Fuel 103 (2013) 387392

Contents lists available at SciVerse ScienceDirect

Fuel
journal homepage: www.elsevier.com/locate/fuel

Comparison of efciency of two methods for tar sampling in the syngas

Sergejs Osipovs
Daugavpils University, Department of Chemistry, Vienibas 13, 5401 Daugavpils, Latvia

h i g h l i g h t s
" The SPA method gives more complete results than the CST method.
" The SPA method is much faster.
" The sampling device is more convenient and simpler in use and maintenance.
" The CST method is more suitable for the quantication of heavy tar components.
" The SPA method is optimal in the majority of cases.

a r t i c l e

i n f o

Article history:
Received 7 March 2012
Received in revised form 8 May 2012
Accepted 9 May 2012
Available online 27 May 2012
Keywords:
Biomass
Tar
Sampling
Solid-phase adsorption method (SPA)
Cold solvent trapping method (CST)

a b s t r a c t
In the present work, a cold solvent trapping (CST) and solid-phase adsorption (SPA) methods for determining concentration of tar compounds have been chosen for comparison. When the cold solvent trapping method is used, the producer gas ows through a series of impingers containing 2-propanol,
whereas in a solid-phase adsorption method it passes through two adsorbent cartridges loaded with
500 mg of aminopropyl-bonded silica, and 100 mg of activated coconut charcoal. During the experiment,
52 compounds were identied by the cold solvent trapping method and 48 compounds by the solidphase adsorption method. The SPA method is more accurate than those using impingers, especially for
determining such volatile organic compounds as benzene, toluene, and xylenes, due to the use of a second sorbent, activated coconut charcoal. By contrast, the CST method proves to be more accurate for
determining components of heavy tar due to a much larger volume of the sampled gas.
2012 Elsevier Ltd. All rights reserved.

1. Introduction
Conversion of abundantly available biomass to synthesis gas
and hydrogen helps protect the environment. Synthesis gas can
be converted into clean liquid fuels, and hydrogen is an encouraging energy producer.
Gasication thus is a promising technique for the production of
energy from biomass with non-catalytic gasication, with air (partial oxidation) or steam at high temperatures being a conventional
method of producing synthesis gas and hydrogen from biomass [1].
The main product of biomass gasication, a mixture of gases
containing mainly carbon oxides, hydrogen, and nitrogen, also contains a small amount of methane and other lighter hydrocarbons.
Ash particles, volatile alkali metals, and tar are biomass gasication
products, too.
As a by-product of biomass gasication, tar is undesirable because of related problems such its condensation and formation of
tar aerosols [2]. Tar is a mixture of acids, aldehydes, ketones,
Tel.: +371 26163221.
E-mail address: sergejs.osipovs@du.lv
0016-2361/$ - see front matter 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.fuel.2012.05.021

alcohols, phenols, and aromatic hydrocarbons, and its composition

depends on the conditions of gasication.
There does not seem to be a consensus about what tar and its
compounds are: rst, Evans and Milne [3], for example, divide
the pyrolysis tar into primary, secondary, and tertiary. Second, Milne et al. think that tar is the organic produced under thermal or
partial-oxidation regimes (gasication) of any organic material
and generally assumed to be largely aromatic [4].
In the gasication process, tar is dened as the condensable
products at ambient temperature, and often implies aromatic compounds and polyaromatics. As far as individual compounds are
concerned, though, benzene is excluded from tar due to high concentration saturation in closed systems at 25 C [57]. Guideline
for Sampling and Analysis of Tar and Particles in Biomass Producer
Gases gives the following denition of tar: Tar: Generic (unspecic) term for entity of all organic compounds present in the producer gas excluding gaseous hydrocarbons (C1C6). Benzene is not
included in tar [8]. There exists a view that benzene is not a problematic compound in the real biomass gasication gas and its complete removal is not required [9] as combustion of benzene is clean
and results in no clogging. Therefore benzene should be treated as

