Jl Bing-yi ,ZHAO Chi-peng ,WU Yue ,HAN Wei ,SONG Ji-qing ,Bai Wen-bo

1 Institute of Environment and Sustainable Development in Agriculture,Chinese Academy of Agricultural Sciences,Beijing 100081,P.R.China

2 Liaoning Province Modern Agricultural Production Base and Construction Engineering Center,Shenyang 110033,P.R.China

3 Shandong General Station of Agricultural Technology Extension,Jinan 250100,P.R.China

Abstract Super-absorbent polymers (SAPs) are widely used chemical water-saving materials,which play an active role in the accumulation of soil water and the improvement of soil structure. Little is known about their performance with repeated usage or about factors influencing their efficiency under alternate wetting and drying cycles. In this study,various concentrations of SAP (0,0.1,0.2 and 0.3%) in soil following three continuous wetting and drying cycles (T1,T2 and T3),were studied to determine effects on soil structure stability and hydro-physical properties. The results indicated that the SAP improved soil water supply capacity under conditions of mild drought (T2) and sufficient irrigation(T3) at concentrations of 0.2 and 0.3%,but a reduction was observed under severe drought conditions (T1),which was negatively correlated with the SAP concentration. The physical adsorption of the SAP by soil and the chemical connection between the SAP and soil mineral colloids as Si-O-Si bonds,-OH bonds and different crystalline silica were the important factors that directly lead to the reduction of water retention capacities of the SAP with alternating wet and dry conditions. Compared with the control,the soil liquid phase ratios of the SAP treatments were increased by 8.8-202.7% in the T1 and T2 cycles,which would have led to a decrease in the soil air phase ratios. After repeated wetting and drying cycles,the SAP treatments increased the amount of ＞0.25 mm soil aggregates and the contents of water-stable macro-aggregate (R0.25),and decreased the amount of ＜0.053 mm soil aggregates,especially with higher concentrations of the SAP. Increases in mean weight diameter (MWD) and geometric mean diameter (GMD),and declines in fractal dimension (D) and unstable aggregates index (ELT) were all observed with the SAP treatments,which indicated an improvement in soil stability and structure. It was concluded that the distribution and stability of soil aggregates and soil water supply capacity was closely related to SAP concentration,soil moisture condition and the interaction between the SAP and soil particles.

Keywords: super-absorbent polymer (SAP),soil water,soil structure,soil aggregate,soil colloid

1.lntroduction

Super-absorbent polymers (SAPs) are high-molecular polymers with the ability to absorb a lot of water from the surrounding medium and retain the water in their network structures (Wuet al.2021). When they are added to the soil,the stored water can be released to the soil,which can increase soil moisture content,prevent soil erosion,reduce soil water evaporation and improve the yield of crops (Yanget al.2014;Houet al.2018;Guoet al.2019;Khanet al.2020). SAPs are crucial chemical agents used in agricultural water-saving technologies in drought areas,and have effectively solved problems of drought resistance and water conservation in those areas (Zhouet al.2012).

As ionic soil structure modifiers,SAPs change soil physical properties because of their strong waterabsorbing capacity and their volume changes during wetting and drying cycles (Aiet al.2021). The hydrophilic groups on the SAP structures have improved adsorption and flocculation effects on soil particles,allowing SAPs to adsorb the fine soil particles to form larger soil macroaggregates (Zhanget al.2014). Studies on the effects of SAPs on soil physical and chemical properties have focused on soil moisture content and bulk density (Baiet al.2010),soil porosity and texture (Yanget al.2018),infiltration performance (Hanet al.2013) and water evaporation rate (Zhang M Cet al.2019). However,the reasons for their ability to repeatedly absorb and release soil water and affect soil hydraulic parameters,and for the gradual decline in these effects are not clear.The specific mechanisms whereby SAPs influence the formation and stability of soil aggregates are also undefined. The positive effects of SAPs on the retention and supply capacity of soil water and crop water adsorption are still controversial. Agabaet al.(2010)indicated that SAPs could reduce the amount of water absorbed by a crop due to their fast water absorption rate,which was not beneficial to crop drought resistance.Baiet al.(2010) suggested that the water retention capacities of SAPs are related to soil moisture content;when soil moisture was lower than a critical threshold,the water retention capacities of SAPs would be sharply reduced.

