A Study on the Optimum Backwashing Method applied to Activated Carbon Process in Waterworks

Journal of Water and Environment Technology, Vol.1, No.2, 2003

A Study on the Optimum Backwashing Method applied to Activated
Carbon Process in Waterworks
Bok-Sil Ko, Ho-Souk Yoon, Sin-Jung Park, Min-Hye Yoon,
Teak-Gyu Kwon, Sun-Koog Kwon and Jong-Woo Kim*
Maegok Water Purification Plant, Water Quality Research Institute*,
Daegu Metropolitan City

Abstract
For applying the optimum backwash method to activated carbon absorption process, this study had
performed an efficiency test of backwash method and a test for determination of backwash period at the M
water purification plant in Daegu metropolitan city.
The minimum fluidization velocity was different according to kinds of carbon like spent carbon and
reactivated carbon. Changing water position before backwashing was more efficient in backwashing than
controlling backwash time. In the case of water position LL(a height of 60cm over the outer layer of
activated carbon) before backwashing, the most efficient backwash method has turned out to be 10 min. of
air wash and 18 min. of water wash.
The turbidity of activated carbon filter outflow water and organic matter change have no big difference
according to the days of seasonal operation after backwashing. As backwash period is very related to
microbiological growth and is influenced by outflow water change, the study has found that it's desirable to
operate in consideration of HPC(Heterotrophic plate counter) distribution of filtered outflow water, water
quality, the condition of a filter basin and the years of activated carbon use.
-------------------Key Words: Optimum backwashing method, Minimum fluidization velocity, Heterotrophic plate counter,
Activated carbon

Ⅰ. Introduction
In advanced water purification, the absorption process of granular activated carbon removes, very
efficiently, not only taste, smell, or color, but every kind of pollutants such as DBPs(Disinfection
By-Products), BDOC(Biodegradable Dissolved Oxygen Carbon), SOCs(Synthetic Organic
Chemicals), and VOCs(Volatile Organic Compounds) organic matter of a small amount in water1).

Granular activated carbon has many angles and irregular shape, and can cause some problems. So it
may create mudball; may leak minute activated carbon and microorganism; its low specific gravity
causes loss in backwashing. Therefore, it requires proper management and careful operation2).
Generally, determining the date of backwashing in sand filter basins is based on the head loss of a
filter layer, the leakage turbidity of processed water, and filter duration. But the quality of water
flowing into the filter basin of granular activated carbon is mostly stable because it has passed
- 189 -


Journal of Water and Environment Technology, Vol.1, No.2, 2003

through sand filtration and later ozone processing; it has a little suspension and head loss doesn't
increase greatly according to filter duration. Therefore, it isn't enough to determine the date of
backwashing only by the head loss of a activated carbon layer and the turbidity of processed
water3).
In filter process, backwashing makes suspension in filter medium dropped off and removed from
filter medium by proper wash methods; can increase filter efficiency after sufficient washing, and
improve productivity because of increase in filter duration and decrease in backwash frequency.
However, insufficient wash effect lessens filter duration, deteriorates the quality of filtered water
because of leaked suspension, and causes other problems, which can have a direct influence on the
quality of purified water.
Thus, with granular activated carbon absorption process of the M water purification plant in
Daegu Metropolitan City as the subject of examination, the study has compared backwash
efficiency according to backwash methods, and analyzed filtered outflow water according to
operation time after backwashing in order to extract factors necessary for determining the optimum
backwash period as a base for efficient management of advanced water purification facilities.

Ⅱ. Experimental conditions and methods
The study selected 5 basins(2 reactivated, 2 spent carbon basins, and 1 virgin carbon basin) of
granular activated carbon from 24 ones in the M water purification plant in Daegu; tried to find out

the optimum condition by changing the time and method of seasonal backwash from October, 2001
to September, 2002.
1. Specifications and operation conditions of Granular activated carbon
24 granular activated carbon contact basins consist of 4 buildings each of which has 6 stationary
downward filter basins. The rate of activated carbon and sand is 250:20(㎝); the under drainage
system is strainer-type. Backwashing uses both air wash and water wash; air wash velocity is 0.83
㎥/min·㎡ and water wash velocity 0.4㎥/min·㎡(Table1).
Table 1. The

present condition of granular activated carbon contact basin facilities
Division

Conditions
3

The charge amount and indexes of activated carbon
The method of current method

250m (8mⅹ12.5mⅹ2.5m), 24 basins
Stationary downward current

Empty bed contact time

10minutes

LV and SV

15m/hr and 6l /hr

Backwash

conditions

Air wash velocity

0.83m3/(min.·m2)