388

S. Osipovs / Fuel 103 (2013) 387392

a separate compound and excluded from the denition of tar. But

we made it a point in this research to consider benzene as a distinct tar component.
Present day specialist literature contains a detailed description
of two principally different methods of tar sampling widely used in
practice: cold solvent trapping, and solid-phase adsorption.
1.1. Cold solvent trapping (CST)
This method extensively uses for analysis the tar obtained during the process of biomass gasication. Equipment with a series
of impingers is employed for tar sampling. In order to develop
widely accepted and standardised Protocols, the Fifth EU Framework Project Tar Protocol (20002001) was carried out by a large
consortium of seven contractors and 10 reviewers. As a result of the
Project, a Guideline for Sampling and Analysis of Tar and Particles in
Biomass Producer Gases was published, and standardization of the
Guideline Protocol has been initiated [8]. In Europe, nowadays, it is
a CEN/TS 15439: 2006 standard [6]. The method is based on the
absorption of tar by an organic solvent. Solid particles are caught
by a hot ceramic lter. The tar is analysed gravimetrically or/and
with the help of gas chromatography. The system for tar and solid
particles sampling consists of heated sampling lines, a heated lter,
and a series of impingers containing a solvent. The impinger with
the collected tar is placed in the thermostatic tub whose regime
is regulated in order to heat or cool down the gas under analysis.
During an indicated time period, gases are drawn through the sampling line and lter. The latter are heated in order to avoid tar condensation. But the temperature should be optimised for avoiding
thermal decomposition of an organic component. The gas volume,
temperature, pressure, and ow are measured during the sampling
process. Immediately after the collection of the sample, the content
of impingers is poured into dark-glass bottles and kept for certain
time before the analysis. The lter with the collected solid particles
is treated with a fresh solvent to extract the heaviest tar components which condense on the lter in spite of its being heated.
The used solvent is then added to the solvent from the series of
impingers, and the resulting substance is left for further analysis
of samples aimed at determining the total tar.
1.2. Solid-phase adsorption (SPA)
The SPA method was initially developed by The Royal Institute of
Technology in Sweden [10] to measure tar compounds ranging in
molecular weight from benzene to coronene. According to this
method, tar is sampled by collecting it on a column with a small
amount of amino-phase sorbent. For each sample, 100 mL of gas is
withdrawn from a sampling line with a syringe or a pump. The temperature in the sampling line is kept between 250 and 300 C in order to minimise tar condensation. But this method does not allow
for determining such volatile organic compounds as benzene, toluene, and xylenes, some of which, because of their high concentration
in biomass tar, do not collect on the amino-phase sorbent. In the previous paper [11,12], an improved system of sampling was suggested
and described, whereby one more adsorbent cartridge loaded with
another sorbent is added. The best results were obtained while using
activated coconut charcoal as the second sorbent [13]. In this study,
a modied sampling device containing 500 mg of amino-phase sorbent and 100 mg of activated coconut charcoal was chosen as optimal for sampling tar (including its volatile organic compounds) in
the synthesis gas produced during biomass gasication.
1.3. Comparison of some tar sampling methods
Related literature contains comparative analysis of various tar
sampling methods. The CST method [5], for instance, is reported

