Effluent stream treatment in a nitrogenous fertilizer factory: An exergy analysis for process integration

ARTICLE IN PRESSPSEP-366; No. of Pages 7

EA

OEa

Ab

c

S

1

Sc(lhrctectti

o

e

0h

Process Safety and Environmental Protection x x x ( 2 0 1 3 ) xxx–xxx

Contents lists available at ScienceDirect

Process Safety and Environmental Protection

journa l h om ep age: www.elsev ier .com/ locate /ps ep

ffluent stream treatment in a nitrogenous fertilizer factory:n exergy analysis for process integration

rlando Jorqueraa,∗, Ricardo Kalida, Asher Kiperstoka, Elias Bragab,merson Andrade Salesc

Department of Environmental Engineering, Clean Technologies Network of Bahia (TECLIM), UFBA – Federal University of Bahia, Ruaristides Novis, 2, 4◦ andar, Polytechnic School, Salvador, BA 40210-630, BrazilFábrica de Fertilizantes Nitrogenados – FAFEN-BA/OT, Rua Eteno 2198, Polo Petroquímico de Camacari, BA 42810-000, BrazilDepartment of Physical Chemistry, Chemistry Institute, Federal University of Bahia, IQ/UFBA Campus Universitário de Ondina,alvador, BA 40170-290, Brazil

a b s t r a c t

The industrial processes used for the production of nitrogenous fertilizers are the main generators of reactive nitrogen

compounds, chemicals and effluents that ultimately impact the biosphere. Exergy analysis has been performed to

a nitrogen fertilizer factory in the State of Bahia, Brazil, where the Anaerobic Ammonium Oxidation (Anammox)

and other physical–chemical processes are used to partially or totally handle the feed streams normally sent to a

stripping tower.

The results showed that the combined use of physical–chemical and biological process can improve the overall

exergetic efficiency and avoid the emission of reactive compounds to the atmosphere allowing the recovery of the

condensate lost as effluent, so that it can be reincorporated in the production of steam network, increasing energy

efficiency and environmental performance of the process.

© 2013 The Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.

Keywords: Nitrogen; Ammonia; Fertilizer plant; Exergy analysis

of about 30 million tons per year of reactive N2 compounds.

. Introduction

ince the industrial revolution, anthropogenic processes havehanged the environment on a global basis. Rockström et al.2009) showed that three processes at global level have vio-ated the boundaries that define a safe operating space forumanity in relation to the earth’s system: climate change,ate of biodiversity loss and interference with the nitrogenycle due to the removal of atmospheric N2 to generate reac-ive nitrogen species for human use (ammonia, urea, nitrates,tc.). Extensive agriculture is the main vector of environmentalontamination by these species (Galloway et al., 2004), con-aminating rivers, coastal and terrestrial systems, stimulatinghe production of gases by microorganisms that also have an

Please cite this article in press as: Jorquera, O., et al., Effluent stream treatmintegration. Process Safety and Environmental Protection (2013), http://dx

mpact on global warming. For instance N2O is over 296 times

∗ Corresponding author at: Department of Environmental Engineeringf Bahia, Rua Aristides Novis N◦. 2, 4◦ andar, Polytechnique Institute, S

E-mail addresses: [email protected] (O. Jorquera), [email protected]@petrobras.com.br (E. Braga), [email protected] (E.A. Sales).

Received 8 April 2013; Received in revised form 22 July 2013; Accepted957-5820/$ – see front matter © 2013 The Institution of Chemical Engittp://dx.doi.org/10.1016/j.psep.2013.07.003

more impacting per unit of weight than CO2 (Crutzen et al.,2007).

The manufacturing processes of fertilizers fix around 120million tons of atmospheric N2 into reactive forms per year,and this amount is larger than all natural processes combined(Galloway et al., 2004). The contribution of the industrial unitstudied in this paper, a nitrogenous fertilizer factory in theState of Bahia, Brazil accounts for around 0.40% of this total.This implies the estimate of 250 fertilizer plants of the samesize in the world. To solve this excessive anthropogenic inputof reactive nitrogen, Rockström et al. (2009) have suggestedreducing of the current amount of N2 fixation by about 75% toreturn to a safe limit for humanity. This means a production

ent in a nitrogenous fertilizer factory: An exergy analysis for process.doi.org/10.1016/j.psep.2013.07.003

, Bahia Center for Clean Technologies (TECLIM), Federal Universityalvador, BA 40.210-630, Brazil. Tel.: +56 71 32839790.

