As a major legume crop in the world, soybean (Glycine max L.) is considered for its high protein content, balanced amino acid profile and broad use in human and animal nutrition (Rizzo et al., 2018). Leguminous crops’ ability to fix nitrogen biologically (BNF) has attracted attention as a nitrogen supply, thus it can reduce the need for N fertiliser to maintain crop yields. BNF from legumes, either alone or in conjunction with fertiliser, can be used as a sustainable farming method (Iannetta et al., 2021). Although soybean may symbiotically fix nitrogen (BNF) with Bradyrhizobium spp. to meet a large portion of their nitrogen needs, the effective-ness of this process depends on a number of agronomic and environmental conditions (Szpunar-Krok et al., 2023). Among these, root nodulation and nitrogen uptake are significantly influenced by soil temperature, water availability and nitrogen fertilisation (Zhang et al., 2023). Both native soil populations and commercial inoculants given to seeds during planting are the sources of rhizobia that colonise legume root nodules (Tamagno et al., 2018). At present, inoculation with effective strains enables the soybean plant to get all of its nitrogen requirements from BNF (Getachew and Dagnaw, 2020). This is a natural process of major relevance in global agriculture and has enormous potential for smallholder farmers in Vietnam. Leguminous crops like soybean, common bean, chickpea and peanut also exhibit extraordinary growth and yield response to rhizobia inoculations in various agro-ecologies, according to many field demonstrations conducted in Vietnam (Doan et al., 2017). It is important to emphasise that there are still insufficient B. japonicum populations in coastal sandy soils, thus farmers or seed manufacturers must inoculate their own seed. Relatively few studies have examined the relationships between split N fertiliser application and rhizobium inoculation on nitrogen fixation in field-grown soybeans in Central Vietnam under coastal sandy soil conditions, most of studies were only conducted in the North and the South of Vietnam for a long time ago (Hiep et al., 2002; Doan et al., 2017). Hence, this study was implemented to fill in existing knowledge gaps about the split N fertilisation and rhizobia inoculation combination and nodulation efficiency as well as N₂ fixation from soybean in this area.

Description of the research area
 
Field experiments were carried out in Quang Loi commune (old) (16.60.34^oN, 107.48.18^oE), Hue city, during spring season (February to June) and summer season (June - August), 2025. The FAO/UNESCO soil classification indicates that Haplic Arenosols is the predominant soil type in the area which > 80% sandy content (light texture soil) and poor nutrients (OC <1%; N <0.05% and CEC <5 cmolc kg^-1) (Hoang et al., 2019).
 
Experimental set up
 
Rhizobium strain (10⁹ CFU g^-1), a well-known commercial rhizobial inoculant provided by SIAMB (Vietnam), was employed in the experiment with the soybean variety namely DT90. Split times of N fertiliser application (1, 2, 3 application times in corresponding with 1 time (at basal), 2 times (at basal and top dressing at 2-3 fully leaves - 12 DAS) and 3 times (at basal, top dressing at 2-3 fully leaves (12 DAS), 5-6 fully leaves -35 DAS were evaluated as subplot treatments and two inoculation levels (inoculated (1) and non-inoculated (0)) were used as the main plot in the split-plot design with three replications. For every plot, two weeks before sowing, 500 kg ha^-1 lime (CaCO₃-56% CaO) was broadcast and mixed into the soil to a depth of 20 cm. All plots received a basal application of cattle manure at 8 t ha^-1 (1.05-1.15% N, 0.67-0,70% P₂O₅, 1.21-1.32% K₂O) and phosphorus (SSP, 16% P₂O₅) at 90 kg P₂O₅ ha^-1. Potassium (60 kg K₂O ha^-1) and nitrogen (40 kg N ha^-1) were utilised as potassium chloride (60% K₂O) and urea (46% N). Following row treatments, the K fertilizer was additionally applied at two times in the plant’s growth cycle: 12 days after seeding, one-third of the total at the time the 2-3 fully leaves expanded and 35 days after seeding, with the remaining two-thirds immediately before blooming (5-6 fully leaves). The N fertiliser was applied following treatments as follows: In 1 time (100% as basal), 2 times (1/3 N at basal and 2/3 N at the 2-3 fully leaves expanded) and three times (1/3 N at basal; 1/3 N at the at the 2-3 fully leaves expanded and 1/3 N at the at the 5-6 fully leaves expanded). Six rows made up the 3 m by 5 m (15 m²) plot, with two rows designated for nodulation and shoot dry weight characteristics and the remaining rows-aside from the borders-being harvestable rows. Plots were spaced 0.3 m apart, with blocks separated by 0.5 m. Each plot was made into a bed by preparing canals around it. The soybean seeds are soaked and mixed with the inoculant at the rate of 500 ml of the inoculant per one kilogram of the soybean seeds and air-dried for 30 minutes before planting. To ensure that every seed had a thin layer of the inoculant, the thick slurry of the inoculant was carefully combined with dried seed. To preserve the viability of the bacterial cells, all inoculations were carried out right before planting in a shaded area. The infected and un-inoculated seeds were then planted at a spacing of 10 cm between plants and 30 cm between rows making 33 plants m^-2. To identify the N_fixed, local maize variety as a reference crop was used, which was grown in the rows outside the experiment plots in an area of 50 m².
 
