3.1. Distribution Changes of Temperature, Humidity and Soil Gas Benzene Concentration
In order to simulate and monitor the specific change of temperature in space during the treatment process, 27 thermocouples were used to continuously monitor the temperature of the soil at depths of 5, 20, and 35 cm in the lower right part of the system. Data was recorded by the data terminal every 30 s. After screening, five typical time points (0, 90, 240, 300, 1200 min) were intercepted. Taking the left front end of the box as the origin, the monitored temperature in the lower right area was processed with central symmetry, and the isotherm diagram of each time point on the corresponding plane in the whole box was drawn to describe the change of temperature field during the experiment. The result is shown in
Figure 2.
To describe the distribution of benzene content in soil gas over time, according to the results of soil gas measurement, the contour map of benzene concentration in soil gas was drawn in three layers (depth 5, 20, 35 cm) with the left front end of the box as the coordinate origin. The result is shown in
Figure 3.
In addition, the benzene concentration in the air above the device was measured at the same time, as shown in
Figure 4. The concentration above the device was basically proportional to time in the heating stage, and remained unchanged after reaching about 7500 mg∙m
−3, it decreased rapidly after extraction. This was consistent with the benzene concentration in the exhaust gas. However, it was worth noting that the benzene concentration in the gas above the device was slightly lower than that in the exhaust gas.
Before treatment, after repeated mixing, the soil moisture was almost uniform, which was 36%. After 24 h heating and 20 h extraction, the soil was divided into three layers (depth 5, 20, 35 cm), and the humidity was detected according to the soil sampling position. After treatment, the average humidity of the soil at the depths of 5, 20 and 35 cm was 20.28%, 29.84%, 32.56%, respectively, the overall average humidity was 29.39%. The treatment process took away 6.61% of the soil moisture, which was 14.54 kg. The contour map of humidity was drawn with the left front end of the device as the origin, and the result was shown in
Figure 5. The soil moisture after treatment was low in the surface layer, high in the bottom layer, low in the middle, and high around.
In this research, 75 treated soil samples were collected in three layers. The benzene content in soil was detected by gas chromatography. The detection results showed that most of the sites were not detected (the detection limit was 1.9 μg∙mg
−1). The highest detection content was 7.3 μg∙kg
−1, which was 0.04% of the initial soil concentration. Similarly, the contour map of benzene content in the treated soil was drawn with the left front end of the device as the origin. The result is shown in
Figure 6. The distribution of benzene residues in treated soil was low in surface layer, high in bottom layer, high in center, and low around.
Soil has high specific heat capacity; it is a poor conductor of heat. In T-SVE, direct thermal contact heating with resistance is usually a slow process [
24]. At the beginning, the soil was just stirred with water, the overall temperature was relatively uniform, about 28 °C. After the system was heated up for 90 min, the temperature of the middle soil layer around the heating rod began to change obviously. At 240 min, the temperature of most sensors reached the predetermined 80 °C. After 1200 min, the highest temperature in the middle soil layer reached 110 °C, then the temperature and its distribution basically remained in this state. On the plane, the temperature distribution was obviously higher around the heating rod and lower around it. It was worth noting that at about (200, 100) and its four symmetrical positions, although the position is closer to the heating rod, the temperature was lower most of the time, especially after the ventilation starts at 240 min. This is because the ventilation pipe is near there, the ventilation pipe has obvious influence on its heat conduction nearby. After the start of ventilation, cold air enters the soil continuously. This shows that the influence of ventilation on temperature is obvious. However, the temperature at the center extraction pipe did not drop significantly due to the start of extraction. This may be because extraction concentrates the energy of the system to the middle area through thermal convection. The box is provided with four heating rods with symmetrical centers. Therefore, on the plane, there are four obvious high temperature centers. In the vertical direction, because the heating rod is 5 cm deep and 25 cm long, the temperature field is obviously high in the middle soil layer and low in the top and bottom soil layers. The temperature of topsoil is slightly lower than that of subsoil because of contact with air above and condensed water on the tank cover.
