JOURNAL OF BIOSCIENCE AND BIOENGINEERING Vol. 87, No. 2, 258-260. 1999

LETTER TO THE EDITOR

Lighted Upflow Anaerobic Sludge Blanket SHIGEKI SAWAYAMA, * TATSUO YAGISHITA, AND KENICHIRO TSUKAHARA

Biomass Division, National Institute for Resources and Environment, AIST, MITI, 16-3 Onogawa, Tsukuba, Ibaraki 305-8569, Japan

Received 9 September 1998/Accepted 23 October 1998

The uptiow anaerobic sludge blanket (UASB) method has been developed as an efficient anaerobic wastewater treatment process; however, the performance of this process in the removal nitrogenous compounds and phosphate is not high. Here, we present the water treatment performance of a lighted upflow anaerobic sludge blanket (LUASB) reactor and propose a novel LUASB concept. A population of phototrophic bacteria was induced from UASB granules under light conditions (100 ,&.m-2-s-1). The ammonium and phosphate ion removal efficiencies of the LUASB reactor were higher than those of a UASB reactor. The difference in the results from runs under light and dark conditions suggests that the efficiencies of ammonium and phosphate ion removal were improved by an increase in the phototrophic bacteria in the LUASB reactor. The UASB granule can decompose a variety of organic substances: therefore, the LUASB method could also be effective for producing phototrophic bacterial biomass and polyhydroxyalkanoates (PHAs) from various wastewaters.

[Key words: anaerobic digestion, lighted UASB method, phototrophic bacteria, wastewater treatment]

Because anaerobic digestion has the advantages of a small amount of sludge production, low energy consump- tion, and methane production, this process has been widely studied and implemented for the treatments of organic wastes and wastewater (1). However, the conven- tional contact process of anaerobic digestion cannot deal with a high rate of organic loading and is not highly stable. The UASB concept was developed to resolve these problems (2, 3). This concept is based on the formation of well-settling granules under dark conditions, natural agitation caused by gas production, and a well-designed gas-solids separator; however, the UASB reactor cannot efficiently remove nitrogenous compounds and phos- phate (4). Improvement in phosphate removal using the UASB method by the addition of Ca2+ and Mg2+ to the influent has been reported (5), but there have been no reports on a biological removal method. Biomass produc- tion during wastewater treatment may be necessary to simultaneously remove nitrogenous compounds and phos- phate. Non-sulfur purple phototrophic (photosynthetic) bacteria used to aerobically and anaerobically decom- pose organic compounds and to simultaneously consume nitrogenous and phosphate ions have been studied in wastewater treatment systems; however, phototrophic bacteria can only decompose a limited number of organic substances (6). The effect of light on the UASB method has not been actively studied (l), and there appears to be a possibility of improving the inorganic compound removal performance in UASB reactors under light con- ditions.

In the study reported here, two cylindrical glass reac- tors (80x 180mm, Able, Tokyo), each having a volume of 850ml were respectively used as LUASB and UASB reactors. To each reactor were added 530ml of the same UASB seed granules, kindly provided by Ebara Co., Kanagawa. The lower part of the glass surface (up to 10.5 cm from the bottom) of the LUASB reactor and the

* Corresponding author.

entire glass surface of the UASB reactor were covered with aluminum foil to prevent light penetration (Fig. 1).

Both the LUASB and UASB reactors were supplied with two kinds of media: (i) an organic medium made up of sodium acetate 2.5 g-f-l (Wako Pure Chemical In- dustries, Osaka), sodium lactate 1.25 g .I-’ (Wako), and sodium propionate 1.25 g .I-’ (Wako); and (ii) a starch medium consisting of starch soluble 4 g . f-l (Wako). The following chemicals were respectively added to these two media: NH&l 200 mg.I-I, KH2P04 16 mg. f--l, CaClz. 2Hz0 25 mg.f-l, MgC12.6Hz0 25 mg.l l, Fe-EDTA 30 mg . I-- l, CoCl, . 6Hz0 5 mg . I-- I, MnClz . 4Hz0 5 mg . I-- I, and yeast extract (containing 10.2 wt% of nitrogen and 0.99 wt% of phosphorus) 300 mg .I-*.

The reactors were maintained at 35( + l)‘C with con- tinuous incandescent light illumination of 100 ,DE .mm~2.

Gas -

glass reactor

UASB granules

FIG. 1. Schematic diagram of the laboratory-scale LUASB reac- tor. The reactor was maintained at 35(? 1)“C with continuous incan- descent light illumination of 100 ,uE. rnmzt s-l and was supplied with the organic acid and starch media.

