The forces between mica surfaces in ammonium chloride solutions

The Forces between Mica Surfaces in Ammonium Chloride Solutions W. A. D U C K E R *'l AND R. M. PASHLEY? *Department of Applied Mathematics, Research School of Physical Sciences, and tDepartment of Chemistry, The Faculties, Australian National University, Canberra, ACT2600, Australia

Received June 16, 1988;accepted September 30, 1988 The total interaction forces between two molecularlysmooth mica crystals have been measured in a range of aqueous NH4C1 solutions. The NH~-ion was found to adsorb onto the mica basal plane with a binding energysimilar to that of the original lattice ion, K +. The chargingproperties at the mica/water interface were analyzed using a model previously developed by one of us which gives an estimated exclusion area of the adsorbed NH~ ion of about 0.52 nm 2. At low concentrationsthe measured interaction forceswere, at all distances, found to be in goodagreement with DLVOtheory, whereas at concentrations above about 10-3 M, a short-range (<1.5 rim) repulsion was measured. This was oscillatory with a periodicityofabout 0.4 nm. This short-rangesolvationforcewas mostlikelycausedby a combination of the hydrated state of the adsorbed NH~-ion and the molecular packing properties of water in thin films.

© 1989 Academic Press, Inc.

INTRODUCTION The basal plane surface of muscovite mica is negatively charged in water because of the dissociation of surface K + ions. However, when electrolyte is added other cations may adsorb at the surface thereby altering the surface charge and other surface properties. The ion-exchange behavior of mica in the presence of alkali and alkali-earth metals is now quite well understood (1, 2). If the size of the ion is included, then a simple ion-exchange model allowing competitive adsorption of metal ion and proton can account for the surface dissociation and charging properties. The hydration of the mica surface in the presence of adsorbing alkali and alkali-earth metals has also been studied. Short-range repulsive forces in addition to those predicted by DLVO theory have been observed and have been attributed to water structure (1, 3). However, when protons adsorb to the same surface no hydration force is measured, probably because some unique chemical bonding ' To whom correspondenceshould be addressed.

of the proton at or beneath the mica surface prevents the proton from perturbing the water structure. This observation, together with the fact that the onset of hydration forces occurs over a narrow concentration range, led Pashley ( 1 ) to propose that the hydration force is determined by the extent to which hydrated ions remain on the surface of mica as it is forced against another surface and thus on the relative bulk concentrations of protons and other cations (4). The extent of proton binding and its exchange are thus important in determining both the charge and the hydration properties of the surface. The adsorption of acidic cations is therefore of interest because there is the possibility of the dissociation of the acid on the surface to leave a bound proton and a neutral group in solution. This allows the possibility of ion exchange simply by the transfer of a neutral molecule from the surface to solution. This paper reports an investigation of the hydration and ion-exchange properties of such an ion, N H +, using the surface forces apparatus developed at the Australian National University (5).

433 0021-9797/89 $3.00 Journal of Colloid and Interface Science, Vol. 131, No. 2, September 1989

Copyright © 1989 by Academic Press, Inc. All rights of reproduction in any form reserved.

434

DUCKER AND PASHLEY

P r e v i o u s e x p e r i m e n t s have s h o w n t h a t the h y d r a t i o n force can be a c o m p l i c a t e d function o f distance. W h e n m o n a t o m i c cations r e m a i n a d s o r b e d to the surface o f m i c a the resultant repulsive force is a n oscillatory f u n c t i o n o f distance with the a m p l i t u d e o f each oscillation increasing as the two surfaces a p p r o a c h contact ( 6 ) . This t y p e o f force has b e e n m e a s u r e d in t h i n films o f o t h e r liquids b e t w e e n m i c a surfaces a n d is t h o u g h t to be a result o f the finite size o f the liquid molecules. However, in highly h y d r o g e n - b o n d e d liquids such as water, these oscillations are n o t c e n t e r e d o n zero force or the e x p e c t e d v a n d e r W a a l s force b u t o n a large repulsive force, which is t h o u g h t to be caused b y residual h y d r a t i o n o f the ions a d s o r b e d to the surface. N o oscillatory forces are m e a s u r e d b e t w e e n lipid bilayers w h e n adsorbed to m i c a ( 7 ) o r in l a m e l l a r phase ( 8 ) , p r e s u m a b l y b e c a u s e the surface is sufficiently r o u g h for the oscillatory structure to be averaged out. T h e forces b e t w e e n surfaces cont a i n i n g a d s o r b e d a m m o n i u m ions is thus o f interest b o t h because the i o n is p o l y a t o m i c a n d because the i o n can h y d r o g e n b o n d with the solvent.

