In this study, the effect of powdered detergent content in grey water on the evaporation and cracking characteristics of soils used in agricultural irrigation was investigated. Soil samples treated with different concentrations of detergent solutions (mass of powder detergent to soil mass of 0, 0.2%, 0.4%, 0.8%, 1.6%, and 3.2%) were used in laboratory evaporation testing to obtain the variation water content and cracking images during the evaporation process. Through image processing technology, the crack ratio and fractal dimensions of the cracks were used as evaluation indexes to quantitatively analyze the crack development of samples during the evaporation process. The results show that the water content of the soil samples treated with the detergent solution slightly increases compared to the control soil sample with no detergent. The maximum difference in water content between the samples with and without 0.8% detergent is 7.8%. The total time of the appearance of initial cracks of soil samples treated with detergent was shorter at a higher water content. The final crack ratio and fractal dimensions of the samples treated with different detergent solutions were all larger than the control sample. The soil samples with 1.6% and 3.2% detergent have precipitation of the grains during the evaporation process and form a flocculent layer on the surface of the sample. The dispersion of the soil samples was enhanced significantly after testing. The purpose of this study is to provide some valuable insights into application and management of grey water irrigation in arid and semi-arid areas.

1. Introduction

With the development of global warming, the supply of water for industrial, agricultural, and domestic use in arid and semi-arid lands is facing a serious shortage. The reuse of grey water for agricultural irrigation may be one of the best ways to conserve water. Grey water refers to the portion of domestic sewage that contains no human waste. The main sources of grey water are from washing laundry, bathing, and kitchen drainage. This type of wastewater has an extraordinary potential to be reused [1, 2]. However, there is a large quantity of wastewater from washing that contains pollutants, which include surfactants from detergents and other pollutants such as organics and salts used in the washing process [3]. One of the more important organic pollutants, laundry detergent, enters the soil and groundwater through various passages with adverse effects on the environment [4, 5].

The long-term use of gray water in irrigation not only causes soil pollution through the accumulation of surfactant in the soil but will also change the physical, chemical, and biological properties of the soil with potential negative impacts on plant growth [6]. Abu-Zreig et al. examined the effects of surfactants on the hydraulic properties of soil. They showed that anionic surfactants can significantly reduce capillary rise and permeability of the soil and increase the solid-liquid contact angle (i.e., enhancing the hydrophilicity) [7]. Pinto et al. experimented on grey water irrigation in glasshouses. Compared to drinking water, grey water significantly increased the pH and electrical conductivity (EC) of the soil. Increases in the pH and salinity of the soil did not favor the growth of plants [8]. Travis et al. studied the properties of different soils after grey water irrigation. Their results showed that the hydrophobicity of sandy soil and loam was increased through grey water irrigation [9]. Furthermore, grey water irrigation would also lead to the enhancement of soil dispersion, since the presence of surfactants in grey water reduced the surface tension of the water and the existence of sodium salt, which was a typical component in surfactants [10]. The above research work has mainly focused on the effects of grey water irrigation on soil properties and plant growth. However, since evaporation in arid and semi-arid lands is high, the reduction in water content during the evaporation process of soil results in volume shrinkage and surface cracking [11]. At present, there have been few research studies that have investigated the development of desiccation cracks due to the drying of soil that contains grey water.

The appearance of cracks would destroy the soil structure and change the hydraulic and mechanical properties of soil. In geotechnical engineering, cracks destroy the integrity and increase the compressibility of soil, which may lead to surface subsidence and damage buildings and infrastructures [1214]. The hydraulic conductivity of soil with cracks will increase by several orders of magnitude, which can lead to the failure of the landfill liner and also have adverse effects on the stability of the soil slope during rainfall [1517]. In agricultural engineering, cracks increase the rate of water evaporation in soils and change the characteristics of surface runoff which affects irrigation efficiency. Cracks also affect the rate of the transport of water, solutes, and microorganisms in the soil which have significant impacts on crop growth. Besides, pollutants can contaminate groundwater through preferential flow generated by cracks in soils [1820]. Therefore, the cracking mechanism, controlling factors and parameters, and quantitative descriptions of cracks in soils have been the subject of interest of many research studies [2123]. Tang et al. examined the geometric pattern of soil cracks under the influence of temperature, soil type, dry-wet cycles, and other factors on the rate of evaporation through laboratory experiments. Quantitative descriptions of the crack parameters (number of cracks, crack length, crack width, crack surface area, etc.) were made by using the image processing technology [2427]. Baer et al. studied the cracking phenomenon of soil in the field at different locations by using image processing technology and fractal geometric theory to analyze crack characteristics [28]. The results presented that argillic horizon high in smectitic clay produced a more heterogeneous crack pattern and larger crack areas.

