Methods of measurement for the evaluation of monolayer properties : development and applications

Methods of measurement for the evaluation of monolayer

properties : development and applications

Citation for published version (APA):

Gieles, P. M. C. (1987). Methods of measurement for the evaluation of monolayer properties : development and

applications. Technische Universiteit Eindhoven. https://doi.org/10.6100/IR271108

DOI:

10.6100/IR271108

Document status and date:

Published: 01/01/1987

Document Version:

Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers)

Please check the document version of this publication:

• A submitted manuscript is the version of the article upon submission and before peer-review. There can be

important differences between the submitted version and the official published version of record. People

interested in the research are advised to contact the author for the final version of the publication, or visit the

DOI to the publisher's website.

• The final author version and the galley proof are versions of the publication after peer review.

• The final published version features the final layout of the paper including the volume, issue and page

numbers.

Link to publication

General rights

Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain

• You may freely distribute the URL identifying the publication in the public portal.

If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, please follow below link for the End User Agreement:

www.tue.nl/taverne

Take down policy

If you believe that this document breaches copyright please contact us at: openaccess@tue.nl

METHOOS OF MEASUREMENT

FOR THE

EVALUATION OF MONOLAYER PROPERTIES

DEVELOPIVIENT AND APPLICATIONS

PROEFSCHRIFT

TER VERKRIJGING VAN DE GRAAD VAN DOCTOR AAN DE TECHNISCHE UNIVERSITEIT EINDHOVEN, OP GEZAG VAN DE RECTOR MAGNIFICUS, PROF. DR. F. N. HOOGE, VOOR EEN COMMISSIE AANGEWEZEN DOOR HET COLLEGE VAN DEKANEN IN HET OPENBAAR TE VERDEDIGEN OP

DINSDAG 15 SEPTEMBER 1987 TE 16.00 UUR

DOOR

PAULUS MARIA CORNELIS GIELES

GEBOREN TE BERGEN OP ZOOM

DIT PROEFSCHRIFT IS GOEDGEKEURD DOOR DE PROMOTOREN:

PROF. DR. J.A. POULIS EN

PROF. DR. J. BENNEBROEK GRAVENHORST CO-PROMOTOR: DR. IR. C.H. MASSEN

Page

SUMMARY 3

LIST OF SYMBOLS 5

ABBREVIATIONS 1

1 . GENERAL INfRODUCTION

1. Surfactants and surface tension 9

2. Pulmonary surfactant and the respiratory distress syndrome 11

3. Problems in lung surfactant research 13

4. Purpose of the study 16

References 1B

Notes 1B

2. TiiE LANGMUIR - WILHELMY METHOD

1. Introduction, principle and short historica! review of the

Langmuir- Wilhelmy methad 19

2. The surface-tension measurement of monolayer-covered interfaces

2.1 Introduetion 21

2.2 The Wilhelmy plate technique 22

2.3 Contact-angle phenomena: capillarograms

2.3.1 Introductory theoretica! considerations 23

2.3.2 Capillarography 30

2.3.3 Capillarograms measured in the absence of a

monolayer 35

2.3.4 Capillarograms: the influence of a DPPC monolayer 39 3. Leakage problems in the Langmuir-Wilhelmy methad

3.1 Causes for and conventional solutions to leakage 45

3.2 The second harrier methad 46

3.3 The elastic band method: test experiments 51 4. The relevanee of the standardization of the measurement and

characterization of lungsurfactant compression curves 59

References 70

~~s

n

3. TiiE ASYMMETRIC METHOD

1. Introduetion and principle of the asymmetrie methad 73 2. Longitudinal wave theory

2.1 Dilatation, general theory 11

2.2 Special cases B1

3. Sourees of errors in the asymmetrie methad

3.1 Introduetion B2

3.2 Combined dilatation and shear B2

3.2.1 Including shear in the theory B3

3.2.2 Moving side walls B4

3.3 The influence of a nonuniform surface tension variation B5 3.4 The influence of the subphase thickness 91

-4. The asymmetrie method: an automated apparatus

4. 1 The apparatus 93

4.2 Hardware 95

4.3 The software and the measurement procedure 95

4.4 Data processing 100

5. Measurements with the asymmetrie method: methodical aspects

5.1 Balance transfer 101

5.2 The influence of a nonuniform surface tension 103

5.3 Shear influence due to side walls 104

5.4 The influence of the subphase thickness 106 5.5 Addition or compression of a monolayer 108

5.6 Other aspects 109

6. Discussion of the asymmetrie methad 111

7. Measurements on phosphatidylcholine monolayers 113

References 118

Notes 118

4. THE BENJAMINS - DE FEIJTER METHOD

1. Introduetion and principle of the Benjamins-de Feijter methad 119 2. Theoretica! aspects of the BDF methad 120 3. Sourees of errors in the BDF methad

3.1 The influence of a nonuniform surface tension variatien 123

3;2 The influence of shear 124

3.3 Wetting of the elastic band 126

4. The BDF method: a semi-autornaeed apparatus

4.1 The apparatus 128

4.2 Hardware 130

4.3 Software and measurement procedure 131

4.4 Methodical aspects 132

5. Measurements on PC monolayers 133

References 138

5. THE OSCILLATING BUBBLE METHOD

1. Introduetion and principle of the asciilating bubble method 139

2. The experimental set-up 140

3. The measurement of the bubble size

3.1 The pos i tioning of the bubble 145

3.2 The influence of the capillary size 148 4. Determination of the Laplace pressure 151 5. Calibration procedure and test experiments

5.1 Pressure calibration 154

5.2 Calibration of the bubble size 154

5.3 Surface tension of liquids 155

6. Surface tension of monolayer covered interfaces 158 7. Discussion

7. 1 Discussion of the methad 160

7.2 Some considerations concerning LWM and OBM 162 7.3 LWM and OBM: reflection of the in-vivo situation? 164

References 166

SAMENVATTING 167

SUMMARY

In a number of industrial and biologica! processes such as foaming and respiration, the properties of surfactants (surface active agents) present at liquid-gas interfaces play a determining role. They cause the interfacial tension

(a)

todependon the surface area (A).

In the lungs, the presence of so-called lung surfactant is thought to be jointly responsible for a normal breathing pattern. However, in premature babies respiratory failure may occur which is attributed to an impraper lung surfactant function. In order to understand the role of lung surfactant and i ts components, and in order to develop criteria for an artificial surfactant or for the maturi ty of lung surfactant, methods of measurement are needed.

The research in question has been concentrated on four methods of measurement, all of which are concerned with the physico-chemical properties of surfactant. They establish the a - A relation. The methods investigated are:

1. The Langmuir - Wilhelmy method {LWM)

Attention has been paid to two important sourees of error: contact-angle phenomena at the Wilhelmy plate, and leakage of monolayer material.

