In the process of microbial curing of desert aeolian sandy soil, we thought of the water-holding properties of straw flour in view of the high proportion of fine particles and poor water retention of desert aeolian sandy soil, and therefore designed an experiment to add straw flour to enhance the effect of microbial curing of desert aeolian sandy soil. The sand columns prepared under different curing stages were analysed by low-field NMR techniques. The test results show that: (1) the curing product calcium carbonate can effectively fill the pores of the sand and reduce the total porosity, and the addition of straw powder increases the total porosity of the sand column, especially the number of large pores (5–60µm) increases significantly; (2) The addition of straw powder increased the original sand column pore volume and pore size, significantly increasing the water holding capacity of the cured sand column, and the free water content of the sand column was significantly greater than that of the original sand column;(3) The pore filling rate of the sand column decreased after mixing with straw powder, which improved the uniformity of calcium carbonate precipitation within the sand column, but the shear strength was lower than that of the original sand column. The results of the study provide a theoretical basis and data support for optimising the curing effect of desert aeolian sandy soil and its water-holding capacity.
Research Article
Analysis of the effect of wheat straw flour on the pore structure of microbial induced carbonate precipitation (MICP) cemented desert aeolian sandy soil based on low field NMR
https://doi.org/10.21203/rs.3.rs-2243966/v1
This work is licensed under a CC BY 4.0 License
Version 1
posted
You are reading this latest preprint version
Microbial induced carbonate precipitation
Nuclear magnetic resonance
Straw powder
Desert aeolian sandy soil
Pore characteristics
Water-holding capacity
According to the results of China's fifth desertification monitoring, the country's desert area is 594300 square kilometres, accounting for 6.19% of the national territory. China has proposed to place more importance on the construction of ecological civilisation while developing its economy, and has put forward the development goal of " Clear waters and green mountains are as good as mountains of gold and silver.". Under the dual carbon policy and the rapid development of the desert photovoltaic industry, it is necessary to change the traditional concept of sand management, and actively address the unsustainable problems of traditional measures, to provide an effective way for sustainable recovery of sandy land and sustainable improvement of the ecological environment (Wang T 2016). For the desert management aspect, many scholars have done a lot of research work to find a fast, efficient and economical method to fix the drifting sand and prevent wind erosion (Lai J H et al. 2017; Qu J J et al. 2014; Tang C S et al. 2020). Compared to traditional curing methods, biomaterials can exhibit unique properties of spontaneity, remodelling and regeneration in all types of soils. Among them, microbial induced calcium carbonate precipitation technology, as a new soil consolidation method, has broad application prospects and economic benefits due to its non-pollution, low disturbance and simple operation (Charles Wang Wai Ng et al. 2022).
The application of microbial grouting techniques can be effective in improving poor foundations. The addition of a bacterial, urea solution mixed with a calcium salt phase to loose sandy soils can improve the integrity and mechanical properties of the soil (Whiffin V S et al. 2007; Van Paassen L A et al. 2010). Also, soils treated with a high concentration of urea-calcium chloride cementing solution showed a more significant reduction in permeability than soils treated with a low concentration of cementing solution under the same grouting rounds (Yasuhara H et al. 2012). The MICP technique has a positive impact on the improvement of soil erosion resistance. During the curing process, the amount of calcium carbonate precipitated in the soil increased simultaneously with the concentration of the cementing solution, but the distribution of the precipitation gradually became uneven (Jiang N J et al. 2016). MICP technology can also be used for the improvement of organic soils. The shear strength and calcium carbonate production of organic soils were tested before and after the amendment, and the mineralisation by micro-organisms increased the calcium carbonate content in the pores by 20% and increased the shear strength (Canakci H et al. 2015). In practice, in order to adapt MICP technology to a wide range of different soil environments, external admixtures are often used to mix with the original soil to optimise the physical and chemical properties of the soil to be cured, thus enhancing the microbial improvement effect on the soil. The combination of 3% rubber granules and MICP technology for the consolidation of sandy soils can increase the strength of the soil mix (Wang Y et al. 2020). Considering the porous structure and activity of volcanic ash, volcanic ash also can be used as an external admixture to enhance the effect of microbial consolidation of sandy soils. It has been found that a 10% admixture of volcanic ash can increase calcium carbonate production by 29.55% compared to conventional MICP, while increasing the compressive strength of cured sand columns by 52.26%-62.96% (Xiong Y et al. 2022). In recent years, the use of natural materials as external admixtures has often been favoured by many researchers. Plant straw is a major by-product of crops, and straw is rich in nitrogen, phosphorus, potassium and trace elements, an important renewable resource, as well as a large amount of chemical energy, which is an energy source for the life activities of soil microorganisms (Liu C et al. 2014). The straw powder is a product of natural plant straw that has been air-dried and crushed and has a larger specific surface area than natural straw fibres, which makes it effective in increasing soil permeability (Larney F J et al. 2012).
At the current stage, the evaluation indicators for the curing effect of MICP technology are divided into two main directions, namely macro-mechanical property testing and microstructural analysis. At the micro level, the pore structure of the soil influences its basic properties. The number, size distribution, shape and three-dimensional spatial configuration of the pores in the soil determine the macroscopic mechanical properties of the soil, while the soil pores are also a channel for biological movement and an important place for life (Shao MA 2006). In practice, the soil-water characteristic curve of a soil body is usually determined by the pressure film gauge method and the pore space structure inside the test soil is indirectly deduced from the pressure values (Hong W T et al. 2016). However, these methods are time-consuming, cumbersome, costly and damaging to the test soil (Kong C et al. 2016). In a microscopic study of the MICP technique, the curing effect of the soil at different temperatures was observed by scanning electron microscopy (SEM) and it was initially found that: the production of calcium carbonate was three times higher at 50°C than in a 25°C environment (Almajed A et al. 2021). (Mahawish A et al. 2019) used X-ray computed tomography (XCT), SEM and energy dispersive spectroscopy (EDS) to measure the amount of calcium carbonate production and changes in the soil pore structure.
Nuclear Magnetic Resonance (NMR) is an advanced non-destructive testing technique that can be used to obtain information on the distribution of water molecules within a cured saturated sand column so that the pore structure of the cured sand column can be calculated and the internal structure of the soil can be imaged by relying on NMR imaging. (Hinedi Z R et al. 1993) made the first use of 1H-NMR relaxation to determine the pore size distribution in soil samples. (Jaeger F et al. 2009) calculated the surface relaxation strength values of the soil and thus the pore distribution of the soil by quantitatively comparing the 1H-NMR relaxation method with the conventional water retention curve, and the two data were very close. Natural sediments are often composed of materials with high magnetisation and to reduce the effect of magnetisation artefacts, measurements can be made on soils by using NMR instruments with low magnetic fields. Low-field NMR instruments are also lightweight and convenient compared to high-field NMR instruments (Guo J C et al. 2020).
