water research 43 (2009) 3177–3186

Available at www.sciencedirect.com

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

Toxicity of fluoride to microorganisms in biological

wastewater treatment systems

Valeria Ochoa-Herrera, Qais Banihani, Glendy León, Chandra Khatri, James A. Field,
Reyes Sierra-Alvarez*
Department of Chemical and Environmental Engineering, University of Arizona, P.O. Box 210011, Tucson, AZ 85721-0011, USA

article info abstract

Article history: Fluoride is a common contaminant in a variety of industrial wastewaters. Available

Received 8 December 2008 information on the potential toxicity of fluoride to microorganisms implicated in biological
Received in revised form wastewater treatment is very limited. The objective of this study was to evaluate the
19 April 2009 inhibitory effect of fluoride towards the main microbial populations responsible for
Accepted 20 April 2009 the removal of organic constituents and nutrients in wastewater treatment processes. The
Published online 3 May 2009 results of short-term batch bioassays indicated that the toxicity of sodium fluoride varied
widely depending on the microbial population. Anaerobic microorganisms involved in
Keywords: various metabolic steps of anaerobic digestion processes were found to be very sensitive to
Fluoride the presence of fluoride. The concentrations of fluoride causing 50% metabolic inhibition
Microbial inhibition (IC50) of propionate- and butyrate-degrading microorganisms as well as mesophilic and
Wastewater treatment thermophilic acetate-utilizing methanogens ranged from 18 to 43 mg/L. Fluoride was also
inhibitory to nitrification, albeit at relatively high levels (IC50 ¼ 149 mg/L). Nitrifying
bacteria appeared to adapt rapidly to fluoride, and a near complete recovery of their
metabolic activity was observed after only 4 d of exposure to high fluoride levels (up to
500 mg/L). All other microbial populations evaluated in this study, i.e., glucose fermenters,
aerobic glucose-degrading heterotrophs, denitrifying bacteria, and H2-utilizing metha-
nogens, tolerated fluoride at very high concentrations (>500 mg/L).
ª 2009 Elsevier Ltd. All rights reserved.

1. Introduction Fluoride is ubiquitous in the environment as it is a compo-

nent of most types of soils. Concentrations of inorganic fluoride
Fluoride is a widespread environmental contaminant and is in unpolluted surface water generally range from 0.01 to
estimated to be the 13th most abundant element on the 0.30 mg/L, but considerably higher concentrations may be
earth’s crust (Mason and Moore, 1982). The public health found in regions impacted by geothermal or volcanic activity
benefits and risks of fluoride from drinking water and other (Camargo, 2003). Human activities can also contribute to
sources has been a source of controversy in recent years. increase the concentration of fluoride in aquatic environments.
Although fluoride is beneficial to human health at low Fluoride is often present in a variety of untreated industrial
concentrations (0.7–1.2 mg/L) by affording protection against effluents, including those from chemical plants manufacturing
dental caries, at concentrations exceeding the U.S. federal organofluorine compounds, aluminum smelters, phosphate
drinking water standard (4 mg/L), it has been reported to fertilizers, semiconductor manufacturing, glass and brick-
cause skeletal and dental fluorosis (US-EPA, 2006). making industries, and coal power plants (Fuge and Andrews,

* Corresponding author. Tel.: þ1 520 626 2896; fax: þ1 520 621 6048.
E-mail address: [email protected] (R. Sierra-Alvarez).
0043-1354/$ – see front matter ª 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.watres.2009.04.032
3178 water research 43 (2009) 3177–3186

1988; Skjelkvale, 1994; Sujana et al., 1998; Camargo, 2003; Shen methanogenic studies which was obtained from an anaerobic
et al., 2003). The concentrations of fluoride in untreated sludge digestor operating at the Hyperion municipal waste-
industrial wastewaters vary widely, and concentrations as high water treatment plant, Los Angeles, CA.
as 500 to 2,000 mg/L have been reported in effluents from The inoculum utilized in the nitrification assays, Randolph
semiconductor industry operations in Taiwan (Huang and Liu, Park sludge I, was obtained from the nitrification stage of
1999; Hu et al., 2005). In contrast, fluoride levels in municipal a sewage treatment facility (Randolph Park Wastewater Recla-
wastewaters receiving fluoride-bearing industrial effluents are mation Facility, Tucson, AZ) and contained 4.8% VSS. An anaer-
generally low because industrial effluents will be diluted by obic glucose-degrading enrichment cultures was developed from
wastewaters from residential and industrial sources. In addi- Aviko sludge by successive transfer of the culture supernatant to
tion, significant removal of fluoride can occur by precipitation fresh glucose containing basal medium at a rate of 5% (v/v) upon
with calcium(II) (CaF2, Ksp ¼ 3.9  1011), which is a common glucose depletion. The aerobic sewage sludge, Randolph Park
wastewater contaminant. sludge II (0.13% VSS), was obtained from the aeration tank of the
Although fluoride is often present as a wastewater Randolph Park Wastewater Reclamation Facility. Sludge samples

