GNP 27

1 Effects of Warm-Mix Additives, Anti Stripping Agent, and Graphene
ed
2 Nanoplatelet on the Cracking Resistance, Moisture Susceptibility, and Cost
3 Effectiveness of Stone Mastic Asphalt
iew
4 Md. Tanvir A. Sarkar1, Mostafa A. Elseifi2, Zahid Hossain3
5 ABSTRACT
ev
6 Moisture resistance and cracking are two of the most common asphalt pavement distresses in the
7 United States. The use of Stone Matrix Asphalt (SMA) mixture has gained popularity due to its
r
8 high durability, resistance to permanent deformation and cracking, and reduced noise pollution.
10
er
Warm-mix additives (WMA) and nanomaterials have also gained significant interest as
promising asphalt binder additives. This study investigated the effects of three warm-mix
pe
11 additives, an anti-stripping agent, and one nanomaterial (Graphene Nanoplatelet, GNP) on the
12 cracking and moisture resistance performance of SMA. The Fourier Transform Infrared
13 Spectroscopy (FTIR) and sessile drop tests were used to evaluate the functional groups and
ot
14 moisture susceptibility of the modified asphalt binder blends, respectively. The Indirect Tensile
15 Asphalt Cracking Test (IDEAL-CT) and the Modified Lottman (ML) tests were used to evaluate
tn
16 the SMA mixtures' resistance to cracking and stripping damage. A cost-effectiveness analysis as
17 well as a two-dimensional performance interaction diagram were employed based on the

rin
18 laboratory performance of the mixtures. Results indicate that the GNP modification of the
19 asphalt binder had the highest wettability potential, adhesion and debonding properties, and
ep
1 Graduate Research Assistant, Department of Civil and Environmental Engineering, Louisiana State University.
Pr
2 Occidental Chemical Corporation Distinguished Professor, Department of Civil and Environmental Engineering,
Louisiana State University
3 Professor, Department of Civil and Environmental Engineering, Arkansas State University
This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4796985
20 moisture resistance potential compared to other additives. In terms of mixture cracking
ed
21 performance, GNP and a chemical WMA significantly improved the SMA cracking
22 performance. However, GNP showed inadequate stripping damage resistance in the laboratory.
23 Overall, all the SMA mixtures with warm-mix additives showed adequate cracking and stripping
iew
24 damage resistance performance and are expected to be cost-effective.
25
26 Keywords: Graphene Nanoplatelet, Warm-mix additives, Moisture damage, Cracking, SMA.
r ev
er
pe
ot
tn
rin
ep
Pr
27 INTRODUCTION
ed
28 Asphalt concrete (AC) endures repeated traffic loading under complex environmental conditions
29 throughout its service life, leading to the manifestation of rutting, cracks, and moisture-induced
iew
30 damage. Moisture-induced damage, or AC stripping, in a pavement structure is a matter of great
31 concern as it is also responsible for accelerating other significant distresses such as fatigue
32 cracking, rutting, and poor durability. Asphalt concrete stripping, which is a chronic problem in
ev
33 flexible pavements in Louisiana and Arkansas, refers to the loss of bond between the aggregate
34 and asphalt binder, usually caused by the accumulation of moisture underneath the pavement
35 surface (Brown et al. 2009; Hossain et al. 2015; Kim et al. 2020). This loss of bond strength
r
36 occurs either by adhesive, cohesive, or a combination of both failure modes. Cohesive failure, a
37
38
er
less common failure phenomenon, occurs in the asphalt binder film, resulting in a decrease in
asphalt mixture stiffness, whereas, adhesive failure, a more common failure phenomenon, occurs
pe
39 at the asphalt binder-aggregate interface due to the presence of water (Birgisson et al. 2003;
40 Santucci 2010).
ot
41 In Louisiana, shallow groundwater table (GWT) and heavy rainfall conditions (the
42 second wettest state in the United States with an average annual rainfall of 60 in.) throughout the
tn
43 year make the pavement highly vulnerable to water entrapment and moisture damage. Due to
44 rising groundwater levels, linked air voids, and shoulders of the pavement, moisture can infiltrate
rin
45 and cause significant damage through cyclic hydraulic pressure resulting from repeated traffic
46 loads and freeze-thaw mechanisms (Kennedy et al. 1983). The effect of AC stripping is
ep
47 manifested on the roadway through poor durability of the mixes and a shorter service life. These
48 pavement distresses greatly affect user safety, pavement performance, and the cost-effectiveness
Pr
49 of maintenance and rehabilitation activities. Moisture-induced damage, typically originating at
50 the asphalt concrete layer’s base and progressing upwards, often remains undetected until
ed
51 significant under-the-surface damage has occurred (Kandhal 1992; Abohamer et al. 2021).
52 Moisture and cracking resistances of AC are controlled by a range of design and
iew
53 production factors, including the properties of the binder and aggregate, mixture characteristics,
54 and the type of additives used. The binder acts as the medium that holds the aggregate in the
55 mixture, contributing to its performance against stripping and cracking. Asphalt binder is
ev
56 generally modified to increase its performance against major pavement distresses. Ground tire
57 rubber-modified asphalt binder was found to show higher total surface free energy (SFE) and an
58 increased compatibility ratio (CR) with aggregate, indicating higher fatigue cracking and
r
59 stripping resistances. The compatibility ratio was also found to increase with an increase in the
60
61
er
dosage rate of ground tire rubber (Hossain et al. 2015). Styrene–butadiene–styrene (SBS)
modified asphalt binders were also found to show enhanced moisture resistance than unmodified
pe
62 asphalt binders in terms of work of debonding (Guo et al. 2022). Several studies also found that
63 the use of WMA increases the cracking and moisture susceptibility resistances of the mixture.
ot
64 Xiao et al. reported that warm-mix asphalt treated with hydrated lime had a higher tensile
65 strength ratio than with liquid anti-stripping agents (Xiao et al. 2014).
tn
66 Although numerous studies were conducted on enhancing asphalt mixtures performance
67 against cracking and moisture damage, these failure mechanisms still remain one of the major
rin
68 asphalt pavement distresses across the US, and that is causing a significant loss of funds to
69 transportation agencies, DOTs, and other entities involved in road maintenance and rehabilitation
ep
70 activities. Therefore, there is a need to evaluate the current state of practice and specifications in
71 asphalt pavements, to identify the reasons for the frequent appearance of this failure mechanism,
Pr
72 and develop cost-effective asphalt mixtures that can provide adequate performance in terms of
ed
73 moisture resistance and cracking.
74 OBJECTIVES AND SCOPE
iew
75 The objective of this study was to assess the effects of warm-mix additives, anti-stripping agents,
76 and graphene nanoplatelet on the laboratory cracking and moisture resistance performance, as
77 well as the cost-effectiveness of SMA. To achieve these objectives, a series of laboratory tests
ev
78 focused on asphalt mixture and asphalt binder were conducted. Laboratory test methods
79 including Fourier Transform Infrared Spectroscopy (FTIR), Sessile Drop, Indirect Tensile
r
80 Asphalt Cracking Test (IDEAL-CT), and Modified Lottman (ML) were utilized to evaluate the
81 effects of the modification on the binder molecular composition, surface free energy (SFE),
82
er
cracking resistance, and moisture resistance of asphalt mixtures, respectively. The results were
pe
83 processed using statistical analysis to quantify the effects of various additives on the overall
84 performance and durability of the asphalt mixes. In addition, based on the IDEAL-CT and
85 Modified Lottman test results, a cost-effectiveness analysis was performed to assess the
ot
86 economic viability of using these additives in SMA. A two-dimensional performance interaction
87 diagram (2-D PID) was also developed to identify the best-performing mixture in terms of
tn
88 laboratory performance and cost-effectiveness.
89 BACKGROUND
rin
90 Stone matrix asphalt is a gap-graded mix that has gained popularity in the US and Europe due to
91 its remarkable durability, substantial resistance to rutting and cracking, and ability to lower
ep
92 traffic noise. SMA mixtures typically contain modified asphalt binder, a large amount of coarse
93 aggregate, and sufficient fine aggregate to fill the voids. Although similar to an Open-Graded
Pr
94 Friction Course (OGFC) course in appearance, it has a low air void (around 4%), similar to Hot-
ed
95 Mix Asphalt (HMA). However, the optimum binder content in an SMA mixture is higher than in
96 a typical HMA mixture. On the other hand, warm-mix additives have gained wide acceptance
97 owing to increased concerns related to non-renewable energy usage, global warming, and the
iew
98 movement towards sustainable transportation infrastructure. The use of WMA reduces mixing
99 and compaction temperatures, which in turn contribute to lower air pollutant emissions and
100 reduced energy consumption during mixture production and construction (Zhao et al. 2013).
ev
101 Khedmati et al. evaluated the performance of SMA mixtures with different warm-mix
102 additives (i.e., Zycosoil and Sasobit) and their interaction in the mixture. Results indicated that
r
103 Zycosoil could enhance the moisture resistance of the SMA mixture when siliceous aggregates
104
105
er
were used. Additionally, the content of Sasobit was identified as an important factor affecting the
TSR parameter, with a 1.5% content showing the highest TSR values among the evaluated
pe
106 mixtures (Khedmati et al. 2017).
107 The use of nanotechnology has also gained interest in the research community to enhance
ot
108 the mechanical properties of asphalt concrete. The low-temperature cracking performance,
109 stiffness, and the resistance of the asphalt binder to rutting, moisture, and aging were positively
tn
110 affected when nanomaterials were introduced into the asphalt binder (Yao et al. 2011; Yusoff et
111 al. 2014). The addition of nanosilica and carbon naotubes (CNTs) was found to improve the
rin
112 adhesive strength, cracking resistance, and stripping resistance of asphalt binder (Yang 2013;
113 Yusoff et al. 2014; Yan et al. 2015). Fakhri and Shahryari examined the impact of nanozinc
ep
114 oxide (ZnO) and nano-reduced graphene oxide (RGO) on the moisture resistance performance of
115 SMA (Fakhri 2021). Results suggested that both nano ZnO and nano RGO could enhance the
Pr
116 moisture resistance performance of SMA, with nano ZnO demonstrating a more significant
117 improvement than nano RGO. A recent investigation assessed the use of graphene nanoplatelets
ed
118 (GNPs) as a modifier for asphalt binders (Le et al. 2016). Findings indicated that GNP was able
119 to improve the compaction properties and low-temperature flexural strength of the asphalt binder
120 and could be a potential alternative to WMA. Due to its nanoscopic size and higher SFE, it could
iew
121 potentially improve the fracture energy, indirect tensile strength, and moisture resistance
122 performance of SMA.
ev
123 A review of the literature revealed that previous studies did not evaluate the effects of
124 GNP on the cracking resistance, moisture susceptibility, and cost-effectiveness of SMA. While a
125 few studies were conducted to evaluate the effects of WMA on the moisture susceptibility of
r
126 SMA, a thorough investigation is still needed with more focus on cracking resistance, SFE, and
127
128
er
the cost effectiveness of these additives with SMA. Moreover, no study was found on the
comparative analysis of the laboratory performance and the cost effectiveness of warm-mix
pe
129 additives, anti-stripping agents, and GNP with SMA.
130 EXPERIMENTAL PROGRAM

