ENVIRONMENTAL TESTING OF BITUMINOUS COATINGS - Industrial

Ind. Eng. Chem. , 1966, 58 (6), pp 55–58. DOI: 10.1021/ie50678a008. Publication Date: June 1966. Note: In lieu of an abstract, this is the article's...
0 downloads 0 Views 3MB Size
ENVlRONMENTAl TESTMI OF BITUMINOUS COATINGS J. J. Meany, Jr.

nvironmental testing programs should reflect the Although there is a great variety of testing programs in progress, most of them are conducted for only a few reasons. We first consider some of the purposes.

E purposes of the tests.

Production Quality Control

This is probably the least frequently encountered reason for a n environmental testing program because few environmental tests yield data quickly enough to be of use in controlling production quality. In one of the few examples of such a program, an applicator of protective coatings for underground piping removes samples of the pipe from each production run. This run is not released from the mill until the samples are tested in a salt water solution, under cathodic protection for a 24- to 48-hour period. Disbonding of the coating due to cathodic protection must not exceed prescribed limits set by the applicator. Though the value of this procedure may be suspect because of the highly artificial environment, it does represent a n effort to control end quality through an environmental testing procedure.

misleading, will remain valuable, especially when conducted by an experienced evaluator. Substantiation of Sales Claims

This purpose may frequently lead to unobjective testing, but is quite common. There is no lack of ethics in providing a sales department with test data which emphasize the good properties of a particular product. Testing managers, however, can easily overstep the limits of credibility, thus making their findings suspect in the eyes of the salesmen or the customer. Compliance with Purchaser’s Specifications

Such testing programs often tie the hands of the investigator. The widespread dependence of some governmental agencies, in the past, upon specific tests which were unrelated to end use frequently resulted in hostility toward environmental testing. This does not reflect on the validity of the tests, but it does show the importance of proper selection of tests and programs, as well as of the enlightened interpretation of the results.

An Aid in Developing or Modifying Products

Selection of a Coating for a Specific Need

Here we are again required rapidly to obtain results for quick evaluation. Frequently, attempts to accelerate environmental effects overemphasize some effects a t the expense of others. Unfortunately, the intensification of one or more environmental factors does not produce the same short-term effects as a less intense factor operating over a longer period of time. Consequently, accelerated tests must be developed with care and interpreted with caution born of experience. Obviously, each new development cannot be evaluated by full-term, natural environmental testing. Accelerated tests, though often

The remainder of the paper will be devoted to exploring the use of environmental testing for this purpose.

John J . Meany, J r . , is Vice President of Ocean City Research Corp., Ocean City,N . J .

AUTHOR

ATMOSPHERIC ENVl RON MENT Most of the procedures available for laboratory evaluation of coatings utilize artificial sources of radiation to simulate sunlight, and fog-chamber devices. The use of such apparatus is governed by ASTM Standard D529-62, “Accelerated Weathering Tests of Bituminous Materials.” While the use of this apparatus will result in more rapid deterioration than will be found in natural environments, there is no reliable means of equating the time of artificial exposure to the corresponding time of VOL. 5 8

NO. 6 J U N E 1 9 6 6

55

natural exposure involving failure of the coating. Thus, it is possible to arrive a t “artificial” endurance rankings substantially different from those obtained in natural environments. Modification of the ASTM procedure by means of a xenon-arc apparatus has yielded more meaningful data according to Martin ( 6 ) , who compared carbon-arc, xenon-arc, and natural radiation. Martin increased the sensitivity of evaluation for asphalt degradation by the use of the change in absorbance of the carbonyl functional groups. His ranking of degradation obtained from the xenon arc compared directly with that obtained from natural sunlight with radiation monitored by the Eppley pyroheliometer. There were discrepancies between degradations obtained from carbon-arc and natural sunlight. While the use of weatherometer-type apparatus has its limitations, careful evaluation of its results can be useful in classifying material with gross differences in weathering resistance qualities. Other laboratory tests are arbitrary and are typified by the procedures of Flournoy (3). He suggested three tests as follows : -Thickness: a minimum of 0.005 inch for corrosive atmospheres -Flaws and Holidays: electrical resistance of a coated, 4 x 2 inch panel must be at least initially 1,000,000 ohms and must be at least 750,000 ohms after 24 hours of water immersion. -Flexibility: the test panel, after being bent 90” around a 1-inch diameter rod, must have no obvious coating failures and must retain an electrical resistance of 750,000 ohms. These tests seem to be arbitrary criteria for acceptance, and are applicable only to certain circumstances. The short-term environmental test, immersing the test coupon in water, could give a rough idea of initial coating condition, but a 24-hour test seems of little value other than to eliminate the obviously unsuitable coatings. Natural Environmental Tests

