Introduction
Pervious concrete has been successfully used as a pavement to reduce and treat runoff through infiltration as well as recharging groundwater resources [
1,
2]. A pervious concrete mix has the same basic constituents as traditional concrete except for the gradation of the coarse aggregates, reduction or elimination of fine aggregates, and less water than the normal concrete mix. This provides about 15 to 30 percent porosity in the pervious concrete pavement layer which allows water to flow freely through the pavement structure and be stored in an aggregate bed below and/or infiltrated into the subgrade [
1,
3].
In addition, there are also wintertime benefits of many pervious concrete systems. Melting snow may percolate into the system reducing the potential for icing when the temperature drops. The surface texture of the pervious concrete layer may also play a role in preventing icing, particularly for pedestrian applications. There is a need to study its surface characteristics to promote its wintertime performance, particularly the microtexture. It is hypothesized that a rougher microtexture of pervious concrete might reduce the amount of deicing materials needed during icy conditions without degradation of its structural performance [
4].
However, there are only a few studies which focus on the friction of pervious pavements such as those conducted by Martinez & Poecker, and Huber [
5,
6]. In both studies, it was observed that open graded pavements improve friction in comparison with traditional pavements except when the voids of the porous pavement are filled, and when the surface of porous pavement is covered with compacted snow [
5,
6]. Kevern et al. [
7] studied how the surface texture of pervious concrete might reduce slip-related falls.
There are several testing methods to determine skid resistance on pavements, but these have mainly been developed for vehicles [
8,
9]. Some studies have also shown the importance of the microtexture [
10]. However, for placements such as sidewalks, or for laboratory specimens, it may be useful to be able to distinguish between various surface microtextures on these smaller areas.
One way to quantitatively evaluate a surface’s microtexture is through friction testing. The values of the coefficients of friction from a resting position and during relative surface motion are called the “static coefficient of friction” (SCOF) and “dynamic coefficient of friction,” respectively. Although the dynamic coefficient of friction during walking varies in a complex and non-uniform way, the static coefficient of friction may provide a close approximation of the slip resistance of a sidewalk [
11].
The adapted methodology for the classification of the surface texture is mainly based on qualitative assessment by means of a visual inspection. However, this type of evaluation may be insufficient, as well as inaccurate, due to its dependence on subjective judgement [
12]. For this reason, developing a practical method to evaluate the surface texture of pervious concrete with different treatments may be helpful for comparing those treatments.
In this study, the possibility of using a spring balance to evaluate the surface microtexture with respect to the SCOF of five different surface treatments was investigated. In addition, specimens with these treatments were placed outdoors throughout a winter season and evaluated subjectively in a qualitative method by various members of the research groups under different weather conditions. This was performed next to a placement that had received some icing under severe conditions [
13]. The results of the quantitative spring balance test are compared with the qualitative ones as a further evaluation of the efficacy of using the static coefficient of friction as an initial testing methodology.
Material and methods
Description of applied surface treatments
Based on previous experiments in the Washington State University laboratory and discussions with various concrete pavement professionals, five surface treatments were chosen to be tested. These are listed and described in Table 1 and photographs can be seen in Fig. 1. Two specimens of pervious concrete receiving each type of treatment were prepared. They were all made in molds to have surface dimensions of 350 mm (13.7 in.) by 200 mm (7.8 in.) and an approximate depth of 100 mm (3.9 in.). All of the specimens were surface compacted using a flat board and a mallet to effect a porosity around 20%-27% based on the compacted volume and the mass of concrete placed in the mold.
Based on ASTM C1688 and a modified ASTM C1701 [
14,
15], the specimen porosities and initial infiltration rates were measured and are also listed in Table 1. The modification to ASTM C1701 was that the inner diameter of the ring was 100 mm instead of 300 mm and one liter of water was used for the test instead of 20 L.
Then, the specimens were placed outdoors in a landscape bed next to the previously mentioned pervious concrete sidewalk [
13]. During this period the specimens were evaluated qualitatively as described later (Qualitative Evaluation), and also allowed to weather under normal wear and exposure conditions for approximately a half year. Afterwards, the specimens were transported back to the laboratory and evaluated with the proposed quantitative method (Spring Balance Method) for friction comparisons as also further detailed in the next section.
Surface evaluations
Since weather conditions may be dry or wet, it may be important to study the SCOF or “slipperiness” under both dry and wet conditions. To analyze the effectiveness of the surface treatment, the two following methods were developed. They are the quantitative spring balance method and the qualitative evaluation which was subjectively performed by members of the research team.