ot
131 Material selection

tn
132 The asphalt mixture in this study was prepared using coarse limestone, fine limestone, and fine
133 river sand (PineBluff 78, Vulcan 78s, Vulcan 89s, and DonnaFill), all of which met Louisiana's
134 specifications, with a nominal maximum aggregate size (NMAS) of 12.5 mm. A styrene-
rin
135 butadiene-styrene (SBS) polymer-modified PG 76-22 asphalt binder, conforming to Louisiana's
136 Standard Specifications for Roads and Bridges, was utilized to prepare the mixes. Table 1(a)
ep
137 presents the physical characteristics of the aggregate blend used in the study. A number of
138 WMA, including two chemical warm-mix additives (Che1, Che2) and one organic warm-mix
Pr
139 additive (Org1) approved by the Louisiana Department of Transportation and Development
ed
140 (LaDOTD) were selected for evaluation in this study. Additionally, an anti-stripping agent
141 (Asa1) and a nanomaterial (GNP) were incorporated to assess their potential for enhancing the
142 performance of the SMA mixture. The properties of GNP are provided in Table 1(b).
iew
143
144 Table 1 (a) Physical Properties of combined aggregate and (b) properties of GNP
ev
145 (a)
Physical Properties Method Results
r
Apparent Specific Gravity AASHTO T 84 and T 85 2.680
Bulk Specific Gravity
Absorption
er
AASHTO T 84 and T 85
AASHTO T 84 and T 85
2.629
0.7
pe
Fine aggregate angularity 47
DOTD TR 121
Coarse aggregate angularity ASTM D 5821 100
Flat Elongation ASTM D 4791 0

ot
146 (b)
tn
GNP Type Properties Value
Carbon Content 99.82 (%)

rin
Ash 0.18 (%)

Nano27
Moisture 0.97 (%)
Surface Area 250 m2/g

ep
147
Pr
148 Preparation of SMA mixes
ed
149 Five mixes, prepared with five different additives and with a NMAS of 12.5 mm, were fabricated
150 using the materials described above. A Level 2 design (Ninitial= 7, Ndesign = 65, and Nfinal = 105
iew
151 gyrations) was performed, adhering to the guidelines outlined in AASHTO M 325: Standard
152 Specification for Stone Matrix Asphalt (SMA), AASHTO R35: Standard Practice for Superpave
153 Volumetric Design for Hot Mix Asphalt (HMA), and Section 502-6 of the Louisiana Standard
ev
154 Specifications for Roads and Bridges. Table 2 presents details of the job mix formula of the
155 SMA evaluated in this study.
r
156
157 Table 2 Job mix formula (JMF) of the evaluated mixtures
Properties
er SMA
pe
JMF Gradation Specification Limits
¾ in. (19.0 mm) 100 96 - 100
½ in. (12.5 mm) 95 91 - 99

ot
3/8 in. (9.5 mm) 73 69 – 77
No. 4 (4.75 mm) 32 28 – 36

tn
No. 8 (2.36 mm) 19 16 – 22

Aggregate Gradation
No. 16 (1.18 mm) 18 16 – 20
No. 30 (600 µm) 15 13 – 17

rin
No. 50 (300 µm) 13 11 – 15
No. 100 (150 µm) 10 8 – 12
No. 200 (75 µm) 7.3 6.6 - 8

