APIZD MicROBIoLoGy, Dec. 1975, p. 905-908 Copyright 0 1975 American Society for Microbiology
Vol. 30, No. 6
Printed in U.S.A.
Removal of Algae from Florida Lakes by Magnetic Filtration G. BITTON,* J. L. FOX, AND H. G. STRICKLAND
Department of Environmental Engineering Sciences, University of Florida, Gainesville, Florida 32611
Received for publication 25 June 1975
Magnetic filtration was used for the removal of algal populations present in five lakes located in the vicinity of Gainesville, Fla. It was found that the use of this technique enabled a good removal (94%) of algal cells from three lakes where the pH was around 7. The other two lakes, with a higher pH, displayed a lower removal. However, the treatment was greatly improved by lowering the pH from 9.5 to 6.5.
Abundant algal growth in lakes and other bodies of water leads to nuisance conditions (odor, taste, turbidity, etc.) and to the deterioration of the water quality in general. Florida lakes are particularly known for their eutrophic condition. Their eutrophication is the consequence of favorable climatic conditions and enrichment by abundant quantities of nutrients, particularly nitrogen and phosphorus. Algal cells are a source of organic and inorganic pollution, and their removal is of extreme importance whenever one needs to insure the quality of water supplies in general and waste stabilization pond effluents in particular. The methods that have been proposed for the removal of algae include centrifugation, microstraining, coagulation, in-pond chemical precipitation, filtration, flotation, and ion exchange (7, 8, 12, 16). High-gradient magnetic separation has been used by various industries (15), and this prompted an investigation of the suitability of this technique in water pollution control (Mitchell et al., unpublished data; 5, 10). The process consists of seeding the polluted water with magnetite, a ferric oxide. The pollutant, under appropriate conditions, becomes associated with the iron oxide and is trapped in a filter placed in a background magnetic field. It was found that this technique was useful in the removal of phosphate (3), bacterial viruses (2), enteric bacteria, turbidity, and color (5, 13). The purpose of this research was to investigate the removal of algal populations from Florida lakes by magnetic separation under controlled conditions.
in Alachua County and range in size from 4.3 to 2,562 hectares. To assure adequate algal populations, all except Little Orange were eutrophic or senescent. Little Orange, although classified by Brezonik et al. as mesotrophic in 1969 (4), now supports dense blooms of Anabaena and Aphanizomenon. Table 1 provides the location, depth, area, and trophic status of the five lakes sampled. The samples were quickly returned to the laboratory, and their pH was measured by using a Beckman Expandomatic model SS-2 pH meter. Magnetite (300 l±g/ml) (purchased from Fisher Scientific Co., Fair Lawn, N.J.) and aluminum sulfate [Al2(SO4)3' 18H201 (50 ug/ml) (supplied by Mallinckrodt Chemical Works, St. Louis) were added to 1 liter of lake samples and mixed by using a multiple stirrer supplied by Phipps and Bird, Richmond, Va. The stirring was carried out at 100 rpm for 1 min followed by a 4-min slower mixing at 40 rpm. The flocculated mixtures were then poured through a magnetic separator, the Frantz ferrofilter model 31 purchased from S. G. Frantz Co., Trenton, N.J. This separator (Fig. 1) consists of an open bowl placed upon an electromagnet that surrounds a stainlesssteel wool filter. The electromagnet is energized by a Frantz model S-3 rectifier. The magnetite, along with any adsorbed pollutant, is trapped in the stainless-steel wool matrix and may be washed out when the electromagnet is turned off. Each filtration experiment was carried out in triplicate. After filtration, 100-ml aliquots were taken and preserved with 4% formaldehyde. Algal counts were performed by the strip count method with a Sedgwick Rafter counting cell (1). Color and turbidity were determined with a Bausch & Lomb Spectronic 88 spectrophotometer (X = 420 nm) and a Hach model 2100A
turbidimeter, respectively.
RESULTS AND DISCUSSION Water samples from Little Orange Lake, MATERIALS AND METHODS Lake Wauberg, Lake Alice, Clear Lake, and One-gallon (about 3.8-liter) water samples were Newnan's Lake were treated with 300 ,ug of taken from five lakes located in the vicinity of magnetite per ml and 50 pg of aluminum sulGainesville, Fla. The lakes chosen for study are all fate (alum) per ml and were poured through 905
906
BI1TON, FOX, AND STRICKLAND
APPL. MICROBIOL.
