Skip to content
Frontiers
Search

Original Research

Frontiers in Chemical Engineering ·

Volume 8 - 2026 | https://doi.org/10.3389/fceng.2026.1892312

Mechanistic evaluation of water removal dynamics in stabilised human urine for circular fertiliser production

  • Charles B. Niwagaba1*corresponding author
  • Anastasija Vasiljev2
  • Mathias Kasirye1
  • Andrew Lubuulwa1
  • Alex Y. Katukiza1
  • Swaib Semiyaga1
  • Musa Manga3,4
  • Prithvi Simha2,5*corresponding author
  1. 1Department of Civil and Environmental Engineering, Makerere University, Kampala, Uganda
  2. 2Department of Energy and Technology, Swedish University of Agricultural Sciences, Uppsala, Sweden
  3. 3The Water Institute at UNC, Department of Environmental Sciences and Engineering, Gillings School of Global Public Health, University of North Carolina, Chapel Hill, NC, United States
  4. 4Department of Construction Economics and Management, Makerere University, Kampala, Uganda
  5. 5Gobabeb Namib Research Institute, Walvis Bay, Namibia
Article metrics0 viewsView details
PDFCiteMetrics

Abstract

Drying source-separated, stabilised human urine offers a route to produce circular fertilisers with low environmental impact, with potential to reduce dependence on synthetic fertilisers and close nutrient loops. This work examines the behaviour of water during the drying stage of the urine-to-solid fertiliser value chain. Specifically, it evaluated how stabilisation chemistry influences water activity (a w) and moisture transport dynamics. Freshly excreted urine was treated with citric acid, calcium hydroxide, or wood ash, and subjected to controlled isothermal drying at 50 °C, with continuous mass loss data fitted to ten thin-layer drying kinetics models. The Two-term model best described wood ash–stabilised urine (k 1 = 8.7 × 10 −5 s -1, k 2 = 1.2 × 10 −4 s -1), suggesting water removal was initially rapid, followed by a slower drying phase. In contrast, water removal in calcium hydroxide- and citric-acid-stabilised urine slowed progressively as drying advanced and was best described by the Page model, with k values of 7.3 × 10 −6 and 3.3 × 10 −8 s -1, respectively. Peleg isotherm fits revealed a non-linear relationship between a w and moisture content, with water activity declining sharply only after substantial water removal. Fresh and acid-stabilised urine showed Type II-like desorption behaviour, whereas alkalised urine showed Type III-like behaviour, with a w remaining >0.7 even after majority of the water had been removed. Considering water activity thresholds for microbial inhibition reported in the literature, and recent evidence that urine concentration supresses urease activity, we propose a five-fold mass concentration, corresponding to a w <0.6, as a design target for limiting microbial growth and ureolysis in urine concentrating systems.

Keywords

1 Introduction

Urine drying has emerged as a promising technology in the developing field of source-separated sanitation (Aliahmad et al., 2022), in which freshly excreted human urine is converted into a stable, transportable fertiliser through a defined sequence of treatment steps (Simha et al., 2020a). The process begins at the point of generation, where urine is collected undiluted from waterless urinals or urine-diverting toilets. The collected urine is immediately stabilised by dosing with acids, bases, or oxidisers to inhibit urease-catalysed urea hydrolysis and reduce the risk of ammonia volatilisation (Randall et al., 2016; Simha et al., 2023; Arve et al., 2024; Zhang et al., 2013). This preserves dissolved nitrogen, phosphorus, potassium, and micronutrients, creating a liquid matrix that can be further processed by controlled water removal, typically by evaporation or distillation, to achieve a mass reduction exceeding 95% (Gao et al., 2025). The resulting concentrate can be applied as liquid fertiliser or processed further by blending with compatible co-substrates (Perez-Mercado et al., 2024), and granulating or pelletising to produce a dry fertiliser that is physically and chemically stable, easy to store, and compatible with conventional agricultural application equipment. Recent life cycle assessments indicate that such urine-derived fertilisers can significantly reduce greenhouse gas emissions and eutrophication compared to mineral fertilisers, especially when integrated with black- and greywater recycling, although these environmental benefits depend on how source-separating sanitation systems are configured (Aliahmad et al., 2025).

Much of the research and development on urine drying has focused on optimising urine stabilisation, improving nutrient recovery, and understanding the interactions between urine composition and process chemistry. These studies have shown that acid and alkaline dehydration can produce nutrient-rich fertiliser products with high retention of nitrogen, phosphorus, potassium and other major ions (Vasiljev et al., 2022; Gao et al., 2025). Related work has identified strategies to prevent mineral scaling and pipe blockages (Simha et al., 2025), use precipitation deliberately to capture volatile compounds (Yu et al., 2021), and selectively remove or degrade endogenous and exogenous compounds in urine (Courtney et al., 2025; Mehaidli et al., 2024; Almuntashiri et al., 2021). However, the largest constituent of urine, water, has received little mechanistic evaluation. Water removal is the largest mass flux in the process chain, since urine contains about 95% water, and it can dominate the energy requirement for treating urine. Previous estimates for active urine evaporation indicate energy demands of approximately 650 Wh L−1, decreasing to 150 Wh L−1 when heat recovery is incorporated (Aliahmad et al., 2025). However, recent field testing of a passive solar-thermal evaporator showed that urine could be concentrated in arid conditions up to 30-fold without external energy input (Simha et al., 2026). Beyond energy use, water removal also determines processing time, physical form and stability of intermediate and final urine-derived fertiliser products. These in turn affect handling, storage, and the ease with which concentrated urine can be applied in agriculture.

Drying of biological materials is governed by a sequence of physical regimes, typically beginning with a constant rate period in which water evaporates freely from a saturated surface, followed by one or more falling rate periods during which internal moisture transport becomes the rate-limiting step (Srikiatden and Roberts, 2007). The transitions between these regimes depend on the physical state of water within the material, whether as thermodynamically free water, weakly bound water through hydrogen bonding, or strongly bound water associated with salts, organics and other hygroscopic compounds (Mathlouthi, 2001). Water activity measurements provide a powerful means to characterise these states, linking moisture content to the thermodynamic availability of water for evaporation (Barbosa-Cánovas et al., 2020). Chemical stabilisation likely alters urine drying dynamics in fundamental ways. As water is removed from urine, the concentration of dissolved salts and organic compounds rises, increasing the likelihood of salt crystallisation, colloidal aggregation, and phase separation of hydrophobic compounds (Gao et al., 2025). Evaluating this behaviour is essential for designing efficient dryers for treating urine.