The measurement of capillarograms, i.e. force on the Wilhelmy plate versus penetration depth, showed that filter paper exhibits only very small contact-angle effects. This material therefore seems the best choice for use as Wilhelmy plate.

Leakage problems were solved by using either a closed elastic band or a secend moving harrier.

a - A curves of sheep surfactant were measured using different Wilhelmy plates and different compression speeds. The surface tension at the end of compression, the average loop area and Clements stability index which were used to characterize the resul ts, turned out to depend on Wilhelmy plate material and compression speed. Th is indicates the necessi ty of standardizing the measurement of lung surfactant compression curves. Initia! suggestions for such a standardization are given.

-2. The asymmetrie method

Wi th this method the surface di latational elastici ty {e) of monolayers can be determined. A completely automated set-up was realized, after which several sourees of errors were examined:

the influence of nonuniform surface tension variations is circumvented by locating the Wilhelmy plate at x /L = 0.42,

wp

shear effects due to fixed side walls are eliminated by using elastic side walls rnaving along with the monolayer,

- the influence of the subphase thickness has shown to be negligible if the penetration depthof longitudinal waves is less than half the subphase thickness,

- the results were corrected for the dynamic behavior of the electrobalance.

Next, measurements were performed on a homologous series of phosphatidyl cholines {PC). Liquid-condensed PC monolayers showed higher

lel

values than the liquid-expanded monolayers. For all PC's the phase angle was close to zero, implying an almost purely elastic behavior.

3. The Benjamins- de Feijter method

This method is different from the asymmetrie method in that it avoids the introduetion of shear effects. In this case also the set-up has been real ized and the signal processing automated. Different sourees of error were studied, after which measurements on the PC monolayers were carried out. We found no difference in shear contribution between the asymmetrie method and the Benjamins - de Feijter method.

4. The oscillating bubble method {OBM)

This method was realized in a newly built set-up, designed as to give the same information as the Langmuir - Wilhelmy method. Different sourees of error in the determination of bubble radius and Laplace pressure were analyzed. The measurement and signal processing were automated. Test experiments with pure liquids showed a good accuracy and reproducibility. In a pilot experiment a - A curves of sheep surfactant were measured and compared with those obtained with 'the LWM under virtually the same conditions. Both minimal surface tension and direction of the loops were different in the two methods. Apparently, the processes in the LWM and OBM are not yet completely understood. It is therefore argued that drawing conclusions about the in-vivo behavior of lung surfactant, based on a - A measurements is not yet justified.

LIST OF SYMBOLS A A c A CS B CR D

E

H surface area

area of capillarogram (Chapter 2)

cross-sectional area of bubble (Chapter 5) transfer function of electrobalance compression rate

penetration depthof the Wilhelmy plate force

average hysteresis of capillarogram (Chapter 2) height difference between pressure transducer and

2 m Nm 2 m cyclimin m N N/m

lower end of capillary (Chapter 5) m Hox hysteresis of compression curve to x% of initial area N/m I inhomogeneity function L trough length m V

w

w

_WO a b d g h k p

number of monolayer molecules radius of curved interface or bubble Clements stability index

temperature volume weight

weight of water maximally taken up by Wilhelmy plate half of the troughwidth

width of elastic band subphase thickness gravitational constant height of the Wilhelmy plate complex wavenumber

length of elastic band

perimeter of the Wilhelmy plate pressure 5 -m m N N m m 3 m m -1 m In m

t u y x, y, z

AA

A

a f3 ö ö p K w 9 p a a 0 T

thickness of Wilhelmy plate time

local surface displacement in x-direction liquid veloei ty

harrier speed

speed of the contact line wetting speed

width of the Wilhelmy plate

local surface displacement in y-direction coordinates of a rectangular coordinate system x-coordinate of the position of the Wilhelmy plate height of the contact line

amplitude of surface area variation amplitude of surface tension variation

Laplace pressure surface pressure

damping coefficient

thickness of water layer attached to the WP penetration depth of surface waves

surface dilatational elasticity phase angle

viscosity of the liquid surface dilatational viscosity

real wavenumber wavelength

angular frequency contact angle

density of the liquid

surface (interfacial) tension

surface tension of a clean air-liquid interface time constant of monolayer relaxation

time constant of balance transfer function

m s m m/s m/s m/s m/s m m m m 2 m N/m N/m2 N/m -1 m m m N/m Ns/m2 Ns/m -1 m m -1 s kg/m3 N/m N/m s s

ABBREVIATIONS BDF Benjamins - de Feijter DBPC dibehenayl phasphatidylchaline DDPC didecanayl phasphatidylchaline DMPC dimirystayl phasphatidylchaline DPPC dipalmitayl phasphatidylchaline DSPC distearayl phasphatidylchaline

ESP equilibrium spreading pressure

FP filter paper G glass CF glass filter Cg graunded glass LC liquid-candensed LE liquid-expanded

LWM Langmuir - Wilhelmy methad

UIT langi tudinal wave technique

OBM asciilating bubble methad

PC phasphatidylchaline

Pt Platinum

WP Wilhelmy plate

-OIAYfER 1

GENERAL INTRODUCriON

1. Surfactants and surface tension

We know that attempts to blow bubbles from a hoop will not be successful when one simply uses water. Also, washing up wi th pure water will not produce very clean dishes and pans. Only i f soap or detergent is added stable bubbles can be blown, the dishwater can foam, and dirt can be removed more easily. Apparently, the properties of an air-water system are drastically changed by the addition of a small arnount of soapor detergent.

A factor of paramount importance in this drastic change is the structure of a molecule of soap or detergent. In general such a molecule consists of two parts: a hydrophilic and a hydrophobic part. The hydrophilic part {the headgroup H, Fig. l.la,b) is highly soluble in water, while the hydrophobic part (consisting of one or two hydracarbon tails, Fig. l.la,b) is highly insoluble. The consequence of the combination of these two parts within one molecule is that these molecules accuroulate at the air-water interface, because the headgroup 'prefers' the surrounding of water, while the tails 'prefer' that of the air (Fig. l.lc). In this way a layer of one molecule thick is formed at the air-water interface. Substances that are able to farm monolayers are called surfactants (surface active agents).

The physical quantity that is influenced by the presence of a monolayer is the surface tension. Water molecules deep in the liquid are equally attracted in all directions by the surrounding molecules, in contrast with those at the surface. Therefore the water molecules deep in the liquid are in an energetically more favourable position. Since the system 'seeks' for a si tuation wi th the least energy, i t will try to reduce the surface area. Surface tension can therefore he regarcled as the tendency of a liquid to lower its {free) energy by reducing its surface area1 · 2 . For a surfactant

molecule the situation is different: because of the two parts the energetically more favourable position is at the interface rather than in the liquid. Consequently, its presence at the interface will lower the free energy. Hence, surfactant molecules at the air-water interface reduce the surface tension. The higher the surfactant concentration in the monolayer, the lower the surface tension. The actual reduction of the surface tension depends on the size of the hydrophilic and hydrophobic part, and on their relative solubilities.

a) b) c)

T T T

AIR

H

WATER

Fig. 1.1 a) Example of a surfactant molecule consisting of a hydrophilic headgroup (H) and hydrophobic tails

(T).

