|
Controlling Nutrient Leaching Profile of Urea Granules through Structural Modification
Camila Jange1, Rhonda Graef2, Chad Penn2, Carl Wassgren3, Kingsly Ambrose1,*
Published in Journal of the ASABE 66(6): 1415-1424 (doi: 10.13031/ja.15675). 2023 American Society of Agricultural and Biological Engineers.
1Agricultural and Biological Engineering, Purdue University, West Lafayette, Indiana, USA.
2National Soil Erosion Research Laboratory, USDA ARS, West Lafayette, Indiana, USA
3Mechanical Engineering, Purdue University, West Lafayette, Indiana, USA.
*Correspondence: rambrose@purdue.edu
The authors have paid for open access for this article. This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License https://creative commons.org/licenses/by-nc-nd/4.0/
Submitted for review on 17 May 2023 as manuscript number NRES 15675; approved for publication as a Research Article by Associate Editor Dr. Durelle Scott and Community Editor Dr. Kati Migliaccio of the Natural Resources & Environmental Systems Community of ASABE on 23 August 2023.
Mention of company or trade names is for description only and does not imply endorsement by the USDA. The USDA is an equal opportunity provider and employer.
Highlights
Abstract. Ammonium and nitrate are byproducts of urea fertilizer hydrolysis in soil. Ammonium is highly unstable and can volatilize in the form of ammonia, a greenhouse gas. Meanwhile, nitrate is highly hydrophilic and can contaminate surface and groundwater systems. This study investigated the influence of a biopolymer binder (a mixture of xanthan and konjac gums) and granule microstructure on urea dissolution and hydrolysis in soil to improve fertilizer release rates. The study compared urea leaching profiles in disturbed soil columns for dry (uniaxial compression), layered wet granulated, and market urea granules. A power-law model of total dissolved nitrogen versus cumulative volume ensured robust estimation of the release rate constants. There was 50% and 20% less total dissolved nitrogen, respectively, for binder-added core (CB) granules produced from the uniaxial compression method and bilayer binder-added (bLB) granules compared with market urea samples (NU). However, no significant reduction in dissolved ammonium and nitrate was observed based on formulation and process changes using a power-law model. However, it is noteworthy that the power-law model overpredicted the initial leaching profiles of binder-added core (CB) and bilayer binder-added (bLB) samples. In conclusion, the microstructure of the core granules compacted at 100 MPa and binder-formulated (CB) granules can delay urea dissolution and suggests a partial reduction of urea hydrolysis in soil.
Soil properties and agricultural practices, including tillage, crop type, and fertilizer application methods, affect the leaching rate of fertilizer hydrolytic byproducts, such as nitrate and ammonium from urea (Shah and Wolfe, 2002; Jiao et al., 2004). Higher nitrate (NO3--N) movement occurs due to tillage, affecting soil nutrient concentrations and water flow patterns (Randall, 1990). Nitrate leaching contributes to a surplus of organic matter growth in surface water systems, resulting in eutrophication and oxygen depletion zones in estuaries such as the Gulf of Mexico. Nitrogen (N) is poorly retained by the soil due to weak ion exchange mechanisms, unlike phosphorus. For the most part, N that is not used by the plant over the growing season will be lost to the environment through leaching, volatilization, or denitrification, which is why N is applied on an annual basis. Traditional N fertilizers are either nearly 100% soluble or applied in a solution such as urea-ammonium-nitrate (Trenkel, 2010). Due to the high aqueous soluble nature of N, the development of slow-release N fertilizers that match a plant’s uptake profile could considerably decrease N leaching. Solid urea fertilizer is less prone to leaching compared to other chemical N sources since conversion to ammonium requires urease enzyme and microbiological mineralization, which makes it an ideal candidate for applying various coatings or binders in the development of a controlled release or slow-release fertilizer, respectively.
