Influences of ammonium and phosphate stimulation on metalworking fluid biofilm reactor development and performance

Influences of ammonium and phosphate stimulation on metalworking fluid biofilm reactor development and performance

New BIOTECHNOLOGY xxx (xxxx) xxx–xxx Contents lists available at ScienceDirect New BIOTECHNOLOGY journal homepage: www.elsevier.com/locate/nbt Full...

678KB Sizes 0 Downloads 22 Views

New BIOTECHNOLOGY xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

New BIOTECHNOLOGY journal homepage: www.elsevier.com/locate/nbt

Full length Article

Influences of ammonium and phosphate stimulation on metalworking fluid biofilm reactor development and performance ⁎

Shivashkar Singh , Lakshmi Manjoosha Adapa, Nicholas Hankins Department of Engineering Science, University of Oxford, Parks Road, Oxford, OX1 3PJ, United Kingdom

A R T I C L E I N F O

A B S T R A C T

Keywords: Biofilm Biostimulation Metalworking fluids Bioreactor Biodegradation

In this study, the effects of common wastewater stimulants, namely NH4Cl and KH2PO4, on the development and performance of metalworking fluid biofilm bioreactors are presented. It is shown that biofilms flourished only when one of these components was present in limiting quantities. Biofilm yields significantly declined when both of the components were withheld from the bioreactors or when both components were provided in excess. Stimulations to the reactors using NH4Cl significantly reduced the total carbon removal performance, while stimulations using KH2PO4 resulted in significant increases in performance. Chromatographic analyses showed that the NH4Cl stimulation enhanced the removal of saturated fatty amides and diethylene glycol butyl ether from the metalworking fluid, but inhibited the removal of diisoproponolamine. Furthermore, NH4Cl additions inhibited the oil/water separation carbon removal mechanism and resulted in the re-dispersion of recalcitrant organic material. The results from this study show that metalworking fluid practitioners should take care in choosing the nutrients used for stimulating bioreactor performance and microbe development. Incorrect stimulations with NH4Cl may result in negative treatment performances due to the inhibition of amine utilisation and enhancing emulsion stability.

Application of fixed-film reactors for treating chemical wastes When considering the design of microbial reactors, the mode of growth is an important parameter that must be taken into account [1]. Microbes such as bacteria are able to grow in both a planktonic form, or in a fixed-form that is associated with a solid surface through the formation of biofilms [1–4]. From an engineering perspective, physically fixed modes of growth within bioreactors offer the advantages of increased biomass concentration in a reactor, protection against hazardous waste, and inherent biomass retention [5–7]. The biological treatment of waste metalworking fluids is a cost-effective means of remediation [8]. However, both the presence of inhibitory components and the slow degradation kinetics associated with complex organic components hinder the effectiveness of the treatment process [9,10]. Since the application of biofilms to the treatment of hazardous wastes is a suitable means of compensating for these disadvantages, several studies have utilised fixed-film reactors for the treatment of metalworking fluids [11–14]. This mode of treatment is typically applied to effluents which have had their inhibitory components degraded with time and to formulations that contain biodegradable components. For formulations containing inhibitory, or non-biodegradable components, the biological treatment of metalworking



fluids is applied after a pre-treatment step (which could be a physical or chemical treatment). Since the biological process is typically slower than other forms of treatment, it is more feasible for treating large volumes of metalworking fluids [8]. It is common practice to add nutrients to wastewater treatment systems to facilitate bioreactor development and improve treatment performance [15,16]. While there is an abundance of data on how biostimulation may influence biofilm development in industrial wastewater treatment reactors [17–20], there are a limited number of studies looking at how it may influence the development and performance of biofilm reactors treating metalworking fluids. Since different nutrient conditions may trigger different biofilm reactor responses [21], a study looking into the effects of stimulation on the development of an indigenous consortium biofilm in metalworking fluid reactors is warranted. In this study, the effects of KH2PO4 and NH4Cl stimulation and limitation on the production of biofilm biomass and bioreactor performance are presented. These compounds were chosen since they are commonly added to wastewater systems to promote growth and development of biomass and to improve reactor performance. The novelty of the investigation lies in the combination of investigating how stimulation affects both biofilm development and reactor performance simultaneously and explaining the trends observed through an in-depth

Corresponding author. E-mail address: [email protected] (S. Singh).

http://dx.doi.org/10.1016/j.nbt.2017.09.002 Received 13 November 2016; Received in revised form 27 August 2017; Accepted 11 September 2017 1871-6784/ © 2017 Published by Elsevier B.V.