This study therefore focused specifically on the drying stage of the urine-to-fertiliser process chain, aiming to develop a mechanistic understanding of how urine stabilisation influences water removal. We addressed three research questions: (i) how do three stabilisation strategies commonly used in urine recycling research and practice (citric acid, calcium hydroxide, and wood ash) alter isothermal drying behaviour and drying rate transitions; (ii) which thin-layer drying models best describe water removal from stabilised urine under controlled conditions; and (iii) how does stabilisation chemistry influence the relationship between water activity and moisture content during concentration? By linking these parameters to observed drying behaviour, this work provides a basis for designing decentralised urine dryers.

2 Methodology

2.1 Urine collection

Fresh human urine was collected from four male urinals at the College of Engineering, Design, Art and Technology, Makerere University. Before collection, the urinals were disconnected from the building’s water and sewage lines to enable undiluted urine collection. Flexible tubing redirected the urine into 20 L jerry cans preloaded with stabilising agents to immediately inhibit enzymatic urea hydrolysis (Randall et al., 2016). Urine from each collection period was therefore pooled as stabilised urine from multiple anonymous donors. Jerry cans filled within 2–3 days, after which they were replaced, transported to the laboratory and stored for 1–2 days before analysis. During storage, the stabilised urine was mixed several times daily to maintain homogeneity. In total, 720 L of urine was collected over the 10-week study period. For alkaline stabilisation, jerry cans were dosed with either 100 g L-1 wood ash or 15 g L-1 calcium hydroxide, targeting pH > 11. For acid stabilisation, citric acid was added at 5 g L-1 to target pH < 4. These dosing protocols were adapted from previous studies (Simha, 2021; Simha et al., 2023) and selected to achieve urine pH ranges known to inhibit urease-catalysed urea hydrolysis. Informed consent signage was displayed above the urinals, and donors retained the option to use nearby conventional water-flushed toilets instead. No personal data, images or recordings were collected.

2.2 Thermogravimetric analysis

The drying kinetics of acid- and base-stabilised urine were evaluated under isothermal conditions using a precision thermogravimetric analyser (ELTRA Thermostep TGA, Germany), featuring a temperature resolution of ±0.01 °C and mass accuracy of ±0.0001 mg. Stabilised urine samples (1.0 ± 0.1 g) were dried at 50 °C with a controlled air flow of 1.0 m s-1 to allow isothermal comparison among stabilisation treatments under conditions previously used for urine evaporation (Simha et al., 2020b). Urine stabilisation was performed using citric acid (acidified), calcium hydroxide, or wood ash (alkalised) as described in Section 2.1. Thermogravimetric drying experiments were performed in triplicate using independently weighed subsamples from the same stabilised urine batch.

The mass loss over time was continuously recorded to determine the moisture content on a dry basis (M *db*, mg H2O mg-1 solids; Equation 1) and the drying rate (DR, kg d-1 m-2); Equation 2, following the methodology in Mawejje (2021). These were calculated as: where, m i and m i+1 are the sample masses (g) at consecutive time points t *i* and t *i+1* respectively; m<sub>d</sub> is the dry solids mass remaining after drying; and A is surface area available for evaporation.

M d b = m i − m d m d (1)

D R = m i − m i + 1 t i + 1 − t i × A (2)

2.3 Mathematical modelling of drying kinetics

To interpret the drying behaviour of stabilised urine, moisture loss data were fitted to ten commonly used thin-layer drying models, previously applied to materials of similar physical character such as faecal sludge and sewage sludge (Mawejje, 2021). Model selection aimed to identify the most appropriate empirical representation of moisture transport during isothermal drying at 50 °C.

The Moisture Ratio (MR) at any time t was calculated using Equation 3:where, M *t* is the moisture content at any time t, M *0* is the initial moisture content and M *e* is the equilibrium moisture content. Given the low ambient relative humidity, small sample mass (1 g) spread in a thin layer, and drying to near constant final mass in the thermogravimetric system, M *e* was assumed to be negligible relative to the initial moisture content. This assumption is commonly applied in thin-layer drying analyses and simplifies MR to Mt/M0, although it may affect late-stage MR values near the end of drying.

M R = M t − M e M 0 − M e (3)

Non-linear regression was performed in Microsoft Excel SOLVER using the Generalized Reduced Gradient nonlinear algorithm, with residual sum of squares between experimental and predicted MR values as the objective function. Convergence was accepted when Solver reached a stable minimum RSS value with no further reduction in the objective function. Parameter uncertainty, including confidence intervals, was not evaluated; fitted parameters were therefore interpreted as descriptive coefficients for comparing model performance among treatments.

Model performance was evaluated using three statistical indices: root mean square error (RMSE) (Equation 4), coefficient of determination (R 2) (Equation 5) and reduced chi-square (χ2) (Equation 6), calculated as: where, M R exp ⁡ , i and M R p r e , i are the experimental and model-predicted moisture ratios at time step i, M R a v , ⁡ exp is the mean moisture ratio over the experimental period, N is the number of observations, and n is the number of model parameters. The model with the highest R *2* and the lowest RMSE and χ2 was deemed most suitable for describing the drying behaviour of each stabilised urine type (Castro et al., 2018).

R M S E = ∑ i = 1 N M R exp ⁡ , i − M R p r e , i 2 N (4)

χ 2 = ∑ i = 1 N M R exp ⁡ , i − M R p r e , i 2 N − n (5)

R 2 = 1 − ∑ i = 1 N M R exp ⁡ , i − M R p r e , i 2 ∑ i = 1 N M R exp ⁡ , i − M R a v , ⁡ exp 2 (6)

2.4 Effective moisture diffusivity

Effective moisture diffusivity (D *eff*) is a fundamental transport parameter that describes the internal movement of water during drying, especially in the falling rate period where diffusion dominates over surface evaporation (Makununika, 2016). Given the planar geometry of the samples and their uniform spread during thermogravimetric analysis, a flat slab diffusion model was applied to estimate effective diffusivity, as it best approximates the physical conditions of the experiment. Samples of wood ash- and calcium hydroxide-stabilised urine were dried in shallow circular containers, forming uniform layers. This enabled the assumption of one-dimensional moisture transport, perpendicular to the slab surface, simplifying the diffusion analysis compared with spherical or cylindrical geometries (Mawejje, 2021).

