms for the anisotropic overlap potential. Detailed expressions for these potentialsaxe given in the original papers. Multipole moments, the polarizability of xenon andthe hyperpolarizability of HC1 were taken from independent experimental measure-ments. The remaining like-pair potential parameters (6, 0, n, 6) were obtained bycomparing theoretical and experimental properties of the orthobaric liquids.Theunlike-pair interactions CXeHCI, YXeHCl and 8keHCl were obtained by comparing experi-mental and theoretical results for GE and VE. Here cap and yab are defined byThe parameter nXeHC) is obtained from the geometric mean rule.' Values of theparameters used are given in the previous papers.l* However, the values given for cXeHC1 and qXeHC1 for model B in table 4 of Clancy et al.' were incorrect ; the correctvalues are 0.986 and 0.989, respectively.A comparison of the experimental and theoretical HE curves for these two modelpotentials is shown in fig. 2. Curves A and B give the results for potentials A and B,using for the reference system properties the procedures described in previouspapers.l* This involves relating the reference system properties to those of a pureLennard-Jones (LJ) fluid and using the argon equation of state (in dimensionlessform) due to Gosman et al.1° for the LJ properties. Both models A and B give HEvalues that are substantially higher than experiment, model A being high by a factorof s32 and model B by w63 %.Models A and B both give a good description of GE and VE, the addition of aniso-tropic dispersion and overlap terms in model B yielding only a marginal improvementin VE.By contrast, these potential terms have a large effect on HE. The discrepan-cies in fig. 2 may be due to a variety of simplifications in the form of the potential.Forthe HCl-HCI model these include the omission of all induction terms and imprecis178 HE FOR HCl(1) AND Xe(1)descriptions of the direct electrostatic, dispersion and overlap potentials. Thepresent paucity of thermodynamic data for HCl precludes a meaningful test of thesevarious models ; more extensive studies of the pure liquid would be particularly useful.For Xe-HCl we have omitted multibody induction terms as well as highcr order dis-persion and overlap terms. Both theoretical l 1 and computer simulation l2 resultshave shown recently that multibody induction terms are important in liquids. In arecent theory due to Wertheim these effects may be estimated by replacing the truela2/I I I I0 0.2 0.4 0.6 0.8 1 .oXHClFIG. 2.-Comparison of the theory for potential models A and B (lines) with experiment (points).(gas-phase) dipole moment p by a renormalized (liquid-phase) value p’ that takes intoaccount these multibody terms.The theory provides a prescription for calculating p’in a purely dipolar liquid (ie., no quadrupole or higher poles). For pure HCl at thestate condition considered here, Wertheim’s theory gives p ’ / p z 1.28 if the effect ofthe quadrupoles is neglected. If this renormalized dipole moment is used in modelB (the other parameters being readjusted to fit the pure liquid HCl and GE and VE forthe mixture) the predicted value of Hf is lowered by ~6 %. This calculation isvery rough in that it neglects effects of the quadrupole in HCl and also of multibodyinduction effects in the Xe-HC1 interaction.We are in the process of investigating themultibody induction effects further.The participation of one of us (L. Q. L.) in this work was made possible by a Fellow-ship from C.P. Invotan (J.N.I.C.T./Portugal). P. C . and K. E. G. thank the NationalScience Foundation and the Petroleum Research Fund, administered by the AmericanChemical Society, for support of this workLOBO, STAVELEY, CLANCY AND GUBBINS 179J. C. G. Calado, C. G. Gray, K. E. Gubbins, A. M. F. Palavra, V. A. M. Soares, L. A. K.Staveley and C. H. Twu, J.C.S. Faraday I, 1978,74, 893.K. L. Lewis and L. A. K. Staveley, J. Chem. Thermodynamics, 1975, 7, 855. (See also K. L.Lewis and L. A. K. Staveley, J. Phys. E, 1975, 8, 811.)K. L. Lewis, G. Saville and L. A. K. Staveley, J. Chem. Thermodynamics, 1975,7,389.B. Schramm, personal communication, 1977.G. Glockler, C. P. Roe and D. L. Fuller, J. Chem. Phys., 1933, 1, 703.J. Brewer, Determinutian of Mixed Virial Coeficients, 1967, U.S.-AFOSR No. 67-2795.J. C. G. Calado, A. F. Kozdon, P. J. Morris, M. N. da Ponte, L. A. K. Staveley and L. A.Woolf, J.C.S. Faraday I, 1975, 71, 1372.For a review of the theory see : K. E. Gubbins and C. H. Twu, Ckm. Eng. Sci., 1978,33, 863,879; C. G. Gray, K. E. Gubbins and C. H. Twu, J. Chem. Phys., 1978,69,182.P. Clancy, K. E. Gubhins and C . G. Gray, Faraday Disc. Chern. Soc., 1979, 66, 116.lo A. L. Gosman, R. D. McCarty and J. G. Hust, Nat. Bur. Stand. Ref. Data Ser. (U.S. Govt.Printing Office, Washington D.C., 1969), p. 27.l1 M. P. Wertheim, Mol. Phys., 1973,26,1425; 1977,34,1109; 1979,37, 83.l2 G. N. Patey and J. P. Valleau, Chem. Phys. Letters, 1976,42,407 ; G. N. Patey, G. M. Torrieand J. P. Valleau, J. Chem. Phys., 1979, 71, 96.(PAPER 91324

ISSN:0300-9599    

