Synthesis of biopolymeric particles loaded with phosphorus and potassium: characterisation and release tests

Abstract

The authors synthesised nanoparticles (NPs) loaded with P and K from KH₂ PO₄ using gelatin type‐A and type‐B, and sodium alginate as carriers. Using type‐A and type‐B gelatin, quasi‐spherical particles were obtained, with average sizes of 682 and 856 nm, respectively; with sodium alginate, the resulting NPs exhibited spherical shapes and 600 nm particle average size. The authors found an interaction between KH₂ PO₄ and alginate via the hydrogen bonds existent among the carboxylic groups of the carbohydrate and the OH‐groups of the H₂ PO₄ ‐; interactions among gelatin types with the OH‐groups and the H₂ PO₄ ‐ion were also observed. Adding trypsin to the distilled water solutions of the NPs coated with type‐A gelatin increased the concentration of P in the solution by threefold, while increasing that of K increased by 2.6‐fold. Conversely, adding α ‐amylase to the water solutions with sodium alginate increased the P and K concentrations in the solution by nearly 1.3‐ and 1.1‐fold, respectively. Thus, sodium alginate resulted in NPs with smaller sizes and better spherical formations, though with a high polydispersity index and lower release rate of P and K. This low release rate represents an advantage since plants demand nutrients for long periods, and conventional fertilisers display low use efficiency.

1 Introduction

With rising demand for food, feed, fibres, and biofuels, optimum use of plant nutrients is becoming a paramount priority. The inefficient use of conventional fertilisers may result from chronically late delivery of program fertilisers, non‐responsive soil conditions, poor management practices, and inefficient use of complementary inputs [1]. Conversely, slow‐release fertilisers may increase the efficient use of such agricultural inputs, while minimising their environmental impacts [2]. The so‐called nanonutrients have become important in agriculture because their nano‐structured formulae can increase the efficiency in the use of fertilisers and the nutrient absorption rate in soils, thus reducing production costs and negative environmental impacts [3, 4].

Nanonutrients delivery has been performed using different biopolymeric nanoparticles (NPs) as carriers, including proteins such as albumin, collagen, and gelatin as well as polysaccharides like alginate and chitosan [5, 6]. Biopolymeric NPs can be prepared using three basic methods: (i) Emulsification; (ii) Nanoprecipitation; and (iii) Drying [7]. The synthesis of alginate NPs may include desolvation and counter ion induce aggregation, while gelatin NPs may involve phase separation in aqueous medium and pH‐induced aggregation [8]. Ultrasonic and microwave tools have been successfully used to prepare biopolymeric NPs and nanonutrients [5, 9].

The total consumption of major fertilisers (N + P₂ O₅  + K₂ O) was estimated to be 190.7 million metric tons in 2015. Since 2012, demand has grown annually by 1.8%; thus, it is predicted that by 2020, fertiliser usage will exceed 200 million tons [1, 10].

Phosphorous (P) accounts for ∼0.2% of a plant's dry biomass. P is a component of key molecules (i.e. nucleic acids, phospholipids and ATP); thus, plants with low P availability exhibit limited growth. P also controls enzymatic reactions and vital metabolic pathways [11].

Potassium (K) is the most abundant cation in plants [12]. K plays a key role in the plant cell, primarily in functional processes such as enzymatic activation, transport, osmoregulation, synthesis of starch and proteins, stomatal movement and ion charge balance [13]. After its absorption, K⁺ is stored in the plant's cells and constitutes between 2 and 10% of the dry matter weight. High K concentration in leaf tissues increases photosynthesis as well as the translocation of photoassimilates and amino compounds from source organs to reproductive organs through the phloem [14]. Although K⁺ determines yields and crop quality, excessive use of K‐fertilisers leads to their leaching into the soil, contributing to environmental contamination [15]. This study attempted to synthesise biopolymer NPs loaded with phosphorus and potassium from potassium dihydrogen phosphate, using the nanospray‐drying technique. This study also aimed to estimate the release time of these macronutrients in nutrient solutions.

