Investigation of Using Waste Welded Tuff Material as Mineral Filler in Asphalt Concrete

In this paper, the welded tuff waste- known as koyke in Isparta region - was used in the hot mix asphalt (HMA) as mineral filler for reduction of the moisture susceptibility of HMA. Optimum binder content was assessed with Marshall Design Method. First of all, welded tuff was substituted as filler with limestone filler in proportion of 50% and 100%. After that Marshall Stability test was performed on specimens. The results showed that the 50% substitution was more effective than the 100% substitution. Therefore, welded tuff was substituted with limestone filler in proportion of 25%, 50%, 65% and 75%. Next, Indirect Tensile Strength test was practiced on the fabricated specimens and the results were assessed. According to the Indirect Tensile Strength results, welded tuff with 65% was given higher strength than the limestone filler. As a result, it has come up that welded tuff can be used as mineral filler in the hot mix asphalt.


Introduction
Water existence causes moisture damage by decreasing stability in asphalt mixtures. Presence of water weakens the strength of the pavement. Moreover, moisture damage increases the probability of raveling which decreases the skid resistance on the surface of the road [1]. The Modified Lottman Test [2] was used to specify the moisture damage. And the specification criterion required for the Tensile Strength Ratio (TSR) of a mixture at optimum binder content (OBC) is minimum 80%.
Because of the high costs, the requirement to use waste materials are increasing. Also new pavement constructions need more and more virgin aggregates. Because of the depletion of natural resources and increased demand, waste materials become more popular on pavement design. At the same time by modifying the bitumen, pavement performance can be improved [3][4][5][6][7]. To improve the performance virgin materials can be used. Using of the virgin materials are damaging environment and very expensive. To improve the pavement performance more sustainable, use of waste materials on pavement become more popular in the last decades. There are many waste material types tried to improve the pavement performance such as; plastic bottle [8][9][10], crumb rubber [11][12], glass [13][14].
The use of waste materials provides sustainable, ecofriendly and economical highways. It is important to obtain the waste materials where the highway will be constructed because of the transportation costs. To reduce the transportation costs for a possible highway construction, a waste material called welded tuff in Isparta region is used in this paper. Decreases in eruption activity or overloading by continued eruption can cause the eruptive column to collapse either continuously or sporadically. The hot pyroclastic material falls from the column and flows outward from the vent following topography, and may flow more than tens or hundreds of square kilometers. Such pyroclastic flows can retain enough heat to fuse or weld the particles together after movement stops. The names applied to these rocks historically have included tuffs, welded tuffs, ash flows, and ignimbrites [15].
Welded tuff has a porous structure and used for industrial purpose. The porous structure of the welded tuff lowers the freezing point temperature [16]. Welded tuff is used as an insulation material for buildings. And, by forging the welded tuff, a large amount of waste material occurs. Occurred waste material is very convenient to replace virgin aggregates. By using this waste material, given damage to the environment is minimized.
Tuff is of interest for use as an isolation material for high heat producing wastes because it provides highly sorptive minerals and suitable thermomechanical properties. Also, tuff is widespread in areas that offer long and deep groundwater flow paths [17]. This paper focused on the usability of welded tuff as waste material in Hot Mix Asphalt (HMA). Usability evaluation is performed by Modified Lottman Test. Moisture susceptibility of welded tuff mixed specimens were obtained via Tensile Strength Ratio (TSR).

Materials
In this section, materials (aggregate, binder and welded tuff) which were used in this paper were described in detail.

Aggregate
Limestone aggregates (LS) used in the study were obtained from Isparta region. The nominal aggregate size for wearing course was selected as 12.5 mm. The properties of mineral aggregate properties were shown in Table 1. Grading curve of HMA was selected in accordance with General Directorate of Turkish Highways [18] (Figure 1).

Binder
Standard tests were applied for the purpose of identifying physical characteristics of binder. Test results were given in Table 2. prepared with ±0.5% and ±1.0% to pre-optimum binder content. According to test results optimum binder content was obtained 5%. All specimens were produced according to this optimum binder content.

