J. Chem. SOC., Faraday Trans. 1, 1982, 78, 61-73 Controlled Wettability of Quartz Surfaces BY ROBERT N. LAMB AND D. NEIL FURLONG*? Colloid and Surface Chemistry Group, Department of Physical Chemistry, University of Melbourne, Parkville, Victoria 3052, Australia Received 4th November, 1980 The wettability of flat quartz-crystal surfaces has been assessed by measurement at 25 "C of contact angle at the water-vapour/water-drop/quartz-plate three-phase line. Plates pretreated by heating in vacuum gave angles, measured through the drop 1 min after removal from vacuum, of 0-44 " as pretreatment temperature was increased from 200 to 1000 "C. Fully hydroxylated, therefore hydrophilic, quartz surfaces are progressively rendered hydrophobic by mutual condensation of surface hydroxyls to form siloxane bridges.Hysteresis was at a maximum after heating at 700-800°C, indicating that maximum surface chemical heterogeneity was produced by heating in this temperature range. Plates methylated subsequent to heat treatment gave angles that were constant at ca. 80" up to treatment at 600 "C and that decreased from 80 to ca. 47 O as the pretreatment temperature was further increased to 1000 OC. This variation with temperature is consistent with a mechanism for methylation in which only non-hydrogen-bonded surface hydroxyl groups on quartz are reactive towards the methylating reagent. Contact angles on both heat-treated and methylated plates were observed to decrease following extended exposure to water vapour. The wettability of solid surfaces, in particular those of finely divided minerals, is central to collection processes such as froth flotation, selective agglomeration and transfer between bulk fluid phases.The complexity of such processes has led to much emphasis being put on the use of model solids1 in wetting studies in much the same way as model colloids (polymer latices, for example) have been used in studies of colloid Various forms of silica and silicate glasses have for some time retained their popularity as model substrates in wetting ~tudies.~ The hydrophilic/ hydrophobic nature of silica surfaces has often been investigated6 and it is well-known that silica, like many other inorganic oxides, owes its inherent hydrophilicity to surface hydroxyl groups, groups which can be removed by treatment at elevated temperature' or by chemical reaction.*Y9 Such treatments have often been assessed on high- surface-area silica powders or granular materials whose amorphous and porous nature sometimes leads to uncertainty in interpretation.The aim of the present study was to determine the wettability of well-defined macroscopic quartz surfaces via measurement of contact angle following outgassing at up to 1000 OC and/or chemical modification by reaction with trimethylchlorosilane. In particular, it was intended to examine how these two types of treatment could be systematically varied and combined so as to produce surfaces with a range of wettability. The study was also undertaken to increase understanding of the inter- pretation of the wettability of heterogeneous silica surfaces and to assess their value as model surfaces in wettability studies.t Currently at C.S.I.R.O. Division of Applied Organic Chemistry, G.P.O. Box 4331, Melbourne, Victoria 3001, Australia. 6162 CONTROLLED WETTABILITY OF SURFACES EXPERIMENTAL Laboratory-reagent trimethylchlorosilane (99 %) was used. All other organic liquids and all inorganic cleaning reagents used were analytical grade (A.R.) chemicals. Triply distilled water produced from an all-glass still in which the second stage contained alkaline permanganate was used. This water had a conductivity < 1.3 x R-'cm-l and a pH of 5.4f0.2 when equilibrated with air. High-purity nitrogen gas (99.99%) was used as supplied. FIG. 1 .