Fresh and dry weight of the nodulation
 
At full flowering stage, ten randomly selected plants were taken from each plot at border rows. To extract undamaged roots and nodules for nodulation parameters and plant dry weight, the entire plant was gently uprooted using a spade. To prevent nodule loss, uprooting was accomplished by exposing the entire root system. The roots with intact nodules were gently washed with water over a metal screen to remove the clinging soil. Nodulation, nodule dry weight and the number of nodules per plant were recorded using the same 10 plants from each plot. Three plants from each plot were cut at ground level to determine the shoot dry weight after nodules were removed. An electronic balance was used to determine the total weight of fresh shoots. Using brown envelopes, subsamples of shoots were oven-dried for 48 hours at 70^oC. Weighing the dry materials allowed us to record the average shoot dry weight for each plant. N shoots and roots of soybean were also analysed following the digestion, distillation and titration processes of the Kjeldahl method.
 
N fixation of soybean
 
Biological nitrogen fixation (BNF) was calculated by formula as follows: %BNF = (N_fixed/N_total) × 100; N_fixed = N_uptake (soybean) – N_uptake (reference); %N_dfa = (N_fixed/N_uptake (soybean)) × 100; N_uptake = N concentration in shoot and root (mg kg^-1) (Mogale et al., 2023).
 
Data analysis
 
The 2 factors ANOVA analysis method offered by Statistix statistical software version 10 was used. The effects of treatment parameters (split times of N fertiliser application and inoculation levels) and their interactions on nodulation and N fixation were assessed by using analysis of variance. The LSD technique was used to compare means among treatments (p<0.05).

Soybean nodulation characteristics
 
There was effects of rhizobium inoculation and split N fertiliser application on the number of nodules (Table 1); the highest number of nodules were found in treatment with inoculation and 3 times of N fertiliser application (32.40 and 28.43 nodules plant^-1); however, the highest number of nodules was found at non-inoculation and 2-3 times of N fertiliser application (23.59 and 22.70 nodules plant^-1). Number of effective nodules plant^-1 were recorded a significant higher in spring season than in summer season from 14 - 28% at 3 times of N fertiliser application both in inoculation and non-inoculation. At both cropping seasons, fresh and dry weight of nodules were exhibited the highest values on combination treatment of 3 times of N fertiliser application with inoculation with 1.69 - 1.94 and 0.73 - 0.74 mg plant^-1, respectively. There were significant differences between fresh and dry weight of nodules for the most treatments. Therefore, N fertiliser application with rhizobium inoculation had an impact on nodulation quantity and weight. This outcome is consistent with prior research showing that co-inoculation increased the number of large nodules and the volume of nodules (Bais et al., 2023). Rhizobia seed inoculation outperformed uninoculated treatments by a wide margin. Both rhizobia inoculation and N rates administered separately and enhanced soybean nodulation and yield (Getachew and Dagnaw, 2020; Szpunar-Krok et al., 2023). Khaitov and Abdiev (2018) showed a significant advantage of rhizobium inoculation over uninoculated treatments at all N fertilisation rates (50, 75 and 100 kg N ha^-1). Jarecki and Bobrecka (2019) indicated an increase in nodulation on soybean roots after seed inoculation with Bradyrhizobium japonicum. The legume fixation at lower N rate may not be sufficient to meet its N demand, an additional high N rate as a starter to create a symbiosis with rhizobia, which could explain the split times of N fertiliser application reported enhanced nodule yield after inoculation (Ntambo et al., 2017; Budiastuti et al., 2025). When soybeans were treated with 3 times of N application, the number of nodules at the was significantly increased, although we found that split times of N fertiliser application inhibits nodulation in our study, as evidenced by a decrease in nodule number and volume with the less times of N fertiliser application, the mechanism by which N inhibits nodulation is not well understood and may be due to a variety of factors including cultivation technology, soil and climate (Xia et al., 2017; Jarecki and Bobrecka, 2019). Several theories have been put out to explain how N inhibits nodulation, including lower O2 diffusion into nodules that limits bacteroids’ ability to breathe, feedback inhibition caused by a byproduct of nitrate metabolism and carbohydrate shortage in nodules (Bais et al., 2023). When process of nodulation formulation by native soil rhizobia is either nonexistent, very low, or restricted in soils lacking of homologous rhizobia, the number of nodules increases following inoculation (Thilakarathna and Raizada, 2017). Incompatibility between the host micro-symbiont and the function of both known and undiscovered biomolecules, such as flavonoids, polysaccharides and hormones, may be the cause of the absence or the development of inefficient nodules on soybean roots (Daayf et al., 2012).