ep
Voids in Total Mix (VTM), % 3.5 2.5 - 4.5
Gsb 2.629
Pr
Binder Performance Grade 76-22
%AC 6.2
ed
Additives Vary
NMAS (mm) 12.5
iew
Design Type Wearing Course
158
159 All the mixtures were laboratory-mixed and laboratory-compacted and were assigned unique
ev
160 Mix ID for ease of analysis. The dosage rate, blending temperature, and compaction temperature
161 of the mixes prepared with WMA and adhesion promoters were selected either based on the
r
162 manufacturers’ recommendations or were adopted from the literature (20-22). The details of the
163 SMA mixes evaluated in this study are presented in Table 3.
164
er
pe
165 Table 3 Description of the evaluated SMA mixtures
Mix
Components
SMAChe1 SMAChe2 SMAOrg1 SMAAsa1 SMAGNP
ot
Binder PG 76-22
Additive Che1 Che2 Org1 Asa1 GNP

tn
Additive Type WMA WMA WMA ASA Nanomaterial
Dosage rate 0.6% 0.1% 2.5% 0.6% 3%

rin
Mixing Temperature 150°C 150°C 150°C 163°C 150°C
Compaction Temperature 140°C 140°C 140°C 155°C 140°C
166
ep
Pr
10
167 Experimental Test Plan
ed
168 In this study, the FTIR and the sessile drop test methods were used to investigate the asphalt
169 binder modification properties and the SFE parameters of the modified asphalt binder. On the
iew
170 other hand, the Indirect Tensile Asphalt Cracking Test (IDEAL-CT) and Modified Lottman
171 (ML) test (AASHTO T283) were used to quantify the cracking and moisture-damage resistance
172 performance of the mixtures, respectively. Based on a partial test factorial, five SMA mixtures
ev
173 were fabricated using three warm-mix additives (Che1, Che2, and Org1), one anti-stripping
174 agent (Asa1), and one nanomaterial (GNP). A cost-effectiveness analysis and a 2-D performance
175 interaction diagram (PID) were used to identify the most cost-effective additive for the SMA mix
r
176 design employed in this study, based on the mixture’s performance test results. Figure 1
177
178
er
illustrates the experimental plan, test methods, and performance parameters adopted in this
study.
pe
ot
tn
rin
ep
Pr
11
ed
iew
r ev
er
pe
179
ot
180
181 Figure 1. Experimental plan for the study

tn
182 EXPERIMENTAL METHODS

rin
183 Asphalt Binder Experiment
184 Fourier Transform Infrared Spectroscopy (FTIR) Test

ep
185 FTIR analysis is a widely-recognized test method for detecting the presence or changes in
186 functional groups during the modification process of asphalt binders by operating on the
Pr
187 principle that each molecular structure of a substance either absorbs or emits energy at a specific
12
188 wavenumber, resulting in unique IR absorption patterns for each molecular compound (Hou et
ed
189 al. 2018). In this study, disposable Real Crystal IR cards, containing a KBr substrate, were
190 coated with the asphalt binder sample. A KBr beam splitter was then used to run approximately
191 50 scans at a resolution of 4 cm-1, with a spectrum range of 400-4000 cm-1 over the samples. All
iew
192 the FTIR tests were conducted using a Nicolet 8700 spectrometer, and data collection and
193 analysis were carried out with the Omnic 6.2 software.
ev
194 Sessile Drop Test
195 The sessile drop test is a widely employed testing technique to determine the surface free energy
r
196 of the materials. This method involves measuring the contact angle of the materials under
197 investigation by dropping a liquid with a known SFE onto a surface. In this study, three different
198
er
liquids (water, ethylene glycol, and formamide) with known SFE were dropped into the surface
pe
199 of the glass plate coated with asphalt binder to determine their contact angles and hence the
200 surface free energy (24). At least three droplets of the same liquid were utilized, and for each
201 droplet, more than 100 contact angles on each of its sides were measured using an Optical
ot
202 Contact Analyzer (OCA) to evaluate the properties of the binder. A consistent liquid drop
203 volume of 3 μL was employed for all the samples and carefully regulated while testing. Using
tn
204 the contact angles obtained from the OCA device, the three different components of SFE were
205 computed: a monopolar Lewis acidic component (Γ+), a monopolar Lewis base component (Γ-),
rin
206 and a nonpolar Lifshitz-van der Waals (ΓLW) component, or the dispersive component. The total
207 SFE (Γtotal) as defined by the total amount of work required to increase the surface of the
ep
208 materials by a unit area was obtained by combining ΓLW and ΓAB as follows (Little and Bhasin
209 2006):
Pr
210 Γtotal = ΓLW + ΓAB (1)
13
211 where, ΓAB, acid-base or polar component = 2 √(Γ+Γ−);
ed
212 The polar components (Γ+ and Γ−) of SFE refer to the amount of energy that arises when the
213 polar components of two materials interact with each other. On the other hand, the nonpolar
214 Lifshitz-van der Waals (ΓLW) component or the dispersive component of SFE, refers to the
iew
215 amount of energy that arises from the interaction of the non-polar molecules of two materials at
216 the surface (Little and Bhasin 2006).
ev
217 𝑤𝑐 = 2𝛤𝑡𝑜𝑡𝑎𝑙 (2)
218 where, 𝑤𝑐= work of cohesion, which represents the energy required to initiate a fracture within
r
219 the asphalt binder.
220 Based on the numerical values of the above components, the SFE parameters (work of cohesion,
221 er
work of adhesion, work of debonding, and energy parameters) of different aggregate (limestone,
pe
222 sandstone, and gravel) and binder systems were then estimated by following the Good-van Oss-
223 Chaudhury theory and the Young-Dupre equation as detailed elsewhere (Van Oss et al. 1987).
224 Asphalt Mixture Experiment

ot
225 Modified Lottman (ML) Test

tn
226 The modified Lottman test was conducted in accordance with AASHTO T283 test procedure to
227 evaluate the moisture susceptibility of the SMA samples by mimicking the effect of saturation
rin
228 and accelerated water conditioning with the use of freeze-thaw cycles. A total of six samples,
229 each with a diameter of 150 mm and a thickness of 95 mm, while targeting an air void content of
230 7.0 ± 0.5%, were fabricated for all the mixes and were then divided into two sets: a saturated
ep
231 (wet) set and an unsaturated (dry) set. Following the freezing process, the samples were then
232 transferred to a water bath at a temperature of 60°C for 24 ± 1 hours and then again at 25°C for
Pr
14
233 2h ± 10 min before testing. A monotonic universal loading frame is used to determine the
ed
234 indirect tensile strength of the samples. The numerical value of the tensile strength ratio, an
235 indicator of the mixtures’ resistance to moisture-induced damage, was obtained through a
236 comparison between the average indirect tensile strength of the saturated and the unsaturated
iew
237 sample groups as follows:
𝑠𝑡, 𝑤𝑒𝑡
238 𝑇𝑆𝑅 = 𝑠𝑡,𝑑𝑟𝑦 (3)
ev
239 where,
240 𝑠𝑡, 𝑤𝑒𝑡= average indirect tensile strength of the saturated or wet sample group; and
r
241 𝑠𝑡,𝑑𝑟𝑦= average indirect tensile strength of the unsaturated or dry sample group.
242 er
Indirect Tensile Asphalt Cracking Test (IDEAL-CT) test
pe
243 The Indirect Tensile Asphalt Cracking Test (IDEAL-CT) test was conducted at 25 ± 0.5°C
244 according to the test procedure described in ASTM D 8225, Determination of Cracking
245 Tolerance Index of Asphalt Mixtures using the Indirect Tensile Cracking Test at Intermediate
ot
246 Temperatures. IDEAL-CT requires samples with a diameter of 150 ± 2 mm, a thickness of 62 ±
247 1 mm, and an air void of 7.0 ± 0.5%. At least three replicates were prepared for each mix type.
tn
248 After each test, a load-displacement curve was generated, which was then used along with the
249 sample dimensions to calculate the CT index as follows:

