RDF-GASIFICATION PART 1:
CHARACTERIZING THE USE OF RDF AS
FUEL AND SOLVING THE TAR PROBLEM
BY AN IN-DEPTH LABORATORY STUDY.
H.J.M. VISSER*, R. ZWART*, J-W KöNEMANN**, A. BOS* AND J.
KUIPERS*
* Energy research Centre of the Netherlands (ECN), Westerduinweg 3 1755LE
Petten, The Netherlands. **Dahlman Industrial Group BV, Maassluis, The
Netherlands.
SUMMARY: Source separation of waste and higher recycling percentages in waste streams
result in better sorted waste, e.g. in RDF with a high plastic content. An optimized, dedicated
installation can be designed for this high calorific waste. Preliminary laboratory tests showed
that gasification of RDF has some advantages over combustion in a fluidized bed system. The
drawback of gasification, however, is the production of tars. The tar chemistry was studied
during laboratory-scale gasification tests at ECN with emphasis on RDF specific tars and their
effect on the OLGA tar-cleaning system designed for biomass. The first tests encountered
polymerization in the OLGA collector. The influence of foam polystryrene in the fuel was
studied after observing extremely high styrene concentrations. However, polystyrene by itself
did not lead to fouling. A new hypothesis is currently experimentally tested. It is thought that the
polymerization can be attributed to the presence of polystyrene in the RDF fuel, chlorine as a
catalyst for the polymerization reactions and at least one other tar component, likely released
from another RDF component. Future efforts will be aimed at understanding the influence of the
main individual RDF components and building up a knowledge base to use as a predictive tool to
avoid operational problems.
1. INTRODUCTION
Source separation of waste and higher recycling percentages in waste streams result in better
sorted waste, e.g. in RDF with a high plastic content. Instead of burning this fraction in a general
Municipal Solid Waste (MSW) Incinerator, where it often will cause problems when not properly
mixed, a more optimized dedicated installation can be designed for this high calorific waste.
Preliminary laboratory tests showed that gasification of a relatively high calorific value fuel, such
as RDF, has some advantages over combustion in a bubbling fluidized bed system. Especially the
feeding of the fuel in the bed was very problematic under combustion conditions while this
problem was clearly less problematic when gasifying the fuel. In addition, the efficiency of the
gasification process is higher and the use of the product gas allows a more versatile end-use than
steam generation only, for example by upgrading towards a synthetic natural gas or a
transportation fuel. The drawback of gasification, however, is the production of tars
Tar related problems are the most important obstruction for the introduction of gasification
and are responsible for major operational problems due to tar condensation (fouling) and
poisonous condensate leading to unacceptable waste water costs. Over the past decade, intensive
research lead to the development of the OLGA cleaning technologie.
OLGA is an advanced scrubbing technology for tar removal that was developed and patented
by ECN. Dahlman Industrial Group BV, Maassluis (the Netherlands), brings the technology to
market. The technology does not concentrate on the tar quantity, but on the unwanted behavior
of tar (i.e. the properties). The performance of OLGA was demonstrated by a 1000 hours
duration test with a 0.5 MW(t) pilot plant at ECN. The process showed to be stable and reliable
and gas quality was on specification for a gas engine that ran smoothly. Results on tar dew point
(<10 C), phenol removal (completely) and naphthalene removal (>99%) were excellent.
Currently the technique is being demonstrated in France on a 4 MW(t) scale gasifier.
The above described 1000 hours success story was performed using clean wood as fuel.
Whether or not similar behavior can be expected when the fuel was not clean wood but a
difficult waste fraction such as RDF is not known. Because the detailed characterization of the
tar content depends on very stable and well known operating conditions it was decided to start
work on RDF in a lab-scale bubbling fluidized bed with a lab-scale version of OLGA, as was
done at the beginning of the biomass work. A test program was performed of which the results
will be presented. Furthermore, the RDF fuel used in test series 1 was also tested in a pilot-scale
installation in Tzum (Netherlands) as a prelude for a commercial scale gasifyer-OLGA
installation (now in the procedure of permissions) that is planned to use the same RDF. (See
presentation “Gasification of RDF part 2: Pilot-scale results”).