Welded tuff
The used welded tuff was obtained from Isparta region. The properties of the welded tuff were given in Table 3. Determining welded tuff elemental properties, Energy-dispersive X-ray spectroscopy (EDS) was carried on welded tuff. Energy dispersive-ray spectra have been taken in a scanning electron microscope (SEM) Cambridge S 200 equipped with a Si (Li) detector. SEM/EDX detects the characteristic X-rays that are emitted from the first few micrometers beneath the specimen's surface after inner shell ionization by the primary electrons. Welded tuff properties obtained by EDS analyses were shown in Table 4.

Method
The strengths of the specimens were obtained by Marshall Stability and Modified Lottman [2] test procedure. The results of the Modified Lottman test were used to obtain the moisture susceptibility of the specimens.

Marshall stability test
Marshall Design procedure was handled to design asphalt concrete mixtures using regional materials [19].
Where; is saturation level (%), . is saturated surface dry weight (g), is weight in air (g), is air voids (%), is volume of the specimen, is theoretical gravity, is bulk specific gravity and is the weight of the specimen in water.
When the specimens reached the saturation level, they are put into freeze cabin at -18⁰C for 16 hours. After 16 hours they are put into water bath at 60⁰C for 24 hours. Finally, they are put into 25⁰C water bath for 2 hours. After 2 hours they are loaded at a load speed as 50,8mm/min. The failure load values are recorded as IDTwet (conditioned) strengths. The IDT strength is calculated by Equation 5. The ratio of the wet specimen strength to dry specimen strength is Tensile Strength Ratio (TSR) (Equation 6). TSR is used to determine the moisture susceptibility. A minimum TSR value of 80% is recommended by Turkey General Directorate of Highways [18].
Where; is the average strength value of the conditioned specimens and is the average strength value of the unconditioned specimens.

Results
First of all, welded tuff (WT) was substituted as filler with limestone (LS) mineral filler in proportions of 50% and 100% by weight. Following, Marshall Stability test was performed on specimens. It was observed that stability value increases with 50% substitution of welded tuff. Diversion of dissimilar Marshall Stability value versus filler proportion was given Figure 2.  It was observed that practical specific gravity increases for 50% WT+50% LS substitution up to 100% welded tuff; then decreases. Change of unit weight with dissimilar welded tuff proportion versus different filler was given in Figure 4.
It was observed that for 50% welded tuff substitution air void decreases. Variation of air void with different welded tuff proportion was given in Figure 5.
It was observed that for 50% welded tuff substitution voids in mineral aggregate decrease. Variation of VMA with different binder content was shown in Figure 6.
It was observed that for 50% welded tuff substitution voids filled with bitumen increase. Variation of VFB with different welded tuff proportion was shown in Figure 7.  The results showed that the 50% substitution was more effective than the 100% substitution. Therefore, in this paper welded tuff was substituted with limestone filler in proportion of 25%, 50%, 65% and 75% by weight. was performed on the fabricated specimens and the results were assessed.
In Figure 8, the conditioned and unconditioned strength values were shown. According to the Figure  8, max IDTdry and IDTwet values were obtained with 65% welded tuff added specimens. The values increase till 65% welded tuff added specimens and then they were tended to decrease. Figure 9 shows the TSR values of the specimens. As seen in the Figure 9, all specimens were provided the specification limit value and max TSR value was obtained at 65% welded tuff added specimens.
In Figure 10, the comparison between conditioned and unconditioned samples IDT strengths were shown. As seen in the Figure 10, the 65% and 75% welded tuff added specimens were close to the 45⁰~line of equality. The meaning of being close to the line is that the conditioned and unconditioned IDT strength values were close to each other. So 65% and 75% of welded tuff added specimens have less moisture susceptibility.

Discussion and Conclusion
First of all, welded tuff was substituted as filler with limestone filler in proportion of 50% and 100% by weight. After that Marshall Stability test was performed on specimens. The results showed that the 50% substitution was more effective than the 100% substitution. Therefore, in this study welded tuff was substituted with limestone filler in proportion of 25%, 50%, 65% and 75%.
According to the results, max IDTdry and IDTwet values were obtained with 65% welded tuff added specimens. Moisture susceptibility of welded tuff mixed specimens were obtained by Tensile Strength Ratio (TSR). As a result, 65% and 75% of welded tuff added specimens have less moisture susceptibility. TSR results ensure the IDT Strength test results.
So that, welded tuff can be used in HMA as waste material in appropriate proportions. Because of these properties, welded tuff provides sustainability to the pavement material industry.