-Heat-treatment/methylation apparatus : 1, thermometer; 2, distillation bulb; 3, glass-to-metal seal; 4, kel-F tap; 5, copper tubing; 6, vacuum line; 7, water-circulation cooling coils; 8, O-ring joint; 9, quartz reaction tube; 10, quartz plate-holder; 11, quartz plate; 12, cylindrical furnace.Natural transparent quartz crystals (Arkansas rock crystal, courtesy of Prof. D. W. Fuerstenau) were z-cut into plates of thickness 1 .O-1.5 mm using a diamond-cut saw. Each plate was mounted in wax on a microscope slide for ease of handling and the exposed face wet-polished on a rotating disc. The sequence of polishing media was carborundum powder, fine diamond paste and a series of cerium oxide powders (ca. 4, 3.2 and 2.8 pm average particle size). TheR. N. LAMB AND D. N. FURLONG 63 plates so polished showed no sign of scratches under an optical microscope at 150 times magnification.Following polishing, the plates were washed in warm hexane to remove the wax and then rinsed in warm ethanol and allowed to dry. Plates were then cleaned with ammoniacal hydrogen peroxide solution for ca. 4 min prior to rinsing with copious amounts of triply distilled water and drying under a jet of nitrogen. Plate cleanliness was assessed by the so-called 'steam test',10 in which water vapour was condensed onto the cleaned silica plate and the nature of the condensed film observed. Vig'l has shown that interference fringes occur during evaporation when the contact angle of water on silica is < 4 O. All cleaned plates used in this study gave excellent fringes during the steam test. The cleaning procedure described above, whilst being very efficient for the removal of contaminants, does not produce macroscopic etching of quartz surfaces.12 Glassware exposed to trimethylchlorosilane (TMCS) was cleaned by soaking in warm 30% KOH solutions for 30s.For all other surfaces the washing sequence chloroform, water, concentrated HNO,, water, concentrated NH, and water was used. All surfaces were blown dry with nitrogen. Plates were heated under vacuum ( < 0.13 N m-2) and/or treated with TMCS using the apparatus shown in fig. 1. The temperature of the quartz tube was raised at CQ. 1 O C mine', a rate sufficiently slow to prevent fracture of the quartz plate as it passed through the a-p transition temperature at ca. 570 OC. The plate was kept at the desired temperature for ca.24 h before being cooled to room temperature at ca. 1 "C min-l. The entire system was then purged with nitrogen. For experiments involving heat treatment only, the quartz plate was then transferred immediately to the contact-angle apparatus. For ' methylation' the plate was completely immersed for 15 min in a 10% by volume solution of TMCS in hexane. TMCS in hexane was distilled directly into the quartz reaction tube via a special kel-F tap. The direct distillation methylation procedure was used to minimise the possibility of polymeric material being deposited onto the quartz plates. Methylated plates were washed in hexane to remove excess TMCS, blown dry and transferred to the contact-angle apparatus. Contact angles (8) were measured at 25 f 2 OC by the sessile-drop technique with a system isolated from vibration.The cell for measurement was sealed from the laboratory atmosphere and contained a reservoir of water to aid in saturation of its volume and to restrict evaporation from the sessile drop. Advancing (8,) and receding (8,) contact angles were determined by adding and removing water, respectively, from the sessile drop using a microsyringe. Angles > 50 O were determined using the arc extension method of Neumann and Good1, to locate accurately the tangent to the drop at the line of solid/liquid/gas contact. Smaller angles were determined by fitting the photographed drop profile to the known shape of sessile drops. A modified version of a computer program developed by Maze and BurnetlO was used to perform this curve fitting and to calculate the contact angle from the drop shape.The methods of analysis used enabled contact angles > 5' to be determined from photographs to within lo. RESULTS AND DISCUSSION All contact angles were measured at the quartz-plate/water-drop/water-vapour three-phase contact