OPTICAL PROPERTIES COLOUR AND LIGHT
Man ability to discriminate colors increases both his enjoyment and appreciation of the world around him. Because color is such a significant factor in the appearance of an object, it is an important characteristic of any paint. Appearance is one quality of a product that every customer can judge for himself. No matter how good the physical properties of a paint, if the color does not meet the expectation of the customer, he will think the finished product unsatisfactory. Color is often thought to be a property of the paint itself, and to a degree, it is. Actually it is more complex than that for color of a paint depends on three things (1) the spectral reflectance of the paint, (2) the spectral composition of the light in which the paint is viewed, and (3) the spectral sensitivity of the eye of the observer for these reasons matching the color of a paint or other material requires considering all three and not merely the spectral characteristic of the material.
This chapter deals with ways in which the color analyst can obtain consistent measurements of the colors of his specimens. It will also touch briefly on the mechanism of vision where this helpful to understand the requirements for evaluation of appearance. The physics, physiology, and psychology of color, however, are broad subjects, and only enough disground for understanding the development of test methods. Readers desiring to purpose these subjects in detail should consult an appropriate text.
Light is electromagnetic radiation weighted by the response of the normal human eye. It involves the portion of the electromagnetic spectrum between approximately 380 and 780 nanometers (nm) (fig 1). Note that visible radiation occupies a very narrow band in the electromagnetic spectrum between ultraviolet and infrared radiation.
In the past the most important sources were daylight and candlelight, and in spite of the construction of many windowless buildings, daylight is still an important source since most objects at some time or other are viewed outdoors. The composition of daylight however is quite variable, depending upon the hour of day, the season of year, and very importantly, the amount of cloud cover. Other sources must replace daylight at the end of the day or when there are no windows to illuminate dwellings and offices. Flame sources were first used but of course, have now been supplanted by electric lamps. But incandescent lamplight is generally preferred of use indoors because it imparts the same soft, mellow effect as that of candlelight. In offices and in places of business where high level of illumination are needed, duplication of daylight is usually preferred, and for this purpose fluorescent lamps can simulate daylight to a degree, but the spectral character is not identical. It should be noted that while fluorescent lamps can be used for visual color evaluation they are generally unsuitable for use in instrumental color measurement fluorescent lamps are made the most common of which is known as cool white. Even this color has deluxe and superdeluxe versions the distinct improvement being an increase if radiation in the red portion of the spectrum light sources are described most accurately by curves are described most accurately by curves or tables of their relative spectral power distributions.
In order to standardize color evaluation work, the CIE (commission International de Eclairage international commission on illumination) in 1931 established three standard Illuminants designated A,B, and C (fig2). Illuminants A is intended to represent ordinary tungsten lamplight, Illuminants B and illuminants C are derived by using liquid filters specified by the CIE with Illuminant A illuminant C does not have provided data for what has been called reconstituted daylight of various correlated color temperature. D65 is suggested for most color measurement, but D 75 can be used when a bluer illuminant is desired, and d55 can be used when a more yellow illuminant is desired. Other distributions of D illuminant can be also used note that color evaluation for D illuminants is easy when spectral data are available. On the other hand, it is difficult to produce a D illuminants for use with a colorimeter, and most measurements with colorimeters continue to be made for CIE illuminant C.
The term color temperature has been often used to identify radiation from incandescent lamps and other sources. As a body is heated it begins to radiate, first in the infrared, then in the visible. As heating continues the color shifts from red, to yellow, to white, to blue white at the highest temperature. Figure 3 shows the blackbody locus of color temperatures potted on the CIE chromaticity diagram. The chromaticity is a function of temperature. The temperature is expressed in kelvins and abbreviated K. A radiator whose emissivity is less than that of a blackbody, such as a tungsten lamp, called a gray body. Since its emissivity is a function of wavelength it will not have exactly the same chromaticity as a blackbody. However, when its chromaticity most nearly matches that if the black body, it is said to have a correlated color temperature of the same numerical value as that of the blackbody. The color temperature of the sun is about 5000k that of a cloudless summer sky is between 20,000 and 25,000 k. The color temperature of a household type 100 W tungsten lamp is about 2850k (CIE Illuminant A). Tungsten melts at about 3600k, so the color temperature of an incandescent tungsten lamp cannot exceed this value and practically is limited to about 3200k.
Color Matching Booth
Because of the varied chromaticities of different natural and artificial sources, it is essential that visual color matching boot. This device allows the color analyst to make a visual match at any hour of the day or night with the confidence that the match will be equally satisfactory in a plant thousands of miles away.
A paint of specified color should be matched with a paint having the same combination of pigments. Frequently a different combination of pigments will be used. Matches of this type of illumination such as incandescent lamplight, but may not match under another type of illumination such as daylight. This phenomenon is known as metamerism, and the colors are said to be metameric. Such paints are found to have different spectral reflectance curves (fig 4). Only when colors have identical spectral characteristics can they be expected to match under all types of light this is why the same pigment formulation should be used when remarking the color. Whenever pigment used for the match have different spectral characteristics from the sample, the resultant color match should be verified under at least two different types of illumination, for example, illuminant A and illuminant C. if the match is not satisfactory under each type, spectrophotometric analysis of the two formulation should be adjusted to minimize these differences.
The sensation of color is evoked by visible radiation, but ultraviolet and infrared radiation can have important effects in paint. Ultraviolet radiation promotes de composition of binder, is the principal stimulus of fluorescence of certain pigments, and is an aid to identification and analytical determination of certain ingredients of paint. Infrared radiation is a factor in heat transfer and also an aid in the identification and analysis if chemical compounds.
Fluorescence some materials encountered in nature have the unique property of fluorescing when irradiated by ultraviolet energy (invisible to the eye) contained in the source. In other words, the object will emit light in the visible range even though the radiation used to energize it is not itself visible. The effect of fluorescence is to increase the apparent reflectance. Most colorimeters and spectrophotometers have not been designed to properly evaluate the colors of fluorescent materials, and most test methods are applicable only to nonfluorescing materials. It is usually sufficient in most applications to expose the paint or other coating to ultraviolet radiation to determine that it does not fluorescence appreciably.
There are recently developed pigments that use fluorescence to convert ultraviolet radiation, as well as visible blue light, to green orange, or red light, thus increasing the luminance in these regions to levels much higher than would otherwise be obtained in daylight fluorescent they have low hiding power and should be applied over a white background for good results. Table 1 lists the fluorescence of some paint materials.
Infrared Radiation Every hot object emits infrared radiation, and while this radiation is not visible to the eye, it is reflected by certain materials including paint. Object painted to conceal them from observation need to reflect infrared radiation to the same extent as their surroundings in order to escape detection. For this reason evaluation of infrared reflectance is important, and devices for detecting infrared radiation have been developed, especially for military operations.
Absorbed radiant energy, invisible as well as visible, is converted into heat. Hence, to the extent that painted surfaces absorb, reflect fluence the temperature of a building, increase heat transfer from a radiator, reduce heat transfer into a refrigerator, or reduce evaporation from a gasoline storage tank.
The perception if objects and their colors depend on the reaction of both the eye and the brain. Vision is thus psychophysical phenomenon physical in that light must reach the eye, psychological in that the brain must interrupt the nerve signals to form an impression of the light in the mind. The psychological factor determines whether a given color combination is pleasing or displeasing. The mechanism of seeing is physical the impression is psychological. Objective color measurement is confined to the physical.
Psychological factors cannot be evaluated objectively. For example, the apparent color of a specimen may be changed by the color of the adjacent area. The effect can be observed by placing one half of a sheet of pastel colored paper in the middle of a large area of a high reflectance color, such as white, and the other half on a low reflectance color, such as black. The specimen on the white background will appear stronger and darker than the identical specimen on the black background. Reversing the specimen reverses the direction of their apparent color difference. The kind of color evaluation cannot be made instrumentally.
Another example of a psychological color factor involves a phenomen called chromatic adaptation or color constancy. When one goes outdoors from a windowless office, he requires a few minutes to adapt to a large increase in light level. Also there probably will be a change in the spectral composition of the light reflected from his friend s clothing, hair, and face will be different. But one has learned to expect this kind of change and, therefore, thinks little about it, except under spectral composition of the light incident on them. By the same token the eye brain combination cannot be trusted to pick date. Most persons will select a color considerably stronger than the original they think they remember.
The human eye is still an excellent device for comparing colors, but only for direct side by side comparisons. Even here we find that result depend on the observe in that some observes report large differences where others claim that there is hardly any difference at all. What the reason for such differences is among observes? The explanation lies in the fact that the color sensitivity of the human eye differs among individuals. Some individuals can be identified or the by tests for color vision such as those devised by Ishihara2, Hardy et al or the ISCC.
The standard observer
In order to evaluate colors consistently, a standard observe was defined by the CIE in 1931. One might assume that to define a standard observe representative of entire human population might require evaluation of the spectral sensitivities of thousands of individuals. Actually the difficulty of obtaining such data precluded using many observe, and the standard observer is based on data from only17 individuals (all trained color matchers). The spectral responses of the CIE 1931 standard observer are defined by the tri stimulus values of the spectrum plotted in fig. 5. At any wavelength the values of are the amount of primaries, x (red), y (green), and z (blue) required to produce the color of the spectrum at that wavelength. Any part of the spectrum at that be produced by the positive amounts of these primaries, which are imaginary because no additive mixture of real lights will duplicate all of the spectrum colors. In the actual experiment with spectrum lights negative amounts of some standard lights were required to match other spectrum lights. That is in the attempt to match a spectrum color with an additive mixture of the three primary lights, it was found necessary to add some light from the primaries too the test color. In effect, this subtracted light from the primaries, producing negative amount however, the CIE decided not to use any negative quantities, and the data were transformed so that all quantities would be positive. Another restriction was imposed, namely, that all of the luminosity should be contained in one of the standard lights, which these restrictions on the experimental data, the CIE 1931 standard observer was born and it has performed very successfully, if not perfectly, for 40 years.
The data for the 1931 observer were obtained with commonly used visual photometer of that era in which the field of view subtended an angle of only 2 deg at the eye of the observer. Actually the usual field of view in commerce is more nearly 10 deg, and data showed that the spectral sensitivity of the eye is different for these two viewing fields. For this reason additional data were obtained with a 10 deg, field and the CIE established the 1964 Supplementary Observer. However, it should be pointed up that, for many colors, differences computed for the two fields are small and that most filter colorimeters are still based on the 1931 Standard Observer.
No individual is average and very few have color vision identical with that of the Standard Observer. This circumstance does not make the Standard Observer less useful, but it must be kept in mind when making visual examinations of colored objects. Color blindness is a term used to denote color vision grossly different from that of the average individual. About three people in 1000 are completely colorblind. There are other types of color defectives who see color slightly, though significantly, different than the average observer although about 95 percent of all people see color approximately the same as the Standard Observer. Since, in many instances, Color matches are made by visual observation, it is important to know just how well a particular observer compares with be average. As mentioned before, color aptitude tests have been devised to test the color matching ability of individuals required to make such judgements.
Reflectance and Transmittance
What happens to light when it strikes an object Some may be reflected, some may be absorbed, and if the object is transparent, some may be transmitted. The reflected light may be concentrated at an angle numerically equal to the angle of the incident ray, or scattered uniformly in all directions, or distributed between these two extremes which are known as specular (minor) reflection and diffuse reflection. A highly polished white metal, such as silver, reflects as much as 99 percent of the incident light in a specular direction. A white powder, such as magnesium oxide or barium sulfate, scatters light in all direction, and it, too, reflects as much as 99 percent of the incident light. Specular reflection is related to gloss diffuse reflection is related to lightness and color. Transmission can also be regular (specular) or diffuse, depending on whether or not light is scattered in passing through a material. The reflectance or transmittance of light by an object is usually wavelength dependent. A spectro photometer is used to provide information on the spectrally selective character of a material. Figure 6 shows several examples of spectral curves of colored paints. This type of chart provides a profile of the spectral reflectance or transmittance of an object. Trained analysts can look at a spectral curve and obtain valuable information, but spectral data alone are unsatisfactory means for color identification.
Gloss is often described as the attribute responsible for the shiny or lustrous appearance of an object. A more formal definition is given by the CIE.
Gloss (of a surface) Directionally selective reflecting properties responsible for the degree to which reflected highlights or the images of object may be seen as superimposed on the surface.
Gloss is second only to color in importance as an appearance characteristic of an object. Like color, gloss can be subdivided into several aspects depending on the particular characteristic that is being considered. Hunter identified five aspects of gloss and function of reflectance by which they can be evaluated (table 1). Experience has shown that no single objective measurement requirement of an object be determined and the most useful aspect selected. It must be pointed up, however, that the measurement techniques for the various aspects are not developed equally well.
Aspect of gloss
Each aspect of gloss is associated with the shininess of the object, but depending on other appearance characteristics or the use of the object, one aspect of gloss may predominate, or the distribution of reflected flux may have to be investigated. Goniophotometry is the technique used for investigating the geometric distribution of reflected light. Goniophotometry is required in any research type investigation in order to determine the nature of the geometric distribution of reflected light.
Specular gloss or specular reflectance is simply the fraction of the incident light reflected from a surface in the mirror direction within a specified angular tolerance. This is the aspect of gloss that is measured most frequently because it is the one for which an instrument is constructed most easily. In practice the angle of incidence is specified as well as the divergence angle or aperture of source and receptor (fig 1). For simplicity the geometry is identified frequently by the giving only the incidence angle, but the associated apertures play a vital role in instrumental readings.
Sheen is often described as specular gloss at a near grazing angle of incidence. However, sheen is defined in ASTM Designation E 284, standard Definitions of terms relating to Appearance of Materials, as specular gloss at a large angle of incidence or an otherwise mat specimen, that is one whose specular gloss is very low at low angle of incidence such as a mat wall paint. The usual angle for sheen measurement is 85 deg from the perpendicular to the specimen. This is about the maximum angle that can be used without excessive difficulty in positioning the optics to illuminate and view the specimen.
Contrast gloss is defined as the ratio of specular to diffuse reflectance. It is the aspect that takes into account the relation between specular gloss, reflectance factor, and glossy appearance. It can be used to explain why a black object appears to be more glossy than a white object having the same specular reflectance. In the case of the black object the reflected highlight is not diluted by high, diffuse reflectance from the surround, as is the case with the white object. Contrast gloss can be used as a means to make this differentiation, but is seldom required to compare gloss readings with visual estimated of glossiness for both black and white specimens.
Absence of Bloom Gloss
This type of gloss is associated with high gloss paints and similar materials that on occasion exhibit a hazy or milky appearance adjacent to the reflected highlight. It is observed most easily in a darkened room by viewing the image of the filament of an unfrosted lamp (operated at reduced voltage) reflected in the surface of the specimen. Some people try to differentiate between haze and bloom. Bloom is said to appear on the surface as efflorescence and can be rubbed off. Haze on the other hand is attributed to scatting by pigment particles just beneath the surface. for either haze or bloom the measurement technique is the same. basically it requires a specular gloss aperture to differentiate between the small differences in near forward scattering associated with bloom or haze.