Following (Koukouch et al., 2017), the estimation of D *eff* was based on the analytical solution to Fick’s second law for slab geometry (Equation 7):where t is the drying time (s), and L is the half-thickness of the slab (m), and MR is the moisture ratio defined in Equation 3. For longer drying times (MR < 0.6), the series can be approximated by the first term only (Equation 8):

MR = 8 π 2 ∑ n = 0 ∞ 1 2 n + 1 2 exp − 2 n + 1 2 π 2 D e f f t 4 L 2 (7)

MR = 8 π 2 exp − π 2 D e f f t 4 L 2 (8)

Taking natural logarithms of both sides yields a linear form (Equation 9):

ln ⁡ MR = ln 8 π 2 − π 2 D e f f t 4 L 2 (9)

From the slope of ln (MR) versus time, D *eff* can be computed. The following assumptions were adopted to simplify the analysis: (a) moisture transport during the selected falling rate period can be approximated as diffusion controlled; (b) effective diffusivity and drying temperature remain constant over this interval; (c) the stabilised urine layer is homogeneous with uniform initial moisture content; (d) the sample can be approximated as a one-dimensional slab over the fitted region. Given these assumptions, D *eff* should be interpreted as apparent effective diffusivity since salt crystallisation, crust formation, and in alkaline treatments, particulate mineral or ash phases may alter moisture transport and influence D *eff* estimates, especially at low moisture contents.

2.5 Water activity measurements

To complement the drying kinetics results, additional experiments were conducted to determine equilibrium relationships between water activity (a w) and moisture content in freshly excreted (unstabilised) and stabilised urine. Aliquots of each urine type were dried in a temperature-controlled oven at 25 °C to systematically generate samples spanning a wide range of moisture contents. Drying continued until each target moisture level was reached, after which moisture content was determined gravimetrically on a wet basis. For a w determination, samples were homogenised, and 1 g was evenly spread across the bottom of a disposable sample cup. The cup was immediately loaded into a calibrated water activity meter (AQUALAB TDL, METER Group, United States; accuracy ±0.005 a w at 25 °C), and a w was recorded once the difference between three consecutive readings was <0.0005. This procedure was applied to five urine types: fresh (unstabilised) urine and urine stabilised with citric acid, sulfuric acid, potassium hydroxide, or fumaric acid.

2.6 Modelling of moisture desorption behaviour

The experimental moisture content versus a w data for each urine treatment were fitted to the four-parameter Peleg desorption isotherm (Arslan and Toǧ;rul, 2006):where M is the equilibrium moisture content on a wet basis, k 1 and n 1 represents moisture loss at low to intermediate a w, where water is predominantly bound, whereas k 2 and n 2 represents the moisture loss in the high a w region, where water is largely unbound and more readily removed (Equation 10). The Peleg model was chosen for its empirical flexibility in reproducing both sigmoid and non-sigmoid desorption isotherm over the full a w spectrum (Turhan et al., 2002). Model parameters were estimated by non-linear least squares regression, and model performance was evaluated using the coefficient of determination (R 2) and root mean square error (RMSE).

M a w = k 1 a w n 1 + k 2 a w n 2 (10)

3 Results and discussion

3.1 Isothermal drying behaviour of stabilised urine at 50 °C

Under isothermal conditions at 50 °C, urine stabilised with citric acid, calcium hydroxide, or wood ash followed a triphasic drying trajectory: an initial rising rate period, a constant rate period and a falling rate period (Figure 1). Such patterns are characteristic of thin-layer drying of many biosolids, reflecting the sequential transition from heat-up and surface wetting to surface-controlled evaporation and finally, to diffusion-limited drying (Getahun et al., 2020; Somorin et al., 2021). The rising rate phase (0–0.5 h) was most apparent for the alkaline treatments and may reflect delayed surface saturation influenced by suspended and precipitated solids in Ca(OH)2- and ash-stabilised urine. The constant rate phase varied with stabilisation chemistry. Citric acid-treated urine showed a well-defined and extended plateau in drying rate (2 kg d-1 m-2 for 5 h), indicating prolonged surface-controlled evaporation from a relatively homogeneous liquid matrix. By contrast, the alkaline treatments had much shorter constant rate periods (<1 h). (Somorin et al., 2021) also observed that faecal sludges from ventilated improved pit latrines exhibit only brief constant rate periods, while fresh faeces and material from urine-diverting dry toilets lack any discernible constant-rate stage, shifting almost immediately to diffusion-controlled drying regime.

Drying characteristics of urine stabilised with citric acid (blue), wood ash (red), and calcium hydroxide (green) under isothermal conditions at 50 °C. Panels show **(a)** drying rate over time. **(b)** Drying rate as a function of moisture ratio (MR), showing shifts in the transition points between drying phases. **(c)** MR over time, and **(d)** moisture content over time. Curves show mean values from three analytical replicates per treatment (n = 3). The raw data underlying Figure 1 are provided in Supplementary Dataset S1.
Figure 1. Drying characteristics of urine stabilised with citric acid (blue), wood ash (red), and calcium hydroxide (green) under isothermal conditions at 50 °C. Panels show (a) drying rate over time. (b) Drying rate as a function of moisture ratio (MR), showing shifts in the transition points between drying phases. (c) MR over time, and (d) moisture content over time. Curves show mean values from three analytical replicates per treatment (n = 3). The raw data underlying Figure 1 are provided in Supplementary Dataset S1.

For the calcium hydroxide treated urine, the drying curve had at least two distinct quasi-constant drying stages, interspersed between falling rate stages. This suggests that, in this matrix, there may be a temporary balance between evaporation and internal water diffusion that allows for brief sustained evaporation. This behaviour is unlikely to arise from salt precipitation during urine concentration alone, as crystallisation can accelerate the onset of falling rate drying by blocking pores (Rad and Shokri, 2012). The slower decline in drying rate may have been influenced by precipitation reactions and changes in the particulate mineral phase during drying. One possible contributing process is carbonation of Ca(OH)2 to form CaCO3 (Vasiljev et al., 2022), which can modify pore structure and water transport pathways in calcium-rich matrices (Galan et al., 2015).