TABLE 1. Characteristics of Alachua County lakes studied (modified from Brezonik et al. (4) Lake
Location
Little Orange Newnan's Alice
S.E. of Hawthorne E. of Gainesville Univ. of Fla. Cam-
Clear Wauberg
S.W. Gainesville N.W. of Micanopy
Depth (m)
Area (hectares)
Trophic status
3 2 2
241 2,562 37
Mesotrophic Hypereutrophic Senescent
pus
a
la
5
4.3 103
Eutrophic Eutrophic
As determined in 1974 by Fox and Bitton. IN FL UE NT
*
1~ 'I
EFFLUENT FIG. 1. Magnetic separator used for algal moval.
re-
the magnetic separator after a 5-min mixing time. Table 2 shows the removal of algae, turbidity, and color by magnetic filtration. It was shown that the removal of algal populations was accompanied by a removal of turbidity (45 to 80% decrease) and color (48 to 83% decrease). Lake Alice, Lake Wauberg, and Little Orange Lake showed good to excellent removal of algae (94%), whereas Clear Lake and Newnan's Lake displayed relatively poor removal (64 and 55.1%, respectively). Examination under the light microscope showed that the magnetite particles were adsorbed on the sur-
face of algal cells. It was thought that the low algal removal obtained in Clear Lake and Newnan's Lake was probably due to the high pH (8.5 and 8.7) and high algal counts (1.24 x 105 and 5.18 x 10W cells/ml, respectively) in these lakes. Table 3 shows the effect of pH and magnetite concentration on the removal of algae from Newnan's Lake. The water was sampled on 5 June 1975 and was found to have a pH of 9.5 and a total algal count of 0.99 x 105 cells/ml. When the pH was decreased from 9.5 to 6.5, the algal removal increased from 61.5 to 87%. However, an increase in magnetite concentration from 300 to 500 ug/ml did not improve the removal of algae. The pH of the water is an important factor in the flocculation of algae with alum (11, 12). Golueke and Oswald (7) reported that, in the presence of alum, the removal of algae wFs maximum at pH 6.5 and decreased when the pH was higher than 7.0. The pH was also found to be important in the interaction of algal cells with polyelectrolytes (16) and with sand particles (6). Total and differential algal counts were undertaken in the five lakes. Except for Lake Alice, blue-green algae (Cyanophyta) were the dominant group in these lakes (Fig. 2). A close look at the cyanophytes shows that Microcystis was dominating in Clear Lake and Newnan's Lake. Lake Wauberg was populated by a mixture of Microcystis and Lyngbya, whereas in Little Orange Lake, Aphanizomenon and Anabaena were the dominant blue-green algae. In general, cyanophytes were the least removed when compared with chlorophytes and chrysophytes (Fig. 2). Moreover, within the phylum Cyanophyta, the filamentous organisms such as Anabaena, Aphanizomenon, or Lyngbya were better removed than the colonial types such as Microcystis. These observations are of preliminary nature because of the lack of information on the cell surface charge of blue-green algae. Most of the studies dealing with this particular subject (6, 8, 9, 14, 16) have indeed focused on green algae such as Chlorella and Scenedesmus. Therefore, more studies are being undertaken on the removal of pure cultures of
VOL. 30, 1975
907
REMOVAL OF ALGAE
TABLE 2. Removal of algae, turbidity, and color from five Florida lakes by magnetic filtrationa pH
Turbidity (NTU)
Color (color units)
Lake
Little Orange Wauberg Alice Clear Newnan's
Before
After
Before
After
Before
After
Algal removal (%)
6.6 7.0 7.2 8.5 8.7
5.9 6.7 7.5 7.7 6.7
22.0 4.9 3.7 5.5 11.0
7.3 1.2 1.7 1.1 6.0
330 101 70 120 235
100 24.6 30 20.3 120
94.0 93.5 79.2 64.0 55.1
a One-liter samples were mixed for 5 min with 300 ,ug of magnetite and 50 ,ug of alum per ml and passed through a magnetic filter.
LL
0
A
w m z
0
J
C1
CYANOPHYTA
Z
CHLOROPHYTA
am F
C HRYSO PHYTA
0
TOTAL COUNTS
-j
MISCELLANEOUS
0~~~~~~~
Vol. 30, No. 6
Printed in U.S.A.