The role of coatings in controlled release fertilizers and binders in slow-release fertilizers in reducing dissolution onset has been extensively investigated (Naz and Sulaiman, 2016; Wang et al., 2022). Nonetheless, a mechanistic understanding of granule pore structure through process and formulation design is crucial in the field of precision agriculture. Optimizing formulation and process design could reduce leaching and improve the functional properties of granular fertilizers (Mehrez et al., 2014). Structural modification of granules by layer-wise agglomeration has been shown to reduce the rate of dissolution of urea granules (Ambrose et al., 2021). The addition of binders reduced the gelation at the solid-liquid interface, which in turn reduced solvation and dissolution and delayed nutrient release from urea composites (Jange et al., 2021). Jange et al. (2023) reported that the dry granulation process at high compaction loading and the use of biodegradable polymer binders reduced the dissolution rate of urea granules in water bath studies.
This study aims to understand the effect of structural modification through granulation processing and binder addition on the dissolution onset and hydraulic pathway to ammonium and nitrate in urea fertilizer granules. The broader impact and novelty rely on using microstructure design to improve the performance and cost-effectiveness of biodegradable binders in smart urea fertilizer technology towards mitigating nitrate leaching. Thus, the objective of this research was to study the effect of a fertilizer granule’s internal density distribution and formulation design on total dissolved nitrogen and ammonium (NH4+-N) and nitrate (NO3-N) conversions in soil.
Materials and Methods
Materials
In this study, technical urea powder (100% purity, Rose Mills Co., West Hartford, CT) was used to prepare urea granules. A mixture of konjac and xanthan gums was used as the binder (Ingredion Inc., Westchester, IL). Since the binder composition is proprietary information of the industry, the chemical composition is not provided. The methods of producing granules are described in the next section. The different types of urea granules produced and tested in this study are: control core granules (100% technical area; CC), core granules with binder (CB), control bilayer granules (CbL), and bilayer granules with binder (bLB).
Production of Core and Bilayer Granules
The control core granules (CC) and control+binder (CB) (with 5% binder, w/w) compacts were produced using a flat faced punch-die set in an MTS uniaxial compression machine (Machine Testing Systems Corporation, Eden Prairie, MN). Before the compaction of urea, the punch wall was lubricated with magnesium stearate to reduce inconsistencies in radial stress distribution. The lubrication procedure consisted of applying a fixed normal load of 2.5 MPa and ejection of the magnesium stearate compact. The punches were cleaned to remove any excess magnesium stearate adhered to the punch wall. Finally, the compaction to produce control core (CC) and control + binder (CB) compacts was performed at a strain rate of 0.5 mm/s, the data acquisition rate was fixed at 10 Hz, and the load endpoint was fixed at a maximum pressure of 100 MPa. The selected pressure was based on our previous dry compaction optimization studies, in which granules produced from compacts at 100 MPa and a mixture of xanthan and konjac gum binder showed the best urea dissolution performance (Jange et al., 2023). The compacts were 19 mm in diameter and 6 mm in thickness (d/h = 3.33). The compacts were stored for 24 h at 40% (±10%) relative humidity to allow for stress relaxation. Compacts were then milled to produce control core (CC) and core+binder (CB) granules using a coffee grinder and sieved to select particle sizes between 1-2.8 mm.
The core granules from sieve cuts of 0.71-1.7 mm were drum granulated along with technical urea fine powder (diameter based on sieving analysis is 100 µm) to produce bilayer granules using water as a liquid binder. The core to fine ratio was fixed to 15.0% (w/w). Two types of bilayer granules were produced: control bilayer granules (only urea as fine powder; CbL) and bilayer + binder granules (95% weight basis urea and 5% weight basis binder mix, bLB). The drum granulation was conducted in a 6 cm in diameter and 4.62 cm in height (D/H = 1.3) drum operating at 10 RPM for 100 revolutions, 50% fines (w/w- relative to the mass of core granules added to the drum), 9% liquid to solid ratio, and 5% fill level. The selected process parameters are based on a series of iterations reported in Jange et al. (2023). Both core and bilayer granules with or without binder were sieved to achieve particle sizes in the range of 2-2.8 mm for further dissolution analysis on soil-column experiments. The particle size range of 2-2.8 mm was selected as it exhibited better layered granule yield (Jange et al., 2023).
