Calculating specific denitrification rates in pre ...

Environmental Technology

ISSN: 0959-3330 (Print) 1479-487X (Online) Journal homepage: http://www.tandfonline.com/loi/tent20

Calculating specific denitrification rates in predenitrification by assessing the influence of dissolved oxygen, sludge loading and mixed-liquor recycle Massimo Raboni, Vincenzo Torretta, Paolo Viotti & Giordano Urbini To cite this article: Massimo Raboni, Vincenzo Torretta, Paolo Viotti & Giordano Urbini (2014) Calculating specific denitrification rates in pre-denitrification by assessing the influence of dissolved oxygen, sludge loading and mixed-liquor recycle, Environmental Technology, 35:20, 2582-2588, DOI: 10.1080/09593330.2014.913690 To link to this article: http://dx.doi.org/10.1080/09593330.2014.913690

Published online: 22 May 2014.

Submit your article to this journal

Article views: 141

View related articles

View Crossmark data

Citing articles: 8 View citing articles

Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=tent20 Download by: [National Research Centre for Citrus ]

Date: 08 August 2017, At: 01:19

Environmental Technology, 2014 Vol. 35, No. 20, 2582–2588, http://dx.doi.org/10.1080/09593330.2014.913690

Calculating specific denitrification rates in pre-denitrification by assessing the influence of dissolved oxygen, sludge loading and mixed-liquor recycle Massimo Rabonia , Vincenzo Torrettab∗ , Paolo Viottic and Giordano Urbinib a Department

of Chemistry, Materials and Chemical Engineering, Politecnico di Milano, Via L. Mancinelli, 7, I-20133 Milan, Italy; of Biotechnologies and Life Sciences, Insubria University of Varese, Via G.B. Vico, 46, I-21100 Varese, Italy; c Department of Civil and Environmental Engineering, University of Roma La Sapienza, Via Eudossiana 18, I-00184 Rome, Italy b Department

Downloaded by [National Research Centre for Citrus ] at 01:19 08 August 2017

(Received 5 November 2013; final version received 6 April 2014 ) This article presents the results of an experimental study on the correlation among the specific denitrification rate (SDNR), the dissolved oxygen concentration (DO), the F:M ratio (F:M) and the mixed-liquor (ML) recycle in the pre-denitrification reactors fed by domestic sewage. The experimental curves reveal a 28.8–32.0% reduction in the SDNR at 20◦ C (SDNR20◦ C ) with DO equal to 0.1 mgO2 L−1 and F:M in the range 0.2–0.4 kgBOD5 kgMLVSS−1 d−1 . The SDNR reduction increases to 50.0–55.9% with DO = 0.3 mgO2 L−1 . A mathematical correlation of these results and an equation for calculating SDNR20◦ C as function of the F:M as well as the average DO and BOD5 in the total flow rate fed in the denitrification stage are proposed. The conducted experience gives useful suggestions for practical usage, in particular regarding the denitrification reactor design, and represents a good starting point for future applications with the aim to optimize the biological process in domestic sewage treatment plants. Keywords: conversion yield; denitrification rate; mathematical correlation; nitrogen control; plant design, wastewater treatment

Introduction Although unwanted, dissolved oxygen concentration (DO) is always present in the biological pre-denitrification reactors. The daily average concentrations measured in realscale plants are mostly in the range 0.2–0.4 mgO2 L−1 , with much higher peak values during the day, especially in small sewage treatment plants characterized by strong fluctuations in flow and quality.[1,2] The DO concentrations within anoxic reactors are the result of two opposing factors: (i) the intake of oxygen loads associated with the raw sewage, the sludge recirculation and, above all, the mixed-liquor (ML) recycling and (ii) the oxygen consumption due to heterotrophic bacteria. The sizing of the biological pre-denitrification reactor is normally based on the denitrification rate (rDEN , gNO3 –N kgMLVSS−1 h−1 ) [3] assuming zero-order kinetics (in relation to both nitrate concentration, NO3 –N, and organic substrate) and considering the significant effect of the temperature (T , ◦ C): (rDEN )T = (rDEN )20◦ C · θ T −20 ,

(1)

where (rDEN )20◦ C is the denitrification rate at 20◦ C (standard values: 2.9–3.0 gNO3 –N kgMLVSS−1 h−1 ) [4–13] and θ is the temperature coefficient (θ = 1.026 [14]; θ = 1.07 [15]). ∗ Corresponding

author. Email: [email protected]

