Basaltic rock characterization

The characterization of basaltic rocks is widely documented in the literature. For example, ^{Borazjani, Ghiasi-Freez, and Hatampour (2016)} reported that the use of petrographic thin sections is the first step to understand the characteristics of a rock. In this sense, ^{Paulsen et al. (2013)} established that thin sheets are principally used for the identification of minerals and aspects such as fractures, mineral zoning, and rock textures, as well as to concisely identify the formation of rocks and their classification. Samples of basaltic rock pieces reduced to a thickness of 30 microns (μm), with their polished surfaces, were used to determine that the aggregate used in permeable concrete is a basalt. The thin sheets were observed through a petrographic microscope model MOTIC BA310, such samples are analyzed using transmitted light for the determination of mineral content, rock texture, and mineral alteration. It was observed that basalt is mainly composed of pyroxenes and plagioclase rich in Ca, its texture is aphanitic and its structure is vesicular as illustrated in Figure 1.

Figure 1 Microscopy of thin sections of basaltic rock at a scale of 0.5 mm. 

Specimens of Permeable Concrete

The dosage method was based on the volume of the paste required to join the aggregates and maintain the design void content according to the American Concrete Institute recommendation (^{ACI, 2010}). Permeable concrete cylinders were prepared using a mixture of CEMEX CP30 compound composite Portland cement, the maximum nominal aggregate size of 9.5 mm (obtained from the La Cañada material bank located in the city of Querétaro, Mexico), and water. Table 1 summarizes the proportions of the above-mentioned materials used in 1 m³ of the mixture. Such materials were combined in a concrete mixer and sneaked into cylindrical molds.

Once mixed, the cylinders were covered and cured for 7 days in a controlled environmental chamber with 50 % humidity at 23 °C, before testing the absorption batches. The dimensions of the cylinders were 50x100 mm. Tests were conducted to determine the compressive strength of cylindrical specimens of permeable concrete according to ASTM C39 as shown in Figure 2 and shown in the lines below in Table 5 (^{ASTM, 2015}). To decrease the experimental variability, every cylinder was prepared in the same batch of concrete using similar sources of cement and aggregates. The process of the test was designed to determine if the permeable concrete may be able to eliminate heavy metals and measure the level of retention that concrete may have at different initial concentrations of As, Cd and Zn.

Figure 2 Simple compression Permeable concrete cylinders. 

Batch absorption method

Three solutions were prepared with standard reagents of As, Cd, and Zn (^{ASTM, 2015}). The standard solutions had a concentration of 1000 mg l-1. From each standard solution, dilutions were made to obtain uniform concentrations of Cd and Zn of 10, 30, 50, and 90 mg/l, respectively for each contaminant. This batch test was performed by immersing the concrete cylinders in a 1-L beaker containing the solutions of As, Cd and Zn for 96 h. All beakers were washed with acid in a 5 % nitric acid solution for 12 h and rinsed with distilled water before the test (see Figure 3).

Figure 3 Deposition of heavy metal precipitates in concrete in a batch reactor. 

After 96 h, the concrete cylinders were removed from the solution, drip-dried, lightly covered by a cellulose-based plastic wrap, and finally dried inside the environmental chamber at 50 % humidity at 23 °C. To avoid cracking and preserve surface reactions, the analysis was subsequently performed on the total reflection X-ray fluorescence spectrometer (TXRF).

Total Reflection X-Ray Fluorescence (TXRF)

In this investigation, a TRFX spectrometer (S2 PICOFOX, “Buker” AXS Microanalysis GmbH) was used. The instrument calibration was improved with a standard of chemical elements concentration used in this research such as As, Cr, Cd, and Zn of 1 mg/l. The time for data acquisition for each sample during the calibration process was 1 000 s. Each spectrum was analyzed. This device has a low-power X-ray ceramic metal tube cooled with a molybdenum lens, working at a maximum power of 50 W, and a silicon drift detector (SDD) without liquid nitrogen with the compact S2 PICOFOX TXRF instrument (^{Bruker, 2007}). The technical specifications according to the Bruker manufacturer of the S2 PICOFOX TXRF instrument are summarized in Table 2.

