F emissions from wood burning facilities

Chemosphere, Vol. 29, Nos. 9--11, pp. 1927-1938, 1994

Pergamon 0045-6535(94)00318-1

Elsevier Science Ltd Printed in Great Britain 0045-6535/94 $7.00+0.00

D E T E R M I N A T I O N A N D REDUCTION OF PCDD/F E M I S S I O N S F R O M W O O D BURNING FACILITIES

Kolenda J.A, Gass H.A, Wilken M. ~, Jager j.c, Zeschmar-Lahl B. D

A

MPU GmbH, Kottbusser Damm 86, D - 10967 Berlin, FRG

B

ITU GmbH, Abt. ITU-Forschung, Kolonnenstr. 26, D - 10829 Berlin, FRG

C

Institut WAR, TH Darmstadt, Petersenstr. 13, D - 64287 Darmstadt, FRG

D

MPU GmbH, Lindenstr. 33, D - 28876 Oyten, FRG

Abstract Seven wood burning facilities were analyzed for their emissions of PCDD/F and other parameters while the input was varied. PCDD/F emission concentrations ranged from 0.004 to 9.820 ng ITEQ/Nm 3, with the majority of values > 0.1 ng I-TEQ/Nm 3, even if the input was untreated wood. Highest concentrations occurred after addition of halogenated materials (NH4CI hardened or PVC coated plywood) to the input. Using typical input materials, facilities with smaller thermal capacity (< 1 MW) showed emission freights ranging from 0.03 to 0.4 pg I-TEQ/h, while the large ones (> 1 MW) emitted 0.07 to 0.5/Jg I-TEQ/h. Using treated input materials, emission freights of smaller facilities reached 1.6 pg I-TEQ/h, while the large plants released up to 7.2 pg I-TEQ/h.

1.

Objectives

PCDD/F are known to be generated in many thermal processes like e.g. waste incineration or metallurgical processes. Analyses of chimney soot from different heating systems [1] showed wood combustion to be a possible generation process, too. Within the research and development project of the German Federal Environmental Agency (Umweltbundesamt) "Determination and Reduction of Halogenated Dioxin and Furan Emissions from Thermal Processes", in 1992/93 various wood burning facilities were monitored with regard to their potential to produce and release PCDD/F [2].

1927

1928

2. Approach and methods For the intended survey, seven wood burning plants were selected for clean gas analyses under aspects of capacity, feeding and furnace type and input material. At two plants, burning material was varied in order to get information on the influence of the input material, of special organic compounds (PVC-coating) and the inorganic hardener [ammonium sulphate = (NH4)2SO4, and ammonium chloride = NH4CI] of plywood on dioxin emission. Clean gas was analyzed for PCDD/F, dust, HCI, SO 3, (all discontinuously) and 02, SO 2, NOx, CO, CO 2, and in some cases TOC (all continuously).

PCDD/F analysis: Sample collection followed VDI [3] (condensation method). Samples are spiked with a mixture of a 13C12-1abelled PCDD- and PCDF-standard per degree of chlorination and then extracted with toluene. Clean-up of the raw extract and fractioning by adsorption chromatography on silicagel- and aluminum oxide columns follow. The PCDD/F containing fraction is concentrated and a defined aliquot of the final volume is injected into a GC/MS. Screening analysis and detection of higher chlorinated congeners is performed with a non-polar DB-5 capillary column, isomer specific detection uses a polar capillary column (CP Sil 88), the MS performing in MID-mode (multiple ion detection). Quantification is based on standard addition, with a recovery rate of the internal standard of at least 70 %.

3. Results

Tables 1 to 7 compile some technical data of the plants analyzed and the respective results. Emission concentrations ranged from 0.004 to 9.820 ng I-TEQ/Nm 3, with the majority of values > 0.1 ng ITEQ/Nm 3.

4. Conclusions

Even the use of natural input materials (untreated wood) produced emissions above 0.1 ng I-TEQ/Nm z, the emission limit for waste incineration plants (17. BlmSchV), considering facilities with insufficient combustion conditions (high emissions of CO). Similar findings are reported by Schatowitz et aL, who observed emission concentrations at automatic wood chip furnaces (110 - 1,800

kW) of 0.066 to 0.214 ng I-TEQ/Nm 3 [4].

Highest dioxin concentrations of up to 9.82 ng I-TEQ/Nm 3 occurred after addition of halogenated materials ((NH4)2SO 4 hardened and PVC coated plywood) to the input (table 8), but a statistically significant correlation between chlorine input and PCDD/F output cannot be derived. Wurst et al.

1929

[5] gained similar results after PCV addition to the input of a small wood burning plant (thermal capacity: 50 kW, performance level: 18 - 23 kW).

Table 1: PCDD/F and CO in wood burning facility no. 1

Plant Type

Test Run No.