In recent years, there has been a large-scale production of synthetic detergents and surfactants in China (Figure 1) for domestic consumption. With the discharge of domestic wastewater, especially untreated or inadequately treated wastewater, these detergents eventually enter the soil in different ways, which has a significant impact on contaminating soils [29].

Therefore, this study explored the effect of grey water with different detergent concentrations on the desiccation and cracking properties of soil. Phosphorus-free detergent powder, which is the most common detergent among all of the synthetic detergent products, was used to prepare grey water. In order to evaluate the effects on the desiccation and cracking characteristics of soils of the different concentrations of phosphate-free detergent due to drying, digital imaging technology was used to quantitatively analyze the crack ratio and fractal dimensions of the cracks. This study is intended to provide some valuable insights into grey water irrigation field.

2. Materials and Methods

2.1. Materials

The soil samples were obtained from Xiangcheng County which is located in the central part of Henan Province, China (see Figure 2). The total average annual hours of sunshine in Xiangcheng County are 2,281.9 hours, the average annual temperature is 14.7°C, and annual precipitation is 570 mm while the average evaporation for many years is 946.2 mm. The soil samples were obtained at a depth of 65 cm below the surface of a wheat field. The specific gravity, Atterberg limits, and particle size distribution of the soil were measured by using a pycnometer, a liquid-plastic limit combined tester (FG-3, Shanghai Lei Yun Test Instrument Manufacturing Ltd, CN), and a laser particle size analyzer (Mastersizer 2000, Malvern Panalytical Ltd, UK), respectively. The physical properties of the soil samples are summarized in Table 1. According to the international standard for soil texture classification, the soil is classified as clay loam.

The phosphorus-free detergent powder was used to prepare gray water for laboratory experiments. The main ingredients of the detergent powder are surfactants, a variety of detergent builders, dirt suspending agents, phosphorus-free water softener, fragrance, and biological enzymes. Among them, the active substance is mainly sodium linear alkylbenzene sulfonate (LAS), which accounts for 14.3%; the inorganic builders are mainly sodium carbonate and sodium sulfate, which account for 13.2% and 66.1%, respectively, and other ingredients account for 6.4%. Detergents are widely used in household and industrial applications. The source, discharge, and impact of grey water are shown in Figure 3.

2.2. Experiments

Six groups of saturated mud samples with different mass contents of powder detergent were prepared. In order to ensure the reliability and repeatability of the test results, each test with a specific amount of powder detergent was conducted three times which resulted in a total of 18 tests, and the sample with no detergent was considered the control.

The soil was air-dried and then crushed and passed through a 2 mm sieve to remove impurities such as small stones and plant roots. In preparing the soil samples, dry soil was weighed and placed in a cylindrical, uncovered organic glass container with an inner diameter of 18 cm and a height of 4 cm. Detergent solutions with different concentrations were prepared by adding powder detergent of 0, 0.2%, 0.4%, 0.8%, 1.6%, and 3.2% of the dry soil mass to distilled water; the mass of distilled water was equal to the dry soil in sample. Then, the saturated slurry samples with an initial water content of 100% were prepared by mixing the power detergent solution with dry soil thoroughly. Mixing was carried out at the lowest possible speed to avoid air entrainment into the mud which could affect the test results. The slurry in the container was sealed in plastic bags to ensure subsequent stability settling of the mud. Then, the soil samples were taken out of the plastic bags and placed in the laboratory for air drying. The final thickness of the sample was 5 mm.

A weighing and photographic system was designed to take measurements with time as shown in Figure 4. The slurry samples were dried at room temperature, alternating with day and night. The change in mass was recorded by an electronic analytical balance (HC2004, with the accuracy of 0.01 g), which was connected to a computer. The water content () and the rate of evaporation () of the sample during the drying process could be calculated by using Equations (1) and (2): where is the mass of the dry soil, is the initial mass of the sample, is the mass of the sample at time t, is the mass of the container, is the surface area of the sample, and are the changes in mass and time between two successive measurements.

