Elsevier

Water Research

Volume 37, Issue 11, June 2003, Pages 2667-2677
Water Research

Estimating the precipitation potential in urine-collecting systems

https://doi.org/10.1016/S0043-1354(03)00071-XGet rights and content

Abstract

Precipitation in urine-separating toilets (NoMix toilets) and waterless urinals causes severe maintenance problems and can strongly reduce the content of soluble phosphate. In this study, we present a computer model for estimating the precipitation potential (PP) in urine-collecting systems. Calculating the PP enables to predict the composition and mass concentration of precipitates. We used our computer model for investigating how urea hydrolysis and dilution with flushing water affect precipitation. In a previous study, we found that microbial urea hydrolysis (ureolysis) triggers precipitation and that the amount of precipitates is limited by calcium and magnesium. With the present simulations, we could confirm these findings. We determined that only a small fraction of urea has to be hydrolysed for reaching 95% of the maximum PP. Since urease-positive bacteria are abundant in urine-collecting systems, strong precipitation is very likely. In further simulations, we determined that struvite (MgNH4PO4·6H2O) and hydroxyapatite (HAP, Ca10(PO4)6(OH)2) are the main precipitate compounds. If urine is highly diluted with tapwater, calcite (CaCO3) occurs as well. HAP is the only calcium phosphate mineral, although several others were supersaturated. Additionally, the simulations indicated that urine dilution diminishes the risk of blockages, since the mass concentration of precipitates decreases with the volume of flushing water. Rainwater flushing is more effective than flushing with tapwater. Moreover, flushing with tapwater leads to high phosphate fixation, because the total amount of calcium and magnesium ions increases, while the total amount of phosphate keeps constant. Finally, we compared simulation results with field measurements and found good agreement at low and very high urine dilution.

Introduction

Urine contributes by far most nutrients to wastewater, and contains most micropollutants excreted from human metabolism. Therefore, separate collection and treatment of urine is a possible way to cope with today's challenges in wastewater management [1].

Urine-separating toilets (NoMix toilets) and waterless urinals are used for collecting low diluted urine. First experiences with urine-collecting systems showed that blockages caused by mineral precipitation are a major maintenance problem [2]. Precipitation occurred in traps, connecting pipes, and storage tanks. Apart from causing blockages, precipitation restricts the later treatment and use of source-separated urine, as phosphate is incorporated in solids [2].

Investigations on running systems and laboratory experiments showed that enzymatic urea hydrolysis (ureolysis) triggers precipitation [2], [3]. Bacteria that produce the urea hydrolysing enzyme urease are ubiquitous [4]. They also colonise urine-collecting systems [3].

Urease decomposes urea into ammonia and bicarbonate causing a pH increaseNH2(CO)NH2+2H2ONH3+NH4++HCO3.

The release of ammonia and bicarbonate, as well as the pH increase, promotes precipitation. In field measurements we identified struvite, calcite, and hydroxyapatite (HAP) as precipitate compounds. Ureolysis is the initial trigger for precipitation, but dilution with flushing water determines the precipitate composition [2].

In the present work, we set up a computer model for simulating how ureolysis, urine dilution, and flushing water characteristics influence precipitation in urine-collecting systems. As indicator, we used the precipitation potential (PP). Loewenthal et al. [5] defined the PP as mass concentration of the mineral that will precipitate from solution to establish an equilibrium state between species in aqueous and solid phases. Accordingly, if several minerals precipitate, their overall mass concentration is the total PP. Calculating the PP enables to predict the composition and mass concentration of precipitates.

Another way for assessing precipitation risks is determining the mineral saturation. Precipitation is only possible if a solution is supersaturated with respect to a mineral. Thermodynamically, supersaturation means that the mineral's ion activity product (IAP) exceeds the solubility product (Ksp) [6]. An often used indicator is the saturation index SI [7].

Several computer models can calculate the mineral saturation, e.g. MINTEQA2 [7] for geochemical problems, or EQUIL93 [8] for evaluating urinary stone risks. However, estimating the saturation is only of limited use for assessing the risk of precipitation. Not all supersaturated minerals will actually precipitate and the degree of saturation is not directly related to the expected amount of precipitates.