2. TAR CHEMISTRY AND MEASUREMENT METHODS
2.1 Tar chemistry versus tar quantity
Tar formation during gasifcation is not avoidable when using fluidized bed technology at
temperatures typically below 900 oC. Tar is formed due to the depolymerization of the building
blocks of biomass (e.g. wood and paper) and the plastic components in RDF. The term 'tar'
includes thousands of single substances. A general definition of 'tar' often used is: all organic
contaminants having a molecular weight higher than benzene.
It is widely recognised that tars are the most problematic species in product gas derived from
biomass gasification. Tars easily lead to fouling and largely depends on the tar dewpoint and
composition of the tars and not so much on the total tar concentration. This is shown in Figure 1.
For example just some tar components that have 4 or more aromatic rings raise the dewpoint of
the productgas from 120 oC to more then 200 oC, bringing it in the temperature range of gas
cleaning components.
220
100%
100%
98%
dew point [°C]
180
80%
68%
140
60%
100
40%
60
20%
20
12%
-20
All tar
1-4 ring
1-3 ring
1-2 ring
1 ring
tar mass fraction from total (wt%)
93%
0%
Figure 1: Dewpoint versus mass fraction of tars with different amounts of aromatic rings. The
red bars show the dew point temperature depending on the composition. The pink line shows the
total amount of tars as indicated on the Y-axis on the right hand side.
2.2 Tar sampling methods
There is a tar guideline sampling and analysis method that is the CEN standard and is often used
at ECN both in the lab and during commercial scale measurement campaigns. However, it
requires long stable operation conditions and in addition solid-phase adsorption (SPA) is often
used for its quick and easy application. Because the lab-scale experiments are performed under
frequently varying conditions, SPA was the method of choice. The method comprises of ≈ 100
ml of hot producer gas being passed through a disposable cartridge containing 500 mg of aminophase (modified silica). After preparation, samples are analysed on a chemical lab for individual
compounds by GC-FID/MS. Its main advantage is the easy and fast sampling (2 minutes).
Disadvange is that only GC detectable components ranging in molecular weight from 78
(benzene) to 300 (coronene) are determined. The method is less suitable for the very volatile tar
components like benzene and toluene. Experience is that there is an overall good fit between
results of the SPA method compared to the Guideline method for most of the tars and where the
concentrations of SPA are (much) lower in the range of the volatile components, the values are
at least reproducible and are used as a rough first qualitative indication (See e.g. “Gasification of
RDF part 2: Pilot-scale results”).
3. EXPERIMENTAL LABORATORY STUDY
3.1 Aims
The first aim of the experimental program was to test RDF as a fuel under gasification conditions
in a fluidized bed reactor. The feeding of RDF, the bulk gas conditions and the formation of tars
were aimed to be studied, but also the behaviour of the fly ash and the potential to agglomerate
the bed material (quartz sand). Due to the enormous variety of RDF it was also aimed to try at
least a few different streams of RDF. In case gasification would be succesfull, the aim of the
experiments would further focus on the gas cleaning, necessary to bring the gas up to
specification for its end use. In particular experiments were designed to test the suitability of the
OLGA tar removal system for a fuel such as RDF.
3.2 Apparatus
The experimental set up is shown in Figure 2 and consists of the bubbling fluidised bed
(BFB) reactor ”WOB” on the left with subsequently a cyclone and a hot gas filter for solids
removal, the columns (collector and absorber/stripper) of the Olga tar cleaning system and the
wet scrubber on the right side of the drawing after which the cleaned gas is send to the flare.
