摘要:

J. Chem. Soc., Faraday Trans. 1, 1984, 80, 99-109 Characterization of Zeolite Acidity An Infrared Study using Ammonia, Methylamine and n-Butylamine BY ASHIM K. GHOSH AND GEOFFREY CURTHOYS* Department of Chemistry, University of Newcastle, New South Wales 2308, Australia Received 19th April, 1983 Characterization of the acidity of an HY and a number of mordenite zeolites (SiO,/Al,O, ratios 8-26) has been studied by infrared spectroscopy using ammonia, methylamine and n-butylamine. The characteristic bands of adsorbed ammonia appeared at 1445 cm-l (Br~nsted type) and 1623 cm-l (Lewis type), with the disappearance of the acidic OH band at 3609 cm-l for mordenites and of the acidic OH bands at 3643 and 3568 cm-l for HY. When methylamine and n-butylamine were adsorbed on the catalysts the acidic hydroxyls were found to be accessible to the amines and a band of S,,,(NH,+) (RNH, adsorbed on a Brnrnsted site) appeared in the 1500-1540 cm-l region with a band at ca.1610 cm-l, which is a combination of RNH, on Lewis sites and daSym(NH,+). Brnrnsted acidity of the catalysts as determined from the band intensities of the adsorbed ammonia, methylamine and n-butylamine, although not in numerical agreement, showed the same trend. The alkylamines were decomposed into alkenes at elevated temperatures and the decomposition was prominent in the case of n-butylamine. Infrared spectroscopy is one of the most valuable techniques for the characterization of an acidic surface. Pyridine (p& 8.75) and NH, (p& 4.75) have been used frequently as bases to determine the nature (Brarnsted or Lewis), number and strength of acid sites. n-Butylamine (pK, 3.23), which is a relatively strong base, is commonly used in the titration methods using Hammett indi~atorsl-~ and therm~metry,*-~ and it will react with weaker acid sites than either pyridine or ammonia.The use of n-butylamine in infrared spectroscopy for the characterization of solid surfaces like silica, alumina, silica-alumina' and montmorillonitea has been reported. Various amines were adsorbed on a number of cation-exchanged X and Y zeolites and the adsorption studied by infrared spectros~opy.~ It is considered that basic molecules enter the pores of a zeolite to interact with all the acid sites, and the number of acid sites is determined from the amount of base required to reach the end point.There are problems with the determination of the acidity of zeolites using non-aqueous n-butylamine titration either by thermometryg* lo or by using Hammett indicators,lo-l2 and this may be due to a steric effect in the zeolite channels. In the case of Hammett indicators Deeba and Hall1, have cast serious doubt on the method, their experiments showing that equilibrium is not achieved between n-butylamine and silica-alumina catalysts. The aim of the present work has been to study the use of n-butylamine as a base in the characterization of zeolite acidity by means of infrared spectroscopy. The results are compared with those obtained using a similar strong base, methylamine (p& 3.34), and a weaker base, ammonia. The steric effect of the alkylamine bases in the zeolite channels has been investigated.99100 I.R. CHARACTERIZATION OF ZEOLITE ACIDITY EXPERIMENTAL A hydrogen mordenite (HM), a hydrogen zeolite Y (HY) and three dealuminated HMs with different Si02/A1203 ratios were used in the present investigation. HM from the Norton Co. was used without further treatment. The dealuminated HMs were prepared by heating the parent HM or NaM with 10 mol dm-3 HCl at its boiling temperature and then washing until free of chloride ion. HY was prepared by exchanging NaY (Linde LY-Y52) with aqueous ammonium chloride and then washing until free of chloride ion. The zeolites were dried at 423 K in air. The unit-cell compositions of the zeolites are shown in table 1 . n-Butylamine was distilled and dried over 5A molecular sieves and finally degassed by the conventional freeze-pumpthaw technique. NH, (purity 99.99 %) and methylamine (purity 98 %) were obtained from Matheson Co.and were used without further purification. Table 1. Unit-cell composition of the zeolites zeolite Si02/A1203 unit cell HY (3.58) Na12.27 H35.31 A147.58 si144.42 O384 HM (Norton) (8.43) Na0.31 H5.58 "5.89 si42.11 '96 dealuminated HM (14.81) Na0.06 H3.48 "3.54 si44.46 '96 dealuminated HM (1 7.9 1) Na0.04 H2.92 "2.96 si45.04 '96 dealuminated HM (25.7 2) Na0.07 H2.03 "2.10 si4.5.90 '96 The cell used for infrared absorption spectroscopy was a greaseless cell fitted with calcium fluoride windows and connected to the vacuum system. The ground zeolite powder was pressed into self-supporting wafers (ca.2.5 mg cmW2) under a pressure of 2 x lo8 N m-2 and introduced into the i.r. cell. The wafers were heated under vacuum while increasing the temperature to 773 K, kept under oxygen for 2 h to burn off any organic species present on the zeolite surface and finally the cell was evacuated for 2 h at the same temperature. The infrared absorption spectra were measured by a Nicolet MX- 1 Fourier-transform infrared spectrometer. The spectral regions between 4000-2600 and 1700-1300 cm-l were recorded at room temperature in all cases. After the background spectrum of the zeolite surface was obtained the base (ca. 60 Torr)? was admitted into the cell for 1 h at room temperature. The cell was evaluated successively at room temperature, 373,473 and 573 K for 1 h each time and the spectrum was recorded at room temperature.The background spectrum of the surface was subtracted from that of base on the surface for the acidity determination. RESULTS HYDROXYL GROUPS The background of all the mordenites studied showed two distinct hydroxyl stretching bands at 3609 and 3738 cm-l and a shoulder or a weak band at 3650 cm-l. The intensity of the low-frequency hydroxyl band decreases with the removal of aluminium, and consequently this hydroxyl group is associated with the presence of A1 in the mordenite.14-16 The low-frequency hydroxyls are considered to be active sites in catalytic activity and are responsible for Bramsted-acid sites, whereas higher- frequency hydroxyls are considered to be associated with silanols, SOH," and are weakly reactive to bases.16 On adsorption of ammonia the 3609 cm-l OH band of the non-dealuminated mordenite disappeared on adsorption of ammonia and the 3738 cm-l band shifted to a lower frequency and was broadened (see fig.1). With methylamine and n-butylamine neither of the hydroxyl bands disappeared completely but each band weakened in intensity. In the case of the dealuminated mordenites (see fig. 2) the 3609 cm-l OH band disappeared on adsorption of each of the bases and t 1 Torr = 101 325/760 Pa.A. K. GHOSH AND G. CURTHOYS 3609 3609 3609 (6) 3738 101 1 3668 3334 3668 333L 3668 333L wavenumber/cm-' Fig. 1. Change in the hydroxyl bands of non-dealuminated HM (Norton) due to the adsorption of bases: (A) ammonia, (B) methylamine and (C) n-butylamine under different conditions: (1) background of zeolite.Base desorbed at (2) room temperature, (3) 373 and (4) 573 K. 3800 3600 3800 3600 3800 3600 3LOO wavenumber/cm -* Fig. 2. Change in the hydroxyl bands of highly dealuminated HM (SiO,/Al,O, ratio 26). Conditions are the same as in fig. 1.102 I.R. CHARACTERIZATION OF ZEOLITE ACIDITY the 3738 cm-l band shifted to a lower frequency. The reappearance of the 3609 cm-l OH band on desorption of the bases occurred at a lower temperature (573 K) with ammonia, the weakest base used. The spectrum of HY showed four hydroxyl bands at 3568, 3643, 3692 and 3745 cm-l. The 3643 and 3568 cm-l OH bands disappeared on adsorption of the bases but the other two OH bands remained at the same positions with a slight decrease in intensity and a band appeared at 3600 cm-l.On desorption at elevated temperatures the background OH bands of the zeolites reappeared with the disappearance of the characteristic bands of adsorbed base. ADSORPTION OF AMMONIA After adsorption of ammonia on the zeolite surface the characteristic absorption bands of adsorbed NH, appeared at 3250, 1623 and 1465 cm-l. The band at 1465 cm-l, which gradually shifted to 1445 cm-l after ammonia was desorbed at elevated temperatures, is characteristic of NH,+ (Brsnsted-acid site) and the band at 1623 cm-l is characteristic of NH, adsorbed on Lewis-acid sites.18 The number of Brmsted-acid sites was determined from the characteristic band intensity after NH, was desorbed at 373 K and is shown in table 2 and compared with that determined by using alkylamines, and with the acidic hydroxyls of the zeolites in fig.3. Table 2. Acidity (Brsnsted) of zeolites determined by infrared spectroscopy acidity (arb. units) zeolite" NH3 CH,NH, BuNH, HY 6.07 HM (Norton) 15.58 dealuminated HM ( 1 4) 7.55 dealuminated HM (1 8) 4.49 dealuminated HM (26) 4.43 2.48 4.60 2.20 I .72 1.28 4.9 1 5.33 2.69 2.27 1.96 a Numbers in parentheses indicate Si0,/Al,03 ratio. ADSORPTION OF METHYLAMINE The assignment of the infrared absorption bands of methylamine and n-butylamine has been made in the present study following Chenon et aI.,l9 Jacobs et aL20 and Morimoto et aL7 The spectrum of methylamine gas or n-butylamine vapour shows a band at ca. 1620cm-l assigned to NH, bending vibration. It is well known that -NH$ has two angle-deformation vibrations, an antisymmetric vibration [6,,,,(NH$)] at 1625-1560 cm-l and a symmetric vibration [6,,,(NH$)] at 1550-1507 cm-l.The S,,,,(NH:) band appears in the same region as that of the NH, bending vibration and this band is not applicable to ascertain whether the NH$ species are present or not on the surface. On the other hand, Ss,,(NH,+) band seems to be of great advantage for the same purpose, as it is in the region which is independent of the NH, bending vibration. A preliminary experiment of adsorption of n-butylamine or methylamine on alumina showed a band at ca. 1600 cm-l which shifted to lower frequency on desorption at elevated temperatures, and this band was assigned to RNH, on Lewis sites.7 Adsorption of methylamine was carried out on an HY, an HM and three dealuminated HMs.The spectra of methylamine adsorbed on these zeolites are essentially the same, and the -NH,+ angle-deformation vibration bands at different evacuation temperatures are shown in table 3. The spectra of methylamine on mordenites show the symmetrical -NH$ bending vibration at 1507-1 5 12 cm-l, whichA. K. GHOSH AND G . CURTHOYS 103 8 - 6 - 4 - 3 . 0 4 - 3 - 2 - t - I I I 1 SOz /A1203 I 12 16 20 24 28 O L d 0 8 Fig. 3. Variation of the absorbance of (a) (LF) hydroxyls, (b) NHi- and (c) RNH; ( x , R = Me; 0, R = Bu) as a function of SiO,/Al,O, ratios of the mordenites. is considered to be characteristic of RNH, adsorbed on protonic sites. The 6,,,(NH:) band is composed of a number of constituents showing the heterogeneous nature of the protonic sites.This d,,,(NH,+) band in the spectrum of methylamine adsorbed on HY was observed at 1539 cm-l. The d,,,(NH~) band shifts a few cm--l towards lower frequency with decreasing intensity on increasing the evacuation temperature. A band observed at 1605-1623 cm-l is a combination of RNH, on Lewis sites and dasym(NH;). The Brsnsted acidity was calculated from the intensity of the d,,,(NH$) band after the base was desorbed at 373 K and is shown in table 2 and in fig. 3.104 I.R. CHARACTERIZATION OF ZEOLITE ACIDITY c, cd Ccl 0 n I C ." W n +m X E 6 5 Q -u C cd n l E In +m X z z W E .o nn m m w w z g l I I I \ o \ o n m W If!I I If! n m W n m W 21 1 2 1 I I I 2 If! n W If!I I If! n n W m W 21 1 2 1 I I I 2 2 I n m W 21 I 141 n m W If!I I s 1 I I I If! 2 n m 4 W E =e 0 n 0 b) 0 c, .r( 4 c, 8 0 E z 0 E .C( c, C .d m b) D 5 z D d 3 2 ...en 0 5; m UTable 4. daSym(NH,) and dSym(NH3) band positions (in cm-l) of adsorbed methylamine at different outgassing temperaturesn evacuation temperature/K room temperature 373 573 zeoliteb Gasyrn(NH,f) 8 s y m (NW) dasyrn(NH,+) &,(NH,+) 6asy m (NH,+) 4ym(NH,+) ;p 1623 (s) 1540 (s) 1621 (s) 1539 (s) 7c HM (Norton) 1605 (s) 1512 (s) 1617 (s) 1512 (s) 1617 (s) 1511 (s) a HY 1623 (s) 1540 (s) - - - - 1533 (w) 8 - - - X - - 1534 (w) 1534 (w) 5 - - - - - 1539 (w) dealuminated HM (14) 1607 (s) 1512 (s) 1621 (s) 1512 (s) 1623 (s) 1510 (s) - 1539 (w) - 1539 (w) - 1539 (w) I;1 dealuminated HM (1 8) 1616 (s) 1507 (s) 1617 (s) 1507 (s) 1623 (s) 1507 (s) 2 2 8 s - - - 1512 (w) 1512 (w) - - 1539 (w) - 1539 (w) - 1623 (s) 1539 (w) 1523 (w) - - - - 1539 (w) - - - - - - 1533 (w) 1533 (w) - 1511 (s) dealuminated HM (26) 1616 (s) 1512 (s) 1616 (s) 1512 (s) 1623 (s) - - a s, Strong; w, weak.Numbers in parentheses indicate SiO,/AI,O, ratio.106 I.R. CHARACTERIZATION OF ZEOLITE ACIDITY ADSORPTION OF n-BUTYLAMINE The spectra of n-butylamine adsorbed on the catalysts are essentially the same as those of methylamine. The S,ym(NH:) and S,,,,(NHi) band positions of n-butylamine desorbed at different temperatures are shown in table 4. The Brarnsted acidity was calculated from the S,,,(NH~) band and is shown in table 2 and in fig. 3. The possibility of decomposition of adsorbed n-butylamine should be considered for the zeolite samples.The infrared spectra adsorbed n-butylamine in the CH stretching zone are shown in fig. 4. The spectra show bands at 2966, 2936, 2870 and 2827 cm-l. On increasing the desorption temperature the 2870 cm-l band disappeared with the formation of two bands at 2885 and 2860cm-' and the intensity of the remaining bands increased. The band at 2860 cm-l, which was masked by the band at 2885 cm-l at lower desorption temperatures, may be assigned to the formation of intermediate butoxy groups. The thermal analysis of mordenites on which n-butyl- amine had been adsorbed shows two strong endothermic peaks at ca. 358 and 678 K which are due to the desorption and decomposition of n-butylamine.22 Analysis of the higher-temperature peak for the desorption of n-butylamine from acidic surfaces has been reported and shows the presence of butene and ~ r o p e n e .~ ~ ~ ~ ~ A mass- spectrometric analysis of the desorption of n-butylamine (temperature-programmed) from mordenites at high temperature (ca. 678 K) shows the presence of substantial amounts of butene and small amounts of n-butylamine, propane, CO, and compounds of molecular weight 78, 92, 106 and 120, whose exact identity was not investigated.,, The spectra in the CH stretching region of adsorbed methylamine (not illustrated) at different outgassing temperatures are similar to those of n-butylamine but the formation of the band at ca. 2865 cm-l is not prominent in this case. In the present study the desorption product of methylamine was not investigated. DISCUSSION ACIDITY The spectra of ammonia adsorbed on the different catalysts used showed the characteristic bands of NH,+ at 1445cm-l with the disappearance of the acidic hydroxyl band (at 3609 cm-l) of the mordenites.It has been reported that the number of acidic hydroxyls showing the OH band at 3609 cm-l decreases on dealumination of the mordenite and this results in a decrease in Brarnsted acidity found by pyridine adsorption.16 The dependence of the number of acidic hydroxyls on dealumination is reproduced from our earlier paper16 and it shows that Brarnsted acidity determined by NH, adsorption decreases on dealumination (fig. 3), and the present study shows that Brarnsted acidity determined from the characteristic bands of dSyr?(NH3+) of adsorbed methylamine and n-butylamine also decreases on dealumination. Non- aqueous titrations using Hammett indicators or thermometry in the characterization of zeolite acidity were found to be un~atisfactory~-~~ and doubts have been raised that a steric effect of n-butylamine in zeolite channels exists. The molecular diameter of n-butylamine is 0.49-0.54 nm (a = 0.44 x 0.77 nm) which makes it possible for n-butylamine molecules to enter and leave the large cavities of zeolite Y (free pore aperture 0.75 nm).Mordenite has virtually a unidimensional channel system (free pore aperture 0.67 nm) and it might be expected that n-butylamine molecules could enter the mordenite channels but may not move from one channel to another. It is important to investigate the accessibility of the base n-butylamine or methylamine to the acid sites of the zeolites. Mordenites possess two types of hydroxyls showing OH bands at 3609 and 3738 cm-l, the former being responsible for Brarnsted acidity.l4t l5 TheA. K.GHOSH AND G. CURTHOYS I07 1 I I 1 3000 2925 2850 2775 wavenumber/cm-' Fig. 4. CH stretching zone of adsorbed n-butylamine at different evacuation temperatures: (1) room temperature, (2) 373, (3) 473 and (4) 573 K.108 I.R. CHARACTERIZATION OF ZEOLITE ACIDITY former band was found to be more strongly acidic to pyridine while the latter band was shifted to a lower frequency.16 The present study also shows that the OH group which gives rise to the 3609 cm-l band of all the mordenites is strongly acidic as this band disappeared on adsorption of ammonia, while the OH group which gives rise to the 3738 cm-l band showed weak interaction (it may be hydrogen-bonded) as this broadened and shifted to a lower frequency.On adsorption of methylamine or n-butylamine these hydroxyl bands weakened in intensity in the case of the non- dealuminated mordenite and disappeared in the case of the dealuminated mordenites. This indicates that the amines used are accessible to the acidic hydroxyls. It is certain that n-butylamine or methylamine molecules can enter the channels of mordenite. However, the incomplete disappearance of the acidic hydroxyl band on adsorption of the bases (in the case of non-dealuminated mordenite) indicates that some hydroxyls were unreacted. Fig. 3 shows that the non-dealuminated mordenite possesses ca.4 times more hydroxyls than the highest dealuminated mordenite and this was reflected in the acidity determined by ammonia. However, the acidity determined by the adsorption of n-butylamine and methylamine shows a relative low value (ca. 2 or 3 times less, respectively). This can be interpreted as being due to the ' volume and space arrangement '24 of the alkylamine preventing adsorption occurring simultaneously on two adjacent sites, as shown below. RNH, RNH, H H H 0 0 0 I I I I I I I I I For this reason adsorption is not possible on site S,. Dealumination decreases the hydroxyl concentration and lessens this steric problem, depending on degree of dealumina tion. An alternative interpretation is as follows. Mordenite structurally has a two- dimensional channel system for small molecules and is considered unidimensional for larger molecules. The large pore has a cross-section of 0.67 x 0.70 nm while the small pore has a cross-section of 0.29 x 0.57 nm.Ammonia has a kinetic diameter of 0.26 nm which suggests ammonia should be able to penetrate the small-pore system, whereas n-butylamine molecules cannot penetrate the small-pore system and interact only with the acid sites available in the larger channels. It may be possible that the mouth of the unidimensional channel (for n-butylamine) is blocked by n-butylamine molecules so that other n-butylamine molecules cannot interact with the remaining acid sites inside the channels. It is known that dealumination widens the channels of the mordenite, which may remove or decrease the pore-blockage problem.The hydroxyls of HY giving rise to bands at 3650 and 3560 cm-l have been reported as being acidic to ammonia and ~yridine.,~ The present results show that the 3643 and 3568 cm-l hydroxyls are acidic, and the bands disappeared on adsorption of the bases showing that HY has no steric hindrance towards the bases used.A. K. GHOSH AND G. CURTHOYS 109 DECOMPOSITION OF n-BUTYLAMINE The desorption of n-butylamine at elevated temperatures shows that the desorption product contains mainly butene (maximum at 623 K) which suggests a Hofman degradation reaction :,O CH,-CH,-CH CH, NHg-Ozeolite -+ NH, + CH,-CH,-CH=CH2 + HOzeolite. 12 1 J - H The same reaction mechanism has been proposed for the decomposition of adsorbed n-butylamine on sili~a-alumina~~ and alumina.24 In the present study our mass- spectrometric analysis was not able to detect NH, because of its low molecular weight.From an infrared study of the decomposition of n-butylammonium-Y and dimethylammonium-Y, Jacobs and Uytterhoeven20 reported the presence of alkoxy groups and proposed the following mechanism for their formation : CH,-(CH,),-NHi-Ozeolite -+ CH,-(CH,),-Ozeolite + NH,. In the present study propane and CO, were observed during desorption of amine indicating the elimination of butoxy groups by hydrolysis.21 In the present study we were not able to detect CO because the computer program for the mass spectrometer deleted the mass of 28. M. W. Tamele, Discuss. Faraday Soc., 1950, 8, 270. 0. Johnson, J. Phys. Chem., 1955,59, 827.H. A. Benesi, J . Phys. Chem., 1957, 61, 970. Y. Trambouze, C.R. Acad. Sci., 1951, 233, 648; Y. Trambouze, L. de Mourgues and M. Perrin, J . Chim. Phys., 1954, 51, 723; C.R. Acad. Sci., 1953, 236, 1023. K. V. Topchieva, I. F. Moskovkaya and N. A. Dobrokhotova, Kinet. Catal., 1964, 5,910. ' K. Tanabe and T. Yamaguchi, J. Res. Inst. Catal., Hokkaido Univ., 1966, 14, 93. ' T. Morimoto, J. Imai and M. Nagao, J. Phys. Chem., 1974,78, 704. * R. Bezman, J. Catal., 1981, 68, 242. lo A. K. Ghosh and G. Curthoys, J . Chem. Soc., Faraday Trans. I , 1983,83, 147. l1 W. F. Kladnig, J. Phys. Chem., 1979, 83, 765. l2 D. Barthomeuf, J. Phys. Chem., 1979,83, 766. l3 M. Deeba and W. K. Hall, J . Catal., 1979,60, 417. l 4 P. E. Eberly Jr, C. N. Kimberlin Jr and A. Voohies Jr, J . Catal., 1971, 22, 419. l5 B. I. Shikunov, L. 1. Lafer, V. I. Yakerson and A. M. Rubinshtein, Izv. Akad. Nauk SSSR, Ser. Khim. l6 A. K. Ghosh and G. Curthoys, J . Chem. Soc., Faraday Trans. I , 1983, 79, 805. l 7 R. M. Barrer and D. L. Peterson, Proc. R . SOC. London, Ser A, 1964, 280, 466. l9 B. Chenon and C. Sandorfy, Can. J. Chem., 1958,36, 1181. 2o P. A. Jacobs and J. B. Uytterhoeven, J. Catal., 1972, 26, 175. 21 E. L. Wu, T. E. White and P. B. Venuto, J. Catal., 1971, 21, 384. 22 A. K. Ghosh and G. Curthoys, J. Phys. Chem., in press. 23 M. Takahashi, Y. Iwasawa and S . Ogasawara, J. Catal., 1976, 45, 15. 24 J. Koubek, J. Volf and J. Pakk, J. Catal., 1975, 38, 385. 25 P. A. Jacobs, Carboniogenic Activity of Zeolites (Elsevier, Amsterdam, 1977), p. 47. 26 J. R. Kiovsky, W. J. Goyette and T. M. Notermann, J. Catal., 1978, 52, 25. J. J. Fripiat, A. Servais and A. Leonard, Bull. SOC. Chim. Fr., 1962, 635. Nauk, 1973, 16, 449. J. W. Ward, Adv. Chem. Ser., 1970, 101, 380. (PAPER 3/629)

ISSN:0300-9599    

DOI:10.1039/F19848000099

出版商:RSC    

年代:1984

数据来源: RSC

摘要:

J. Chem. SOC., Faraday Trans. I, 1984,80, 1 I 1-1 17 Electron Mean Free Paths from Quantitative X-Ray Photoelectron Spectroscopic Studies of a Modified Thionine Electrode BY W. JOHN ALBERY* AND A. ROBERT HILLMAN Department of Chemistry, Imperial College of Science and Technology, South Kensington, London SW7 2AY AND RUSSELL G. EGDELL AND HOWARD NUTTON Inorganic Chemistry Laboratory, South Parks Road, Oxford OX1 30R Received 22nd April, 1983 Thionine multilayer films deposited on Pt and Sn-doped In,O, electrodes at oxidizing potentials display electrochemical activity similar to that of free thionine; the film thickness may be measured from cyclic voltammetry of the surface-bound redox centres. Attenuation by the thionine films of substrate core photoemission in X-ray photoelectron spectroscopy is shown to conform to a Beer-Lambert law.Electron mean free paths in the films for different electron kinetic energies have been measured and found to be in reasonable agreement with theoretical predictions. This agreement may be contrasted with pathlengthsmeasured in Langmuir-Blodgett films, which are considerably larger than those predicted by theory. The surface sensitivity of X-ray photoelectron spectroscopy (X.P.S.) derives from the short pathlength (typically < 100 A) of low-energy electrons in solids; development of the technique as a method for quantitative analysis demands a knowledge of the attenuation behaviour of electrons in solids. With a continuum model for the solid, the flux of electrons with energy E, I(E), emerging from a slab of material of thickness 1 may be related to the incident flux Io(E) through a Beer-Lambert relationship of the type where A(E) is the mean free path of electrons with energy E.Several compilations of mean free pathlengths may be found in the literature,lP3 along with critical discussion of their method of determination. Particularly attractive is the overlayer technique, wherein one measures the attenuation of a substrate core photoelectron peak as a function of the thickness of a well characterized overlayer. Recent interest in this method has centred on multilayer films of medium- or long-chain carboxylic acids and their salts deposited on substrates by Langmuir-Blodgett techniques.l-' These films can be characterized by capacitance measurements4 or by ellipsometry .5 Pathlengths of ca.50 A at an electron energy of ca. 1000 eV appear typical for these films, although a pathlength closer to 100 A has been reported for 1 keV electrons in a polydiacetylene carboxylic acid film.6 These values are all substantially larger than the mean free path of ca. 10 A estimated from the empirical 'universal curve' of Seah and D e n ~ h ; ~ they are also larger than measured pathlengths in graphite.8* In the present work we use the electrochemical oxidative polymerisation of thionine1*l3 to prepare films for attenuation studies of known thickness. Thionine = exp [ - W91 (1) 111112 MODIFIED THIONINE ELECTRODES (Th) and leucothionine (L) form a two-electron redox couple as shown in reaction (1) Th L At potentials > 1.0 V [relative to the saturated calomel electrode (SCE)] thionine may be irreversibly oxidized, leading to the build up of a surface-bound polymer film on the electrode.As described below the total number of surface-bound redox centres may be found by integrating the charge from a cyclic voltammogram obtained at a scan rate of < 100 mV s-l. In the preparation of the electrodes it is found that the longer the coating time the greater is the number of redox centres attached to the electrode. Thus this system provides a method for preparing reproducible films of different and known thicknesses. We report here attenuation studies for thionine overlayers on platinum and tin-doped In,O, electrodes. These two substrates provide reasonably strong core signals in X.P.S.over a wide range of electron-kinetic energies, enabling us to explore the energy dependence of the electron mean free paths. EXPERIMENTAL All electrochemical experiments were carried out at 25 OC and all potentials are quoted relative to the SCE. Polymer films were deposited on either platinum or tin-doped In,O, by holding the electrodes at a potential of 1.1 V in 0.05 mol dm-, H,S04 containing ca. 30 pmol dm-, thionine for varying times. The thickness of the film was estimated by cyclic voltammetry of the washed coated electrode in 0.05 mol dm-3 H,SO, electrolyte. At slow scan rates the peak height varied linearly with sweep rate, showing that the whole of the layer was reduced or oxidised.ll Following correction for double-layer charging, the number of surface-bound redox centres could be obtained from the area of the cyclic voltammograms.We have previously shown12 that there is good agreement between the number of redox centres and a measurement from the ring current of a ring-disc electrode of the amount of thionine deposited on the disc electrode. To convert the number of redox centres (as measured from the charge) into an overlayer thickness we estimate from the molecular geometry that each redox centre in the film occupies a volume of 200 A. We assume that there are sulphate counterions in the 1a~er.l~ Preliminary X-ray photoelectron spectra were measured in an AEI ES200B spectrometer with a Mg anode operating at 228 W and a base pressure of 2 x lo-@ Torr.* It was established in the preliminary experiments that desorption and decomposition of the films was of minor importance: cyclic voltammograms taken after X.P.S.experiments indicated that exposure to u.h.v. for ca. 12 h and X-irradiation for up to ca. 8 h only reduces the coverage of surface redox centres by ca. 10%. Quantitative attenuation studies were conducted in an ESCALAB 5 spectrometer (V. G. Scientific) again using Mg Ka radiation at an anode power of 240 W. The total time for accumulation of spectra was typically < 4 h per sample. The spectrometer base pressure was 5 x lo-" Torr, although owing to rapid sampling throughout typical operating pressures were closer to 5 x Torr. Photoelectrons emerging within a cone 5' from the surface normal werecollected by the analyser input lens; we assume strictly normal photoemission in the derivation of pathlengths. Following stripping of satellite structure and the background of inelastically scattered electron^,'^ areas were estimated by numerical integration of narrow (25 eV) scans over the peaks of interest.* 1 Torr = 101 325/760 Pa.W. J. ALBERY, A. R. HILLMAN, R. G. EGDELL AND H. NUTTON 113 RESULTS AND DISCUSSION Typical wide-scan photoelectron spectra for thionine films of various thicknesses on a platinum electrode are shown in fig. 1 along with the cyclic voltammograms obtained in 0.05 mol dm-3 H,SO, background electrolyte. Increasing film thickness is monitored by the increasing area of the cyclic voltammograms. Note the progressive attenuation of Pt 4fand Pt 4d signals in X.P.S., together with the growth of the S 2p I I 1 100 PA I H 100mV 0 250 500 750 binding energy/eV Fig.1. Wide-scan photoelectron spectra of Pt coated with thionine films of varying thickness: (a) 4; (b) 19 and (c) 49 A. Cyclic voltammograms of the films are shown adjacent to the spectra: film thickness is proportional to the area of the voltammogram. and N 1s signals characteristic of the thionine overlayer. Fig. 2 shows a Beer-Lambert plot for the attenuation of the Pt 4fsigna1, demonstrating that eqn (1) is obeyed over almost two orders of magnitude of variation in the electron flux I(E). These data yield L(E) = 16 A at E = 1180 eV. We can also use the N 1s and S 2p peaks, which are caused by atoms in the overlayer itself. In fig. 1 these peaks increase in intensity as the coat becomes thicker.However,114 3 - MODIFIED THIONINE ELECTRODES (0) - 1 - -2 - n s c 3 - 3 - -4 - - 5 - 1 1 I 1 0 20 LO 60 Fig. 3. Data for the intensities of S 2p and N 1s peaks plotted according to eqn (2). the strength of these signals is not linear in film thickness as self-attenuation of the photoelectron flux from atoms in the film leads to a limiting intensity Im(E). For a film thickness I the photoelectron intensity I(E) is given byW. J. ALBERY, A. R. HILLMAN, R. G . EGDELL AND H. NUTTON 115 Hence a plot of log [Ioo/(loc - I ) ] against I should be linear with slope l / A . Fig. 3 confirms that the N 1s and S 2p peaks conform to this behaviour, enabling us to derive values for the electron pathlengths at the kinetic energies characteristic of atoms in an overlayer itself.The range of electron kinetic energies is further extended by the study of thionine on Sn-doped In,O,; similar results were obtained to those above. Here the metal 4d, 3d and 3p photoelectron and MNN Auger signals span a range of kinetic energies from 1230 to 400 eV. The results are collected in table 1. In fig. 4 we plot the variation with Table 1. Kinetic energies and pathlengths 2 . 0 ’ core peak kinetic energy/eV pathlength/A 1237 1227 1180 1089 8 54 809 80 1 767 758 587 550 400 25.5 29.0 15.6 15.8 12.6 17.8 18.4 13.9 13.7 14.6 8.5 8.3 V V 2 . L 2.6 2.8 3 .O 3.2 log (EIeV) Fig. 4. Electron mean free paths in thionine films on Pt (0) and In203 (0) electrodes. Experimental data for graphite (A)8, and Langmuir-Blodgett multilayers (V) are also The empirical ‘universal curve’ of Seah and Dench is shown by the broken line assuming that rn = 3 %i in eqn (3).The full line is calculated for the model of Leckey and ~ o w o r k e r s ~ ~ * ~ ~ from eqn (4). The following parameter values are used: p = 2.70 g A = 325, Z = 112 and Eg = 2.1 eV. 5 FAR 1116 MODIFIED THIONINE ELECTRODES kinetic energy of the electron mean free path ( A ) in thionine films on both Pt and In,O, substrates together with additional experimental data points for Langmuir-Blodgett m~ltilayers~-~ and for graphite.s. We have also plotted the empirical ‘universal curve’ of Seah and D e n ~ h , ~ which is given by ( 3 ) where A(E) is in A, E in eV and m is the thickness of one monolayer. This empirical curve is constructed from the scattered data for some 300 different compounds.Our data lie within the scatter of the points used to construct the curve. A more fundamental equation based on electron scattering by bulk jellium which relates the electron pathlength to the kinetic energy has been derived by Leckey and coworkers:15y l6 In this equation all energies are in electronvolts and Ep is the plasma energy, given A(E) = 538 mE++0.13 (m3 E): A(E) = 1.8 E E f / E t . (4) Ep = 28.8 ( p Z / A ) : ( 5 ) by where p is the density, A is the molecular weight and 2 is the number of valence electrons per molecule. In eqn (4) Eis the centroid of the optical loss function, which for the present purpose can be approximated byl5t l6 E = Ep+Eg (6) where the bandgap Eg is taken as the energy of the first U.V.absorption band. A plot of eqn (4) is shown in fig. 4, and it is gratifying that our data for films of the same material, but interrogated with electrons of different energy, lie close to the theoretical line, thereby confirming the theoretical model. The similarity of the pathlengths in our films to those in graphites* is striking, and provides support for a model of the polymer layer in which the aromatic units are oriented with the plane of their rings parallel to the surface. It is also apparent from fig. (4) that the pathlengths in thionine films are very much shorter than those in Langmuir-Blodgett multilayers. This is in agreement with the hypothesis that the long pathlengths in Langmuir-Blodgett films are caused by channelling effects peculiar to systems where long-chain molecules are oriented normal to a surface and are not a general feature of organic films.In conclusion, this work has shown first that electron scattering in compact organic films obeys a simple Beer-Lambert model and secondly that electron pathlengths for thionine films are in reasonable agreement with the model of Leckey and coworkers.15~ l6 The equipment was funded by the S.E.R.C. We thank Shell Research Ltd and I.C.I. for the provision of two Research Fellowships for R. G. E. and A. R. H., respectively. This is a contribution from the Oxford Imperial Energy Group and the Wolfson Unit for Modified Electrodes. C. J. Powell, SUI$ Sci., 1974, 44, 29. C. R. Brundle, J . Vac. Sci. Technol., 1974, 11, 212. M. P. Seah and W. A. Dench, Surf: Interface Anal., 1979, 1, 2. D. T. Clark, Y. C. T. Fok and G . G. Roberts, J. Electron Spectrosc., 1981, 22, 173. S. M. Hall, J. D. Andrade, S. M. Ma and R. N. King, J . Electron Spectrosc., 1979, 17, 181. B. Hupfer, H. Schupp, H. Ringsdorf and J. D. Andrade, J . Electron Spectrosc., 1981, 23, 103. C. R. Brundle, H. Hopster and J. D. Swalen, J. Chem. Phys., 1979,70, 5190. R. G. Steinhardt, J. Hudis and M. L. Perlman, Phys. Rev. B, 1972, 5, 1016. 13 K. Jacobi and J. Holzl, Surf: Sci., 1971, 26, 54.W. J. ALBERY, A. R. HILLMAN, R. G. EGDELL AND H. NUTTON 117 lo W. D. Albery, A. W. Foulds, K. J. Hall, A. R. Hillman, R. G. Egdell and A. F. Orchard, Nature l 1 W. J. Albery, W. R. Bowen, F. S. Fisher, A. W. Foulds, K. J. Hall, A. R. Hillman, R. G. Egdell and l2 W. J. Albery, A. W. Foulds, K. J. Hall and A. R. Hillman, J. Electrochem. SOC., 1980, 127, 654. l 3 W. J. Albery, M. G. Boutelle, P. J. Colby and A. R. Hillman, J. Electroanal. Chem., 1982, 133, 135. l4 N. Beatham and A. F. Orchard, J. Electron Spectrosc., 1976, 9, 129. l5 J. Szajman and R. C. G. Leckey, J. Electron Spectrosc., 1981, 23, 83. l6 J. Szajman, J. Liesegang, J. G. Jenkin and R. C. G. Leckey, J. Electron Spectrosc., 1981, 23, 97. (London), 1979, 282, 793. A. F. Orchard, J. Electroanal. Chem., 1980, 107, 37. (PAPER 3/642) 5-2

ISSN:0300-9599    

DOI:10.1039/F19848000111

出版商:RSC    

年代:1984

数据来源: RSC

摘要:

J . Chem. Soc., Furuduy Trans. 1, 1984,80, 119-128 Mechanism of Photo-oxidation of Propene over Vanadium Oxide Supported on Silica BY SATOHIRO YOSHIDA,* TSUNEHIRO TANAKA, MITSUO OKADA AND TAKUZO FUNABIKI Department of Hydrocarbon Chemistry, Kyoto University, Kyoto 606, Japan Received 25th April, 1983 The nature of the active oxygen species and intermediates in the photo-oxidation of propene over vanadium oxide supported on silica has been investigated and a possible mechanism is proposed. The phosphorescence spectra confirm that the active species is the photoexcited V=O bond in the catalyst. Aldehydes produced by the oxidation of propene with 1 8 0 2 contained little l80, showing that excited lattice oxygen is the species directly involved in attacking propene. A kinetic study of the photoadsorption of oxygen on a sample of the catalyst having preadsorbed propene showed that the first intermediate comprises one molecule of propene and one molecule of oxygen.The formation of a n-ally1 intermediate is suggested from the results of the photo-oxidation of deuterated propene. Photoassisted reactions over n-type semiconductors are of current interest, yet studies of the photo-oxidation of hydrocarbons have not been widely reported.l-' In particular, only a few papers have been published on the oxidation of alkenes, although the formation of alkenes as intermediates has been proposed in the oxidation of alkanes and alcohols.' In a previous communication5 we reported the formation of aldehydes by the oxidation of propene over u.v.-irradiated vanadium oxide supported on silica (V,O,/SiO,).The distribution of products was different from that in the oxidation of propene by 0; on partially reduced V,05/Si0,.8 The prereduction of the catalyst led to a decrease in the conversion of propene and formation of propenal. These facts suggest the direct participation of lattice oxygen excited by photons. We proposed that propene might be converted to surface complexes over the photoirradiated catalyst followed by the thermal decomposition of the complexes to aldehydes. In this paper we report experiments performed to verify these assumptions by using l80, and deuterated propene and we discuss the reaction mechanism. EXPERIMENTAL The catalyst was the same as that reported previously (V,O,/SiO,, V,O,; 5 wt %)., Commercial propene and normal oxygen gases were purified by vacuum distillation using a liquid-nitrogen trap.Heavy oxygen ( 1 8 0 2 ) was supplied from CEA-CEN Saclay (purity 99.88%) and deuterated propene was prepared by the method described in ref. (9). Its purity was found to be ca. 90% by infrared and mass spectroscopies. The apparatus was the same type as that described elsewhere,'O with a slight modification made by attaching a sampling chamber connected to a quadrupole mass spectrometer (Shimadzu MASPEQ-070) for an analysis of the products. The procedures for pretreatment of the catalysts and for photoirradiation were the same as those reported previously ; however, the procedure for collecting the products was modified. Thus after 0.5 h irradiation of the catalyst under a circulated mixture of propene and oxygen, the gaseous products were frozen 119120 PHOTO-OXIDATION OF PROPENE OVER VANADIUM OXIDE out by a liquid-nitrogen trap (step A); then the catalyst was irradiated again for 10 min to check photodesorption of the adsorbed products (step B).After the reactor had been evacuated for 10 min to 0.1 Pa, the catalyst was heated in the dark to 473 K (step C) and then to 573 K (step D). The desorbed products collected in each step were analysed in the mass spectrometer. The photoadsorption and photodesorption of propene and oxygen were investigated by using an apparatus comprised of a small quartz vessel and a calibrated Pirani gauge. Phosphorescence spectra of the catalyst were recorded by a Shimadzu RF-501 spectrometer.RESULTS AND DISCUSSION ACTIVE OXYGEN SPECIES Negatively charged adsorbed oxygen species [0-(ads) and O;(ads)] are often considered to be the active oxygen species in photo-oxidation over n-type semiconductors. On the other hand, the role of excited lattice oxygen has also been discu~sed.~*~~ So far the identity of the active oxygen species in the photo-oxidation of alkenes over metal oxides has not been clarified.* We attempted this for the photo-oxidation of propene over V,O,/SiO, in an oxidised state by obtaining e.s.r. spectra of the u.v.-irradiated catalysts using a JEOL JES-3BS-X spectrometer, but no evidence for the formation of 0- and 0; was obtained. We also found that there was no equilibration between lSO, and 180, over the u.v.-irradiated catalyst.These facts strongly suggest that the excited lattice oxygen species participates in the photo- oxidation, since 0- is known to be an active species in the equilibrium reaction.16 Direct evidence for the participation of excited lattice oxygen was obtained from an analysis of the products of photo-oxidation using 180,. In fig. 1 a mass spectrum of the products is shown. Very small peaks were observed at m/e 31 and 45, corresponding to CH180+ and CH,C180+ fragments, respectively, produced from aldehydes. Since very pure leO, was used in this experiment this result shows that the 40 h m * ._ c 'j: 30 -e W c1 .c Do .- 2 20 Y W a 10 + 0 2 r u + 0 u % N r 2 + 0 0 7 u t + a 3 I m 0 c o r u " '0 m " I 5 : : a3 c I V 29 30 31 43 44 45 46 mle Fig. 1. Mass spectrum of products obtained from the photo-oxidation of propene using l*O,.S.YOSHIDA, T. TANAKA, M. OKADA AND T. FUNABIKI 121 oxygen atoms incorporated into the aldehydes do not come from gaseous oxygen, but rather from lattice oxygen in the catalyst. After repeating the photo-oxidation with lS0, several times over the same catalyst, l6O, was used. The aldehydes produced in the last run contained l*O, showing that the oxygen species involved in direct attack on the propene is lattice oxygen. In this connection note that the conversion of propene was much influenced by the partial pressure of oxygen, as shown in fig. 2, and the level of propene conversion was very low in the absence of gaseous oxygen. This result suggests that gaseous oxygen is essential to photo-oxidation.I 1 I I 1 I I I 1 10 20 30 40 50 60 70 80 initial amount of oxygen/pmol Fig. 2. Influence of initial amount of oxygen on conversion of propene after 30 min irradiation. Initial amount of propene, 37 pmol. Recently, Cunningham et aZ. 1 7 9 l8 reported an enhancement of the photo-oxidation of alcohols by gaseous oxygen over ZnO and TiO, in which photoexcited lattice oxygen was assumed to be the active oxygen species. To clarify the role of gaseous oxygen we have investigated in detail the photoadsorption of propene and oxygen on our catalyst. PHOTOADSORPTION OF PROPENE AND OXYGEN The photoadsorption of propene and oxygen was measured at low pressures (ca. 1 Pa) to prevent oxidation as much as possible. A small amount of oxygen was adsorbed onto the oxidised Catalyst in the dark at room temperature, but no enhancement of adsorption by U.V.irradiation was observed. The sites for adsorption in the dark are surface vanadium@) ions formed in the pretreatment procedure (heating at 673 K under an oxygen atmosphere for 2 h and evacuation at 673 K for 30 min). On the other hand, U.V. irradiation greatly enhanced the adsorption of propene. Adsorption appears not to be very extensive since almost all the adsorbed propene was removed during 30 min evacuation in the dark. Phosphorescence spectroscopy can provide information about the adsorption sites of gases.19 The spectrum of V,O,/SiO, was first reported by Kazanskii et aZ.,,O who assigned it to the transition122 PHOTO-OXIDATION OF PROPENE OVER VANADIUM OXIDE from the excited triplet state to the ground state of the V=O bond.Similar spectra were obtained for the present catalyst at room temperature in vacuo. When propene was contacted with the catalyst, the phosphorescence was quenched significantly, indicating that the site of adsorption of propene is the excited V=O bond. Although no appreciable photoadsorption of oxygen was observed on the bare catalyst, the catalyst preadsorbing propene was found to cause the photoadsorption of oxygen as shown in fig. 3. After propene was photoadsorbed, gaseous propene was . o i I I I n I \ . o . o I I I I I I I I I I I I D ' UV-39 I UV - 3 1 I D k l '3c)a I I I I I 0 10 20 30 40 time/min Fig. 3. Change in the pressure of oxygen following u.v.-irradiation over V,O,/SiO, with preadsorbed propene (41 nmol).D, in the dark; UV-39, UV-31, see text. evacuated for 1 min and oxygen was introduced into the vessel. The recording of pressure change was started immediately. Photoirradiation with a UV-39 glass filter (transmitted wavelength > ca. 390 nm) caused a small pressure change, but irradiation with a UV-3 1 glass filter (transmitted wavelength > ca. 3 10 nm) caused an appreciable change. This shows that the effective wavelength is in the range 390-310 nm. The adsorption curve using UV-3 1 was analysed as follows. Let us assume the photoadsorption process as hv v=o e (V=O)* (v=o)* + C3H6 -+ (v=o) ' C3H6 (v=o) C3H6 + 0, -+ (v=o) ' C3H6 ' 0,. (2) (3) hv Provided that there is no interference between (V=O) - C3H6 and (v=o) * C3H6 - o,, the rate equation for the photoadsorption of oxygen, which we observed in this experiment, should be rate = -- dp - k[C3H6(adS)b (4) dtS.YOSHIDA, T. TANAKA, M. OKADA AND T. FUNABIKI 123 where k, [C,H,(adS)] and p are the rate constant, the concentration of (V=O)-C,H, and the pressure of oxygen, respectively. Let [C,H,(adS)] and p be u and po at t = 0, respectively, then and the following integrated form of eqn (4) is obtained: ( 5 ) - [C3H6(ads)1 = c(PO In - = k(Cp,+v)t. Po VlC+P-Po Fig. 4 shows that the experimental data fit eqn (6) within the first 10 min. The deviation from the expected line after 10 min would result from the oxidation of propene. 8 . 6. 4 . 2 . 0 5 10 15 20 time/min Fig. 4. Verification of rate equation, eqn (6). When a mixture of propene and oxygen was introduced and the catalyst was u.v.-irradiated, the pressure rapidly decreased in the initial stages, but the pressure decrease became slow after 3 min.The first rapid change would correspond to fast propene adsorption and the following gradual change would result from the same phenomena described by eqn (3). The results mentioned above show that first propene is adsorbed on the excited V=O bond and then oxygen is adsorbed to form an intermediate comprising one molecule of propene and one molecule of oxygen. THERMAL EFFECTS Although the photo-oxidation of hydrocarbons and alcohols in the steady state over TiO, and ZnO has been studied at room temperat~re,~? l7 it seems probable that polar products such as aldehydes are partially adsorbed and that the desorption of products requires heat treatment of the catalyst after photoirradiation. In fact heating is necessary to collect all the products in our case, as reported previ~usly.~ The heating procedure may enhance not only the desorption of aldehydes but also the oxidation of intermediates, if photoirradiation mainly affects the formation of the latter.Thus it is important to distinguish photo-oxidation from the thermal oxidation of124 PHOTO-OXIDATION OF PROPENE OVER VANADIUM OXIDE intermediates in order to clarify the essential effects of photoirradiation. Quantitative discrimination is very difficult, because even at room temperature the thermal effects may not be negligible. In this work we attempted a qualitative discrimination by analysing the products collected stepwise as described in the experimental section.The substances collected were unreacted propene, H,O, CO,, ethanal, propenal and propanal; the amount of propanal was much smaller than those of ethanal and propenal. In fig. 5 the relative amounts of ethanal and propenal are shown. In step A only 12 and 13 % of the total amounts of propenal and ethanal respectively, were collected. In step B the photodesorption of adsorbed species was examined. In this procedure the temperature of the catalyst rose by ca. 10 K. It is characteristic that the amount of photodesorbed ethanal is 2.7 times that of propenal. The heat treatment of the catalyst which followed (step C ) resulted in the evolution of amounts of aldehydes. This shows that the formation of aldehydes (especially propenal) from intermediates requires thermal energy.Some ethanal should be formed by a pure photoprocess. l 4 4 . 0 1 I I CH,=CHCHO C H ,CHO +-’ 5 3.0 2 CQ I B 1 D In Fig. 5. Relative amount of propenal and ethanal normalised to the amount of propenal collected by step B. Steps A-D, see text. PHOTO-OXIDATION OF DEUTERATED PROPENE In the thermal oxidation of propene over transition-metal oxides the formation of n-ally1 intermediates has been established,,l but no definite information has been reported concerning intermediates in the photo-oxidation. [2H3JPropene has been used to investigate the structure of these intermediates;,l, 22 and we adopted this technique in the present work. The analysis of the distribution of deuterium was carried out for the products collected by a combination of steps A and B (step A + B) and by step C .The pattern coefficients of the mass spectra were calibrated for [2H3]propene, ethanal and propenal prior to the analysis of the product mixture. The results are shown in table 1 . Obviously, the distribution of deuterium in propenal is the same in step A + B as in step C . The result suggests that propenal is formed thermally from a photogenerated intermediate. Since the thermal reaction may be slow at room temperature, the yield of propenal in step A + B is expected to be much smaller than that in step C. This is consistent with the results in fig. 5, which shows the low yield of propenal in step A + B. The distribution of deuterium in propenal is reasonably explained by scheme 1 , inS.YOSHIDA, T. TANAKA, M. OKADA AND T. FUNABIKI Table 1. Deuterium distribution in the products propenal and ethanal distribution (%) product step A + Ba step Ca CH,=CHCHO 40.4 39.8 CHD=CHCHO 37.8 36.4 CH,=CHCDO 21.8 23.8 CH,CHO 13.3 33.8 CH,DCHO 86.6 64.1 CHD,CHO trace 2.1 125 a See text. which the formation of a n-ally1 intermediate and the relative rate of hydrogen elimination k D / k , = 0.734 are assumed. In scheme I, the numbers show the relative probability of the preceding species reacting along the indicated path. CH2=CH-CH2D Z = k , / k , = 0.734 Y Y CHZ"CH"CH2 ,CH2-CH-CHD CH2=CHCHO- CHD=CHCHO CH2=CHCDO 41.2 % 39.2 % 19.6% Scheme 1. We have also estimated the deuterium distribution based on other possible paths, i.e. the path involving an epoxide intermediate accompanied by the abstraction of an allylic hydrogen or the parallel process via the epoxide and n-ally1 intermediates.However, the estimated values were not consistent with those found experimentally. In contrast, the different deuterium distribution in ethanal depending on the collecting procedure suggests that ethanal is formed via at least two paths as shown in scheme 2. Kubokawa et al. observed the double-bond fission of propene over irradiated molybdenum oxide supported on porous Vycor glass (PVG).' If the exchange of deuterium between ethanal molecules can be ignored, the path of double-bond fission (path 1) may form only CH,DCHO. Neither path 1 nor path 2 (via n-ally1 intermediates) can explain the obtained deuterium distribution.However, we have found that the deuterium distribution in table I can be reproduced as shown in table 2 by assuming that both paths operate in the ratios 3.2: 1 and 0.57: 1 for the products collected by step A + B and step C, respectively. Table 2. Estimated deuterium distribution of ethanal assumed ratio distribution (%) collecting procedure path 1 :path 2 CH,CHO CH,DCHO CHD,CHO A+Ba 3.2 1 Ca 0.57 1 1 1 . 1 86.6 2.3 29.6 64.1 6.3 ~~~ ~ a See text.126 PHOTO-OXIDATION OF PROPENE OVER VANADIUM reaction path 1 reaction path 2 CH,=CH --CH2D CHz=CH- CH2D I double-bond CHZ-CH-CHZ CHZ-CH OXIDE 'CHD iz CH,DCHO CH3CHO CHZDCHO CHD2CH0 100 (L 46.4 5.: 43.8 '% 9.8 % Scheme 2. The expected deuterium distribution was derived as follows. Let a and b be the numbers of molecules of ethanal formed via paths 1 and 2, respectively. If the deuterium distribution expected from each path is assumed as shown in scheme 2, the overall distribution is calculated as CH,CHO: CH,DCHO: CHD,CHO = 0.464b/(a+b):(a+0.438 b)/(a+b):0.098 b/(a+b).(7) Equating the second term to the experimental value (0.866) leads to the ratio a/b = 3.19. The first and third terms are calculated using this ratio. The calculated deuterium distribution is consistent with the experiment results as shown in table 1. The ratio of path 1 to path 2 is 3.2: 1 for step A+B, in which the thermal effect is expected to be very low, reflecting the fact that the contribution of path 1 is much higher than that of path 2. On the other hand, the ratio 0.59: 1 for step C shows that the relative contributions of paths 1 and 2 are inverted. Thus this result suggests that double-bond fission is mainly caused by a photoprocess. REACTION MECHANISM From the results mentioned above we propose a probable reaction mechanism.The active oxygen species in the photo-oxidation of propene is considered to be the excited triplet state of the V=O bond of the V,O,/SiO, catalyst. The same active species has been proposed for the photo-oxidation of CO19 and CH,.23 Scheme 3 shows the probable reaction paths in the photo-oxidation of propene. Propene should interact with the oxygen atom in the excited V=O bond, since the n-interaction of propene may be favoured by the photogenerated, less negatively charged oxygen. Photoadsorption of molecular oxygen occurs only on the catalyst having preadsorbed propene to form a one-to-one complex of propene and oxygen.This adsorption of oxygen may result from an increase in the electron density of the vanadium atom by electron donation from the adsorbed propene. Products may be formed from the one-to-one complex by two paths. The first involves the direct fission of the propene double bond to yield ethanal. The second involves the formation of a n-ally1 intermediate. The present results indicate that thermal energy is required for the conversion of the n-ally1 intermediate to propenal and ethanal. A characteristic of the photo-oxidation of propene over V,O,/SiO, seems to be the formation of a n-ally1 intermediate at low temperatures, leading to aldehydes with fairly high selectivity.S. YOSHIDA, T.TANAKA, M. OKADA AND T. FUNABIKI 127 01 II hv o'b'o h v V - HzCyCHCH3 + HCHO + CI13CHOL H C- CH -C H . . . . . . . . 0 II / I \ 1 " , V + HCHO + CH,CHO[ Scheme 3. We thank Prof. Yamamoto and Dr Tsuchida of the Department of Polymer Chemistry of Kyoto University for their help in obtaining the phosphorescence spectra. This work was partially supported by a grant-in-aid for scientific research from the Japanese Ministry of Education. M. Formenti and S. J. Teichner, in Catalysis (Specialist Periodical Report, The Chemical Society, London, 1978), vol. 2, p. 87. * M. Formenti, F. Juillet, P. Meriaudeau and S. J. Teichner, Bull. SOC. Chim. Fr., 1972, 1, 69. M. Formenti, F. Juillet, P. Meriaudeau and S . J. Teichner, Chem. Tech., 1971, 1, 680.P. Pichat, J-M. Herrmann, J. Disdier and M-N. Mozzanga, J. Phys. Chem., 1979, 83, 3122. S. Yoshida, Y. Magatani, S. Noda and T. Funabiki, J. Chem. SOC., Chem. Commun., 1981, 601. Y. Kubokawa, M. Anpo and C. Yun, 7th. Int. Congr. Catal. (Tokyo, 1981) preprint B36. M. Anpo, I. Tanahashi and Y. Kubokawa, J. Chem. Soc., Faraday Trans. I , 1982,78, 2121. 1564. C. D. Hurd and J. L. Azorlosa, J. Am. Chem. SOC., 1951, 73, 33. lo S. Yoshida, Y. Matsumura, S. Noda and T. Funabiki, J. Chem. SOC., Faraday Trans. 1,1981,77,2237. V. S . Zakhrenko, A. E. Cherkasin, N. P. Keiler and G. F. Gerasimova, Kinet. Kaial., 1975, 16, 174. l 2 A. Thevenet, F. Juillet and S . J. Teichner, Jpn J. Appl. Ph-vs., Suppl. 2, 1974, 2, 529. A. Walker, M. Formenti, P. Meriaudeau and S. J. Teichner, J. Catal., 1977, 50, 237. 13 S . Yoshida, T. Matsuzaki, T. Kashiwazaki, K. Mori and K. Tarama, Bull. Chem. SOC. Jpn, 1974,47,128 PHOTO-OXIDATION OF PROPENE OVER VANADIUM OXIDE l4 J. Cunningham, D. J. Morrissey and E. L. Goold, J. Catal., 1978, 53, 68. l5 J. Cunningham, €3. Doyle and E. M. Leahy, J. Chem. SOC., Faraday Trans. I , 1979, 75, 2000. l7 J. Cunningham and B. K. Hodnett, J. Chem. Soc., Faraday Trans. I , 1981,77, 2777. V. V. Nikisha, B. N. Shelimov, V. A. Shvets, A. P. Griva and V. B. Kazansky, J. Catal., 1973,28,230. J. Cunningham, M. Ilyas and E. M. Leahy, J. Chem. Soc., Faraday Trans. I , 1982,78, 3297. M. Anpo, I. Tanahashi and Y. Kubokawa, J. Phys. Chem., 1980,84, 3440; 1982,86, 1 . 2o A. M. Gritscov, V. A. Shvets and V. B. Kazansky, Chem. Phys. Lett., 1975,35, 51 1 . 21 D. J. Hucknall, Selective Oxidation of Hydrocarbons (Academic Press, London, 1974), chap. 3. 22 C. R. Adams and T. J. Jennings, J. Catal., 1963, 2, 63. 23 S. L. Kaliaguine, B. N. Shelimov and V. B. Kazansky, J. Catal., 1978, 55, 384. (PAPER 3/651)

ISSN:0300-9599    

DOI:10.1039/F19848000119

出版商:RSC    

年代:1984

数据来源: RSC

摘要:

J. Chem. SOC., Faraday Trans. 1, 1984,80, 129-134 The Role of [2 + 21 Cycloaddition-type Reactions in Catalysis Activation of H-H, C-H and C=C Bonds by Metal Complexes BY JAMES G. HAMILTON AND JOHN J. ROONEY* Department of Chemistry, The Queen’s University, Belfast BT9 5AG, Northern Ireland Received 25th April, 1983 It is argued that direct [2 + 21 cycloadditions and cycloreversions involving Mt=C and C=C species and quasi-metallacyclobutanes are the key steps in olefin metathesis. This theoretical concept is then extended in a novel fashion to describe the mechanisms of many other important reactions in homogeneous and heterogeneous catalysis, and especially the activation of C=C, H-H and C-H bonds by metal-alkyl complexes where the metal ion is electron deficient and the alkyl ligand assumes ‘carbenoid ’ character.Detailed investigations of the stericl and kinetic2 aspects of the ring-opening polymerization of norbornene and its various derivatives using olefin metathesis catalysts, including EtAlCl, have convinced us2y that the mechanism involves a direct [2 + 21 cycloaddition and cycloreversion, as shown in scheme 1 . C C c - c c=c [Mtl C [M tl- - -C [Mtl-C [ M t l - c C [Mt]=C ‘ I - + Scheme 1. Since this theory requires that a metallacyclobutane per se be only present as a transition-state species between reactant and product quasi-metallacyclobutane inter- mediates, it is in sharp contrast to the generally accepted idea of independent coordination of olefin and metallacyclobutane formation. If these steps are so distinct an initial orthogonal orientation of the Mt=C and C=C x orbitals is required, and this is implicit in various theories5 that contend that puckering of the metallacycle is the source of various stereospecific features of the overall process (scheme Scheme 2.Scheme 1, in contrast to scheme 2, has also recently received strong support from theoretical calculations,6 and indeed is clearly the only one possible when transients such as Al=C3 and P=C7 mediate metathetic reactions of olefins. That an Al=C species seems to form from EtAlC1, in norbornene and participates in formal 129130 [2 + 21-CYCLOADDITION REACTIONS IN CATALYSIS [2 + 21-type reactions indicates that the n orbitals, HOMO and LUMO, are essentially located on the A1 and C centres, respectively, such that the canonical representation of the mechanism in scheme 3 is a good approximation. c=c + [All 1 C [All-- C’ c ’ c c - c I + I / = / I - [All- C [All- ‘C Scheme 3.When the overlap integral for the p* atomic orbitals is small there is an inherent theoretical problem still unresolved uis-&uis the degree of concertedness and the application of Woodward-Hoffmann theory as opposed to a stepwise mechanism (cf scheme 3), and this has recently been discussed* for [2 + 21 cycloreversions. Indeed Hoffmann and coworkers6 have recently considered this problem in respect of the olefin metathesis mechanism and conclude that canonical forms depict better the quasi-metallacyclobutane predicted by the interaction of the Mt=C and C=C species. A case in point, which well illustrates this difficulty, is that the highly strained olefin adamantene,g which is essentially a 1,2-diradical, reacts with simple alkenes to form cyclobutane derivatives. The main criterion therefore for such [2 + 21-type reactivity of a general A-B species of this type is associated with a weak n component in the bonding in the unit and not its polarity, cf.Al+-C-, P+-C- and C’-C’, although polarity is obviously important in influencing energetic and steric aspects4? of the reaction. Such A-B species are also expected to be widespread in transition-metal chemistry where the n bond is now a @-p* or 8-8 rather than a pk-pn component, but the basic theory for [2+2]-type reactions still seems to apply. Our main purpose now is to argue that A-B species of the above type may not only participate directly in olefin insertion reactions (cf.schemes 1 and 3) but also in H-H and C-H activation by a similar [2 + 21 mechanism. Further examples of such reactive units are Mt-Mt pairs as found on metal surfaces and in metal cluster compounds, even dinuclear compounds, cation-anion pairs, e.g. the Al-0 redox sites in the surface of y-Al,O,, and perhaps even ‘carbenoid’ metal-alkyl units in electron-deficient complexes, as well as metallacarbenes. In all cases the n component in the A-B bond can be well represented by zwitterionic and diradical canonical forms and we will now use these to describe reactions where a [2 + 21-type mechanism seems to apply for each of these examples. \ [ M t l + H R\C/,H I - H, [Mtl ‘IM 11 ri“‘ R \ / I1 H-[Mtl R’C H~ RLC I = .(I [Mtl [ M t l Scheme 4.J.G . HAMILTON AND J. J. ROONEY 131 Because of the very close relationship of the ring-opening polymerization of cy- cloalkenes to Ziegler-Natta polymerization of alk- 1 -enes it was SuggestedlO that a hy- drido-metallacarbene is generated from a metal-alkyl species and that it is the former and not the latter which propagates the addition polymerization. However, there is now growing evidence that, while this may be correct in some cases,ll a ‘carbenoid’ metal-alkyl with an a-H bridging over to the metal, but not fully transferred to it, can be a ground-state species12 which may undergo olefin insertion,13 as shown in scheme 4. The a-H bridging species can first be considered as a three-membered ring. Complete rupture of the C-H bond results in one extreme, the hydridometallacarbene, but severance of the C-Mt bond gives a zwitterion or diradical.It is thus clear why such canonical representations which illustrate the II. character of the C-Mt bonding, >CCH H \ [M t i R H \ [Mtl’ are an important part of the description of the bridging ‘carbenoid’ metal-alkyl unit. It may even be useful to consider the analogous formation of a metallacyclopentane from two alkenes and a metal ion, e.g. tantalum,14 in the same manner, with emphasis on the three-membered ring representations of the metal-alkene complex, as it engages the second alkene molecule (scheme 5). c= c c- c C-$ c /-, +. ll--Pc C [M tl iMt1 iM t j - L “MtI/ 1 - \ / - - I Scheme 5. A most attractive feature of the above mechanism for Ziegler-Natta polymerization is that it may explain the very high stereoselectivities often observed, since there is evidence15 that H migration towards the metal from the a-methylene group (scheme 4) can be remarkably selective.The widening of the C-C-Mt angle required of the ‘carbenoid’ mode of behaviour of the metal-alkyl is also diametrically opposed to the narrowing of this angle which would be required if the chain transfer step of cis-P-H elimination is to take place, so a strong impetus in many cases towards formation of high polymer is understood. The [2+2] theory also leads naturally, without participation of dz orbitals, even though that may be highly beneficial for transition-metal catalysis, and without prior coordination of olefin, to a good mechanism for the oligomerization of ethylene mediated by Al(alkyl),, where the electron-deficient propagating species may be depicted by the following canonical forms : I R \ c / H H/I ‘*I / \ R R R‘ \ /H H 2- Several ‘carbenoid’ metal-alkyl species also seem to be capable of activating H-H and C-H bonds, as evidenced by a growing number of examples described in the132 [2 + 2]-CYCLOADDITION REACTIONS IN CATALYSIS , recent literature.Thus several do-Zr complexes, e g . (Cp),Zr(alkyl)Cl, react directly with dihydrogen,16 which is also used as a chain-transfer agent in typical Ti-based alk-1-ene polymerization systems.17 In the absence of d electrons H-H cannot be cleaved by the usual oxidative addition mechanism so a direct [2+2]-type attack also seems warranted16 (scheme 6).The molecular-orbital treatments of [2, + 2,] and [2, + 2,J cycloadditions are essentially equivalent. R CH, Scheme 6. Another very interesting example is the recently described metal-alkyl compound, Lu(q5-C5Me,),R, which not only catalyses propylene polymerizationla but also cleaves C-H bonds in tetramethylsilane at 20 OC.19 The most interesting example of all may well be that of alkylcobalamins, where there is a long-standing mystery of how paraffinic C-H bonds in substrates are activated in such a highly selective fashion in the vicinal interchange processes catalysed by B,, coenzyme.2o Since the C-C-Co bond angle is ca. 125” the alkyl group attached to the Co ion may well be ‘carbenoid’ in character and quite reactive in a [2+2]-type reaction with substrate (scheme 7), in the same way that the lutetium-alkyl compound reacts with tetramethylsilane.l9 [C 01 R’ [C 01 Scheme 7. Just as described for Ziegler-Natta polymerization (scheme 3), highly stereospecific behaviour will have its origins in the selectivity with which one of the two hydrogen atoms of the a-methylene group moves towards the metal thereby turning this group into a highly reactive chiral centre. Other homogeneous examples are intramolecular addition of a C-H bond across a Ta=C bond21 and dissociative addition of dihydrogen to a Ta=Ta pair of ions2, in a couple of tantalum compounds. In the heterogeneous field there are many such reactions, e.g. the carbanionic activation of paraffins even at room temperature, for deuterium exchange23 by the redox sites on y-Al,03 (scheme 8).+ R-H - A[-0 Al-0 Scheme 8. The [2+2]-type reactions have long since been recognized as occurring on metal surfaces but are called dissociative adsorption, e.g. of H,, or associative adsorption, e.g. of etmlene, on a contiguous pair of sites. An exciting extension of this type ofJ. G. HAMILTON AND J. J. ROONEY 133 mechanism may well reside in the recent discoveries of homologation reactions of paraffins24 and of 0lefins~~9 26 on various catalysts, one of simultaneously promotes olefin metathesis. This implies that adsorbed methylenes are formed from ethylene etc. on a pair of metal sites according to scheme 9, and then react with alkene by another [2 + 21 cycloaddition which is a key step in the homologation process.c c c-c [Mtl C MI- c Etc - l l + l l --I I Scheme 9. Even acetylenes are now known to fall apart to give metallacarbynes upon addition across a W=W triple bond as has been observed in recently described reactions of a dinuclear complex of tungsten.28 Heteronuclear units such as C=N and C=O behave in like fashion. The insight which this recent branch of catalysis and organometallic reactions gives into the mechanism of Fischer-Tropsch synthesis is extremely imp~rtant.~' Note added in proof: B e n ~ z e ~ ~ has reported that carefully purified EtAlCl, does indeed catalyse the cross-metathesis of norbornene and cis-pent-2-ene. Huu Thoi Ho, K. J. Ivin and J. J. Rooney, J. Mol. Catal., 1982, 15, 245. Huu Thoi Ho, B. S. R. Reddy and J. J. Rooney, J.Chem. SOC., Faraday Trans. I , 1982,78, 3307. K. J. Ivin, J. J. Rooney and C. D. Stewart, J. Chem. SOC., Chem. Commun., 1978, 603. J. G. Hamilton, K. J. Ivin, J. J. Rooney and L. C. Waring, J. Chem. SOC., Chem. Commun., 1983,159. M. Leconte and J-M. Basset, J. Am. Chem. SOC., 1979, 107, 7296. 0. Eisenstein, R. Hoffmann and A. R. Rossi, J. Am. Chem. Soc., 1981, 103, 5582. E. Schaumann and R. Ketcham, Angew. Chem., Int. Ed. Engl., 1982, 21, 225. W. Bums and M. A. McKervey, unpublished results pN. Bums, Ph.D. Thesis (Queens University of Belfast, 1976)l. lo K. J. Ivin, J. J. Rooney, C. D. Stewart, M. L. H. Green and R. Mahtab, J. Chem. SOC., Chem. Commun., 1978, 604. l 1 H. W. Turner and R. R. Schrock, J. Am. Chem. SOC., 1982, 104, 2331. l2 Z. Dawoodi, M. L.H. Green, V. S. B. Mtetwa and K. Prout, J. Chem. SOC., Chem. Comrnun., 1982, 1410. l3 D. T. Laverty and J. J. Rooney, J. Chem. SOC., Faraday Trans. 1, 1983, 79, 869. S. J. McLain, J. Sancho and R. R. Schrock, J. Am. Chem. SOC., 1979, 101, 5457. l5 W-K. Wong, W. Tam and J. A. Gladysz, J. Am. Chem. SOC., 1979, 101, 5440. l6 K. I. Gell, B. Posin, J. Schwartz and G. M. Williams, J. Am. Chem. SOC., 1982, 104, 1846. *' See J. Boor Jr, Ziegler-Nutta Catalysts and Polymerizations (Academic Press, New York, 1979), p. ' U. Khlabunde, N. F. Tebbe, G. W. Parshall and R. L. Harlow, J. Mol. Catal., 1980, 8, 57. 251. P. L. Watson, J. Am. Chem. SOC., 1982, 104, 337. I* P. L. Watson, J. Chem. SOC., Chem. Commun., 1983, 276. 2o See J. M. Pratt, in BZ2, ed. D. Dolphin (Wiley, New York, 1982), vol 1, p. 325. 22 A. P. Sattelberger, R. B. Wilson and J. C. Huffman, Znorg. Chem., 1982, 21, 4179. 23 C. S. John, C. Kemball, E. A. Pearce and A. J. Pearman, J. Chem. Res., 1979, (S) 400, (M) 4830. L. Chamberlain, I. P. Rothwell and J. C. Huffman, J. Am. Chem. Soc., 1982, 104, 7338.134 [2 -k 21-CYCLOADDITION REACTIONS IN CATALYSIS 24 C. O’Donohoe, J. K. A. Clarke and J. J. Rooney, J. Chem. Soc., Farada-v Trans. I , 1980, 76, 345; 25 F. Hughes, B. Besson, P. Gussiere, J. A. Dalmon and J-M. Basset, Nouv. J . Chim., 1981, 5, 207. 26 R. A. Strehlow and E. C. Douglas, J . Chem. Soc., Chem. Commun., 1983, 259. 27 P. P. O’Neill and J. J. Rooney, J . Am. Chem. SOC., 1972,94,4383. 29 L. Bencze, ISOM 5 (Graz, Austria. August 1983). J . Chem. SOC., Chem. Commun., 1979, 649. R. R. Schrock, M. L. Listemann and L. C. Sturgeoff, J. Am. Chem. Soc., 1982, 104, 4291. (PAPER 3/652)

ISSN:0300-9599    

DOI:10.1039/F19848000129

出版商:RSC    

年代:1984

数据来源: RSC

摘要:

