Examination of nanoparticulate phosphate rock as both a liming agent and phosphorus source to enhance the growth of spinach in acid soil

ABSTRACT

To examine the effect of nanoparticulate phosphate rock (NPR) as both a liming agent and phosphorus source in a tropical acid soil. The study examined five rates of NPR (0, 250, 500, 1000, or 2000 kg ha⁻¹), which supplies 30, 60, 120 or 240 kg P ha⁻¹, in a randomized complete pot experiment design with 3 replications. The pots of soil were incubated in a climate-controlled greenhouse for 21 days and then spinach was grown for 49 days. Soil parameters (pH, available P and exchangeable acidity), spinach parameters (leaf area, root hair surface area, root length and dry matter yield) and the effectiveness of NPR dissolution were estimated. The soil and plant parameters and the effective NPR dissolution all increased to the same degree at 1000 and 2000 kg NPR ha⁻¹. Therefore, the use of 1000 kg ha⁻¹ was most economically justified. Although, NPR has been appeared as an effective liming agent and phosphorus source in tropical acid soil. However, a regular application of NPR and further research for economic comparison between NPR and both of lime and superphosphate, as well as the original PR, will be needed.

KEY WORDS:

• Nanotechnology
• liming effect
• soil reaction
• root hair surface area

1. Introduction

Soil acidity problems have become an environmental and economic concern, especially in humid tropical regions, because they are difficult and expensive to ameliorate (Gazey et al. 2014). Acid soils occupy approximately 43% of the world’s tropical land area, mostly in the Americas (68%), Asia (38%) and Africa (27%), and have high potential for agriculture and agroforestry production (Buni 2014). However, plant growth in such soils is hindered by low availability of P, Ca and Mg, and by acidity related to toxicity of Al, Fe and Mn (Rafael et al. 2018). The application of lime can mitigate soil acidity and thus prevent Al, Fe and Mn toxicity, while providing Ca and Mg to the soil (Álvarez et al. 2012; Rafael et al. 2018; Viadé et al. 2011). The application of soluble P fertilizer can satisfy the P deficiency (Chien et al. 2011). Such applications are considered general practice for correcting acidity and nutrient deficiencies in acid tropical soils, and should be applied together, as P without lime does not lead to increased yields (Opala 2017; Sarker et al. 2014). As the costs of lime and P fertilizers have increased markedly worldwide (Aishwath et al. 2015; Chien et al. 2011), there is an urgent need for new materials that achieve both goals at the same time at a reduced cost.

Phosphate rock (PR) has been used as an alternative to superphosphate for supplying P in acid soils (Adhikari et al. 2014; Adorolo et al. 2015; Dodor 2016; Onwonga et al. 2013) being cheaper and usable with or without lime (Adetunji et al. 2005). It also has a liming effect (Arifin et al. 2018; He et al. 2005; Loganathan et al. 2005; Dodor 2016; Sinclair et al. 1993; Sikora 2002; Yang et al. 2012) attributed mainly to the dissolution of PO₄^3‒, CO₃^2‒ and F^‒ from apatite in the PR, which associate with protons to displace acidity (Arifin et al. 2018; He et al. 2005; Loganathan et al. 2005; Sikora 2002), and partly to small amounts of gangue carbonates, such as calcite and dolomite (Loganathan et al. 2005). But the liming action could be small from a single application at rates typically used for maintaining soil P status. It also depends on the dissolution rate, which is slow at a large particle size (Adhikari et al. 2014; He et al. 2005; Ghosal and Chakraborty 2012; Rafael et al. 2018). As the dissolution rate, the application rate and the liming effectiveness all depend mainly on the particle size (Pagani and Mallarino 2012; Rafael et al. 2018), reducing the particle to nanoscale and thus increasing the total surface area could offer a mechanism to increase the efficiency of PR and make it a more effective P fertilizer (Devnita et al. 2018; Liu and Lal 2015). Although very fine grinding of PR, especially at the nanoscale, is expensive, and the extreme fineness makes handling difficult (Ibrahim et al. 2010), the potential for the use of smaller amounts could limit such problems and reduce costs (Devnita et al. 2018). However, published data on the effectiveness of nanoparticulate PR as both a liming agent and a P fertilizer is limited or non-existent.