U H Q w a t e r purification u n i t with the reverse o s m o s i s u n i t disconnected. Finally, the water was d e a e r a t e d to p r e v e n t b u b b l e f o r m a t i o n in the stainless-steel a p p a r a t u s . T h e resulting water h a d a c o n d u c t i v i t y o f 0.4/~S cm -1 i m m e d i a t e l y after p r e p a r a t i o n a n d a p H in the range 5.5-5.7 after e q u i l i b r a t i o n with a t m o spheric c a r b o n dioxide. A n a l y t i c a l - g r a d e NH4CI f r o m A J A X was used w i t h o u t further treatment. RESULTS AND ANALYSIS

Forces at Various NH-~ Concentrations T h e forces b e t w e e n two m i c a surfaces i m m e r s e d in a range o f NH4C1 solutions at 21 °C are given in Figs. 1 to 4. In each figure t h e solid line is t h e best fit o f the e x p e r i m e n t a l d a t a b y D L V O theory. T h i s was c a l c u l a t e d using a n exact n u m e r i c a l s o l u t i o n o f the Poisson-Boltzmann equation (9) and a nonret a r d e d H a m a k e r a t t r a c t i o n given b y t h e equation (10)

METHOD AND MATERIALS

Forces between the surfaces of mica were measured using the method developed by Israelachvili and Adams (5). In this technique two thin mica sheets, glued to cylindrical silica support disks and oriented so that the axes of the cylinders are orthogonal, are pushed together under a force (between about 0.01 N and 0.1/~N) which is measured by the deflection of a weak double-cantilever spring. The distance between the surfaces is measured by analysis of interference fringes (FECO) produced by multiple reflection from the silvered back surface of the mica sheet. This technique allows the distance between the surfaces to be measured with an accuracy of about 0.05 nm under the best conditions. Water used in all experiments was passed through an ion-exchange resin and then distilled once. It was then processed in an Elga Journal of Colloid andlnterface Science, Vol. 131, No. 2, September 1989

lC .50

160 Dlnm

1'50

FIG. 1. Forces between mica surfaces in dilute NI4_4C1 solutions. Note that the forces have been scaled by the radius of curvature of the surfaces and so are proportional to the energy/unit area between equivalent fiat surfaces. The solid circles represent measurements at 1.0 × 10-5 M and the open circles measurements at 1.1 X 10-4 M. The solid lines are the best fits to DLVO theory with exact numerical solutions of the Poisson-Boltzmann equation at constant charge and a nonretarded van der Waals contribution with a Hamaker constant of 2.2 × 10 -20 J and demonstrate that the surfaces interact almost at constant charge. At 1.1 X 10-4 M the surfaces interact with a surface potential at an infinite separation of-90 mV, an effective surface charge of -0.0038 C m-2, and a Debye length of 26 nm. At 10-s M the corresponding values are -100 mV, -0.0013 cm -2, and 96 nm.

FORCES BETWEEN MICA SURFACES

io~I3×163M

o 1

435

°

,

10

20

~ 3

"10

1'0

30

2'o

FIG. 2. Forces between mica surfaces in 1.1 X 10 -3 M

NH4CIsolution. The open and solidcirclesrepresentrepeat measurements. The solid line represents the best fit to DLVO theory with the constant charge boundary condition with a surface potential o f - 8 0 mV, a surface charge of -0.0093 C m-Z, and a Debye length of 8.7 nm. At separations less than about 2 nm the strongly repulsive hydration force causes the surfaces to move beyond the DLVO limit at constant charge. This can be seen in the inset where the constant charge and constant potential interactions (together with a van der Waals contribution) are shown by the upper and lower lines, respectively.The two points at 1.5 nm and 103, 1.5 × 103 uNm -1 indicate the presence of a secondary minimum.