contaminant, published data on its inhibitory effect to were stored under N2 gas in a refrigerator at 4 C.
microbial populations present in wastewater treatment
systems are very scarce. Nitrification is the only process that 2.2. Culture media
has been studied in some detail (Whang, 1985; Clarkson et al.,
1989; Carrera et al., 2003). In contrast with the limited under- The composition of the basal mineral medium supplied in the
standing of the potential inhibitory effects of fluoride on nitrification bioassays (BM-1) was (in mg/L): NaH2PO4 (1500); and
wastewater treatment microorganisms, the impact of fluoride Na2HPO4 (894). The basal mineral medium employed in the
on oral bacteria of interest to dentistry is well documented by denitrification bioassays (BM-2) contained (in mg/L): K2HPO4
a vast body of literature (Marquis et al., 2003; Wiegand et al., (250); (NH4)HCO3 (417); NaHCO3 (2678); and yeast extract (10). The
2007). Oral bacteria are inhibited by fluoride at concentrations basal mineral medium utilized in both the aerobic heterotrophic
in the range of 10–1600 mg/L. Inhibition of soil microorgan- toxicity and fermentation bioassays (BM-3) contained (in mg/L):
isms by inorganic fluoride, resulting in increased accumula- NH4Cl (280); NaHCO3 (3000); K2HPO4 (250); KH2PO4 (2050);
tion of soil organic matter, has also been reported in several CaSO4 $ 2H2O (10), MgSO4 $ 7H2O (100), and yeast extract (50).
studies (Rao and Pal, 1978; Wilke, 1987; Tscherko and Kand- The basal mineral medium used in all other anaerobic bioassays
eler, 1997). In one study microbial biomass and enzymatic (BM-4) contained (in mg/L): K2HPO4 (37); CaCl2 $ 2H2O (10);
activity in soil were decreased substantially at water- MgSO4 $ 7H2O (10); MgCl2 $ 6H2O (78); NH4Cl (669); NaHCO3
extractable fluoride concentrations exceeding 20 mg/g soil (3000); and yeast extract (20). All media were supplemented
(Tscherko and Kandeler, 1997). These findings suggest that, if with 1 mL/L of a trace element solution containing (in mg/L):
present at sufficient concentration, fluoride might have H3BO3 (50), FeCl2 $ 4H2O (2000), ZnCl2 (50), MnCl2 $ 4H2O
a negative impact on biological wastewater treatment (50),(NH4)6Mo7O24 $ 4H2O (50), AlCl3 $ 6H2O (90), CoCl2 $ 6H2O
systems. (2000), NiCl2 $ 6H2O (50), CuCl2 $ 2H2O (30), NaSeO3 $ 5H2O (100),
The objective of this study is to evaluate the inhibitory EDTA (1000), resazurin (200) and 36% HCl (1 mL/L), and they were
effect of inorganic fluoride towards the main microbial pop- adjusted to pH 7.2 with HCl or NaOH, as required.
ulations involved in organic matter and nitrogen nutrient
removal in wastewater treatment plants. 2.3. Batch microbial toxicity assays

The experimental conditions utilized in the various toxicity

2. Materials and methods bioassays are summarized in Table 1. Batch experiments were
conducted in triplicate using glass serum flasks (165 mL) sup-
2.1. Sludge sources plemented with 50 mL of medium, unless otherwise indicated.
The desired amount of fluoride was added to flasks containing
Three different mesophilic inocula were evaluated in this the growth medium and inoculum using neutralized concen-
study, including two types of methanogenic granular sludge trated stock solutions of sodium fluoride. Flasks lacking fluo-
(Eerbeek and Aviko sludge) and anaerobically digested sewage ride were also included and served as uninhibited controls. In
sludge (Ina Road sludge). Eerbeek sludge was obtained from an anaerobic and anoxic bioassays, flasks were sealed with butyl
industrial anaerobic sludge blanket (UASB) reactor treating rubber stoppers and aluminum crimp seals and, subsequently,
recycle paper effluent (Industriewater, Eerbeek, The the headspace was flushed with a N2/CO2 mixture (80:20, v/v) to
Netherlands), and the Aviko sludge from a UASB reactor exclude oxygen from the assays. The headspace of the aerobic
treating potato processing wastewater (Aviko, Steenderen, heterotrophic and nitrification bioassays was atmospheric air,
The Netherlands). Both inocula were washed and sieved to and it was replenished daily to prevent oxygen limitation.
remove fine particles and they were stored under nitrogen gas In the methanogenic assays, the methane content in the

at 4 C. The content of volatile suspended solids (VSS) in the headspace of each flask was determined periodically until 80%
Eerbeek and Aviko sludges was 12.9% and 11.5%, respectively. or more of the substrate in the controls was depleted. Subse-
The Ina Road sludge (1.5% VSS) was obtained from an anaer- quently, a second feeding of the substrate (acetate or H2/CO2)
obic sewage sludge digester at the Ina Road municipal was supplied in some assays to test the possible impact of
wastewater treatment plant, in Tucson, AZ. A thermophilic extended exposure to fluoride on the methanogenic activity of
inoculum (Hyperion sludge, 1.6% VSS) was also utilized in the the anaerobic sludge.
 water research 43 (2009) 3177–3186 3179

Table 1 – Summary of experimental conditions utilized in the various inhibition batch bioassays.
Inhibition Substrateb Substrate Inoculum Fluoride Basal Headspace Monitoring
bioassaya concn. concn. concn. mediumc contentd
(g/L-liquid) (g VSS/L) (g F/L)

Methane production H2 0.222 1.5 0–1500 BM-4 H2/CO2 (80/20 v/v) CH4
Acetate 1.88* 1.5 0–300 N2/CO2 (80/20 v/v)

Propionate- Propionate 1.32* 1.5 0–539 BM-4 N2/CO2 (80/20 v/v) Propionate, acetate,
degradation and CH4

Butyrate degradation Butyrate 0.98* 1.5 0–539 N2/CO2 (80/20 v/v) Butyrate, propionate,
acetate and CH4

