Journal of Environmental Management 105 (2012) 53e60

Contents lists available at SciVerse ScienceDirect

Journal of Environmental Management

journal homepage: www.elsevier.com/locate/jenvman

Degradation and detoxication of 4-nitrophenol by advanced oxidation

technologies and bench-scale constructed wetlands
J.A. Herrera-Melin*, A.J. Martn-Rodrguez, A. Ortega-Mndez, J. Araa, J.M. Doa-Rodrguez,
J. Prez-Pea
FEAM-C.I.D.I.A., Departamento de Qumica, Universidad de Las Palmas de Gran Canaria, Edicio de Ciencias Basicas, Campus de Tara, 35017 Las Palmas de Gran Canaria, Spain

a r t i c l e i n f o a b s t r a c t

Article history: The degradation and detoxication towards the duckweed Lemna minor of 4-nitrophenol (4NP)
Received 19 August 2011 was studied by means of bench-scale constructed wetlands (CWs), TiO2-photocatalysis and
Received in revised form Fenton photoFenton reactions. The main goal of this work was to compare the three treatment
7 March 2012
techniques to evaluate their possible combination for the efcient, low cost treatment of 4NP efuents. In
Accepted 21 March 2012
Available online 22 April 2012
CWs, adsorption on the substrate of 4NP was found to achieve 34e45%. Low concentrations (up to
100 ppm) of 4NP were successfully treated by CWs in 8e12 h. The microbial degradation of 4NP started
after a lag phase which was longer with higher initial concentrations of the pollutant. The greatest
Keywords:
4-Nitrophenol
degradation rate was found to occur at initial concentrations of 4NP between 60 and 90 ppm. Solar TiO2-
Toxicity photocatalysis was faster than the CWs. The greatest removals in terms of mass of 4NP removed after 6 h
Constructed wetland of irradiation were found to occur at 4NP concentrations of about 200 ppm. Fenton reaction provided
TiO2-photocatalysis complete 4NP degradation up to 500 ppm in only 30 min but TOC was removed by only about 40%. The
Fenton resulting toxicities were below 20% for initial 4NP concentrations below 300 ppm. It was the
photoFenton Fenton photoFenton combination (180 min in total) that provided TOC reductions up to 80% and
negative L. minor growth inhibition for almost all the 4NP concentrations tested. The combination of
solar TiO2-photocatalysis (6 h) with CWs (16 h) was able to completely treat and detoxify 4NP efuents
with concentrations as high as 200 ppm of the organic.
2012 Elsevier Ltd. All rights reserved.

1. Introduction non-biodegradable organics. AOPs generate a highly reactive, non-

specic oxidant namely hydroxyl radical (OH), capable of degrad-
Nitroaromatic compounds are widely employed in the manu- ing and mineralizing many organic pollutants in water (Kavitha and
facture of pharmaceuticals, pigments, dyes, plastics, pesticides and Palanivelu, 2005). Great interest has been focused in recent years
fungicidal agents, explosives and industrial solvents (Spain, 1995). on TiO2-photocatalysts, Fenton and photoFenton reactions, for the
However, considering its industrial production and the potential oxidative degradation of many organic pollutants. One of the most
environmental pollution, p-nitrophenol or 4-nitrophenol (4NP) has interesting features of these methods is the possibility of using
been regarded as the most important of the mononitrophenols solar light as UV source. This way, the economic cost of the treat-
(Gemini et al., 2005). 4NP is toxic to many living organisms, it may ment would be signicantly lower and the environmental value of
accumulate in the food chain (Donlon et al., 1996), and it has been the technique would be notably increased (Malato et al., 2009).
classied as a priority pollutant by the EPA of the United States of Although AOPs are effective methods, their energy and reagent
America, which restricts its maximum concentration to 10 ppb in consumption and operational costs are relatively high compared to
natural waters (Kulkarni and Chaudhari, 2006). 4NP is stable in the those of biological treatments. Thus, the combination of biological
environment, hardly biodegradable and can be leached easily treatment and AOPs has been proposed as a suitable way of
through the soil into surface and groundwater (Tomei et al., 2006). reducing toxicity and augmenting biodegradability for different
Advanced oxidation processes (AOPs) can be regarded as industrial efuents (Sarri et al., 2003; Suryaman et al., 2006; Oller
a suitable alternative or complement for the treatment of toxic and et al., 2011).
Constructed wetlands (CWs) are mainly employed for the
secondary treatment of domestic and municipal wastewater,
* Corresponding author. Tel.: 34 928 45 44 38; fax: 34 928 45 29 22. constituting an optimal alternative to conventional systems for
E-mail address: [email protected] (J.A. Herrera-Melin). the sanitation of small communities (<2.000 h.e.). CWs have also

0301-4797/$ e see front matter 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jenvman.2012.03.044
54 J.A. Herrera-Melin et al. / Journal of Environmental Management 105 (2012) 53e60