J . Chem. SOC., Faraday Trans. I , 1984, 80, 135-152 X-Ray Photoelectron Spectroscopic Studies of the Iridium Electrode System BY H. YVONNE HALL AND PETER M. A. SHERWOOD* Department of Inorganic Chemistry, University of Newcastle upon Tyne, Newcastle upon Tyne NEl 7RU Received 28th April, 1983 X-ray photoelectron spectroscopy has been used to investigate the surface of iridium electrodes studied under both steady-state and cyclic conditions in 0.5 and 5 mol dmV3 sulphuric acid solutions. The results show that pulsing into the anodic region produced very small amounts of oxide in 0.5 mol dm-3 acid and slightly more in the 5 mol dm-3 solutions. The oxide is thought to be predominantly Ir02. When the electrode has been cycled up to 1.5 V (us RHE) substantial amounts of oxide were observed, the amount being dependent upon the number of cycles and the sweep rate.The oxide was not a simpIe oxide as in the case of polarization, but is explained as a mixture of iridium oxy/hydroxide species. Two other sweep limits (1.2 and 1.8 V) were briefly investigated. Possible mechanisms explaining these results are discussed. The pure-metal spectra (including angle-resolved studies) have been examined at ambient and elevated temperatures with particular emphasis upon the 0 1s region and the methods for removing any residual oxide from the metal surface. There has been considerable interest in the iridium electrochemical system as a result of the unusual cyclic voltammetric behaviour it exhibits. Cycling is thought to produce a film of IrO, which acts as a better catalyst than the metal in the oxygen-evolution r e a ~ t i 0 n . l ~ ~ Fig.1 shows the cyclic voltammogram for an iridium electrode, after argon-ion etching, in both 0.5 and 5 mol dm-3 sulphuric acid. As can be seen from fig. 1 (a) the electrode in concentrated acid shows the type of behaviour observed for the other noble metals (Pt, Pd, Rh and Au), which display voltammograms that are essentially superimposable8 on multiple scans (i.e. the current remains constant). Fig. 1 (b) shows that when the electrolyte is diluted there is a marked change in the appearance of the voltammogram of iridium. It can be seen that as the number of cycles increases there is a corresponding steady increase in current, leading to an almost symmetric voltammogram about the potential axis.This cannot be explained simply by the formation of a phase oxide, since Pt, Pd, Rh and Au are also known to form these oxide~.~-l~ A more complex mechanism must therefore be occurring on the iridium surface. A proper understanding of the nature of this oxide film is not only important for interpretation of the cyclic voltammetric results, but also is required to explain the electrocatalytic behaviour with respect to oxygen evolution and the well known electrochromic properties of the electrochemically produced oxide X-ray photoelectron spectroscopy (X.P.S.) is being increasingly used as a method for the examination of electrode s~rfaces.~9 11-137 20-22 The characteristic valence bands of solids may be studied and the binding energies of the core electrons measured. The latter depend upon the element involved, the charge on the atom or ion concerned and its chemical environment.Furthermore, the technique can be applied to very thin films (5-50 A). film .14-19 135136 X.P.S. STUDY OF IRIDIUM ELECTRODES 1 1 1 1 1 1 1 1 0.2 0 4 0 6 0 8 10 1 2 1 4 16 E/V us RHE Fig. 1. Cyclic voltammograms for etched iridium in (a) 5 and (b) 0.5 mol dm-3 sulphuric acid. The current increases with the number of cycles for (b). Voltammograms are shown for 150, 350, 1500 and 2500 cycles for (b). EXPERIMENTAL All the spectra were obtained using an AEI(Kratos) ES200B X-ray photoelectron spectrometer, operated in the FRR mode, using unmonochromatized Mg Ka X-radiation. The data were collected using an Apple I1 microcomputer system linked to an IBM 370/168 computer, the latter being used for most of the data 24 and for all the molecular-orbital calculations.The base pressure in the sample cbmber varied from lo-* to Torr.* Samples were etched using an Ion Tech B24 mechanically scanned saddle ion source. Calibration was based upon the Ir 4f7,2 metal peak (60.28 eV) and the carbon Is electron peak (284.6 eV) from residual pump oil on the sample surface. Current against voltage curves were recorded using a triangular voltage sweep at three sweep rates (100, 31.25 and 10 mV 0). The electrochemical cell has been described previou~ly.~ All the solutions used triply distilled water and AristaR grade sulphuric acid. Goodfellow Metals 99.9% purity iridium of 0.125 mm thickness was used for the electrodes, which were cut to have * 1 Torr = 101 325/760 Pa.H.Y. HALL AND P. M. A. SHERWOOD 137 a cross-section of CQ. 5 mm x 20 mm, this size being chosen mainly for ease of mounting samples for the spectrometer. All potentials were recorded using a saturated calomel reference electrode (SCE), but are reported with respect to the reversible hydrogen electrode (RHE). Spectra were fitted using a non-linear least-squares method with a Gaussian/LorentzianZ5 peak shape, including the effect of radiation satellites. Molecular-orbital calculations were carried out using the CNDO charge-iterated extended- Huckel program described previously.26 The Basch, Viste and Gray parameters for the s, p and delectrons of iridium were taken to be (in electron volts) -0.75, -0.90 and - 1 .OO for A values, -7.00, -6.00 and -8.90 for B values and -7.49, -4.89 and -7.46 for C values. Burn’s integrals were used2’ and the Ir-0 bond length was taken as 1.905 A in Ir-0 bonds and 2.005 8, in Ir-OH bonds.The OH bond length was taken as 0.99 A. Expectation values for l / r for use in the ground-state potential model were taken from Mann’s integrals.28 RESULTS ELECTRODE PRETREATMENT The electrode was cleaned of oxide by using argon-ion etching, complete removal of oxide being confirmed by X.P.S. studies. The effect of etching with pure oxygen was studied, but this was found to have no effect on the metal region of the spectrum even on turning the sample to a highly surface sensitive angle.29 The pure-metal spectrum was fitted to a Gaussian/Lorentzian mixing ratio of 0.783 with an exponential tail slope of 0.02284 for a ca.360 channel spectrum. The separation of the spin-orbit split components was 3.03 eV and the width was 0.96 eV. These fitting parameters were used in all the cases below for the metal peak in spectra containing metal and oxide. Various workers have claimed that iridium metal can be cleaned by electrochemical 3(t34 such as polarization above 1.63 V or cycling to a similar potential in 0.5 mol dm-3 sulphuric acid electrolyte. These methods are shown below to be unsuitable for completely removing the oxide and in some cases actually form some oxide. We have found that thin oxide films are often left or even produced on many metal surfaces when they are cleaned chemically or electrochemically.13 In contrast, the 0 1s region for etched iridium is interesting in that it shows three peaks at binding energies of 529.32, 531.01 and 532.26 eV, even after extensive argon-ion etching.When the argon is replaced with pure oxygen the overall ifitensity of the oxygen region increases slightly and the binding energies of these three components remain constant. However, the relative percentage areas of the total envelope do show changes with the peak at lowest binding energy increasing in intensity at the expense of the other peaks [fig. 2 (a)-(c)]. Angle-resolved experiments with steadily increasing surface sensitivity [fig. 2(d) and (e)] showed the peak at highest binding energy (which is probably due to chemisorbed oxygen) to increase steadily with respect to the other two peaks.This suggests that the other two peaks must lie further towards the bulk. There was no effect seen in the Ir 4fregion on surface sensitization of the samples. Previous workers have suggested the presence of sub-surface oxygen specie^^^-^^ on iridium and other noble metals which could explain these observations, especially as the etching process would be expected to remove surface oxygen species. The peaks occur in a similar region to those on treated electrodes, but their contribution to the oxygen spectrum in such cases would be expected to be very small as a result of the substantial oxide layers covering the metal surface. Further information was provided about the 0 1s region in the metal by heating in situ to 150, 350 and 600 O C and recording the spectra at the relevant temperature (table 1).These results show that the peak at lowest binding energy, assigned to138 X.P.S. STUDY OF IRIDIUM ELECTRODES l . . . . I . . . . I . . . . I . . . . 535 530 535 530 chemical shift/eV Fig. 2. Effect of etching with pure oxygen on the 0 1s spectral region. (a) Argon-ion-etched surface. (b) Sample (a) after 5 min pure oxygen etch. (c) Sample (a) after 15 min pure oxygen etch. (d) Sample (c) at a moderately surface-sensitive angle. (e) Sample (c) at a highly surface-sensitive angle. Table 1. Effect of heat on etched iridium, oxygen 1s spectra binding energies/eVa percentage areas T/OC peak 1 peak 2 peak 3 width/eV peak 1 peak 2 peak 3 ambient 529.3 531.0 532.4 1.84 24.2 52.3 23.5 150 530.3 531.7 533.6 2.56 31.5 57.8 10.7 350 529.8 531.3 532.7 1.95 41.1 45.2 13.7 600 529.1 530.9 533.0 2.12 53.6 41.6 4.8 ambient 529.5 531.2 532.8 2.06 43.6 45.4 11.0 (after heating to 600 "C) a W.r.t.Ir 4f,,2 peak at 60.28 eV.H. Y. HALL AND P. M. A. SHERWOOD 139 chemisorbed oxygen, is steadily reduced with respect to the other two peaks. This would correspond to the loss of chemisorbed oxygen and the diffusion of oxygen (in the form of 0, or some combined form such as water) from the bulk to the surface. The metal region showed no change on heating. 8 6 4 2 0 8 6 4 2 0 chemical shift/t.V Fig. 3. Iridium 4fregion after steady-state polarization for 15 min in 0.5 mol dmV3 sulphuric acid. (a) Argon-ion-etched iridium surface. (b) Sample (a) after polarization to 1 .O V.( c ) Sample (a) after polarization to 1.4 V. ( d ) Sample (a) after polarization to 2.0 V. ELECTRODES PULSED TO VARIOUS POTENTIALS A number of experiments were carried out where the electrode was pulsed to and held at a steady potential for 15 min and the electrode surface subsequently studied by X.P.S. The ten potentials examined in 0.5 mol dm-3 acid were 0.5,0.8, 1 .O, 1.2, 1.3, 1.4, 1.5, 1.6, 1.8 and 2.0 V, the first four of which showed the same metal spectrum [fig. 3(6)] as for etched iridium [fig. 3(a)]. A sample held at the rest potential of the cell (ca. 0.85 V) also showed this typical Ir 4f spectrum. Samples pulsed to 1.3, 1.4 and 1.5 V showed some oxide formation (table 2). Fig. 3 (c) shows the spectrum for the 1.4 V electrode which had 21 % oxide intensity. Previous workers have reported that the oxide formed on iridium is soluble at potentials > 1.63 V but this was found to be incorrect (table 2).Fig. 3(d) shows the spectrum obtained at 2.0 V indicating the presence of oxide. The behaviour at above 1.63 V was further studied by examining the effect on an oxide-covered electrode of polarizing to 1.8 V. The spectrum of a completely oxide-covered electrode (which must be > 20 A thick since no metal peak is seen) is shown in fig. 4(a). The second oxide peak seen in this spectrum is discussed140 X.P.S. STUDY OF IRIDIUM ELECTRODES Table 2. Steady-state behaviour in 0.5 mol dm-3 acid, iridium 4fspectra potential binding energy width of percentage area /V us RHE of oxideleva oxide/eV of oxide no oxide peaks present in these spectra 1 .o 1.2 1.3 61.4 4.28 18.3 1.4 61.6 2.46 21.1 1.5 61.6 3.17 20.2 1.8 61.3 2.60 16.2 1.8 (1 h) 61.4 2.46 18.3 2.0 61.5 3.14 16.3 a W.r.t. Ir 4f,,2 peak at 60.28 eV.3 8 6 4 2 0 8 6 4 2 0 chemical shift/eV Fig. 4. Iridium 4f region showing the effect of polarization above 1.63 V in 0.5 mol dm-3 sulphuric acid. (a) Oxide-covered electrode. (b) Sample (a) held at 1.8 V for 15 f i n . (c) Argon-ion-etched electrode held at 1.8 V for 15 min. (d) Argon-ion-etched electrode held at 1.8 V for 1 h.H. Y. HALL AND P. M. A. SHERWOOD 141 below. This electrode was then polarized at 1.8 V for 15 min [fig. 4(b)], and its spectrum shows the removal of most of the oxide film. Fig. 4(c) is for a clean electrode polarized at 1.8 V for 15 min and fig.4(d) for 1 h. These results clearly show that oxide is removed at this potential, but at least 15% of the spectral intensity is always found to be due to oxide. The 0 1s spectra for the above experiments generally showed only two peaks, whose positions varied with conditions as shown in table 3. These small but significant differences in 0 1s binding energies may correspond to subtle differences in the nature of the oxide species (vide infra). I Table 3. Steady-state behaviour in 0.5 mol dm-3 acid, oxygen Is spectra binding energies/eVa widths/eV percentage areas /VvsRHE peak 1 peak 2 peak 1 peak 2 peak I peak2 potential 0.5 0.8 1 .o I .2 1.4 1.8 1.8 (1 h) 2.0 2.29 - 529.7 530.9 - 529.8 53 1.4 2.37 1.51 2.17 - 530.8 532.7 - 530.7 53 1.6 2.01 3.33 2.19 - 530.5 53 1.9 - 530.6 532.4 2.25 1.85 529.7 53 1.4 - 529.6 530.9 1.95 2.65 1.93 - 530.3 53 1.9 - 530.3 531.9 1.92 2.22 2.41 - 529.2 532.2 - 529.2 532.2 2.4 1 2.00 2.71 - 528.8 530.2 - no fit was possible with variable-width oxide peaks 2.11 - 530.2 532.5 - 530.2 532.5 2.1 1 1.91 2.07 - 78.8 92.7 87.9 63.6 85.1 93.0 78.7 63.1 84.6 83.1 97.3 97.7 3.5 95.2 95.7 21.2 7.3 12.1 36.4 14.9 7.0 21.3 36.9 15.4 16.9 2.7 2.3 96.5 4.8 4.3 a W.r.t.Ir 4f,,2 peak at 60.28 eV. ELECTRODES CYCLED OVER VARIOUS POTENTIAL RANGES Experiments were carried out where the initially argon-ion-etched electrode was cycled from 0.05 V to one of three different anodic potentials, namely 1.2, 1.5 and 1.8 V in 0.5 mol dm-3 sulphuric acid. 1.2 V was studied because previous workers have used this limit as a means of cleaning the electrode and also because this is the upper limit commonly used in studying the electrochromic propertiesl4-l99 40 of anodic iridium oxide films (AIROFS) (although such films are much thicker, typically 2000-3000 A).In agreement with other ~ ~ r k e r ~ ~ ~ ~ ~ ~ ~ ~ ~ we find that a clean surface shows no oxide formation after cycling twenty times at a rate of 31.25 mV s-l as shown in fig. 5(a). Fig. 5(b) and (c) indicate that an oxide-covered electrode does show some removal of the layer on cycling but only a very small amount on each scan. This is not therefore a suitable method for cleaning the surface, but it could have a significant effect on thick AIROFS, where cycling is often prolonged. Another limit investigated was 1.8 V, again as this has been used as a method to clean an oxide-covered surface.A clean electrode cycled to 1.8 V shows oxide142 X.P.S. STUDY OF IRIDIUM ELECTRODES A I I 1 I I I I , 6 4 2 0 chemical shift/eV Fig. 5. chemical shift/eV Fig. 6. Fig. 5. Iridium 4fregion showing the effect of cycling to 1.2 V at 31.25 mV s-l in 0.5 mol dm-3 sulphuric acid. (a) Argon-ion-etched electrode cycled 20 times. (b) Oxide-covered electrode cycled 100 times. (c) Oxide-covered electrode cycled 200 times. Fig. 6. Iridium 4fregion showing the effect of cycling to 1.8 V at 3 1.25 mV s-' in 0.5 mol dm-3 sulphuric acid. (a) Argon-ion-etched electrode cycled 20 times. (b) Oxide-covered electrode cycled 50 times. (c) Oxide-covered electrode cycled 20 times. formation with the amount increasing as the number of cycles increases [fig.6(a) and (b)]. Starting with an oxide-covered sample does result in most of the oxide being removed [fig. 6 ( c ) ] but there is always a residual amount left similar to that seen on steady-state polarization at 1.8 V. The bulk of the experiments were carried out with a limit of 1.5 V since this is the optimum potential for oxide formation and growth. Fig. 7 shows a plot of the amount of oxide formed against the number of cycles for all three sweep rates and fig. 8 shows the iridium 4f regions for samples cycled at 3 1.25 mV s-l. It can be seen from fig. 5 and 8 that in some cases the oxide region of the metal spectrum is fitted to two peaks and the areas of these have been added together to obtain the total oxide coverage.The second peak appears only when there is > 20% oxide with a contribution to the total peak envelope which can be less than or greater than that of the first peak. The total amount of oxide, however, increases steadily as shown in fig. 7. The first oxide peak has a binding energy of 61.62 (+0.17) eV and the second oxide peak has a binding energy that varies from 61.9 to 63.7 eV. The width of both peaks varies andH. Y. HALL AND P. M. A. SHERWOOD 143 A I L I 1 I T $0 20 sb do 100 I no. of cycles Fig. 7. Area of the iridium metal 4fpeak expressed as a percentage of the total iridium 4farea (% metal) plotted against the number of cycles. The figure shows the amount of oxide formed on cycling argon-ion-etched iridium to 1.5 V at three different sweep rates: ., 10; 0, 31.25 and A, 100 mV s-l.a plot of peak width against the number of cycles for the first peak indicates that there is a minimum for all scan rates of ca. 1.54 eV (fig. 9). There is no such pattern for the second peak where widths vary unpredictably. The spectra were then all refitted with the width of the first peak fixed at this value and the results are shown in fig. 10 for the same scan rate as in fig. 8. The binding energy of peak one remains constant at 61.62 (f 0.16) eV and again the second peak varies over the range 61.82-63.87 eV. Comparison of fig. 10 with fig. 8 shows that fixing this width has the main effect of broadening the second peak to compensate for any narrowing of the first peak. In all cases the spin-orbit splitting between the Ir 4fcomponents was fixed at 3.03 eV for all peaks.Oxygen 1s spectra are similar to those seen for pulsed electrodes, usually with two components, and fig. 11 shows those corresponding to the metal spectra in fig. 8 and 10. ELECTRODES CYCLED TO 1.5 v (US RHE) AND THEN PULSED TO VARIOUS POTENTIALS A range of experiments were carried out where the electrode was cycled 5 times and then pulsed for 15 min to one of five anodic potentials, namely 1.2, 1.3, 1.4, 1.5 and 1.8 V. The results are summarized in table 4 and indicate that a potential of 1.2 or 1.3 V has little effect upon the oxide layer. The sample held at 1.4 V shows more oxide144 X.P.S. STUDY OF IRIDIUM ELECTRODES , , I , I I I 1 I I I I I I a 6 4 2 0 8 6 4 2 0 chemical shiftlev Fig. 8. Iridium 4f region showing the effect of cycling to 1.5 V at 31.25 mV s-' in 0.5 mol dmP3 sulphuric acid.Electrodes cycled (a) 5, (b) 10, (c) 20, ( d ) 50, (e) 100 and (f) 200 times. (74%) than would be expected from simply adding effects of polarization (21 %) and cycling (27%). Although the 1.5 V sample has more oxide after polarization than before, the increase is consistent with the amount expected on polarization at this potential. ANGLE-RESOLVED STUDIES Angle-resolved studies were carried out where the take-off angle from the sample surface was made very small in order to enhance the intensity of the surface species.29H. Y. HALL AND P. M. A. SHERWOOD 145 I I I 1 I 1 210 40 Go do 60 no. of cycles Fig. 9. Graph showing the effect of change in width of the first oxide peak in the iridium 4f region on cycling to 1.5 V in 0.5 mol dme3 sulphuric acid.., 10; 0, 31.25 and A, 100 mV 0. These studies show that when oxide was detected in the Ir 4fregion, the relative intensity of the oxide peaks increased with respect to the metal peaks (fig. 12), and the presence of more than one oxide peak at surface-sensitive angles showed that the second oxide peak lay nearer the surface than the first oxide peak [fig. 12(c) and (41. EXPERIMENTS IN 5 mol dm-3 ACID Electrodes were polarized at four potentials (0.6, 1 .O, 1.4 and 1.8 V) in 5 rnol dm-3 sulphuric acid which previous workers3** 4 2 have suggested should give similar behaviour to other noble metals in acid solutions. Table 5 shows that this is probably the case for polarization since the amount of oxide increases as the potential increases, although there is never a large amount on the iridium surface in contrast to other noble metals.The behaviour with cycling is different from the other noble metals since oxide formation is seen. Cycling in 5 mol dm-3 acid was carried out only at the medium rate of 31.25 V s-l. Again a limit of 1.2 V showed no oxide formation [fig. 13(a)] and an upper limit of 1.8 V showed ca. 20% oxide as seen in 0.5 mol dm-3 acid [fig. I3 (b)]. With an upper limit of 1.5 V electrodes were cycled 5,20 and 50 times. The first showed no oxide formation, the other two showed 22 and 45 % oxide, respectively [fig. 13 (c) and (41, which compare with 66 and 80.8% for 0.5 mol dm-3 acid. Note that the second oxide peak is never seen in these studies in 5 mol dmA3 acid, which suggests that it may play an important role in the oxide-formation process occurring in the dilute electrolyte.146 X.P.S.STUDY OF IRIDIUM ELECTRODES n 1 . 1 I 1 1 . 1 I I 8 6 4 2 0 8 6 4 2 0 chemical shift/eV Fig. 10. Iridium 4fregion of electrodes cycled to 1.5 V at 3 1.25 mV s-l in 0.5 mol dm-3 sulphuric acid. F.w.h.m. of first oxide peak is fixed at 1.54 eV. Electrodes cycled (a) 5, (b) 10, (c) 20, ( d ) 50, (e) 100 and cf) 200 times. DISCUSSION There has been extensive study and debate on the iridium electrode system, and in trying to analyse all the above results it is helpful to consider mechanisms suggested by previous workers to see if they are consistent with our observations. STEADY -STATE POLARIZATION There has not been a great deal of work concentrated on this aspect of the behaviour of iridium, but there are a number of points shown by our results.H0are~~9 44 indicatesH. Y. HALL AND P. M. A. SHERWOOD 147 1 . . , . 1 , . , . 1 l * , , , l n , , , 535 530 525 535 530 chemical shift/eV Fig. 11. Oxygen 1s region of electrodes cycled to 1.5 V at 31.25 mV s-l in 0.5 mol dmP3 sulphuric acid. Electrodes cycled (a) 5, (b) 10, (c) 20, ( d ) 50, (e) 100 and cf) 200 times. that a very thin chemisorbed layer forms initially on the surface in the potential range 0.8-1 .O V, and at higher potentials a metal/metal oxide equilibrium of 1r/1r02 occurs, as seen with other noble metals (Au/Au,O, and Pd/PdO,). Our work substantiates this although we do not see oxide until 1.3 V.' Hoare further states that these films are not stable indefinite1~ll-l~ in acid and, unlike other noble rnetal~,~ the amount seen on iridium does not increase steadily as the anodic potential increases.Indeed, applying a potential of 1.8 V to an oxide-covered electrode results in loss of most of the oxide film, but there is always a residual amount present. This indicates that at higher potentials there is competition between the formation and dissolution of the oxide, as has been suggested previou~ly,~~ 34 and that the electrode cannot be cleaned by such treatment. Field-ion microscopy studies459 46 and a previous X.P.S. in~estigation~~ of iridium 6 FAR 1148 X.P.S. STUDY OF IRIDIUM ELECTRODES Table 4. Electrodes cycled to 1.5 V (us RHE) and then pulsed to various potentials, iridium 4f spectra po ten ti a1 width of percentage area /V us RHE binding energy of oxideleva oxide/eV of oxide cycled alone 61.6 1.2 61.5 1.3 61.6 1.4 1.5 2.82 27.4 2.46 21.2 3.17 27.8 1.66 (2 pk.fit) 2.28 2.15 z!j:: (2 pk’ fit) 2.80 63.4 61.5 (1 pk. fit) 2.74 36.1 a W.r.t. Ir 4f,,2 peak at 60.28 eV. 1 I I 1 l I I 1 8 6 4 2 0 8 6 4 2 0 chemical shift/eV Fig. 12. Angle-resolved studies on iridium electrode with oxide coverage. (a) and (c) correspond to the optimum angle; (b) and (d) correspond to spectra run at a surface-sensitive angle.H. Y. HALL AND P. M. A. SHERWOOD 149 Table 5. Steady-state behaviour in 5 mol dm-3 acid, iridium 4fspectra potential binding energy width of percentage area /V us RHE of oxide/eV" oxide/eV of oxide 0.6 no oxide peak present in this spectrum 1 .o 62.1 1.86 16.7 1.4 61.7 3.04 15.7 1.8 61.6 3.03 18.3 a W.r.t. Ir 4f,,2 peak at 60.28 eV.I I 1 1 1 I 1 1 1 1 1 l 1 1 1 1 1 . I 8 6 4 2 0 8 6 4 2 0 chemical shift/eV Fig. 13. Iridium 4f region showing the effect of cycling argon-ion-etched electrodes in 5 mol dm-3 sulphuric acid at 3 1.25 mV s-l. (a) Upper limit 1.2 V, cycled 20 times. (6) Upper limit 1.5 V, cycled 20 times. (c) Upper limit 1.5 V, cycled 50 times. ( d ) Upper limit 1.8 V, cycled 20 times. electrochemistry both show formation of Ir02, and Kim et aZ.*' also suggest that some Irvr may be present on the surface. We see no evidence for the latter, and from the binding energies of the oxide peaks it is clear that one oxide species is predominant at all times.IrO, is the most likely oxide to form, and we therefore suggest that polarization of iridium electrodes forms an IrO, film which is partially soluble above 1.6 V and which may contain small amounts of other Ir oxides as discussed below. 6-2150 X.P.S. STUDY OF IRIDIUM ELECTRODES CYCLING IRIDIUM ELECTRODES Early results on the cyclic voltammetric behaviour of iridium were explained via a chemisorption mechanism similar to that seen for other platinum metals. The increase in current passed was thought by Otten and V i ~ c h e r ~ ~ 9 ~ ~ to be due to the formation of pits in the metal surface where oxide could be formed on the anodic sweep and reduced on the following cathodic scan. Later work has shown that oxide is present on the surface at all times, and this has been described as a mixture of oxy/hydroxide ' 7 32-349 419 4 2 7 50 Our results clearly show that there is in fact more than one type of oxide present.It is well known that iridium exhibits a range of oxidation-state oxides, of which the most stable is Ir02. The oxygen 1s spectra normally show two peaks which have binding energies in the range expected for oxide and hydroxide groups (table 3) whose relative intensities and absolute binding energies vary in a complex manner. These observations are consistent with a complex oxide/hydroxide system which would lead to species with a wide range of formal oxidation states. The full analysis of the binding energies observed is complicated by the limited availability of reasonable model compounds and the lack of detailed crystallographic data, which makes calculations on such compounds difficult.Nevertheless, in order to get some idea of what binding energies might be expected for such a range of oxyhydroxides we have carried out molecular-orbital cluster calculations for the octahedral ions IrOi-, II-O,(OH)~-, Ir04(OH):- and Ir03(OH)!-, which correspond to iridium in formal oxidation states Irvl, IrV, IrTV and I F , respectively. We have assumed that the relaxation-energy differences between the different oxyhydroxides will be small since the solid-state environment is very similar in all cases. If this is so then the relative electronic potentials within the framework of the ground-state potential model using eqn (10) of ref. (26) should follow the reverse order of the binding energies.The calculated electronic potentials (eV) : 111, (2 16.9), IV (21 6. l), v (2 18.9) and VI (21 7.6), indicate that the oxides should lie within a binding energy range of ca. 3 eV, with the binding energies increasing with formal oxidation state in the order v < VI < 111 < IV. It would appear that any mechanism to explain the unusual behaviour must feature reactions involving a series of oxidation-state changes probably involving different oxyhydroxides. Several such mechanisms have been suggested and these will need to explain the complex electrochemical behaviour and the surface chemistry as revealed by our X.P.S. studies. The main feature of the electrochemical behaviour that must be explained is the way in which the current density rises symmetrically on the anodic and cathodic peaks in the forward and reverse sweeps as the number of cycles is increased.This has been well commented upon and recent workers have suggested that there must be an irreversible oxide-formation process occurring in addition to the reversible oxidation process. Explanation of the reversibility seen in the voltam- metric peaks is more difficult and has been ascribed to reversible stoichiometric changes within this irreversibly formed oxide. If one accepts that there may be a reversibly and an irreversibly formed oxide, then it is not going to be simple to distinguish clearly the oxides in our X.P.S. studies, first because it will be seen that there may be no difference or only a small difference in binding energy between the oxides and secondly because there may be enough dehydration in the vacuum of the spectrometer to change the oxides, although we do not observe any change in the spectra with changing exposure time to the X-rays or the vacuum.The pulsed experiments would be expected to show mainly the irreversible oxide, as is observed. The first oxide peak seen in the cycled spectra may therefore be due to this irreversible oxide, especially as it must rise to 15 % of the total area before the other oxide is seen.H. Y. HALL AND P. M. A. SHERWOOD 151 The cycling experiments would be expected to show peaks for both types of oxide, as would the cycling and pulsing experiments. If the first peak can again be assigned to IrO,, then the second peak with variable width and position is consistent with a mixture of oxy/hydroxide, as suggested by our calculations and observations discussed above.One could consider the formation of the irreversible oxide by the following process: Ir + 2H,O -+ 4H+ + IrO, + 4e. The initially greater width for the first oxide peak (table 2) might be explained by the formation of another irreversible oxide such as 1r203, which would occur at a similar potential to the above reaction: 21r + 3H,O -+ 6H++ 1r203 + 6e. The reversible oxide, which one might associate primarily with the second oxide peak, may be associated with a range of complex oxy/hydroxide reactions such as: IrOJOH), e II-O~+~(OH),-~ + cH+ + ce as suggested by Rand and WOO^,^^^, where c was calculated to be 2, corresponding to an overal reaction of: Our results do not support the idea of total conversion from one iridium oxide at one potential to a second at the other end of the voltammetric excursion.We have shown that there is always a mixture whose composition is almost impossible to predict at any specific potential, but we can add no comment as to whether it is indeed a two-electron reaction. The inability to give an unambiguous identification of all the possible oxy/hydroxide compounds means that we can make no further comment in the debate about the origins of the change in colour of the oxide films (which has been It is very interesting to note that whilst cyclic voltammograms in 5 mol dm-3 acid do not show the peaks seen in the dilute electrolyte (fig. I), the electrodes do show some oxide formation. The second oxide peak ascribed mainly to the oxy/hydroxide is not seen, and so it is likely that IrO, is again the predominant oxide.There seems to be competition again between formation and dissolution of the oxide in the more concentrated media. Finally, it is clear from the results above that great care must be taken to clean the electrode properly, since we have shown that various methods (such as polarization above 1.63 V) which have been used previously actually form oxide. Ir(OH), IrO, + 2H+ + 2e. suggested as being due to protons, hydroxide ions or bothlg? 3 0 7 40 1. CONCLUSIONS be a complex system with a variety of oxy/hydroxide species on the electrode surface. Such an observation is consistent with those electrochemical mechanisms that have been suggested which are sufficiently flexible to account for the complex behaviour observed. While the behaviour in 0.5 mol dm-3 acid is complex and has previously been reported as such, we report for the first time from observations of the surface chemistry that this complex behaviour is still partly present in 5 mol dm-3 acid.The iridium electrode system in acid is confirmed through X.P.S. surface studies to ' We thank the S.E.R.C. for the provision of equipment and for a studentship to H. Y. H. and Dr J. P. G. Farr for helpful comments.152 X.P.S. STUDY OF IRIDIUM ELECTRODES A. Damjanovic, A. Dey and J. O’M Bockris, J. Electrochem. Soc., 1966, 113, 739. A. Damjanovic and M. K. Y. Wong, J. Electrochem. Soc., 1967, 114, 592.D. N. Buckley and L. D. Burke, J. Chem. Soc., Faraday Trans. 1, 1976, 72, 2431. E. J. Frazer and R. Woods, J. Electroanal. Chem., 1979, 102, 127. S. Gottesfeld and S. Srinivasan, J. Electroanal. Chem., 1978, 86, 89. S. Hackwood, L. M. Schiavone, W. C. Dautremont-Smith and G. Beni, J. Electrochem. Soc., 1981, 128, 2569. L. D. Burke and E. J. M. O’Sullivan, J. Electroanal. Chem., 1981, 117, 155. R. Woods, Electroanalytical Chemistry, A Series of Advances, ed. A. J. Bard (Marcel Dekker, New York, 1976), vol. 9, pp. 98-108. T. Dickinson, A. F. Povey and P. M. A. Sherwood, J. Chem. Soc., Faraday Trans. I , 1975,71, 298. lo D. A. J. Rand and R. Woods, J. Electroanal. Chem., 1971, 31, 29. l1 K. S. Kim, N. Winograd and R. E. Davis, J. Am. Chem. Soc., 1971,93, 6296. l2 K.S. Kim, A. F. Gossmann and N. Winograd, Anal. Chem., 1974,46, 197. l3 H. Y. Hall and P. M. A. Sherwood, to be published. l4 S. Gottesfeld, J. D. E. McIntyre, G. Beni and J. L. Shay, Appl. Phys. Lett., 1978, 33, 208. l5 C. E. Rice, Appl. Phys. Lett., 1979, 35, 563. l6 J. L. Shay, G. Beni and L. M. Schiavone, Appl. Phys. Lett., 1978, 33, 942. l7 G. Beni and J. L. Shay, Appl. Phys. Lett., 1978, 33, 567. l8 L. M. Schiavone, W. C. Dautremont-Smith, G. Beni and J. L. Shay, Appl. Phys. Lett., 1979,35, 823. G. Beni, C. E. Rice and J. L. Shay, J. Electrochem. Soc., 1980, 127, 1342. 2o P. M. A. Sherwood, Contemporary Topics in Analytical and Clinical Chemistry, ed. D. M. Hercules, G. M. Hieftje, L. R. Snyder and M. A. Evenson (Plenum Press, New York, 1982), vol. 4, pp.205-293. 21 P. M. A. Sherwood, Am. Lab., 1983, 15, 14. 22 P. M. A. Sherwood, J. Microsc. Spectrosc. Electron., 1980, 5, 475. 23 A. Proctor and P. M. A. Sherwood, Anal. Chem., 1980, 52, 2315. 24 A. Proctor and P. M. A. Sherwood, Anal. Chem., 1982,54, 13. 25 R. 0. Ansell, T. Dickinson, A. F. Povey and P. M. A. Sherwood, J. Electroanal. Chem., 1979,98,79. O6 P. M. A. Sherwood, J. Chem. Soc., Faraday Trans. 2, 1976, 72, 1791. 27 G. Burns, J. Chem. Phys., 1964, 42, 1521. 28 J. B. Mann, Atomic Structure Calculations. II, Hartree-Fock Wavefunctions and Radial Expectation Values: Hydrogen to Lawrencium, Los Alamos Scientific Laboratory, Report LA-3691, 1968. R. 0. Ansell, T. Dickinson, A. F. Povey and P. M. A. Sherwood, J. Electron Spectrosc. Relut. Phenom., 1977, 11, 301. 30 J. D. E. McIntyre, W. R. Peck, Jr. and S. Nakahara, J. Electrochem. Soc., 1980, 127, 1264. 31 J. L. Ord, J. Electrochem. Soc., 1982, 129, 335. 32 D. Michell, D. A. J. Rand and R. Woods, J. Electroanal. Chem., 1977, 84, 117. 33 D. N. Buckley and L. D. Burke, J. Chem. Soc., Faraday Trans. 1, 1975,71, 1447. 34 D. N. Buckley, L. D. Burke and J. K. Muleahy, J. Chem. Soc., Faraday Trans. 1, 1976,72, 1896. 35 P. A. Zhdan, G. K. Boreskov, A. J. Boronin, W. F. Egelhoff Jr and W. H. Weinberg, Surf. Sci., 1976, 36 M. Salmeron, L. Brewer and G. A. Somorjai, Surf. Sci., 1981, 112, 207. 37 V. P. Ivanov, G. K. Boreskov, V. I. Savchenko, W. F. Egelhoff Jr and W. H. Weinberg, Surf Sci., 38 J. L. Taylor, D. E. Ibbotson and W. M. Weinberg, Surf. Sci., 1979, 79, 349. 3e H. J. Conrad, J. Kuppers, F. Nitschke and A. Plagge, Surf Sci., 1977, 69, 668. 40 S. Gottesfeld and J. D. E. McIntyre, J. Electrochem. Soc., 1979, 126, 742. 41 J. 0. Zerbino, N. R. de Tacconi and A. J. Arvia, J. Electrochem. Soc., 1978, 125, 1266. 42 D. A. J. Rand and R. Woods, J. Electroanal. Chem., 1974, 55, 375. 43 J. P. Hoare, J. Electrochem. Soc., 1964, 111, 988. 44 J. P. Hoare, J. Electroanal. Chem., 1968, 18, 251. 46 C. C. Schubert, C. L. Page and B. Ralph, Electrochim. Acta, 1973, 18, 33. 46 J. P. G. Farr,H. N. Southworthand A. G. Tyson, Proc. 10th Plansee-Seminar, 1981,ed. H. M. Ortner O7 K. S. Kim, C. D. Sell and N. Winograd, Proc. Symp. Electrocatal., ed. M. W. Breiter (The Electro- 48 J. M. Otten and W. Visscher, J. Electroanal. Chem., 1974, 55, 1. 48 J. M. Otten and W. Visscher, J. Electroanal. Chem., 1974, 55, 13. 50 S. H. Glarum and J. H. Marshall, J. Electrochem. Soc., 1980, 127, 1467. 61, 25. 1976, 61, 207. (Metallwerke Plansee GMBH, Reutte), vol. 3, p. 143. chemical Society, Princeton N.J., 1974), pp. 242-257. (PAPER 3/670)

ISSN:0300-9599    

DOI:10.1039/F19848000135

出版商:RSC    

年代:1984

数据来源: RSC

摘要:

J . Chem. SOC., Faraday Trans. 1, 1984,80, 153-181 Thermodynamics of Solution of Homologous Series of Solutes in Water BY MICHAEL H. ABRAHAM Department of Chemistry, University of Surrey, Guildford, Surrey GU2 5XH Received 9th May, 1983 Values of AGF (g --+ aq) and AGF (liq -+ aq) at 298 K are documented for 14 homologous series of gaseous and liquid solutes, and corresponding enthalpies of solution listed for 7 homologous series. It is shown by a thermodynamic argument that only parameters for the process g -+ aq can be used to assess solute-water interactions and that the standard state of pure liquid solute includes a differsnt solute-solute interaction term for each solute standard state. For most of the homologous series, parameters for the process g --+ aq are linear in the number of carbon atoms in the solute; from such linear equations, methylene and group contributions are obtained.It is shown that the methylene increments to AG?(g + aq) and to AHF(g -+ aq) are not constant but vary from one homologous series to another. In a few homologous series the methylene increment is not constant, the most outstanding examples being the alkan-1-01s and n-alkanes. Above dodecan- 1-01, AGF(g -+ aq) becomes gradually more negative than expected, so that octadecan-1-01 is 16 times as soluble as calculated from results on the low alkan-1-01s. A similar, but much larger, effect is observed for the n-alkanes: n-octadecane is more soluble than expected by a factor of 5 x lo3 (the factor for n-hexatriacontane is 2 x 10'") and it is deduced that the n-alkane C,,H,,, will be as soluble in water as in the non-aqueous solvents ethanol and phenol.There are a number of concentration scales used to express the solubility of gases in liquids, and a corresponding number of sets of standard states used to define standard Gibbs energies (and standard entropies) of solution of gases in liquids. Hine and Mookerjeel in their extensive compilation used standard states of 1 mol dm-3 gas and 1 mol dm-3 in solution, as did also Cabani et aL2 in their tabulation of Gibbs energies of solution, AGP, of gases in water. Other workers have reported solubilities or AGP values in terms of the process ideal gas (1 atm) + ideal solution (unit mol fraction solute), and these are probably the most common standard states u ~ e d .~ - l ~ Other standard states that have been used are 1 mmHg for the gas and unit mol fraction solute in ~ o l u t i o n , ~ ~ - ~ ~ 1 mmHg for the gas and 1 mol dm-3 for the solute in and more recently 1 kPa for the gas and 1 mol m-3 for the solute in solution.20 Since gas solubilities, or AGP values, reported in terms of one set of standard states are normally easily converted into corresponding values on any other set of standard states, it does not usually matter which set is employed.7 In table 1 are given numerical values to convert Gibbs energies of solution of gases on various standard states to those on the standard state of ideal gas (1 atm) and ideal solution (unit mole fraction solute). Because the same conversion quantities, table 1, apply to all solutes, differences in standard Gibbs energies of solution between two solutes in a given liquid are However, note that because the solvent molecular weight is used in calculating the mole fraction solute in solution, the standard state of unit mole fraction solute cannot be used with solvents that have no well defined molecular weight, e.g.biological fluids and polymeric materials. 153154 THERMODYNAMICS OF SOLUTION IN WATER independent of these standard states. Thus for the solution of methane and ethane in water at 298 K, AGF (methane, gas) - AGP (ethane, gas) = 0.17 kcal mo1-I on any of the standard states shown in table 1. This is most important in considerations of the methylene increment to the solution of gaseous solutes in a homologous series, because this increment will also be independent of the standard states shown in table 1.Likewise, although values of AS? for gaseous solutes will differ on various standard states, differences between AS? will be independent of the standard states in table 1, as will also the entropic methylene increment to solution of gaseous solutes in a given solvent. Table 1. Effect of standard states on values of AGp(g + solution), in kcal mo1-I at 298 K standard states 6AGFa gas solvent water methanol hexane ~ ~ ~~ 1 atm unit mol fraction 0 0 0 1 atm 1 mol dm-3 2378 1896 1201 1 atm 1 mol kg-' 2380 2039 1452 1 mol dm-3 1 mol dm-3 4272 3790 3095 1 kPa 1 mol dm-3 3734 3252 2557 solvent molecular weight 18.015 32.04 86.18 solvent density 0.997 1 0.7865 0.6548 a Defined so that AG?( 1 atm + unit mole fraction) = AGF (any other standard state) plus the given correction term.However, a quite different set of standard states has been advocated by Tanford2l9 22 and has been used by several w o r k e r ~ , ~ ~ - ~ ~ especially in discussion of the methylene in- crement to solution of solutes in 25 Tanford's standard states are those of the pure liquid solute and the ideal solution (unit mole fraction solute). The relationship of these standard states to those given in table 1 may most simply be discussed with reference to that of the ideal gas (1 atm) and the ideal solution (unit mole fraction solute). For solution in a given solvent, say water, the two standard-state processes g -+ aq and liq -+ aq are related through the standard Gibbs energy of vaporisation of the pure liquid solute to the ideal gas at 1 atm: AG3g --+ aq) ideal gas at 1 atm- solute in solvent water.AGF 2 AGp(liq -+ aq) (1) pure liquid solute From eqn (1) it may be seen that AGp(g -+ aq) + AG? = AGP(1iq -+ as). (2) Similar equations may be set up in terms of enthalpy, entropy etc. Now since AGF (or AH? etc.) will differ from solute to solute, the conversion from AGP(1iq + aq) to AGP(g --+ aq) will differ numerically even for solutes as similar as n-hexane and n-heptane. Thus the standard state of pure liquid solute amounts to assigning a different standard state to each solute. This has been pointed outM. H. ABRAHAM 155 before,l2? l3 and the relationship shown in eqn (1) and (2) has been used by Butler et al.15 and subsequently by other worker~.~**-ll In spite of this, Tanford22 has argued again that the standard state of pure liquid hydrocarbon is equivalent to assigning the same standard state to the various liquid solutes concerned, for example n-hexane liquid and n-heptane liquid.Tanford22 advances a thermodynamic argument to support his views, as follows. For the solution of a small liquid n-alkane solute in a long-chain n-alkane solvent such as n-hexadecane or n-octadecane, the difference between pe for the small n-alkane in the pure liquid (itself) and pe for the small n-alkane in the long-chain n-alkane is only ca. 60calmol-1, when the small n-alkane solute and long-chain n-alkane solvent differ in chain length by some 10 carbon atoms. This amounts to only ca. 6 cal mol-l per methylene group, which is negligible.The validity, or otherwise, of Tanford’s argument can rigorously be demonstrated. Let py be the standard chemical potential of an alkane solute of carbon number rn in an alkane solvent of carbon number 1: thus p& represents the standard chemical potential of n-hexane in n-hexadecane, with mole fraction standard states, and pz represents the standard chemical potential of n-hexane in n-hexane itself. Then for the solution of liquid n-hexane solute in n-hexadecane (Raoult’s law activity coefficient = 0.87) and for the solution of liquid n-heptane solute in n-hexadecane (Raoult’s law activity coefficient = 0.91) Thus Tanford22 is quite correct in stating that numerically p& -p: or --pi amounts to ca. 60 cal mol-l.However, this value has nothing to do with the relative energy of the standard states of n-hexane in n-hexane (pfi) and of n-heptane in n-heptane (pi), which is given by the term p i - p z . The relationship of this term to the ideal-gas standard states of n-hexane and n-heptane can be obtained by considering the solution of gaseous alkanes (ideal gas, 1 atm) into solvent n-hexadecane (unit mol fraction solute). Denoting the ideal gas standard state asppat,, then for the solution of n-hexane gas into n-hexadecane (Henry’s law constant = 0.173 atm) ( 5 ) and for the solution of n-heptane gas into n-hexadecane (Henry’s law constant = 0.055 atm) ,&-&atm = RTln (0.055) = - 1720 cal mol-l. (6) @I6 -/.4!6) - @: atm -p! at,) = - 680 cal m0l-l (7) (p: - pg) - - p! = - 708 cal mol-l.(8) j.46--& = RTln (0.87) = - 84 cal mol-1 (3) = RTln (0.91) = - 56 cal mol-l. (4) & 6 - / 4 atm = RTln(0.173) = - 1040 cal mol-1 Then from eqn ( 5 ) and (6) and now combining eqn (7) with eqn (3) and (4) Thus with reference to the ideal-gas standard states, in which there exist no solute-solute interactions, the standard states of pure liquid n-hexane and pure liquid n-heptane (& -pz) differ by 708 cal rnol-l, identifiable [see eqn (l)] as being due to solute-solute interactions in the pure liquids. As I have pointed out before,l29l3 this difference is given by the ratio of the saturated vapour pressures (or more correctly, fugacities) of n-hexane and n-heptane, uiz: RTln(P7/P6) = -(AGp7-AGy), see eqn (1). To summarise, if Tanford’s standard state of pure liquid n-alkane, or more generally pure156 THERMODYNAMICS OF SOLUTION IN WATER liquid solute, is used then this will lead to the incorporation of a different solute-solute interaction term for each solute standard state.In such a case, little meaning can be attached to the variation of AGP [AGP (liq + solution)] with carbon number along a homologous series, because, defined in this way, the AGP values contain various contributions from solute-solute interactions in the pure liquid solute. A more readily interpretable quantity is AGp(g + solution) because only solute-solvent interactions now contribute. It is the purpose of this paper to set out values of AG?(g -+ aq) and AGP(1iq + aq) for solution of solutes of various homologous series in water and to show how the calculated methylene increment depends on these standard states ; where possible, similar quantities will be tabulated in terms of enthalpy and entropy.In addition, methylene increments will be analysed carefully to determine whether or not the increment is indeed constant along an homologous series, or whether, as found by Beezer et ~ 1 . ~ ~ for a series of monoalkylated resorcinol compounds, there is an alternation in the methylene increment with carbon number. There are two main methods of obtaining AGP values. First, Henry’s law constants maybeobtainedforsolutionofgaseoussolutesinwater ; thenAGP(g -+ aq) = RTln KH. If the solute vapour pressure (or fugacity) is known, AG? = -RTlnP and then through eqn (2) AGP(1iq -+ aq) may be deduced. Secondly, the solubilities of sparingly soluble solutes may be obtained. If expressed on the mole fraction scale, then AGP(1iq -+ aq) = -RTln A‘, and combination with AG? will now yield corresponding values of AG?(g -+ as).For solutes of high carbon number, it is almost impossible to obtain Henry’s law constants, and even the determination of solubilities becames difficult and subject to considerable error. For solutes of low carbon number, solubilities are much easier to obtain, but AGP(1iq -+ aq) may only be set equal to -RTln Xprovided that the secondary medium activity coefficient of the solute in the saturated solution is unity. Corresponding AH? values are again obtained by two main methods. First, variation of Henry’s law constants with temperature will yield AH?(g + aq), and then combination with AH? will give AHP(1iq + aq).Secondly, calorimetric determina- tions of AHp(1iq + aq) may be carried out and values of AHp(g -+ aq) obtained by again using the AH? values. A large number of gas and liquid solubilities have been carried out on the n-alkanes. For the series methane to n-octane there is general agreement on the AG?(g -+ aq) values,2$13* 2o and in table 2 the values are 1i~ted.l~ Vejrosta et ~ 1 . ~ ~ have recently measured Henry’s law constants, as Cwater/Oas, for the n-alkanes from n-pentane to n-nonane over a range of temperature. They summarised all their results in one equation, but I thought it useful to fit the given values of ln(Cwater/Oas) to an equation in A / T+ B + L In T for each alkane separately in order to obtain values of Caterloas and of AH* at 298.15 K.Conversion to the standard states used in this work leads to AGp(g --+ aq) values as follows$ n-pentane (6.59), n-hexane (6.78), n-heptane (7.00) and n-octane (7.21), in kcal mol-l. These are in very good agreement with those in table 2, and confirm the generally accepted values. Combination of the AGP(g + aq) values with vapour pressures27 or fugacities then leads to values of AGP(1iq + as), see also table 2. For the n-alkane series n-nonane to n-hexadecane, there are available the liquid solubilities, listed by Mackay and Shiu,20 that lead directly to AGP(1iq -+ aq) and thence t@AGp(g -+ as). A number of investigations into the solubilities of the solid n-alkanes have been reported,20 the best documented study being that of Sutton and Calder.28 From the solid mole fraction solubilities, values of t Determinations for n-nonane were carried out at 287.95 and 293.20 K only.M.H. ABRAHAM 157 Table 2. Standard Gibbs energies of solution and of vaporisation of n-alkanes, in kcal mol-l at 298 K n-alkane Platma AGPb AGp(g -+ aq)c AGp(1iq -, aq)" methane 87.3 ethane 24.07 propane 7.68 butane 2.19 pentane 0.674 hexane 0.199 heptane 6.028 x lo-, octane 1.847 x lo-, nonane 5.724~ lo-, decane 1.797 x lo-, undecane 5.645 x dodecane 1.737 x lo-* tridecane 5.226 x tetradecane 1.542 x lov5 pentadecane - hexadecane - hep t adecane - octadecane - eicosane, C,, - hexacosane, C,, - hexatriacontane, C,, - - 2.65 - 1.88 - 1.21 - 0.46 0.23 0.96 1.66 2.36 3.06 3.75 4.43 5.13 5.84 6.56 (7.26)g (7.96)Q (8.67)Q (10.77)g (1 5.0)Q (22.0)g (9.37)g 6.28 6.1 1 6.23 6.35 6.61 6.82 6.90 7.16 7.42" 7.44" 8.25" 7.72" 6.64" 5.88h 3.88h 2.19h - - - - 3.25h - 12.55h 3.61 4.23 5.02 5.89 6.84 7.78 8.56 9.52 10.48f 11.19 12.68f 12.89 13.20f 13.84f 13.25i (13.40y 12.96i (13.59y 11.7S (1 3.77Y 9.4Y (13.94y - - - a From Dreisbach,,' except for values from methane to butane which are fugacities, see J. P. Montfort and J. R. Varela H., Chem. Eng. J., 1976, 12, 1. Standard states, 1 atm gas and pure liquid hydrocarbon. Standard states, 1 atm gas and unit mole fraction solute in solution. Values from Abraham13 except where indicated. Standard states, pure liquid hy- drocarbon and unit mole fraction solute in solution. " From AGP(1iq -+ aq) and the tabulated value of AGP.f From the solubility of the liquid alkane, given in ref. (16). Q Estimated from plots of log P or AGP against carbon number. From AGp(1iq -+ aq) and the estimated value of AGP. From the solubility of the solid alkane,,* corrected using the data of A. A. Schaerer, C. J. Busso, A. E. Smith and L. B. Skinner, J. Am. Chern. SOC., 1955,77, 2017, on fusion and solid-solid transitions of n-alkanes. * Value of AGF(so1id -+ as). From the solubility of solid hexatriacontane given by E G. Baker, Science, 1959, 129, 871, corrected as in footnote (i). AGF(so1id + aq) may be calculated, and these values then corrected through the following equation to AGP(1iq + aq) values. In eqn (9), AHg is the molar enthalpy of fusion of the solute in question, Tm is the solute melting point and T is the temperature of the solubility determination (298.15 K in the present instance). If the solute undergoes a solid-solid transition between T and Tm then the second correction term in eqn (9) must also be applied; AHt is the molar enthalpy of transition and Tt is the transition temperature.For solutes that undergo two such transitions between Tand Tm, two such transition correction terms are needed.? t Note that in eqn (9) it is assumed that AH% and AH? are temperature independent over the range T, to T and to T, respectively.158 THERMODYNAMICS OF SOLUTION IN WATER In table 2 are listed AGP(1iq -+ aq) values determined through eqn (9), together with corresponding AGp(g -+ as) values obtained from AGP(1iq -+ aq) and the extrapolated values of AGP.With the exception of the value for n-undecane, which seems much too positive, there is a reasonable straight-line plot of AG?(g -+ aq) against N , the carbon number, from N = 2 to N = 12. Beyond n-dodecane, however, the AGp(g -+ aq) values become rapidly much more negative than expected. Nelson and de Ligny,14 who had available values only up to n-octadecane, attributed this phenomenon to the coiling of alkyl chains in aqueous solution, but the magnitude of the effect is far greater than previously imagined. Results are availablel33 29 for enthalpies of solution and vaporisation of n-alkanes up to n-octane, table 3, although the AH? values for n-octane are rather doubtful. The given AHg(g -+ aq) values may be compared with those I have calculated from the results of Vejrosta et aZ.,26 after a minor correction (45 cal mol-l) from the molar Table 3.Standard enthalpies and entropies of solution and of vaporisation of n-alkanes, in kcal mo1-1 and cal K-l mo1-I at 298 K methane ethane propane butane pentane hexane heptane octane nonane decane undecane dodecane - 2.3 1 3.60 5.02 6.39 7.54 8.74 9.92 11.10 12.28 13.46 14.65 - 3.30 -4.72 - 2.41 - 5.38 - 1.78 -6.21 - 1.19 - 6.5 -0.1 1 - 7.5 0.04 -8.1 0.64 - 9.5 0.42 - 32.1 14.1 - 36.3 - 22.2 16.1 - 38.9 - 22.8 18.4 - 42.1 - 23.7 20.7 - 44.0 - 23.3 22.1 - 48.0 - 25.9 23.7 - 50.3 - 26.6 25.4 - 55.9 - 30.5 a From Ducros et al.,29 except for ethane and propane from Dreisba~h.~’ Values given by From AG? and AH?, standard states 1 atm gas and From AGF and AH?, standard states pure liquid Abraham.I3 From AGF and AH?.unit mole fraction solute in solution. hydrocarbon and unit mole fraction solute in solution. to mole fraction aqueous solution standard state: n-pentane ( - 6.43), n-hexane (- 7.3 I), n-heptane (- 8.76) and n-octane (- 9.82), in kcal mol-l. When taken together with values for ethane to n-butane, table 3, these new values do not give as good a straight line as the values in table 3 on plotting against N , the solute carbon number.? I have therefore carried out all subsequent calculations with the results in table 3, but use of these new values would not affect the general conclusions reached in this work. The corresponding entropy values are in table 3, and the results of the various plots of enthalpy and entropy against carbon number (C, to C,) are in table 4.Although the solution results are not accurate enough to probe any alternation effects, they can be used to obtain the various methylene contributions. The AHF values for the n-alkanes in water show how misleading Tanford’s standard states can t Plots of AHp(g --* aq) against N have correlation coefficients of 0.9776 (new values) and 0.9946 (table 3) for ethane to n-heptane, or 0.9812 (new values) and 0.9882 (table 4) for ethane to n-octane. Values for the corresponding AHp(1iq -+ aq) plots are 0.9394 (new values) and 0.9872 (table 3) for ethane to n-heptane, or 0.9085 (new values) and 0.9586 (table 3) for ethane to n-octane.M. H. ABRAHAM I59 be. Whereas the increment to AHP(1iq -+ aq) is +0.62 kcal mol-l, it is actually -0.67 for the solution process g -+ aq.For solution of alk-l-enes in water, only the Gibbs energies are available for any extended set of solutes. There is quite good agreement on AGF values between various compilations,'. 2* 6* 2o and these values are also compatible with recent solubility meas~rements.~~ Details are in table 5, with the methylene increments given in table 4. There seems to be a slight alternation effect in AGP, although the differences (ca. 0.1 kcal mol-l) are very small. Cabani et aI.2 have surveyed results for Gibbs energies of solution of alk-l-ynes; details are in table 6. For this series of homologues there is a slight alternation effect in AGP(1iq -+ as), again very small in magnitude. The various methylene increments are in table 4.The n-alkylbenzenes have been studies by numerous sets of workers who have obtained the solubilities of the liquid solutes in order first to obtain AGP(1iq -+ as). Combination with AGP then leads to the AGp(g + aq) values often tab~lated.l-~ In table 7 are the mol fraction solubilities from a number of reviews,'? 11* 2o together with some recently determined values;30-32 in some cases recorded solubilities in mol dm-3 have been recalculated to mole fraction solubilities. In general there is reasonable agreement between the various sets. The values taken, final column table 7, are mostly based on the average values listed by Mackay and Shk2O It should be noted that the solubilities of Ben-Naim and Wilfj' are significantly larger than the average 'best' values, so that AGP values calculated from Ben-Naim and Wilf's data will be smaller than the 'best' values of AGF.This is significant when comparing values of AGF(g -+ aq)for then-alkylbenzenes, see table8, because thereisamarkeddisagreement between the values givent by Ben-Naim and Wilf3I and those calculated from the data of Tewari et aL30 Whereas the latter, as is usual for homologous series up to at least the C,, or C,, compound, increase with carbon number, the former data show a continual decrease after n-propylbenzene. The data of Tewari et al. are preferred, and in table 9 are given the various sets of Gibbs energies; the results of plots of these energies against carbon number are summarised in table 4. have obtained values of AHp(1iq -+ aq) for a number of alkylbenzenes calorimetrically, and combination with AH? values29 leads to the AHp(g -+ aq) values in table 10.Also in table 10 are the corresponding entropy values, and in table 4 are listed the regression equations for plots of all these values against carbon number. There is as good agreement as can be expected between the calorimetrically determined AH? values in table 10 and values obtained3* 3 1 9 32 from the temperature dependence of solubilities$ As well as the above results on solution of hydrocarbons, there are available also results for various other homologous series. Details of such series, in which a reasonable number of homologues have been studied, follow. Cabani et aL2 and Rytting et al.19 give values for solution of gaseous n-alkylamines from which AGp(g -+ aq) may be calculated; the two sets of data are in reasonable agreement, see table 1 I .Results are available3*- 35 for calculation of the corresponding Wadso and t Ben-Naim and WilP' list polynomials from which the Ostwald absorption coefficient may be calculated. This coefficient is then easily converted into Henry's law constant (atm/mol fraction). t Note added in proox A recent value of AGP(1iq + aq) for n-hexylbenzene (9.49 kcal mol-l) is very close to that gven in table 9 (see W. E. May, S. P. Wasik, M. M. Miller, Y. B. Tewari, J. M. Brown-Thomas and R. N. Goldberg, J . Chem. Eng. Data, 1983, 28, 197). The corresponding value for AHP(liq -+ aq) of 1.83 _+ 0.33 kcal mol-* when combined with data in table 10 leads to methylene incrementsof0.30 kcal mol-' for AHP(1iq -+ aq), -0.77 kcal mol-I for AHp(g + aq), - 1.7 cal K-I mol-I for A e ( l i q -+ aq), and -3.1 cal K-' mol-I for ASp(g -+ aq), and to Ph group contributions (see table 29) of -6.44 kcal mol-l to AHp(g aq) and - 23.0 cal K-' mol-I to A@(g -+ as).1 60 THERMODYNAMICS OF SOLUTION IN WATER Table 4.Regression equations for thermodynamic parameters of solution regression equation ra sb nc solute ranged n-alkanes AGp(g -+ aq) = 5.71 1 +(0.177+0.009) N AGp(1iq +as) = 2.399+(0.887+0.010) N AHP(g -+ aq) = -3.374-(0.673&0.035) N AHp(1iq + aq) = - 3.603 + (0.623 f 0.050) N 1@(g+ aq) = -30.51 -(2.834+0.106) N Aw(liq +as) = -20.11 -(0.883_+0.167) N AGp(g - aq) =4.895 + (0.150 5 0.024) N AGP(1iq -+ aq) = 1.424 + (0.876 & 0.028) N AGp(g -+ aq) = 3.079+(0.247_+0.010) N AGp(1iq -+ aq) = 0.129 + (0.900 0.009) N n-a1 k ylbenzenes AGp(g + aq) = 3.323 +(O.146 5 0.003) N AGp(1iq -+ aq) = 4.579+(0.809$0.010) N AHp(g -+ aq) = - 7.767 - (0.91 5 k 0.020) N AI@(liq + aq) = 0.340 + (0.070 & 0.0oO) N A@(g - aq) = -37.27-(3.500f0.116) N Aw(1iq -+ aq) = - 14.53-(2.300f0.058) N AGp(g + aq) = -0.539+(0.144+0.002) N AGp(liq -+ aq) = - 1.998 + (0.799_+ 0.004) N AHP(g - aq) = - 1 1.278 - (0.725 k 0.032) N AHP(1iq -+ aq) = - 7.102 + (0.368 k 0.095) N A@(g -+ aq) = - 36.12 - (2.890 k 0.125) N A@(liq -+ aq) = - 17.08 - (1.460 f 0.088) N AGP(g + aq) = -0.380 + (0.229 0.008) N alk- 1 -enes alk- 1 -ynes n-alk ylamines alkan-Zones 0.9927 0.9997 0.9946 0.9872 0.9972 0.9354 0.05 1 0.054 0.147 0.2 10 0.445 0.698 ethane to n-octane ethane to n-heptane 0.9420 0.9975 0.126 0.148 7 7 prop- 1 -ene to non- 1 -ene 0.9962 0.9998 0.05 1 0.047 7 7 prop- 1 -yne to non- 1 -yne 0.9992 0.9997 0.0 12 0.042 0.029 0.0oO 0.163 0.082 toluene to n-hexylbenzene toluene to n-propylbenzene 0.9993 0.9999 0.9970 0.990 1 0.9972 0.9946 0.013 0.020 0.103 0.095 0.394 0.278 ethylamine to n-octylamine ethylamine to n-hexylamine 0.9965 0.05 I 8 butan-2-one to undecan- 2-one AGP(1iq -+ aq) = - 1.610+(0.848f0.009) N AHp(g +as) = -7.547-(0.818f0.028) N 0.9997 0.998 1 0.057 0.111 8 5 butan-2-one to nonan- 2-one AHP(1iq -+ aq) = - 3.479 +(0.210 k 0.027) N A@(g -+ aq) = -23.82-(3.561 k0.128) N A@(liq + aq) = -5.89-(2.215k0.074) N 1-chloroalkanes AGp(g + aq) = 3.390+(0.163+0.019) N 0.9755 0.998 1 0.9983 0.105 0.492 0.286 5 5 5 0.9745 0.078 6 1-chloroethane to l-chloro- heptane AGP(1iq -+ aq) = 1.730+(0.870_+0.021) N 1 -bromoalkanes AGp(g --+ aq) = 3.070+(0.219+0.009) N AGp(1iq -+ aq) = I .975 + (0.920 k 0.009) N AGp(g -+ aq) = 3.163+(0.200+0.015) N 1 -iodoalkanes 0.089 0.9988 6 0.9953 0.050 7 1-bromoethane to l-bromo- octane 0.9998 0.047 7 0.9946 0.055 4 1 -iodoethane to 1 -iodo- heptane AGP(1iq -+ aq) = 2.832+(0.876+0.014) N 0.9997 0.053 4M. H.ABRAHAM 161 Table 4. (con?.) alkan- 1-01s AGp(g -+ aq) = - 1.094 + (0.163 f 0.005) N AGP(1iq + aq) = - 0.928 + (0.822 k 0.004) N b@(g + aq) = - 1 1.245 - (0.851 f 0.021) N AHp(g -, aq) = - 11.245-(0.851 k0.021) N AhHp(1iq + aq) = -3.347+(0.297f0.013) N A@(g -+ aq) = -34.07-(3.390f0.083) N ASP(liq -+ aq) = -8.17-(1.750*0.038) N methyl alkanoates AGp(g + aq) = 0.575+(0.226f0.017) N AGp(1iq -+ aq) = -0.01 5 f (0.855 k 0.020) N AHp(g -+ aq) = - 8.203 -(0.825 k0.003) N AHP(1iq -+as) = -2.570+(0.165f0.009) N A@(g + aq) = - 30.00 - (3.400 f 0.173) N A@(liq -+ aq) = - 9.20 - (2.150 f 0.260) N n-alkyl acetates AGp(g -+ aq) = 1.046 + (0.144 f 0.003) N AGP(liq -+ aq) = 1.032 + (0.768 f 0.014) N ethyl alkanoates AGp(g + aq) = 1.294+(0.099f0.019) N AGP(1iq +as) = 1.317+(0.688f0.018) N dGp(g + aq) = 1.239+(0.186f0.019) N n-alkyl propanoates ACP(1iq -+ aq) = 2.08 + (0.680 0.052) N AHP(Mg -+ aq)= 10.27-(1.01 f0.37) N alkanoic acids AHp(1iq -+ aq) = - 1.391 +(0.262+0.015) N 0.9960 0.046 0.9999 0.031 0.9991 0.066 0.9991 0.066 0.9970 0.042 0.9991 0.263 0.9993 0.120 0.9825 0.134 0.9984 0.154 0.004 0.012 0.245 0.367 0.9994 0.008 0.9997 0.031 0.8943 0.145 0.9977 0.137 0.040 0.112 0.8894 0.82 0.9968 0.033 9 9 5 5 5 5 5 propan- l-ol to dodecan- l-ol propan- l-ol to heptan- l-ol 8 methyl propanoate to 8 3 methyl propanoate to 3 3 3 methyl pentanoate methyl pentanoate 4 ethyl acetate to n-pentyl- 4 acetate 9 ethyl acetate to ethyl 9 decanoate 3 ethyl propanoate to 3 n-pentyl propanoate 4 butanoic acid to heptanoic 4 acidf a Correlation coefficient.Standard deviation. Number of solutes. Range of homologous solutes covered. If methyl decanoate is excluded, slope = (0.200&0.012), t = 0.991 1 and s = 0.075 (n = 7). f If the smoothed AH9 values are used to calculate AHp(g-+aq), the resulting equation is AHp(g + aq) = - 11.24-(0.814+0.016) N , with r = 0.9996 and s = 0.036.enthalpy and entropy of solution, see table 12. The regression equations for various plots against carbon number are in table 4. There is no sign of any alternation effect in the n-alkylamine series of results. The solution of alkanones and alkanals in water may be complicated by formation of the corresponding gem-diols ; equilibrium constants for such hydrate formation are 1 x for propanone and 1.06 for ethana1.36 The latter value is so large that observed solution parameters for alkanals cannot be used without correction for hydrate formation, but the equilibrium constants for alkanones seem small enough for hydrate formation to be disregarded. Buttery et aZ.37 have determined Henry’s law constants for a series of n-alkylmethylketones (alkan-2-ones) from which AGp(g -+ as) values may be obtained.Combination with vapour pressures38 (or AGP values) enables AGp(liq -+ aq) to be calculated. For the less soluble ketones, AGP(1iq -+ aq)may also be162 THERMODYNAMICS OF SOLUTION IN WATER Table 5. Standard Gibbs energies of solution and of vaporisation of alk-1-enes, at 298 Ka alk- 1 -ene P/atmb AG? AGp(g -+ aq) AGg(1iq -+ aq) ethene prop- 1 -ene but- 1 -ene pent-1-ene hex- 1 -ene hept- 1 -ene oct- 1 -ene non- 1 -ene 59.91 11.29 2.92 0.854 0.246 7.41 x 10+ 2.29 x 7.02 x 10-3 - 2.42 - 1.44 - 0.63 0.09 0.83 1.54 2.24 2.94 5.55 5.59 5.65 5.94 5.85 5.93 6.19 6.33 2.93 3.85 5.02 6.03 6.6P 7.47d 8.43d 9.27d a Standard states and units as in table 2. Data from ref. (1) and (6) unless shown otherwise. Ref. (27), except for the value for ethene from ref.(20). Average of values in ref. (1) and (30). Ref. (30). Table 6. Standard Gibbs energies of solution and of vaporisation of alk-1-ynes, at 298 K" alkyne P/atmb AGP AG?(g -+ aq) AGP(1iq + aq) ethyne prop- 1 -yne but- 1 -yne pent- 1 -yne hex- 1 -yne hept- 1 -yne oct- 1 -yne non- 1 -yne 48.52 5.51 1.86 0.568 0.179 6.91 x 1.79 x 8.24 x 10-3 - 2.30 - 1.01 -0.37 0.34 1.02 1.58 2.38 2.84 4.26 3.79 4.1 1 4.29 4.56 4.87 4.98 5.32 1.96 2.78 3.74 4.63 (4.85)c 5.58 (5.21)c 6.45 7.36 8.16 a Standard states and units as in table 2. Data from Cabani et aL2 unless shown otherwise. Ref. (20) and (27). Ref. (30). Table 7. Mole fraction solubilities of liquid n-alkylbenzenes, at 298 K ref. 1 31 11 20 32 30 values n-alkylbenzene (1975) (1980) (1981) (1981)a (1982) (1982) taken benzene ( x 4.14 4.21 4.03 4.12 (8) 3.73 - 4.12 toluene ( x 1.02 1.21 1.06 1.05 (7) 1.03 1.13 1.05 ethylbenzene ( x lo+) 2.61 3.61 2.69 2.85 (6) 2.87 3.18 2.85 n-propylbenzene ( x lop6) 8.26 - 8.28 - 7.64 7.84 8.00 n-butylbenzene ( x lo+) - - 1.64 1.94 (4) - 1.86 1.94 n-pentylbenzene ( x lo-') - - - - - 4.68 4.68 n-hexylbenzene ( x lo-') - - - - - 1.13 1.13 a Average values given, the number of investigations covered is in parentheses.determined directly from solubility measurements of the liquids,30* 39 there being good agreement between the two sets of values, see table 13. There is a suggestion of an alternation effect in AGp(liq--+aq), but such an effect must be very small. The regression equations yielding methylene increments are in table 4.Della Gatta et aL40 have determined AH? and AHP(1iq + aq) values for the alkan-2-ones. Their enthalpiesM. H. ABRAHAM 163 Table 8. Comparison of tabulated AGp(g -+ aq) values for n-alkylbenzenes, in kcal mol-l at 298 K ref: 3 1 31 2 from tables n-alkylbenzene (1 972) (1 975)" (1 980)" (198 1)" 7 and 9b benzene 3.40 toluene 3.39 e thy1 benzene 3.48 n-propylbenzene - n-butylbenzene - n-pentylbenzene - n-hex ylbenzene - n- hep t yl benzene - n-oc t yl benzene - 3.39 3.51 3.66 3.74 3.88 3.37 3.35 3.47 3.50 3.42 3.09 2.53 1.91 1.37 3.39 3.39 3.48 3.48 3.74 3.61 3.88 3.75 - 3.91 - 4.04 4.2 1 - a After conversion to standard states of 1 atm gas and unit mol fraction solution. Using the mole fraction solubilities in the final column of table 7 and the solute vapour pressures in table 9.Table 9. Standard Gibbs energies of solution and of vaporisation of n-alkylbenzenes, at 298 Ka P/atmb AG: AG?(g + aq)" AG?(liq + as)" n-a1 ky benzene benzene 0.1252 1.23 3.39 4.62 toluene 3.742 x 1.95 3.48 5.43 ethylbenzene 1.259 x 2.59 3.61 6.20 n-prop ylbenzene 4.524 x 3.20 3.75 6.95 n-pen t yl benzene 4.316 x 4.59 4.04 8.63 n- bu t y 1 benzene 1.428 x 10-3 3.88 3.91 7.79 n-hexylbenzene 1.383 x 5.26 4.21 9.47 a Standard states and units as in table 2. Ref. (27). Table 10. Standard enthalpies and entropies of solution and of vaporisation of n-alkylbenzenes, at 298 K" From tables 7 and 8. AH? AH? A@ A* n-alkylbenzene AHFb (g -, (liq + aq)d A* (g + aq) (liq -, aq) benzene 8.09 -7.59 0.50 23.0 -36.8 -13.8 toluene 9.08 - 8.67 0.41 23.9 -40.7 -16.8 ethylbenzene 10.10 - 9.62 0.48 25.2 -44.4 -19.2 n-propylbenzene 11.05 - 10.50 0.55 26.3 -47.7 -21.4 a Units and standard states as in table 3.Determined calorimetrically. 33 Ref. (29). From AH: and AHF(liq + aq). of vaporisation are not quite the same as those listed by Ducros et aZ.,35 but in table 14 are given all the relevant results of Della Gatta et aL40 for consistency, with the methylene increments in table 4. The aqueous solubilities of a number of liquid 1-halogenoalkanes are known, see table 15, and lead directly to AGP(1iq -+ as). Combination with values of AG? gives164 THERMODYNAMICS OF SOLUTION IN WATER Table 11. Standard Gibbs energies of solution and of vaporisation of n-alkylamines, at 298 K a n-alk ylamine P/atmb AGP AG?(liq -+ aq)d me thy lamine ethylamine n-propylamine n-butylamine n-pentylamine n-hexylamine n-hep tylamine n-octylamine 3.527 - 0.75 1.398 - 0.20 0.405 1 0.54 0.1367 1.18 4.54 x 10-2 1.83 1.50 x 2.49 5.20 x 10-3 3.12 1.76 x 10-3 3.76 - 0.29' - 0.23' -0.12' 0.03" (-0.02)' 0.18" (0.18)' 0.32" (0.24)' 0.4P 0.62" - 1.04 - 0.43 0.42 1.21 2.01 2.81 3.60 4.38 a Standard states and units as in table 2.Ref. (19) and (27). ' Ref. (2). From the previous two columns. Ref. (19). Table 12. Standard enthalpies and entropies of solution and of vaporisation of n-alkylamines, at 298 Ka AH? AH? A* A* n-alkylamine AH? (g -+ aq) (liq --+ aq) A,!@ (g --+ aq) (liq -+ aq) methylamine 5.80 - 10.82 -5.02 22.0 ..: 35.3 - 13.3 ethylamine 6.36 - 12.83 -6.47 22.0 -42.3 - 20.3 n-butylamine 8.53 -14.11 -5.58 24.7 -47.4 - 22.7 n-pentylamine 9.58 - 14.85 -5.27 26.0 - 50.4 - 24.4 n-hexylamine 10.78 - 15.72 -4.94 27.8 - 53.8 - 26.0 n-propylamine 7.49 - 13.38 -5.89 23.3 -44.5 -21.2 a Standard states and units as in table 3.Results from ref. (34) and (35). Table 13. Standard Gibbs energies of solution and of vaporisation of alkan-2-ones, at 298 Ka alkan-2-one P/atmb AGP AG?(g -+ aq)" AG?(liq -+ aq)d propanone but an-2-one pentan-2-one hexan-2-one heptan-2-one octan-Zone nonan-2-one decan-2-one undecan-2-one 0.304 0.71 0.1 19 1.26 4.66 x 1.82 1.53 x 2.48 5.07 x 10-3 3.13 1.78 x 10-3 3.75 - (4.34)f - (4.96)5 8.42 x 10-5 5.56 0.46 0.56 0.75 0.98 1.23 1.39 1.78 1.928 2.1 1 1.17 1.82 2.57 (2.61)" 3.46 (3.44)e 4.36 (4.32)e 5.14 (5.18)e 6.12 (5.90)" 6.88e 7.67 a Standard states and units as in table 2.From Henry's law constants in ref. (37). Ref. (38) except for undecan-2-one [ref. (37)]. Calculated from solubilities of the liquid alkan-2-0nes.~~. 39 f Estimated from a plot of AG? against carbon number (butan-2-one to undecan-Zone). g From AG?(liq --+ aq) and AGP. From the previous two columns.M. H. ABRAHAM 165 Table 14. Standard enthalpies of solution and of vaporisation of alkan-2-0~, at 298 Ka AH$' AH? A* A* alkan-2-one AH? (g 3 aq) (liq -B aq) A* (g + aq) (liq + aq) ~ ~~ propanone 7.29 -9.72 -2.43 22.0 - 34.1 - 12.1 butan-2-one 8.35 - 10.91 -2.56 23.8 - 38.5 - 14.7 pentan-2-one 9.20 -11.64 -2.44 24.8 -41.6 - 16.8 hexan-2-one 10.12 - 12.38 -2.26 25.6 -44.8 - 19.2 heptan-2-one 1 1.02 -13.16 -2.14 26.5 -48.3 -21.8 nonan-Zone 13.51b - 15.01' - 1.Y 30.7 - 56.3 - 25.6 a Units and standard states as in table 3; all results from Della Gatta et aL40 (in enthalpies) except for nonan-2-one.Ref. (35). From AH? and AHP(1iq --+ aq). R. Bury, M. Lucas and P. Barberi, J. Chim. Phys.. 1978, 75, 575. Table 15. Mole fraction solubilities of 1-halogenoalkanes, at 298 K 1-halogenoalkane ref: 1 39 30 taken values 1 -chloropropane 1 -chlorobutane 1 -chloropentane 1 -chloroheptane bromoethane 1 -bromopropane 1 -bromobutane 1 -bromopentane 1 -bromohexane 1 -brornoheptane 1-bromo-octane iodomethane iodoe t hane 1 -iodopropane 1 -iodobutane 1-iodoheptane 6.26 x 1.31 x 10-4 3.36 x 10-5 5.37 x 10-4 1.30 x 10-4 1.50 x 10-3 3.60 x 10-4 7.89 x 10-5 1.59 x 10-3 3.34 x 10-4 7.78 x 10-5 1.81 x 10-3 4.54 x 10-4 1.14 x 10-4 2.07 x 10-5 - 1.81 x 10-3 4.54 x 10-4 9.27 x 10-5 1.98 x 10-5 - 1.70 x 10-4 - 1.82 x - 1.14 x 10-4 1.51 x 10-5 1.56 x 10-7 2.82 x 10+ 6.70 x - 2.80 x 10-7 - 5.82 x 10-4 1.31 x lo-* 3.36 x 1.82 x lo+ 1.55 x 10-3 3.47 x 10-4 7.84 x 10-5 1.51 x 10-5 6.70 x 10-7 1.81 x 10-3 4.54 x 10-4 1.03 x 10-4 2.02 x 10-5 2.80 x 10-7 2.82 x 1.56 x lov7 AGp(g-+aq) as shown in table 16; the methylene increments are listed in table 4.Unfortunately there are too few AH? values2 to determine methylene contributions. The alkan-1-01s form a particularly important series of solutes. They have been studied a number of times, and the values of AGp(g ---* aq) have been used by Beezer and Hunter25 as an example of the alternation effect; that is, the alternation or oscillation of methylene increments along an homologous series.Beezer and Hunter25 used the old data of Butler et a[.,14-17 but there have been many subsequent investigations and it seemed essential to set out the best available values to date. Solubilities of the higher alcohols have been determined not only by Butler et al.14-17 but also by Amidon et U Z . , ~ Tewari et al. 30 and Kinoshita et aL41 The AGP(1iq as) values calculated directly from mole fraction solubilities are in table 17, together with values from Raoult's law activity coefficients given by Pierotti et ~ 1 . ~ ~ In order to convert these AGP (liq -+ aq) values into AGp(g -+ aq) it is necessary to know the166 THERMODYNAMICS OF SOLUTION IN WATER Table 16. Standard Gibbs energies of solution and of vaporisation of 1-halogenoalkanes, at 198 K" 1-halogenoalkane P/atmb AGg AGp(g + aq)" AG?(liq + aq)d chloromethane chloroethane 1 -chloropropane 1 -chlorobutane 1 -chloropentane 1 -chlorohexane 1 -chloroheptane 1-chloro-octane bromomethane bromoe t hane 1 -bromopropane 1 -bromobutane 1 -broniopentane 1 -bromohexane 1 -bromoheptane 1-bromo-octane iodomethane iodoethane 1 -iodopropane 1 -iodobutane 1 -iodopentane 1 -iodohexane 1 -iodoheptane I-iodo-octane 5.672 1.577 0.454 0.135 4.09 x 1.27 x 3.99 x 2.148 0.617 0.182 5.43 x 10-2 1.66 x 1.25 x 10-3 - 1.03 -0.27 0.47 1.19 1.89 2.59 3.27 3.96 - 0.45 0.29 1.01 I .73 2.43 3.72e 3.64g 3.94 4.1 1 (4.05)h 4.21 (4.26)h (4. 27)h 4.56 - 3.46e 3.54 3.71 3.87 4.18 5.13 x 1.61 x 5.00 x 0.533 0.179 5.67 x 1.82 x 0-3 3.12 4.4s 0-3 3.81 4.61 0-4 4.50 4.79 0.37 3.31 1.02 3.54 0 - 2 1.70 3.74 0-2 2.37 4.03 - 5.78 x 3.05 1.84 x I 0-3 3.73 - 5.92 x 10--4 4.40 4.54 1.84 x lo-* 5.10 __ 2.69f 3.37f 4.4 1 5.30 6.10 6.86f 7.83 - 3.01f 3.83 4.72 5.60 6.58 7.57 8.42 9.29 3.74-f 4.56 5.44 6.40 - 8.94 - a Standard states and units as in table 2.Ref. (27). Obtained from AGF(1iq + aq) and From the solubilities given in table 15 except where shown. D. N. Clew and E. A. Moelwyn-Hughes, Discuss. Faraday Soc., 1953, 15, 150. f From From Henry's law constants, D. T. Leighton Jr AGF except where shown. AGg(g + aq) and AGP. and J. M. Calo, J . Cheni. Eng. Data, 1981, 26, 382. ' Ref. ( I ) and (2). Table 17. Values of AGF(liq + aq) from mole fraction solubilities of alcohols, at 298 K ref: 14 42 8 41 30 values alcohol (1933) (1959) (1974) (1958) (1982) taken pen tan- 1 -01 3.19 3.04 3.19 3.19 3.57 3.19 hexan- 1-01 4.03 4.03 4.06 4.26 4.03 heptan- l-ol 4.85 - 4.85 4.88 5.03 4.85 5.58 octan- 1-01 5.58 5.5 1 5.58 5.69 nonan- l-ol - _.6.47 6.49 6.65 6.48 decan- 1-01 - 7.26 7.42 7.33 7.34 undecan- 1-01 - - - - - - 8.93 dodecan- l-ol - - 8.71 (8.93)" - - - - - a Given in ref. (46).M. H. ABRAHAM 167 vapour pressures of the alcohols at 298 K. These are quite well established for the alkan-1-01s up to h e p t a n - l - ~ l , l * ~ ~ but those for the higher alkan-1-01s are not so reliable. There have been several sets of determinations or evaluations of vapour pressures of a l k a n - l - ~ l s , ~ ~ - ~ ~ but only in the study by Davies and K ~ b e t t ~ ~ were temperatures close to room temperatures used.In table 18 are listed AG? values derived from the vapour pressures given by Rytting et aZ.19 and by Davies and K ~ b e t t . ~ ~ In other cases, the required AGP values have been obtained by interpolation of the (good) straight line plot of AGP against carbon number. For the lower alkanols, the solubility method is not appropriate, but Henry’s law constants have been obtained by several w o r k e r ~ . l ~ - ~ ~ ? l9 These are all in reasonable agreement and table 18 presents AGp(g --+ aq) values for the C, to C, alkan-1-01s derived from results of Rytting et a1.19 Table 18. Standard Gibbs energies of solution and of vaporisation of alkan-1-ols, at 298 Ka alkan- l-ol P/atmb AG? AGF(g --* aq)“ AGF(liq -+ aq)d methanol ethanol propan- 1-01 butan- 1-01 pentan- 1-01 hexan- l-ol heptan- 1-01 octan- l-ol nonan- 1-01 decan- 1-01 undecan- 1-01 dodecan- 1 -01 tridecan- 1-01 tetradecan- l-ol pentadecan- l-ol hexadecan- 1-01 octadecan- 1-01 0.1 672e 0.0776 0.0264 2.85 x 8.92 x 10-3 9.47 x 10-9 1.07 x 10-4i 1.20 x 10-5” - - 1.1 1 x 10-6” - 1.36 x 10-7“ - 1.57 x 1.06 1.51 2.15 2.80 3.47 4.13 4.799 5.42 6.099 6.7 I 7.40s 8.12 8.719 9.37 10.029 10.65 12.00s -0.83 - 0.73 -0.58 - 0.45 -0.30 -0.14 0.06h 0.18j 0.39h 0.63h 0.81h 0.54l 0.53h 0.27l 0.211 - - 0.23 0.78 1.57 2.35 3.17 (3.19)f 3.99 (4.03)f 4.89 5.60 (5.58)f 6.48f 7.34f 8.9Y - 9.91l (10.34)m 10.5P (1 1.03p 10.92l (1 1.93)m 12.21l (13.84)m a Standard states and units as in table 2.From ref. (19) and (43) except where shown.From AGF and AGp(g --* aq) except where shown. Ref. (45). From the mole fraction solubiIities in table 17. 9 Estimated from a plot of AGF against carbon number. From AG? and AGp(1iq -+ aq). G. Geiseler, J. Fruwert and R. Huetting, Chem. Ber., 1966. 99, 1594 [see ref. (45)]. j From a direct determination3? of Henry’s law constant. From the solubility of the solid Value of AGF(so1id -+ aq). From a corrected value of the solubility given in ref. (8). From Henry’s law constantslg except where shown. Ref. (43). corrected through eqn (9) and eqn (lo), see text. The solubilities of the solid alkanols, tetradecan-1-01, pentadecan- 1-01 and hexadecan- 1-01, have been determined by Amidon et aZ.,* who corrected the observed solubilities to the hypothetical solubilities of the liquid alkanols using literature values for enthalpies of fusion of the alkanols.More recently, Yalkowsky and V a l ~ a n i ~ ~ listed molar solubilities of the alkanol-1-01s as averages of literature values but gave no literature references. The listed46 values for tetradecan- 1-01 and pentadecan- 1-01 are the same as those given before (log S = -5.84 and -6.35, respectively),? but new t S is the solubility in mol dmP3.168 THERMODYNAMICS OF SOLUTION IN WATER values for hexadecan-1-01 (log S = 7.00) and octadecan-1-01 (log S = -8.40) were given. After recalculation to the mol fraction scale, these values lead directly to AGp(so1id --+ aq), and application of eqn (9) affords the required values of AGP(1iq + as). Unfortunately the various solid-solid transitions of these alkanols are not well documented and the only correction that can be applied is that of an ‘overall’ AHg value that includes any transition AH? values.For the alkanols tetradecan-1-01, hexadecan-1-01 and octadecan-1-01 such a correction has been made using the AHg and T, values given by Davies and K ~ b e t t . ~ ~ Another procedure is to calculate AG?