REFUSE DERIVED FUEL (RDF)

by

S. H. Russell

Henningson, Durham & Richardson, Inc. Omaha, Nebraska

Introduction

The term "RDF" is a relatively new member of the advancing mass of resource recovery jargon which is exchanged freely between those involved daily with the problems of recovering energy from solid waste. It is one of those three-letter acronyms that rolls easily off the tongue, but is dangerously ambiguous. Taken literally, "Refuse Derived Fuel" means any fuel derived from refuse including shredded fuel, pyrolysis gas and oil, landfill gas, etc. However, RDF in this paper is defined as that fraction of municipal solid waste composed mainly of cellulose and petroleum derivatives, whether mixed with other non-combustibles in its un-processed form, or in its various processed forms. The purpose of this paper is to discuss the direct combustion of this fraction in a steam generator.

Direct Combustion

Figure 1 shows where the direct combustion technique fits in the context of the rest of resource recovery technology. Two general groups can be defined; Materials Separation and Chemical Conversion. Materials Separation is the physical separation and/or sizing of the various com­ponents of solid waste without changing the chemical makeup of any com­ponent for the· purpose of preparing them for further physical changes, chemical conversion, or disposal. A few of the many techniques in this general group are shredding, magnetic separation, density separation, screening, nonferrous separation, and various "wet" processes using water as a transport fluid. The second general group of which direct combustion is a part, is Chemical Conversion. This group includes techniques which change the chemical makeup of either processed or un-processed solid waste for the purpose of recovering the heat released in the chemical reaction and/or for changing the solid waste into a more usable form. Pyrolysis and Bioconversion are included as well as direct combustion.

This paper will discuss the two general forms of the direct com­bustion of RDF (as it has been defined earlier); the use of RDF as a primary fuel, and its use as a supplemental fuel. This paper will also attempt to identify needed areas of research for each of these direct combustion forms.

300

RDF as a Primary Fuel

This form of direct combustion uses RDF as the source of greater than fifty percent of the heat input to the steam generator. This requires a steam generator designed specifically for the combustion of refuse when using RDF in its un-processed form. In certain situtations, depending on the type of solid waste processing used to produce the RDF, existing boilers can be modified to burn processed RDF as a primary fuel. In either case, the RDF heat input is usually supplemented by a "conventional" fuel such as gas, oil or coal. Figure 2 shows the typical processing steps required for the different ways of utilizing RDF as a primary fuel; unprocessed fuel firing (mass burn) and processed fuel firing. These are the two types of primary refuse fuel direct combustion now being practiced by certain facilities in this country and in Europe.

Unprocessed RDF Firing (Mass Burn)

There are several operating facilities of this type in the U. S. , Canada and Europe. Table 1 gives the details of most of these facilities. Figure 3 shows the general arrangement of the equipment used in these facilities. The waste is dumped from the collection or transport vehicles into a pit for surge storage, then is moved to a hopper which feeds a traveling/r�rocating grate arrangement on which the refuse is burned. Heat is transferred by radiation and convection in the traditional manner to the boiler walls and hanging tube sections. Some systems separate bulky items such as refridgerators before firing and recover ferrous metal from the ash with magnetic separators.

Processed RDF Firing

In contrast to unprocessed fuel firing, this technique subjects the raw refuse to a· number of different steps of materials separation. before combustion. Most of the systems whether operating or in the final design o f construction phases utilize a new, specially designed RDF boiler with a travel ing grate, spreader stoker design originally used for crushed coal (see Figure 4). The refuse is typically shredded to a 3 to 6 inch (7.62 to 15.24 em) nominal particle size and magnetically separated to recover a ferrous metal fraction before firing ( some systems use air density separation as an additional step). When processed refuse is to be used as a primary fuel in an existing boiler, the raw refuse is typically subjected to additional shredding for smaller particle sizes,and density classified to remove additional noncombustibles. Boilers originally designed to use a solid fuel (coal, bark, etc. ) can then be modified to burn this RDF as the primary fuel.

Table 2 gives the details of many of the systems operating, under design or under construction.

301

>

Comparison and Research Needs

Both the unprocessed RDF firing and processed RDF firing primary fuel systems have been constructed and have operating experience. The question of which form is more desirable is unanswered. A comparison between the two not only points out the advantages and disadvantages of each, but identifies certain areas for needed research.