A very sensitive measure of haziness can be obtained by providing a glossmeter with a specular stop, so that only the reflected rays that are, scattered beyond the usual specular aperture are accepted for measurement.
In general high gloss surfaces are best measured for specular gloss with an incidence angle near the perpendicular, and conversely low gloss surfaces are best differentiated by using an incidence angle close to grazing. This axiomatic truth is found in a paper on gloss measurement by fund. High gloss measurement also requires that the divergence of rays accepted for measurement be as small as possible consistent with ease of specimen preparation and positioning on the instrument. With small apertures warped of curved specimens are likely to give erroneous readings.
The light flux entering a glossmeter receiver after reflection from a perfectly diffusing surface is a direct function of the size of the receptor entrance window. Readings for low gloss specimens not perfectly polished will be a function of the size of the source image formed at the receptor entrance window. If the source image formed by reflection from a polished specimen nearly fills the window, instrument readings. The effect on gloss readings of variations in 60 deg glossmeter geometry is discussed by Hammond and Nimeroff.
Another consideration that influences the choice of the receiver source aperture ratio is the planarity or flatness of specimens. For specimens having a wavy surface, the receiver source aperture ratio should not be too small otherwise specimens ranked by eye as having high gloss will be rated too low by the instrument. The gloss of a warped specimen is always difficult to measure accurately because the flux reflected from the specimens ranked by eye as having high gloss will be rated too low by the instrument. The gloss of a warped specimen is always difficult to measure accurately because the flux reflected from the specimen is not likely to be centered on the receptor entrance window. A warped specimen distorts the reflected flux distribution so that the glossmeter is likely to indicate an erroneous gloss value.
Standardization of specular gloss measurements necessitates specification of the angles of illumination and receptors field angles or apertures. The aperture sizes must be specified with tolerance. Specification for these apertures was often omitted from the appartures requirements of early test methods. In some cases, instrument type was specified, but this was found to be inadequate.
Early instruments for gloss measurement were visual instruments, and, as a result objective gloss evaluation was not carried out with great frequency until the production of the photovoltaic cell in the 1930 s. Since then a large number of photoelectric glossmeters have been designed and built, and several designs of photoelectric goniophotometers have been marketed. Distinctness of image evaluation. However, has yet to be accomplished to complete satisfaction by a means other than visual.
Gonio means angle, and a photometer is an instrument that measures light so a goniophotometer is an instrument for measuring the angular distribution of reflected or transmitted light. This type of instrument is used regularly in the research laboratory to investigate the geometric distribution of light flux (fig 3). This instrument can be also used to provide gloss data for a wide variety of angles and apertures. When goniophotometric measurement are desired, reference should be made to ASTM Recommended Practice E 166, Goniophotometry of Transmitting objects and Materials, and Recommended Practice E 167, Goniophotometry of reflecting objects and Materials. Analysis of goniophotometric curves was treated by Nimeroff.
Goniophotometry is not new. A goniophotometer was devised by Bouguer in an attempt to corroborate Lambert s (consine) law for uniform diffusers by experimental evidence. A number of goniophotometers have been described since this earliest known instrument. Only a few will be mentioned here. McNicholas described a sophisticated visual goniophotometer with provision for diffuse or unidirectional illumination and Multiplan view. Wet laufer and Scott described a photoelectric goniophotometer with rectangular source and receiver apertures however they receiver apertures, whereas for replication of data the separate dimensions are required.
Two types of goniophotometers are currently available commercially. In the Gardner goniophotometer the receptor direction is fixed, and the source and the specimen angles are set manually as desired. In the Hunterlab instrument the geometry is selected speeds, and the receptor signal is plotted on rectangular coordinates as a function of angle.
All current specular glossmeters are photoelectric instruments. They embody the geometric requirements set forth in the various test methods. In the paint industry the geometeries of 20, 60, and 85 deg are used exclusively. The geometeries are designated by the angle of incidence, but there are also receptor aperture changes for each geometry. Some materials are measured regularly for gloss with geometeries different from those used for paint, such as 45 deg for ceramic material, 75 deg for paper. A good way to evaluate an instrument is to use it in a particular application.
Instruments for gloss measurement are made in various degrees of complexity and configuration. One class of instruments is known as portable, by which is meant that the instrument in this class is battery operated so that they can be used where the usual electric power line is not available. Some instruments have a self contained photometer and readout device (microammeter), notably the portable class.
Other instruments are constructed with the optical unit separate from the photometric unit. This type of construction permits use of a common photometric unit to read out the signals from a number of optical units that can be either plugged into the photometric unit is provided with a switch so that the operator can change easily from one optical unit to another. Some instruments have been constructed with several optical units built into a single housing, thereby permitting a change from one geometry to another by the flick of a switch. This multiangle design also has the advantage of eliminating the connecting cables that are required when separate optical units are used with a common photometric unit.
Instrument of different manufactures vary in their complexity, precision of measurement, and cost. schreckendgust and Gowing described a high speed, direct reading. Digital readout is convenient, rapid and less likely to be recorded incorrectly, but it is also more costly. However, the saving in operator time may offset the increase in instrument cost.
Pellegrini patented a visual instrument, which was later improved by Harrison. This instrument has been used widely in the porcelain enamel industry and is also useful for evaluating the image quality of paint specimens.
Specular Gloss Methods
Special specular gloss methods have been developed for a number of different materials such as ceramics, paper, plastics and the like however. The discussion here will be confined to paint methods.
A method for classifying paints according to gloss was developed originally by Hunter and Judd and this same year the first edition of ASTM Method D 523, Test for specular Gloss, was published. This method originally contained only the 60 deg geometry for general classification, subsequently it was expanded to include the 20 deg geometry for better differentiation of high gloss specimens and the 85 deg geometry for sheen or low gloss evaluation.
The ASTM method has been adopted by the America National standards Institute (ANSI) and the basic requirements of apertures and angles have been incorporated into Federal Methods and into an Australian method. They are being currently considered for use by the International standards organization (ISO). Other geometries have been adopted by some other countries, notably Great Britain, where a 45 deg geometry designated as British standard (BS) 3900 has been used for high gloss evaluation.
Two Parameter Methods
It has long been recognized that a measurement of specular gloss at one angle of incidence and for one receptor was insufficient to characterize the appearance of a surface. Yet the recording of a goniophotometric curve takes considerable time and equipment and may provide more data than are needed. The logical compromise is to make gloss measurements at two angles of incidence or for two different apertures or a combination of both and combine the result in some form to provide a two parameter index. A procedure of this type was developed by Nimeroff originally to provide a more satisfactory evaluation of the gloss of clear finishes over wood.
Distinctness of Image Methods
Images gloss measurement by objective means has not provided sufficient reproducibility to warrant publication of a method. The only published method is ASTM C 540 67, Test for Image Gloss of porcelain Enamel surfaces, a visual method. Although the method was developed for porcelain it can be applied to organic enamels.
The general requirements for gloss standards are similar to those for color standards namely, the standard should be uniform over its surface, should have permanence, and should not be damaged easily in use. In addition gloss standards should be quite planar and should have surfaces that provide reflected light flux distributions closely duplicating those of paint specimens.
Material for standards
Vitreous or gloss like materials is preferred for use as standards because of their permanence. Polished black gloss is used as the primary gloss standard because its specular reflectance can be computed from its index of refraction. Liquid surfaces of different indexes of refraction were proposed by Moore and Hunter. ceramic tiles of the 4 ½ in. (11cm) square type commonly used for bathroom walls have been used in large quantities. The main difficulty is to obtain tiles that are reasonable flat. For bathroom walls a convex surface is preferred. Specimen of ground sand blasted and acid etched glass have been used as standard. The first two frequently have flux distribution that depart markedly from paint specimen having the same glossmeter readings an acid etched glass surface is particularly good for providing high gloss standard of high planarity, but it is difficult to obtain with high uniformity, in fact it is difficult to obtain with high all because the glass must be etched with a hot combination of concentrated hydrofluoric and sulfuric acids.
Porcelain enamel on steel can be also made in a wide range of gloss. Such standards are less fragile than ceramic tile, and there seems to be little difficulty in obtaining uniform and relatively plane specimens.
Calibration of standards
The upper portion of each gloss scale is set by using polished black glass standards whose specular gloss has been computed from its index of refraction by using the Fresnnel snell equation. A factor may be introduced to provide the desired scale value. All other gloss standards are calibrated relative to a polished black glass primary standard by measuring them on a glossmeter whose apertures and angles conform accurately to the requirements of the published method.
MASS COLOR AND TINTING STRENGTH
The color, when viewed by reflected light, of a pigment vehicle mixture of such thickness as to obscure completely the background is called the mass color (MC) of the pigment. Sometimes this is called mass tone, over tone, or self colors. MC encompasses lightness, hue, and saturation. Although generally applied to white pigments.
The power of a pigment to color a standard paint or pigment is called the tinting strength (TS) of a pigment. When applied to white pigments, Haslam suggested a better term would be lighting power or brightening power, but this is not ordinarily done. Sometimes, for white pigments, the definition for TS is given as the ability to resist discoloration by colorants.
Basic Factors Producing MC and TS
The TS of pigments is often used as a guide for estimating hiding power. Consequently, it is not surprising to find that the same basic factors given for hiding power apply for MC and TS. Thus the refractive Index of the pigment is the primary factor followed by its particle size and its degree of dispersion in the vehicle. For colored pigments, the inherent light absorbing ability is also a primary factor.
Mixing Pigment and Vehicle
TS determinations involve through mixing of definite quantities of white pigment and colored pigment. The preliminary mixing is sometimes done dry, after which vehicle is added, but it is probably better practice to first prepare separate pastes in the vehicle, and then to mix the pastes. Thus, for colored pigments, MC is determined first after which tinting strength is determined with the paste prepared for the MC test. For white pigment lampblack or ultramarine blue is the usual tinting pigments, white for colored pigment, it is zinc oxide.
The mixing of pigment and vehicle may be carried out by a number of methods as will be described for all methods great care must be exercised to avoid contamination of one color with another.
Spatula and Muller Methods
Ayers fond that a muller is more reliable than a spatula and that the texture of the rubbing surface may vary considerable without affecting the accuracy of the test.
Stutz in a study of the TS of white pigments also found a muller to be better than a spatula but that a weighted muller is no better than an unweighted one, as shown in table 1. In the rubbing operation, the muller should travel over areas about 3 in. wide and 12 in. long, being pushed up one side and pulled down the other side of the strip so that all color particles receive the same amount of rubbing. By one rub is meant one complete up and down circuit.
The number of rubs depends upon the type of pigment and actual factory grinding operation. For example, Ayers found that excessive mulling changes the MC and TS of many iron oxide pigments and some chemical pigments. Table 2 shows relations between different methods of paste preparation obtained by stutz on TS of zinc oxide and basic carbonate of lead.
Hoover Automatic Muller
In the Hoover Automatic Muller the mixture of pigment and oil is rubbed between two glass disks, one stationary the other rotated by a motor. Since the rotation is about the centers of the disks, paste at the center receives less mulling than paste near the edge. To minimize this it is helpful to spread the paste in a ring halfway between center and edge. This method of paste preparations is becoming favored over either type of hand mulling.
Laboratory Roller Mill
a small three roller mill that has been found useful for grinding small laboratory batches of paint. The rolls of this mill are 4 in. in diameter by 8 in. in length. Batches as small as 5 g have been prepared.
Pall Glass Mill
The pall glass Mill was designed for preparing small quantities of pigment pastes without contamination by metal spatulas or metal rolls. It consists of a ground glass stopper and the female section of a heavy glass joint.
The mixed but unground ink is placed in the female section and the plunger inserted. The male section is rotated by a small motor at about 150 rpm. The grinding pressure ranges from 20 to 30 psi. Quantities of 1 to 8 g can be handled.
The mill is said to be superior to hand milling by saving time and by greater development of TS it is superior to roller mill grinding in freedom from contamination although not as efficient in dispersing very hard grinding pigments. No comparison with automatic mullers is available.
Consistency of the paste influences the efficiency of the rubbing procedure and consequently the degree to which the color is developed. In a study of the relationship between plant grinding and laboratory testing on the color of iron oxides, Ayers showed that the higher the consistency of the paste, the more rapid is the development of color. Above a certain point, increase of consistency had no appreciable effect. At the point of maximum reflectance, 700nm (fig 1), the low consistency paste had a reflectance of 26 percent, while the high consistency paste had a reflectance of 23 percent.
Stutz gives the relationship found in table 3 between the amount of oil used during rubbing and the tinting strength of a sample of zinc sulfide. Note that the final amount of oil is the same in all cases, and that the thickest paste is 4 percent greater in strength than the thinnest one.
To test at equal consistency, preliminary tests would have to be made. However, for convenience in routine work, definite weights of pigment and vehicle may be used for each type of pigment. The weights vary from, for example, 0.3 g for carbon and lampblack to 2.0 g for chromium oxide and white pigments. The proper amount of vehicle depends upon the oil absorption of the pigment and the type of vehicle. Refined linseed oil is generally acceptable. Experience has shown, however that for best correlation with factory findings, better results are obtained if the dispersion is carried out in the same vehicle or at least one more comparable to factory practice. For example, polymerized linseed oil or solvent free alkyds have been used.
After rubbing is done, apply liberal portions of each paste (standard and sample) side by side, on glass, tinplate or lacquered chart. Draw a wide spatula, French scraper, or doctor blade over both daubs at least about 1 by 2 in. edges should touch but not overlap. Judge colors immediately. If panel is glass, view colors from above, not through the panel.
Venuto designed a scraper (fig 2) for leveling surfaces of rubouts. A spring steel blade is attached to a rigid steel plate which in turn is fastened to a table. Adjustable sides are attached to the plate to guide the panel along the flexible blade whose tension is regulated by an adjusting screw S. The panel with the rubouts facing down ward is placed against the upper portion of the guides and then drawn over the spring steel wiping blade.
One of the properties of oil color that affects its suitability is the speed with which it may be dispersed into white paints. A method for measuring this speed has been directed by Paul and Diehlmen. It consists in mechanically rotating a bottle containing a white paint, the oil color, and No. 11 lead shot. The bottle is changed with 550g of shot, 2 ml of the oil color, and 75 ml of white paint. It is then closed with a cork (concave in the inner end to match glass end of the bottle) and placed in the holder of the rotating machine.
Immediately after charging the test is run. The tube is first tilted and rotated slightly by hand to wet the walls with the white paint. The mechanical rotation is continued until no streaking is noted. The time in seconds for mixing is taken with a stop watch. Results on a number of samples by one operator are shown in Table 4.