In general, OLGA is operated downstream of a low-efficient solids removal step (e.g. a
cyclone) and a wet electrostatic precipitator between collector and absorber. However, in the
lab-scale set up the a high efficient solids removal step (i.e. a hot gas filter) is applied upstream
OLGA. In the OLGA the product gas is cooled with oil in the collector, where the liquid tars are
collected. Further downstream, gaseous tars are absorbed in the scrubbing liquid of the second
column
The liquid tars in the collector are separated from the scrubbing liquid and returned to the
gasifier. A small amount of the scrubbing liquid is bled and recycled to the gasifier. For the
absorption step, scrubbing columns were selected that are interacting with each other in a
classical absorption-regeneration mode. The scrubbing liquid from the Absorber with the
dissolved tars is regenerated at high temperature in the Stripper. In case of air-blown
gasification, air is used to strip the tar. Subsequently, the air with the stripped tars is used as
gasifying medium. The loss of scrubbing liquid in the Stripper by volatilisation is minimised by
the use of a condenser.
OLGA
Absorber
OLGA
Collector
To flare
Biomass
BFB
Gasifier
Cyclone
Hot
Gas
Filter
Coarse
solids
Fine
solids
Aqueous
Condenser
& Scrubber
OLGA
Tar
removal
Gasification air fan
Heat
Heavy tar
Light tar
Condensate water
Inorganic impurities
(NH3 – HCl etc)
Figure 2. Process flow diagram.
3.3 Experimental procedure
Three different test series gasifying RDF were run in March, May and June 2008. The first series
was aimed at establishing the gasifyer optimum conditions and deal (when necessary) with
feeding problems. To make best use of the experiments the OLGA collector was in operation and
tar charactierisation using SPA sampling was performed in case stable enough operating
conditions could be established. Series 2 and 3 were aimed at testing different RDF streams as
well as operating the full gas cleaning train and characterizing the gas and tar composition after
each component in the experimental set up. Due to unexpected polymerisation problems the aims
of series 2 and 3 were slightly modified, as will be explained in Section 4 describing the results.
Test series 1 results led to a more fundamental procedure change as well, also for future work.
Instead of testing different streams of RDF all over again, the efforts are aimed at truly
understanding the influence of the main individual RDF fuel components and building up a
knowledge base. Series 2 with the expanded (foam) polystyrene (EPS) fuel is an example of this.
For the lab-scale installation the RDF fuel, delivered as chips, is first dried overnight at 85
oC, then milled under liquid nitrogen and sieved to a maximum size of 2 mm. The size of the fuel
particles is mainly resitricted due to the small screw feeder that transports the fuel from the
bunker into the bed of the gasifyer. The initial moisture content is re-introduced in the form of
steam in the gasifyer. For all tests the steam was set equivalent to 25% of moisture in the fuel
and the additional air for the gasification process was calculated to result in an equivalence ratio
(ER) of 0.35. The amount of fuel fed in the reactor was roughly 500 g/h and always checked
afterwards by weighting the remaining fuel. Two principle test temperatures were used, i.e. 850
oC as the temperature conversion temperature for waste incineration and 750 oC as the lower
temperature alternative, also used in the pilot scale tests in Tzum. The higher heating value of the
RDF was roughly 25 J/g.
In Table 1 the different fuels of the three test series are given, the temperatures of gasification
and some remarks on the test itself.
Table 1. Fuels, test conditions en remarks on the performed experiments.
Test
Dates
Fuel
WOB
Remarks
series
temp.
1
27th of
 RDF1 and RDF2 from 750oC
Focussed on gasification proces and
March –April
demolition and
and
feeding (bunker problems).