(R. Kalid), [email protected] (A. Kiperstok),

23 July 2013

This analysis does not consider the process operations where

neers. Published by Elsevier B.V. All rights reserved.

dx.doi.org/10.1016/j.psep.2013.07.003

http://www.sciencedirect.com/science/journal/09575820

www.elsevier.com/locate/psep

mailto:[email protected]





dx.doi.org/10.1016/j.psep.2013.07.003


2 Process Safety and Environmental Protection x x x ( 2 0 1 3 ) xxx–xxx

Nomenclature

dEcv,sys

dtrate of exergy change in the system (kW)

Qj rate of heat transfer in the system (kW)T0 reference ambient temperature for exergy (con-

sidered here 298 K)Tj temperature of the “j” process (K)Wcv rate of energy transfer by work in the system

(kW)p0 reference pressure (1 bar)dVcv

dtvolume change of the system (L)

miefi rate of exergy change related to the inlet massflow (kW)

meefe rate of exergy change related to the outlet massflow (kW)

ECh rate of chemical exergy change (kW)ED rate of destruction of exergy due to the irre-

versibilities inside the control volume (kW)T temperature (K)h specific enthalpy (kJ/kg)s specific entropy (kJ/kg K)g gravitational acceleration (9.8 m/s2)z height (m)

a comparative analysis in terms of exergy balance, the effluent

flue gases or liquids emit considerable amounts of reactivenitrogen species into the environment. Reducing the currentfixation of nitrogen by man by 25% is unrealistic in the shortterm, but a reduction in industrial emissions may be morefeasible and will improve energy efficiency as well as the envi-ronmental performance of such processes.

Fertilizer plants generate multiple effluents from theprocess of getting products such as ammonia, nitric acid,hydrogen, carbon dioxide and urea, with significant loads ofreactive nitrogen, usually present as ammonium. In this casestudy, most of these effluents are sent to a stripping tower(Fig. 1), which operates with multiple inputs, primarily pro-cess condensate effluent stream, CO2 station effluent stream andother minor sources. Other minor sources are discontinuousstreams containing ammonia and methanol. The CO2 stationmeans the CO2 compressor facilities, outside the battery lim-its.

This combined stream is treated with low pressure steam(0.35 MPa) volatilizing and sending compounds such as ammo-nia and methanol into the atmosphere, about 500 and100 ton/year, respectively. In the stripping process a neweffluent is generated still containing reactive nitrogen andmethanol. The latter is finally sent to a treatment plant tobe disposed of in the environment at concentrations that arewithin the prevailing environmental limits. This practice hasbeen applied to date, but the use of stripping towers is con-sidered an outdated technology, an “end of pipe” treatmentwithout attacking the problem at the source, and withoutthinking about the reuse of these reactive nitrogen species andwater which is being wasted. Applying the concepts of cleantechnologies will lead to the segregation, reduction, treatmentand reuse of the wastewater with considerable environmentaland economic gains.

Bioprocesses for treating wastewater with high nitrogenloading have been developed and used industrially, such as:SHARON, Anammox, InNitri, BABE, MAUREEN, Oland, and


CANON (Abma et al., 2006, 2007; Ahn, 2006; Third et al., 2001).

The Anammox process (Anaerobic Ammonium Oxida-tion) was chosen for this study due to the advantages itpresents when compared to other technologies such as: (a)higher treatment capacity (more than 6-fold in comparisonto other technologies, 6–12 kg of nitrogen/m3/day); (b) noneed for external carbon source; (c) very low sludge pro-duction; (d) lower power consumption (around 1 kWh/m3

effluent) and is a consolidated technology, commerciallyavailable, with several plants in operation worldwide(http://en.paques.nl/pageid=66/ANAMMOX%C2%AE.html,03-26-2013).

This process consists of the oxidation of ammonia andnitrite in nearly equal proportions generating gaseous nitro-gen and nitrate formation with low biomass production (Eq.(1)), using microorganisms of the group planctomyocetes (Strouset al., 1998; Ahn, 2006).