DOI:10.1039/F19807600174

出版商:RSC    

年代:1980

数据来源: RSC

摘要:

J. C. S. Faraday I, 76,180-1 95Zeolite SorbentsModification by Impregnation with SaltsBY RICHARD M. BARRER,* DAVID A. HARDING (IN PART) AND ARVIND SIKANDPhysical Chemistry Laboratories, Chemistry Department, Imperial College,London SW7 2AYReceived 1st March, 1979A study has been made of the effect of salt inclusion in certain zeolites upon the kinetics of sorp-tion of, and the molecule sieving behaviour shown towards, n-hexane, 2- and 3-methyl pentanes,2,2- and 2,3-dimethyl butanes, cyclohexane and benzene. The zeolites selected provided threetypical one dimensimal channel systems. They were zeolite L, offretite and H-mordenite. Thesalts used to impregnate the channels were KCl, KBr, KzS04 and K2Cr04, all comparatively stableto heating. The effects of salt concentration and salt type were investigated in each zeolite and alsoof washing and extraction of salt-impregnated sorbents.Among the above salts K2Cr04 gave thelargest effects. Access to the channels could be selectively controlled and changed by salt imbibitiongiving rise to various separation possibilities based on differences in size and shape of sorbatemolecules.Molecule sieving by a given zeolite sorbent can be altered in various ways. Oneof these is ion exchange : the ions may be of about the same size but have differentcharges (2Naf $ Ca2+); or they may have the same charge but different radii(Na + K+ ~t Rbf + Cs+). Exchange can modify zeolite sorbents very effectivelywhen the cations occupy window positions in three-dimensional channel networks orsit in one-dimensional (i.e.parallel non-intersecting) channel systems. 1-3 Secondly,one may change selectively the accessibility of intracrystalline pores to molecules ofdiffering dimensions by direct syntheses in which the A1 : Si ratio is altered and hencethe cation density in windows and cl~annels.~ Thirdly, one may pre-sorb into thepores strongly sorbed polar molecules which are immobile at the temperature atwhich separation of other mobile sorbates is to be effe~ted.~. Finally, one may bychemkorption attach groups to the anionic frameworks, as illustrated by reactions ofSiH4 with acidic silanols of H-zeolites and de-aluminated H-zeolites. Attachedgruups were found strongly to influence sorption and molecule sieving.has had success ininterpreting the accessibility of the intracrystalline pore space.* One-dimensionalchannel systems block more easily however and their behaviour has been consideredin terms of the effect of high energy barriers at intervals along each channel uponmolecule migration along it. ORecently changes in mobility of n-hexane and 2,2-dimethyl butane were investi-gatedl in the one-dimensional channels of offretite, after introducing, respectively, theexchange ions Lif, Naf, Kf, Csf, MeNHz, Me,NH;, Me3NH+ and Me4N+. Theinorganic ions were all too small greatly to influence sorption kinetics, but adequateamounts of Me,N+ blocked even n-hexane. However, organic exchange ions are notvery stable to heating and so a further method of selective channel blocking wasdesired.Zeolites soaked in salt solutions may imbibe the salts, according to aDonnan membrane equilibrium.12 Thermally stable salts can be selected and theIn three-dimensional channel networks percolation theory18R . M. BARRER, D . A. HARDING AND A . SIKAND 181anion dimensions varied systematically. It was thus of interest to establish whetherimbibed salts can function within one-dimensional channels as controllable intra-crystalline barriers selective for molecule sieving.EXPERIMENTALMATERIALSSynthetic offretite, zeolite L and mordenite (as H-Zeolon) were selected as sorbents. Thesorbates were the Cs hydrocarbons n-hexane, 2- and 3-methyl pentanes, 2,2- and 2,3-dimethylbutanes, cyclohexane and benzene. The salts used to impregnate the channels of thezeolites were KCl, KBr, KzS04 and K2Cr04.Crystallographic dimensions l3 of the ionswere taken as : K+, C1- and Br-, 2.66, 3.6 and 3.9A in diameter, respectively ; SO$- andCrOi-, 4.8 and 5.5 A for the respective tetrahedron heights and 5.5 and 6.3 A for the spheresjust circumscribing each tetrahedron.Some relevant properties of the Cs hydrocarbons (> 99 % pure) and of the zeolitesorbents are given in tables 1 and 2, respectively. Liquid densities, p, of hydrocarbons atTABLE 1 .-HYDROCARBONSrelative pressure at liquid densitymolecule 30°C and 63 Torr at 30°C dimensionsa/An-lrexane 0.34 0.649 4.0 ( t ) 4.9 (b) 10.3 (I)2-Me-pent ane 0.24 0.643 4.65 (h) 6-18 (6) 9.0 ( I )3-Me-pentane 0.27 10.643 b] 4.65 (h) 6.18 (6) 9.0 ( I )2,2-di-Me-butane 0.16 0.639 5.9 (h) 6.18(b) 7.77(1)2,3-di-Me-butane 0.22 0.652 4.65 (h) 6.18 (b) 7.77 ( I )cyclohexane 0.52 0.769 4.9 ( t ) 6.4 (b) 7.2 ( I )benzene 0.55 0.869 3.7 ( t ) 6.7 (b) 7.4 ( I )0 t = thickness ; b = breadth ; h = height ; I = length.The lengths refer to fully stretchedchains. * b Assumed density.TABLE 2.-zEOLITE SORBENTSwater and salt-free corresponding cell free dimensionszeolite cell compositions cell weights dimensionso of channels1A IA(K,TMA)-offretitea K J . O T ~ I * ~ [ A ~ ~ . ~ S ~ 14.203d 1290(K,TMA)-offretitea K~.