Download Concrete Pavement Material Surface Wear Test under Wet-Dry Weathering and more Summaries Construction in PDF only on Docsity!
TRANSPORTATION RESEARCH RECORD 1234 87
Effects of Cyclic Wetting-Drying
Weathering on Wear Resistance of
Concrete Pavement
T. F. FwA
This paper describes the development of a laboratory test proce· dure to assess the weathering effects of a wet tropical climate on wear resistance of concrete pavement ·. An accelerated weathering mode.I was fabricated to prO'ide cyclic wetting and drying treat· mcnl in a laboratory simulation of the weathering effects of Sin· gapore's climate. The effects of weathering on wear resistance were evaluated by means of a rotating drum test conducted on plain cement mortar specimens with and without weathering treatment. The tudy showed that cyclic welting and drying weathering ig· nificantly reduced the wear resistance of cement mortar test spec- imens. Although no correlation relationship is a ailable at thi stage between test data of the proposed procedure and field per· formance of in -service pavements, results obtained in lhl tudy suggest that the te st procedure could find useful application ill evaluating the relative surface wear resista nce a nd durability of various concrete pavement materials.
The provision of a durable traveled surface is an important consideration in highway pavement design. Deterioration of pavement surfaces, such as raveling, spalling, and surface wear, although usually causing negligible loss in structural capacity of pavements, is a major concern to highway engi- neers. Such surface defects can seriously affect riding quality and may considerably reduce the useful service life of the pavement affected. The task of providing a durable pavement surface is not an easy one. Highway pavements are exposed to environmental weathering throughout their service periods and are con- stantly subjected to repetitive wearing actions caused by mov- ing vehicular traffic. Once a surface distress is initiated, by either traffic or environmental factors, both will act together to contribute to the continuing and, in many cases, acceler- ating deterioration of the pavement surface. A complete study of pavement surface durability would therefore require anal- ysis of the impacts of both traffic loadings and environmental weathering. On concrete pavements, according to Burwell (J), the fol- lowing two types of wear mechanism are prominent. One is known as surface fatigue wear, which is caused by vertical dynamic wheel loads of vehicles. The other is abrasive wear, which causes damage through rubbing actions between tires and pavement surface and through scratching and gouging actions caused by particles caught between moving wheels and the road surface. This suggests that a meaningful surface wear evaluation of concrete pavement materials must include the effects of both wear mechanisms. The study of damaging effects of environmental weathering
Department of Civil Engineering, National University of Singapore, Kent Ridge, Singapore 0511.
has been a topic of research for many decades (2, 3). This problem is complicated because outdoor exposure tests are time-consuming, and it is difficult to control and standardize test conditions. As a result, laboratory-accelerated tests are often employed to predict the in-service durability of various types of materials. A number of testing devices known as
weatherometers are available in the commercial market (4).
They are typically equipped for test cycles requiring alternate light-dark and wet-dry exposures with temperature and rel- ative humidity controls. Regardless of the type of equipment used, a laboratory-accelerated test must be designed to sim- ulate as closely as possible the in-service environment in which the test material is intended to be used, so as to produce results that correlate well with material performance under outdoor service conditions. This paper describes a laboratory test procedure for eval- uating the relative durability of concrete pavement materials against surface wear. A laboratory-accelerated weathering model was developed to simulate the wetting and drying cycles of a wet tropical climate. The effects of wetting and drying weathering on material surface wear resistance were then determined by means of a rotating drum wear test conducted on specimens with and without weathering treatment.
BACKGROUND
Concrete pavement have been extensively used for bus bay , bus lane, and bus terminal con !ruction in Singapore since the late 1970s. The switch from bituminous to concrete pave- ment was made after more than 20 years of un atisfactory performance of bituminou pavements on roads dominated by bus traffic (5). The local experience with concrete pave- ments ·o far indicates that, although these pavement in gen- eral perform satisfactorily structurally, surface deterioration is a common distress problem that could drastically reduce the useful service life of those pavements affected. The surface deterioration is characterized by loss of cement mortar from the road surface along wheel paths, thereby expo ing coarse aggregates and giving the surface a rough and shabby appearance. The ride on such deteriorated surface i rough, noisy, and uncomfortable. Field observations indicated that the deterioration of concrete pavements involved initial wear of the hardened cement mortar at the surface. As the cement mortar was worn away, coarse aggregates were ulti- mately exposed. Depending on the bond between the cement mortar and coarse aggregates, the progressive wear could lead to raveling.