been proved to be efcient for treating farming and industrial 2.5. Photocatalytic experiments
efuents (Gessner et al., 2005; Herrera-Melin et al., 2008a and
2008b, Khan et al., 2009). Low cost of operation and mainte- The photocatalytic reactors were 250 mL Pyrex cylinders.
nance, high effectiveness even under high ow uctuations and Aqueous solutions were prepared with Milli-Q water. After the
contaminant loading and low or nil energy requirements are addition of the photocatalyst, pH was adjusted at the beginning of
some of the advantages of these systems (Vymazal, 2009). 4NP the reactions to 4.5e5 with sulphuric acid and/or sodium
can be eliminated in CWs by different mechanisms such as hydroxide.
adsorption, bacterial degradation, photolysis or plant uptake. The 60-W UV source was a Solarium Philips HB175 UV lamp,
However, to our knowledge, no information on the treatment of equipped with four 15-Watt Philips CLEO uorescent tubes (emis-
4NP wastewaters and the possible mechanisms of elimination sion spectrum from 300 to 400 nm, maximum at 365 nm), which
has been reported. gave an irradiation level of 8.9 milliWatt/cm2 (Graseby 5370
Thus, this paper explores some of the main degradation path- photometer) at a distance of 10 cm. For high irradiation TiO2-
ways of 4NP in CWs, namely biodegradation, adsorption and plant photocatalytic experiments an 800-Watt xenon UV lamp (Solarium
uptake. Additionally, a comparative study between TiO2-photo- Facial UV-A HPA 8068, 210e380 nm, ENCO) was employed.
catalysis, Fenton and photoFenton, and bench-scale CWs for the The reactors were magnetically stirred and continuously aerated
degradation of aqueous solutions of 4NP at concentrations between with an air ow of 0.4 L/min provided by a RESUN AC-3100 air
25 and 500 ppm is presented. From the obtained results, the pump.
optimal experimental conditions were explored to determine the Solar TiO2-photocatalyisis experiments were performed under
feasibility of their combination for the complete removal of the optimal weather conditions (clean skies without clouds) and at the
pollutant and detoxication of the efuent with respect to the same hours (those of highest irradiance) of the day for the solar
duckweed Lemna minor. plant and the solar reactor. The solar results shown on Fig. 6 were
obtained the 25th of May of 2008 and were performed between
2. Materials and methods 12 h and 16 h (solar hour).
For Fenton and photoFenton experiments, pH was set at
2.1. Reagents 3.0  0.2 and 0.45 mM Fe2 and a 12:1 H2O2:4NP molar ratio
(Kavitha and Palanivelu, 2005) was used. Previous to toxicity
Degussa P25 TiO2 (anatase-rutile: 80e20), with a particle size of measurements and/or resuming treatment in the CWs, the excess
30 nm and BET surface area of 50 m2 g1 was employed as pho- of H2O2 was eliminated by adding catalase (made from Micrococcus
tocatalyst. 4NP was purchased from Aldrich and all reagents were lysodeikticus, 1AU destroys 1 mmol H2O2 at pH 7, 243.075 U/mL,
analytical grade. Fluka). The volume of catalase (100e500 mL of 1% w/v to 200 mL of
sample) to be added was calculated according to the H2O2 used as
2.2. Chemical analysis reagent. Thus, the catalase was added in excess. The effect of this
excess of the enzyme on L. minor toxicity tests was proved to be nil.
4NP concentration has been measured with a Varian Prostar
9050 HPLC equipped with a CC-250/4 Nucleosil 100-5 C18 HC 2.6. Experiments in bench-scale CWs
column and a UV detector (l 270 nm). Acetonitrile-water (40:60)
was used as mobile phase. Samples were ltered using 0.45 mm The lab-scale CW reactors were operated in batch and consisted
syringe lters before analysis. TOC was measured using a Shimadzu of 500 mL glass vessels containing 300  0.5 g of gravel and
5000 A. Nitrite ions were quantied with a Thermoelectro Corpo- 100e150 mL of water. To achieve a certain water ow the reactors
ration model Helyos gamma UVeVis spectrophotometer (Millero, were placed on orbital stirrers set at their lowest stirring speed. No
1996). aeration was provided. The gravel (porosity: 49%, average diameter:
6.5  2.5 mm) was taken from a CW in operation for the treatment
2.3. Microbial analysis of urban wastewater generated in the Campus of Tara (Gran
Canaria, Spain).
Total bacterial counts were performed by using R2A agar Bacteria acclimatization to 4NP in CW reactors was performed
medium (APHA, 2005). The agar and nutrient salts were from for 21 days. For this, increasing concentrations of 4NP, from 10 to
PANREAC (Spain). The microorganisms were extracted from the 200 ppm, were added as unique source of carbon and nitrogen,
gravel by sampling about 10 g of wet gravel from the reactors to until complete elimination of the pollutant. Degradation was
which 100 mL of sterile mineral salt medium (Lockhead and monitored by means of TOC and 4NP measurements.
Chase, 1973) were added and stirred for 1 h. After sedimenta-
tion of the suspended particles for 30 min, 1 mL of the superna- 2.7. 4NP uptake by the macrophytes
tant was used for subsequent serial dilution. The lters were
incubated for 48 h at 37  C. The gravel sampled was later weighed To determine the degree of 4NP uptake by common CW plants,
to obtain the exact wet weight. The concentration of bacteria is hydroponic cultures containing three similar-size exemplars of
expressed as colony forming units per gram of gravel, CFU/g. The Phragmites sp., Cyperus sp., Pontederia sp., and Scirpus sp., were
concentration of bacteria in the CWs water was made by the withdrawn from a CW located at the Campus of Tara of our
adequate dilution of the sample and subsequent incubation under university, washed with tap water and placed in plastic recipients
the same conditions. containing 3 L of tap water. Samples of approximately 13 g wet
weight of L. minor specimens were collected. To minimize the
2.4. Toxicity tests photo-oxidation of 4NP by sunlight, the recipients with the plants
and the control were located on the roof of the building but
Toxicity was determined by using the duckweed (L. minor) avoiding the direct exposure to the sunlight. The plants were
toxicity test (APHA, 2005) as specied by Herrera-Melin et al., acclimated for a month with increasing concentrations of 4NP
2008a. L. minor specimens were obtained from LemnaTec GmbH, (10e50 ppm) and ampicillin (10e40 ppm) to inhibit microbial
Germany. activity (Xiong et al., 2008). A control with water and ampicillin but
 J.A. Herrera-Melin et al. / Journal of Environmental Management 105 (2012) 53e60 55