~TMAI.I[AI~.~S~ 14.20 36] 1 1 262 hexag.(K,H)-offretitea K2.7H i.iW3.7Si 14.20361(outgassed 300°C)(K,H)-offretite(salt-impregnated :zeolite Lc JL~Na3[AWi270721 2456 hexag.zeolite L(outgassed 300°C)u Z 13.3 Z 6.4 1 ::: c 7.6K 3.8[AI 3& 14-20 3-61outgassed 360°C) JJa ," 18.4 ," 7.1(salt impregnated : Kg[A19Si270721 1 2504 c 7.5outgassed 360°C)outgassed 360°C)H-mordenite (H-Zeolona &*6[A16.61Si400961 1 2843 orthorhombica ", 18.1b ", 20.5 ", 6.7x7.0c ", 7.5 1 3093 J mordenite as above (salt K6*61[A16*6lSi4009dimpregnated : outgassed 360°C)a Approximate values because of treatments given or compositional differences from typeComposition of type material, ref.(19). materials, ref. (18). b Compositions from ref. (11).Compositions from ref. (17)182 ZEOLITE SORBENTS30°C were obtained from those at 20°C,'* assuming (ap/aTp) = 0.001 g ~ r n - ~ K-'. Forn-hexane, benzene and cyclohexane the values so found agreed with values at 30°C lS towithin < 0.25 %.The dimensions are based on those given for CH3, CH2 etc. by Pauliag.16The parent offretite of composition given in table 2 contained a small excess of base(eg., NMe40H), almost all of which was removed by outgassing at 360°C.'' The parentH-Zeolon (H-mordenite) was partly dealuminated l7 and the ideal composition when free ofmolecular water was H6.61[A16.61Si400,3.,2fOH)5.56]. It is assumed in the mordenitecompositions of table 2 that on outgassing at 360°C the -OH groups in " nests " resultingfrom de-alumination release all their water, although this is not exactly true.8 Also, aftertreatment with the potassium salts and outgassing at 360°C in presence of ex- solid salt,it is assumed that acidic H is replaced by K.The channels in offretite and mordenite are ofnearly uniform cross-section along their lengths. Those in offretite are cross-linked via 14-hedral gmelinite cages with 8-ring windows, so that if tetramethylammonium (TMA) ionsare removed from these cages it might be possible for n-hexane to enter and so migratebetween channels. None of the other Ca hydrocarbons could do so. The channels inmordenite are lined with side pockets, access to which is through an 8-ring window. Theends of n-hexane molecules might enter such pockets. In mordenite and in zeolite L thereare no lateral openings between channels large enough for passage of any c6 hydrocarbon.In zeolite L the free dimension of E 7.1 A refers to the narrowest parts along each channel,occurring at inter-vals of M 7.5 A.Between these parts the channel broadens considerably.L-PROCEDURES - .DI Parts of the parent (K,TMA)-offretite were calcined in air at 520°C for 16 h to convertthem to (K,H)-offretites. Samples of the parent (K,TMA)-form and the (K,H)-offretiteswere then soaked in the appropriate salt solutions for 16 h at 80°C. They were next filteredand similarly soaked in the solution twice more before being finally air-dried without re-moving adhering salts. (K,H)-offretites were outgassed at 360°C for at least 16 h prior touse as sorbents. Samples containing Th4A ions were, however, outgassed at 300°C becausehigher temperatures tended to decompose TMA. The TMA and water contents were foundby microanalysis and thermogravimetry.Several of the salt-impregnated sorbents werewashed on a Buchner funnel and associated residual salt determined. In these instances 0.5 gof sorbent were first washed for FZ 0.5 min with 10 cm3 of 0.1 mol dm-3 salt solution andsimilarly with 10 cm3 of distilled water, in order to remove as much external salt as possiblewith the least elution of intracrystalline salt. In other instances the salt-impregnated zeolitewas more severely extracted. = 0.5 g of zeolite was washed for E 1 min with 500 cm3 ofdistilled water. The sampIe was then filtered and stirred with 50 cm3 of distilled water for16 h at 80°C. This stage was repeated twice more. Various salt-impregnated mordeniteand zeolite L samples were made by the same impregnation procedures and some werewashed or extracted in the same way.Thus samples of each salt-bearing zeolite withadhering salt, in lightly washed or in extracted form, were available, as well as parent formsbefore any salt impregnation.Sorption measurements were made gravimetrically using silica spring balances, withsample weights in the range 190-280mg. The sample temperatures were 30°C and thepressure of sorbate vapour was 63 & 2 Torr.RESULTS AND DISCUSSIONSome of the experimental observations are summarised in tables 3 and 4. It isimmediately apparent that sorbent modification by salt impregnation can readily beeffected in ways that are selective as regards the shapes of the sorbate molecules.Kinetics of uptake (cm3 of sorbed liquid hydrocarbon against time, in a weight cor-responding with 1 g of outgassed salt-free zeolite) are compared for the various hydro-carbons in parent (K, TMA)- and (K, €3)-offretites in fig.l(a) and (b), respectively.The kinetics in the same two sorbents treated with 1 mol dm-3 K,CrO, are similarlR. M. BARRER, D. A . HARDiNG AND A . SIKAND 183expressed in fig. l(c) and (d) and may be compared inter se and with the correspond-ing behaviour in the parent zeolites. Only parts of the measured curves are shown ;runs were continued for + 1-4 days, according to the rate of uptake, before the finalsorptions in tables 3 and 4 were recorded, so that most of these refer to, or are veryclose to, equilibrium. Similarly fig. 2 and 3, respectively, show kinetics of hydro-carbon uptakes in parent zeolites L and H-mordenite and in these sorbents variouslytreated with KCl and K,CrO,.TABLE 3 .