Spring balance method
Among different friction measurement methods, the spring balance approach was chosen due to its simplicity and being a practical approach for horizontal (not sloped) placements. It had also been previously tested in the Washington State University laboratory, along with other methods on different concrete specimens, and appeared to be easy to perform and to train another operator on, with fairly consistent results. The idea is that if one puts a solid object on a rough surface and starts to pull on it, there is a point at which the object will start to slide. The required tools as shown in Fig. 2 are:
1) spring balance,
2) set of weights, and
3) a smooth, flat 127 mm × 76 mm × 13 mm (5 × 3 × 0.5 in.) wood board with a hook with the total mass of 73 g.
In this friction measurement method, the board is placed on a rough surface which is attached to a spring. Then, a gradually increasing external force is applied on the board to reach the point in which the board starts sliding. The reading on the spring balance scale when the load begins to slide is a measure of the static friction force. In this experiment, the mass of the wood block and the weights placed on it totaled to three different values which were 0.9, 0.8, and 0.7 kg (1.98, 1.76, and 1.55 lb). Then, the coefficient of friction is calculated as:
where
m is the coefficient of friction,
F is the external force applied on the board as measured by the spring balance, m is the mass of the weights and board, and g is acceleration due to gravity [
16].
By using different combinations of weights on the board and their corresponding external forces, several SCOFs are evaluated. Similar to other experimental measurements, it is suggested to conduct the test in at least two different locations and three times for each location to increase the accuracy of the results. In this experimental set-up, the tests at each location were done with the three different normal forces, one test for each, and then the SCOFs were averaged at each location. In addition, conducting the test by different operators can be helpful for this purpose.
It is worth mentioning that according to Amonton’s second law (1699), the force due to friction is generally independent of the contact area between the two surfaces. This means that even if one has two heavy objects of the same mass, where one is half as long and twice as high as the other, these objects still experience the same frictional force when dragged over the ground. That is why if the area of contact doubles, one may think that there should be twice as much friction. But as the length of an object doubles, the force on each square centimeter is halved, because the weight is distributed over a larger area [
16].
For this experiment, 360 friction tests were conducted by three different operators on pairs of specimens with five different textures under both dry and wet surface conditions. To conduct the friction test on a wet surface, the specimens were immersed in water for five minutes. However, due to the fast drainage characteristics of pervious concrete, water was also poured on the surface just before performing the friction test to re-wet the specimen.
Qualitative evaluation
As previously mentioned, to conduct qualitative tests, the laboratory specimens were placed in a landscape bed outside near a sidewalk to observe how they perform under ambient weather conditions. In this method, a person (operator) slides his or her foot on the specimen (with no label describing the treatment type to avoid any possible prejudice) in order to gain a general idea about the performance of different treatments in terms of skid resistance. The person then records a score of 5, 4, 3, 2, or 1 as corresponding to “considerable”, “high”, “medium”, “low”, and “negligible” skid resistance, respectively, on a checklist designed for this purpose. Then, the average score of 74 daily sets of observations (1/25/2015 to 4/7/2015) were calculated. The person performing the test also noted the weather conditions so that the observations could be separated into either wet or dry surface conditions. Three of these daily observations were performed by three other members of the research team, while the remaining observations over the winter were made by the principal author. Please note that it may have been beneficial to perform these observations under icy or snow covered conditions, but this winter period was very mild and only a few observations were noted with ice on the specimens. In those cases, these observations were included in the wet category. Since these data are not absolute values, but rather only comparative within each set of observations, no detailed requirements were given for weight of the operator, shoe type, etc. other than they were the same for each set of observations.
Results and discussion
Surface evaluation results
Figure 3 is the summary of average values of measurements on two locations of each specimen (three trials for each location) reported by the three different operators for the spring balance method. The standard deviations are also provided in Fig. 3 by the whiskers on the bar tops. Figure 4 provides similar information on these specimens for the qualitative evaluation as averaged for all observations on each specimen. The standard deviation is included on Fig. 4.
To visually compare the two methods, the values in Figs. 3 and 4 were first averaged for both specimens in each treatment type, and then normalized. Normalization was done by dividing each surface treatment by the highest averaged value for either the wet or dry condition. Figures 5 and 6 compare the normalized values of the surface evaluations for both the spring balance and the outside observations under dry and wet conditions, respectively. The actual numbers for each average are also depicted in Figs. 5 and 6.