The drying and wetting cycles of soils are a common phenomenon in arid and semi-arid areas. Many studies have observed the negative effects of alternate drying and wetting conditions on soil water retention capacity and the stability of soil aggregates (Chenet al.2018;Wenet al.2020). SAPs applied to the soil undergo repeated water absorption and rehydration under natural rainfall and evaporation processes,resulting in frequent shrinkage and expansion. Baiet al.(2010) showed that the positive effects of SAPs on soil water holding capacity would reduce after four continuous drying and wetting cycles.The long-term stability and effect of SAPs directly affect drought resistance and soil moisture conservation. Until now,the long-term effects of SAPs on soil structure and soil water holding and supply capacities have been rarely reported under alternating dry and wet conditions.

The main purpose of this study was to evaluate the effects of different concentrations of SAPs on soil structure and hydro-physical properties over three wetting and drying cycles. The functional groups and morphological characterizations of the SAPs and soil mixtures were analyzed to show the interaction between the SAPs and soil particles,and their influences on soil hydro-physical capacities. An additional objective was to determine if changes in soil-water retention and other soil properties following amendment with SAPs would persist for at least three wetting and drying cycles.

2.Materials and methods

2.1.SAP and soil

A commercial synthetic SAP with organic macromolecule polymer commonly used in agriculture and forestry in China was selected for this study. The basic characteristics and water absorbencies of the SAPs in different solutions are shown in Table 1.

The soil used in the study was sampled from an experimental field of the Institute of Environment and Sustainable Development in Agriculture (IEDA),Chinese Academy of Agricultural Sciences (CAAS) (116.65°E,40.13°N,altitude of 30 m). The soil was taken from the tillage layer (0-30 cm depth) and classified as a clay loam(45.1% sand,24.5% silt and 30.4% clay) according to Soil Taxonomy developed by the United States. The soil pH was 7.5,and bulk density and electric conductivity were 1.14 g cm-3and 768 μS cm-1,respectively. Soil organic matter and field capacity were 13.2 g kg-1and 30.5%,respectively. Soil total N,available P and available K were 2.34 g kg-1,56.2 mg kg-1and 81.2 mg kg-1,respectively.

2.2.Experimental methods

The study was carried out in a greenhouse at the IEDA,CAAS,from October 2018 to January 2019. Four SAP treatments,replicated five times,were organized in a randomized complete block design. Different amounts of the SAP were mixed with the soil samples to give concentrations of 0% (CK),0.1% (C1),0.2% (C2) and 0.3% (C3),expressed as the percentage of SAP to the weight of dry soil sample. A total of 3.5 kg of air-dried soil was added to plastic pots (diameter 14 cm,height 15 cm). Un-amended soil was first added to the bottom of the pots to a depth of 2 cm. Next,the SAP was mixed with the soil and the pots were filled with the mixture to a depth of 10 cm. The pots were fully irrigated to maintain a water depth of 2 cm on the surface of soil,making the soil samples fully saturated. The pots were then placed in a rain shelter and the soil water was allowed to evaporate naturally. On sunny days,the rain shelter was moved to the side,while on rainy days,it was repositioned. The pots were fully irrigated after each sampling,and three continuous wetting and drying cycles were conducted.Soil samples were collected from the pots on October 16(T1),October 26 (T2) and November 5 (T3). For each sampling,the relative soil water contents (RSWCs) of the CK treatments were 29.3% (T1),44.8% (T2) and 69.6% (T3),which simulated the soil conditions of severe drought (T1),mild drought (T2) and sufficient irrigation(T3),respectively.

2.3.Soil properties

Soil pH was measured using a soil-water ratio of 1:5.Gravimetric soil moisture content was calculated from mass loss after drying for 48 h at 105°C. RSWC was determined as the ratio of gravimetric soil moisture to field capacity.The moisture content at field capacity was measured with a pressure plate apparatus at -33 kPa (Cassel 1986). Soil bulk density (g cm-3) was determined by the core method(100 mm in diameter). Soil solid,liquid and gas contents were measured with a three-phase meter (DIK-1130,Daiki Rika Kogyo,Saitama,Japan). Soil organic matter was determined by the H2SO4-K2Cr2O7oxidation method (Nelson and Sommers 1982). Total N was determined by steam distillation after Kjeldahl digestion at 370°C. Available P was extracted with NaHCO3(Olsenet al.1954). Available K was extracted with 1 mol L-1NH4OAc (pH 7.0) solution and then determined by atomic absorption spectrometer(Jackson 1964).