Steam flow variation.--It has been shown that steam flow variations are less severe in a processed RDF primary fuel system than in an unprocessed RDF system.l Processed fuel has a higher degree of homogeneity than raw refuse creating a more uniform heat release, and a fire which is easier to control. Unprocessed fuel units have fuel firing systems which are modi­fications of systems originally designed for incineration of raw refuse. Many times, the goals of'refuse incineration conflict with the goals of steam generation from unprocessed refuse. A goal of incineration is not necessa­rily to create a uniform heat release rate which is important in steam generation. An area for research is to identify how to control steam flow variations in an unprocessed fuel system. Areas for investigation are: boiler control technology, traveling or reciprocating grate design, and auxiliary fuel modulating techniques.

Corrosion.--Refuse derived fuel has been shown to be more corrosive than conventional fuels.2 The small amount of testing which has been completed indicates that greater amounts and rates of chloride corrosion occurs in the fireside of the boiler apparently because of the polyvinal­chloride polymer which exists in municipal waste in the form of trash bags, plastic wrap, etc. This problem is perhaps more acute in a system using RDF as primary fuel than in a supplemental fuel system. There is also speculation that this type of corrosion is more severe in a mass burn system than in a processed RDF system.

It is clear that only limited knowledge exists about the actual corrosion mechanism, and the conditions which control it, such as tempera­ture, gas flow rate, excess air rate, boiler geometry, RDF composition, and firing system design. Research directed toward answering boiler corrosion questions could result in the design of more reliable refuse direct combustion systems.

Stack emissions.--Only limited stack emissions testing has been completed for steam generators using "refuse derived fuel as a primary fuel. Considering the number of RDF systems either unde� const�ction or in the design or planning stages, a high priority research need should be the identification of stack emissions from RDF systems.

There is speculation about which of the primary RDF fuel systems has more severe emissions control problem. Since a higher gas flow is needed for complete combustion in a mass burn system, the problem is potentially greater. The East Hamilton SWARU (a processed fuel system)

302

boilers are designed for 37% excess air leaving the boiler as compared to an excess air figure of 84% at the boiler exit in the Nashville mass burn system. Higher air flow per Btu released means more gas to clean up. Research should address the effects on stack emissions of refuse fuel homogeneity, fuel composition, firing system design, boiler geometry, fire control with auxiliary fuel, and air pollution control equipment design and operation.

RDF as a Supplemental Fuel

The major differences between an RDF primary fuel system and a supplemental fuel system are that in the supplemental system, a conven­tional fuel (gas, oil or coal) supplies over 50% of the heat input and an existing boiler is usually modified for RDF supplemental firing. In certain situations, RDF can be used as a primary fuel in existing boilers (see Table 2), but modification costs are usually prohibitive. The primary attractiveness of the supplemental fuel concept is that when RDF is only used as a fuel supplement, considerable investment capital can be saved because an existing boiler can be modified for use at a relatively small cost. EVen thoug�amore extensive materials separation system is required to prepare an RDF with better fuel properties, the cost of these components is usually more than offset by the savings realized by the elimination of the necessity for a new, specially-designed boiler.

Figure 2, shows that there are two basic types of supplemental RDF systems; the shredded, :classified fuel system and the densified RDF (d-RDF) system.

Shredded, Classified Fuel Supplemental System

The materials separation involved with this type of supplemental fuel system involves size reduction to typically under 1-1/2 inch (3.8lcm) particle size (possibly two stages of shredding or milling) and density classification (usually air density separation). The fuel is then burned either on a traveling grate or in suspension. Table 3 shows the details of the steam generators which now or will fire supplemental shredded RDF. Supplemental rates range from 10% to a possible 50% in these existing steam generators. The only steam generators operating with a shredded RDF supplemental fuel on a day-to-day basis are those operated by the Ames Municipal Electric Service in Ames, Iowa. These three units have been utilizing RDF constantly for about one year at the date of this writing.