The molecul.e shown is dipalmitoyl phosphat idylcholine (DPPC). b) schematic representation of the same molecule. c) Surfactant molecules accumulate at the air-water interface, the headgroups being partly surrounded by water molecules, while the tails point into the air.

So far, the picture has been a static one. When stability of foams or bubbles is considered we have to cope wi th the dynamic aspects of a monolayer at an air-water interface as well.

Two processes are of importance. The first one is the fact that a thin film of surfactant molecules will try to contract when it is stretched, i.e. it behaves like an elastic membrane. This effect is called the Gibbs effect and is expressed in the Gibbs elasticity. The second aspect is the Marangoni effect. I f a surfactant film is stretched new surface area is

-created to which surfactant molecules can diffuse from the bulk liquid. The Marangoni effect implies that some time is required for this diffusion process to take place, and that only after some time a newly stretched film will have reached an equilibrium surface tension.

Acting together, the Gibbs elasticity and the Marangoni effect stabilize foams against small deformations. A pure liquid has no Gibbs elasticity, which is the main reason for the fact that a pure liquid cannot faam [1]: such a foam cannot withstand small deformations.

Al though our picture of surfactants and surface tension is strongly simplified, it contains the basic features that are the background of the numerous applications of surfactants.

2. Pulmonary surfactant and the respiratory distress syndrome

Surfactants also play an important part in biophysical systems and processes. Typical examples are cell membranes and respiration. We will focus on the latter and therefore consider an alveolus and surfactant function according to one of the present theories.

Fig. 1.2a shows an isolated alveolus of spherical shape. It consists of lung alveolar (epithelial) tissue on which a continuous liquid layer is present, called the liquid lining layer. Since the liquid-air interface is curved, a pressure difference across the interface exists. In the case of a spherically shaped interface this pressure difference (ApL)' given by Laplace's law, reads

2n

R

where pa is the pressure of the air, p 1 the pressure of the liquid, a the surface tension, and

R the radius of the curved interface (radius of the alveolus).

Due to the surface tension of the 1 iquid layer the alveolus wi 11 try to reduce i ts size. This resul ts in a l.aplace pressure ApL which increases with increasing surface tension and with decreasing size (Eq. (1.1)). During respiration the size of the alveolus changes; at the end of exhalation the size is minima!, and the l.aplace pressure is maximal. This l.aplace pressure can partly be counteracted by the tissue. However, there is experimental evidence that when the water present is pure, collapse of alveoli cannot be prevented by the tissue alone [2]. The existence of stable alveoli is therefore attributed to the fact that the liquid lining layer is covered with a monolayer (Fig. 1.2b). Since the si ze of the alveolus changes during respiration the monolayer is compressed and expanded, i.e. the surface concentratien increases and decreases. As pointed out in §1 a monolayer lowers the surface tension with increasing surface concentration. The consequence is that, consiclering the stability, the increase in ApL due to the reduction of R during exhalation is counteracted by a decrease in a. In this way a monolayer of lung- or pulmonary surfactant is thought to prevent alveolar collapse.

a) b)

T

~

XX XXXXXX

Fig. 1.2 a) Schematic representation of an isotated alueoLus; A = air, L

=

Liquid Lining Layer, T

=

tissue. b) a monolayer of puLmonary surfactant present at the air-water interface.

If the surfactants ability to lower the surface tension during exhalation is inadequate, the alveoli cannot be stabilized and collapse of alveoli follows. This implies a reduction of alveolar surface available for gas exchange (uptake of oxygen and delivery of carbondioxide}, and results in respiratory problems, known as the respiratory distress syndrome (RIJS). Such a si tuation may occur wi th premature babies, which suffer from a

-surfactant deficiency. I f RDS occurs the cl in i cal treatment includes an artificial breathing procedure using air at a high pressure in order to open the collapsed alveoli. Th is procedure a lso has some drawbacks. The high pressure may e.g. cause damage to the well functioning part of the lungs. This problem can be coped with in several ways. First, the technique of high frequency ventilation [3] has been developped in which there is no convective transport and therefore the risk of mechanica! damage is reduced. Second, artificial surfactant can be applied to the airways.

I f premature babies die, this is - in the developped countries - for approx. 50% due to RDS [4]. Apart from surfactant deficiency, surfactant disfunction may be caused by aneasthesia. The latter complication is one of the reasans for the occurrence of respiratory problems with adults (ARDS) [5]. It is the goal of pulmonary surfactant research to find out the actual role of surfactant, to develop an artificial surfactant, or to find other ways to reduce or prevent the complications in case of surfactant disfunction.

3. Problems in lung surfactant research

The actual picture of lung surfactant function is not as straightforward as presented in §2. The spherical shape of the alveol i, the continuous liquid lining layer and the relevanee of surface tension are subject to discussion.

The importance of surface-tension farces was shown by von Neergaard who measured pressure-volume (pa - V) curves of air-fi lled and liquid-fi !led lungs [2]. He also found that the tissue elasticicty 'ceases to produce tension ( .. ) in the range of the normaL voLume at the end of expiration. If the voLume is further decreased the tension of the tissue can actuaUy oppose compression' [2]. Von Neergaard therefore compared the lung with a sponge, and tried to develop a conceptual model of alveolar mechanics based on the Laplace equation. He proposed two models (Fig. 1.3}. The first model considers an alveolus of spherical shape with a size less than a hemisphere

in the absence of a monolayer (Fig. 1. 3a.). If the volume is increased the radius decreases (R2

<

R1). According to Eq. (1.1} this leads to an increase in Laplace pressure. The second model considers a spherically shaped bubble (alveolus) with a size larger than a hemisphere (Fig. 1.3b). In this case the radius increases wi th an increase in volume, and this configuration would therefore lead to a decrease in Laplace pressure. Von Neergaard experimentally found that 8pa/8V

>

0, and he therefore suggested

that the physiological situation would correspond to the first model. In 1957 Clements found that when lung-lavage fluid was deposited on an air-water interface and the available surface area was reduced. the surface tension became lower [6]. This was in favour of the second model, because Clements suggested that this finding might again imply 8(ApL)/8V

>

0 according to the Laplace equation. When two years later Avery and Mead discovered that the occurrence of RDS correlated with surfactant deficiency

b)

~

L

~

~

Fig. 1.3

Two

configurations of a sphericaLly shaped air-liqutd interface attached toa capillary (cf. Fig. 5.1) a) sized Less than a hemisphere, b) sized Larger than a hemisphere. An increase in airvoLume Leads toa decrease in

R in a).

to

an

increase in

R

in b).

