Dataset of mechanical properties of coarse aggregates stabilized with traditional and nontraditional additives: Stiffness, deformation, resistance to freezing and stripping

The dataset derives from a thorough laboratory characterization of all existing stabilization technologies suitable for coarse-graded aggregates. They include two traditional binders (based on cement and bitumen) and eleven nontraditional binders (based on brine salt, clay, organic non-petroleum, organic petroleum and synthetic polymer). The dataset derives from four laboratory test operations: repeated load triaxial test performed both before and after exposing the investigated samples to ten freeze-thaw cycles, weight measurement of Marshall specimens during ten freeze-thaw cycles and a modified version of rolling bottle test. Repeated load triaxial tests assess the resilient modulus and the resistance to permanent deformation of both unstabilized and stabilized specimens. The mass loss of Marshall specimens expresses the susceptibility of each additive to lose its binding property when exposed to freezing action. The modified version of the rolling bottle test characterizes the propensity to stripping for each additive coating the aggregates subjected to mechanical stirring action. Given the surging necessity to improve the construction and maintenance operations for road pavements worldwide, this dataset containing information about several stabilization technologies can be very useful for transport agencies, contractors, industry and university researchers as well as companies manufacturing and supplying stabilization technologies.


Specification
Civil and Structural Engineering Specific subject area Road stabilization, Traditional stabilizers, Nontraditional stabilizers, Unbound granular materials, Pavement geotechnics, Freeze-thaw cycles, Repeated load triaxial test, Rolling bottle test Type of data Table  Image How data were acquired The data were collected performing the following laboratory tests: Repeated Load Triaxial Test (RLTT) both before and after the action of 10 Freeze-Thaw (FT) cycles, weight measurement of Marshall specimens during 10 FT cycles and a modified version of Rolling Bottle Test (RBT) for 14 different time intervals, namely 1 h, 2 h, 3 h, 4 h, 5 h, 6 h, 7h, 8 h, 10 h, 12 h, 14h, 16 h, 20 h and 24 h. The total number of samples created and tested was: 28 RLTT samples (2 replicates, particle size between 0 mm and 32 mm), 39 Marshall samples (3 replicates, particle size between 4 mm and 8 mm) and 588 RBT samples (3 replicates, particle size between 8 mm and 11.2 mm). Data format Raw Description of data collection The research comprised all existing types of traditional and nontraditional stabilizers for coarse aggregates. Creation and testing of RLTT, Marshall and RBT samples were based on the codes "EN 13286-7 Cyclic load triaxial test for unbound mixtures", "EN 12697-30 Specimen preparation by impact compactor" and "EN 12697-11 Determination of the affinity between aggregate and bitumen", respectively.

Value of the Data
• Connected with the surging necessity to improve the construction and maintenance operations for road pavements worldwide, this dataset derives from a thorough laboratory characterization and comparison of all the existing binder technologies suitable for stabilizing coarse-graded aggregates. • Considering the global relevance of road infrastructures, the dataset derived from an independent investigation of several stabilization technologies can become a very useful resource for transport agencies, contractors, industry and university researchers as well as companies manufacturing and supplying stabilization technologies. • The data can be used to evaluate and compare the mechanical behavior associated to all the types of existing binder technologies that can stabilize coarse-graded aggregates. Furthermore, the data can be interpreted according to various regression models and their analysis can indicate the directions for possible further laboratory or field tests. • As unpaved low-volume roads form most of the road infrastructures worldwide and are often in very poor condition due to lack of maintenance operations, the dataset formed in this single independent laboratory testing campaign represents a valuable resource for objective comparison across several stabilization technologies.

Objective
This work reports on the dataset "Mechanical properties of coarse aggregates stabilized with traditional and nontraditional additives: stiffness, deformation, resistance to freezing and stripping" ( https://data.mendeley.com/datasets/xvb2dtjdch ) [1] obtained by means of laboratory tests performed at the Department of Civil and Environmental Engineering (Norwegian University of Science and Technology, Trondheim, Norway) in 2020 and 2021. The major reason for performing such testing campaign has been to investigate and compare the stabilization potential of traditional and nontraditional binders that are used in road pavement engineering. Very few previous investigations have comprehensively compared the mechanical properties of both traditional and nontraditional technologies in a single independent study. The dataset has a size of 3 772 MB and revolves around the binder application to coarse-graded aggregates as thoroughly discussed in the two corresponding research articles [ 2 , 3 ]. The objective of this data article is three-fold: to describe how the data have been obtained in the laboratory, to explain the taxonomy adopted to store the dataset on the public repository Mendeley Data and to highlight how stakeholders (e.g., transport agencies, contractors, researchers, manufacturers of the stabilizers) can benefit from it.