Visual Mass color Methods
The first methods developed for mass color measurement, and those requiring the least equipment involve subjective, visual comparisons instrumental methods will be considered later. Thus, for routine examination, it is generally sufficient to determine, visually, if a sample is lighter or darker, redder or bluer, etc., than the standard.
ASTM Method D 387, Mass color and Tinting Strength of color pigments, is based on the experience reported in the foregoing. Two methods of rubbing are provided (1) slab and Muller and (2) Hoover Muller. Appropriate quantities of pigment and oil are given in Table 5. unless otherwise specified, the oil is No. 2 transparent lithographic varnish.
Other methods for MC of colored pigments are in general agreement with the ASTM method. The National printing Ink Research Institute (NPIRI). Method E 1 specifies No 1 lithographic oil for most pigment to oil for a few. Ratios of pigment to oil are about the same as those in Table 5. The Technical Association of the pulp and paper Industry (TAPPI) method is limited to inorganic pigments and pastes it includes procedures for using a casein vehicle for water dispersible pigments and also for extending the color pigment with clay.
Mass color of white Pigments
The criteria of quality for the MC of white pigments are the lightness (that is Y value in terms of colorimetery sometimes called brightness) and tone (hue) of the pigment paste. All assessment are generally given as lighter or darker, and bluer or yellower than standard. Although the color dry powder has some finance in the paper trade, it does provide a realistic evaluation of the color obtained by dispersing the pigment into a vehicle. Like other optical properties, Mc is influenced by the color of the medium pigment concentration, and degree of the pigment dispersion
One MC color test for all grades of titanium dioxide pigments uses a 50 50 mixture of refined linseed oil and soybean oil. Four grams of standard pigment are transferred to the bottom plate of a hoover muller, and 2.4 ml of color oil is added from a syringe. After a preliminary mix with a spatula the paste is mulled 25 revolutions and transferred to a glass panel or a cardboard panel. The pastes are then drawn down in juxtaposition with a 0.015 in. applicator blade. The sample is rated against the standard in north daylight or a color matching booth.
Visual Tinting strength Methods
The sample is visually rated as lighter as or darker than the standard. Identify the standard should have the same tone (under tone, hue) as the sample or else the visual estimate of lightness is affected, if a numerical estimate of TS is required, the relative proportions of white and colored pastes are varied for one of the pastes (either standard or sample, depending upon the particular method) until the lightness of the two pastes match.
ASTM Method for colored pigments
The ASTM method for TS of colored pigments is combined with the method for MC. In addition to pastes prepared for determination of MC, it requires a reduction paste. If prepared in advance, the paste may be kept in collapsible tubes. There are three alternative pastes the most frequently used one is prepared by grinding 100 lb green or white seal zinc oxide in 25 lb of No.2 transparent lithographic varnish.
Weigh small amounts of sample paste and standard paste on counterbalanced watch glass. Add an appropriate amount of reduction paste as given in Table 5. transfer pastes to separate sheets of glass or other impervious substrates and mix thoroughly with spatulas. Apply sample and standard in side by side position as directed for MC. If the colors are not the same repeat using smaller amounts of the stronger paste (sample or standard) until reduced pastes match. The amount of reduction paste is held constant.
VISCOSITY AND CONSISTENCY
Viscosity or consistency is an extremely significant property that is included routinely in specifications and is an important factor in the purchase and use of not only formulated paint products but also the countless raw materials used in the industry.
The successful formulation of a surface coating or paint requires that it be able to be satisfactorily applied to the object to be protected or decorated in a prescribed or predetermined manner or both. It is necessary, therefore, that the paint be formulated to have a proper level of a property that may be designated by any of several terms consistency, body, flowability, viscosity that will allow it to be handled and applied as intended. Measurement of this property is carried out usually with anyone of a variety of different instruments developed for the purpose of measuring it either accurately, approximately, or practically. Also since this property, which is basically its viscosity or its ability to flow, influences certain performance properties such as leveling or flowout, resistance to sagging, brushability, etc., some instruments have been devised with the intent of measuring these particular performance properties directly.
As the paint industry has matured over the past three or four decades and its technology is gradually becoming less of an art and more of a science, researchers have come to realize that a single viscosity determination on a paint is not sufficient. A series of viscosity determinations, a viscosity profile, is required to give a complete picture of rheological properties of the material. Instruments capable of making these more sophisticated measurements have been developed and have become available. As a result many of the subjectively evaluated properties, the attainment of which was not always accompli shed and hence considered part of the art are now attained quickly, measured precisely, explained more easily, and then altered at will.
The subject of rheology, which includes viscosity and consistency measurement, is broad, and the reader is urged to consult any of the many more recent references cited throughout this section for a more comprehensive treatment of the subject.
A subjective term descriptive of the flow properties of a liquid usually measured in an empirical manner and in arbitrary units. Often used as a synonym for viscosity.
Non Newtonian Liquid
Any liquid that does not satisfy the requirements for a Newtonain liquid. A liquid that displays plastic, pseudoplastic, or dilatant flow characteristics. Most paints are non Newtonian liquids.
A material displaying pseudoplastic flow has no yield value, and the curve of the plot of shear stress versus shear rate is nonlinear with the shear rate increasing faster than the shear stress.
A type of flow characterized by an increase in viscosity as shear stress is increased. The curve of the plot of shear stress versus shear rate is nonlinear with the shear stress increasing faster than the shear rate.
Thixotropic materials are those whose consistency depends on the duration of shear as well as on the rate of shear. Thixotropy is a reversible process, and after resting the structure of the material builds up again gradually. The thixotropy or thixotropic material meant when the terms are used in the paint industry is generally a thixotropic Newtonian or a thixotropic pseudoplastic material.
A term used to distinguish types of thixotropic behavior in materials exhibiting plastic now (Bingham bodies).
Countless instruments are available for measurement of the rheological properties of liquids. Many of these instruments have been accepted by and found considerable use in the paint industry. A few of these are described or noted in this chapter.
Broadly one may consider viscometers to fall into two categories those that are capable of a fundamental measurement of the property and those that are approximate to varying degrees, or arbitrary. In the fundamental group, Patton includes capillary (tube and orifice), concentric cylinders (rotary motion and axial motion), falling sphere, band, and cone and plate viscometers. In the approximate or arbitrary group he includes bubble, rotating disk, orifice (single and multiple), and rotating paddle. This classification and grouping is somewhat similar to classification into the following groups by Weltman those that express consistency in arbitrary terms, those that can represent flow conditions mathematically for Newtonian and liquids that can be tested at low shear and are not applicable to non Newtonian materials, and those that can measure consistency under controllable flow conditions and establish accurate relationship between shearing force and rate of shear. Myers groups viscometers according to the method of imposing shear such as efflux, viscous drag (falling weight), rotational, telescopic flow, and transverse flow. Le Sota classifies instruments as capillary, efflux, bubble, rotational, and perforated disk. Rotational is further divided into cup, bob, paddle, disk, and cone and plate. Westgate also categories instruments in essentially the same manner. Karam lists viscometers as capillary, falling sphere, rotational, vibrational, and compression or extension viscometers.
In the following paragraphs, instruments are grouped by simple physical type capillary, rotational, falling ball, band, efflux, bubble, and orifice. The overall listing goes from fundamental to approximate however, within anyone class of instrument there may be also a range between fundamental and approximate.
Hercules Capillary Viscometer
This is an early capillary viscometer (Fig. 1) developed for the purpose of determining the viscosity of the low viscosity types of nitrocellulose (that is, 18 25 centipoise and 30 35 centipoise types) where the standard solution viscosity was too low to be determined by the Hercules Falling Ball method. The instrument is calibrated by the user with oil having a viscosity in the range in which the instrument will be used. For nitrocellulose solutions, castor oil having an absolute viscosity of about 665 centipoise was used as the calibrating fluid. To use the instrument, tilt it so that the reservoir is vertical. Fill to the etched line with the sample being tested, allowing time for the material to reach an equilibrium level in the capillary. Then, holding a finger over the other capillary end, lower the instrument to the original horizontal position. Remove finger and determine the time for the material to travel from the first mark on the capillary to a second mark suitably located further down the capillary. The second mark (or marks) is determined in calibrating the viscometer.
Bingham Green Plastometer
The instruments rely on gravity to move the liquid through the capillary. Instruments were also developed that used either pressure or vacuum to move the liquid through the capillary. One such instrument is the Bingham Green Plastometer (Fig. 2). Many early studies of consistency were made with this apparatus. Because it combines air under pressure and capillary tubes of various sizes, almost any substance that exhibits now may be run in it. This apparatus was used primarily in research work, being too costly and time consuming for general routine work. It consists of a container for the paint, at the bottom of which a carefully calibrated glass capillary tube may be attached, Air pressure is used to force the paint through the capillary into a receiver. Rotational devices have largely displaced it today.
Caster Severs Viscometer
This instrument was devised for measuring the rheological properties of plastisols. Basically, the instrument utilizes air or gas pressure to force the test material through an orifice. The time required for a given weight or volume in poises is calculated from basic equations. Apparent Viscosity of Plastisols at High Shear Rates contains a recommended procedure for determinations using this instrument.
Gardner Pressurized Flow Cup
While this device may be looked upon as a simplified plastometer, the dimensions of its orifice placed it more in the category of a pressurized simple efflux cup rather than in the class of the capillary instrument. It comprised a cylinder, inside diameter about 2 in., length 1.7 in., with a conical bottom terminating in a detachable orifice, and a cover with fittings to admit air under pressure. Three different sizes of orifice were provided.
Efflux Type Viscometers
The capillary viscometers are efflux type instruments. However, the term efflux instrument or efflux viscometer has come to mean a cup type instrument containing an orifice. Whatever precision and accuracy an efflux instrument may have is essentially dependent upon the dimensions of the orifice. The closer an orifice resembles a capillary, the more accurate is the instrument. Theoretically, the behavior of the Newtonian liquid in an efflux viscometer can be expressed by the last equation.
Normally, however, the viscosities obtained with an efflux viscometer or viscosity cup are expressed in time units such as seconds, No.4 Ford Cup, have a limited useful viscosity range, and are not precise. However, they are practical instruments, easy to use, and easy to clean, features that account for their widespread popularity.
One of the more accurate cup type instruments, the Saybolt viscometer utilizes two capillary type orifices to measure viscosity in terms of Universal seconds or Furol seconds. The time of measurement for this instrument is that required for 60 cm3 of liquid to pass through the orifice. The Saybolt viscometer instrument has been adopted by organizations, such as the National Bureau of Standards, American Petroleum Institute, and American Society for Testing and Materials. The procedure may be summarized as follows. Pour some of the sample through the outlet tube and insert cork stopper into air chamber below tube, sufficiently tight to prevent escape of air, but not to touch tube. Heat material in separate container to slightly above 25 C (or specified temperature), stirring with thermometer. Pour material into viscometer until its fills gutter. Maintain temperature for one minute, remove thermo meter, pipet material from gutter until level in gutter is below level in viscometer proper. Place receiver (60 cm3 flask) below viscometer so that stream of material from outlet tube will strike neck of receiver, thus avoiding foaming. Snap cork stopper from its position and simultaneously start a stopwatch. Stir bath liquid continually during test and maintain temperature. Stop the watch when bottom of meniscus of material reaches mark on neck of receiver. Time in seconds is Saybolt Universal Viscosity.
Free Interfacial Energy
The creation of more surfaces, as in the atomization of a liquid, requires an investment of energy. Part of this energy goes into the surface as free energy and is supplied by the mechanical agent creating the surface. The rest is contributed by the environment, generally as the heat influx required to maintain the experiment at its initial temperature, since surface formation causes a temperature drop.
If neither of the two phases is a gas or vapor, the boundary between them is an interface. However, the terms surface and interface are used interchangeably.
The superficial similarity between blowing a soap bubble and inflating a balloon, which suggests the existence of a skin under tension, led early investigators to the assumption of a surface tension. But surface effects can be correlated more readily in terms of the thermodynamics of interfacial energetics. It is easily shown that the numerical value of the hypothetical surface tension of a liquid is identical with its very real free surface energy. Surface tensions are expressed in dynes per centimeter and free surface energy in ergs per square centimeter.
Coatings technologists deal with interfaces of all classes gas liquid as in an aerosol spray liquid liquid, as in an emulsion gas solid, as in a dry pigment before its immersion in a vehicle liquid solid, as in a pigment dispersion and solid solid, as when the crystal faces of two different pigment particles are in tight contact. Paint scientists are particularly interested in the formation of liquid solid interfaces that are stable in the package, that is, in the permanent replacement of the air at the air solid interface of the pigment by the vehicle to give the liquid solid interface of the dispersion.
Wetting is not merely the submersion of the pigment in the vehicle. If it were, any pigment forcibly submerged in any liquid would be considered as wetted by the liquid, and wetting would lose all discriminatory meaning since it would be universal. Wetting is the unforced, spontaneous spreading of a liquid over a solid, or a second liquid, and is synonymous with a decrease in free interfacial energy. For each 1 cm2 of the solid surface that becomes wet, 1 cm2 of air liquid surface and 1 cm2 of liquid solid interface is formed and 1 cm2 of air solid surface disappears. The change in the free interfacial energy balance sheet is called the spreading coefficient, S
Although air solid and liquid solid interfaces are of the utmost importance to coatings technologists, the corresponding free energies cannot be measured directly because existing methods require that the interface be mobile, as between two liquids, or a liquid in a gas, or a liquid in a vacuum saturated with its own vapor, and not rigid, as when one of the two phases is a solid. Hence, the scope of surface and interfacial tension measurements is limited. Nevertheless, much can be deduced from related determinations for example, observation whether a liquid spreads over a solid in air or in a second liquid and, if not, what the magnitude of the resulting finite contact angle is.
Surface tension ranges of common organic liquids are not broad. At room temperatures, the saturated aliphatic hydro carbons extend from 18 to 28 dynes per cm the saturated cyclic hydrocarbons, 22 to 28 the aromatic hydrocarbons, 28 to 30 the alcohols, ketones, and esters, 21 to 28 the fatty acids, 25 to 38 other oxygenated aliphatic liquids, 17 to 28 the aromatic amines, 33 to 43 and practically all the halogenated liquids, 21 to 35. The cause of this narrowness of range is not far to seek. Molecules orient themselves in the surface of a liquid in the general direction that leaves the least possible free surface energy under the circumstances. For organic compounds, this means that the non polar ends of the molecules at the surface point toward the air phase, while the polar ends, if any, point into the liquid. Thus, organic liquids expose a hydrocarbon exterior, and their surface tensions are those of hydrocarbons somewhat elevated by the influence of any polar groups immediately below.