88
The present study is confined only to the first phase of the deterioration process, during which the wear resistance of cement mortar governs. This is important because a highway pavement that suffers deterioration of surface cement mortar is undesirable because it would not provide an acceptably smooth ride.
TEST METHODOLOGY
The aim of the test program in this study was to develop a test procedure that could be used by local pavement designers and maintenance engineers in their evaluation of material quality and durability for new construction or resurfacing of concrete pavements. The laboratory test was conducted on 100-mm (3 .937-in.) cubical specimens. The procedure devel- oped was composed of two parts. First, specimens were treated with laboratory-simulated wetting and drying cycles. Next, surface wear resistance tests were conducted on treated and untreated specimens to evaluate their relative wear resistance. Basis and details of the two parts are described below.
Weathering Simulation
The wet tropical climate of Singapore, as depicted in Figure 1, is characterized by an abundance of rainfall as well as bright sunshine and a relatively uniform temperature accompanied by high humidity throughout the year. The region generally has rainfall throughout the year but tends to be particularly wet during the monsoon season from November to January. The average annual rainfall is around 2,200 mm (86 in .), with
precipitation of mo.re than 50 mm a day occurring about nine
time a year (6). Tht! pavements in the region are also exposed to sunshine extensively each year. The annual average number of hours each day under bright sunshine is more than 5 hours. The abundance of both rainfall and sunlight results in a rel-
15
10
.--..,M..clAN OAILV MAXIMUM
M(AN
r M A M J l A s 0 N a MONTH
MUIMUM MONTHLY AVERAGE
(^21) F M A M I I A S 0 N 0 MONTH
TRANSPORTATION RESEARCH RECORD 1234
atively large number of wetting and drying cycles on the pave- ment surface. Precipitation falling on sun-heated pavements and intense sunlight drying up rain-soaked road surfaces are common scenes in Singapore. On a tropical day, the maximum air temperature is around 31°C and the minimum 24°C. The surface temperature vari- ation on most road pavements is much larger. Concrete pave- ment surface temperatures, measured by using thermocouple wires, usually range from around 25°C in the early morning to as high as 60°C on wheel paths on a hot afternoon. Thus, a wetting or drying process with a temperature change of 30 to 35°C is not unusual. A laboratory model designed to provide alternating wetting and drying was used to simulate weathering caused by the fre- quent alternating rainfall and sunshine exposure of in-service pavements. The accelerated weathering model is essentially a concrete tank with an enclosed space that measured 915 mm (36 in.) in height and 940 mm by 1,420 mm (37 in. by 56 in.) in plane cross-section. Wetting of test specimens was achieved by spraying tap water, at about 28°C, through eight shower heads that were fitted on the interior walls of the weathering tank. The number of shower heads was more than sufficient to keep test specimens wet throughout the wetting phase. Heating during the drying phase was provided by four 500-W ceramic heaters located at the underside of the ceiling of the weathering tank. A surface temperature of 60°C was achieved on 100-mm concrete cube specimens placed on the floor of the tank, after about 100 minutes of exposure to heating in the tank. This study adopted a 4-hour weathering treatment cycle that consisted of 2 hours of wetting followed by 2 hours of drying. The length of the wetting phase was selected to rep- resent approximately the mean duration of rainfall in Sin- gapore (6). The 2-hour drying period was chosen so as to attain the desired maximum temperature of about 60°C during the drying period. Due to high humidity in the weathering tank, the 2-hour drying treatment was, in terms of moisture
_ !:!. ML\N^ DAILY^ MAXIMUM ~ 0 90 ':li: ~ 60
~JOO
~zoo
~ 150 ~
MEAN
MEAN DAILY MINIMUM
fMAMJJASONO MONTH
FIGURE 1 Characteristics of Singapore climate based on data from 1967-1986.
90
Studies conducted by Fwa and Paramasivam (7) found the test to have satisfactory repeatability and good sensitivity against variations in the properties of concrete pavement materials tested. Their results indicated that, by conducting three tests for each material type tested, a confidence level of 95 percent can be achieved so that the difference, between sample mean and population mean surface wear, would not exceed an allowable error of 5 percent. In addition, they observed that the test produced worn speci- men surface with good resemblance to a field-deteriorated surface similar to the distress described in this paper. They also found that surface wear of cement mortar test specimens was similar to the initial phase of surface wear of concrete test specimens; they concluded that cement mortar specimens could be effectively used to study the crucial initial phase of surface deterioration, which is of major concern in highway pavement engineering.