2 Materials and methods

2.1 Preparation of nanoparticles

We first evaluated the use of type‐B gelatin as a coating polymer. To achieve this, 0.2 g of KH₂ PO₄ was weighed and dissolved in 20 ml of deionised water. Subsequently, 0.4 g of type‐B gelatin was dissolved in 80 ml of deionised water at 120°C. Both of the resulting solutions were filtered through 0.45‐µM‐pore PVC filters and mixed. The mix was then shaken for 60 min at 140°C, followed by sonication for 60 min using an ultrasonic bath (TI‐H‐5; Elma, Germany). After sonication, 4 mg of glutaraldehyde was added to the solution, which was then shaken for 20 h at room temperature. Finally, the solution was sonicated again for 60 min and filtered through a Whatman 542 filter paper and after that using a 0.22‐µm‐pore PVC filter. To form NPs, a Nano Spray Dryer B‐90 (Flawil, Switzerland) was used with a 4‐µm mesh. The drying temperature inside the machine was 100°C at 35 Mbar with 100% aspersion. The resulting powder was collected from the bottom of the machine and stored for the subsequent analyses of size, morphology, concentration, and P and K release. This procedure was repeated using only the polymer (i.e. type‐B gelatin) to synthesise unloaded NPs (i.e. negative control).

A second assay evaluated the use of type‐A gelatin as a coating polymer based on the methodology described for type‐B gelatin. However, a few parameters were different: the shaking time of the KH₂ PO₄ and type‐A gelatin mixture was increased to 180 min, and the machine was set to a drying temperature of 95°C with 80% aspersion.

Finally, a third experiment used sodium alginate as a coating polymer based on the aforedescribed methodology.

2.2 Morphology, particle size analysis and polydispersity index (PDI)

The size and morphology of the NPs obtained in all three assays were evaluated using a scanning electron microscope (SEM) [JOEL model JSM‐6390LV; Tokyo, Japan]. Previously, the samples were covered with gold‐palladium via sputtering (Desk IV, Denton Vacuum; Moorestown, NJ, USA). The micrographs were taken at 7000x zoom using an accelerating voltage of 20 kV.

According to Rane and Choi [16], polydispersity index (PDI) was determined with the following formula: PDI = (the square of the deviation standard of diameter data)/(the square of mean diameter).

Given the morphology and size characteristics observed in the NPs with the use of type‐A gelatin and sodium alginate as a polymer, the determinations described below were only performed for these NPs.

2.3 FTIR analysis

A Cary 630 FTIR spectrometer (Agilent Technologies; Santa Clara, CA, USA) was used to evaluate the possible interactions between KH₂ PO₄ and the biopolymers (alginate and type‐A gelatin) inside the NP. These analyses were performed using an attenuated total reflectance (ATR) cell over a range of 4000–600 cm⁻¹ with a resolution of 4 cm⁻¹ for 32 scans.

2.4 Differential scanning calorimetry (DSC)

By using a Pyris 1 differential scanning calorimeter (DSC; Perkin Elmer; Waltham, MA, USA), the thermal properties of NPs and their precursors were determined. Samples of 3 mg were introduced in aluminium capsules, and subsequently sealed. The heating velocity in the equipment was 10°C/min, with a temperature interval of 30–200°C. Temperature and heat flow calibrations were performed with a reference material (Indium, No. 0319‐0033). An empty capsule was used as a reference in all measurements performed.

2.5 Release of P and K in water, with enzyme additions

To increase the release of P and K from the type‐A gelatin nanocapsules, trypsin was added. Trypsin is a pancreatic serine protease with substrate specificity on the positively charged lateral chains of lysine and arginine. This enzyme catalyses the hydrolysis of the ester and amide bonds as well as those from peptides on the carboxylic side of arginine, lysine, and ornithine [17]. The enzyme α‐amylase was added to the sodium alginate NPs. This enzyme catalyses the hydrolysis of α‐1,4 glycoside bonds in starch and other related carbohydrates. The active site has several charged groups, including three amino acids (Asp231, Glu261, and Asp238, BLA numbering), which are vital for catalytic activity [18]. Of the NPs coated with both polymers (i.e. type‐A gelatin and sodium alginate), 1.5 g were weighed and dissolved in 200 ml of deionised water. During the first 72 h, the P and K concentrations were determined at 24 h intervals. After 72 h, the corresponding enzyme was added to the solutions at a concentration of 0.4 mg/ml. The solutions were shaken continuously, and samples were taken 5 min after the addition of the enzyme at 10 min intervals during the subsequent 30 min (72 h 35 min) to determine the P and K concentrations present using inductively coupled plasma atomic emission spectroscopy (Agilent, ICP‐ES 725; Mulgrave, Australia). In parallel, the release of P and K from the synthesised NPs was quantified (0 to 72 h 35 min) without the addition of enzymes.

