Experimental dataset on the residual performance of fiber-reinforced cementitious composite subjected to high temperature

The present dataset refers to the research article entitled “A multiscale investigation on the performance improvement of fiber-reinforced cementitious composites after exposure to high temperatures” [1]. Supplementary data on raw materials characterization, temperature recording, mass loss, water absorption, compressive strength, flexural behavior, pull-out response, fiber-matrix interface, and surface, microstructure and hardness of fibers are presented here. The continuous matrix was produced from cementitious grout containing Portland cement, sand, silica fume, superplasticizer, and water. The heating was carried out in an electric oven up to 260 °C. The bending tests was performed for fiber-reinforced cementitious composite (FRCC) with steel fiber contents of 1%, 3%, and 5% by volume, and for non-fibrous matrix. The pull-out test was performed using single fiber embedded in the matrix. The water absorption and axial compression tests was performed for non-fibrous matrix. The fiber-matrix analysis was performed from polished sections of fibers embedded in cementitious matrix. The fiber analysis was performed from steel fibers. The data refer to the residual properties after heating and slow cooling or to the reference condition without heating. The data can help in understanding residual performance of FRCC after exposure to high temperatures and may be useful for developing resilient building materials.

from steel fibers. The data refer to the residual properties after heating and slow cooling or to the reference condition without heating. The data can help in understanding residual performance of FRCC after exposure to high temperatures and may be useful for developing resilient building materials.
© 2022 The Author(s). Published by Elsevier Inc. This is an open access article under the CC BY license ( http://creativecommons.org/licenses/by/4.0/ ) Table   Subject Civil and Structural Engineering Specific subject area Construction materials Type of data Table  Image Graph Figure  How the data were acquired

Specifications
The data presented in this article were obtained from the following laboratory tests: - The age of 7 days was used for all tests. The high temperature submission was set at 260 °C. For FRCC, fiber contents of 1%, 3% and 5% by volume were used. Description of data collection The data were collected from the test of specimens, produced through the methodology of slurry infiltrated fibers for bending, cementitious grout (non-fibrous) for compression and water absorption, and single fibers for pull-out. The fiber characterization was performed from embedded fibers (microstructure and hardness) and individual fibers (surface). The fiber-matrix interface was evaluated from fibers immersed in the cement matrix. Half of the specimens were exposed to high temperature and tested in a residual condition, after slow cooling. The other half was tested without heating. Data