Soil acidity is a major growth-limiting factor for plants in many parts of the world (Kidd and Proctor 2001), where it causes significant losses in crop production (Yang et al. 2005). It also restricts root development (Haling et al. 2011; Meriño-Gergichevich et al. 2010; Nduwumuremyi 2013). As a result, plants are stunted and show symptoms of nutrient deficiency, especially P deficiency (Anetor and Akinrinde 2007; Zapata and Zaharah 2002; Ayalew 2011). Spinach is an important and popular vegetable crop that is extremely sensitive to soil acidity, and needs a soil pH in the range of 6.4 to 6.8 (Laboski and Peters 2012; Wyenandt et al. 2018). It also has a relatively high requirement for P during early growth (Grant et al. 2001). So we grew spinach as a test crop in this study.

Given the importance of liming to alleviate the constraints posed by soil acidity in addition to the high prices of lime and P fertilizers, we examined the effect of nanoparticulate PR to mitigate soil acidity while increasing P availability in a pot experiment.

2. Materials and methods

2.1. Location and soil

The study was conducted at the Tropical Agriculture Research Front of the Japan International Research Centre for Agricultural Sciences on Ishigaki Island, Okinawa Prefecture. The soil used in the study was brought from a mountain on the island. It is classified as a Red Soil in the Japanese soil classification system (NIAS 1996), which corresponds to Ultisols in the USDA soil taxonomy (Hamazaki 2005), with the following physicochemical characteristics: pH (water), 5.31; exchangeable (exch.) Al³⁺, 2.67 cmol_c kg⁻¹; effective cation exchange capacity (CEC), 5.69 cmol_c kg⁻¹; and texture class, sandy clay loam with 30.56% clay.

2.2. Preparation and properties of nanoparticulate phosphate rock (NPR)

The NPR was prepared as described by Adhikari et al. (2014) with some modifications. Burkina-Faso PR was crushed, passed through a 2-mm sieve, and ground to ≤74 µm in a high-speed vibrating sample mill (TI-200, CMT Co., Ltd, Tokyo, Japan) at 1500 rpm for 5 min. This process was performed 5 times. The material was then ball-milled in a mixer mill (MM 400, Retsch GmbH, Germany) at 1500 rpm for 10 min with a 2-min pause in the middle. Table 1 shows its mineralogical and chemical properties. Scanning electron microscopy showed that the resultant nanoparticles tended to agglomerate in several diameters of 10 to 100 nm (Fig. 1).

Table 1. Chemical analysis of nanoparticulate phosphate rock

Parameters	Value
pH	8.60
Element	(%)
P2O5	27.59
CaO	34.40
MgO	0.180
Na2O	0.190
K2O	0.530
SiO2	23.47
Fe2O3	2.980
Al2O3	4.230
SO3	0.060
Co2	1.940
F	2.640
Cl	0.040
Ca:P ratio	2.040
Mineral composition	(%)
Apatite	62.00
Quartz	20.00
CaCO3 Equivalent (%)	61.45
Water soluble P (%)	0.109
Extractable P (2% citric acid) (%)	22.33
Extractable P (2% formic acid) (%)	21.56

Figure 1. Scanning electron microscopy image of nanoparticles of phosphate rock

2.3. NPR requirement for liming

NPR requirement for liming was estimated based on the amount of NPR required to raise soil pH to 6.0 in the top 15 cm surface layer. Taking into consideration the target limit of pH for spinach crop is 6.0 (Laboski and Peters 2012). This was determined by shaking soil (50 h; 160 rpm min⁻¹) with various amounts of NPR in distilled water intermittently for three days and then measuring the pH. The amount of NPR required was 6000 kg NPR ha⁻¹; that is referred as the NPR requirement needed to raise soil pH to 6.0 in the top 15 cm surface layer. Accordingly, we examined four economically affordable rates of NPR at 250, 500, 1000, or 2000 kg ha⁻¹, which supplies 30, 60, 120 or 240 kg P ha⁻¹ (@ 12% P; Table 1).

2.4. Experimental design and treatments

The soil was first crushed and sieved (<2 mm). Portions of 3.5 kg (3.3 kg dry weight) were weighed and thoroughly mixed with the specified dose of NPR. The mixture was packed in 1/5000-are Wagner pots (4000 cm³) and brought to the same bulk density by compacting to the same height. The experiment used a randomized complete design with a five rates of NPR at 0 (control), 250, 500, 1000 or 2000 kg ha⁻¹, giving 5 treatments × 3 replications = 15 pots.