Wood ash-stabilised urine dehydrated fastest, reaching negligible moisture in about 4 h, followed by calcium hydroxide-stabilised urine (≈5.5 h), while citric acid-treated urine required more than 6 h for complete dehydration (Figure 1). Acidification can increase water retention via protonation of solutes, delaying water release (Gao et al., 2025), whereas alkaline treatments may promote moisture removal through precipitation reactions, changes in ionic composition, and possible ammonia volatilisation, which together reduce water binding and the solids content of urine (Gao et al., 2025; Simha et al., 2022). Moreover, the partial insolubility of Ca(OH)2 and ash may introduce suspended and precipitated solids that help maintain surface wetting during drying, supporting faster evaporation than in completely soluble citric acid-stabilised urine (Simha et al., 2020c). From a process design perspective, these mechanistic differences imply trade-offs. Alkaline stabilisers offer higher evaporation rates but require higher dosages and active mixing to overcome their poor solubility (Gao et al., 2025). The maximum achievable mass concentration factor was about 2.6 for wood ash and 4.5 for Ca(OH)2, reflecting the stabiliser mass added to urine and the earlier cessation of evaporation once unbound water had been removed. In contrast, citric acid–treated urine reached a mass concentration factor of 50, consistent with previous work showing that no salt precipitation occurs until >99% of the water is removed (Gao et al., 2025). This likely helps maintain surface-controlled evaporation for longer, delaying the onset of diffusion-limited drying.

3.2 Drying kinetics modelling

To interpret the drying behaviour of stabilised urine, ten thin-layer drying models were evaluated against experimental data (Table 1). Because these models are empirical descriptions of drying-curve behaviour, they can guide interpretation of dominant drying regimes but do not resolve moisture transport mechanisms. Drying kinetics differed between the three stabilisation approaches, and this was reflected in the best-fit models (Figure 2). For wood ash–stabilised urine, the Two-term exponential model achieved the highest agreement with the data (R2 = 0.99, χ2 ≈ 0, RMSE = 0.015). The fitted drying constants (k1 = 8.76 × 10−5 s-1 and k2 = 1.2 × 10−4 s-1) point to two concurrent mechanisms: a rapid phase dominated by surface evaporation of free water and a slower phase linked to diffusion of more tightly bound water within the porous ash–urine matrix. Such drying behaviour has been reported in other biosolids and porous matrices where both external evaporation and internal desorption occur simultaneously (Ertekin and Yaldiz, 2004). By contrast, calcium hydroxide- and citric acid-stabilised urine were both best described by the Page model, with excellent fits (R2 > 0.94), in line with previous reports on drying of faecal sludge (Somorin et al., 2021). The estimated drying rate constants were 7.3 × 10−6 s-1 for calcium hydroxide and 3.3 × 10−8 s-1 for citric acid. The Page empirical exponent was >1 in both cases, indicating that water release from these matrices progressively decreased as drying progressed. Calcium hydroxide, though only sparingly soluble at the applied dosage (Randall et al., 2016), was associated with higher apparent moisture diffusivity than citric acid. This likely reflects the role of undissolved Ca(OH)2 particles, which remain dispersed in the matrix and generate a porous structure during drying, even if it is less heterogeneous than that formed by wood ash.

TABLE 1. Thin-layer drying model parameters and statistical performance for wood ash-, calcium hydroxide-, and citric acid-stabilised urine under isothermal drying at 50 °C. The table lists model constants for each drying model tested, along with statistical goodness-of-fit indicators: residual sum of squares (RSS), coefficient of determination (R2), reduced chi-square (χ2), and root mean square error (RMSE).

Drying modelModel constantsStatistical parametersReferences
​Wood ashCa(OH)2Citric acid​Wood ashCa(OH)2Citric acid
Newtonk6.2 × 10−57.4 × 10−57.3 × 10−5RSS0.1820.1230.692Akpinar & Bicer, (2005)
MR = exp (-kt)​​​​R20.9320.9720.933
​​​​χ20.0430.0730.043
​​​​RMSE0.0530.0440.054
Pagek3.3 × 10−47.3 × 10−63.3 × 10−8RSS0.1280.0430.051Ortiz-Jerez et al. (2022)
MR = exp (-ktn)n0.821.251.81R20.9520.9910.992
​​​​χ20.0020.0010.001
​​​​RMSE0.0450.0260.028
Hendersonk5.9 × 10−58.3 × 10−58.7 × 10−5RSS0.3480.0570.453Inyang et al.(2018), Onwude et al.(2016), Erbay & Icier, (2010)
MR = a exp (-kt)a0.971.091.16R20.9350.9870.929
​​​​χ20.0060.0010.007
​​​​RMSE0.0740.0300.085
Logarithmick1.4 × 10−48.3 × 10−58.7 × 10−5RSS0.0380.0570.453Ortiz-Jerez et al. (2022), Minaei et al. (2012)
MR = a exp (-kt) + ca0.761.081.16R20.9860.9870.929
c0.3100χ20.0010.0010.008
​​​​RMSE0.0250.0300.085
Wang and singha0.0055 × 10−118.7 × 10−7RSS7.2 × 101319.502.333Mirzaee et al. (2010), Fang et al. (2015), Kumar et al. (2012)
MR = 1 + at + bt2b0.0059 × 10-14−2 × 10-9R2−2.7 × 1013−3.2880.633
​​​​χ21.2 × 10120.320.038
​​​​RMSE0.1820.1230.692
Diffusionk6.2 × 10−57.4 × 10−57.3 × 10−5RSS0.9330.9730.891Alara et al. (2018)
MR = a exp (-kt) + (1-a) exp (-kbt)a111R20.0430.0730.103
b111χ20.0540.0440.105
​​​​RMSE0.1280.0430.051
Vermak5.9 × 10−58.4 × 10−58.8 × 10−5RSS0.1740.0490.423Karaaslan et al. (2016)
MR = a exp (-kt) + (1-a) exp (-gt)a0.971.101.18R20.9350.9890.934
g111χ20.0030.0010.007
​​​​RMSE0.0530.0280.082
Two term-exponentialk6.2 × 10−57.4 × 10−57.3 × 10−5RSS0.1820.1230.692Metwally et al. (2024), Eltawil et al. (2018)
MR = a exp (-kt) + (1-a) exp (-kat)a0.990.991R20.9330.9730.891
​​​​χ20.0030.0020.003
​​​​RMSE0.0540.0440.054
Midillik6.2 × 10−56.2 × 10−56.2 × 10−5RSS0.1700.3310.869Sansaniwal et al. (2018), Ertekin & Yaldiz, (2004)
MR = a exp (-ktn) + bta111R20.9370.9270.863
b1.1 × 10−600χ20.0030.0060.015
n111RMSE0.0070.0730.117
Two-termk18.8 × 10−58.3 × 10−58.7 × 10−5RSS0.0140.0570.453Karaaslan et al. (2016), Özbek & Dadali, (2007), Inyang et al. (2017)
MR = a exp (-k1t) + b exp (-k2t)k21.2 × 10−45.7 × 10−60R20.9950.9870.929
a1.031.091.16χ20.0010.0010.007
b0.01800RMSE0.0150.0300.085

