Você está na página 1de 5

Energy 35 (2010) 4531e4535

Contents lists available at ScienceDirect

Energy
journal homepage: www.elsevier.com/locate/energy

Biogas quality upgrade by simultaneous removal of CO2 and H2S in a packed


column reactor
N. Tippayawong*, P. Thanompongchart
Department of Mechanical Engineering, Faculty of Engineering, Chiang Mai University, Chiang Mai 50200, Thailand

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 13 October 2009
Received in revised form
5 April 2010
Accepted 7 April 2010
Available online 10 May 2010

Biogas from anaerobic digestion of biological wastes is a renewable energy resource. It has been used to
provide heat, shaft power and electricity. Typical biogas contains 50e65% methane (CH4), 30e45%
carbon dioxide (CO2), moisture and traces of hydrogen sulphide (H2S). Presence of CO2 and H2S in biogas
affects engine performance adversely. Reducing CO2 and H2S content will signicantly improve quality of
biogas. In this work, a method for biogas scrubbing and CH4 enrichment is presented. Chemical
absorption of CO2 and H2S by aqueous solutions in a packed column was experimentally investigated.
The aqueous solutions employed were sodium hydroxide (NaOH), calcium hydroxide (Ca(OH)2) and
mono-ethanolamine (MEA). Liquid solvents were circulated through the column, contacting the biogas in
countercurrent ow. Absorption characteristics were examined. Test results revealed that the aqueous
solutions used were effective in reacting with CO2 in biogas (over 90% removal efciency), creating CH4
enriched fuel. H2S was removed to below the detection limit. Absorption capability was transient in
nature. Saturation was reached in about 50 min for Ca(OH)2, and 100 min for NaOH and MEA, respectively. With regular replacement or regeneration of used solutions, upgraded biogas can be maintained.
This technique proved to be promising in upgrading biogas quality.
2010 Elsevier Ltd. All rights reserved.

Keywords:
Biogas
Chemical absorption
Fuel upgrade
Gas scrubbing
Renewable energy

1. Introduction
Renewable energy deriving from biomass sources has great
potential for growth to meet our future energy demands. Biogas is
a very important source of renewable methane. It is produced from
anaerobic biodegradation of biomass in the absence of oxygen and
the presence of anaerobic microorganisms. Anaerobic digestion is
the consequence of a series of metabolic interactions among
various groups of microorganisms. The process is carried out in
digesters that are maintained at temperatures ranging from 30 to
65  C. Biogas is rich in CH4 with typical range between 40 and 70%,
and its lower heating value is between 15 and 30 MJ/Nm3. In
Thailand, biogas resources are from industrial wastewater and live
stock manure, which have potential of 7800 and 13,000 TJ/year,
respectively [1]. Conversion of chemical energy in biogas to heat or
electricity is possible via combustion. Apart from direct combustion
in burners or boiler units, gas engines are usually employed as
prime movers in utilization of biogas [2e4]. There is even greater
potential for biogas if it can be made viable as a transport vehicle

* Corresponding author.
E-mail address: n.tippayawong@yahoo.com (N. Tippayawong).
0360-5442/$ e see front matter 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.energy.2010.04.014

fuel. Biogas may be made transportable via pipelines, or by compressing in gas cylinders. This is possible only after removing CO2,
H2S and water vapour. In all cases, quality of biogas is crucial in both
its CH4 content and purity. CO2 is present in large quantities in
biogas and is inert in terms of combustion. Its presence therefore
decreases the energetic content of biogas. The CH4 content in
biogas will be increased if CO2 is removed. The most common
contaminant in biogas is H2S and other S-containing compounds
that come from S-bearing organic matters [5]. Depending on the
composition of the organic matter, the H2S content in biogas can
vary from 100 to 10,000 ppm. This contaminant is highly undesirable in combustion systems due to its conversion to highly corrosive and environmentally hazardous compounds. Its removal is
essential before any eventual utilization of biogas.
Reducing CO2 and H2S content will signicantly improve the
quality of biogas. Various technologies have been developed for
separation of CO2 from gas streams in the past. These include
absorption by chemical solvents, physical absorption, cryogenic
separation, membrane separation and CO2 xation by biological or
chemical methods [5e7]. There are also a number of techniques to
remove H2S. Examples are chemical absorption in aqueous solutions, physical absorption on solid adsorbents and conversion to
base S or low solubility metal sulphides [8,9]. These techniques are