-SORPTIONS OF & HYDROCARBONS AT 30°C AND 63 2 TOIT IN MODIFIED OFFRETITES.SORPTIONS ARE GIVEN IN MOLECULES PER UNIT CELLhydrocarbon1no.sorbentn-hexane 2-Me- 3-Me- 2,2-diMe- 2,3-diW- cyclo-pentane pentane pentane pentane hexane benzene123456789101112131415(K,TMA)-OFF 1.03(K,TMA)-OFF in 3 mol dm-3 KCI 0.651(K,TMA)-OFF in 1 mol dm-3 K2CkO4 1.02(K,H)-OFF 1.156(K,H)-OFF in 3 mol dm-3 KC1 0.697(K,H)-OFF in 0.1 mol dm-3 KCl 1.136(K,H)-OFF in 3 rnol dm-3 KCl, washed 1.119(K,H)-OFF in 0.1 mol dm-3 K2Cr04 1.043(K,H)-OFF in 1 mol dm-3 K2CrO.q 1.012(K,H)-OFF in 1 mol dm-3 K2Cr04,(K,H)-OFF in 1 rnol dm-3 K2CrO4.washed 0.986extracted 1.23 1(K,H)-OFF in 0.1 mol dm-3 KBr 1.138(K,H)-OFF in 3 mol dm-3 KBr 0.376(K,H)-OFF in 0.1 mol dm-3 K2SO4(K,H)-OFF in 0.5 mol dm-3 K2SO41.1081.11 10.5710.0850.2220.7300.5100.0700.2780.1630.0550.6600.0820.22~0.2270.8620.1300.4390.1350.3310.1100.10~0.0980.1930.1000.0850.8390.38,0.083 -0.1120.0790.4010.1250.4880.3600.164 1.030.11~ 0.4510.126 0.4380.891 1.330.540 0.7410.325 0.5030.117 0.2g70.126 0.2060.080 0.1100.05~ 0.09~0.800 1.120.425 0.767- 0.2800.515 0.7440.358 0.543TABLE 4.-sORPTIONS OF cg HYDROCARBONS AT 30°C AND 63 & 2 Ton IN MODIFIED ZEOL- LAND MORDENITES.SORPTIONS ARE GIVEN IN MOLECULES PER UNIT CELLhydrocarbon1no.sorbent2-Me- 2,2-di-Me 2,3-di-Me cyclo-n-hexane pentane butane butane hexane benzene123456789101112131415zeolite Lzeolite L in 0.1 mol dm-3 KC1zeolite L in 3 mol dm-3 KClzeolite L in 0.1 mol dm-3 K2Ci-04zeolite L in 1 mol dm-3 K2Cr04zeolite L in 1 rnol dm-3 K2Cr04,zeolite L in 0.1 mol d m 3 KBrzeolite L in 0.1 mol dm-3 KzSO4extractedH-MORH-MUR in 0.1 mol dm-3 KClH-MOR in 0.1 mol dm-3 KCl, extractedH-MOR in 3 mol dm-3 KC1H-MOR in 0.1 rnol dm-3 &Cm4H-MOR in 0.1 mol dm-3 K~Cr04,H-MOR in 1 rnol dm-3 K2CrO4extracted2.22 1.87 1.171.82 1.40 0.980.535 0.255 0.2880.497 0.382 0.2920.062 0.081 0.03~2.165 1-76 1.3542.10 1.61 1.1741.64 1.10 0.9042.00 1.23 0.1611.97 1.76 1.361-18 0.85 0.830.73 0.324 0.2920.128 0.144 0.0921.15 0.90 0.780.175 0.144 0.1%1.62 ---0.1120.2501.76 -0.2721.131.160.2960.3160.1151.551.320.9440.4241.751.180.3910.1331.050.0092.291.530.4680.3870.1982.251.822.563.091.861.011.3440.2611.790.281840.12-0.10ZEOLITE SORBENTSap"""""" = - = - 0-0.040.0202/qh*FIG.l.-Sorption behaviour of C6 hydrocarbons, at 30°C and the relative pressures of table 1, in(a) (K,TMA)-offretite, (b) (K,H)-offretite, (c) (K,TMA)-offretite treated with 1 mol dm-3 K2Cr04and (4 (K,H)-offretite treated with 1 mol dm-3 K2Cr04. 0, n-hexane ; A, 2-Me-pentane ; V, 3-Me-pentane ; n, 2,2-di-Me-butane ; m, 2,3-di-Me-butane and X, benzene. Va is the volume sorbedexpressed as cm3 of liquid hydrocarbon per g of outgassed salt-free zeoliteR.M. BARRER, D . A. HARDlNG AND A . SIKAND0.120.100.080.06b 0.04-0.02185- - - L v - " n - - n Aw -- A : - A 3 # LL- *-U)(==-X-.---o---.- 0 fA- (dl -I . I I I I IFIG. 2.-Sorption behaviour of C6 hydrocarbons in zeolite L, at 30 "C and the relative pressures oftable 1 , in (a) parent L, (b)L treated with 0.1 mol dm-3 K2Cr04, (c) L treated with 3 rnol dm-3 KCL and(d) L treated with 1 mol dm-3 K2Cr04. The ordinates and symbols denoting the hydrocarbonsare as in fig. 1186 ZEOLITE SORBENTSAt 30°C and the relative pressures in table 1 equilibrium sorptions should ap-proach the saturation capacity of each C6 hydrocarbon because intracrystalline sorp-tion isotherms of these hydrocarbons in zeo1ites at this temperature are rectangular inform.At the same time the relative pressures are such that sorption on the smallexternal surfaces of the crystallites should be limited. Thus the uptakes, expressed intables 3 and 4 as molecules per unit cell, refer primarily to intracrystalline hydrocarbonuptakes except for the smallest of these, for which sorption on external surfaces couldbe a significant part.THE SALT-FREE ZEOLITESPassing through a unit cell of offretite there is one channel section of length E 7.6 A(table 2), whereas the full length of n-hexane is 10.3A (table 1). Despite this themaximum uptake of n-hexane in (K, TMA)-offretite was 1.03 molecules per unit cell.Accordingly, because the gmelinite cages are believed to be fully occupied by TMAions, the n-hcxane may either be coiIed as a helix against the channel wall, or bepuckered. In (K, H)-offretite, where the gmelinite cages me free of TMA, the maxi-mum uptake of n-hexane was 1.23 molecules per unit cell and some part of this couldrepresent insertion of a portion of the n-hexane into the cages, additionally to itspresence