The PP is the final mass concentration of precipitates at solid–solute phase equilibrium. However, kinetic factors such as activation energy barriers, nucleation seeds, or inhibitors determine outset and velocity of precipitation [6]. Therefore, fresh precipitates may not accord with the PP. Nevertheless, their composition and mass concentration will converge to the PP, while the solid–solute phase equilibrium establishes.

Computer programs that can calculate PPs are available for special cases. Loewenthal et al. [5] presented a program for struvite precipitation in digester effluents. Ashby et al. [9] used a similar approach for assessing the composition of urinary stones. Several computer programs or models are available to calculate the calcium carbonate precipitation potential CCPP in tapwater [10], [11]. Minteqa2 can be used as well for calculating mineral mass concentrations at solid–solute phase equilibria.

Compared to the programs and models above our model has additional features:

  • Effects of dynamic processes such as ureolysis and carbon dioxide degasing can be simulated.

  • Concentrations and pH of a mixed solution can be calculated.

  • Data and processes can easily be changed or added.

  • No special programming skills are necessary.

Musvoto et al. [12] have published a similar model. With it, the authors could successfully reproduce batch aeration tests on anaerobic digester supernatants [13]. We will shortly compare the two models at the end of the Model section.

Section snippets

Model

Our model basically calculates a thermodynamic equilibrium state. It considers solubility equilibria, acid–base and complex formation reactions.

The solutes integrated in the model are urea, calcium, magnesium, potassium, sodium, ammonia, carbonate, phosphate, sulphate, chloride, citrate, oxalate, protons, and hydroxylions, their species and complexes. The modelled solids are amorphous calcium phosphate (ACP, Ca3(PO4)2), dicalcium phosphate anhydride (DCPA, CaHPO4), dicalcium phosphate dihydrate

Materials and methods

We took samples of precipitates and wastewater in one NoMix toilet (Type DS, Wost Man Ecology AB, Saltsjö-Boo, Sweden), two waterless toilets (Urimat AG, Tann-Rüti, Switzerland), and one conventional urinal (trap from Geberit AG, Jona, Switzerland, Art. No. 152.936.11.1). The pipe of the NoMix system between toilet and collection tank had a length of 4.0 m, of which about 2.0 m were nearly horizontal. The inner diameter of the pipe and trap was 25 and 20 mm, respectively. The urine retention time

Effects of flushing with tapwater

Conventional and NoMix toilets are usually flushed with tapwater. For our simulations, we used tapwater concentrations typical for the Swiss Midlands (Table 3). We neglected complexation with citrate and oxalate assuming that bacteria have degraded these substances before the solid–solute phase equilibrium is reached.

In our simulations, flushing with tapwater had two effects.

  • 1.

    The total amount of precipitates increased, because tapwater supplied additional calcium and magnesium (Fig. 1 and Table 3

Conclusions

Modelling the precipitation potential (PP) is a promising approach for predicting precipitation effects in urine-collecting systems. PP predicts both the mineral composition and the expected mass concentration of precipitates. However, it does not consider further effects, which govern the occurrence of blockages, such as precipitation kinetics, attachment of minerals on pipe walls or foreign solids. Calculating PP is particularly applicable for old precipitates, since they are supposed to be

References (35)

  • Allison, JD, Brown DS, Novo-Gradac KJ. Minteqa2/Prodefa2, a Geochemical Assessment Model for Environmental Systems:...
  • C.M. Brown et al.

    EQUIL93a tool for experimental and clinical urolithiasis

    Urol Res

    (1994)
  • R.A. Ashby et al.

    Urine is a saturated equilibrium and not a metastable supersaturated solutionevidence from crystalluria and the general composition of calcium salt and uric acid calculi

    Urol Res

    (1999)
  • T.R. Holm et al.

    Computing SI and CCPP using spreadsheet programs

    J Am Water Works Assoc

    (1998)
  • G.H. Nancollas et al.

    Formation and dissolution mechanisms of calcium phosphates in aqueous systems

  • J.C. Elliott

    Structure and chemistry of the apatites and other calcium orthophosphates

    (1994)
  • Martell AE, Smith RM, Motekaitis RJ. Critical Selected Stability Constants of Metal Complexes, Version 4.0. NIST...
  • Cited by (170)

    View all citing articles on Scopus
    View full text