The results from both the quantitative and the qualitative approaches show that the broom finishing has the highest ranking in terms of friction or lack of “slipperiness”. In addition, both sand treatments appeared to increase the surface microtexture under both dry and wet conditions. However, the pre-compaction method appears to not be as effective as post-compaction. (As previously described in the methods section, the molds were filled and then surface compacted. The difference in these two sand methods was when the sand was applied to the surface, either before or after compaction.) Finally, the coating treatment seems to affect the surface microtexture negatively, providing a lower SCOF, especially under the wet condition.
Reliability analysis
Measuring reliability is an important step in the development of a testing method to demonstrate that the method can be applied similarly by different operators. Consistency is a term which describes how much these operators agree on ranking of the items, but not necessarily the actual value [
17]. For this study, Kendall's coefficient of concordance (KCC) was used to evaluate the extent of consistency of multiple raters (more than two operators) for reliability evaluations of the spring balance method [
18].
In addition, it is important to evaluate the extent of correlation between the results of the spring balance method and the qualitative results, as the qualitative results are based on the actual performance in the field and the idea is to see if the quantitative test may be used instead. To measure this, Spearman’s Rho, which is based on paired comparisons (the two methods) was used to analyze the level of consistency in ranking between the two methods. This statistic is calculated based on taking the measured continuous data and ranking each test. For example, to use Spearman’s Rho to compare the spring and subjective measurements, the value of every spring measurement was replaced with the rank, and similarly for the subjective values. If the two methods are consistent, the rankings should be similar. Again, it is the relative ranking that is being evaluated, not necessarily the absolute value.
Tables 2 and 3 present the results of inter-rater reliability and parallel-form reliability, for comparing the consistency between the two methods, and the operators in the quantitative results, respectively. Based on the rankings in Table 2 the Kendall’s coefficient of concordance was calculated as 0.739.In this reliability analysis, coefficients of 0.1, 0.3, and 0.5 are generally interpreted as low, medium, and high correlations, respectively [
19]. Based on the rankings in Table 3, Spearman’s Rho was calculated as 0.74. The interpretation scale is similar, implying that there is also high consistency between the rankings in the two methods [
19].
Conclusions
As noted in the previous section, the results of the reliability analyses show high levels of consistency among the operators for the quantitative (spring balance) approach, and similarly a high level of consistency between the qualitative results and the spring balance method. This implies that the spring balance method may be an adequate substitute for other qualitative methods of evaluation of the various surface treatments, and that the spring balance method holds promise for adequate consistency with respect to comparisons between different operators. These high levels of consistency are supported by similar trends in Figs. 5 and 6. The quantitative and qualitative methods reported similar ranking trends in which the following surface treatments were ranked from highest to lowest under both dry and wet conditions: broom, post-sand, pre-sand, control, and coating.
Considering these findings, as well as the fact that brooming may also be a cost-effective treatment, one might conclude that the broom finish is the most sustainable solution among different treatment methods in this study. However, further study is needed in the field to see if the post-placement curing period with plastic sheeting over the slabs might impact the brooming. In addition, the impacts of the sand treatments might change as the sand is worn off of the surfaces, as may be the ridges from the broom. Again, only further long-term study in the field will provide these additional answers. The coating method with the lowest ranking appeared to be the least cost-effective and sustainable solution, and may not be better than the control. The broom and the post-sand treatments were chosen along with the control for a future sidewalk installation experiment. It was thought that both the broom finish and the post-compaction sand treatment could be easily implemented after traditional compaction performed by certified installers in a short time with little interruption to the construction process.
There may also be other enhanced testing protocols to explore for evaluating surface texture [
20]. In addition, evaluations should be performed in the field under icing conditions such as with hoarfrost. Performance of the quantitative spring balance test on placements in the field and under other extreme winter conditions may provide even more information on its efficacy.
The work performed for this study does appear to 1) constitute a preliminary step toward an enhanced surface evaluation for ranking in terms of static friction and improvement of understanding of where these surface treatment methods may or may not be used effectively and 2) provide the initial development of a test method of different surface treatments in terms of microtexture. The methods developed do not provide definite answers, but rather a way to initially compare various surface treatment options for pervious concrete placements, as each placement might be made with a different mix design and be subject to varying environmental and operational conditions. If this quantitative method is further evaluated into a test method, then more detailed protocols on board type, smoothness, etc. should be developed to provide more consistency between different laboratories and applications. At this time the spring board method does not provide an absolute scale, and can only be used for comparisons under identical conditions.
Higher Education Press and Springer-Verlag Berlin Heidelberg
PDF
(1156KB)