in the through channels.For benzene, of maximum length w7.4 a and,because of its dimensions (table l), no chance of entering gmelinite cages, the maxi-mum uptake also exceeds one molecule per unit cell. Thus at the closest packing ofbenzene the planes of the benzene rings may be tilted relative to the channel axes.For all branched chain C6 paraffins and for cyclohexane access to the salt-free(K, TMA)-offretite is restricted, often very greatly so (table 3, sorbent 1).Thisbehaviour is attributed to the presence of residual TMA ions in the through channelswhich are bulky enough to limit or hinder access of the equally bulky iso-, neo- andcyclo-paraffin molecules. However, in (K, H)-offretite where no TMA is presentthese bulky molecules were sorbed in amounts between 0.73 and 0.89 molecules perunit cell (table 3, sorbent 4). 2- and 3-Me-pentanes have fully extended lengths of9 A (table 1) and are less flexible within the channels than n-hexane, while 2,3- and2,2-di-Me-butanes have lengths of = 7.8 A ; within the channels their bulk allowslittle flexibility.Cyclohexane [dimensions 4.9 by 6.4 by 7.2 A (table l)] can also havelittle room for movement or flexibility in channels of 6.4A free diameter. When inaddition one considers the loose van der Waals contact between molecules, uptakesof less than one per unit cell (table 3, sorbent 4) are seen to be reasonable for iso-, neo-and cyclo-hexanes.In zeolite L and mordenite two wide channels traverse each unit cell so that up-takes in molecules per unit cell are about twice those in offretite (table 4, sorbents 1and 9). Allowing for this, for n-hexme and benzene the numbers of moleculessorbed are such that the arguments of the previous paragraph should also apply to thepacking of these hydrocarbons in each channel. In mordenite the 14-hedral gmelinite-type cages of offretite are replaced by side-pockets. In mordenite, however, a newfactor appears in that the uptakes of the most globular molecules, 2,2- and 2,3-di-Me-butane and cyclohexane, are drastically reduced (table 4, sorbent 9)’ even though thefree dimensions of the channels are greater than in offretite (table 1).It is suggestedthat, in this partially dealuminated H-mordenite, Al-bearing fragments derived fromthe framework by dealurnination are present in the wide channels in sufficient mountsto hinder or prevent entry of the most globular of the C6 hydrocarbons. Rates ofuptake of the various CB hydrocarbons in the parent H-mordenite are compared infig. 3(a). For n-hexane, 2-Me-pentane and benzene the sorption is rapid but the uptakeR .M. BARRER,TD. A. HARDING AND A . SIKAND 187differ considerably (table 4, sorb& 9). For 2,2- and 2,3-di-Me-butanes sorption isvery slow indeed, so that this sarbent has differentiated kinetically between the twogroups of molecules in a very clear manner. Zeolite L, having channels the freediameters of which vary from ~ 7 . 1 to x 13 A, with periodicity along the channel of0.080.060.04va 0.02A n n A - 1'002 c z0*02t- (Cl0 OB-- - L x-x- -- A+-b b - - 002- (d 10. I I 1 I I I I I I I I 102 0 4 06 0 8 10 12 14 16 18 2 0 2.2 2.4&lh+FIG. 3.-Sorption behaviour of Cg hydrocarbons, at 30°C and the relative presrmres of table 1, inH-mordenite partially dealuminated. (a) The parent zeolite ; (6) the same treated with a1 mol dm3KC1 ; (c) the same treated with 0.1 mol d ~ n - ~ KC1 and extracted and (a) the same treated with0.1 mol dm-3 Kt CrO4 and extracted.The ordinates and symbols are as in fig. 1I88 ZEOLITE SORBENTS257.5A between pairs of narrow or pairs of wide points, does not so differentiate(table 4, sorbent 1), although uptakes of 2,2-di-Me-butane and cyclohexane in par-ticular are reduced.THE SALT-BEARING ZEOLITESAfter impregnation with 3 mol dm-3 KC1 the (K, TMA)- and (K, H)-offretitesbecame similar (table 3, sorbents 2 and 6). Thus, with the possible exception ofbenzene, a given hydrocarbon was taken up to nearly the same extent in both sorbentswhile the sorption of each hydrocarbon was characteristic of that species and dependedstrongly on its shape.It is therefore likely that treatment with 3 mol dm-3 KCl dis-places TMA from the wide channels of offretite and that after outgassing [at 300°Cfor the salt-bearing (K, TMA)-form and at 360°C for the salt-bearing (K, H)-form]both forms contain only K+ and KCl in the wide channels and so behave much alike.