1 NH+4 + 1.32 NO−

2 + 0.66 HCO−3 + 0.13 H+ → 1.02 N2

+ 0.26 NO−3 + 0.66 CH2O0.5N0.15 + 2.03 H2O (1)

Exergy balances are useful to calculate the exergydestruction of the system components and identify thethermodynamic inefficiencies, and therefore can help inchoosing the best available technology (Wall and Gong, 2001;Tsatsaronis and Cziesla, 2004). Changes of exergy are not nec-essarily equal to the gross exergy transferred, as this may bedestroyed due to irreversibilities in the system during opera-tion (Moran and Shapiro, 2006).

The purpose of this study was to evaluate the use ofAnammox technology combined with some physicochemi-cal processes through mass, energy and exergy balances inthe system that involves the stripping tower in a nitrogenousfertilizer factory located in Camacarí, in the State of Bahia,Brazil, to achieve environmental gains. Two potential solutionswere analyzed, one of which eliminates the use of the cur-rent stripping tower, preventing atmospheric emissions andminimizing the generation of reactive nitrogen effluents.

2. Materials and methods

2.1. Characterization of the streams

Physicochemical characterization of the streams entering andleaving the stripping tower were made in the laboratories ofthe industry in Camacari, Bahia, Brazil, determining the pH,flow rate, temperature, concentration of ammonia (mg/L) andmethanol (mg/L).

2.2. Cases for analysis

Three cases were analyzed:Case 1: Actual operation: treatment of the streams process

condensate effluent and CO2 station effluent by the use of a strip-ping tower. This generates two streams (Fig. 1): a gas that goesinto the atmosphere containing ammonia and methanol anda liquid with methanol and ammonia that goes (after mix-ing with other streams) to the effluent treatment plant of thepetrochemical complex of Camacari. The process consumesenergy in the form of saturated steam at 0.35 MPa. To carry out


generated by the stripping tower was hypothetically used to

dx.doi.org/10.1016/j.psep.2013.07.003

http://en.paques.nl/pageid=66/ANAMMOX%C2%AE.html,03-26-2013

http://en.paques.nl/pageid=66/ANAMMOX%C2%AE.html,03-26-2013


Process Safety and Environmental Protection x x x ( 2 0 1 3 ) xxx–xxx 3

Fig. 1 – Process condensate stripping tower of the fertilizer factory. The capital letters next to arrows are explained inT

gl

psmd

bbco

(o

d

oW

2

2

I(

0

able 1.

enerate steam after pretreatment by reverse osmosis (dottedine in Fig. 1).

Case 2: Segregation of streams and the use of the Anammoxrocess and more reverse osmosis through a boiler generatingteam for the treatment of CO2 station effluent stream whileaintaining the stripping tower for treating the process con-

ensate effluent stream.Case 3: Removal of the stripping tower and treatment of

oth streams process condensate effluent and CO2 station effluenty using a chemical heat pump coupled to the Anammox pro-ess followed by reverse osmosis and subsequent generationf steam.

Process flowcharts were made by using the program VisioMicrosoft) to visualize the different possibilities of treatmentf the effluents considered better.

Mass balances were calculated with laboratory and plantata using Excel software (Microsoft).

An analysis of the reverse osmosis process at the outputf the Anammox process was performed using the softwareinFlow (General Electric), considering 80% of efficiency.The efficiency of the boiler used to generate steam in cases

and 3 was considered to be 40%.

.3. Exergy balance

n a control volume, the exergetic balance is expressed asMoran and Shapiro, 2006):

dEcv,sys

dt=

∑j

(1 − T0

Tj

)Qj −

(Wcv − p0

dVcv

dt

)

+∑

i

miefi −∑

e

meefe − ECh − ED (2)

At steady state:

∑ ∑ ∑


=j

EQj − Wcv +i

miefi −e

meefe + ECh − ED (3)

With

EQj =(

1 − T0

Tj

)Qj (4)

ef = h − h0 − T0(s − s0) + V2

2+ gz (5)

The rational exergetic efficiency was calculated by(Shukuya and Hammache, 2002):

ε =∑

Efout,real∑Efin

(6)

where the numerator is the exergy of the process with poten-tial as useful work or interconnection between processes. Thedenominator indicates the input exergy in the system.