Laboratory Flow-Cell Experiments
The total N content of the fertilizer granules was measured by dry combustion (LECO, St. Joseph, MI, USA) prior to amending in soils (Nelson and Sommers, 1996). The total N concentrations of fertilizer were then used in calculating soil application rates for achieving 168 kg total N Ha-1, based on a typical soil-mixing depth of 10 cm and a soil bulk density of 1.35 g cm-3. Table 1 displays the measured total N (TN) content of each fertilizer and the subsequent mass applied to each flow-cell soil column. In addition, a control trial without soil fertilizer application was used to compare the other treatments.
| Table 1. Total nitrogen (TN) and application mass required for achieving equivalent application rates of 168 kg N Ha-1 in the flow-cell soil column experiments. | ||||
| Processing Method | Sample | Sample Acronym | Nitrogen Percent% (LECO, w/w) | Sample Mass Required in Flow-Cell (g) |
| Drum granulation | Normal urea | NU | 46.4 | 0.0243 |
| Dry compaction | Core control | CC | 46.2 | 0.0244 |
| Core+binder | CB | 39.5 | 0.0286 | |
| Layer-wise granulation | Control bilayer | CbL | 46.2 | 0.0244 |
| Bilayer+binder | bLB | 41.3 | 0.0273 | |
Many authors have used the disturbed column measurement technique to study leaching (Smith et al., 1992; Pappas et al., 2008). The method described here consists of a continuous flow of water through a soil column. Silt loam soil samples obtained from sieve analysis (< 2 mm) were used in the laboratory flow-cell experiments (Penn et al., 2022) to characterize the loss of N as influenced by the urea granules’ formulation design. A Whatman 42 filter was installed at the bottom of each flow cell to prevent the loss of soil material. Fifty grams of soil sample was placed in the cell and then pre-wetted with deionized water to achieve a 25% (v/w) moisture content. Approximately 0.024 g of urea sample was then placed in each cell, and an additional 50 g of soil was added on top of the urea sample. The soil was again pre-wetted with deionized water. Each cell had a nozzle at the bottom that was connected to a peristaltic pump. While a Mariotte bottle maintained a constant head on the soil column/cell, the pump pulled water through the cell at a fixed flow rate (0.20 mL min-1). Experiments were performed with and without incubation, i.e., immediate leaching compared to 2 days of incubation prior to the initiation of leaching. Figure 1 shows a schematic of the flow-through experimental set-up.
 |
| Figure 1. Schematic of the laboratory flow-through cell following the Penn et al. (2022) experimental setup. |
All leachate was collected; samples were separated with a fraction collector every 180 min to obtain 15 samples at a total time of approximately 45 h. Leachate samples were stored at 4°C until further analysis. The flow-cell experiments were performed in duplicate.
Leachate samples were analyzed for total dissolved nitrogen, ammonium, and nitrate contents. Ammonium and nitrate were determined colorimetrically using a Gallery Discrete Analyzer (Thermo Fisher Scientific, West Palm Beach, FL) following the ISO 15923-1 standard (Nelson, 1987). Total dissolved N (TDN) was determined by digesting leachate samples with persulfate (ISO, 2013) prior to colorimetric analysis of ammonium and nitrate. The difference between total dissolved N and ammonium plus nitrate is dissolved organic N. The TDN, ammonium, and nitrate losses were compared between the urea fertilizer treatments using nonlinear models by JMP software analysis (SAS Institute Corporation, Cary, NC), as described in the next section. The TDN, ammonium, and nitrate concentrations were expressed as a mass lost per unit mass of urea in equation 1:
(1)
where
x = mass fraction of the N lost (in mg g-1 of urea)
V? = volumetric flow rate (mL min-1)
tj – ti = time interval with j > i (min)
ppmN = N concentration in leachate over that time (106 mg mL-1)
murea = mass of fertilizer amended to the soil column (g).
Nonlinear Models and Statistics
The data obtained from the soil column experiments were fit to logarithmic and power-law relations, respectively, given as equations 2 and 3:
(2)
where k0 is the release rate constant (mg g-1 of urea s-1) and k1 is the intercept value.