FIO. 3. Forces between mica surfaces in 3 X l0 -3 M NH4CI solutions. The open and solid circles represent repeat measurements. The solid line is the best fit as ex-

plained in the legend to Fig. 1, with a surface potential of -82 mV, a surface charge of-0.015 C m-z, and a Debye length of 5.5 nm.

crossed cylinder geometry this value is equal to 27r times the c o r r e s p o n d i n g energy per u n i t area between flat surfaces ( 1 2 ) . F o r the 10-5 M solution, the theoretical curve was calcu-

•

Fvdw/R

-

3'0

D/nm

D/nm

-A 6D 2 ,

r

o

•

-2

1"2x10 M

o

where D is the distance between the mica surfaces, R is the (geometric) m e a n radius of the crossed cylinders a n d the H a m a k e r constant, A, has a value o f 2.2 × 10 -2o J for the m i c a / w a t e r / m i c a system (5). The m a g n i t u d e o f the double-layer repulsion is d e t e r m i n e d b y the surface potential a n d the decay o f these forces by the D e b y e length. T h e double-layer equat i o n c a n be solved for various b o u n d a r y conditions, with all cases lying between the u p p e r l i m i t o f c o n s t a n t surface charge a n d the lower l i m i t o f c o n s t a n t surface potential ( 11 ). Each set of measured points lies close to the c o n s t a n t charge case so o n l y this limit has b e e n shown. Figure 1 shows the i n t e r a c t i o n i n 10 -s a n d 1.1 × 10-4 M NH4C1 solutions. T h e plotted forces are scaled b y the radius because i n the

vE EL 10

102

;

;o

1'5

D/nm

FIG. 4. Forces between mica surfaces in 1.2 X 10-2 M NH4C1 solutions. The open and solid circlesrepresent repeat measurements. The solid line is the best fit as explained in the legend to Fig. 1, with a surface potential of -65 mV, a surface charge of -0.022 C m-2, and a Debye length of 2.7 nm. Journal of Colloid and InterfaceScience, Vol. 131, No. 2, September 1989

436

D U C K E R AND PASHLEY

lated with a surface potential of - 1 0 0 mV, a surface charge of -0.0013 C m-2, and a Debye length of 96 n m (as expected at this concentration). In the 1.1 × 10 -4 M solution, the D L V O curve was calculated using a surface potential o f - 9 0 mV, a surface charge of - 0 . 0 0 3 8 C m -z, and a Debye length of 26 n m (compared to the value of 29 n m expected for this concentration). In both cases, the forces c o u l d not be measured beyond the force maxi m u m because of the nature of the measurem e n t technique. When the slope of the attractive force becomes greater than the spring constant, the system is unstable so the surfaces j u m p to the next stable position (in this case primary m i n i m u m contact at D = 0). For the relatively weak spring used in this experiment (spring constant = 148 N m - l ) , the j u m p starts roughly at the position of m a x i m u m force. In the 10-5 M solution, the last measurement before the j u m p was made at 4.2 n m separation, which is close to the predicted j u m p distance of 3.7 nm, and in the 1.1 X 10 -4 M solution, where the surfaces would j u m p from 2.2 n m apart, the last measurement was made at 3.3 nm. The interfacial energy for the m i c a / a m m o n i u m solution was also measured at these two concentrations and was found to be close to that measured for mica in water. At 1.1 N 10 -3 MNH4C1 (Fig. 2), the measured forces agreed well with those predicted by D L V O theory for a surface potential, G0, of - 8 2 mV, a surface charge, a0, of - 9 . 3 × 10 -3 C m -2, and a Debye length, K -~, of 8.7 n m (calculated 9.2 n m ) at distances greater than about 1.5 n m separation. Below this distance the measured force was greater than that predicted by D L V O theory, indicating an additional repulsive force. This force was large enough to prevent the surfaces from being brought into contact because the spring used for the measurement was too weak. This type of force has been identified as a hydration force or, more precisely, a secondary hydration force (14). When the applied force was reduced, the surfaces j u m p e d apart only a small distance (about 10 rim) which, together with the fact that the surfaces were not in contact, indicated Journal of Colloid and Interface Science, VoL 131,No. 2, September1989