Densified Refuse Derived Fuel d-RDF Supplemental System

The advantages of densified or pelletized RDF are: 1) densified RDF will store for longer periods of time than shredded RDF before decomposing and 2) the higher bulk density and shape of the densified fuel pellets

303

may allow this fuel to be handled with exisitng coal-handling equipment which is not compatible with shredded fuel, reducing boiler modification costs. Consequently, several companies now market systems or equipment which produce various forms of densified fuel. Figure 2 shows that the d-RDF is typically produced with either wet or dry methods involving size reduction to less than I inch (2.54 cm) nominal particle size, magnetic separation, density separation, drying, and densification.

No densified RDF supplemental fuel system is currently operating on a full-time daily basis. However, a number of tests of the firing of d-RDF as a supplement to coal have been done in existing steam generators. The details of some of these tests are shown in Table 4. All of the tests shown on the Table were conducted with no �rior modifications to the fuel handling systems or to the boilers.

Comparison and Research Needs

Economics.--It appears that in certain situations, the capital and operating costs for a shredded, classified RDF supplemental system are lower than those for a d-RDF system. However, many of the special costs of supplemtntal firing with either shredded or densified RDF have not been identified such as: extra ash handling costs, extra maintenance costs due to higher boiler tube corrosion, shredded fuel conveying system main­tenance costs (potentially high in pneumatic systems), etc. Research to identify these costs is needed.

Combustion.--Especially in a suspension firing arrangement, RDF particle size is very critical. Very little is known about what the optimum particle size is for the suspension firing of supplemental shredded, classified RDF for various sizes of steam generators. The optimum particle size for traveling grate firing systems is also not known for different systems and sizes of boilers. Research in this area could lead to the more efficient combustion of RDF. There has also been recent effort to test d-RDF in a mixture with coal as a fuel in a pulverized coal system. This would require that the d-RDF pellets be pulverized along with the coal in the existing pulverizing. mills. Very little is known about the technical and economic feasibility of this concept. Research is needed in this area.

Conclusion

It is certain that the direct combustion of Refus�Derived Fuel will receive increasing scrutiny in the future as a technically and economically viable method for both solid waste disposal and energy production. Both forms of primary fuel RDF systems; unprocessed fuel firing (Mass Burn) and processed fuel firing, along with both forms of supplemental fuel RDF systems, shredded, classified fuel and densified fuel, have shown initial technical feasibility. Research in the areas of steam flow Variation,

304

corrosion, stack emissions, �conomics, and combustion efficiency, could serve to reduce-the uncertainty inherent in this new technology.

305

Footnotes

lSommerlad, Robert E. Quality and Characterisitcs of Steam Produced

From Wastes, unpublished paper presented at the Engineering Foundation Conference on Present Status and Research Needs in Energy Recovery from Solid Wastes, September 21, 1976.

�aughan, D. A. Corrosion Mechanisms in Municipal Incinerators Versus Refuse Composition, unpublished paper presented at the Engineering Foundation Conference of Present Status and Research Needs in Energy Recovery from Solid Wastes, September 20 1976.

306

Lo

ca

tio

n

Ch

ica

go

, 1l

1in

01s

Br

ain

tre

e,

Ma

ss

.

Ha

rr

isb

urg

, P

A

w

0

Mo

ntr

ea

l,

-..J

Qu

e

Na

sh

vill

e,

Te

nn

.

No

rfo

lk,

VA

Qu

eb

ec

Cty

, Q

ue

.

Sua

gus

, M

as

s.

TA

13L

£

I

OP

ER

AT

ING

ST

EA

M G

EN

ER

AT

OR

S F

1RIN

C;--U

NP

HO

CE

SSE

D R

EF

US

E

(MASS

BU

RN

)

Ca

pac

ity

S

tar

t-U

p

sh

or

t to

ns

/d

ay

°

Eer

ato

r

Ye

ar

(m

etr

ic t

on

s/

da

:z:y

Cit

y o

f 1

97

2

180

0

Ch

ica

go

( 1

63

3)

To

wn

of

19

71

2

40

Br

ain

tre

e

(21

8)

Cit

y o

f 19

72

7

20

Ha

rr

isb

urg

(6

53

)

Cit

y o

f 1

97

0

12

00

Mo

ntr

ea

l ( 1

08

8)

Na

sh

vil

le

19

74

7

20

Th

er

ma

l (6

53

) T

ra

nsfe

r

Co

rp

.