2.4.Soil water characteristic curves

Soil water characteristic curves (SWCCs) were determined by the centrifuge method (Guoet al.2021).The soil samples were compressed with 100 cm3cutting rings and then soaked in distilled water for 48 h before weighing. Finally,the soil samples were centrifuged to simulate the gravimetric water content under different water suctions with various centrifuge speeds and the corresponding centrifugal equilibrium times using a highspeed refrigerated centrifuge (H-1400 pF,Kokusan,Kyoto,Japan). Centrifuge speeds and centrifugal equilibrium times corresponding to various soil water suctions are shown in Table 2.

Table 1 The basic characteristics and water absorbencies of the super-absorbent polymer (SAP)

Table 2 The centrifuge speeds and centrifugal equilibrium times corresponding to soil water suctions

The empirical formula proposed by Gardner (1958)was used to fit SWCCs:

whereθ(%) is the soil water content;S(kPa) is the soil water suction;a and b are parameters of the level and the trend of the SWCC,respectively.

The Gardner equation was used to derive the following formula for soil specific water capacity (SSWC):

whereC(θ) (mL g-1) is the soil specific water capacity.

2.5.Distribution and stability of soil aggregates

The distribution and stability of soil aggregates were determined by the dry sieving method (Elliott 1986) using sieves with pore sizes of 2,1,0.5,0.25 and 0.053 mm stacked from top (2 mm) to bottom (0.053 mm). Soil samples weighing 100 g were placed on the top sieve.The sieves were immersed in water for 5 min,and then vibrated for 5 min by an aggregate analyzer (XY-100,Beijing Xiangyu Weiye Instrument Equipment Co.,Ltd.,Beijing,China),at a speed of 40 times min-1,and an amplitude of 5 cm. The sieves were taken out of the water slowly and the aggregate size fractions of＜0.053 mm,0.053-0.25 mm,0.25-0.5 mm,0.5-2.0 mm and ＞2 mm were collected,dried at 55°C and weighed.

The water-stable macro-aggregate contents (R0.25) and unstable aggregates index (ELT) were calculated using the following formulae (Zhang D Bet al.2019):

whereR0.25(%) indicates the content of aggregates that are larger than 0.25 mm;Mr＞0.25(g) andMT(g) indicate the weight of aggregates that are larger than 0.25 mm,and the total mass of aggregates,respectively.

Mean weight diameter (MWD) and geometric mean diameter (GMD) were calculated by the following formulae(Houet al.2012):

Fractal dimension (D) of the aggregates was measured using the model of Saiediet al.(2017):

2.6.Characterization techniques

Attenuated total reflection fourier transform infrared technique (ATR-FTlR) analysisThe ATR-FTIR spectra were obtained using a Nicolet iZTM10 FTIR spectrometer(Thermo Fishe,Waltham,America) to determine the functional groups of a dry SAP sample (D-SAP),the unamended loamy soil (L) and the mixed SAP and soil samples (L-SAP). All samples were dried at 40°C,and placed directly on the germanium crystal. The samples were then compression molded into a thin layer of files at 230°C,with a pressure of 0.3 kPa. The samples were cooled to room temperature (23°C) in the mold under pressure and then scanned in the range of 400-4 000 cm-1.

Environmental scanning electron microscope (ESEM)analysisMorphological characterizations of the L-SAP samples were made using a Quanta 250 ESEM (FEI,Hillsboro,America) at an acceleration of 25 kV. The samples were sputtered with a thin layer (～20 nm) of gold prior to ESEM observation.

2.7.Data analysis

All reported values are means of the five replicates of each treatment. The RSWC,MWD,GMD,ELTandDdata were analyzed with a one-way analysis of variance(ANOVA) using the SAS 9.4 Software. Regression analysis was used to fit the SWC curves using SPSS 19.0 Software. The Duncan test was used to separate the means when the differences were significant (P＜0.05).