Water wash velocity

0.40m3/(min.·m2)
- 190 -


Journal of Water and Environment Technology, Vol.1, No.2, 2003

2. The characteristics of a granular activated carbon
In a granular activated carbon contact basin, virgin carbon is domestic activated carbon made
from palm shell, reactivated carbon means activated carbon produced in the compound regenerative
facilities, and spent carbon is activated carbon used for over 3 years. The specification of three
carbons are expressed in Table 2.
Table 2. The specification of activated carbons
Spent*

Reactivated**

Virgin**

(m2/g)

1,066

1,157

1,030

MB absorption

(㎎/g)

100

190

150

Iodine value

(㎎/g)

680

1,060

1,130

Items
Specific surface area

* Spent : examined in October, 2001 ** Reactivated, Virgin: examined in June, 2001

3. Experimental methods

1) An efficiency test of backwashing
In order to examine backwash efficiency in a granular activated carbon contact basin, the study
has measured the minimum fluidization velocity and backwash discharged-water turbidity of
backwashing by changing backwash methods as in Table 3. And through a test of the minimum
fluidization, the study has measured head loss values, and regarded as the minimum fluidization
velocity the time when their measurements are constant.
Table 3. Backwash methods in a granular activated carbon contact basin
Process

Methods
8 min. of air wash and 18 min. of water wash

Backwash time change

Water position
backwashing

change

12 min. of air wash and 20 min. of water wash
before

A height of 110㎝ over the outer layer of activated
carbon(water position L).
A height of 60㎝ over the outer layer of activated
carbon(water position LL)

2) A test of determining the date of backwashing
In order to determine the proper date of backwashing for a granular activated carbon contact
basin, the study has divided 4 seasons like this - spring(March to June), summer(July to September),

autumn(October and November) and winter(December to February); at the beginning of every
season, for 10 days the study just picked outflow water from 3 basins(spent, reactivated and virgin
carbon) every day and examined 6 items like turbidity while operating and not backwashing them.
The analysis of filtered outflow water was based on the official test methods of water
pollution4), Standard Methods5), and the Japanese waterworks test methods6; each-item analysis
equipment and test methods are as follows:

- 191 -


Journal of Water and Environment Technology, Vol.1, No.2, 2003

(1) Turbidity
On picking water, turbidity was measured by Turbidimeter(HACH 2100)
(2) UV254
UV254 was analyzed at 254nm by UV/Vis spectrophotometer(JASCO V-560).
(3) KMnO4 consumption
KMnO4 consumption was experimented in accordance with the official test method of water
quality.
(4) TOC(Total organic carbon)
On picking water, TOC was measured by TOC analyzer(SHIMAZU 5000A).
(5) THMFP(Trihalomethane formation potential)
Until free residual chlorine became 1.0∼2.0㎎/l , chlorine was poured in; pH was controlled
into 7±0.2 by phosphoric acid buffer solution; it was settled at 20±1℃ for 24±1 hours. Then
the remaining chlorine quantity was measured; the water was picked into 50㎖ vial. Right
after that, arsenious acid sodium and phosphoric acid(1+10) were added there and THMs were
measured; the early THM values were deducted from their measurements and the remaining
values were THMFP (Purge&Trap/HP5890 GC).
(6) HPC(Heterotrophic plate counter)
On picking water, a sample of 1㎖ was diluted step by step and put on R2A agar; it was

cultured at 20±1℃ for 7 days; HPC was measured.
(7) The quantity of germs attached to activated carbon
A sample was picked by an activated carbon picker inserted, by less than 1m, into the filter
layer of a granular activated carbon contact basin once every month; picked granular carbon of
50g was put into 100㎖ of sterilized and distilled water; while the water was stirred for 1 min.,
the carbon was washed 5 times and then dried naturally for about 4 hours. After that, 20㎖ of
sterilized saline solution was poured to the dried activated carbon of 1g, and the carbon was
processed ultrasonically(40㎑, 180W) for 5 min.; 1㎖ of the sample was diluted step by
step in R2A; was cultured for 7 days at 20±1℃.