Data Description
The dataset derives from a thorough investigation assessing the mechanical properties of all the existing binder technologies suitable for stabilizing coarse-graded aggregates [ 2 , 3 ]. The data were collected by performing four laboratory test operations: Repeated Load Triaxial Test (RLTT) both before and after the action of 10 Freeze-Thaw (FT) cycles, weight measurement of  Marshall specimens during 10 FT cycles and a modified version of Rolling Bottle Test (RBT). The content of the dataset includes both raw data and pictures for all the tested specimen ( https://data.mendeley.com/datasets/xvb2dtjdch ) [1] . As reported in Table 1 , thirteen stabilization technologies are investigated and represent both traditional additives (cement and bitumen) and nontraditional additives (brine salt, clay, organic non-petroleum, organic petroleum and synthetic polymer) [4][5][6][7] .

Data from repeated load triaxial tests
The data derived from RLTTs are reported in the folder "Data from RLTT before 10 FT cycles" and "Data from RLTT after 10 FT cycles". 14 subfolders contain the data for each stabilization treatment listed in Table 1 , two replicate specimens (denoted as "01" and "02") were tested for each additive binder. RLTT samples created using UGM materials were only tested before the exposure to 10 FT. For each specimen, three pieces of information are reported according to the nomenclature reported in Table 2 .
The content of all the spreadsheets is structured according to the same logic as described elsewhere [8] . Each spreadsheet contains five sheets ("Sequence 1", "Sequence 2", "Sequence 3", "Sequence 4", "Sequence 5") corresponding to as many RLTT loading sequences. Each loading sequence is made of six steps and their number is reported in column A. The time t since the sequence started, temperature T (namely room temperature), deviatoric pulse number and frequency f (fixed to 10 Hz) are listed in column B, C, D and E, respectively. In the RLTT device the specimen is subjected to a triaxial stress state achieved by means of a hydraulic piston acting vertically and pressurized water acting in all the directions. The deviatoric stress σ d exerted by the hydraulic piston is made of two components, a dynamic part ( σ d,dyn ) and a static part ( σ d,st ); the values of the former one are reported in column F and the values of the latter one (always approximately equal to 5 kPa) are shown in column G. In a similar way for the triaxial stress σ t , the dynamic part ( σ t,dyn , always approximately equal to 0 kPa) and the static part ( σ t,st ) are specified in columns H and I, respectively. Six Linear Variable Displacement Transformers (LVDTs) measure the deformations of the sample. Three LVDTs assess the vertical deformations classified as elastic components ( ε a,el,01 , ε a,el,02 , ε a,el,03 ) and plastic components ( ε a,pl,01 , ε a,pl,02 , ε a,pl,03 ); their values are reported in columns J, L, N and in columns K, M, O, respectively. Similarly, three LVDTs probe the radial deformations referred to as elastic components ( ε r,el,01 , ε r,el,02 , ε r,el,03 ) and plastic components ( ε r,pl,01 , ε r,pl,02 , ε r,pl,03 ); they are reported in columns P, R, T and in columns Q, S, U, respectively.
The two main mechanical properties assessed by RLTTs are the resilient modulus M R (definition reported in Section 2 ) and the resistance against permanent deformation. For an instance, the M R and the development of axial plastic deformation for all the RLTT specimens tested before the exposure to 10 FT is displayed in Figs. 1 and 2 , respectively, according to the number of load cycles N ; each colour corresponds to one of the five loading sequences ( σ t = 20 kPa, 45 kPa, 70 kPa, 100 kPa, 150 kPa).

Data from modified version of the rolling bottle test
The data derived from the modified version of the Rolling Bottle Test (RBT) are reported in the folder "Data from modified version of RBT" which contains two files, namely "Weight of RBT specimens.xlsx" and the folder "RBT sample pictures". The spreadsheet file "Weight of RBT specimens.xlsx" contains the weight of all the 588 RBT samples at three main stages: dried    Table 4 Designation of Marshall samples and corresponding images for each FT cycle.
FT cycle Specimen name Name of corresponding image  Table 6 Designation of RBT samples and corresponding images for each tested time interval.
Tested time Specimen name Name of corresponding image