without plants was used. Finally, the degradation of 50 ppm of 4NP wastewater. One was sterilized in autoclave (15 min at 120  C).
in each recipient in the presence of 47 ppm of ampicillin was After this, 50 ppm of 4NP was added to both reactors, covered with
monitored for 96 h. The concentration of 4NP was measured daily aluminum lm and placed on an orbital stirrer. The concentration
in each recipient and control. After the experiment, samples of the of 4NP and nitrite ion was monitored for 36 h. The obtained results
roots, stalks and leaves, with the exception of L. minor for which the are shown in Fig. 2. During the rst 12 h of the experiment, the
complete mass of the plant was considered, were triturated and concentration of 4NP evolved in a similar way for both reactors,
dried at 30  C for 24 h to determine dry weight. Then, a known with a strong initial reduction during the rst hour, from 47.5 down
volume of bi-distilled water was added to each sample and stirred to 35.7 ppm. After this, 4NP concentration remained almost stable
for 24 h. The concentration of 4NP was measured in the ltered around this value for about 6 h. At this time, the concentration of
supernatant. 4NP in the non-sterilized reactor started to differentiate from that
of the sterilized one as it became more readily reduced. After 24 h,
3. Results and discussion there was almost no 4NP left in the non-sterilized reactor, while in
the sterilized one its concentration was stabilized at about 32 ppm.
3.1. Toxicity studies These results indicate that during the rst 6e12 h, biodegradation
of 4NP was almost negligible, being reduced only by adsorption. In
The obtained results from the L. minor toxicity test for 4NP and fact, almost no nitrite was produced. It is after this time that the
the photocatalytic intermediates observed in our analysis and cited bacteria were able to degrade 4NP as no further adsorption seemed
in the literature, i.e. hydroquinone, 4-nitrocatechol and nitrite ion to occur, biodegradation of 4NP progressed as the concentration of
are presented in Fig. 1. nitrite ion was notably increased. In the sterilized reactor, the
As can be observed, concentrations of 4NP and 4-nitrocatechol stabilization of the concentration of 4NP coincides with the lack of
lower than 100 and 10 ppm, respectively, resulted in negative nitrite ion production showing that 4NP molecule was not broken
growth inhibition for the duckweed. This means faster growth than down.
that of the control without the pollutant and the possible utiliza- In this case, adsorption of 4NP measured as the pollutant
tion of these organics as nutrients by the plant. We have also reduction in the sterilized reactor, accounted for 34%, from 47.5 to
observed this effect in experiments with other pollutants (Araa 31.4 ppm. These experiments were repeated with a different
et al., 2008). Hydroquinone and nitrite ions did not show any sample of gravel and the concentration of 4NP was monitored for
nutrient effect on L. minor as they inhibited the plants growth at all 14 h. In this case, adsorption accounted for 45%, from 48 to
the concentrations tested. 26.4 ppm. Thus, these results show that adsorption of 4NP can be
The resulting EC50, the concentration of pollutant which inhibits an important removal mechanism in CWs.
the duckweed growth rate by 50%, for 4NP, 4-nitrocatechol,
hydroquinone and nitrite ions were 202.5, 57, 11 and 105 ppm, 3.2.2. Bacteria acclimation
respectively. Therefore, these results indicate that all these inter- Most nitroaromatic compounds, including 4NP, are toxic to
mediates are more toxic than 4NP itself and that the elimination of microorganisms and generally considered highly resistant to
4NP during the treatment of the sample does not imply a lower microbial degradation (Kuscu and Sponza, 2005). Moreover, the
toxicity, but most probably a greater one as long as these inter- nitro group of 4NP enhances the resistance of the aromatic ring to
mediates remain in solution. biodegradation, although several bacteria strains able to utilize 4NP
as sole carbon and nitrogen sources have been described (Gemini
3.2. Bench-scale constructed wetland et al., 2005).
Fig. 3 shows the variation in bacteria population and added 4NP
3.2.1. Adsorption versus degradation of 4NP concentration during the acclimation period. At the beginning of
Adsorption onto the surface of the CWs substrate (gravel, sand, the experiment, the concentration of bacteria was reduced from
organic debris) can be a signicant removal mechanism for organic 3.1  109 to 1.5  106 CFU/100 mL. This was probably caused by
pollutants in constructed wetlands. To determine if adsorption toxicity of 4NP and/or the death of those microbes that could not
could be a relevant removal mechanism of 4NP in CWs, a set of two develop an adaptation mechanism to live in a media containing
reactors were prepared with gravel and untreated urban 4NP as unique source of nutrients. After this, as 4NP concentration

100

80
Toxicity, growth inhibition %

60

40

20

0
0 50 100 150 200 250 300 350 400 450

-20 Concentration, ppm

-40

Fig. 1. Toxicity as Lemna minor growth inhibition for 4NP (-o-) and some of its Fig. 2. Evolution of the concentration of 4NP (solid lines) in a sterilized reactor (->-)
degradation intermediates: hydroquinone (-,-), 4-nitrocatechol (-A-) and nitrite ion and non-sterilized reactor (-,-), and nitrite ions (dotted lines) in a sterilized reactor
(--). (-:-) and non-sterilized reactor (--).
56 J.A. Herrera-Melin et al. / Journal of Environmental Management 105 (2012) 53e60