Please cite this article as: Singh, S., New BIOTECHNOLOGY (2017), http://dx.doi.org/10.1016/j.nbt.2017.09.002

New BIOTECHNOLOGY xxx (xxxx) xxx–xxx

S. Singh et al.

through a 0.22 μm syringe filter to ensure sterility and through a humidifier made from a 50 mL centrifuge tube, before being bubbled into a 100 mL bioreactor (100 mL Duran bottles). The air was humidified so that evaporation was negligible during the course of the experiment. The air was pumped at 0.5 L/min. The reactor was submerged in a water bath kept at 27° C. Biofilms were grown on a 3 cm × 5 cm polyethylene matrix which is commercially sold as BioBlok 300. A picture of the matrix is given in Fig. 1B.

chemical analysis of the metalworking fluid. The choice of metalworking fluid formulation and reactor system makes this possible and allows for a more in-depth understanding of the process to be obtained. Materials and methods Micro-organisms A mixed consortium used for the industrial treatment of metalworking fluid wastes was supplied by Microbial Solutions Ltd. (a company specialised in the disposal of metalworking fluids). The mixed consortium was provided in freeze-dried form, and originated from an aged reactor treating spent metalworking fluid wastewaters. The inoculum used for this community was described by van der Gast and Thompson [22]. In order to resuscitate the freeze dried micro-organisms, 2 mL of phosphate buffered saline solution (PBS) was added to the vials in which they were contained. The vial was left to stand for two hours, before the contents were transferred to a 250 mL flask containing 95 mL of Luria–Bertani (LB) Media and 5 mL of 10% metalworking fluid solution (recipe for the metalworking fluid is given below). The flask was incubated in an orbital shaking incubator set at 28 °C and 120 rpm for 18 h. After incubation, the contents of the flask were centrifuged at 4100 rpm to collect the cell pellet and to discard the supernatant. The cell pellet was washed with deionised (DI) water twice, before being resuspended once more in DI water. This suspension served as the inoculum for all experiments described.

Bioreactor operation In order to achieve a significantly measurable amount of biomass, the batch biofilm reactors were operated for 2 cycles. The first cycle lasted for 4 days and the second for 3 days (since nutrients are consumed faster in the second cycle). At the end of each cycle, the total carbon was measured, and the total carbon removal efficiency of the bioreactor was calculated using the following equation:

TCcyc1 + TCcyc 2 ⎞ *100 TC removal efficiency (%) = ⎛1 − 2*TCinitial ⎠ ⎝ ⎜



TCcyc1 and TCcyc2 are the total carbon concentrations at the end of the first and second batch cycles respectively. TCin is the initial total carbon concentration of the artificial metalworking fluid used for the experiments (4600 mg/L). Biofilm dry weight and biofilm yield Biofilm dry weight measurements were taken at the end of the second cycle of bioreactor operation. To measure the weight of the biofilm attached to the substratum, the matrix was taken out of the reactor and placed within a 50 mL centrifuge tube. The matrix was gently rinsed with deionised (DI) water twice and then the biofilm was dislodged into 20 mL of PBS solution through a combination of vigorous shaking and vortexing using a Vortex Genie 2. 0.2 mL of a 10% Triton X-100 solution was added to the centrifuge tube to re-emulsify the oils that had dislodged from the biofilm and the matrix. The tube containing the dislodged biofilm was centrifuged at 4100 rpm for 15 m, which resulted in a distinct biomass layer floating on the top of the tube and a cell pellet. The pellet was found to be of a negligible mass compared to the floating biomass and thus only the floating biomass was considered in dry-weight determinations. To determine the mass of the floating layer, the supernatant of the tube and the floating biomass were passed through a piece of Whatman filter paper (Qualitative Number 1). The filter paper was dried overnight at 60 °C and the difference in its weight before and after the biomass addition was taken as the dry-weight of the biofilm. The biofilm yield can be calculated using the total carbon (TC) removed from the bioreactor over both cycles of operation, and the biofilm dry weight obtained:

Metalworking fluid An artificial, semi-synthetic concentrate was developed and used for all of the experimental work. The formulation used was modified from that given by Childers [23]. The formulation was developed since metalworking fluid wastewaters and commercial concentrates usually contain a mixture of organics which are difficult to identify and characterise. By developing an artificial proxy, observed trends can be investigated in further detail. To prepare the semi-synthetic concentrate, 13.5 g of Naphthenic Mineral Oil (base oil), 4.5 g of Sodium Sulfonate (emulsifier), 10 g of Tall Oil Fatty Amide (emulsifier), and 1.8 g of Diethylene Glycol Butyl Ether (coupler) were mixed together. 28 g of this mixture was added to 75 g of water to create the final semi-synthetic concentrate. All experiments described used a 2.5% (v/v) metalworking fluid concentration prepared by diluting this metalworking fluid using an artificial tap water [24]. The chosen components are common ingredients that are found within metalworking fluid formulations to add confidence that the results obtained will be applicable across different formulations [23]. To ensure that the developed consortium is capable of treating formulations containing biocides, a developed bioreactor was applied to formulations containing varying concentrations of sodium orthophenyl phenate (Na-OPP).

mg Dryweight ⎞⎟ = Biofilm Yield ⎜⎛ (2*TCin − TCcyc1 − TCcyc 2)* 0.1 ⎝ mg TOC removed ⎠

Reagents

TCcyc1 and TCcyc2 are the total carbon concentrations at the end of the first and second batch cycles respectively. TCinitial is the initial total carbon concentration of the artificial metalworking fluid used for the experiments (4600 mg/L).

Analytical grade inorganic salts (CaSO4·2H2O, MgSO4·7H2O, NaNO3, NaCl, FeSO4·7H2O, KH2PO4, K2HPO4, (NH4)2SO4) Triton X-100, acetonitrile (99.9% purity), acetic acid and Benzoyl chloride were purchased from Sigma Aldrich. Samples of naphthenic mineral oil were provided by Nynas Base Oils. Samples of sodium sulfonate were provided by Sonneborn. Tall oil fatty amides were provided by Colonial Chemical.

HPLC HPLC analyses were used for the determination of the concentrations of amides, DGBE, DIPA within the reactors. All analyses were done on an Agilent 1120 compact HPLC system equipped with an Agilent C18 Eclipse Plus Column and a UV–vis detector. All analyses were done at 210 nm using an isocratic elution made up of a ratio of acetonitrile acidified with 0.2% acetic acid and DI water (for amides, the ratio was

Bioreactor system A schematic for the bioreactor system is provided in Fig. 1A and a photo of the set-up is provided in Fig. 1C. Briefly, air was passed 2

New BIOTECHNOLOGY xxx (xxxx) xxx–xxx

S. Singh et al.

Fig. 1. A-Schematic of bioreactor set-up. B-Polyethylene Matrix before and after growth. C-Photo of experimental set-up.

90:10, for DGBE and DIPA, the ratio was 70:30). 2.5 μL of amide samples were directly injected into the system for analysis. DGBE and DIPA samples were derivatised before 5 μL was injected into the system. A method for HPLC sample derivatisation developed by Sinjewel et al. [25] was modified and applied to samples with DGBE and DIPA.Briefly, 100 μL of sample was added to an Eppendorf tube, together with 100 μL of propylene glycol (which served as the internal standard). To this tube, 250 μL of a 300 g/L NaOH solution was added with 15 μL benzoyl chloride. The tube was then vigorously shaken for 15 min using a TOC X-5 shaker. After this, 100 μL of 100 g/L glycerol solution was added to terminate the derivatisation reaction. The sample tube was then shaken for another 5 min. After this, 400 uL of Hexane was added to the tube, which was then shaken for another 5 min. After shaking, the tube was centrifuged for 5 min at 14,500 rpm. The hexane layer was extracted and dried under a stream of air. To the residue, 300 uL of acetonitrile was added. This acetonitrile sample was injected into the HPLC for analysis. All calibration curves made were linear with an R2 value of greater than 0.98.