Test Specimen Types
Based on the reasonings of earlier discussions, cement mortar specimens were used in the tests. The cement mortar tested corresponds to that of a concrete with the following mix pro- portions: cement/sand/coarse aggregate ratio = 1:1.5:2.5 by weight. Test specimens were prepared with the following five water/cement ratios: 0.45, 0.50, 0.55, 0.60, and 0.65. In all the mixes prepared, an accelerator was added at 1 L per 22 kg (0.27 gal per 50 lb) of cement to give the desired strength at the age of 7 days. The 7-day, 100-mm (3.937-in.) cube strength values of water-cured plain cement mortars, in order of increasing water/cement ratio, were 49.7 N/mm 2 (7, psi), 46.0 N/mm 2 (6,670 psi), 41.8 N/mm 2 (6,060 psi), 31. N/mm 2 (4,580 psi), and 30.1 N/mm 2 (4,365 psi).
Test Specimen Exposure Conditions
To evaluate the significance of wetting and drying weathering effects on test materials, three exposure conditions were spec- ified for each material type. Specimens from a batch of mix were divided into three groups, each subjected to a different exposure condition, followed by a rotating drum surface wear test. The three exposure conditions are described below:
- Initial condition. Cubical test specimens of 100 mm were cast, then demolded after 24 hours, and cured in a water tank for 5 days. The specimens were removed on the 6th day and left in room conditions at approximately 28°C for 1 day. Sur- face wear test and compressive cube strength test were con- ducted on the 7th day after casting. The relative humidity varied from about 70 percent in the afternoon to about 95 percent at night.
- Weathering treatment condition. Test specimens were cast and water-cured as in step 1 above. The specimens were air-dried for 1 day in room conditions at 28°C before they were subjected to alternating wetting and drying in the lab- oratory weathering tank for 150 cycles. Each weathering cycle consisted of 2 hours of wetting followed by 2 hours of drying. The specimens went through 6 weathering cycles per day, and it took 25 days to complete the 150 cycles. Specimens were again air-dried for 1 day in room conditions after the weath-
TRANSPORTATION RESEARCH RECORD 1234
ering treatment and before rotating drum surface wear tests were carried out on the 32nd day after casting.
- Control condition. Test specimens were cast and water- cured as in step 1 above. After removal from a water tank, the specimens were left in room conditions at 28°C for the same duration as the weathering treatment described in step 2 above. Rotating drum surface wear tests were conducted on the 32n<l <lay afle1 casting.
The control condition provides a basis of comparison on which the effect of weathering can be determined from test results of specimens subjected to weathering treatment. Because of the length of treatment duration, some gain in strength took place during this period. The initial condition was there- fore included to offer a more thorough examination of changes in surface wear resistance of test specimens.
ANALYSIS OF TEST RESULTS
The surface wear test data are recorded in Table 1. As an illustration, Figures 4 and 5 show examples of typical pro- gre sive wear plot obtained from the surface wear tests. Each curve in the figurei repre ·ents th average surface wear obtained from three rotating drum tests. Figure 6 summarizes the final results of the test program. Discussed below are findings derived from analyses of the test results.
Initial Versus Control Conditions
Specimen that underwent the control conditions were te ted
at an age 15 days older than those tested after the initial water-
curing. Table 2 records the strength comparison of spec im ens tcste.d under these two condition. All five mixes studied had about a 10 percent increase each in their compressive strength. The surface wear test results in Table 1 show corresponding increases in surface wear resistance of all materials from initial to control condition. The increase in wear resistance of each mix type was found to be statistically significant at a confi- dence level of 99 percent. As can be computed from the data in Table 1, each material type had approximately 20 percent less in surface wear. This positive correlation of surface wear resistance with strength is in agreement with the findings of other researchers (9, 10) who studied the resistance of con- crete against various forms of abrasion.