3. Results and discussion

3.1. Mass balance

The first alternative for treating the effluent streams enteringthe stripping tower (Case 2) is shown in Fig. 2.

In this flowchart the Anammox process treats the streamCO2 station effluent stream. This led to a 80% reduction inammonia and a 93% reduction in methanol in the emissionsto the atmosphere from the stripping tower. A new effluentstream with low concentrations of ammonia and methanol isgenerated which fit the requirements of the effluent streamtreatment facilities. The Anammox process treating the CO2

station effluent stream releases 41 kg/h of N2 gas to atmosphereand generates an effluent stream containing a nitrate load of4.4 kg/h. The reverse osmosis process generates an effluentstream concentrated in nitrate (∼1000 mg/L) at a flow rate ofabout 4 m3/h. The main stream leaving the reverse osmosisprocess reaches nitrate concentrations of the order of 2.6 mg/Lwith a total dissolved solids concentration of about 11 mg/L


and a pH of around 5, according to the simulations carriedout. These characteristics mean that this new stream, after

dx.doi.org/10.1016/j.psep.2013.07.003



Fig. 2 – Flowchart of the Anammox process for treatment of the CO2 station effluent stream. The capital letters next to
arrows are explained in Table 1.
adjusting the pH and degassing, can be used for steam gen-eration, reducing considerably the consumption of boiler feedwater.

In Case 3 (Fig. 3), the mixing of the process condensate andthe CO2 station effluent streams resulted in a temperature ran-ging from 80 to 105 ◦C which is outside the operating rangeof the Anammox process (between 30 and 40 ◦C). The useof a chemical heat pump of the type isopropanol–acetone(Karaca et al., 2002) applied to recover this heat could gen-erate a gain of about 7 kWh/m3 of effluent stream. This wouldallow this combined stream to be conditioned to the workingtemperature of the Anammox process. This would produce54 kg/h of N2 gas and a new effluent stream with a con-centration of nitrate around 150 mg/L. The reverse osmosisprocess generates 8 m3/h of a concentrated effluent streamwith 700 mg/L of nitrate concentration and 32 m3/h of anew stream with less than 1 mg/L of nitrate with a totaldissolved solids concentration of 2 mg/L which, after pHadjustment and degassing, can be used for steam genera-tion.

Chemical analysis of the input streams of the actual strip-ping tower are performed periodically by the company andindicate that the concentration of metals such as Fe and Cu,total organic carbon and silicate have values far below thelimits required for boiler feed water (limits shown in Fig. 1).As the processes examined here do not alter these values,these streams were considered adequate to generate steam.The dashed arrows are used in Fig. 1 to represent a hypotheti-cal case, only for comparison of exergy analysis with the othertwo cases.

Table 1 shows the main results of the exergy analysis per-formed for the three cases considered.

3.2. Exergy balance

In the exergy balance calculated for Case 1 (Fig. 1) the gaseousstream sent to atmosphere was considered to have zero exergy


due to its high dispersion. The rational exergetic efficiency cal-culated and shown in Table 1, 62% considers the possibility

of heat recovery from the liquid stream leaving the strippingtower through steam generation after conditioning (reverseosmosis). As stated in the previous section this hypothesiswas made to allow effective comparison of the three casesstudied. The calculations indicated in this case high powerconsumption and high exergy destruction compared to Cases2 and 3. This is a polluting process, with continuous emissionsof ammonia and methanol to atmosphere.

In Case 2, the separation of the process condensate effluentand CO2 station effluent streams and treatment of the latterby the Anammox process, reverse osmosis and subsequentsteam generation kept the exergetic efficiency comparable toCase 1 (Table 1). The existing stripping tower is still in opera-tion, but only with the process condensate effluent stream, andit exhibits a similar behavior to that discussed in Case 1. TheAnammox process calculations indicate a high rate of exergydestruction, mainly due to the loss of chemical exergy withthe formation of nitrate. It is noteworthy that this compoundcan be used in another process, such as a nutrient in the culti-vation of microalgae. The reverse osmosis process showed lowexergy destruction and high rational exergetic efficiency, con-sistent with the objective of this process, which is to recoverwater for some useful application (in this study, steam pro-duction). The boiler equipment was the major consumer anddissipator of thermal energy, in other words, there was a highrate of exergy destruction.