Existing research has concentrated on the improvement of coarse-grained soils such as sandy soils, and there are still some shortcomings in the curing of fine-grained desert aeolian sandy soil. Desert aeolian sandy soil is characterised by fine grains, poor grading, low cohesion and poor water retention, which present new challenges for microbial curing effects. There is a strong correlation between the pore structure and macro-mechanical properties of cured desert aeolian sandy soil (Yang G et al. 2021). In order to comprehensively understand the mechanism of microbial mineralization of windy sands and to visually analyze the changes of mineralized windy sands pore space under the action of MICP, this experiment uses the low-field nuclear magnetic resonance (LF-NMR) technique to analyze the pore structure of microbial cured desert aeolian sandy soil from the perspective of nuclear magnetic dynamics with the help of nuclear magnetic relaxation characteristics of H protons. At the same time, the enhancement effect of straw powder on microbial curing of desert wind-deposited sand was systematically analysed in conjunction with macroscopic physico-mechanical tests and calcium carbonate curing rates.
2.1 Experimental materials
The desert aeolian sandy soil: In this experiment, the desert aeolian sandy soil was taken from the Gurbantunggut Desert near Sangonghe Kazakh National Township, Fukang City, Changji Hui Autonomous Prefecture, Xinjiang. The desert aeolian sandy soil was dried to a constant weight at 105°C and subsequently, the dry density was determined by the cutting ring method and the sand was sieved for particle size using a vibrating sieve apparatus. The particle size distribution was found to be mainly between 0.075 and 0.3mm, accounting for 93.18% of the total. The apparent density of the desert aeolian sandy soil is 2640 kg/m3, pH 8.36, and the chemical composition is mainly SiO2, Al2O3, Fe2O3 and CaO, with the four compounds accounting for about 95% of the population. In addition, there are some compounds containing Na, K, Mg, etc., of which SiO2 contains about 72.1%. The mineral composition of the desert aeolian sandy soil is mainly quartz, with some feldspar and muscovite.
| Sand shape | Specific gravity | Natural porosity | Mud content | d50(mm) |
|---|---|---|---|---|
| Ovoid | 2.64 | 43.20% | 0.4% | 0.131 |
Wheat straw powder: The experimental material was selected from wheat straw from Wujiaqu farm in Xinjiang, dried to constant weight in a drying oven at 60°C and crushed to a particle size of less than 180µm, as shown in Fig. 1.
Sporosarcina pasteurii: S. pasteurii were obtained from Shanghai Bioresource Collection Center (NO. ATCC 11859).
Trypticase soy broth (TSB): The culture medium was used from Guangdong Huankai Microbial Sci & Tech. Co. Ltd. The specific composition is shown in Table 2.
Test cementing solution: A 1.5 mol/L CaCl2 solution was mixed with a 1.5 mol/L urea solution in an equal ratio of 1:1 to produce the cementing solution used for the curing test.
| Composition | Quantity per litre of deionised water |
|---|---|
| Pancreatic digest of casein Papaic digest of soya bean (Soybean) Sodium chloride Dipotassium hydrogen phosphate Glucose monohydrate Final PH | 17.0g 3.0g 5.0g 2.5g 2.5g 7.3 ± 0.2 |
2.3 Experimental preparation
2.3.1 Experimental solution preparation
To begin, the experiment was prepared by inoculating 2 ml of S. pasteurii in TSB medium, which was subsequently placed in a thermostatic shaker at 30°C and 170 rpm for an activated value-added culture to produce a bacterial solution with OD600 = 2.0. The cementing solution was also made in a ratio of 1.5 M calcium chloride to 1.5 M urea to produce the cementing solution required for the test (Duo L et al. 2018).
2.3.2 Experimental sample preparation.
(1) Sand column production
Preparation of the original sand column, indicated by OS for this test. The dried desert sand was filled through a funnel into a 25mm diameter by 60mm high test mould and a 61.8mm diameter by 60mm high layered shear mould, filled and pounded several times to confirm the sand column was dense.
A sand column with wheat straw flour was prepared, denoted by WS in this experiment. In the preparation of the externally blended straw powder samples, the straw powder was added at 0.2% of the mass of the original sand column (Li H et al. 2021), and the desert aeolian sandy soil was mixed with the straw powder sufficiently to distribute the straw powder uniformly in the desert aeolian sandy soil, after which the mixture was filled into the mould in the same manner as the original sand column.
(2) Sand column curing scheme
Bacterial solution with OD600 = 2.0 was injected from the top of the sand column at a rate of 0.278 mol-L-1.h-1 until the column was saturated and the bacterial solution filled the internal pores of the sand column and left to stand for 3h after the injection was completed. Next, the cementing solution was injected into the sand column continuously, at 12h intervals (Al Imran M et al. 2018). The number of injections of bacterial and cementing solutions for each test group is shown in Table 3.
| Curing round | OS1-1 | OS1-5 | OS2-1 | OS2-5 | WS1-1 | WS1-5 | WS2-1 | WS2-5 |
|---|---|---|---|---|---|---|---|---|
| The first round of curing | ★ | ★ | ★ | ★ | ★ | ★ | ★ | ★ |
| ▲ | ▲ | ▲ | ▲ | ▲ | ▲ | ▲ | ▲ | |
| ▲ | ▲ | ▲ | ▲ | ▲ | ▲ | |||
| ▲ | ▲ | ▲ | ▲ | ▲ | ▲ | |||
| ▲ | ▲ | ▲ | ▲ | ▲ | ▲ | |||
| ▲ | ▲ | ▲ | ▲ | ▲ | ▲ | |||
| The second round of curing | ★ | ★ | ★ | ★ | ||||
| ▲ | ▲ | ▲ | ▲ | |||||
| ▲ | ▲ | |||||||
| ▲ | ▲ | |||||||
| ▲ | ▲ | |||||||
| ▲ | ▲ | |||||||
| Note: ▲Add a cementing solution | ||||||||
| ★Add a bacterial solution | ||||||||
(3) Drying and demoulding of sand column
Once the curing is complete, the sand column is placed in a constant temperature drying oven at 105°C to dry the process. When the specimen is completely dry, it is placed at 25°C and cooled down until the temperature is reduced to room temperature.
2.4 Experimental methods
2.4.1 NMR test
The finished specimens were subjected to a water retention treatment and subsequently tested for the amount of material H proton signal using a MesoMR23-060H-l NMR instrument. 1H NMR measurements were performed at a low magnetic field strength of 0.047 T using a CPMG sequence with the following test parameters: intrinsic parameters: SF = 12MHz, O1 = 556624.31Hz, P1 = 5.8us, P2 = 9.6us; adjusted parameters: SW = 200KHz, TW = 5000ms, RFD = 0.08ms, RG1 = 20; DRG1 = 3, NS = 8, DR = 1, PRG = 3, NECH = 15000, TE = 0.3ms. As the time required to obtain the T2 relaxation time is shorter than that required to obtain the T1 relaxation time in practice, the transverse relaxation time T2 will be used in this test to characterise the pore distribution within the curing sand column. Eventually, the transverse relaxation time (T2) of 1H protons in soil samples was obtained experimentally.
