RESEARCH & DEVELOPMENT

REPORT NO. RD 2076

Determination of Hydrogen Sulfide Capacity of

Different Activated Carbon Types at Various Operating
Conditions

(Final Report)

Research and Development Section

Electrical & Mechanical Projects Division
Drainage Services Department

Apr 2016

Final Report endorsed by R&D Steering Committee Meeting

No. 2-2016 (E&M Part) on 1 Apr 2016
 DRAINAGE SERVICES DEPARTMENT

EVALUATION REPORT

PROVISION OF CARBON TESTING UNIT TO DETERMINE HYDROGEN SULFIDE

ADSORPTION CAPACITY OF DIFFERENT ACTIVATED CARBON TYPES AT
VARIOUS OPERATING CONDITIONS

(Final Version)

(CONTRACT NO.: DEMP/2013/06)

(HKPC REF: M:\emd\4103\Water_Team\PROJECTS\40179572\FinalReport\DEMP1306)

Prepared by

Environmental Management Division

Hong Kong Productivity Council

1st April 2016

 This summary is abridged. Names of brand and model
Executive Summary of activated carbon tested have been removed..

1. Background

1.1 Activated carbon filter technology is an adsorption process which can be used in deodorizing.
Hydrogen sulfide (H2S) generated from sewage treatment facilities is a common source of
malodour, which can be adsorbed by activated carbon. In order to eliminate the malodor, it is
valuable for Drainage Service Department (DSD) to study H2S adsorption by activated carbon.

1.2 The objectives of the study are:

(i) To design and deliver a carbon testing unit that allows concurrent evaluation of H2S
adsorption capacities of multiple activated carbons;
(ii) To evaluate the effects of varied operating conditions (that is, high and low relative
humidity) on H2S capacities of different carbons in laboratory tests and field tests to
ascertain the effect of relative humidity on H2S capacity as this information is not available
from the manufacturers;
(iii) To evaluate the effect of water wash regeneration of exhausted carbons on H2S capacities of
the two catalytic activated carbons (CAC) in laboratory tests.

2. Carbon Testing Unit

2.1 DSD has commissioned HKPC to supply a carbon testing unit, a mobile system which can be
moved to sewage treatment plants for investigations on genuine foul air, and to carry out the
laboratory study to compare the H2S adsorption capacity of different carbon at different relative
humidity (%RH). The system was first set up and fabricated in a well-ventilated workshop in
HKPC for laboratory study. Then it was moved to the Stonecutters Island Sewage Treatment
Works (SCISTW) for the field test.

2.2 The carbon testing unit in this study was composed of 7 main parts, including the Foul air
stream intake system w/ H2S gas cylinder, the Humidification system, the Dehumidification
system, the Cylindrical plastic AC filters, the Supporting framework, the Humidity and H2S Gas
detection system, and the Activated carbon scrubber.

2.3 The carbon testing unit in this study is capable of:

(i) Simulating different conditions of relative humidity (<50%, 60–70%, 70–80%, 80–90%,
i
 90-99% and 100% with water condensation);
(ii) Maintaining a steady gas stream to attain the designed face velocity and empty bed retention
time (EBRT);
(iii) Providing a steady feed of H2S to maintain an average of 90 ppmv H2S for four AC filters.

2.4 After completing the laboratory and field tests, it can be concluded that the carbon testing unit
can allow concurrent evaluation of H2S adsorption capacities of multiple activated carbons. The
designed conditions of face velocity at 0.3 m/s, EBRT 1–2 sec and inlet H2S at 90 ppmv can be
met and kept steady with a little fluctuation throughout each set of test. Besides, the carbon
testing unit is also capable of producing air of a specified %RH range. The %RH can be
controlled within the preset range most of the time in each set of test. A breakthrough time of
8–20 days for the carbon with the highest capacity was observed in each set of test.

2.5 During the laboratory tests, the carbon testing unit was modified to allow in-situ water
regeneration. The regeneration conditions were tested according to manufacturer’s
recommendations and confirmed effective.

2.6 Compared with the ASTM D6646 accelerated test with 10,000 ppmv inlet H2S and other
laboratory studies found in literature, the designed conditions of this carbon testing unit is closer
to typical design of AC filters (0.3 m/s face velocity, EBRT 2–3 sec, inlet H2S around or below
10 ppmv). Therefore, the results are expected to simulate the real applications better. The
breakthrough time is not excessively long for practical evaluation.

3. Types of Activated Carbons

3.1 Four types of carbons were selected by DSD in this study. The first activated carbon is (CG)
which is a kind of virgin activated carbon (VAC) without impregnation. The second type of
carbon tested is (IK) which is a caustic-impregnated activated carbon (IAC) in use by DSD. The
remaining two carbons are (UN) and (CA) which are catalytic activated carbons (CAC). The
two CAC carbons selected are the two major commercially available CAC which can be
regenerated with water-wash.

ii
4. Study Programme

4.1 The study consisted of the laboratory tests and field tests.

4.2 The laboratory tests were separated into two rounds. In Round 1, four activated carbons packed
in the four AC filters were tested under the relative humidity conditions of %RH <50%,
60–70%, 70–80%, 80–90%, 90–99% and 100% with water condensation.