 
N content in shoots and roots and uptake of soybean
 
There were increasing of shoot and root N accumulation following the split times of N fertiliser application with inoculation of rhizobium in soybean (Table 2). Both non-inoculation and inoculation treatments had the highest values of shoot N (2.31 -2.36% and 2.41 - 2.56%) and root N (2.23 - 2.40% and 2.50 -2.62%) at 3 times of N fertiliser application in both cropping seasons. Omari et al., (2022) reported that under greenhouse experiment, shoot N levels were consistently greater after GMM36 and GEM96 inoculation. When compared to the non-inoculated control, the GMM36 and GEM96 inoculation dramatically raised the shoot N content by an average of 79%. Under field conditions, there was an average 39% increase in shoot N between the inoculated and non-inoculated control. N uptake was found the highest values at treatment of inoculation combined with 3 application times of N fertiliser in both crop seasons (200.02 - 247.34 kg ha^-1). Number of nodules and N uptake were noticeably higher in the greenhouse than in the field conditions, suggesting that abiotic factors have a substantial impact on nodulation and soybean growth (Goyal et al., 2021; Abiyot et al., 2022).


 
N fixation ability of soybean
 
Across all data sets, BNF had a close relationship with N fixation, N_fda and it tended to increase as N fertiliser application time together with inoculation (Table 3). BNF data was ranging from 43.18 to 59.06% in spring season and from 32.64 to 47.51% in summer season. It was found higher than 9.3 - 14.8% in spring season and 12.9 - 15.9% in summer season at inoculation treatment compared to non-inoculation at 3 times of N fertiliser application. Generally, BNF was the highest values at 3 application times of N fertiliser combined with inoculation in both cropping seasons. Similarly with result of Martins et al. (2022) who reported that inoculated soybean exhibited higher levels of nodulation, plant biomass, BNF and yield components compared to non-inoculated by preserving BNF at subsequent growth stages, further inoculation in the V3 growth stage demonstrated many benefits. When non-inoculation was added, the maximum amount of N₂ fixation reached at 3 times of N fertiliser application (66.62 kg ha^-1 and 49.89 kg ha^-1 in both cropping seasons). N₂ fixation was higher than 29.1-37.8% in inoculation compared to non-inoculation. Similarly, it was also found that N_dfa was 29.04-33.95% at non-inoculation treatments in both seasons and 32.06-37.04% at inoculation in both seasons. These results suggested that N uptake, BNF, N₂ fixation and N_dfa were the highest at treatment of N fertiliser application at basal, 12 and 35 days after emergence combined with rhizobium inoculation, though they were not significantly differences from that of inoculated and uninoculated rhizobium on soybean. Plant growth and dry matter production are the main factors influencing how much N₂ legumes fix. However, at both sample intervals, plant %N_dfa and N accumulation did not react to further inoculation (Martins et al., 2022). If no new inoculations are produced, less effective strains with high saprophytic capacity may eventually dominate the soil population, reducing the benefits of BNF, even though the soybean crop responses to inoculation which is demonstrated successful in the first few years (Zilli et al., 2021). Since symbiotic N2 fixation requires nodulation, soybeans may have stimulated free-living N₂-fixers, or any type of biological N2 fixation that does not involve a clearly defined symbiotic relationship between plants and microorganisms, which is typical in undisturbed soil (Reed et al., 2011). Methodo-logical issues may also be the cause of the BNF activity that we saw in non-inoculated on soybeans (Martins et al., 2022). The soil used in this experiment was coastal sandy soil with high sand content (>80%), implying that the soil was well aerated, low water and nutrient holding capacity, then the number of nodules were low in less times of N fertiliser application and non-inoculation of rhizobium resulting in low BNF (Thanni et al., 2017). N₂-fixing systems can thrive in soils poor in N that they are a source of proteins and they provide N for soil fertility (Cordeiro and Echer, 2019).

Fertilisation of N fertiliser application up to 40 kg ha^-1 at three times (as basal, top dressing at 12 and 35 days after sowing) may be the ideal time for efficient nodulation and better N uptake, which would increase soybean BNF, N₂ fixation and N_dfa. The current findings show that employing rhizobia inoculant in conjunction with split N fertiliser application into 3 times at 40 kg N ha^-1 has the potential to improve soybean nodulatio and N₂ fixation. Further research will consider on how the N₂ fixation related with soybean yield and N use efficiency.

“PhD student: Tran Thi Anh Tuyet was funded by the PhD Scholarship Programme of Vingroup Innovation Foundation (VINIF), VinUniversity, code VINIF.2025.TS.62”. This work was partially supported by Hue University under the Core Research Program, Grant No. NCTB.DHH.2025.06.

All authors declare that they have no conflicts of interest.