Fig. 3. Evolution of the concentration of bacteria (>) and 4NP (,) concentrations in water during the acclimatization period of the biological reactors.

was incremented, that of bacteria started to raise up to

1.5  1012 CFU/100 mL, when the concentration of 4NP reached its
maximum (209.7 ppm). The nal reduction of the concentration of
bacteria, down to 1.7  1010 CFU/100 mL could be caused again by
toxicity of 4NP. a
50 1.E+09
4NP, TOC and nitrites, ppm

3.2.3. The effect of the initial concentration of 4NP

Bacteria, CFU/100 mL
The biodegradation of increasing concentrations of 4NP was 40
studied in bench-scale CWs by triplicate. Fig. 4 shows the evolution
30
of 4NP, TOC, nitrites and the initial and nal concentration of
1.E+08
bacteria during these experiments for one of the reactors. As can be 20
observed, removal of 4NP was slow or nil at the beginning of the
experiment. This is particularly clear for the higher initial concen- 10
trations of 4NP. A similar behavior was also observed by Gemini
et al. (2005) for the degradation of 4NP with Rhodococcus wra- 0 1.E+07
tislaviensis, for which 4NP biodegradation kinetics comprised two 0 2 4 6 8 10
phases: a rst lag phase when either the appropriate metabolic
mechanisms are being selected or an active population is produced, b Time, h

or both instances, and a second one with the molecule being 100 1.E+09
4NP, TOC and nitrites, ppm

actively metabolized. This phase is longer when high pollutant

Bacteria, CFU/100 mL
concentrations are treated, and less signicant with lower 80

concentrations. In fact, Tomei et al. (2004), reported that after an 1.E+08

60
initial lag period 4NP biodegradation showed the trend of
a substrate inhibited kinetics, with an initial low removal rate when 40
the substrate concentration is high, and subsequent faster kinetics 1.E+07
with reduced substrate concentration. Zhang et al. (2009) also 20
found a lag phase of 4 h when Rhodococcus sp.CN6 was used to
0 1.E+06
degrade high initial 4NP concentrations.
0 5 10 15
The obtained TOC reductions were also very high, ranging from
74 to 92%. This agrees with toxicity results as negative L. minor
c Time, h
growth inhibition values (from 37 to 75%) were measured,
160 1.E+11
showing the nutrient rather than toxic effect of the remaining
4NP, TOC and nitrites, ppm

Bacteria, CFU/100 mL

intermediates.
120
The production of nitrites in these experiments was almost 1.E+10
stoichiometric with theoretical to experimental ratio ranging
between 8 to 16%. This indicates that nitrites derived mainly from 80
the degradation of 4NP and the fact that no nitrication occurred, at 1.E+09
least under the experimental conditions employed. 40
As a result of the biodegradation of 4NP, the concentration of
bacteria was incremented (Fig. 4) for the three reactors. However, 0 1.E+08
the highest efciency of 4NP removal, determined as mg of 4NP 0 10 20 30 40
degraded divided by the initial concentration of bacteria and time,
was achieved at initial 4NP concentrations between 50 and Time, h

100 ppm (Fig. 5), with a strong reduction of removal efciency at Fig. 4. Effect of the initial concentration of 4NP (a) 39 ppm, b) 84 ppm and c) 152 ppm)
higher concentrations. Results for TOC were quite similar (not on the evolution of: 4NP (->-), TOC (--) and nitrites (-D-) in bench-scale CW. The
shown). initial and nal concentrations of bacteria (,) are also shown.
 J.A. Herrera-Melin et al. / Journal of Environmental Management 105 (2012) 53e60 57

1.0
4NP removal , mg/CFU h

Initial 4NP degradation rate,

1.2E-03 0.78
0.8

0.6

ppm/min
8.1E-04
0.47

0.4
4.1E-04 0.27 0.28

0.2

1.0E-05
0.0
0 50 100 150 200 Sunlight Solar pilot plant 60-W UV-lamp 800-W UV-lamp
UV source
Initial 4NP concentration, ppm
Fig. 6. Effect of different UV sources and reactors on the initial degradation rate of 4NP.
Fig. 5. Effect of the initial concentration of 4NP on removal efciency in bench-scale [4NP]: 50 ppm, [TiO2]: 1.2 g/L, pH: 5.
CW.

These results indicate that CWs can be a very economic, efcient The photocatalytic degradation of 4NP has been described in the
method for the treatment of 4NP at concentrations up to literature by using the widely known LangmuireHinshelwood
100e150 ppm in less than 40 h, which is a suitable hydraulic model (Chen and Ray, 1998; Essam et al., 2007). However, at low
retention time for CWs. substrate concentration, a pseudo-rst-order pattern can be used
(Ahmed et al., 2011). For comparison purposes, initial degradation
3.2.4. Uptake of 4NP by CWs macrophytes rate values were obtained by multiplying the initial 4NP concen-
Plants can help in the removal of xenobiotics by means of trations by the obtained pseudo-rst-order kinetic constants.
different processes such as their uptake and/or accumulation into The optimal concentration of TiO2 was found to be 1.2 g/L
their tissues. However, this could enhance the transfer of toxic and/ though little differences in removal performance were found
or bioaccumulative pollutants to upper trophic levels. between 1 and 1.5 g/L. These results agree with those of Chen and
To test the possible uptake of 4NP by CW plants, hydroponic Ray (1998). Thus, all the TiO2-photocatalytic experiments were
cultures of Cyperus sp., Phragmites sp., L. minor, Scirpus sp. and Pon- then performed with 1.2 g/L of the catalyst.
tederia sp. and a control without plants were prepared by triplicate In TiO2-photocatalysis, the O2 present in air, dissolved or
with 47 ppm of 4NP and ampicillin, an antibiotic that would mini- gaseous, can react with the photogenerated electron to yield O2