Glucose fermentation Glucose 1.00–3.00 1.5 0–539 BM-3 N2/CO2 (80/20 v/v) Glucose

Aerobic glucose Glucose 1.00 1.5 0–539 BM-3 Air Glucose

degradation

Nitrification NHþ
4 0.055 0.50 0–300 BM-1 Air Ammonium

Denitrification Acetate 0.45 1.00 0–800 BM-2 N2/CO2 (80/20 v/v) Nitrate
Nitrate 1.01

* These concentrations are equivalent to 2 g COD/L.

 
a All bioassays were incubated at 30  2 C, with the exception of thermophilic methanogenic assays, which were conducted at 55  2 C.
Assays were incubated in an orbital shaker (110 rpm).
b Acetate, propionate and butyrate were added as the corresponding sodium salts, nitrate as KNO3, and ammonium as NH4(HCO3).
c The composition of the basal media is detailed in Section 2.
d The headspace pressure in all the bioassays using a N2/CO2 or air atmosphere was 102.3 kPa. When using a H2/CO2 atmosphere the pressure
was 121.6 kPa.

The maximum specific glucose consumption (mg glucose- headspace of the serum flasks was determined by gas
degraded/(g VSS d)), nitrifying (mg NHþ 4 -consumed/(g VSS d)), chromatography using an HP5290 Series II system (Agilent
denitrifying (mg NO 3 /(g VSS d)) and methanogenic (mg CH4- Technologies, Palo Alto, CA) equipped with a flame ionization
COD/(g VSS d)) activities were calculated from the slope of the detector (GC-FID). The GC was fitted with a DB-FFAP column
glucose consumption, ammonium concentration, nitrate (J&W Scientific, Palo Alto, CA) capillary column. The temper-
concentration and cumulative methane production; respec- ature of the column, the injector port and the detector was

tively, and biomass concentration versus time (d), as the mean 140, 180 and 275 C, respectively. The carrier gas was helium at
value of triplicate or duplicate assays. In each case, the a flow rate of 9.3 mL/min and a split flow of 32.4 mL/min.
maximum specific activity at a given fluoride concentration Samples for measuring methane content (100 mL) in the
was determined during the time period when the fluoride-free headspace were collected using a pressure-lock gas syringe.
control displayed maximum specific activity. The inhibition Sugars were determined colorimetrically (Dubois et al., 1956).
observed was calculated as shown below. The initial concen- Briefly, 0.5 mL of centrifuged sample was transferred into

 
Maximum Specific Activity at the Tested Concentration
Inhibition,ð%Þ ¼ 100  100,
Maximum Specific Activity of the Control

trations of fluoride causing 20%, 50% and 80% reduction in a test tube and then 0.5 mL of 5% (v/v) phenol and 2.5 mL
activity compared to an uninhibited control were referred to of concentrated sulfuric acid was added. The sample was
as IC20, IC50 and IC80, respectively. These values were calcu- vortexed and allowed to rest for 7 min. Subsequently, the tube

lated by interpolation in the graph plotting the inhibition was heated at 45 C for 20 min. The solution was allowed to
observed (expressed as percent) as a function of the inhibitor cool down and then the concentration of glucose in the
concentration. Unless otherwise indicated, reported inhibi- bioassays was determined by measuring the color intensity of
tory concentrations are average values of triplicate assays and the sample at 490 nm.
corresponding standard deviations. Nitrate, nitrite and fluoride were analyzed by suppressed
conductivity ion chromatography using a Dionex 3000 system
2.4. Analytical methods (Sunnyvale, CA, USA) fitted with a Dionex IonPac AS18
analytical column (4  250 mm) and an AG18 guard column

The concentration of acetate, propionate, and butyrate in (4  40 mm). The column was maintained at 35 C. The eluent
liquid samples as well as the methane content in the used was 10 mM KOH at a flow rate of 1.0 mL/min, and the
3180 water research 43 (2009) 3177–3186

injection volume was 25 mL. Before measurement, all samples control displayed maximum ammonium oxidizing activity
were either, centrifuged (10,000 rpm) for 10 min or passed (i.e., time 17–91 h). The normalized activity of nitrifying
through a membrane filter (0.45 mm). Routine analyses of microorganism as a function of the initial fluoride concen-
fluoride were conducted using a VWR SympHony fluoride- tration is plotted in Fig. 2A. The same procedure was utilized
selective combination electrode. Ammonium was determined to calculate the microbial activities of the different microor-
using an Orion Thermo combination ion-selective electrode. ganisms evaluated in this study. The IC20, IC50 and IC80 values
The pH was determined immediately after sampling with an determined are summarized in Table 2.
Orion model 310 PerpHecT pH-meter with a PerpHecT ROSS Fluoride was found to inhibit nitrifying microorganisms in
glass combination electrode. Volatile suspended solids and aerobic sewage sludge. Nonetheless, nitrifying bacteria
other analytical parameters were determined according to appeared to become acclimated rapidly to this contaminant.
standard methods APHA (1998). The metabolic activity determined in assays that were initially
inhibited by fluoride increased substantially with exposure
2.5. Chemicals time (Fig. 1). Assays with fluoride concentrations of 230 mg/L
and higher were completely inhibited during the initial 100 h
Potassium nitrate (>99.0% purity) and sodium acetate (99.0%) of exposure, but their activity increased sharply thereafter
were obtained from Spectrum Chemicals and Laboratory reaching levels close to those observed in the fluoride-free
Products (Gardena, CA, USA), respectively. Sodium fluoride controls. The observed recovery may be due to physiological
(99.0%), sodium nitrite (>99.5%), acetic acid (99.7%), pro- acclimatization to fluoride or to a shift in the microbial pop-
pionic acid (99.5%) and butyric acid (99.0%) were purchased ulation to another nitrifying strain which is less sensitive to
from Sigma Chemical Co. (St. Louis, MO, USA). Ammonium the contaminant.
bicarbonate was obtained from MP Biomedicals (Solon, OH,
USA). Phenol (99.0% ACS grade) was obtained from EMD
Chemicals (Gibbstown, NJ). D-Glucose anhydrous was A 100
purchased from Mallinckrodt Chemical (Paris, KY).
Activity (% of control)