line. Contact angles on a clean quartz plate for increasing residence time in the contact angle cell are given in fig. 2. The sessile drop used to assess contact angle was removed from the plate following each measurement. The plate, initially water-wet, remained so for a period > 20 h. At longer times both OA and 8, increased, most probably due to atmospheric contamination. It was therefore considered that 20 h represented the upper limit for time-dependent studies using the present experimental set-up.Contact angles are shown with an uncertainty of *2*. This represented the combined effects of drop asymmetry (eleft - Oright), surfacc non-reproducibility, varying drop position on a plate and parallax errors in drop photography. The latter error was estimated from the geometry of the experimental apparatus while the other errors were determined from repeated measurements on numerous quartz plates.64 CONTROLLED WETTABILITY OF SURFACES time/h FIG. 2.4ontamination test on quartz plates: 5, advancing contact angle; Q , receding contact angle. HEAT TREATMENT Contact angles (el) on quartz plates outgassed at temperatures from 200 to 1000 OC are given in fig. 3. These angles were measured ca. 1 min after the plates were removed from the outgassing/methylation cell, hence the superscript. One minute was the minimum total time required for transfer of a plate to the contact-angle cell, positioning of the sessile drop and for photographing advancing and receding angles.Some results of White15 for fused silica plates heated in oxygen are also shown in fig. 3. Although the contact angles reported by White appear to be in good agreement with the advancing angles determined in the present study, comparison is difficult because White gave no details of his methods or accuracy of measurement or if the angles he reported were advancing or receding or if, in fact, he observed hysteresis at all. Both the advancing and receding contact angles, O i and Ok in fig. 3, increased with increasing outgassing temperature, the largest increases occurring between 650 and 700 OC.Contact-angle hysteresis exhibited a maximum in the temperature range 700 to 800 OC (fig. 4). The increase in hydrophobicity of the quartz plates with outgassing temperature is believed due to the progressive removal of surface silanol groups by mutual condensation to form surface siloxane bridges. There is a good deal of reported evidence, for example from studies of interactions with water vapour16- l7 and heats of immersion,18 showing siloxane bridges to be inherently hydrophobic at a molecular scale relative to silanols. Detailed discussion of the nature of this molecular hydro- phobicity will therefore not be given here. Fig. 3 represents the only study in which the ' macroscopic' hydrophobicity of quartz surfaces (i.e.the hydrophobicity detected via contact-angle measurement) has been systematically studied as a function of treatment temperature. The changes in macroscopic hydrophobicity shown in fig. 3 have resulted from changes in the molecular nature of the quartz plates, in particular the removal of surface silanol groups. However, because of the relatively low number of surface groups present on a macroscopic quartz-crystal surface, it was not possible in the present study to perform analyses of surface populations of silanol and siloxane groups after heat treatment at various temperatures. In order to relate the contact-angleR. N. LAMB AND D. N. FURLONG I FIG. 3 . 4 o n t a c t 200 LOO 600 800 1000 outgassing temperature/OC 65 , receding (Bk); (---) data of 200 Loo 600 800 loo0 outgassing ternperature/"C FIG. 4.-Contact-angle hysteresis on heat-treated quartz plates.66 CONTROLLED WETTABILITY OF SURFACES data to the population of surface groups on silica it is necessary therefore to use population data reported for silica powders.Twenty different studies on silica powders have recently been summarized by Knozinger' in the form of a plot of surface number-density of silanol groups as a function of outgassing temperature over the range 0-1000 O C . The data plotted in this manner show considerable scatter due, no 1.0- 0.9 - 0.8 - $ 0.7- h U m L-( - 5 0.6- .