Data Description
The data presented in this dataset refers to the supplementary data from the article entitled "A multiscale investigation on the performance improvement of fiber-reinforced cementitious composites after exposure to high temperatures" . This section has been divided into subsections related to: Raw materials characterization (1.1); Ambient temperature record (1.2); Mass loss (1.3); Water absorption (1.4); Compressive strength (1.5); Flexural behavior (1.6); Pull-out response (1.7); Fiber microstructure (1.8); Fiber hardness (1.9); Fiber surface (1.10); and Fibermatrix interface (1.11). The no-heat condition was defined as AM and the data obtained after submission to high temperature of HT. The raw and analyzed data are available in Mendeley Data (see Ref [ 2 , 3 ].). Tables 1-5 show the detailed characterization of the raw materials used. All data were obtained from manufacturers and suppliers. Table 1 shows the physical-chemical characteristics of Portland cement, as well as tests performed and methodology adopted, with 0.57% ( < 1.0) of insoluble residue, 3.75% ( < 4.5) of loss of ignition, 1.48% ( < 6.5) of magnesium oxide, 2.73% ( < 4.5) of sulfur oxide, 2.61 ( < 3.0) of carbon     dioxide, 4.507 cm2/g ( > 30 0 0) of surface area (Blaine), 0.06% ( < 6.0) of sieve residue #200, 0.87% of residue on the #323 sieve, 30.4% of normal consistency water, 142 min of start setting ( > 60), 191 min of end setting ( < 600) and 0.00 mm of hot expandability ( < 5.0). Table 2 shows the characteristics of steel fibers. The fiber nomenclature was designated as (FF3) according to the manufacturer, the material is low carbon steel, the manufacturing method was cold-forming (drawing), the section shape is circular with length (l) of 50 mm and diameter (d) 0.75 mm, form factor equals to 67 (l/d) and tensile strength (limit of strength) of 1200 MPa. Table 3 shows the chemical composition of silica fume, which had 91.04% silicon dioxide, 0.10% aluminum oxide, 0.70% iron oxide, 1.10% calcium oxide, 1.50% magnesium oxide, 0.39% sodium oxide, 4.40% potassium oxide, 0.16% sulfur trioxide and 0.61% loss of ignition. Table 4 shows the properties of the superplasticizer, which has a homogeneous viscous liquid appearance, siena color, 47.8% solid residue content, 3.1 pH and ≤0.15% chloride content. Table 5 shows the properties of the water used in the mixture of the fluid mortar, which had a colorless and clear appearance, at room temperature around 25 °C, 1.09 mg/L of chloride and 0.8 mg/L of fluoride.   Tables 6 and 7 show the raw data for mass loss and relative mass loss, both due to high temperature submission. The average mass loss of the six specimens was 19.17 g after drying in an oven and 7.50 g after heat treatment. The total mass loss was 26.66 g. The average relative mass loss was 4.88% after oven drying, 2.02% after heat treatment and 6.90% in total.   Tables 8 , 9 and 10 show the data from the water absorption test. Note: M dry -dry mass; M ssd -saturated mass dry-surface; M sub -submerged-saturated mass. Table 9 Data of water absorption, dry bulk density, saturated bulk density and porosity for the unheated condition. Note: WA -water absorption; DBD -dry bulk density; SBD -saturated bulk density; PR -porosity.

Table 10
Data of water absorption, dry bulk density, saturated bulk density and porosity after high temperature. Note: WA -water absorption; DBD -dry bulk density; SBD -saturated bulk density; PR -porosity. Table 8 shows the weighing data of the six specimens used, in the following ways: dry mass in oven or after HT (M dry ), saturated mass dry-surface (M ssd ) and submerged-saturated mass (M sub ). For the AM condition, the M dry data are comprised between 363.20 g and 380.40 g, M ssd between 383.17 g and 399.25 g, and M sub between 191.67 g and 209.24 g. For the HT condition, the M dry data are comprised between 355.52 g and 373.53 g, M ssd 375.11 g and 393.53 g, and M sub between 183.34 g and 202.85 g. Table 9 shows the water absorption (WA), dry bulk density (DBD), saturated bulk density (SBD) and porosity (PR) data for the unheated condition. The WA data are between 4.96% and 5.50%, average of 5.14%. DBD data are between 1.90 g/cm 3 and 2.00 g/cm 3 , average of 1.96 g/cm 3 . The SBD data are between 2.00 g/cm 3 and 2.10 g/cm 3 , average of 2.06 g/cm 3 . PR data are between 8.98% and 9.44%, average of 9.13%. Table 10 shows the WA, DBD, SBD and PR data after high temperature submission. The WA data are between 4.70% and 5.89%, average of 5.43%. DBD data are between 1.85 g/cm 3 and 1.96 g/cm 3 , average of 1.92 g/cm 3 . The SBD data are between 1.96 g/cm 3 and 2.07 g/cm 3 , average of 2.03 g/cm 3 . PR data are between 8.44% and 10.30%, average of 9.44%.

Pull-out response
Fig . 5 shows the single fiber pull-out response. Two load-slip (P-s) curves are shown for each condition, each curve from one specimen. The blue data represents the AM condition and the red data the HT condition. PL means pull-out. Table 16 shows the pull-out test parameters. The maximum load was between 412.75 N and 395.40 N without heating, and between 450.75 N and 479.01 N after heat treatment. The displacement in peak load was on average 1.2364 mm for AM and 1.3415 mm for HT. The pull-out energy was between 3140.53 N.mm and 4044.96 N.mm for AM, and between 3490.56 N.mm and 4293.77 N.mm after heating.        Fig. 10 shows an image obtained by OM during the nanoindentation test, in which the typical microstructure of steel fibers and the impressions left by the indenter can be seen. Parallel bars were used as a reference. Table 17 shows the hardness data obtained from six samples tested by nanoindentation. The fibers as received (AR) was an average hardness of 275 MPa. After flexural testing without heating, the average of hardness was 323 MPa. After submission to high temperature, the average of hardness was 277 MPa.