(3)
where k0 is the release rate constant (mg g-1 of urea s-1) and n is the constant dependent on the geometry of the system and release mechanism, which varies from 0.45 < n < 0.89. The nonlinear regression analyses were performed using JMP software. The goodness of fit was evaluated based on the predicted versus measured profile of normalized urea concentration released. Statistical analyses were performed using analysis of variance (ANOVA) and Tukey’s mean comparison test at a significance level of 5% to compare the fitted model parameters.
Results
Flow-Cell Experiments without Incubation
The urea granular samples produced were designed to maintain a similar range of total solid fraction as reported in our previous study (Jange et al., 2023). All the modified urea samples were produced with core granules from dry compaction with and without a mixture of xanthan and konjac gums as a binder (Jange et al., 2021, 2023).
Total Dissolved Inorganic Nitrogen (NO3-N + NH4-N)
The N loss expressed per unit mass of soil and as a function of leaching time and cumulative leachate volume are presented in figures 2 and 3, respectively. As expected, there was significantly less inorganic N leached from non-amended soils compared with fertilizer-amended soils (p < 0.05). For instance, the release rate constant, represented by the slope of the logarithmic model curve of core+binder (CB) systems measured in mg of nitrogen released per g of soil, was 4.06 times higher than the release rate constant for the blank samples (0.0016 mg g-1 of soil min-1 for blank sample compared to 0.0065 mg g-1 of soil min-1 for the core+binder sample). This trend indicates negligible background N loss from the soil when comparing the results from fertilizer-amended soils. The degree and pattern of dissolved inorganic N from drum granulated urea (NU) were not significantly different from the other fertilizer granules (figs. 2 and 3). Yet, the inorganic dissolved nitrogen rates shown in figure 2 for non-incubated soils suggest an increased inorganic N loss from core+binder (CB) and bilayer granules produced with binder (bLB) (figs. 2b-d and 3b-d) compared with control core (CC), bilayer (bL), and market urea (NU). The logarithmic coefficients from the data fitting represent the leaching rates when N loss is expressed as a function of time (fig. 2). The insignificant differences are likely due to the lack of incubation time between soil and fertilizer before
 |
| Figure 2. Cumulative dissolved inorganic nitrogen (N) lost from non-incubated soil columns as a function of time: (a) control soil (no amendment), and fertilizer-amended soils, (b) control core (CC), core+binder (CB), and normal urea samples (NU), (c) control bilayer (CbL), and bilayer+binder (bLB). |
 |
| Figure 3. Cumulative dissolved inorganic nitrogen (N) lost from non-incubated soil columns as a function of cumulative volume: (a) control soil (no amendment), and fertilizer-amended soils, (b) control core (CC), core+binder (CB), and normal urea samples (NU), (c) control bilayer (CbL), and bilayer+binder (bLB). |
leaching was initiated, i.e., urea dissolution depends on the interaction time between fertilizer and soil microbiota for adequate hydrolytic conversion of urea into ammonium, nitrite, and nitrate. The lack of interaction hypothesis between soil and fertilizer was confirmed by performing flow tests in incubated samples, as discussed in section Flow-Cell Experiments with Incubation.
Total Dissolved Nitrogen (TDN)
The cumulative mass of dissolved nitrogen (organic + inorganic) expressed as a function of time and cumulative volumetric flow rate is presented in figure 4. The results include measurements conducted on market urea fertilizer (NU), core (CC), and core+binder (CB) fertilizers produced from compacts. Similar to total dissolved inorganic N, TDN losses were not significantly different between fertilizer types. Based on these results, two hypotheses can be made: (1) the binder and granule density are affected by physical and chemical variations, (2) the time for hydrolytic conversion and dissolution of urea was insufficient with no pre-incubation before leaching ensued.