[7]. this seemed to emphasize the justification of the second model. Yet, objections were raised. The equilibrium of a collection alveoli of different size requires different surface tensions in each of the alveoli (Eq. (1.1}). However. this doesnotmake sense i f neighbouring alveoli of different size would be lined with a continous liquid layer, since surfactant flow would occur in order to reduce the surface tension differences; this again would lead to collapse.

-This problem has led to serious doubts about the original concept and to new ideas. Since the first model (Fig. 1.3a) had fallen into oblivion, Reifenrath recalled both models, but he doubted both configurations and proposed a more foam-like structure (Fig. 1.4). Where Reifenrath doubted the geometry of the models, Barrow and Hills proposed a non-continuous liquid lining layer [9]. Although this concept can solve the problem outlined above, Barrow and Hills' suggestions contained more. To their opinion the main function of dipalmitoyl phosphatidylcholine (DPPC) - the

Fi.g. 1.4 Rei.fenrath's suggestion for a different al.ueol.ar structure.

Fi.g. 1.5 The concept of Barrow and Hi.l.Ls of a dry al.ueotar

watt,

coated wi.th surfactants and havi.ng surfactant pool.s.

A=

air,

P =

surfactant pool.,

AS=

adsorbed surfactant,

T =

tissue.

main constituent of lungsurfactant - is to coat the alveolar wall in order to provide a hydrophobic layer which inhibits the spreading of surfactant pools toa continuous lining layer (Fig. 1.5). So not only the concept of a continuous lining layer but also the original concept of the variation of

the surface tension as stahilizing factor was subjeeeed to discussion. Al though i t is now generally accepted that lungsurfactant plays an

essential role in the respiratory system, na agreement has yet been reached with respect to its actual functioning. Avery and Mead discovered that the surface tensions of lung fluid obtained from heal thy lungs was significantly lower than those obtained from lungs of premature babies that died due to ROS [7]. This led to the idea that a low surface tension under compression is a prerequisite for a proper lung- and artificial surfactant. However, an important reason for the doubts raised was the fact that the results obtained with physico-chemical methods showed ártifacts which made i t sametimes hard to decide whether the measured effects were due to surfactant behavior ar not. It has for instanee been argued that the low surface tensions, as measured with lung surfactant in the Langmuir-Wilhelmy methad (Chapter 2), are due to an artifact rather than to surfactant behavior [9].

4. Purpose of the study

As pointed out in the previous sections, one of the most important consequences of surfactant is that the surface tension

(a)

becomes dependent on the available surface area (A). As a consequence, a considerable amount of information concerning (natural as well as artificial) surfactanes is derived from physico-chemical experiments in which a is measured in dependenee of A. Fig. 1.6 shows a general schematic representation of the methods applied in most of those experiments. At the interface - e.g. of liquid and gas - a surfactant {S} is applied. A harrier (B) is used to limit the available surface area and to realize the surface area variation. A measurement device (M) detects the changes in the monolayer, which are mostly recorded in terms of surface-tension variations, but which can also be surface-velocity ar surface-displacement variations. Anyhow, the purpose is to come to conclusions about the surfactant behavior.

The a - A relation, however, can be influenced in many ways. These influences can be of 'physico-chemical nature', such as temperature, pH,

-speed and amplitude of the surface area change. Also artifacts due to the geometry of the actual set-up, and due to the measurement device may seriously hamper the interpretation of the measurements. Wi th respect to lung surfactant research these influences and artifacts (may) well have led

GAS

B M

OliTPliT

Fig. 1. 6 Generat schematie representat ion of the me thod.s used to establish the a - A retation of surfactants. B =

harrier, S

=

surfactant at the interface, M = mensurement deuice.

to conflicting results and complications in the rnadelling of the alveolar

processes as described in §3.

Therefore, the main purpose of our work was a critica! assessment of the methods used to measure and evaluate the a - A relation. The methods studied are:

1. The Langmuir - Wilhelmy method {Chapter 2) 2. The asynnnetric method (Chapter 3}

3. The Benjamins - de Feijter method {Chapter 4} 4. The oscillating bubble method (Chapter 5}

These methods were chosen in view of the biomedical application {RDS). The Langmuir-Wilhelmy method and the oscillating bubble method are the most widely used ones, while the other two methods. were developped because of their ability to measure dynamic properties. The biomedical background has also inf luenced the frequency range of the set-ups and the choice of surfactants used. It should be noted that the methods (may} also find application in a number of research fields that are beyond the scope of

Heferences

I. Aubert, J.H., Kraynik, A.M., Rand, P.B., Scientific American, 254(5):58 (1986)

2. von.Neergaard, K., Z. Ges. Exptl. Med. 66:373 (1929)

3. Kol ton, M., Cattran, C.B., Kent, G., Volgyes i, G., Froese, A.B., Bryan, A.C., Anesth. Analg. Curr. Res., 61:323 ( 1982)

4. Verloove-Vanhorick, S.P., Verwey, R.A., Thesis University of Leiden, The Netherlands, 1987

5. Lachmann, B., Danzmann, E., Chaptei- 18 in 'Pulmonary Surfactant', Eds. B. Robertson, L. van Colde, J. Batenburg, Elsevier Science Publishers, Amsterdam, 1984

6. Clements, J. A., Proc. Soc. Exptl. Biol. Med. 95:170 (1957) 7. Avery, M.E., Mead, J .. Amer. J. Dis. Child., 97:517 (1959) 8. Reifenrath, R .. Respir. Physiol. 24:115 (1975)

9. Hills, B.A .. J. Physiol .. 325:175 (1982)

10. Rowlinson, J.S., Widom, B .. "Molecular theory of capillarity', Ciarendon Press, Oxford. UK (1982).

1. The term 'surface tension' will be assumed to refer to liquid-gas interfaces only. The term 'interfacial tension' will be used in case solid-liquid or solicl-gas interfaces are included.

2. Of course, this picture of surface tension is far from complete. Many other ways to look upon surface tension are possible. For a more detailed description of the concept of surface tension the reader is referred to e.g. [10].

-OIAPTER 2

TIIE LANGMUIR - WIUIEI...MY METHOD

1. Introduction. principle and short historica! review of the

Langmuir-Wi lhelmv method

The interfacial tension of a liquid-gas or a liquid-liquid interface is, in the case of pure liquids, a constant which is determined by the molecular properties of the liquids (and gases) involved, and by temperature and pressure. Surfactants that are present at an interface may have a drastic influence on the interfacial tension, mainly since they cause it to depend on the specific surface area (area per molecule). One of the methods used to evaluate the properties of surfactants is the Langmuir-Wilhelmy method {LWM), which offers the possibility to determine

the surface tension 1 (a) - area {A) relation of a monolayer.