Fiber surface
Figs. 12 , 13 and 14 show images obtained by OM of the fibers surface, as received, after bending test and after heat treatment, respectively. The x100 and x400 magnifications were used.    Fig. 15 shows a specimen prepared for fiber-matrix interface analysis by OM, after sanding and polishing. The fibers sections are the bright spots indicated.

Fiber-matrix interface
Figs. 16 , 17 and 18 show images obtained by OM of the fiber-matrix interface, as received, after heat treatment, and after heat treatment with previous preparation of the samples, respectively. The x100 and x400 magnifications were used.

Experimental Design, Materials and Methods
The experimental phase was divided into three major stages. The first consisted in the selection and characterization of materials, as well as in the development of the cementitious composite. In the second stage, the specimens were produced. In the third stage, the specimens were tested to investigate the bending and compression behaviors, water absorption, porosity, density, fiber-matrix interface and surface, microstructure and hardness of the fibers. All tests were performed at the 7 days, before and after heat treatment. For additional information see related article in Ref. [1] .

Molding of specimens
For molding the specimens, the slurry infiltrated fiber method was used, which consists of placing the fibers in the mold and then casting a fluid mortar, which fills the empty spaces between the fibers [4] . Prismatic mold with dimensions of 40 × 40 × 160 mm was used for bending. The fibers were placed in the molds according to the percentages of 1%, 3% and 5% by volume; finally casting a fluid mortar. The pull-out specimens were produced using cylindrical molds with dimensions of 50 × 100 mm. One of the hook-end of the fibers was sectioned to facilitate fitting into the test equipment. The other end of the fiber was embedded in the matrix. The length of the fibers embedded in the matrix was equal to 15 mm. The same cylindrical molds were used for compression and water absorption. The specimens for microstructural investigation and fiber characterization are described below.
After demolding within 24 h of mixing the materials, the specimens were placed in submerged curing in water saturated with calcium hydroxide, where they remained for seven days. After curing, the specimens were placed in an oven for a period of 24 h, at an average temperature of 65 °C, to remove excess water and carry out the tests. After the oven-drying, the specimens were removed and left at room temperature. Finally, half of the specimens were tested without heating, while the other half was subjected to high temperature and subsequent testing.
To produce cementitious grout, the literature suggests a water-cementitious materials ratio of 0.30; silica fume-cement ratio of 0.20; and sand-cementitious materials ratio of 1.23 [5] . Thus, the dosage was adjusted for the raw materials. The sand was dried in an oven for 24 h, at an approximate temperature of 105 °C, to completely remove the moisture. Then, the sieving process was adopted to obtain grains with a maximum dimension of 1.2 mm.
The variables proposed in this study were fibers content (1%, 3% and 5%) and thermal condition (with or without heat treatment). Thus, to differentiate the composites, a nomenclature was set up as follows: fiber percentage (F0, F1, F3 and F5 for 0%, 1%, 3% and 5% of fibers, respectively) and presence (HT) or absence (AM) of heat treatment.

Heat treatment
The heat treatment (HT) of the specimens was performed in an electric oven, to simulate a controlled condition for temperature variation with time. The temperature measurement was made using thermocouples attached to the specimens and the use of a glass wool thermal blanket. The other end of the thermocouple was connected to the digital meter, which converts the electrical signal into temperature values. To perform the heat treatment, the specimens were inserted in the oven along with sensors to measure the temperature, so that the heat flow occurred equally over the entire surface. Heating in the range of 200 °C to 300 °C promotes maintenance of properties. However, at higher temperatures the properties are deteriorated [6] . Around 400 °C, the propagation of micro-cracks in the matrix occurs, compromising the microstructure. At 560 °C, the decomposition of the structures of calcium silicate hydrate (C-S-H) occurs, porosity increases even more, and the microstructure is very cracked, which causes a rapid drop in strength. Thus, a temperature close to 260 °C was chosen to study the properties of FRCC in this specific condition and to evaluate the increment of residual strength.