2.4.2 Shear strength test
The shear test was carried out using a quadruple automatic straight shear apparatus, and the cured sand column was divided into three parts: the upper, middle and lower parts, each with a diameter of 61.8 mm and a height of 20 mm. Align the upper and lower shear boxes before starting the test, place the hard plastic film and permeable stone at the bottom of the sand column and place the ring knife blade facing upwards. A hard plastic film and a permeable stone are placed on top of the sand column while it is pressed into the shear box, after which the sensor is placed and secured, and finally the apparatus is reset and the test begins. The test shear rate was set at 0.8 mm/min and a normal stress of 100 kPa was applied to each set of specimens, with the test ends when the shear displacement reached 6 mm.
2.4.3 CaCO3 curing rate
The sand column was divided into 3 parts, the upper, middle and lower parts, according to the finished curing. 20g of each part was taken and the content of calcium carbonate in the specimen was determined by the acid-washing method. This is done as follows: before the test, the specimens are soaked in pure water to rinse off any soluble salts left over from the curing process and are subsequently dried and weighed in an oven. The test was started by placing the dried specimen in an excess of hydrochloric acid to fully dissolve the calcium carbonate until the sand particles were all dispersed and no air bubbles were generated and then filtered when the reaction was complete. Next, the soluble salts on the surface of the sand particles are washed with pure water and finally, the washed sand particles are dried to a constant weight and weighed. The mass loss before and after treatment of the specimen is the calcium carbonate content of the specimen. The calcium carbonate curing rate is obtained by the following formula:
C(CaCO3)=\(\frac{{m}_{1}-{m}_{2}}{{m}_{1}}\)×100% formula 1
In formula 1:
m1 is the dry weight of the sample before pickling
m2 is the dry weight of the sample after pickling
3.1 Relationship between sand column water absorption mass and T2 spectral area
The LF-NMR test is based on the use of H protons as a probe to detect changes in the amount of T2 spectral signal during water uptake, which shows a positive correlation with the quality of the water uptake. Figure 4 shows the relationship between the measured water absorption mass and the T2 spectral area of the OS1-1 sand column specimen and the fitted equation for its water absorption mass and T2 spectral area as a calibration line for the curing material. The fitted equation shows a correlation coefficient (R2) of 0.99219, showing a high linear correlation. The different masses of water can be determined by quantifying the NMR equation of the specimen and the porosity of the corresponding material is calculated from this fitted equation.
The fitted formula for the absorbed mass and T2 spectral area is
Y = 16914.49X + 360.89 formula 2
In the formula 2: Y is the T2 spectral area and X is the NMR water mass of the composite.
The NMR method measures the porosity of the material i.e. the ratio of the volume of water in the pores to the volume of the saturated specimen, as shown in formula 3:
In the formula 3, P is the porosity, X1 is the NMR mass of the composite when it is full of water, X2 is the NMR mass of the composite when it is absolutely dry, ρw is the density of water, taken as 1 g/cm3, and Vw is the volume of the full water sample.
| Test number | T2 spectral area | X 1/g | X2/g | Vw/cm3 | NMR Porosity/% | Weighing method Porosity/% | Difference in porosity/% |
|---|---|---|---|---|---|---|---|
| OS2-5 | 63170.26 | 24.54 | 20.83 | 12.56 | 29.56 | 28.22 | 1.34 |
| WS2-5 | 79709.59 | 25.41 | 20.72 | 12.56 | 37.35 | 38.97 | 1.62 |
As can be observed from Table 4, the porosity test results calculated by both the NMR and the weighing methods are in high agreement for desert aeolian sandy soil in its original state and desert aeolian sandy soil with external straw powder.
3.2 Effect of microbial mineralisation on the overall test signal amplitude of the sand column
3.2.1 Variation in the overall signal amplitude of OS and WS columns
The LF-NMR test is based on the use of H protons as a probe to detect changes in the amount of T2 spectral signal during aspiration, with the signal showing a positive correlation with the quality of the aspiration.
The LF-NMR test is based on the use of H protons as a probe to detect changes in the amount of T2 spectral signal during the water absorption process, with the amount of signal showing a positive correlation with the quality of the water absorption. In the inverse spectra of T2 relaxation times, each peak is generally considered to represent the same class of volume of water, the peak area is the moisture content of the corresponding pore, and the highest point of the peak represents the average relaxation time of the pore in this state (Zhao H et al. 2021). The tests reflect the structure of the pores at different casting rounds and different stages of material curing by analysing the change in the peak area of different sand column specimens during the curing process. The sand columns OS1-1, OS1-5, OS2-1, OS2-5, WS1-1, WS1-5, WS2-1, WS2-5 were cured and then subjected to low field NMR T2 relaxation time testing. The obtained T2 relaxation signal was inverted to obtain the peak area in this state. Figure 5 shows the T2 relaxation time and signal amplitude of the colluvial sand column at different water absorption times, and Table 5 corresponds to the T2 relaxation time and peak area for each peak. The horizontal coordinate is the T2 relaxation time, the longer the relaxation time, the larger the pore radius; the peak area (A) reflects the water content of the saturated soil at each stage after microbial curing, the larger the peak area, the higher the water content (Jaeger F et al. 2010).
| Soil types | Group NO. | T21/ms | T22/ms | A21/cm2 | A22/cm2 | A/cm2 |
|---|---|---|---|---|---|---|
| OS | 1–1 | 0.30 | 27.364 | 22963.096 | 59869.583 | 82832.679 |
| 1–5 | 0.30 | 33.701 | 20357.742 | 54089.699 | 74447.441 | |
| 2 − 1 | 0.30 | 31.44 | 16191.263 | 51192.399 | 67383.662 | |
| 2–5 | 0.35 | 25.529 | 16401.862 | 46768.399 | 63170.261 | |
| WS | 1–1 | 0.30 | 22.219 | 24592.627 | 67988.048 | 92580.675 |
| 1–5 | 0.30 | 29.332 | 22730.281 | 64279.032 | 87009.313 | |
| 2 − 1 | 0.30 | 23.817 | 21383.531 | 60298.104 | 81681.635 | |
| 2–5 | 0.30 | 23.817 | 21644.191 | 58065.395 | 79709.586 |
Figure 5 and Table 5 show the relaxation times and peak areas corresponding to the various stages of curing for both OS and WS columns, where T21 and T22 correspond to A21 and A22 respectively, and the total peak area corresponds to A. T21 relaxation time in the range of 0.28-0.35ms, T22 relaxation time in the range of 22.21-33.70ms. The T21 and T22 of the two types of sand columns vary with the number of additions of the cementing solution, with T21 varying in a small range and T22 moving in the direction of short relaxation. As microbial mineralisation continued, the total peak area A decreased continuously, with the total peak area A of the OS column decreasing by 22.13% and the total peak area A of the WS column decreasing by 12.09%.