The conventional set-up of the method is schematically shown in Fig. 2.1. It consists of a Teflon (Langmuir) trough which is filled to the brim with the subphase liquid. The surface-active material is deposited at the interface; the area available for the surfactant is limited by a (movable) harrier that lies across the trough. This harrier is used to apply surface area variations; the surface tension is simultaneously measured by means of a Wi lhelmy plate (WP). Th is plate is dipped into the subphase 1 iquid so that a meniscus is formed and a vertical force, proportional to the surface tension, is exerted on the plate. The result of a compression-expansion experiment is usually presented in a surface tension - area

(a -

A) curve, but also other types of presentations are possible [1].

The method is in fact the result of numerous adaptations of some basic parts. Peckels was the first who used a trough with harriers [2]. Her design was later adapted by Langmuir to become the classica! 1917 set-up [3). Fora long time troubles with the cleanliness of the set-up have been

5

Fig 2.1. Schematic uiew of the Lnngmuir·Wi.thetmy method. Teflon trough, 2

=

Withetmy plate, 3

=

movabte harrier, 4 = monolayer on water, 5 =water.

a souree of annoyance. This was partly due to the materials used: glass, metal ar perspex troughs wi th different kinds of coating to provide the necessary hydrophobici ty. The application of Teflon put an end to these problems and nowadays most troughs are milled from a solid piece of Teflon. In Langmuir's original set-up, a floating harrier was meant to separate the film-covered surface from the clean surface and to measure the surface tension difference - being the surface pressure - by means of a torsion wire. However, the separation was cumbersome, and sa problems with leakage of monolayer material led to numerous adaptations and even completely new designs [4-13]. In most cases the floating harrier has been replaced by a WP. By doing sa the problem of leakage was displaced to the movable harrier. This problem led to adaptations that differ in the way the clean surface is separated from the monolayer covered surface (10, 12, 13].

For an accurate measurement of monolayer properties, a careful examination of all sourees of errors is necessary. Such an examination starts with the design of the set-up: the choice of materials, the type of harriers, etc. We will nat pay attention to these factors since considerations concerning the design as well as those concerning the necessary scrupulous handling to arrive at a perfectly clean interface are described by Gaines [14]. In the present Chapter we shall deal wi th some factors we payed special attention to. These factors include contact-angle (§2} and leakage effects (§3). and the influence of the compression rate (§4}.

-2. The surface tension measurement of monolayer covered interfaces

2.1 Introduetion

In the study of monolayers at liquid-gas or liquid-liquid interfaces one is interested in the monolayer 's abi 1 i ty to lower the surface tension a with respect to the surface tension a0 of the clean, uncovered surface, i.e. the surface pressure {IT):

11 {2.1)

Although there are many methods to measure the surface tension of a pure liquid {see e.g. [14, 15]), only a few are sui table for use in monolayer studies2 . The first methad of measurement used was the Langmuir floating harrier {Fig. 2.2). A harrier floats on the subphase liquid, providing the separation of monolayer-covered and clean surface. This float is connected to the side walls of the trough through flexible paraffin-coated threads or

3

Teflon sheets, in order to prevent leakage It is also connected to a torsion wire or strip and a scale. The monolayer-covered surface and clean surface exert different farces on the float, and the net (horizontal) force is proportional to the surface pressure (Eq. (2.1)). The elegance of this device lays in the direct measurement of 11, and the independency of the net horizontal force on contact angle, as was shown by Harkins and Anderson [16].

Several disadvantages are to be mentioned:

- the force is averaged over the whole lengthof the floating harrier, which is .located at the end of the trough; this implies that no local measurement can be performed,

- it is assumed that the conneetlans to the side walls are identical in shape and do nat take up farces; usually, part of the gaps is taken into account to calculate the effective lengthof the float,

- the device is quite sensitive to changes in water level, and - a nonzero surface pressure will displace the float, and thereby

influence the surface area.

These sourees of errors can mostly be accounted for, for instanee by adjusting the water level. by putting the torsion wire at some distance

from the surface, and by applying a compensating force and a servo mechanism to keep the float at its original position [17].

The most serious error reported, however, is leakage of monolayer material due to a hazardous construction between float and side walls. This

leakage is more likely to occur when surface tension falls to very low

3

Fig. 2.2 The fLoating ba.rrier principle. 1 = trough; 2 monolayer; 3 = float; 4 = conneetion to the side wall; 5

torsion wire; 6

=

scale.

values (§3). I f one is interested in surfactants that do not attain low surface-tension va lues, or do not adsorb ( the lat ter would disturb the clean surface and thus the reference surface tension). and i f one is content wi th the non-local measurement. the floating harrier may be a proper way of evaluating surface pressures, free from contact-angle artifacts. Otherwise, the alternative is the Wilhelmy plate (WP) method.

2.2 The Wilhelmy plate technigue

The most widely used procedure to measure the surface tension of monolayer-covered interfaces is to locate a WP, suspended from an electrobalance, at the interface {Fig. 2.3). A meniscus is formed and a force is exerted on the plate. This force is proportional to the surface tension according to

22-2.a.(w + t).cos

e

(2.2) in which w is the width of the WP.

t the thickness of the WP, and

e

the contact angle between liquid and plate.

The monolayer-covered surface is disturbed by the presence of the WP. The increase in surface area [18]. the influence of a non-straight contact line [19] and other sourees of errors are of minor influence. Contact-angle phenomena however, are considered to be the most serious shortcoming of the WP technique. When the WP is applied it is usually assumed that the contact angle is zero. As we will see, the occurrence of B

=

0 is exception rather

than rule.

2.3. Contact angle phenomena: capillarograms

2.3.1 Introductory theoretica! considerations

Users of the WP method generally consider the contact angle as an artifact, al though i t should be regarcled as a fundamental property of solid-liquid interactions. Contact-angle phenomena are found in numerous research fields (e.g. determining the surface energy of solids). daily applications like lubrication and wetting (raincoats, canvas). and in biologica! processes (e.g. agricul ture). In all these examples surface tension and contact angle are the basic quantities. We shall first pay attention to some theoretica! aspects, in order to grasp the factors that determine the contact-angle behavior, and to decide which measures might minimize contact-angle influences on results obtained with the WP method.

When a liquid is brought into contact with a solid and then allowed to attain equilibrium. this equilibrium is governed by the surface free energies per unit area (a. referred to by surface tension in the above). The case of a meniscus formed to a vertical plate dipped into a liquid, is illustrated in Fig. 2.3. Under special conditions, which will bedealt with below, the surface free energies and contact angle are interrelated by Young's equation:

(2.3}

in which the subscripts s, v, 1 indicate the solid, vapour and liquid phase respectively, and

Be is the equilibrium contact angle.