U.S

. N

avy

1

96

7

36

0

(32

7)

Que

be

c U

rb

an

1

97

4

10

00

Co

nirnu

ri it

l'

(90

7)

RE

SC

O

19

75

1

50

0

( 13

61

)

Bo

ile

r D

es

ign

Dat

a

Ib/

hr

x

10

00

(Kg/

hr

x

10

00

)

4

unit

s,

ea

ch

11

0

(50

)

2 u

nit

s,

ea

ch

30

(13

.6

)

2 u

nit

s,

ea

ch

92

.5

(37

.4

)

4

unit

s,

ea

ch

10

0

(45

.3

)

2

un

its

, e

ac

h 1

35

(61

. 2

)

2

un

its

, e

ac

h 5

0

(22

.7

)

2

unit

s,

ea

ch

81

(36

.-7

)

2 u

nit

s,

ea

ch

18

5

(83

.9

)

ps

ia,

° [

(Kg

/c

m

,oC

)

27

5,

41

4

(19

.3

,2

12

)

25

0,

sa

t.

(17

.5

, s

at.

)

25

0,

45

6

(17

.5

, 2

36

)

22

5,

50

0

(15

.8

,2

60

)

40

0,

60

0

(28

. 1

,3

16

)

27

5,

sa

t.

(19

.3

, s

at)

68

0,

60

0

(47

.8

,3

16

)

89

0,

87

5

(62

.5

, 4

86

)

Aux

illa

ry

F

uel

ga

s

ga

s

oil

oil

ga

s

oil

oil

oil

Bo

ile

r M

an

ufa

ctu

re

r

Wa

lth

er

Gm

bH

R H

ey

Sto

ke

r

Co

rp.

Wa

lth

er

Gm

bH

Do

min

ion

Br

idg

e C

o.

B&

W

Fo

ste

r-

Whe

ele

r C

or

p.

Do

min

ion

Br

idg

e C

o.

Do

min

ion

Bri

dg

e C

o.

w

0

CO

Le-

cAAion

O

pe

ra

tor

NEW'B

ol

le

t:s

:

Akro

n O

hio

C

ity

of

Ak

ro

n

Ea

s�

Ham

ilto

n C

ity

of

Ont· •

. � S

WA

RV

H

amil

ton

Hempfl. t

ead;-

He

mp

ste

ad

N.Y

. R

eso

ur

ces

R

ec

ov

er

y

&is

ting

Bo

iler

:

Ro.ch

este

r,

Ko

da

\{

N..Y

.

TAB

LE

2

STEA

M G

ENE

RA

TO

RS F

IRIN

G P

ROC

ESS

r':D I,

nt"

AS

A P

RJM

A1�Y

FU

EL

Sta

tus

und

er

co

nstr

uct

ion

in o

pe

rat

ion

sin

ce

19

72

und

er

Re

fus

e C

ap

acit

y,

sh

or

t to

ns

/d

ay

(M

etr

ic t

on

s/

da

y)

10

00

(90

7)

60

0

(544

)

20

00

co

nstr

uct

ion

( 18

14

)

ope

ra

tin

g 1

80

(16

3)

Bo

ile

r D

es

i gn

Da

ta

1b/h

r x

10

00

'

(kg

/hr

x 1

00

0)

3 u

nits

, ea

ch 1

26

(57

)

2

unit

s,

ea

ch 1

06

(48

. 1

)

40

0

( 18

1)

77

(34

.9

)

u

p;;ia

, fO

(K

g/c

m,2

C

)

56

0,

47

9

25

0,

sa

t.

(17

.5

, s

at)

Su

ppl

em

en

tal

Fue

l

#2

oil

oil

oil

.

#6

oil

Co

rnrne

nts

den

sit

y s

ep

arat

io:-:

inc

lud

ed

in

fue

l pr

oc

es

s in

g

No

ste

arn

ma

rk

et

Ste

am

so

ld t

o L

on

! Is

lan

d L

ight

ing

C"

for

ele

ctr

icit

y p

ro

duc

tio

n.

