Milk vetch (Astragalus sinicus L.) seeds for the experiment were purchased from Royal Garden Greening Engineering Co., Ltd. On 8 October 2021, we sowed 5 g of milk vetch seeds in the substrate [Jiffy^®, consisting of 3 parts of peat soil and coconut husk, 1 part of vermiculite, and 1 of part perlite (v:v)] with a volume of 60 × 35 × 8 cm and placed them in an artificial climate incubator (temperature, 23 ± 2°C; humidity, 75%; 12-h/12-h light-dark cycle), keeping the substrate soil moist. To keep in line with the growing habits of the field, no fertilizer treatment was applied during the growth of milk vetch seedlings. The seedlings were planted in Jiffy substrate before harvest. On 8 December of the same year, before the milk vetch flowered, the aboveground parts of milk vetch were harvested with fresh and dry yields of 1.01 and 0.32 Kg m^-2, respectively. The harvested milk vetch was vacuum freeze-drying at −50°C and 100 Pa for 48 h and crushed; the powder was sealed and stored in a −20°C refrigerator for later use. Goosegrass seeds were collected in the Experimental Base of South China Agricultural University (23°14′18.42″ N, 113°38′8.06″ E).

We used milk vetch straw powder as material, mixed it according to the following ratio: powder, fresh paddy soil, and ultra-pure water (1:1:30 by weight), and put them into a plastic bottle with an inner lid. Soil solution without powder was used as the control (CK), and each treatment set three duplications. The bottle was placed in a constant temperature shaker at 28°C for 200 rpm for 12 h, 9 days, 12 days, 15 days, and 18 days to obtain the decomposition leachates. All the decomposition leachates were stored in a −40°C refrigerator and were used for the test together after the last decomposition.

In the goosegrass-suppressed active substance identification assay, we determined three duplications of each time point decomposition leachates, and the average of the chromatographic signals from the three replicates at each time point was used to re-plot the signal intensity of the substance content. In the bioassay test, we determined six germination indicators, i.e., germination potential, germination rate, germination index, vital index, plant height, and fresh weight. Ten seedlings with uniform growth were selected from each treatment to measure plant height. Twenty plants were randomly selected from three of the five replicates to determine fresh plant weight (the number of samples was reduced by a certain proportion, and the final result was multiplied by the corresponding multiple after weighing when the sample was insufficient). Germination-related indicators were calculated according to the following formula:

where Gr is the germination rate, Gp is the germination potential, Gi is the germination index, and Vi is the vital index.

The allelopathic effects of milk vetch were evaluated using means of response index (RI). When T ≥ C, R I = 1 − C T ; T < C, R I = T C − 1 (C: the result of CK; T: the result of treatment, use the mean values as the result). RI > 0, indicating allelopathic promotion; RI < 0, indicating allelopathic inhibition; absolute value was consistent with the degree of allelopathy (Williamson and Richardson, 1988). RI accumulate value was the sum of six RI values of germination indicators. RI accumulate values of each indicator were calculated, and the allelopathic effects of milk vetch were evaluated comprehensively. RI accumulate values reach −6, indicating that germination and growth of goosegrass were completely inhibited.

Replicated experiments were conducted in a completely randomized design. For the parameters of goosegrass in vitro trials, determination of total phenolic content, and antioxidant tests, statistical significance was analyzed by one-way ANOVA. We verified normality and homogeneity of variance using the Shapiro–Wilk test and Levene’s test, respectively. Significant differences were further compared using the post-hoc Fisher LSD test. SPSS 24 was used for significance analysis of the data, which were expressed as mean ± standard deviation. GraphPad Prism 7 was used to plot. For parameters of pH and EC between CK and decomposition leachates, the statistical significance of treatment effects was determined by independent samples Student’s t-test (P ≤ 0.05)

The pH value of the decomposition leachates decreased by 24.1%, 37.8%, 40.3%, 47.2%, and 43.2% at 12 h, 9 days, 12 days, 15 days, and 18 days, respectively, compared with that of the CK ( Table 1 ). The EC of the 12-h decomposition leachate was 10.5% higher than that of the CK. In contrast, the EC of 9-day decomposition leachate was 23.7% lower than that of CK. The EC of 12- to 18-day decomposition leachates showed no differences compared with that of CK. The increase in EC at 12-h decomposition leachate was mainly due to the release of chemical elements from milk vetch, whereas the subsequent decrease may be due to the uptake of the elements by the colony and the chelation of certain elements such as manganese ions by organic substances such as oxalic acid and citric acid. The trend in EC changes was consistent with the findings of the previous study (Liu et al., 2022). Researchers found that in the paddy-upland rotation system, milk vetch incorporation as green manure decreased soil pH value, which is consistent with the findings of this study (Zhong et al., 2021). We estimated the lowest pH value at 15 days may be related to the increased acid content, whereas the increased pH value at 18 days may be related to the decrease in acid, and whether the pH can be restored to its pre-decomposition state needs to be further investigated.