Control Conditions Versus Weathering Treatment
Since specimens that were exposed to control conditions and tho e exposed to weathering treatment were cast fr m the same batch of mix and tested at the same age for surface wear, a comparison of their wear test results offers an indi- cation of the significance of the weathering effect on surface wear resistance of test materials. An examination of the plots in Figures 4-6 reveals that, compared to specimens in control conditions, specimens exposed to laboratory cyclic wetting and drying treatment suffered increased surface wear. Individual statistical tests conducted separately for each of the five mixes concluded that, at 99 percent confidence level, the effect of weathering
Fwa (^) 91
TABLE 1 SUMMARY OF SURFACE WEAR TEST RESULTS
Treatment (^) Water/Cement Ratio to Specimen 0.45 0.50^ 0.55 0.60^ 0.
56.4 60.9 64.9 71.0 76. Initial ';4.1 58.2 64.0 67.8 78. Condition 52.1 64.3 64.4 69.9 78. (54.2%) (61.1) (64.4%) (69.6%) (77.7%)
44.3 48.5 53.2^ 5e.4 62. Control 47.6 47.9 51.3 54.0 61. Condition 46.4 44.3 51.6 54.4 57. (46.1%) (46.9%) (^) (52.0%) (55.6%) (60.5%)
54.6 58.8 63.9 70.9 73. Weathering 58.2 63.4 66.5 73.3 76. Condition 53.8 57.6 66.6 72.8 74. (55.5%) (59.9%) (65.7) (72.3%) (74.8%)
Notes: 1. Values in table represent percentage wear by mass after 2000 revolutions in rotating drum test.
- Each value in parentheses represents average result of 3 rotating drum tests.
!a 70 < ::E i;; 60 ..... ~...., 50 a: ..... .._ LO :!:
3:: ....,^ .....^20 ~
g; en 10
WATER I CEMENT RATIO = o. 55
o INITIAL CONDITION
- CONTROL CONDITION A CONDITION AFTER WEATHERING
200 400 600 BOO 1000 1200 lLOO 1600 1800 2000
NUMBER OF REVOLUTIONS IH ROTATIHG DRUM TEST
FIGURE 4 Results of surface wear test on cement mortar specimens with water-cement ratio of 0.55.
treatment was significant in reducing the surface wear resis- tance of test materials.
Initial Conditions Versus Weathering Treatment
It is of practical interest to compare the wear resistance of a concrete pav eme nt at the initial stage and at a later pha e f it service li fe aft er exposure to weathering. On one hand ,
there will be improvement in wear re i tance because of an
incrnase in concrete strength with age; on the otner hand , environmental weathering causes deterioration of concrete and reduces its wear resistance. For materials to be used for highway pavement constructioo , it is hi ghly de. irable th at its gain in wear re istance with age should at least be sufficient to offset any loss caused by weathering.
~J ....... D 3:: t:j 20 <....... a:: =en
WATER/CEMENT RATIO = 0 65
o INITIAL CONDITION
- CONTROL CONDITION A CONDITION AFTER WEATHERING
200 400 600 ' BOO 1000 1200 1400 1600 lBOO 2000
H!IMBER OF REVOLUTIONS IN ROTATING DRUM TEST
FIGURE 5 Results of surface wear test on cement mortar specimens with water-cement ratio of 0.65.
For the mixe tested in this st ud y, Figure 6 hows that the ga in in wear resistance of test pecirnens brought abo ur by the growth of mort ar strength were more or .less offset by the reduction of wear resistance cau. ed by weathering. Although these findi ngs can not be indiscriminately extended l cover other concrete or cement mortar mixes before similar tests are conducted, it appears logical to state th at designer must not simply assume in their designs any increa e in wear resis- tance with age for concrete pavements exposed to weathering.
Effects of Water/Cement Ratio
Past studies (7, 9, 10) have found that the abrasion resistance of concrete or cement mortar of a given mix can be improved by reducing its water/cement ratio. In the present study, for
Fwa
shown, however, that such a relationship is likely to be non- linear in nature, and it varies among various mixes of cement mortar. One must therefore refrain from using concrete or mortar compressive strength as a relative indicator of the wear resistance of various concrete materials. The test results in this study provide further confirmation of this view. In Figure 7 the surface wear of test specimens exposed to various conditions are plotted against their compressive strength. Looking at the test results of each of the three treat- ment conditions independently, one can conclude that among specimens there existed a trend of decreasing surface- wear with increasing compressive strength. The same however can- not be said when one combines the test data of specimens treated under different exposure conditions. At any com- pressive strength level, it can be seen that surface wear values could differ by as much as 15 percentage points. There was clearly no unique relationship between compressive strength and surface wear in this case. It is interesting to note that the compressive cube strength of specimens exposed to weathering and that of specimens under control conditions were similar. There was, however, a large difference in their respective surface wear resistance. It is possible that although the weathering treatment had caused deterioration of the surface layer of test specimens, as reflected by the increased surface wear in these specimens, it was not intensive enough to have adverse effects on the interior of the sp.ecimens and their overall compressive strength.
