2313
ULTRAHIGH-VACUUM TECHNIQUES
Ultrahigh-Vacuum Techniques in the Measurement of Contact Angles. 11.
Water on Gold' by Malcolm E. Schrader Nasal Applied Science Laboratory, Flushing and Washington Avenues, Brooklyn, New York 12861 (Received January 6 , 1970)
The contact angle of water on gold was measured in ultrahigh-vacuum and conventional vacuum systems by means of the vapor phase transfer technique. Moderate surface activation (heating in oxygen and vacuum) of a polished gold disk resulted in a large contact angle hysteresis, with a gradual decrease in both the receding and advancing angles occurring upon increased activation of the surface. In the conventional vacuum system a limit was reached at which a low receding but relatively high advancing angle was obtained. In the ultraclean (ultrahigh vacuum) system increasing activation resulted in a aero receding angle at first, and ultimately in a zero advancing angle as well. Contact angle measurements in the ultrahigh-vacuum system on gold films evaporated in situ yielded zero advancing and receding angles without prior surface activation. The hysteresis on the solid gold surfaces was due to the presence of areas of hydrophobic contamination on the hydrophilic gold surface, with the decrease in contact angle resulting from gradual removal of the contamination during surface activation.
Introduction Fox and Zisman2 have classified solid surfaces into the categories of high and low energy with respect to characteristics affecting their wettability. The highenergy surfaces include metals, metal oxides, and siliceous glasses, while those of low energy consist mainly of organic materials. As a rule, compounds which are liquid a t room temperature spread on high energy surfaces, since their surface tensions are considerably less than the surface energies of these solid substrates. An important characteristic of these high energy surfaces from the practical point of view is the tendency they have to lower their surface free energy by picking up hydrophobic organic contamination. It is consequently necessary to subject them to some sort of cleaning procedure before measuring their wettability. I n fact, the ability to spread water is widely used as a criterion of the cleanliness of a high-energy surface in this respect. I n 1964 White3 reported results of observations of the wetting of gold surfaces by water under atmospheric conditions. He obtained dropwise condensation from air saturated with water vapor under conditions of cleanliness during surface preparation and wettability measurement which were adequate to render base metals hydrophilic, Le., able to spread water. He ultimately reported the contact angle as 60 5 O S 4 I n 1965 Erb6 reported continuous dropwise condensation of water vapor in a gold-plated still a t atmospheric pressure. Fused quartz and nonnoble metals were observed to yield filmwise condensation under these same conditions. Shortly thereafter, Bewig and Zismane reported a water contact angle of 0" on a gold disk which had previously been heated to near melting in a flowing
*
atmosphere of hydrogen (at atmospheric pressure) purified of organic contaminants. When the precautions against contamination of the flowing hydrogen were relaxed, the contact angle was no longer zero. I n 1968 Erb7 reported the results of contact angle measurements a t atmospheric pressure on a series of gold surfaces prepared in a variety of ways. There was a very large scatter to the data, with an average contact angle of about 12" under certain conditions and 63" under different conditions. He chose 63" as his best result and attributed lower angles to various types of hydrophilic contamination. Wes have previously investigated the contact angle of methylene iodide on glass, using ultrahigh-vacuum techniques in surface preparation followed by in situ measurement of the contact angle by means of a vapor phase transfer procedure. The purpose of the ultrahighvacuum technique was to eliminate all traces of water in order to obtain and measure contact angles of methylene iodide on truly anhydrous surfaces. I n the present work the same technique is utilized, but with the objective of measuring the contact angle of water on a gold surface completely devoid of organic contamination. This is made possible by the complete absence
(1) Presented at the 157th National Meeting of the American Chemical Society, Minneapolis, Minn., April 1969. (2) H. W. Fox and W. A. Zisman, J. Colloid Sci., 5,514 (1950). (3) M. L.White, J. Phys. Chem., 68, 3083 (1964). (4) M. L.White and J. Drobek, ibid., 70,3432 (1966). (5) R.A. Erb, ibid., 69, 1306 (1965). (6) K.W. Bewig and W. A. Zisman, ibid., 69, 4238 (1965). (7) R.A. Erb, ibid.,72,2412 (1968). (8) M.E.Schrader, J. Colloid Interface Sci., 27, 743 (1968). The Journal of Physical Chemistry, Vol. 74, N o . 11, 1970
MALCOLM E. SCHRADER
2314 of any organic components in the all metal and glass ultrahigh-vacuum system.