mize 4NP bacterial degradation. After 96 h, the plants were collected 

and reduce the e /h recombination (Chen and Ray, 1998). In
and analyzed for 4NP in their roots, stems and leaves (Table 1). experiments with and without aeration, the degradation of
With the exception of L. minor, 4NP was detected in almost all 100 ppm of 4NP after 3 h with the same UV lamp was 70% (0.6 ppm/
the samples analyzed. This experiment was designed as a test to min) with aeration and 50% without aeration (0.4 ppm/min). Thus,
determine if CWs macrophytes could uptake 4NP and no quanti- all the photocatalytic degradation experiments were performed
tative conclusions are intended to be drawn from these data. with aeration.
However, these results should be considered as a warning when One of the most interesting features of TiO2-photocatalysis and
deciding the use of plants in CWs for the treatment of wastewater photoFenton is the possibility of using solar light as UV source. The
containing 4NP as its toxicity and potential accumulation in the use of this renewable energy would notably reduce the economic
food chain (Donlon et al., 1996) could enhance its impact on human cost of the technique and make it more sustainable. The goal of
beings and biota. these experiments was to compare sunlight to UV lamps for the
photocatalytic degradation 4NP. The solar photocatalytic pilot plant
3.3. Advanced oxidation technologies, AOTs (Herrera Melin et al., 2008b) was used and compared to vertical
250 mL glass reactors under solar radiation without any light
3.3.1. TiO2-photocatalytic degradation of 4NP concentrator, and UV lamps (60 W and 800 W) for the degradation
The optimization of the TiO2-photocatalytic degradation of 4NP of 50 ppm of 4NP. The obtained results expressed as initial degra-
included studying the effect of UV source, aeration and hydraulic dation rate (ppm/min) for UV source are shown in Fig. 6.
regime (batch versus continuous feeding of the reactor) and As expected, the UV lamps provided faster degradations. In fact,
concentration of the catalyst. The reaction time was 6 h as this was the initial degradation rate achieved by the 800 W UV lamp
considered to be an average photocatalytically active time for solar (0.71 ppm/min) was 2.4 times higher than that of the solar pilot
irradiation during a day. plant (0.29 ppm/min). However, the merit of using these lamps
should be considered as the improvement obtained by increasing
Table 1
Concentration of 4NP (average  standard deviation) in different parts of the CW
the power of the lamp by 13 times (from 60 to 800 W) resulted in
macrophytes and 4NP removal after 96 h in hydroponic cultures and control. Initial a degradation increment of only 1.6 times.
concentration of 4NP was 47 ppm. Another interesting result of these experiments is the fact that
4NP removal Roots Stalks Leaves
the 250 mL reactor under solar light and the pilot plant provided
% mg/Kg mg/Kg mg/Kg comparable degradations, even though they had quite different
Control 3.3 e e e congurations. This indicates that the results obtained with the
Lemna minor 54.2  19.0 0.0  0.0 0.0  0.0 0.0  0.0 small reactors can be up-scaled with reasonable condence. Similar
Cyperus sp. 54.8  30.8 5.5  7.8 8.5  4.9 6.5  9.2 results have been obtained for phenol (Herrera Melin et al.,
Phragmites sp. 54.5  7.8 0.0  0.0 31.0  29.7 10.0  14.1 2008b).
Scirpus sp. 58.0  5.8 1.0  1.4 21.5  9.2 2.0  1.4
Pontederia sp. 80.9  13.9 0.0  0.0 0.5  0.7 0.5  0.7
Regarding the effect of the initial concentration of 4NP, the results
obtained in this study indicate that the LangmuireHinshelwood
58 J.A. Herrera-Melin et al. / Journal of Environmental Management 105 (2012) 53e60

100 Fe3 H2O hg / Fe2 OH

150
Mass of TOC and 4NP removed, ppm

80 The high oxidation potential of the hydroxyl radical explains the

120 efciency of these processes.
60
Kavitha and Palanivelu observed that almost all the nitrite

Toxicity, %
90 40 formed in the degradation of nitrophenols was oxidized to nitrate.
Srensen and Frimmel (1997) indicated that nitrate reduced the
20 absorption of UV by H2O2 leading to lower reduction of the ef-
60
ciency of UVeH2O2 at the degradation of EDTA. Mack and Bolton
0
(1999) considered that nitrates are not efcient OH radical scav-
30 enger but can be reduced to nitrites by direct photolysis:
-20

0 -40 NO 
3 hg / NO2 O2
0 100 200 300 400 500
However, nitrites can be efcient  OH radical scavengers,
Initial concentration of 4NP, ppm
directly:
Fig. 7. Effect of initial 4NP concentration on the mass removal of TOC (-), toxicity ()
NO
2 OH / NO2 OH
 
and mass removal of 4NP (A) by TiO2-photocatalysis. [TiO2]: 1.2 g/L, pH: 5. 