the presence of a weak hydration m i n i m u m ( 1 ). This region of the force curve is discussed in more detail later. When the concentration was increased to 3 X 10 -3 M , the measured forces still agreed well with D L V O theory (G0 = - 8 2 mV, ~0 = - 0 . 0 1 5 C m -2, K -1 = 5.5 n m ( K -1 calculated = 5.6 n m ) ) up to about 1 n m separation; then again the additional hydration force prevented the surfaces from jumping into adhesive contact. However, despite the threefold increase in concentration from the previous measurement, the fitted potential did not decrease. The significance of this is discussed in the next section. Figure 4 shows the results for 1.2 × 10 _2 M NH4C1, where once again the forces agree well with D L V O theory (Go = - 6 5 mV, a0 = - 0 . 0 2 2 C m -2, K -1 = 2.7 n m ( K -1 calculated = 2.8 n m ) ) until the hydration force becomes significant.

Mass Action Model of Ion Binding The surface potentials obtained from the theoretical fits in Figs. 1 to 4 are summarized in Fig. 5. Three theoretical curves are also shown in this figure. The dashed curve was calculated from a model in which the only potential determining ion is the proton, and thus the a m m o n i u m ion influences the system only

~.-~ a

.

bulk NI44conc/Moles/I FIG. 5. The surface potentials obtained in Figs. 1 to 4 are plotted here as a function of concentration. The solid line was calculated using the model described in the text with the following parameters: An = 0.52 nm2; An = 0.48 nm2; PKN = 3.12, and pH 5.4 and the intermittent line

with the same binding constant, but with AN = An = 0.48 nm 2. The dashed line was calculated by assuming that there is no binding of NH~(KN infinite), only electrolyte screening.

FORCES B E T W E E N

by screening the electrical interaction. When this model was extended to also allow the amm o n i u m ion to bind to the mica surface in competition with the proton, the fit to the experiment points was also poor (dotted and dashed line). However, as observed previously for Li +, Na +, and K +, a good fit to the experimentally measured potentials can be derived from an ion-exchange model in which the adsorbing ion occupies a larger area than the mica binding site (the full line in Fig. 5). Pashley ( 1 ) has shown that when the ion size is included in the model, small changes in its magnitude cause large changes in the form of the potential curve, thus allowing sensitive determination of the ion size. This model considers surface dissociation reactions represented by the equations SNH4 = S

+ NH~

SH= S-+H

[ 1]

+,

[2]

where S-, SNH4, and SH represents a negative site on the mica surface which is unoccupied, bound to an a m m o n i u m ion, and bound to an adsorbed proton, respectively. The corresponding mass action equations for these equilibria are KN =

Kn =

[ S - ] [NH~lsurface [SNH41 [ S - ] [ H + ]surface

[SH]

[3]

[cation ]~.rf~ = [ cation ]bulkexp ( ~ 0 ) ,

[4]

[5 ]

where e is the magnitude of the charge on an electron, ~I'0 is the surface potential, k is the Boltzmann constant, and Tis the temperature. The number of vacant sites per unit area on the mica surface, V, is given by the equation -

-

ions adsorbed per unit area, A is the effective area occupied by an adsorbed ion, and the subscripts N and H refer to the a m m o n i u m ion and proton, respectively. Values of pKH = 6.0 and AH = 1/Ns = 0.48 nm 2 were obtained from previous work (2). The surface charge, ~r0, is related to the surface potential by the G o u y - C h a p m a n equation ~o = { 8~oDkT([H+ ]bulk 1/2 •

+ [NH~-]bulk) }

exit0

slnh(2kT )

[7]

(where e0 is the permittivity of free space and D is the dielectric constant of water) and also to the surface density of adsorbed ions by the equation ~o = - e ( N s - nN -- nH).