2.5. Incubation of soil + NPR

The pots of soil were incubated in a climate-controlled greenhouse (50% relative humidity; 25°C day/20°C night) for 21 days during December 2017. They were watered every 2 days with distilled water according to their weight to maintain field capacity.

2.6. Spinach cultivation and agronomic management

Seeds of spinach (Spinacia oleracea L.) ‘Apollo’ were germinated on moist medical cotton in a refrigerator for about 48 h at 4°C and then in the laboratory at 21 ± 1°C. Seeds that germinated first were directly sown into the incubated pots on 25 December 2017 at eight seeds per pot. Seedlings were thinned to leave the four healthiest, at 4 cm between plants. Tap water (pH 7 ± 1; conductivity 0.26 ± 1 dS m⁻¹) was applied twice a week to field capacity by weight until the end of the growth period (49 days). All pots received 0.7 g urea (equivalent to 160 kg N ha⁻¹) and 0.39 g KCl (equivalent to 102 kg K ha⁻¹), which were mixed into the soil. NPR was applied in the doses explained above.

2.7. Soil samples and analyses

Soil samples were taken after incubation and after harvest from the top 15 cm in each pot, air-dried, crushed, sieved to <2 mm, and analyzed for pH and available P. The soil exchangeable (exch.) acidity (H⁺ and Al³⁺) was estimated after harvest.

2.8. Evaluation of soil P

The Truog method is considered to be the best method to estimate P in soil enriched with calcium phosphate (Aye et al. 2011). We used the target (35 mg kg⁻¹) and upper limit (56 mg kg⁻¹) soil test P values (Olsen P) for spinach indicated by Prasad et al. (1988), converted from Truog P values as:

(1) $\mathrm{O}\mathrm{l}\mathrm{s}\mathrm{e}\mathrm{n}\mathrm{P}=14.3+\left(1.47\times \mathrm{T}\mathrm{r}\mathrm{u}\mathrm{o}\mathrm{g}\mathrm{P}\right)$(1)

This equation was derived from the results of a study in which 90 samples of different soil types were analyzed by both methods (R² = 0.74; SE a = 1.87; SE b = 0.093).

We used the diagnostic ranges for P in spinach leaves to evaluate the general condition of the soil P, where ≤0.1% is low, 0.25% to 0.35% is sufficient, and >0.35% is high (Barker and Pilbeam 2007).

2.9. Estimation of spinach growth parameters

To assess the impact of NPR on yield, we harvested the leaves at maturity and weighed them to measure the dry matter yield (DMY) and leaf P concentration.

We estimated leaf surface area (LSA, cm²) by analysis of scanned images of leaves in ImageJ v. 1.47 software (Aboukarima et al. 2017; Chaudhary et al. 2012; Supp. material S1). A leaf count was done to find the total number of leaves in all the plants. To get the average number of leaves per plant, the total number of leaves was divided by the total number of plants whose leaves were counted. To calculate the leaf surface area index (LSAI, cm² cm⁻²) the following Eq. (2)(2) $\mathrm{L}\mathrm{S}\mathrm{A}\mathrm{I}=\left(\mathrm{Y}\times \mathrm{N}\times \mathrm{L}\mathrm{S}{\mathrm{A}}_{\mathrm{a}\mathrm{v}\mathrm{g}}\right)/\left(\mathrm{A}\mathrm{P}\right)$(2) was used:

(2) $\mathrm{L}\mathrm{S}\mathrm{A}\mathrm{I}=\left(\mathrm{Y}\times \mathrm{N}\times \mathrm{L}\mathrm{S}{\mathrm{A}}_{\mathrm{a}\mathrm{v}\mathrm{g}}\right)/\left(\mathrm{A}\mathrm{P}\right)$(2)

where Y = Number of plants per pot; N = Average number of leaves per plant; LSA_avg = Average leaf surface area per plant; AP = Area of pot.

