Recovery of nitrogen-fertilizer from centrate of anaerobically digested sewage sludge via gas-permeable membranes

https://doi.org/10.1016/j.jwpe.2020.101630Get rights and content

Highlights

  • Gas permeable membranes enabled 98 % recovery of NH3-N from municipal centrate.

  • Achieved flux ranging 21−133 g N/d per m2 that is better than literature reports.

  • Achieved N-fertilizer yield of 2.1 g ammonium sulfate per liter of centrate.

  • N-fertilizer recovered from centrate met EPA regulations for land application.

Abstract

It has been estimated that about 15 % of the current nitrogen-fertilizer demand could be served by recovering the nitrogen-content in domestic sewage. In most sewage treatment plants employing the activated sludge process, the concentrated liquid stream (referred to as centrate) resulting from anaerobic digestion of the waste sludge offers a good opportunity for recovering nitrogen for use as fertilizer. This paper presents a non-pressurized gas-permeable membrane-based approach capable of recovering 98 % of the ammoniacal-nitrogen in the centrate as high purity fertilizer suitable for crop cultivation. In the proposed approach, ammoniacal-nitrogen in the centrate on the feed-side of the hydrophobic gas-permeable membrane is preferentially separated and absorbed into sulfuric acid on the product-side to yield ammonium sulfate solution. A tubular gas-permeable membrane reactor developed in this study was able to recover 21−133 g N/d per m2 of membrane surface area, which is higher than the typical values of 2–62 g N/d per m2 reported for more concentrated feeds. The highest yield recorded in this study was 2.1 g of ammonium sulfate per liter of centrate. Scanning electron microscope equipped with an energy dispersive x-ray spectrometer confirmed the presence of nitrogen, oxygen, and sulfur in the recovered fertilizer. Inductively coupled plasma optical emission spectrometer results confirmed that heavy metal levels in the recovered fertilizer were well below the regulatory levels for land application.

Introduction

This study is based on the premise that nitrogen-fertilizers could be recovered from the centrate resulting from anaerobic digestion of the waste sludge generated at sewage treatment plants (STP). The motivation for the study is that, such recovery could offset about 15 % of the demand for nitrogen-fertilizers manufactured by the energy-intensive Haber-Bosch process, lowering its dependence on fossil fuels and emissions of greenhouse gases [1]. Recovering nutrients as fertilizers for crop cultivation instead of discharging them can also lower the potential for eutrophication in receiving waters [2,3]. The aim of this study is to demonstrate the feasibility of a gas-permeable membrane reactor in recovering ammonium from centrate for use as crop fertilizer. Performance of this reactor in recovering nitrogen-fertilizers from centrate generated at an STP in Las Cruces, NM, is quantified and compared with that reported for current recovery technologies.

Current technologies for recovering N from centrate include air-stripping, ion-exchange, struvite precipitation, and pressure-driven membrane separation [[4], [5], [6], [7], [8], [9], [10], [11], [12], [13]]. Performance of the air-stripping process depends on the degree of dissociation of ammoniacal-N in the centrate to gaseous ammonia which, in turn, depends on pH and temperature. Provolo et al. [6] found that, at pH of 9.05, N-recovery could be increased from 64 % at 30 °C to 87 % at 40 °C; but recovery of 84 % could be achieved at a lower pH of 7.81 if operated at 50 °C. Tao and Ukwuani [7] reported 50 % N-recovery in 5 h at 97 °C by thermal stripping and complete stripping in 3−4 hr at 101−102 °C. Most air-stripping studies had concluded that heating of the feed can be beneficial in improving recovery. Limoli et al. [8] reported that instead of air-stripping, mixing at a pH of 12 can recover ∼51 % of N.

Thornton et al. [9] reported >95 % removal of ammonium by ion-exchange with mesolite as the exchange medium and 5 % NaOH as the regenerant solution. Another study using bentonite as the medium reported 52.3 % removal of ammonium from municipal centrate at pH 7.8 [10]. A study using natural zeolite reported ∼50 % removal of ammonium from centrate at pH 8.3 [11]. Although N could be recovered from centrate by precipitation as struvite by adding magnesium chloride (MgCl2), its recovery is limited by the available phosphate due to the composition of struvite (Mg2+:NH4:PO4 of 1:1:1). Recovery of only 4.5 % of ammonium has been reported at a pH of 7.5 [12]. Jia et al. [13] showed that by adding supplemental potassium phosphate, >99 % of N in the centrate could be recovered.

Pressure-driven membrane processes such as microfiltration, nanofiltration, and reverse osmosis have been proposed for separation of N (and other chemicals of value), mostly from high-strength wastes (e.g. livestock wastes). These processes concentrate the target chemicals into the retentate for further downstream processing for recovery [[14], [15], [16], [17], [18]]. Gerardo et al. [14] have reported on recovery of NH3-N from anaerobically digested dairy farm sludge by cross-flow ceramic microfiltration and subsequent recovery of nutrient solutions of desired N:P ratios and metals via diafiltration under varying pH conditions. Ledda et al. [18] reported that 72 % of ammonium from centrate of digested cattle manure could be concentrated into the retentate of an integrated ultrafiltration (3.5–4.5 bar) and reverse osmosis system. Since the bulk medium permeates through the membrane in pressure-driven membrane systems, their performance could decline with time due to membrane fouling [16,18]. Hence, the feed is often pretreated by a series of solid-separation processes to minimize membrane fouling. For example, in the study by Ledda et al. [18], the feed was pretreated by a screw press and centrifuge. Yet, gradual buildup of foulants on the membrane surface can increase the energy consumption, maintenance cost, and reduce membrane lifetime making it undesirable for continuous operation [16].