The formation and evolution of the cracks on the surface of the samples were recorded by using a digital camera (Nikon D7500, 20.88 million pixels). In order to ensure that the same image size was obtained for the different samples in recording the pattern of the cracks during the drying process, the digital camera was fixed at the same distance above the sample for all of the tests, and the lens was positioned perpendicular to the center of the surface of samples. The focal length of the camera lens was adjusted to maximize the size of the sample image with sufficient and stable lighting to eliminate dark spots or shadows. The digital camera was connected to a computer through WiFi, and the collected images were automatically transmitted and stored in the computer. Weight measurements and photographs were taken hourly until the difference in mass between two successive readings is less than 0.1 g. During the experiment, temperatures were recorded by using an electronic thermometer.

2.3. Image Processing

The images taken by the camera were RGB color images (see Figure 5(a)). Processing a full-color image requires a longer time while greyscale images can capture soil cracks well with a much shorter processing time. Therefore, the image was first converted to greyscale, see Figure 5(b), by using Equation (3) [30]: where , , and represent the red, green, and blue color components of a pixel, respectively.

In the greyscale images, cracks are usually darker than the other areas, and the value of the greyscale pixel is significantly different from its surroundings. The global threshold method could be used to quickly and effectively binarize the image to identify the cracks [31]. Pixels larger than the threshold value were marked as 1 (white), which represented the soil matrix, and pixels less than the threshold value were marked as 0 (black), which represented the crack (see Figure 5(c)).

Due to the impurities on the surface of the soil sample and image noise, the image contains a large number of impurities after binarization. Some are isolated small black spots, which are not real cracks (see Figures 5(c) and 5(d)). These often appear in the white areas which represent the uncracked soil mass. Although the area of a single black spot due to image noise is not significant, when they are all added up, they will statistically affect the crack ratio (the total crack area to the total surface area). Nevertheless, the single area of the black spots due to image noise is much smaller than the total area of the crack; therefore, the area of the black spots can be removed by setting an area threshold. The optimal area threshold can be obtained by manually comparing the largest black spot in the soil blocks image with the total area of the cracks. When there are some large noise areas in the image, a better binary image can be obtained by adjusting the local brightness of the image visually. Also, since parts of a crack may penetrate the entire soil layer and expose the bottom of the container, the grey value at that location becomes higher. Some small isolated white areas will form in the middle of a crack (Figure 5(d)), which are not uncracked soil. Due to the illumination, while taking photographs, a continuous crack may appear to be broken into several segments after binarization. Therefore, it is necessary to bridge the broken cracks with the closed operation in the morphological during image processing to connect the narrow gaps by filling the holes with crack pixels if they are smaller than a certain size. The processed image is shown in Figure 5(e).

The crack analysis is usually based on measurements of the geometric parameters of cracks [3234]. In this paper, the crack ratio is used to reflect the cracking of soil treated with a detergent solution. The crack ratio is defined as the ratio of the crack area to the total area of the surface of the sample given by: where is the crack ratio; is the surface area of the crack; is the total area of the soil sample, and and are the number of black and white pixels in the binarized image, respectively. Note that is not exactly equal to since is the projected area of on a two-dimensional (2D) image. However, for a thin sample with a small thickness, is approximately equal to .

The characteristics of cracks in soils are highly nonlinear. By applying the fractal theory, Decarlo and Shokri showed that the patterns of surface cracks suit self-similarity [35]. When the fractal theory is used to analyze surface cracks, the degree of crack development can be evaluated by the fractal dimensions, that is, a larger fractal dimension means that the cracks are more developed. The box dimension is a widely used fractal dimension. Consider a square grid with the length of the sides , by counting the minimum number of grids that cover the entire cracking area. The fractal dimensions of the cracks could be calculated from the slope of the versus plots as follows [36]: where is the fractal dimension, is the size of the square grid, and is the number of grids that cover all of the cracks.

3. Results

3.1. Effects of Detergent Powder Solution on the Evaporation of Soil

Changes in the average water content of the samples with time for different detergent contents are shown in Figure 6. The changes in the average water content of soil samples treated with different concentrations of the detergent solution were basically the same. The slope of the curve was equal to the rate of evaporation. The process of drying could be divided into three stages. Stage I was the evaporation of the free surface water and the rate of evaporation of all the samples was basically the same. Stage II was the evaporation of the capillary water on the soil surface with no free surface water before cracking of the soil surface. In this stage, the rate of evaporation slowed down and the water content of the samples treated with different concentrations of detergent differed. This presented the effects of detergent on the rate of evaporation since all of the samples treated with detergent had a higher water content than the control sample. In Stage II, the maximum difference in the water content was 6.8% when compared with the control sample, which was found in the sample treated with 0.8% detergent. In Stage III, surface cracks gradually developed, which led to an increase in the surface area for evaporation due to the development of cracks. Compared with the second stage, the decrease rate of water content increased, and the differentiation of water content of each sample continued to expand. The water content of the samples with 1.6% and 3.2% detergent decreased most rapidly with the shortest time for completion of drying. At the end of the test, the residual water content of the samples is between 0.22% and 0.86%.