model can be applied only at initial concentrations of 4NP below or through the reaction of a nitrite photolysis product (NO):
300 ppm for which correlation coefcients (R2: 0.965e0.996) were
high enough. However, at higher initial 4NP concentrations (400 and NO OH / HNO2
500 ppm) the obtained correlation coefcients were very low (0.76
and 0.68, respectively), tting more properly to zero order kinetics. These reactions can notably reduce the concentration of OH
Thus, the simplication of the LangmuireHinshelwood model available to take part in oxidation reactions with organic pollutants
cannot be assumed in this case. Similar results were obtained by (Mack and Bolton, 1999).
Dimitrakopoulou et al. (2012), when studying the photocatalytic In this study, the Fenton reaction (without light) was applied
degradation of amoxicillin. during the rst 30 min. Then, the UV lamp was turned on (photo-
As a consequence, 4NP removal was expressed in terms of Fenton) during the rest of the time (150 min). The application of the
percentage or mass eliminated instead of initial degradation rate. Fenton reaction to increasing 4NP concentrations, (from 50 to
The obtained results, in addition to the toxicity of the treated 500 ppm) led to remarkable degradation of the organic pollutant,
samples are shown in Fig. 7. being complete at concentrations below 400 ppm. For 500 ppm the
It must be noted that the evolution of removal percentage versus removal of 4NP was up to 97.5% (Fig. 8).
initial 4NP concentration was very similar to that of initial degra- However, mineralization measured as TOC reduction remained
dation rate versus initial 4NP concentration, with a maximum at around 41.4%. Considering the higher toxicity of the intermediates
100 ppm of 4NP (data not shown). The highest removal efciencies with respect to that of 4NP, it is important to achieve a high degree
(expressed in %) of 4NP and TOC were achieved at 4NP initial of mineralization. As shown in Fig. 8, this was obtained by adding
concentrations below 100 ppm. At these levels, almost complete the photoFenton reaction to the process that provided an average
detoxication of the sample (below 10% inhibition) could be ach- TOC removal of 79.4%.
ieved. In fact, when the initial concentration of 4NP was 45 ppm, Toxicity towards L. minor of the treated samples was also studied
removal of 4NP and TOC was almost complete and the inhibition (Fig. 9). As noted above, concentrations of 4NP below 100 ppm are
growth was negative, indicating complete toxicity elimination and not toxic to the duckweed but exert a nutrient effect on it. The
a certain nutrient effect towards the duckweed of the remaining treatment of these samples by Fenton and/or photoFenton
intermediates. increased their toxicity, though 4NP was almost completely
However, if the mass degraded of the organic pollutant is
considered instead of percentage degraded, it is obvious that the
best results are obtained at higher 4NP concentrations, between 100
Removal of TOC and 4NP, %

150 and 300 ppm, approximately. In this case, the toxicity of the
treated sample is still very high, with inhibition values between 30 80
and 70%. Thus, the sample requires further treatment to eliminate
toxicity and the remaining pollutants, i.e., 4NP and intermediates 60
which in this study were identied as nitrite ions, hydroquinone
and 4-nitrocatechol. It must be noted that these organic interme- 40
diates exert a remarkable growth inhibition on L. minor (Fig. 1).
20
3.3.2. Fenton and photoFenton reactions
Fenton and photoFenton processes have been successfully
0
applied to a wide range of organic pollutants and industrial waste-
waters (Kavitha and Palanivelu, 2005). In the Fenton process, OH 0 100 200 300 400 500
radicals are generated by the reaction of Fe2 ions with H2O2 at pH 3: Concentration of 4NP, ppm

Fig. 8. Effect of Fenton and Fenton photoFenton on the removal (%) of 4NP and TOC:
Fe2 H2O2 / Fe3 OH OH
4NP by Fenton (>), TOC by Fenton (-) and TOC by Fenton photoFenton (). Data
corresponding to 4NP by Fenton photoFenton coincide almost completely with those
Under irradiation (photoFenton) Fe3 ions are reduced to Fe2 to of 4NP by Fenton and are not shown for clarity. Reaction time for Fenton and photo-
provide a new hydroxyl radical and re-start the Fenton process: Fenton were 30 and 150 min, respectively.
 J.A. Herrera-Melin et al. / Journal of Environmental Management 105 (2012) 53e60 59

100 Table 2
Toxicity as Lemna minor growth

Degradation of high concentrations of 4NP by combining TiO2-photocatalysis (6 h of

80 reaction time) and bench-scale CWs. The efuent of the photocatalytic reactor was
the inuent of the bench-scale CW.
60
inhibition, %

TiO2-photocatalysis Bench-scale CW (after TiO2-photocatalysis)

40
Inuent Efuent Efuent
20
[4NP] [4NP] Toxicity Reaction 4NP removal TOC removal Toxicity
0 ppm ppm % time h % % %
0 100 200 300 400 500 205 49.9 43 16 100 39.9 71.4
-20
289 142.9 68 21 100 49.5 40.0
-40 396 271.0 68 45.5 99.1 83.0 8.0