After the soil was filled, the benzene concentration in the soil gas was uniform, 1000 mg∙m
−3. At the beginning of the heating stage, the benzene concentration changed obviously, the overall concentration almost doubled every 20 min. At 90 min, the central concentration reached about 8500 mg∙m
−3, and the corner concentration was also higher than 4000 mg∙m
−3. At this moment, the distribution of benzene concentration in the middle soil layer was almost the same as that of temperature. As the diffusion of soil gas is obviously faster than the transfer of temperature in soil, although the temperature of surface and bottom soil is still low, the benzene concentration in soil gas is almost the same as that in the middle layer. At 180 min, the concentration reached its peak, the distribution of high concentration in the center and low concentration around it reached the extreme meanwhile. The highest concentration in the center reached 11,200 mg∙m
−3, with an average value of 7521.3 mg∙m
−3. At this time, the distribution was slightly lower in the middle and higher around on the plane. Considering the rapid diffusion of soil gas, it was speculated that the benzene around the heating rod was almost completely thermally desorbed at 180 min, but the benzene in the surrounding soil was still desorbing. At 240 min, the overall concentration decreased to about 7500 mg∙m
−3, with an average value of 7629.4 mg∙m
−3, slightly higher than that at 180 min. This indicated that the thermal desorption between these two times was inefficient or nearly completed, and the change of soil gas in this stage was mostly caused by the free diffusion of soil gas. Li [
25] removed benzene from the soil through a similar stirable steam-enhanced SVE device, and the properties of the tested soil were similar to this study. The study found that at 80 °C, adding mineral powder and stirring, with a gas flow rate of 6 L∙min
−1, the removal rate of benzene reached 90% within 2 h. This was basically consistent with the change of soil gas concentration in the early stage of heating in this treatment.
After the extraction started, the benzene concentration in the system decreased rapidly in a very short time. As a result, in the contour map of 250, 260 and 400 min, the scale of color scale had to be changed in order to show the distribution change of concentration obviously. After extraction for 10 min (total process of 250 min), the concentration of all monitoring points was lower than 1000 mg∙m
−3. After 20 min (260 min), the whole concentration decreased by two orders of magnitude, the distribution in plane space conformed to the law of central extraction, it had no obvious correlation with the distribution of temperature. It was worth noting that the concentration in the upper right corner of the device was abnormally high. Combined with the analysis of inlet flow in
Figure 7, it was deduced that the ventilation condition of the upper right ventilation well was inferior to that of other ventilation wells. At 400 min, the system concentration was relatively uniform, all below 100 mg∙m
−3, with the central concentration about 40 mg∙m
−3 and the lowest concentration reaching 14 mg∙m
−3, it was basically consistent with the monitoring situation of tail gas benzene concentration in
Figure 8. In the extraction stage, there was no significant difference of benzene concentration in the vertical direction, and a slow decline in the upper middle layer and a rapid decline in the middle bottom layer. Liao [
26] used a two-dimensional columnar experimental device to perform nonthermally enhanced SVE on benzene contaminated soil, and found that the removal rate was only 80% after 45 h at a gas flow rate of 15 L∙min
−1. Moreover, because the vent holes were arranged on the outer side of the lower end, there was an obvious dominant channel, and the benzene content of the inner soil had obvious residues. In this study, the design of the extraction pipe was improved, and the position of the ventilation pipe was optimized to prevent the occurrence of dominant channels as much as possible. Thanks to it, the dominant channel effect was not obvious in the treatment.
Combined with the residual benzene concentration in the treated soil, further analysis was made. On the plane, there was basically no detection in the outermost periphery of the three soil layers, and the residual benzene concentration was higher in the middle and lower in the periphery, higher in the right side and lower in the left side. The reason for the imbalance between the left and right residues was the differences in the ventilation of the ventilation pipe. The lower ventilation flow obviously slowed down the extraction process and reduced the final treatment effect. Therefore, it was speculated that the critical extraction flow rate of benzene in the completely treated soil may be in the middle of C pipe and B pipe flow rate in this experiment. Although there was a big difference in benzene residues between topsoil and bottom soil, the difference in temperature was small. It also showed that in this experiment, the temperature should exceed the critical temperature for completely treating benzene in this soil, which was not a limiting factor of the remediation.
3.2. Airflow Changes and Remediation Efficiency
In the experiment, the running power of vacuum pump was adjusted to the maximum. It was found that the flowmeter at the tail end of pneumatic pipeline showed only 22.1 L∙min−1, this was significantly lower than that at idling. After confirming that the pipeline setting and airtightness were all right, it was speculated that the resistance of soil system in the box is large. In the follow-up monitoring, it was found that some obvious changes occurred in the flow rate, so the flow rate was monitored and recorded. The acquisition used the mobile phone time-lapse shooting function, it was set to shoot once every 10 min, recording the flow change every 10 min, with a total of 108 records.