Water can expose only polar groups therefore, it shows a higher surface tension, around 70 dynes per cm at laboratory temperatures. However in an aqueous solution of an organic compound, the solute molecules are able to decrease the free surface energy by collecting in the surface, oriented with their nonpolar groups out, even if this means depleting the solution of much of the solute in the case of a very dilute solution. Aqueous solutions of organics have surface tensions intermediate between that of water and those of the organics. The slightest contamination drops the surface tension of the water. The surface tension of the aqueous phase of paints is about half of that of pure water.
On the other hand, molten salts show very high surface tensions. Aqueous solutions of strong electrolytes reveal surface tensions somewhat higher than that of pure water. In these solutions, the solute leaves the surface since this desorption drops the free surface energy. But there is some electrolyte sufficiently dose to the surface to give a net effect above that of pure water.
Thus, the surface and interfacial tensions of pure liquids have limited significance in the coatings industry. The liquids in which the paint chemist is interested are, first, the liquids as supplied and, second, the liquids with their interfacial characteristics modified after they have become saturated by the other ingredients of a coating.
The extent to which the dass of otganic derivatives known as surface active agents lowers the surface tension of water and, to an even greater degree, depresses the interfacial tension of a mineral oil emulsion is summarized graphically in Figs. 1 and 2.
The interfacial tension of organic liquids against water, mutually saturated, ranges from below 10 to over 50 dynes per cm.
Surface Tension Measurements
The methods used to measure surface and interfacial tensions may be classed either as absolute or empirical. The readings made by an absolute method may be converted directly into surface tension values. An example is the capillary rise method. An empirical method must first be standardized by a series of measurements with liquids of currently known surface tension, after which the previously unknown surface tension of additional liquids may be determined. Examples are the ring method and the drop weight method. Many empirical methods have been developed. As more becomes under stood about the fundamental nature of matter, theoretical developments will convert empirical into absolute methods by supplying rational duplications of the correction curves used with empirical approaches.
A schematic representation of the arrangement is shown in Fig. 3. This method is obviously direct since one need only to determine r, d, and h and know g to calculate. However, to obtain the accuracy inherent in this method, several factors, in addition to those common to other physical measurements, must be considered. As with all surface tension determinations, contamination must be scrupulously avoided as the slightest impurities may concentrate in the surface and greatly affect the results. The vessel in which the bulk liquid is held must be relatively wide or else a correction must be introduced for the minor capillary effect caused by the vessel the wider the vessel in proportion to the capillary, the smaller the correction. The meniscus in the narrow tube is a part of the capillary rise, and the true value of h is the rise to the bottom of the meniscus plus an allowance for the liquid in the meniscus calculated with sufficient accuracy by assuming the meniscus to be hemispherical. The density, d, is actually not quite that of the liquid but the difference in densities between the liquid and the gas or vapor above it. Most important, the liquid must wet the capillary that is spread spontaneously to its very top, so that the column of liquid is suspended not from the dry material of which the capillary is made but from a very thin layer of the liquid spread as a so called duplex film, not as a monolayer, over it. If the liquid does not wet the capillary wall, it shows a contact angle q at the junction which will diminish h because only the vertical component of the surface tension is effective in raising the column of liquid.
The capillary rise method is not confined to surfaces between a liquid and a gas or vapor. the two phases involved may both be liquids, with the liquid of lower density replacing the gas as the supernatant phase. The experimental ingenuity required has been recorded in the literature.
While overtly a simple procedure, capillary rise, except for meticulous research investigations, does not rank as a practical method for the determination of interfacial energies in the coatings industry.
Maximum Bubble Pressure Method
For a static bubble held at the tip of a capillary, the dimensions of the bubble, the pressure within it, and the surface tension of the liquid are interrelated. Properly contrived measurements, therefore, may permit the calculation of the surface tension in an absolute manner. However, the procedures may be simplified by slowly increasing the size of the bubble until it breaks away from the tip and measuring only the pressure maximum required to achieve this rupture.
Drop Weight Method
The drop weight method (Fig. 4) centers around a vertical tube conveniently about 7 mm in diameter with a narrow capillary along its axis. The bottom of the tube is ground flat, perpendicular to the axis. The liquid is permitted to flow very slowly down the capillary from an upper reservoir and to spread along the circular tip of the tube, from which it hangs. As the drop is allowed to enlarge, it ultimately becomes infinitesimally too heavy to continue to be supported by the tip and falls. For greater accuracy, a number of drops are collected under the same carefully controlled conditions, weighed, and the average weight of a drop is calculated.
The correction factor has been painstakingly determined empirically and relates to the fact that, once the drop has just grown to the size where it is destined to fall, only a specific fraction of it is severed, that fraction being governed only by the dimensionless ratio of the radius of the tip r to the cube root of the drop volume. The drop weight method is thus not an absolute method, since it must be standardized by establishing the values of through Eq 5 for many liquids and solutions of known r with tips covering a range of rvalues. The simple apparatus involved is easy to construct. Obviously, the method is adaptable to interfacial tension measurements, since the drops may be allowed to fall from an immersed tip through a second, lighter, immiscible liquid or rise up through a second heavier liquid from an inverted tip. In such interfacial studies, m of Eq 5 is not the weight of the drop in air or vapor but the differential weight which takes into account thc buoyancy of the second liquid. Particularly in interfacial experiments it is the drop volume rather than the weight that is measured and which can be converted into the weight by knowledge of the difference in densities between the two fluid phases. Simple as the method is, it has been over simplified with a sacrifice in accuracy by adaptions which, for example, count drops from relatively primitive tips for a predetermined volume of the liquid under study.
Perhaps the most widely employed method for measuring surface tension is the ring method. It is based on the ideal concept that if a submerged thin horizontal wire 1 cm long is to be pulled up through a surface, a force of 2 r dynes must be exerted on it against the surface forces, the factor 2 appearing because there is an interface on each of the two sides of the thin film. To avoid corrections for the effect of the wire ends, the wire may be bent into a flat ring, in which case the force to pull it up through the surface is ideally 4 Rr where R is the radius of the ring in centimeters. For this method to apply the liquid must wet and spread over the wire. However, the geometry of the attachment of the film to the ring is not ideal. Consequently, a correction factor p enters the consideration as a measure of the extent to which the system departs from the ideal.
Obviously, the method may be applied to the measurement of interfacial tension, in which case the composition of the ring must be such that it is preferentially wet by one of the liquids and not by the other. If the wetting liquid is the denser, the ring is pulled up from it into the other. If the wetting liquid is the lighter, the ring is pulled down from it into the lower liquid. Instruments are available from several commercial sources for the practice of the ringless method.
Wherever interfaces occur in a structure to an extent great enough to influence its physical behavior, interfacial energy or its equivalent, interfacial tension, appears as a variable in the equations governing some aspects of the system. Each such equation, therefore, may be adapted into a method for determining interfacial tension provided the other variables are readily controlled and are measurable with sufficient accuracy. A number of other methods have been proposed on this basis for measuring interfacial tension in addition to those described above. For example, the shape parameters of a hanging drop have been so used, and also those of a jet of liquid passing through an elliptical orifice, which imparts an oscillation to the stream. However, except for special applications, such methods are not popular.
It has been pointed up that since the paint industry deals with stable mixtures of powders and liquids as articles of commerce, liquids that wet the surfaces of the powders involved are of paramount interest. Nevertheless, situations arise in coatings technology where wetting does not occur. Examples are the specific pairing of a liquid and a powder, and using a particular liquid over an intended substrate. In such instances of nonspreading, a finite contact angle is involved, and occasions arise where it is desirable to measure it. As in all measurements concerned with interfacial energy, the greatest precaution must be taken to ensure that the surface used is actually the one of interest, or that all contaminants are avoided. Thus, the magnitude of the angle of contact shown by a small drop of a highly pure non spreading liquid, resting on the clean horizontal facet of a large crystal of rutile has nothing to do with the practical realities in a sample of paint in which the surface of commercial titanium dioxide pigment is anything but pure titanium dioxide and the liquids used in coatings are anything but pure single chemical species.
PARTICLE SIZE MEASUREMENT
In coatings and similar products, the criteria for best performance of particulate ingredients inorganic, organic, extender, and metallic flake pigments, and the dispersed phase of latexes depends in large measure on the size and shape of the particles composing the particulate materials. Examples of these relationships appear in other parts of this book. The importance of particle size measurement including the amount of coarse oversize particles is evident. In this book it is practical to cover in detail the many procedures employed in the measurements, but the brief descriptions are supported by references to the literature. Also cited are textbooks for the reader who desires to acquaint himself with general theory. The recent book by Allen can be particularly helpful.
Pigments and Extenders
Size, size distribution, and shape of pigment particles greatly influence the optical and other physical properties of paints. Relatively large particles may protrude through the surface of the film and detract from its gloss. But extreme fineness is not always desirable. There is an optimum size below and above which hiding power decreases.
Eide showed that coarse acicular zinc oxide imparted greater durability to a paint than did round zinc oxide. Morris and Nelson found evidence that acicular asbestine of a wide distribution of sizes favorably influenced the durability of paints. A review by Jacobsen cites many examples of the significance of particle size and shape of pigments and extenders on the optical and physico chemical properties of coating systems.
A good generalization of the relationship between particle dimensions and film properties is shown in Fig. 2
The national paint, Varnish, and Lacquer Association has published a Pigment Index that gives available information of the particle size of pigments on the American market.
Metallic Flake pigments
Particle size and shape of metallic flake pigments are also important. The different tone effects in polychrome finishes are obtained by using different sizes of non leafing aluminium flakes. The covering area of leafing pigments is a function of size and thickness. For nonleafing flakes, microscopical and surface area measurements are used for leafing pigments, a method based on the Langmuir balance principle is used.
The particle size of the dispersed phase of emulsions (latexes) influences the performance of paints made with these materials. Martens show with graphs how the many parameters of paint performance are influenced by the particle size of the dispersed polymeric materials. Brown considers the particle size of the polymer as one parameter involved in the coalescence of the particles, and Kreider connects rheological properties with particle size. Extensive investigations into light scattering reflect the great interest in influence of particle size.
Methods for the determination of the particle size of emulsions are numerous. Bradford and Vanderhoft cite references concerned with the following techniques light scattering, ultracentrifuge, soap titration, low angle X ray scattering and microscopy. Except for soap titration, these methods also apply to pigments.
Treatment of Data
Measurements are commonly expressed in terms of frequency against size, or weight percent against size, that is, the number of particles in each of the various size fractions, or the weight percent of the fractions that compose the specimen. Another way of representing the data is by average diameter, of which there are several kinds. The arithmetic mean diameter is commonly used. Others, with their symbols and mathematical definitions, are given in Table.1
For reporting particle size ctiaracteristics, Reporting Particle Size characteristics of Pigments, offers three parameters size, coarseness, and dispersion. These were selected because it appeared that they could be derived from data obtainable by most analytical methods.
The size parameter (SSD) is the specific surface diameter the diameter in micrometers that the particle would have if they were particles of uniform size. It is also known as the surface mean diameter.
Particle Size with Light Microscope
Above about 0.2 m, the microscope is a basic tool for determining particle size. The lower limit is set by its resolving power and the upper limit by its depth of focus. The microscope is especially useful for measurement of platelike and needle shape particles that do not obey Stokes law on which the sedimentation methods are based. Disadvantages of the method are that it is slow and laborious. Hence, it is used chiefly for the calibration of the more rapid relative methods. Microscopy and in ASTM Recommended Practice E 20, Analysis by Microscopical Methods far Particle Size Distribution of Particulate Substances of Subsieve Sizes, and to a paper by Loveland. Following are same general comments.
Direct Measurement Method
Green was the first investigator to systematize the microscopical method. He dispersed the pigment in a medium on a microscope slide, photographed the dispersion at a known magnification, and projected the image on a screen so as to increase the magnification for measurement. Dunn bypassed the photomicrograph by projecting the image of the particles directly from the slide on the screen.
Ideally, the pigment should be dispersed in the medium in which it is used, but this is rarely done. Green used turpentine, Allen recommended a viscous vehicle, and Eide used fused resins. Gehman and Morris milled the pigment in rubber, dissolved the mix in a solvent, and applied the suspension to the slide.
For some purposes, comparison under the microscope with linear scales, or with circles or ellipses in graduated sizes, may be advantageous. The comparison scale may be a micrometer eyepiece, an eyepiece reticle, or a scale engraved on the micro scope slide. For an approximate comparison of coarse particles, Gardner used the grooves of phonograph records.
Dark Field Technique
The technique of dark field illumination uses the detecting power of the microscope rather than the resolving power, which is a limiting factor. Therefore it is used essentially for sizes below the resolving power of the light microscope.
The slit ultramicroscope, as well as the Cardioid condenser dark field microscope, is also used for estimating particle size.
Particle Size by Sieving
Sieving appears to be a simple means for separating pigments into fractions according to size. A significant addition to the physical laws governing the process has been made by Whitby who investigated variables such as size distribution, mesh size, sieve loading, sieve motion, sieve material, and relative humidity. A comprehensive manual on sieving methods has been published by the ASTM.
For routine testing, sieves conforming to ASTM Specification E II, Sieves for Testing Purposes, are recommended, Table 3. These sieves are made of woven wire cloth, supported in frames 8 in. in diameter and 2 in. high, having below the sieve a skirt that nests into the frame of another sieve. Frames for sieves 149 m (No. 100) and lower may be 3 by 1 in.
The openings in successive sieves progress from a base of 1.00 mm in the ratio when selecting a range of sieves from the series, it is recommended that each sieve, each alternate sieve, or each fourth sieve be taken. In this way, the basic ratio between successive sieves remains constant.
The U.S. Series of sieves, Table 3, is patterned after the Tyler Series that was introduced in 1910. Infact, the two are now interchangeable, the only difference between them being the designations of the individual sieves. Those in the U.S. Series are identified preferably by the sizes of the openings in millimeters or micrometers. An alternate means is by a number approximately equal to the mesh. Tyler sieves are identified by mesh. Equivalent sieves are available in both series.
The sieves proposed as standard by the International Standards Organization (ISO) correspond too many of the sieves in the U.S. Series.
A recent development, especially useful for sizes under 100 m is a sieve made by a photoengraving and electroplating process. Such sieves are intended to serve mainly as primary reference standards.
If the sieves are used singly, the following directions that appears in several ASTM methods of test. Sieve Analysis of Mineral Filler, may be used.
After transferring the specimen to the sieve, Hold the sieve, with pan and cover attached, in one hand in a slightly inclined position so that the specimen will be well distributed over the sieve, and at the same time gently strike the side about 150 times per minute against the palm of the hand on the upstroke. Turn the sieve every 25 strokes about one sixth of a revolution in the same direction. Continue the operation until not more than 0.05 g passes through the sieve in 1 min of continuous sieving.