As shown in Table 2 and Figure S1 , 4-ethylphenol, N-acrylimorpholine, and allyl isothiocyanate appeared on the 9 days and in subsequent leachates. 2-Hydroxyethyl acrylate was present in the 12-h decomposition leachates, with a concentration of 1.0 µg ml⁻¹. The highest concentrations of 4-ethylphenol and 2-hydroxyethyl acrylate at different decomposition times were at 15 days, with concentrations of 20.2 and 127.1 µg ml⁻¹, respectively, whereas the highest concentrations for allyl isothiocyanate and N-acrylimorpholine were on 18 days, with concentrations of 32.4 and 247.2 µg ml⁻¹, respectively. Goosegrass-suppressed active substances have a pattern of “Hormesis effect”, i.e., dose-dependent, which has been reported in related articles (Álvarez-Iglesias et al., 2018; Eladel et al., 2019; Ojija et al., 2019). The green manure straw takes time to release the goosegrass-resistant potential substances and these substances may be produced with concomitant degradation, which gives rise to a peak in concentration, preceded by a gradual increase in goosegrass suppression potential and followed by a decrease. In a study exploring the effects of Brassica napus decomposition on the growth of mung beans, the decomposition period changed the phytotoxic effect of residues; more inhibitory effects were shown at 14 days of decomposition, whereas decomposition for 21 days reduced the inhibitory effect (Mehmood et al., 2018). Puig et al. (2013) suggested adopting a relay planting of maize after 12 to 15 days from eucalyptus incorporation, as the early decomposition of eucalyptus leaves returned as green manure inhibits maize growth. The decomposition patterns, and the persistence of phytotoxic chemicals differ between plants. Researchers found a rule in the cover crop decomposition studies that the phenolic acids released from cereal rye increased in soils during the first 3–7 days after cereal rye termination and then decreased to initial concentrations after 56 days (Otte et al., 2020). Previous studies have shown that the return of milk vetch to the field promoted crop growth, so our focus should be on the point in time when the concentration of goosegrass-resistant active substances in the milk vetch decomposition leachate was at its highest.

All four identified compounds, i.e., 4-ethylphenol, allyl isothiocyanate, N-acrylimorpholine, and 2-hydroxyethyl acrylate, inhibited the growth of goosegrass to varying degrees at high concentrations (≥30%), especially N-acrylimorpholine and 2-hydroxyethyl acrylate ( Pictures S1C, D ). The weaker inhibition of goosegrass by allyl isothiocyanate and 4-ethylphenol ( Pictures S1A, B ) may be related to their lower concentrations. The four mixtures inhibited the goosegrass germination rate of 14.4% and 41.3% at 2% and 5% decomposition leachate treatments, respectively, which was stronger than the same concentration of the single compound ( Picture S1E ).

4-Ethylphenol is a strong odorant produced by microbial activity that is described as off flavors in several foods (Santamaría et al., 2018), especially wine production (Nieto-Rojo et al., 2014; Branco et al., 2021). Low concentrations (0.2 mM) of 4-ethylphenol promote crop growth and have the ability to disrupt the cell membranes of pathogens, so it can be used as an oomycete biological control agent for soybean root rot and tobacco black shank diseases (Ge et al., 2021). 4-Ethylphenol at concentrations ≥ 2% had a significant inhibitory effect on goosegrass germination ( Figures 1I–IV, A ), especially at high concentrations (≥ 30%), but the plant height was not affected at 2% concentration ( Figure 1V, A ). Treatment with 4-ethylphenol at 2% concentration increased the fresh weight of goosegrass seedlings by 13.3% compared with CK ( Figure 1VI, A ), which was inhibited with increasing concentrations, with 37.4% inhibition at 80% concentration.