© 2014 Taylor & Francis

The possible inhibitory effects of DO on the denitrification process kinetics were postulated in 1975 by US-EPA.[4] The following US-EPA reports [13,14] highlighted this effect by inserting the inhibition factor K0 /(K0 + DO) in the expression of rDEN :

rDEN

1 − 1.42Y K ·S ·X = · 2.86 KS + S NO3 − N K0 · · · η, KN + NO3 − N K0 + DO

(2)

where rDEN is the denitrification rate, that is the NO3 –N removal by dissimilation (mgNO3 –N L−1 h−1 ); Y is the heterotrophic bacteria synthesis yield (mgVSS mg−1 substrate consumed); K is the maximum specific rate of substrate utilization (h−1 ); X is the biomass concentration (mgMLVSS L−1 ); S is the soluble degradable substrate concentration (mg L−1 ); KS is the substrate utilization half-velocity coefficient (mg L−1 ); NO3 –N is the nitrate concentration as N (mgNO3 –N L−1 ); KN is the nitrate half-velocity coefficient (mgNO3 –N L−1 ); K0 is the DO inhibition constant for nitrate reduction (mgO2 L−1 ); and η is the fraction of heterotrophic bacteria that use nitrate in lieu of oxygen (dimensionless).

Environmental Technology Equation (2) was extended considering the complete denitrification in both the dissimilation and assimilation (cell synthesis) processes as follows [15]:


rDEN

1 − 1.42Y K ·S ·X NO3 − N = · · 2.86 KS + S KN + NO3 − N K0 NO3 − N · · η + K0 + DO KN + NO3 − N K0 1.42 · (3) · · Kd · X · η, K0 + DO 2.86

where Kd is the endogenous decay coefficient (h−1 ). Dawson and Murphy highlighted the DO inhibition on denitrification at concentrations of 0.20 mgO2 L−1 .[16] K0 is considered a variable in a wide range (0.02– 0.20 mgO2 L−1 ) depending on the floc size and structure. [15] In any case, the mere presence of 0.2 mgO2 L−1 theoretically decreases rDEN up to 40% compared to the maximum value in the absence of inhibition.[17,18] Other studies reported the effect of inhibition.[19–22] Oh and Silverstein noted a significant denitrification rate reduction (35%) at DO of only 0.09 mg L−1 .[20,21] The specific denitrification rate (SDNR, kgNO3 – N kgMLVSS−1 d−1 ) is introduced as the ratio between the removed nitrite and the amount of biomass in the denitrification reactor: SDNR =

Q · NO3 − N , VDEN · X

(4)

where Q is the sewage flow rate (m3 d−1 ); NO3 –N is the NO3 –N removal in the denitrification tank (kgNO3 – N m−3 ); VDEN is the denitrification reactor volume (m3 ); and X is the biomass concentration in the denitrification reactor (kgMLVSS m−3 ). For the practical calculation of the denitrification reactor volume, a semi-empirical relationship which correlates SDNR at 20◦ C (SDNR20◦ C ) to the sludge loading in denitrification, that is the food-to-microorganism ratio (F:M; kgBOD5 applied kgMLVSS−1 d−1 ) has been proposed [15, 23,24]: SDNR20◦ C = 0.029 + 0.03 · F:M.

(5)

US-EPA proposed a modified version of Equation (5) with a correction factor applied to the F:M ratio just to take into account the deviation of the biomass active fraction in the ML from the reference value of 0.3.[17] SDNR observed in the pre-anoxic reactors of fullscale installations ranges from 0.04 to 0.42 kgNO3 – N kgMLVSS−1 d−1 ,[15,24,25] while US-EPA [17] reported a more limited range (0.05–0.15 kgNO3 –N kgMLVSS−1 d−1 at 20◦ C).

2583

As the temperature strongly affects the denitrification kinetics, the well-known correlation was proposed: SDNRT = SDNR20◦ C · θ (T −20) .