Removal of Algae from Florida Lakes by Magnetic Filtration G. BITTON,* J. L. FOX, AND H. G. STRICKLAND
Department of Environmental Engineering Sciences, University of Florida, Gainesville, Florida 32611
Received for publication 25 June 1975
Magnetic filtration was used for the removal of algal populations present in five lakes located in the vicinity of Gainesville, Fla. It was found that the use of this technique enabled a good removal (94%) of algal cells from three lakes where the pH was around 7. The other two lakes, with a higher pH, displayed a lower removal. However, the treatment was greatly improved by lowering the pH from 9.5 to 6.5.
Abundant algal growth in lakes and other bodies of water leads to nuisance conditions (odor, taste, turbidity, etc.) and to the deterioration of the water quality in general. Florida lakes are particularly known for their eutrophic condition. Their eutrophication is the consequence of favorable climatic conditions and enrichment by abundant quantities of nutrients, particularly nitrogen and phosphorus. Algal cells are a source of organic and inorganic pollution, and their removal is of extreme importance whenever one needs to insure the quality of water supplies in general and waste stabilization pond effluents in particular. The methods that have been proposed for the removal of algae include centrifugation, microstraining, coagulation, in-pond chemical precipitation, filtration, flotation, and ion exchange (7, 8, 12, 16). High-gradient magnetic separation has been used by various industries (15), and this prompted an investigation of the suitability of this technique in water pollution control (Mitchell et al., unpublished data; 5, 10). The process consists of seeding the polluted water with magnetite, a ferric oxide. The pollutant, under appropriate conditions, becomes associated with the iron oxide and is trapped in a filter placed in a background magnetic field. It was found that this technique was useful in the removal of phosphate (3), bacterial viruses (2), enteric bacteria, turbidity, and color (5, 13). The purpose of this research was to investigate the removal of algal populations from Florida lakes by magnetic separation under controlled conditions.
in Alachua County and range in size from 4.3 to 2,562 hectares. To assure adequate algal populations, all except Little Orange were eutrophic or senescent. Little Orange, although classified by Brezonik et al. as mesotrophic in 1969 (4), now supports dense blooms of Anabaena and Aphanizomenon. Table 1 provides the location, depth, area, and trophic status of the five lakes sampled. The samples were quickly returned to the laboratory, and their pH was measured by using a Beckman Expandomatic model SS-2 pH meter. Magnetite (300 l±g/ml) (purchased from Fisher Scientific Co., Fair Lawn, N.J.) and aluminum sulfate [Al2(SO4)3' 18H201 (50 ug/ml) (supplied by Mallinckrodt Chemical Works, St. Louis) were added to 1 liter of lake samples and mixed by using a multiple stirrer supplied by Phipps and Bird, Richmond, Va. The stirring was carried out at 100 rpm for 1 min followed by a 4-min slower mixing at 40 rpm. The flocculated mixtures were then poured through a magnetic separator, the Frantz ferrofilter model 31 purchased from S. G. Frantz Co., Trenton, N.J. This separator (Fig. 1) consists of an open bowl placed upon an electromagnet that surrounds a stainlesssteel wool filter. The electromagnet is energized by a Frantz model S-3 rectifier. The magnetite, along with any adsorbed pollutant, is trapped in the stainless-steel wool matrix and may be washed out when the electromagnet is turned off. Each filtration experiment was carried out in triplicate. After filtration, 100-ml aliquots were taken and preserved with 4% formaldehyde. Algal counts were performed by the strip count method with a Sedgwick Rafter counting cell (1). Color and turbidity were determined with a Bausch & Lomb Spectronic 88 spectrophotometer (X = 420 nm) and a Hach model 2100A
turbidimeter, respectively.
RESULTS AND DISCUSSION Water samples from Little Orange Lake, MATERIALS AND METHODS Lake Wauberg, Lake Alice, Clear Lake, and One-gallon (about 3.8-liter) water samples were Newnan's Lake were treated with 300 ,ug of taken from five lakes located in the vicinity of magnetite per ml and 50 pg of aluminum sulGainesville, Fla. The lakes chosen for study are all fate (alum) per ml and were poured through 905
906
BI1TON, FOX, AND STRICKLAND
APPL. MICROBIOL.