Figure 7 presents that the rate of evaporation of the samples is greatly affected by the temperature which fluctuates between day and night. The rate of evaporation is directly correlated with the temperature change. As the evaporation continued, the fluctuation in the rate of evaporation decreases with changes in temperature between day and night. In the final stage of the test, the rate of evaporation decreased continuously and is not affected by the change in temperature.

3.2. Effect of Detergent Powder Solution on Desiccation Cracks

The crack ratio can reflect the cracking degree of the soil surface as a whole. Figure 8(a) shows that the overall trend of crack development with different concentrations of detergent was basically the same. The development of cracks could be divided into three stages: crack initiation, rapid crack development, and slow crack development. Crack initiation occurred at about 50 to 55 hours of evaporation which was followed by rapid crack development. The rapid development stage was short, between 10 and 15 hours, and then the soil entered the stage of slow crack development until the end of the testing as shown in Figure 8(b). In general, soils treated with detergent had a higher crack ratio than the control samples.

With an increase in detergent content, the completion time of cracking decreased with a higher rate of crack growth. The sample with 0.8% detergent had the largest crack ratio and the sample with 1.6% detergent had the highest rate of crack growth in the stage with rapid crack development.

The fractal dimension can be used to evaluate the irregularity of the cracking network, that is, the complexity. A larger fractal dimension means that the cracking is more developed. After the cracking of the samples treated with different amounts of detergent cracks, the general trend of the fractal dimension with time was similar to that of the crack ratio (Figure 9). After crack initiation, the fractal dimension increased rapidly in a short period. This was followed by slow crack development and the fractal dimension was more or less constant with time. The fractal dimension of the soils treated with detergent was larger than that of the control sample. The sample treated with 0.8% detergent had the largest final fractal dimension of 1.767, compared to the final fractal dimension of 1.45 of the control sample which was the lowest among all the samples.

Figure 10(a) shows the crack ratio versus the average water content. The overall trend of all of the samples was basically the same. An initial decrease in the water content caused a rapid rate of increase in the surface cracking which was followed by stable cracking. The water content of the samples treated with detergent was higher than that of the control sample. The soil treated with detergent more than 40% compared to the control sample of 32.7%. For the same water content, the crack ratio of the soil treated with detergent was larger than that of the control sample. A comparison of Figures 6 and 10(a) presents that the samples treated with detergent showed an increase in the water content with an increase in the rate of surface cracking. Figure 10(b) shows the same trend in the fractal dimension with water content that is similar to Figure 10(a).

4. Discussion

Macroscopically, the evaporation of water is the process of transforming liquid water into a gaseous phase. Microscopically, due to increases in the kinetic energy, water molecules overcome the forces between the liquid water molecules and escape from the liquid surface.

Temperature is the main factor that affects water evaporation. At higher temperatures, the kinetic energy in the water molecules is higher, with increased molecular motion. At the same time, the viscosity and surface tension of the liquid decrease with increases in the temperature which results in a decrease in the resistance of the water escaping from the liquid surface thus increasing evaporation [37]. Therefore, the rate of evaporation has a positive correlation with the temperature between day and night (Figure 7).

The control soil sample, which was formed by the natural deposition of saturated mud, had a low density, high porosity, and loose soil particle arrangement, which led to the formation of rich capillary channels with different pore diameters [26]. During the evaporation process, there is a water gradient between the surface and the interior of the soil sample. Under the action of the capillary, the water is easily transported in these channels with little resistance to evaporation. Therefore, the control soil sample can easily lose water and has a lower water content.