This equation may be derived by minimizing the free energy of the system when the meniscus is formed. Neumann [20] distinguished three terms contributing : the change of solid-vapour to solid-liquid interface, the increase in liquid-vapour interface, and the work done against gravi ty by the rising liquid. Other derivations are also possible [21, 22].

For our purpose the assumptions made in the derivation are of special interest:

1. the solid surface is supposed to be perfectly smooth 2. the solid surface is supposed to be chemically homogeneaus 3. the solid part exposed to the vapour phase is and remains dry

-;/~/;--/.,-/-r-/--r/-r/~/----;-z = 0

Fi.g 2.3 Locati.ng a Wilhel.my pl.ate at an interface (z=O). When the contact i.s establ.i.shed, the areas of sol.i.d-uapour, sol.id-l.i.qui.d and l.i.qui.d-uapour interface have changed. Al.so work is done in rising of the l.iquid. At equil.i.bri.um l.i.qui.d

and sol.id meet at a contact angl.e 6. The hei.ght of the contact l.i.ne i.s zei..

4. Eq. (2.3} is derived in the absence of a monolayer; if surfactants are present, they should be taken into account

5. a force component perpendicular to the solid surface is also present; it is assumed that this force does not affect the solid surface [23]

6. Eq. (2.3) is val id on macroscopie scale only; on microscopie scale deformation of the profile occurs

-7. the equilibrium contact angle is assumed to be independent of the history of the contact line4 . This implies reversibility when the contact angle is advanced or receded from its original position, i.e. Eq.

(2.3)

is valid under static conditions only. When the con-tact line is not at rest, the hydrodynamic behavior of the liquid should be taken into account.

Although these points restriet the use of Eq.

{2.3),

it still contains an important fact. Since cos ae ~

1

Eq.

{2.3)

implies that spontaneous spreading occurs if

(2.4)

The value for which o₁v = osv- osl is called the critica! surface tension (oe) introduced by Zisman [24]. Neglecting remark 4, Eq. (2.4) clarifies for instanee the 'leakage' of monolayer material to Teflon parts of the trough (oe~ 18 mN/m [24]), when compressed to low surface tensions.

In cases where o1v

>

asv - asl we deal with a partially wetted solid surface and a nonzero contact angle between solid and liquid.

We will now discuss the influence of the most important of the remarks mentioned above on contact-angle phenomena and their consequences for the WP method.

1 . Surface roughness

It was Wenzei [25] who first incorporated surface roughness, which resulted in

where

cos a w

a is the

w contact angle at rough surfaces according to Wenzel, A is the true surface area,

r

A is the

g geometrical surface area, and a is the equi 1 ibrium

e contact angle for a perfectly smooth solid surface.

{2.5)

smaller than 90, but increases contact angles larger than 90.

Another important factor is the difference between the so-called microscopie and macroscopie contact angle. Where the microscopie angle (Bm) is the true contact angle between solid and liquid, it is the macroscopie one (BM) which is the quant i ty measured in experiments {Fig. 2.4). The macroscopie or 'phenomenological' [20] contact angle determines the meniscus profile, and is hence the principal quant i ty in characterizing wetting phenomena [20].

air

Fig. 2.4 Ittustration of the difference between microscopie

contact angLe Bm and macroscopie contact angte BM, as a consequence of surface roughness.

Wenzel's formula is basedon the idea that the surface free energy of the

solid-liquid interface is proportional to the surface area. However, it does not include the shape or direction of the surface roughness, nor does it account for contact-angle hysteresis found in experiments. The remairring

gap between theoretica! description and experimental results has been

narrowed in various steps. The description of ideal surface roughness (triangular and sinuoidal parallel grooves, [26, 27]). and of random roughness in one dimension [28] have resulted in a better onderstanding of contact angle hysteresis. Recently, jansons [29] gave a description of the influence of surface roughness in two dimensions.

Studies on surface roughness theoretica! [28, 29] as well as experimental [28] - have revealed the necessi ty to distinguish between

static and dynamic contact angles, and between advancing and receding contact angles. The preserree of surface roughness causes a moving contact line to experience ob~tacles in the surface in a different way when

-advancing than when receding. Hence, contact-angle hysteresis is caused. Moreover, the contact angle depends on the speed of the contact line, also called the wetting speed. Several studies have shown the increase of (advancing) contact angle with increasing wetting speed [30].

2. InJw.ogeneous surfaces

In analogy wi th Wenzei 's treatment of surface roughness, Cassie [31] showed the effect of chemica! inhomogenei ties. For an inhomogeneous but perfectly smooth surface the Cassie equation reads

cos 9 _c

where 9c is the contact angle for the heterogeneous surface, fi is the surface fraction of the i-th component

(2.6)

ai is the intrinsic equilibrium contact angle for the i-th component. This equation is also based on the surface free energy of the solid -liquid interface. Like Young's and Wenzel's equations,

Eq.

(2.6) does not offer an explanation for the experimentally observed hysteresis. Yet,

Eq

.

(2.6) demonstrates that the WP should be perfectly free of whatever types of patches other than the intrinsic material.

3. Dry surfac.e of the sol.id }ila.se

To satisfy this requirement seems especially difficult in the case of a receding contact line, where a wet solid surface will be hard to avoid. Even in case the WP is dry, adsorption of gases from the vapour phase can also influence the surface of the solid.

We already want to mention here that the presence of a liquid layer on the solid surface will give rise to additional complications. They concern the o measurement with a WP, due to the weight of this layer (§2.3.2).

4. The presence of surfactant at the air-Liquid interface

In thi s case

a:n

extra phase is added to the sys tem of liquid, gas and solid. Since the surface tension may be drastically influenced and become dependent on the area of the interface, Young's equation should then be reconsidered (which we shall not do here). For low a values Eq.

(2.3)

prediets decreasing contact angle. The opposi te has been observed when campressing a DPPC monolayer to low surface tension values

[32].

Problems that ·ar i se in dynamic si tuations may be responsible for this observation. These problems concern - among other things - the deposition of the monolayer on the solid in the case of a receding contact line, which may cause chemical inhomogeneities (Fig. 2.5).

~Oiid

liquid

Fig. 2.5 Deposition of a monoLayer on the solid, initially present at the air-water interface onLy, Leads to a chemically different soLid surface after a receding contact

Line.

Consequences for the LWlf

During a monolayer compression-expansion measurement the height of the contact line (zcl' Fig.

2.3)

depends on surface tension and contact angle according to [18]

J2a

.

(1 - sin ee) pg

in which pis the density of the liquid (p 103 kg/m3 for water), and g the gravitational constant (g

=

10 mls2 ).