Sources of error in hand sieving are fatigue and mere casual attention to directions. Machine sieving eliminates these errors, but some specifications require hand sieving unless it can be shown that machine sieving gives the same results.
Machine sieving has the advantages of uniformity of treatment and saving time since the operator is free to perform other tasks while the machine is working. Several types of machine are available. The conventional ones impart to the sieve an oscillating or rotating motion or both, with regular tapping. None appear to reproduce the motions of hand sieving specified in the preceding section. But we might ask what is sacred about hand sieving? Perhaps the restriction should be reversed permit hand sieving provided the results agree with machine sieving.
Ro Tap Sieve Shaker This machine manipulates a series of sieves, graduated with respect to mesh size, so as to permit separation of a specimen into sizes. As the name implies, the sieves are given a special rotary motion accompanied at regular intervals by a tap or jar. The nature of the specimen dictates the size of the sieve openings and the timing cycle. This type of machine is recommended in ASTM Test E 276, Particle Size or Screen Analysis at No.4 Sieve and Finer for Metal Bearing Ores and Related Materials.
Besides the machines that agitate the specimen by rotary and tapping actions, those that use air are also available. Two of these are described next.
OIL ABSORPTION OF PIGMENTS
A common preliminary test applied to a pigment is mixing it into a paste with linseed oil. The test gives a rough idea of color and texture and, if done quantitatively, tells how much oil is needed to make a stiff paste. According to some authorities the oil absorption (OA) test predicts consistency of ready mixed paint. Mills writes when equal volumes of paste (oil absorption paste) are thinned with equal volumes of thinners, the resulting paints have equal consistencies. Stieg shows that OA by the rubout method, expressed on a volume basis, is proportional to critical pigment volume, judged by uniformity of flat alkyd paints made with varying percentages of pigment, or determined by the Asbeck Van Loo method.
Nature of Oil Adsorption
The most reasonable explanation of what happens during an OA test appears to be
The surface of each particle of pigment is wetted with oil, that is, each particle becomes enclosed in a shell the thickness of which has been estimated to be as much as eight molecules. The amount of oil required depends on the specific area of the pigment, that is, particle size, roughness of the surface of the particle, and presence of cracks and pores in the particles. Other things being equal, it depends on the vigor and duration of rubbing and on the wetting power of the oil for the pigment. With the addition of more oil, the interstices between the particles (with their shells of oil) now become filled with oil. The amount of oil required for this stage depends on the type of packing assumed by the pigment particles, which may range from rhombe hedral to cubic. It is also influenced by the presence of aggregates (clusters of pigment not broken up by the procedure) and agglomerates (clusters formed after particles have been wetted). Agglomeration is influenced by the nature of the oil. Figure l is a stylized representation of conditions existing in an oil absorption paste.
Methods for Determining Oil Absorption
ASTM Rubout Method
This is the classical method, and the following directions are essentially those in ASTM Method D 281, Oil Absorption of Pigments by Spatula.
The apparatus required for the test includes a glass plate or marble slab, about 10 by 20 in. in size a dropping bottle or buret, Fig. 1, graduated in 0.1 ml a stiff spatula and raw linseed oil, having an acid value of I to 3.
One to 0.05 g, or multiple thereof, of pigment is transferred to the rubbing plate. Oil is added, drop by drop, the mix being thoroughly rubbed with the spatula between additions until a very stiff putty like paste is obtained. This paste should not fall apart when separated from the rubbing plate with the spatula. Results are expressed as pounds of oil per 100lb of pigment, or grams of oil per 100 g of pigment. The amount of pigment should be regulated to require at least 1 g of oil, for example, 20 g of white lead or 1 g of carbon black.
A, B, and C are each 1 g of zinc oxide with increasing amounts of oil. A is still granular and crumbles when separated from the plate the end point has not been reached. B may be rolled without crumbling the end point has just been reached. C may not be rolled the end point has been passed. Table 1 gives some typical results.
Unless extreme care is exercised, precision is not particularly good. Results by a single operator are often relatively satisfactory, within ± 5 percent. On the other hand, results among different operators may differ by as much as ± 25 percent. Contributing to the low order of reproducibility among different operators are several uncontrolled or poorly controlled conditions, such as spatula, acid value of oil, rate of addition of oil or pigment, sequence of additions, energy of rubbing, duration of rubbing, end point, and use of standard specimen.
The following variations of the method show how several authorities have tried to surmount the difficulties.
This method differs from Method D 281 mainly in the acid value of the oil (7.5 to 8.5) and in a requirement that the rubbing shall occupy from 20 to 25 min. Also, the spatula is a palette knife with a tapered blade, 140 mm long, 20 to 25 mm wide at its widest point, and not less than 12.5 mm at its narrowest point, but this difference would seem to be minor. The method allows the addition of several drops of oil to the pigment between rubbings until agglomerates are formed but does end the test with single drop increments. The final paste should be just spreadable without cracking.
This method is practically the same as the ASTM method, except that the end point is defined more rigorously as the point where the paste just adheres to the spatula. Rubbing is continued for about 20 min.
In the course of many determinations of OA, Azam noted that sticking of the paste to the spatula seemed to coincide with a definite transition point. Investigation indicated that a complete paste absorbs no more oil when immersed in oil but that an incomplete paste absorbs oil in an amount equal to the deficiency. This check of the end point was made by immersing the rubout in a known amount of oil in a graduated cylinder and, after 2 or 3 days, noting any decrease in the volume (Table 2).
This method introduced two changes. Pigment is added to the oil, and much smaller quantities are used in order to reduce the labor. Some operators get better results while others find no practical difference.
The Smith stead method, below, also adds the pigment to the oil and works with larger quantities, but makes the rubout with a mechanical muller.
Smith Stead Method
This method eliminates the variable and poorly defined rubbing effort of the rubout method by the use of a mechanical muller. It follows the Hoffman practice of adding the pigment to the oil on the hypothesis that this procedure more readily displaces occluded air.
Exactly 1 g of oil. is transferred to the lower plate of the muller. A watchglass or dish containing an excess of pigment is weighed. About half of the required quantity of pigment is transferred to the oil and mixed with a spatula until no dry pigment remains. The muller is closed and run for 50 revolutions. Another portion of pigment is transferred to the muller and mixed with the spatula, and the muller is closed again and run for 25 revolutions. This procedure is repeated with smaller quantities of pigment until the end point is reached. At the end point, the paste may be lifted without distortion when the spatula is drawn across the plate at an angle of about 20 deg. The leftover pigment is weighed, and the OA is computed in the usual way. The test should be scheduled to occupy from 30 to 35 min and requires from 8 to 12 additions of pigment.
National Lead Company Method
This method is similar to the ASTM and Azam methods but differs in a few respects. It attempts to regulate the pressure of rubbing and the area of contact of the spatula with the rubbing plate. The spatula is calibrated by holding it at an angle of 45 deg against the pan of a torsion balance, under a load of about 1500 g, under which conditions it should flex sufficiently to make contact with the balance pan for a distance of about 1 in. The blade of a 4 in. spatula may need to be shortened to about 3 in. to meet this requirement. Dutch 37 AA linseed oil, or equivalent, as the liquid, and 5 g of pigment are used. A slight excess of oil is first added to a portion (about two thirds) of the pigment to make a thin paste, increments then being added until a stiff paste is formed. Oil and pigment are then alternately worked into the paste until a dry, crumbly mass is formed. Finally, oil is added until the crumbs are converted to a coherent paste that adheres slightly to the glass plate. The pressure during the rubbing should flex the spatula to the same degree as in the calibration, and the rubbing should normally occupy from 2 to 5 min, depending on the type of pigment.
Finally, a correction factor should be applied. This is obtained from a determination on a pigment whose OA has been established by experienced operators.
Standard deviation for 72 degree of freedom has been found to be 0.66 for a single determination.
Density End Point Method
In this method in the absence of a chemical reaction between oil and pigment, it is postulated that the density of a mixture of oil and pigment is an additive function of the densities of the components (Curve AB of Fig. 2). Bulk (apparent) densities of mixtures having oil contents lower than those required to produce the putty like pastes corresponding to oil absorption end points lie along a curve C similar to Curve CD. The volume of oil corresponding to Curve D is the true OA. With greater proportions of oil, bulk density falls under Curve AB, probably because all air is not removed from the sticky mass. If mixing is inadequate, the entire Curve CD is displaced downward, and a lower oil absorption is found.
Bessey Lammiman Method
This method is an indirect one for determining OA. A known volume weight of the pigment is compressed into a wafer, and the volume of the wafer is determined. From this value is subtracted the actual volume computed from the density of the pigment. The difference is the volume of the pores.
The apparatus consists of a 10 ton hand operated hydraulic press, cylindrical mold, rams, and upper and lower plates. The mold is 3 in. long, with an inside diameter of 1 in. and an outside diameter of 2 in. The lower plate fits into a hole in the bottom plate. The upper plate is bolted to the press.
With the lower ram sandwiched between the bottom plate and the mold 10 g of pigment is transferred to the mold and distributed uniformly by tapping the assembly against the table top a few times. The upper ram is inserted in the mold and pressed against the pigment under hand pressure with a slight twist. The assembly is placed in the press, and the pressure is applied. The thickness of the compacted wafer is obtained from a dial micrometer gage mounted in the upper plate.
The pressure is released, and the bottom plate is lowered to make room for a small beaker (tared) inside a metal ring, slightly taller than the beaker, under the mold. Pressure is applied to eject the pigment wafer into the beaker. The void volume of the specimen is then obtained.
The time required for a test, including cleanup, is about 15 min.
For whitings, the void volume at 12,800 psi is numerically equal to true oil absorption, which is the OA obtained by the density end point given by the density end point. True OA is substantially under the values obtained by the rubout method, as specified in British Standard Specifications.
For other pigments, different pressures may be required.
Gardner Coleman Method
In this method oil is stirred into the pigment until a soft paste is formed. The labor of the rubout method and its accompanying errors are avoided. The following directions are essentially those of ASTM Method D 1483, Oil Absorption of Pigments by the Gardner Coleman Method.
Apparatus for the test includes a 250 ml round bottom glass container, buret graduated in 0.1 ml, spatula with a blade approximately 18 by 100 ml, and raw linseed oil with acid number of 3 to 4.
Variations in the method have been proposed but none appears to have received much support, for example, stirring with a ½ in. steel rod in a metal dish. The end point is reached when a Vicat needle, fitted with a disk, 1.875 by 0.125 in. under a load of 280 g, penetrates the paste to distance of 0.5 in. within 1 min. Another investigator preferred a glass rod as a stirrer, and still another used a mortar and pestle. In the last named test, the end point is reached when a single drop of oil changes the mixture from a mealy mass to a sticky paste. At the end point, the paste, lightly touched with the finger, will adhere to it.
Another concept related to OA is that of testing the performance of exterior house paint at equal free binder content. The property is strictly comparative and is measured by consistency, usually at a low rate of shear in an apparatus, such as the Krebs Stormer viscometer. Paints are considered to have equal free binder when they are made with identical vehicle and have equal consistency.
The method requires a paint of known performance. This is represented by Paint 1 in Fig. 3. It is desired to replace the magnesium by calcium carbonate. The first step is to replace it with an equal volume of the calcium carbonate. This gives Paint 2. Next, Paint 3 is prepared with an extra volume of calcium carbonate replacing an equal volume of complete vehicle. Step 2 gives a paint of lower consistency than that of Paint 1. While Step 3 gives one of higher consistency. Finally, Paints 2 and 3 are blended to give Paint 4, having tho consistency of Paint 1. It has the same content of titanium oxide and zinc oxide as paint 1.
Liquid Absorption by Pigments
Oil absorption concepts may be extended to liquids other than oil. For example, pigments for cellulosic lacquers may be dispersed in plasticizer and the paste so formed be stirred into the other ingredients of the lacquer. Some typical data are given in Table 3.
Critical Pigment Volume
The application of scientific principles in the paint industry upsets the long held fallacy that oil is the life of the paint and with the almost universal practice of formulating paints on a weight basis.
In 1926, as a result of a statistical study of the North Dakota house paint tests, Calbeck concluded that the pigment/binder system should contain at least 28 volume percent of pigment. A short time later, Wolff in Germany, related optimum performance of exterior paints to their critical oil contents. Wolff and Zeidler devised a simple graphical method that is illustrated in Fig. 4. Elm also studied this relationship and, in general confirmed Wolff s conclusions, Table 4, but reversed the point of reference to express the critical composition in terms of pigment content, as Calbeck had done.
In 1929, and again in 1961, Vannoy used consistency, determined with the Krebs Stormer viscometer, to monitor the effects of replacing, on a volume basis, certain pigments by others in experimental formulations.
Measurement of FILM THICKNESS
Control of film thickness is an important part of paint testing. Film thickness measurements are a check on other associated tests in the laboratory, on the skill and will of the journeyman painter, and on the performance of automatic painting machinery.
A number of methods for measuring film thickness are available. The choice depends on where the film is located (laboratory or field), the substrate (wood, plaster, or metal, and the electrical or magnetic properties of the metal), of the surface texture of the substrate (smooth or rough, or flat or wavy), and whether the film is wet or dry.
Wet Film Thickness
The principal reason for determining the thickness of a wet film is to check on the spreading rate at which the paint is being applied. Two devices for producing this information have been standardized in ASTM Method D 1212, Measurement of Wet Film Thickness of Organic Coatings, the Inmont gage (formerly known as the Inter chemical), and an alternate, the Pfund gage. These are described in the sections following. Two nonstandardized methods are also described.
Measurements on wet films must be made without delay after the film has been prepared in order to avoid shrinkage due to evaporation of volatile thinner.
Inmont Wet Film Gage
This gage is essentially a pair of wheels and a ring of shorter radius mounted eccentrically between them so that on one radius the wheels and the ring are tangent. Thus, when placed on a plane surface at the tangent radius, the clearance of the ring is zero. At the opposite radius, the clearance is maximum. Two symmetrical scales on one face of the gage show the clearance of the ring at any point on the circumference. With the maximum clearance at the bottom, the gage is placed on the specimen and rolled. The point on the circumference of the ring where the paint and the ring first meet denotes the thickness of the film. Best practice is to start from the maximum clearance. Rolling from zero clearance may pile up paint ahead of the ring which would indicate thickness greater than actual.
Reproducibility is stated to be within one half of the smallest graduation of a gage.
It has been observed that a substantial proportion of paints do not obey the 1 to 1 relationship. The actual thickness, obtained by independent methods, may be several times, or only a fraction of, the thickness calculated by the equation. A small amount of thinner added to a paint may increase the diameter of the spot on the lens and give a corresponding increase in the calculated thickness. This phenomenon has been ascribed to the effects of surface tension. Hence, for best results, a correction factor should be established for each type of paint based on the known thickness of a freshly prepared film measured by the Inmont gage.