The germ quantity attached to activated

7,8,9)

carbon was expressed the number of germs per 1g

.

Ⅲ. Results and consideration
1. The results of an efficiency experiment according to backwash methods
The turbidity of discharged water from backwashing is used as one of the important factors
evaluating backwash efficiency. Generally, increase in water temperature needs raising backwash
velocity10,11), but the backwash equipment of granular activated carbon in the M water purification
plant is uncontrollable because the condition of air is fixed in 0.83㎥/min·㎡ and that of water in
0.4㎥/min·㎡. Also, the water-position regulator was divided into 4 steps like LL(a height of 60cm
over the outer layer of activated carbon), L(a height of 110cm over it), H(a height of 210cm over
- 192 -


Journal of Water and Environment Technology, Vol.1, No.2, 2003

it), and HH(a height of 300cm over it); until now, backwashing has been performed at L water
position. Accordingly, as a method for raising backwash velocity according to increase in water
temperature, water position before backwashing will be controlled downward to LL and backwash
effect be improved.
1) The turbidity of discharged water according to backwash time and changing water position
before backwashing
Table 4 shows the maximum turbidity of discharged water caused by change in backwash time.
The maximum turbidity of discharged water from backwashing by air for 12 min. and by water for
20 min. at the L water position before backwashing was 4.2∼17.3NTU and higher than that from
backwashing by air for 8 min. and by water for 20 min. It was 4.2NTU in spent carbon with 4
days of backwash period, 17.3NTU in spent carbon with 6 days, 7.6NTU in reactivated carbon with
6 days, and 7.3NTU in reactivated carbon with 8days.
Table 4. The turbidity of discharged water by change in backwash time
The maximum turbidity of discharged
water (NTU)
Increase
and
8-min, of air wash
12-min. of air
and 18-min. of
wash and 20-min. decrease
water wash
of water wash

Division

4days of back wash
6days of back wash
8days of back wash

Spent

27.3

31.5

+ 4.2

Spent

31.2

48.5

+17.3

Reactivated

45.6

53.2

+ 7.6

Reactivated

53.5

60.8

+ 7.3

Remarks
*Water temp. at
the time of
measurement

: 4∼9℃

Table 5 shows the turbidity of discharged water caused by change in water position before
backwashing. The maximum turbidity of discharged water from backwashing by air for 12 min.
and by water for 20 min. was 15.6∼40.2NTU and higher at LL water position than at L water
position. And it was 15.6NTU in spent carbon with 4 days of backwash period, 18.3NTU in spent,
25.7NTU in reactivated carbon with 6 days, and 40.2NTU in reactivated with 8 days.
Table 5. The turbidity of discharged water according to changing water position
before backwashing
The maximum turbidity of
discharged water(NTU)
Water position L
before backwashing

Water position LL

before backwashing

Increase
and
decrease

Spent
Spent
Reactivated

31.5
48.5
53.2

47.1
66.8
78.9

+15.6
+18.3
+25.7

Reactivated

60.8

101.0

+40.2

Division
4days of back wash
6days of back wash
8days of back wash

- 193 -

Remarks
*Water temp. at
the time of
measurement
: 4∼9℃


Journal of Water and Environment Technology, Vol.1, No.2, 2003

Fig. 1 shows changes in the turbidity of discharged water from backwashing caused by backwash
time and water position change before backwashing. All the 4 basins, experimental targets, were
the most efficient in backwashing by air for 12 min. and by water for 20 min. As a result, when
backwash efficiency is evaluated by the turbidity of discharged water, it is judged to be more
efficient by controlling water position before backwashing than by controlling backwash time.
S p e n t c a rb o n ( 4 d a y s o f b a c k w a s h p e rio d )

Turbidity(NTU)

120
100
80
60
40
20
0
2

6

10

15

16

17

W a te r p o s itio n L *

18
19
21
B a c k w a s h tim e ( m in )

23

W a te r p o s itio n L *

25

27

29

31

29

31

29

31

W a te r p o s itio n L L **

S p e n t c a rb o n ( 6 d a y s o f b a c k w a s h p e rio d )