3 Results and discussion

3.1 Size and morphology of the nanoparticles

The SEM micrographs showed that the type‐B NPs loaded with KH₂ PO₄ have a quasi‐spherical morphology (Fig. 1 a). The size of the NPs varied between 100 and 1400 nm with a mean diameter of 755 nm. A total of 50% of the particles of the analysed sample exhibited a size smaller than or equal to 680 nm (Fig. 1 b). The type‐B gelatin NPs loaded with KH₂ PO₄, displayed a moderately polydisperse distribution type with a PDI value of 0.324.

Fig. 1.

Micrograph of the type‐B gelatin nanoparticles loaded with KH₂ PO₄ (a), and particle size distribution histogram (b)

Similar to the use of type‐B gelatin as a polymer, the use of type‐A gelatin caused the formation of quasi‐spherical particles (Fig. 2 a). The size of the NPs in this second assay varied between 100 and 1100 nm, with an evident reduction in the mean particle size to 615 nm. It was also observed that 90% of the NPs had sizes between 100 and 800 nm (Fig. 2 b). The PDI value was of 0.227, which indicated a moderately polydisperse distribution type of these NPs.

Fig. 2.

Micrograph of the type‐A gelatin nanoparticles loaded with KH₂ PO₄ (a), and particle size distribution histogram (b)

Conversely, using sodium alginate as a polymer, the formation of spherical particles was observed (Fig. 3 a), as well as NP sizes varying between 100 and 1200 nm, with a mean diameter of 526 nm.

Fig. 3.

Micrograph of sodium alginate nanoparticles loaded with KH₂ PO₄ (a), and particle size distribution histogram (b)

Similar to the use of type‐A gelatin, a significant concentration of the particle size was observed in values below or equal to 800 nm (87%, Fig. 3 b). These NPs displayed a PDI of 0.569, which means a highly polydisperse distribution.

NPs are commonly described as solid colloidal particles that range in size from 10 nm to 1 µm [19]. Harsha [20] developed gelatin NPs loaded with amoxicillin using a Büchi Nano Spray Dryer B‐90. This method proved to be gentle, continuous, and scalable, allowing the transformation of liquids into dry powder with high yields. The NPs obtained in the Harsha [20] study showed a rough morphology similar to that obtained in this study (Figs. 1 and 2). Such morphology can be caused by quick drying during pulverisation and could be advantageous because it translates into a larger surface area. The gelatin NPs loaded with amoxicillin have a heterogeneous particle‐size distribution. This result agrees with those obtained using type‐B gelatin, with a size interval of 100 to 1400 nm, while the size interval obtained using type‐A gelatin ranges between 100 and 1100 nm. When NPs are of uniform size, the space between them is larger, but when they exhibit diverse sizes, the empty spaces between the NPs are filled by the smaller NPs. According to Harsha [20], the mean diameter of the amoxicillin NPs was 571 nm, while those of the type‐A and type‐B gelatin NPs loaded with KH₂ PO₄ were 755 and 615 nm on average, respectively. Additionally, the fit value R ²  = 0.69 for the type‐B NPs shows that the model explains nearly 70% of the phenomenon between the ratio of the particle size and the number of particles (Fig. 1 b). In the type‐A gelatin NPs, the ratio is higher because R ²  = 0.92, which indicates that there is a larger total variation explained by the model (Fig. 2 b). Finally, for the alginate NPs, the prediction model showed a good fit with R ²  = 0.77 (Fig. 3 b).