Reproducibility is within about 2 percent for films 2 mils thick, decreases to about 10 percent for films about 5 mils thick, then becoming better as thickness increases.
This gage is a small square or rectangle of metal or plastic with teeth of graduated clearances cut in the edges. It is a simple, low cost gage, useful when approximate values are satisfactory. The gage is simply dipped vertically in the film until it rests firmly on the substrate. The thickness of the film lies between the highest tooth wet by the coating and the next highest tooth. The gages are available in many ranges.
This method was used to study the relationship between the clearance of a doctor blade and the thickness of the wet film left by the blade. A needle is attached vertically to the objective holder of a microscope. The barrel is lowered until the needle just touches the film that has been spread on a plain metal panel. The contact is observed through a horizontal microscope. When contact is made the needle and its image reflected by the film just meet. The needle is then lowered into the film until it just touches the metal panel. This contact is noted by the deflection of a galvanometer in series with the panel, a dry cell, and the needle. The thickness of the film is calculated from the number of turns made by the focusing screw of the vertical microscope between the two points of contact.
Dry Film Thickness
Many methods are in use for measuring dry film thickness, but only a few are suitable for all types of specimens and circumstances. Some require removal of the film from the substrate and thereby destroy or damage the film others make small holes in the film but, in some instances, may be considered nondestructive a large number depend on magnetic or electrical functions and are nondestructive. Naturally, these types are limited to metallic substrates, specifically ferromagnetic metals for the magnetic methods.
This tool may be used in the conventional manner to obtain the thickness of a free chip of paint by direct measurement, or of a film on a substrate by measuring the total thickness, removing the film by solvent or scraper, and measuring the thickness of the substrate at the spot where the film was removed. The thickness of the film is the difference between the two measurements. ASTM Method D 374, Thickness of Solid Electrical Insulation, contains instruction for calibrating and using machinists micrometers, with and without ratchet, and with dead weight loading.
A preferred procedure is, set forth in ASTM Method D 1005, Measurement of Dry Film Thickness of Organic Coatings. This method directs the use of a dial micrometer rigidly mounted on a support and the use of clamps to hold the specimen firmly during the measurement, and sets forth approved manipulation and adjustment of the micrometer. The method is recommended only for thicknesses over 0.5 mil and is accurate within about 0.1 mil.
Gardner Needle Thickness Gage
This instrument is designed to measure the thickness of electrically nonconducting films on metal (conducting) substrates. It is small enough to be used in the field where the substrate can be made part of the electric circuit. The needle makes only a minute puncture in the film. In many instances, particularly in a go no go determination, the damage is so slight that the method may be considered nondestructive for many end uses.
the cast aluminium housing contains the needle, screws for forcing the needle through the film, and a lamp to signal when the needle contacts the substrate. For use in the field and for occasional use in the laboratory, the electric circuit comprises the needle, the substrate, a dry cell, the lamp, and a cord that connects with the substrate. In the laboratory, if many measurements are to be made, it is advisable to use a step down transformer and to connect to a 110 V source.
The zero setting of the needle is obtained by retracting the needle within the housing, placing the gage on a plane metal block, and lowering the needle until the lamp signals contact. The block is replaced by the specimen, and the needle is lowered until the lamp again signals contact. The difference between the two readings is the thickness of the film.
Gardner Carboloy Drill Thickness Gage
In many instance films are so hard that they successfully resist penetration by the needle penetrometer described above. This difficulty has been overcome by substituting a Carboloy drill for the needle. The drill is a needle terminating in a pyramid having three faces. The drill is secured in a chuck that can be rotated and advanced independently. The rotation is controlled by finger action on a knob at the upper end of the chuck shaft. In all other respects the operation is the same as that of the needle gage.
This method has heen standardized by ASTM as Procedure B in Method D 1400. Measurement of Dry Film Thickness of Nonmetallic Coatings of Paint, Varnish, Lacquer and Related Products Applied on a Nonmagnetic Base.
Gardner Gage Stand
Although the Gardner Needle Gage and the Carboloy Drill Gage may be operated by manually holding the gage against the specimen, using the Gardner Gage Stand is less tiring and more accurate, especially when many measurements are to be made. The stand provides constant known pressure (up to 10 lb) on the specimen and ensures that the needle or drill is always perpendicular to the specimen.
Gardner Micro Depth Gage
Measurement is not restricted to nonmetallic films on metal any type of film on any type of substrate may be measured, and the film is always damaged.
In this gage, a chisel replaces the needle of the gage. The zero setting having been established, the chisel is advanced by an amount estimated to be less than the thickness of the film. The gage is placed on the specimen and drawn toward the operator through a distance of a few millimeters. If the scratch made by the chisel does not penetrate the film, the chisel is advanced by a small increment, and another scratch is made. The procedure is repeated until the substrate is reached and exposed. Inspection is best made with the aid of a low power magnifier. The range is 0 to 40 mils. Repeatability depends on the magnitudes of the increments and the compressibility of the film and substrate. For example, specimens with poor adhesion may be torn off and expose the base even if the chisel does not penetrate the film.
Microscope for Film Thickness
Classical Method The classical method of using the microscope for determining film thickness is to measure the width with a micrometer eyepiece. For approximate results a chip of paint, which may or may not include the substrate, is used. For more accurate results, the specimen should be mounted in a block of wax, and the face of the mount is cut or ground to a smooth surface. This is a common method of preparing many kinds of material for microscopical examination.
ASTM Method D 2691, Microscopic Measurement of Dry Film Thickness of Coatings on Wood Products, is based on the above procedure. It states that the specimen may be sliced with a microtome or a razor blade.
Whitehead points up that brittle paint films are usually shattered when cut with a microtome. Also, the microtome blade is easily damage by hard paint films. He recommends grinding and polishing, somewhat as metallurgical specimens are prepared. Molten wax is flowed around the specimen and the face is leveled with a dead smooth file and finally polished with a fine hone lubricated with dilute soap solution and alcohol. Whitehead does not recommend abrasive paper as the particles of abrasive become lodged in the film and cannot be removed.
On the other hand, others have had success with abrasive paper. The face of the specimen is rubbed on a sheet of No. 80 wet or dry paper lubricated by a stream of cold water. When a suitable cross section has been exposed, it is polished successively with finer grades of paper.
Various waxes have been tried, such as a mixture of equal parts of beeswax and carnauba (carnauba is too brittle, beeswax is too soft), bayberry, montan, and others. A synthetic wax (Acrawax B) is very good. It does not fill the pores of the abrasive paper excessively.
Brightwell Method This method does not require removal of a chip and elaborate mounting and preparation. A tiny furrow is made in the film or a small chip is removed. A prism or ribbon of light is projected on the selected area at an angle of 45 deg. The distortion of the beam is examined with a microscope equipped with a micrometer eyepiece. Apparatus for this is available in the Schmaltz Optical Surface Analyzer.
Another type somewhat easier to manipulate. A balanced beam is attached to the shaft of a light duty motor spring. At the end of one arm is the magnet, which is counterbalanced by a rider on the other arm. The free end of the spring is attached to a ring. Rotation of the ring raises or lowers the magnet.
In the determination the gage is placed on the specimen and is held firmly with the thumb of one hand on the housing near the magnet with the thumb or finger of the free hand, the ring is rotated counterclockwise to bring the magnet into contact with the specimen the ring is then rotated clockwise until the magnet breaks loose. The thickness is read on the scale on the rim of the ring.
Probably no test on coatings has given as much trouble as drying time, unless it is hardness. One reason is the subjective nature of the tests. Another is the unawareness of the influence of environment and of the mechanical details of instrumental aids.
Subjective methods have been used since time immemorial. Says Werner It s been a long time since our illustrative and artistic ancestor touched the first paint film on the wall of his cave. Little did he know that he was the initiator of the world s longest established test method.
Authorities may not agree that all of the stages actually exist and, if they do, that they appear in the order named.
For many purposes, the early stages may be identified by that universal tool, the finger, especially if experienced. One exception is the dust free stage, which obviously needs dust of some kind. However, for specification purposes, instruments or tools are usually specified.
Widely used directions for determining drying time by finger methods appear in ASTM Method D 1640, Drying, Curing, or Film Formation of Organic Coatings at Room Temperature, and in Method 4061.1, Drying Time of Coatings, of Federal Test Method Standard No. 141 (FTMS). Except in a few minor respects, the two versions agree in the technical requirements. Two methods dust free and tack free times call for simple accessories dust, or lint, and Kraft paper, respectively.
The directions include expressions such as touch lightly, feel sticky, and firm pressure, which mean different things to different people. Mr. Brobdingnad has a bigger finger than has Miss Lilliput, and it would be amazing if they agreed in their subjective observations.
Lately, there has been added the International Standards Organization (ISO) Method CHM 02 64, Determination of the Degrees of Drying of Paint Films. Three degrees dust dry, tack free, and hard dry require only simple accessories ballotini (fine glass beads), paper, and wire gauze and do not depend on the finger.
Effects of Environment
As already mentioned, drying rates are influenced by the environment, mainly, temperature, relative humidity, circulation of air, and light. Frequently these factors exert more influence than the method of test.
Among the first investigators to examine the effects of environment were Schmutz and Palmer who studied the drying rates of exterior oil base house paint in environments consisting of various combinations of temperature, 32 and 100 F, relative humidity, under 10 and over 90 percent, and light, with and without, Table 1. In general, low temperature retards drying more than does excessive humidity or absence of light. Light accelerates drying, but, under very humid conditions at high temperatures, the effect is slight.
Sheetz studied the drying time and hardness of alkyd finishes over a much smaller range of temperature, 70 to 77 F, and relative humidity, 50 to 95 percent. In these ranges he found that relative humidity was a greater factor than the temperature.
In a study of factors influencing the drying time of paint, Algeo observed a difference of 4 h for a paint dried at 73 and 77 F, both at 50 percent relative humidity. The two temperatures were within the limits specified in ASTM Method D 1640.
For films that dry by oxidation the rate of drying should be a function of the concentration of oxygen at the interface. Since oxygen can reach the surface only by diffusion, the rate of drying is a function of the thickness of the stationary air film.
For films that dry by solvent evaporation, in whole or in part, the continuous removal of solvent laden air hastens drying. It is well established that the thickness of such stationary films is related to the velocity at which the main body of air is moving. Hence, the reaction rate (drying time) is profoundly influenced by local air currents. Since it is almost impossible to provide absolutely stationary air, Scofield provided a uniform current by mounting the panels on a turntable that carried them through the air at approximately 6 ft/min, a rate higher than that of local air currents, so that the latter may be disregarded.
Algeo went a step further. They regulated the flow of fresh air through the drying area to that of the environment that they wished to reproduce. Uniform composition of the air was achieved by laminar movement.
Typical standard conditions are given in Table 2. If such space is not available, a control specimen should be tested at the same time.
Besides controlling the environment, it is usually desirable to control other parameters, such as type of substrate, method of preparing the film, and the thickness of the film.
The panels for the test may be glass, metal, or other material agreed upon. The surface should be smooth. Glass, for example, should be ground and polished plate. Ordinary window glass may deviate as much as 1 mil from flatness.
Ordinarily no restriction is placed on the method of preparing the film, but doctor blade preparation is often specified. The base of the doctor blade bearing on the panel should be lubricated with a drop of high flash naphtha.
Unless otherwise specified, the film thicknesses given in Table 3 are suggested.
Set to Touch Time
According to ASTM Method D 640 and FTMS 4061.1, a film is set to touch when it clings weakly to the finger under gentle pressure, but none of the film transfers to the finger. To confirm, press the finger tip on clean glass. There should be no transfer to the glass.
A test for the efficiency of liquid drier is somewhat more severe.
Dust Free Time
According to ASTM Method D 1640 and FTMS 4061.1, prepared dusts are used for this test cotton fibers and a powder, usually calcium carbonate pigment for good visibility on white opaque coatings, the dust may be darkened by mixing with it a trace of carbon black. Other colors or clears should give no trouble.
Cotton Fiber Method
Individual fibers of absorbent cotton are separated with tweezers. At regular intervals several fibers are dropped onto the specimen from a height of 1 in. The film is considered to be dust free when a gentle current of air removes the fibers.
Rossmann proposed a similar method whereby a thread is unwound automatically from a reel or otherwise placed on the wet film. A machine for dispensing the thread is available.
The powder (Standard Multifex, Diamond Alkali Co.) is deposited on the film at regular intervals with the aid of the New York Dust Free Tester.
Charge the tester with the powder to a depth of about ½ in., place tester over the film, and operate the vibrator for 2 s. immediately blow air under about 5 psi on the deposit and gently rub the spot with a soft cloth. The film is at the dust free stage when it is clean after the rubbing. If not clean repeat at regular time intervals on different spots.
Glass Bead Method
According to ISO Method CHM 02 64, ballotini (glass beads), 0.2 mm in diameter, are sprinkled on the film. After 10 s, the specimen is tilted 20 deg from the horizontal, and the beads are brushed gently with a soft hair brush. The film is dust dry or surface dry if no trace of beads is left. Machines that automatically dispense the beads have been made, but none provides for removal of the beads in the manner mentioned. The nearest is the Paraffin machine in which the specimen is oriented at 45 deg to the stream of beads (sand), and the excess rolls off immediately.
During a study of artists, colors. The Federal Art Project of the Works Progress Administration for Massachusetts devised a metering device for sand used to measure surface dry time. It was essentially a small funnel having a valve in the stem. Sand from a reservoir is admitted through a similar valve and then discharged onto the specimen oriented at an angle of 45 deg below the funnel. All operations are manual.
Tack Free Time.
Tack is the tenacity with which an accessory, such as a piece of paper that has been pressed into contact with the film, clings to the film. The accessories and the weights used to impress them on films hardly merit being classed as instruments.
This tester is made of sheet metal, 0.015 to 0.018 in. thick,. I in, wide, and about 3 in. long, bent as shown in Fig. 3. The base is 1 in. square and the angle between the two arms is approximately 135 deg, such that a 5 g weight placed on the geometric center of the base just balances the side arm. Aluminum foil is wrapped around the base cushioned on the underside with felt, blotting paper, or smooth white paper. The New York Production Club recommends 2 SO plain coiled aluminum, 0.5 mil thick, dull side out. ASTM Method D 1640, however, specifies the shiny side out.
To make a test, place a weight (300 g for most films) on the base of the tester and set the tester on the film. Remove the weight after 5 s. The film is tack free if the tester tips over immediately upon removal of the weight.
Table 5 compares tack free times obtained with the Zapon tester, the bare finger, and the finger wrapped in aluminum foil. The shorter times of the bare finger method may have been caused by moisture on the finger interfering with good contact.