to make use of four impingers lled with methanol and operates

at the cooling temperature of 60 C. The adsorption method employs Carbotrap 300 as a sorbent which presents a mixture of the
following sorbents, Carbotrap C (graphitized carbon black), Carbotrap B (graphitized carbon black), and Carbosieve SIII (carbon
molecular sieve). This method envisages that adsorption is followed by thermal desorption. The authors claim that the adsorption method is more precise, particularly in case of light tar;
besides, the sampling time is noticeably shorter and the determination limit lower. The CST method is more convenient for sampling tar from the generator gas with a high content of tar.
Williams and Phillips [14] compare CST, SPA, and gravimetric
method of analysing tar. When the gravimetric method is used,
heavy tar components condense on the glass bre of the lter.
For determining the total tar, it is recommended to combine
adsorption and gravimetric methods. The authors further conclude
that from the point of view of effectiveness, this combination of
methods is equivalent to the CST method. Mrsch et al. [7] also
compare the tar sampling methods suggested in the present
article.
Comparison of the two methods, CST and SPA, allows for identifying their strengths and weaknesses. The CST method, for example, has its shortcomings; the biggest of them are as follows:
 long sampling time (from 15 to 10 min) and quite a lengthy
preparation period which prevents from efciently following the gasication process;
 for taking each sample, use of minimum 500 ml of 2-propanol is required; it is harmful for health, and it has to be utilised after the sampling process;
 formation of tar and solvent aerosol, which results in the
loss/waste of adsorbed tar;
 evaporation of the solvent together with which part of the
tar also evaporates;
 incomplete detection of the tar components with a low boiling point, with a possibility of mass discrimination, i.e.
light tar compounds ow out of the system in larger
amounts than the heavy compounds, which results in
imprecise determination of compound proportion in tar;
and
 the sampling system is bulky and complicated.
The SPA method has a number of following advantages over the
CST one:
 a short sampling time (normally 1 min) allows for effectively conducting generator gas analysis and adjusting to
the conditions of a gasier, particularly when conditions
are being changed and/or necessity to determine tar content
at a required moment;
 each case of sampling requires at least a hundred times
smaller amount of solvent;
 there is no loss of tar caused by evaporation of the solvent
and formation of aerosol; and
 the sampling system is simple and easy to operate.
2. Experimental
2.1. Gasier
For research in a real-life context, the Circulating Fluidised Bed
(CFB) gasier situated in eastern Latvia (Rezekne region) was
chosen. Peat extracted several kilometres from the gasier was
employed as biomass. The main characteristics of the gasier are
as follows: heat output is 600 kWth, reaction temperature reaches
8001050 C, the mass of dry peat is 250 kg h 1, and the mass

389

S. Osipovs / Fuel 103 (2013) 387392

Fig. 2. The total tar mass concentration with standard deviation (%) determined by
different sampling methods depending on the generator gas volume.

2.2. Tar sampling

The product gas was sampled by both methods from the outlet
tube after the gasier where the product gas temperature reached
150 C.
Fig. 1. A tar sampling device. 1 connection to a pump, 2 adapter (polypropylene),
3 sorbent column (polypropylene, 1 mL), 4 fritted disc (polyethylene),
5 activated coconut charcoal, 6 sorbent column (polypropylene, 4 mL),
7 amino-phase sorbent, 8 septum nut (polyethylene), 9 rubber/silicone
septum, 10 hypodermic needle (stainless steel), 11 glass Tee tube, 12 heating
tape, 13 product gas inlet, 14 connection to electric pump.

of air 120 kg h 1. The peat biomass has the following fuel characteristics calculated as for dry basis: gross caloric value is
20.00 MJ kg 1, moisture 13.0%, C 53.23%, H 7.63%, N 0.86%, S
0.10%, O 38.18% (by difference). The product gas comprises H2
(13.43%), N2 (50.52%), O2 (0.38%), CO (12.72%), CH4 (2.41%), and
CO2 (16.91%). Average concentration of tar in the product gas is
1400 mg m 3.

2.3. Tar sampling device according to the SPA method

A tar sampling device consisting of two consecutively joined
columns with adsorbents was made especially for the present research (Fig. 1).
The rst column was a 4 mL solid-phase extraction adsorbent
cartridge with 500 mg of loosely packed aminopropyl-bonded silica adsorbent (the surface area was about 400600 m2 g 1, the particle size 50 lm, and the average pore size 60 (Alltech). The
second column was a 1 mL SPE adsorbent cartridge (Alltech) with
100 mg of loosely packed activated coconut charcoal (the surface
area was 1070 m2 g 1, the particle size 20/40 mesh, 420840 lm
(SUPELCO).