16th 2008
construction waste
850oC
Polymerisation problem in OLGA
(high EPS/0.23 wt%
collector
Cl and high EPS/ 0.38
wt% Cl)
2
21-23rd  RDF3 from recycler
750 oC
RDF3 experiment steam forgotten in
May 2008
(low EPS/1.8 wt% Cl) and
first part. Low EPS/high Cl
 EPS10% mix with
850oC
EPS tested as single RDF component,
wood (no Cl)
no Cl
o
3
11-12th  RDF4 from household 750 C
RDF4 without EPS, complete smooth
June 2008
packaging waste
and
operation
o
minus EPS/ 1.1 wt%
850 C
Cl
4. RESULTS
4.1 First test series
During the first test days several feeding problems and plugging problems occurred.
Reconstruction of the outlet piping of the WOB was nesscecary due to plugging with ash in
horizontal parts of the piping and in sharp bends. Also the small particles < 0.7 mm were sieved
out of the fuel. After these actions most problems were gone although stable feeding of this
inhomogenous material stays difficult at lab scale and many adjustments of the fuel feeding
system and bunker were done during this test series.
During the first test series it was established that no problems considering agglomeration of the
sand bed took place. Limited amount of ash (8%) accumulated in the bed section, roughly 50%
was collected in the cyclone and 26% in the hot gas filter. The remaining 16% was present after
the filter but reduced to 0 ± 10 mg/mn3 after the OLGA collector. In Figure 3, the temperatures
measured at various locations in the gasifier are shown with time. The irregular pattern shows
the difficulties with the feeding system. Similar graphs of test series 2 and 3 show a more smooth
behaviour. The same feeding problems are reflected in the bulk gas and minor gas compositions
shown in Figures 4 and 5. The gases were measured by on-line analysers and micro-GC.
1st April Time vs. temperature
850
800
TI3-3
TI3-4
Temperature (oC)
750
TI3-5
TI3-6
TI3-7
TI3-8
700
650
TI3-2
TIC4
TIC5
600
550
500
7:12
8:24
9:36
10:48
12:00
13:12
14:24
Time
Figure 3. WOB temperature measurements with time.
1st April Time vs. gas concentrations
25
20
mol %
mol %
mol %
mol %
mol %
15
10
O2
H2
CO
CO2
mol % CH4
5
0
7:12
8:24
9:36
10:48
12:00
13:12
14:24
Time
Figure 4. The main gas concentrations of the producer gas after the WOB with time. N2 is
excluded to benefit the scale of the other gas concentrations. The few high oxygen intervals
show the problems with the feeding system
MicroGC data RDF1, 750 oC,
01-04-2008
2
C2H4
C2H6
C2H2
1.8
1.6
mol% (dry)
1.4
1.2
1
0.8
0.6
0.4
0.2
0
10:33
11:02
11:31
12:00
12:28
12:57
13:26
Time [Hr:min]
Figure 5. Minor gas concentrations, i.e. C2H4, C2H6, C2H2 in the producer gas with time.
SPA tar sampling was performed only if gasification lasted at least 1 hour.. The concentration is
lower after the collector than before, which is unusual for a volatile tar. It is very typical fuel
related peak not seen in this concetration with any of the biomass streams tested in the past.
Naphtalene is the next highest peak but shows a reverse behaviour with the higher concentration
after the collector
Test 3, 01-04-08 Spa taranalysis RDF/Fluff tests OLGA collector
14000
12000
Concentration [mg/Nm3]
10000
8000
A, downstr. WOB avg.