CONCLUSIONS
Based on the findings of the test program, the following conclusions may be drawn:
- As the strength of concrete increased with age, the Jim-
en en :::i;:^ <Z:
70
c... :z:
=<Z:
;: ...... 50 '-' ~ == en 45
0
0
lill!ill
0 INITIAL CONDITION A CONTROL CONDITION e CONDITION AFTER WEATHERING
0 0
0 0
0 • A
4 0 --~.~~~--~ ~-'-~~--'~ ,,__-'-~ 30 35 40 45 50 55 COMPRESSIVE STRENGTH OF 100 mm CUBE (H/mm 2)
FIGURE 7 Relationship between compr ess ive strength and surface we ar resista nc e of test specimens.
93
ited test results in this study suggest that there was a corre- sponding gain in wear resistance as cement mortar matured.
- The laboratory-accelerated wetting and drying weath- ering treatment caused statistically significant reductions in surface wear resistance of all cement mortar mixes tested. The test was found useful in providing an indication of the detrimental effects that adverse weather conditions have on the wear resistance of concrete pavements.
- For a highway concrete pavement that is exposed to weathering in its service life, one must not assume that its wear resistance would improve with time as the concrete ages. The present study showed that the reduction in wear resis- tance caused by weathering could be higher than the gain brought about by maturation of cement mortar.
- Test results showed that an improvement in wear resis- tance of the test specimens under weathering conditions could be effected by lowering the water/cement ratio of the cement mortar mix.
- The wear resistance of a concrete pavement is essentially a function of its surface properties. It cannot be uniquely related to its compressive strength measured from conven- tional cube tests.
- Although no correlation can be made at this stage with field performance of in-service pavements, the test results obtained in this study indicate that the test procedure adopted could be useful in evaluating the relative surface wear resis- tance and durability of various concrete pavement materials.
REFERENCES
- J. T. Burwell, Jr. Survey of Possible Wear Mechanisms. Wear, Vol. 1, 1957, pp. 119-141.
- J. B. Rau hut , R. L. Lytton, and M. r. Darter. Paveme nt D11111age Functions for Cos1 Allocmion. Report FHWA!RD-82/126, Vol.
- FHWA U.S. Department or Transportation , 1982.
- T. F. Fwn and K. C. Sinha. Effec1s of Load and Non-Load- Refated Fac1ors 011 Rigid Pavement Performance. Special Publi- cation SP-93. American Concrete Institute, De1roi1 , Mich., 1986, pp. 107 - 124.
- J. L. Scott. Laboratory Accelerated Tests Can Work for You. Proc. , 4th lntematio11al Conference on Durability of Buil,li11g M(lt erials and Components, Singapore, Vol. 2, 1987, pp. 833- 841.
- T. F. Fwa and C. Y. Tan. Concrete Road Construction in Sin- gapore. Proc., 2nd l11tematio11al Co11fere11ce on Concrete Tech- 110/ogy for Developing Countl'ies, Tripoli, Libya, 1986.
- Summary of Observation. Singapore Meteorological Service, in- gaporc, 1967- 1986.
- T. F. Fwa and P. Paramasivam. Surface Deterioration Resistance of Concrete Pavement Materials. Presented at First International Symposium on Sur(ace Characteristics, Pennsylvania State Uni- versity, University Park, Pa., 1988.
- Standard Method of Te t Cor Resistance 10 Abrasion of Large Size Coar e Aggregate by Use of the Los Angeles Machine. Designation C515-81, Am111al Books of AST 1 W Standards, Vo l. 4, 1986. 9. R. 0. Lane. Abrasi.on Resistance. ln Significance of Tests and Properties of Concrete and Concr(!Je-Making M(lterials, ASTM TP 1698, 1978 Chapter 2, pp. 332-368.
- ACI Committee 201. Guide to Durable Concrete. AC/ Journal, Vol. 74, No. 12, Dec. 1977, pp. 573-569.
This paper reports the results of research funded by the National Universily of Singapore. The author is, however, solely responsible for its findings and conclusions.
Publication of this paper sponsored by Committee on Performance of Concrete.