Table 1
Mass spectrometric data and limits of detection for quantied tar and internal standards.
Compound

Chemical formula

Molecular weight (g mol

Benzene
Benzene-d6
Toluene
m-p-Xylene
o-Xylene
Phenol
Phenol-d6
Indane
Indene
o-Cresol
m-p-Cresol
Naphthalene
Naphthalene-d8
Acenaphthylene
Acenaphthene
9H-Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Chrysene
Benzo(a)pyrene

C6H6
C6D6
C7H8
C8H10
C8H10
C6H6O
C6D6O
C9H10
C9H8
C7H8O
C7H8O
C10H8
C10D8
C12H8
C12H10
C13H10
C14H10
C14H10
C16H10
C16H10
C18H12
C20H12

78
84
92
106
106
94
100
118
116
108
108
128
136
152
154
166
178
178
202
202
228
252

Mass spectrum (NIST 08) (mass) (abundance)

Detection limits (ng)

78 (999), 77 (283), 51 (221)

84 (999), 52 (261), 56 (259)
91 (999), 92 (776), 65 (121)

0.0026
0.0075
0.0042
0.0045
0.0061
0.0127
0.0101
0.0040
0.0069
0.0093
0.0087
0.0023
0.0071
0.0099
0.0107
0.0049
0.0184
0.0143
0.0062
0.0077
0.0051
0.0030

91 (999), 106 (501), 105 (206)

94 (999), 66 (387), 65 (266)
99 (999), 71 (482), 42 (334)
117 (999), 118 (692), 115 (266)
116 (999), 115 (792), 89 (100)
108 (999), 107 (673), 79 (253)
128
136
152
153
166
178
178
202
202
228
252

(999),
(999),
(999),
(999),
(999),
(999),
(999),
(999),
(999),
(999),
(999),

129
108
153
154
165
176
179
203
203
226
253

(109),
(116),
(152),
(827),
(844),
(202),
(156),
(173),
(170),
(271),
(215),

127 (107)
137 (103)
151 (137)
152 (507)
167 (140)
179 (150)
176 (140)
200 (153)
200 (152)
229 (203)
250 (172)

390

S. Osipovs / Fuel 103 (2013) 387392

2.3.1. Sample preparation

There are two desorption procedures, column extraction from
the amino-phase sorbent by with dichloromethane/acetonitrile,
and solvent extraction from the activated charcoal sorbent. The following applies to the rst column. The inlet of the adsorbent cartridge was connected to a reservoir (an empty adsorbent
column). The cartridge was positioned vertically in a stand exactly
above a vial (1.8 mL) placed on a laboratory jack. The internal standards dissolved in dichloromethane were added to the reservoir.
Then elution of the tar compounds with 3  600 lL of dichloromethane/acetonitrile (4:1 v/v) took place. Dichloromethane/acetonitrile was drawn through the amino-phase adsorbent by means of
a 100 mL syringe. The sorbent from the second column was placed
in separate vials, 1.8 mL each; 1.0 mL of carbon disulphide was
added to each vial, and crimp caps were immediately attached to
each vial. The vials were allowed to stand for at least 30 min with
occasional agitation.
2.4. Equipment for the CST tar sampling method
Fig. 3. Mass concentrations with standard deviation (%) of the total tar, tar without
benzene, and tar without BTX analysed by several sampling methods.

Using an insulated heating tape to minimise tar condensation,

the Tee tube was heated to 250 C. The product gas was drawn
through the heated tube with the help of an air sampling pump
(SKC). A required volume (depending on tar concentration) of
product gas at approximately atmospheric pressure was sampled
with a mass ow rate regulator and a specic gas vane-type pump
at the constant ow rate of 100 mL min 1. The product gas with tar
was drawn into the rst adsorbent cartridge with the amino-phase
adsorbent. The core amount of the compounds such as indene,
naphthalene, acenaphthylene, acenaphthene, uorene, phenanthrene, anthracene, uoranthene, and pyrene was adsorbed in
the rst adsorbent cartridge, while volatile organic compounds,
mainly benzene and toluene, were drawn with the syngas into
the second adsorbent cartridge loaded with activated coconut
charcoal. After sampling, the inlet and outlet sides of the cartridges
were sealed with plastic stoppers.