Case 2 simulations resulted in 380 t/day of water for usein steam generation and twice this value was calculated forCase 3, 760 t/day, corresponding to 79% of the original effluentstream to be treated.

In Case 3 with the withdrawal of the stripping tower, theemission of reactive compounds to atmosphere is eliminated,the only gaseous emission is inert N2 from the Anammoxprocess. This option requires a larger energy input, but italso comprises a process for increasing exergy (chemical heatpump), besides the same steps mentioned in Case 2 (reverseosmosis and steam generation). Case 3 presents the highest


global rational exergetic efficiency (66%) and the lowest rateof exergy destruction when compared to Cases 1 and 2.

dx.doi.org/10.1016/j.psep.2013.07.003

Please cite

this

article in

press

as: Jorq

uera,

O.,

et al.,

Efflu

ent

stream treatm

ent

in a

nitrogen

ous

fertilizer factory:

An

exergy an

alysis for

process

integration

. Process

Safety an

d En

vironm

ental

Protection (2013),

http

://dx.d

oi.org/10.1016/j.psep

.2013.07.003

AR

TIC

LE

IN P

RE

SS

PSEP-366;

No.

of Pages

7

Process

Safety

an

d En

viro

nm

enta

l Pro

tection

x

x x

(

2 0

1 3

)

xxx–xxx

5

Table 1 – Exergy analyses comparison between the three case analyzed.

Exergy input (kW) Exergyassociatedto work(kW)

Exergyassociatedto heat(kW)

Chemicalexergyinput (kW)

Exergy output (kW) Chemicalexergyoutput (kW)

Totalexergyinput(kW)

Totalexergyoutput(kW)

Rate ofexergydestruction(kW)

Rationalexergeticeffi-ciency(%)

Effluent Steam

Case 1A B

Stripping tower 423 454 1301 386 0 616 2177 1002 1175 46.03Reverse osmosis 386 30 616 306 76 616 1002 998 4 91.97Steam generator 306 15,840 616 5822 4644 16,761 10,467 6295 62.45

Total 19,940 12,467 7474 Globalefficiency

62.14

Case 2C D

Stripping tower 261 227 538 191 0 305 1026 496 530 48.35Reverse osmosis (1) 193 15 308 153 38 308 501 499 2 91.97Steam generator (1) 153 7920 308 2911 0 2322 8381 5233 3147 62.45

E FAnammox 6 20 581 9 0 275 607 283 324 46.64

G HReverse osmosis (2) 47 3 275 21 7 274 325 283 42 69.71

ISteam generator (2) 7 7920 220 I 2911 2322 8170 5233 2937 64.06


62.03

Case 3J K L

Chemical Heat Pump 423 939 284 18 89 939 1361 1330 32 91.12

M NAnammox 18 41 939 0 18 564 997 582 416 58.33

O PReverse osmosis 18 30 564 3 14 550 612 567 45 74.06

QSteam generator 14 15,840 440 5822 4644 16,293 10,467 5827 64.24


66.15

dx.doi.org/10.1016/j.psep.2013.07.003



Fig. 3 – Flowchart of the Anammox process replacing the stripping tower. The capital letters next to the arrows are
explained in Table 1.
Preliminary performance tests carried out in Holland by theowner of the Anammox process with a sample of the strippingtower input stream indicated a positive result for the use ofthis technology.

The implementation of the process suggested in Case 3would increase the environmental performance of the nitroge-nous fertilizer factory studied because of:

(a) the heat recovered by the chemical heat pump, whichcould be used to generate medium pressure steam

(b) the treatment of streams with a high reactive nitro-gen load by using a biotechnological process (Anam-mox), avoiding atmospheric emissions of ammonia andmethanol

(c) the use of the effluent stream concentrated from reverseosmosis for biomass production or direct use as a fertilizer

(d) the use of the final dilute effluent stream for steam gen-eration. This corresponds to 79% of the original effluentstream to be treated. Part of the thermal energy requiredcan be supplied by the chemical heat pump, using theconcept of energy integration.