Statistical analysis Single-factor ANOVA was used to test for significance of a response as a function of a varied parameter. The 2-tail student t-test with equal variance was used to test for significance between two individual responses. Tests were deemed to be significant for results were p < 0.05. Results and discussion Ammonium influences For the experiments in this section, the artificial metalworking fluids were stimulated with 100 ppm of KH2PO4 to ensure that there was an ample phosphorus source available to the microbes. Fig. 2A shows the effect of NH4Cl addition on both the extent of biofilm development and the total carbon removal performance. It can be seen that the addition of NH4Cl significantly reduced biofilm formation within the reactor. An increase of NH4Cl stimulation from 10 ppm to 200 ppm resulted in a significant biomass decline from 79.7 ± 2.0 mg to 19.2 ± 2.1 mg (p < 0.001). Further to the absolute biomass decline, there was also a significant decline in the yield of the biofilm. Fig. 2 B shows that the increase of the addition of NH4Cl from 10 ppm to 200 ppm resulted in a significant decline in biofilm yield from 0.012 ± 0.0006 mg/mg TC removed to 0.0047 ± 0.0004 mg/mg TC removed (p < 0.001). This suggests that there was an anabolic mechanism which resulted in increased biofilm growth under NH4+ limiting conditions. Similar results were obtained by Thompson et al. who showed that C:N:P ratios with smaller amounts of nitrogen resulted in

TOC analyses Total carbon was measured using a Shimadzu TOC-VCPH. When measuring the total carbon within the bulk fluid of the reactors, samples were centrifuged to remove biomass. The resulting supernatant was diluted 10×, and sonicated for 2 m using a Branson 3510 sonicator set at 41.5 KHz before being injected into the instrument. 3

New BIOTECHNOLOGY xxx (xxxx) xxx–xxx

S. Singh et al.

Fig. 2. A-Effects of NH4Cl stimulation on biofilm development and total carbon removal in reactors treating the defined artificial metalworking fluid. B-Effect of NH4Cl addition on biofilm yield in reactors. Increased NH4Cl levels resulted in lower amounts of biofilm biomass and total carbon removal within the reactors. Error bars represent standard deviation of triplicate measurements.

had no effect or even an inhibitory effect on others. As the bio-treatment process progressed, a distinct oil-layer began to form on the top of the bioreactor. This suggested that a possible mechanism for removal within the bioreactors was oil/water separation induced by the biodegradation of the surfactants within the metalworking fluid. If this mechanism was largely responsible for carbon removal, then the addition of an oil-emulsifying reagent to the reactor would result in the re-dispersion of removed carbon. Fig. 4 shows the effects of adding 1 mL of 10 wt% Triton X-100 to the bioreactors after a single week-long cycle. The addition of the emulsifier resulted in the redispersion of a large amount of the removed carbon. While there was a large significant difference (32.8%) in carbon removal between the 10 ppm and 200 ppm reactors before re-emulsification, the difference was much smaller (only 3.7%) after re-emulsification. This suggests that the addition of NH4Cl has a significant impact on the oil/water separation mechanism. This could be inhibition of surfactant degradation, or resulting in stimulated production of bio-surfactants [26,27]. Investigations on the influences that NH4Cl stimulation has on the

more biofilm growth of Enterobacter cloacae and Citrobacter freundii [18]. Punal et al. also showed that nitrogen limitation led to increased cell attachment rates in an anaerobic upflow sludge blanket reactor [17]. In addition to biofilm decline, NH4Cl stimulation also resulted in a significant decline in total carbon removal performance. Surprisingly, the total carbon removal efficiency declined from 72.4 ± 2.6% to 44.6 ± 1.4% (p < 0.01) when the NH4Cl concentration increased from 10 ppm to 200 ppm. This was not due to a toxic effect (the number of colony forming units suspended in the bulk liquid was much greater in the 200 ppm condition than the 10 ppm condition- data not shown. Furthermore, as shown in a later section, NH4Cl stimulation accelerated the removal of some of the constituents of the metalworking fluid). Fig. 3 shows the kinetics of carbon removal. While the addition of NH4Cl resulted in a significant decline in overall performance, it resulted in an increase in the rate of carbon removal during the early stages of treatment. This suggests that NH4Cl may have stimulated the removal of some components of the metalworking fluid, but may have

Fig. 3. Kinetics of total carbon removal as a function of NH4Cl stimulation. Data is presented for cycle 1 of the 2 cycle batch process. Increased levels of NH4Cl lowered the rate of total carbon removal within the reactors. Error bars represent standard deviation of triplicate measurements.