Bloom Drying Time Tester
A 1 cm2 disk of chrome tanned leather attached to one end of a balance beam is impressed on the film under a 50 g load for 30 s. A counterbalancing rider is drawn along the beam by a motor until it pulls the leather disk away from the film. A pen attached to the counterbalance makes a time load record on a chart.
Gardner Magnetic Tack Tester
This consists of a leather disk to be impressed on the specimen and an arrangement for pulling the disk from the specimen with magnetic force.
The disk, cross section 5 cm2, is located at the bottom of a rod hanging from the left arm of a balanced beam and counter balanced with a rider on the right arm, the right arm terminates in a fixed magnet. Below the fixed magnet is another magnet vertically adjustable with a worm and gear. A dial calibrated in millimeters indicates the extent of the vertical movement.
The specimen is placed under the disk, which is lowered into contact with the film and further impressed with a 75 g load. The dial is manipulated to raise the movable magnet until its pull on the other magnet separates the disk from the film. When the distance between the two magnets is 20 mm or more the film is tack free. When more than about 31 mm, the film may be considered dry.
MECHANICAL PROPERTIES OF FILMS
Concept and Definition
Before it is possible to arrive at a definition of hardness, one must realize that hardness is conceptual. It means different things to different people and is measured in different ways. Dealing with the hardness of relatively thin organic finishes over various substrates presents other complications not experienced when one deals with materials in bulk. That is, the apparent or measured value of hardness is dependent on factors outside of the organic finish itself. Included among these factors are the thickness of the organic finish and the nature of the substrate. Another way of expressing this is to say that hardness is not an absolute or intrinsic material property of organic finishes. It is not an absolutely definable value such as mass, length, velocity, etc. It is descriptive of many things and is variable. Trying to limit the hardness of organic finishes to a particular, intrinsic material property, is like trying to confine mercury in one s fist.
Over the years many methods and devices have been employed to measure the hardness of organic finishes. In 1946 ASTM conducted a survey to determine what methods were being used by the paint industry to measure hardness. They found that some forn of scratching or abrasion was used 21 times, pendulum or damping hardness (also referred to as entropy hardness) 14 times, and indentation hardness 5 times. The modem trend has been towards an increasing use of indentation methods. Indentation hardness has been universally accepted throughout the scientific and engineering communities. In addition to those methods just mentioned, some work has been done using impact and rebound. However, these latter methods have not been seriously considered for measuring paint hardness.
All of this leads us to a definition of hardness based on practical concepts. A material is considered to be hard if it resists indentation or is not easily scratched. Therefore, hardness is resistance hardness is a resistance to indentation or scratching. However, the latter part of this definition presents more problems than it solves.
Resistance to strictly surface scratching is within the province of mar resistance, which is covered later on Organic finishes are often doped with silicones or waxy additives to increase this surface scratch resistance, while maintaining the same resistance to indentation. Scratching through to the substrate has been used to evaluate adhesion. Perhaps the distinguishing difference here is that cutting through with a relatively wide instrument or tool, wherein removal of the finish from the substrate occurs, is more in the realm of adhesion while resistance to scratching through with a relatively narrow or pointed tool is considered indicative of hardness. The use of scratch tests for assessing the hardness of organic finishes was undoubtedly influenced by the use of scratch tests for minerals and metals.
However, at the expense of being redundant, we may note that the modem trend in all areas has been towards the increasing use and adaptation of indentation hardness tests. These tests are far more objective and precise than the scratch tests. If resistance to indentation is used as a measure of hardness, then such hardness should be related to the rigidity of the material or more precisely, the elastic modulus. With certain Qualifications this is true.
Laurie Baily Hardness Tester
This apparatus was among the first to be developed for measuring the scratch hardness of varnish films and was invented by A. P. Laurie and F. G. Baily of Heriot Watt College, Edinburg. The apparatus consists essentially of a hardened, blunt steel point upon which pressure is exerted by a vertical coil spring. The spring tension is controlled by an adjusting screw and can vary from 0 to 2000 g. The finish to be tested is placed under the point and slowly moved horizontally by hand as the pressure is increased until a scratch is made. Then the reading on the scale is recorded.
Graham Linton Hardness Tester
This device might also be considered an adhesion tester. As shown in the figure, it is essentially a small, circular blade upon which pressure is exerted by a coil spring. A scale, graduated in 100 g increments from 0 to 2000 g, indicates the load on the blade.
Clemen Hardness Tester
consists of a balanced beam, on one arm of which is a short knife and a post upon which weights are placed. The specimen rests on a tray that is drawn under the knife during the test.
duPont Scratch Testing Machine
This type of instrument was once used at the duPont lacquer plant at Parlin, N.J., for determining the hardness of lacquers and their resistance to scratching. It was one of the first to electrically signal the end point of the test.
The device consists of a wooden base on which are mounted (1) a fulcrum holding a graduated lever equipped with weights and a needle point, (2) a transformer, (3) a 6 V lamp, and (4) a metal plate. These parts are connected in series, the transformer being used to step down ordinary light voltage to that of the small lamp.
The metal panel, coated with the lacquer, is placed coated side up on the metal plate under the needle. The weight is adjusted, and the panel is drawn along the plate by the operator in the direction of the long axis of the instrument. This operation is repeated, each time using an increased weight, until the needle penetrates the film. When this happens, the electric circuit is closed, and the lamp lights.
Hoffman Scratch Tester
this to be a low carriage with a weighted lever on one end. The scratching tool is a sharp edged, hardened steel cylinder with its axis at an angle of 45 deg to the plane of the film. This cylinder is attached to the lever arm, and the load is varied by varying the position of the weight on the lever. This has also been used for adhesion and mar resistance tests
Taber Shear, Scratch Tester
In this tester, the tool is fixed to the underside of a beam, pivoted on ball bearings. Riders provide for adjusting the load on the tool between zero and 1000 g. The test films for this tester are prepared on panels containing a hole in the middle for locating on a turntable. In making a test, the panel is rotated counterclockwise. Three tools are provided thumb nail contour shear tool (S 20) lapped to a 25 mm radius with a 30 deg clearance, a diamond cut to the shape (diamond pyramid) of a corner of a cube, and a diamond cut to the shape of a cone.
This device uses the corner of a diamond cube (diamond pyramid) as a scratching tool. It is a rather elaborate device consisting of a microscope, stage, and diamond tool on a balanced arm. Bierbaum scratch hardness is the ratio of the load on the diamond, in kilograms, to the square of the scratch width, in millimeters. Although it was designed for testing metals, it had been used on plastics for a while. It never gained much acceptance for use on organic finishes.
Schopper Hardness Tester
This device was one of the first to provide for automatically increasing the load on the scratching tool while the scratch is being made. Arms extending upward from the panel carrier end in slots above the beam carrying the scratching tool A roller resting on the beam is guided by the slots. As the panel carrier is drawn along, the roller travels with it, thereby increasing the load on the scratching tool. Provision is made for automatically lifting the load from the specimen at the end of each trip and also for a sidewise displacement of the specimen in order to provide a new path for repeat tests. The interpretation of results is the same as with other types of scratching devices, that is, according to the character of the mark under a particular load or tool, or the load at which a particular tool makes a mark.
Parker Siddle Scratch Tester
This tester is a very simple form of the Schopper type in which the load on a needle is increased as the scratch is made. The point of the needle is a hemisphere 0.2 mm in diameter. The panel carrier is moved by hand, the speed recommended by Parker and Siddle being 30 cm per min. appreciably greater speeds give inconsistent results.
Simmons Scratch Tester
This is another tester of the increasing load type and is suitable only for films on metal. When the stylus breaks through the film, a relay stops the machine. Hardness is reported as the weight necessary to penetrate the film.
Dantuma Scratch Tester
This device employs a novel means of increasing the load during the travel of the scratching tool across the film. as arm B, with tool G resting on panel H, is lowered, arm A follows. The scratching tool travels from H to L, a distance of 5 cm. The load varies from 0 to 5000 g. Operation may be by hand crank or by motor.
Four types of hardened steel scratching tools are provided a ball 1 mm in diameter, a simulated finger nail, and two wedges.
Rondeau Scratch Tester
This device shown in Fig. 1, also belongs to the type where the load on the scratching tool automatically increases as the test is being made. The tool, parabolic in shape, is mounted on the free end of a cantilever spring, one end being movable in a slot in the frame. At the start of a test, the tool rests on the test surface under zero load. The finish end of the slot is 0.100 in. nearer to the test surface than it is at the start. At the finish end, the load is the rated value of the spring. At intermediate distances, the load is proportional to the distance. Three springs are provided, giving loads of 300, 600, and 1200 at the finish end.
Sheppard Schmitt Scratch Dynamometer
The principle of this device is similar to that of the Schopper tester. The scratching tool is a hardened steel, 45 deg tetrahedron.
Measurement of scratch resistance is expressed either as the threshold load producing a scratch, or by a curve expressing the relation between the load and size (width) of the scratch.
this device in which the scratching tool, a 60 deg diamond point, rests nearly vertically on the test panel which is clamped to a movable stage. In making a scratch, the assembly containing the diamond point is moved by means of an electric motor. The load is adjusted until two strokes in the same place just cut through the coating and a third stroke cuts into the supporting panel. This load (in grams) is reported as, the hardness of the film. By means of a micrometer screw, the stage may be displaced for successive determinations. The Arco Microknife used in a different manner has been standardized as an adhesion tester. This is covered in the section on adhesion.
Rating the hardness of an organic finish according to the hardness of a lead (graphite) pencil that will just scratch it was probably an old custom when described by Wilkinson. Gardner studied the method using pencils sharpened to different shapes sharp cones, rounded cones, and chisels. He found that the principal source of error lay in the character of the point, because it was difficult to reproduce points. Other sources of error were the pressure on the pencil and the angle at which the pencil was held while it was moved over the organic finish. Gardner built a device to hold eight pencils at one time, at an angle of 45 deg to the panel, but found that it was impossible to align all pencils uniformly.
Hold the pencil in a writing position, that is, at approximately 45 deg, and push forward against the film. Use pressure short of breaking the lead. By turning the pencil after a test a new edge is available for use, and three or four trials may be made with one dressing of the lead.
Clean the marks with a soap or artgum eraser. Any marring of the surface, visible at an oblique angle in strong light, indicates that the pencil is harder than the film. The hardness is expressed as the grade of the next softer pencil.
In developing this method, Smith determined the pencil hardness of 14 different organic finishes that varied widely in hardness. Five different brands of pencils were used, and the results are shown in Table 1. The results indicated that there were variations in the hardness between different brands of pencils and that it was necessary to use only one brand in order to obtain reproducibility.
Mechanical Pencil Method
The Erichsen Company in Germany has marketed a mechanical Hardness Test Pencil (DBGM), Model 318, which somewhat resembles a mechanical pencil in appearance, but whose point is a spring loaded tungsten carbide hemisphere or ball point of a 0.75 mm diameter. Two interchangeable helical springs are supplied with each Hardness Test Pencil to give ranges of 0 to 400 or 0 to 1000 g load on the ball point. The hardness is defined as the minimum load or force, in grams, on the ball point, that leaves a mark in the surface just visible to the unaided eye.
Pendulum Rocker (damping) Hardness
As indicated by the title, this hardness is measured by some sort of pendulum or rocker, the mechanism being the damping of the oscillations due to the hysteresis losses of the material, rolling friction, shear modulus, and in the case of relatively thin organic coatings, the hardness of the substrate. While damping of the oscillations due to rolling friction and hysteresis losses are self explanatory, one can readily see that, all other things being equal, a lower modulus (stiffness) will result in deeper indentation of the ball or rocker into the material, resulting in a greater rate of decrease of the oscillations. It logically follows, therefore, that this type of hardness should be related to the shear modulus of a material. It is, as will be shown in the section on the Sward Rocker.
Walker Steele Swinging Beam
The principle of this beam is that of the Herbert Pendulum Hardness Tester, a device for testing metals. This is a metal form weighing from 2 to 4 kg, supported by a 1 mm ball of ruby or steel. Normally the center of gravity is adjusted to a fraction of a millimeter below the center of the ball. With the ball resting on the test surface, the pendulum is set into oscillation through a small angle, and the time required for ten oscillations is determined. This time depends on the nature of the surface. The pendulum may also be tilted from the horizontal through a selected angle and then released. The angle reached at the swing in the other direction is called the scale hardness number. A curved spirit level and scale on the pendulum allow angles to be read.
The harder the film, the longer it will require for the beam to come to rest, after having been set to oscillating through any specific angle. However, instead of the dead stop scheme of measurement, it is easier to determine the time for the amplitude to decrease by a specified fraction, say 50 percent, or what amounts to the same thing, to count the number of oscillations. Amplitude may be measured on a circular scale behind the end of one of the arms. The scale may be divided in degrees or in arbitrary divisions.
It is necessary to have a reference standard of known hardness. Since, to date, no organic composition can be relied upon, polished plate glass is used.
The problem of resistance to abrasion exists all around us on floors, walls, furniture, automobiles, highways, signs, military gear, spacecraft, etc. Because abrasion or wear resistance is a basic factor in the durability of a coating, its measurement is of practical importance to both producer and consumer.
Abrasion resistance is not a unique or isolated property of a material but is related to other physical properties, such as hardness and mar resistance cohesive and tensile strength modulus of elasticity, and toughness. From the standpoint of retaining its protective or decorative function, the thickness of a coating is an important factor in addition to its intrinsic abrasion resistance.
When we seek to measure abrasion resistance, we are measuring a complex composite of many different but inter related properties. It is not surprising, therefore, that a wide variety of testing devices and machines have been developed in the quest for a practical test method that correlates with actual service performance.
Abrasion resistance may be defined generally as the ability of a material to withstand mechanical action, such as rubbing, scraping, or erosion that tends progressively to remove material from its surface. This definition applies also to organic coatings or paint films. Implicit in the definition is the concept of a gradual or progressive wearing away of the entire thickness of the coating. In this respect, it is distinguished from mar resistance, which pertains only to surface effects manifest as visual changes in appearance as a result of mechanical action.
Relation to Other Physical Properties
Let us consider briefly the relationship of abrasion resistance to the other physical so properties mentioned in our introductory paragraph. This may help us to understand the mechanism of abrasion and the problems involved in its measurement.
While this property is sometimes confused with abrasion resistance, those concerned with the problem in the American Society for Testing and Materials (ASTM) have found it practical to consider mar resistance as a surface property, whereas abrasion resistance involves the body of the material as well. For example, in an automobile finish, where good appearance is a prime requirement, mar resistance would be a very important property. On an army tank, however, where surface appearance is of secondary importance, it is the abrasion resistance of the coating throughout its thickness that determines how long it will remain in place to perform its primary function of protecting the underlying steel from corrosion.