(g + aq) through the equation AGP(g --+ aq) = AGP(so1id --+ aq) - AGgbl (10) where AGgbl is the standard Gibbs energy of sublimation of the solid at the experimental temperature, T. This method, whilst theoretically impeccable in that no solid-solid transitions need be considered, is subject to possibly large errors in AGg?,,, obtained as AGgbl = - R T h Psubl, where Psubl is the sublimation vapour pressure at temperature T, in atm.Davies and K ~ b e t t ~ ~ recorded &,l for the even alkan-1-01s and so it is possible to compare AGP(g --+ aq) obtained via eqn (10) with values obtained through eqn (9) together with eqn (2). These are, respectively, in kcal mol-1 : tetradecan-1-ol(O.51 and 0.57), hexadecan-l-ol(0.29,0.25) and octadecan-1-ol(O.41, 0.01). Considering that the first set of figures includes AGgbl, and the second set in- cludes not only the correction via eqn (9) but also an extrapolated value of AG?, agreement between the two sets of values is remarkably good. Table 18 gives AGP(g + aq) as average values of the two sets for the three even alkan-1-01s studied, and the value given by Amidon et aZ.* for pentadecan-1-01. Plots of AGP(g + aq) or AGp(1iq --+ aq) against N are reasonably linear for N = 3 to N = 12, but beyond dodecan-1-01 the AGP values become more negative than expected, so that, as for the n-alkanes, the very-long-chain alkan- 1-01s are more soluble in water than expected.This effect, possibly due to coiling of the alkyl chains,8 although large, is not of the same magnitude as for the n-alkanes. Thus n-octadecane is more stable in solution by 5.0 kcal mol-l, corresponding to an increased solubility by a factor of 4800, but octadecan-1-01 is more stable in solution by 1.65 kcal mol-l, corresponding to an increased solubility by a factor of only 16. Inspection of the results in table 18 for the C, to C,, alkanols indicates that any alternation effect must be within experimental error. In table 19 are set out three series of values for AGP(g + aq), the original ones of Butler et aI.,l7 the more recent values of Amidon et aL9 and those from table 18.The slight alternation effect observed by Beezer and Hunter2, in the results of Butler et al. is not found at all with the results on the C , to C, alkanols from table 18. Enthalpies of solution of the alkanols have also been studied by several sets of w o r k e r ~ . ~ ~ - ~ ~ The usual method is to obtain AHp(1iq -+ aq) by direct calorimetry and to combine this with AH? to yield AHP(g --+ as). Fortunately, AH? values are well known,29 and there is good agreement between at least four sets of calorimetrically determined AHP(1iq + aq) values for the C , to C , a l k a n - l - ~ l s ~ ~ - ~ ~ and three sets of AH? values for pentan-1 For hexan- l-ol and heptan- l-ol, Hill and White53 have obtained AHp(liq --+ aq) both by calorimetry and by the temperature variation of solubility, and from a graph given by Hill and White53 a value of - 0.76 kcal mol-1 for AHp(liq + aq) for octan-1-01 may be deduced.has obtained a value of AHp(1iq --+ aq) from the temperature variation of the solubility of dodecan- 1-01 but unfortunately gives no numerical data: from an expanded reproduction of a graphM. H. ABRAHAM 169 Table 19. Examination of the alternation effect in AGp(g --+ as) for alkan- 1 -01s’ at 298.1 5 K alkan-1-01 Butler et a1.l’ Amidon et aL9 this work methanol ethanol propan- l-ol - 0.57 butan- l-ol - 0.44 pentan-1-01 -0.21 I hexan- l-ol heptan- 1-01 octan- l-ol 0.17 nonan- 1-01 decan- l-ol - 0.10 0.17 0.13 0.23 0.12 0.12 0.14 0.72 0.21 0.13 0.22 0.21 -0.28 0.33 0.38 0.26 -0-83 1 -0.73 \ I -0.45 \ -OS8 1 0.63 0.10 0.15 0.13 0.15 0.16 0.20 0.12 0.21 0.24 in Benjamin’s paper, a value of 2.60 kcal mol-l was estimated.? The AHp(g -+ aq) and AHP(1iq + as) values are set out in table 20.Benjamin4, showed that in a plot of AHp(liq + aq) against carbon number, points for the first three or four alkanols did not fall on the more-or-less straight line from butan-1-01 to dodecan-1-01. An equivalent graph constructed by Hill and White, however, is clearly a continuous curve with no straight section at all. However, from the numerical data in table 20, it seems that for the C, to C , alkan-1-01s it is possible to construct a reasonable straight line.It is rather unfortunate that the main outlying point, for dodecan-1-01, is derived from Benjamin’s graph and not from results of any rigorously presented experiment. On the assumption that the C , to C , alkan-1-01 points do lie on a straight line, the methylene increments listed in table 4 have been calculated. The solubilities of a very large number of esters have been determined,l. 2 y 3 0 7 39 but there are only four series of these compounds for which a reasonable number of homologues have been studied. Buttery et aZ.37 determined Henry’s law constants for a series of methyl alkanoates, leading to AGp(g --+ as), and combination with AGP values obtained from several ~ o u r c e s ~ ~ ~ 54* 55 leads to AGP(1iq + as). The latter are in reasonable agreement with values derived from sol~bilities~~ of the liquid esters.In addition, Tewari et al.,O have recorded solubilities of the higher esters, methyl nonanoate and methyl decanoate, leading to the various AG? values as set out in table 21. Kieckbusch and King56 have determined Henry’s law constants for several n-alkyl acetates, from which the various AGP values may be obtained, see table 22. There t Note ihat numerical values of AHP(1iq + aq) from Benjamin4’ or from Hill and White53 are often quoted, but Benjamin gives no numerical values at all, and Hill and White numerical values for only hexan- 1-01 and heptan- 1-01.170 THERMODYNAMICS OF SOLUTION IN WATER Table 20. Standard enthalpies and entropies of solution and of vaporisation of alkan- 1 -ols, at 298 Ka A€€?- AH?- A*- Ah?- alkan- 1-01 AH$'b (g --+ aq)" (liq --* aqId A* (g --+ aq) (liq -+ aq) methanol ethanol propan- 1-01 butan-1 -01 pentan- l-ol hexan- 1-01 heptan- l-ol octan-1-01 nonan- l-ol decan- 1-01 undecan- 1-01 dodecan- 1-01 8.95 10.10 11.35 12.51 13.61 14.75 15.97 16.96 18.37 19.48 20.688 21.98 - 10.69 - 12.53 - 13.77 - 14.73 - 15.46 - 16.30 - 17.24 - 17.72 - 1.74 - 2.43 - 2.42 - 2.22 - 1.85 - 1.55" - 1.27" - 0.76f - - 19.38 - 2.60h 26.5 28.8 30.8 32.6 34.0 35.6 37.5 38.7 41.2 42.8 44.5 46.5 - 33.1 - 39.6 -44.2 - 47.9 - 50.8 - 54.2 - 58.0 - 60.0 - - 6.6 - 10.8 - 13.4 - 15.3 - 16.8 - 18.6 - 20.5 -21.3 - - 67.0 - 20.5 a Standard states and units as in table 3. Average values from ref. (49)-(52), unless shown otherwise. Ref. (29). From AH$' and AH?(liq --* as).Average of values in ref. (53) obtained by calorimetry and temperature variation of solubility. f Taken from a graph given in ref. (53). 8 Interpolated value from a plot of AH$' against carbon number. Taken from a graph given in ref. (47). Table 21. Standard Gibbs energies of solution and of vaporisation of methyl alkanoates, at 298 Ka methyl alkanoate P/atmb AG? AG?(g -+ aq)' AGP(1iq -+ aq)d methyl formate methyl acetate methyl propanoate methyl butanoate me thy1 pent anoate methyl hexanoate methyl heptanoate methyl octanoate methyl nonanoate methyl decanoate 0.822 0.2845 0.1136 4.24 x - 4.29 x 10-3 5.25 x 10-4 - 6.44 x 10-5 0.12 0.74 1.29 1.87 2.54-f 3.15 3.82f 4.47 5.08f 5.72 1.49 1.10 1.34 1.44 1.70 1.78 2.23 2.588 3.059 - 1.61 1.84 (1.67)e 2.63 (2.51)" 3.31 (3.44)e 4.24 4.93 6.70 - (7.66)h (8.77)h a Standard states and units as in table 2. From ref. (I), (46), (54) and (55). Ref. (37), From AG$' and AG?(g -+ aq) f Estimated from a plot of AG$' From solubilities of the liquid except for methyl formate, ref. (2), from Henry's law constants. except where shown. against carbon numbers. From solubilities of liquid From AG$' and AG?(liq + as). is reasonable agreement between the indirectly determined AGP(1iq -+ aq) values and those calculated from solubilities of the liquid esters.'* 30- 57 For a rather long series of ethyl alkanoates, on the other hand, it is the liquid ester solubilitiesl9 39 that lead to a prime set of AG?(liq -+ as) values. There are only a limited number of AG? values available,lY 4 5 9 58 but it is possible to obtain the remaining valuesM.H. ABRAHAM 171 Table 22. Standard Gibbs energies of solution and of vaporisation of n-alkyl acetates, at 298 Ka n-alkyl acetate P/atmb AGF AGF(g + aq)" AGP(1iq --* aq)d methyl acetate 0.2845 0.74 1.16 1.90 (1.67)" - (1.59)g n-propyl acetate 4.438 x 1.85 1.48 3.33 (3.28)" (3.33)s (3.30)s n-butyl acetate 1.45 x lop2 2.51 1.63 4.14 (4.25)" (4.07)f (4.13)g n-pentyl acetate 5.39 x 10-3 3.09 1.76 4.85 (4.92)" - (4.90)g ethyl acetate 0.1244 1.23 1.33 2.56 (2.43)" (2.57)s - a Standard states and units as in table 2. Ref. (1) and (55). From directly determined From AGF and AGp(g + aq); values in parentheses are from Henry's law constants.56 solubilities of the liquid esters. Ref. (1). f Ref. (30).9 Ref. (57). "able 23. Standard Gibbs energies of solution and of vaporisation of ethyl alkanoates, at 298 Ka ethyl alkanoate P/atmb AGV AGg(g -, aq)" AGP(1iq + aq)d ethyl formate ethyl acetate ethyl propanoate ethyl butanoate ethyl pentanoate ethyl hexanoate e thy1 hep tanoa te ethyl octanoate ethyl nonanoate ethyl decanoate 0.3218 0.67 0.1244 1.23 4.89 x 1.79 1.98 x 2.32 6.32 x 10-3 3.00 8.95 x 10-4 4.16 - 3.568 -_ 4.74s - 5.339 - 5.929 1.70 1.33" 1.59 1.77 1.77 2.03 1.95 2.26 2.23 2.04 2.27 2.56" (2.57)f 3.38 (3.51)s 4.09 4.77 5.59 6.1 1 7.00 7.56 7.96 a Standard states and units as in table 2. Ref. (l), (46) and (58). From AGF and From solubilities of the liquid esters'. 39 except where Estimated from plots of AG$? against carbon number. AGp(1iq --* aq) except where shown.shown. From table 22. f Ref. (30). from plots of AGF against carbon number, see table 23. A much shorter series of n-alkyl propanoates can also be constructed, see table 24. In all four series of esters, the methylene increments are not as regular as in most of the other series, but there seems to be no unusual effects in AGp(g -+ as) or in AGp(liq + as). There are a few esters for which values of AHp(liq + as) have been determined calorimetrically, and in table 25 are given results of Della Gatta et aL40 for a series of methyl alkanoates. The only other homologous series for which enthalpy data is available is that of the n-alkyl acetates, but values of AHF(1iq -+ as) given by Richon and Viallard59 and by Cross and McTigueGo are not in good agreement, viz, respectively, methyl acetate (- 2.11, - 1.70), ethyl acetate (- 2.35, - 2.09), propyl acetate (-, - 1.90) and butyl acetate (- 2.24, - 1.87), all in kcal mol-1 at 298.15.f.There have been several investigations into the solution of alkanoic acids in water,48* 6 4 9 65 but there are considerable difficulties in interpretation of results due to t The value for propyl acetate is at 297.05 K. Note also other values for methyl acetate ( - 1.87 kcal mol-I) (table 25), and ethyl acetate ( - 2.23,61 -2.346* and - 2.3663 kcal mol-l).172 THERMODYNAMICS OF SOLUTION IN WATER Table 24. Standard Gibbs energies of solution and of vaporisation of n-alkyl propanoates, at 298 Ka n-alkyl propanoate P/atmb AG? AGF(g -+ aq) AG?(liq + aq) methyl propanoate 0.1136 1.29 1 .34c 2.63c ethyl propanoate 4.89 x 1.79 1 .59d 3.38d n-propyl propanoate 1.79 x 10+ 2.38 1 .83e 4.21f n-butyl propanoate 5.55 x 10-3 3.08 - - n-pentyl propanoate 3.87 x 10-3 3.29 2.1 6e 5.49 a Standard states and units as in table 2.From AGF and AG?(liq + aq). f Ref. (1). Ref. (l), (46) and (58). Table 21. Table 23. Table 25. Standard enthalpies and entropies of solution and of vaporisation of methyl alkanoates, at 298 Ka AH?- AH?- A@- A*- methyl alkanoate AH? (g -+ aq) (liq + aq) A* (g + aq) (liq + aq) methyl acetate 7.57 -9.44 -1.87 22.9 - 35.3 - 12.4 methyl propanoate 8.61 - 10.68 -2.07 24.5 - 40.3 - 15.8 methyl butanoate 9.58 -11.50 -1.92 25.9 - 43.4 - 17.5 methyl pentanoate 10.59 -12.33 -1.74 27.0 - 47.1 - 20.1 a Standard states and units as in table 3. All data on enthalpies from ref.(40). Table 26. Enthalpies of solution and of vaporisation of alkanoic acids, at 298 Ka AH? AH?(g+Wb alkanoic acid Id IIe If 11s AHg(1iq + as)" methanoic acid ethanoic acid propanoic acid butanoic acid pentanoic acid hexanoic acid heptanoic acid octanoic acid 11.1 12.3 13.1 13.9 14.9 17.5 17.2 19.8 10.93 12.00 13.08 14.15 15.23 16.30 17.38 18.45 -11.3 - 12.6 - 13.5 - 14.2 - 15.0 - 17.3 - 16.8 - 19.1 - 11.09 - 12.28 - 13.45 - 14.50 - 15.32 - 16.08 - 16.96 - 17.75 -0.16 - 0.28 - 0.37 -0.35 - 0.09 0.22 0.42 0.70h a Standard states and units as in table 3; all values refer to the monomeric alkanoic acids. From AH$? and AHF(1iq + aq). Ref. (48) and (65). Observed values from ref. (65) and (67), corrected where necessary to vaporisation to monomeric acid.Smoothed values from a plot of AH? against carbon number, table 27. f From AHg(1) and AH?(liq + as). g From AHF(I1) and AHP(1iq + as). Estimated value.M. H. ABRAHAM 173 dimerisation of the acids in the vapour phases5-s7 and possibly also in aqueous The latter is not so much of a problem, and AHP(1iq -+ aq) values are known for all the n-alkanoic acids up to heptanoic 65 Dimerisation in the vapour phase, however, is extensive especially for the lower acids (acetic acid for example is 90% associate6 at 298 K)ss, but Konicek and Wadsos5 have corrected observed AHv values to vaporisation from the liquid to the gaseous monomer, and de Kruif and Oonks7 have obtained the required AH? values for the homologues pentanoic to octanoic acid. In table 26 are set out the various enthalpies of solution of the pure liquid and the monomeric gaseous acids.There are unfortunately not enough data on AGP valuess4 to construct a reasonable series of homologues. Inspection of table 26 shows that, as in the case of the alkan-1-ols, the AHP(1iq -+ aq) values for the lower homologues are more positive than expected. It is not so easy to discern such a trend in the AHP(g -+ aq) values because the observed AH? values are subject to considerable random error. However, de Kruif et aLs8 have determined AH? values for a large number of the higher n-alkanoic acids. A plot of AH? against N is linear from N = 1 to N = 15, and if smoothed values of AH? are obtained from such a plot, the resulting AHp(g + aq) values show exactly the same trend as do the AHp(1iq -+ aq) values, see table 26.Wadso et aZ.69 have obtained values of AH?, AHp(liq + aq) and AHp(g --* aq) for two series of difunctional compounds, ROCH,CH,OH and ROCH,CH,OMc (where R = methyl, ethyl, n-propyl and n-butyl), but neither series is extensive enough to ascertain if the methylene increments are constant or not. However, in a related series of RO(CH,CH,),H (where R = H, ethyl, n-butyl, n-hexyl, n-octyl, n-decyl and n-dodecyl) the methylene increment in AHP(1iq + aq) is definitely not constant but varies in quite a similar manner to AHP(liq -+ aq) for the alkan-1-01s. The solubiluty of a similarly extensive series of n-alkyl p-aminobenzoates has been determined by Yalkowsky and Valvani ;46 for the homologues methyl to n-dodecyl (except the n-decyl and n-undecyl compounds) the methylene increment for solution of the solid esters is constant at 0.82 kcal mol-l, a value quite comparable to those of AGP(1iq + aq) for the various series in table 4.DISCUSSION THERMODYNAMICS OF VAPORISATION The parameters for vaporisation in the various homologous series are quite linear with respect to carbon number, N . Regression equations for AG?, AH? and A* against N are collected in table 27; the correlation coefficients and standard deviations indicate that these simple equations are good enough for the estimation of vaporisation parameters within any homologous series. Indeed, if a datum point deviates signifi- cantly from the regression equation, it is almost certainly due to experimental error.7 The slopes of these regression equations differ from series to series because of the different solute-solute interactions in the pure liquid state, so that it is not possible to construct a generalised equation in, say, AG? that will cover all the homologous series.Because AG? and AH? are both linear in N , it follows that AH? is linearly related to AGP in any given series. This is the condition for the so-called compensation phenomenon to a ~ p l y , ~ ' ~ 72 and wherever reasonable sets of AH? and AGP values are available, there is found an excellent linear relationship. The slopes of the regression t Except for the first member of the homologous series, where the vaporisation parameters do often deviate from the given regression equations.174 THERMODYNAMICS OF SOLUTION IN WATER Table 27.Regression equations for thermodynamic parameters of vaporisation regression equation ra n-alkanes AG? = - 3.2802 + (0.7027 & 0.001 5) N AH? = 0.06424-(1.2247 k0.0125) N A* = 11.221 +(1.7502&0.0373) N alk- 1 -enes AG? = -3.4768+(0.7146&0.0026) N AH? = 0.1832+(1.1861 k0.0066) N AS$' = 12.276+(1.5812f0.0183) N alk- 1 -ynes AG? = -2.9281 +(0.6504&0.0185) N AH? = 2.2355 + (0.891 7 k 0.0387) N AS? = 17.319+(0.8094&0.1268) N n-alkylbenzenes AG? = 1.2576+(0.6601 kO.0058) N AH? = 7.6490+(1.1263f0.0201) N AS? = 21.437 + (1.5636 & 0.0720) N n-alkylamines AGF = - 1.4759 + (0.6595 & 0.0074) N AH? = 4.1222+(1.1091 k0.0139) N AS? = 18.776+(1.5080&0.0591) N alkan-2-ones AGF = - 1.2321 +(0.6192&0.0043) N AH? = 3.7771 +(1.0790+0.0292) N AS? = 16.801 +(1.5423+0.1034) N 1 -chloroalkanes AG? = - 1.6469 + (0.7036 k 0.0042) N AH? = 3.2815+(1.1768f0.0051) N AS? = 16.530 + (1.5868 k 0.03 1 1) N 1 -bromoalkanes AG? = - 1.0968+(0.7019~0.0030) N AH? = 4.0949 + (1.I837 & 0.0059) N AS? = 17.413+(1.6161 k0.0297) N 1 -iodoalkanes AGF = -0.3377 + (0.6783 & 0.0009) N AH? = 5.1604+(1.1465+0.0115) N AS? = 18.440+(1.5703f0.0374) N alkan- 1-01s AG$' = 0.1912+(0.6552f0.0019) N AH? = 7.6911+(1.1863+0.0067) N AS? = 25.155+(1.7813+_0.0213) N 1 .oooo 0.9995 0.9980 1 .oooo 0.9999 0.9997 0.9984 0.9962 0.9542 0.9998 0.9990 0.9937 0.9997 0.9997 0.9969 0.9999 0.9978 0.9868 0.9999 1 .oooo 0.9990 1 .oooo 0.9999 0.9992 1 .oooo 0.9997 0.9986 1 .0000 0.9998 0.9985 sb nc solute ranged 0.020 0.130 0.391 0.0 14 0.035 0.097 0.078 0.162 0.530 0.037 0.130 0.466 0.03 1 0.058 0.247 0.028 0.189 0.670 0.022 0.027 0.165 0.0 16 0.03 I 0.157 0.004 0.061 0.198 0.027 0.098 0.310 13 ethane to n-tetradecane 11 ethane to n-dodecanee 1 1 7 but-1-ene to dec-1-end 7 7 6 but- 1 -yne to non- 1 -yne 6 6 8 ethylbenzene to n-nonyl- 8 8 benzend 6 ethylamine to n-heptyl- 6 6 amine 8 butan-2-one to undecan- 8 8 2-ones 7 1-chloroethane to 1 -chlorooctanee 7 7 7 1-bromoethane to 1 -bromooctanee 7 7 7 1-iodoethane to 1-iodo- 7 7 octanee 12 ethanol to hexadecan- 12 12 1 -olhM.H. ABRAHAM Table 27. (cont.) 175 methyl alkanoates AGV = -0.65 17 + (0.6374 & 0.0045) N 0.9999 AH? = 5.4534 + (1.0347 k 0.0071) N A S = 20.447 + (1.3324 & 0.0251) N 0.9999 0.9993 n-alkyl acetates AG? = -0.01 39 + (0.6242 & 0.0099) N 0.9998 AG? = 6.5050+(0.9750+0.0116) N 0.9999 AS? = 21.864 + (1.1765 k 0.0293) N 0.9994 ethyl alkanoates AG? = 0.0282+(0.5888 kO.0110) N 0.9995 n-alkyl propanoates AG? = 0.7776 + (0.5294 k 0.0424) N 0.9905 n-alkanoic acids AH? = 9.8538 + (1.0747 & 0.01 22) N 0.9994 0.026 0.042 0.146 0.022 0.026 0.065 0.042 0.134 0.201 6 methyl propanoate to 6 6 methyl decanoatei 4 ethyl acetate to n-pentyl 4 4 aceta teb 5 ethyl acetate to ethyl heptanoa te 5 methyl propanoate to n-pentyl propanoate 11 methanoic acid to penta- decanoic acidi a Correlation coefficient.Standard deviation, defined as (b(obs.) -y(calc.)12/(n - 2)}:. All values for the solutes taken from Additional AH? values from ref. (27). f Additional Additional Additional AH: values from R.Fuchs and L. A. Peacock, Can. Number of solutes used in the regression equation. previous tables, unless shown otherwise. AG? and AH? values from ref. (27). AH? values from ref. (29). J. Chem., 1980, 58, 2796. j Additional AH? values from ref. (68). Additional AH? values from ref. (35). equations differ from one homologous series to another, and therefore there is no common compensation temperature, p, amongst these series. Furthermore the standard deviations in AH? from plots of AH? zgainst AG? are usually rather larger than the standard deviations from plots of AH? against N , so that for the prediction of new AH? values, the latter plots are generally to be preferred. THERMODYNAMICS OF SOLUTION For nearly all the homologous series, there are found excellent linear plots of AGP or AH? against N , as summarised in table 4.Before considering the important deviations from such linearity, it seemed first useful to discuss the information that can be derived from the linear regression equations in table 4. The slopes of the regression equations represent the contribution of a methylene group to the particular process considered; thus for solution of the n-alkanes from ethane to n-octane, the methylene contribution to AGP(g -+ aq) is 0.178 kcal mol-l, and to AGP(1iq --+ aq) is 0.887 kcal mol-l. As shown in the introduction, the difference between these two quantities is due to solute-solute interactions in the pure liquid, and in order to concentrate only on solute-solvent interactions it is the term AGp(g -+ aq) that must be considered.This is even more evident for enthalpies of solution, because for most of the homologous series listed in table 4 AH?(g -+ aq) is negative but AH?(liq --+ aq) is positive. Of course, the regression equations in AGP(1iq -+ aq) or in AHP(liq -+ aq)176 THERMODYNAMICS OF SOLUTION IN WATER can be used to correlate and to predict values just as well as the corresponding equations for the process g -+ aq, the standard deviations between calculated and observed values being about the same for the liq -+ aq regression equations and the g -+ aq equations, see table 4.f It follows from the very good linear regression of AGP and AH? with N , that if a liq -+ aq regression equation is linear in N then so will be the respective g -, aq regression equation and vice versa.Which process will lead to the best fit, in terms of the standard deviation, depends to a large extent on the nature of the experimental observations. Thus if AGp(g -+ aq) is determined directly through Henry’s law constants, the fit of the AGp(g -+ aq) regression equation will be better than the indirectly determined AGp(liq -+ aq) regression, as is the case for the methyl alkanoates and n-alkyl acetates (see table 4). If, on the other hand, it is the AGP(1iq --* aq) values that have been directly determined through solubilities of the liquid solutes, then the fit of the AGp(liq -+ aq) regression equation will be better than that of the indirectly determined AGp(g -+ aq) regression, see for example the ethyl alkanoate regressions in table 4.There are available enough results on extended homologous series to test the possibility of alternation effects on the methylene contributions, as outlined for alkan-1-ols, see table 19. In any given series, methylene contributions will fluctuate around a mean value merely through experimental error, and it is to be expected that such fluctuations will lead to alternation effects purely on a statistical basis. Thus for a series of say, six consecutive methylene contributions, three of which are higher than the average (H) and three of which are lower than the average (L), the alternation sequences HLHLHL or LHLHLH will occur by chance in about one case out of ten, and ‘partial’ sequences such as LHLHHL in one case out of three. In my view, all the alternation effects shown in the various homologous sequences studied in this work can be regarded as arising through random experimental error.This in no way contradicts the observation of Beezer and of alternation effects in a bifunctional homologous series, for partition between water and octan- 1-01. In order to probe solute-water interactions, it is necessary to deal only with parameters for the g -+ aq process. Wolfenden and Lewis4 first reported on methylene contributions to AGp(g -+ aq) for a series of homologous compounds. Although the contribution did vary somewhat from one series to another, Wolfenden and Lewis regarded this variation as being within experimental error,$ and calculated an overall methylene contribution of 0.150 kcal mol-l. From the more extensive data given in table 4 and summarised in table 28, it seems that the variation in methylene contribution is outside experimental error, ranging from 0.099 f 0.019 (ethyl alkan- oates) or 0.146 k 0.003 (n-alkyl benzenes) up to 0.229 k 0.008 (alkan-2-ones) or 0.247+0.010 (alk-1-ynes), in Gibbs energy.There is less information on the en- thalpic methylene increment, see tables 4 and 29, but there is substantial variation in the contribution to AHp(g -+ aq), viz. from - 0.673 f 0.035 (n-alkanes) to -0.915f0.02 (n-alkyl benzenes). A number of schemes for the estimation of AGp(g -+ aq) or AHp(g -+ aq) have been constructed on the basis of constant group contributions,lP 2 v 74 and variation of methylene contribution from one series to another might seem to negate such schemes.However, unless long-chain homologues t The correlation coefficients for the regression equations in ACP(1iq + aq) are nearly always better than the coefficients for equations in AGp(g -+ aq), but this merely reflects the larger values of the slopes in the AGP(1iq -+ aq) regression equations. For comparison purposes, the standard deviation is a better ‘goodness-of-fit ’ parameter than the correlation constant, at least in the present case. The extreme limits were 0.137 +_ 0.094 kcal mol-I for n-alkanes and 0.179 +_ 0.078 kcal mol-’ for n-alkyl acetates .M. H. ABRAHAM 177 Table 28. Methylene and group contributions to AG? values for homologous series, in kcal mol-1 at 298 K CH, incrementa group contribution* AG?(g -+ aq) AG?(liq -+ aq) group nc AG?(g+aq) series n-alkanes alk- 1 -enes alk- 1 -ynes n-alkylbenzenes n-alkylamines alkan-2-ones 1 -chloroalkanes 1 -bromoalkanes 1 -iodoalkanes alkan- 1-01s methyl alkanoates n-alkyl acetates ethyl alkanoates n-alkyl propanoates 0.178 0.150 0.247 0.146 0.144 0.229 0.163 0.219 0.200 0.163 0.226 0.144 0.099 0.186 0.887 0.876 0.900 0.809 0.799 0.848 0.870 0.920 0.876 0.822 0.855 0.768 0.688 0.680 CH3 C6H5 NH2 CH,=CH CHEC CH,CO c1 Br I HO MeO. CO CH,CO, EtO.CO CH,CH,CO, 7 7 7 6 7 8 6 7 4 9 8 4 9 3 3.03 k0.02 2.40 f. 0.1 1 0.79 k 0.05 0.44 f 0.01 - 3.42 f 0.01 - 2.72 f. 0.05 0.52 f 0.07 0.26 _+ 0.05 0.33 f 0.04 - 3.96 f. 0.04 - 2.00f 0.12 - 1.84+0.01 -1.54k0.14 - 1.61 & 0.03 a Taken from the regression equations in table 4. Calculated on the basis of a constant Number of members of the homologous series used methyl contribution of 3.03 kcal mol-I.to obtain the given group contribution. Table 29. Methylene and group contributions to AH? and A@ values for homologous series, in kcal mol-, or cal K-I mo1-I at 298 K CH, incrementa group contributionb series AHg(g -, aq) AH?(liq + aq) group nc AHg(g-+aq) n-alkanes -0.673 0.623 n-alk ylbenzenes -0.915 0.070 n-alk ylamines - 0.725 0.368 alkan-2-ones -0.818 0.210 alkan- 1-01s - 0.85 1 0.297 methyl alkanoates - 0.825 0.165 n-alkanoic acids -0.814 0.262 A@& -+ as) A W i q -+ aq) CH3 C6H5 NH2 CH,CO HO Me0 . CO H0,C group n - 2.36 f 0.07 - 6.32 k 0.02 - 9.64 _+ 0.09 - 7.64 0.10 - 9.74 f 0.06 - 7.49 0.01 - 10.51 f0.03 A*(g -+ 4 ) n-alkanes - 2.83 - 0.88 CH3 6 -18.1k0.2 n-alk ylbenzenes - 3.50 - 2.30 C6H5 3 -22.7k0.1 n-alk ylamines -2.89 - 1.46 NH2 5 -20.9k0.3 alkan-2-ones - 3.56 - 2.21 CH,CO 5 -16.4f0.4 alkan-1 -01s - 3.39 - 1.75 HO 5 -19.4kO.2 methyl alkanoates - 3.40 -2.15 Me0 .CO 3 -18.7k0.2 a Taken from the regression equations in table 4. Calculated on the basis of a constant methyl contribution of - 2.36 kcal mol-l to AHF(g + aq) and of - 18.1 cal K-I to A@(g + aq). Number of members of homologous series used to obtain the given group contribution.178 THERMODYNAMICS OF SOLUTION IN WATER are considered, any error due to the assumption of a constant methylene increment is probably within the claimed accuracy of the schemes put f0rward.f. The great advantage of the postulate of the constancy of group contributions is that it enables quantitative measures of group-water interactions to be calculated. In the present work it is clearly not logical to regard the methylene contributions as constant, but if the methyl contribution to solution of the gaseous n-alkanes is taken as constant, then other group contributions may be obtained.$ Group contribution values obtained in this way are in tables 28 and 29; because the basis for the calculation is not the same as in the schemes previously reported,l! 2, 73 the numerical values of the group contributions are not the same.However, in all the group or atomic contribution schemes, hydrophilic groups such as NH,-, HO-, RC0,- etc. all make negative contributions to AGP(g -+ aq) as expected. What is remarkable about the results in table 29 is that both hydrophobic groups (CH,, C6H5) and hydrophilic groups (CH,CO,NH,,OH) are transferred from the gas phase to aqueous solutions exo- thermally.Gianni et aZ.73 have suggested that there is a clear separation between AH? for hydrophobic and hydrophilic groups, but the results in table 29 do not bear this out: compare for example values for the C6H5 group and the NH, group. For the homologous series (table 4) for which both AGP(g -+ aq) and AHP(g -+ aq) values are available, there are good linear correlations between these two quantities, demonstrating the existence of the compensation effect. As for the corresponding vaporisation parameters, the slopes of the plots of AH? against AGP vary from one series to another, so that no general conclusion can be drawn about the mechanism of the solution process.ANOMALOUS SOLUTION EFFECTS In a number of the homologous series, linearity of AGF or AH? with respect to carbon number, N , is not completely observed. For the alkan- 1 -ols, there is a deviation from linearity at low carbon number both in AGP and in AH? as has been pointed out b e f ~ r e , ~ ~ ? ~ ~ and for the alkanoic acids there is also a similar deviation in AH?; unfortunately there is not enough information to determine AGP values for an extended series of alkanoic acids. The origin of these effects is not clear, but may be due to incorporation of the small, hydrophilic solutes into the general water structure. More important are deviations as N becomes very large. Tanford22 has discussed this effect in connection with the related transfer of alkanoic acids from heptane to water and concludes that AGP (heptane + aq) is linear with N up to a carbon number of at least 22.In the present work, results on AGF values for two series of compounds, the alkan-1-01s and the n-alkanes, indicate that linearity with N is not maintained for values of N above ca. 12. The effect in the alkan- 1-01 series is not so large ; octadecan- 1-01 is 16 times as soluble as expected, from the linear regressions in table 4, but should the effect continue above octadecan- 1-01, the deviation factor will become progressively larger. For the g -+ aq transfer, the methylene increment from tetradecan-1-01 to octadecan-1-01 is -0.1 kcal mol-l, each extra methylene group now increasing the alkanol solubility.If the enhanced solubility is due to coiling of the long alkyl chains, with methylene-methylene interactions replacing methylene-water interactions, the methylene increment when coiling takes place might be expected to be intermediate between the methylene-water contribution in the lower alkan-1-01s (+ 0.16 kcal t In the scheme of Hine and Mookerjee', the standard deviation between calculated and observed AGp(g + as) is 0.16 kcal mol-I, in the scheme of Cabani et al.,* it is 0.17 kcal mol-I for AGp(g + aq) and 0.40 kcal mol-I for AHF(g + aq), and in the differentiated atom scheme of Gianni et uI.,'~ it is 0.55 and 0.76 kcal mol-I, respectively. $ Note that all gas-to-water standard-state effects are included in the methyl contribution.2+ l3M.H. ABRAHAM 179 mol-l) and the methylene-methylene contribution for solution of n-alkyl chains in n-alkanes. The methylene increment to AGP for solution of gaseous n-alkane in solvent n-hexadecane is -0.74 kcal mol-l* l3 and so a net value of -0.1 kcal mol-l for the higher alkan- 1-01s is not unreasonable. The deviations for the solution of the n-alkanes in water above about n-dodecane are very much larger in magnitude, the C,, compound, n-octadecane, being 4.9 x lo3 times as soluble as expected (cf. the factor of only 16 for octadecan-1-01). For the n-alkanes between n-hexadecane and n-hexatriacontane, there is a good correlation of AGP(g -+ aq) with N , the slope being - 0.92 kcal mol-l. This methylene contribution is even more negative than that for solution of n-alkanes in n-hexadecane, so that it is easier to introduce a methylene group into a hydrocarbon-like volume surrounded by water (i.e.the coiled n-alkane) than it is into pure hydrocarbon.7 It is possible to compare the solution of n-alkanes in water with solution in non-aqueous solvents. Pierotti et al.42 have recorded activity coefficients for n-alkanes up to eicosane (C,,) in ethanol and up to triacontane (C3,,) in phenol, from which AGF (g + solvent) values may be obtained using the AGF values in table 2; for the g+ethanol transfer, additional AGp(g + solvent) values are a~ai1able.l~ With both solvents the AGp(g -+ Table 30. Comparison of the mol fraction solubility of n-alkanes in water and in ethanol and phenol, at 298 K AGp(g + solvent) solubility factora alkane waterb ethanolc phenold ethanol/water phenol/water methane ethane propane butane pentane hexane heptane octane nonane decane undecane dodecane tetradecane hexadecane octadecane eicosane, C,, hexacosane, c 2 6 triacontane, C3, hexatriacontane, '36 - 6.28 3.95 6.1 1 2.98 6.23 2.33 6.35 1.65 6.6 1 1.15 6.82 0.52 6.90 -0.13 7.16 - 0.66 7.42 - 1.25" 7.44 - 1.77 7.66f - 2.44e 7.72 - 3.03" 6.64 - 4.22e 5.88 - 5.39 3.88 - 6.60" 2.19 - 7.80 - 3.25 - 1 1.35e - 13.73" 12.55 - 17.30" - 3.26" 2.63e 2.00" 1.37 0.66 0.07 - 0.52" - 1.15" - 1.71 - 2.40 - 3.03 - 4.29e - 5.47 - 6.80" - 8.01 - 1 1.83" - 14.42 - 18.12" 5.1 x lo1 2.0 x 102 7.2 x lo2 2.8 x 103 1.0 x 104 4.1 x 104 1.4 x 105 5.4 x 105 2.5 x 107 7.6 x 107 9.1 x 107 4.8 x 107 2.1 x 107 3.0 x 103 2.3 x lo6 5.6 x lo6 1.8 x lo8 8.6 x lo5 - 1.2 x 102 4.4 x 102 1.5 x 103 6.9 x 103 3.3 x 104 1.0 x 105 4.3 x 105 2.4x 107 7.6 x 107 6.7 x 107 3.0 x 107 1.2 x 104 1.9 x lo6 5.1 x lo6 1.0 x 108 2.1 x 108 1.9 x lo6 - a Mole fraction solubility of gas (or liquid) in ethanol or phenol/mole fraction solubility of gas (or liquid in water).From table 2. Ref. (1 3) and (42). Ref. (42). Estimated from linear plots of AGF against carbon number. f This is the value calculated from the regression in table 4. The observed value, 8.25 kcal mol-I, (table 2) seems much to high. t An alternative possibility to the coiling of long alkyl chains is the formation of micelles in aqueous solution. However, the solubilities of the solid n-alkanes at 298 K seem far too small for micellar formation, for example 6 x mol dmP3 (eicosane) or 4 x loP9 mol dmP3 (hexacosane). 7 FAR 1180 THERMODYNAMICS OF SOLUTION IN WATER solvent) values are accurately linear with N and hence other values for the n-alkanes may be estimated with reasonable certainty up to triacontane, and probably well beyond.The various g + solvent transfer values are collected in table 30 and are then combined to yield solubility factors that represent the enhanced solubility of alkanes in ethanol or phenol over the solubility in water.7 For both solvent/water systems, the solubility factors steadily increase from methane onwards, level off at about n-hexadecane and then steadily decrease due to the more negative AGp(g + aq) values. Judging from results on the C , to C , n-alkanes, both sets of solubility factors lead to the conclusion that n-hexacosane and n-hexatriacontane are more soluble in water than expected by factors of no less than 1.3 x 1Olo and 2.0 x 1Ol8, respectively.Furthermore, if the trend in solubility from the C,, compound onward is continued beyond n-hexatriacontane, it may be deduced that the C,, n-alkane will be as soluble in water as in the non-aqueous solvents ethanol and phenol. Certainly these predictions depend on results of the solubility of n-alkanes in water from but few investigations, and further experimental work would be most desirable, not only on the long-chain n-alkanes but also on other homologous series. I am grateful to Dr A. R. Beezer for communicating results prior to publication, and to a referee for valuable suggestions which have greatly benefitted the paper.J. Hine and P. K. Mookerjee, J. Org. Chem., 1975,40, 292. S . Cabani, P. Gianni, V. Mollica and L. Lepori, J. Solution Chem., 1981, 10, 563. R. D. Wauchope and R. Haque, Can. J. Chem., 1972, 50, 133. R. Wolfenden and C. A. Lewis, J. Theor. Biol., 1976, 59, 231. E. Wilhelm and R. Battino, Chem. Rev., 1973, 73, 1 . E. Wilhelm, R. Battino and R. J. Wilcock, Chem. Rev., 1977, 77, 219. G. L. Amidon, S. H. Yalkowsky and S. Leung, J. Pharm. Sci., 1974,63, 1858. G . L. Amidon, R. S. Pearlman and S. T. Anik, J. Theor. Biol., 1979, 77, 161. ’ T. R. Rettich, Y. P. Handa, R. Battino and E. Wilhelm, J. Phys. Chem., 1981, 85, 3230. lo G. L. Amidon and S. T. Anik, J. Phys. Chem., 1980,84,970. l1 G. L. Amidon and S. T. Anik, J. Chem.Eng. Data, 1981, 26, 28. l2 M. H. Abraham, J. Am. Chem. Soc., 1979, 101, 5477. l3 M. H. Abraham, J. Am. Chem. Soc., 1982, 104, 2085. l5 J. A. V. Butler, C. N. Ramchandani and D. W. Thomson, J. Chem. Soc., 1935,280. l6 J. A. V. Butler and W. S. Reid, J. Chem. Soc., 1936, 1 1 71. J. A. V. Butler, D. W. Thompson and W. H. Maclennan, J. Chem. Soc., 1933,674. J. A. V. Butler, Trans. Faraday Soc., 1937, 33, 229. H. D. Nelson and C. L. de Ligny, Reel. Trav. Chim. Pays Bas, 1968,87, 528; 623. l9 J. H. Rytting, L. P. Huston and T. Higuchi, J. Pharm. Sci., 1978, 67, 615. 2o D. Mackay and W. Y. Shiu, J. Phys. Chem. Re$ Data, 1981, 10, 1175. 21 C. Tanford, The Hydrophobic Eflect (Wiley-Interscience, New York, 1973). 22 C. Tan ford, The Hydrophobic Eflect : Formation of Micelles and Biological Membranes (Wiley- 23 S.J. Gill and I. Wadso, Proc. Natl Acad. Sci. USA, 1976, 73, 2955. 24 K. Hallinga, J. R. Grigera and H. J. C. Berendson, J. Phys. Chem., 1980,84, 2381. 25 A. E. Beezer and W. H. Hunter, J. Med. Chem., in press: A. E. Beezer, W. H. Hunter and D. E. Storey, 26 J. Vejrosta, J. Novak and J. A. Jonson, Fluid Phase Equilibria, 1982, 8, 25; 1982, 9, 279. 27 R. R. Dreisbach, Physical Properties of Chemical Compounh [American Chemical Society, Washing- 28 C. Sutton and J. A. Calder, Environ. Sci. Technol., 1974, 8, 654. 29 M. Ducros, J. F. Grunson and H. Sannier, Thermochim. Acta, 1980, 36, 39. 30 Y. B. Tewari, M. M. Miller, S. P. Wasik and D. E. Martire, J. Chem. Eng. Data, 1982, 27, 451. Interscience, New York, 1980). J. Pharm. Pharmacol., in press. ton D.C., 1955 (vol. l), 1959 (vol. 2), 1961 (vol. 3)]. t Because AGP cancels between water and the non-aqueous solvent, these solubility factors refer to both the gaseous and the liquid n-alkane. Furthermore, an error in extrapolated AGP values also cancels out.M. H. ABRAHAM 181 31 A. Ben-Naim and J. Wilf, J. Phys. Chem., 1980,84, 583. 32 I. Sanemasa, M. Araki, T. Deguchi and H. Nagai, Bull. Chem. SOC. Jpn, 1982,55, 1054. 33 S. J. Gill, N. F. Nichols and I. Wadso, J. Chem. Thermodyn., 1976, 8, 445. 34 F. M. Jones and E. M. Arnett, Prog. Phys. Org. Chem., 1974, 11, 263. 3b M. Ducros, J. F. Grunson and H. Sannier, Thermochim. Acta, 1981,44, 131. 36 P. A. Cullimore and P. J. Guthrie, Can. J. Chem., 1979, 57, 240. 37 R. G. Buttery, L. C. Ling and D. G. Guadagni, J. Agric. Food Chem., 1969, 17, 385. 38 D. Ambrose, J. H. Ellender, E. B. Lees, C. H. S. Sprake and R. Townsend, J. Chem. Thermodyn., 39 C. Hansch, J. E. Quinlan and G. L. Lawrence, J. Org. Chem., 1968,33, 347. 40 G. Della Gatta, L. Stradella and P. Venturello, J. Solution Chem., 1981, 10, 209. 41 K. Kinoshita, H. Ishikawa and K. Shinoda, Bull. Chem. SOC. Jpn, 1958, 31, 1081. 42 G. J. Pierotti, C. H. Deal and E. L. Derr, Ind. Eng. Chem., 1959, 51, 95. 43 M. Davies and B. Kybett, Trans. Faraday SOC., 1965, 61, 1608. 44 D. Ambrose and C. H. S. Sprake, J. Chem. Thermodyn., 1970,2,631; D. Ambrose, J. H. Ellender and 46 T. Boublik, V. Fried and E. Hala, The Vapour Pressures of Pure Substances (Elsevier, Amsterdam, 46 S. H. Yalkowsky and S. C. Valvani, J. Pharm. Sci., 1980, 69, 912. 47 L. Benjamin, J. Phys. Chem., 1964, 68, 3575. 48 R. Aveyard and R. W. Mitchell, Trans. Faraday SOC., 1968, 64, 1757. 49 D. M. Alexander and D. J. T. Hill, Aust. J. Chem., 1969, 22, 347. 50 E. M. Arnett, W. B. Kover and J. V. Carter, J. Am. Chem. SOC., 1969,91,4028. 51 C. V. Krishan and H. L. Friedman, J. Phys. Chem., 1969,73, 1572. 52 A. C. Rouw and G. Somsen, J. Chem. Thermodyn., 1981, 13, 67. 53 D. J. T. Hill and L. R. White, Aust. J. Chem., 1974, 27, 1905. 54 A. Rose and W. R. Supina, J. Chem. Eng. Data, 1962,2, 174; A. Rose and V. N. Schrodt, J. Chem. 55 D. Ambrose, J. H. Ellendar, H. A. Gundrey, D. A. Lee and R. Townsend, J. Chem. Thermodyn., 56 T. G. Kieckbusch and C. J. King, J. Chromatogr. Sci., 1979, 17, 273. 57 A. Skrzeczand A. Maczynski, Pol. J. Chem., 1979,53,715; A. Skrzecz, Pol. J. Chem., 1980,54,1101. 58 T. E. Jordan, Vapour Pressures of Organic Compounds (Interscience, New York, 1954). 59 D. Richon and A. Viallard, Can. J. Chem., 1976,54, 2584. 6o R. F. Cross and P. T. McTigue, Aust. J. Chem., 1977,30, 2597. J. H. Stern and A. Herman, J. Phys. Chem., 1967,71, 306. 62 B. G. Cox, J. Chem. Sci., Perkin Trans. 2, 1973, 607. 63 E. M. Arnett, W. G. Bentrude, J. J. Burke and P. McC. Duggleby, J. Am. Chem. SOC., 1965,87,1541. 64 J. A. V. Butler and C. N. Ramchandani, J. Chem. SOC., 1935,952. 65 J . Konicek and I. Wadso, Acta Chem. Scand., 1971, 25, 1541. 1975, 7, 453. C. H. S. Sprake, J. Chem. Thermodyn., 1974,6, 909. 1973). Eng. Data, 1963, 8, 9. 1981, 13, 795. C. H. D. Calis-van Ginkel, C. H. M. Calis, C. W. M. Timmermans, C. G. de Kruif and H. A. J. Oonk, J. Chem. Thermodyn., 1978, 10, 1083. 67 C. G. de Kruif and H. A. J. Oonk, J. Chem. Thermodyn., 1979, 11, 287. 68 C. G. de Kruif, R. C. F. Schaake, J. C. van Miltenburg, K. van der Klauw and J. C. Blok, J. Chem. 69 K. Kusano, J. Surrkuusk and I. Wadso, J. Chem. Thermodyn., 1973,5, 757. 70 J. M. Corkhill, J. F. Goodman and J. R. Tate, Trans. Faraday SOC., 1967, 63, 773. 71 R. R. Krug, Ind. Eng. Chem., Fundam., 1980, 19, 50. 7 2 E. Tomlinson, Int. J. Pharm., 1983, 13, 115. 73 P. Gianni, V. Mollica and L. Lepori, Z. Phys. Chem. ( Wiesbaden), 1982, 131, 1. Thermodyn., 1982, 14, 791. (PAPER 3/736) 7-2