4

New BIOTECHNOLOGY xxx (xxxx) xxx–xxx

S. Singh et al.

Fig. 4. Extents of total carbon removal within a metalworking fluid bioreactor before and after the addition of the oil-emulsifying reagent Triton X-100. Addition of Triton X-100 causes re-dispersion of removed carbon within the bioreactors. Error bars represent standard deviation of triplicate measurements.

degradation of metalworking fluid constituents are provided in the Supplementary information and are discussed in the final section. Briefly, those results showed that NH4Cl stimulation did not inhibit the degradation of surfactants (tall oil amides), but resulted in the re-dispersion of other components. This suggests that the stimulated production of bio-surfactants may be the more likely reason for the observed behaviour.

research has shown that KH2PO4 limitation may enhance specific types of EPS production [28,29] or may promote attachment of bacteria on surfaces [30]. Both of these mechanisms would result in a relative increase in biofilm production. However, having both nutrients present in relatively large amounts quenched biofilm formation in the reactors. The effects of KH2PO4 stimulation on reactor performance in both low and high NH4Cl reactors are presented in Fig. 6A. It can be seen that there was a significant increase in carbon removal performance with KH2PO4 for both the low (p < 0.01) and high (p < 0.01) NH4Cl conditions. An addition of 25 ppm of KH2PO4 to the 10 ppm NH4Cl reactor enhanced total carbon removal from 38.7 ± 0.9% to 80.9 ± 0.8%. An addition of 50 ppm KH2PO4 to the 200 ppm NH4Cl reactor increased the total carbon removal from 24.4 ± 2.7% to 50.8 ± 2.9%. This behaviour was anticipated since the artificial metalworking fluid wastewater was KH2PO4 deficient, which was representative of real metalworking fluid wastewaters [8]. Similar results of KH2PO4 improving the treatment efficiency of bioreactors have been reported [8,31]. The trend of increasing NH4Cl concentration reducing total carbon removal performance was consistent with previous findings. Increasing KH2PO4 increases carbon removal efficiency since providing a phosphorus source allows for a greater extent of microbial growth [32]. It was observed that the number of colony forming units present in both the 10 ppm and 200 ppm NH4Cl reactors increased with increasing KH2PO4 concentration (data not shown). This suggested that while KH2PO4 addition quenched biofilm growth in the 200 ppm NH4Cl condition, it stimulated the production of cells in the suspended state. This could explain why there was a carbon removal increase in the 200 ppm NH4Cl reactor, even though biofilm biomass was reduced. Fig. 6B shows that the yield in the reactor was maximised when either NH4Cl or KH2PO4 was limited. Limiting both of the nutrients or having both of the nutrients present in an excess amount significantly

Phosphate requirements Having investigated the effects of varying the concentration of NH4Cl in the reactor, this section proceeds to investigate the effects of varying the concentration of KH2PO4. The effects of KH2PO4 addition on conditions using 10 ppm and 200 ppm NH4Cl addition are presented in Fig. 5. Systems stimulated with 10 ppm NH4Cl showed different trends from those stimulated with 200 ppm. For 10 ppm NH4Cl systems, the addition of KH2PO4 resulted in a significant increase in biofilm biomass production (p < 0.01). Specifically, reactors stimulated with 100 ppm of KH2PO4 produced a biofilm biomass of 60.0 ± 6.9 mg as compared to reactors with no KH2PO4, which had only 7.4 ± 0.4 mg (p < 0.01). This suggests that at relatively low levels of NH4Cl stimulation, KH2PO4 addition results in growth stimulation. This was consistent with the previous results, but shows further that limiting both nutrients did not stimulate biofilm growth. The results indicate that at relatively high levels of NH4Cl stimulation, KH2PO4 addition leads to a significant decline in biofilm growth (p < 0.05). In reactors stimulated with 200 ppm NH4Cl, reactors containing no KH2PO4 had 28.3 ± 2.3 mg of biofilm biomass, which was significantly more than the 5.0 ± 3.1 mg of biomass found in reactors containing 100 ppm of KH2PO4 (p < 0.01). This suggests that KH2PO4 limitation may be a means to enhance biofilm growth. Previous