Experimental Section Ultrahigh-VacuumRuns. The vapor phase transfer technique for contact angle measurements of true liquids under ultrahigh-vacuum conditions is described elsewhere.* Briefly, the method consists of activating the surface in a clean ultrahigh-vacuum system, admitting vapor t o the system, condensing the vapor in the sample chamber by means of a cold finger, and depositing a drop on the sample surface through magnetic manipulation. The cold liquid in the finger is removed either immediately after or before deposition of the drop on the sample surface. The contact angle is then read by means of a goniometer eyepiece mounted on a telescope. The 15 1. per sec ion pump, ion gauge, and bakeable valves were the same as used previously.8 I n this work a quartz sample chamber and manifold were used for the experiments with polished gold disk surfaces. Graded seals joined the manifold t o the Pyrex breakseals and adaptor flange on the metal cut-off valve. The graded seals were situated approximately 22, 23, and 30 in., respectively, from the sample. The entire system was capable of attaining a pressure of 2 X 10-lO Torr or lower. The distilled water was thoroughly degassed in vacuo before sealing off into the break-seal tubes. For the experiments on evaporated gold films, a spherical Pyrex sample chamber was utilized. Tungsten electrical leads were spot welded t o a tungsten conical basket type heater placed in the chamber between the walls and cold finger. A ball of 99.999+% pure gold was placed in the basket and the system given a preliminary bake-out,, The tungsten basket was degassed through resistance heating and the entire system baked out until a vacuum pressure reading in the 10-lo Torr decade could be obtained a t room temperature. The basket was again degassed electrically a t red heat for 2 hr before evaporating the gold. The evaporation procedure consisted of raising a polished disk of fused silica or carbon magnetically to the top of the cylindrical portion of the sample chamber, then increasing the heater current until gold evaporated and condensed onto the substrate surface. Towards completion of evaporation the pressure was approximately 9 X Upon cessation of evaporation it dropped rapidly to the 10-lO decade. The contact angle was then measured in the usual fashion. Conventional High-Vacuum Runs. I n some experiments a manifold with conventional high-vacuum glass stopcocks lubricated with high-vacuum grease was used instead of the bakeable all-glass and metal ultrahighvacuum system. This manifold was evacuated with a two-stage oil diffusion pump backed by a mechanical forepump. One liquid nitrogen trap was sealed between The Journal of Physical Chemistry, Vol. 74, No. 11: 1970
the two pumps and another between the diffusion pump and the Pyrex manifold. A quartz sample chamber, identical in design with that used in the ultrahighvacuum system, was connected to the manifold by means of a graded seal. Preliminary Preparation of the Gold Disk Surfaces. The samples consisted of polished gold disks of 99.999+% purity. The gold disks were abraded, polished to a mirror finish with an aqueous slurry of diamond powder on clean Gamal cloth, then rinsed in distilled water. Residual diamond powder was then removed with a fresh piece of Gamal cloth wet with distilled water, followed by wiping with wet untreated lens tissue. The surface activation (heating in oxygen and vacuum) performed in the vacuum apparatus will be described separately for each sample in the tables in the Results section. I n all runs where heating in air was followed by heating under vacuum (at the same temperature) the temperature was maintained during evacuation.
Results Hysteresis. A striking feature of the results was the hysteresis of the water contact angle which was observed during the various stages of surface activation of the gold disk. The hysteresis effect was observed in a number of different ways. ( a ) Freezing out the water vapor in the vacuum system after the drop was on the gold surface. As the drop evaporated, a receding angle was observed. Manipulations of the vapor pressure in vacuum eliminated the possibility that this was due to the effect of vapor pressure on the contact angle. ( b ) Observing the contact angle change with time. The drop may continue t o spread for many hours. ( c ) Agitation of the drop. This caused immediate partial spreading. Experiments with Gold Disks. Upon raising the temperature and time of heating of gold in air followed by evacuation, the receding angle decreased to zero, while the advancing angle decreased t o about 20-30". With more activation, the advancing angle decreased further, while the receding remained zero. I n the conventional vacuum system (Table I), a limit was reached in the decrease of the advancing angle. Further activation either raised the angle or ceased to lower it. I n the ultrahigh-vacuum system (Table 11) on the other hand, increased activation decreased the advancing angle to zero degrees. Experiments on Gold Films Evaporated I n Situ. The contact angle of water on a gold film (Table 111) evaporated onto the surface of a polished fused silica disk was zero. The system was evacuated again and another layer of gold deposited on top of the original. The contact angle was zero once more. A few minutes was sometimes required for the drop to reach the equilibrium zero value.
2315
ULTRAHIGH-VACUUM TECHNIQUES Table I1 : Contact Angles on Gold Disk in Ultrahigh-Vacuum System
Table I : Contact Angles on Gold Disk in Conventional Vacuum System
Surface activation
Drop no.
Vacuum, looo, 3 hr
1
Air, 570°, 1 hr, followed by vacuum, 580", 1 hr Air, 580", 2 hr
1
2 1
2
Vacuum, 600°, 1 . 5 hr
1
2
Air, 720°, 2 hr
1
2
Drop history after initial reading: time lapse from Advancing previoua reading or mechanical angle, agitation deg
29 19
20 18
28 25 14 23 11 22.5 11 35 27 20 15 30 28 19
1
2
*
0
Mechanical agitation
...
14
Mechanical agitation
... Mechanical agitation Mechanical agitation
0
28 22 22 6 26 11
Advancing angle, deg
Surface activation
Drop no.
Vacuum, 560", 2 hr
1
Air, 1 Torr, 550°, 1 . 5 hr Air, 710°, 2 hr, followed by vacuum, 710", 1 hr Air, 715", 2 . 5 hr, followed by vacuum, 715O, 2 . 5 hr
1 2
G
...
5
.,.
1
2 0 3 0 5 0 6.5 2
31 16
2
1 2
1 0
Mechanical agitation
...
0
Mechanical agitation *
...
#
.
.*.
1 min
... 5 min
...
...
5 min
... 1 min 5 min 30 rnin
,,.
Mechanical agitation 10 min 30 rnin
~~
Table 111: Contact Angles on Gold Film Evaporated in Ultrahigh-Vacuum System
1 min Mechanical agitation
48
37 Vacuum, 700", 2 hr
...
Receding angle, deg
Drop history after initial reading: time lapse from previous reading Receding or mechanical angle, agitation deg
Nechanical agitation
...
Run no.
*..
3 min 7 min Mechanical agitation
1
2
Mechanical agitation
3
I n another experiment the gold was deposited on a polished graphite surface. A zero contact angle was again observed a few minutes after deposition of the water drop. The contact angle of water on the graphite disk surface which was shielded from the gold vapor flux was approximately 22".
Substrate
Polished silica disk Polished silica disk Polished graphite disk
FiIm no.
Advancing angle, deg
Drop
1
First
2
Subsequent First
1
First
1
Subsequent First Second
Spread with low angle 0 10 0 Spreading at low angle 0 0 Spread at