Sample Preparation for Total Reflection X-Ray Fluorescence Analysis

Polished spectrosil (synthetic) quartz discs have 30 mm in diameter and 3.0 mm of thickness. They were used for sample preparation. Before starting the test, a cleaning protocol was followed, comprising several stages. First, a cleaning solution RBS 50TM (SigmaTM, Germany) 5:50 to ultrapure water (18 MΩ cm, Milli-Q, Millipore) was used. Then, nitric acid solution at 10 % under heating, rinsing with ultrapure water, drying in an oven at 80 °C was applied. Finally, the siliconization of the surface was performed with a solution of silicon in isopropanol (ServaTM, Germany) so that the surface becomes hydrophobic and causes a small evaporation point. The successful removal of the previous sample was verified by carrying out blank test measurements, the absence of element peaks, apart from silicon (Si), quartz, and molybdenum (Mo) from the source was observed. Figure 4 illustrates the process of preparation for TXRF analysis samples, where Figure 4(a) shows the addition of silicone in isopropanol and Figure 4(b) illustrates the preparation of the sample including 10 μL solution of the concrete sample exposed to Arsenic, Zinc and Cadmium metals for TXRF analysis.

Figure 4 Preparation of the samples for analysis with S2 PICOFOX: (a) Addition of silicone in isopropanol Y; (b) sample preparation for TXRF analysis (^{De-La-Calle, Cabaleiro, Romero, Lavilla, & Bendicho, 2013}). 

Finally, the quantitative analysis was performed by internal standardization of the analysis samples. Gallium was chosen as the internal standard due to its low abundance in the type of samples studied and good sensitivity in the determination of TXRF.

Total Reflection X-Ray Fluorescence Measurements

The instrument calibration test was performed with a multi-element standard containing a certified concentration of Ar, Cr, Cd, Zn, and Mr of 1 mg/l (^{Bruker, 2011}). The data acquisition time for each sample measurement was 1000 s (live time). The spectra were processed with the S2 PicofoxTM software (^{Bruker, 2011}). Subsequently, the lines of the elements to be found in the response spectrum are identified to verify which elements are in the sample to be analyzed. Finally, the concentration of the elements was calculated from Eq. (1) using the intensities and relative sensitivities of the elements to be analyzed, which are dimensionless values that indicate the maximum intensities of the respective elements about the quantity of sample (^{Bruker, 2011}).

Where Ci is the concentration of the element to be analyzed; CIS is the concentration of the internal standard; N_i is the number of the net pulse within the measurement spectrum of the element to be analyzed; N_IS the number of the net pulse within the measurement spectrum of the internal standard; S_i the relative sensitivity of element i; S_IS is the relative sensitivity of the internal standard element.

Energy-dispersive X-ray Fluoresce (EDXRF)

A commercially available benchtop EDXRF spectrometer (S2 PUMA Carousel, Bruker AXS, MbH, Germany) was used in the present study. This instrument is equipped with a Pd target t X-ray tube (max. Power 50 W) and a XFLASH LE Silicon Drift Detector (SDD), ultra-thin beryllium window (0.3 μm thickness) with a resolution lower than 129 eV at Mn-Kα line for a count rate of 100,000 counts-per second. In this LE configuration of SDD detectors, the intensities for Na K-alpha and Mg K-alpha are, respectively close to 8 and 4 times higher than the intensity recorded by conventional SDD detectors. The instrument is also equipped with nine primary filters that can be used in front of the tube before the X-ray beam impinges the sample surface to improve measuring conditions for the elements of interest and it can operate under vacuum conditions. The technical specifications according to the Bruker manufacturer of the S2 PUMA EDXRF instrument are summarized in Table 3.

The software used to control the equipment, build the calibrations, and perform the data treatment was SPECTRA EDX (Bruker AXS, GmbH, Germany). Such software can perform the full line profile fitting, deconvolutions when lines overlap, intensity corrections for inter-element effects, and full-quantitative routines.