Fig. 5. Biofilm dry weight as a function of KH2PO4 addition to reactors treating artificial metalworking fluid. Reactors were further stimulated with either 10 ppm or 200 ppm NH4Cl. Results showed that biofilm growth is optimised under NH4Cl limiting conditions. Error bars represent standard deviation of triplicate measurements.

5

New BIOTECHNOLOGY xxx (xxxx) xxx–xxx

S. Singh et al.

Fig. 6. A-Total carbon removal performance as a function of KH2PO4 addition. B-Biofilm yield obtained in MWF reactors as a function of KH2PO4 addition. Results are given for 10 ppm and 200 ppm NH4Cl stimulation conditions. Results show that yields are maximised in nutrient limiting conditions. Error bars represent standard deviation of triplicate measurements.

components of the metalworking fluid was investigated and are presented within the Supplementary data. The artificial metalworking fluid contains diethylene glycol butyl ether (DGBE), saturated and unsaturated fatty amides (from the tall oil that is used in the formulation), diisopropanolamine (DIPA), naphthenic mineral oil (NMO) and sodium sulfonate. Kinetics for DGBE, the amides, and DIPA are presented in the Supplementary data. Within Fig. S1, it is shown that NH4Cl stimulation resulted in the accelerated removal of DGBE, and saturated amides such as Palmitamide-DIPA and Stearamide-DIPA. However, in Fig. S2, it is shown that NH4Cl stimulation resulted in the inhibition of diisoproponolamine removal. This suggests that NH4Cl stimulation enhances the removal of some components in the metalworking fluid, and limits the removal of others. A possible explanation for the inhibition trend observed can be found by considering that micro-organisms use organic amines (such as DIPA) as an NH4+ source when no other alternative is present. By stimulating the metalworking fluid with NH4Cl, the need for organic amine utilisation was suppressed. Fig. S4 shows that stimulating the metalworking fluid with NH4Cl resulted in the re-dispersion of recalcitrant unsaturated amides into the bulk liquid. This supports the hypothesis used to explain the observations of Fig. 4, namely that bio-surfactants were being produced in reactors stimulated with NH4Cl. Biosurfactant production would result in the re-dispersion of components that are removed due to oil/water separation (such as recalcitrant unsaturated fatty amides). This does not occur in reactors with limited NH4Cl stimulations. In these reactors, the unsaturated fatty amides are removed from the bulk, and are not redispersed. Fig. S3 further illustrates this by showing the extents of removal before and after the addition of the oil-emulsifying reagent. Emulsifier addition resulted in the re-dispersion of unsaturated amides, but not saturated amides, DGBE or DIPA. These results thus show that unsaturated amides are removed through oil/water separation.

reduced the yield (p < 0.01 for both stimulatory conditions). Thus nutrient limitation may be an effective strategy that can be utilised for the development of fixed-film reactors treating metalworking fluids. Several studies have shown how nitrogen limitation and phosphorus limitation may be used to stimulate biofilm growth [17,18,30,33]. Nutrient limitation may have created a stress response that led to more biofilm development [21] or may have stimulated EPS production as a storage mechanism for the component in excess [34]. The results indicate that fixed-film reactors for the treatment of metalworking fluids are best developed with both KH2PO4 stimulation, and with NH4Cl limitation. Practitioners who wish to employ the fixedfilm bio-treatment process may use this information to simultaneously maximise the treatment performance and the yield of biofilm obtained in the reactor. The results of this investigation were obtained using a mixed consortium of microbes that had been acclimated to treating metalworking fluids. Should a defined or uninitiated consortium be utilised, it is advised that similar recommendations of this report be applied at a bench scale before being rolled out to larger systems. Dynamics of metalworking fluid component removal as a function of NH4Cl concentration In Fig. 3, it was shown that NH4Cl stimulation resulted in a significant decline in the total carbon removal efficiency of the metalworking fluid. NH4Cl is a common stimulant that is added to wastewater treatment processes and thus this effect was an unexpected one. A possible explanation was that addition of NH4Cl resulted in the production of biosurfactants, which resulted in the re-dispersion of removed carbon in metalworking fluid waste. In order to determine if NH4Cl stimulation had any stimulatory or inhibitory effect of degradation, a chemical analysis of the metalworking fluid during treatment was required. The effects of NH4Cl on the removal of the organic 6