Mechanical characterization of cementitious composite
The 4-point bending test was adopted to evaluate the mechanical behavior of FRCC, using a universal testing machine (UTM) from the manufacturer EMiC, located in the Department of Transports Engineering of CEFET/MG. The midpoint deflection was accurately measured by a deflectometer while the loading was measured by a load cell with a capacity of 300 kN. The resulting load and deflection values were captured and analyzed by TESC software, also from EMiC, generating a complete database. These data were collected and analyzed, resulting in the loaddeflection (P-d) diagrams for each specimen. The bending stress were calculated using Eq. (1) , considering the precepts of the bending test and considering the acting efforts, as follows: Where σ 4 P F lex is the 4-point bending stress, P is the load, d is the span ( d = 132 mm), b is the width and h is the height ( b = h = 40 mm).

Toughness
Toughness was calculated considering the area under the load-deflection curve (P-d). The ASTM C1609/C1609M [7] standard suggests the adoption of notable points to calculate the toughness when the deflections are equal to L/600 and L/150, where L is the distance between the supports. Kim, Naaman and El-Tawil [8] indicate the adoption of the other point (L/100) to fully characterize the behavior of FRCC using different fibers. In view of the conditions, the characteristics of the composite, the different fiber content and the extension of the P-d curves, it was necessary to adopt another notable point corresponding to the loads of 0.9P (before and after peak load), P (load of peak), 0.7P (post-peak) and 0.5P (post-peak). Therefore, the toughness points were T 0.9P,bp , T P,pc , T 0.9P,pp , T 0.7P,pp and T 0.5P,pp .

Single fiber pull-out test
The present study adopted a specific setup for single fiber pull-out test, coupled to a universal testing equipment (See related article in Ref. [1] ). A load cell with a capacity of 20 kN and accuracy of 0.001 N was used to measure the load. The specimen was attached by an adapter. A clamp was used to pull-out the fiber. To measure the fiber displacement accurately, an extensometer with accuracy of 0.001 mm was used. The adopted displacement rate was 10 μm/s, in agreement with Abdallah, Fan and Cashell [9] .

Compressive strength of cementitious matrix
The mechanical strength of the cementitious matrix was evaluated by the axial compression test. Six cylindrical specimens were cast, half of which underwent heat treatment. All specimens were tested in compression using the universal testing equipment.

Fiber-matrix interface analysis
The fiber-matrix interface analysis was performed using a Kontrol optical microscope (MO), IM713. The fiber-matrix interface was studied using polished sections by MO to reveal the contour of the interface.

Characterization of steel fibers
The steel fibers were characterized under three different conditions. The first refers to the fiber as received, collected in the plastic container for conditioning materials. The second condition represents the fiber after the bending test, collected from the fractured specimen. The third condition represents the fiber subjected to heat treatment and after bending test. All fibers were chosen randomly, to obtain a reliable sample. Then, the fibers were divided in half with cutting pliers and inserted in silicone molds for cold mounting, which was done with transparent acrylic resin. The samples were prepared by sanding and polishing, then chemically attacked (Nital 3% reagent) to reveal the microstructure of the steel. The samples were taken under the Kontrol optical microscope, IM713, to obtain images of the microstructure of the steel. Finally, the samples were placed in the Shimadzu microdurometer, HMV, to evaluate the Vickers microhardness of the steel fibers. The microhardness test was performed with a load application of 1961 N (HV 0.2) and a holding time of 15 s.

Ethics Statements
This work adheres to ethical publishing standards and does not include human studies, 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
Raw data on fiber-reinforced cementitious composite subjected to high temperature (Original Data) (Mendeley Data) Analyzed data on fiber-reinforced cementitious composite subjected to high temperature (Ori ginal Data) (Mendeley Data).