80

3. Results and discussion 60

3.1. Microbial toxicity assays 40

3.1.1. Nitrification and denitrification

20
An illustrative example of the time course of ammonium
consumption in nitrifying activity assays amended with
0
fluoride concentrations ranging from 0 to 300 mg/L is shown
0 100 200 300 400 500 600
in Fig. 1. The specific nitrifying activities in treatments con-
Fluoride (mg/l)
taining fluoride were normalized based on the activity of the
control treatment lacking fluoride. In each case, the nitrifying
activity was determined during the time period when the B 100
Activity (% of the control)

80
60
NH4+ concentration (mg/l)

60

40 40

20

20
0
0 200 400 600 800
Fluoride (mg/l)

0 Fig. 2 – Inhibitory effect of sodium fluoride on the specific

0 40 80 120 160 200
ammonium-oxidizing activity of a mixed microbial culture
Time (hours)
obtained from the nitrification stage of a full-scale
Fig. 1 – Time course of ammonium consumption by municipal wastewater treatment plant (A); and on the
a mixed microbial culture obtained form the nitrification denitrifying specific activity of an industrial anaerobic
stage of a municipal wastewater treatment plant in the granular sludge as determined by nitrate depletion (B).
presence of increasing fluoride concentrations (in mg/L): Error bars (shown if larger than the symbols) represent
(C) 0,(B) 50,(J) 90,(-) 130,(,) 180,(6) 230, and (:) 300. standard deviations of duplicate assays.
Table 2 – Inhibitory effect of sodium fluoride on the key microbial populations in biological wastewater treatment systems. IC20, IC50 and IC80 are the concentrations of
fluoride causing 20%, 50% and 80% decrease in the activity of the target microbial population, respectively.
Substrate Redox conditions Inhibitory effect Inoculuma 1st feeding (mg/L) 2nd feeding (mg/L)

IC20 IC50 IC80 IC20 IC50 IC80

Mesophilic methanogens
Acetate Anaerobic Methanogenesis Eerbeek granular sludge 40.0 160.0 500.0 15.0 30.3 60.0
Acetate Anaerobic Methanogenesis Ina Road sludge 17.5 34.5 93.0 12.4 25.8 35.7

water research 43 (2009) 3177–3186

H2 Anaerobic Methanogenesis Eerbeek granular sludge 815.0 >815.0 >815.0 520.0 645.0 795.0
H2 Anaerobic Methanogenesis Ina Road sludge 30.0 82.0 390.0 805.0 1005.0 1125.0
Thermophilic methanogens
Acetate Anaerobic Methanogenesis Hyperion sludge 7.2 18.1 39.2 29.5 42.6 62.9
H2 Anaerobic Methanogenesis Hyperion sludge 218.6 432.6 >600.0 >600.0 >600.0 >600.0
Anaerobic propionate-utilizing microorganisms
Propionate Anaerobic Propionate degradation Eerbeek granular sludge 10.5 36.5 62.0
Anaerobic butyrate-utilizing microorganisms
Butyrate Anaerobic Butyrate degradation Eerbeek granular sludge 17.5 25.5 34.8
Denitrification
Nitrate Anoxic Nitrate reduction Eerbeek granular sludge >800.0b >800.0b >800.0b
Nitrification
Ammonium Aerobic NHþ 4 oxidation Randolph Park I 104.3 148.8 179.8
Heterotrophic aerobes
Glucose Aerobic Glucose degradation Randolph Park II >539.0c >539.0c >539.0c
Glucose fermenters
Glucose Anaerobic Glucose degradation Aviko granular sludge 150.5 >539.0 >539.0
Glucose Anaerobic Glucose degradation Enrichment culture 87.0 325.0 539.0

a All inocula used consisted of dispersed biomass unless otherwise indicated.

b Not toxic at the highest concentration tested, i.e., 800 mg/L.
c Not toxic at the highest concentration tested, i.e., 539 mg/L.