Also, as shown in Table 5, the WS column showed a gradual decrease in A21 as the number of microbial cures increased, but there was a small increase in peak area A21 in both the second curing round. However, the peak area A22 corresponding to both soils showed a decreasing trend throughout the microbial mineralisation, with a decrease of 21.88% in the OS column and a decrease of 14.59% in the WS column.
WS columns are prepared by adding 0.2% wheat straw powder to desert aeolian sandy soil. From the test results, the peak areas A21 and A22 corresponding to the WS sand column were higher than those of the OS in the original desert aeolian sandy soil at all stages of the microbial mineralisation curing sand column, indicating that the straw powder enhanced the porosity of the material and that the straw powder in the voids stored a large amount of water in a water retaining state, thus enhancing the water holding capacity of the original desert aeolian sandy soil.
3.2.2 Comparison of peak area trends between OS and WS columns
According to Table 5, it was found that the trend of change in the peak area of OS and WS showed a large difference, and we calculated the difference in the trend of change in the peak area of A21, A22 and A of OS and WS by Eq. 8 and plotted the results as Fig. 6.
R=\(\frac{{A}_{WS}-{A}_{OS} }{{A}_{OS}}\) ×100% formula 4
In formula 4: R is the ratio of the difference between the peak areas of the two soils.
AWS is the total peak area A of WS.
AOS is the total peak area A of OS.
As can be shown in Fig. 6(a), the total peak areas of OS and WS both showed a decreasing trend with the increase in the number of injections, but the rate of decrease of OS and WS were different. RA is the ratio of the difference between the total peak areas of OS and WS, which showed an increasing trend with the increase in the number of injections, indicating that the total peak area of WS decreased slowly relative to OS. Figure 6(b) shows the trend of the difference in peak area between WS and OS. As can be seen from Fig. 6, the difference in RA21 is significantly larger, indicating that the difference in peak area between OS and WS corresponding to A21 has increased significantly after multiple injections; the difference in RA22 has decreased.
The results of the above tests showed that the straw powder had a small effect on the trend of the A21 peak area of OS and WS and a large effect on the trend of the A22 peak area.
3.3 Influence of microbial mineralisation on the overall pore size distribution of desert aeolian sandy soil
The results of 1H-NMR tests on soils are T2 distribution curves for lateral relaxation times, indicating relative pore size distributions, rather than absolute true pore size distributions. There are usually two methods for converting the NMR T2 distribution into a true pore size distribution: the T2 cut-off method for centrifugation experiments and the surface relaxation rate method. The surface relaxation rate method is a simple and convenient way to obtain the pore size distribution directly from the surface relaxation rate value of the soil compared to the T2 cut-off method of centrifugation experiments (Ge X et al. 2021).
In the NMR testing of sand columns, due to the irregularity of the true shape of the pores and the variable form of contact between sand particles, we assume the pore volume to be the geometry of a cylindrical pore and the "pore diameter" to be the radius of the equivalent cylindrical pore. Also, different soil types have different values of surface relaxation strength ρ2. For sandy soil materials, the value of ρ2 is generally in the range of 400–700 µm/s. In this paper, the value ρ2 = 550 µm/s is taken (Duschl M et al. 2015), and then a link can be established between T2 relaxation time and pore radius, as shown in formula 5.
r = ρ2 T2 Fs formula 5
In the formula 5: r is the pore radius in µm and Fs is the geometry factor (for spherical pores, Fs = 3; for columnar pores, Fs = 2).
In order to reasonably analyse the properties of the sand column, the pores of the sand column were divided into micropores (0–30µm), mesopores (30–75µm), and macropores (> 75µm) according to the pore size in this experiment (Strandberg A et al. 2021). And the T2 spectra of the different stages in the mineralisation of the soil by the 2 microorganisms were converted to plot the corresponding pore distribution according to formula 5. as shown in Fig. 7.
| Pore Type | OS1-1 | OS1-5 | OS2-1 | OS2-5 | WS1-1 | WS1-5 | WS2-1 | WS2-5 |
|---|---|---|---|---|---|---|---|---|
| Micropores | 63.59% | 59.51% | 58.33% | 64.05% | 66.54% | 61.31% | 64.65% | 65.65% |
| Mesopores | 23.57% | 25.92% | 24.41% | 23.49% | 22.23% | 25.00% | 22.69% | 21.68% |
| Macropores | 12.84% | 14.57% | 17.26% | 12.46% | 11.23% | 13.69% | 12.66% | 12.67% |
The pore radius of desert aeolian sandy soil is mainly distributed between 0µm and 150µm, with a very small percentage of pore volume with a radius greater than 150µm. Combined with Table 6, we can see that the OS column has the largest number of micropores, accounting for 58–64% of the total pores, with 0.1µm -0.5µm, 5.0µm-15.0µm and 15.0µm-30.0µm being more abundant and higher than the percentage of other pore sizes. The mesopores content ranges from approximately 22–26% of the total pore space, with a decreasing trend in the pore percentage with increasing pore size. The sand column has the least proportion of macropores with a pore size distribution > 75 µm, with a content of between 12% and 18%. The pore structure of the WS column was changed after the external mixing of straw powder compared to the original desert aeolian sandy soil OS. Macropores in the WS column were relatively small accounting for 11–14% of the total pores, while the proportion of micropores increased significantly, accounting for 61–67% of the total pores; the proportion of mesopores was close to that of the OS column.
Figure 7 shows the pore size distribution of the sand column at different stages of curing. It can be seen that the volume of micropores in the OS column decreases during the first three stages of microbial curing, OS1-1, OS1-5 and OS2-1, while the volume of macropores remains constant as curing progresses. The volume of Mesopores tends to decrease gradually throughout curing. This phenomenon is probably due to the fact that microorganisms tend to accumulate in the finer pores and use the hydrolysis of urea to form urease near the small pore size to solidify calcium ions to produce calcium carbonate, resulting in a continuous reduction in pore volume. However, during the filling of the cementing solution, the percolation of the cementing solution within the soil will have a scouring effect on the macropores class of large pores, thus increasing the content of this pore. The volume of micropores was elevated during the OS2-5 stage, while the content of macropores decreased. This phenomenon is due to the deeper development of curing, the gradual increase in the connectivity between sand particles, the gradual decrease in the scouring effect of the cementing solution and the gradual increase in the precipitation of calcium carbonate filled by microorganisms in the macropores and mesopores with, resulting in a gradual decrease in their macropore size, leading to an increase in the volume share of micropores. This phenomenon is consistent with the conclusion reached by (Lei D 2022) in his tests on the curing of calcareous sands that the large pores of the sand column are filled into small pores as the number of grout cycles increases.