Arsenic absorption lots with permeable concrete and concentration measurement with spectrometry (EDXRF)

The preparation of concrete and basalt specimens followed a methodology similar to that of thin sheets for petrography. First, the manufacture of concrete cubes was carried out with the proportion of the mix design for permeable concrete, except that, without the added basalt stone, this was to be able to perform the analysis separately for concrete and basalt. A cutter of MK-500 material was used to give the samples a regular shape. Once the solution was prepared at a concentration of 90 mg l-1 of As in water, the suspension of concrete samples of 5 cm x 5 cm sideways with a thickness of 5 mm and basalt with measurements of 3 cm x 2 cm was taken, in Falcon centrifuge tubes up to a height of 10 ml of As solution. The test time was 8 days at a constant temperature of 25 ° C in an oven. After 8 days, the samples were removed from the solution, to subsequently perform the analysis on the S2 PUMA spectrometer with the EDXRF technique. Sampling was performed at 2, 4, 6, and 8 days, respectively for both liquid basalt and concrete samples. The solid samples were dried in the oven at a temperature of 25 ° C for 24 h before taking their measurement. For the liquid samples, the process was different, since for its preparation plastic cups were used to store 7.0 g of liquid sample already contaminated with As, at the bottom of this glass a transparent Mylar membrane was used. After 8 days of taking and preparing the samples, all samples were measured. For this, a Bruker S2 PUMA brand device was used to measure up to 11 samples. Plastic cups containing liquid samples were placed in the equipment carousel. Once all the samples were mounted on the S2 PUMA, the type of sample (solid or liquid) and the characteristics of the sample are selected. The sample ID was introduced, then the diameter of 3.5 cm of the plastic cup, with a Myler membrane with a thickness of 3.60 microns (µm), to then perform the quantification with the equipment with a duration of 5 minutes per sample. The response spectrum of the equipment is shown in Figure 5.

Figure 5 Response spectrum for As 90 mg l-1 solution. 

Permeable Concrete Mix Design for Contaminant Absorption Tests

A mixture of simple permeable concrete containing a coarse aggregate of basaltic stone, composite Portland cement, and water was used for the experiment. The maximum nominal size of the aggregate was 9.5mm (3/8”). The specific gravity of the aggregate was 2.90, and the absorption was 1.5 % (^{ASTM, 2015}). All mixtures were designed for empty spaces of 20% and were produced with a water-cement ratio of 0.40. Table 4 summarizes the mechanical properties of basalt used in the permeable concrete mixture.

Additional information about the resistance of the samples of concrete used in this research is presented in Table 5.

The results obtained are within the expected range of 140 kg/cm2 to 180 kg/cm2 after 28 days of setting (^{ACI-522R.6, 2006}), which are appropriate for the use of this permeable concrete as drainage infrastructure on a land route, or streets or roads with little traffic.

Heavy metal retention with permeable concrete

This study has been carried out to determine the efficiency of As, Zn, and Cd removal from the synthetic aqueous solution using permeable concrete with basaltic rocks. For batch tests with permeable concrete, the removal efficiencies were as follows: As 55-81 %, Zn 75 % -96 %, and Cd 50-89 %. As the concentration of metals increased, retention efficiencies decreased, except for Cd, which had better disposal levels at a higher concentration. The percentage of removal of individual heavy metals for each solution is shown in Figure 6. On the other hand, Figure 7 and Figure 8 are grouped by initial concentration, final concentration, and efficiency percentage. In general, all samples showed high levels of metal removal. For concentrations less than 10 mg/l of As and Zn, it is observed a higher absorption efficiency, and for concentrations greater than 10 mg/l Cd, higher levels of heavy metal retention are noted.

Figure 6 Arsenic retention efficiency with the use of permeable concrete. 

Figure 7 Cadmium retention efficiency with the use of permeable concrete. 

Figure 8 Zinc retention efficiency with the use of permeable concrete. 