Turbidity(NTU)

120
100
80
60
40
20
0
2

6

10

15

16

17

W a te r p o s itio n L *

18
19
21
B a c k w a s h tim e ( m in )

23

W a te r p o s itio n L *

25

27

W a te r p o s itio n L L **

R e a c tiv a te d c a rb o n ( 6 d a y s o f b a c k wa s h p e rio d )

Turbidity(NTU)

120
100
80
60
40
20
0
2

6

10

15

16

17

W a te r p o s itio n L *

18
19
21
B a c k w a s h tim e ( m in )

23

W a te r p o s itio n L *

25

27

W a te r p o s itio n L L **

R e a c tiv a te d c a rb o n ( 8 d a y s o f b a c k w a s h p e rio d )

Turbidity(NTU)

120
100
80
60
40
20
0
2

6

10

15

16

17

W a te r p o s itio n L *

*: 8 min. of air wash,

18
19
21
B a c k w a s h tim e ( m in )
W a te r p o s itio n L *

18 min. of backwash **: 12 min. of air wash

Fig.1. The turbidity of discharged
changing water position before backwashing

water

- 194 -

23

25

27

29

31

W a te r p o s itio n L L **

20 min. of backwash

according

to

backwash

time

and


Journal of Water and Environment Technology, Vol.1, No.2, 2003

2) The minimum fluidization velocity according to changing water position before
backwashing
The minimum fluidization velocity, that at the beginning of fluidization, is also the smallest
velocity for expanding filter medium2). The point of the minimum fluidization was when air inflow
gradually increases loss head and then its difference keeps constant without increasing. Table 6
shows the results of examining the minimum fluidization by 12-minute air wash or 20-minute water
wash. In order to reach the minimum fluidization at the L water position, spent carbon has 6min. of
air wash and reactivated has 10min. of air wash. At the LL water position, spent carbon requires
4-minute air wash; reactivated carbon 8-minute air wash.
Like this, difference in the point of the minimum fluidization between spent and reactivated
carbon results from the height of an activated-carbon layer and the condition of activated carbon,
etc. Also, backwashing at the LL water position required less time for the minimum fluidization
than at the L water position; the loss head of both spent and reactivated carbon at the LL water
position was 12㎝ higher than at the L water position.
Generally, backwashing is done by higher than the minimum fluidization velocity; increasing
water temperature needs much more increase in backwash velocity because of decreasing water
viscosity and lessening attraction between filter media.
Thus, lowering water position before backwashing, not backwashing by extended time, lessens
time for the minimum fluidization, which can raise backwash velocity, strengthen its force over the
filter medium of activated carbon, and lessen the loss of activated carbon by backwashing.
The above results put together, the method of increasing backwash effect is thought to lower
water position(LL) before backwashing and to backwash at over the minimum fluidization velocity.
Reactivated carbon can reach the minimum fluidization by 8-minute air wash; as 18 min. of water
wash goes down to less than 5NTU, 10-minute air wash and 18-minute water wash has turned out

to be proper.
Table 6. The
backwashing

minimum

fluidization

according

Spent carbon(㎝)
Division

to

changing

water

position

before

Reactivated carbon(㎝)

Water position L Water position LL Water position L Water position LL
before
before
backwashing before backwashing backwashing before backwashing

Air wash 0min

0

0

0

0

Air wash 2min

30

44

22

28

Air wash 4min

34

48

30

38

Air wash 6min

36

48

31

41

Air wash 8min

36

48

32

48

Air wash 10min

36

48

36

48

Air wash 12min

36

48

36

48

- 195 -

Remarks

*Water temp. at
the time of
measurement
: 4∼9℃


Journal of Water and Environment Technology, Vol.1, No.2, 2003

2. The results of an experiment for determining backwash period
The test of determining backwash period has compared the changes of water quality factors and
tried to find out the right date of backwash and control factors, while operating and not
backwashing for 10 days every early season.
1) Changes in the water-quality of raw water
Table 7 show changes in the water quality of raw water. Changes in water temperature are
obviously different according to seasons: the average tmeperature of autumn is 13.4℃; that of
winter is 3.2℃; that of spring 10.2℃; that of summer 27℃.