New BIOTECHNOLOGY xxx (xxxx) xxx–xxx

S. Singh et al.

particulate biofilm reactors. J Biotechnol 2000;80:1–33. [6] Ødegaard H. The moving bed biofilm reactor. Water Environ Eng Reuse Water 1999;25:250–305. [7] Das N, Basak LVG, Salam JA, Abigail EA. Application of biofilms on remediation of pollutants – an overview. J Microbiol Biotechnol Res 2012;2:783–90. [8] Cheng C, Phipps D, Alkhaddar RM. Treatment of spent metalworking fluids. Water Res 2005;39:4051–63. [9] Thill PG, Ager DK, Vojnovic B, Tesh SJ, Scott TB, Thompson IP. Hybrid biological, electron beam and zero-valent nano iron treatment of recalcitrant metalworking fluids. Water Res 2016;93:214–21. [10] Jagadevan S, Graham NJ, Thompson IP. Treatment of waste metalworking fluid by a hybrid ozone-biological process. J Hazard Mater 2013;244:394–402. [11] Dong Z, Lu M, Huang W, Xu X. Treatment of oilfield wastewater in moving bed biofilm reactors using a novel suspended ceramic biocarrier. J Hazard Mater 2011;196:123–30. [12] Borghei SM, Hosseini SH. The treatment of phenolic wastewater using a moving bed biofilm reactor. Process Biochem 2004;39:1177–81. [13] Li H, Han H, Du M, Wang W. Removal of phenols, thiocyanate and ammonium from coal gasification wastewater using moving bed biofilm reactor. Bioresour Technol 2011;102:4667–73. [14] Tziotzios G, Teliou M, Kaltsouni V, Lyberatos G, Vayenas DV. Biological phenol removal using suspended growth and packed bed reactors. Biochem Eng J 2005;26:65–71. [15] Jefferson B, Burgess JE, Pichon A, Harkness J, Judd SJ. Nutrient addition to enhance biological treatment of greywater. Water Res 2001;35:2702–10. http://dx. doi.org/10.1016/S0043-1354(00)00553-4. [16] Ammary BY. Nutrients requirements in biological industrial wastewater treatment. Afr J Biotechnol 2004;3:236–8. [17] Puñal A, Trevisan M, Rozzi A, Lema J. Influence of C:N ratio on the start-up of upflow anaerobic filter reactors. Water Res 2000;34. [18] Thompson LJ, Gray V, Lindsay D, Von Holy A. Carbon:nitrogen:phosphorus ratios influence biofilm formation by Enterobacter cloacae and Citrobacter freundii. J Appl Microbiol 2006;101:1105–13. [19] Miqueleto AP, Dolosic CC, Pozzi E, Foresti E, Zaiat M. Influence of carbon sources and C/N ratio on EPS production in anaerobic sequencing batch biofilm reactors for wastewater treatment. Bioresour Technol 2009;101:1324–30. [20] Zhang W, Seminara A, Suaris M, Brenner MP, Weitz DA, Angelini TE, et al. Nutrient depletion in Bacillus subtilis biofilms triggers matrix production. New J Phys 2014:16. [21] Jefferson KK. What drives bacteria to produce a biofilm? FEMS Microbiol Lett 2004;236:163–73. [22] van der Gast CJ, Thompson IP. Effects of pH amendment on metal working fluid wastewater biological treatment using a defined bacterial consortium. Biotechnol Bioeng 2004;89:357–66. [23] Childers J. The chemistry of metalworking fluids. In: Byers J, editor. Metalwork Fluids. 2nd ed.Taylor and Francis Group; 2006. p. 127–46. [24] EAI Corporation. Preparation of Synthetic Tapwater for the JSAWM Program n.d. http://studylib.net/doc/7577859/preparation-of-synthetic-tapwater-for-thejsawm-program1. [25] Sinjewel A, Swart EL, Lingeman H, Wilhelm AJ. LC determination of propylene glycol in human plasma after pre-column derivatization with benzoyl chloride. Chromatographia 2007;66:103–5. [26] Md F. Biosurfactant: production and application. J Pet Environ Biotechnol 2012;3(124). http://dx.doi.org/10.4172/2157-7463.1000124. [27] Mulligan CN, Gibbs BF. Correlation of nitrogen metabolism with biosurfactant production by Pseudomonas aeruginosa. Appl Environ Microbiol 1989;55:3016–9. [28] Mendrygal KE, Gonzalez JE. Environmental regulation of exopolysaccharide production in sinorhizobium meliloti. J Bacteriol 2000;182:599–606. [29] Fang W, Hu JY, Ong SL. Influence of phosphorus on biofilm formation in model drinking water distribution systems. J Appl Microbiol 2009;106:1328–35. [30] Xu J, Kim J, Danhorn T, Merritt PM, Fuqua C. Phosphorus limitation increases attachment in Agrobacterium tumefaciens and reveals a conditional functional redundancy in adhesin biosynthesis. Res Microbiol 2012;163:674–84. [31] Schreyer HB, Coughlin RW. Effects of stratification in a fluidized bed bioreactor during treatment of metalworking wastewater. Biotechnol Bioeng 1999;63:129–40. [32] von Sperling M. Basic principles of wastewater treatment. IWA Publishing; 2007. [33] Liu J, Prindle A, Humphries J, Gabalda-sagarra M, Asally M, Lee DD. Metabolic codependence gives rise to collective oscillations within biofilms. Nature 2015;523:550–4. [34] Flemming H-C, Wingender J. The biofilm matrix. Nat Rev 2010;8:623.