(6)

In order to consider the effects of the sludge recycle on the BOD5 and DO concentrations in the anoxic reactor, Tchobanogolous introduced a variant of Equation (5) when the ML recycle ratio (MLRR) is higher than 1 [15]: SDNR = 1 − 0.0166 · ln(F:Mb ) − 0.0078 for MLRR = 2 SDNR = 1 − 0.029 · ln(F:Mb ) − 0.012 for MLRR = 3 − 4, where F:Mb is the sludge loading in the pre-anoxic tank referred to as the active biomass (kgBOD kgMLVSS active−1 d−1 ). As an alternative to the use of SDNR, more detailed calculations for the anoxic reactor volume are now commonly performed with models that simulate the fate of the degradable particulate, the soluble substrate, the active biomass and the nitrogen species considering the effect of temperature and DO concentration.[26–30] Both of the above two calculation methods have limitations. In fact, the modelling methods are excessively based on theoretical background while the empirical SDNR-based methods approximate the DO inhibitory influence and the ML recycle effect. The aim of this research is to develop a practical calculation of SDNR for the design of pre-denitrification reactors fed by domestic sewage. The influence of DO concentration, sludge loading and MLRR in denitrification on SDNR has been investigated during an experimental six-month study carried out on an activated sludge pilot plant with pre-denitrification. Materials and methods Pilot plant description The research was based on the use of an activated sludge pilot plant (Figure 1) with a biological pre-denitrification (DEN) and a biological oxidation-nitrification (OX-NIT) stage followed by a final sedimentation (SED). The pilot plant was fed by pre-treated (screening and aerated grit chamber) sewage coming from 80,000 inhabitants’ basin in northern Italy.[31,32] Aeration of DEN and OX-NIT is guaranteed by four slow vertical-axis mixers (power input: 11 W m−3 ) and a micro-bubble aeration system, respectively. The features of the pilot plant include: • DEN tank: volume (VDEN ) 10 m3 ; liquid height 1.8 m; • OX-NIT tank: volume 20 m3 ; liquid height 1.8 m; • SED: diameter 2 m; volume 6 m3 ;

2584


Figure 1.

M. Raboni et al.

The pilot plant layout.

• sewage flow rate (Qsewage ): 1.5–2.5 m3 h−1 (variable throughout the experiment); • ML recycle flow rate (QML ): 2–6 m3 h−1 (variable throughout the experiment); • sludge recycle flow rate (q): 1.5 m3 h−1 . Pilot plant operating conditions and testing methods The pilot plant ran for a continuous period of six months. Different F:M ratios in the denitrification stage were tested (range: 0.2–0.4 kgBOD5 kgMLVSS−1 d−1 ). With the different F:M ratio conditions, variations of DO (increasing the level of aeration in OX-NIT) and QML in nitrification were made. The following analytical parameters were measured: • BOD5 , COD, TKN and NO3 –N (automatic daily average samplings) in the: (a) pre-treated sewage entering the pilot plant; (b) total flow rate entering DEN; (c) flow rate at the end of DEN and in the middle of OX-NIT (filtered samples); (d) pilot plant effluent; • ML volatile suspended solids (MLVSS) and ML suspended solids (MLSS) in DEN and OX-NIT (manual sampling); • temperature in DEN and OX-NIT (fixed probe with continuous sampling; Figure 1); • DO concentration in pre-treated sewage, sludge recycle, ML recycle, total flow rate entering in DEN, inside DEN (four locations) and inside OX-NIT (two locations) with fixed probe and continuous registration (Figure 1); • pH in the pre-treated sewage as well as at the end of DEN and OX-NIT (fixed probe and continuous registration; Figure 1). The mean quality of the domestic sewage fed to the pilot plant was: • BOD5 = 135 mg L−1 (standard deviation (SD) = 25 mg L−1 );

Table 1. Operating condition of the pilot plant in biological pre-denitrification (DEN) and biological oxidation-nitrification (OX-NIT) during the experiments. DEN Parameter MLVSS MLSS DO T pH

Unit kgMLVSS m−3 kgMLSS m−3 mgO2 L−1 ◦C –

M

OX-NIT SD

M

2.53 0.12 2.53 3.95 0.14 3.95 0.1–1.0a 2.2 15.0 0.95 15.0 7.85b 0.10b 7.54

SD 0.12 0.14 0.2 0.92 0.08

Note: M , mean; SD, standard deviation. a Variable throughout the experiment. b At the end of DEN stage.