pH was 7.8∼8.6 on the average,

and especially 8∼9 in spring; it's because the dry season can produce a very large quantity of algas.
And the density of chlorophyll-a has been found the highest as 67.5ppb. The average turbidity
ranged from 8 to 18NTU as a typhoon, rainfall, and more caused high turbidity. Fluctuations in
TOC, THMFP, UV254, KMnO4 consumption were comparatively high in spring and summer
influenced by the dry season, rainfall, etc.; the consumption of KMnO4 averages 6.3∼9.1㎎/l ;
UV254 averages 0.040∼0.068㎝-1; TOC averages 2.74∼3.53㎎/l; THMFP averages 0.0820∼
0.1388㎎/l.
Table 7. Changes in the water quality of raw water
Division
Average
Autumn Maximum
Minium
Average
Winter Maximum
Minimum
Average
Spring Maximum
Minimum
Average
Summer Maximum
Minimum

Water
temperatur
( ℃ )
13.4
18.4

11.0
3.2
5.9
0.8
10.2
13.5
7.2
27.0
30.0
24.8

pH
7.8
8.4
7.3
8.0
8.2
7.6
8.6
9.0
8.1
8.0
8.5
7.1

Turbidity Chlorophyll-a
(ppb)
(NTU)
8
33

5
11
44
4
13
17
10
18
66
9

18.6
26.1
10.8
6.1
6.9
4.6
67.5
113
35
21.3
40.1
7.9

KMnO4
comsuption
(㎎/l )
6.5
9.6
5.3

6.3
12.9
4.7
9.1
10.6
7.0
8.4
13.7
5.6

UV254
-1
(㎝ )

TOC
(㎎/l )

THMFP
(㎎/l )

0.047
0.068
0.030
0.040
0.067
0.032
0.045
0.056
0.036
0.068

0.094
0.055

2.74
3.22
2.00
3.18
4.06
2.77
3.53
3.80
3.17
3.39
4.16
3.13

0.0890
0.1210
0.0650
0.0820
0.1080
0.0559
0.1388
0.2249
0.1090
0.0878
0.1170
0.0728

2) Changes in the water quality of outflow water from activated-carbon filtration according to days

of operation after backwashing
(1) Changes in turbidity
Fig. 2 shows the turbidity of outflow water from spent-, reactivated-, and virgin-carbon filtration,
almost the same as or a little lower than that of ozonized water: it's 0.08∼0.13NTU in spring; 0.06
∼0.10NTU in summer; 0.06∼0.10NTU in autumn; 0.06∼0.11NTU in winter.

Particularly,

turbidity is somewhat higher in spring than in any other season, which results from gradual rise in
water temperature and a very large quantity of generation of algas during the dry season; that seems
to require much care in waterworks. However, considering the above-mentioned results, the
- 196 -


Journal of Water and Environment Technology, Vol.1, No.2, 2003

turbidity of outflow water from activated carbon-filtration has little change according to days of
operation after backwashing; turbidity can't be a factor of operation in determining the date for
backwashing.

Turbidity(NTU)

0.16
0.12
0.08
0.04
0.00
1

2

3

4

5

6

7

8

9

10

1

2

3

4

5

Autumn

6

7

8

9

10

1

2

3

4

Winter

Days
R ea ctiv a ted carbo n

S pent carbo n

5

6

7

8

1

2

3

Spring

4

5

6

7

8

Summer

Virg in ca rbo n

O zo nized water

Fig. 2. Changes in the turbidity according to days of seasonal operation
(2) Changes in organic matter
Fig. 3 shows changes in organic matter according to days of seasonal operation after
backwashing. As for spent carbon, the removal rate of KMnO4consumption, UV254, TOC,

THMFP, etc. has turned out to be just 10% by examination; the low rate has resulted from a
falling-off in absorption. In autumn, 3 months after the beginning of operation, reactivated carbon
has, in removal rate, 50% of KMnO4 consumption, 43% of UV254, 40% of TOC, 27% of THMFP
and virgin carbon has, in removal rate, 45% of KMnO4 consumption, 33% of UV254, 39% of TOC,
25% of THMFP; the removal rate of reactivated carbon is higher than that of virgin carbon. The
longer days of operation, the less removal rate. During the 10-day operation after backwashing, the
removal rate of organic matter had little difference.
As a result, the removal efficiency of KMnO4 consumption, TOC, UV254, and THMFP seems to
be directly influenced by the quality of flowing-in water and the degree of breakdown of activated
carbon more than by days of operation after backwashing; it's improper to see changes in the
removal rate of organic matter as a source determining the time for backwashing.
K M n O 4 c o n s u m p tio n

