Optimizing activated carbon size and ratio in bitumen modification

The aim of the study is to investigate the effects of activated carbon (CA) on the rheology of bitumen, and to optimize the activated carbon size and ratio in bitumen modification. The use of CA produced from waste hazelnut shells in bitumen modification is evaluated. Different sizes and ratios of CA are used. Test results show that a CA particle size smaller than 0.063 mm is the most effective, and that the CA modification increases the bitumen consistency and high temperature performance, while reducing the temperature susceptibility and weight change.


Introduction
Bitumen is an organic mixture of various chemical elements and compounds.It is widely used for the construction of roads due to its good adhesion to mineral aggregates and its viscoelastic properties [1][2][3].Unfortunately, bitumen assumes the form of liquid at high temperatures and becomes brittle at low temperatures, which can limit its application by causing high temperature rutting and low temperature cracking of pavement [1,4].Continuous increase in traffic volume and load on roadways, combined with adverse climatic effects, result in serious rutting and cracking occurrences.The improvement of bitumen and bituminous mixture properties by means of appropriate additives is an important issue in bitumen industry [5].The additives used for bitumen modification include polymers, carbonaceous materials, and other materials of various origin [6].Pavements with modified bitumen exhibit greater resistance to permanent deformation, fatigue damage, thermal cracking, stripping, and temperature susceptibility [1,4].C A, the additive used in this research, is a highly porous and amorphous carbon-based material.Due to its high degree of microporosity, it is used in the absorption of gases and soluble substances from water.C A is produced from carbonaceous source materials such as nut shells, coconut husk, wood, and coal [7][8][9][10].
The purpose of the present study is to investigate the effects of C A addition on the rheology of modified bitumen through conventional softening point and penetration tests, thin film oven test (TFOT), and Superpave methods (rotational viscometer (RV) and dynamic shear rheometer (DSR)).The C A used in this study has been obtained by pyrolysis of hazelnut shells.The effects of C A size and ratio on the rheology are also evaluated.

Materials and methods
The 50/70 penetration grade bitumen (original bitumen) obtained from the Kırıkkale terminal of the Turkish Petroleum Refinery was used throughout the study.Physical and rheological properties of the original bitumen are given in Table 1.

Production and characterization of C A
C A was produced from hazelnut shells obtained from the Giresun region of Turkey.The hazelnut shells were dried, crushed, and sieved to a particle size fraction of 1.0-2.0millimetres.Chemical activation was applied in order to obtain a porous surface texture.During the activation procedure, hazelnut shells were saturated with 30 % H 2 SO 4. The resulting chemical-loaded sample was placed in a furnace and heated (12.5°C min -1 ) to the final carbonization temperature of 450°C for 2 hours [7,8,10,11].After cooling to room temperature, the resulting products were taken out and leached with distilled water until the pH 7.0 was reached [7,12,13].The porous surface texture of C A,

Properties Standard Results
Penetration (0. where Pen 25 is the penetration at 25 °C.and SP is the softening point temperature.Optimizing activated carbon size and ratio in bitumen modification with macropores, micropores, and mesopores that disperse to the surface heterogeneously, is shown in a scanning electron microscope photograph (Figure 1).The specifications of C A are summarized in Table 2.

Preparation of modified bitumens
Modified bitumens were prepared at a mixing temperature of 150°C by means of a laboratory Marshall mixer rotating at 500 rpm.C A was added to hot bitumen at ratios varying between 1 and 25 % (w/w) and mixed for 45 minutes.C A size categories were chosen as size 1 (< 0.063 mm), size 2 (0.063-0.125 mm), and size 3 (0.125-0.25 mm) (Figure 2).To avoid agglomeration of C A in the bitumen, and to ensure the homogeneity of mixtures, C A was added in small amounts [4,6].

Softening point and penetration test results
Original and modified bitumens were subjected to a standard ring and ball softening point test [14,15] in order to determine consistency parameters.The softening point of the original bitumen amounted to 48°C.As shown in Figure 3, softening point values of modified bitumens are higher than those of the original bitumen.The increase in softening point, especially at 10, 15, and 20 % ratios, is an indicator of the stiffening effect of C A .The most effective C A sizes and ratios were determined based on the softening point test results for use in the latter tests.These values are given in Table 3.
The standard 100 gram, 25°C, 5 second penetration test was carried out on the bitumens given in Table 3.In addition, the temperature susceptibility of the bitumens was calculated in terms of penetration index (PI) using the results obtained during the penetration and softening point tests.
Temperature susceptibility is defined as the change of consistency parameter as a function of temperature [14].
The effect of C A modification on the properties of original bitumen can be seen in Figure 4, a decrease in penetration values and an increase in softening points occurs with an increase in C A content.  penetration and softening point values.The increase in softening point is favourable since bitumen with a higher softening point is more resistant to rutting.C A modification reduced the temperature susceptibility (as determined by PI) of the original bitumen, especially at the 20 % ratio.Lower values of PI indicate higher temperature susceptibility [14].Size 1 was determined as the most effective size for PI increase.The 20 % C A ratio with size 1 increased the PI value from -1.41371 to -0.96942.