The addition of wheat straw powder to the WS column resulted in a change in the pore structure of the material and a significant increase in the volume of the micropores. The pore volume of 0.1-1.0µm does not change significantly during curing; the pore volume of 1.0–15.0µm shows an increasing trend except for the WS1-1 stage. The WS1-5 stage shows a rapid decrease in pore size in the interval 0.1–30 µm compared to the WS1-1 stage, while the content of other pore sizes does not change significantly, and the OS sand column shows the same regularity. This indicates that curing of desert aeolian sandy soil by the bacterial solution occurs preferentially in the micropores, before filling the mesopores and macropores. Between the WS1-1 and WS1-5 stages, the WS column macropores content decreased by 3.4%, while during the second round of bacterial curing the macropores content decreased by only 0.22%. For the mesopores, there was an increase in the mesopores content during the first round of microbial curing, but the increase was limited to 0.52% and when the second round of bacterial curing was carried out, the mesopores content gradually started to decrease.
3.4 OS and WS column MRI image analysis
Compared to microbial reinforcement of sand and gravel materials with larger grain sizes, desert aeolian sandy soil has characteristics such as small grain size and poor water retention (Bu C et al. 2022), which will affect the injection of microorganisms and cementing solutions, and is often accompanied by difficulties in grouting and uneven reinforcement effects during the curing process. To this problem, we conducted MRI tests on the longitudinal sections of the top and bottom of the sand column at each stage of the curing of the two sand columns to collect top and bottom NMR imaging data of the test sand columns to make a preliminary determination of the internal homogeneity of the microbial cured sand columns. The coloured bar on the right shows the relative intensity range of the 1H content of the fluid in the pore space. Black indicates that the imaged sand column pores contain no 1H atoms and therefore do not have an imaging signal, while red (high proton density) indicates more fluid in the pores and a stronger NMR imaging signal.
The combination of Fig. 8 and Fig. 9 shows that as the number of microbial cures increases, the signal of the bright spots at the upper and lower parts of both sand columns gradually decreases and the signal of the dark spots gradually increases. The pore water content in the saturated state of both sand columns is gradually decreasing under the effect of microbial mineralisation, indicating that the pore ratios at the top and bottom of both sand columns are gradually decreasing.
However, it is worth noting that there is some variability in the variability of pore space between the top and bottom of the OS column and the WS column. In the OS1-1 stage, the bright spot signal in the upper part of the OS column is significantly higher than the lower bright spot signal, and the red bright spots are more dense. This illustrates the higher porosity in the upper part of the original sand column than in the lower part of the sand column during the OS1-1 stage. With the continued addition of subsequent cementing solutions, there is still some imbalance between the bright spot signals of the upper and lower parts of the three sand columns OS1-5, OS2-1 and OS2-5, but the difference is significantly improved compared to the initial stage of curing OS1-1.
Comparing Fig. 8 with Fig. 9, we observed that the difference between the bright spots of the upper and lower MRI images of the WS column was not significant at all stages of microbial curing, indicating that the microbial curing of desert aeolian sandy soil by external doping with wheat straw powder could improve the inhomogeneity problem during the curing process. However, according to the results of the bright spot comparison between OS2-5 and WS2-5 at the final curing completion, the MRI image bright spot signal of the WS column with external wheat straw powder doping was more than that of the OS column, which indicated that the external straw powder doping increased the porosity of the sand column (Zhou X F et al. 2011).
Combining Table 5, the total peak areas of the OS and WS columns, it was found that the addition of straw powder to desert aeolian sandy soil increased the total peak area of WS1-1 by 11.7% compared to OS1-1 and WS2-5 by 26.1% compared to OS2-5. It indicates that the NMR measurement image analysis has similar results to the T2 spectra. It further indicates that the incorporation of straw powder can enhance the water-holding capacity of desert aeolian sandy soil during microbial curing, which is important for the survival and effective curing of bacteria in the arid and hot desert environment.
3.5Analysis of the variation in pore size structure at different heights of the sand column at various stages of curing
To address the variation of pore space in each part of the sand column in the height direction, we divided the cured sand column into three layers at the top, middle and bottom (as in Fig. 3(b)) and examined them using 1H-NMR. The T2 spectra of the upper, middle and lower layers of the microbially mineralised sand column were transformed according to formula 3 to plot the corresponding superimposed histograms of pore content, see Fig. 11, and the porosity of each layer is shown in Table 7 and Fig. 10.
| Type of sand column | Curing stage | Porosity ratio(%) | ||
| Upper | Middle | Lower | ||
| OS | 1–1 | 34.82 | 43.01 | 33.99 |
| 1–5 | 31.40 | 35.34 | 33.76 | |
| 2 − 1 | 28.64 | 31.90 | 29.50 | |
| 2–5 | 25.55 | 30.65 | 28.22 | |
| WS | 1–1 | 38.14 | 45.40 | 41.44 |
| 1–5 | 37.12 | 40.61 | 39.73 | |
| 2 − 1 | 34.00 | 38.12 | 37.03 | |
| 2–5 | 32.11 | 37.61 | 36.79 | |
As can be seen from Table 7 and Fig. 11, there are differences in the magnitude of pore decline in the longitudinal sections of the OS sand column after the MICP treatment. The pore content of the upper part of the OS decreases uniformly with increasing rounds of microbial grouting, with an average decrease of 3.09% in the porosity of the upper sand column at all stages. The rate of decline in porosity in the upper part of the sand column is relatively close between stages; the rate of change in pore space in the middle of the sand column slows down with increasing grouting frequency; the rate of decline in porosity in the lower part of the sand column is significantly greater after initial grouting, and the subsequent rate of decline in porosity is generally consistent.
Compared to the OS column, the magnitude of pore variation is relatively low in all parts of the WS sand column. From the whole process of microbial curing, the upper pores of the WS column behaved similarly to the upper pores of the OS column, both showing a more uniform downward trend, but the rate of pore filling was smaller than that of the OS column, with an average reduction in pore size of 2.01% at each stage. The change in pore size in the middle of the WS column was also similar to that of the OS column, but the reduction was smaller compared to the OS column, with the first curing WS1-1 and WS1-5 pore sizes decreasing by 4.8%, WS1-5 and WS2-1 pore sizes decreasing by 2.49%, and WS2-1 and WS2-5 pore sizes decreasing by 0.51% in the final stage. The lower pore changes in WS occur mainly in WS2-1, with an average decrease in porosity of 2.21% in WS1-1 and WS2-1 and a similarly small decrease of 0.24% in WS2-5. The change in porosity in the lower part of the WS column occurs mainly in the three phases WS2-1, WS1-1 and WS2-1, with an average decrease in porosity of 2.21% and a smaller decrease of 0.24% in WS2-5.