Removal rate(%)

80. 0

40. 0

0. 0
1

2

3

4

5

6

7

8

9

10

1

2

3

4

Autumn

5

6

7

8

9

10

1

2

3

W inter

4

5

6

7

8

1

2

Spring

3

4

5

6

Summer

Da ys
S pent c arbo n

R eac tivated c arbo n

Virgin c arbo n

Fig. 3. Changes in the removal rate of organic matter according to days of seasonal operation

- 197 -

7

8


Journal of Water and Environment Technology, Vol.1, No.2, 2003

UV

254

7

8

Removal rate(%)

80. 0

40. 0

0. 0
1

2

3

4

5

6

7

8

9

10

1

2

3

4

5

6

Autumn

9

10

1

2

3

4

5

W inter

6

7

8

1

2

3

Spring

4

5

6

7

8

6

7

8

6

7

8

Summer

Da ys
S pent c arbo n

R eac tivated c arbo n

Virgin c arbo n

TO C
Removal rate(%)

80. 0

40. 0

0. 0
1

2

3

4

5

6

7

8

9

10

1

2

3

4

5

6

Au tu mn

7

8

9

10

1

2

3

4

W in te r

5

6

7

8

1

2

3

S p rin g

4

5

S u mme r

Da y s
S pe nt c a rbo n

R e a c tiv a te d c a rbo n

Virgin c a rbo n

THM FP

Removal rate(%)

80. 0

40. 0

0. 0
1

2

3

4

5

6

7

8

9

10

1

2

3

4

5

A u tu m n

6

7

8

9

10

1

2

W i n te r

3

4

5

6

7

8

1

2

S p ri n g

3

4

5

Summer

Da ys
S p e n t c a rb o n

R e a c tiv a te d c a rb o n

V irg in c a rb o n

Fig. 3. Changes in the removal rate of organic matter according to days of seasonal operation
(3) HPC changes
Fig. 4 shows the results of HPC according to days of seasonal operation after backwashing; in
autumn, spent carbon has 570∼7,400 CFU/㎖; activated has 420∼5,200CFU/㎖; virgin has 380∼
8,300CFU/㎖.

In winter, spent carbon has 2,200∼12,300CFU/㎖ in HPC; reactivated has 1,400

∼9,400CFU/㎖; virgin has 1,500∼8,700 CFU/㎖. In spring, the dry season, over-generation of
algas and activated carbon-attached germs(Table 8) seem to get the number of flowing-out germs to
greatly increase: spent carbon has 20,200∼97,500 CFU/㎖; reactivated has 7,400∼90,300CFU/㎖;
virgin has 13,600∼95,300CFU/㎖.

In summer, in HPC, spent carbon has 3,100∼7,200CFU/㎖;

reactivated has 2,400∼8,800CFU/㎖; and virgin has 2,300∼8,400CFU/㎖. The average HPC of
flowing-out water from activated carbon-filtration is 1.1∼1.5ⅹ104CFU/㎖, the same as that, 1.5∼

3.0ⅹ104CFU/㎖, examined by Mr. Park et al.,(2001)9), but lower than that by Servails et
al.,(1991)12), 4.1ⅹ104∼1.5ⅹ107CFU/㎖.

The above experimental results put together,

seasonlessly, on the first day after backwashing, HPC increases and then gradually decreases; the
longer days of operation, the higher HPC. That seems to have some relation to the growth of
- 198 -


Journal of Water and Environment Technology, Vol.1, No.2, 2003

microorganism in the absorption of activated carbon; therefore, prevention of flowing-out
microorganisms by their over-propagation requires periodical backwashing based on observation of
HPC.
As a result, in autumn as microorganisms begin to increase on the 6th day of operation, autumn
needs backwashing on the 5th or 6th day; winter needs backwashing on the 4th or 6th day of
operation. In spring, HPC begins to increase on the 4th day, but spring has much more
flowing-out germs, so backwashing should be done on the 2nd or 4th day. Summer has a little
fewer flowing-out germs; however, high turbidity of water flows in because of typhoon or the rainy
season, which results in rise in water temperature and the density of organic matter. Considering
these factors, summer requires backwashing on the 3rd or 5th day of operation. On the other hand,
spent carbon has a little more flowing-out germs than reactivated or virgin carbon does; it's
desirable to make the backwashing date in spent carbon earlier than in activated or virgin carbon.