Values are empirical least-squares fitting coefficients used for descriptive comparison of thin-layer drying models. Negative R2 values indicate poor model adequacy relative to the experimental mean. Model suitability was assessed jointly using RMSE, R2 and reduced χ2.

Thin-layer drying model fits for stabilised urine under isothermal conditions at 50°C: Panels show **(a)** wood ash-treated urine fitted with the Two-term exponential model, **(b)** calcium-hydroxide-treated urine fitted with the Page model, and **(c)** citric acid-treated urine fitted with the Page model. Dotted markers represent experimental moisture ratio data and solid lines represent model-predicted values.
Figure 2. Thin-layer drying model fits for stabilised urine under isothermal conditions at 50°C: Panels show (a) wood ash-treated urine fitted with the Two-term exponential model, (b) calcium-hydroxide-treated urine fitted with the Page model, and (c) citric acid-treated urine fitted with the Page model. Dotted markers represent experimental moisture ratio data and solid lines represent model-predicted values.

The effective moisture diffusivities (D *eff*), calculated from the linear portion of the falling rate period using Fick’s second law for slab geometry, were 1.1 × 10-14 m2 s-1 for wood ash, 1.07 × 10-14 m2 s-1 for calcium hydroxide, and 4.5 × 10−15 m2 s-1 for citric acid, respectively. These values are much lower than those reported for drying of many food and biological materials (Koukouch et al., 2017). While ash and calcium hydroxide have similar D *eff* values, consistent with their modelled rate constants, the corresponding drying curves reveal distinct behaviours. Ash-treated urine dried more rapidly through a single falling rate stage, whereas calcium-hydroxide-treated urine exhibited a tiered falling rate profile. In citric acid-treated urine, D *eff* was 3-fold lower, yet the falling rate stage was short, accounting for less than 20% of the total drying time. These differences, when interpreted alongside the best-fit drying models, indicate that similar bulk diffusivity values do not necessarily correspond to comparable drying regimes in concentrated urine residues.

3.3 Moisture desorption isotherm and water binding regimes

Fresh and stabilised urine, regardless of stabilising agent, initially exhibited a w values above 0.99, consistent with urine being predominantly water. As desorption began from this fully hydrated state, most treatments retained high a w until the moisture content fell below 30%. This indicates that a large fraction of the initial water was unbound or only weakly associated with solutes, and could therefore be removed readily during early drying stages (Figure 3). By analogy with the IUPAC classification of physisorption isotherms, Type II-like curves are sigmoidal, whereas Type III-like curves are convex across their range and lack a clear inflection point or plateau (Bernet et al., 2026). Fresh urine and acid-stabilised urines (citric, fumaric, and sulfuric acid) showed Type II-like desorption behaviour, typical of biological matrices in which water transitions progressively from unbound to multilayer-bound and, finally, to tightly bound states as moisture content decreases. In these treatments, the steep moisture loss at a w >0.7 reflects the evaporation of unbound/free water; the intermediate region (a w = 0.3–0.7) corresponds to removal of water loosely bound within pores or to surfaces; and the gradual decline at a w <0.3 represents water that resists desorption.

Water activity–moisture content relationships for **(a)** fresh urine, **(b)** potassium hydroxide-stabilised urine, **(c)** fumaric acid-stabilised urine, **(d)** citric acid-stabilised urine, and **(e)** sulphuric acid-stabilised urine. Data points represent experimental measurements, and solid lines show Peleg model fits. Potassium hydroxide-stabilised urine showed Type III-like behaviour, whereas fresh and acid-stabilised urines showed Type II-like behaviour. Peleg model parameters and goodness-of-fit statistics are provided in Supplementary Table S2.
Figure 3. Water activity–moisture content relationships for (a) fresh urine, (b) potassium hydroxide-stabilised urine, (c) fumaric acid-stabilised urine, (d) citric acid-stabilised urine, and (e) sulphuric acid-stabilised urine. Data points represent experimental measurements, and solid lines show Peleg model fits. Potassium hydroxide-stabilised urine showed Type III-like behaviour, whereas fresh and acid-stabilised urines showed Type II-like behaviour. Peleg model parameters and goodness-of-fit statistics are provided in Supplementary Table S2.

In contrast, potassium hydroxide-stabilised urine showed Type III-like desorption behaviour. In this treatment, >85% of the initial moisture was lost while a w remained >0.7, indicating that the remaining water was still relatively mobile despite its low absolute content. This behaviour is common in matrices containing deliquescent salts, which hold little water at lower humidities and therefore lose most of their moisture only at high a w during dehydration. It also implies that dehydrated KOH-stabilised urine, if exposed to humid air, could rapidly reabsorb water, leading to stickiness and clumping, and would need to be stored in airtight, moisture-proof packaging to maintain product stability.

From a microbiological perspective, reducing the a w of urine to <0.9 inhibits the growth of most pathogenic bacteria (Rockland and Beuchat, 1987), including ureolytic strains, as most bacteria require high a w to support enzymatic metabolism; for example, Staphylococcus aureus is inhibited below a w = 0.86 (Ballesteros et al., 1993). Further reducing a w below 0.8 suppresses most yeasts and moulds (Peleg, 2022), while broad inhibition of microbial growth generally requires a w <0.6 (Peleg, 2022). Recent work has shown that evaporative concentration of urine suppresses urease activity, with enzymatic urea hydrolysis decreasing substantially at higher concentration factors as water activity declined and ionic strength increased (Melhus et al., 2026). According to our experimental data, drying urine to a w <0.6 equated to 20% residual moisture content on wet basis, or about a five-fold mass concentration. Therefore, considering water activity thresholds for microbial inhibition reported in the literature, and recent evidence that concentrating urine suppresses urease activity, we propose that treatments involving water removal from urine target a w <0.6 to limit microbial activity and ureolysis, assuming that low a w can be maintained during storage.