After the MICP treatment, the curing effect of both the sand column in its original form (OS) and the sand column mixed with wheat straw powder (WS), from the NMR test results, can be indicated from strong to weak curing effect of the sand column as the best curing effect in the upper part, followed by the lower part of the sand column, and the worst curing effect in the middle of the sand column. Analysed from a one-dimensional longitudinal perspective, when the sand column is reinforced using the top-to-bottom saturated grouting method, the bacteria continuously move towards the lower part of the sand column due to the continuous addition of the cementing solution, resulting in an uneven spatial distribution of bacteria, making the lower part of the sand column more bacterial than the upper part of the sand column, thus causing the lower part of the sand column to cure better. This is consistent with the findings of Huang R P et al. (2019) who studied microbial curing of inhomogeneous sands to obtain a higher production of calcium carbonate at the bottom of the trough than at the top. However, at the same time, this test used 2 rounds of slurry injection of bacterial solution to ensure the overall activity of the bacteria within the sand column. As the first round of MICP reduced the overall porosity of the sand column, at the end of the second addition of the bacterial solution pending the resting period, the accompanying addition of the cementing solution may have flocculated the microorganisms at the grouting port and rapidly produced calcium carbonate precipitation, thus causing blockage of the upper pores of the sand column and enhancing the denseness of the upper structure of the sand column, which eventually exhibited less pores in the upper part of the sand column than in the middle.
According to Fig. 11, the content of micropores and mesopores in the upper, middle and lower parts of the in original sand column OS showed a decreasing trend during the first round of microbial grouting and curing, with a particularly significant decrease of 4.76% and 3.56% in micropores in the middle and lower parts, respectively. However, the change in macropores content was not consistent across the sand column, with the upper and lower macropores content gradually increasing with the addition of the cementing solution, and the middle macropores content showing some decrease. This is probably due to the fact that although the cured sand column has a large overall porosity in the initial stage, the micropores and mesopores are less susceptible to the scouring action of the seepage, with CaCO3 preferentially generating crystals in the smaller pores. In contrast, due to the saturated grouting method used for the test pouring, the upper macropores may be subject to the scouring action of the solution, while bringing a small amount of fine sand particles with bacteria attached to the middle, which increases the curing efficiency in the middle and thus fills the middle macropores and makes them less. As the solute molecules continue to precipitate out of the cementing solution, the concentration of the lower cured cement decreases and the filling effect of CaCO3 in the macropores is negatively affected, while the scouring of the solution continues, eventually increasing the content of the lower macropores. During the second round of bacterial infusion curing, the micropores content in the upper and middle parts of the sand column showed some increase, while the micropores content in the lower part continued to decrease. It is worth noting that the content of mesopores and macropores in the upper and middle sections tended to decrease with the second round of curing, while the content of mesopores and macropores in the lower section did not change significantly.
For the straw powder-doped desert aeolian sandy soil, the pore variation in the upper part of the WS column is similar to that in the upper part of the OS column. However the change in pore size in the middle of the WS sand column was mainly concentrated in the first round of bacterial injection stage, with a large change in pore size in WS1-1 and WS1-5. The change in pore size was mainly distributed between 5µm-30µm pore size in micropores, with a 3.6% reduction in pore size in 5µm-15µm pore size and a 1.4% reduction in 15µm-30µm pore size. However, for WS2-1 and WS2-5 during the second round of bacterial curing, there was little change in the porosity of each pore size in the middle of the sand column. the content of micropores in the lower part of WS showed a gradual decrease during the first round of bacterial curing, while the content of mesopores and macropores showed a small increase. When the second round of bacterial curing was carried out, the content of micropores in the lower part of the WS sand column increased with the number of colloidal injection, while the content of mesopores and macropores gradually decreased.
3.6 Distribution of CaCO3 content in sand columns at various stages of curing
| Type of sand column | Curing stage | CaCO3 curing rate | ||
| Upper | Middle | Lower | ||
| OS | 1–1 | 7.38% | 6.25% | 7.47% |
| 1–5 | 13.25% | 11.23% | 13.67% | |
| 2 − 1 | 18.18% | 15.11% | 17.06% | |
| 2–5 | 21.03% | 17.34% | 19.19% | |
| WS | 1–1 | 6.02% | 5.56% | 5.80% |
| 1–5 | 12.68% | 10.34% | 11.52% | |
| 2 − 1 | 17.17% | 14.78% | 16.43% | |
| 2–5 | 19.01% | 16.65% | 18.41% | |
The tests used the acid wash method to measure the amount of calcium carbonate produced at each stage of curing for both sand columns in layers, as shown in Table 8 and drawn as Fig. 12.
From Table 8 and Fig. 12, it can be seen that the amount of calcium carbonate generated from the sand column specimens for the two curing regimes increases with the number of injections. Overall the WS calcium carbonate curing rate was less than OS. it can be inferred that the incorporation of straw powder reduced the calcium carbonate curing rate of the sand column under the test conditions, which corroborates with the results of the NMR analysis. The results of calcium carbonate curing rates in Table 8 show that WS calcium carbonate curing rates show WS -Upper > WS - Lower > WS -Middle; while OS columns show OS-Lower > OS-Upper > OS-Middle in the first grouting stage. After carrying out the second round of grouted bacterial curing, the internal calcium carbonate content was consistent with the WS column performance, OS-Upper > OS-Lower > OS-Middle. Because of the saturation addition method of grouting used for the experiments, the OS column has good overall permeability during the bacterial curing phase of the first round of grouting, with the cementing solution seeping from top to bottom along the soil pores and the gap between the sand column and the mould. And the bottom of the specimen discharges the liquid at a slower rate, the cementing solution in the pore space is subject to gravity and appears to aggregate at the bottom of the OS column, causing the slurry to stagnate at the bottom, resulting in a larger amount of calcium carbonate generation in the lower part. When the pores within the sand column were filled after the second round of bacterial refilling, the overall permeability decreased and the rapid deposition of calcium carbonate in the upper part of the sand column caused by the addition of the cementing solution blocked the pores in the upper part of the specimen, resulting in a much greater calcium carbonate production in the upper part than in the lower part. In comparison to the OS column data, the addition of straw powder to the WS column improved the uniformity of calcium carbonate precipitation produced by the microorganisms in the vertical longitudinal section, reducing the variability in the amount of calcium carbonate precipitation in the upper, middle and lower sections.