HPC(CFU/ml)

1 .0 E + 0 5

1 .0 E + 0 4

1 .0 E + 0 3

1 .0 E + 0 2
1

2

3

4

5

6

7

8

9

10

1

2

A utum n

3

4

5

6

7

8

9

10

1

2

3

W in t e r

4

5

6

7

8

1

S p rin g

2

3

4

5

6

7

8

S um m er

D a ys
S p e n t c a rb o n

R e a c tiva te d c a rb o n

Virg in c a rb o n

Fig. 4. Changes in HPC according to the days of seasonal operation after backwashing
(4) The number of activated carbon-attached germs
The number of granular activated carbon-attached germs has been found as 1.4ⅹ105∼5.8ⅹ
108 CFU/g in case of spent carbon, 1.1ⅹ105∼1.6ⅹ108 CFU/g in case of reactivated carbon, and
6.6ⅹ105∼5.8ⅹ108 CFU/g in virgin carbon. The number with these three kinds of activated
carbon was comparatively high in November, March to April, and September; decreases from
December when water temperature began to fall; is the lowest in January to February and a little
low in July to August with over 30℃ of water temperature(Table 8, Fig. 5). This is higher than
the number by Mr. Kim(1995)13), 106∼107CFU/g, and similar to that by Mr. Park et al.,(2001)9),
0.6∼9.8ⅹ108CFU/g.
The number of germs attached to reactivated or virgin carbon usually remains over 106CFU/g
from the 60th day of operation on, so its granular carbon contact basin was found to be managed by
BAC; the number of granular activated carbon-attached germs has revealed that free residual
chlorine or residual ozone doesn't influence BAC process very much. In spring with increasing
granular activated carbon-attached germs, HPC flowing into the filter water of granular activated
carbon also increases; therefore, it's necessary to advance the date of backwashing, to observe if
backwashing is going well, and to take care of the existing process including a sand filter basin.

- 199 -


Journal of Water and Environment Technology, Vol.1, No.2, 2003

Table 8. The monthly number of activated carbon-attached germs
The number of attached germs (CFU/㎖)
Spent carbon
Reactivated carbon
Virgin carbon
7.3 106

2.7 106
5.4 106
8
8
1.2 10
1.6 10
3.2 108
7
7
3.7 10
5.3 10
1.0 107
2.0 106
4.5 106
1.3 107
6
5
1.4 10
1.1 10
6.6 105
5.8 108
1.1 108
5.8 108
7
8
5.6 10
1.4 10
5.6 107
6
6

6.4 10
5.3 10
3.1 106
1.1 107
2.9 106
2.1 107
8
8
1.0 10
1.1 10
5.0 107
1.8 107
1.3 107
3.8 107

Division
2001. 10
2001. 11
2001. 12
2002. 1
2002. 2
2002. 3
2002. 4
2002. 7
2002. 8
2002. 9
2002. 10

HPC(CFU/g)

1 .E + 0 9
1 .E + 0 8
1 .E + 0 7
1 .E + 0 6
1 .E + 0 5
1 .E + 0 4
10

11

12

1

2

3

4

7

8

9

10

Month
S pent c arb on

R eac tivated c arbon

V irg in c arb on

Fig. 5. Monthly changes in the number of activated carbon-attached germs

Ⅳ. Conclusion
For the efficient operation of granular-activated-carbon absorption process in advanced water
purification, the study had performed a efficiency test of backwashing and a
backwash-period-determining test, with 5 granular activated carbon(spent, reactivated, and virgin
carbon) basins as the subjects of study, among 24 ones in the M water purification plant in Daegu
metropolitan city, from October 2001 to September 2002. And its results are as following;
1. An efficiency experiment according to backwash methods
1) The turbidity of flowing-out water according to changes in backwashing has proved to be 4.2∼
17.3NTU higher in 8-minute air wash or 18-minute water wash than 10-minute air wash or
20-minute water wash.
2) The turbidity of flowing-out water according to changes in water position before backwashing
has turned out to be 15.6∼40.2NTU higher at water position LL before backwashing than at water
position L.
- 200 -