Exposure to a natural environment is the best method of environmental testing of coatings for atmospheric exposure. The major disadvantage is the excessive time required for a full evaluation. There is also some difficulty in knowing exactly what factors influence the exposure. Caryl (2) suggests the use of mirrors to intensify solar radiation. His studies with a system which maintained the test panels facing the sun (including tests with and without periodic washings with distilled water) indicated that only 14 weeks were required to simulate natural exposure for 3 years a t a 45” southern exposure. Other investigators utilize special panels with angles, crevices, rivet heads, weld splatter, and other severe surfaces to subject the coatings to the most difficult conditions leading to early failure in service. Such panels have considerable merit and can lead to early elimination of inferior coatings. The evaluation of the coatings on these special panels is more complicated and timeconsuming, but the data are of greater value. 56

INDUSTRIAL A N D ENGINEERING CHEMISTRY

Because natural environments vary substantially from place to place, it is necessary to devise test methods which reflect this fact. Solar radiation is a factor of major concern. The evaluation of radiation is relatively simple, however, through the use of the Eppley pyroheliometer and appropriate recording and integrating devices. Generally records are kept of the total langleys, langleys/0.823, and hours of radiation over 0.823. Martin demonstrated a close correlation between total langleys and asphalt oxidation. The pyroheliometer, together with further development of microtechniques, seems to offer a path for very meaningful, relatively short-term evaluations using natural sunlight. Other factors of major importance are temperature variations and relative humidity. Atmospheric pollutants may also be important in some localities. UNDERGROUND AND UNDERWATER EXPOSURE Underground and underwater environments impose a necessity for greater physical strength and higher electrical resistance, and usually require much longer useful lives than those of coatings intended for exposure to the atmosphere. For purposes of further discussion, we shall consider only those applications where the minimum life of the coatings is in excess of 30 years. Laboratory TOE!*

The U. S. Bureau of Reclamation (4) utilizes a series of tests to evaluate coatings for use in submerged service. These tests include : fresh water immersion ( 7 ) , salt-spray box (7), outdoor weathering (8),and weatherometer ( 9 ) . For specific problems, the Bureau designs specific tests. There have been some standards developed for the quality evaluation of specific bituminous coatings, but we lack standardization of tests, even for comparing various coatings with each other. Generally, investigators develop their own tests and procedures in much the same manner that the Bureau of Reclamation does. I t appears that this situation will not change in the immediate future. Let’s consider the development test program for evaluation of an underground protective coating intended to serve for 30 years. The structure to which the coating is to be applied might possibly be cathodically protected. The investigator’s first problem is finding a method to eliminate the obviously inferior coatings. Experienced judgment is usually the fastest method, but judgment must be substantiated with test data. The “salt-crock” test, or a modification of it, is usually employed in the early stages of testing. In general, the test consists of placing a coated specimen in a container of salt water and measuring the electrical resistance between the specimen and a fixed electrode. The measured resistance must exceed somearb itrary value for an arbitrary amount of time. The duration of the salt-crock tests is usually of the order of 3 to 6 months. I n some cases the specimen is maintained electrically negative to the solu-

Figure 1. Apparatus far tnodjfed sall-nock krt

tion in which it is immersed. This permits some evaluation of the coating’s performance on a cathodically protected structure. As a rough test, the salt-crock test has considerable merit. Many factors may lead to erroneous results, however. Many coatings are not easily applied to small samples of metal, and hand-applied coatings can be expected to fail in a shorter time than machine-applied coatings. There are some coatings for which the reverse is true. A short-term c m k test would reveal inherent porosity but would probably not, in ski months, necessarily show a tendency for moisture adsorption that could lead to failure in 5 to 10 years. There is also the hazard that electrical connections to the sample might bedifficult to coat and electrical leakage caused by the connection would be damaging to the test. T o illustrate one method of conducting the salt-crcck test, consider that use by the author’s company. A pipe sample is coated with the same equipment used in commercial work. The sample is selected so that no accidental flaws exist. The coated sample, generally about 18 inches in length, is placed concentrically in a second uncoated pipe of a convenient larger size, and the ends are fitted with rubber stoppers as shown in Figure 1. The annular space is filled with highly conductive brine and the electrical resistance between the sample and the casing used as a measure of coating effectiveness. High quality coatings will maintain resistances of 1000 megohms or more for long periods of time. If a value of about 10 megohms per square foot is used as a criterion, most grossly inferior coatings will be detected in a few weeks’ time. The decision to give further consideration to a given coating that passed the salt-crodr test must also consider the physical requirements for the coating. Resistance to impact, shock, and sustained loads must be determined by appropriate testing. Resistance to underfilm water migration should be evaluated by the use of samples with scratched or impacted areas. The salt-crock test can be used for this evaluation, although resistance measurements must be interpreted with this in mind. Generally,