Other elements that were found in the measurement of the solutions with the total reflection X-ray fluorescence method and the S2 PICOFOX equipment are presented in Table 6.

Based on the previously presented results, Zn, As, and Cd are attracted by potassium, calcium, and phosphorus, which causes ionic exchange between these elements. It was also observed that there was a process of decalcification of the permeable concrete caused by the presence of As, Zn, and Cd in the solution, leaving basalt exposed in some areas of the cylinder. In this sense, Figure 9 illustrates this phenomenon after exposing the concrete cylinder for 96 h with heavy metals.

Figure 9 Decalcification of permeable concrete exposed to As, Cd, and Zn. 

Subsequently, when analyzing each contaminant separately, a general mixture was made with Cd, As, and Zn in the percentages where the highest retention efficiency was obtained for each of them. Next, Table 7 shows the initial and final concentrations of the elements found in the complete mixture of all contaminants.

As well as what was mentioned by ^{Zhao and Zhao (2014)}, where the analysis is performed for the eficence of lead and zinc removal of basalt materials, Limestone and limestone with basalt, it is determined that basalt can retain up to 80.50 % Zinc, which for analysis performed for this research was up to 96 %, thus increasing the retention for Zinc by 15.50 %.

Arsenic absorption lots with concrete and basalt with concentration measurement with spectrometry (EDXRF)

After the eight days of proof of exposure of the material to the solution with As for concrete and basalt and after performing the measurements with the S2 PUMA equipment, it was verified that the calculation of the solution for 90 mg l-1 of As was correct and thus to obtain the initial concentration with which reference was made to know the efficiency of each of the materials. The measured concentrations for each of the test days are shown in next Figure 10.

Figure 10 Measurement of arsenic concentration (lower line) and retention efficiency (upper line) with concrete. 

As illustrated in Figure 10, the concrete after having been exposed to the solution with As for 8 days was able to achieve 99.75 % retention, which proves that this material has excellent characteristics for the retention of As dissolved in water. The measurements of concentrations for basalt are shown in Figure 11.

Figure 11 Measurement of arsenic concentration (lower line) and retention efficiency (upper line) with basalt. 

It can be observed in Figure 10 and Figure 11 that basalt has a lower percentage of As retention concerning concrete of up to 14.50 % versus 99.75 % of concrete, so it can be verified that the material has higher qualities for retention is concrete.

Concrete and basalt as an ion exchanger

According to the Cemex building guidelines (Cemex, 2015), the composition for the manufacture of cement consists mainly of limestone, clay, and plaster in different proportions. With the results obtained from the analysis with the EDXRF technique, it was possible to extract the chemical composition of the cement and basalt reference before being exposed to As for 8 days. Table 8 shows the chemical composition of cement and basalt, these data were obtained with the S2 PUMA Bruker equipment.

As summarized in Table 8, concrete and basalt contain some similar chemical elements. However, some chemical compounds such as calcium oxide, which was the most released in the TXRF tests, present a higher percentage compared to basalt. In addition to this, these materials are very similar to the chemical composition that contains the Zeolites and has been more studied for use in water purification filters because of their high ion exchange capacity.

In some other investigations as the one documented by ^{Ponce-Lira et al. (2017)} it has been shown that materials containing aluminum oxide (Al2O3), ferric oxide (Fe2O3), manganese oxide (MnO), and silicon oxide (SiO2), as well as phosphates and silicates, have ion-exchange properties, these constituent elements of concrete and basalt are involved in the removal of As, Cd and Zn used in this investigation

The presence of calcium oxide (CaO) in the cement, interferes with the removal of metal ions, this element directly influenced the removal of As in the concentrations used in the batch absorption test performed in this paper. In Table 9 and Table 10, the percentages obtained from the chemical composition for solid specimens of concrete and basalt are presented.

The measurements of the chemical compositions obtained before and after the contamination of As for basalt and concrete with the EDXRF technique reveal that this heavy metal is retained in the solid part, due to the ionic exchange that exists through the elements that are related to As. This result could not be obtained with the TXRF technique.