4532

N. Tippayawong, P. Thanompongchart / Energy 35 (2010) 4531e4535

CO2 RR0 -NH / RR0 -NCOOe H

Nomenclature
A
A0
C
Ci
k
t

(4)

In the aqueous solution, the dissolved CO2 reacts with alkaline


or amine following a complex reaction mechanism. This depended
strongly on the pH, CO2 concentration and other factors. In this
work, the temperature, pressure and ionic strength of solutions
were assumed to be constant. The packed column reactor was
considered as a continuously stirred tank reactor and the rate of
CO2 absorption can then be obtained through a material balance for
a complete mixing reactor. Correlation may be developed for
empirical modeling of absorption process. The fraction of CO2 in the
gas mixture that is absorbed or removed by the solution in the
packed column at time t was denoted as;

fraction of specic gas absorbed or removed from


the gas mixture after time t (e)
fraction of specic gas absorbed or removed from
the gas mixture after t s (e)
outlet CO2 concentration (% v/v)
inlet CO2 concentration (% v/v)
absorption constant (e)
absorption time (s)
characteristic absorption time (s)

of signicant industrial importance, generally applied during


natural gas sweetening as well as in the removal of CO2 from ue
gases of power plants. Apart from the use of gas permeation
membranes [10], applications of such technologies to biogas have
been quite rare. To the authors knowledge, there were relatively
few reports on biogas purication and upgrading, especially in
small-scale applications. The majority of previous studies were
carried out on water scrubbing systems [11,12], which is the
simplest and cheapest method involving the use of pressurised
water as an absorbent. The process performance is dependent on
factors such as the dimensions of the scrubbing tower, gas pressure,
composition of raw biogas, water ow rates and purity of water
used. It is possible to produce high quality CH4 enriched gas from
biogas by chemical absorption where a packed bed column and
bubble column are normally used to provide liquid/gas contact [13].
However, there remain several drawbacks of using liquid solutions
for CO2 and H2S removal. These include high energy requirement
for regeneration, stability and selectivity of chemicals used, environmental impact from waste liquids, requirement for large
equipment sizing and high equipment corrosion rate. Even though
this conventional method is complex and expensive, it may be
adopted and modied for employment in small-scale biogas
production. There is a paucity of literature available on small scale
sizing of biogas upgrading systems. This work is among the rst
attempt to investigate the technical feasibility of this concept and
its application to a small farm.
The objective of this work was therefore to determine technical
feasibility of a chemical absorption system designed for small-scale
biogas production. In this work, absorption by chemical solvents was
used to capture CO2 and H2S in biogas. A laboratory scale, packed
column reactor was used to investigate absorption performance by
NaOH, Ca(OH)2 and MEA solutions at different gas-to-liquid ow
rates. The main objective was to evaluate chemical absorption
technology for biogas purication, and to generate data for scale-up
calculation and techno-economic analysis of the process.

A 1

C
Ci

(5)

where Ci and C are, respectively, the inlet CO2 concentration and


CO2 concentration in the gas exiting the column after time t. The
removal rate is a transient effect for this particular setup. Rate of
declining CO2 absorption was assumed to be proportional to the
CO2 fraction that was absorbed in the packed column, and the
CO2 fraction that passed through. It can then be expressed
as [15];

dA
kA 1  A
dt

(6)

Integration of the above equation yields:



A 1  A0
ln
ks  t
A0 1  A
After rearrangement, we get:



1
C
s
ln
k
Ci  C

Chemical absorption is an efcient technology for the removal of


CO2 and H2S from gas mixture. In an absorption column, the
pollutant is transferred from the gas to the gas/liquid interface, and
then to the bulk of the liquid phase, where reactions take place.
Alkaline and alkanolamines are among the popular reagents for
practical applications. In the case of CO2 absorption, the following
reactions take place [14]:
(1)

e
CO2 CO2e
3 H2O / 2HCO3

(2)

e
CO2 R-NH2 H2O / R-NH
3 HCO3

(3)

(8)

where k is an absorption constant and s is the characteristic


absorption time when 50% absorption of CO2 occurs. This implies
that the solutions in the packed column should be completely
saturated after elapsed time of 2s.