Fig. 4. Content of different compounds in mg m

The CST tar sampling method uses the equipment described in

Tar Protocol [6].
2.5. Analysis
A Shimadzu GCMS-QP2010 system (Shimadzu Corporation,
KYOTO, Japan) was used for the analysis. The gas chromatograph
was equipped with an electronically controlled split/splitless injection port. GC was carried out on a 5% diphenyl-/95% dimethylpolysiloxane fused-silica capillary column (Rtx-5SIL-MS,
30 m  0.32 mm, 0.25 lm lm thickness; Restek). Helium
(99.999%) was used as the carrier gas at a constant ow of
1.6 mL min 1. The injection (injection volume of 1 lL) was performed at 250 C in the split mode, with the split ratio 1:10. The
oven temperature program was as follows: the temperature was
held at 30 C for 5 min, then 30180 C at the rate of 10 C min 1,
180300 C at the rate of 15 C min 1, and nally held at 300 C
for 5 min. The mass spectrometer was operated in the electron ionisation mode (ionisation energy of 70 eV). The source and transfer

with standard deviation (%) depending on sampling method.

391

S. Osipovs / Fuel 103 (2013) 387392

line temperatures were 200 and 310 C, respectively. Detection

was carried out in the scan mode: m/z 35300. The analysis time
for dichloromethane/acetonitrile was 31 min and for carbon disulphide 15 min.
2.6. Calibration and quantication
Quantied compounds are given in Table 1. Benzene-d6, naphthalene-d8, and phenol-d6 were used as internal standards. Calibration curves were performed with ve points, each of them realised
in triplicate. Five calibration mixtures were made using pure standards of the compounds to be quantied and a certain amount of
internal standards added to each calibration mixture and to the
studied samples. All areas were measured and referenced to the
area of internal standards. The instrumental limits of detection
(LOD) were estimated from the chromatograms at the signal-tonoise ratio of 3. They ranged from 0.0029 ng for naphthalene to
0.0184 ng for phenanthrene (see Table 1).
3. Results and discussion
3.1. Determination of the total tar content depending on the generator
gas volume
Sampling according to both methods was performed simultaneously and under identical conditions; the following samples
were taken at the same time: 100 mL of generator gas by the SPA
method and 50 L by the CST one; then 200 mL and 100 L, and nally 300 mL and 150 L, respectively. In Fig. 2, one can see the total
content of tar obtained by using the SPA method of sampling and
the CST method, with different volumes of the generator gas taken
for analysis.
It can be seen in Fig. 2 that concentration of the total tar mass
obtained by the SPA sampling method is higher than that obtained
by the CST method irrespective of the generator gas volume. On
average, the difference between the results obtained by the two
methods is 22.3%. These results could be predicted as the CST
method is not suitable for determining such volatile tar components as benzene (whose quantity in tar may reach 60%). Use of
the SPA method of sampling with varying generator gas volumes
gave comparable results, whereas use of the CST method demonstrated lessening of the total tar concentration with an increase
of the generator gas volumes from 100 to 150 L. This can be explained by the formation of the tar and solvent aerosol that ows
out from the sampling device, and by evaporation of the solvent.
3.2. Determination of the total tar, tar without benzene, and tar
without benzene, toluene, and xylenes
The total mass concentration of tar in comparison to its mass
concentration without benzene, and the tar mass concentration
without BTX (benzene, toluene, and xylenes) are presented in
Fig. 3.
Fig. 3 demonstrates that if there is a need to know the total tar
content and its volatile components including benzene, the SPA
method of sampling is preferable. Comparing to the SPA tar content, the results obtained with the CST method are lower by
19.8%. Anyway, during the sampling process, part of the sample
forms aerosol with the solvent, which ows away from the sampling device uncontrolled. If it is necessary to evaluate the total
tar mass content without benzene, the difference between the
results obtained by different methods is smaller; however the best
results are still with the SPA method. In comparison to it, the
results obtained with the CST tar sampling method are lower by
11.5%. This once again conrms that with the CST method, the

Table 2
Compounds identied by SPA method. Average values and standard deviations are
given for three replicate runs.
Nr

Retention time
(min)

Compounds

Mass part (%)