rin
𝑡 𝐺𝑓 𝑙75
250 𝐶𝑇𝑖𝑛𝑑𝑒𝑥 = 62 × |𝑚75| × 𝐷
× 106 (4)
251 where,
ep
𝑝85 ― 𝑝65
252 𝑚75 = 𝑙85 ― 𝑙65 (5)
253 where,
Pr
15
254 𝑡 = sample thickness, mm;
ed
255 𝐷 = sample diameter, mm;
256 Pmax = peak load;
257 |m75| = post-peak slope;
iew
258 𝑝85 = corresponding to 85% of 𝑃𝑚𝑎𝑥 at the post-peak stage;
259 𝑝65 = corresponding to 65% of 𝑃𝑚𝑎𝑥 at the post-peak stage;
ev
260 L85 = displacement corresponding to 85% of the peak load (𝑃𝑚𝑎𝑥) at the post-peak stage;
261 l75 = displacement corresponding to 75% of the peak load (𝑃𝑚𝑎𝑥) at the post-peak stage;
r
262 l65 = displacement corresponding to 65% of the peak load (𝑃𝑚𝑎𝑥) at the post-peak stage; and
263 Gf = failure energy or the total area under the load-displacement curve divided by the product of
264 the sample dimensions. er

pe
265
266 The Fracture strain tolerance (FST) parameter was also calculated from the load-displacement
267 curves obtained from the IDEAL-CT test results for all the samples. Numerous studies have
ot
268 suggested that FST offers improved repeatability of performance attributes and, hence, could
269 serve as a promising alternative to the CT-index parameter in gauging the cracking susceptibility
tn
270 of asphalt mixtures (Habbouche et al. 2021; Boz et al. 2021). The FST is calculated as follows:
271
rin
𝐺𝑓
272 Fracture Strain Tolerance (FST)= 𝑆𝑡 × 106 (6)
273 where,
ep
2000𝑃max
274 St = 𝜋𝑡𝐷
× 103 (7)
275 where,
Pr
16
276 St = Indirect tensile strength in kPa
ed
277
278 Cost Effectiveness Analysis
iew
279 Cost-effectiveness analysis was used in this study to provide a quantitative assessment of the
280 benefits gained from the considered additives as compared to their cost. Several methods are
281 available to evaluate the cost-effectiveness of asphalt mixtures. This study adopted a simple
ev
282 approach that divided the cost of the mixture by its expected performance. Since the focus of the
283 study is on the cracking and moisture resistance performance, the results of the Cracking
284 Tolerance Index (CTindex) and Tensile Strength Ratio (TSR) were used as indicators of mixture
r
285 performance to calculate the cost effectiveness of the evaluated mixtures. The cost-effectiveness
286 er
(CE) of any given mixture was calculated as follows:
pe
287
𝐸𝑖
288 𝐶𝐸𝑖 = 𝐶𝑖 ∗ 100 (8)
289
ot
290 where,
tn
291 Ci = Mixture unit cost (per ton); and
292 Ei = Expected performance.
293
rin
294 The expected performance, Ei was calculated using the following equation:
295
ep
𝑇𝑆𝑅𝑖 𝐶𝑇𝑖
296 𝐸𝑖 = 𝑇𝑆𝑅max + 𝐶𝑇max (9)
297
Pr
17
298 where,
ed
299 𝑇𝑆𝑅𝑖 = mixture TSR value;
300 𝑇𝑆𝑅max = maximum TSR value obtained from all the mixes;
301 𝐶𝑇𝑖 = mixture CT-index value; and
iew
302 𝐶𝑇max = maximum CT-index value obtained from all the mixes.
303
ev
304 Based on Equation (9), the most cost-effective mixture is the one that provides a higher CE value
305 i.e., a higher ratio of performance to cost, as compared to other mixtures. The estimated initial
306 costs per ton with the dosage rate that was used in the experimental program for the SMA mixes
r
307 with Che1, Che2, Org1, Adhere, and Graphene Nanoplatelet, were $88.8, $86.0, $84.2, $86.0,
308 and $241.5, respectively. er

pe
309 Statistical Analysis
310 The laboratory performance results of the SMA mixtures were statistically analyzed using the
311 analysis of variance (ANOVA) technique at a 95% confidence interval (α of 0.05) with the aim
ot
312 of assessing if the performance of the SMA mixtures varied significantly. It should be noted that,
313 before conducting the ANOVA of the test samples, the Shapiro-Wilk test and Levene test were
tn
314 employed using Statistical Analysis System (SAS) 9.4 software to validate that the test
315 assumptions were met: namely, that each sample is derived from a normally-distributed
rin
316 population, and that the variances are equal. Following ANOVA, a multiple comparison
317 procedure known as the Tukey’s Honest Significant Difference (HSD) test was conducted to
ep
318 identify statistically significant mean performance values. Based on the significance of the
319 results from Tukey’s HSD test, statistical groups were assigned different alphabetical letters: A,
Pr
320 B, C, and so forth, symbolizing the performance from best to worst. The assignment of multiple
18
321 letters, such as A/B or A/B/C suggests that the variation in the averages is not significant enough
ed
322 to be statistically different.
323 RESULTS AND DISCUSSION
iew
324 Fourier Transform Infrared Spectroscopy Test Results
325 Figure 2 presents the FTIR spectra of the PG 76-22 SBS asphalt binder modified with different
326 warm-mix additives, an anti-stripping agent, and a nanomaterial additive (GNP). Depending on
ev
327 the type of molecular structures or functional groups present in the asphalt binder, different
328 peaks were observed and identified. As seen in Figure 2 (a-d), the original PG 76-22 asphalt
r
329 binder, which was SBS-modified, displayed peak responses at wavenumbers 690-870, 966, 1030,
330
331
er
1160, 1310, 1370, 1460, 1510-1540, 1580, 1600, 1650, 1680-1730, 2850-2960, and 3060 cm-1,
representing C-H bending, strong C=C bending from alkene, strong S=O stretching from
pe
332 sulfoxide, strong C-O stretching from tertiary alcohol, strong C-O stretching from aromatic ester
333 or strong C-N stretching from aromatic amine, strong S=O stretching from sulfonate or medium
334 O-H bending from phenol, medium C-H bending from alkane, strong N-O stretching from nitro
ot
335 compound, medium N-H bending from amine, medium C=C stretching from cyclic alkene,
336 strong C=O stretching or weak C-H bending from aromatic compound, strong C=O stretching,
tn
337 medium C-H stretching from alkane, strong O-H stretching from carboxylic acid or strong N-H
338 stretching from amine salt or medium C-H stretching from alkene, and medium C-H stretching
rin
339 from alkene or weak O-H stretching from alcohol, respectively.
340 On the other hand, all the modified asphalt binders displayed new peaks or peak areas
ep
341 that overlapped with the original PG 76-22 asphalt binder or exhibited reduced absorption peaks
342 compared to the original asphalt binder, indicating modification and change in the molecular
Pr
19
343 groups. For example, Org1-modified PG 76-22 asphalt binder (PG 76-22 + Org1) manifested a
ed
344 strong C=C bending from alkene, indicated by peaks at around 733 cm-1. GNP modified PG 76-
345 22 asphalt binder (PG 76-22 + GNP) did not show any peaks at wavenumber of 1030, 1160,
346 1310, 1510-1540, 1680-1730, and 3060 cm-1. It also showed lower absorption at 1370, 1460,
iew
347 1580, 1600, 1650 cm-1 wavenumbers, and showed higher absorption at 2850-2960 cm-1
348 wavenumbers, indicating that some of the S=O, C-O, N-O, and C=O stretching could not be
349 detected when PG 76-22 asphalt binder was modified with GNP. Asa1-modified asphalt binder
ev
350 (PG 76-22 + Asa1) did not display any peaks in the 1500-1550, 1650-1750 cm-1 wavenumber
351 range, indicating the absence of N-O and C=O stretching. Additionally, it exhibited lower
r
352 absorption at around the 2960 cm-1 peak, suggesting fewer O-H, N-H, and C-H bonds in that
353
354
er
spectral region. Interestingly, the Org1-modified asphalt binder displayed nearly all the peaks of
the original PG 76-22 binder, with strong peaks around 1370, 1460, and 2960 cm-1, indicating a
pe
355 higher amount of O-H, S=O, C-H, and N-H bonds. Similar to PG 76-22 + Org1 asphalt binder,
356 Che2-modified asphalt binder (PG 76-22 + Che2) did not show any distinct peaks other than the
357 peak at 1460 cm-1 wavenumber. Furthermore, in the case of Che1-modified asphalt binder (PG
ot
358 76-22 + Che1), almost all the peaks from the original binder remained undetected, except for the
tn
359 peak at 1370, 1460, and 2960 cm-1 wavenumbers, where it showed lower absorption. Overall, the
360 FTIR results showed that, depending on the additive type, each additive demonstrated specific
rin
361 absorption reactions at different regions and induced notable changes in the original asphalt
362 binder (PG 76-22) spectra, thereby confirming asphalt binder modification.
ep
363
Pr
20
ed
iew
ev
364
(a) (b)
r
365
er
pe
ot
tn
366
rin
(c) (d)
367
ep
368 Figure 2. FTIR test results
369
Pr
21
370 Sessile Drop Test Results
ed
371 Surface Free Energy Components of Asphalt Binder Blends
372 In this study, the sessile drop test method was utilized to examine the impact of the additives on
iew
373 the SFE components of the asphalt binder blends, as well as the potential for moisture-induced
374 damage in the asphalt mixes. The asphalt binder's ability to adequately coat the aggregate and
375 establish a strong bond is a function of its SFE components, specifically, the non-polar, acidic,
ev
376 and base components. Figure 3 presents the SFE components of PG 76-22 asphalt binder
377 modified with different additives evaluated in this study. According to the test results, GNP-
r
378 modified PG 76-22 asphalt binder showed the highest total SFE among all the modified asphalt
379 binders evaluated in this study. The total SFE of PG 76-22 + GNP was approximately 8 mJ/m2
380
er
higher than the Che1-modified PG 76-22 asphalt binder, (PG 76-22 + Che1). This could be
pe
381 attributed to the nanomaterial modification of the PG 76-22 asphalt binder, which in turn
382 increased the non-polar component (ΓLW) values of the PG 76-22 + GNP binder. These results
383 agreed with the results of a previous study (Yang 2013).