• • • •

COD = 295 mg L−1 (SD = 45 mg L−1 ); TKN = 27.2 mg L−1 (SD = 5.6 mg L−1 ); SS = 188 mg L−1 (SD = 29 mg L−1 ); pH = 7.45 (SD = 0.09).

The operating conditions during the pilot plant functioning are listed in Table 1. Sampling and analyses were carried out in compliance with official methods.[33] Fixed immersed electrochemical probes were used for DO measurements (resolution: 0.01 mgO2 L−1 ; automatic calibration and temperature compensation). Results and discussion SDNR as a function of dissolved oxygen and the F:M ratio in denitrification The experiment results shown in Figure 2 confirm the strong dependence of SDNR, computed using Equation (4), on the DO concentration. This dependence is at the same time influenced by the F:M relative to the volume of denitrification. Within the experimented F:M ratio range, the operating values of the real-scale denitrification plants are normally found. The curves representing the experimental SDNR have a decreasing trend, which seems to be exponential,

Environmental Technology

2585

dissolved oxygen (SDNRASS het ): SDNRASS = SDNRASS den + SDNRASS het .

(10)


The first contribution, due to denitrifying bacteria, is dominant in the absence of DO. Introducing in Equation (9) the BOD-to-Nitrogen removal ratio for dissimilative denitrification (BOD5 /NO3 − N = 4 [17]) and applying the definition of RDEN , Equation (8) becomes:

15◦ C

Figure 2. Specific denitrification rate at (SDNR15◦ C ) as a function of the dissolved oxygen concentration (DO) in the denitrification reactor at two different values of the F:M ratio (dashed and dash-dotted lines represent the curves obtained with regression analysis of experimental data regarding F:M at 0.2 and 0.4 kgBOD5 kgMLVSS−1 d−1 , respectively). The grey band represents the 95% confidence interval.

with a strong initial gradient (in the range DO = 0.1 − 0.3 mgO2 L−1 ) that tends to fade towards an asymptotic behaviour in the presence of high DO concentrations. Increasing the DO concentration from 0.1 to 0.3 mgO2 L−1 , there is a SDNR drop of 31.9% for F:M = 0.4 kgBOD5 kgMLVSS−1 d−1 . A more pronounced behaviour (40.5% SDNR drop) was found for F:M = 0.2 kgBOD5 kgMLVSS−1 d−1 . Moreover, the two curves tend to diverge with increasing DO. A sound theoretical basis can be ascribed to this experimental result. Indeed, SDNR can be represented as the sum of two contributions: SDNR = SDNRDISS + SDNRASS ,

0.05Q · BOD5 X · VDEN

(11)

SDNRASS het = 0.05F:M · ηBOD ·

DO , 0.2 + DO

(12)

where ηBOD is the BOD removal yield (ηBOD = 0.90–0.95 for F:M ranging from 0.4 to 0.2 kgBOD5 kgMLVSS−1 d−1 ) and DO/(0.2+DO) is the so-called ‘switching function’ which describes the DO influence on ηBOD . The switching function is necessary because the dissolved oxygen is a limiting factor which affects BOD removal when it has concentrations below 2 mgO2 L−1 .[17] Substituting Equations (8) and (10)–(12) in Equation (7), we obtain:

K0 SDNR = 1.2RDEN · K0 + DO + 0.05F:M · ηBOD ·

DO . 0.2 + DO

(13)

(8)

where RDEN represents the specific rate of dissimilative denitrification with zero-order kinetics (RDEN = Q · NO3 − −1 N · X −1 VDEN ), while SDNRASS =

0.2Q · NO3 − N = 0.2RDEN . X · VDEN

The second contribution, due to heterotrophic bacteria, is dominant in the presence of high DO concentration. It can be represented by the term:

(7)

where SDNRDISS and SDNRASS represent the dissimilative and assimilative (cell synthesis) denitrification, respectively. In the absence of inhibition factors: SDNRDISS = RDEN ,

SDNRASS den =

(9)

that is the SDNRASS is estimated considering a NO3 –N consumption for a biological synthesis equal to 5% of the removed BOD5 ( BOD5 ). More specifically, assimilative denitrification (cell synthesis) is given by two contributions due to the nitrogen consumption related to the growth of denitrifying bacteria (SDNRASS den ) and heterotrophic bacteria that use the