3181
3182 water research 43 (2009) 3177–3186

Literature data indicates that fluoride exerts relatively low assay inoculated with the dispersed enrichment culture,
inhibition towards nitrifying bacteria in activated sludge microbial growth has a minor impact on the glucose utiliza-
systems. Complete oxidation of ammonium (400 mg/L) in tion rate of the granular sludge due to the high biomass
a synthetic coal gasification wastewater was observed in concentrations utilized in the latter assay (1.5 g VSS/L). Inhi-
batch assays amended with 300 mg F/L (Whang, 1985). CSTR bition of fermentation in anaerobic wastewater treatment
experiments inoculated with a nitrifying culture enriched sludge has not been reported in the literature. Nonetheless,
from municipal sewage sludge showed that fluoride concen- there are numerous studies that confirm the toxic action of
trations of up to 2430 mg /L did only exert moderate inhibition fluoride on glucose incorporation and carbohydrate metabo-
(39%) of nitrification activity (Clarkson et al., 1989). Carrera lism by oral bacteria (Marquis et al., 2003; Wiegand et al., 2007).
et al. (2003) reported that elevated fluoride levels (630 mg/L) The minimum inhibitory concentrations reported to impair
were required to cause 50% inhibition of the nitrifying activity carbohydrate degradation by Streptococcus species and other
of biomass in a lab-scale activated sludge reactor treating oral bacteria vary widely from 5 to 1600 mg/L (Maltz and
a high-strength ammonium wastewater (546 mg NHþ 4 /L). The Emilson, 1982; Eisenberg et al., 1991; Lenander-Lumikari et al.,
wide variation observed among the various studies in the 1997; Ekenback et al., 2001; Bradshaw et al., 2002).
toxicity response of nitrification to fluoride may be due to
differences in the experimental conditions, including 3.1.4. Anaerobic propionate- and butyrate-degrading
ammonia concentration and other medium components, as microorganisms
well as variations in the microbial community structure of the Anaerobic microorganisms that utilize propionate and
inoculum, among others. butyrate were very sensitive to fluoride as evidenced by the
Regarding the toxic action of fluoride on denitrifying relatively low IC50 values determined for these microbial
microorganisms, the contaminant did not exert any signifi-
cant inhibitory effect on the activity of the mixed culture at
concentrations as high as 800 mg/L (Fig. 2B). We are not aware
of any previous study from the literature reporting on the A
100
inhibitory effect of fluoride on denitrifying bacteria during
Activity (% of control)

biological wastewater treatment. However, fluoride in

soils (up to 3,700 mg/kg) has been reported to affect nitrate 80
reduction, but the inhibitory effect varied considerably
depending on the soil characteristics (Ottow and Kottas, 1984; 60
Wilke, 1987).
These results suggest that biological processes for the 40
removal of nitrogen from wastewater are not expected to
be affected by fluoride unless the concentrations are exceed- 20
ingly high.

0
3.1.2. Aerobic heterotrophs 0 100 200 300 400 500 600
Sodium fluoride did not display any inhibitory effect towards Fluoride (mg/l)
glucose-utilizing bacteria in aerobic sewage sludge at fluoride
concentrations of up to 540 mg/L (Fig. 3A). In agreement with B
our results, the only other study assessing the impact of 100

fluoride (50–100 mg/L) on the removal of organic matter

Activity (% of control)

(ethanol and acetate) in the activated sludge process reported 80

no significant effect on cell growth and chemical oxygen
demand (COD) removal efficiency (Singh and Kar, 1989). In the 60
latter study, the presence of fluoride resulted in deterioration
of the sludge settling properties as indicated by a 100 to 200%
40
increase of the sludge volume index. The mechanisms
contributing to the deterioration of the sludge settling ability
20
in the presence of fluoride are unknown. Several studies from
the dental literature have reported that fluoride can inhibit the
formation of biofilms and other cell aggregates (Embleton 0
0 100 200 300 400 500
et al., 2001; Cao and Doyle, 2002; Marquis et al., 2003).
Fluoride (mg/l)

3.1.3. Glucose fermenting bacteria Fig. 3 – Inhibitory effect of sodium fluoride on the specific
Fluoride was inhibitory to glucose-fermenting microorgan- glucose-degrading activity of aerobic sewage sludge (A);
isms in different microbial consortia (Fig. 3B). The somewhat glucose fermentation by different mixed anaerobic culture
higher inhibition observed in the assay with the enrichment (B). Legends for panel B: Anaerobic granular sludge (-), and
culture as compared to the anaerobic granular sludge suggests an anaerobic glucose-degrading enrichment culture (C).
a greater impact of fluoride on the growth as compared to the Error bars (shown if larger than the symbols) represent
metabolic activity of glucose fermenters. In contrast with the standard deviations of triplicate assays.
 water research 43 (2009) 3177–3186 3183

populations, 36.5 and 25.5 mg/L, respectively (Table 2). As the first substrate feeding were 34.5 and 18.1 mg/L, respec-
shown in Fig. 4A and B, the utilization rate of VFA steeply tively. Mesophilic acetoclastic methanogens in granular
decreased with increasing fluoride concentrations. This sludge appeared to be more tolerant to fluoride as demon-
decrease was accompanied by a reduction in the rate of strated by the IC50 value of 160 mg/L during the same feeding
methanogenesis. The high toxicity of fluoride towards (Table 2). The impact of extending the time of exposure on the
anaerobic microbial populations involved in VFA degradation methanogenic inhibition of fluoride was investigated by
is a concern because anaerobic digestion is commonly applied supplying a second feeding of substrate to the bioassay once
for the management of excess sludge in municipal waste- that the initially supplied substrate had been depleted. The
water treatment systems (Chen et al., 2008). The inhibition of inhibitory concentrations determined for the mesophilic
acetogenic microorganisms would prevent the effective granular sludge following the second substrate feeding were
degradation of VFA (e.g. propionate, butyrate, etc.) into acetate lower and relatively similar to those determined for the
and, therefore, compromise the anaerobic digestion process. mesophilic sewage sludge (Table 2). These results suggest that
time was required to enable for fluoride penetration in the
3.1.5. Methanogens thick anaerobic granular sludge biofilm and, thus, for effective
Acetoclastic methanogens were particularly susceptible to the exposure of the methanogens to the toxicant. It is well
inhibitory effect of fluoride. Fig. 5A and B reveal a sharp established in the literature that internal mass transfer limi-
decrease in the specific acetoclastic methanogenic activity of tations can impact the transport of solutes in anaerobic bio-
mesophilic and thermophilic microorganisms in digested films (Dolfing, 1985; Pavlostathis and Giraldo-Gomez, 1991;
sewage sludge with increasing fluoride concentrations. The Wu et al., 1995; Gonzalez-Gil et al., 2001).
IC50 values determined for those microbial populations during In the case of H2-utilizing methanogens, fluoride caused
a decrease in the rate of methane production by mesophilic
and thermophilic anaerobically-digested sludges but only
when present at very high concentrations (Fig. 5C and D), and
A 100
the inhibitory effect decreased sharply with extended incu-
bation time. The IC50 values determined for those microbial
80 populations during the first substrate feeding were 433 mg/L
Activity (% of Control)