In support of the first hypothesis, Fertahi et al. (2021) discussed in detail the incongruencies among the nutrient release profiles observed in solvent and soil studies due to physical (i.e., fluid flow profiles) and chemical (i.e., pH, enzymatic activity, microbial activity) differences. This trend can be observed from the slope of the logarithmic curve presented in figure 4a, which describes the leaching rate. The rate of TDN was higher for the core+binder fertilizer granules shown in figures 4a and 4b. This result also suggests that the soil type and microbiome might affect the binder gelling strength and increase N leaching (Cao and Mezzenga, 2020). Conversely, the rate of dissolved organic nitrogen was lower for market urea systems, although not significant.
 | ||||
| (a) | (b) | |||
| Figure 4. Total dissolved nitrogen (TDN) leached from non-incubated samples as a function of (a) time and (b) cumulative volume for urea (NU), control core (CC), and core+binder (CB) granules. | ||||
Flow-Cell Experiments with Incubation
Figure 5 shows soil + fertilizer samples incubated for 48 h before the leaching experiments. The soil sample containing gum-added granules (CB) had some partially disintegrated granules (fig. 5a). Meanwhile, no granules were visually noticeable after 48 h for core granules (CC) and market urea (NU) granules in soil samples (figs. 5b and 5c), which indicated a faster disintegration process. This finding suggests that granule densification and binder addition might hinder microbial activity. Total inorganic dissolved nitrogen and total dissolved nitrogen analyses were conducted to further understand this phenomenon.
Total Inorganic Dissolved Nitrogen (NO3-N + NH4-N)
Figure 6 displays the inorganic dissolved nitrogen content after 48 h of incubation as a function of leaching time, and figure 7 shows the same data as a function of the cumulative volume of leaching. Figures 6b and 6d and figures 7b and 7d represent the goodness of fit for logarithmic and power-law models, respectively. In this study, the goodness of fit in figures 6b and 6d and figures 7b and 7d demonstrated that the power-law is a better fit for the data compared to a log function due to the higher adjusted coefficient of determination and lower residual sum of squares.
Incubating the fertilizer samples for 48 h reduced variability between replicates compared with non-incubated soils, resulting in significant differences between treatments, unlike the previously discussed results from non-incubated soils in section ‘Flow-cell experiments without incubation’. These results suggest that the non-significant differences observed for dissolved ammonium and nitrate without incubation might be due to the lack of interaction time between fertilizer samples and soil microbiota (discussed in section ‘Total dissolved nitrogen’). Core+binder (CB) granules significantly reduced nutrient leaching compared with control core (CC) granules and normal urea (NU) samples. No significant difference in nitrate and ammonium content was observed between normal urea and control core samples (p > 0.05). Figures 7a and 7c also show no difference as a function of cumulative flow volume of leaching as observed from the overlapping standard deviations among cumulative flow volume. However, the leaching rate of core granules with gums (CB) was significantly lower compared to the gum-added bilayer (bLB) granules (p < 0.05).
The results indicated that the binders delay urea dissolution and, consequently, nutrient leaching due to the rise of a stable gel network at the solid-liquid interface (figs. 6 and 7). Yang et al. (2017) used scanning electron microscopy to demonstrate the effect of pore constriction at the solid-liquid interface in delaying urea dissolution. This trend suggests a positive relationship between granule densification and binder addition. Ye et al. (2020) reported a delay in urea dissolution onset due to smaller gaps between urea-polyester associations attributable to polyester chain crystallization. The disadvantage of using scanning electron microscopy to quantify porosity is that it is a surface technique and, hence, does not quantify internal porosity but rather from the
surface. However, our previous studies on urea granule microstructure design demonstrated and quantified the relationship between higher granule densities (i.e., low porosity distributions) and formulation design using bulk porosity analysis and X-ray microcomputed tomography (Jange et al., 2021). An increase in granule density and binder addition reduces granule disintegration and the rate of diffusive mass transport, as reported by Markl et al. (2017).