4 Conclusion

This work demonstrates that the choice of chemical stabiliser strongly influences the drying dynamics and water-binding behaviour of human urine. Under controlled isothermal conditions, wood ash-, calcium hydroxide- and citric acid-stabilised urine showed distinct drying rate profiles and moisture-transport behaviour. The kinetic profiles captured by the Two-term and Page models indicate that dominant drying behaviour changes with urine chemistry and matrix properties. Modelling of water activity further showed that urine can display Type II-like or Type III-like desorption behaviour, which has direct implications for stability and safe storage of urine-derived fertiliser products. For practical application, we propose targeting a w <0.6, equivalent to approximately 20% residual moisture content on wet basis or a five-fold mass concentration, as a design threshold for limiting microbial and ureolytic activity, provided this low aw state is maintained during storage. Overall, these findings provide a mechanistic basis for selecting urine stabilisation strategies and matching them with dryer design and fertiliser storage requirements. Future work should examine how drying kinetics influences nutrient recovery, particularly nitrogen retention, under varying stabiliser chemistries, and whether faster drying in alkaline matrices can reduce the risk of urea hydrolysis. The storage and fertiliser application stages of the value chain also warrant further targeted investigation.

Generative AI statement

The author(s) declared that generative AI was used in the creation of this manuscript. During preparation of this manuscript, generative AI was used only for language polishing and editing. The authors reviewed and approved all AI-assisted edits, verified the scientific content, and take full responsibility for the final manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fceng.2026.1892312/full#supplementary-material

Supplementary material Supplementary material at the source

Footnotes

Edited by: Chirangano Mangwandi, Queen's University Belfast, United Kingdom

Reviewed by: Surya Teja Varanasi, Tamil Nadu Agricultural University, India

Sanjarbek Shamuratov, Urgench State University, Uzbekistan

Statements

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

The author PS declared that they were an editorial board member of Frontiers at the time of submission. This had no impact on the peer review process and the final decision.

Funding

The author(s) declared that financial support was received for this work and/or its publication. This work was supported by generous funding from Vinnova (Grant No. 2020-04176) and Jane and Dan Olssons Stiftelse för Vetenskapliga ändamål.

Data availability

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding authors.

Author contributions

CN: Conceptualization, Formal Analysis, Funding acquisition, Methodology, Project administration, Resources, Supervision, Writing – original draft, Writing – review and editing. AV: Conceptualization, Data curation, Formal Analysis, Investigation, Methodology, Writing – original draft, Writing – review and editing. MK: Conceptualization, Data curation, Formal Analysis, Investigation, Methodology, Software, Visualization, Writing – original draft. AL: Conceptualization, Data curation, Formal Analysis, Investigation, Methodology, Software, Visualization, Writing – original draft. AK: Conceptualization, Methodology, Resources, Supervision, Writing – review and editing. SS: Conceptualization, Methodology, Resources, Supervision, Writing – review and editing. MM: Conceptualization, Methodology, Supervision, Writing – review and editing. PS: Conceptualization, Formal Analysis, Funding acquisition, Methodology, Project administration, Resources, Visualization, Writing – original draft, Writing – review and editing.