2. Theory

CO2 2OHe / CO2e


3 H2O

(7)

Fig. 1. Experimental apparatus.

N. Tippayawong, P. Thanompongchart / Energy 35 (2010) 4531e4535

3. Experimental
The semi-batch experimental apparatus is shown in Fig. 1, with
(1) gas inlet measurement point, (2) ow meter, (3) ow distributor, (4) packed column, (5) packing material, (6) outlet measurement point, (7) are, (8) solvent container, (9) drainage, (10) pH
meter, (11) pump, (12) ow meter, (13) pressure gauge, (14)
injector. The apparatus consists of a Pyrex glass cylinder of 70 mm
in diameter and 1.0 m high. The absorption column was randomly
packed with a packing material (commonly known by the trade
name plastic bioball, having an overall spherical shape, with
uniform and structured spikes around the body and high surface
area to volume ratio of 1895 m2/m3, as shown in Fig. 2) to a height
of 700 mm, to facilitate gas/liquid interaction. These bioballs are
normally used in anaerobic lter reactors for biogas production.
The packed bed was found to have a void fraction of 0.916. NaOH,
CaO and MEA were employed in the present investigation. They
were obtained from Fisher Scientic. Biogas was obtained from
a local chicken farm. At the beginning of a typical run, 10 l of the
liquid solvent was placed in the vessel. The biogas purication
process took place in the packed column reactor where the gas was
continuously fed from the bottom of the column, and the liquid
solvent was sprayed from the top, creating counter current ow.
The solvent was circulated between the column and a liquid vessel
by means of a peristaltic pump. The liquid ow rate was controlled
to maintain a smooth liquid lm. Gas ow rates were regulated
using a gas ow meter. The pressure was slightly above atmospheric level. The CO2 and H2S concentrations in the biogas
entering and exiting the column were constantly monitored by an
IR gas analyzer. The pH of the liquid solvent was measured using
a pH meter. All measurements were repeated at least three times.

Table 1
Compositions of biogas immediately after the treatment with different solvents.

CH4
CO2
H2S

(%)
(%)
(ppm)

Inlet

NaOH

Ca(OH)2

MEA

53.1
46.8
2150

95.5
3.2
0

95.0
4.0
0

98.0
1.3
0

proceeded with time, the CO2 and H2S were continuously accumulated in the solvents. Due to its much higher concentration in
comparison with H2S, the CO2 started to evolve in the outlet stream
after a certain time. The end of each run was determined when the
liquid solvent became completely saturated or neutralized (pH
7e8). The corresponding breakthrough curves obtained are presented in Fig. 3, showing variation of the dimensionless CO2
concentration and changes in pH level with time. It was found that
NaOH, Ca(OH)2 and MEA showed similar patterns. NaOH and MEA
were found to become saturated in about 100 min. However, rate of
Ca(OH)2 saturation appeared to be much faster than for the other
two. This was observed after around 50 min, compared with about
100 min observed for NaOH and MEA. It should be noted that MEA
possessed lower basicity than NaOH and Ca(OH)2 but its rate of
change in pH level was also slower. The relatively fast saturation
time realised in this investigation may be attributed to the fact that
biogas used here contained very high concentration (47%) of CO2. It
should be noted that each data presented was an average value.