3.Results

3.1.Relative soil water content (RSWC)

Compared with the CK,the RSWC of soil with various SAP concentrations were increased to different degrees at T1 and T2 (Fig.1). At T1,the RSWC of SAP treatments were increased by 64.5% (C1),106.5% (C2)and 121.8% (C3). This meant that the soils received sufficient irrigation in the C2 and C3 treatments,while light drought was apparent in the C1 treatment. At T2,the RSWC of SAP treatments were increased by 49.6% (C2)and 73.2% (C3),while there was no significant difference between the C1 and CK treatments. At T3,no significant differences in RSWC were observed between the SAP treatments and the control.

3.2.Three-phase ratios of soil

Compared with the CK,the soil liquid phase ratios (SLPRs)in the SAP treatments were increased by 111.8% (C1),168.6% (C2) and 202.7% (C3) at T1,and by 17.9% (C1),73.9% (C2) and 88.5% (C3) at T2 (Fig.2). At T3,they were 12.3% (C1),8.8% (C2) and 17.2% (C3) greater than the CK treatment,but there were no significant differences in SLPRs between various SAP concentrations. The effects of SAPs on SLPRs were gradually decreased with increasing water absorption and dehydration. The trends of the soil air phase ratios (SAPRs) were opposite to those of the SLPRs,and their differences were relatively small.

3.3.Valence bond structure

The functional groups of different concentrations of the SAP and soil samples (L-SAP) and dry SAP (D-SAP)characterized by ATR-FTIR exhibited similar patterns(Fig.3). The characteristic absorption bands of loamy soil(L) in the peak area and fingerprint area (Ellerbrocket al.1999) were observed in all L-SAP treatments. The bands at approximately 1 018-1 097 cm-1were attributed to the Si-O-Si bonds in loamy soil,and those at approximately 783-804 cm-1to different crystalline silica (Farmer 1982).Additionally,broad peaks were apparent at 3 333 cm-1from the stretching vibration of -OH bonds in SAP treatments,except for the L-SAP-C3 treatment at T3. The intensities of spectral bands of L-SAP were significantly higher than those of the D-SAP. Moreover,the intensities of these absorption peaks increased with the increasing SAP concentrations at T1 and T2,and decreased with the process of dehydration and rehydration. These trends were the same as those of SLPRs and SAPRs. This indicated that the stronger reactions between the SAP and soil mineral colloids were observed with greater SAP concentrations,and the interactions were gradually weakened following continuous drying and wetting cycles.

For the ATR-FTIR spectra of the L-SAP-C3 treatment at T3,the original basic characteristic absorption peaks that represented the hydrophilic groups of the SAP,as-OH,C-H,C=O and -COO bonds (Pourjavadiet al.2008)disappeared,indicating that soil clay minerals might enter into the three-dimensional structure of the SAPs during the repeated dry and wet cycles. This might result in the damage of the SAP network structure,which might be an important reason for the reduction of its water absorbing and holding capacity.

3.4.Morphological characteristics

The ESEM visualizations of different L-SAP samples at 578-766× magnifications are shown in Fig.4. The different magnifications in all treatments were adjusted appropriately to obtain clear images due to the differences in morphology and thickness of the samples. Different degrees of folds were observed on the surface of L-SAP samples with alternate dehydration and rehydration. At T1,the surfaces of the L-SAP samples were compressed and uneven with all of the polymer concentrations,which might be induced by different types of fine clay particles attaching to the networks or surfaces of the SAP. The surfaces of the L-SAP were more coarse and irregular with increasing SAP concentration,which meant that stronger adsorption occurred between the SAP and soil particles (Fig.4-A-C). At T2 and T3,patchy soil clay particles were apparent on the ESEM visualizations of the L-SAP samples with prolonged reaction time,and there were more obvious wrinkles at C2 and C3 concentrations than those at T1 (Fig.4-E,F,H and I). Differences in clay particles that adhered to the SAP were apparent at T1,but no significant differences were observed at T2 and T3(Fig.4-D-I). Large amounts of SAP seemed to adsorb more soil clay particles,and such adsorption effects increased by repeated dehydration and rehydration. The differences in soil particles adsorbed by the SAP could be reduced gradually with prolonged reaction. An obvious rupture phenomenon on the SAP surface was observed,this might cause the reduction of water absorption and retention properties of the SAP.