Journal of Water and Environment Technology, Vol.1, No.2, 2003

3) The point of the minimum fluidization differed according to kinds of activated carbon like spent
or reactivated carbon: at the L water position before backwashing, spent carbon(the 16th basin)
required 6 min. of air wash, and reactivated carbon(the 19th) 10 min. of air wash; at the LL, spent
carbon(the 16th) required 4 min. of air wash, and reactivated carbon(the 19th) 18 min. of air wash.
4) Backwashing should be done at over the minimum fluidization velocity, but changing water

position could raise the efficiency of backwashing more than controlling backwash time; the LL
water position before backwashing had 10-minute air wash and 18-minute water wash as the most
proper condition.
2. A test determining backwash period
1) The turbidity of outflow water from activated-carbon filtration and changes in organic matter
had no big difference according to days of seasonal operation after backwashing.
2) Changes in HPC according to days of operation after backwashing was a little high on the first
day after backwashing, and decreased gradually; the more days of operation, the higher HPC.
Particularly, HPC was the highest in spring.
3) The number of granular activated carbon-attached germs was 1.1ⅹ105∼5.8ⅹ108CFU/g, and
granular activated carbon proved to be operated by BAC. The number of spent carbon-attached,
reactivated carbon-attached, and virgin carbon-attached germs were high in November, March, and
September; the lowest in February; especially high from March to April.
4) The date of backwashing was closely connected with the growth of microorganism, and greatly
influenced by changes in water quality; therefore, HPC range, the condition of water quality, the
condition of a filter basin, and the years of activated-carbon use should be considered for operation.

References
1. Carlson, M.A. et al., "Comparing two GACs for Adsorption and Biostabilization", J. AWWA,
86(3): 91-102. 1994.
2. Cho, W.H. et al., A Study on the Optimization of Activated-carbon Backwashing. Water
technology Research Institute of Seoul Metropolitan Government in Korea, 219-275. 1999.
3. The Ministry of Environment, Development of Automation and Related Facilities for Water
System Using Ozone and Activated Carbon. Korea, 858-884. 2001.
4. The Ministry of Environment, The Official Test Method of Drinking-Water Quality. Korea,
1999.
- 201 -


Journal of Water and Environment Technology, Vol.1, No.2, 2003

5. APHA, AWWA and WPCF, Standard Methods of the Examination of Water and Wastewater.
17th ed., Washington, D.C., USA., 1987.
6. The Japanese Waterworks Society, The Japanese Test Method of Waterworks. 1985.
7. Rice, RG. and C.M. Robson., Biological Activated Carbon. Lewis publishers, Boca raton, Florida.
1982.
8. Nagazawa, The Movement of Microorganism in a Granular Activated Carbon Layer. The 41st
Japanese waterworks research announcement meeting. 1990.
9. Park, HG. et al., Improving Water Quality and Bacterial Characteristics during water Treatment
Process Using Biologica Activated Carbons on Downstream of the Nakdond River. J. Korean
Environmental Sciences, 10(2): 105-111. 2001.
10. Kim, B.G., A Study on the Efficiency of Backwashing in the Adoption Process of Activated
Carbon. Kyeongnam university graduate school. 2001.
11. Yu, P.J., A Study on the Optimum Backwash Velocity by Water Temperature Change. Seoul
industrial university graduate school. 2000.
12. Servais, P., G. Billen, C. Ventresque and G.P. Bablon., Microbial activity in GAC filters at the
Choisy-Roi treatment plant. J. AWWA. 75: 62-68. 1991.
13. Kim, J.Y., A Study on the Intensive Waterworks of Raw Water by Activated Carbon., Dohoku
university, Japan. 1995.
14. Park, J.H., Drinking Water Microbiology. Chemical Engineering Corporation, Korea. 109-110.
1990.

- 202 -