References

  1. Akpinar E. K., Bicer Y. (2005). Modelling of the drying of eggplants in thin-layers. J. Sci. Food Agric. 85 (2), 273 – 281. doi:10.1111/j.1365-2621.2004.00886.x
  2. Alara O. R., Abdurahman N. H., Abdul Mudalip S. K., Olalere O. A. (2018). Mathematical modeling of thin layer drying using open sun and shade of Vernonia amygdalina leaves. Agric. Nat. Resour. 52 (1), 53 – 58. doi:10.1016/j.anres.2018.05.013
  3. Aliahmad A., Harder R., Simha P., Vinnerås B., McConville J. (2022). Knowledge evolution within human urine recycling technological innovation system (TIS): Focus on technologies for recovering plant-essential nutrients. J. Clean. Prod. 379, 134786. doi:10.1016/j.jclepro.2022.134786
  4. Aliahmad A., Lima P. de M., Kjerstadius H., Simha P., Vinnerås B., McConville J. (2025). Consequential life cycle assessment of urban source-separating sanitation systems complementing centralized wastewater treatment in lund, Sweden. Water Res. 268, 122741. doi:10.1016/j.watres.2024.122741
  5. Almuntashiri A., Hosseinzadeh A., Volpin F., Ali S. M., Dorji U., Shon H. (2021). Removal of pharmaceuticals from nitrified urine. Chemosphere 280, 130870. doi:10.1016/j.chemosphere.2021.130870
  6. Arslan N., Toǧrul H. (2006). The fitting of various models to water sorption isotherm of tea stored in a chamber under controlled temperature and humidity. J. Stored Prod. Res. 42 (2), 112 – 135. doi:10.1016/j.jspr.2005.01.001
  7. Arve P. H., Mason M., Randall D. G., Simha P., Popat S. C. (2024). Concomitant urea stabilization and phosphorus recovery from source-separated fresh urine in magnesium anode-based peroxide-producing electrochemical cells. Water Res. 256, 121638. doi:10.1016/j.watres.2024.121638
  8. Ballesteros J., Chirife J., Bozzini J. P. (1993). Specific solute effects on Staphylococcus aureus cells subjected to reduced water activity. Int. J. Food Microbiol. 20 (2), 51 – 66. doi:10.1016/0168-1605(93)90094-W
  9. Barbosa-Cánovas G. V., Fontana A. J., Schmidt S. J., Labuza T. P. (2020). Water Activity in Foods: Fundamentals and Applications (Wiley).
  10. Bernet T., Canu C., Jackson G. (2026). Revisiting the classification of physisorption isotherms with classical density functional theory. J. Phys. Chem. B 130, 1933 – 1949. doi:10.1021/acs.jpcb.5c08129
  11. Castro A. M., Mayorga E. Y., Moreno F. L. (2018). Mathematical modelling of convective drying of fruits: a review. J. Food Eng. 223, 152 – 167. doi:10.1016/j.jfoodeng.2017.10.016
  12. Courtney C., Al-Saadi A., Simha P. (2025). Balancing the oxidation of endogenous organics and macronutrient recovery from human urine treated with Fenton’s reagent: a targeted metabolomics study. Chem. Eng. J. Adv. 23, 100772. doi:10.1016/j.ceja.2025.100772
  13. Eltawil M. A., Azam M. M., Alghannam A. O. (2018). Energy analysis of hybrid solar tunnel dryer with PV system and solar collector for drying mint (Mentha viridis). J. Clean. Prod. 181, 352 – 364. doi:10.1016/j.jclepro.2018.01.229
  14. Erbay Z., Icier F. (2010). A review of thin layer drying of foods: theory, modeling, and experimental results. Crit. Rev. Food Sci. Nutr. 50 (5), 441 – 464. doi:10.1080/10408390802437063
  15. Ertekin C., Yaldiz O. (2004). Drying of eggplant and selection of a suitable thin layer drying model. J. Food Eng. 63 (3), 349 – 359. doi:10.1016/j.jfoodeng.2003.08.007
  16. Fang S., Wang L. P., Wu T. (2015). Mathematical modelling and effect of blanching pretreatment on drying kinetics of Chinese yam (Dioscorea opposita). Chem. Industry Chem. Eng. Q. 21 (4), 511 – 518. doi:10.2298/CICEQ140816007F
  17. Galan I., Glasser F. P., Baza D., Andrade C. (2015). Assessment of the protective effect of carbonation on portlandite crystals. Cem. Concr. Res. 74, 68 – 77. doi:10.1016/j.cemconres.2015.04.001
  18. Gao Y., Vinnerås B., Simha P. (2025). Partitioning behavior and crystallization of urea, salts and water during stepwise dehydration of acidified human urine by evaporation. Sci. Total Environ. 966, 178709. doi:10.1016/j.scitotenv.2025.178709
  19. Getahun S., Septien S., Mata J., Somorin T., Mabbett I., Buckley C. (2020). Drying characteristics of faecal sludge from different on-site sanitation facilities. J. Environ. Manag. 261, 110267. doi:10.1016/j.jenvman.2020.110267
  20. Inyang U., Oboh I., Etuk B. (2017). Drying and the different techniques. Int. J. Food Nutr. Saf. 8 (1), 45 – 72.
  21. Inyang U. E., Oboh I. O., Etuk B. R. (2018). Kinetic models for drying techniques—Food materials. Adv. Chem. Eng. Sci. 8 (2), 27 – 48. doi:10.4236/aces.2018.82003
  22. Karaaslan S., Uysal O., Uysal F. O., Ekinci K., Kumbul B. S. (2016). Mathematical modelling of drying of chlorella sp., Neochloris conjuncta and Botryococcus braunii at different drying conditions. Eur. J. Sustain. Dev. 5 (4), 421 – 430. doi:10.14207/ejsd.2016.v5n4p421
  23. Koukouch A., Idlimam A., Asbik M., Sarh B., Izrar B., Bostyn S. (2017). Experimental determination of the effective moisture diffusivity and activation energy during convective solar drying of olive pomace. Renew. Energy 101, 565 – 574. doi:10.1016/j.renene.2016.09.015
  24. Kumar N., Sarkar B. C., Sharma H. K. (2012). Mathematical modelling of thin layer hot air drying of carrot pomace. J. Food Sci. Technol. 49 (1), 33 – 41. doi:10.1007/s13197-011-0266-7
  25. Makununika B. S. N. (2016). Thermal Drying of Faecal Sludge from VIP Latrines and Characterisation of Dried Faecal Material. [ Master’s thesis ], University of KwaZulu-Natal. Available online at:.
  26. Mathlouthi M. (2001). Water content, water activity, water structure and the stability of foodstuffs. Food Control. 12 (7), 409 – 417. doi:10.1016/S0956-7135(01)00032-9
  27. Mawejje M. N. (2021). Solar Thermal Convective Drying Characteristics and Kinetics for Faecal Sludge Treatment. [ Master’s thesis ].
  28. Mehaidli A. P., Mandal R., Simha P. (2024). Selective degradation of endogenous organic metabolites in acidified fresh human urine using sulphate radical-based advanced oxidation. Water Res. 257, 121751. doi:10.1016/j.watres.2024.121751
  29. Melhus C. P., Vinnerås B., Simha P. (2026). Concentrating human urine intrinsically inhibits urease activity. Front. Environ. Sci. 14, 1805779.
  30. Metwally K. A., Oraiath A. A. T., Elzein I. M., El-Messery T. M., Nyambe C., Mahmoud M. M. (2024). The mathematical modeling, diffusivity, energy, and enviro-economic analysis (MD3E) of an automatic solar dryer for drying date fruits. Sustainability 16 (8), 3506. doi:10.3390/su16083506
  31. Minaei S., Motevali A., Ahmadi E., Azizi M. H. (2012). Mathematical models of drying pomegranate arils in vacuum and microwave dryers. J. Agric. Sci. Technol. 14 (2), 311 – 325.