or higher (Table 2). On the other hand, hydrogenotrophic

methanogens in mesophilic anaerobically-digested sludge
60
were more sensitive to the presence of fluoride during the first
feeding, but halogen ion caused little methanogenic inhibition
40 during the second feeding at concentrations below 900 mg/L
(Fig. 5C). The mechanisms responsible for the substantial
decrease in the inhibitory effects of fluoride towards hydro-
20
genotrophic methanogens with incubation time are unclear.
To the best of our knowledge, there are no previous liter-
0 ature reports concerned with the inhibitory effects of fluoride
0 100 200 300 400 500 600 towards methanogenic microorganisms. The relatively low
Fluoride (mg/l) fluoride inhibiting values observed in this study are of
particular concern in view of the slow growth kinetics of
B 100 acetate-utilizing methanogens (doubling times range from 1
to 7 d (Mara and Horan, 2003)). An active acetoclastic meth-
anogenic population is essential for the degradation of acetate
80
which, in turn, is required to attain adequate removal of
Activity (% of Control)

biodegradable organic matter during wastewater treatment.

60 In practice, this means that fairly long time periods might be
required for the recovery from an incidental toxicity exposure.
For instance, Zayed and Winter (2000) reported that recovery
40
of methanogenesis following exposure to 10–20 mg/L of
copper was attained after 35 or 47 d after the toxic shock. The
20 apparent tolerance of H2-utilizing methanogens to fluoride
does not preclude inhibition of anaerobic sludge digestion by
this ion, since only about one-third of the electron equivalents
0
in complex organic matter are channeled through H2 gas.
0 100 200 300 400 500 600
The mechanisms responsible for the microbial toxicity of
Fluoride (mg/l)
fluoride towards acetate, propionate and butyrate-utilizing
Fig. 4 – Inhibitory effect of fluoride on the maximum microorganisms in anaerobic reactors are not well under-
specific activity of propionate- (A) and butyrate-degrading stood. Nonetheless, fluoride can bind to many microbial
microorganisms (B) in a mesophilic anaerobic consortium. enzymes, including heme-containing enzymes, copper-based
Error bars (shown if larger than the symbols) represent oxidases and other metalloenzymes, affecting metabolism
standard deviations of triplicate assays. (Hamilton and Ellwood, 1978; Marquis et al., 2003). Fluoride
3184 water research 43 (2009) 3177–3186

A C
100
120

Activity (% of control)

Activity (% of control)
80 100

60 80

60
40
40
20
20

0 0
0 50 100 150 200 250 300 0 250 500 750 1000 1250 1500
Flouride (mg/l) Fluoride (mg/l)

B D 100
100
Activity (% of control)

Activity (% of control)
80
80
60
60
40
40

20 20

0 0
0 50 100 150 200 250 300 0 100 200 300 400 500 600 700

Flouride (mg/l) Fluoride (mg/l)

Fig. 5 – Inhibitory effect of fluoride on the specific acetoclastic and hydrogenotrophic activity of mesophilic anaerobically-
digested sludge, (panels A and C, respectively), and on the specific acetoclastic and hydrogenotrophic activity of
thermophilic anaerobically-digested sludge (panels B and D, respectively). Legend: First substrate feeding (C), and second
substrate feeding (B). Error bars (shown if larger than the symbols) represent standard deviations of triplicate assays.

can form complexes with metals such as aluminum or (Frenzel and Bosse, 1996; Janssen and Frenzel, 1997; Conrad
beryllium, leading to compounds that can mimic phosphate and Klose, 1999). The different sensitivity of acetate- and
(e.g. AlF 
4 and BeF3 ), and interfere with a variety of enzymes, H2-utilizing methanogens towards CH3F is in agreement with
e.g., phosphatases and pyrophosphatases (Marquis et al., the toxic response observed in this study for these two meth-
2003). anogenic trophic groups to fluoride, suggesting the possibility
Fluoride is a potent inhibitor of pyrophosphatase (PPase) in that microbial methylation of fluoride might have occurred
various microorganisms (Tominaga and Mori, 1977; Smirnova leading to the formation of toxic CH3F. However, this biotrans-
and Baykov, 1983). PPase plays an important role in the energy formation has not been reported earlier (O’Hagan et al., 2002;
metabolism of methanogens (Roth and Bachofen, 1994; Smith Murphy et al., 2003).
and Ingram-Smith, 2007), and variation in the susceptibility to
fluoride of PPase enzymes from acetoclastic and hydro-
genotrophic methanogens might have contributed to the 4. Conclusions
considerable differences observed in the response of these
methanogens to fluoride. A PPase has been isolated from Fluoride is inhibitory towards microbial populations involved
Methanothrix soehngenii, a common acetoclastic methanogen in various metabolic steps in anaerobic digestion processes,
in anaerobic reactors that is not inhibited by fluoride (Jetten i.e., mesophilic and thermophilic acetoclastic methanogens,
et al., 1992). The composition of the medium could also affect as well as propionate- and butyrate-degraders, at concentra-
inhibition by fluoride. Studies with the methanogen, Meth- tions lower than those found in some fluoride-containing
anococcus jannaschii, have shown that prior exposure to industrial effluents. In contrast, very high concentrations of
specific metallic ions (e.g., Mn2þ or Co2þ) increased the toler- soluble fluoride (>500 mg/L) can be tolerated by microbial
ance against inhibition by sodium fluoride, to which the communities involved in the aerobic activated sludge and in
enzyme was otherwise very sensitive (Kuhn et al., 2000). denitrification processes without significant inhibitory
The inhibitory effect of organic compounds containing impact. Nitrification processes are somewhat more sensitive
fluoride such as methyl fluoride (CH3F) is well documented. but they appear to acclimate rapidly to fluoride. In conclusion,
CH3F is known to be a specific inhibitor of acetoclastic the high susceptibility of key microbial populations involved
methanogens and anaerobic mixed cultures that produce in the anaerobic metabolism of volatile fatty towards inhibi-
acetate as an intermediate, but this organofluorine compound is tion by fluoride indicates that measures to reduce the
not or only mildly inhibitory to hydrogenotrophic methanogens concentration of this ion (e.g., pretreatment of the wastewater
 water research 43 (2009) 3177–3186 3185