Loading

Max. Fuel

Dust

and

Index

Throughput

Filter

Burning Material

(%)

(kg/h)

95-110 %

10

Hand fed

1

wood blocks, coated and

chute

+

uncoated plywood, wood

incinerator

2

residues

none

95-110 %

48 kW Emission concentrations referring to 13 vol,-% 0 2 PCDD/F in ng I-TEQ/Ncbm 10

CO in g/Ncbm 10

1-

1

0.1-

0.01-

0.1

1.1

1.2

0 01

Test Run No.

l

EE PCDD/F [ ] CO

The high emission concentrations observed after burning (NH4)2SO4 hardened and PVC coated plywood) are in the same order of magnitude Schatowitz et al. [4] observed when waste wood chips from demolition of buildings, presumably containing chlorinated wood treating agents (PCP etc.) were incinerated (up to 14.42 ng I-TEQ/Nma).

1930

Table 2: PCDD/F and CO in wood burning facility no. 2

Plant Type

Prefiring system

Test Run No.

Loading

Max. Fuel

Dust

and

'ndex

Throughput

Filter

Burning Material

(%)

(kg/h)

80 %

45

1 briquets of chopped woo

multi

+

cyclone 85 %

2

211 kW Emission concentrations referring to 13 vol.-% 0 2 CO in g/Ncbm

PCDD/F in ng I-TEQ/Ncbm 10

10

0.1

0.1

0.01

i 0.01 2.1

2.2 Test Run No. •

PCDD/F [ ] CO

Very clear relationships were determined between CO and PCDD/F emission concentrations. This was especially true for production specific wood wastes and when halogenated input materials were added. Therefore, it can be concluded that the addition of halogenated input materials produced a worser combustion quality at the facilities monitored. The dominant influence of combustion quality on PCDD/F emission concentrations can be seen from the increasing CO concentrations and simultaneously decreasing flue gas temperatures.

1931

Table 3: PCDD/F and CO in wood burning facWtty no. 3

Test Run No.

Loading

Max. Fuel

Dust

and

Index

Throughput

Filter

Burning Material

(%)

(kg/h)

all:

2,400

Plant Type

Fuel

1

mix of untreated wood and

injection

+

wood based product

incinerator

2

residues

3

untreated pine wood

9 0 - 110 %

filter

+ 4

9.6 MW

5

plywood, hardened with

+

(NH4)2S04

6 7

plywood, hardened with

+

(NH4)2SO4 and coated with

8

PVC

Emission concentrations referring to 11 vol.-% 0 2 PCDD/F in ng I-TEQ/Ncbm 10

CO in g/Ncbm 10

0.1

0.1

0.01 0.001

0.01

3.1

3.2

3.3

3.4

3.5

Test Run No.

3.6

3.7

electro

3.8

0.001

1932 Table 4: PCDD/F and CO in wood burning facility no. 4

Plant Type

Test Run No.

Loading

Max. Fuel

Dust

and

Index

Throughput

Filter

Burning Material

(%)

(kg/h)

90 %

5,300

electro

(grate)

filter

Moving

1

wood pieces, coated and

grate

+

uncoated plywood, wood,

incinerator

2

bark, saw dust

90 %

3,000 (dust)

39 M W

Emission concentrations referring to 11 vol.-% 0 2 CO in g/Ncbm 10

PCDD/F in ng I-TEQ/Ncbm 10-

0.1

0.1

0.01

4.2

4.1

0.01

Test Run No. IIPCDD/F DCO I Two of the three large facilities (> 1 MW) showed distinctly lower PCDD/F emission concentrations than the plants with smaller thermal capacity. Thus, the facility size tends to have an influence on PCDD/F concentration due to more accurately managed combustion conditions. With the addition of input materials containing large quantities of halogen at one of these large facilities, a significant increase in dioxin concentration could also be measured in the emissions. However, these concentrations remained below those determined at small plants using typical plant input and even natural woods.

1933

Table 5: PCDD/F and CO in wood burning facility no. 5

Test Run No.

Loading

Max. Fuel

Dust

and

Index

Throughput

Filter

Burning Material

(%)

(kg/h)

all:

50

Plant Type

1

mix of untreated wood and

+

wood-based residues

Stoker incinerator

about

2

234 kW

cyclone

30 - 60 %

3

plywood, hardened with

+

(NH4)2S04

4 5

plywood, hardened with

+

(NH4)2SO4 and coated with

6

PVC

7

plywood, hardened with

+

NH4CI

8 9

untreated pine wood

+ 10

Emission concentrations referring to 13 vol.-% 0 2 PCDD/F in ng I-TEQ/Ncbm

10

CO in g/Ncbm

10

=

1

1

0.1

0.01

0.1

5.1

5.2

5.3

5.4

5.5

5.6

5.7

Test Run No.

I= coo, []co ]

5.8

5.9 5.10

0.01

1934

Table 6: PCDD/F and CO in wood burning facility no. 6

Plant Type

Stoker incinerator

3.5 M W

Test Run No.