TFOT results
The aging of bitumens was performed using the TFOT (thin film oven test) method (ASTM D 1754).Standardized conditions, that is, 163°C and 5 hours, were used [16].The aged bitumens were evaluated by penetration and softening point tests.TFOT test results are presented in Figure 5 and Table 5. C A modification to the original bitumen reduced the weight loss percentage.As can be seen in Figure 5, the aging characteristics of original bitumen improve with an increase in the C A content.

Figure 5. Weight loss percentages
This improvement can be explained by interactions between the functional groups of C A and volatile components of bitumen.As is evident from Figure 5, size 1 is the most effective size in terms of characteristics.The penetration and softening point tests conducted according to the TFOT method show that penetration values decrease, while softening point and PI values increase, in comparison with original bitumen.

DSR test results
The DSR test was performed on original and C A -modified bitumens using a Bohlin DSRII rheometer under the 120 Pa controlled stress at temperatures varying between 58-76°C with an increment of 6°C and the 10 rad/s frequency, using a 25 millimetre diameter plate with a 1 mm gap opening.In order to determine the high temperature properties of bitumen, complex shear modulus (G*), and phase angle (δ), the principal viscoelastic parameters were determined during the testing.G*/sinδ, which indicates the bitumen resistance to rutting under elevated temperatures, was calculated and compared with the Superpave asphalt bitumen test specifications as defined in AASHTO TP5.The specification shows that the G*/ sinδ parameter is less than 1000 Pa for unaged bitumens [17,18].DSR test results (G*/sinδ values) are given in Figure 6-7.Optimizing activated carbon size and ratio in bitumen modification Based on the Superpave PG grading system, the original bitumen exhibits 1000 Pa of G*/sinδ at 64°C.The high temperature performance of mixtures 1, 4, and 7 is 70°C, while it amounts to 76° for all other mixtures.As can be seen in Figures 6 and  7, the rutting parameter improves with an increase in the C A content.Correlations between the C A content and improvement of G*/sinδ values are given in Figure 8.The G*/sinδ values at 76°C with the 20 % additive ratio show that sizes 1 and 2 have similar effects on the rutting parameter, while the worst value is obtained with size 3.This can be explained by the separation of coarse C A and the resulting deterioration of homogeneity.

RV test results
The viscosity obtained by RV test is the measure of internal friction in bitumen.The viscosity at 135°C is usually used to measure workability according to Superpave specifications.The RV also measures rheological properties of bitumens to evaluate their ability to be pumped during delivery and plant operations.Therefore, elevated temperatures of 135° and 165°C were used in this study for testing purposes [17,19].The viscosity was determined by measuring the torque required to maintain constant rotation speed (20 rpm) of a cylindrical spindle submerged in bitumen maintained at a constant temperature [20].RV test results at 135 and 165°C are given in Figure 9.

Figure 9. RV results of bitumens
The viscosities give a clear indication of the stiffening effect of C A modification with a high additive content.Size 1 is the most effective size in terms of viscosity increase for all ratios.It was determined that the viscosity for size 3 C A mixtures with a 20 % additive ratio at 135°C is by 26 % smaller compared to the remaining two sizes.The observed decrease in viscosity can be explained by the non-homogenous mixture evidenced in DSR results.

Conclusions
The following conclusions were drawn based on the test results: -Bitumen stiffness and consistency clearly increase with an

Figure 3 .
Figure 3. Softening point test results for modified bitumens

Figure 4 .
Figure 4. Correlations between softening point and penetration values and C A content

Figure 2
Figure 2. C A sizes

Figure 7 . G*/sinδ for mixtures 4 , 5 , 6 , 7 , and 8 (Sizes 2 and 3 )Figure 8 .
Figure 7. G*/sinδ for mixtures 4, 5, 6, 7, and 8 (Sizes 2 and 3) addition of C A .-According to softening point test results, the optimum additive ratio varies between 10 and 20 %. -PI values show that the temperature susceptibility of original bitumen decreases with C A modification.In particular, size 1 and the 20 % additive ratio are the most effective conditions for preventing temperature susceptibility.-C A modification affects the ageing characteristics of original bitumen and decreases the weight loss percentage.-In the DSR test, C A modification yields higher G*/sinδ values.When comparing the three different sizes, size 3 performs worse than the other two sizes.The reduction of rutting parameter in case of size 3 is a result of separation of coarse C A and deterioration of homogeneity.-Viscosity increases and workability decreases when bitumen is modified with C A .