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45

2.242
4.223
7.028
7.233
7.388
7.682
9.178
9.255
9.304
9.459
9.526
10.001
10.128
10.231
10.436
10.498
10.729
10.803
10.894
10.939
11.830
12.387
12.972
13.126
13.510
13.782
14.043
14.061
14.182
14.226
14.297
14.387
14.408
14.503
14.830
14.924
15.175
15.660
15.788
15.850
16.207
16.296
17.859
17.974
19.182

66.8 2.8
14.1 0.5
0.035 0.003
1.28 0.08
0.19 0.01
2.41 0.11
0.13 0.01
0.44 0.04
0.40 0.03
0.69 0.04
0.070 0.004
0.013 0.002
2.04 0.01
0.014 0.004
0.053 0.004
0.045 0.009
0.037 0.002
0.040 0.003
0.086 0.005
0.041 0.002
7.83 0.31
0.13 0.01
0.76 0.03
0.64 0.04
0.035 0.003
0.29 0.02
0.007 0.001
0.008 0.001
0.032 0.004
0.020 0.003
0.023 0.002
0.007 0.001
0.003 0.001
0.99 0.08
0.055 0.008
0.013 0.002
0.106 0.009
0.008 0.001
0.010 0.002
0.16 0.01
0.005 0.001
0.003 0.001
0.36 0.02
0.012 0.002
0.006 0.001

46
47
48

19.259
20.558
21.058

Benzene
Toluene
Ethylbenzene
m-p-Xylene
Phenylethyne
Styrene + o-Xylene
Aniline
Phenol
Benzonitrile
Benzofuran
m-Methylstyrene
Indane
Indene
o-Cresol
2-Methylbenzonitrile
m-p-Cresol
4-Methylbenzonitrile
2-Methylbenzofuran
3-Methylbenzofuran
7-Methylbenzofuran
Naphthalene
Quinoline
2-Methylnaphthalene
1-Methylnaphthalene
1H-Cyclopropa[g]quinoline
Biphenyl
1-Ethylnaphthalene
2,6-Dimethylnaphthalene
1,3-Dimethylnaphthalene
1,6-Dimethylnaphthalene
2-Ethenylnaphthalene
1,4-Dimethylnaphthalene
1,5-Dimethylnaphthalene
Acenaphthylene
Acenaphthene
4-Methylbiphenyl
Dibenzofuran
Acenaphthenone
1H-Phenalene
9H-Fluorene
4-Methyldibenzofuran
9H-Xanthene
Phenanthrene
Anthracene
4HCyclopenta[def]phenanthrene
1-Methylanthracene
Fluoranthene
Pyrene

0.008 0.001
0.009 0.001
0.008 0.001

largest losses are caused by the imprecise determination of benzene and other volatile compounds (toluene and xylene), which
considerably inuences the results of analysis. The situation
changes when it is necessary to determine the total tar concentration without taking into account the most common volatile components (benzene, toluene, and xylene). Then the results of all
investigated methods are comparable. The amount of tar obtained
by the CST method is by 1.1% higher than that obtained by the SPA
method. The difference between the SPA and CST methods of tar
analysis, taking into account volatile components, can be explained
by the fact that the latter is slightly more sensitive exactly to heavy
tar components.
3.3. Determination of various tar compounds
Concentrations of most common tar compounds are presented
in Fig. 4 where the part of each compound under study in the total
tar is not less than 1%, as well as the total concentration of all other
compounds.