TABLE 1. Characteristics of Alachua County lakes studied (modified from Brezonik et al. (4) Lake
Location
Little Orange Newnan's Alice
S.E. of Hawthorne E. of Gainesville Univ. of Fla. Cam-
Clear Wauberg
S.W. Gainesville N.W. of Micanopy
Depth (m)
Area (hectares)
Trophic status
3 2 2
241 2,562 37
Mesotrophic Hypereutrophic Senescent
pus
a
la
5
4.3 103
Eutrophic Eutrophic
As determined in 1974 by Fox and Bitton. IN FL UE NT
*
1~ 'I
EFFLUENT FIG. 1. Magnetic separator used for algal moval.
re-
the magnetic separator after a 5-min mixing time. Table 2 shows the removal of algae, turbidity, and color by magnetic filtration. It was shown that the removal of algal populations was accompanied by a removal of turbidity (45 to 80% decrease) and color (48 to 83% decrease). Lake Alice, Lake Wauberg, and Little Orange Lake showed good to excellent removal of algae (94%), whereas Clear Lake and Newnan's Lake displayed relatively poor removal (64 and 55.1%, respectively). Examination under the light microscope showed that the magnetite particles were adsorbed on the sur-
face of algal cells. It was thought that the low algal removal obtained in Clear Lake and Newnan's Lake was probably due to the high pH (8.5 and 8.7) and high algal counts (1.24 x 105 and 5.18 x 10W cells/ml, respectively) in these lakes. Table 3 shows the effect of pH and magnetite concentration on the removal of algae from Newnan's Lake. The water was sampled on 5 June 1975 and was found to have a pH of 9.5 and a total algal count of 0.99 x 105 cells/ml. When the pH was decreased from 9.5 to 6.5, the algal removal increased from 61.5 to 87%. However, an increase in magnetite concentration from 300 to 500 ug/ml did not improve the removal of algae. The pH of the water is an important factor in the flocculation of algae with alum (11, 12). Golueke and Oswald (7) reported that, in the presence of alum, the removal of algae wFs maximum at pH 6.5 and decreased when the pH was higher than 7.0. The pH was also found to be important in the interaction of algal cells with polyelectrolytes (16) and with sand particles (6). Total and differential algal counts were undertaken in the five lakes. Except for Lake Alice, blue-green algae (Cyanophyta) were the dominant group in these lakes (Fig. 2). A close look at the cyanophytes shows that Microcystis was dominating in Clear Lake and Newnan's Lake. Lake Wauberg was populated by a mixture of Microcystis and Lyngbya, whereas in Little Orange Lake, Aphanizomenon and Anabaena were the dominant blue-green algae. In general, cyanophytes were the least removed when compared with chlorophytes and chrysophytes (Fig. 2). Moreover, within the phylum Cyanophyta, the filamentous organisms such as Anabaena, Aphanizomenon, or Lyngbya were better removed than the colonial types such as Microcystis. These observations are of preliminary nature because of the lack of information on the cell surface charge of blue-green algae. Most of the studies dealing with this particular subject (6, 8, 9, 14, 16) have indeed focused on green algae such as Chlorella and Scenedesmus. Therefore, more studies are being undertaken on the removal of pure cultures of
VOL. 30, 1975
907
REMOVAL OF ALGAE
TABLE 2. Removal of algae, turbidity, and color from five Florida lakes by magnetic filtrationa pH
Turbidity (NTU)
Color (color units)
Lake
Little Orange Wauberg Alice Clear Newnan's
Before
After
Before
After
Before
After
Algal removal (%)
6.6 7.0 7.2 8.5 8.7
5.9 6.7 7.5 7.7 6.7
22.0 4.9 3.7 5.5 11.0
7.3 1.2 1.7 1.1 6.0
330 101 70 120 235
100 24.6 30 20.3 120
94.0 93.5 79.2 64.0 55.1
a One-liter samples were mixed for 5 min with 300 ,ug of magnetite and 50 ,ug of alum per ml and passed through a magnetic filter.
LL
0
A
w m z
0
J
C1
CYANOPHYTA
Z
CHLOROPHYTA
am F
C HRYSO PHYTA
0
TOTAL COUNTS
-j
MISCELLANEOUS
0~~~~~~~