using ion exchange or precipitation with calcium(II), waste- Gonzalez-Gil, G., Seghezzo, L., Lettinga, G., Kleerebezem, R., 2001.
water dilution, etc.) may be required to prevent microbial Kinetics and mass-transfer phenomena in anaerobic granular
inhibition during the anaerobic treatment of fluoride-con- sludge. Biotechnol. Bioeng. 73, 125–134.
Hamilton, I.R., Ellwood, D.C., 1978. Effects of fluoride on
taining streams.
carbohydrate-metabolism by washed cells of Streptococcus
mutans grown at various pH values in a chemostat. Infect.
Immun. 19, 434–442.
Hu, C.Y., Lo, S.L., Kuan, W.H., Lee, Y.D., 2005. Removal of fluoride
Acknowledgements from semiconductor wastewater by electrocoagulation–
flotation. Water Res. 39, 895–901.
This study was conducted with financial support from the Huang, C.J., Liu, J.C., 1999. Precipitate flotation of fluoride-
Semiconductor Research Corporation/Sematech Engineering containing wastewater from a semiconductor manufacturer.
Water Res. 33, 3403–3412.
Research Center for Environmentally Benign Semiconductor
Janssen, P.H., Frenzel, P., 1997. Inhibition of methanogenesis by
Manufacturing.
methyl fluoride: Studies of pure and defined mixed cultures of
anaerobic bacteria and archaea. Appl. Environ. Microbiol. 63,
4552–4557.
Jetten, M.S.M., Fluit, T.J., Stams, A.J.M., Zehnder, A.J.B., 1992. A
references fluoride-insensitive inorganic pyrophosphatase isolated from
Methanothrix soehngenii. Arch. Microbiol. 157, 284–289.
Kuhn, N.J., Wadeson, A., Ward, S., Young, T.W., 2000.
APHA, 1998. Standard Methods for the Examination of Water Methanococcus jannaschii ORF mj0608 codes for a class C
and Wastewater. In: Clesceri, L.S., Greenberg, A.E., Eaton, A.D. inorganic pyrophosphatase protected by Co2þ or Mn2þ
(Eds.), 20th ed. American Public Health Association, ions against fluoride inhibition. Arch. Biochem. Biophys. 379,
Washington DC. 292–298.
Bradshaw, D.J., Marsh, P.D., Hodgson, R.J., Visser, J.M., 2002. Lenander-Lumikari, M., Loimaranta, V., Hannuksela, S.,
Effects of glucose and fluoride on competition and metabolism Tenovuo, J., Ekstrand, J., 1997. Combined inhibitory effect of
within in vitro dental bacterial communities and biofilms. fluoride and hypothiocyanite on the viability and glucose
Caries Res. 36, 81–86. metabolism of Streptococcus mutans, serotype C. Oral Microbiol.
Camargo, J.A., 2003. Fluoride toxicity to aquatic organisms: A Immun. 12, 231–235.
review. Chemosphere 50, 251–264. Maltz, M., Emilson, C.G., 1982. Susceptibility of oral bacteria to
Cao, J.K., Doyle, R.J., 2002. Fluoride modifies adhesion various fluoride salts. J. Dental Res. 61, 786–790.
of Streptococcus pyogenes. FEMS Immunol. Med. Microbiol. 32, Mara, D., Horan, H.J., 2003. Handbook of Water and Wastewater
175–177. Microbiology, tenth ed. Academic Press.
Carrera, J., Torrijos, M., Baeza, J.A., Lafuente, J., Vicent, T., 2003. Marquis, R.E., Clock, S.A., Mota-Meira, M., 2003. Fluoride and
Inhibition of nitrification by fluoride in high-strength organic weak acids as modulators of microbial physiology.
ammonium wastewater in activated sludge. Proc. Biochem. FEMS Microbiol. Rev. 26, 493–510.
39, 73–79. Mason, B., Moore, C.B., 1982. Principles of Geochemistry. Wiley,
Chen, Y., Cheng, J.J., Creamer, K.S., 2008. Inhibition of anaerobic New York.
digestion process: A review. Bioresource Technol. 99, Murphy, C.D., Schaffrath, C., O’Hagan, D., 2003. Fluorinated
4044–4064. natural products: the biosynthesis of fluoroacetate and
Clarkson, W.W., Collins, A.G., Sheehan, P.L., 1989. Effect of 4-fluorothreonine in Streptomyces cattleya. Chemosphere 52,
fluoride on nitrification of a concentrated industrial waste. 455–461.
Appl. Environ. Microbiol. 55, 240–245. O’Hagan, D., Schaffrath, C., Cobb, S.L., Hamilton, J.T.G.,
Conrad, R., Klose, M., 1999. How specific is the inhibition by Murphy, C.D., 2002. Biosynthesis of an organofluorine
methyl fluoride of acetoclastic methanogenesis in anoxic rice moleculedA fluorinase enzyme has been discovered that
field soil? FEMS Microbiol. Ecol. 30, 47–56. catalyses carbon-fluorine bond formation. Nature 416 (6878)
Dolfing, J., 1985. Kinetics of methane formation by granular 279–279.
sludge at low substrate concentrations. Appl. Microbiol. Ottow, J.C.G., Kottas, P., 1984. Einflup anthropogener
Biotechnol. 22, 77–81. Fluorkontamination auf den Bodenmetabolismus. Daten
Dubois, M., Gilles, K.A., Hamilton, J.K., Rebers, P.A., Smith, F., Dokumente Unweltschultz 36, 145–154.
1956. Colorimetric method for detection of sugars and related Pavlostathis, S.G., Giraldo-Gomez, E.G., 1991. Kinetics of
substances. Anal. Chem. 28, 350–356. anaerobic treatment. Water Sci. Technol. 24, 35–59.
Eisenberg, A.D., Oldershaw, M.D., Curzon, M.E.J., Handelman, S.L., Rao, D.N., Pal, D., 1978. Effect of fluoride pollution on the organic
1991. Effects of fluoride, lithium, and strontium on growth and matter content of soil. Plant Soil 49, 653–656.
acid production of mutans streptococci and Actinomyces Roth, R., Bachofen, R., 1994. Characterization and
viscous. Caries Res. 25, 179–184. partial-purification of an ATPase and inorganic
Ekenback, S.B., Linder, L.E., Sund, M.L., Lonnies, H., 2001. Effect of pyrophosphatase of the archaebacterium Methanobacterium
fluoride on glucose incorporation and metabolism in biofilm thermoautotrophicum. Biochim. Biophys. Acta Gen. Subjects
cells of Streptococcus mutans. Eur. J. Oral Sci. 109, 182–286. 1201, 271–283.
Embleton, J.V., Newman, H.N., Wilson, M., 2001. Amine and tin Shen, F., Chen, X.M., Gao, P., Chen, G.H., 2003. Electrochemical
fluoride inhibition of Streptococcus sanguis adhesion under removal of fluoride ions from industrial wastewater. Chem.
continuous flow. Oral Microbiol. Immun. 16, 182–184. Eng. Sci. 58, 987–993.
Frenzel, P., Bosse, U., 1996. Methyl fluoride, an inhibitor of Singh, M., Kar, D., 1989. Effect of fluoride on the settleability of
methane oxidation and methane production. FEMS Microbiol. biomass from an activated sludge plant treating
Ecol. 21, 25–36. petrochemical effluent. J. Ferment. Bioeng. 67, 366–368.
Fuge, R., Andrews, M.J., 1988. Fluorine in the UK environment. Skjelkvale, B.L., 1994. Water chemistry in areas with high
Environ. Geochem. Health 10, 96–104. deposition of fluoride. Sci. Total Environ 152, 105–112.
3186 water research 43 (2009) 3177–3186