Root hair surface area (RHSA, cm²) was also estimated by image analysis (Supp. material S1). One plant from each pot was randomly selected. After the shoot was cut off, the root was collected in a soil core, washed, and then carefully separated from the soil. The clean root was suspended in water and photographed with a digital camera. RHSA was estimated in ImageJ (Aboukarima et al. 2017; Chaudhary et al. 2012; Supp. material S1). The root hair surface area index (RHSAI, cm² cm⁻²) was calculated as RHSA divided by the cross-sectional area of the soil core.

Root length (RL, cm) was measured by a ruler from cutting point to root tip with an error of ±1 mm.

2.10. Physicochemical analyses

All soil physicochemical analyses were done as outlined by the Editorial Boards of Methods for Soil Environment Analysis (1997) as follows: After the removal of organic matter, particle size was analyzed by the hydrometer method. Soil pH was measured in distilled water at a 1:2.5 (w/v) ratio with a pH meter (LAQUAact D-73 pH/ORP/ion meter, Horiba Scientific, Japan). Exchangeable acidity was extracted with 1 M KCl and then titrated with 0.01 M NaOH to pH 7. After the addition 1 M KF to the previous solution, the exch. Al³⁺ was determined by titration with 0.1 N HCl to a phenolphthalein endpoint and to an isoelectric endpoint. CEC was determined by the ammonium acetate (pH 7) method. Effective CEC was estimated as CEC + exch. acidity. The calcium carbonate equivalent of the NPR was determined by the AOAC method. Available P was extracted by the Truog method (0.002 N H₂SO₄, pH 3). Leaves were oven-dried for 48 h at 65°C and weighed, and the DMY was calculated. The dried leaves were then milled to pass through a 1.0-mm sieve and digested in a 1:3 mixture of perchloric acid (HCLO₄) and nitric acid (HNO₃). P concentrations in soil and leaves were determined on a spectrophotometer (U-2001, Hitachi, Japan) by the ascorbic acid reduction – molybdenum blue method. Total major and minor element composition of NPR samples were determined by inductively coupled plasma-optical emission spectrometry (Kalra and Maynard 1991). The mineralogical composition of NPR was determined by powder X-ray diffraction (Pansu and Gautheyrou 2006).

2.11. Calculation of NPR dissolution and increased in plant phosphorus (IPP)

NPR dissolution (NPRD) is calculated as follows (Baligar et al. 1997; He et al. 2005):

(3) $\mathrm{N}\mathrm{P}\mathrm{R}\mathrm{D}\left(\mathrm{m}\mathrm{g}\mathrm{k}{\mathrm{g}}^{-1}\right)=\left[{\left(AvailableP\right)}_{NPR}-{\left(AvailableP\right)}_0\right]$(3)

where ${\left(AvailableP\right)}_{NPR}$ and ${\left(AvailableP\right)}_0$ are the available P (mg kg⁻¹) in NPR-treated soil and untreated soil after harvest, as converted to Olsen P.

IPP is the difference between plant P in soil with NPR and soil without NPR. Plant P (mg plant⁻¹) is calculated as follows:

(4) $\mathrm{P}\mathrm{l}\mathrm{a}\mathrm{n}\mathrm{t}\mathrm{P}=\left[LeafP\left(\mathrm{\%}\right)-DMY\left(\mathrm{m}\mathrm{g}\mathrm{p}\mathrm{l}\mathrm{a}\mathrm{n}{\mathrm{t}}^{-1}\right)\right]$(4)

2.12. Statistical analyses

The differences between means were tested by univariate analysis of variance (ANOVA), followed by Tukey’s honestly significant difference (HSD; P ≤ 0.05) when a significant difference was detected, in SPSS v. 17.0 software (SPSS, Chicago, IL, USA).

3. Results

3.1. Soil parameters

NPR significantly affected soil pH (P < 0.001) both after incubation and after harvest. pH increased with increasing NPR (Table 2). After incubation, pH increased from 5.32 with 0 NPR to 6.51 and 6.54 with NPR at 1000 and 2000 kg ha⁻¹, respectively. After harvest, although pH followed a similar trend, it was 5.90 and 5.96, respectively. Thus, irrespective of the NPR rate, the soil pH became more acidic again. NPR application represented liming at low rates that did not completely ameliorate soil acidity, as a pH of 5.3–5.6 is still acidic. After incubation, NPR at 250 and 500 kg ha⁻¹ increased soil pH, though not to the levels required to promote spinach growth, while NPR at 1000 and 2000 kg ha⁻¹ increased pH to within the optimum range (typically 6.4–6.8), suggesting the inefficiency of NPR at <1000 kg ha⁻¹ at increasing pH. Interestingly, after harvest, the pH at all NPR rates was below the critical level, but that at the high rates was nearest to the lower limit (pH 6) required for most crops. In general, pH increased by ~33 × 10^–5 units per kg of applied NPR.