At a temperature of 20◦ C, assuming RDEN 20◦ C = (rDEN )20◦ C = 3.0 gNO3 − N kgMLVSS−1 h−1 = 0.072 kg NO3 –N kgMLVSS−1 d−1 , we have:

SDNR20◦ C

K0 = 0.0864 · K0 + DO DO . + 0.05F:M · ηBOD · 0.2 + DO

(14)

Bringing the two equations to the average temperature of 15◦ C detected in the pilot study using Equation (6) and assuming typical values for parameters θ = 1.07 and

2586

M. Raboni et al.

ηBOD = 0.95, we have:

material balance of the oxygen:

SDNR15◦ C

K0 = 0.0616 K0 + DO DO + 0.00677 0.2 + DO

SDNR15◦ C = 0.0616

K0

Qtot · DO0 = Qtot · DO + RDO · VDEN · X , for F:M = 0.2, (15)

K0 + DO DO + 0.01283 0.2 + DO


where Qtot (m3 d−1 ) is the total flow rate entering the denitrification reactor (Qtot = Qsewage + QML + q); DO0 (mg L−1 ) is the DO in Qtot (DO0 = (Qi DOi )/Qtot ); and RDO (mgO2 kgMLVSS−1 d−1 ) is the average specific oxygen consumption rate in denitrification (RDO = −1 Qtot DO VDEN · X −1 ). Rearranging Equation (18) we obtain:

for F:M = 0.4. (16)

Equations (15) and (16) converge for DO = 0 mgO2 L−1 and tend to diverge for increasing values of DO, exactly as in the experimental curves (Figure 3). These equations are in a good agreement with the experimental data for K0 = 0.18 mgO2 L−1 , as shown in Figure 3. In denitrification tanks, an average DO concentration of at least 0.2–0.3 mgO2 L−1 is realistic. Considering DO = 0.25 mgO2 L−1 and ηBOD = 0.95 and applying Equation (14), we would have: SDNR20◦ C = 0.036 + 0.026 · F:M.

(17)

This result is not dissimilar from Equation (5), but is a consequence of a general equation that gives a more precise calculation of SDNR.

Estimation of the average DO concentration expected in denitrification The possibility of calculating the expected DO average concentration in denitrification has been evaluated via a

Figure 3. Comparison between theoretical curves (calculated with the proposed equation) and experimental data fitting curves at two different F:M ratios.

(18)

DO = DO0 −

RDO · VDEN · X . Qtot

(19)

Applying the definition of the F:M ratio (F:M = −1 Qtot B0 VDEN · X −1 ), Equation (19) becomes: DO = DO0 −

RDO · B0 , F:M

(20)

where B0 is the BOD5 concentration entering the denitrification stage. RDO is influenced by the concentration of both the DO (which depends on the oxygen load carried by the flow rates) and the BOD5 (which depends on the wastewater strength). Figure 4 shows the experimental values of RDO as a function of the DO0 at different B0 . Maintaining the DO concentration at about 2 mgO2 L−1 in OX-NIT, it is reasonable to expect that DO0 is in the range 1.0–1.5 mgO2 L−1 (mainly depending on QML ). Therefore, RDO is on average in the range of 3– 6 mgO2 kgMLVSS−1 d−1 . In the range of DO0 considered, RDO can be represented by a first-order equation: RDO = K · DO0 ,

(21)

Figure 4. Average rate values of the oxygen-specific consumption in the denitrification reactor (RDO ) as a function of the average DO in the total flow rate fed into the reactor (DO0 ), and for different values of BOD5 in the same total flow rate (B0 ).

Environmental Technology where K is the first-order constant, depending on B0 as follows: K = 0.11

for

B0 = 20 − 29 mg L−1

K = 0.14

for

B0 = 30 − 49 mg L−1

K = 0.18

for

B0 = 50 − 69 mg L−1


Therefore, substituting Equation (21) in Equation (20) we obtain: K · B0 DO = DO0 · 1 − . (22) F:M Proposal for a general equation for calculating SDNR Substituting Equation (22) into Equation (14) we obtain the following equation for the calculation of SDNR20◦ C : K0 SDNR20◦ C = 0.0864 · K0 + DO0 · (1 − K · B0 /F:M) + 0.05F : M · ηBOD 1 · , 0.2/DO0 · (1 − K · B0 /F:M) + 1