ot
384 Comparing the results, the ΓLW component of PG 76-22 + Org1 was the highest, whereas
385 PG 76-22 + Asa1 showed the lowest (37.14 mJ/m2). Overall, the following order was found
tn
386 when the PG 76-22 asphalt binder blends modified with different additives were ranked
387 according to their ΓLW components value from highest to lowest: PG 76-22 + Org1, PG 76-22 +
rin
388 GNP, PG 76-22 + Che1, PG 76-22 + Che2, and PG 76-22 + Asa1 (ΓLW value of 47.24 mJ/m2,
389 46.93 mJ/m2, 40.91 mJ/m2, 38.56 mJ/m2, and 37.14 mJ/m2, respectively). Asphalt binders with
ep
390 higher ΓLW are expected to show improved elastic properties, as it is well recognized that the
391 non-polar molecules of the asphalt binder act as a matrix for their counterparts, thereby
Pr
22
392 increasing the work of adhesion and indicating a superior binder-aggregate bond under dry
ed
393 conditions (Jones and Kennedy 1991).
iew
r ev
394
395
er
Figure 3. Sessile Drop test results
pe
396 The acid (Γ+) and base (Γ-) components of the SFE measurements for all the asphalt binder
397 suggest that introducing Org1 to the PG 76-22 asphalt binder might have decreased its base SFE
398 component, and on the contrary, introducing Asa1 might have decreased the acid component of
ot
399 the PG 76-22 asphalt binder. It is to be noted that the presence of high polar components in the
tn
400 asphalt binder could increase the potential for fatigue cracking, rutting, and moisture-induced
401 damage in the asphalt mixture (Jones and Kennedy 1991). However, the stripping potential of the
rin
402 asphalt mixture also depends on the wettability, adhesion, debonding, and energy ratio
403 parameters of the asphalt binder-aggregate system, which were also evaluated as part of this
404 study. The work of cohesion, which is twice the surface free energy of the asphalt binders, is
ep
405 presented in Figure 4(a) and follows the same trend as the total SFE presented in Figure 3.
406
Pr
23
407 Effect of Additive Type on Asphalt Binder Wettability
ed
408 Figure 4(b) presents the spreading coefficients of the PG 76-22 asphalt binder modified with
409 different types of additives, and aggregate. The higher the spreading coefficient, the easier it is
iew
410 for the asphalt binder to coat the aggregate surface, hence strengthening the bond. As illustrated,
411 regardless of the aggregate type, the PG 76-22 asphalt binder modified with GNP had a higher
412 spreading coefficient (i.e., 101.56 mJ/m2 for limestone aggregate) than any other PG 76-22
ev
413 warm-mix additive or anti-stripping agent-modified asphalt binder evaluated in this study. On
414 the contrary, the PG 76-22 asphalt binder with Asa1 exhibited the lowest spreading coefficient
415 (i.e., 69.18 mJ/m2 for limestone aggregate) among all the modified binders investigated,
r
416 indicating that it might not coat the aggregate well enough to form a strong bond.
417
er
Effect of Additive Type on Asphalt Binder Work of Adhesion
pe
418 Work of adhesion is defined as the amount of energy needed to detach the asphalt binder from
419 the aggregate-binder interface (Bhasin and Little 2007). Therefore, to create a strong bond
420 between an asphalt binder and the aggregate in dry conditions, it is desirable to have a greater
ot
421 work of adhesion. Figure 4(c) presents the work of adhesion among the three types of aggregate
422 (limestone, sandstone, and gravel) and all the PG 76-22 asphalt binder modified with Che1,
tn
423 Che2, Org1, Asa1, and GNP. Similar to the trend observed for wettability results, PG 76-22
424 asphalt binder with GNP exhibited the highest work of adhesion among the tested samples,
rin
425 whereas PG 76-22 asphalt binder with Asa1 showed the lowest work of adhesion between all
426 three aggregate types. In general, the gravel and the limestone aggregate had the highest and
ep
427 lowest work of adhesion, respectively, with all the modified binders evaluated in this study.
Pr
24
428 Effect of Additive Type on Asphalt Binder Work of Debonding
ed
429 The likelihood of detachment occurring between an asphalt binder and aggregate in the presence
430 of moisture can be assessed quantitatively by measuring the work of debonding. Asphalt binders
iew
431 that exhibit a greater degree of work of debonding, characterized by a more negative value, are
432 generally more prone to experiencing sensitivity to stripping in comparison to asphalt binders
433 with a lower work of debonding, indicated by a less negative value (Ghabchi et al. 2014). Figure
ev
434 4(d) presents the work of debonding for all the asphalt binders evaluated in this study. It is
435 evident from the results that GNP might lead to improved stripping resistance of the PG 76-22
436 asphalt binder as it showed the lowest (i.e., least negative) work of debonding compared to the
r
437 other additives investigated in the study. Comparing the work of debonding for all the asphalt
438
439
er
binders also revealed that Asa1 demonstrates a consistently higher magnitude of debonding
energy for all the cases, indicating increased susceptibility to stripping. Overall, gravel and
pe
440 limestone aggregate showed the highest and lowest work of debonding, respectively, with all the
441 asphalt binders modified with different additives.