Table 2. Post-incubation and post-harvest effects of nanoparticulate phosphate rock (NPR) on soil pH, available P and exchangeable (exch.) acidity

	Post-incubation^‡		Post-harvest^‡
NPR rate (kg ha^−1)	pH	Available P (mg kg^−1 soil)	pH	Available P (mg kg^−1 soil)	exch. acidity (cmolc kg^−1 soil)
0	5.32^e	23.57^e	5.30^d	18.56^d	3.50^a
250	5.58^d	29.70^d	5.40^d	24.22^d	3.37^b
500	5.67^c	41.20^c	5.63^c	34.35^c	1.85^c
1000	6.51b^b	108.56^b	5.90^b	92.43^b	0.87^d
2000	6.54^a	117.81^a	5.96^a	104.36^a	0.87^d
Analyses of variance^†
Sign	***	***	***	***	***
CV	0.188	1.990	0.137	5.550	0.370
SEM	0.006	0.737	0.004	1.756	0.004
HSD	0.030	2.431	0.021	8.171	0.021

^‡Means of column under each subheading followed by different superscript letters are significantly different from each other according to Tukey’s honestly significant difference (HSD; P ≤ 0.05).

^{† ***} is the significant degree at P< 0.001; SEM is the standard error of mean; CV is the coefficient of variance.

NPR also significantly decreased the soil exch. acidity (P < 0.001, Table 2). NPR at 1000 and 2000 kg ha⁻¹ produced the minimum exch. acidity (0.87 cmol_c kg⁻¹), reducing it by about 75% from the initial value (3.5 cmolc kg⁻¹). Both rates gave comparable results. In general, the exch. acidity after harvest was low, at 0.87 cmol_c kg⁻¹, when the soil pH was around 6, but it rose sharply to 1.85 cmol_c kg⁻¹ when the pH fell below 5.6, and rose very sharply to 3.37 cmol_c kg⁻¹ at pH 5.4. Below pH 5.6, the exch. acidity increased by 0.13 cmol_c kg⁻¹ per 0.1 unit decrease in pH.

Increasing NPR significantly increased available P both after incubation and after harvest (P < 0.001, Table 2). The available P contents of all NPR-treated soils were significantly higher than the control. NPR at 1000 and 2000 kg ha⁻¹ gave the maximum available P, which was >4.5× to >5.5× the initial value. Although the available P after harvest followed a similar trend, the values were reduced by 22.63% at 250 kg ha⁻¹, by 19.94% at 500 kg ha⁻¹, by 17.45% at 1000 kg ha⁻¹ and by 12.87% at 2000 kg ha⁻¹. Available P was higher at the high NPR rates than at the low rates. Generally, the higher rates of NPR were more effective in releasing more P in the acid soil. NPR at 2000 kg NPR ha⁻¹ released 56 mg P kg⁻¹ (upper limit for spinach), while NPR at 1000 kg ha⁻¹ released 35 kg P ha⁻¹ (target limit).

3.2. Spinach growth parameters

NPR also significantly altered LSAI, RHSAI and RL (P < 0.001, Table 3). NPR at ≥1000 kg ha⁻¹ increased LSAI by >2×, RHSAI by >4× and RL by >3× the initial value, while NPR at <1000 kg ha⁻¹ increased them by >1×, >2× and >1.5×. But these increases did not differ significantly between 1000 and 2000 kg ha⁻¹. In addition, RL did not differ significantly from the initial value when NPR = 250 kg ha⁻¹.