2587

with DO = 0.3 mgO2 L−1 (the values of the ranges shown relate to F:M = 0.4 kgBOD5 kgMLVSS−1 d−1 and F:M = 0.2 kgBOD5 kgMLVSS−1 d−1 , respectively). An even more general equation was developed. It provides the correlation between SDNR20◦ C and the values of dissolved oxygen and BOD5 detected in the total flow rate entering the denitrification reactor on an experimental basis, considering: • K: 0.11–0.18 mgO2 L−1 , in dependence of DO0 e B0 ; • DO0 : DO concentration in the total flow rate entering the denitrification reactor; • B0 : BOD5 concentration in the total flow rate entering the denitrification reactor. The conducted experiment can give some useful suggestion for practical usage, in particular regarding the reactor design destined to the denitrification process, and can represent a good starting point for future applications with the aim to optimize the biological process in domestic sewage treatment plants. This aspect has a great importance considering the central role of the biological stage in the overall efficiency and costs of treatment plants.

(23) where K0 is the 1.8 mgO2 L−1 ; K the 0.11–0.18 kgMLVSS mg d L−1 gNO3 –N−1 , depending on the values assumed by B0 ; and ηBOD is 0.90–0.95, depending on the values assumed by F:M. Conclusions The paper highlights the strong dependence of the SDNR on the DO and the F:M ratio in the pre-denitrification stage. The results, supported by theoretical evaluations, represent SDNR at 20◦ C with the following equation: gNO3 − N K0 SDNR20◦ C = 0.0864 gMLVSS · d K0 + DO + 0.05F:M · ηBOD DO · , (24) 0.2 + DO where K0 = 0.18 mgO2 L−1 is the dissolved oxygen (DO) inhibition constant and ηBOD is the BOD5 removal efficiency in the denitrification reactor, which depends on the F:M ratio (ηBOD = 0.90 for F:M = 0.4 kgBOD5 kgMLVSS−1 d−1 and ηBOD = 0.95 for F:M = 0.2 kgBOD5 kgMLVSS−1 d−1 ). By controlling the DO in the ML recycle, the average DO concentration in the denitrification reactor is expected on average to be in the range 0.2–0.4 mgO2 L−1 . With these data it is easy to calculate SDNR20◦ C . The equation indicates a reduction of 28.8–32.0% in the SDNR20◦ C value with DO = 0.1 mgO2 L−1 and a reduction of 50.0–55.9%

References [1] Raboni M, Torretta V, Urbini G. Influence of strong diurnal variations in sewage quality on the performance of biological denitrification in small community wastewater treatment plants (WWTPs). Sustainability. 2013;5:3679–3689. [2] Raboni M, Torretta V, Viotti P, Urbini G. Pilot experimentation with complete mixing anoxic reactors to improve sewage denitrification in treatment plants in small communities. Sustainability. 2014;6:112–122. [3] Torretta V, Ragazzi M, Trulli E, De Feo G, Urbini G, Raboni M, Rada EC. Assessment of biological kinetics in a conventional municipal WWTP by means of the oxygen uptake rate method. Sustainability. 2014;6:1833–1847. [4] United States Environmental Protection Agency. Process design manual for nitrogen control. Jefferson (MO): Water Resource Center; 1975. [5] De Fraja Frangipane E, Cossu R, Urbini G. Modern trend in plant design for the removal of nutritive substances from sewage. Proceedings of the 2nd international symposium on advanced wastewater treatment; 1978; Krakow, Poland. [6] Christensen HM, Harremoes PA. Literature review of biological denitrification of sewage. Proceedings of the international nitrogen as a water pollutant; 1975 August 18–20; Copenhagen, Denmark. [7] Christensen HM, Harremoes PA. Nitrification and denitrification in wastewater treatment. In: Mitchell, editor. Contribution to water pollution microbiology, Vol. 2; New York: Wiley; 1976. p. 71–110. [8] Sutton PM, Murphy KL, Jank BE. Nitrogen control: a basis for design with activated sludges systems. Proceedings of IAWPR conference; 1975 August 18–20; Copenhagen, Denmark. [9] Barnard JL. Biological denitrification. Water Pollut Control. 1975;74:178–186. [10] Collivignarelli C, Bertanza G. Ingegneria SanitariaAmbientale (Environmental Engineering). 1st ed. Milan, Italy: Città Studi; 2012.