ot
442 Effect of Additive Type on Asphalt Binders’ Energy Ratio Parameters
443 Figure 4(e-f) presents the energy parameters ER1 and ER2 calculated for different aggregate types
tn
444 (limestone, sandstone, and gravel) and asphalt binder combinations. Asphalt binder and
445 aggregate combinations with higher energy parameters value are expected to show better
rin
446 moisture induced damage resistance than asphalt binder and aggregate combination with lower
447 energy parameter values. As illustrated, PG 76-22 asphalt binder modified with GNP showed the
ep
448 highest ER1 and ER2 values as compared to others for all the aggregates, whereas PG 76-22
449 asphalt binder modified with Asa1 showed the lowest.

Pr
450
25
ed
iew
451
ev
452
(a) (b)
r
453
er
pe
ot
454
tn
(c) (d)
455
rin
ep
Pr
26
ed
iew
456
ev
(e) (f)
457 Figure 4. Surface Free Energy (SFE) parameters (a) Work of Cohesion; (b) Spreading
r
458 Coefficient or Wettability; (c) Work of Adhesion; (d) Work of Debonding; (e) Energy ratio, ER1;
459 er
(f) Energy ratio, ER2
pe
460 IDEAL-CT Test Results
461 The IDEAL-CT test was conducted at 25°C with at least three replicates for each SMA mixture.
462 As shown in Figure 5, the average CT-index of the SMA mixtures varied from 253 to 468, with a
ot
463 Coefficient of Variation (COV) ranging from 5.6% to 24.2% without the outliers. It should be
tn
464 noted that the error bars represent one standard deviation above and below the mean value of the
465 three replicates selected for each mix. Among all the SMA mixtures evaluated, Che2-modified
466 SMA mixtures (SMAChe2) showed the highest CT-index, while Asa1-incorporated mixture
rin
467 (SMAAsa1) showed the lowest. Higher CT index values typically suggest superior cracking
468 resistance, thereby implying a reduced potential for cracking in the field. Although slightly
ep
469 lower, Graphene Nanoplatelet modified SMA mixture (SMAGNP) showed almost the same level
470 of CT-index result as Che2-modified SMA mixture (468 and 454 respectively for SMAChe2 and
Pr
471 SMAGNP) and hence were both assigned the same statistical group, (A). As observed, the
27
472 modification of the binder with Che2 and GNP increased the CT-index of the SMA mix by 34.1
ed
473 and 30.0%, respectively, compared to the control mix (SMAChe1), which contained Che1.
474 As shown in Figure 5, none of the mixes failed according to the CT-index performance
iew
475 criterion of 70 set by the Virginia Department of Transportation (VDOT) to reduce the cracking
476 potential of the surface mixtures, indicating that all the warm mix additives, anti-stripping agent,
477 and the nanomaterial evaluated in this study are expected to exhibit sufficient resistance against
ev
478 cracking in the field (Boz et al. 2021).
479 Interestingly, the trend of the CT-index results for all the asphalt binders mirrored the
r
480 surface free energy parameters (i.e., total SFE, work of adhesion, work of debonding, and energy
481 ratios) presented in this study, except for the cases involving Che2 and Org1. This suggests that
482
er
the performance of the mixture does not only depend on the surface energy components of the
pe
483 asphalt binder and aggregate, but it also depends on the characteristics of the mixture itself (e.g.,
484 NMAS, asphalt content, passing No. 200 materials, etc.).
485
ot
tn
rin
ep
486
487 (a)
Pr
28
ed
iew
488
ev
489 (b)
490 Figure 5. IDEAL-CT test results at 25°C: (a) CT-Index results and (b) FST results
r
491
492
er
As illustrated in Figure 5(b), FST index values of the SMA mixtures evaluated ranged from
12.37 to 15.15 with a COV between 3.4% to 8.1%, indicating greater repeatability of the test
pe
493 results as compared to CT-index parameter. The results indicate that, Che2 and GNP increased
494 the FST index performance of the SMA mixtures by about 11.5 and 13.9%, respectively,
495 compared to Che1. However, the difference between the average FST index values of all three
ot
496 additives was not statistically significant, therefore, they were all categorized under either 'A' or
497 'A/B' statistical groups. Interestingly, unlike the CT-index results, Org1 exhibited a slightly lower
tn
498 FST index than Asa1, and collectively they were placed into the lowest statistical performance
499 category among all the additives studied.

rin
500 Modified Lottman (ML) Test Results
501 The average indirect tensile strength (ITS) of both conditioned and unconditioned samples, as
ep
502 well as the tensile strength ratio of the SMA mixtures and their respective statistical analyses, are
503 presented in Figure 6. It can be observed that, SMAChe1 exhibited the highest TSR value,
Pr
29
504 followed by SMAOrg1, SMAAsa1, SMAChe2, and SMAGNP. According to the results, all the
ed
505 warm mix additives (Che1, Che2, and Org1), and anti-stripping agent (Asa1) modified SMA
506 mixtures had a TSR value greater than 0.80, whereas the GNP-modified SMA mixture had a
507 TSR value of 0.76. Given the minimum acceptance threshold of 0.80 set by AASHTO T 283, it
iew
508 suggests that the SMA incorporating GNP may exhibit inadequate moisture resistance. Yet,
509 SMAGNP exhibited the highest unconditioned ITS among all the mixtures evaluated.
r ev
er
pe
510
ot
511 Figure 6. Modified Lottman (ML) test results

tn
512 Cost-Effectiveness Results
513 As the selection of additives depends not only on their performance but also on their associated
rin
514 cost, a cost-effectiveness analysis was conducted to compare the overall benefits of all the
515 evaluated SMA mixtures. Figure 7 presents the cost-effectiveness results of all the SMA
ep
516 mixtures prepared with the five different additives evaluated in this study. It can be seen that, the
517 SMA mixture prepared with Che2 showed the best cost-effectiveness, followed by SMA
518 mixtures prepared with Che1, Org1, Adhere, and GNP. Overall, the low cost-effectiveness (CE)
Pr
30
519 values of the SMA mixtures prepared with warm-mix additives (i.e., SMAChe2, SMAChe1,
ed
520 SMAOrg1) suggest that the use of these additives is a more cost-effective strategy for improving
521 the cracking and moisture resistance performance of asphalt mixtures. However, further research
522 is needed on their field performance, maintenance cost, and end-of-life options in order to
iew
523 evaluate the overall cost-effectiveness of all the additives over the entire lifespan of the surface
524 mix.
ev
SMAGNP
r
SMAAsa1
SMAOrg1
er
pe
SMAChe2
ot
SMAChe1
tn
0 1 2 3 4
Cost Effectiveness
525
rin
526 Figure 7. Cost-Effectiveness of the evaluated SMA mixtures
527 2-D Performance Interaction Diagram (PID) Analysis