Abrasion resistance obviously is related to hardness, yet the relationship is not a simple one. It might seem, at first thought, that the harder a coating or other material, the better would be the abrasion resistance. This is not necessarily true. Steel is much harder than rubber, but stell tires on automobiles, besides giving a very rough ride, would not last long on concrete roads compared to the service life of a good rubber tire. The ability of a material, like a rubber tire, to undergo elastic deformation or to ride with the blow is associated with good abrasion resistance.
The energy transferred to an elastic material by an impacting object or particle is returned largely to the particle, though redirected, instead of being expended in the destruction of (separation and removal of material from) the impacted surface. From a fundamental point of view, this is a consequence of the smaller deceleration and hence smaller force generated when the impact is with a material that will deform or give, if the deformation is not elastic, of the material will yield and flow. Hence, soft materials of low tensile strength are not abrasion resistant.
The fact that elastic materials often are abrasion resistant does not mean that hard materials are not abrasion resistant. Theoretically, however, given two materials of equal tensile strength, the material of lower modulus should have the better abrasion resistance. The deceptive factor here is that a hard material usually has a much higher tensile strength than a soft one. Thus, when we compare rubber with steel, we are comparing different orders of magnitude in tensile strength. The fact that rubber is abrasion resistant does emphasize the value of a low modulus of elasticity, coupled with adequate tensile strength, as factors in good abrasion resistance.
In theory, a very hard material, of hardness and cohesive strength adequate to completely resist any impact force it might encounter, would not be dependent upon rubber like elasticity to reduce impact stresses, and would be more abrasion resistant than the best elastic but weaker material. In practice, the writer feels that hardness bears a closer correspondence to mar resistance than to abrasion resistance, while abrasion resistance is related more closely to toughness than to hardness.
Modulus of Elasticity and Tensile Strength
Since hardness and modulus are closely interrelated, what has been stated about the relation of abrasion resistance to hardness holds quite well for modulus of elasticity also but we must not forget the role of tensile strength. Other factors being equal, the higher the tensile strength (or cohesive strength) of a material, the more abrasion resistant it is likely to be.
Correlation with Service Performance
From the foregoing, it is evident that the measurement of abrasion resistance involves measuring a complex composite of many interrelated properties among which there is no 1 to 1 correspondence. The task of devising a test methodology that will correlate with service performance is therefore complex and difficult but not impossible. If the test method we use, whether by chance or by design, subjects the material under test to a composite of destructive forces similar to those encountered in service, then the test method will correlate with or predict the service performance of the material, at least with respect to the ranking of materials.
With all accelerated tests, even those that rank materials in the same sequence as actual service tests, a quantitative correspondence with actual service is seldom expected or obtained. Actual service tests, while the most reliable in providing an indication of the probable long term durability of a material, suffer from the difficulties of ensuring equivalent usage and measurement, especially when the inter comparison of different materials is attempted. Because of such difficulties and because service tests are usually very time consuming, a wide variety of test machines have been developed to provide an accelerated indication of the abrasion resistance of coatings and related materials such as plastic tile flooring, linoleum, and wall coverings. An investigation by the International Study Committee for Wear Tests of Flooring Materials in which seven commercial, organic flooring materials were systematically tested on 21 abrasion machines of 17 different types, indicated that very few of these machines were capable of providing a reliable inter comparison of the abrasion resistance of widely different materials that could be correlated with service performance nor did the different machines correlate well with each other.
On the basis of a round robin conducted by ASTM on six different clear floor coatings evaluated by six different abrasion test methods only two of the methods were found to correlate with actual service performance and to have the reproducibility necessary for acceptance as ASTM standards. These were the falling sand method and the air blast abrasion tester. The jet abrader which has since become available appears to correlate well with the two aforementioned methods and with various types of service performance while offering greater speed and precision of measurements. All three of these methods are discussed in the accompanying text.
Mechanism of Abrasion
The success of the particle impingement types of abrasion testers in correlating with service performance is perhaps not surprising if one considers the abrading mechanism. Although a comprehensive discussion of the mechanism of abrasion is beyond the scope of this article, some useful observations can be made.
Whether or not a particular type of abrasion test correlates with service performance depends not only on a similarity of abrading mechanisms in the two cases but also on the extent to which that mechanism is maintained during the course of the abrasion test. It is on the latter point that many of the available methods fail.
Rubbing (friction) and scraping methods obviously wear away the test surface in a different way than methods in which abrasive particles are blasted at the surface. One aspect of the mechanistic difference lies in the angle of attack upon the surface. Abrasive particles striking the surface of a coating at near normal incidence would tend to compress, scar, and cut into the coating, so that minute portions of the coating eventually would be crosscut and displaced. On the other hand, the rubbing and scraping types of abrasion that take uplace at near grazing incidence would tend to undercut and to shear through minutely thin layers of the coating in successive slices that would ultimately wear it away. Different machines might incorporate various degrees of these basic processes, depending on the angle and force of attack. Apart from the nature of the mechanisms considered before, it should be apparent that whatever the mechanism, it is not maintained uniformly in friction methods. The latter inevitably suffer from change in the abrading conditions as the testing proceeds, either because of heating of the specimen or clogging of the abradant or both.
If, to avoid undesirable friction effects, we choose an abrasion test employing the direct impingement of free particles, we might have some doubt about whether the method would correlate well with a type of service in which the coating, for example, is walked upon. Yet, the previously mentioned ASTM round robin of abrasion tests on floor coatings clearly established the validity of particle impingement type tests for evaluating this type of service.
It must be concluded that until the mechanism of abrasion is more fully the studied and understood, we must continue to rely to a large extent on experimentation, intuition, or genius in devising abrasion test methods and we must depend upon properly designed test programs to establish the extent to which a given method correlates with actual service performance.
Classification of Test Methods
Despite the failure of most abrasion testing machines to rank a widely differing range of materials consistently or to correlate with actual service performance, many of the machines and methods give useful results when comparisons are limited to similar types of materials or it when the results are performed and interpreted on a single machine in a single laboratory experienced in its use. A wide a variety of such useful abrasion test methods and machines is described in the following sections.
Abrasion testing devices for coatings may be classified either by the type of abrading action they exert (for example, abrasive blast, rotating wheels, etc.) or according to the nature or function of the material they are designed to test (for example, traffic paints, automotive finishes, etc.). In this chapter, we have elected to classify the test methods according to type of abrading action, except for traffic paints and a miscellaneous category. Traffic paints are mentioned here only briefly.
The following sections present a description and discussion of a selected group of abrasive testing devices that have found acceptance and use in the coatings field.
Concept and Definition
Before one can evaluate various instruments and methods for measuring the adhesion of organic coatings, it is important to have a fundamental understanding of what measuring adhesion means in the light of modern science, since so many significant discoveries have been made within the past decade. Only after we have this understanding can we appreciate which methods are useful and which have little or no practical value. ASTM Designation D 907, Definition of Terms Relating to Adhesion, defines adhesion as the state in which two surfaces are held together by interfacial forces which may consist of valence forces or interlocking action, or both. If we speak of adhesion as these interfacial forces, then we must abandon hopes of trying to measure adhesion. That is, we cannot measure these interfacial forces by means of mechanical tests. However, even if we could, knowledge of the magnitude of these forces would not necessarily mean anything in a practical sense, since adhesive failures have invariably proven to be cohesive.
To review briefly the aspects involved in trying to measure adhesion as defined in ASTM Designation D 907, assume for the moment that interfacial failure is possible and that the organic coating material has freely wet the substrate. Also assume that the substrate is flat and planar (on an atomic scale) and that we can apply a force to the coating in such a way that it is normal to the substrate surface, and that no energy is expended in hysteresis or deforming the coating or substrate.
With regard to what is wetting the substrate, we know that during application organic finishes are liquid mixtures (even dry resin powders must achieve a fluid state in order for them to flow and adhere) of resins, pigments, solvents, diluents, and various additives, any of which can result in weak boundary layers at the coating substrate interface. Even with the resins themselves, there often exist low molecular weight fractions which can form weak boundary layers. By a weak boundary layer we mean inherently weak substance at or near the coating substrate interface, which will substantially reduce the apparent adhesion. Some examples are stearates, heat and ultraviolet stabilizers, plasticizers, mar reducing additives (such as silicones, waxes, and polyethylene), low molecular weight fractions of polymeric resins, and others mentioned in the following paragraph. Even though surface chemistry tells us that organic materials in the liquid state should freely wet and therefore adhere to comparatively high free energy surfaces such as metals and metal oxides, one of the weaker substances in our organic coating mixture may preferentially wet the substrate. Zisman has shown that preferential wetting can and does occur with mixtures of organic liquids. Hence, there is a problem in ascertaining just which component of the mixture is wetting the substrate.
Identification of the substrate is also a problem. As an example, one rarely has a pure iron, aluminum, zinc, etc., surface. Instead, there exist layers of water, mixed oxides, carbonates, carbides, water soluble salts, and various other materials referred to as dirt. Vapor degreasing applies a layer of organic material to the surface and perhaps metal carbonyls. Any of these can result in weak boundary layers. As an illustration, our laboratory experienced difficulty in obtaining strong adhesive joints with brass. Investigation revealed that a zinc compound was causing the weak boundary layer. The brass, therefore, was treated to etch the zinc away from the surface, leaving a copper surface which was then treated to form a strong layer of copper oxide. This resulted in adhesive joints so strong that the brass broke instead of the joint. Zinc by itself is usually treated to form a phosphate layer at the surface. Aluminum and iron can be treated to form strong oxide layers on the surface. However, even if the surface is treated to form a relatively strong surface layer materials in the atmosphere impinge, fall, and are adsorbed onto the treated surface, thereby contaminating it. In some cases where an organic coating is applied to a thermoplastic substrate, an interface, as such, may not even exist. Instead, there may be a zone of mutually inter diffused material between the coating and substrate, due to solution of the substrate by solvents in the liquid coating material.
In light of the above, we cannot really answer the two questions previously posed. To carry the argument further, it is obvious that we can never have an absolutely flat, planar substrate surface. This consideration precludes the possibility of ever obtaining a force that is normal to the interface. Also, the existence of true interfacial failure has never been proven. Studies employing electron microscopy and other techniques have shown that there is usually a layer of polymer left on the substrate surface, although this may not be apparent under an optical microscope.
In practice, we can never be sure of the substrate area that has been wet or covered by the coating. Even a highly polished surface looks like a rough mountain range on a microscopic scale. As an example, fire polished glass has an average peak to valley distance of about 400 A. If we have a non wetting liquid, interfacial contact area is low. If we have a wetting liquid, the interfacial contact area is usually high. Even then, high viscosity may limit contact to the peaks. Capillary pressure may prevent the liquid polymer from flowing into the pits and similar irregularities.
The state or configuration of the polymer chains in solution can also influence the interfacial contact area. The solvent system determines whether the dissolved polymer chains are in coil form or relatively flat and open. Adsorption studies of polystyrene on graphited, nonporous carbon black illustrated the effect of solvent on configuration and surface orientation. In poor solvents wherein the polymer forms a loop or coil structure, only part of the polymer segments was attached directly to the carbon surface. In good solvents, the polymer attached itself to the carbon surface as a flat, oriented monolayer.
However, even if all of our initial assumptions were correct and we knew the answer to the two questions, we still could not measure the interfacial forces by means of mechanical tests, because no test has ever been, devised that can eliminate the influence of the rheological and other physical properties of both coating (or adhesive) and substrate, joint geometry, stress concentrations, etc., nor can any test method ensure the same manner or mode of failure for all materials. For instance, a direct tension test of adhesive joints may have many of the features of a rapid stripping or peeling. Further, in peeling, an increase in the elastic modulus of the coating or adhesive will result in a sharper angle of peel and apparent lower adhesion due to the higher stress concentrations at the root of the peel apex. This can also change the mode of failure from ductile shear to brittle fracture. One can cause an increase in the elastic modulus of the coating merely by increasing the speed of peeling or by lowering the temperature.
Therefore, if we cannot measure the interfacial forces of adhesion as defined by, ASTM Designation D 907, what can we measure? The answer is that we measure a force of removal, which is probably totally unrelated to the interfacial forces. Let us therefore define adhesion in a practical sense, as the force required to remove an organic coating, break an adhesive joint, peel a tape, etc., the nature of the test being dictated by the expected mode of stress or removal in service. To repeat, when we measure what we refer to as adhesion, what we are really measuring is a force of removal.
If we were testing adhesive tapes, then some sort of stripping or peel test would be used to measure the peel strength. If we were using adhesives to bond metals together to form cabinets, we would at least want to perform lap shear tests to obtain the joint strength.
We would not normally use a peel test for organic coatings unless this is the expected mode of failure and either the coating or substrate is flexible and strong enough to be peeled. Similarly, organic coatings are not used to make adhesive joints and we would not normally use tests that measure joint strength. However, there are exceptions. One exception would be in the case of relatively thick, hard, strong, tile like coatings that may defy other commonly used methods. To summarize then, we could say that the type of test used should be dictated by the mode of failure expected in service, which may require more than one type of test.
Classification of Test Methods
For the most part organic coatings are removed in service by abrasion, chipping, cutting, or scraping off through the use of knives, coins, and other instruments, picking away at exposed edges, corrosion of the substrate, impact or impingement by stones, etc. Most failures start at the edges or corners of objects and not in the middle of a flat section.
At one time the automotive industry made extensive use of the Gravelometer, because they believed that it best duplicated the kind of impact and chipping failure (chipping at door edges, etc.) encountered in service. Over the past decade they have placed less reliance on it, and this writer believes that most people who own a late model automobile will agree that they should place more reliance on it. However, most technicians judge how well an organic coating adheres to a substrate by how difficult it is to remove with a penknife. A penknife is used almost universally because, among other things, its mode of removing coatings is fairly representative of the type of removal failure encountered in service. Many devices developed for testing the adhesion force of removal of organic coatings have been, literally speaking, mechanized knives. Of course there have also been other types of cutting devices, scratching devices and a wide variety of exotic and gimmick types of devices. We must bear in mind that the usefulness of any device or method should not be judged by its complexity, impressiveness, or novelty. Indeed, in this writer s opinion, some of the latter are the most useless.
Rossmann Chisel Adhesion Test
The first record of a chisel adhesion test appears to be that of Rossmann. A chisel under a definite load is caused to bite into the film at a fixed angle. Meanwhile the panel supporting the film is drawn against the edge of the chisel by means of a spring. The tension of the spring diminishes during the test until it is nolonger sufficient to separate the film from its support. The principle of the device is shown in Fig. 1, but in practice the chisel is a safety razor blade, a new one being used for each test. It is turned to such an angle that it cuts a path 1 cm wide. This allows convenient loads and forces to be used. The razor blade is inclined at an angle of 45 deg to the panel, the load on it is usually 10 kg, and the starting tension on the spring varies from 2 to 10 kg. A time tension curve is recorded on a rotating drum. From this curve, various characteristics of the adhesive forces may be read or calculated.