Smith, K.S., Ingram-Smith, C., 2007. Methanosaeta, the forgotten Wu, M.M., Criddle, C.S., Kiceky, R.F., 1995. Mass-transfer and
methanogen? Trends Microbiol. 15, 105–155. temperature effects on substrate utilization in brewery
Smirnova, I.N., Baykov, A.A., 1983. Two step mechanism of granules. Biotechnol. Bioeng. 46 (5), 465–475.
inhibition of inorganic pyrophosphatase by fluoride. Zayed, G., Winter, J., 2000. Inhibition of methane production from
Biokhimiya 48, 1414–1423. whey by heavy metals - Protective effect of sulfide. Appl.
Sujana, M.G., Thakur, R.S., Rao, S.B., 1998. Removal of fluoride Microbiol. Biotechnol. 53, 726–731.
from aqueous solution by using alum sludge. J. Colloid Whang, G.D., 1985. Effects of fluoride, boron, arsenic and
Interface Sci. 206, 94–101. thiocyanate on ammonia and thiocyanate oxidation in
Tominaga, N., Mori, T., 1977. Purification and characterization of extended aeration activated sludge process. PhD dissertation,
inorganic pyrophosphatase from Thiobacillus thiooxidans. J. Texas University, Austin, TX.
Biochem. 81, 477–483. Wiegand, A., Buchalla, W., Attin, T., 2007. Review on fluoride-
Tscherko, D., Kandeler, E., 1997. Ecotoxicological effects of releasing restorative materials - Fluoride release and uptake
fluorine deposits on microbial biomass and enzyme activity in characteristics, antibacterial activity and influence on caries
grassland. Eur. J. Soil Sci. 48, 329–335. formation. Dental Mater. 23, 343–362.
US-EPA, 2006. Fluoride in drinking water: a scientific review Wilke, B.-M., 1987. Fluoride-induced changes in chemical
of EPA’s standards. The National Academies Press, properties and microbial activity of mull, moder and nor soils.
Washington, DC. Biol. Fert. Soils 5, 49–55.