Concentration of 4NP, mg/L

Fig. 9. Toxicity towards Lemna minor of 4NP before (A) and after treatments with this study the Fenton photoFenton process reduced the toxicity of
Fenton (,) and Fenton photoFenton (D).
300 and 400 ppm 4NP from 60 to 100%, respectively to 20% after 3 h
of treatment. For concentrations between 100 and 200 ppm of the
eliminated. This can be attributed to the presence of degradation compound, TiO2-photocatalysis can be a more competitive method
intermediates as TOC was reduced only by 41.4%. At higher as it does not require as many reagents as Fenton and derived
concentrations, the application of Fenton reduced toxicity to some techniques. For concentrations below 100 ppm, the most suitable
extent but it was the later application of photoFenton that provided technique according to this study would be a CW, as these
almost complete detoxication of the samples with growth concentrations were successfully detoxied in less than 1 d, which
inhibitions ranging from 29.4% to 11.8%. These results agree is a proper hydraulic retention time for these reactors.
with the higher TOC removal obtained by the combined In addition to this, the efcient but more expensive AOTs can be
Fenton photoFenton process. implemented with the less efcient but economic CWs, particularly
Given that this treatment alone achieved the complete detoxi- in the case of the AOTs that require irradiation and where solar light
cation of 4NP, its combination with the biological method was not is used.
studied.
3.5. Combination of TiO2-photocatalysis and bench-scale CW
3.4. Comparing the biological and AOTs techniques
According to the data of this study, TiO2-photocatalysis could
Comparing the efciency of the three techniques studied for not achieve satisfactory detoxication of 4NP at concentrations
4NP degradation is not a simple task because they are very different above 100e200 ppm and after 6 h of treatment (Fig. 5). However,
in terms of reactor conguration, hydraulic retention time, 4NP can be found at higher concentrations in industrial wastewa-
requirement of reagents and radiation or bacteria concentration ters (Tomei et al., 2006) and the number of hours of photo-
and selection in the case of the biological method. However, to catalytically active sunlight can be scarce on a daytime basis. In
compare the kinetics of the three techniques the degradation of this regard, the combination of TiO2-photocatalysis with a low cost
100 ppm of 4NP was studied in batch reactors (Fig. 10). biological method such as CW can offer a remarkable exibility and
The obtained results clearly show that the Fenton reaction efciency to the treatment process (Araa et al., 2008; Herrera
provided the fastest degradation of 4NP, followed by TiO2-photo- Melin et al., 2008b) In addition to this, the use of AOPs as
catalysis and the CW. However, when selecting the optimal method a pretreatment can greatly enhance wastewater biodegradability
for the treatment of wastewater containing 4NP, one cannot (Sarri et al., 2003).
consider only kinetic information but also other factors such as Table 2 shows the results obtained with the combination of
maintenance, input of energy and reagents, land availability in the TiO2-photocatalysis (6 h) with bench-scale CWs for the treatment
case of the CW, and concentration of 4NP, among others. of 4NP concentrations above 200 ppm. According to these data,
The results of this study suggest that Fenton and/or complete detoxication towards L. minor can be achieved within
Fenton photoFenton are the most suitable methods for the 1e2 days for 4NP concentrations up to 480 ppm. Additionally, the
treatment of high concentrations (200e500 ppm) of 4NP. In fact, in results suggest that the photocatalytic treatment enhanced biode-
gradability as the bench-scale CW alone required about 37 h to treat
152 ppm of 4NP. However, when a concentration of 289 ppm of 4NP
was treated by TiO2-photocatalysis the remaining 143 ppm were
eliminated and detoxied in only 21 h. This result could be
explained in the particular case of 4NP by considering the nutri-
tional value of some of the degradation intermediates, particularly
for nitrites and nitrates and the fact that no nutrients were added to
the CWs.
This combination provided an efcient treatment that seems to
be faster than the solution reported by Essam et al. (2007), which is,
to the best of our knowledge, the only study based on coupled TiO2
photocatalytic-biological systems for the treatment of 4NP.

4. Conclusion

Efuents having 4NP at concentrations up to 150 ppm can be

Fig. 10. Degradation of 100 ppm of 4NP with bench-scale CW (,), TiO2-photocatalysis successfully treated in CW within the typical hydraulic residence
(>) and Fenton photoFenton (:). time for these systems (2e3 days). Adsorption onto the substrate
60 J.A. Herrera-Melin et al. / Journal of Environmental Management 105 (2012) 53e60