 | ||||
| (a) | (b) | (c) | ||
| Figure 5. Fertilizer-added soil samples after 48 h of incubation (before the leaching experiment): (a) Core+binder (CB), (b) core (CC), and (c) normal urea (NU). | ||||
 | ||||
| Figure 6. Cumulative dissolved inorganic nitrogen leached for 48 h incubated samples as a function of time (a) fitting logarithmic model, (b) goodness of fit for the logarithmic model, (c) fitting power-law model, and (d) goodness of fit for the power-law model. NU- normal urea samples, CC- control core, CbL- control bilayer, CB- core+binder, bLB- bilayer+binder. | ||||
Total Dissolved Nitrogen
Figures 8 and 9 display TDN leaching after 48 h of incubation as a function of leaching cycle time and cumulative flow volume, respectively. Figures 8b and 8d and figures 9b and 9d represent the goodness of fit for logarithmic and power-law models, respectively. As expected, the total dissolved nitrogen for the incubated samples presented significant differences between the core+binder (CB) samples produced from compacts and the other samples (p < 0.05) (figs. 8 and 9). The total dissolved nitrogen also showed a delay in leaching rates for core+binder (CB) granules. Pereira et al. (2022) also reported a decrease in inorganic and total dissolved nitrogen leaching in biochar-based organomineral urea fertilizer produced using a mechanical pelleting process. The pelletized organomineral urea fertilizer in Pereira et al. (2022) reduced total dissolved nitrogen from 27% to 60% compared to conventional fertilizer. This trend proves that urea densification and binder addition mitigated nitrogen mineralization and, therefore, have the potential to reduce nutrient leaching in soil. Likewise, the total dissolved nitrogen profiles, as in the ammonium and nitrate studies presented in section ‘Total inorganic dissolved nitrogen’, exhibited a power-law model as best fitting based on the higher adjusted coefficient of determination and the lower residual sum of squares compared to a log equation.
 |
| Figure 7. Cumulative dissolved inorganic nitrogen leached for 48 h incubated samples as a function of cumulative volume (a) fitting logarithmic model, (b) goodness of fit for the logarithmic model, (c) fitting power-law model, and (d) goodness of fit for the power-law model. NU- normal urea samples, CC- control core, CbL- control bilayer, CB- core+binder, bLB- bilayer+binder. The standard deviations of cumulative dissolved ammonium and nitrate and total dissolved nitrogen contents are represented as a continuous interval in figures 7a and 7c. |
Comparing Parameter Estimates of Power-Law Models
The release rate constants estimated from the power-law models for dissolved inorganic N and TDN as a function of time and volumetric flow rate are displayed in figure 10. Figure 10a showed no clear differences between the estimated release rate constants from the soil dissolution profiles from dissolved inorganic nitrogen versus leaching cycle time. The core granules with binder (CB) exhibited high values of the release rate constant (figs. 10a and 10c). However, the constant “n” for core granules with binder was the lowest compared to the other samples (figs. 10 a and 10c). It is noteworthy that the power-law model for total ammonium and nitrate content overpredicts the initial data points for core + binder (CB) and bilayer + binder (bLB) granules and underpredicts the trend for market urea (NU) samples, as shown in figures 6d and 7d. These regions of overprediction and underprediction might affect the interpretation of the fitted parameters. Nevertheless, the TDN data presented significant differences for the release rate constants for time and cumulative flow volume factors (figs. 10b and 10d). In figures 10b and 10d, core+binder (CB) granules showed the lowest release rate constants compared with market urea (NU), control core granules (CC), and bilayer granules with or without binder produced from core granules (bLB and bL). Once again, the power-law model underpredicted the initial and final leaching regions as a function of leaching cycle time in the normal urea samples (NU), as shown in figure 9c. This trend explained the unusually low release rate constant estimated for normal urea (NU), as displayed in figure 10b. Nonetheless, figure 10d demonstrated a 20% and 50% reduction in the estimated release rate constants for bilayer granules with a binder (bLB) and control + binder (CB) compared to normal urea (NU) and control+binder (CB) groups, respectively. The model of leaching versus cumulative volume fitted the data trends robustly, ensuring better estimation of the release rate constants. This result agrees with the significant differences observed for water bath dissolution studies in our previous studies (Jange et al., 2023). The residual core + binder granules shown in figure 5 also demonstrated the delayed dissolution observed in figures 10b and 10d.