(2.7)

For the ease of notation the macroscopie contact angle will from now on

-be denoted by a.

If a and a are time dependent, the contact line will move with a speed

dzcl_daji-sina d a j a

dt - dt' 2pga - cos a.dt' "'"2-pg __ {;:-l---s-i-=-n-a"")<"" (2.8)

vel can be related to the speed vb of the harrier in the LWM, by using the surface dilatational elasticity (é, Chapter 3)

da (2.9)

where Lis the trough length. Then Eq. (2.8} can be rewritten as

é.Vbjl-sina d a j a

-L- 2pga - cos

e.

d t. "2p-g--:(.-:1...:;.--s-i:-n---::9") (2.10) With Eq. {2.10) the speed of the contact line duringa small time interval in a compression - expansion measurement can be estimated. To give a quantitative indication of vel' we consider a DPPC monolayer with é = 100 mN/m at a

=

50 mN/m (Chapters 3 and 4}. Taking vb

=

1 mmls, L

=

100 mm, a

=

5 degrees and d9/dt

=

0 Eq. (2.10) yields ~ 2 mmlmin. This illustrates that

the contact line can move up and down the plate with a considerable speed, and that surface roughness and 'wetting speed' may indeed be of influence during a a - A measurement. In parts of the compression where. é becomes small (e.g. phase transitions; squeeze-out plateaus), the impressed motion may become smaller.

The difference between advancing and receding contact angle is also of influence on the LWM. During the compression of a monolayer the surface tension is reduced and the height of the contact line will decrease (Eq. (2.7)). Thus the contact line will move towards regions previously wetted by the meniscus. We then deal with receding contact angle. Expansion of the monolayer leads to an increase in zcl and makes the contact 1 ine move towards previously unwetted parts of the plate, i.e. we are dealing with advancing contact angle.

After lowering of the contact line during monolayer compression. the WP is - for the greater part - exposed to the vapour phase. If sufficient time is given to let previously wetted parts dry out, the advancing contact

-angle during subsequent expansion may be considerably larger than when the drying out does not occur.

During the lowering of the contact line monolayer molecules may be deposited on the WP (Fig. 2.5). Consequently, the meniscus, when traveling upward during expansion of the monolayer. meets a chemically different surface which may drastically influence the wetting behavier (Eq. (2.6)). It is difficult to understand that monolayer deposition should be absent in the WP method, as this deposi tion is the basis of the Langmuir~Blodgett technique [33]. In that technique a slide is pulled up and down through a monolayer-covered surface in order to farm mono- or mul ti layers on the slide. The presence of surfactant makes the hydrodYJ)ID!lics of a rnaving contact line - which is qui te complex for pure liquids - even more complicated.

Often the surface of a WP is treated mechanically or chemically in order to. imprave wetting properties and to minimize contact angle phenomena. A treatment usually includes roughening of the surface. Same studies have shown the importance of the magnitude and direction of roughness [34, 35]. Al though Huh and Mason [36] have shown that receding contact angles are reduced ·by surface roughness. they also showed that advancing angles rnay be increased, thereby increasing the hysteresis. Treatment of the plate by chemieals such as chromic acid will remave greasy spots and imprave wettability. Flaming also rernoves grease, but at the same time dries the plate's surface and may therefore have a bad influence on wetting.

From the above discussion it can be understood that we can only try to minimize contact-angle effects on the WP, but the measures taken cannot guarantee a LWM to be free of contact-angle phenomena. To this end much can be done by measurements additional to the LWM. In the literature we find suggestions for such addi tional measurements [28. 37. 38, 39]. We shall refer to these additional measurements as capillarography.

2.3.2 Capillarography

In our capillarography experiments a WP is moved up and-down. We call the resulting diagram of force versus penetratien depth a capillarogram.

Let us consider a WP wi th dimensions w. h, and t (width, height and thickness respectively). which is suspended from an electrobalance. We

-start by wetting the plate by total immersion, after which it is located at the interface (z=O, Fig. 2.3). At t

=

0 the plateis moved downwards with a constant speed; we shall call this the wetting speed vw. The z-value of the

lower end of the plate, the penetration depth, is indicated by D. At a certain penetration depth Dmax the direction of motion is reversed and again stopped at D

=

0.

Four forces contribute to the total (vertical) force F measured:

whe-re

w

_p F a Fb Ww Si nee

w

_p F is is is is W + F - F + W p a b w

the weight of the dry Wilhelmy plate, the force due to surface tension, the buoyancy force, and

the weight of water above the contact line. is constant throughout the measurement we shall

F F - W

m P

(2.11)

consider

(2. 12)

as the measured quantity. Fa satisfies

Fa = p.a.cos

e

(2.2)

where pis the plate's perimeter: p 2.(w+t). The buoyancy force Fb equals

Fb = p.g.w. t.D (2.13)

Since the plate is wetted by immersion before t = 0 (a normal procedure in nionolayer experiments) water attached to the plate at z

>

zcl may also contribute to Fm. We estimate this contribution Ww by (Fig. 2.6)

(2. 14)

where Wwo is the weight of the water that can maximal1y be taken up by the

WP

(zcl

=

Ö,

D

=

0).

Wwo can be determined by measuring the change in force on the initially dry p1ate, when after total immersion and successive retractton the liquid detaches from the plate. The height of the contact

line- if allowed to attain equilibrium- is given through Eq. (2.7), which makes Ww depend on a, 9 and D. If we assume that the water taken up surrounds the plate with a uniform layer of thickness ó we may approximate WWO by (Fig. 2.6)

WWO

=

p.g.p.ó.h

r---zc, /

/-;?;

~

~

/

_

["

Fig. 2. 6 Contribut ion of water present at the Wil he lmy plate aboue the contact line. For expLanat ion of symbols see text.

(2. 15}

Using Eq. (2.2), and (2.12)-(2.15} into (2.11} gives the expression for F m(D}:

p.a.cos e - p.g.(w.t + ó.p).D + p.g.ó.p.(h- zcl) (2. 16)

w

z

e

(

wo) D W (1 _.El.)

p.a.cos - p.g.w.t + ~ · + wo· - h (2.17}

At the start of a capillarogram measurement we have a situation similar to the normal use of a WP in a monolayer compression experiment (in the LWM), and we measure

p.a.cos eo + p.g.p.ó.(h- zclo> z

(

~)

p.a.cos eo + wwo' 1 - h (2. 18)

in which 90 is the initial macroscopie contact angle, related to the

initial height of the contact line zclo through Eq. (2.7). The next step is to move the WP downwards and upwards, now measuring a force given by Eq. (2. 16}.