(K, H)-offretite treated with 3 mol dm-3 KBr sorbed less n-hexane and benzene thandid this form treated with 3 mol dm-3 KCl, but 3-Me-pentane was sorbed nearlyequally in both sorbents.When (K, TMA)- and (K, H)-offretite were each treated with 1 mol dm-3 K2Cr04the effect on the final uptake of n-hexane was small (table 3, sorbents 1 and 3 andsorbents 4 and 9, respectively). In strong contrast, however, both these two salt-impregnated sorbents took up remarkably little cyclohexane, 2,2- or 2,3-di-Me-butane.The salt-bearing (K, H)-offretite also differentiated sharply between benzene md n-hexane (sorbeat 9, columns 3 and 9).The treatment of (K, TMA)- and (K, H)-offretites with 1 mol dm-3 K,Cr04resulted in marked changes in the rates of sorption of n-hexane [cf.fig. l(a) and (c)and fig. l(b) and (d)] and in very great changes for benzene and 2-Me-pentane in(K, TMA)-offretite and for all the hydrocarbons save n-hexane in (K, H)-offretite.On the other hand, in (K, H)-offretite treated with 0.1 mol d ~ n - - ~ KC1 sorption rateswere not sensibly diminished, although uptakes were altered for all the C6 hydro-carbons save n-hexane [cf. fig. l(b) and fig 4(a) and table 3, sorbents 4 and 51.In zeolite L treated with 3 mol dm-3 KCl uptakes were all much reduced withoutmarked differentiation between the hydrocarbons (table 4, sorbent 3).Whenimpregnated with 1 mol dm-3 K2Cr04 the uptakes, including that of n-hexane, werevery small, with benzene showing the best result (table 4, sorbent 5).The behaviour of H-mordenite after the treatment with 0.1 mol dm-3 KCI wasunexpected because after the treatment the sorptions of benzene, cyclohexane and theiso- and neo-parafEbs were much enhanced (table 4, sorbents 9 and 10) and therewere large increases in rates of uptake of 2,2- and 2,3-di-Me-butanes and cyclohexane[cf. fig. 3(a) and (b)]. Whatever had blocked the salt-free parent H-mordenite to thesemolecules appeared largely to have been removed by the treatment.On the otherhand, the H-moirdenite impregnated with 3 mol dm-3 KCl showed the expected reduc-tion in sorption capacity (table 4, sorbent 12). Although shape-selectivity is evidentthis selectivity is less marked than for the offretites 2 and 6 of table 3, but rather moreso than for zeolite L, sorbent 3 of table 4.EFFECTS OF WASHING AND OF EXTRACTIONWashing the (K, H)-offretite as described in the Experimental section, aftertreating it with 3 mol dm-3 KCl, considerably increased the uptakes of all the hydro-carbons and especially improved the rates of sorption of n-hexane [cJ: fig. 4(b) and (c)].There was at the same time some loss of selectivity as between n-hexane and the otherhydrocarbons (table 3, sorbents 6 and 7).In contrast, washing the (K, H)-offretit189after treating it with 1 mol dm-3 K2Cr04 if anything improved the selectivity for n-hexane (table 3, sorbents 9 and 10). As observed earlier, the washing was designed toremove externally adhering salt but as little as possible of that within the zeolite. Themore severe process of extraction (cf Experimental) of the (K, H)-offretite treated with1 rn~ljldrn-~ K,Cr04 restored much of the capacity of the parent salt-free (K, €3)-offretite to sorb 2-Me-pentane, cyclohexane and benzene (table 3, sorbents 9, 10 andR. M . BARRER, D. A . HARDING AND A . SIKAND0 a12n fi “ U U- o c - n A C h *U Y ” v0 I I I I I I I 1 I I I I ,0.640.02--0.06O * O 8 IFIG. 4.-Sorption behaviour of C6 hydrocarbons, at 30°C and the relative pressures of table 1 in(K,H)-offrdtites.(a) Treated with 0.1 mol d ~ n - ~ KCI ; (b) treated with 3 mol dm-3 KCl and (c)treated with 3 mol dm-3 KCl and washed. The ordinates and symbols are as in fig. 1190 ZEOLITE SORBENTS.\/@Fx;. 5.--Sorption behaviow of C6 hydrocarbons, at 30°C and the relative pressures of table I, in(K,H)-offretite treated with (a) 0.1 mol d m 3 K2Cr04 ; (b) 1 mol d w 3 K2CrO4 and washed and(c) 1 mol dm-3 K2Cr04 and extracted. The ordinates and symbols are as m fig. 1R . M. BARRER, D. A . HARDING AND A . SIKAND 19111). Again, while impregnation with 0.1 mol dm-3 K,CrO, caused a decrease inthe rate of sorption of n-hexane [fig. l(b) and fig. 5(a)] and treatment with 1 mol dm-3K2Cr04 diminished the rate still more [fig.5(a) and (b)], extraction produced a sorbentwhich once more sorbed n-hexane rapidly [fig. 5(b) and (c)]. The extraction, asdescribed in Experimental, was intended to remove both external and intra-crystallinesalt.When zeolite L previously impregnated with 1 mol d ~ n - ~ K,CrO, was extractedthe hydrocarbon uptakes became similar to those in the parent salt-free zeolite, asseen on comparing sorbents 1 and 6 of Table 4. Sorbent 6 can in turn be comparedwith the salt-treated, unextracted sorbent 5 in which the uptakes of the hydrocarbonsare greatly reduced. Zeolite L and mordenite treated with 0.1 mol dm-3 K2Cr04also gave reduced uptakes (sorbents 4 and 13 of table 4, respectively). Extraction ofthe mordenite sorbent 13 restored much of the ability to take up n-hexane, 2-Me-pentane and benzene and resulted in a higher uptake of 2,2-di-Me-butane andcyclohexane than in the parent salt-free H-mordenite (table 4, sorbents 14, 13 and 9).On the other hand, extraction of H-mordenite treated with 0.1 mol dm-3 KC1 causeda reduction in uptake with no marked improvement in selectivity (table 4, sorbents 11and 10).Evidently more than one factor operates in the extraction process.RELATIVE BLOCKING BY DIFFERENT SALTSThe relative ability of the four salts, K2Cr04, K2S04, KCl and KBr, to modify theuptakes of the c6 hydrocarbons is assessed in Table 5, at concentrations of 0.1 mold ~ n - ~ . Potassium chromate was consistently much the most effective in reducing theamounts sorbed, except for n-hexane in (K, H)-offretite where none of the 0.1 mol dm-3salts had any marked effect.Accordingly the (K, H)-offretite treated with 0.1 moldm-3 K2Cr04 can differentiate very well between n-hexane and the other c6 hydro-carbons (table 3, sorbent 8). In zeolite L a single sequence of blocking ability isobserved for all four salts towards each of the hydrocarbons. In offretite there aresome variations in the sequences but these variations do not represent large changes.TABLE 5.