Allyl isothiocyanate is a naturally occurring plant volatiles with insecticidal properties and can be 100% inhibition against larval emergence from spotted-wing drosophila eggs in blueberry fruits after 24 h of exposure under an air level of 0.5 µl L⁻¹ (Jabeen et al., 2021). Using a novel analysis of protein target-assisted computational tools for homology modelling, the researchers found that the antibacterial potency of Brassica nigra essential oil is due to the preferable binding efciency of allyl isothiocyanate, the major active ingredient of B.nigra essential oil (Mandal et al., 2021). In this study, allyl isothiocyanate elevated goosegrass germination at concentrations ≤ 30%, with the strongest promotion of 44.3% at 5% treatment ( Figures 1I–IV, B ), whereas the germination index, fresh weight, and plant height ( Figures 1III,V,VI, B ) of goosegrass were inhibited by 41.6%, 25.7%, and 12.6%, respectively, at 80% concentration.

N-acrylimorpholine is a water-soluble compound with an unsaturated carbon-carbon double bond and is an important raw material for the synthesis of polymers. Because of its non-toxic nature and good biocompatibility, homopolymers of N-acrylomorpholine can be used as drug retardants, water treatment agents, and cosmetic support agents (Wang and Zhang, 2003). Dimethomorph, a new type of fungicide, for the control of diseases caused by fungi of the oomycetes of the Subphylum Flagellaria (Lu et al., 2019), has a similar structure to N-acrylimorpholine. The treatment with a 2% concentration N-acrylimorpholine elevated germination-related indicators, except for the vital index of goosegrass ( Figures 1I–III, C ). Briefly, the germination rate, germination potential, and germination index of goosegrass were improved by 9.7%, 8.1%, and 9.3%, respectively, and the vital index was inhibited by 4.1%. At 30% concentrations treatment, all germination indicators, plant height, and fresh weight of goosegrass were significantly (P < 0.05) inhibited. Briefly, the germination rate, germination potential, germination index, vital index, plant height, and plant weight of goosegrass were inhibited by 78.8%, 84.7%, 84.3%, 90.5%, 39.7%, and 27.7%, respectively. One hundred percent of the inhibition of goosegrass germination was recorded at 80% concentration of N-acrylimorpholine treatment ( Figures 1I–V, C ).

2-Hydroxyethyl acrylate is a raw material for the development of high-performance composite biomaterials in the materials field (Bahmani et al., 2019). No study is showing that 2-hydroxyethyl acrylate inhibits weed growth. In this study, 2-hydroxyethyl acrylate did not affect goosegrass germination at 2% concentration treatment ( Figures 1I–IV, D ). Compared with CK, treatment with a 5% concentration of 2-hydroxyethyl acrylate increased the germination potential and germination index of goosegrass by 21.2% and 14.3%, respectively ( Figures 1I,III, D ), but the germination rate and vital index were inhibited by 4.8% and 6.5%, respectively ( Figures 1II,IV, D ). 2-Hydroxyethyl acrylate inhibited goosegrass fresh weight, as reflected by 18.5% and 17.4% decreased goosegrass fresh weight at 2% and 5% concentration treatments, respectively ( Figure 1VI ). Similarly, the plant height of goosegrass decreased with increasing concentrations of 2-hydroxyethyl acrylate ( Figure 1V, D ), with 5.9% and 18.2% reduction in plant height for 2% and 5% concentrations, respectively.

The mixture of the four components inhibited the germination and growth of goosegrass, with all concentrations except for the 2% treatment, which did not inhibit the germination potential and germination index of goosegrass. The germination and growth of goosegrass were completely inhibited at concentrations of ≥ 30% of the four components ( Figures 1I–VI, E ).

The RI accumulation value plot shows that 4-ethylphenol ( Figure 2A ) inhibited the growth of goosegrass as the concentration increased. Concentrations of allyl isothiocyanate ( Figure 2B ) up to 30% improve germination and growth of goosegrass but inhibit it at 80%. The growth inhibition in goosegrass by N-acrylimorpholine ( Figure 2C ) increased with increasing concentration. The RI values of 2% and 5% 2-hydroxyethyl acrylate treatments were −0.17, whereas the RI values of 30% and 80% treatment reached −6, which suggested that the goosegrass suppression potential of 2-hydroxyethyl acrylate ( Figure 2D ) was much stronger at high concentrations than at low concentrations. The inhibitory effect of the mixture of the four components ( Figure 2E ) on the growth of goosegrass increased with increasing concentration.