Table 3. Post-harvest effects of nanoparticulate phosphate rock (NPR) on spinach dry matter yield (DMY), leaf area (LSAI), root hair surface area index (RHSAI), root length (RL) and leaf P concentration

NPR rate (kg ha^−1)	Plant parameters^‡
	DMY	LSAI	RHSA	RL	Leaf P
	(g plant^−1)	(cm^2 cm^−2)	(cm^2 cm^−2)	(cm)	(%)
0	0.203^d	2.828^d	0.114^d	3^c	0.090^c
250	0.417^c	3.369^c	0.247^c	4^bc	0.110^bc
500	0.647^b	4.083^b	0.317^b	6^b	0.130^b
1000	2.303^a	6.284^a	0.502^a	10^a	0.513^a
2000	2.330^a	6.309^a	0.506^a	10^a	0.530^a
Analyses of variance^†
Sign	***	***	***	***	***
CV	5.762	1.956	0.485	13.964	3.760
SEM	0.039	0.052	0.001	0.537	0.006
HSD	0.182	0.240	0.004	2.501	0.028

^‡Means of column under each subheading followed by different superscript letters are significantly different from each other according to Tukey’s honestly significant difference (HSD; P ≤ 0.05).

^{† ***} is the significant degree at P < 0.001; SEM is the standard error of mean; CV is the coefficient of variance.

NPR significantly increased DMY (P < 0.001, Table 3). NPR at ≥1000 kg ha⁻¹ increased it by ≤11.5× the initial value, while NPR at <1000 kg ha⁻¹ increased it by ≤3×. DMY did not increase significantly from 1000 to 2000 kg ha⁻¹. Therefore, the critical threshold or application rate that produced the maximum yield (2.3 g DMY plant⁻¹) is 1000 kg ha⁻¹.

NPR also significantly increased leaf P (P < 0.001, Table 3). NPR at 2000 kg ha⁻¹ increased it the most, but not significantly more than 1000 kg ha⁻¹. The leaf P concentration varied from 0.11% to 0.53% among NPR rates. That when NPR was ≥1000 kg ha⁻¹ was about ≤5.8× the initial value. Interestingly, the effects did not differ significantly between 1000 and 2000 kg ha⁻¹. According to the diagnostic range for P in spinach leaf, NPR at ≥1000 kg ha⁻¹ increased available soil P to at least the upper limit.

3.3. NPR dissolution (NPRD) and increased in plant P (IPP)

As NPR increased, NPRD and plant P both increased (Table 4). Most dissolution occurred at NPR ≥1000 kg ha⁻¹; consequently, plant P increased from 0.27 mg plant⁻¹ at 250 kg NPR ha⁻¹ to 12.17 mg plant⁻¹ at 2000 kg NPR ha⁻¹. NPRD in soil increased similarly to plant P: from 5.66 mg kg⁻¹ soil at 250 kg NPR ha⁻¹ to 85.80 mg kg⁻¹ soil at 2000 kg NPR ha⁻¹.

Table 4. Post-harvest nanoparticulate phosphate rock dissolution (NPRD) and increased plant P (IPP) with different rates of NPR

	NPRD^†	IPP^‡
NPR rate (kg ha^−1)	(mg kg^−1 soil)	(mg plant^−1)
250	5.66	0.27
500	15.79	0.64
1000	73.87	11.64
2000	85.80	12.17

^†$\mathbf{N}\mathbf{P}\mathbf{R}\mathbf{D}\left(\mathbf{m}\mathbf{g}\mathbf{k}{\mathbf{g}}^{-1}\right)=\left[{\left(\mathbf{A}\mathbf{v}\mathbf{a}\mathbf{i}\mathbf{l}\mathbf{a}\mathbf{b}\mathbf{l}\mathbf{e}\mathbf{P}\right)}_{\mathbf{N}\mathbf{P}\mathbf{R}}-{\left(\mathbf{A}\mathbf{v}\mathbf{a}\mathbf{i}\mathbf{l}\mathbf{a}\mathbf{b}\mathbf{l}\mathbf{e}\mathbf{P}\right)}_0\right]$, where ${\left(\mathbf{A}\mathbf{v}\mathbf{a}\mathbf{i}\mathbf{l}\mathbf{a}\mathbf{b}\mathbf{l}\mathbf{e}\mathbf{P}\right)}_{\mathbf{N}\mathbf{P}\mathbf{R}}$ and ${\left(\mathbf{A}\mathbf{v}\mathbf{a}\mathbf{i}\mathbf{l}\mathbf{a}\mathbf{b}\mathbf{l}\mathbf{e}\mathbf{P}\right)}_0$ are the available P in soil (mg kg⁻¹) treated with and without NPR at post-harvest, respectively.