ep
528 A 2-D performance interaction diagram was developed based on the laboratory mechanical
529 performance and the cost-effectiveness results of all the SMA mixtures evaluated in order to
Pr
31
530 select the best performing additive in this study. Numerous studies have proposed a minimum
ed
531 CT-index value of 70 as the cutoff for acceptable performance of the wearing or surface course
532 and also suggested a minimum TSR value of 0.8 as the threshold for satisfactory moisture
533 resistance in asphalt mixtures. Figure 8 presents the 2-D PID developed in this study for the
iew
534 evaluated mixtures. Mixtures placed in the upper-right-hand corner of the green region in the 2-
535 D PID are considered the best, and vice versa. As illustrated, although GNP modified SMA mix
536 (i.e., SMAGNP) showed excellent cracking resistance, it failed to pass the moisture resistance
ev
537 threshold (i.e., TSR= 0.8), and also showed the lowest cost-effectiveness (CE) value, thereby
538 placing it outside of the green region. On the other hand, SMA mixture prepared with Che2 (i.e.,
r
539 SMAChe2) exceeded both the cracking resistance and moisture susceptibility thresholds, and
540
541
er
also showed the highest cost-effectiveness value, therefore being placed in the green region.
When ranked according to the position of the mixtures in the 2-D PID, the following order was
pe
542 found: SMAChe2, SMAChe1, SMAOrg1, SMAAsa1, and SMAGNP, suggesting a performance
543 range from the best to the least effective SMA mixture.
ot
tn
rin
ep
Pr
32
ed
iew
r ev
544
545
er
Figure 8. 2-D performance interaction diagram (PID) between TSR, CT-index, and the cost-
pe
546 effectiveness value of the mixtures
547 SUMMARY AND CONCLUSION

ot
548 The objective of this study was to evaluate the laboratory performance and the cost effectiveness
549 of the SMA mixtures and to evaluate the effects of warm-mix additives, anti-stripping agent, and
tn
550 graphene nanoplatelet on their performance. Based on the results of the experimental program,
551 the following conclusions can be drawn:

rin
552  Investigation of the functional groups of the asphalt binder blends suggests that each additive
553 demonstrated unique absorption reactions and caused alterations in the original PG 76-22
ep
554 asphalt binder spectra, confirming asphalt binder modification.
555  Sessile drop test results show that GNP modification of the asphalt binder increased the total
Pr
556 SFE, wettability, adhesion, and debonding properties, as well as the energy ratios, indicating
33
557 greater resistance to cracking and moisture resistance. However, mixture performance against
ed
558 pavement distresses also depends on its characteristics, such as NMAS, asphalt content,
559 aggregate gradation, passing No. 200, etc.
iew
560  According to the cracking resistance test results, SMA mixture prepared with Che2 and GNP
561 (i.e., SMAChe2, SMAGNP) are expected to show superior performance against cracking in
562 the field, whereas the anti-stripping agent (Asa1) modified SMA mixture (i.e., SMAAsa1) is
ev
563 expected to show the worst resistance as compared to the mixes evaluated in this study.
564  Moisture susceptibility test results suggested that warm-mix additives would show superior
r
565 performance in terms of moisture resistance as compared to other additives evaluated. It also
566
567 
er
indicates that GNP might not exhibit adequate performance against moisture in the field.
In terms of cost-effectiveness, the Che2-modified SMA mixture was found to be the most
pe
568 cost-effective. Overall, all the warm-mix additives showed good cost-effectiveness,
569 indicating greater benefits from their use in the asphalt pavements.
570  A 2-D performance interaction diagram suggests the following order of the evaluated SMA
ot
571 mixtures when ranked according to their laboratory performance and cost-effectiveness:
tn
572 SMAChe2, SMAChe1, SMAOrg1, SMAAsa1, and SMAGNP.
573
rin
574 Based on the findings of this study, Che2 is expected to deliver satisfactory field performance for
575 SMA mixtures in terms of cracking and moisture susceptibility resistance, while also being the
ep
576 most cost-effective. However, it should be noted that the GNP-modified SMA mixture exhibited
577 promising results in terms of surface free energy and superior cracking resistance compared to
578 other evaluated mixtures. It also demonstrated a higher unconditioned indirect tensile strength
Pr
34
579 and a TSR value close to the minimum threshold limit. Therefore, further research into its
ed
580 potential use in asphalt mixtures (e.g., dosage rate, performance with other types of mixtures,
581 etc.) is needed for a more comprehensive assessment.
iew
582 ACKNOWLEDGMENTS
583 The researchers of this study are appreciative of the financial support of Tran-SET through the
584 21BLSU03 grant. Furthermore, the authors would like to acknowledge the assistance of Sean
ev
585 Maher and Barry Moore of the Louisiana Department of Transportation and Development.
586 REFERENCES
r
587 Abohamer, H., M.A. Elseifi, N. Dhakal, Z. Zhang, and C. Fillastre. Effects of Asphalt Concrete
588 Stripping on the Performance and Cost-Effectiveness of Chip Seal and Asphalt Concrete
589
590
591
Transportation Research Board, 2021. er
Overlay in Hot and Humid Climates. Transportation Research Record: Journal of
Abohamer, H., M.A. Elseifi, C. Mayeux, S.B. Cooper III, and S. Cooper Jr, S. Laboratory
pe
592 Evaluation of Warm-Mix Asphalt Open-Graded Friction Courses. Transportation
593 Research Record: Journal of the Transportation Research Board, 2022. 2676: 507-519.
594 Bhasin, A., D.N. Little, K.L. Vasconcelos, and E. Masad. Surface free energy to identify
595 moisture sensitivity of materials for asphalt mixes. Transportation Research Record:
596 Journal of the Transportation Research Board, 2007. 2001:37–45.
ot
597 Birgisson, B., R. Roque, and G. C. Page. Evaluation of Water Damage Using Hot Mix Asphalt
598 Fracture Mechanics. Journal of the Association of Asphalt Paving Technologists, 2003.
599 72: 516–562.
tn
600 Boz, I., J. Habbouche, S. D. Diefenderfer, and Y. K. Bilgic. Precision Estimates and Statements
601 for Performance Indices from the Indirect tensile Cracking Test at Intermediate
602 Temperature. Transportation Research Record: Journal of the Transportation Research
603 Board, 2021. 2676: 225–241.
rin
604 Brown, E. R., P. S. Kandhal, F. L. Roberts, Y. R. Kim, D.-Y. Lee, and T. W., Kennedy. Asphalt
605 Refining, Uses, and Properties. In Hot Mix Asphalt Materials, Mixtures Design, and
606 Construction, Lanham, Maryland, 2009.
607 Fakhri, M. The effects of nano zinc oxide (ZnO) and nano reduced graphene oxide (RGO) on
ep
608 moisture susceptibility property of stone mastic asphalt (SMA). Case Studies in
609 Construction Materials, 2021. 15, e00655.
Pr
35
610 Ghabchi, R., D. Singh, and M. Zaman. Evaluation of moisture susceptibility of asphalt mixes
611 containing RAP and different types of aggregates and asphalt binders using the surface
ed
612 free energy method. Construction and Building Materials, 2014. 73, 479-489.
613 Guo, F., J. Pei, J. Zhang, R. Li, P. Liu, and D. Wang. Study on adhesion property and moisture
614 effect between SBS modified asphalt binder and aggregate using molecular dynamics
615 simulation. Materials, 2022. 15(19): 6912.
iew
616 Habbouche, J., I. Boz, S. D. Diefenderfer, and Y. K. Bilgic. Round Robin Testing Program for
617 the Indirect Tensile Cracking Test at Intermediate Temperature: Phase I. Report No.
618 VTRC 22-R3. Virginia Transportation Research Council, Charlottesville, 2021.
619 Hossain, Z., M. Zaman, M. Saha, and T. Hawa. Evaluation of Viscosity and Rutting Properties of
620 Nanoclay- Modified Asphalt Binders. Geo-Congress, 2014, 3695–3702.
ev
621 Hossain, Z., B. Bairgi, and M. Belshe. Investigation of moisture damage resistance of GTR-
622 modified asphalt binder by static contact angle measurements. Construction and Building
623 Materials, 2015. 95: 45-53.
624 Hou, X., S. Lv, Z. Chen, and F. Xiao. Applications of Fourier transform infrared spectroscopy
r
625 technologies on asphalt materials. Measurement, 2018. 121: 304-316.
626 Jones D.R., and T.W. Kennedy. The asphalt model. Results of the SHRP asphalt research
627
628
629
University of Texas; 1991. er
program. SHRP A-1001 contract. Austin, TX: Center for Transportation Research,
Kandhal, P.S. (1992). Moisture susceptibility of HMA mixes: identification of problem and
pe
630 recommended solutions. NCAT Report 92-01, Auburn, AL.
631 Kennedy, T. W., F. L. Roberts, and K. W. Lee. Evaluation of Water Effects on Asphalt Concrete
632 Mixtures. Transportation Research Record: Journal of the Transportation Research
633 Board, 1983. 911: 134–143.
634 Khedmati, M., A. Khodaii, and H. F. Haghshenas. A study on moisture susceptibility of stone
ot
635 matrix warm mix asphalt. Construction and Building Materials, 2017. 144: 42-49.
636 Kim, S. S., J. J. Yang, and R. A. Etheridge. Effects of mix design variables on flexibility index of
637 asphalt concrete mixtures. International Journal of Pavement Engineering, 2020.
tn
638 21(10):1275-1280.
639 Le, J., M. Marasteanu, and M. Turos. Graphene nanoplatelet (GNP) reinforced asphalt mixtures:
640 A novel multifunctional pavement material (No. NCHRP IDEA Project 173), 2016.
rin
641 Little, D.N., and A. Bhasin. Using surface energy measurements to select materials for asphalt
642 pavement, 2006. NCHRP Project 9-37.
643 Santucci, L. Minimizing Moisture Damage in Asphalt Pavements. Pavement Technology
644 Update, University of California Pavement Research Center, Transfer Program, Vol. 2,
ep
645 No. 2, 2010, pp. 1–12.