3.5.Soil water characteristic curve (SWCC) and soil specific water capacity (SSWC)

The soil water characteristic curve (SWCC) is the curve describing how soil water suction changes with soil moisture content. The curve reflects the relationship between soil pore condition and soil water content.In this study,the SWCCs in all treatments changed similarly with different sampling stages,but the soil water contents changed significantly with changes in soil water suction (Fig.5). The soil water holding capacities were strong with high soil water suction,and the soil water content didn’t change significantly with increasing suction head. The changes in the SWCCs were steep,and the soil water maintained in small or medium-sized pores was difficult to dewater. The soil water holding capacities were weak where soil water suction was small,and the soil water content changed markedly with decrease in suction head. Where the changes of SWCCs were relatively gentle,the soil water was discharged through soil macropores,resulting in easy dehydration. With the same soil water suction,the soil water contents of the SAP treatments were almost larger than the CK treatment,especially at T1,and the increasing water contents were positively correlated with the SAP concentrations. The slopes of the SWCCs in the SAP treatments were more moderate than those of the CK and followed the order of C3＞C2＞C1≥CK.Under a certain soil water suction,the higher the SAP concentration,the stronger the soil water holding capacity,andvice versa. According to the SWCC changes at different sampling stages,the differences in soil water contents between various treatments declined following continuous wetting and drying cycles,indicating that the soil water retention capacity would decrease with the reuse of the SAP.

The SWCC data under different soil water suctions were fitted with the Gardner equation,and the corresponding expressions for different treatments are shown in Table 3. The correlation coefficients of each expression were larger than 0.98,which confirmed that the expressions could be used for quantitative analysis.In the expressions,parameter a represents the soil water holding capacity,the greater the value of a,the stronger the soil water holding capacity. Parameter b represents the trend of the water characteristic curve,that is,the change in soil water content with change in soil water suction. The larger the value of b,the steeper the curve and the greater the change in SWCC (Leong and Rahardjo 1997). Compared the parameters a with b,the more obvious effects of the SAP on soil water holding capacity were observed with greater SAP concentrations,and the effects weakened with increasing sampling time.At soil water suctions of 30,500 and 1 500 kPa,different soil water constants for field water holding capacity,effective water content and permanent wilting point were calculated using the SWCC expressions (Table 4). All of the above soil water constants for the SAP treatments were larger than those of the CK,and the differences between the SAP and control treatments increased with the SAP concentrations.

The soil specific water capacities (SSWCs) under different soil water suctions are shown in Table 5. SSWC decreased with increasing suction,and the reduced rates of SSWC were negatively correlated with soil water suction. At T1,when the soil water suction was lower than 100 kPa,the SSWCs under different treatments declined in the following order of CK＞C1＞C2＞C3,which indicated that the water supply capacity decreased with increasing SAP concentration. At T2 and T3,the SSWCs showed almost opposite trends to those observed at T1,and they increased by 4.0-29.9% (T2) and 1.0-11.6% (T3)with the SAP concentrations of C2 and C3. Combined with the changes in RSWC (Fig.1),it could be inferred that the SAP was not conducive to increasing the SSWC,and the SSWC would be reduced under extreme drought condition (T1). The soil water supply capacities were improved with SAP application under mild drought (T2)and sufficient irrigation (T3).

3.6.Distribution and stability of soil aggregates

The distribution of soil aggregates in CK treatment was changed during the process of dehydration and rehydration (Fig.6). At T1,the percentage of soil waterstable macro-aggregates (R0.25) in the SAP treatments were 15.7-33.3% greater than in the CK treatment,however there were no regular changes with the SAP concentrations. At T2 and T3,larger values ofR0.25were obtained with higher SAP concentrations. Application of the SAP increased the proportion of large aggregates(0.25-2.0 mm) above that of the CK treatment and sharply decreased the proportion of water-stable microaggregates less than 0.053 mm. The MWD and GMD values in all SAP treatments were significantly larger than those of the CK,and opposite changes ofELTandDwere observed,except at the SAP concentration of C1 at T3 (Table 6). Especially at T2,the MWD and GMD values increased significantly with increasing SAP concentration,while theELTandDvalues showed opposite trends.