1.0
0.8

4. Results and discussion

0.6
C/C0

Aqueous solutions of NaOH, Ca(OH)2, and MEA were used as


chemical solvents to demonstrate the experimental apparatus
function and ability to absorb CO2 and H2S. Table 1 shows
composition of biogas downstream of the experimental setup after
about three minutes of operation. The packed column reactor was
observed to simultaneously remove high proportion of CO2 and H2S
(over 90% removal efciency), resulting in CH4 enriched biogas.
However, this was a time dependent process. These gaseous
concentrations were found to decrease with time. This effect will be
discussed later in the section. Initially, the liquid solvents reacted
rapidly with, and almost completely absorbed CO2 and H2S. Their
concentrations at the outlet stream were practically very small
compared to their original values. As the absorption process

4533

0.4
NaOH
Ca(OH)2
MEA

0.2
0.0

20

40

60

80

100

120

Time (min)

13.0
12.0

NaOH
Ca(OH)2

11.0

MEA

pH

10.0
9.0
8.0
7.0
6.0

20

40

60

80

100

120

Time (min)

Fig. 2. Plastic bioball as packing material.

Fig. 3. Variations of normalized outlet CO2 concentration and pH of the solutions with
time; solvent concentration of 0.1 M, gas to solvent ow ratio of 1.0, 47% inlet CO2
concentration and temperature of 30  C.

4534

N. Tippayawong, P. Thanompongchart / Energy 35 (2010) 4531e4535

Time (min)

120
100

NaOH
Ca(OH)2

80

MEA
NaOH
Ca(OH)2

60

MEA

40
20
0
-2.0

Fig. 5. Total CO2 absorption and loading by different chemicals used.

-1.0

0.0

1.0

2.0

3.0

ln[C/(Ci-C)]
Fig. 4. Kinetics of CO2 absorption by different solvents: solvent concentration of 0.1 M,
gas to solvent ow ratio of 1.0, 47% inlet CO2 concentration and temperature of 30  C.

Their coefcient of variations, dened as a ratio between standard


deviation and mean value were found to be within about 10%.
Plots of absorption time against dimensionless biogas concentration for CO2 absorption by NaOH, Ca(OH)2 and MEA are shown in
Fig. 4. These curves may be approximated by straight lines, as
shown by Eq. (6). These linear relationships were obtained within
a different range of operation times for different solutions used. The
ranges were 100, 50, and 110 min for NaOH, Ca(OH)2 and MEA,
respectively. Table 2 lists model parameters for these aqueous
solutions, as well as correlation coefcient R2 of each linear
approximation. For NaOH and Ca(OH)2, R2 values of greater than
0.96 were obtained, showing good linear relationships. However,
an R2 value of only 0.84 was obtained for MEA. It appeared that the
MEA curve was not exactly linear. A single linear relationship may
not be sufcient to describe the behaviour of MEA. This observation
was in line with the previous reported work [15] on absorption
with primary and tertiary amines, where two line sections were
proposed. The reasons for the observed characteristics of MEA as
well as for the difference from NaOH and Ca(OH)2 are not yet
known. From the reaction kinetic parameters of NaOH and linearly
approximated MEA, they were found to be similar. The saturation
time of the solutions in the packed column was predicted to be 2s.
While saturation time for Ca(OH)2 observed from the experiment
was closely predicted by the theoretical model, signicant theoretical underestimations of saturation time were observed for
NaOH and MEA.
With respect to absorption capacity of the solvents, Fig. 5
depicts the total amount of CO2 absorbed by these solutions in kg
CO2 per kg chemicals used. Error bars show standard deviations of
each case. It was observed that the total amounts of CO2 absorbed
were in the range between 0.18 and 0.22 kg/kg, with highest
loadings occurring for NaOH solution. The better loading for NaOH
over MEA did not imply that NaOH was a better absorbent for CO2
removal. Other properties such as cost, ability to regenerate,
chemical stability, etc. must be taken into account. For a packed bed