Conversely, Hasaneen et al. [21] obtained spherical NPs loaded with NPK with a homogenous size distribution of 20 ± 2 nm by synthesising chitosan NPs using methacrylic acid polymerisation in a two‐step process. Similarly, Corradini et al. [22] synthesised chitosan NPs via the polymerisation of methacrylic acid to incorporate NPK fertilisers using calcium phosphate [Ca(H₂ PO₄)₂ H₂ O], urea [CO(NH₂)₂], and potassium chloride (KCl) as sources. The NPs showed a spherical shape with a homogenous size distribution, and the mean diameter of the chitosan NPs in their dry state was ∼78 ± 1.5 nm.

It has been shown that the concentration of cross‐linker is positively associated with particle size [23]. Likewise, Aramwit et al. [24] reported the glutaraldehyde‐cross‐linked type‐B gelatin NPs (40.39%) showed higher crosslinking degree than the glutaraldehyde‐crosslinked type‐A gelatin NPs (36.98%). These data are in full agreement with our results reported herein, since the type‐B gelatin NPs showed not only a greater average particle size than the type‐A gelatin NPs but also a higher PDI.

The PDI values of type‐A and type‐B gelatin NPs classified as moderately polydisperse stand out, as well as that of sodium alginate NPs classified as highly polydisperse. One strategy for obtaining smaller particles, with controlled polydispersity and limited formation of large aggregates, is the adjustment in the relationships between the biopolymer and the cross‐linking agent [25, 26]. Another factor affecting these properties is the pH value, since small and extremely stable gelatin NPs can be obtained at pH 3, while at pH 2 NPs with lower polydispersity can be synthetised [25].

3.2 Evaluation of the interaction between KH₂ PO₄ and the biopolymers

The KH₂ PO₄ spectrum showed characteristic bands at 2321 cm⁻¹ (tension vibrations of the O‐H bond), 1270 cm⁻¹ (bending vibrations of the O‐H bond), and 1050 and 830 cm⁻¹ (tension vibrations of the PO₄). On the other hand, the alginate NPs (NPAlg) spectrum revealed characteristic absorption bands at 3256 cm⁻¹, corresponding to the tension vibration of the O‐H bond, 2925 cm⁻¹, belonging to the tension vibration of the O‐H bond in the carboxylic group, 1591 and 1401 cm⁻¹, corresponding to the asymmetric and symmetric tension vibrations, respectively, of the carboxylate anion (COO^‐), and 1021 cm⁻¹, corresponding to the asymmetric tension of the C‐O‐C bond. The type‐A gelatin NPs (NPGA) spectrum revealed characteristic absorption bands at 3275 cm⁻¹, corresponding to the amide A (tension vibrations of the N‐H bond), 1632 cm⁻¹, corresponding to the amide I (tension vibrations of the C=O bond), 1524 cm⁻¹, corresponding to the amide II (bending vibrations of the N‐H bond and tension vibrations of the C‐N) and, finally, 1240 cm⁻¹, which is characteristic of the amide III (bending vibrations of the N‐H bond) (Fig. 4).

Fig. 4.

Fourier transform infrared (FTIR) spectra of KH₂ PO₄, alginate nanoparticles (NAlg), alginate nanoparticles loaded with KH₂ PO₄ (NPAlg‐KH₂ PO₄), type‐A gelatin nanoparticles (NPGA) and type‐A gelatin nanoparticles loaded with KH₂ PO₄ (NPGA‐KH₂ PO₄)

In the spectrum of type‐A gelatin NPs loaded with KH₂ PO₄ (NPGA‐KH₂ PO₄), the band belonging to the bending vibrations of the O‐H bond in H₂ PO₄ (1270 cm⁻¹) was absent. Similarly, it is not possible to visualise the bands located at 1050 and 830 cm⁻¹ in the KH₂ PO₄ spectrum (Fig. 4). In the FTIR analysis, the KH₂ PO₄ spectrum showed characteristic bands that coincide with those described by Hayashi and Mukamel [27]; the NPAlg spectrum also showed characteristic bands [28, 29].

3.3 Differential scanning calorimetry (DSC)

Fig. 5 displays the thermograms of the particles analysed.

Fig. 5.