Table 3 Soil water characteristic curve and specific water capacity expressions under different super-absorbent polymer (SAP)concentrations

Table 4 Effects of the super-absorbent polymers (SAPs)concentrations at three sampling times on various soil water constants

Table 5 Effects of the super-absorbent polymers (SAPs) concentrations on soil specific water capacities under different water suctions at three sampling times

4.Discussion

4.1.Functional groups and morphological characterizations of the SAP and soil mixtures

For most SAPs,the main chain or graft side chain contains strongly hydrophilic carboxyl groups and hydroxyl groups. These groups can ionize and combine with water molecules to form hydrogen bonds,allowing SAPs to absorb water in their network structures. Under natural alternating dry and wet cycles,SAPs applied to soil will produce frequent shrinkage and expansion,go through a series of decomposition and transformation phases,and interact with soil particles. In the present study,with drying and wetting cycles the RSWCs of the SAP and soil mixtures decreased gradually (Fig.1).This result is consistent with previous research (Baiet al.2010;Hanet al.2013;Banedjschafie and Durner 2015). The changes in physical and chemical structure of L-SAP samples were observed by ESEM and ATRFTIR analysis. It was found that a large amount of the soil mineral colloids were adsorbed on the SAP surface with the drying and wetting cycles,which destroyed the smoothness of the SAP (Fig.4). ATR-FTIR spectra showed that the Si-O-Si bonds,-OH bonds,and different crystalline silica that belonged to soil mineral colloids of the L samples appeared in the L-SAP samples,indicating that the soil mineral colloids had reacted with the SAP,resulting in changes in the functional groups (Fig.3). The SAP could absorb more soil mineral colloids following the interactions between the SAP and soil particles allowing the soil mineral colloids to enter into the SAP,destroying its molecular structure and reducing its water absorbing capacities (Fig.4-G-I). Yuet al.(2012) showed that the water release capacity of SAPs might be related to the contact between soil particles and SAP. Aiet al.(2021)found that soil moisture content played a role in water absorbing and releasing capacity of SAPs. This study provided direct evidence that the physical and chemical interactions between SAPs and soil particles was one of the most important factors influencing soil moisture holding capacity of soil treated with SAPs.

4.2.SAP effects on structure and stability of soil aggregates

The improvement of soil structure after application of the SAP might be due to the cementing effect of the SAP,granular shrinkage after water absorption and release cycles,and the promotion of crop root growth (Yanget al.2021). Previous studies have shown that the effect of application of a SAP on soil properties under natural rainfall and evaporation influences increased over time.In the present study,the proportions of 0.25-2.0 mm soil aggregates were mostly increased following three drying and wetting cycles (Fig.6),which indicated that the SAP enhanced the cementation effect on soil aggregates,especially the proportion of ＞2.0 mm soil aggregates.This result is consistent with that of Yanget al.(2021).Wanget al.(2009) also showed that the proportion of＞2.0 mm soil aggregates was significantly increased with an increase in SAP concentration,in agreement with our results,indicating that suitable amounts of SAPs can improve the proportion of soil aggregates.