The inflow numbers are a, b, c, and d, which correspond to the vents of upper left, upper right, lower left, and lower right ventilation wells, respectively. The changes in the flow rate of each in flow and the outflow rate are shown in
Figure 7. It can be seen that the extraction flow rate began to increase significantly after the extraction began; It reached its peak of 28.9 L∙min
−1 at about 140 min, and then began to decrease significantly. It reached a low point of 10.2 L∙min
−1 at about 640 min, and then fluctuated up and down at this value until the end of the test.
No similar experiment has mentioned the change of air velocity during the treatment. Theoretically, extraction and heating take away water from the system, the soil porosity should be increased, the air resistance should be reduced, and the flow rate should be gradually increased. This situation of rising first and then falling is presumably because the vacuum pump continues to operate, this causes the equipment to overheat and reduces the operating efficiency of the electric motor. The sum of the flow rate of the inflow is lower than the pumping flow in the whole process, especially at the beginning of extraction. One of the possible reasons is that the negative pressure in the box after the start of extraction makes the front sampling hole or the port reserved at the top air leak. Among the vents, the flow rate of port B on the right is significantly lower than other vents in the whole process, and the aeration capacity of the surface B pipe is inferior to other vent pipes, which affects the distribution of soil gas benzene concentration and soil residual benzene distribution.
After heating for 4 h, the extraction started, and after the vacuum pump, a PID detector was connected in parallel to monitor the benzene in the tail gas. After testing, connecting the detector in parallel at this position can effectively monitor the tail gas concentration. Compared with the monitoring data of the detector directly connected to the extraction pipe outlet, the error is less than 1%. Due to the limitation of equipment storage, and for the convenience of calculation, the detector is set to record data once every 6 s to represent the average gas concentration in the first 6 s, and a total of 10,800 data were recorded.
The change of benzene concentration in tail gas during the test is shown in
Figure 8. As the concentration changes greatly with time, the data is divided into two parts: the first hour (
Figure 8A) and the last 19 h (
Figure 8B). It can be seen from the figure that the benzene concentration in the tail gas at the beginning of extraction is extremely high, reaching 7812 mg∙m
−3. Once extraction started, the concentration dropped sharply. It dropped to 100 mg∙m
−3 after 11.7 min. After about 60 min, it decreased to 35 mg∙m
−3 and maintained for about 200 min. After that, it gradually decreased, and finally remained at 15 mg∙m
−3.
The method of heating first and then extracting leads to the extremely high benzene concentration in soil gas in the box at the beginning of extraction. This shows that the system has basically completed the thermal desorption of benzene and maintained the dynamic equilibrium at high temperature. After the extraction started, the air containing benzene was pumped out quickly, and completed 43% of the pollutant extraction volume in 2% of the total treatment process.
After 60 min, the benzene concentration in the extracted gas was still about 45 mg∙m−3. At this time, the extraction can still effectively remove the benzene concentration in the box. At 600 min, the benzene concentration in the exhaust gas reached 15 mg∙m−3 and basically no longer decreases. The amount of pollutants extracted from 60 to 600 min accounts for about 45% of the total.
According to the results of soil monitoring before the test, the total mass of soil benzene is about 5340. Based on the Formula 1 calculation, approximately 5389 mg of benzene was extracted at the end of the treatment and the removal rate was 100.9%.
The benzene concentration test results of the treated soil showed that benzene was detected at 18 points. The maximum content is 7.3 μg∙kg−1, which is 0.04% of the 17.8 mg∙kg−1 soil content before the treatment. After treatment, the average residual benzene concentration of the soil is only 0.916 μg∙kg−1, the total content is about 274.8 μg, and the removal rate reaches 99.995%
In a similar device experiment, Li [
27] adopted a one-dimensional columnar device for thermally enhanced SVE of benzene-contaminated soil, which was first thermally enhanced and then extracted. The study showed that in 7 days after heat intensification at 80 °C and extraction at 3 L min
−1 for 12 h, the removal rate of benzene with an organic content of 3.5% soil reached 93%. The organic matter content of the tested soil was 7.5%, which was not suitable for gas-phase extraction [
28]. However, due to better heat preservation conditions and higher heating power, and the benzene removal rate was over 99% within 24 h.
It may bring some significance for further simulation research and engineering application. It means that for volatile organic compounds with a lower boiling point, heating throughout the process may be unnecessary and uneconomical. In addition, heating first and then pumping may be of great help in saving pumping costs and treatment time. In the follow-up research, more detailed energy and economic analysis, more types of pollutants, diversified pollution situations, and comparative study of computer models should be included, in order to obtain important parameters for the engineering application of heat conduction enhanced SVE and build a generalized model for site remediation.