* k .d 2 0.5- & c1 u 0.4 - (d 0.3 - 0.2 - outgassing temperature/'C FIG. 5.-Calculated silanol populations on silica surfaces: 1111, summary of literature data on silica powders; (a) from equilibrium contact angles on heat-treated quartz plates; (b) from equilibrium contact angles on heat-treated/methylated quartz plates.doubt, to variations in the origin, pretreatment, porosity and cleanliness of the materials studied, as well as the relative precisions of the various spectroscopic and thermogravimetric techniques used. As rigorous discussion of the wettability of heterogeneous surfaces is based upon surface-area heter~geneity,'~ the silanol number- density data summarized by Knozinger have been converted to 'area fraction of silanols' and, as such, are shown as a band in fig. 5 . For the conversion, it was assumed that one siloxane bridge occupied the area of two silanols, an assumption consistent with the formation of siloxane bridges via the mutual condensation of silanols. The scatter in the area fraction plot (fig.5) is small and indicates that the variations in the reported silanol number density/temperature data stem in the main from variations in the total number of hydroxyls on a fully hydrated surface.2o The data band in fig. 5 predicts that silica surfaces outgassed at 1000 *C should beR. N. LAMB AND D. N. FURLONG 67 chemically homogeneous on a molecular scale, in that they consist almost entirely of siloxane bridges. As it is customary to consider the major causes of contact-angle hysteresis to be surface roughness and chemical heterogeneity,21 the degree of hysteresis observed (fig. 4) following outgassing at 1000 OC, ca.5 O, can therefore be attributed to surface roughness alone. This level of hysteresis is quite sma1119 and is a good indication that the polishing techniques used were successful in producing essentially smooth surfaces. The data band in fig. 5 also predicts that the molecular surface chemical heterogeneity passes through a maximum as the outgassing temperature is raised from 200 to 1000 OC ; i.e. with increasing temperature the surface changes from one consisting entirely of silanols to one with a mixture of silanols and siloxanes to one dominated by siloxane bridges. The general form of the contact-angle hysteresis (fig. 4) is consistent with this prediction. However, maximum contact-angle hysteresis occurred at 700-800 O C , compared with the predicted maximum in surface chemical heterogeneity ‘ seen ’ in contact-angle studies varied with outgassing temperature in a similar manner to powdered silicas previously studied, the nature of the surface at any given temperature was not as previously determined.It has been proposed by Cassie22 that the equilibrium contact angle (8,) on a heterogeneous surface consisting of microscopic patches of two homogeneous com- ponents can be related to the composition of that surface as [a(sioH) - a(si-o--si)] at 400 O C . It therefore appears that, whilst the quartz surface cos 6E = a1 cos 8 E l + a2 cos 8E2 (1) where a, and a, are the fractions of components 1 and 2, respectively, and 8E1 and 8,2 the equilibrium contact angles on homogeneous surfaces of components 1 and 2.Although application of eqn (1) is sometimes limited,lW it has been used to describe the nature of surfaces such as those of interest in the present study, i.e. to derive a surface population profile of a mixed surface (al,a2) from observed contact angles (8E,8El,8E2). The quartz plates of fig. 3 were bifunctional from a wettability point of view, consisting of silanol groups of various typess for which &(SiOH) = &@OH) = 8l,(SiOH) = 0’ and siloxane bridges. In the present study of treated quartz surfaces, advancing and receding, rather than ‘equilibrium’, contact angles were measured. Thus in order to use eqn (1) it is