In general,＞0.25 mm aggregates are referred to as water-stable macro-aggregates,while those of ＜0.25 mm aggregates are referred to as water-stable microaggregates. The greater the amount of water-stable macro-aggregates,the more stable the soil structure.MWD and GMD may accurately reflect the stability of soil aggregates with different grain sizes. Zhenget al.(2018) found that the larger the MWD and GMD values,the more stable the soil structure.Dis also an important tool to evaluate soil quality. The smaller theDvalue of aggregates,the more stable the soil structure is. TheR0.25,MWD and GMD values in all SAP treatments were significantly larger than those of the CK,and opposite changes ofELTandDwere observed (Table 6),which indicated that the proportion of ＞0.25 mm aggregates and the stability of soil aggregates were enhanced. Some studies were consistent with our results (Parsakhoo and Mostafa 2020;Sunet al.2020;Yanget al.2021),but they did not consider the differences in distribution and stability of soil aggregates during wetting and drying cycles. We found that the values ofR0.25,MWD,GMD,ELTandDdidn’t change regularly with different SAP concentrations at T1. However,larger values ofR0.25were obtained with higher SAP concentrations at T2 and T3. The proportion of ＜0.053 mm soil aggregates was inversely related to the change inR0.25at T2 and T3. It was speculated that the SAP induced bonding of the water-stable micro-aggregates (＜0.053 mm) to waterstable macro-aggregates (＞0.25 mm) particle sizes,reducing the possible damage to soil aggregates caused by the alternation of dry and wet cycles,and improving soil structure stability. After three drying and wetting cycles,there were no significant differences in MWD,GMD,ELTandDvalues between CK and C1 treatments,which indicated that small SAP concentrations had inconspicuous effects on soil aggregates. As discussed above,the effect of SAPs on the distribution and stability of soil aggregates were closely related to suitable SAP concentrations,and the interaction times between SAPs and soil particles. Yanget al.(2021) reported that excessive SAPs would accumulate between soil particles or macro-pores,and expand after water absorption,then the number of aggregates in soil would decrease resulting in deterioration of the soil structure. A preliminary conclusion was that the SAP improved the stability of soil aggregates,and induced greater uniformity of soil texture.The SAP concentration and the time for interaction between the SAP and the soil particles might be the main factors that affect the formation and stability of soil aggregates.

4.3.SAP effects on soil water holding and supply capacities

Many studies have shown that SAPs can significantly improve soil water holding capacity,and that there is a positive correlation between soil water holding capacity and SAP application (Yuet al.2012;Xi and Zhang 2021).Our former studies showed that the contributions of a small amount of SAP to soil water holding capacity was relatively small,and the ability to sustain soil water content was also weak (Baiet al.2013). We achieved similar results by visually observing the SWCCs in this study.The trends of the original SWCCs were not obviously changed with SAP application,but the gravimetric water contents were almost increased within 0-1 500 kPa,which might be related to the improved water holding capacity of the soil capillary pores. The slopes of the SWCCs in the SAP treatments were more moderate than in the CK treatment and changed in the order of C3＞C2＞C1≥CK(Fig.5). This indicated that when the soil water suction decreased,the SSWCs of the SAP treatments decreased more slowly than that of the CK. By fitting SWCCs,we further verified that the higher the SAP concentration,the greater the change in the soil water holding capacity,and that the change was reduced by drying and wetting cycles. The SAP increased the effective water content and permanent wilting point,and such soil water constants were positively correlated with the SAP concentrations.These results were consistent with that found by Vesnaet al.(2013) in a field experiment.

SSWCs are one of the important indicies for evaluating soil water supply capacity and drought resistance. In this study,when the soil water suction was less than 100 kPa,the SAP failed to increase the SSWCs,and reduce the soil water supply capacity under the condition of severe drought (T1). Under slight drought (T2) or sufficient irrigation (T3),the SAP was helpful in improving the soil water supply capacity. Under severe drought,although SAP application could increase the soil water holding capacity,the water absorption rate of the SAP was greater than that of soil,and the limited water in the SAP and soil mixture might have been preferentially absorbed and maintained in the SAP three-dimensional network structure,so it was difficult to effectively supply water to the soil for timely drought mitigation. The risk of water competition between SAP and soil intensified with increased SAP concentration. Under slight drought or sufficient irrigation,the water could satisfy the requirement of both SAP and soil,and the soil water supply capacity was improved to different degrees. Therefore,when applying SAP for increasing soil water supply capacity,attention should be paid to both the applied concentration and the soil moisture condition. Our former study found that maintaining adequate soil moisture was a precondition for water retention and improvement of soil properties with SAP application in agriculture (Baiet al.2010). Further research is need to focus on the above factors.