 |
| Figure 8. Cumulative total dissolved nitrogen (TDN) leached for 48 h incubated samples as a function of time (a) fitting logarithmic model, (b) goodness of fit for the logarithmic model, (c) fitting power-law model, and (d) goodness of fit for the power-law model. NU- normal urea samples, CC- control core, CbL- control bilayer, CB- core+binder, bLB- bilayer+binder. |
Conclusions
The soil leaching column experiments without incubation did not show differences between the fertilizer samples with regard to dissolved inorganic N or TDN losses. However, when soils were pre-incubated for 48 h, N leaching was significantly different between various N fertilizers, indicating
the initiation of mineralization through soil microbiological processes. Granular microstructure design using dry compaction and formulation design yielded a significant reduction in TDN after 48 h of incubation for the core+binder (CB) samples. An increase in internal density delayed nutrient leaching rates more significantly in gum-added core granules (CB) compared with gum-added bilayer (bLB) granules. However, a sole increase in internal density did not reduce nutrient leaching in disturbed soil flow-cell studies when comparing core granules (CC), bilayer granules (bL), and market urea (NU). Moreover, core granules with binder (CB) did not present a reduction in total ammonium and nitrate content as assessed using the power-law model. Nonetheless, it is noticeable that this model overpredicted the initial leaching data points. The model of TDN as a function of cumulative volume was also more reliable in discriminating leaching behavior. Additional investigation on the effect of soil type, pH levels, and water flow rate would help better understand nutrient leaching rates in samples with distinct solid fractions and binders. Optimization of granulation methods to produce bilayer granules might also reduce water percolation at the solid-liquid interface and, consequently, alter urea dissolution as well as its breakdown into ammonium and nitrate.
 |
| Figure 9. Cumulative total dissolved nitrogen (TDN) leached for 48 h incubated samples as a function of cumulative volume (a) fitting logarithmic model, (b) goodness of fit for the logarithmic model, (c) fitting power-law model, and (d) goodness of fit for the power-law model. NU- normal urea samples, CC- control core, CbL- control bilayer, CB- core+binder, bLB- bilayer+binder. The standard deviations of cumulative dissolved ammonium and nitrate and total dissolved nitrogen contents are represented as a continuous interval in figures 9a and 9c. |
 |
| Figure 10. Release rate constants derived from parameter estimation of a power-law relation for the profiles of 48 h incubated samples: (a) dissolved inorganic nitrogen versus leaching cycle time, (b) total dissolved nitrogen versus leaching cycle time, (c) dissolved inorganic nitrogen versus cumulative flow volume, and (d) total dissolved nitrogen versus cumulative flow volume. NU- normal urea samples, CC- control core, CbL- control bilayer, CB- core+binder, bLB- bilayer+binder. |
References
Ambrose, R. P., Wassgren, C., Pai, D., & Chen, Y. (2021). Layer-wiser agglomerated urea granules. U.S. Patent No. 11021410.
Cao, Y., & Mezzenga, R. (2020). Design principles of food gels. Nat. Food, 1(2), 106-118. https://doi.org/10.1038/s43016-019-0009-x
Fertahi, S., Ilsouk, M., Zeroual, Y., Oukarroum, A., & Barakat, A. (2021). Recent trends in organic coating based on biopolymers and biomass for controlled and slow release fertilizers. J. Control. Release, 330, 341-361. https://doi.org/10.1016/j.jconrel.2020.12.026
ISO. (2013). Water quality - Determination of selected parameters by discrete analysis systems - Part 1: Ammonium, nitrate, nitrite, chloride, orthophosphate, sulfate and silicate with photometric detection. ISO 15923-1:2013.