necessary to evaluate an equilibrium angle from the measured advancing and receding angles. Wolfram and F a u ~ t ~ ~ have recently provided experi- mental and simple theoretical justification for the use of the equation for systems wherecontact-angle hysteresis is observed. r is the Wenzel surface-roughness which can be set equal to 1 1 9 7 25 when hysteresis due to surface roughness alone is < loo, as concluded above for the quartz plates used in the experiments of fig.3. Eqn (2) can simply be regarded as a means of averaging 8 A and 8,; the key to its use in combination with eqn (1) is the identification of OY, the angle defined by Young’s equation, with the equilibrium contact angle. Johnson and Dettre19 have proposed a model to describe contact-angle hysteresis on heterogeneous surfaces and show how 8, is related to BAand 8, for some surface types. The essential point of their theory is that many metastable angles exist for heterogeneous surfaces, the energy barriers between states becoming smaller the smaller the size of surface heterogeneities and68 CONTROLLED WETTABILITY OF SURFACES the more random their distribution.Dettre and Johnson have advancing and receding angles on variously coated titanium dioxide surfaces for which OEl = 54' and OEz = 0"; calculation of 6, using eqn (2) shows it to be within 10' over a wide range of surface coverage, with 6, calculated using eqn (1). Assuming the quartz plates of the present study to be completely dehydroxylated during outgassing at 1000 O C gives &(Si-O-Si) = 44' and 6k(Si-0-Si) = 39O.These values are similar to the OE2 value of Johnson and Dettre. The system of variously dehydroxylated quartz surfaces has then similar OEl and 8,, values to those of the experimental system of Dettre and and we have therefore used eqn (2) in conjunction with the data of fig.3 to calculate Ol, at the various outgassing temperatures used. Eqn (1) can be rewritten for these experiments as (3) cos el, - 0.743 0.257 a ( S i O d = TABLE 1 .-CONTACT-ANGLE-TIME STUDIES ON HEAT-TREATED QUARTZ outgassing 9,io wo temperature 1°C 1 min 2.5 h 10 h 1 min 2.5 h 10 h ~~ 625 20 20 20 0 3 5 865 35 26 26 21 18 17 1000 44 44 42 39 40 38 and used to calculate the variation of surface silanol area coverage with outgassing temperature from the contact-angle data of fig. 3. The resultant curve (a) in fig. 5 shows that the quartz plates contained a significantly higher density of surface silanols at all outgassing temperatures when compared with the silica powders studied previously.Curve 5 (a) also indicates that maximum chemical heterogeneity occurred at ca. 800 O C , in good agreement with the occurrence of maximum contact-angle hysteresis at 700-800 O C (fig. 4). It therefore seems that the nature of the quartz plates as seen by contact-angle measurements is well-represented by curve (a) in fig. 5, despite the approximations used in its derivation. Note that the contact-angle data of White15 would also correspond to a surface more hydroxylated at any outgassing temperature than would be predicted from studies on high-surface-area silicas. It is possible that flat silica surfaces, the quartz plates of the present study or the fused discs used by White, do in fact respond differently to heat treatment than do silica powders. It is more likely, however, that in some instances rehydroxylation has occurred in the time interval between heating and measurement of the contact angle. The contact angles on quartz plates in fig.3 were measured at between 1 min, for advancing angles, and 1.5 min, for receding angles, after removal from vacuum and after ca. 30-60 s, respectively, of water-drop residence. Some examples of the subsequent stability of the contact angle on exposure to near-saturated water vapour are given in table 1. Following treatment at 625 O C (and for all temperatures below) no changes were detected in 6, during 10 h of ageing, whilst 6, increased slightly. By contrast, both 6, and 6, decreased significantly during