3.7 Analysis of the shear strength of sand columns at various stages of curing
The sand column curing experiment used top-down microbial injection, and the molecular diffusion and migration of the bacterial solution and solute along the direction of solution flow had a large effect on the spatial distribution of calcium carbonate. The inhomogeneity of the spatial distribution of calcium carbonate caused inhomogeneity in the filling of the microbially cured sand column with calcium carbonate. In this experiment, the cured sand column was divided into three parts: upper, middle and lower, and shear strength tests were carried out on OS and WS cured sand columns respectively, and the shear strength test results obtained are shown in Fig. 13.
The sand column curing experiment used top-down microbial injection, and the molecular diffusion and migration of the bacterial solution and solute along the direction of solution flow had a large effect on the spatial distribution of calcium carbonate. The inhomogeneity of the spatial distribution of calcium carbonate caused inhomogeneity in the filling of the microbially cured sand column with calcium carbonate. In this experiment, the cured sand column was divided into three parts: upper, middle and lower, and shear strength tests were carried out on OS and WS cured sand columns respectively, and the shear strength test results obtained are shown in Fig. 13.
As can be seen from Fig. 13, the shear strengths of the upper, middle and lower parts of the OS column and the modified WS column under the action of MICP showed an increasing trend with the increase of the number of treatments of the cementing solution. The shear strength of the lower part of the OS column was higher than the shear strength of the upper part when the first round of bacterial curing was carried out. When the second round of bacterial curing was carried out, the increase in shear strength of the upper structure was higher than the increase in shear strength of the lower part, and the shear strength of the upper part was greater than that of the lower part at this stage. From a strength point of view, comparing the shear strength of the two sand columns, we found that the addition of straw powder decreased the shear strength of each cured portion at different stages. The shear strength of the upper part of the WS column was the highest in all stages of curing, followed by the substructure and finally the central part. Looking at the development of shear strength between OS and WS columns, it can be found that the difference in shear strength between the upper, middle and lower parts of OS columns is greater than that of WS columns, and the shear strength of sand columns mixed with straw powder increases more gently under the action of MICP. This indicates that the addition of straw powder material to the desert wind-deposited sand reduces the curing strength of the sand column, but allows for an effective increase in the overall uniformity of microbial curing.
In this paper, from the pore characteristics of colluvial desert aeolian sandy soil, wheat straw flour was used to optimize the curing effect of desert aeolian sandy soil, and the changes in calcium carbonate curing rate and pore characteristics of colluvial sand columns under different curing regimes were analysed by low-field NMR techniques. The main findings and conclusions of this paper are summarised as follows.
-
The product of microbial curing calcium carbonate can fill the pores of desert aeolian sandy soil, cement the sand particles, change the pore structure, so that micro, small and large pores gradually reduced, the total porosity decreased. T2 relaxation time of OS and WS shows that with the increase of the number of curing treatment, the pores in the specimens are continuously filled to increase, the water content of the sand column in the saturated state decreased. The total peak area of the NMR signal volume corresponding to 1H decreased continuously, with 22.13% reduction in total peak area after curing of the original desert aeolian sandy soil and 12.09% reduction in total peak area after curing of the desert aeolian sandy soil mixed with straw powder. The addition of straw powder enhanced the total porosity of the sand column, especially the number of large pores (5–60µm) increased significantly.
-
The pore size distribution data of sand columns obtained by T2 relaxation time inversion of OS and WS showed that the pore sizes of OS and WS columns mainly ranged from 0µm-150µm, and the pore size distribution showed a bimodal pattern with the highest content of micropores, followed by mesopores and the lowest content of macropores. The straw powder changed the internal pore structure of the OS columns and increased the porosity. Compared to the OS column, the WS column showed an increased proportion of micropores and a decreased proportion of macropores in the pores. During the curing process of MICP for both sand columns, calcium carbonate crystals preferentially filled the pores of micropores and mesopores, and the filling effect of macropores pores was not obvious.
-
NMR images showed that the addition of straw powder increased the pore volume and pore size of the sand column pores, and the free water content in the sand column was significantly greater than that of the pure sand column, which obviously increased the water holding capacity of the cured sand column, and the increase in water holding capacity was of greater significance to the curing efficiency in the desert environment.
-
Sand column stratification tests showed that the rate of pore filling in the sand column decreased after mixing with straw powder, and the overall porosity change trend was relatively similar, improving the uniformity of calcium carbonate precipitation along the height direction of the WS sand column. The stratified shear strength showed that the change trend of the shear properties of each layer after the sand column was mixed with straw powder was basically the same, and the overall uniformity of the cured sand column was better, but the shear strength was lower than that of the original sand column.
Acknowledgements We are very grateful to the College of Water Resources and Civil Engineering of Xinjiang Agricultural University for providing us with the experimental platform.
Author contribution All authors contributed to the study conception and design. Manxu Zhou participated in laboratory experimentations, data analysis, and drafting the manuscript. Qiang Jin participated in the manuscript review and revision. Zihua Li and Fuhao Zhu made important suggestions. All authors read and approved the final manuscript.
Finding This research was supported by the National Natural fund [Grant no. 52269028]. [2019E0241] and the Postgraduate Research Innovation of Xinjiang Agricultural University [no. XJAUGRI2021013].
Data availability All data generated or analyzed during this study are included in this published article.
Ethics approval and consent to participate Not applicable.
Consent for publication I give my consent for the publication that the research details in the paper are to be published in this journal.
Competing interests All authors certify that there are no conflicts of interest regarding the publication of this paper.