The main active ingredient in powder detergent is an anionic surfactant called linear alkylbenzene sulfonate (LAS). The molecular structure of LAS is shown in Figure 11(a). This surfactant is amphiphilic. LAS is a compound with a structure that comprises a benzene ring attached to one end as a linear alkyl chain of different lengths (hydrophobic part of the sulfonic acid group) and the other end as a sodium sulfate group (hydrophilic part of the sulfonic acid group), as shown in Figure 11(b). LAS tends to accumulate at the two-phase interface (liquid/gas, liquid/solid) in a solution, which increases the distance between the water molecules at the interface. This reduces the surface tension at the liquid/gas and liquid/solid interfaces to enable cleaning or solubilization [38].

On the one hand, the samples treated with the powder detergent solution had a decrease in the surface tension of the solution, which reduced the resistance of the water molecules to escape from the liquid and therefore this was conducive to evaporation. On the other hand, surfactants are amphiphilic, and their hydrophilic end could interact with free water in the soil and bounded water on soil particles. The interaction with the hydrophobic end would result in the hydrophobicity of the water-conducting channels and could hinder the migration of capillary water and the rising of water from the bottom of the sample. Besides, as the concentration of the surfactant in the water increased, the surfactant molecules in the form of monomer could aggregate into micelles and block some of the capillary channels, thereby reducing the amount of evaporation. This process is shown in Figure 12(a). The two mechanisms together affect the final water content of the soil. As shown in Figure 6, the water content of the samples with different contents of detergent was greater than that of the control sample, which indicated that in this test condition, the dominant process was the inhibition of water evaporation.

During the evaporation process, the water on the surface of the sample was continuously decreasing which generated suction when the soil changes from a saturated to an unsaturated state. This led to the formation of a tensile stress field on the surface of the sample. Cracks appear when the tensile stress is greater than the tensile strength between the soil particles. The formation and development of soil cracks are the results of matric suction (internal force) and a form of tensile stress failure [39]. The tensile strength of soil depends on the matric suction. When the soil has a high water content, the tensile strength of the soil is also low although the suction in the soil is small. Clay is prone to crack under the action of small suction. According to Peron et al. [40], the saturation of fine-grained soil was close to 100% when the initial cracks appear.

It can be seen from Figure 10 that the crack ratio and fractal dimension of the cracks of the soil treated with detergent were both higher than those of the control sample. This is because the surfactant in the powder detergent solution increased the dispersibility and changes the properties of the soil [10]. After the surfactant enters the soil, it will be adsorbed at the soil interface through ion exchange, hydrogen bonding, static electricity, van der Waals attraction, etc. to reduce the interfacial tension of the soil particles, so that the agglomerated soil particles are dispersed. On the other hand, the anionic surfactant in the detergent can increase the potential of a negatively charged on the surface of the soil colloids, thereby increasing the repulsive force and stability among the colloids. This results in enhancing the dispersion of the soil particles. Finally, these factors lead to a change in the soil structure [41]. The samples treated with 1.6% and 3.2% detergent both had grains that precipitated during the evaporation process, thus forming a flocculent layer on the surface of the soil sample (Figure 13). After the testing, the soil samples with the powder detergent contents 1.6% and 3.2% were in a powder state with low strength, while the control sample had a higher strength.

5. Conclusions

The purpose of this study is to investigate the effects of grey water used in agricultural irrigation on the characteristics of evaporation and desiccation cracks in the soil during drying. Free evaporation tests that have no controls on environmental temperature and humid of saturated slurry samples treated with different contents (0, 0.2%, 0.4%, 0.8%, 1.6%, and 3.2% of the soil dry mass) of powder detergent were carried out in the laboratory.

The results presented that the soil samples treated with powder detergent solution have a lower inhibitory effect on the evaporation characteristics, and the rate of evaporation shows a positive correlation with temperature which fluctuates between day and night. The soil samples treated with detergent have higher crack ratios and larger fractal dimensions of the cracks than the control samples which have a shorter time for the initial crack to appear at a higher water content.

However, the composition of the powder detergent solution is complex, and the soil itself also contains different types of ions and organic matter. The impact of the interaction process among the various components of the detergent and soil on the test results has not been fully resolved in this study. Therefore, further research work in this respect will be carried out later.

Data Availability

The data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of 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.


The laboratory portion of this study was carried out in the Soil Mechanics Laboratory of the School of Civil & Architecture Engineering, Xi’an Technological University. The authors sincerely thank the authority of the School of Civil & Architecture Engineering. This research is financially supported by National Natural Science Foundation of China (NO. 51979225); National Natural Science Foundation of China (NO. 51679199); Xi’an Key Laboratory of Civil Engineering testing and destruction analysis on military-civil dual-use technology.