Loading

Max. Fuel

Dust

and

Index

Throughput

Filter

Burning Material

(%)

(kg/h)

5 0 - 60 %

726

1

coated and uncoated ply-

+

wood, hardened with

2

(NH4)2SO4

electro filter

6 0 - 70 %

1935

Table 7: PCDD/F and CO in wood burning facility no. 7

Plant Type

Stoker incinerator

1

Test Run No.

Loading

Max. Fuel

Dust

and

Index

Throughput

Filter

Burning Material

(%)

(kg/h)

60 - 70 %

25

briquets of chopped wood

none

+ 2

60-70

%

117 kW

Emission concentrations referring to 13 vol.-% 02 PCDD/F in ng I-TEQ/Ncbm 10

0.1

0.01

CO in g/Ncbm 10

0.1

7.1

7.2 Test Run No. BB PCDD/F ~ CO J

0.01

1936 Table 8: PCDD/F emission concentrations in relation to the input material

Referring Plants

Fuel Type

Emission concentrations ng I-TEQ/m 3

untreated wood usual plant input* plywood hardened with NH4CI and/or

2,3,5

0.004 - 0.697

1,2,3,4,6,7

0 . 0 0 6 - 1.55

1,3,4,5,6

0.028 - 3.30

(NH4)2S04 (NH4)2SO4 plus PVC

3,5

0.297 - 9.82

* untreated wood, coated and uncoated plywood, mix of the mentioned fuel types

Facilities with smaller thermal capacity (< 1 MW) using typical input materials generally showed emission less than 0.5 pg I-TEQ/h (Fig. 1).

Fig. 1 :

PCDD/F emissions freights of smaller wood burning plants (thermal capacity < 1 MW)

Test Run No. ~:~ ~

wood and plywood coated and uncoated } briquets of c:hopped wooJ

| mix of untl eated wood tnd woo J-based prod ~ct residues } plywoc ~d hardened ~vith (NH4)2S 04 ... and coat( ,d with PVC } plywood h~hrdened with NH4CI

I

s!i

~} untreated pine wood

7.1 L . } briquets of chopped ~ood 7.2m 0

0.5

1

pg I-TEQ/h

1.5

2

1937 Comparable loads in the large facilities could only be determined at the one which had lowest PCDD/F concentrations when typical input materials were incinerated. With the addition of atypical input materials at this as well as at the other two large facilities, emission loads of up to 7.2 pg ITEQ/h were determined (Fig. 2). Based on these data, an annual load of 1 to 50 mg I-TEQ/a per plant could be estimated. For a less populated and industrialized region such a plant can contribute up to 15 % of the total PCDD/F emission.

Fig. 2:

PCDD/F emissions freights of large wood burning plants (thermal capacity > 1 MW)

Test Run No. 3.1 ~ 3.2 | 3.3 3.4 3.5 3.6 3.7 3.8

mix of t lntreated wooc and wood-ba;ed produ1:t residues

~ } untreated )ine wood ~ BB

} plywood hardened with (NH4)2S04

~

~

4.1 ~ 4.2 ~

~

~ ~

6.1 6.2

~

~=m

... ~nd coated with PVC woodp eces, coated im and uncoap l ~ ~ J ~ ~ t e d ~ d u s t

coated and uncoated plywooL,, hardened with (NH4)2SO4 0

2

4 pg I-TEQ/h

6

8

As raw gas analyses have not been performed, the influence of dust removal on dioxin emission concentrations cannot be quantified. An optimized particle removal system can play a positive role in emission reduction. But more effective measures seem to be the optimization of the combustion process to improve combustion quality, and the substitution of halogenated materials in the wood working industry for decreasing the input of halogens into the incineration process.

An emission standard of 0.1 ng I-TEQ/Nm 3 could be considered as a goal for advanced wood burning facilities using these reduction techniques.

5. Acknowledgement

The authors wish to thank the German Federal Environmental Agency for financial support of this research program.

1938

6. References

[1 ]

Thoma H.: PCDD/F concentrations in chimney soot from house heating systems. Chemosphere 1988;17:1369-79

[2]

Wilken M., Kolenda J., Gass H.: Ermittlung und Verminderung der Emissionen von halogenierten Dioxinen und Furanen aus thermischen Prozessen - Holzfeuerungsanlagen. UBA-FB 104 03 365/06, 10.5.1993

[3]

Draft, VDI guideline 3499, page 2, Condensation Method. Society of German Engineers, Duesseldorf, Germany

[4]

Schatowitz B., Brandt G., Gafner F., Schlumpf E., BOhler R., Hasler P., Nussbaumer T.: Dioxin emission from wood combustion. OrganohalogenCompounds 1993; 11:307-10

[5]

Wurst F., Prey T., Boos R., Scheidl K.: Untersuchungen zur PCDD/F-Emission bei der Holzverbrennung. OrganohalogenCompounds 1991 ;7:197-9