646 Van Oss C.J., M.K. Chaudhur, R. J. Good. Monopolar surfaces. Advances in colloid and
647 interface science, 1987. 28:35-64.
648 Yang, J., S. Tighe. A Review of Advances of Nanotechnology in Asphalt Mixtures. Procedia—
Pr
649 Social and Behavioral Sciences,2013. 1269–1276.

36
650 Yan, Y., C. Cocconcelli, R. Roque, T. Nash, J. Zou, D. Hernando, G. Lopp. Performance
651 Evaluation of Alternative Polymer-Modified Asphalt Binders. Road Materials and
ed
652 Pavement Design, 2015. 16: 389–403.
653 Yao, H., L. Li, H. Xie, H. Dan, and X. Yang. Microstructure and Performance Analysis of
654 Nanomaterials Modified Asphalt. Road Materials and New Innovations in Pavement
655 Engineering, 2011. 220–228.
iew
656 Yusoff, N., A. Breem, H. Alattug, A. Hamim, and J. Ahmad. The Effects of Moisture
657 Susceptibility and Ageing Conditions on Nano-Silica/Polymer-Modified Asphalt
658 Mixtures. Construction and Building Materials, 2014. 72: 139–147.
659 Xiao, F., W. Zhao, T. Gandhi, and S. N. Amirkhanian. Laboratory investigation of moisture
660 susceptibility of long-term saturated warm mix asphalt mixtures. International Journal of
ev
661 Pavement Engineering, 2014. 13(5): 401-414.
662 Zhao, S., B. Huang, X. Shu, and M. Woods. Comparative evaluation of warm mix asphalt
663 containing high percentages of reclaimed asphalt pavement. Construction and Building
664 Materials, 2013. 44: 92-100.
r
er
pe
ot
tn
rin
ep
Pr
37

GNP 27

Uploaded by

Copyright:

Available Formats

GNP 27

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

GNP 27

Uploaded by

Copyright:

Available Formats

1 Effects of Warm-Mix Additives, Anti Stripping Agent, and Graphene

3 Effectiveness of Stone Mastic Asphalt

17 well as a two-dimensional performance interaction diagram were employed based on the

26 Keywords: Graphene Nanoplatelet, Warm-mix additives, Moisture damage, Cracking, SMA.

49 of maintenance and rehabilitation activities. Moisture-induced damage, typically originating at

52 Moisture and cracking resistances of AC are controlled by a range of design and

66 Although numerous studies were conducted on enhancing asphalt mixtures performance

74 OBJECTIVES AND SCOPE

86 economic viability of using these additives in SMA. A two-dimensional performance interaction

88 laboratory performance and cost-effectiveness.

122 performance of SMA.

130 EXPERIMENTAL PROGRAM

131 Material selection

135 butadiene-styrene (SBS) polymer-modified PG 76-22 asphalt binder, conforming to Louisiana's

Physical Properties Method Results

Bulk Specific Gravity

Flat Elongation ASTM D 4791 0

GNP Type Properties Value

Carbon Content 99.82 (%)

Ash 0.18 (%)

Surface Area 250 m2/g

155 SMA evaluated in this study.

157 Table 2 Job mix formula (JMF) of the evaluated mixtures

¾ in. (19.0 mm) 100 96 - 100

½ in. (12.5 mm) 95 91 - 99

3/8 in. (9.5 mm) 73 69 – 77

No. 4 (4.75 mm) 32 28 – 36

No. 8 (2.36 mm) 19 16 – 22

No. 30 (600 µm) 15 13 – 17

No. 50 (300 µm) 13 11 – 15

No. 100 (150 µm) 10 8 – 12

No. 200 (75 µm) 7.3 6.6 - 8

Voids in Total Mix (VTM), % 3.5 2.5 - 4.5

NMAS (mm) 12.5

163 SMA mixes evaluated in this study are presented in Table 3.

Additive Che1 Che2 Org1 Asa1 GNP

Additive Type WMA WMA WMA ASA Nanomaterial

Dosage rate 0.6% 0.1% 2.5% 0.6% 3%

Mixing Temperature 150°C 150°C 150°C 163°C 150°C

Compaction Temperature 140°C 140°C 140°C 155°C 140°C

181 Figure 1. Experimental plan for the study

182 EXPERIMENTAL METHODS

183 Asphalt Binder Experiment

184 Fourier Transform Infrared Spectroscopy (FTIR) Test

210 Γtotal = ΓLW + ΓAB (1)

216 the surface (Little and Bhasin 2006).

224 Asphalt Mixture Experiment

225 Modified Lottman (ML) Test

249 sample dimensions to calculate the CT index as follows:

256 Pmax = peak load;

257 |m75| = post-peak slope;

259 𝑝65 = corresponding to 65% of 𝑃𝑚𝑎𝑥 at the post-peak stage;

264 the sample dimensions. er

278 Cost Effectiveness Analysis

291 Ci = Mixture unit cost (per ton); and

292 Ei = Expected performance.

301 𝐶𝑇𝑖 = mixture CT-index value; and

308 and $241.5, respectively. er

323 RESULTS AND DISCUSSION

339 from alkene or weak O-H stretching from alcohol, respectively.