2588

M. Raboni et al.

[11] Ekama GA, Marais GVR, Siebritz IP, Pitman AR, Keay GFP, Buchan L, Gerber A, Smollen M. Theory, design and operation of nutrient removal activated sludge process. Pretoria, South Africa: Water Research Commission Pretoria, South Africa Government; 1984. [12] Ekama GA, Wentzel MC. Denitrification kinetics in biological N and P removal activated sludge systems treating municipal wastewaters. Water Sci Technol. 1999;39:69–77. [13] Ekama GA. Biological nutrient removal. Reference module in earth systems and environmental sciences treatise on water science. Water Qual Eng. 2011;4:409–526. [14] United States Environmental Protection Agency. Nutrient control design manual: state of technology review. Report EPA/600/R09/012, Jefferson (MO); January 2009. [15] Tchobanogolous G, Burton FL, Stensel HD. Wastewater engineering: treatment and reuse. 4th ed. New York: McGraw Hill Ed.; 2003. [16] Dawson RN, Murphy KL. The temperature dependency of biological denitrification. Water Res. 1972;6:71–77. [17] United States Environmental Protection Agency. Nutrient control design manual. EPA/600/R10/100, August 2010, Jefferson (MO); 2010. [18] Oh J, Silverstein J. Oxygen inhibition of activated sludge denitrification. Water Res. 1999;33:1925–1937. [19] Plósz BG, Jobbágy A, Grady CPL. Factors influencing deterioration of denitrification by oxygen entering an anoxic reactor through the surface. Water Res. 2003;37:853–863. [20] Badstreet KA, Jhonson GR. Study of critical operational parameters for nitrogen reduction at municipal wastewater treatment plant. Proceedings of water environment federation 67th annual conference and exposition; October 15–19; Chicago (IL); 1994. [21] United States Environmental Protection Agency. Nitrogen control manual. EPA/625/R-93/010. Washington (DC): Office of Research and Development; 1993. [22] Burdick CR, Refling DR, Stensel HD. Advanced biological treatment to achieve nutrient removal. J Water Pollut Control Fed. 1982;54(7):1078–1086. [23] Henze M. Capabilities of nitrogen removal process from wastewater. Water Sci Technol. 1991;23:669–679.

[24] Esposito G, Fabbricino M, Lens P, Pirozzi F. Mathematical model for sizing combined nitrification and predenitrification activated sludge systems. Environ Technol. 2007;28:391–399. [25] Aytimur G, Di Palma L, Merc C. Experimental validation of a model describing the cycle of nitrogen in a step sludge recirculation activated sludge system with denitrification. Environ Technol. 2008;29:591–601. [26] Ribas F, Rodríguez-Roda I, Serrat J, Clara P, Comas J. Development and implementation of an expert system to improve the control of nitrification and denitrification in the Vic wastewater treatment plant. Environ Technol. 2008;29: 583–590. [27] Luciano A, Viotti P, Mancini G, Torretta V. An integrated wastewater treatment system using a BAS reactor with biomass attached to tubolar supports. J Environ Manage. 2012;113:51–60. [28] Viotti P, Luciano A, Mancini G, Torretta V. A wastewater treatment using a biofilm airlift suspension reactor with biomass attached to supports: a numerical model. Int J Environ Sci Technol. 2014;11(3):571–588. doi:10.1007/s13762013-0256-6 [29] Torretta V. PAHs in wastewater: removal efficiency in a conventional wastewater treatment plant and comparison with model predictions. Environ Technol. 2012;33: 851–855. [30] Torretta V, Katsoyiannis A. Occurrence of polycyclic aromatic hydrocarbons in sludges from different stages of a wastewater treatment plant in Italy. Environ Technol. 2013;34:937–943. [31] IRSA-CNR (Italian National Research Center). [cited 2013 July 10]. Available from: http://www.irsa.cnr.it/ShPage. php?lang=it&pag=metod/ [32] Henze M, Gujer W, Takashi M, van Loosdrecht M. Activated sludge models ASM1, ASM2, ASM2d and ASM3. London (UK): IWA; 2000. [33] Rada EC, Istrate IA, Ragazzi M, Andreottola G, Torretta V. Analysis of electro-oxidation suitability for landfill leachate treatment through an experimental study. Sustainability. 2013;5:3960–3975.