Liquid chromatography (LC) analysis revealed that the identified acid fractions showed variation at different decomposition times ( Figures S2 , S3 ). The highest total acid content was found in the 12-day decomposition leachate at 4147.7 µg ml⁻¹, followed by 12 h (3,162.5 µg ml⁻¹) and 15 days (3,141.9 µg ml⁻¹), and the lowest at 1,868.1 µg ml⁻¹ for 9 days. The highest oxalic acid concentration of 341.2 µg ml⁻¹ was recorded in the 12-h treatment and fluctuated with increasing decomposition time, with the lowest concentration of 43.2 µg ml⁻¹ at 9 days. The highest tartaric acid content of 429.1 µg ml⁻¹ was found in the 12-day decomposition leachates, followed by the 12-h decomposition leachates (279.0 µg ml⁻¹), whereas no tartaric acid was detected in the 18-day decomposition. Within 15 days, the acetic acid concentrations increased with increased decomposition time, from 740.3 to 2,308.1 µg ml⁻¹, and decreased to 2,126.4 µg ml⁻¹ on 18 days. The highest citric acid content of 1,436.4 µg ml⁻¹ was recorded in the 12-h decomposition leachate, followed by 12-day decomposition leachate, with a concentration of 420.1 µg ml⁻¹. The highest level of succinic acid was found in the 12-h decomposition leachate, reaching 348.1 µg ml⁻¹, followed by 12 days (246.1 µg ml⁻¹), and the lowest concentration was found in 18 days, at 64.4 µg ml⁻¹. No fumaric acid was detected in the 12-h decomposition solution, and the highest fumaric acid content was recorded in the 15-day decomposition solution, reaching 1.1 µg ml⁻¹. Within 12 days, the propionic acid content increased with increasing decay time, with the highest propionic acid content (1335.1 µg ml⁻¹) in the 12-day decomposition leachate. Beyond 12 days, the propionic acid content of the decomposition leachates decreased, with 53.9 µg ml⁻¹ of propionic acid at 18 days. The tannic acid content was highest in the 12-day decomposition leachates, reaching 3.0 µg ml⁻¹, followed by the 12-h, 9-day, and 15-day decomposition leachates, with concentrations of 2.1, 1.8, and 0.8 µg ml⁻¹, respectively, and no tannic acid was detected in the 18-day decomposition leachate ( Table 3 ). From the data on acid content and type, we can tentatively conclude that the decrease in soil pH caused by the return of milk vetch to the field in the previous study is related to the acid produced by decomposition. The incomplete agreement may be due to the unmeasured acid fractions and acidic substances such as 4-ethylphenol, as well as the magnitude of acidity between the acid fractions.

The acid fractions measured in this experiment are consistent with Gu et al. (2021); the proportion of citric, malic, and oxalic acids in the total organic acids of six green manure crop decompositions was closely related to the activation level of Al-P; the proportion of tartaric acid was related to the activation level of Fe-P, whereas milk vetch, which contains higher amounts of oxalic, citric, and succinic acids, had stronger activation of Al-P (Gu et al., 2021). Similar studies have shown that the return of leguminous green manures such as milk vetch and hairy vetch to the field can reduce soil pH at the initial stage and rebounded at a later stage; the remaining three types of green manure, however, do not lead to such a result (Su et al., 2022b), which is consistence with our study. Within reasonable limits, a lower pH value facilitates nutrient uptake by plants. Woody composts are more suitable for use in growing media due to their lower pH, EC, and inorganic C content comparison with regular green manure composts (Vandecasteele et al., 2022). This study was carried out under laboratory conditions and further research is needed to determine whether it has an effect on soil pH in the field.

Green manure decomposition is a simultaneous process of decomposition and synthesis, as shown by the changes in the content of the four goosegrass-resistant active substances and the acid fractions. The single components of the four goosegrass-resistant active substances have less effect on the germination and growth of goosegrass compared with the mixture treatment and do not fully reproduce the effect of the decomposition leachate treatment. In experiments exploring possible synergies between volatile organic compounds (VOCs) and water-soluble compounds released into the soil matrix by Cytisus scoparius, the presence of the aqueous extract significantly increased the phytotoxicity of volatile on weed root growth compared with the volatiles emitted alone. In addition, soil enhanced the synergistic interactions among VOCs and water-soluble compounds (Pardo Muras et al., 2022). In this study, the presence of the high concentration of phenolic and antioxidant activity in the decomposition leachates and therefore whether the presence of these compounds may reduce the goosegrass suppression effect of the four goosegrass-resistant active substances need to be further investigated, as well as whether the organic acids present in the decomposition leachates may also affect the effect of goosegrass-resistant and their effect on the validity of nutrients in the soil and therefore crop growth.