column reactor technology, comparison of CO2 absorption capacity


with data from various references are summarised in Table 3. It can
be seen that most studies were undertaken at relatively low CO2
concentrations of about 15% or less, compared to that found in
typical biogas (>40%). Various chemicals used include NaOH, NH3,
MEA and MDEA (methyl-diethanolamine). Results obtained from
this work appeared to be in similar range to those in the literature.
With respect to MEA, our results were observed to be lower than
those from Lin and Shyu [15] and Aroonwilas et al. [17] results. This
may be attributable to the fact that they used higher MEA
concentrations, 0.1 M in our case compared to 1.8 M and 3 M for Lin
and Shyu [15] and Aroomwilas et al. [17] cases, respectively. Published data for NaOH and Ca(OH)2 was rather scarce. However, with
regards to NaOH, we obtained higher absorption than that reported
by Georgiou et al. [14].
Attempts were made to generate high quality CH4 enriched gas
for a prolonged period. This was conducted by replacing used NaOH
and Ca(OH)2 solutions with fresh ones at about 20% of total volume
to maintain basicity level above pH 11.9. However, preliminary
testing revealed that the addition had to be done in the rst 15 min.
And after 30 min, addition of fresh solutions was not able to
produce greater than 80% purity of CH4 enriched gas. Therefore,
sole spent solutions had to be replaced in a 30-min interval. For
MEA, a similar procedure was also performed. Replacement of
a 20% volume fraction of used MEA solution with clean solution was
carried out at regular intervals of 15 min. This was done to mimic
partial regeneration of the spent MEA solution. Within the operation time considered, the basicity level of recirculated MEA was
found to reduce only slightly from its original value prior to when
the rst mixing commenced. Fig. 6 shows CH4 concentration
proles obtained, and comparison with the case where the absorbents were employed until saturated. It was clear that all solutions

Table 3
Comparison of absorption performance reported in literature.
References

Gas composition

Aqueous solvents

Capacity
(mol/mol)

This work

47% CO2, 53% CH4

[14]
[15]

10% CO2, 90% N2


5% CO2, 95% N2

0.10 M NaOH
0.10 M Ca(OH)2
0.10 M MEA
0.10e0.25 M NaOH
10% wt MEA
10% wt MDEA
10% wt MEA
10% wt MDEA
10% wt MEA
10% wt MDEA
7e14% wt NH3
3 M MEA

0.20
0.32
0.25
0.05e0.12
0.44
0.37
0.51
0.44
0.50
0.46
0.04e0.06
0.25e0.55

Table 2
Kinetic parameters for CO2 absorption by different solvents.
1

solvents

k (min

NaOH
Ca(OH)2
MEA

23.74
13.00
26.71

8% CO2, 92% N2

s (min)

42
29
36

0.962
0.975
0.844

15% CO2, 85% N2


[16]
[17]

15% CO2, 85% N2


15% CO2, 85% Air

N. Tippayawong, P. Thanompongchart / Energy 35 (2010) 4531e4535

4535

used. Chemical absorption by alkali aqueous solutions did not


appear to be promising for biogas quality upgrade due to the nonregenerable nature, requirement of large liquid solvents volume
and the environmental impact. However, amine solutions may be
worth exploring further, due to the regeneration capability. Additional experimental results are required to provide the basis for
system selection and design, as well as subsequent up-scaling. The
next phase of experimental work will involve setting up a regeneration system for amine absorbent, evaluating its long term
performance, and undertaking cost analysis of the system, as well
as a possible scale up study and eld test in a biogas farm.
Acknowledgements

Fig. 6. Comparison of CH4 concentration proles in the treated biogas (i) with and (ii)
without fresh solvent addition.

used may be operated in such a way that purity of greater than 80%
CH4 enriched biogas could be obtained over a period of time. In
terms of practical applications, these chemicals proved to be
potentially suitable absorbents. However, it should be pointed out
that although high concentrations of CH4 were obtained after
a short time (as shown in Table 1), these high concentrations were
found to decrease rapidly. To maintain high absorption rate,
a substantial fraction of original volume of the solution needed to
be replaced or regenerated. Frequent replacement would lead to
a uctuation in concentrations, as observed in Fig. 6. One of the
limitations of biogas upgrading lies in economics of the bulk
separation of CO2. The main disadvantage of using alkaline solvents
was that they are very difcult or impossible to regenerate for
reuse. Even though they are relatively low cost materials, large
amount of chemicals would be required to satisfy purity requirements of CH4 enriched gas. Use of MEA has the advantage of being
able to be regenerated. However, for a small scale biogas production
as considered in this study, initial investment cost for constructing
the set up, as well as energy requirement may be excessive and
outweigh the benet of using high purity biogas. Further work may
thus be required.
5. Conclusions
Removal of CO2 and H2S from biogas by aqueous solutions in
a packed column was investigated. NaOH, Ca(OH)2 and MEA were
employed in the present study and their absorption characteristics
were examined. A simple empirical model was also adopted for CO2
absorption rate prediction. Test results indicated similar absorption
patterns between the chemical solvents used. Chemical absorption
by solvents in a packed column was an effective technique for
removing CO2 and H2S over a short operation time, but their
absorption capability declined rapidly with time. Ca(OH)2 appeared
to become saturated more rapidly than the other solvents. The CO2
loading ranged between 0.18 and 0.22 kg CO2 per kg chemicals