ISSN:0300-9599    

DOI:10.1039/F19848000153

出版商:RSC    

年代:1984

数据来源: RSC

摘要:

J . Chem. SOC., Faraday Trans. 1, 1984,80, 183-189 Photochemical Reduction of Oxygen Catalysed by Colloidal Cadmium Selenide BY JAMES R. DARWENT Department of Chemistry, Birkbeck College, University of London, Malet Street, London WClE 7HX Received 12th May, 1983 Colloidal CdSe sensitises the photoreduction of 0, and methyl viologen (MV2+) by cysteine. In the absence of cysteine the semiconductor is rapidly photo-oxidised. Rates of 0, photore- duction and yields of MV'+ tend towards a maximum value for MV2+ concentrations of 1 O-, mol dm-3. Flash-photolysis and steady-state measurements of 0, reduction are interpreted by a mechanism which shows that the low quantum yields are due to rapid recombination of photogenerated charge carriers. Only 2 of the photogenerated electrons are available to reduce 0,.Photosensitized electron-transfer reactions at the surface of semiconductor particles may provide a route for the conversion of solar These reactions are also important in photography and the photo-oxidation of pigments. Recent research has shown that dispersions of semiconductor particles can photodissociate water into H, and 0,.3-5 Most of this work has focussed on oxide semiconductors, which have large band gaps and absorb very little visible radiation, but a number of research groups have also shown that CdS can sensitise the photoreduction of water with visible light. Krasnovskii et aL6 were the first to demonstrate this reaction using CdS, methyl viologen (MV2+) and the enzyme hydrogenase. More recent work has shown that the enzyme can be replaced by metal catalysts (Pt, Rh or RuO,) deposited on the CdS particles, in which case the viologen relay is no longer Although CdS absorbs light up to 550 nm this represents only a small fraction of the solar spectrum, and ideally the absorption threshold should be in the region of 900 nm.12 Platinised semiconductor particles with band gaps in this region (Si, CdSe and copper phthalocyanine)s can sensitize the reduction of water, but the quantum efficiency is invariably orders of magnitude less than that obtained with CdS.The photosensitised reduction of 0, by colloidal CdSe has been investigated to provide a better understanding of the poor efficiencies of these low-band-gap semiconductors, and our results are presented below. EXPERIMENTAL Steady-state experiments were performed with an Applied Photophysics UV90 photoirrad- iation system using a 900 W Xe lamp and a grating monochromator.Light intensities were measured with a calibrated thermopile and absorption spectra were recorded with a Perkin-Elmer 200 spectrophotometer. Conventional flash-photolysis measurements were made with an Applied Photophysics K200 system using a nitrate filter to remove wavelengths < 400 nm as described previ0us1y.l~ Solutions for the flash-photolysis experiments were purged for 1 h with N,. Oxygen concentrations were measured with a Clark membrane oxygen electrode. A detailed 183184 PHOTOREDUCTION OF 0, USING COLLOIDAL CdSe CATALYST 0 I I 1 0 5 10 15 20 tlmin Fig. 1. Photosensitised reduction of 0, (MV2+, lo-, mol dm-,; cysteine, lo-, mol dm-3; CdSe, 2 x lo-, mol dm-3; NaLS, mol dm-3; phosphate buffer, lo-, mol dm-3; pH 6.8; A,, = 475 nm; IA = 3 x loA7 ein s-l).description of the experimental arrangement, sensitivity and calibration of this instrument is given in ref. (14). The solution (37 cm3) was stirred and thermostatted at 35 OC. Colloidal CdSe was prepared by passing H,Se gas into a solution of Cd(NO,), (5 x mol dm-,) in phosphate buffer (pH 7.0, lo-, mol drn-,) with lo-, mol dm-, sodium lauryl sulphate (NaLS) as a supporting surfactant. H,Se was generated by adding solid CdSe to dilute hydrochloric acid. Both the acid and the cadmium solution were continuously purged with N, to avoid reaction of 0, with H,Se. The solution was then purged overnight to remove excess H,Se.The average radius of the particles was determined by transmission electron microscopy. Cd(NO,), (AnalaR) and NaLS (specially purified for biochemical work) were purchased from B.D.H. Solid CdSe (99.999%) was from Koch-Light. Water was doubly distilled. RESULTS In the last two years a number of research groups have investigated electron-transfer reactions in colloids of semiconductor particles of either Ti0215-17 or CdS.,? 159 These colloids are transparent and stable for a period of weeks and so provide a suitable medium in which to study photoredox reactions at the semiconductor/liquid interface. No previous work has been reported on the photochemical reactions of colloidal CdSe, which has a band gap of 1.7 eV and represents an attractive semiconductor for the conversion of solar energ~.~~q 26 Colloidal CdSe was prepared by bubbling H,Se into an oxygen-free solution ofcadmium nitrate ( 5 x lod3 mol dm-3) and sodium lauryl sulphate (NaLS) ( lo-, mol dm-3).This resulted in a relatively clear brown colloid containing particles with an average radius of 100 nm. Optical-absorp- tion studies showed that these colloids were stable for several weeks. Photosensitised reduction of 0, provides a simple model system with which to study electron-transfer reactions of semiconductor particle^.^^-^^ The reaction can be3 d I v) 3 2 E E m . n 0 7 S ' 0 0 J. R. DARWENT 5 10 [MV2+]/10-3 mol dm-3 185 5.0 T E a - ii! I- = 5 . - t 2.5 > 0 Fig. 2. Yield of MV'+ (a) and rate of 0, photoreduction (0). (Unless otherwise stated conditions as in fig.1.) monitored continuously with a membrane polarographic detector, and unlike the photoreduction of water no additional catalysts are needed. When colloidal CdSe is illuminated in the presence of 0, the semiconductor is photo-oxidised and a red precipitate of Se forms in a period of minutes. This reaction can be prevented by the presence of a sacrificial electron donor, such as cysteine, which is preferentially oxidised. This approach has been particularly successful with conventional semiconductor e l e c t r ~ d e s . ~ ~ ~ 26, 30 Colloidal CdSe will then sensitise the photoreduction of 0, by cysteine with no appreciable loss of CdSe. The rate of this reaction is dramatically increased by low concentration of methyl viologen. Fig. 1 shows a typical example of the concentration-time profile for the photoreduction of 0,.The rate is essentially independent of 0, concentration for the first two half-lives. In the dark, the reduction of 0, by cysteine is insignificantly slow at this pH and temperature. The effect of MV2+ concentration on the rate of 0, reduction is presented in fig. 2. This shows that the rate tends to a maximum value for [MV2+] > lo-, mol dm-3, at which point the rate is 7.5 times faster than in the absence of MV2+. Microsecond flash photolysis was used to provide a better understanding of the mechanism of this system. When CdSe colloids were illuminated with a short flash of visible light (3, > 400 nm) in the presence of cysteine ( lo-, mol dmP3) and MV2+ ( 10-4-10-2 mol dmP3) the absorption spectrum of reduced viologen (MV'+) (Amax 395 and 605 nm) was observed.In the absence of 0,, MV'+ was formed as a permanent product. All the MV'+ was produced during the 10 ps photoflash. The yield of MV'+ was proportional to the flash intensity and showed a similar dependence on [MV2+] to that found for the rate of 0, reduction (fig. 2).186 PHOTOREDUCTION OF 0, USING COLLOIDAL CdSe CATALYST DISCUSSION The results described above can be described by the mechanism in scheme 1. hv + CdSe Cd Se( h +e -) CdSe(h+e-) + cysSH CdSe(e-) + MV2+ MV'++O, CdSe(e-) (cysSH (light absorption) (1) -+ CdSe(h+e-) -+ CdSe (charge recombination) (2) -+ CdSe(e-) + $(cysS), + H+ (oxidation of cysteine) (3) (4) -+ CdSe + MV'+ -+ MV2++0i- (oxygen reduction) ( 5 ) + CdSe (electron loss) (6) (electron trapping) = cysteine). Scheme 1.Absorption of light by the CdSe particles leads to the formation of an electron-hole pair (e- h+). In the absence of a sacrificial electron donor h+ will recombine with e- or oxidise selenide, leading to the formation of selenium and the destruction of the semiconductor. Cysteine can act as a suitable electron donor and prevent photo- oxidation of the particle^.^^ 8 y 2 5 1 2 6 q 30 A fraction of the photogenerated electrons are then trapped at a high negative potential in the conduction band of CdSe (ECB = - 1.2 V us NHE)31 and are available to reduce 0, to 0,- (E, = - 0.16 V).279 32 This is a slow reaction, presumably because of the low solubility of 0, in water (ca. 2 x 1 0-4 mol dm-3 under these experimental conditions) and alternative reactions, such as the reduction of cystine and Cd2+, will result in electron loss.The rate of 0, reduction is increased by low concentrations of MV2+ ( 10-3-10-2 mol dm-3) which will bind to the anionic surfactant close to the surface of CdSe and accept electrons from the conduction band of the semiconductor. Subsequent electron transfer from MV'+ to 0, is known to be a diffusion-controlled reaction.33 The energetics for these reactions are summarised in fig. 3. Under these experimental conditions, 0;- is known to disproportionate, so that the overall reaction is oxidation of cysteine and reduction of oxygen to H,0,.32 Applying the steady-state approximation to the concentration of e- and MV'+ leads to the following equation for the rate of oxygen reduction: 1 k6 -+ - 1 R - R, aI, [MV' +]k,al,' (7) R and R, (mol s-l) are the rates of 0, reduction in the presence and absence of MV", a is the fraction of photogenerated electrons which are trapped [this will depend on the relative rates of reactions (2) and ( 3 ) ] , I , (ein s-l) is the rate of absorption of photons, k,(s-l) is the rate of electron loss, and k , is the rate of electron transfer to MVS2+.This equation assumes that the rates of reactions (2) and (3) are essentially constant (the concentration of cysteine will only drop by 2% during the reduction of 0,) and that reaction (2) is fast compared with the rate of MV'+ generation. The equation predicts correctly that the rate will be independent of the 0, concentration. A plot of eqn (7) is shown in fig.4, from which cc = (2.0f0.1) x lo-, and k 6 / k , = (1.4k0.2) x mol dmP3. This shows that the low efficiency of colloidal CdSe results from the small value of a, i.e. the high rate of charge recombination compared with the reaction of h+ with cysteine.J . R. DARWENT 187 Fig. 3. Redox diagrams for the photosensitised reduction of 0,. 0 1 2 3 l/[MV2']/10-3 dm3 mol-' Fig. 4. Reciprocal plot of MV'+ yield (0) and rate of 0, photoreduction (0) against 1/[MV2+] (variable concentrations of MV2+, otherwise conditions as in fig. 1).188 PHOTOREDUCTION OF 0, USING COLLOIDAL CdSe CATALYST This kinetic scheme can also describe the flash-photolysis results, from which [MV'+]- [e30+k,[e],[MV2+] -- 1 1 k6 where [el, is the concentration of electrons trapped during the photoflash by reaction (3) and [MV'+] is the total concentration of MV'+ formed for a given concentration of MVe2+.The results of the flash-photolysis experiments are also shown in fig. 4, where there is good agreement between the two sets of data. From this figure k,/k, is again (1.4 If: 0.2) x mol dm-3 and [el,, is (4.8 -f 0.5) x lo-' mol dm-3. CONCLUSION Poor quantum yields are observed for electron-transfer reactions sensitised by colloidal CdSe suppported by NaLS, since charge recombination occurs more rapidly than oxidation of cysteine in this system. This problem may be overcome by using a surfactant support which could facilitate oxidation of cysteine. Alternatively it might be possible to reduce the rate of charge recombination by adsorbing metal ions onto the particles, an approach which has been very successful with liquid-junction photovoltaic cells.25* 2 5 y 34 This work was supported by the S.E.R.C.and London University Central Research Fund. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 A. J. Bard, J. Phys. Chem., 1982, 86, 172. M. Gratzel, Acc. Chem. Res., 1982, 15, 376. A. V. Bulatov and M. L. Khidekel, Zzv. Akad. Nauk SSSR, Ser. Khim. Nauk, 1976, 1902. J. M. Lehn, J. P. Sauvage and R. Ziessel, Nouv. J. Chim., 1980, 4, 623. E. Borgarello, J. Kiwi, E. Pelizzetti, M. Visca and M. Gratzel, J. Am. Chem. SOC., 1981, 103, 6324. A. A. Krasnovskii, G. P. Brin, A. N. Luganskaya and V. V. Nikandrov, Dokl. Akad. Nauk SSSR, 1979, 249, 896. J. R.Darwent and G. Porter, J. Chem. SOC., Chem. Commun., 1981, 145. J. R. Darwent, J. Chem. SOC., Faraday Trans. 2, 1981, 77, 1703. K. Kalyanasundaram, E. Borgarello and M. Gratzel, Helv. Chim. Acta, 1981, 64, 362. E. Borgarello, K. Kalyanasundaram, E. Pelizzetti and M. Gratzel, Helv. Chim. Acta, 1982, 65, 243. J. R. Harbour, R. Wolkow and M. L. Hair, J . Phys. Chem., 1981, 85,4026. G. Porter and M. D. Archer, Interdisc. Sci. Rev., 1976, 1, 119. M. A. West, Creation and Detection of Excited States, ed. W. R. Ware (Marcel Dekker, New York, 1976), vol. 4. A. Mills, A. Harriman and G. Porter, Anal. Chem., 1981, 53, 1254. D. Duonghong, J. Ramsden and M. Gratzel, J . Am. Chem. SOC., 1982,104,2977. A. Henglein, Ber. Bunsenges. Phys. Chem., 1982, 86, 241. M. A. Fox, B. Lindig and C. Chen, J. Am. Chem. SOC., 1982, 104, 5828. J. Kuczynski and J. K. Thomas, Chem. Phys. Lett., 1982,88,445. A. Henglein, Ber. Bunsenges. Phys. Chem., 1982, 86, 301. A. Henglein, J . Phys. Chem., 1982, 86, 2291. Z. Alfassi, D. Bahnemann and A. Henglein, J. Phys. Chem., 1982, 86, 4656. R. Rossetti and L. Brus, J . Phys. Chem., 1982, 86, 4470. K. Kalyanasundaram, E. Borgarello, D. Duonghong and M. Gratzel, Agnew. Chem., Int. Ed. Engl., 1981, 20, 987. K. Metcalfe and R. E. Hester, J. Chem. SOC., Chem. Commun., 1983, 133. A. Heller, Acc. Chem. Res., 1981, 14, 154. M. S. Wrighton, Acc. Chem. Res., 1979, 12, 303. J. R. Harbour and M. L. Hair, J . Phys. Chem., 1977,81, 1791.J. R. DARWENT 189 2R J. D. Spikes, Photochem. Photobiol., 1981, 34, 549. 29 R. E. Stephens, B. Ke and D. Tnvich, J . Phys. Chem., 1955, 59, 966. 3o A. Kirsch-de Mesmaeker, A. M. Decoster and J. Nasielski, Solar Energy Materials, 1981, 4, 203. 31 A. B. Ellis, S. W. Kaiser, J. M. Bolts and M. S. Wrighton, J . Am. Chem. SOC., 1977, 99, 2839. 32 D. T. Sawyer and J. S. Valentine, Acc. Chem. Res., 1981, 14, 393. 33 J. R. Darwent and K. Kalyanasundaram, J . Chem. SOC., Faraday Trans. 2, 1981,77, 373. 34 A. Heller, K-C. Chang and B. Miller, J. Am. Chem. SOC., 1978, 100, 684. , (PAPER 3/761)

ISSN:0300-9599    

DOI:10.1039/F19848000183

出版商:RSC    

年代:1984

数据来源: RSC

摘要:

J . Chem. Soc., Faraday Trans. 1, 1984, 80, 191-209 Mechanisms of 1,5-Dehydrocyclisation and Isomerisation of Alkanes on Iridium, Rhodium, Palladium and Platinum Films BY ODILLA E. FINLAYSON AND JOHN K. A. CLARKE Department of Chemistry, University College, Belfield, Dublin 4, Ireland AND JOHN J. ROONEY* Department of Chemistry, The Queen’s University of Belfast, Stranmillis Road, Belfast BT9 SAG, Northern Ireland Received 19th May, 1983 Deuterium tracer studies with a combined gas-liquid chromatography and mass spectrometry analysis of products and using 2,2,4,4-tetramethylpentane (TMP) as model reactant establish clearly that the selective cyclisation (SCM) of five-carbon chain alkanes on Ir, Rh, Pd and Pt is by the heterogeneous counterpart of reductive elimination of alkyls in organometallic chemistry.A comparison of cyclisation of TMP, 3,3-dimethylpentane and n-pentane on films of the same series of metals and an extended study of cyclisation of 3,3-dimethylpentane and of n-pentane on a series of Pt-Cu alloy films with variation of hydrogen partial pressure show that more than one mechanism of non-selective cyclisation (NSCM) takes place. The isotopic studies show clearly that a monoadsorbed intermediate is sufficient for bond-shift rearrangement of TMP to 2,2,5-trimethylhexane on Ir, Rh, Pd or Pt. An important major conclusion to be derived from the present work is that one surface metal atom comprises the active centre of the catalytic site for a variety of reactions in addition to simple hydrogenationdehydrogenation. The mechanisms of 1,5-dehydrocyclisation of alkanes and the reverse hydro- genolysis reaction on platinum surfaces have been debated actively in the recent literature. Two modes of reaction were recognised by Gault,’ namely a non-selective cyclic mechanism (NSCM) and a selective cyclic mechanism (SCM), and these were illustrated by the catalytic interconversion of methylcyclopentane, 2-methylpentane, 3-methylpentane and n-hexane.The NSCM allows interconversion between all four hydrocarbons whereas SCM does not allow methylcyclopentane and n-hexane to interconvert. It has been recognised that the nature of the catalyst and the experimental conditions employed determine the selectivity for NSCM or SCM. Further, while Pt has been found to be active for 1,5-dehydrocyclisation b either NSCM or SCM,l hydrogen only interconvert methylcyclopentane and 2- and 3-methylpentanes, i.e.exclusively SCM;2.3 Pd catalysts were suggested to cyclise by NSCM only.* Because the SCM involves two terminal C atoms Gault proposed a dimetallocarbyne intermediate, i.e. triple bonding of the two terminal carbons of the hydrocarbon chain to two contiguous metal site atoms. For the NSCM Gault proposed a metallo- 1 ,5-dicarbene specie^.^ Serious criticism of Gault’s arguments may be offered even on the basis of presently available experimental results. First, we note that carbenes are intermediates in the iridium catalysts of various particle sizes down to 15 K diameter and in excess 191192 CYCLISATION AND ISOMERISATfON OF ALKANES formation of carbynes.Further, the SCM is more important at relatively lower temperatures and higher hydrogen pressures whereas the NSCM is dominant at relatively higher temperatures and lower hydrogen pressures;lq therefore the NSCM must be more dehydrogenated than the SCM intermediate. Other intermediates suggested include the (highly dehydrogenated) pentadienyl radical for NSCM, alkyl/alkene insertion on Pd7b (implicitly NSCM)8 and an alkyl/carbene insertion reaction for Pt cyclisations8 for which Clarke and Rooney suggestedg reductive elimination of an ae-diadsorbed species by analogy with known organometallic reactionslO (see scheme 1). If an ae-type species were the intermediate for SCM, then cyclisation would be confined to the terminal C atoms for steric reasons only.Scheme 1. The present work is in two parts, first a mechanistic evaluation of the SCM on Ir, Rh, Pd and Pt films using deuterium exchange and combined gas-liquid chromato- graphy and mass spectrometry (g.l.c./m.s.), secondly a mechanistic investigation of NSCM with several reactant alkanes on the same series of metals and partly a comparative test with a series of Pt-Cu alloys. The sequence undertaken in the first part was as follows: (i) to prepare surfaces of Ir, Rh, Pd and Pt having a slow rate of alkane/deuterium exchange at the temperature required for C-C skeletal rearrangements and (ii) to analyse by g.l.c./m.s. the reaction product on these surfaces of a suitable model compound in excess deuterium gas : detailed mechanistic information was sought from such results mainly for the cyclisation reaction but also for isomerisation.The model compound chosen was 2,2,4,4-tetramethylpentane (TMP, 1) for the following reasons : (i) 1,5-dehydrocyclisation is confined to the terminal groups and (ii) the exchange pattern is expected to be simple owing to the presence of quaternary groups, and the exchange of the cyclic compound 1,1,3,3-tetramethylcyclopentane (TMCP, 2) is known to be essentially simple at low temperatures on Rh and Pt films." 2 is the cyclisation product from 1. The expected products of isomerisation are 2,2,4- trimethylhexane (2,2,4-TMH, 3) (by 1,2-methyl shift) and 2,2,5-trimethylhexane (2,2,5-TMH, 4) (by 1,2-neopentyl shift). The isomer 2,2,3,4-tetramethylpentane is much less probable because of the unfavourable 2,3-methyl shift required in its formation.Possible intermediates involved in the cyclisation are shown in table 1, which also gives the minimum number of deuterium atoms that would be incorporated into the product if cyclisation took place in excess deuterium rather than hydrogen. As will become apparent, suitable catalyst surfaces could be formed as described, and a decision on SCM proved to be possible for all four metals. As will be argued from the trends in reactivity of n-pentane, 3,3-dimethylpentane (DMP) and 1 on Ir, Rh, Pd and Pt in parallel with the isotopic work, there are indications that more than one NSCM is possible. Work by de Jongste and PeneP showed that ring opening of methylcyclopentane by SCM occurred on Pt-Cu alloys of 57 at.% Pt and this changes to NSCM between 25 and 5 at.% Pt. Work with C13-labelled pentanes allowed the same authors with Gault to confirm that above 490.E. FINLAYSON, J. K. A. CLARKE AND J. J. ROONEY I93 Table 1. Possible mechanisms of 1,5-dehydrocyclisation by SCM and minimum deuterium incorporation by each mechanism intermediate 2Hmi, ref. a& elimination 0 9 0 aae insertion aam aaaE aaam 1 8 2 Gault NSCM5 2 - - n 3 I l l It M M 5 at. % Pt SCM takes place and gradually this changes to totally NSCM at 21 at. % Pt.13 Other workG shows clearly that NSCM is favoured over SCM on Pt at low hydrogen partial pressure (see above). In view of the finding that Pt-Au films caused cyclisation of n-pentane but not of 3,3-dimethylpentane14 it was thought promising to compare the cyclisation of n-pentane with that of 3,3-dimethylpentane on a series of Pt-Cu alloy films, first with a standard reaction mixture and then with a mixture lean in hydrogen.In this way a clarification might be found of the possible versions of the NSCM. EXPERIMENTAL MATERIALS A standard static reactor (ca. 800cm3) with provision for preparing single or binary evaporated films was connected to an in situ g.1.c. unit and via a glass capillary to an AEI MS 1Oc2 mass spectrometer. Iridium, palladium, platinum and copper wires used to prepare films were Johnson-Matthey ' spectrographically standardised '. Rhodium wire was Materials Research Corporation, Marz grade. 1 was an API standard sample supplied for an earlier study by the API Project at the Carnegie-Mellon University, Pittsburgh.3,3-Dimethylpentane was K and K and the purity was verified by g.1.c. n-Pentane was Fluka puriss. 2 was prepared by Wolff-Kishner reduction of 1,1,3,3-tetramethylcyclopentan-2-one. The sample contained ca. 10% impurity and was used mainly for calibration purposes. All hydrocarbons were subjected to repeated freeze-thaw cycles in vacuum before use. Commercial hydrogen or 99.99% deuterium as appropriate was palladium-diffused. ISOTOPIC STUDIES OF THE SELECTIVE CYCLISATION MECHANISM (SCM) A number of variations of film deposition and subsequent film treatment were used in efforts to slow the rate of the exchange reaction of the hydrocarbon while still allowing skeletal rearrangements to occur. On the one hand the degree of sintering was varied.Treatment of194 CYCLISATION AND ISOMERISATION OF ALKANES the deposited film with air or oxygen, with and without heating, and heating of the film in oxygen mixed with a large excess of H, (or D,) were other variations. The final procedure adopted involved deposition of the metal film at 773 K followed by heat treatment in 3-4 Torrt H, (or D,) at 773 K for a minimum of 2 h. The temperature was then reduced to the reaction temperature prior to admission of the hydrocarbon + H, (or D,) mixture. The reaction mixture was 10: 1 : : H, (or D,): hydrocarbon, with 0.8 Torr hydrocarbon corresponding to 1.30 x 1019 molecules of reactant in the reaction vessel. The cyclisation reaction of 1 was first run in H, to establish the reaction temperature and was then rerun in D,.Additional experiments were on (i) the exchange reaction of 2 on Pd films and (ii) the reaction of 3,3-dimethylpentane on Ir, Rh, Pd and Pt films prepared as before. A 15% squalane column operating at 378 K was used in the in situ g.1.c. to separate 1 and its products. These column conditions were inadequate to separate 2 and the isomer 4. These products could, however, be separated in the g.l.c./m.s. unit. In the mass spectrometer at 18eV 1 gave very small amounts of the parent ion but a large pseudo-parent ion at m/e 113 formed by loss of a methyl group, namely (CH,),C-CH,-C+(CH,),. Peaks at m/e 112 and 11 1 due to the loss of one or two hydrogen atoms were 2.5 and 1.1 %, respectively, of the m/e 113 peak. The reaction of 1 and deuterium was monitored by g.1.c. and by m.s.either until a peak due to the combined cyclic product and isomer was seen in the chromatogram or until there was an increase in the ratio of m/e 11 1 to 112 in the mass spectrum (in practice 5-20 min at the selected temperature). The presence of ,H,-,H, cycloalkanes should be manifest as an obvious outgrowth of the exchange contour, as the cycloalkane had a ten-fold greater sensitivity in the MSlOc2 than did the alkane. In order to obtain a detaild analysis the reaction mixture was then quenched and the hydrocarbon removed for g.l.c./m.s. analysis. For this, the sample was dissolved in ca. 1 cm3 diethylether and injected onto a capillary OVlOl fused silica column temperature-programmed from 293 K at 4 K min-l. The sample was then bled into a mass spectrometer operating at 20 eV.Partial separation of the exchanged hydrocarbons occurs on the g.1.c. column, with the heavier deuteroisomers being eluted first. Therefore several m.s. scans through the g.1.c. peaks were necessary to give the genuine starting point of the deuterium distribution of each compound. The mass spectrometer scanned the selected mass range every In studies of the reaction of 3,3-dimethylpentane, reactant and products were separated on a column of either 15% squalane on Chromosorb P (80-100 mesh) at 333 K or 10% E30 on silanised acid-washed diatomite C (100-120 mesh) at 336 K. 2-3 S. RESULTS As indicated, an extensive programme of tests was conducted in which films (of Pd and Pt, principally) were formed in different ways or treated following formation with oxygen, preadsorbed hydrocarbon etc.The general observation may be made that ‘unsintered’ films gave over-rapid deuterium exchange with 1 relative to cyclisation. Deuterium-number contours had a maximum at a relatively high deuterium number, and peaks in the ,H,-,H, range were small. Thus mechanistically meaningful information could not be obtained from the contours. Karpinski employed platinum15 and palladium16 films sintered for 3 h at 800 K in 4 Torr D, and thereby succeeded in reducing the deuterium exchange rate sufficiently to study the mechanism of isomerisation of neopentane. In the present work sintering of Pt and Pd films for 1 h at 773 K only partly alleviated the exchange/cyclisation relative rate difficulty. For example, exchange of 1 with D, on Pd films sintered for 1 h at 773 K in 3-4 Torr D, gave a composite exchange contour$ centred at 2H2-2H, with maximum shifting to ,H, with time: ,H,, and ,H1 were very small.Significantly, f’ 1 Torr = 101 325/760 Pa. 1 Composite contour means the exchange distribution from 1, 2 and 4, and 3 if present, prior to g.1.c. separation.0. E. FINLAYSON, J. K. A. CLARKE AND J. J. ROONEY 195 sintering Pd films at 773 K for a minimum of 2 h gave a composite exchange contour from 2Ho-2H4 with maximum at 2H, shifting to 2H4 with time but [2Ho]TMP, [2Hl]TMP remain of significant size. Experience was similar with Rh and Ir, although these were much less extensively tested. The conditions described were therefore adopted for the detailed isotopic analysis work.The order of emergence of compounds on the OVlOl column was The first two peaks in the g.1.c. trace were thus due to [2H,l]- and [2H2]-2,2,5- trimethylhexane, respectively. The mass spectrometer scans the g.1.c. peak every 2-3 s and such multiple scanning through each peak is necessary to yield a genuine deuterium distribution pattern. Integration methods are available1' for the derivation of an average concentration of each deuteroisomer. It was not necessary to employ this for the present results, rather what was required was the starting point of the deuterium contour of the cyclic and isomeric products. No account was necessary of the dilution of the deuterium pool by hydrogen1*" because of the sharp cut-offs found at low deuterium number in the catalytic runs, as will be shown.Isotopic and fragmentation corrections were made on the results for TMP/D2 exchange only. With the other products it was ensured that the lowest deuterium number of the contour was genuine and not due to fragmentation of the next deuteroisomer of the contour. Positive identification of each reaction product was achieved by (i) preliminary identi- fication from g.1.c.-trace retention times coupled with the m/e value of the pseudo- parent ion (e.g. 11 1 for TMCP) and (ii) a comparison of the observed mass-spectral fragmentation pattern with that measured in preliminary calibration scans for each hydrocarbon. In particular, measurement of fragment peaks at m/e 70 and 71 is important for complete identification of 2 in the presence of considerable isomerised 1.(Thus at 20 eV the isomeric alkanes give a large peak at m/e 71, whereas 2 gives a large peak at m/e 70 and a smaller peak at m/e 71.) The procedure described is especially necessary because of g.1.c. overlap between compounds when deuterium substitution is extensive. IRIDIUM FILMS Both 3 and 4 as well as the cyclic product 2 are formed from the standard reaction mixture at 400 K (fig. 1). The amounts of 4 and 2 are approximately equal while the isomer 3 is present in much lesser amounts. The exchange contour of the isomer 4 clearly starts at m/e 114, i.e. the 2Hl isomer. This means that the bond shift (of a neopentyl group) requires only a monoadsorbed intermediate. A large mass-spectral peak at m/e 11 1 was present and had a maximum between the g.1.c.retention times of the two isomers 4 and 3. The assignment of this peak to the pseudo-parent of 2 was verified by the observation of a large fragmentation peak at m/e 70 associated with a small peak at m/e 71. Exchange of 2 was simple, and only small amounts of 2Hl were formed. This result proves that there is a mechanism of 1,5-dehydrocyclisation possible on Ir which involves direct ring closure from a di-a-adsorbed intermediate.196 100 0 t CYCLISATION AND ISOMERISATION OF ALKANES mass 120 100 120 I 1 i t I I I 15L I 162 L N I CI u - I u 1 I I I I I I I I I I I I - 170 I Fig. 1. Reaction of 1 on Ir at 400 K. Representative mass spectrometry scans (no. 454170) through a g.1.c. chromatogram. Relative concentration is given as a number for each high peak in this and later figures.The notation [2HI]-4 signifies monodeuteroisomer 4. 100 mass 120 100 120 1 I I 1 I i f - ' ' I L 2 c .- Y 5 s L t 1 6 2 155 L I v I' ", 160 / G I T= I Y L Fig. 2. Reaction of 1 on Rh at 513 K. Representative mass spectrometry scans (no. 155-171) through a g.1.c. chromatogram.0. E. FINLAYSON, J. K. A. CLARKE AND J. J. ROONEY 197 The isomer 3 (identity conclusively established from its fragmentation pattern) gave an exchange contour of the pseudo-parent ion insufficiently separated from the TMP 2Ho-2H, contour to give a mechanistically meaningful distribution. No detailed comment on the mechanism of this (methyl group) bond-shift reaction from the present deuterium contours is therefore possible. RHODIUM FILMS Reaction at 513 K (fig.2) gave 4 in much greater amounts than TMCP. Small amounts of the second isomer 3 were produced. The 4 was exchanged to 2H2, implying an aa-diadsorbed species but also small amounts of [2Hll]i~~mer were observed in accordance with the operation of the ap process. This feature of simultaneous maxima appearing in the [2H,]isomer and in the perdeuteroisomer (that possible by the ap process only) for exchange of alkane has previously been noted using unsintered Rh films at and below room temperature18b and can also be attributed to formation and migration of intermediate 2H, alkene to sites where it can only add back two 2H, atoms, but not undergo the ap process. Significantly, the contour for the 2,2,5-isomer clearly starts at m/e 114, i.e. the [2Hl]isomer.As for Ir, the bond-shift rearrangement reaction involves formation of a monoadsorbed intermediate. The cyclic product 2 was present as m/e 11 1 peak with a small amount of m/e 112. The identity of the cyclic product was confirmed by the fragmentation pattern as discussed for Ir . It may be concluded therefore, as for Ir, that the bond shift (of the neopentyl group) requires a monoadsorbed intermediate and that the 1,5-dehydrocyclisation involves the heterogeneous counterpart of reductive elimination of a 1,5-a-dialkyl intermediate. PALLADIUM FILMS It was found beneficial with both Pd and Pt to conduct runs using deuterium and g.l.c./m.s. analysis at ca. 50 K above the onset temperature of cyclisation rather than in the vicinity of that temperature because the relatively more rapid rate of isomerisation at the lower temperature tended to obscure the cyclisation product.3 and 4 were now found on Pd at 468 K but accompanied by a clear manifesta- tion of the deuterium contour of 2 (fig. 3). The contour of isomer 4 commenced clearly at m/e 114, i.e. the [2Hl]isomer. Therefore as on Ir and Rh a monoadsorbed intermediate permits the bond shift by neopentyl shift. 2 appeared at m/e 11 1, i.e. 2H,-cyclic, with some exchange to m/e 1 12, i.e. 2Hle-cyclic. The identity of the 1 1 1 peak was confirmed by the fragmentation pattern in the region m/e 70. Therefore cyclisation can occur on Pd by reductive elimination of a-diadsorbed intermediate. It is interesting that the isomer 3 was formed at approximately the same rate as 4.The former had exchanged to 2H12 with a maximum at 2H,-2H,. The minimum deu- terium content of this contour was not clear because of inadequate separation of the deuteroisomers of 1 and 3. The pattern of exchange of the parent compound 1 is also unclear owing to this lack of separation. However, in contrast to Ir, Rh and (see below) Pt, the exchange of 1 seems to tend towards multiple exchange with small [2Hl]deuteroisomer, maximum at 2H,-2H8 and exchange to 2Hlo. This suggests that an ay-type multiple exchange may be possible on Pd at these temperatures (where Pd exists as the a-hydride phase under conditions of the catalytic run). The formation of 3 appears to imply then a tendencyg for 1,2-methyl shift on Pd through a 1,3-diadsorbed species and a metallacarbonium-type intermediate (scheme 2).This mechanism permits methyl shift of 1 without having a strong repulsion of the bulky neopentyl group by the surface.198 CYCLISATION AND ISOMERISATION OF ALKANES mass 120 100 100 120 I i. c Fig. 3. Reaction of 1 on Pd at 468 K. Representative mass spectrometry scans (no. 100-116) through a g.1.c. chromatogram. mass 100 120 100 120 I 1 I 1 ' 1 I I 1 4 I n r" \ Y I l l I I I I I I I 1 1 1 1 1 I I I I I I - - c I n 'I 11 0 I . . I" I u I I 1 I l l l l l 0 t Fig. 4. Reaction of 1 on Pt at 523 K. Representative mass spectrometry scans (no. 102-110) through a g.1.c. chromatogram. m/e 1 15-1 25 represent a distribution of 2H,-2H,, deuteroisomers of product 4 (scans 102 and 105) or of products 3 and 4 (scans 107 and 110).0.E. FINLAYSON, J. K. A. CLARKE AND J. J. ROONEY 199 C c c \ / I,c /c\ ;\c - c c b + \w/ b- The reaction temperature was cyclisation (453 K). However, at Scheme 2. 2,2,4 - T M H PLATINUM FILMS raised by ca. 70 K above the onset temperature of this temperature isomerisation of 1 to both 3 and 4was very rapid. To poison the isomerisation reaction the hydrocarbon mixture was first admitted at 493 K and the temperature was raised to 543 K. A second hydrocarbon mixture was then admitted at 523 K and the reaction followed as previously described. Both 3 and 4 occurred as products along with 2 at 523 K (fig. 4). The isomer 4 gave a deuterium contour starting at m/e 114, i.e. the [2Hl]deuteroisomer with a maximum at the [2Hll]i~~mer (cut-off) moving to a maximum at the [2H,]isomer with successive mass-spectral scans.The peak at m/e 11 1 in scan 107 is due to the [2Ho]isomer of 2 as confirmed by the fragmentation pattern as for Ir, Rh and Pd. Some 2 with m/e 112 and 113 was also produced arising from exchange to the cyclic [2H2]isomer. The peak at m/e 111 in earlier scans (e.g. 102) was not due to 2 but to the loss of a methyl group from the exchanged 4 (i.e. m/e 126). The isomer 3 showed as a contour with a local maximum at 2Hl, (e.g. scan 107). Again inadequate separation conditions prevent the minimum deuterium content of this isomer from being determined. We note that the formation of 4 was favoured over 3. It may be concluded that one of the bond-shift reactions on Pt (neopentyl shift) requires a monoadsorbed intermediate.Also, 1,5-cyclisation can occur by elimination of a diadsorbed intermediate. PLATINUM-COPPER ALLOY STUDIES Alloy films were prepared as previously reported for Pt-Au.14 Some difficulty was experienced in preparing films of high Pt content owing to the relatively high volatility of copper. The Pt-Cu alloy system forms a continuous series of solid solutions at equilibrium at all temperature^.'^ The films prepared in the present work, which had undergone a homogenisation stage at 773 K, showed two regions of composition in the nominal composition range 50-80 at. % Pt. This is in contrast also to the results of de Jongste,20 who prepared Pt-Cu alloy catalysts which were single-phased at all compositions and which obeyed Vegard’s law. The two-composition region characteristic of the present films is probably due to an insufficient homogenisation temperature.Higher temper- atures are not feasible because of the Pyrex construction of the vessel. At compositions of < 50 at.% Pt lattice parameters agreed with Vegard’s law, indicating bulk homo- geneity on an atomic scale. No indication of superlattice formation was detectable in X-ray diffraction. From the foregoing, X-ray diffraction could be used to test lateral homogeneity of films. Thus for one film having a nominal composition of 23 at.% Pt the mean deviation was f 1.7 at.% Pt with a maximum deviation of k6.5 at.% Pt in samples of metal sampled from around the vessel.200 CYCLISATION AND ISOMERISATION OF ALKANES Table 2. Reaction of 3,3-dimethylpentane + hydrogen on Pt-Cu films initial product distribution (wt%) (10: 1 hydrogen to hydrocarbon ratio) at.% Pt T/K C,-C6 alkanes MCPa 1 , 1 -DMCPb Xc Tol/MCHd k/% min-l 77 518 550 567 59 5 35 497 535 573 600 638 28 576 603 636 22 18 635 0 695 98.7 1.2 86.1 3.9 54.6 2.0 76.2 - 100.0 - 100.0 - 27.4 - 9.3 95.8 89.8 - 88.7 - 57.7 - inactive to 609 K 100.0 - 100.0 - - - - - 70.9 0.3 90.7 - 3.5 10.2 4.2 4.6 36.9 - - - 10.0 14.9 22.0 - 1.5 0.8 - 2.5 5.3 0.22 0.28 0.33 0.002 0.001 0.0 10 0.0 19 0.007 0.004 0.006 0.007 0.56 0.002 0.002 a MCP, methylcyclopentane; * 1,l-DMCP, 1,l-dimethylcyclopentane; X, other C,-ring isomers including unsaturateds; Tol/MCH, toluene + methylcyclohexane. It is probable that film surfaces are more copper-rich than indicated by the nominal alloy composition even in the range of bulk homogeneity.Two ion-scattering spectroscopy studies2'? 22 vindicate theoretical expectation^^^ in showing moderate surface enrichment with copper. Thus bulk concentrations of 13 and 18 at.% Cu showed surface compositions of 48 and 5 1 at. % Cu. A preliminary Auger electron spectroscopic study on 25 at.% Pt-Cu gave a surface concentration of 15 at.% Pt.24 Surface copper enrichment of Pt-Cu alloys has been found to be dependent on crystallite size. 25 The reactions of 3,3-dimethylpentane and n-pentane were studied on the Pt-Cu films at hydrogen to hydrocarbon partial-pressure ratios of 10: 1 (tables 2 and 3) and 4: 1 (tables 4 and 5). The overall activity of the surfaces (total reaction) for both 3,3-dimethylpentane and n-pentane decreased (by 20-fold or more) as Cu was added in both series of experiments. Cyclisation was generally the main process on all films at both partial-pressure ratios : however, in detail a distinction becomes apparent between the two reactants.By comparison of the 3,3-DMP and n-pentane cyclisation reactions at 10: 1 ratio it is seen that 3,3-DMP failed to cyclise to 1,l-DMCP at 22 at.% Pt whereas n-pentane continued to form cyclopentane on alloys of 18 at.% Pt. [We estimate that transition compositions of alloys for catalysis were significant to ca. 2% Pt even though the composition spread is greater (see above).] 100% Cu films failed to cyclise 3,3-DMP to 695 K. By reducing the hydrogen to hydrocarbon ratio from 10: 1 to 4: 1 the overall activity of the films decreased by a factor of ca.2-4. 3,3,-DMP failed to cyclise at a higher Pt content, i.e. below 39 at. % Pt (cyclic product not produced at 27 at. % Pt) whereas n-pentane continued to undergo cyclisation at lower Pt compositions ( < 23 at. % Pt). There is thus a widening of the range in which 3,3-DMP fails to cyclise while n-pentane continues to do so as the hydrogen pressure is decreased. It may thus be deduced that0. E. FINLAYSON, J. K. A. CLARKE AND J. J. ROONEY 20 1 Table 3. Reaction of n-pentane+ hydrogen on Pt-Cu films initial product distribution (wt %) (1 0 : 1 hydrogen to hydrocarbon ratio) at.% Pt T/K C,-C, i-C, c-dieneu c-eneb c-CSc > CSd SHe SIe SCe k/% min-' 100 546 575 607 30 559 589 603 23 540 57 1 609 18 5 56 580 608 644 673 22.3 26.9 - - 50.8 - 0.22 0.27 0.51 69.9 14.4 0.7 4.0 10.8 - 0.70 0.14 0.16 67.1 7.4 3.7 5.7 16.1 - 0.67 0.07 0.26 8.3 8.9 - 2.6 80.1 - 0.08 0.09 0.83 5.4 5.2 2.9 7.5 79.0 - 0.06 0.05 0.89 9.0 1.1 27.1 46.6 16.2 - 0.09 0.01 0.90 15.1 6.4 - 7.2 71.3 - 0.15 0.07 0.78 8.7 1.0 20.3 26.6 43.2 0.4 0.09 0.01 0.90 13.8 1.4 51.6 19.8 13.4 - 0.14 0.01 0.85 75.9 9.3 - - 11.3 3.5 0.64 0.08 0.10 15.5 3.1 - 4.6 75.5 1.3 0.16 0.03 0.80 8.8 1.0 25.9 15.1 49.2 - 0.09 0.01 0.90 28.7 1.9 56.6 11.5 1.1 0.2 0.29 0.02 0.69 26.8 - 30.4 18.6 24.2 - 0.27 - 0.73 0.29 0.13 0.29 0.01 1 0.05, 0.036 0.003 0.0 16 0.017 0.001 0.005 0.010 0.006 0.008 c-diene, cyclopentadiene; c-ene, cyclopentane; c-C,, cyclopentane; > C, hexane and S,, S,, S,, selectivity for hydrogenolysis, isomerisation and cyclisation, respectively. Table 4.Reaction of 3,3-dimethylpentane + hydrogen on Pt-Cu films benzene; initial product distribution (wt%) (4: 1 hydrogen to hydrocarbon ratio) at.% Pt T/K C,-C, alkanes MCP" 1,l-DMCP' Xu Tol/MCH' k/% min-' 100 483 516 551 589 39 552 576 618 27 581 612 645 28.9 43.5 53.8 1.3 100.0 35.9 51.7 67.5 100.0 77.3 - 7.2 63.9 2.2 51.6 - - 30.7 0.7 51.6 0.8 58.4 6.7 - 39.8 6.7 26.0 6.0 - - - - - - - - - - - - - 2.7 14.8 46.3 - 1.8 6.5 16.7 - 0.032 0.042 0.024 0.034 0.001 0.004 0.003 0.001 0.003 0.008 See table 2. the n-pentane molecule has available to it an additional mechanism of cyclisation not possible for 3,3-DMP, and that this mechanism has an intermediate which is more dehydrogenated than that for 3,3-DMP. It is consistent with this deduction that n-pentane cyclised more rapidly on Pt, Pd, Ir and Rh films (table 6) than did 3,3-DMP (table 7). In the reaction of n-pentane, cyclisation is accompanied by considerable monoene and diene formation both on the present Pt-Cu films and on the Pt-Au films of previous work,14 where two Pt-Au films ( 5 and 33% Pt, respectively) were active in n-pentane cyclisation but inactive for 3,3-DMP cyclisation because development202 CYCLISATION AND ISOMERISATION OF ALKANES Table 5.Reaction of n-pentane + hydrogen on Pt-Cu films initial product distribution (wt%) (4: 1 hydrogen to hydrocarbon ratio) at.% Pt T/K C,-C, i-C, c-dienea c-enea c-C, > CSa k/% min-l 37 533 2.9 5.9 0.6 90.6 - 0.018 566 2.3 3.6 1.4 4.8 87.5 0.4 0.042 599 9.6 - 15.6 53.2 20.1 1.5 0.010 23 538 95.0 - 0.003 5.0 - 97.0 3.0 0.0 10 57 1 603 2.5 0.4 42.7 31.4 12.5 10.5 0.014 619 2.9 0.4 31.9 29.2 32.6 3.0 0.0 14 - - - - - - - a See table 3.Table 6. Reaction of n-pentane + hydrogen on sintered metal films initial product distribution (wt%) metal T/K C,-C, alkanes i-C, c-C,-enea c-C, > C,b k/% min-l ref. Ir 52 1 73.0 2.2 - 24.8 - 0.8, 7 556 49.0 7.2 43.8 - 1.3, - 1.3, 589 100.0 658 91.0 0.3 trace 0.4 8.3 0.75 605 37.8 1.5 13.6 47.1 - 0.06, - - - - Rh 574 99.0 0.5 trace 0.5 - 0.37 26 Pd 573 28.1 3.4 68.5 - 0.18 7 Pt see table 3 - a c-C,-ene cyclopentene and cyclopentadiene: > C , n-hexane and benzene. of the additional unsaturation required is prevented by the gem-dimethyl group. We note, lastly, that de Jongste et a1.12 found for a series of Pt-Cu alloys that ring opening of MCP (hydrogen:MCP: : 17: 1) was by NSCM only between 25 and 5 at.% Pt but because surface compositions are not known for either set of alloy catalysts, so that 'matching' of bulk compositions is not possible, we restrict ourselves to the comment that a NSCM must be operating (at the least) at < 22 at.% Pt on the alloy surfaces of the present work at 10: I hydrogen: hydrocarbon pressure ratio.(In drawing this conclusion the assumption is made that cyclisation and ring scission by NSCM both have the same mechanism). The cyclisation activity with n-pentane, 3,3-DMP and TMP are compared for metal films of Ir, Rh, Pd and Pt (tables 6-8). At each of several temperatures the rates increase in the order0. E. FINLAYSON, J. K. A. CLARKE AND J. J. ROONEY 203 Table 7.Reaction of 3,3-dimethylpentane + hydrogen on sintered metal films initial product distribution (wt%) metal T/K C,-C, alkanes MCP 1,l-DMCP Tola k/% min-I Ir 405 433 49 1 Rh 507 533 557 Pd 439 460 484 514 Pt 473 509 542 580 52.2 66.5 95.1 91.8 75.5 89.3 52.9 20.7 7.1 22.2 34.0 54.1 73.3 - 47.8 10.5 0.9 2.0 1.7 4.6 6.3 15.6 4.6 4.1 1.7 - 23.0 4.0 6.1 22.8 10.7 trace 47.1 74.7 86.6 62.2 57.1 32.6 11.5 - 0.007 - 0.016 - 0.12 - 0.09 - 0.05, - 0.08 - 0.002 - 0.006 - 0.014 - 0.05, 4.3 0.06, 9.2 0.10 13.5 0.09 - - a Tol, toluene, methylcyclohexane and other dimethylcyclopentanes (C, isoalkanes are not resolved from reactant peak in g.1.c.). Table 8. Reaction of 2,2,4,4-tetramethylpentane on sintered metal films initial product distribution (wt%) from g.l.c./m.s.: see footnotes partial: see footnotes 2,2,5-TMH k/% 2,2,5-TMH + k/% 1,1,3,3-TMCP: min-lb metal T/K C,-C, alkanes 1,1,3,3-TMCP min-l 2,2,4-TMH (corr.) Pt 459 487 Pd 442 479 Ir 400 430 462 504 Rh (i) 494 513 Rh (ii) 517 1.9 9.3 100.0 19.3 89.5 90.4 89.2 98.6 96.5 92.2 96.8 98.1 90.7 80.7 10.5 9.6 10.8 1.4 3.5 7.8 3.2 - 0.07, 0.06, 0.003 0.05, 0.003 0.008 0.028 0.26 0.027 0.05, 0.30 78: 1:9 (523 K) 84: 1:94 (468 K) 25:l:l (400 K) - 8: 1:0 (413 K) 0.09 0.08 0.003 0.1 1 0.003 0.01 0.03 0.27 0.03 0.06 0.30 a Estimate for each metal of the ratio of the three rearrangement products (from g.l.c./m.s.): 2,2,5-TMH and 1,1,3,3-TMCP are not resolved in the in situ g.1.c. and 2,2,4-TMH is not resolved from the 2,2,4,4-TMP reactant. Total conversion rate corrected using product ratios in previous column and assuming these ratios to apply at the several reaction temperatures.204 CYCLISATION AND ISOMERISATION OF ALKANES Also the onset temperature for cyclisation decreases in the same order.The a& mechanism of cyclisation known to occur with TMP on these metals is also possible for 3,3-DMP and n-pentane. Pt gives comparable rates of cyclisation for the substituted pentanes while Pd cyclises 3,3-DMP at a temperature 20-30 K lower than that required to cyclise TMP. Muller and Gault similarly showed a lower temperature for cyclisation of 2,2,4-TMP and 2,2,3-TMP than for tetramethylpentane on Pd films.* These results support the existence of a cyclisation mechanism in addition to the a& mechanism, one of which is available to (b) above but not to (a).This further mechanism cannot involve a completely dehydrogenated intermediate as previously proposed7 because the presence of the gem-dimethyl group blocks its formation. Finally, the Pt-Cu alloy results show a clear distinction between cyclisation of 3,3-DMP and n-pentane, the latter having available a route requiring a substantially dehydrogenated intermediate. DISCUSSION SELECTIVE l75-CYCLISATI0N MECHANISM For Pt catalysts the NSCM mechanism is generally favoured over SCM at higher metal dispersion,l higher temperature' and lower hydrogen partial pressure.6 Under various conditions of dispersion, pressure and temperature, however, iridium has been shown to cyclise only by the SCM.2v TMP will necessarily cyclise by the SCM because only the terminal carbon atoms are available for cyclisation. The appearance of sub- stantial cyclic [2Ho]isomer at the onset of the deuterium contour of 2 on Ir (fig.1) shows beyond doubt that cyclisation of 1 occurs by formation of an ae-di-o-adsorbed intermediate as in scheme 3. Organometallic analogues are known for this mechanism Yr ' Ir Scheme 3. of reductive elimination. C,, C, and C, rings, respectively, are reductively eliminated from the corresponding polymethylene di-a-complexes of platinum in which electron withdrawing ligands are also coordinated.l0* 2 7 9 28 Although carbynes are reasonably well documented, molecular dicarbynes have still to be reported requiring necessarily a binuclear or larger metal complex. Rhodium, palladium and platinum have also been found to cyclise 1 to 2 by the same a& reductive elimination mechanism which involves removal of the minimum number of hydrogen atoms that is required before ring closure is possible.Other experimental facts apart from the definitive cyclic 2Ho compound produced on each metal support this mechanism, as follows. Assuming an aam intermediate for the formation of 2 then the hydrogenolysis of the cyclic compound (thermodynamically the more favoured direction of reaction) which involves carbene formation should occur at the same temperature as cyclisa- tion of the alkane. On Pd films 2 failed to hydrogenolyse below 493 K whereas 1 cyclised at 436 K. Nickel films also failed to hydrogenolyse 2 at 419 K.ll The temperature for the hydrogenolysis reaction is experimentally higher then than the temperature for cyclisation. Therefore cracking of 2 and cyclisation of 1 must involve different mechanisms. The law of microscopic reversibility is not violated here.The cycloalkane is chemisorbed by the relatively easy C-H rupture and the resulting monoadsorbed species becomes a/? diadsorbed etc. rather than sustaining the more difficult metal-atom insertion reaction. Cracking may then proceed via the dicarbene0. E. FINLAYSON, J. K. A. CLARKE AND J. J. ROOMY 205 intermediate visualised by Gault. [Significantly, exchange of 2 with deuterium on Pd films at 440 K gave a large percentage of 2Hl (and also some 2H2-2H5) suggesting that carbene formation was not occurring at this temperature.] Gault noted29 that ring closure across the transannular positions of Cg-C,, ring compounds (scheme 4) to give bicycl~alkanes~~ is not possible by dicarbene or dicarbyne formation owing to conformational strain.ThePt-catalysed formation at Scheme 4. 703 K of triamantane by 1,6-cyclisation of a polycyclic alkene structurally related to it reported by Burns et aL31 is also not possible by carbene or carbyne intermediates owing to steric hindrance, Gault suggested a 1,5-di-a-adsorbed intermediate for transannular ring closure. He dismissed the possibility of such a mechanism of ring closure for small molecules such as n-hexane because he considered it not to give an explanation of the changes observed in selectivity on large-sized particles of Pt, on alloying with a Group IB metal, or the changes in selectivity of the different metals Ir, Pd and Pt.29 Hydrogen-pressure dependences were not considered and these indeed provide the key to these problems, as developed in the following paragraphs.The mechanism of SCM on Ir, Rh, Pd and Pt involving reductive elimination of an a&-diadsorbed species requires only one metal atom. Conclusions arrived at previously as to the surface-structure sensitivity of SCM on Pt are affected by the confusion that the ring-opening reaction is in practice by a different preferred mechanism to the ring-closure direction of reaction as already discussed : experimental results were in the main for the former in Gault’s programme. For Ir, however, the cyclisation reaction (SCM) was, qualitatively at least, independent of Neither the SCM nor the NSCM reaction took place on a PtAu/Aerosil alloy having good activity for the bond-shift reaction and ring enla~gernent,~~ suggesting a surface-structure sensitivity which was not satisfied on this severely heat-treated surface. The question of the surface-structure sensitivity of selective 1,5-cyclisation cannot be regarded as settled, but on balance we incline to the view that it is surface-structure sensitive.However, this may be resolved because as the intermediate lies edge-on to the surface, steric hindrance by methyl or larger alkyl groups in the 1 or 5 position of the alkane chain will block the formation of the a&-species, as may readily be appreciated from a molecular model. The need for diterminal carbon atoms for SCM is not therefore due to dissociation of the hydrogen atoms but is a steric factor only.Finally, in disagreement with Gault and coworkers,* who suggested that Pd catalysts cyclise by NSCM only, Pd films have now been shown to cyclise 1 at 436 K upward. Muller and Gault failed to find cyclisation of 1 at 573 K on Pd films.8 NON-SELECTIVE 1 ,SCYCLISATION The NSCM is important at lower hydrogen pressures and generally occurs at higher temperatures than the SCM. The SCM involves an ae-intermediate, and therefore a206 CYCLISATION AND ISOMERISATION OF ALKANES more dehydrogenated intermediate than this is required for NSCM. In the paragraphs which follow the possible subdivision within the general class of non-selective I ,5-cyclisations will be considered. As discussed in the Results section the cyclisation of 3,3-DMP occupies an intermediate position between n-pentane and 1 in the rate of reaction at a given temperature and in the onset temperature for cyclisation.The 3,3-DMP cyclisation may be intermediate also in the degree of dehydrogenation of the reaction intermediate as follows. The results with Pt-Cu films show that 3,3-DMP fails to cyclise on alloys at low % Pt while n-pentane continues to do so. Decrease of hydrogen partial pressure brings about a widening of the region of alloy composition in which this distinction applies. This result implies that the additional route available for n-pentane cyclisation involves a more dehydrogenated intermediate than for 3,3-DMP cyclisation. Ponec et ~ 2 Z . l ~ have concluded that NSCM-type cyclisation takes place at < 21 at.% Pt in PtCu alloys in their work.We suggest that both n-pentane and 3,3-DMP cyclise by NSCM at low percentages of platinum in our Pt-Cu films, the additional mechanism for n-pentane being the pentadienyl -, cyclopentenyl route7 precluded for 3,3-DMP by the gern-dimethyl group. Only under extreme conditions of low hydrogen concentration, e.g. alloying with Cu or Au, or at high reaction temperatures,14 is the completely dehydrogenated (pentadienyl) intermediate of importance. This is the end species in a progressive series of partly dehydrogenated intermediates : a&, a@, a&, a@$&. The ap& mechanism available for 3,3-DMP may be an alkyl/alkene or a& reductive elimination from an initially n-monoadsorbed chain having lengthened surface sojourn time to permit successful chelation to the site atom.The statistical ratios of the hydrogenolysis products of methylcyclopentane by NSCM may be explained by a completely dehydrogenated intermediate (n-bonded to a Pt atom) shown in scheme 5 or by the partly dehydrogenated intermediates alkyl/alkene ( a p ~ ) or alkyl/allyl (aby~) insertion. [We note here that the latter two modes (abc) or (abyc) may effect direct 1,6-closure as well (NSCM).] The conversion from n-ally1 to 7 ‘ I J Scheme 5.0. E. FINLAYSON, J. K. A. CLARKE AND J. J. ROONEY 207 pentadienyl species or from a-adsorbed alkyl to n-adsorbed alkene/a-adsorbed alkyl gives an equal chance of breaking all bonds in the MCP ring. Decreasing the surface hydrogen pressure should favour the NSCM by the above mechanisms over SCM.We may summarise the possible intermediates for various pentane chains as follows : cyclisation intermediate SCM only a& apc apy& SCM, NSCM + a& SCM, NSCM a& as& apy& SCM, NSCM n Some deductions may be noted. For n-pentane cyclisation the ease of formation of the possible intermediates is in the order apyde > ap& > a&, the first of these resulting from progressive dehydrogenation, i.e. a -+ ap -+ apy + apyd -+ apydc, whereas a/% requires the monoadsorbed species to become diadsorbed, i.e. it is not progressive. Therefore n-pentane may cyclise under conditions where 3,3-DMP fails to. Further, factors which alter the surface hydrogen concentration, e.g.carbiding, al- loying, temperature and possibly aromatics, will change the selectivity of the catalyst for the cyclisation reaction. The dispersion of the catalyst may also be important for selectivity for cyclisation; it certainly is for ring opening. Gault and his group have found29 that NSCM (ring opening) occurs at Pt diameters of < 25 A, and not at all for > 25 A. As noted earlier, a PtAu surface subjected to high-temperature reduction retained activity for bond-shift rearrangement and for dehydrogenation but was inactive for n-pentane cy~lisation,~~ suggesting that sites necessary for both SCM and NSCM had been removed. Even highly dispersed Ir catalysts, however, are inactive in NSCM (both ring opening and ring scission).273 From the foregoing remarks the existence of the required sites may be a necessary but not sufficient condition for non-selective cyclisation and the ability of the surface to effect the required degree of dehydrogenation of the reactant may be equally necessary.1,2-BOND-SHIFT REARRANGEMENTS The present work clearly confirms that an adsorbed alkyl is already at a sufficient degree of dehydrogenationl59l6 for bond shift as first claimed from studies of isomerisation reactions of caged hydrocarbon^^^ (scheme 6). This is potentially very - 1 'ct - \ C"3 M CH,- C .- CHYA-C-' RCH2-C -CH3 I M 1 '3 M Scheme 6.208 CYCLISATION AND ISOMERISATION OF ALKANES important when viewed in the light of the fact that various alkyl-cobalt complexes which are intermediates in Vitamin B,, catalysed vicinal interchange reactions, e.g.Q-methyl aspartate e glutamate (cJ neopentane + i~opentane),~~ also undergo 1,2- bond-shift is~merisations.~~ Another very important aspect is that while it was necessary to use heavily sintered films in this work each metal still exhibits its characteristic behaviour (most particularly for the ap process), as noted over twenty years ago for unsintered films at or near ambient ternperat~res.~~ We can therefore be confident that the nature of the sites is the same, only the numbers per m2 have changed. Pd still gives very extensive multiple exchange in contrast to Rh. CONCLUSIONS Firm evidence has been provided here that in the mechanisms of ‘non-destructive’ reactions such as cyclisation and 1’2-bond-shift rearrangements of hydrocarbons one metal atom only is required as the catalytic centre, just as previously argued for hydrogenation and dehydr~genation.~~ Therefore the claim that catalysis is often due to interconversion of various intermediates as ligands of one surface metal atom or ion also extends to the higher-temperature reactions as well.This is a very important philosophy because of the current importance attached to such terms as ‘facile’, ‘demanding’ and ‘structure sensitive’ and an over-ready acceptance of such jargon without critically examining what it means as far as surface sites are concerned. In the hydrocarbon field multiple bonding at the edges or faces of surface ensembles or clusters may only be important when carbene, carbyne and carbide intermediates are involved, i.e.in ‘destructive’ reactions such as hydrocracking. Even in the reverse re- action of hom~logation~~ the key homologation step of addition of methylene to an alkene may still be mediated by one metal atom. We thank Mr D. Clements and Gallaher Ltd for the g.l.c./m.s. analyses. G. Maire, G. Plouidy, J. C. Prudhomme and F. G. Gault, J. Catal., 1965, 4, 556. F. Weisang and F. G. Gault, J. Chem. Soc., Chem. Commun., 1979, 519. F. Weisang, Thbe Docteur-IngLnieur (University of Strasbourg, 1979). M. Hajek, G. Maire, A. 0. Cinneide, C. Corolleur and F. G. Gault, J. Chim. Phys., 1974, 71, 1329. F. G. Gault, Gazz. Chim. Ital., 1979, 109, 255. 0. V. Bragin, Z. Karpinski, K. Matusek, Z. Paal and P. Tetenyi, J. Catal., 1979, 56, 219. (b) F. E. Shephard and J. J. Rooney, J.Catal., 1964, 3, 129. J. M. Muller and F. G. Gault, J. Catal., 1972, 24, 361. J. K. A. Clarke and J. J. Rooney, Ado. Catal., 1976, 25, 125. ’I (a) C. O’Donohoe, J. K. A. Clarke, and J. J. Rooney, J. Chem. SOC., Faraday Trans. I , 1980,76,345; lo G. B. Young and G. M. Whitesides, J. Am. Chem. SOC., 1978, 100, 5808. l1 F. G. Gault, J. J. Rooney and C. Kemball, J. Catal., 1962, 1, 255. H. C. de Jongste and V. Ponec, Proc. 7th Znt. Congr. Catal., Tokyo, 1980, ed. T. Seiyama and K. Tanabe (Elsevier, Amsterdam, 1981), p. 186. l3 H. C. de Jongste, V. Ponec and F. G. Gault, J. Catal., 1980, 63, 395. l4 A. F. Kane and J. K. A. Clarke, J. Chem. SOC., Faraday Trans. I , 1980,76, 1640. l5 Z. Karpinski and L. Guczi, J. Chem. Soc., Chem. Commun., 1977, 563. l6 Z. Karpinski, Nouv. J. Chim., 1980, 4, 561. l7 R. S. Dowie, C. Kemball, J. C. Kempling and D. A. Whan, Proc. R. SOC. London, Ser. A , 1972,327, l8 (a) C. Kemball and J. C. Kempling, Proc. R. SOC. London, Ser. A , 1972, 329, 391 ; (b) F. G. Gault Is A. Schneider and U. Esch, Z. Elektrochem., 1944, 50, 290. *O H. C. de Jongste, F. J. Kuijers and V. Ponec, Proc. 6th Int. Congr. Catal., London, 1976, ed. 491. and C. Kemball, Trans. Furaday SOC., 1961, 57, 1781. G. C. Bond, P. B. Wells and F. C. Tompkins (The Chemical Society, London, 1977), p. 915.0. E. FINLAYSON, J. K. A. CLARKE AND J. J. ROONEY 209 21 M. J. Kelley, D. G. Swartzfager and V. S. Sundaram, J. Vac. Sci. Technol., 1979, 16, 664. 22 H. H. Brongersma, M. J. Sparnaay and T. M. Buck, Surf: Sci., 1978, 71, 657. 23 V. S. Sundaram and P. Wynblatt, Su$. Sci., 1975, 52, 569. 24 A. D. Van Langeveld, cited by F. Stoop, F. Toolenaar and V. Ponec, J. Chem. SOC., Chem. Commun., 25 J. H. Anderson, P. J. Conn and S. G. Brandenberger, J. Catal., 1970, 16, 326. 26 J. F. Taylor, Ph.D. Thesis (National University of Ireland, 1976). 27 P. Foley and G. M. Whitesides, J. Am. Chem. Soc., 1979, 101, 2732. 28 P. W. Hall, R. J. Puddephatt, K. R. Seddon and C. F. H. Tipper, J. Organomet. Chem., 1974,81,423. 2B F. G. Gault, Ad3. Catal., 1981, 30, 1 . 30 B. A. Kazanskii, E. A. Shokova, S. I. Khromov, V. T. Aleksanyan and Kh. E. Sterin, Dokl Akad. 31 W. Burns, M. A. McKervey and J. J. Rooney, J. Chem. Soc., Chem. Commun., 1975, 965. 32 J. K. A. Clarke, A. F. Kane and T. Baird, J. Catal., 1980, 64, 200. 33 A. Heumann, M. Reglier and B. Waegell, Angew. Chem., Znt. Ed. Engl., 1979, 18, 866. 34 P. Parayre, V. Amir-Ebrahimi and F. G. Gault, J. Chem. Soc., Faraday Trans. I , 1980, 76, 1723. 35 M. A. McKervey, J. J. Rooney and N. G. Samman, J . Catal., 1973, 30, 330. 36 J. M. Pratt, in B Twelve, ed. D. Dolphin (Wiley, New York, 1982), vol. 1 , p. 325. 37 R. Hamilton and J. J. Rooney, J. Mol. Catal., 1982, 17, 29. 38 C. Kemball, Bull. Soc. Chim. Belg., 1958, 67, 373. 39 J. J. Rooney and G. Webb, J . Catal., 1964, 3, 488. 1981, 1024. Nauk SSSR, 1960, 133, 1090. (PAPER 3/808)