4.3 In one test of Round 2, the AC filters broken through at %RH=100% in Test 6 of round 1 were
air dried without any water-wash to examine the performance of reviving process of wet spent
carbon.

4.4 In other tests of Round 2, the two CAC were regenerated by water wash and the regeneration
efficiency were evaluated by comparing the H2S adsorption capacities of fresh carbon, carbon
regenerated once and regenerated twice respectively.

4.5 Four sets of field tests in Round 3 were conducted at the selected site in the SCISTW of DSD to
ascertain the performance of the Carbon testing unit in field conditions. To investigate the effect
at the driest and most common moderate %RH conditions, four %RH conditions of <50%,
70–80%, 80–90% and 90–99% were selected in the field test.

5. H2S adsorption on four activated carbons in Carbon Testing Unit

5.1 The H2S breakthrough capacity of each carbon in the Round 1 laboratory tests and the Round 3
field tests under different relative humidity was averaged and shown in the figure below.

iii
Figure: Average H2S adsorption capacities of four carbons in Round 1 and 3 tests

5.2 From the above figure, it can be observed that the four carbons are useable between %RH of
60%99% (non-condensing). The order of H2S capacities are in the order of IK>UN>CA>CG.
It was also observed that the H2S adsorption capacities of IK and UN dropped significantly at a
dry condition of %RH <50%.

5.3 The adsorption capacity dropped significantly when the %RH increased from 90-99% to 100%
(with condensation) in all four types of carbons, showing that condensing conditions is
detrimental to H2S adsorption.

5.4 The H2S adsorption capacities of the four carbons were compared with the claimed value
provided by the supplier. The highest H2S adsorption capacity of IK was determined to be close
to the claimed value while that of the CAC (UN and CA) were far below their claimed values.

iv
6. Comparison with ASTM D6646 result

6.1 The H2S adsorption capacities of the four carbons determined in this study were also compared
with the ASTM D6646 results from a commercial laboratory. It is fair to assume that the results
from this study is closer to practical carbon filter application in full scale when compared to 1%
v/v (10,000 ppmv) H2S inlet concentration in the ASTM D6646 method. The capacities of the
IK and UN carbons are a little higher than the ASTM D6646 results, while that of the CA
carbon is a little lower. In contrast, the capacity of the CG carbon is higher than the ASTM
D6646 results and the difference is huge.

7. Regeneration of AC filters in carbon testing unit

7.1 The two types of catalytic activated carbon (CAC), UN and CA carbons are claimed to be
regenerable and one can recover their H2S adsorption capacities by simple water wash process.
A standard in-situ regeneration procedure including a water-washing step and air-drying step
was developed in the laboratory test based on the procedures recommended by the suppliers.
Both CACs could be regenerated. H2S capacity of the fresh carbon and the regenerated carbon
were compared in the laboratory tests with two regeneration cycles.

7.2 The regeneration efficiency of CA was higher than that of UN. The recovery of 88–100% in CA
and 73–85% in UN were determined from the first two regenerations of their spent carbon
respectively.

8. Revival of AC filters in carbon testing unit

8.1 The H2S capacities was hindered in the presence of water condensation, nevertheless, true
capacity are not depleted and the carbons can be revived by air drying without water wash. The
H2S adsorption performance of the four revived carbons was evaluated in Test 7 laboratory test
after the air-drying process. The adsorption capacities of the UN, CG, and CA carbons were
fully recovered after the reviving process while large reduction was observed for IK.

v
9. Selection of Activated Carbon for full-scale AC filters

9.1 It will be cost effective to use the IAC carbon for H2S removal due to its high adsorption
capacity and low price. However, the CAC carbons were demonstrated to be regenerable to
recover their H2S adsorption capacity assuming 8588% recovery after each regeneration. The
ultimate H2S adsorption capacity of UN and CA after regeneration were determined to be higher
than that of IAC and the unit cost of these two CACs per g H2S adsorbed is about 80% higher
than that of IAC if CAC are regenerated 6 times before replacement (within 10 times
recommended by manufacturer and 6 times would be practical).

9.2 Among the four activated carbons, unit costs by weight are in order CG<IK<CA<UN. Results
on cost effectiveness (in terms of gH2S/cc carbon) showed that IAC is the recommended carbon
type and CG is about twice the material cost. However, if bed fire safety issue is a great concern
and system suspension for regeneration of CAC is acceptable, CAC is also suggested.

9.3 It is important to stress that the above discussions and recommendations are based on the results
from the four selected carbon only. It is uncertain if other members of the same type will behave
the same.

vi