C, downstr. Coll
6000
4000
2000
E
th
yl
b
o- m e n
X /p ze
yl - n
en X e
e+ yle
S ne
ty
re
P ne
he
o- no
C l
re
s
I o
m nd l
/p en
-C e
r
N es
af o
ta l
Q le
2- I uin ne
m so o
et qu lin
1- hy in e
m l- o
et na lin
hy ft e
l-n ale
af ne
E
th
ta
en
le
yl Bip ne
-n h
A ap en
ce ht yl
a
na l
e
A ph ne
ce ty
na len
ph e
t
P Flu ene
he o
na re
nt ne
A hr
n e
Fl thra ne
uo c
B
ra en
en
nt e
zo
he
(a
)- Py ne
an r
B
en
th en
z
r e
B o(b C ace
en ) h n
zo -flu ry e
(k or se
)- an ne
f
B luo the
en ra n
z n e
B o(e the
en ) n
In
zo -py e
de
(a re
no
)- ne
py
D (1
ib 23 P re
en -c e ne
z( d) ry
B ah -pe len
en )- r e
zo an yle
(g thr ne
hi ac
)- e
pe n
r e
C yle
o n
U ro e
nk ne
n n
U ow e
nk n
n s
U ow -1
nk n
n s
U ow -2
nk n
n s
U ow -3
nk n
no sw 4
ns
-5
0
Components
Figure 6. Tar concentrations in the productgas before and after the OLGA collector column
During all the tests in series 1, a significant and unusual temperature increase was registrated in
the upper sections of the OLGA collector column (except for the restricted exit temperature).
Also the pressure drop over the collector column increased and testing had to be stopped.The
increase occurred faster with higher chlorine contents (RDF2) and at lower gasification
temperature (750 0C). The temperature increase is shown in Figure 7 by the light blue line as an
example of one section. The temperature raise is though to be a result of an exothermal reaction.
The OLGA collector was dismounted and the different sections were inspected on fouling.
The lower sections were clean, the upper 3-4 sections showed some fouling with black and
sticky material, thought to be polymerized tars produced during the exothermal reaction.
OLGA Collector RDF (fluff)
Comparison temperature behaviour tests on 27-3-08 and 16-4-08
400
250
350
200
150
200
100
P oilfilter [mbar]
250
o
Temperature [ C]
300
150
100
50
Test 27-3-08
T gasifier 750oC, high styrene
Fluff RDF1 (low Chlorine)
50
0
9:53
Test 16-4-08
T gasifier 750oC, high styrene
Fluff RDF2 (high Chlorine)
0
10:09
10:26
10:43
10:59
11:16
11:33
11:49
8:07
8:23
8:40
8:57
9:13
9:30
Time [H:m]
T Collector inlet
T oil inlet
T segment 8
P Oilfilter
Figure 7. Collector temperature.profiles and oil pressure, RDF1 test1 (27-3) and RDF2 test 7 (16-4). The
dark blue line and the pink line represent the fixed in- and outgoing temperatures, respectively.
On the black fouling deposit, sampled from the OLGA collector some first analyses were
performed. Some preliminary FTIR analyses showed the main component to be a rubber-like
substance . More deta iled analyses will be performed on the deposit sample soon.
4.2 Second test series
The extremely high styrene peak in the first series, related to the RDF fuel, which was reduced
after the collector, the exothermal (polymerization) reaction likely to have resulted in the fouling
as well as the aggrevating effect of chlorine led to the approach of the second series to test the
EPS component seperately and without chlorine and to choose an RDF high in chlorine and low
EPS. RDF3 was a milled sample without any visible EPS, however, the styrene peak during the
test run suggested it was not very low in EPS. A better choice, RDF4 was chosen for test series
3. Here the results of the EPS test are described. Figure 8 shows the bulk gas composition as an
example of the smooth operation. In Figure 9 the light blue linenshows again the exothermal
reaction, although no fouling was established. It shows EPS is an important factor in the problem
but its presence alone is not enough to cause the fouling.