Als

o b

\.!rns

dr

ied

wa

ste

wa

ter

slU:

dge

Ko

da

k u

se

s s

tea

�

inte

rn

a ll

y.

w

o

CO

TA

BL

E

ST

EA

M G

EN

ER

AT

OR

S F

IRIN

G :;

HR

ED

Dt�

D,

CA

LSS

lFIE

D S

UP

PL

EM

EN

TA

L R

DF

Lo

cati

on

, F

ac

ilit

y

Am

es

, Io

wa

U

nit

s 5

, 6

,

&7

Op

er

ato

r

Am

es

Mu

nic

ipa

l E

lect

ric

Se

rv

ice

Er

idg

ep

or

t,

Un

it"

d

Co

nn.,

B

rid

ge

-Il

lum

ina

tin

g

po

rt

Ha

rb

or

S

tati

on

Un

its

HI a

nd #

2

Un

ite

d

St.

L

ou

is,

MO

U

nio

n E

lect

ric

M

eram

ec P

lan

t

Wis

con

sin

Oa

k

WE

PC

O

Cr

ee

k P

ow

er

S

tati

on

Bo

ile

r D

es

ign

Da

ta

lb/h

r x

1

00

0

ps-ia

, 0 F

R

DF

Pr

odu

ced

B

y:

2

0 (K

g/hr

x

100

0)

(Kg/

en:',

C)

Cit

y o

f A

me

s S

oli

d W

ast

e

Re

cove

ry

Sy

st.

Re

sour

ce

Re

cov

er

y

Ass

oci

ate

s

St.

L

ou

is D

em

on

­s

tra

tio

n P

lan

t

Am

er

ico

log

y,

Mil

wa

uke

e R

eso

ur

ce

Re

cove

ry

Pla

nt

#7

36

0

( 16

3)

#6

12

5

(57

)

#5

9

5

(43

)

90

0,

90

0

(63

,4

82

)

72

5,

82

5

(51

,4

41

)

71

0,

82

5

(50

, 4

41

)

2 u

nit

s,

ea

ch 9

25 -

-­

(42

0)

Fir

ing

Sy

ste

m

Ta

nge

nti

al

tra

vell

ing

g

ra

te

tra

vell

ing

g

ra

te

cy

clo

ne

.b

ur

ne

rs

tan

gen

tia

l s

ys

pe

ns

io

n

sus

pe

nsi

on

Sup

ple

me

nta

l R

ate

8%

-

10%

10

%

-5

0%

10%

-

50

%

30

%

-4

0%

9%

-

27

%

100/

.

Pr

ima

ry

F

uel

S"ta

tus

coa

l, g

as

op

er

atin

g r

egu

la rl

y w

ith

coa

l,

ga

s R

DF

sin

ce

No

v •

.

197

5

coa

l,

ga�

coa

l,

oil

R

DF

pla

"t

un

de

r de

sig

n

coa

l,

ga.

s

de

mo

ns

tra

tio

:1.

coa

l

pla

nt

:10

lo

n,;

:.:: �

op

er

a r

ing

re

gu

la d

y

RD

F p

lant

un

der

co

nst

r:;.

H

an

tv

a

TA

BL

E 4

ST

EA

M

GE

NE

RA

TO

RS

TE

ST

ED

W

ITI!

TH

E S

UP

PL

F:M

F.N

TA

L

FIR

IN

G O

F

d-R:

DF

Lo

cati

on

F

ac

ilit

y

Piq

ua,

Oh

io

Boi

ler

No

.4

Wri

ght

Pa

tte

,so

n

AF

B

Boi

ler

Pla

nt

Oo

e,

at

ion

Piq

ua E

lect

ric

U

tili

ties

Pow

er

P

lan

t

U.S

. A

ir

Fo

rce

Osh

ko

sh,

Wis

e.

Un

ive

rsit

y o

f U

niv

er

sit

y o

f W

isco

ns

in

Wis

co

nsi

n,

Osh

ko

sh C

am

pus

B

oil

er

Pla

nt

Da

te a

nd

Du

rat

ion

of

Te

st

Ju

ne

27,

19

75

7 h

our

s

Au

g.

197

5 3

0 h

our

s

Aug

. 1

97

6

8 h

our

s

"Te

ste

d i

or a

bo

ut o

ne

ho

ur

at 1

00

0/. d

-RD

F

Bo

ile

r D

es

ign

Da

ta

Fir

ing

lb/h

r x

10

00

p

sia

, \IF

B

oile

r M

etho

d

(Kg

/hr

x 1

000)

(K

g/c

m2,

DC

) M

fg.