the first 2.5 h of ageing of the 865 OC treated plate.The initial and final values for OA on this plate were within experimental error of those given recently by Pashley and KitcheneP for a plate heated at 875 O C and subsequently aged. No significant changes in 6, or 6, were detected during 10 h ageing of a plate heated in vacuum at 1000 O C .R. N. LAMB A N D D. N. FURLONG 69 It therefore appears that for the quartz plates heated at up to 625 O C considerable rehydroxylation occurred before and/or during the measurement of contact angles. At higher temperatures less rehydroxylation occurred. Such a temperature dependence for the rate of rehydroxylation is consistent with many previous studies on high- surface-area silicas.l79 18* 27 Thus the maximum in chemical heterogeneity observed at 700-800 O C in contact-angle studies represents a surface that at the time ofmeasurement had rehydroxylated to a 1 : 1 silanol: siloxane area ratio from a surface immediately after heat treatment that had a lower area ratio.outgassing temperaturelo C FIG. 6.-Summary of the controlled wettability of quartz: $ 0 , heat-treated quartz; I) @, heat-treated/ methylated quartz. Round symbols, advancing contact angle; square symbols, receding contact angle. HEAT TREATMENT/METHYLATION Contact angles on quartz plates methylated subsequent to outgassing at temperatures from 250 to 1000 O C are given in the upper half of fig. 6. Both advancing and receding angles were constant up to 400 OC and decreased above 700 OC.Both the advancing and receding angles on quartz methylated subsequent to heat treatment at 1000 "C were only 6" greater than on surfaces subjected to the heat treatment alone. This provides confirmation of the assumption used above in application of the Cassie equation to contact angles on heat treated quartz, viz. surfaces treated at 1000 O C were very sparsely populated with surface silanol groups [qSiOH) = 0.0251.70 CONTROLLED WETTABILITY OF SURFACES In an attempt to relate this behaviour to that previously reported, reference can be made to numerous previous studies26*28-33 in which estimates of the air/water contact angle on silica methylated using TMCS have been made (table 2). Often, TABLE 2.-sUMMARY OF CONTACT-ANGLE DATA ON METHYLATED SILICA material method contact angle/O ref.vitreous silica glass microscope vitreous silica Cabosil MY platea slideb platea Aerosil 200a ground quartza quartz crystala quartz crystal captive bubble sessile drop sessile drop ( ?) sessile drop on pressed disc calculation from coverage rise up a packed bed captive bubble sessile drop 70-75 28 8, x 85 eR x o 8, 95- 100 8, 80-90 - 130 29 30 31 70 32 75 33 (x 10% coverage) 8, 80 hysteresis x 5 26 8,88 current work 8,72 a Liquid-phase methylation of Laskowski and Kitchener ; refluxing with liquid chlorosilane. vapour-phase methylation; however, the measurement of contact angle was ancillary to the main aim of each study, which might have been investigation of film thickness,26* 30 water or immersion and partition beha~iour,~~ and as a result not well-described.Published contact angles for various types of methylated silica (without heat treatment) fall between 70 and 1 30°, the one exception being a zero receding angle given by Herzberg et 4Lz9 These workers determined OR using an evaporating sessile drop and their data must therefore be suspect. Three previous 2 9 9 30 in which an advancing angle was measured gave an average value of 88 f go, in excellent agreement with the present value for non-heat-treated quartz. Only Blake and Kitchener30 seem to have also measured a receding angle; their value of 80-90° is higher than that reported here. One study31 in which methylation was achieved by 'refluxing in liquid chlorosilane' reported a contact angle some 40° greater than other studies.It can only be supposed that such treatment resulted in the deposition of polymeric material, particularly as abnormally high surface densities of trimethyl silyl groups were also