[8]

These equations can be solved numerically for given values of AN and KN to give ~0 as a function of concentration. Because An and AN are set either equal to or greater than 0.48 nm 2 the surface is not allowed to change sign. The curve which best fits the experimental data (see Fig. 5) was calculated with an effective surface area for the NH~ ion, AN, of 0.52 nm 2 (cf. K +, 0.53 nm2; Cs +, 0.51 nm 2 (1)) and a KN of 1260 (pKN = 3.1) at p H 5.4. Details of the Hydration Force

,

where the surface concentrations of the adsorbing ions are related to the bulk terms by the Boltzmann equation

V = Ns(1 - nNAN

437

MICA SURFACES

nHAH),

[6]

where Ns is the number of sites per unit area on the bare mica surface, n is the number of

Measurements given in the previous section indicated that the hydration force arose at an NH4C1 concentration of between 1.1 × 10-4 and 1.l × 10 -3 M. Further experiments showed that this force was absent up to 8 × 1 0 - 4 M and that as in previous experiments with K + ions (5) there was fine structure in the force distance relationship (See Fig. 6). Much thinner mica was used to enable greater accuracy in distance measurement, and a much stiffer spring (constant = 1600 N m -~) was used because fairly high pressures were required to push the surfaces together to perform these measurements. This precluded accurate measurements at larger distances where the forces were relatively weak. A pressure of Journal of Colloid and Interface Science,

Vol. 131, No. 2, September 1989

438

DUCKER AND PASHLEY

0]. 0"21 , ~.. 0.1 ] u_ sI ~ 0

j

'

-0.02 -0.04

t 1

2

3

A

D/nm

FIG. 6. The forcesbetween mica surfacesin 1.6 X 1 0 - 3 M NH4C1 at 24.0°C. The forces given in this figure are scaled by the radius at largeseparations,R*, not the actual radius of curvature since this is continuallychangingwhen the surfaces deform under the large forces applied. Note also that the scale on the force axis is linear and there is a different scale for positive (repulsive) and negative (attractive) forces and'the F/R is shown in newtons, not micronewtons. The lower line is the theoretical van der Waals force betweenthe surfaces.On this scalethe doublelayer force lies close to the zero force line. The surfaces jump in from about 0.9 nm (point J) to where they are held in a metastable hydration minimum at D = 0.4 nm.

about 17 a t m was applied to bring the surfaces into primary m i n i m u m and the highest pressure applied in contact was about 28 arm. Because the glue used to attach the thin mica sheets to the silica support disks was relatively elastic, these high applied pressures caused flattening of the surfaces. (In this experiment the greatest flattening was over an area of about 1300/~m2.) Because of this flattening the forces plotted in Fig. 6 are scaled by the undeformed radius, R*, which was measured at large separation. The results show one hydration m i n i m u m at a separation of about 0.4 rim. The dashed regions of the curve show regions which could not be measured because of the nature of the spring device used. The j u m p inward at point J was most likely due to the effect of attractive van der Waals forces. At higher concentrations of a m m o n i u m Journal of Colloid and Interface Science, Vol. 131, No. 2, September 1989

chloride (Figs. 3 and 4), the measured hydration force was monotonic, and no outward j u m p s were recorded. This was probably because the magnitude of the repulsive hydration force was large enough to hide any fine structure (e.g., oscillations) caused by the finite size of the solvent molecules (6). At these higher concentrations the hydrated a m m o n i u m ions must remain between the surfaces on compression, whereas at lower concentrations they can be forced out to be replaced by protons (1, 4), which has the effect of reducing the hydration force. Oscillatory hydration forces have previously been measured when K + (5, 15) or Ca 2+ (16) ions are adsorbed to the surface of mica. The forces shown here for N H ~ are somewhat different in that the distance between consecutive force m a x i m a is about 0.4 n m rather than the 0.3 n m measured for the other ions. However, the method used for distance measurement in this experiment was slightly different from that used previously. With the interferometric technique used, the separation can be measured using either an odd or an even order fringe (5). This is significant because only even order fringes are dependent on the refractive index of the m e d i u m between the surfaces. The data given in Fig. 6 were obtained from measurement o f an even order fringe, and calculated using a pure water refractive index (1.333). In order to make the period of the oscillation the same as those previously reported, an index of about 1.45 is required, but this figure is unreasonably large since our model predicts that the concentration of N H ~ on each surface is only about 2.8 M at infinite separation and decreases as the surfaces approach contact. The fact that such a variety of ions causes the same type of oscillatory force lends further support to the idea that the force is caused by the layering of water molecules in thin films held between molecularly smooth rigid surfaces; however, it is possible that some specific ion-water interaction can vary the periodicity of these oscillations.