^‡IPP is the difference in plant P (leaf P × DMY) between soils with NPR and soil without NPR.

4. Discussion

4.1. Liming effect of NPR

The capacity of a soil to supply H⁺, as estimated from soil pH and exch. acidity, is the most important factor that controls NPR dissolution. NPR increased soil pH as result of the consumption of protons during dissolution (Sikora 2002; He et al. 2005). The major liming action of PR is attributed to the dissolution of PO₄^3‒, CO₃^2‒ and F^‒ from the apatite in the PR and the ability of these anions to associate with protons to remove acidity from soil (Arifin et al. 2018; He et al. 2005; Loganathan 2005; Sikora 2002). The application of NPR as a liming agent can reduce acidity, as indicated by the effect on pH. However, as its liming effect increased quickly during incubation but then decreased slightly until harvest, NPR has a long-term but insufficient liming action to raise the pH to levels favorable to most crops. Certainly, the effect of a single application is small, but repeated application over the long term might produce cumulative effects of practical significance (Loganathan et al. 2005). Therefore, regular NPR application is needed. Similarly, in a 4-month incubation experiment, the addition of NPR to acid soil increased soil pH in the first month, but pH then decreased during further incubation (Arifin et al. 2018). Thus, to confirm the effectiveness of NPR in increasing soil pH, further long-term studies are needed. By increasing soil pH, NPR decreased the Al³⁺ concentration and thus total soil exch. acidity (Loganathan et al. 2005; Sinclair et al. 1993; Sikora 2002). Since Al phytotoxicity in soils is determined by the fraction of Al that is bioavailable, the reduced concentration of exchangeable Al³⁺ suggests the potential of NPR to reduce the toxic effects of Al in soils. The lower pH after harvest than after incubation could be due to the hypo-buffering capacity of the variable-charge acid soils (Arifin et al. 2018; Dikinya and Mufwanzala 2010). Furthermore, the greater liming effect after incubation and during the early growth of spinach raised the soil pH, and consequently decreased the dissolution of NPR because some of the soil acidity was consumed by the free carbonates, resulting in a decrease in pH during the late growth of spinach until harvest (He et al. 2005; Loganathan et al. 2005).

4.2. Phosphorus availability

NPR was more effective in increasing available P in acid soil than in increasing pH, especially during incubation. Its effectiveness decreased to some extent until harvest, indicating that NPR can release P, but for no more than 3 months. This result is consistent with the result that available P increased to the maximum value after 1 month and decreased after 4 months during incubation of NPR-treated acid soils (Devnita et al. 2018). Our explanation for the decreased available P in soil after harvest is that dissolution of the NPR initially raised the soil pH (Table 2), which in turn decreased the activity of Fe and Al hydroxides, thereby increased the available P (Table 2), but when Ca from NPR dissolution reaches a certain level at which re-fixation starts, the available P is reduced to some extent (Table 2), especially during the late growth stage, i.e., before harvesting of spinach (Savini et al. 2015). Additionally, the relatively high P requirement of spinach during early growth stages also could be part of the reason for the later decline in available P (Asomaning et al. 2006; Grant et al. 2001). P deficiency in acid soils is often associated with high P fixation; P uptake rates are highest at pH 5.0 to 6.0, where H₂PO₄⁻ dominates (Onwonga et al. 2013). P adsorption, precipitation and lack of liming could also have been responsible for the declining levels of available P at low NPR application rates (pH < 5.7). Furthermore, the increase in available P at high NPR rates could be due to the continued supply of Ca by NPR, while P availability was ensured by the release of Al- and Fe-bound phosphates from sorption sites (He et al. 2005; Loganathan et al. 2005).