Support from the Commission on Higher Education, Ministry of


Education, and the Fund for Energy Conservation Promotion,
Energy Policy and Planning Ofce, Ministry of Energy are highly
appreciated. The authors wish to thank Drs P. Polchan and M. O. T.
Cole for their valuable technical discussions, and staff at the Energy
Research and Development Institute, Chiang Mai University for
technical and laboratory assistance.
References
[1] Prasertsan S, Sajjakulnukit B. Biomass and biogas energy in Thailand: potential, opportunity and barriers. Renewable Energy 2006;31:599e610.
[2] Borjesson P, Mattiasson B. Biogas as a resource-efcient vehicle fuel. Trends
Biotechnol 2007;26:7e13.
[3] Tippayawong N, Promwungkwa A, Rerkkriangkrai P. Long-term operation of
a small biogas/diesel dual-fuel engine for on-farm electricity generation.
Biosyst Eng 2007;98:26e32.
[4] Tippayawong N, Promwungkwa A, Rerkkriangkrai P. Durability of a small
agricultural engine on biogas/diesel dual fuel operation. Iran J Sci Technol
Trans B Eng 2010;34(B2):167e77.
[5] Abatzoglou N, Boivin S. A review of biogas purication processes. Biofuels
Bioprod Bioren 2009;3:42e71.
[6] Yang H, Xu Z, Fan M, Gupta R, Slimane RB, Bland AE, et al. Progress in carbon
dioxide separation and capture: a review. J Environ Sci 2008;20:14e27.
[7] Granite EJ, OBrien T. Review of novel methods for carbon dioxide separation
from ue and fuel gases. Fuel Process Technol 2005;86:1423e34.
[8] Horikawa MS, Rossi F, Gimenes ML, Costa CMM, da Silva MJC. Chemical
absorption of H2S for biogas purication. Brazil J Chem Eng 2004;21:415e22.
[9] Osorio J, Torres JC. Biogas purication from anaerobic digestion in a wastewater treatment plant for biofuel production. Renewable Energy
2009;34:2164e71.
[10] Favre E, Bounaceur R, Roizard D. Biogas, membranes, and carbon dioxide
capture. J Membr Sci 2009;328:11e4.
[11] Kapdi SS, Vijay VK, Rajesh SK, Prasad R. Biogas scrubbing, compression and
storage: perspective and prospectus in Indian context. Renewable Energy
2005;30:1195e202.
[12] Rasi S, Lantela J, Veijanen A, Rintala J. Landll gas upgrading with countercurrent water wash. Waste Manage 2008;28:1528e34.
[13] Krumdieck S, Wallace J, Curnow O. Compact low energy CO2 management
using amine solution in a packed bubble column. Chem Eng J 2008;135:3e9.
[14] Georgiou D, Petrolekas PD, Hatzixanthis S, Aivasidis A. Absorption of carbon
dioxide by raw and treated dye-bath efuents. J Hazardous Mater
2007;144:369e76.
[15] Lin SH, Shyu CT. Performance characteristics and modeling of carbon dioxide
absorption by amines in a packed column. Waste Manage 1999;19:255e62.
[16] Yeh JT, Resnik KP, Rygle K, Pennline HW. Semi-batch absorption and regeneration studies for CO2 capture by aqueous ammonia. Fuel Process Technol
2005;86:1533e46.
[17] Aroonwilas A, Tontiwachwuthikul P, Chakma A. Effects of operating and
design parameters on CO2 absorption in columns with structured packings.
Separ Purif Technol 2001;24:403e11.

Você também pode gostar