258

LETTER TO THE EDITOR 259

FIG. 2. View of the LUASB reactor under light conditions. The reactor was supplied with the starch medium.

s-l and were supplied with the respective medium at a flow rate of 600 ml.d-l (retention time, 0.9 d). The or- ganic acid medium was supplied to the LUASB reactor from days 1 to 60 and the starch medium from days 61 to 98. When the TOC concentration in the effluent was stable, the reactor was thought to be at a steady state. The LUASB reactor was placed in the dark for 24 h on operational days 1-19, 31, 37, 38, 45, 52, 60, 89, 90, 97, and 98. Effluent samples were collected at room tempera- ture for 3 h and then centrifuged (MOOxg, 5 min) before carrying out the analyses. Any phototrophic bacteria at- tached on the inside glass surface of the LUASB reactor were wiped off once a week to keep the glass transpar- ent.

FIG. 3. Changes in removal ratios of TOC, ammonium and phosphate, and in the bacteriochlorophyll concentration in the effluent from the LUASB reactor. Symbols: n , operational days when the LUASB reactor was placed in the dark for 24 h; 0, TOC removal ratio; q , ammonium removal ratio; a, phosphate removal ratio. Line: ----, bacteriochlorophyll concentration,

The NH4+, N02-, and POd3- concentrations were calorimetrically determined with Nessler reagent, sul- fanilamide and N-1-naphthylenediamine dihydrochlo- ride, and ammonium molybdate, antimony1 potassium tartrate and ascorbic acid, respectively (7). The NOs- concentration in the effluent was determined by a nitrate electrode (93-07; Orion, USA). The concentrations of

organic carbon, inorganic carbon, and total organic car- bon (TOC) were determined using a TOC meter (TOC- 5000A; Shimadzu, Kyoto). The biogas yield from the bioreactor was monitored by the displacement of a satu- rated sodium chloride solution. The biogas composition was determined using a gas chromatograph (GC-8A; Shimadzu) with a WG-100 column (GL Sciences, Tokyo) at 50°C. Bacteriochlorophyll was extracted with acetone- methanol and its concentration was measured using a spectrophotometer (120A; Shimadzu) (8). The absorp- tion spectrum of the effluent was also ascertained with a spectrophotometer (16OOPC; Shimadzu). The dissolved oxygen concentration in the reactor was monitored using an oxygen electrode (CSP-2; Able).

The color of the liquid in the lighted area and gran- ules in the upper surface area of the LUASB reactor turned to red under light conditions (Fig. 2). The LUASB reactor was placed under light conditions from day 20 and bacteriochlorophyll was found in the effluent

TABLE 1. Removal of TOC, NH4+, and Pod3 by the LUASB reactor under light conditions

5- %?I 2.

4 . 2 8

3=r c”

2 3 8 a

1 ‘; 8 2

02 20 30 40 50 60 70 80 90 100

Operational days

’ Organic acid medium ’ Starch medium I

TOC concentration

(mg C.l-r)

Removal efficiency

TOC (%I

NH4+ concentration

(mg N./-l)

Removal Removal efficiency PO& efficiency Methane

NH4+ concentration PO& yield (%I (mg P . I-r) (%I (ml. g-r-added TOC)

Influent: organic acid 13961444 - 52.1-54.3 4.1-4.9 - - medium

Effluent from 48.3a 96.5 36.5 30.0 51.3 721 LUASB reactor (4.4)b (0.3) (3.3) (6.3) (E) (33.0) (307)

Effluent from 45.8 96.8 53.4 UASB reactor (11.7) (0.8) (4.5) (s’.;‘, (E)

- 12.2 885 (12.7) (60)

Influent: starch medium 1364-1383 - 50.6-53.4 - 4.4-4.5 -

Effluent from 69.8 95.0 ,,a;

87.2 LUASB reactor (21.2) (1.5) (12.7) (Z)

25.0 558 (38.7) (80)

Effluent from (“64;

96.8 28.3 (2) (E)

-80.4 726 UASB reactor (0.5) (7.3) (32.8) (41)

a Average values in effluent. The average values of the TOC concentration, ion concentrations, and removal efficiencies in the effluent from the LUASB reactor were calculated from 23-25 successive measurements under only light conditions. The average values of the UASB reactor were calculated from 15 successive measurements.

b Values in parentheses are standard deviations.