OLGA Collector bedreven op 10 wt% Polystyreen in beukensnippers gestookte WOB
Gassamenstelling mGC van test 3, 23-05-08 (750 en 850oC)
25
70
60
20
15
40
30
10
Concentratie N2 [vol%]
Concentratie [vol.%]
50
20
5
10
0
0
14 23 33 42 52 0 2 1 1 2 1 3 0 4 0 5 0 5 9 0 9 1 9 2 8 3 8 4 7 5 7 0 7 1 6 2 6 3 5 4 5 5 5 0 4 1 4 2 3 3 3 4 3 5 2 0 2 1 1 2 1 3 1 4 0 5 0 5 9 0 9 1 9 2 8 3 8 4 7 5 7 0 7 1 6
9: 9: 9: 9: 9: 10: 10: 10: 10: 10: 10: 10: 11: 11: 11: 11: 11: 11: 12: 12: 12: 12: 12: 12: 13: 13: 13: 13: 13: 13: 14: 14: 14: 14: 14: 14: 14: 15: 15: 15: 15: 15: 15: 16: 16:
Tijd [H:m]
H2
CH4
CO
CO2
N2
Figure 8. The main gas concentrations of the producer gas after the WOB with time. N2 is on a
seperate scale on the right axis.
OLGA Collector, 10wt% polystyrene in beechchips
o
Temperature profiles, 23-05-08 (750 en 850 C)
400
300
350
250
200
250
200
150
150
P oilfilter [mbar]
o
Temperature [ C]
300
100
100
50
50
850 oC
o
750 C
0
08:18:00
0
09:08:00
09:58:00
10:48:00
11:38:00
12:28:00
13:18:00
14:08:00
14:58:00
15:48:00
Time [H:m]
T Collector inlet
T oil inlet
T segment 8
P Oilfilter
Figure 9. Temperature profiles of the collector column test 3 on May 23 rd.
4.3 Third test series
On the 11th and 12th of June experiments were performed using RDF4 from household
packaging waste, seperately collected by Hummel Recycling, from which all visible EPS was
removed. However, this type of packaging waste contains other forms of solid (non-foam)
polystyrene which remained present. The test did run smoothly as can be seen from Figure 10
showing the bulk gas composition. The tar composition, sampled before and after the collector of
OLGA (Figure 11) shows that no styrene remains in the collector. Otherwise, most of the tars are
much higher before the collector (in blue) compared to the concetrations after the collector (in
purple) showing a very good perfomance in tar reduction. No temperature rise in the upper
sections of the collector could be detected (see Figure 12) and no fouling took place.
Gasconcentrations Hummel fluff test 12-06-08
H2
CO
CO2
CH4
20
18
16
mol% (dry)
14
12
10
8
6
4
2
0
7:12
7:40
8:09
8:38
9:07
9:36
10:04 10:33 11:02 11:31 12:00 12:28 12:57 13:26 13:55 14:24 14:52 15:21 15:50 16:19 16:48
Time [hr:min]
Figure 10. The main gas concentrations of the producer gas after the WOB with time. N2 is
excluded.
SPA tar results up- and downstream collector Hummel fluff 12-06-2008
6000
avg. upstream coll 11:06 and 11:11 hr
no styrene lost in collector
Concentration [mg/Nm3]
5000
avg. downstream coll 10:33 and 10:37 hr
4000
3000
2000
1000
0
-5
ns
w -4
no ns
nk w 3
U no sn
nk w 2
U no nsnk w 1
U no nsne
nk w
le e
U no e ry cen e
e
nk n p ra en
U one hi)- nth ryl
or (g a pe
C zo h)- d)en (a -c
B z 3
n
2
e
ib (1
e
D no
en
de ne yr
e
In yle )-p ene en
er (a yr th e
P zo )-p ran en
h
en (e o t
B zo )-flu ran
en (k o
B zo )-flu
ne
ce
en (b
B zo e hra
en n nt
B yse )-a
hr (a
C zo
e
en
B ene hen
yr nt
P ra ne
uo ce ne
Fl hra hre
nt nt
A na
he ne e
P re ten e e
uo ph en n
Fl na tyl tale
ce ph h
A na nap
ne
ce lA eny l ale
e
th y ft n
E en l-na ale
h
ip y ft
B eth l-na
m y
1- eth line
m o
2- ui n e
oq in
Is nol e
ui n
Q tale sol
af e
N Cr
/p ol
m r es
C
ne
o- ne
re
de l ty
In n o + S
he ne
P yle ne
X le ne
o- -Xy ze
/p n
m ylbe
th
E
Component
Figure 11. Tar concentrations in the product gas, before and after the OLGA collector.