4. Conclusions

The combined analysis using mass, energy and exergy bal-ances allowed the comparison between different possibilitiesfor process integration. The hypothetical removal of the strip-ping tower and treatment of both streams process condensateeffluent and CO2 station effluent by using a chemical heat pumpcoupled to the Anammox process followed by reverse osmo-sis presented the lower overall rate of exergy destruction andavoided atmospheric emissions of ammonia and methanol.Moreover, the recovery of the condensate lost as effluent to bereincorporated in the production of steam network, was sig-nificant to increase the energy efficiency and environmentalperformance of the process.


The implementation of new equipment logically impliesfixed investment and increased operating costs. However, the

proposed solution offers the most efficient use of energy, thelowest exergy destruction and avoids the disposal of liquid andgaseous pollutant compounds into the biosphere.

Acknowledgment

The authors are grateful to the Nitrogenous Fertilizer Factoryof Bahia for their financial support for the research project.

References

Abma, W.R., Schultz, C.E., Wouters, J.W., Mulder, J.W., vanLoosdrecht, M.C.M., Van der Star, W., Strous, M., Tokutomi, T.,2006. Full Scale Granular Sludge ANAMMOX® process. In: 4thCIWEM Annual Conference, pp. 1–8.

Abma, W., Schultz, C., Mulder, J.W., Loosdrecht, M.V., Van der Star,W., Strous, M., Tocutomi, T., 2007. The advance of Anammox.Water 21, 36–37.

Ahn, Y.-H., 2006. Sustainable nitrogen eliminationbiotechnologies: a review. Process Biochem. 41, 1709–1721.

Crutzen, P.J., Mosier, A.R., Smith, K.A., Winiwarter, W., 2007. N2Orelease from agro-biofuel production negates global warmingreduction by replacing fossil fuel. Atmos. Chem. Phys.Discuss. 7, 11191–11205.

Galloway, J.N., Dentener, F.J., Capone, D.G., Boyer, E.W., Howarth,R.W., Seitzinger, S.P., Asner, G.P., Cleveland, C.C., Green, P.A.,Holland, E.A., Karl, D.M., Michaels, A.F., Porter, J.H., Townsend,A.R., Vörösmarty, C.J., 2004. Nitrogen cycles: past, present, andfuture. Biogeochemistry 70, 153–226.

Karaca, F., Kıncay, O., Bolat, E., 2002. Economic analysis andcomparison of chemical heat pump systems. Appl. Therm.Eng. 22, 1789–1799.

Moran, M.J., Shapiro, H.N., 2006. Fundamentals of EngineeringThermodynamics: SI, fifth ed. John Wiley & Sons, Inc, TheAtrium, Southern Gate, Chichester.

Rockström, J., at, al., 2009. A safe operating space for humanity.Nature 461, 472–475.

Shukuya, M., Hammache, A., 2002. Introduction to the Concept ofExergy – For a Better Understanding of


Low-Temperature-Heating and High-Temperature-CoolingSystems. Espoo, VTT Tiedotteita–Research Notes 2158.

dx.doi.org/10.1016/j.psep.2013.07.003

http://refhub.elsevier.com/S0957-5820(13)00047-5/sbref0005



























































































































































































































Process Safety and Environmental Protection x x x ( 2 0 1 3 ) xxx–xxx 7

S

T

trous, M., Heijnen, J.J., Kuenen, J.G., Jetten, M.S.M., 1998. Thesequencing batch reactor as a powerful tool for the study ofslowly growing anaerobic ammonium-oxidizingmicroorganisms. Appl. Microbiol. Biotechnol. 50,589–596.

hird, K.A., Sliekers, A.O., Kuenen, J.G., Jetten, M.S.M., 2001. The


CANON system (completely autotrophic nitrogen-removalover nitrite) under ammonium limitation: interaction and

competition between three groups of bacteria system. Appl.Microbiol. 24, 588–596.

Tsatsaronis, G., Cziesla, F., 2004. Exergy Balance and ExergyEfficiency, vol. 1. Encyclopedia of Life Support Systems(EOLSS), UNESCO, Eolss Publishers, Oxford, UK.

Wall, G., Gong, M., 2001. On exergy and sustainable


development—Part 1:conditions and concepts. Exergy Int. J. 3,128–145.

dx.doi.org/10.1016/j.psep.2013.07.003









































































































































Documents

Effluent stream treatment in a nitrogenous fertilizer factory: An exergy analysis for process integration