Experimental Design, Materials, and Methods
The tested rock aggregate material derives from Vassfjell, Heimdal, Norway. The thirteen binder technologies are obtained from industrial producers and the categories these additives belong to are representative of all the existing commercial products suitable to stabilize coarse aggregates [2][3][4][5][6][7] . The testing campaign was performed in the laboratories of the Department of Civil and Environmental Engineering (Norwegian University of Science and Technology, Trondheim, Norway) to characterize the mechanical properties of unstabilized and stabilized rock aggregates.
As for road pavement engineering, the investigation and comparison of several stabilization technologies for construction aggregates is a very relevant topic to create more sustainable transportation infrastructures [22][23][24] . The major part of the roads worldwide comprises low-volume unpaved roads directly exposed to trafficking actions [25][26][27] suffering from poor maintenance and displaying several types of premature damage [28][29][30] . Considering the very scant amount of independent published literature regarding the comparison of different road stabilizers and the high relevance of the topic, the target audience is broad in nature. It encompasses transport agencies, contractors, industry and university researchers as well as companies manufacturing and supplying stabilization technologies [31] . In this regard, it may be worth mentioning that Step 1  20  20  45  60  70  80  100  100  150  100  Step 2  20  40  45  90  70  120  100  150  150  200  Step 3  20  60  45  120  70  160  100  200  150  300  Step 4  20  80  45  150  70  200  100  250  150  400  Step 5  20  100  45  180  70  240  100  300  150  500  Step 6  20  120  45  210  70  280  100  350  150  600 there is a myriad of proprietary road stabilizers available on the market globally, whose disclosed information is often subjective and rests on the laurels of the vendor's claims [32] . RLTTs were performed according to the Multi-Stage Low Stress Level (MS LSL) procedure; a RLTT is composed of thirty loading sequences as defined in the standard [33] . The analysed data can lead to the evaluation and comparison between all the binder treatments in terms of resilient modulus and resistance against permanent deformation. The particle size distribution used for the tests varied from 0 mm to 32 mm and thus corresponded to a typical road base layer; different quantities of binder ranging from 1% to 4% were mixed with 12 0 0 0 g aggregates for each RLTT sample [ 2 , 3 ]. The stress path applied is a combination of triaxial stress σ t and deviatoric stress σ d as reported in Table 7 ; each loading step applies 10 0 0 0 loading repetitions.
A sequence is interrupted after the completion of the six loading steps or if the axial permanent deformation measured by the axial LVDT reaches 0.5%. Given a constant value of σ t and a variation in the dynamic deviatoric stress σ d,dyn , the resilient modulus M R is determined as where ε a,el is the mean axial resilient strain measured by the three axial LVDTs.
Marshall specimens containing the different binder technologies were created using the mould and laboratory procedure traditionally adopted for asphalt mixtures, with 50 compaction blows per side [34] . Each sample comprised 850 g aggregates with uniform coarse gradation ranging from 4 mm to 8 mm. The Marshall specimens were exposed to 10 FT cycles and the weight loss was measured after every cycle. Defining the dried mass of the sample recorded initially ( M 1 ) and after ( M 2 ) the selected amount of FT repetitions, the mass loss ML MRS is assessed as and can be expressed as a percentage. The procedures adopted to perform each FT repetition were: specimen submersion in water (23 °C, 5 minutes), retrieval and release of water excess (23 °C, 5 minutes), freezing (-15 °C, 24 h) and thawing (23 °C, 24 h for RLTT samples and 40 °C, 48 h for Marshall samples) [ 2 , 3 ].
The RBT is originally a standardized procedure to evaluate the degree of adhesion between aggregate and bituminous binder covering the aggregate after the application of rotating and stirring actions [35] . The assessment is performed visually and therefore this gives room to possible unprecise results. As an improvement towards unequivocal interpretation, the testing campaign performed a modified version of RBT in the sense that the dried weight of a specimen covered by the additive was recorded before ( M 3 ) and after ( M 4 ) the testing while preserving the same rotating and stirring actions defined by the standard. In addition, the adjective "modified" also applies since several additive types were considered (and not only bituminous binder) and that 14 time intervals, namely 1 h, 2 h, 3 h, 4 h, 5 h, 6 h, 7 h, 8 h, 10 h, 12 h, 14h, 16 h, 20 h and 24 h, were evaluated (the code specifies to run the test only referring to 6 h and 24 h). The aggregates had size comprised between 8 mm and 11 mm; each RBT specimen was fabricated by blending 150 g of aggregates with 3% by mass of binder [2 , 3] . The mass loss ML RBT is assessed as and can be expressed as a percentage.

Ethics Statement
Ethical guidelines have been complied with during the collection of laboratory data. This work did not involve human subjects, animal experiments or data collected from social media platforms.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data Availability
Mechanical properties of coarse aggregates stabilized with traditional and nontraditional addi tives: stiffness, deformation, resistance to freezing and stripping (Original data) (Mendeley Data). Germany. Acrylates (Soil Control) kindly supplied by Sparks AS, Asker, Norway, Evonik Industries, Essen, Germany and Alberdingk Boley, Krefeld, Germany. Styrene butadiene (Eco-Pave H), acetate-A (Eco-Pave E) and acetate-B (Soil Sement Engineered formula 69PBc) kindly supplied by Midwest, Canton, USA. The above information regarding additive trade names and their suppliers are reported for informational purposes only and the inclusion of this information does not imply endorsement of any product or company. The findings and opinions reported are those of the authors and not necessarily those of the additive suppliers.