found. Laskowski and KitchenerZ8 have reported a drop in contact angle from ca. 50' to ca. 40' when a fused-silica disc was methylated following treatment at room temperature and 450 O C , respectively. However, drying conditions were not specified in both experiments and hence their change cannot be truly compared with the results of the present study. It has been p r ~ p o s e d ~ ? ~ ~ that reaction of TMCS with fully hydroxylated silica surfaces proceeds via non-hydrogen-bonded hydroxyl groups. Therefore only removal by outgassing of these isolated and geminal hydroxyls, believed to occur above ca.500 OC, should influence the subsequent population of surface methyl groups. ThisR. N. LAMB AND D. N. FURLONG 71 is confirmed by the contact-angle data in fig. 6. If it is proposed, therefore, that the quartz surfaces heated at above 500 OC and then methylated consist only of siloxane bridges (formed by removal of vicinal hydroxyls below 500 "C) and trimethyl silyl groups, eqn (3) can, for these surfaces, be written as where 8, (Si-0-Si) is taken as 42' (fig. 3), the equilibrium contact angle on a surface consisting only of methyl groups as 1 and where a(SiOH) corresponds to the area density of silanols prior to methylation. The resultant curve of a(SiOH), calculated from the methylation contact-angle data of fig. 6 in conjunction with eqn (2), is included in fig.5 [curve (b)]. Curve 5(b) shows that the maximum area coverage of the quartz surface with trimethyl silyl groups was 54 4 1 %, a coverage in excellent agreement with that calculated by Laskowski and Kitchener2* using Kiselev's area36 of 0.42 nm2 per -Si(CH,), group and assuming reaction with only isolated and geminal silanol groups. Such good agreement indicates that the Cassie treatment leading to eqn (4) does indeed provide a realistic picture of the composition of mixed siloxane-trimethyl silyl surfaces. Curve 5 (b) indicates that heat-treated/methylated quartz surfaces were ca. 1 : 1 in siloxane: trimethyl silyl groups at ca. 500 'C, whilst at higher temperatures they were more heavily populated by siloxane bridges. This prediction of decreasing chemical heterogeneity as the outgassing temperature was increased above 500 OC is confirmed qualitatively by the observed hysteresis in contact angles, although the experimental trend in hysteresis was not very clear. Quartz plates heated at below 500 OC and then methylated will contain siloxane bridges and trimethyl silyl groups as well as residual hydroxyls.There are, however, insufficient contact-angle data in this temperature region (fig. 6) to justify any attempt to describe the degree of surface heterogeneity. The variation with time of contact angles on methylated plates, shown in table 3, TABLE 3 .<ONTACT-ANGLE-TIME STUDIES ON HEAT-TREATEDIMETHYLATED QUARTZ outgassing W0 wo temperature 1°C 1 min 2.5 h 10 h 1 min 2.5 h 10 h ~~ ~~ 82 1 70 63 58 48 39 35 922 55 44 42 48 41 38 indicates that the surfaces became more hydrophilic following exposure to water vapour.Whether such surfaces eventually become fully hydrophilic, as has been previously ~uggested,~~ has not been evaluated in the present study. CONCLUSIONS (1) Quartz plates heated in vacuum at temperatures between 200 and 1000 'C exhibited contact angles that increased progressively from 0 to 44O with increasing temperature. Methylation subsequent to heat treatment gave angles that decreased from ca. 80 to ca. 47'. Therefore, by combination of heat treatment and methylation, quartz can be made to exhibit a wide range of wettability (fig. 6). (2) The wettability of treated quartz surfaces was explained in terms of the relative hydrophobicities of surface silanol, siloxane and trimethyl silyl groups.Analysis of72 CONTROLLED WETTABILITY OF SURFACES contact angles enabled the construction of surface chemical heterogeneity profiles following various treatments, profiles which were in qualitative agreement with observed contact -angle hysteresis, (3) Some contact angles decreased