-RELATIVE EFFECT OF 0.1 m01 dm-3 SALTS IN BLOCKING HYDROCARBON UPTAKES INZEOLITES OF TABLES 3 AND 4zeolite and salt sequence in blockinghydrocarbon offretite zeolite L mordeniten-hexane KzCrO4 > KCI2-Me- pentane K2CrO4> KCI K2Cr04 > K2S04 > KCI > KBr K~Cr04 > KC13-Me-pentane2.2-di-Me-butane KzCrO4> KCI 2.3-di-Me-butane KzCr04 > KCI - -c yclo hexane K2Cr04 > KCIbenzene K2Cfi4 > KCIlittle blocking by 0.1 rnol dm-3 saltsKzCr04 > KBr > KCI > K2S04K2Cr04 > KBr > KCI > KzSO4KZCa4 > KBr > K2S04 > KCIK2Cfi4 > KCI, K2SO4 > KBrK2Cr04 > KzSO4 > KCl > KBrKzCrO4 > K2S01> KCI > KBrK2Cr04 > K2S04 > KCI > KBrK2Cfi4 > K2S04 =- KCI > KBr- -Tables 3 and 4 show that the zeolites all become increasingly blocked as the impreg-nating salt solutions are made more concentrated.The behaviour of H-mordenitetreated with 0.1 mol d ~ n - ~ KC1 as compared with parent salt-free H-mordenite isexceptional (table 4, sorbents 9 and lo), but going from 0.1 to 3 mol dm-3 KC1 reducesthe uptakes markedly (Table 4, sorbents 10 and 12).Further information on con-centration dependence was obtained in an additional study of offretite. Crystals of(K, TMA)-offretite were soaked in KCl solutions at room temperature for xl6 h192 ZEOLITE SORBENTSfiltered and dried and then part of each sample was calcined in air at 650°C to removeTMA. Crystals similarly soaked in K,Cr04 solutions were also prepared from(K, 33)-offretite which had been made from the (K, TMA)-offretite by calcining in air at520°C. The salt-bearing sorbents containing TMA were outgassed as before at 300°Cand those free of TMA at 360°C. Table 6 shows the effects of salt concentrationsupon the uptake of n-hexane, 3-Me-pentane and 2,2-di-Me-butane.For the(K, €3)-offretites the behaviour is regular : as salt concentration increases hydrocarbonuptake decreases. With K2Cr04 the amounts of the two branched chain hydro-carbons sorbed decreases at first very sharply and then more slowly with salt concen-tration.For the (K, TMA)-offretites the behaviour was different, due to the operation oftwo factors. The first of these was progressive elimination of blocking TMA ionsfrom the wide channels by ion exchange with potassium and the second was imbibitionTABLE 6.-SALT CONCENTRATION AND HYDROCARBON UPTAKE IN SALT-BEARING OFFRETITESmoles sorbed per unit celln-hexane 3-Me-pentane 2,Zdi-Me-pentanesolution andconcentration (K,TMA)-form (K,H)-form (K,H)-form (K,TMA)-form (K,H)-form00.5KCI 1 .o2.03.000.05K2Cr04 0.10.5w30.93 1.140.65 1.170.71 1.090.76 0.860.70 0.67- 1.14- 1.12- 1.11- 0.92- 0.570.860.370.280.120.090.15 0.890.11 0.820.19 0.620.07 0.390.22 0.31- 0.89- 0.24I 0.16 - 0.04 - 0.030.02va o.o\ 0.012.0 4.0dqh&FIG.6.-Effects of successive runs at 30 "C and the realtive pressure of table 1 on the sorption be-haviour in ofketite. Curves 1 and 2 refer to the second and eighth runs with parent (K,TMA)-offretite. The sorbate is 3-Me-pentane. Curves 3 and 4 are for benzene, the first and eighth runswith (K,TMA)-offretite treated with 1 mol dm-3 K2Cr04. The ordinate is as in fig. 1T 4t-1,1,IIIIII8 16 24 32 8t*/min+FIG.7.-Plots against l / t (t in min) of the fractional approach to equilibrium for the uptake of n-KCl. The lower curves refer to samples with TMA present and the upper curves to ignited samplesmolarities of the KCI solutions with which the crystals were treated. (b) (K,H)-offretites treated withcurves are the molaxities of the K2Cr04 solutions with which the crystal194 ZEOLITE SOWBENTSof the salt. These two opposed factors partially or largely offset one another as thetable shows.EFFECT OF SUCCESSIVE SORPTIONS ON A GIVEN SORBENTWith the (K, H)-offretites treated with a particular salt at a given concentrationthere was little change in the sorption capacity for a given C6 hydrocarbon before andafter a succession of sorption runs.With (K, T11MA)-offretite, salt-free or salt-bearing,there was a tendency, shown in fig. 6, for the hydrocarbon uptakes to increase withrepeated use. This behaviour may be associated with progressive elimination of TMAfrom the wide channels. With the salt-bearing forms of zeolite L and mordenitechanges in sorption with repeated use were small and no definite correlations wereobserved in that slight increases and decreases were noted.KINETICSFig.1-6 illustrate some kinetic runs and show that sorption rates can vary greatlyaccording to the sorbent, its treatment and the hydrocarbon. A more detailedexamination was made of sorption kinetics of n-hexane at 30°C in (K, TMA)- and(K, H)-offretites treated with KCl solutions at various concentrations [fig.7(a)] and(K, H)-offretiles similarly treated with