Saturated amides, DGBE and DIPA were not re-dispersed, suggesting that these components were removed due to mechanisms other than oil/water separation. Conclusions The role of NH4Cl and KH2PO4 associated with biofilm development and reactor performance were investigated.

• Biofilm



• •

yields were sensitive to both NH4Cl and KH2PO4 concentrations and were found to be significantly higher when only one of the components was present in limiting quantities. When both components were deficient in the metalworking fluid, significantly lower biofilm yields were obtained. Maximum biofilm yields were obtained in NH4+ deficient conditions. For the first time, we have shown how NH4Cl may have dual effects on pollutant removal in metalworking fluids. While NH4Cl increased the rate of removal of saturated amides and DGBE, it was shown to inhibit the removal of diisopropanolamine in the reactor. This suggested that diisopropanolamine was utilised as a nitrogen source under conditions of NH4+ limitation. Reactor total carbon removal performance was shown to increase with increases in KH2PO4 concentration and to decrease with increases in NH4Cl concentrations. While the removal of carbon in the bulk was shown to decline with increases in NH4Cl concentration, it is shown for the first time that the absolute removal of carbon through mineralization was similar. This suggests that addition of NH4Cl led to more stable emulsions in the reactor, possibly due to inducing the production of biosurfactants.

Acknowledgements We acknowledge the support of a collaborative funding scheme between the National University of Singapore, Peking University and the University of Oxford (Singapore-Peking-Oxford Research Enterprise). We thank the University of Oxford Department of Chemistry, especially Dr James Wickens, for their help with characterising the tall oil fatty amides used for creating the metalworking fluid emulsions. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.nbt.2017.09.002. References [1] Leslie CP, Grady J, Daigger GT, Love NG, Filipe DM. Biological Wastewater Treatment. 3rd ed. IWA Publishing; 2011. [2] Dunne WM. Bacterial adhesion seen any good biofilms lately? Clin Microbiol Rev 2002;15:155–66. [3] O’toole G, Kaplan HB, Kolter R. Biofilm formation as microbial development. Annu Rev Microbiol 2000;54:49–79. [4] Donlan RM. Biofilms: microbial life on surfaces. Emerg Infect Dis 2002;8:881–90. [5] Nicolella C, Van Loosdrecht MCM, Heijnen JJ. Wastewater treatment with

7