  32. Mirzaee E., Rafiee S., Keyhani A. (2010). Evaluation and selection of thin-layer models for drying kinetics of apricot (cv. NASIRY). Agric. Eng. Int. CIGR J. 12 (2), 111 – 116. Available online at:.
  33. Onwude D. I., Hashim N., Janius R. B., Nawi N. M., Abdan K. (2016). Modeling the thin-layer drying of fruits and vegetables: a review. Compr. Rev. Food Sci. Food Saf. 15 (3), 599 – 618. doi:10.1111/1541-4337.12196
  34. Ortiz-Jerez M. J., Sánchez A. F., Zapata Montoya J. E. (2022). Drying kinetics and sensory characteristics of dehydrated pumpkin seeds (Cucurbita moschata) obtained by refractance window drying. Heliyon 8 (10), e10947. doi:10.1016/j.heliyon.2022.e10947
  35. Özbek B., Dadali G. (2007). Thin-layer drying characteristics and modelling of mint leaves undergoing microwave treatment. J. Food Eng. 83 (4), 541 – 549. doi:10.1016/j.jfoodeng.2007.04.004
  36. Peleg M. (2022). Models of the water activity effect on microbial growth rate and initiation. Int. J. Food Microbiol. 371, 109670. doi:10.1016/j.ijfoodmicro.2022.109670
  37. Perez-Mercado L. F., Simha P., Moreira A. P., Paulo P. L., Vinnerås B. (2024). Circular fertilisers combining dehydrated human urine and organic wastes can fulfil the macronutrient demand of 15 major crops. Sci. Total Environ. 951, 175655. doi:10.1016/j.scitotenv.2024.175655
  38. Rad M. N., Shokri N. (2012). Nonlinear effects of salt concentrations on evaporation from porous media. Geophys. Res. Lett. 39 (4), L04403. doi:10.1029/2011GL050763
  39. Randall D. G., Krähenbühl M., Köpping I., Larsen T. A., Udert K. M. (2016). A novel approach for stabilizing fresh urine by calcium hydroxide addition. Water Res. 95, 361 – 369. doi:10.1016/j.watres.2016.03.007
  40. Rockland L. B., Beuchat L. R. (1987). Water Activity: Theory and Applications to Food (Marcel Dekker).
  41. Sansaniwal S. K., Sharma V., Mathur J. (2018). Energy and exergy analyses of various typical solar energy applications: a comprehensive review. Renew. Sustain. Energy Rev. 82, 1576 – 1601. doi:10.1016/j.rser.2017.07.003
  42. Simha P. (2021). Alkaline urine dehydration. PhD Thesis. Uppsala: Swedish University of Agricultural Sciences. Available online at:.
  43. Simha P., Karlsson C., Viskari E. L., Malila R., Vinnerås B. (2020a). Field testing a pilot-scale system for alkaline dehydration of source-separated human urine: a case study in Finland. Front. Environ. Sci. 8, 570637. doi:10.3389/fenvs.2020.570637
  44. Simha P., Lalander C., Nordin A., Vinnerås B. (2020b). Alkaline dehydration of source-separated fresh human urine: preliminary insights into using different dehydration temperature and media. Sci. Total Environ. 733, 139313. doi:10.1016/j.scitotenv.2020.139313
  45. Simha P., Senecal J., Gustavsson D. J. I., Vinnerås B. (2020c). “ Resource recovery from wastewater: a new approach with alkaline dehydration of urine at source,” in Current Developments in Biotechnology and Bioengineering: Sustainable Bioresources for Emerging Bioeconomy (Elsevier), 46. doi:10.1016/B978-0-444-64327-0.00015-1
  46. Simha P., Deb C. K., Randall D. G., Vinnerås B. (2022). Thermodynamics and kinetics of pH-dependent dissolution of sparingly soluble alkaline Earth hydroxides in source-separated human urine collected in decentralised sanitation systems. Front. Environ. Sci. 10, 889119. doi:10.3389/fenvs.2022.889119
  47. Simha P., Vasiljev A., Randall D. G., Vinnerås B. (2023). Factors influencing the recovery of organic nitrogen from fresh human urine dosed with organic/inorganic acids and concentrated by evaporation in ambient conditions. Sci. Total Environ. 879, 163053. doi:10.1016/j.scitotenv.2023.163053
  48. Simha P., Ahopalo N., Pay O., Jermakka J., Vasiljev A. (2025). On-site reactor for treating source-separated human urine with sparingly soluble fumaric acid in building-scale sanitation systems. Front. Environ. Sci. 13, 1546396.
  49. Simha P., Malefors C., van der Merwe G., Marais E. (2026). Passive solar-thermal concentration of human urine for decentralised sanitation using an air-closet evaporation architecture. Desalinat. 120406.
  50. Somorin T., Getahun S., Septien S., Kolios A., Buckley C. (2021). Isothermal drying characteristics and kinetics of human faecal sludges. Gates Open Res. 5, 66. doi:10.12688/gatesopenres.13203.2
  51. Srikiatden J., Roberts J. S. (2007). Moisture transfer in solid food materials: a review of mechanisms, models, and measurements. Int. J. Food Prop. 10 (4), 739 – 777. doi:10.1080/10942910601161672
  52. Turhan M., Sayar S., Gunasekaran S. (2002). Application of peleg model to study water absorption in chickpea during soaking. J. Food Eng. 53 (2), 153 – 159. doi:10.1016/S0260-8774(01)00152-2
  53. Vasiljev A., Simha P., Demisse N., Karlsson C., Randall D. G., Vinnerås B. (2022). Drying fresh human urine in magnesium-doped alkaline substrates: capture of free ammonia, inhibition of enzymatic urea hydrolysis and minimisation of chemical urea hydrolysis. Chem. Eng. J. 428, 131026. doi:10.1016/j.cej.2021.131026
  54. Yu C., Yin W., Yu Z., Chen J., Huang R., Zhou X. (2021). Membrane technologies in toilet urine treatment for toilet urine resource utilization: a review. RSC Adv. 11 (56), 35525 – 35535. doi:10.1039/d1ra05816a
  55. Zhang Y., Li Z., Zhao Y., Chen S., Mahmood I. B. (2013). Stabilization of source-separated human urine by chemical oxidation. Water Sci. Technol. 67 (9), 1901 – 1907. doi:10.2166/wst.2013.055

Citation

Niwagaba, C. B., et al. (2026). Mechanistic evaluation of water removal dynamics in stabilised human urine for circular fertiliser production. Frontiers in Chemical Engineering, Volume 8. https://doi.org/
Received
Accepted
Published
Volume
8 - 2026

Copyright

Copyright © 2026 Niwagaba, Vasiljev, Kasirye, Lubuulwa, Katukiza, Semiyaga, Manga and Simha.

License: https://creativecommons.org/licenses/by/4.0/

*Correspondence: Charles B. Niwagaba; Prithvi Simha

Article metrics

Views
0
Downloads
0
Citations
0
Shares
0

Related articles

Original ResearchImproving decision-making in energy system modeling by formalizing the scenario-based approach using knowledge management toolsEdrisi Munoz, Turhan DemirayPerspectiveGlycan-intrinsic monoclonal antibody biomanufacturingAron GyorgypalOriginal ResearchMethylomonas methanica and Methylosinus sp. co-cultivated with methylotrophs under saline conditions for microbial protein production from biogas as alternatives to natural-gas- based production using Methylococcus capsulatusFrederiek-Maarten Kerckhof, Norah Benmeridja, Korneel Rabaey, Nico Boon, Jo De Vrieze, Myrsini Sakarika