4.3. Spinach growth performance

NPR at the highest application rates increased both root development (RL and RHSAI) and plant growth (LSAI and DMY). The increase in DMY was partly related to the liming effect of NPR, which changed the soil pH, adsorption and CEC, which are strongly involved in the availability of acidic ions (e.g. H⁺, Fe, or Al³⁺) and hence in nutrient toxicity. The enhanced plant growth and root development would be expected if Al toxicity were alleviated by the raised Ca/Al ratio in the soil solution (Baligar et al. 1997; Meriño-Gergichevich et al. 2010; Moir and Moot 2010; Sarker et al. 2014), and by the raised pH (Adorolo et al. 2015; Yang et al. 2005). This is confirmed by the low exch. acidity and the high pH of soils limed with higher NPR rates (Table 2). However, the lower NPR rates were not sufficient to eliminate the toxicity of displaced Al, and therefore spinach growth was reduced by the reduced RL and RHSAI reduced plant growth. Resistance to Al toxicity is necessary for spinach to survive in acidic, Al-toxic soils (Yang et al. 2005). Trends in DMY and leaf P roughly followed one another, generally increasing with increased NPR addition. The increase in leaf P could be due to the increase of soil pH to favorable levels and the reduction of the activity of Fe and Al hydroxides, thereby making P available and easy to take up. Also, with high rate of P release by NPR dissolution, soil sorption sites are satisfied and the P level increases to sufficiency for spinach production. This is evident from the reduction of exch. acidity, increase of NPRD and increase of available P as the rate of NPR increases (Table 2, 4). In general, the low exch. acidity and sufficient supply of P due to application of NPR at 1000 and 2000 kg ha⁻¹ favored plant growth, as evident from the high values of growth parameters (LSAI, RHSAI and RL) and the high amounts of P taken up by plants (leaf P), and as evidenced by a high DMY due to the application of NPR or ground PR (Adhikari et al. 2014; Dodor 2016).

4.4. NPR dissolution (NPRD) and increased in plant P (IPP)

The addition of NPR increased soil pH and available P, which indicates NPR dissolution. The increase in soil pH due to the consumption of protons during the acidulation of NPR and the subsequent neutralization of bases released, as well as the decrease in P fixation and promotion of P availability due to the liming effect, is another strong indication of NPR dissolution (Adhikari et al. 2014; Dodor 2016). In addition, the liming effect and its consequent increase of pH favored spinach root growth, which increased the dissolution of NPR through root activities (Baligar et al. 1997; Dodor 2016). We expect that the nanoparticle with larger surface area could increase the PR effectiveness and possibly the final extent of dissolution by increasing the degree of contact between the nanoparticle and the soil solution, thereby could enhance the rate of chemical reaction between H⁺ in the soil solution and the nanoparticle, because dissolution is most rapid when the exposed PR surface area is large and PR is well dispersed throughout the soil (Arifin et al. 2018; Devnita et al. 2018; Rafael et al. 2018; Zhao et al. 2014). Thus, the increased dissolution rate increases the plant P (Table 4) (He et al. 2005). However, a regular application of NPR and further research for economic comparison between NPR and both of lime and superphosphate, as well as the original PR, will be needed.

5. Conclusion

The results of this study show that the application of NPR as a source of both lime and P had good effects on the growth (LSAI, RL and RHSAI) and yield (DMY) of spinach. NPR at 1000 and 2000 kg ha⁻¹ proved equally effective at both goals. So from an economic point of view, 1000 kg ha⁻¹ is most justified. A single application of NPR at 1000 kg ha⁻¹ produced a rapid release of P followed by a slower conversion into less available forms; P in the NPR is most available during the first 3 weeks after application and remains sufficient during the early growth of spinach, but it declines gradually with crop growth, although it is still favorable for crop production. The addition of NPR greatly improved spinach DMY, principally by supplying P, but also by raising pH and reducing Al toxicity. Although these results indicate that NPR has both P fertilizer and liming effects on tropical acid soils. However, a single application cannot sustain soil pH beyond the first crop. Therefore, a regular application of NPR and further research for economic comparison between NPR and both of lime and superphosphate, as well as the original PR, will be needed.

Acknowledgments

The Japan International Research Centre for Agricultural Sciences (JIRCAS) provided funding through its visiting research fellowship program. We thank the staff of the Tropical Agriculture Research Front for the facilities and their cooperation in this study. Special thanks to Dr. Satoshi Tobita and Dr. Takeshi Watanabe for reviewing the first draft of this manuscript.

Supplementary material

Supplemental data for this article can be accessed here.

Additional information

Funding

This work was supported by the Japan International Research Centre for Agricultural Sciences (JIRCAS).