260 SAWAYAMA ET AL. J. BIOSCI. BIOENG.,

FIG. 4. Absorption spectra of the effluent from the LUASB reactor. A: Spectrum of the effluent from the LUASB reactor using the

300 400 600 800 Wave length (nm)

1000

organic acid medium as influent under light conditions. B: Spectrum of the effluent from the LUASB reactor using the starch medium as influent under light conditions. C: Spectrum of the effluent from the LUASB reactor using the organic acid medium as influent before light conditions. The absorption maxima of the purple nonsulfur photo- trophic bacterium R. capsulatus are 311, 482, 514, 593, 809, and 866 nm (9).

after day 23 (Fig. 3). On the other hand, bacterio- chlorophyll was not detected in the effluent from the UASB reactor through the entire incubation period. The TOC removal efficiency of the LUASB reactor was 94- 97% under light conditions using first the organic acid and then the starch medium as the influent and these efficiencies were the same for the UASB reactor (Table 1). The ammonium and phosphate ion removal efficien- cies of the LUASB reactor using the organic acid and starch media were higher than those of the UASB reac- tor (Table 1). The high phosphate concentration in the effluent from the UASB reactor compared with that in the influent could be mainly caused by conversion of the phosphorus contained in the yeast extract (3.0 mg P./-r) and partial release of phosphate ion from the UASB granules (Table 1). The concentrations of nitrate and nitrite ions in the effluent from the LUASB and UASB reactors were under 0.1 mg N .I-’ through the entire incu- bation period. The dissolved oxygen concentration in the LUASB reactor was maintained at 0.0 mg. 1-l through the entire incubation period.

The bacteriochlorophyll concentration in the effluent from the LUASB reactor decreased under dark condi- tions and the removal efficiency of ammonium was also low compared with that under light conditions (Fig. 3). The methane yield from the LUASB reactor was lower than that from the UASB reactor (Table 1). This differ- ence could be caused by the phototrophic bacterial con- sumption of organic substrates, probably organic acids.

The absorption spectra of the effluent from the LUASB reactor (Fig. 4) were similar to that of the purple nonsul- fur phototrophic bacteria Rhodobactor capsulatus (9). These results suggest that phototrophic bacteria were induced in the LUASB reactor from the UASB granules under light conditions, and consumed organic substances and ammonium and phosphate ions.

The mechanism of this coexistence of methane fermen- tation and the growth of phototrophic bacteria in one reactor is ecologically interesting. The LUASB method could be useful for the production of phototrophic bac- terial biomass, PHAs, molecular hydrogen, and other materials as well as wastewater treatment. There is the possibility that a group of specific microorganisms that consume organic acids is induced in the UASB reactor under specific conditions.

We are grateful to Ms. Tae Kimura and Ms. Yukiko Fukuda for their technical assistance.

REFERENCES

1. Fannie, K. F., Conrad, J. R., Srivastava, V. J., Jerger, D. E., and Chynoweth, D. P.: Anaerobic processes. J. Water Poll. Control Fed., 55, 623-632 (1983).

2. Lettinga, G., van Velsen, A. F. M., Hobma, S. W., de Zeenw, W., and Klapwijk, A.: Use of the upflow sludge blanket (USB) reactor concept for biological wastewater treatment, especially for anaerobic treatment. Biotechnol. Bioeng., 22, 699-734 (1980).

3. Lettinga, G. and Pol, L. H.: Advanced reactor design, opera- tion and economy. Wat. Sci. Tech,, 18, 99-108 (1986).

4. Lettinga, G. and van Haandel, A. C.: Anaerobic digestion for energy production and environmental protection, p. 817-839. In Johansson, T. B., Kelly, H., Reddy, A. K. N., and Williams, R. H. (ed.), Renewable energy sources for fuels and electricity. Island Press, Washington D.C. (1993).

5. Chang, Y. and Nishio, N.: Removal of POa and NH4 from wastewaters by UASB methanogenic fermentation. J. Ferment. Bioeng., 77, 450-452 (1994).

6. Sasikala, C. and Ramana, C. V.: Biotechnological potentials of anoxygenic phototrophic bacteria. I, p. 173-226. In Neidle- man, S. L. and Laskin, A. I. (ed.), Advances in applied micro- biology, vol. 41. Academic Press, San Diego (1995).

7. APHA, AWWA, WPCF: Standard methods for the examina- tion of water and wastewater, 17th edition. American Public Health Association, Washington D.C. (1989).

8. Cohen-Bazire, G., Sistrom, W. R., and Stanier, R. Y.: Kinetic studies of pigment synthesis by non-sulfur purple bacteria. J. Cell. Comp. Physiol., 49, 25-68 (1957).

9. Pfennig, N. and Trtiper, H. G.: The Family Chromatiaceae, p. 3200-3221. In Balows, A., Truper, H. G., Dworkin, M., Harder, W., and Schleifer, K. (ed.), The prokaryotes, 2nd edition. Springer-Verlag, New York (1991).