following exposure to water vapour, indicating that some quartz plates had undergone rehydroxylation subsequent to treatment but prior to contact-angle measurement. For heat-treated plates this was also indicated following a comparison of the calculated silanol population with those of previous studies of silica. Rehydroxylation reduces the accessible range of wettability available for such treated surfaces and may limit their usefulness as model surfaces in wettability studies.The authors acknowledge support from the Australian Research Grants Committee (ARGC) and the National Energy Research, Development and Demonstration Council (NERDDC) of the Department of National Development, Commonwealth of Australia. We thank Prof. T. W. Healy for many valuable discussions and for constructive criticism of this work. J. M. Haynes, Wetting, Spreading and Adhesion, ed. J. F. Padday (Academic Press, London, 1978), p. 469. A. Kotera, K. Furusawa and K. Kuoto, Kolloid 2. 2. Polym., 1970, 240, 837, T. W. Healy, A. Homola, R. 0. James and R. J. Hunter, Polymer Colloids 11, ed. R. M. Fitch (Plenum Press, New York, 1980), p. 527. H. Sasaki, E. Matijevic and E. Barouch, J. Colloid Interface Sci., 1980, 76, 319.J. F. Padday, Wetting, Spreading and Adhesion, ed. J. F. Padday (Academic Press, London, 1978), p. 464. I3 R. K. Iler, The Chemistry of Silica (Wiley-Interscience, New York, 1979), chap. 6, pp. 622-729. ' H. Knozinger, The Hydrogen Bond, ed. P. Schuster, G. Zundel and C. Sanderfy (North Holland, Amsterdam, 1976), vol. 3, chap. 27, p. 1270. M. L. Hair and W. Hertl, J. Phys. Chem., 1969, 73, 2372. C. G. Armistead and J. A. Hockey, Trans., Faraday Soc., 1967, 63, 2549. lo M. L. White, Proc. Annu. Freq. Control. Symp., 1973, 27, 79. l1 J. R. Vig, J. W. Lebus and F. L. Filler, Proc. Annu. Freq. Control Symp., 1975, 29, 220. l2 R. M. Pashley, PhD Thesis (Imperial College, London, 1978), chap. 5, pp. 94-104. l3 A. W. Neumann and R. J. Good, Surface and Colloid Science, ed.R. J. Good and R. R. Stromberg l4 C. Maze and G. Burnet, Surf. Sci., 1969, 13, 451. lF, M. L. White, Clean Surfaces, ed. G. Goldfinger (Marcel Dekker, New York, 1970), chap. 18, pp. l6 G. J. Young, J. Colloid Sci., 1958, 13, 67. (Plenum Press, New York, 1979), vol. 1 1 , chap. 2, p. 36. 36 1-373. K. Klier and A. C. Zettlemoyer, J. Colloid Interface Sci., 1977, 58, 216. G. J. Young and T. P. Bursh, J. Colloid Sci., 1960, 15, 361. New York, 1969), chap. 2, pp. 85-153. lB R. E. Johnson and R. H. Dettre, Surface and Colloid Science, ed. E. Matijevic and F. Eirich (Wiley, 2o D. E. Yates, F. Grieser, R. Cooper and T. W. Healy, Aust. J. Chem., 1977, 30, 1655. 21 R. J. Good, Surface and Colloid Science, ed. R. J. Good and R. R. Stromberg (Plenum Press, New 22 A. B. D. Cassie, Discuss. Faraday SOC., 1948, 3, 11. 23 E. Wolfram and R. Faust, Wetting, Spreading and Adhesion, ed. J. F. Padday (Academic Press, 24 R. N. Wenzel, J. Phys. Chem., 1949, 53, 1466. 25 N. K. Adam and G. Jessop, J. Chem. Soc., 1925, 1863. 26 R. M. Pashley and J. A. Kitchener, J. Colloid Interface Sci., 1979, 71, 491. 27 R. H. Dettre and R. E. Johnson, J. Phys. Chem., 1965, 69, 1507. 28 J. Laskowski and J. A. Kitchener, J. Colloid Interface Sci., 1969, 29, 670. 2B W. J. Herzberg, J. E. Marian and T. Vermeulen, J. Colloid Interface Sci., 1970, 33, 164. T. D. Blake and J. A. Kitchener, Trans. Faraday Soc., 1972, 68, 1435. 31 W. T. Yen, R. S. Chahal and T. Salman, Can. Metall. Q., 1973, 12, 231. York, 1979), vol. 1 1 , chap. 1, pp. 10-13. London, 1978), chap. 10, p. 213.R. N. LAMB AND D. N. FURLONG 73 32 L. Alzamora, S. Contreras and J. Cortes, J . Colloid Interface Sci., 1975, 50, 503. 33 S. Garhsva, S. Contreras and J. Goldfarb, Colloid Polym. Sci., 1978, 256, 241. 34 V. Ya. Davydov, A. V. Kiselev and L. T. Zhuravlev, Trans. Faraday SOC., 1964, 60, 2254. 36 N. K. Adam, A h . Chem. Ser., 1964,43, 52. 38 A. V. Kiselev, N. V. Kovaleva, A. Ya. Korolev and K. A. Shcherbakova, Dokl. Akad. Nauk SSSR, 1959, 124, 617. (PAPER O / 1682)