Non-pressurized gas-permeable membrane (GPM) processes have emerged recently for recovering ammoniacal-N from waste streams including centrates. Gas-permeable membranes afford preferential and selective diffusion of gas molecules dissolved in a feed solution on the feed-side of the membrane towards an absorbing solution on the product-side under ambient pressure. When applied to centrate treatment, at pH greater than 9.26, dissolved NH3 in the centrate dissociates to form NH3(g) (pKa,NH3 = 9.26) and diffuses through the GPM towards an acid solution (e.g. sulfuric acid) on the product-side of the GPM, where it reacts rapidly with the acid to form an ammonium salt (e.g. ammonium sulfate).

A limited number of previous studies had reported on the application of non-pressurized GPMs in N-recovery from centrates, but from livestock wastes laden with high levels of N (1000-3500 mg N/L) [[19], [20], [21]]. Garcia-Gonzalez and Vanotti [19] reported recovery of NH4+ in swine manure (2290 mg NH4+/L) by a GPM at 190 mg/L-d where the pH of the feed ranged 7.7-9.0. Dube et al. [20] found that mild aeration of the feed-side was beneficial in recovering 96–98 % of N in 5 days compared to 25 days without aeration. Similarly, Riano et al. [21] integrated both slow rate aeration and mixing to remove 78 % of ammoniacal-N from raw swine manure with an initial concentration of 3425 mg/L in 7 days. Both studies by Dube et al. [20] and Riano et al. [21] used inhibitors to avoid loss of N by nitrification during aeration.

A recent study undertook a multi-criteria analysis of N-recovery processes considering ten performance criteria including pre-treatment, operating conditions, recovery performance, chemical and energy demands, and post-treatment [22]. This study identified the emergent gas-permeable membrane process as the preferred option over air-stripping, ion exchange, struvite precipitation, and reverse osmosis [22].

Building on the limited literature, in this feasibility study, it was presumed that GPMs could be engineered for N-recovery from municipal centrate that typically contains lower levels of ammonia (<500 mg N/L) than in previous studies. Specific goals of this study were to i) quantify N-recovery performance of GPMs; ii) examine the purity of recovered fertilizer; and iii) compare the results of this study with literature reports. Although Dube et al. [20] had suggested aerating the feed to enhance N removal, we opted to raise its pH to >9.26 instead, to minimize energy input, nitrification, and volatilization of N, and maximize dissociation of NH4+ to NH3(g). As part of this study, an empirical model describing the performance of the proposed gas permeable membrane reactor was developed for future scale-up and optimization. Purity of the recovered ammonium sulfate was evaluated against the EPA guidelines for land application in terms of elemental composition and heavy metal content to justify its use in crop cultivation.

Section snippets

Gas permeable membrane reactor (GPMR)

The GPMR developed in this study was of a tubular configuration, fabricated out of expanded polytetrafluoroethylene (ePTFE) tube (outer diameter =11.1 mm; wall thickness =0.2 mm; Zeus Industrial Products Inc., NJ) serving as the GPM. This tubular GPM was submerged in the feed solution (centrate) contained in a capped glass bottle. Sulfuric acid solution (pH = 2) serving as the absorbing solution was circulated from a reservoir through the tubular GPM by a peristaltic pump (Masterflex®) at 25

Removal of N from centrate

Fig. 2 shows the temporal decrease of normalized concentrations of NH3-N in the feed-side (centrate) in Tests A-E, where N-removals ranged 69–98 %. The lowest removal efficiency (= 69 %) was in Test A that was initiated with a pH of 9.3, where conjugate acid and base of ammonia exist in equal concentrations. In comparison, in Test D, under the same test conditions as in Test A, except for the initial pH of 10, higher removal of 97 % was achieved in the same period (12 h). This improvement is

Conclusions

Feasibility of the gas-permeable membrane technology in recovering ammonium sulfate as fertilizer from municipal centrate has been demonstrated here for the first time. Compared to the preferred current practice of air-stripping, the proposed approach enabled similar recoveries, but without the need for any feed-heating. pH adjustment of the feed to 10 and mild mixing on the feed-side were found to be the optimal operating conditions. This feasibility study demonstrated NH3-N recovery

Declaration of authors’ contribution

All authors whose names listed this manuscript certify that they have participated sufficiently in the work to take public responsibility for the content, including participation in the concept, design, analysis, writing, or revision of the manuscript.

Statement of informed consent, human/animal rights

No conflicts, informed consent, human or animal rights applicable.

Declaration of authors agreement to authorship and submission of the manuscript for peer review

All authors whose names are listed in this manuscript have contributed significantly to the work, have read the manuscript, attest to the validity and legitimacy of the data and its interpretation, and agree to its submission to Journal of Water Process Engineering for peer review.

Declaration of Competing Interest

The authors report no declarations of interest.

Acknowledgments

This study was supported in part by the NSF Engineering Research Center for Reinventing the Nation’s Urban Water Infrastructure (ReNUWIt) award # EEC 1028968, the College of Engineering at NMSU, and the Ed & Harold Foreman Endowed Chair. The authors acknowledge the support provided by City of Las Cruces Utilities; Barbara Hunter from the Department of Plant and Environmental Sciences, NMSU for conducting the heavy metal analysis; and Microscopic Imaging Core Suite (MICS) of Core University

References (31)

Cited by (26)

View all citing articles on Scopus
View full text