the test is conducted with and without cathodic protection of the substrate. Coatings surviving the above tests can next be considered for further testing in environments simulating those conditions that will be encountered in service. Where service conditions are such that high soil stress conditions may be expected, tests specifically designed to evaluate soil stress conditions may be used. Samples may be placed in bentonite that is alternately wetted and dried. The performance of the coating can be evaluated electrically by measuring the resistance between the sample and ground. This would be supplemented by physical examination at the end of the test period or when electrical measurements indicated that failure had occurred. This test method seems to be worthwhile only if extreme soil stresses are expected. Natural Environment Testinfl

I t is best to test the coatings under conditions as close as possible to those encountered in service. One approach is to use an outdoor test box filled with soil typical of the area. Such an arrangement is shown in

F&wr 2. A w a t u s far conb.ollrd ssil dmW VOL 5 8

NO. 6 J U N E 1 9 6 6

57

I Figure 3. Arrangmcnt for Mhnal am‘ronnunt soil

Figure 2, with the box being permanently mounted outdoors. Aluminum foil lining is placed in the bottom of the box and test samples are inserted through sleeves with plastic spacers on the samples where they pass through the sleeves. Standard rubber casing end seals are then used to seal the space between the sample and the sleeve. The box is then filled with soil. Stainless steel guard rings are clamped on the pipe about 6 inches outside the casing seals. If the coating includes glass fabric or felt, the guard ring must be in contact with the bottom layer of fabric or felt. These rings must be installed carefully and are essential to reliable data. Use of the soil box simplifies the electrical measurements and keeps the samples relatively accessible for physical examination. The shallow nature of the box c a u m the samples to be exposed to more severe wettingdrying and temperature cycles than would be urperienced under service conditions. If grass is allowed to grow in the soil, the effects of roots would also be more w e r e . These factors must be weighed in evaluating test results. I n one 7-year test, the electrical resistance of high quality coatings was in excess of 50 megohms per square foot. Another approach to soil testing is to bury samples directly in the soil. If samples are treated in this manner, gmat care must be taken to ase.ure that the pipe ends and electrical connections are coated at least as well as the rest of the sample. This may be difficult. One

58

I N D U S T R I A L A N D E N G I N E E R I N G CHEMISTRY

tast

preferred arrangement is shown in Figure 3. The most critical factor is the encapsulation of the ends and connections. As long as this is effective, the resistance between guard rings and earth will be extremely high, and any leakage of water into the encapsulation will be detected by a change in the guard ring resistance to earth. While this test is the most difficult to set up, it also yields the most meaningful data. Probably the largest natural environment test program for protective coatings was conducted by the American Petroleum Institute (5). These tests utilized 16 test sites and involved different coating systems a p plied to 3-inch pipe at each of the sites. There were aLw 19 coatings applied to 14 working pipelines. The test program ran for the 10 years prior to 1940. While the results of this program are not directly applicable to today’s problems, the program is of interest in itself and could be used as a starting point for the resumption of similar industry-wide tests. REFERENCES (1) Bur. Rednmnhn Paint Lab. Rep. No. P-59, 1959. (2) can/l, R., F d w a l h Point T a h d . , O m d o i p t 97(481) (1965). P)ploumoy, P. W., Cormion, UP),12 (1955). (4) Lnri.,..I I. Kimit, 8. R., South Oenfnl NACE conf., OElOba 1959. (5) bean, K. H.,“API P ~ F Mating Tab,” API Div. Pmd. conl.,N o - k 19m. ( 6 ) Martin, K. G., ACS Div. Pctml. Chon., Chicap, Sqtcrnber 1964. (7) Method 6061 01Ped& Teat Method Standard 141. (8) M c W 6161 ofP d a a l Tac Method Stendad 141.

(9) Me-

6152 offddcral Test Method Staadard 141.