Ch

ain

1

50

4

54,

750

C

E

Gr

ate

(6

8)

(31

.9,3

99

) S

tok

er

Tr

ave

llin

g

2 u

nit

s,

ea

ch

12

5, s

at.

E

dg

e M

oor

G

ra

te

80

(3

6.3

) (8

.8 s

at

. )

Iron

Wo

rks

S

pr

ea

de

r

Sto

ker

Vib

ra

tin

g 50

lI

S, s

at.

W

i.ck

es

�ra

te

(22

.7)

(8.

I,

sat.

)

d-R

DF

P

rim

ar

y

Sup

ple

men

t,,1

ma

de

by

Fu

el

Ra

te (

Btu

bas

is)

Bla

ck C

law

-c

oa

l 2

0 -

24

0/.

so

n (

#2 f

uel,

3

/8"

dia

. )

Bl

ac

k C

la

w1:

oal

so

n

({12

f

ue

l,

23

-2

7%

3/

8"

di

a.

)

Gr

uma

n co

al

20 -

24'r.

" (5

/8"

dia

.. )

..,

G)

c

:u

rT\ w

RES

OUR

CE

REC

OV

ERY

TEC

HNO

LOG

Y

,.---S

HR

ED

DI

NG

,..-- M

AG

NE

TI

C

SE

PA

RA

TIO

N

(Fe

)

MA

TE

RIA

LS

S

EP

AR

AT

ION

�

AIR

D

EN

SI

TY

S

EP

AR

AT

ION

--- S

CR

EE

NIN

G

"'--- N

ON

-F

e

SE

PA

RA

TIO

N

'-- W

ET

P

RO

CE

SS

ES

DIR

EC

T

CO

MB

US

TIO

N

CH

EM

ICA

L

CO

NV

ER

SIO

N

( P

YR

OL

YS

IS

� B

IOC

ON

VE

RS

ION

PR

IMA

RY

F

UE

L

SU

PP

LIM

EN

TA

L

FU

EL

W

r-J

DIR

EC

T

CO

M B

US

TIO

N

UN

-P

RO

CE

SS

ED

R

EF

US

E

SP

EC

IA

LL

Y

SH

RE

DD

ED

F

UE

L

DE

SI

GN

ED

�

ST

EA

M

RE

FU

SE

FU

EL

B

OIL

ER

AS

H

� PRI

MA

RY

r;

�O-;:;-D

�71

M

UN

IC

IPA

L

SI

ZE

S

IZ

E

I �I

DE

NS

IT

Y

, -

'DR

YI

NG

' -'D

EN

SIF

ICA

TIO

NI d

-R

DF

RE

FU

SE

R

ED

UC

TIO

I R

ED

UC

TI

ON

�S

EP

AR

AT

IO

N�

�

."

CO)

C

:0

ITI

N

L-_

__

:J

FE

RR

OU

S

r-

--

' I

I -

OP

TIO

NA

L

L _

__

J

1, H

EA

VI

ES

SH

RE

DD

ED

C

LA

SS

IFIE

D

FU

EL

EX

IS

TI

NG

BO

IL

ER

AS

H

ST

EA

M

RE

FU

SE

AS

A

PR

IM

AR

Y

FU

EL

RE

FU

SE

AS

A

SU

PP

L.E

ME

NT

A L

Fu

EL

w

.....

w ."

Cl

C

::u

rrI

�

iA

�CR

ANE

FRAG

MEN

TIZE.R

MOV

ING

GRA

TE S

YST

EM

SO

UR

CE

:W

HE

EL

AB

OR

ATO

R-

FR

YE

PR

OF

ILE

O

F

TO

S\TACK

WH

EE

LA

BO

RA

TO

R -

FR

YE

B

OIL

ER

P

LA

NT

S

AU

GU

S,

MA

SS

.

w

.....

�

"Tl

G)

C

::0

m

�

1.0.

FAN

BANK

EL

ECTR

OSTA

TIC

PREC

IPITA

TOR

AIR

�TE

R-WA

LLEO

FU

RNAC

E

TYP

ICA

L P

RO

CES

SE

D R

DF

PR

IMA

RY

FUE

L S

TE

AM

GE

NER

AT

OR