FORCES BETWEEN MICA SURFACES

439

TABLE I NH4+

PKdiss(mica) Onset of hydration force (M)

K+

3.1

3.5

8 - 1 1 X 10 -4

0 . 4 - 3 . 0 X 10 -4

Exclusion area of hydrated ion on mica (nm2)

0.52

0.53

Cs+

3.1 1 - 1 0 X 10 -4

0.51

Note. The results shown for K + and Cs+ were measured in previous work by Pashley (1).

C o m p a r i o n s o f NH-~ with O t h e r C a t i o n s

Table I shows a c o m p a r i s o n o f the ion-exchange and hydration properties o f K +, Cs +, and NH~- ions. F r o m this it is clear that the properties o f N H ~ closely resemble those o f both K + and Cs +. The dissociation constant for N H ¢ on mica is the same as that for Cs + and the onset o f hydration forces for N H ¢ lies in the range m e a s u r e d for b o t h K + and Cs +. In a n o t h e r study, K + and N H ~ have been found to have similar ion-adsorption constants on a variety o f other surfaces, including layered aluminosilicates (18). Finally, the area over which the hydrated NH~- ion excludes other ions on the surface o f mica was f o u n d to be intermediate between the values f o u n d for K + and Cs +, which is consistent with measurements o f hydrated ion sizes obtained from conductivity m e a s u r e m e n t s (17). CONCLUSIONS The measured forces between mica surfaces in aqueous NH4C1 solutions agree well with those predicted by classical D L V O theory up to a b o u t 8 × 10-4 M . Above this concentration, an additional short-range repulsive force prevents the surfaces from reaching a p r i m a r y m i n i m u m . This force was identified as the " h y d r a t i o n force" previously observed for a range o f other cations on mica surfaces. At concentrations close to where the adsorption o f N H ~ ions give rise to this repulsion, a detailed study o f the force indicates that it is oscillatory and has the periodicity approximately equal to the diameter o f a water molecule. This structure appears to be caused by the layering

o f water molecules in the thin film. The ionexchange and hydration properties o f N H ~ on mica were f o u n d to be similar to the alkali metal ions even t h o u g h this ion is polyatomic and can dissociate at the mica surface. REFERENCES 1. Pashley, R. M., J. Colloid Interface Sci. 83, 531 (1981). 2. Pashley, R. M., and Israelachvili, J. N., J. Colloid Interface Sci. 97, 446 (1984). 3. Attard, P. and Bachelor, M. T., submitted for publication. 4. Claesson, P. M., Herder, P., Stenius, P., Erikson, J. C., and Pashley, R. M., J. Colloid Interface Sci. 109, 31 (1986). 5. Israelachvili, J. N., and Adams, G. E., J. Chem. Soc. Faraday Trans. 1 74, 975 (1978). 6. Pashley, R. M., and Israelachvili, J. N., J. Colloid Interface Sci. 101, 511 (1984). 7. Horn, R. G., Biochim. Biophys. Acta 778, 224 (1984). 8. Lis, L. J., McAlister, M., Fuller, N., Rand, R. P., and Parsegian, V. A., Biochemistry 17, 3163 (1978). 9. Chan, D. Y. C., Pashley, R. M., and White, L. R., J. Colloid Interface Sci. 77, 283 (1980). 10. Hunter, R. J., "Foundations of Colloid Science," Vol. 1, Chap. 4, p. 183. Oxford Science Publications, New York, 1987. 11. Chan, D. Y. C., and Mitchell, D. J., J. ColloidInterface Sci. 95, 193 (1983). 12. Derjaguin, B. V., KolloidZ. 69, 155 (1934). 13. Horn, R. G., Israelachvili, J. N., and Pfibac, F., £ Colloid Interface Sci. 115, 480 (1987). 14. England, D., in "Water, A Comprehensive Treatise" (F. Franks, Ed.), Vol. 5, Chap. 1, p. 47. Plenum, New York, 1975. 15. Israelachvili,J. N., and Pashley, R. M., Nature (London) 306, 249 (1983). 16. Pashley, R. M., unpublished results. 17. Nightingale, E. R., Jr., £ Phys. Chem. 69, 1381 (1959). 18. Krishnamoorthy, C., and Overetreet, R., Soil Sci. 69, 41 (1950). Journal of ColloM and Interface Science, Vol, 131, No. 2, September 1989