ISSN:0300-9599    

DOI:10.1039/F19848000191

出版商:RSC    

年代:1984

数据来源: RSC

摘要:

J . Chem. SOC., Furaday Trans. I , 1984, 80, 21 1-21 6 Electron Addition to Triphenylmethyl Arsonium Iodide An Electron Spin Resonance Study BY MARTYN C. R. SYMONS* AND GLEN D. G. MCCONNACHIE Department of Chemistry, The University, Leicester LE 1 7RH Received 27th May, 1983 Exposure of crystalline Ph,AsMe+I- crystals to s°Co y-rays at 77 K yields an electron-adduct having e.s.r. parameters characteristic of methyl radicals weakly interacting with a Ph,As mol- ecule. We now find that, on annealing, this species dissociates to give normal methyl radicals, whilst solutions in methanol (CD,OD) give, on electron addition at 77 K, the normal arsoranyl radical [Ph,AsMe] together with methyl radicals, there being no trace of the adduct formed in the crystalline material. Reasons for these differences are discussed and it is suggested that the arsoranyl radical rearranges to a o* species as a necessary step in the dissociation process. Geoffroy and Llinaresl have recently established that electron addition to Ph,AsMe+ ions in triphenylmethyl arsonium iodide crystals at 77 K yields a species having e.s.r.parameters (lH, 13C and g) characteristic of methyl radicals, except that there was also a well defined splitting from a single 7 5 A ~ nucleus. They estimated a spin density of ca. 0.75 on carbon for this adduct. No further species were detected on annealing. They drew attention to the surprising difference between this species and the phosphoranyl radicals 'PL, normally formed by electron addition to similar phosphorus compounds. Our interest in these results stemmed from a recent controversy regarding possible structures for the radical anions of alkyl halides and fluorinated alkyl Discussion centred around the fact that (F,C'-ha1)- anions have a well defined o* structure, but similar alkyl halide anions, (R' -hal)-, have never been prepared.We argued that a major contributory factor must be the tendency for alkyl radicals to prefer a planar structure, whereas ' CF, radicals are pyramidal. Clearly another factor is the electron-withdrawing power of the fluorine atoms, which lowers the energy of the carbon o level thus increasing the covalency of the anion. These factors are illustrated in fig. 1. The force of these arguments is nicely illustrated by results for the isostructural ammonia derivatives, [H,N'-hal].12 These species are o* anions, with considerable spin density on the halogen and with significant 2s contribution from nitrogen indicating pyramidality for the H3N group.Delocalisation onto halogen and 2s character on nitrogen increase on going from C1 to I, thus establishing the expected trend in orbital hybridisation. The aim of the present study was to discover conditions under which the As-Me bond does not break on electron addition and to learn more about the structure of the adduct. This work links with our recent study of the (Br-CN)- radical anion which was shown to exist in a linear, o* structure on initial electron addition, but rather than undergoing dissociation this rearranged to a bent, onb structure14 on annealing above 77 K.15 8 21 1 F A R 1212 E.S.R.STUDY OF Ph,AsMeI + adduct SOMO fi (CH,) C(~P,) 2. E Y g m v M .$ (CH3) c(sp3)- s 2 b) 4 u* anion SOMO c .- Y .- (CF,) C(sp3) .- - Ih Y .- z 41 ir \(I Fig. 1. Bonding scheme for H,C’----hal- ‘adduct’ and for (F,C’-ha1)- CT* radical anion. The vertical arrow shows how the rehybridisation that occurs as the CH, unit flattens aids in weakening the C-ha1 bond. 1 3264 G u u u u -Yz -Y* .‘/2 .3/2 T- Fig. 2. First-derivative X-band e.s.r. spectrum for Ph,AsMe+I- after exposure to s°CO y-rays at 77 K showing features assigned to Ph,As- - - -. CH, adducts. EXPERIMENTAL Triphenylmethylarsonium iodide was obtained from Lancaster Synthesis. The pure compound and its solutions in CD,OD were exposed to s°Co y-rays at 77 K in a Vickrad y-cell with doses up to 1 Mrad.E.s.r. spectra were measured at 77 K with a Varian El09 spectrometer calibrated with a Hewlett-Packard 5246L frequency counter and a Bruker B-H 12E field probe. Samples were annealed by decanting the liquid nitrogen from the insert Dewar and recooling whenever significant spectral changes were observed.M.C.R. SYMONS AND G. D . G. MCCONNACHIE i 3250 G v I I 1 3260 G 100 G - H I I - 3 4 L 1 - I ,H 10 G , 213 Fig. 3. First-derivative X-band e.s.r. spectra for a dilute solution of Ph,AsMe+I- in CD,OD after exposure to s°Co y-rays at 77 K, (a) showing features assigned to ph,AsMe] arsoranyl radicals and (b) showing features assigned to methyl radicals. The centre features are largely due to solvent radicals. 8-2214 E.S.R. STUDY OF Ph,AsMeI RESULTS AND DISCUSSION The e.s.r.spectrum obtained from the pure compound is shown in fig. 2. The data derived therefrom are very close to those obtained by Geoffroy and Llinares from their single-crystal experiments. In marked contrast, solutions in CD,OD gave the spectrum shown in fig. 3. The central region of this spectrum is shown in fig. 3(b). The four features shown in fig. 3(a) are assigned to the arsoranyl radical Ph,AsMe. 7 5 A ~ has I = $!, but because of the large magnitude of the hyperfine coupling the four features corresponding to M , = f $, -k$ are unequally spaced. The features exhibit parallel and perpendicular components with a small x, y splitting for the M , = -f line. The MI = +f line is fortuitously almost isotropic. The results have been analysed using the axially symmetric spin Hamiltonian Table 1.E.s.r. hyperfine parameters hyperfine coupling to 7 5 A ~ , g- tensor Br or lZ7I/Gu componentsa lH hyperfine coupling radical All A , Aiso gll g, constant/G ref Ph,AsMe 592 505 54 1 Ph,As 592.6 505 54 1 Ph,As'-Me 102f2 98+2 99k4 525 524.8 Ph,As' -Me 97 88 90 CH,Br- 58 -28 0.7 86 CH,I- 108 -60 -4 1.97 2.0 14 - this work 1.97 2.014 16 2.0028 2.0017 (11)22.7 this work 2.0010 (1)22.0 2.0021 2.0014 - 2.0009 2.027 2.027 - 1 - 20.6 22 - 20.6 22 - - a Calculated using the Breit-Rabi equation. The resulting tensor components are given in table 1, together with those for the Ph,As ' radical16 for comparison. The four narrow features shown in fig. 3(b) are characteristic of 'free' methyl radicals, with no trace of any residual interaction with the Ph,As molecule.In their single-crystal study Geoffroy and Llinares were unable to detect any break-down products from the [Ph,As- - - - -Me'] adduct. However, using our technique of rapid quenching outlined above we have been able to obtain well defined features for normal methyl radicals on annealing, together with features assigned to H$AsPh, radicals, presumably formed by Ph,AsCH, + * CH, -+ Ph,AsCH, + CH,. (2) ASPECTS OF MECHANISM Irradiation of the pure salt, Ph,AsMe+I-, is expected to give electron-gain and electron-loss centres. The former is clearly the methyl radical addult, there being no trace of either the arsoranyl radical or of the substituted benzene anion derivativesM.C.R. SYMONS AND G. D. G. MCCONNACHIE 215 such as are usually formed from the phosphorus ana10gues.l~ The electron-loss centre is almost certainly I;, with very broad, ill defined features such as are often found in ' loose ' matrices in which extensive libration can occur.In marked contrast, electron addition to Ph,AsMe+ ions in a CD,OD matrix gave no sign of the methyl radical adduct, but low yields of methyl radicals together with relatively high yields of arsoranyl radicals. Ph Ph When the electron comes into the sphere of attraction of the 'tetrahedral' cation it can be stabilised either by bond stretching to give a o* radical, structure (I),l8?l9 or by bond-bending to give an arsoranyl radical, structure (11) or (111). We suggest that structure (I) is a necessary precursor to the formation of methyl radicals, whilst structure (11) or (111) must convert to structure (I) before giving methyl radicals.This suggestion is similar to our postulate for the formation of phenyl radicals and halide ions from ha10genobenzenes.l~ Initial electron addition is partitioned between the n* aromatic orbital and the C-ha1 o* orbital. The latter anion can move smoothly to give products, but the former cannot so a n* + o* excitation is required. For the arsine, a similar excitation is required for structure (11), which is probably the favoured structure. However, for structure (111) there is already extensive delocalisation onto the two axial groups so direct loss of methyl could occur without any electronic excitation. Hence the rate-determining step could well be thermal conversion of structure (11) to structure (111).ASPECTS OF STRUCTURE THE ARSORANYL RADICAL We cannot readily distinguish between structures (I)-(111) from the e.s.r. parameters alone. However, note that the parameters are identical, within experimental error, to those for the Ph,As' radical.16 Since the parameters are known to depend primarily on the nature of the axial ligands in such structures, this suggests that structure (111) should be favoured. This is also expected in terms of the observation that the more electronegative substituents favour the axial sites. The o* structure (I) is unlikely because it is not expected to resemble closely the Ph,As' radical. This is unlikely to have a o* structure, and even if it had it would not be identical with structure (I).However, the most important argument against the o* structure (I) is the one we have used to explain the absence of o* alkyl halide anions.10~13~14~20 If this structure represents a minimum along the reaction coordinate leading to Ph,As: + 'CH,, then given that these units are in close contact in the methyl radical adduct, we would expect 'recombination' to give the o* structure. In other words, we think it is unacceptable to have such a high barrier between the o* state and the methyl radical adduct state that the two species can be quite stable at 77 K with no tendency to interconvert. We therefore conclude that the arsoranyl radical centre detected in CD,OD has structure (111).216 E.S.R. STUDY OF Ph,AsMeI THE METHYL RADICAL ADDUCT It was previously concluded that this centre has ca.75% spin density on 'CH, and, presumably, 25% on the Ph,As group. In our view, however, the species is essentially a pure methyl radical undergoing a very weak charge-transfer interaction with the Ph,As molecule. Data for this adduct and for a range of other adducts are given in table 1. The lH hyperfine coupling is seen to be a sensitive probe for delocalisation, in which case there is almost zero loss of spin density for the Ph,As adduct. Certainly the anisotropic 13C coupling is less than that expected for stationary 'CH, radicals, but that can be understood in terms of libratory motions of these small radicals rather than delocalisation. Indeed, this is normal for 'CH, radicals at 77 K. If the isotropic coupling is used to estimate the spin density on carbon it must again be close to unity.Had there been some 25 % delocalisation, indicative of considerable a-bonding, the CH, unit should have become pyramidal, as for the NH, group in the hale-NH, cenfres.l2 This would lead to a reduction in IA(lH)I and a major increase in A(13C), neither being observed. We conclude that the centre is essentially a methyl radical. In that case the small coupling to 7 5 A ~ represents a slight electron transfer. If the isotropic and anisotropic coupling constants are analysed in the usual way21 we find ca. 0.018 4s character and ca. 0.033 4p character, in reasonable agreement with the model. Similar values were actually estimated by Geoffroy and Llinares, but they apparently favoured the 13C results.Since even this degree of delocalisation is not evidenced by the isotropic lH and 13C coupling constants, it is possible that the 75A~ results are due to a spin-polarisation mechanism rather than representing real delocalisation. M. Geoffroy and A. Llinares, Mol. Phys., 1980, 41, 55. E. D. Sprague and F. Williams, J. Chem. Phys., 1971, 54, 5425. S. P. Mishra and M. C. R. Symons, J . Chem. Soc., Perkin Trans. 2, 1973, 391. Y . Fujita, T. Katsu and K. Takahashi, J. Chem. Phys., 1974, 61, 4307. A. Hasegawa, M. Shiotani and F. Williams, Faraday Discuss. Chem. Soc., 1977,63, 157. D. J. Nelson and M. C. R. Symons, Chem. Phys. Lett., 1977,47,436. M. C . R. Symons, J. Chem. Soc., Chem. Commun., 1977,408. M. C. R. Symons, J. Chem. Res. (S), 1978, 360. J. T. Wang and F. Williams, Chem. Phys. Lett., 1980, 72, 556. lo M. C. R. Symons, Chem. Phys. Lett., 1980, 72, 559. l1 J. T. Wang and F. Williams, J . Am. Chem. Soc., 1980, 102, 2861. l2 M. C. R. Symons, J . Chem. Res. (S), 1981, 160. l3 M. C. R. Symons, Proceedings of the 6th International Conference on Radiation Research, 1979,6,238. l4 M. C. R. Symons, Pure Appl. Chem., 1981,53, 223. l6 M. C. R. Symons and S. P. Mishra, J. Chem. Soc., Furaday Trans. I , 1982, 78, 3019. l6 S. A. Fieldhouse, H. C. Starkie and M. C. R. Symons, Chem. Phys. Lett., 1973, 23, 508. S. P. Mishra and M. C. R. Symons, J. Chem. SOC., Perkin Trans. 2, 1976, 21. l* T. Berclaz, M. Geoffroy and E. A. C. Lucken, Chem. Phys. Lett., 1975, 36, 677. lS M. C. R. Symons, Chem. Phys. Lett., 1976,40, 226. 20 M. C. R. Symons, Radial. Phys. Chem., 1980, 15, 453. 21 M. C. R. Symons, Chemical and Biochemical Aspects of Electron Spin Resonance Spectroscopy (Van 22 M. C. R. Symons and I. G. Smith, J . Chem. Soc., Perkin Trans. 2, 1981, 1180. Nostrand Reinhold, Wokingham, 1978). (PAPER 3/862)

ISSN:0300-9599    

DOI:10.1039/F19848000211

出版商:RSC    

年代:1984

数据来源: RSC

摘要:

J. Chem. Soc., Faraday Trans. I, 1984,80, 217-224 Crystal Growth of Strontium Fluoride from Aqueous Solution BY ROBERT A. BOCHNER, ABBAS ABDUL-RAHMAN AND GEORGE H. NANCOLLAS* Chemistry Department, State University of New York at Buffalo, Buffalo, New York 14214, U.S.A. Received 6th June, 1983 The kinetics of growth of strontium fluoride crystals has been studied in aqueous solution at 25 OC using a constant-composition method in which the supersaturation and ionic strength were maintained constant by the addition of titrants consisting of strontium nitrate and potassium fluoride solutions. The rate of crystallization is independent of ionic strength and, at low supersaturation, is best represented in terms of a spiral growth mechanism with a reaction order, n = 2, with respect to relative supersaturation.At higher supersaturations, a greater change in rate of growth with increasing concentration (n x 4.3) suggests a polynuclear mechanism. Inhibition of crystallization by the presence of phosphonates can be interpreted in terms of a Langmuir isotherm. The crystal growth of the alkaline-earth fluorides is of importance in view of their involvement in spectroscopy, electronics, lasers, glass manufacture and in the fluoridation of drinking waters. In the case of strontium fluoride, the crystallization as strontium-90 fluoride offers an attractive method for removing this element from highly radioactive fission products. The development of methods for growing large, easily filtered crystals of this sparingly soluble salt would significantly improve this process.The irreproducibility of spontaneous precipitation studies is well known since both nucleation and crystal growth may be involved concomitantly. These problems were overcome with the development of highly reproducible seeded-growth techniquesly which allowed factors such as temperature, supersaturation, ionic strength and solid: solution ratios to be studied quantitatively. In the present work a constant- composition method3 has been used to investigate the crystallization of strontium fluoride under conditions of constant ionic strength and supersaturation. The method yields kinetics data for the crystallization reaction even at very low supersaturation. The application of the technique to the industrial preparation of large crystals of strontium fluoride has attractive possibilities.EXPERIMENTAL Supersaturated solutions of strontium fluoride were prepared using both ultrapure (Alfa Chemicals) and reagent-grade (J. T. Baker) chemicals with triply distilled deionized water. The concentration of strontium ions was determined (0.2%) by exchanging the metal ion for hydrogen ion on a Dowex-50 ion-exchange resin and titrating the liberated acid with standard base. All fluoride solutions were prepared and stored in polyethylene or polypropylene bottles in order to prevent fluoride attack on glass surfaces. Seed crystals were prepared by simultaneously adding 200 cm3 of 0.956 mol dmP3 potassium fluoride and 190 cm3 of 0.504 mol dmP3 strontium chloride to 250 cm3 of triply distilled water. The slow mixing was made in a nitrogen atmosphere over a period of 2.5 h at 25 OC.The crystals were washed free of chloride ions with triply distilled water by decantation, and the suspension (seed A) was stored 217218 CRYSTAL GROWTH FROM AQUEOUS SOLUTION in polyethylene containers for up to one year before use. Seed suspension B was prepared by adding 1 dm3 of potassium fluoride (1 .O mol dmP3) and 1 dm3 of strontium nitrate (0.50 mol dm-3) to 500 cm3 of triply distilled water over a period of 12 h at 25 "C. The particles were kept in suspension by means of a Teflon overhead stirrer, and the filtered solid was washed with saturated strontium fluoride solution and allowed to age before use for up to six months. A third seed suspension (seed C) was obtained by the rapid mixing (ca.4 min) of 1 dmP3 volumes of potassium fluoride (0.780 mol dm-3) and strontium nitrate (0.390 mol dm-3) solutions. The crystals were washed with distilled water and allowed to age at 25 O C . Seed crystals were confirmed as strontium fluoride by X-ray powder diffraction (Phillips XRG 3000 X-ray diffractometer, Ni filter and Cu Ka radiation). Specific surface areas (s.s.a.) of the seed and grown phases were measured (& 1 %) by a single-point B.E.T. nitrogen adsorption using a 30% nitrogen + helium mixture (Quantasorb 11, Quantachrome Greenvale, N.Y .). The values were l9.7,2.68 and 33 m2 g-l for seeds A, B and C, respectively. Scanning electron micrographs (IS1 model I1 scanning electron microscope) and transmission electron micrographs (Lietz electron microscope) showed the seed materials to consist of small aggregates of particles having cube-like morphology. Crystal-growth experiments were made in a stream of pre-saturated nitrogen gas using a cell consisting of a double-walled Pyrex glass vessel of 500 cm3 capacity fitted with a Teflon lid and polyethylene liner.The cell was maintained at 25 OC by circulating thermostatted water through the outer jacket. Stirring was effected using a Teflon stirrer with an overhead variable-speed motor (model I, Eastern Engineering Co., Conn.). Supersaturated solutions were prepared by slowly mixing strontium nitrate and potassium (or sodium) fluorides in the range of super- saturation (ti = O.OC1.8) in which the supersaturation, ti, is defined by 0 = (([Sr2+] [F-]2)1'3 - ([Sr2+]o[F-]f)1~3})/([Sr2fl,[F-]~)1~3 where [Sr2+], [F-] and [Sr2+],,, [F-1, are the concentrations of lattice ions at time t and at equilibrium, respectively.The latter values were calculated from conditional solubility products at the ionic strengths of the experiments. In the crystallization experiments, following the addition of a known volume of suspension of seed crystals to the metastable supersaturated solutions, the activities of ion species were maintained constant by the addition of titrant solutions consisting of strontium nitrate and potassium fluoride from mechanically coupled burets. The addition was controlled by means of a pH-stat (Metrohm Combitrator model 3D, Brinkmann Instrument Co.) using a fluoride- specific electrode (Orion model 94-09), coupled with a thermal-electrolytic silver, silver chloride reference electrode.This reference electrode was designed to include an intermediate liquid junction containing background electrolyte so as to eliminate errors due to leakage of salt bridge solution into the crystallization cell during the reactions. The reference electrode limb which was immersed in the cell solution was constructed of Teflon. During the crystallization experiments, aliquots were withdrawn from time to time, filtered (0.2 pm Millipore filtration) and analysed for strontium ion by spectrophotometric titration with EDTA and by atomic absorption (Perkin-Elmer model 503). The data confirmed the constancy of the lattice-ion concentrations to within 1 %. The solid phases, collected during the experiments, were also investigated by X-ray diffraction, specific-surface-area analysis and scanning electron microscopy. RESULTS AND DISCUSSION Concentrations of ionic species in the supersaturated solutions were calculated as described previously4 using expressions for mass balance, electroneutrality and the thermodynamic equilibrium constants, K , for the various associated species in the equilibria : K H F e H++ F- 6.61 x ref.(5) HF+ F- s HF; 0.295, ref. (5) SrF+ e Sr2+ + F- 0.147, ref. (6) SrOH+ + Sr2+ +OH- 0.150, ref. (7) H,O e H+ + OH- 1.002 x ref. (8).R. A. BOCHNER, A. ABDUL-RAHMAN AND G. H. NANCOLLAS 219 The computations were made by successive approximations for the ionic strength, I, as described previouslyg using activity coefficients calculated from the extended form of the Debye-Hiickel equation proposed by Davies.l0 For the experiments which were made at constant ionic strength (0.15 mol dm-3), values of the conditional equilibrium constants were also estimated by using activity-coefficient data calculated from the Davies equation.The solubility of strontium fluoride was obtained by allowing growth experiments to proceed to equilibrium; the value at 25 "C was 1.343 x mol dm-3 at an ionic strength of 0.046 mol dm-3. The corresponding thermodynamic solubility product, K,, = 2.82 x m0l3 dm-9. The value compares well with that reported by Talipov and Khadeev, 2.45 x m013 dm-9.11 In the crystal-growth experiments, the rate of reaction was calculated from the rate of addition of titrant solutions.It has been shown that errors introduced by the limited response time of the Orion fluoride electrode were negligible. The results of the crystal-growth experiments are summarized in table 1, in which 7&. and TF are the molar strontium and fluoride concentrations. Typical plots of the Table 1. Crystallization experiments at 25 OC, &.ITF = 0.5 T,r I seed ~/10-7 / 1 0-3 mol / 1 O-* mol mol dmP3 dm-3 ff amount/mg type min-' mP2 10, 11, 12a 20 21 16 18 19 17 107 109 106 111 110 108, 112a 23 29 31 33 35 39 40 49 50 52 54 55 57 59 60 63 65 3.75 3.50 3.30 3.20 3.00 2.70 2.50 3.75 3.50 3.30 3.00 2.75 2.50 1.60 1.80 2.00 2.20 2.30 2.40 2.60 2.80 3.00 1.90 2.00 2.30 . 2.60 2.90 3.30 3.50 3.70 6.86 6.74 6.74 6.59 6.49 6.37 6.24 6.86 6.74 6.64 6.49 6.37 6.24 15.0 15.0 15.0 15.0 15.0 15.0 15.0 15.0 15.0 6.80 6.80 6.80 6.80 6.80 6.80 6.80 6.80 1.81 1.63 1.49 1.42 1.27 1.09 0.9 I 1.81 1.63 1.49 1.27 1.09 0.9 1 0.09 0.22 0.36 0.49 0.56 0.63 0.76 0.90 1.03 0.44 0.52 0.74 0.97 1.20 1 S O 1.65 1.80 48.3 48.3 48.3 48.3 48.3 48.3 48.3 28 1 28 1 28 1 28 1 28 1 28 1 3120 1800 390 2000 780 780 2000 800 800 266 266 133 266 266 266 133 133 A A A A A A A B B B B B B C C C C C C C C C C C C C C C C C 41 30 20 18 11 7.2 4.0 75 51 34 18 14 5.3 0.0 1 0.03 0.10 0.18 0.25 0.45 0.64 0.83 1.15 0.13 0.21 0.38 0.70 2.20 4.10 6.19 8.74 a Effect of stirring rate (in r.p.m.) expt 10 (60); expt 11 (100); expt 12 (80); expt 108 (100); expt 112 (60 and 300); all others 100 r.p.m.220 CRYSTAL GROWTH FROM AQUEOUS SOLUTION amount of strontium fluoride precipitated as a function of time are shown in fig.1 . It can be seen that for given values of CT, the rate of growth is constant for at least 60 min, at which time the amount of precipitation corresponds to as much as 25% of the initial mass of inoculating seed. The rates of crystallization, R, in table 1 were calculated from the slopes of linear plots such as those in fig. 1 . It can be seen that 36 20 40 tlmin 60 Fig. 1. Crystal growth of strontium fluoride. Plots of amount precipitated against time. The numbers refer to the experiment numbers in table 1 . Expt 107, 109, 21 and 16, left-hand ordinate; expt 40 and 50, right-hand ordinate. R, normalized for the initial surface area of each inoculating seed, is constant, confirming that crystallization takes place on the seed crystals without additional nucleation or spontaneous precipitation. The rates of growth of strontium fluoride seed fall in the order seed B > seed A > seed C , suggesting different densities of growth sites on the crystal surfaces.During the reactions, the crystals (ca. 2pm in size) maintained their cubic morphology, as observed by scanning electron microscopy, and were present largely as aggregates ca. 10 pm in size. For many sparingly soluble salts M,A,, the rate of crystallization can be expressed by12 where K is a constant (= ks Kson’(a+b)), k is the rate constant for crystal growth, s is proportional to the number of growth sites available on the seed crystals, K,, is the solubility product at the ionic strength of the experiment and n is the apparent order of reaction.Analysis of the growth data for strontium fluoride is shown in fig. 2, in which -log R is plotted against -log CT. The slopes of the linear plots indicate a change in the value of n as the supersaturation of the solutions is reduced. In the concentration range 0.9 < CT < 2.0 the value of n is 3.5k0.1, while at lower supersaturation (0.09 < CT < 1.03) the apparent reaction order is 2.0 & 0.1. At all supersaturations, the growth rate is insensitive to changes in stirring rate, as can be seen from the results of experiments 10-12 in table 1. l4 hydrated monolayer15 and polynuclear/birth and spread.l6-l* In general, it has been shown13 that for the spiral growth and hydrated-monolayer models, a value of n M 2 would rate = R = d[M,A,]/dt = Kan (1) Crystallization models for solution growth include spiralR. A.BOCHNER, A. ABDUL-RAHMAN AND G. H. NANCOLLAS 22 1 be expected in eqn (1) while for polynuclear crystallization n > 2. The slopes of the linear plots in fig. 2 indicate spiral growth or dehydration mechanisms at lower supersaturation and a polynuclear crystallization at higher concentrations. Similar results have been obtained in the crystallization of other alkaline-earth flu~rides.'~ A typical plot of the specific surface area (s.s.a.) of the grown phases as a function of the extent of reaction is shown in fig. 3. It can be seen that the decrease in s.s.a. is more rapid than that calculated on the basis of an isometric three-dimensional crystal growth.The reduction in s.s.a. during the first 100% of crystal growth probably -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0 -log cr Fig. 2. Plots of -log R against -log 0 for the crystallization of strontium fluoride. Seed A: 0 ; seed B: V; seed C: a, 0 and 0. 0 and 0, Right-hand ordinate, all others left-hand ordinate. 20 16 - I M n E 12 9 > 2 8 4 500 1000 1500 2000 growth (%) Fig. 3. Plots of s.s.a. against growth with respect to original seed. Seed A, 10, 1 1 and 12 : 0 ; calculated values: 7. a = 1.810. Expt222 CRYSTAL GROWTH FROM AQUEOUS SOLUTION reflects crystal-lattice perfection and surface annealing processes. Indeed, the decrease in s.s.a. during this period over-compensates for the general increase in surface area accompanying macroscopic crystal growth. Scanning electron micrographs of the crystals grown to an extent of more than twenty times the mass of inoculating seed showed that although the crystals maintained their orthorhombic form, aggregation into larger particles occurred.These compensating effects of increasing and decreasing available surface area for growth appeared to account for the negligible change in overall surface area during the linear growth plots in fig. 1. Crystallization experiments made in the presence of phosphonate inhibitors, TENTMP (triethylenediaminetetramethylene phosphonic acid, Dequest 205 1, Mon- santo Chemical Co.) and NTA (nitrilotrimethylene phosphonic acid, Dequest 200 1, Monsanto Chemical Co.) are summarized in table 2. The marked reduction in rate Table 2. Crystal growth in the presence of phosphonates.T,, = OST, = 3.30 x mol dm-3, [KNO,] = 5.66 x mol dm-3, seed A expt no. concentration R/ 1 0-6 inhibitor (PPm) mol m i x 1 m-2 21 22 24 26 25 28 27 47 43 44 45 46 - NTMP NTMP NTMP NTMP NTMP NTMP TENTMP TENTMP TENTMP TENTMP TENTMP - 0.02 0.08 0.23 0.463 0.650 0.93 0.084 0.232 0.463 0.650 0.79 1 1.88 1.64 1.51 0.66 0.36 0.1 1 0 1.52 1 .oo 0.72 0.24 0.21 12 ' N E 7 ' 8 2 --. U m a J aJ a * .y .- .- 20 4 0 6 0 t /m in Fig. 4. Crystallization rate plots in the presence of phosphonate inhibitors.R. A. BOCHNER, A. ABDUL-RAHMAN AND G. H. NANCOLLAS 223 of crystallization is illustrated by the rate plots in fig. 4. In these experiments, the phosphonate additives were introduced into the strontium fluoride supersaturated solutions prior to inoculation with seed crystals.It is apparent that constant rates of growth were again observed for at least 60 min of reaction and that both NTMP and TENTMP were effective inhibitors of crystallization. Applying a simple Langmuir adsorption the influence of both inhibitors can be interpreted in terms of their selective adsorption at growth sites on the crystal surface. The decrease in crystallization rate can be related to the crystal surface area covered by the adsorbed inhibitor molecules. If R, and Ri are the rates of crystallization in the absence and presence of inhibitors, respectively, the Langmuir isotherm requires a linear relationship between the relative reduction in rate, Ro/(Ro-Ri), and the reciprocal of the inhibitor concentration.20 Fig. 5 confirms the applicability of this simple adsorption isotherm, with both phosphonates being approximately equally effective in reducing the rate of crystallization of strontium fluoride.6.0 '4 e l < 4.0 % v . 0 2.0 2.0 6.0 1 /[inhibitor] 10.0 Fig. 5. The influence of TENTMP (v) and NTMP (0) on the crystallization of strontium fluoride at 25 "C. We acknowledge financial support from the National Science Foundation in a University/Industry Cooperative research grant (no. CPE 8005345) between the State University of New York at Buffalo and Martin Marietta Laboratories. C . W. Davies and A. L. Jones, Discuss. Faraday SOC., 1949,5, 103. G. H. Nancollas and N . Purdie, Q. Rev. Chem. SOC., 1964, 18, 1. P. Koutsoukos, Z. Amjad, M. B. Tomson and G. H. Nancollas, J . Am. Chem. Soc., 1980,102, 1553. L. J. Shyu and G. H. Nancollas, Croat. Chem. Acta, 1980, 53, 281. A. J. Ellis, J . Chem. SOC., 1963, 4300. R. E. Connick and M. S. Tsao, J . Am. Chem. SOC., 1954,76, 5311. ' F. G. Gimbell and C. B. Monk, Trans. Faraday SOC., 1954, 50, 965. H. S. Harned and W. J. Hamer, J. Am. Chem. Soc., 1933,55, 2194. G. H. Nancollas, Interactions in Electrolyte Solutions (Elsevier, Amsterdam, 1966). lo C . W. Davies, Ion Association (Butterworths, London, 1960). l 1 Sh. T. Talipov and V. A. Khadeev, Zh. Obshch. Khim., 1950, 20, 783.224 CRYSTAL GROWTH FROM AQUEOUS SOLUTION l2 G. H. Nancollas, A h . Colloid Interface Sci., 1979, 10, 215. l3 A. E. Nielsen, Pure Appl. Chem., 1981, 53, 2025. l4 G. M. van Rosmalen, P. J. Dandey and W. G. J. Marchee, J. Cryst. Growth, 1981, 52, 801. l5 C. W. Davies and A. L. Jones, Trans. Faraday SOC., 1955,51, 812. l6 G. H. Gilmer and P. Bennema, J. Appl. Phys., 1972,43, 1347. l7 L. H. E. Madsen and R. J. Boistelle, J. Cryst. Growth, 1979, 46, 681. l9 G. H. Nancollas, R. A. Bochner, E. Liolios, L. J. Shyu, Y. Yoshikawa, J. P. Barone and D. Svrjek, 2o P. G. Koutsoukos, Z. Amjad and G. H. Nancollas, J. Colloid Interface Sci., 1981, 83, 599. W. B. Hillig, Acta Metall., 1966, 14, 1868. Am. Inst. Chem. Eng., Symp. Ser., 1982, 215, 26. (PAPER 3/921)

ISSN:0300-9599    

DOI:10.1039/F19848000217

出版商:RSC    

年代:1984

数据来源: RSC