392

S. Osipovs / Fuel 103 (2013) 387392

Comparing results of different methods presented in Fig. 4, it

can be seen that the biggest differences are observed for benzene;
and as it was previously predicted, the best results are obtained by
the SPA method. The amount of benzene obtained by the CST
method is by 24.8% lower. Comparing the results for toluene and
xylene, it becomes obvious that the difference between the investigated methods gradually decreases, according to volatility of
compounds, in such an order: benzenetoluenexylene. The mass
concentrations of indene analysed by various methods are quite
comparable. The mass concentration of naphthalene obtained by
the CST method is only by 2.7% lower comparing to the SPA method. The highest mass concentration of acenaphthylene, phenanthrene, pyrene, and other heavy tar compounds was received by
the CST method. The number of compounds identied in generator
gas varies similarly when different methods are used. Thus, 48
individual compounds were identied with the SPA method and
52 with the CST method comparing the spectra obtained by manual interpretation with the spectra from the NIST mass spectral library. They are presented in Table 2 as a function of their retention
time.
However, the total content of the four additionally detected
compounds is only 0.19% of the total tar mass determined by the
SPA method. The number of compounds identied with the CST
method is bigger because it allows for identifying additionally
many heavy tar components. For example, this method determines
benzo(a)pyrene (whose mass concentration is slightly higher than
the Method Detection Limit). This gives evidence that the CST
method is more sensitive to the heavy components of tar.
Comparing the SPA method with liquid injection and the
SPA/TD method (Dufour et al. [5]), it can be concluded that the
former is more suitable for sampling in case of tar analysis (due
to a high temperature of the producer gas) because use of sorbents
selected for thermal desorption can lead to the loss of a part of tar
because partial thermal desorption of tar compounds will take
place alongside their adsorption.
4. Conclusions
Comparing CST and SPA methods of sampling tar in the generator gas produced during biomass gasication, advantages and disadvantages of every method were evaluated. The SPA method is

much faster. The sampling time with this method is only 2 min
comparing to the 20 min with the CST method. The sampling device is more convenient and simpler in use and maintenance. The
results of analysis of the total tar (including volatile compounds)
obtained by the SPA method are more complete than the results
obtained by the CST method. Drawing parallels between the methods with regard to individual tar components, it appeared that the
SPA method is more suitable for determination of benzene, toluene, and xylene than the CST one. The CST method, in its turn,
has the advantage of being slightly more suitable for the quantication of heavy tar components and allows for identication of a
bigger number of individual compounds although their mass in
the total tar is relatively small. It can be thus concluded that when
choosing the appropriate method, it is important to take into consideration the necessity to determine as many individual compounds as possible, and existence of limits for the sampling time.
The SPA method is optimal in the majority of cases.
References
[1]
[2]
[3]
[4]

[5]
[6]

[7]
[8]

[9]
[10]
[11]
[12]
[13]

[14]

Bridgwater AV. Appl Catal A: Gen 1994;116:5.

Uddin MA, Tsuda H, Wu S, Sasaoka E. Fuel 2008;87:451.
Evans RJ, Milne TA. Energ Fuels 1987;1:311.
Milne TA, Evans RJ, Abatzoglou N. Biomass gasier Tars: their nature,
formation, and conversion, National Renewable Energy Laboratory report no.
NREL/TP-570-25357. Golden, Colorado; 1998. p. 68.
Dufour A, Girods P, Masson E, Normand S, Rogaume Y, Zoulalian A. J
Chromatogr A 2007;1164:240.
CEN/TS 15439:2006. Biomass gasication tar and particles in product gases
sampling and analysis. European Committee for Standardization, Brussels;
2006.
Moersch O, Spliethoff H, Hein KRG. Biomass Bioenerg 2000;18:79.
Neeft JPA, Knoef HAM, Zielke U, Sjstrm K, Hasler P, Simell PA, Dorrington
MA, Thomas L, Abatzoglou N, Deutch S, Greil C, Bufnga GJ, Brage C,
Suomalainen M. Guideline for sampling and analysis of tar and particles in
biomass producer gases <http://www.tarweb.net./results/pdf/guideline-3.3v2.pdf>.
Engelen K, Zhang Y, Draelants DJ, Baron GV. Chem Eng Sci 2003;58:665.
Brage C, Yu Q, Chen G, Sjostrom K. Fuel 1997;76:137.
Osipovs S. Anal Bioanal Chem 2008;391:1409.
Osipovs S. Int J Environ Anal Chem 2009;89(812):871.
Osipovs S. In: Presented at the 16th European biomass conference & exhibition
from research to industry and markets, Feria Valencia, Spain; 2008.
[unpublished].
Williams K, Phillips S. Comparison of three common methods used to sample
tars in biomass gasication. Student Abstracts: Engineering at NREL, National
Renewable Energy Laboratory, Golden, Colorado; 2004.