Thermogram of the fertiliser (KH₂ PO₄), the type‐A gelatin nanoparticles (NPGA), the type‐A gelatin nanoparticles loaded with fertiliser (NPGA‐KH₂ PO₄), the alginate nanoparticles (NPAlg), and the alginate nanoparticles loaded with fertiliser (NPAlg‐ KH₂ PO₄)

Table 1 shows the fusion temperature and the glass transition temperature (T _g). T _g refers to the temperature at which the mechanical properties of a material radically change due to the internal movement of the polymer chains that form the material. It is the temperature at which the material physically changes from a solid to a liquid or from a liquid to a solid [30]. The T _peak value for KH₂ PO₄ was 273.03°C, with an enthalpy of 682.22 J/g. In NPGA and NPGA‐KH₂ PO₄, the T _g values were 73.17 and 79.97°C, while the T _peak values were 120.01 and 128.69°C, respectively. In both cases, i.e. T _g and T _peak, temperature values rose when gelatin NPs were loaded with KH₂ PO₄. Conversely, the KH₂ PO₄ T _peak value of NPGA‐KH₂ PO₄ dropped to 253.84°C. On the other hand, NPAlg and NPAlg‐KH₂ PO₄ exhibited T _g values of 89.86 and 134.69, and T _peak values of 120.84 and 187.32, respectively. In both groups of samples, there is an exothermic pick with a T _peak of 237.85°C for NPAlg and of 245.85°C for NPGA‐KH₂ PO₄. These responses are attributed to dehydration and depolymerisation reactions due to a possible partial decarboxylation of the protonated carboxylic groups releasing CO₂, which alters the polymer chemical structure [29].

Table 1.

Fusion temperature (T _peak), transition temperature (T _g) and enthalpy (ΔH) of the fertiliser (KH₂ PO₄), type‐A and alginate nanoparticles loaded and non‐loaded with KH₂ PO₄

Sample	KH2 PO4				Polymer
	T onset, °C	T peak, °C	T end, °C	ΔH, J/g	T onset, °C	T peak, °C	T end, °C	T g, °C	ΔH, J/g
KH2 PO4	271.58	273.03	278.04	682.22
NPGA					105.29	120.01	140.03	73.17	64.34
NPGA‐KH2 PO4	247.91	253.84	257.7	60.48	116.40	128.69	143.29	79.97	25.46
NPAlg					104.98	120.84	139.02	89.86	66.8
					229.34a	237.85	254.36		−369.95
NPAlg‐KH2 PO4	185.67	187.32	193.36	134.69	185.67	187.32	193.36	134.69	2.99
					236.36a	245.85	258.07		−27.49

^a Data correspond to the carbonisation of alginate.

Type‐A gelatin nanoparticles (NPGA), type‐A gelatin nanoparticles loaded with fertiliser (NPGA‐KH₂ PO₄), the alginate nanoparticles (NPAlg), and alginate nanoparticles loaded with fertiliser (NPAlg‐KH₂ PO₄).

When combined with the polymers (NPGA‐KH₂ PO₄ and NPAlg‐KH₂ PO₄), KH₂ PO₄ exhibits a lower T _peak value, possibly due to the concomitant interactions among the polymers and the glutaraldehyde molecules (a cross‐linking agent). In particular, the thermogram corresponding to NPAlg‐KH₂ PO₄ exhibited a single endothermic pick, differing from that displayed by NPGA‐KH₂ PO₄, which showed two endothermic picks with the aforementioned values. Thus, NPAlg‐KH₂ PO₄ may form a new chemical structure, and hence the T _peak value for KH₂ PO₄ and for the polymer are the same (187.32°C).

3.4 Release of P and K with the addition of enzymes

P and K concentrations in the type‐A gelatin and sodium alginate NPs loaded with KH₂ PO₄ were 7.59 and 9.58%, respectively. The solutions with NPs for the release rate evaluation had P y K concentrations of 568.93 y 718.27 mg/L, respectively.

In the type‐A gelatin NPs loaded with KH₂ PO₄ (where trypsin was added after 72 h of shaking in distilled water), the concentrations of P and K were evaluated at 5, 15, 25 and 30 min after the addition of the enzyme. In the sampling performed 5 min after the addition of trypsin, the concentration of P increased threefold, while that of K increased 2.6‐fold compared to the mean concentration obtained without the addition of the enzyme at 24, 48 and 72 h. Measurements performed 15, 25, and 35 min after the addition of trypsin revealed no significant changes in the concentrations of P and K. This same tendency was observed in the samples to which trypsin was added, given that P and K concentrations were constant from 24 to 72 h 35 min, with average values of 71 and 91.7 mg/kg, respectively (Fig. 6).