was found to be a signicant removal process. The presence of 4NP Gessner, T.P., Kadlec, R.H., Richard, P., Reaves, R.P., 2005. Wetland remediation of
cyanide and hydrocarbons. Ecol. Eng. 25, 457e469.
in different CW plants can prevent their use for the treatment of
Herrera-Melin, J.A., Ortega-Mendez, A., Araa, J., Gonzalez-Diaz, O., Tello-Rendon, E.,
efuents containing the organic pollutant. 2008a. Degradation and detoxication of formalin wastewater with aerated
Solar TiO2-photocatalysis could efciently remove up to biological lters and wetland reactors. Process Biochem. 43, 1432e1435.
200 ppm of 4NP in 6 h, but the treated efuent was still toxic. When Herrera Melin, J.A., Araa, J., Ortega, J.A., Martn Munoz, F., Tello Rendn, E., Prez
Pea, J., 2008b. Comparative study of phenolics degradation between biological
combined with CW (16 h) the complete elimination and detoxi- and photocatalytic systems. J. Sol. Energ-T. Asme 130, 1003/1e1003/7.
cation of 4NP was achieved. Khan, S., Ahmad, I., Tahir Shah, M., Rehman, S., Khaliq, A., 2009. Use of constructed
Fenton reaction provided complete 4NP degradation up to wetland for the removal of heavy metals from industrial wastewater. J. Environ.
Manage. 90, 3451e3457.
500 ppm in only 30 min but TOC was removed by only about 40%. Kavitha, V., Palanivelu, K., 2005. Degradation of nitrophenols by Fenton and photo-
The resulting toxicities were below 20% for initial 4NP concentra- Fenton processes. J. Photochem. Photobiol. A 170, 83e95.
tions below 300 ppm. It was the Fenton photoFenton combina- Kulkarni, M., Chaudhari, A., 2006. Biodegradation of p-nitrophenol by Pseudomonas
putida. Bioresour. Technol. 97, 982e988.
tion (180 min in total) that provided TOC reductions up to 80% and Kuscu, O.S., Sponza, D.T., 2005. Performance of anaerobic bafed reactor (ABR)
negative L. minor growth inhibition for almost all the 4NP treating synthetic wastewater containing p-nitrophenol. Enzym. Microb. Tech.
concentrations tested. 36, 888e895.
Lockhead, A.G., Chase, F.E., 1973. Quantitative studies of soil microorganisms.
The higher efciency of TiO2-photocatalysis at the degradation Nutritional requirements of the predominant bacterial ora. Soil Sci. 55,
of 4NP can be combined with the much cheaper, light-independent 184e195.
but less efcient CW for the convenient treatment of efuents Malato, S., Fernndez-Ibez, P., Maldonado, M.I., Blanco, J., Gernjak, W., 2009.
Decontamination and disinfection of water by solar photocatalysis: recent
containing the organic pollutant at concentrations up to 200 ppm
overview and trends. Catal. Today 147, 1e59.
on a 1-day-basis. For higher concentrations (up to 500 ppm) or Mack, J., Bolton, J.R., 1999. Photochemistry of nitrite and nitrate in aqueous solution:
shorter reaction times, Fenton photoFenton alone can be the a review. J. Photochem. Photobiol. A 128, 1e13.
optimal treatment method, though at higher economic cost. Millero, F.J., 1996. Chemical Oceanography, second ed. CRC Press.
Oller, I., Malato, S., Snchez-Perez, J.A., 2011. Combination of advanced oxidation
processes and biological treatments for wastewater decontamination. A review.
References Sci. Total Environ.. doi:10.1016/j.scitotenv.2010.08.061
Sarri, V., Kenfack, S., Guillod, O., Pulgarin, C., 2003. An innovative coupled solar-
Ahmed, S., Rasul, M.G., Brown, R., Hashib, M.A., 2011. Inuence of parameters on the biological system at eld pilot scale for the treatment of biorecalcitrant
heterogeneous photocatalytic degradation of pesticides and phenolic contam- pollutants. J. Photochem. Photobiol. A 159, 89e99.
inants in wastewater: a short review. J. Environ. Manage. 92, 311e330. Srensen, M., Frimmel, F., 1997. Photochemical degradation of hydrophilic xenobi-
APHA, 2005. Standard Methods for the Examination of Water and Wastewater, otics in the UV/H2O2 process: inuence of nitrate on the degradation rate of
twenty rst ed. American Public Health Association/American Water Works EDTA, 2-amino-1-naphthalenesulfonate, diphenyl-4-sulfonate and 4-40 -dia-
Association/Water Environment Federation (APHA-AWWA-WEF), Washington, minostilbene-2-20 -disulfonate and 4-40 -diaminostilbene-2-20 -disulfonate.
DC, USA. Water Res. 31, 2885e2891.
Araa, J., Garriga-i-Cabo, C., Fernndez Rodrguez, C., Herrera Melin, J.A., Ortega Spain, J.C., 1995. Biodegradation of aromatic compounds. Annu. Rev. Microbiol. 49,
Mndez, J.A., Doa-Rodrguez, J.M., Prez Pea, J., 2008. Combining TiO2-pho- 523e555.
tocatalysis and wetland reactors for the efcient treatment of pesticides. Che- Suryaman, D., Hasegawa, K., Kagaya, S., 2006. Combined biological and photo-
mosphere 71, 788e794. catalytic treatment for the mineralization of phenol in water. Chemosphere 65,
Chen, D., Ray, A.K., 1998. Photodegradation kinetics of 4-nitrophenol in TiO2 2502e2506.
suspension. Water Res. 32, 3223e3234. Tomei, M.C., Rossetti, S., Annesini, M.C., 2006. Microbial and kinetic characterization
Dimitrakopoulou, D., Rethemiotaki, I., Frontistis, Z., Xekoukoulotakis, N., Venieri, D., of pure and mixed cultures aerobically degrading 4-nitrophenol. Chemosphere
Mantzavinos, D., 2012. Degradation, mineralization and antibiotic inactivation 63, 1801e1808.
of amoxicillin by UV-A/TiO2 photocatalysis. J. Environ. Manage. 98, 168e174. Tomei, M.C., Annesini, M.C., Bussoletti, S., 2004. 4-nitrophenol biodegradation in
Donlon, B.A., Razo-Flares, E., Lettinga, G., Field, J.A., 1996. Continuous detoxication, a sequencing batch reactor: kinetic study and effect of lling time. Water Res.
transformation, and degradation of nitrophenols in upow anaerobic sludge 38, 375e384.
blanket (UASB) reactors. Biotechnol. Bioeng. 51, 439e449. Vymazal, J., 2009. The use constructed wetlands with horizontal sub-surface ow
Essam, T., Amin, M.A., Tayeb, O.E., Mattiasson, B., Guieysse, B., 2007. Solar-based for various types of wastewater. Ecol. Eng. 3, 1e17.
detoxication of phenol and p-nitrophenol by sequential TiO2 photocatalysis Xiong, J., He, Z., Liu, D., Mahmood, Q., Yang, X., 2008. The role of bacteria in the
and photosynthetically aerated biological treatment. Water Res. 41, heavy metals removal and growth of Sedum alfredii Hance in an aqueous
1697e1704. medium. Chemosphere 70, 489e494.
Gemini, V.L., Gallego, A., de Oliveira, V.M., Gmez, C.E., Mano, G.P., Korol, S.E., Zhang, J., Sun, Z., Li, Y., Peng, X., Wen, L., Yan, Y., 2009. Biodegradation of p-nitro-
2005. Biodegradation and detoxication of p-nitrophenol by Rhodococcus phenol by Rhodococcus sp. CN6 with high cell surface hydrophobicity. J. Hazard.
wratislaviensis. Int. Biodeter. Biodegr. 55, 103e108. Mater. 163, 723e728.