4.4.SAP effects on the three-phase ratio of soil

The three-phase ratio of soil is the volume ratio of soil solid,liquid and gas phases,and it is often taken as a measure of soil quality in place of properties,such as porosity,soil moisture and hydraulic conductivity. Maintenance of favorable soil three-phase ratios is one of the most important goals in agriculture. The application of SAPs can change soil moisture characteristics,soil pore size and the soil pore network (Ostrandet al.2020). SAPs may increase soil moisture and decrease soil permeability,this might be dependent on the SAP swelling effect after absorbing water,resulting in compaction of surrounding soil particles and the reduction of soil permeability (Guoet al.2019). In our study,SAP application noticeably increased the SLPRs,resulting in relative decreases in the SAPRs. Such changes were positively related to SAP concentrations,and negatively correlated with the number of drying and wetting cycles. Combining these results with the differences in SWCCs and proportions of soil aggregates between different treatments further confirmed that the SAP could significantly increase the soil water holding capacity in this experiment (Fig.5),and the numbers of soil macropores were reduced,while the medium and small pores were increased with SAP application,resulting in more uniform distribution of soil pores. Wuet al.(2021) found that SAPs could effectively obstruct the flow of the free water existing in the soil and maintain a relatively uniform distribution of soil moisture.Over time,part of the SAP gradually penetrate the pores within soil aggregates,and the SAP concentration affected their ability penetrate soil pores (Sojkaet al.2007). When the application concentrations of SAPs exceed a certain threshold,the SAPs would absorb water and expand,which could block soil pores and cause soil particles to adhere together,and also affect the circulation and exchange of water and air in the soil (Yanget al.2021). It is well understood that SAPs are suitable for soil with more macrospores,and the soil pores are more uniform to satisfy the root growth of crops (Kentaet al.2019).

5.Conclusion

This study simulated the effects of a SAP on soil structure and some soil hydro-physical properties under severe water deficit (T1),slight water deficit (T2) and sufficient soil moisture (T3),with the SAP concentrations of 0%(CK),0.1% (C1),0.2% (C2) and 0.3% (C3). Following continuous wetting and drying,SAP treatments increased RSWCs by 10.3-121.8% above the CK except at T3.The soil water supply capacity was reduced under severe drought,and improved by 1.0-29.9% at higher SAP concentrations under mild drought and sufficient irrigation conditions. Attention should be paid to SAP application concentration and soil moisture condition,when applying SAPs for increasing soil water supply capacity. Through varying valence bond structures and morphological characteristics of the SAP and soil mixtures,it was shown that the physical and chemical reactions between the SAP and soil mineral colloids,such as Si-O-Si bonds,-OH bonds and different crystalline silica were important factors that changed the soil water holding capacity with SAP application. SLPRs were increased significantly and SAPRs were reduced with SAP treatments,and higher SAP concentrations resulted in more marked changes in the three-phase ratios of soil. The SAP applied in this study is recommend for soil with large pores. Excessive SAP application would aggravate the deterioration of soil permeability. Compared with the CK,the proportions of large soil aggregates (0.25-2.0 mm) were increased,while the small aggregates (＜0.053 mm) were decreased gradually with SAP application,which indicated that the SAP mainly adhered to ＜0.053 mm soil aggregates to ＞0.25 mm soil water-stable aggregates. The MWD,GMD,DandELTvalues of soil aggregates were accurate and reliable means by which to characterize soil structure stability,and higher stability was observed with SAP application. The effect of the SAP on the stability of soil structure was closely related to SAP concentration and the interaction between the SAP and soil particles. It was found that even under a relatively extreme dry-wet alternate cycle,SAP treatment improved soil moisture and water supply capacity,increased the soil water-stable macro-aggregate proportion,and enhanced soil structure stability. The results indicated that reasonable application of SAPs could improve the buffer power of soil against wetting and drying in arid and semi-arid farmland. The SAP is not recommended for application in severe drought conditions. Application of the SAP at a concentration of 0.3% is recommended as this concentration seemed suitable under alternating wet and dry conditions in this experiment. Further studies should determine longterm influences of SAPs application on water supply capacity,porosity,and soil structure stability,as well as an economic evaluation in practice,so as to supply a more rational and scientific basis of applying SAPs to farmland.

Acknowledgements

This study was supported by the National Natural Science Foundation of China (41601226) and the Agricultural Science and Technology Innovation Project of the Chinese Academy of Agricultural Sciences.

Declaration of competing interest

The authors declare that they have no conflict of interest.