Hummel Fluff fuel WOB
Collector temperature profile on 12-06-2008
350
250
300
200
150
200
150
100
P Oilfilter [mbar]
Temperature [oC]
250
100
50
50
nothing happens here, stable temp.
0
0
:5
15
0
:3
15
0
:1
15
0
:5
14
0
:3
14
0
:1
14
0
:5
13
0
:3
13
0
13
:1
0
:5
12
0
:3
12
0
12
:1
0
:5
11
0
:3
11
0
:1
11
0
:5
10
0
:3
10
50
30
10
50
30
10
:1
10
9:
9:
9:
8:
8:
8:
50
0
7:
7:
30
0
Time [H:m]
T Collector inlet
T oil inlet
T segment 8
P Oilfilter
Figure 12. Temperature profiles of the colector column on June 12th. Note that the light blue line
does not indicate at an exothermal process.
5. DISCUSSION AND CONCLUSIONS
Despite some feeding and plugging problems during the first test series, a generally good
performance of the fuidised bed gasifier on RDF was shown.The bulk and minor product gas
composition with time show a stable trend. No signs of agglomeration of the sand bed were
observed. Overal, the application of a fluidised bed as dedicated installation for RDF gasification
is promising.
Focussing further on the tar removal system, optimization of the OLGA is needed because the
tar concentrations are different for RDF than for the biomass fuel for which it was designed.
Sample analyses during the experiments on the product gas showed that the most prominent peak
in the tar concentrations is that of xylene/styrene. Cleaning the gas from the gasifier with OLGA,
using RDF as a fuel, resulted in a fouling problem in the OLGA collector column during the first
test series. The temperature rise was probably caused by an exothermic reaction. These problems
occur faster gasifying at 750 oC than at 850 oC. Other RDF from household packaging waste
without expanded (foam) polystyrene (EPS) did not give any problems. The polystyrene itself
used as a fuel in 10% mixture with beech did not result in fouling problems either
The formation of a rubber-like substance within the gas cleaning section seems related to the
plastics in RDF. A typical composition of RDF contains among others polyurethane foam
(PUR), polyethylene (PE), polypropylene (PP), polyvinyl chloride (PVC), acrylonitrile butadiene
styrene (ABS) and polystyrene (PS). By gasifying the RDF material, these components will
decompose into the individual building blocks ethylene, propylene, butadiene and/or styrene,
which via chain reaction polymerization can form an heterogeneous mixture of different difficult
to identify thermoplastic polymers, although polymerization of butadiene and styrene is more
likely to occur than of propylene or even ethylene.
A new hypothesis will be experimentally tested in future work. It is thought that the
polymerization can be attributed to the presence of polystyrene in the RDF fuel, chlorine as a
catalyst for the polymerization reactions and at least one other tar component, likely released
from another RDF component. Future efforts will further be aimed at understanding the
influence of the main individual RDF components and building up a knowledge base to use as a
predictive tool to avoid operational problems.
ACKNOWLEDGEMENTS
SenterNovem is kindly thanked for the financial contribution to the project ISO73013 of which
the results are presented.
REFERENCES
Boerrigter, H.; Van Paasen, S.V.B.; Bergman, P.C.A.; Könemann, H.W.J.; Emmen, R.; Wijnands,
A.,“OLGA” Tar Removal Technology Proof of concept(PoC) for application in integrated biomass
gasification combined heat and power (CHP) systems, ECN Biomass, Coal & Environmental
Research, Dahlman Industrial Group, report ECNC—05-009, January 2005.
Könemann H.W.J.; Egas, F.; van Paasen, S.V.B., TarRemoval in Biomass Gasification Processes;Dahlman Industrial Group, Paper for ERTC Paris November 2006