Fig. 6.

Release of P and K from type‐A gelatin nanoparticles loaded with KH₂ PO₄ without and with addition of the enzyme trypsin

In the tests of P and K release from sodium alginate NPs loaded with KH₂ PO₄, the enzyme α ‐amylase was used. Prior to the addition of α ‐amylase, the concentration of P had a value of 74.4 mg/kg, and the concentration of K was stable with a mean value of 109 mg/kg. Similar concentration values were registered in these NPs that did not receive α ‐amylase in the measurement from 24 to 72 h 35 min. Five minutes after the addition of α ‐amylase, increases in the concentrations of P and K close to 1.3‐ and 1.1‐fold, respectively, were observed (Fig. 7).

Fig. 7.

Release of P and K from sodium alginate nanoparticles loaded with KH₂ PO₄ without and with the addition of the enzyme α‐amylase

The enzyme α ‐amylase was less efficient in the release of P and K from NPs with sodium alginate compared to trypsin in the release from type‐A gelatin NPs (Figs. 6 and 7). Alternatively, the enzyme alginate lyase can degrade alginate by β ‐elimination of glycosidic bonds and produce unsaturated oligosaccharides with double bonds at the non‐reducing end [31]. This enzyme may be used instead of α ‐amylase, in order to ensure a higher fertiliser release. Importantly, a number of alginate lyases have been identified, had their genes cloned, and been purified and characterised from various organisms.

When studying the released fractions from type‐A gelatin NPs loaded with the chemical doxorubicin (DXR) and stabilised with glutaraldehyde, only a small amount of DXR was released as a free chemical (8%) when the particles were placed in saline solution [32]. The degradation of the protease induced the release of 10% of the loaded chemical, while the α‐chymotrypsin and trypsin allowed a higher amount of DXR to be released (15 and 20%, respectively).

These results are in full agreement with those obtained when adding trypsin to NPs loaded with KH₂ PO₄ and type‐A gelatin, where the concentration of P increased three times, while the K concentration increased 2.6 times; both cases were in comparison with the mean concentration obtained without the addition of the enzyme after 24, 48, and 72 h. Conversely, with the addition of α ‐amylase to the sodium alginate NPs, increases in the concentrations of P and K equal to 1.3‐ and 1.1‐fold, respectively, were observed.

The release rate of P with the addition of trypsin and α ‐amylase from type‐A gelatin and sodium alginate NPs was 38.5 and 17.4%, respectively, while that of K was 35.9 and 17.4%, respectively.

The higher particle size of the alginate NPs negatively affects the rate of release of P and K; because with high particle size, the specific surface area of the NPs is reduced, and, therefore, the solubility and dissolution rate can be decreased [33].

The low release rate of P and K from alginate NPs loaded with KH₂ PO₄, even with the addition of the enzyme α‐amylase, represents an advantage, considering the low efficiency of P and K supplied from conventional fertilisers, with values lower than 10 and ∼40%, respectively. These low use efficiencies are due to significant losses of nutrients by leaching, run‐off, gaseous emission, and fixation by soil [34].

4 Conclusion

The NPs loaded with KH₂ PO₄ exhibited smaller diameters and better spherical shapes when sodium alginate was used as the coating polymer. The FTIR tests indicate that there is an electrostatic interaction and hydrogen bonds between KH₂ PO₄ and the polymers tested (sodium alginate and type‐A gelatin).

The addition of trypsin was found to induce a higher release of P and K from the type‐A NPs. Conversely, in the sodium alginate NPs, the α ‐amylase showed a lower release of P and K. Thus, to increase this release, the enzyme alginate lyase, which is specifically designed to degrade this polymer, can be used. However, this low release rate represents an advantage when it comes to fertilisers, given the demand for nutrients by the plant during relatively long periods and the low efficiency in the use of fertilisers.

The synthesis of NPs loaded with KH₂ PO₄ is a process with benefits that include the controlled release of P and K for plants, which produces more efficient nutrition and reduces negative environmental impacts caused by conventional fertilisers.