K,CrO, solutions [fig. 7(b)], the proceduresbeing those described in the paragraph on relative blocking. The rates for n-hexanein (K, 33)-offretite decreased in a regular manner as the salt concentrations increased[fig. 7(a), top series of curves; and fig. 7(6)]. In the case of (K, TMA)-offretite thecompeting effects of TMA displacement and salt uptake are evident because the ratesequence is related to salt concentration in the order : 1 mol dm-3 > 2 mol >3 rnol dm-3 > 0.5 mol drr3.For n-hexane sorbing into (K, H)-offretite treated with 0.5 and 7z3 mol dm-3K2Cr04 the rates were sufficiently slow to obtain the first sections of the curves of(Q,- Q,)/(Q, - Q,) against dt, where Qt, Q, and Q, are uptakes at time t, whent = 0 and at equilibrium, respectively.These sections are linear and if the diffusivity,D, is constant the slope, S, of the plot iss = 4 0 -J-.I n :Thus D/Z2 = 0.196 S2.For the (K, H)-offretites treated with 0.5 and 7z3 mol dm-3 K2Cr04 the values ofD/Z2 at 30°C were, respectively, 7.4 x and 2.5 x min-'. If sorption rates arecontrolled by intra-crystalline diffusion I is the mean length of crystallites measured inthe c direction. If sorption is controlled by inter-crystallite diffusion I is the effectivemacroscopic depth of the bed. In the former instance taking I as 1 pm, which is inthe size range typical of synthetic zeolites, D in cm2 s-l would be 1.2 x 10-l2 and4.2 x for the two treated offretites.CONCLUDING REMARKSThe present study has established that access to the channels of typical zeolites inwhich the channel systems are one-dimensional can be selectively controlled andchanged by soaking the zeolites in solutions of different concentrations and of varioussalts. The sorbents were then often shape- and size-selective for the various C6hydrocarbon sorbates examined and this suggests possibilities for mixture separationR .M. BARRER, D. A. HARDING AND A. SIKAND 195provided interferences of the fast by the slow moving molecules are not dominant.This aspect has not yet been examined. Some possibilities axe :(1) Separation of straight from branched chain and cyclic paraffins. Sorbents 9 and10 of table 3. (2) Separation of straight chain paraffin and benzene from branchedchain paraffin and cycloparaffin.Sorbents 3, 6 and possibly 13 of table 3. (3)Separation of straight chain param, benzene and 2-Me-pentane from the two di-Me-butanes and cyclohexane. Sorbcnt 9 of table 4. (4) Separation of benzene fromcyclohexane. Sorbent 1 of table 3, and sorbent 9 of table 4.One may expect that salt-treated zeolites like offretite, zeolite L, mordenite andmazzite (zeolite Q) could be extended to give a variety of shape selective sorbentsadditional to those developed in this work. Separation possibilities like those indi-cated above need testing because of the possible interferences referred to abovebetween fast and slow moving diffusants, especially in one-dimensional channelsystems. Here much will depend on the relative affinities of slow and fast movingspecies for the intra-crystalline sorption sites and upon the mixture compositions.D. H. and A. S. acknowledge the support of the Wolfson Foundation and of AirR.M.B. acknowledges a recent award Products and Chemicals for part of this work.of a Leverhulme Emeritus Fellowship.R. M. Barrer, J. SOC. Chem. Ind., 1945,64,130, 132, and R. M. Barrer and L. Belchetz, J. SOC.Chem. Ind., 1945,64,131.R. M. Barrer, Trans. Faraday Soc., 1949, 45, 358.T. Takaishi, Y. Yatsurugi, A. Yusa and T. Kuratomi, J.C.S., Faraday I, 1974,70,97.R. M. Barrer and J. W. Baynham, J. Chem. SOC., 1956,2892.R. M. Barrer and L. V. C. Rees, Trans. Faraday SOC., 1954,50,852, 989.L. V. C. Rees and T. Berry in Molecular Sieves (SOC. Chem. Ind., London, 1968), p. 149.1977, Chicago.R. M. Barrer and J. C. Trombe, J.C.S. Faraday I, 1978,74,2798.J. M. Hammersley in Methods in Computational Physics (Academic Press, N.Y., 1963), vol. 1,p. 281.lo R. M. Barrer, Zeolites and Clay Minerals as Sorbents and Molecular Sieves (Academic Press,London, 1978), p. 30 et seq.l1 R. M. Barrer and D. A. Harding, Separation Sci., 1974, 9, 195.l2 R. M. Barrer and A. J. Walker, Trans. Furaday Soc., 1964, 60, 171.Interatomic Distances (Spec. Pub. Chem. SOC., London, 1958), no. 11 ; L. Pauling, The Nature ofthe Chemical Bond (Cornell, Ithaca, 1940), p. 345.l4 J. A. Riddick and W. B. Bunger, Techniques of Chemistry, ed. A. Weissburger (Wiley, N.Y.,3rd edn, 1970), vol. 2.International Critical Tables (McGraw Hill, N.Y., 1926).R. M. Barrer and J. Klinowski, J.C.S. Faraday I, 1975,71, 690.London, 1978), p. 472.(Amer. Chem. SOC, 1971), no. 101, p. 155.' R. M. Barrer, R. G. Jenkins and G. Peeters in 4th Int. Con$ on Molecular Sieve Zeolites, Aprill6 L. Pauling, Nature of the Chemical Bond (Cornell, Ithaca, 1940), p. 187.lS R. M. Barrer, Zeolites and Clay Minerals as Sorbents and Molecular Sieues (Academic Press,l9 W. M. Meier and D. H. Olson, in Molecular Sieve Zeolites-.& Adv. Chem. Ser., 1971, 101(PAPER 9/339

ISSN:0300-9599    

DOI:10.1039/F19807600180

出版商:RSC    

年代:1980

数据来源: RSC