- Al Imran M, Shinmura M, Nakashima K et al (2018) Effects of various factors on carbonate particle growth using ureolytic bacteria. Mater Trans 59:1520–1527. https://doi.org/10.2320/matertrans.M-M2018830
- Almajed A, Lateef MA, Moghal AAB et al (2021) State-of-the-art review of the applicability and challenges of microbial-induced calcite precipitation (MICP) and enzyme-induced calcite precipitation (EICP) techniques for geotechnical and geoenvironmental application. Crystals 11:370. https://doi.org/10.3390/cryst11040370
- Bu C, Lu X, Zhu D et al (2022) Soil improvement by microbially induced calcite precipitation (MICP): a review about mineralization mechanism, factors, and soil properties. Arab J Geosci 15:1–25. https://doi.org/10.1007/s12517-022-10012-w
- Canakci H, Sidik W, Kilic IH (2015) Effect of bacterial calcium carbonate precipitation on compressibility and shear strength of organic soil. Soils Found 55:1211–1221. https://doi.org/10.1016/j.sandf.2015.09.020
- Ng CWangW, Zhang Q, Zhou C et al (2022) Eco-geotechnics for human sustainability. Sci China Technological Sci 1–37. https://doi.org/10.1007/s11431-022-2174-9
- Duo L, Kan-liang T, Hui-li Z et al (2018) Experimental investigation of solidifying desert aeolian sand using microbially induced calcite precipitation. Constr Build Mater 172:251–262. https://doi.org/10.1016/j.conbuildmat.2018.03.255
- Ge X, Myers MT, Liu J et al (2021) Determining the transverse surface relaxivity of reservoir rocks: A critical review and perspective. Mar Pet Geol 126:104934. https://doi.org/10.1016/j.marpetgeo.2021.104934
- Guo JC, Zhou HY, Zeng J et al (2020) Advances in low-field nuclear magnetic resonance (NMR) technologies applied for characterization of pore space inside rocks: a critical review. Pet Sci 17:1281–1297. https://doi.org/10.1007/s12182-020-00488-0
- Hinedi ZR, Kabala ZJ, Skaggs TH et al (1993) Probing soil and aquifer material porosity with n-uclear magnetic resonance. Water Resour Res 29:3861–3866. https://doi.org/10.1029/93WR02302
- Hong WT, Jung YS, Kang S et al (2016) Estimation of soil-water characteristic curves in multiple-cycles using membrane and TDR system. Materials 9:1019. https://doi.org/10.3390/ma9121019
- Huang RP, Shao GH, Chen FY (2019) Morphological evolution characteristics of cementation area in sand solidified by single point biogroutin. J Forestry Eng 4:158–164. https://doi.org/10.13360/j.issn.2096-1359.2019.06.022
- Jaeger F, Bowe S, Van As H et al (2009) Evaluation of 1H NMR relaxometry for the assessment of pore-size distribution in soil samples. Eur J Soil Sci 60:1052–1064. https://doi.org/10.1111/j.1365-2389.2009.01192.x
- Jaeger F, Shchegolik hina A, Van As H et al (2010) Proton NMR relaxometry as a useful tool to evaluate swelling processes in peat soil. Open Magn Reson J 3:27–45. https://doi.org/10.2174/1874769801003010027
- Jiang NJ, Soga K (2016) The Applicability of Microbially Induced Calcite Precipitatio (MICP) for Internal Erosion Control in Gravel-Sand Mixtures. Geotechnique 67:42–55. https://doi.org/10.1680/jgeot.15.P.182
- Kong C, Wang MY, Shi XZ et al (2016) Study on water holding capacity and pore characteristics of soils based on LF-NMR. Acta Pedol Sin 53:1130–1137
- Lai JH, Zhang KW, Wei S, Wang YK, Xu XL, Qu JJ, Xiao JH (2017) Research advances and prospect in chemical sand-fixing materials. J Desert Res 37:644–665
- Larney FJ, Angers DA (2012) The role of organic amendments in soil reclamation: A review. Can J Soil Sci 92:19–38. https://doi.org/10.4141/cjss2010-064
- Lei DL (2022) Experimental study on meso-structures of MICP-treated calcareous sand based on CT image processing. Guangzhou University, Guangzhou China. https://doi.org/10.27040/d.cnki.ggzdu.2022.000123
- Li H, Tang CS, Yi LY et al (2021) Experimental study on surface erosion resistances and mechanical behavior of MICP-FR-treated calcareous sand. Chin J Geotech Eng 43:1941–1949
- Liu C, Lu M, Cui J et al (2014) Effects of straw carbon input on carbon dynamics in agricultural s-oils: a meta-analysis. Glob Change Biol 20:1366–1381. https://doi.org/10.1111/gcb.12517
- Mahawish A, Bouazza A, Gates WP (2019) Unconfined compressive strength and visualization of the microstructure of coarse sand subjected to different biocementation levels. J Ge-otechnical Geoenvironmental Eng 145:04019033. https://doi.org/10.1061/(ASCE)GT.1943-5606.0002066
- Qu JJ, Yu WB, Qin XB (2014) Wind-protecting Efficiency of HDPE Functional Sand-fixing Barriers. J Desert Res 34:1185–1119
- Shao MA (2006) Soil physics. Higher Education Press, Beijing, China
- Strandberg A, Skoglund N, Thyrel M (2021) Morphological characterisation of ash particles from co-combustion of sewage sludge and wheat straw with X-ray microtomography. Waste Manag 135:30–39. https://doi.org/10.1016/j.wasman.2021.08.019
- Tang CS, Yin L, Jiang N et al (2020) Factors affecting the performance of microbial-induced carbonate precipitation (MICP) treated soil: a review. Environ Earth Sci 79:1–23. https://doi.org/10.1007/s12665-020-8840-9
- Van Paassen LA, Ghose R, van der Linden TJM et al (2010) Quantifying biomediated ground improvement by ureolysis: large-scale biogrout experiment. J Geotech GeoEnviron Eng 136:1721–1728. https://doi.org/10.1061/(ASCE)GT.1943-5606.0000382
- Wang T (2016) Study on the coordinated development of ecosystem and socio-economic system in desertification control: a case study of desertification control in semiarid area in North China. Acta Ecol Sin 36:7045–7048. https://doi.org/10.5846/stxb201610172107
- Wang Y, Tong HW, Qiu RK, Yuan J (2020) Research on mechanical properties of rubberparticle-improved soil cemented by MICP. Industrial Constr 50:8–14. https://doi.org/10.13204/j.gyjzG20062207
- Whiffin VS, Van Paassen LA, Harkes MP (2007) Microbial carbonate precipitation as a soil imp-rovement technique. Geomicrobiol J 24:417–423. https://doi.org/10.1080/01490450701436505
- Xiong Y, Deng HF, Li JL, Cheng L, Zhu WX (2022) Experimental study on improving effect of microorganism solidifying sand by pozzolan. Rock and Soil Mechanics 43:1–13. https://doi.org/10.16285/j.rsm.2022.0134
- Yang G, Zuo S, Wu D et al (2021) Experimental study on the mechanical behavior and mesoscopic damage of Guiyang red clay during triaxial loading. Arab J Geosci 14:1–13. https://doi.org/10.1007/s12517-021-09234-1
- Yasuhara H, Neupane D, Hayashi K et al (2012) Experiments and predictions of physical properties of sand cemented by enzymatically-induced carbonate precipitation. Soils Found 52:539–549. https://doi.org/10.1016/j.sandf.2012.05.011
- Zhao H, Wu X, Huang Y et al (2021) Investigation of moisture transport in cement-based materials using low-field nuclear magnetic resonance imaging. Magazine of Concrete Research 73:252–270. https://doi.org/10.1680/jmacr.19.00211
- Zhou XF, Jie YH, Pei YH, Ge YH, Fen X (2011) Effect of Adding Straw Powder on the Physical and Chemical Properties of Saline Alkali Soil. Hubei Agricultural Sciences 50:4806–4808. https://doi.org/10.14088/j.cnki.issn0439-8114.2011.23.032