At first, the capillarogram in the absence of a monolayer is considered;

-so a is taken constant. We will concentrate on the area of the capillarogram (Ac}. Buoyancy does not contribute to Ac' and of the third term on the right hand side of Eq. (2.16} only zcl contributes. Using Eq. (2.7} we get

Ac

==

f

Fm(D).dD = f[p.a.cos 9-

W~o.J~{l-

sin 9) ].dD (2. 19}

We will assume Wwo to be constant, so we neglect evaporation effects for the duration of the experiment.

We see that in general two terms contribute to Ac. and it depends on the specific properties of the plate whether the Wwo term can be neglected~ Eq. (2.19} also shows that Ac depends on the dimensions of the plate, the surface tension and maximum penetration depth (Dmax}. In order to arrive at a more practical quantity we introduce H, defined by

A

H c

p.Dmax (2.20}

H

indicates the averaged hys teresis, and has the dimension of surface tension. H can play an important role when we consider the measurement of capillarograms as additional to the

LWM

.

H

gives an indication of the error that occurs during monolayer compression - expansion measurements with the

LWM,

due to the moving contact-line.

If we assume a constant surface tension Eq. (2.19) may be combined with Eq.

(2.20} to

D

H

=

~-

I

max(cos ea- cos er}.dD max 0

W f2ä

IDmax

D w\

.v

.:::::...

(../

1 - sin e

max· .p pg 0 a ..; 1 -sin er}.dD (2.21} For smal! values of ea and er the second integral can be rewritten and Eq.

(2.21} then becomes

- - [WWO

~

1 - 1 - ]

H = a.(e - 9 }. h-.v ",....-!--_{2 - - 2}

-.e - -

₂

.e

Eq. (2.22) illustrates the relation between H and contact-angle hysteresis

(6 - 6 )

.

It should be noted that if H = 0, this does not mean that the

a r

contact angle is zero, but that the

6

and

6

are equal.

a r

So far a monolayer was not taken into consideration. Allowing a monolayer to be present at the liquid - air interface leads to several complications. The first one is the variation in total surface area when the meniscus profile is deformed by the forced motion of the WP. This causes the surface tension to change. Secondly, when compressed beyond the equilibrium surface pressure, a monolayer shows relaxation. This relaxation also influences the force measured, which may cause erroneous interpretation of contact angle phenomena. To circumvent this problem we apply a second WP which is located and kept at D

=

0. Since this plate also measures leakage and/or relaxation effects we expect the difference between the farces acting on the plates to be independent of these influences. This procedure has always been applied when the capillarograms were measured in the presence of a monolayer (§2.3.4). The second plate in fact serves as a control; the force F m2 on the second plate reads:

(2.23}

in which the subscript 2 refers to the second WP, and the subscript o indicates a constant value.

Although the main part of the surface-tension variation during the measurement of a cap i llarogram may be compensated for by the use of the second plate. it will not be removed completely. This is due to different contact angles at the two plates. In order to minimize the introduetion of extra contact-angle effects at the second plate we always took filter paper (FP) as the reference plate. This choice will be justified by the measurements (§2.3.4).

In conclusion, we can say that a solid is suitable for use as WP in the LWM, if the capillarogram is and remains free of contact-angle hysteresis.

-2.3.3 Capillarograms measured in the absence of a monolayer

In §2.3.1 factors of influence on contact-angle behavior were mentioned. In this section we present measurements in which the effect of several of these factors on the solid-liquid interactions bas been investigated.

Jfate:rf.al.s

The factor surface roughness was included in the choice of solids to be tested: microscope cover-glass (G), grounded glass (Gg), platinum (Pt, sandblasted from Prolabo, France). filter j:)aper (FP, Schleicher & Schull 589/1) and glass f i 1 ter (GF, po re si ze 40 - 80 j.llll). The dimensions and properties of these materials are listed in Table I. The materials were selected for several reasons. Microscope cover glass was chosen since it was eXpected to show considerable contact-angle artifacts. The Gg plate was

included because of its eXpected roughness effect. The Pt plate, being sandblasted. is commonly used as WP plate in monolayer studies. FP and GF are eXpected to exhibit small contact-angle effects.

The solids were cleaned either by a thorough washing procedure (FP) or by a cleaning procedure which included treatment with chromic acid, ethanol, and triple distilled water.

TABLE I. THE MATERIALS USED AS WILHELMY PLATE

MATERIAL w (mm) h (mm) t (mm)

w

+ st. dev. (mN) n WO -G 14.9 15.0 .15 0.029 +

_-

0.023 6 Gg 14.9 15.0 .15 0.021

_:!:

o.oos

6 Pt 19.7 15.0 .05 0.05 + _- 0.{)3 9 FP 15.0 15.0 .14 0.4 +

_-

0.1 9 GF 14.4 16.8 .35 0.34

_:!:

0.02 6

n d.enotes the ru.unber of d.eterminations of _wwO. The

Set-up

The triple distilled water was deposited in a Teflon Langmuir trough, and just befare each measurement i ts surface was sucked in order to remove impurities. The WP was suspended from an electrobalance (Beckman LM 600). which was placed on a vertically movable plateau. The plateau was driven by a stepping motor and gear. providing speeds from 3. 75 up to 56.25 mm/min. The penetration depth was measured through a linear potentiometer, fixed at

the rnaving plateau. Cap i llarograms we re measured on a XY - recorder (HP 7045). The temperature was 21 ± 1 0 C for all measurements.

The influence of 'time' (euaporation, adsorption, relatiue humidity)

A proper use of a WP requires the absence of evaporation and adsorption effects at the plate's surface. The effect of 'time' on the contact-angle hysteresis was measured for all plates. Immediately after total immersion, a cap i llarogram was measured. Then the plates were located at D

=

0 (in contact wi th the 1 iquid) and 1eft at rest for periods of &tr = 0. 10. 20, 30, 40 minutes. After each period of time another capi1larogram was measured. The pH value of the water was 5.3, the wetting speed vw

=

0.91 mm/s, the relative humidity was 40 ± 5 %. Fig. 2.7 shows the resulting H as a function of &t .

r

The G, Gg and Pt plate show an increase in H with increasing &tr. The GF and FP show no significant change in H with &tr. This may be caused by the capi11ary farces of the porous materials which cause the 1 iquid to rise easi1y to the upper end of the WP. Striking is the fact that the H values for FP and GF are positive. which implies a larger force measured during immersion than during retract ion. Repeti tive measurements confirmed the pos i tive values. In terms of contact angle this wou1d imp1y a receding contact angle exceeding the advancing one, which is hard to understand. Hydrodynamic effects are expected to cause a 1ower force on immersion and a higher one on retraction, which would thus workout the opposite way. Since contact-ang1e effects are sma11 for these plates, the measurement may be quite sensitive to changes

considerable (Table I).

in W •

w The W wo value for these plates is

At a higher relative humidity, 70 ± 5 %, the results of the FP and GF