Jange, C. G., Wassgren, C. R., & Ambrose, R. P. (2021). Disintegration and release kinetics of dry compacted urea composites: A formulation and process design study. EFB Bioeconomy J., 1, 100020. https://doi.org/10.1016/j.bioeco.2021.100020
Jange, C. G., Wassgren, C. R., & Ambrose, R. P. (2023). Investigating the role of dry compaction and layer-wise agglomeration to control the dissolution of granular urea fertilizer. Powder Technol., 420, 118382. https://doi.org/10.1016/j.powtec.2023.118382
Jiao, Y., Hendershot, W. H., & Whalen, J. K. (2004). Agricultural practices influence dissolved nutrients leaching through intact soil cores. Soil Sci. Soc. Am. J., 68(6), 2058-2068. https://doi.org/10.2136/sssaj2004.2058
Markl, D., Yassin, S., Wilson, D. I., Goodwin, D. J., Anderson, A., & Zeitler, J. A. (2017). Mathematical modelling of liquid transport in swelling pharmaceutical immediate release tablets. Int. J. Pharm., 526(1), 1-10. https://doi.org/10.1016/j.ijpharm.2017.04.015
Mehrez, A., Ookawara, S., Ali, A. H., & Suzuki, M. (2014). A numerical study on cooling-solidification process of urea particles in prilling tower. J. Chem. Eng. Jpn., 47(8), 628-634. https://doi.org/10.1252/jcej.13we019
Naz, M. Y., & Sulaiman, S. A. (2016). Slow release coating remedy for nitrogen loss from conventional urea: A review. J. Control. Release, 225, 109-120. https://doi.org/10.1016/j.jconrel.2016.01.037
Nelson, D. W., & Sommers, L. E. (1996). Total carbon, organic carbon, and organic matter. In D. L. Sparks, A. L. Page, P. A. Helmke, R. H. Loeppert, P. N. Soltanpour, M. A. Tabatabai,... M. E. Sumner (Eds.), Methods of soil analysis: Part 3 Chemical methods, 5.3 (pp. 961-1010). Soil Science Society of America. https://doi.org/10.2136/sssabookser5.3.c34
Nelson, N. S. (1987). An acid-persulfate digestion procedure for determination of phosphorus in sediments. Commun. Soil Sci. Plant Anal., 18(4), 359-369. https://doi.org/10.1080/00103628709367826
Pappas, E. A., Kanwar, R. S., Baker, J. L., Lorimor, J. C., & Mickelson, S. (2008). Fecal indicator bacteria in subsurface drain water following swine manure application. Trans. ASABE, 51(5), 1567-1573. https://doi.org/10.13031/2013.25313
Penn, C. J., Williams, M. R., Camberato, J., Wenos, N., & Wason, H. (2022). Desorption kinetics of legacy soil phosphorus: Implications for non-point transport and plant uptake. Soil Syst., 6(1), 6. https://doi.org/10.3390/soilsystems6010006
Pereira, V. V., Morales, M. M., Pereira, D. H., de Rezende, F. A., de Souza Magalhães, C. A., de Lima, L. B.,... Petter, F. A. (2022). Activated biochar-based organomineral fertilizer delays nitrogen release and reduces N2O emission. Sustainability, 14(19), 12388. https://doi.org/10.3390/su141912388
Randall, G. W. (1990). Nitrate-N in the soil profile and tile drainage water as influenced by tillage. Am. J. Ind. Med., 18(4), 457-460. https://doi.org/10.1002/ajim.4700180417
Shah, S. B., & Wolfe, M. L. (2002). Nitrate leaching impact of urea pellets versus urea granules. Appl. Eng. Agric., 18(1), 57. https://doi.org/10.13031/2013.7711
Smith, W. N., Prasher, S. O., Khan, S. U., & Barthakur, N. N. (1992). Leaching of 14C-labelled atrazine in long, intact soil columns. Trans. ASAE, 35(4), 1213-1220. https://doi.org/10.13031/2013.28722
Trenkel, M. E. (2010). Slow- and controlled-release and stabilized fertilizers: an option for enhancing nutrient use efficiency in Agriculture. Paris, France: International Fertilizer Industry Association.
Wang, C., Luo, D., Zhang, X., Huang, R., Cao, Y., Liu, G.,... Wang, H. (2022). Biochar-based slow-release of fertilizers for sustainable agriculture: A mini review. Environ. Sci. Ecotechnol., 10, 100167. https://doi.org/10.1016/j.ese.2022.100167
Yang, L., An, D., Wang, T.-J., Kan, C., & Jin, Y. (2017). Swelling and diffusion model of a hydrophilic film coating on controlled-release urea particles. Particuology, 30, 73-82. https://doi.org/10.1016/j.partic.2016.03.005
Ye, H.-M., Li, H.-F., Wang, C.-S., Yang, J., Huang, G., Meng, X., & Zhou, Q. (2020). Degradable polyester/urea inclusion complex applied as a facile and environment-friendly strategy for slow-release fertilizer: Performance and mechanism. Chem. Eng. J., 381, 122704. https://doi.org/10.1016/j.cej.2019.122704