Gas chromatographic-mass spectrometric analysis of tar ...

Journal of Chromatography, 562 (1991) 531-545 Biomedical Applications Elsevier Science Publishers B.V., Amsterdam

CHROMBIO. 5557

Gas chromatographic-mass spectrometric analysis of tar compounds formed during pyrolysis of rice husks

V. A. C. HAANAPPEL

Biomass Technology Group, University of Twente, P.O. Box 217, 7500 AE Enschede (The Netherlands')

and

T. W. STEVENS, A. HOVESTAD, V. SKOLNIK and R. VISSER*

Laboratory for Chemical Analysis, University of Twente, P.O. Box 217, 7500 AE Enschede (The Nether- lands)

ABSTRACT

Pyrolysis of agricultural waste to produce fuel gas involves formation of tars as noxious by-products. In this paper the qualitative analysis of tars formed during pyrolysis of rice husks is presented, based on identification by gas chromatography-mass spectrometry and interpolation of retention times on a polyaromatic hydrocarbon index scale. The influence of some reaction parameters on product formation is briefly discussed.

INTRODUCTION

Biomass can be converted by gasification into a gaseous fuel (producer gas) [1]. The fuel can be used to operate engines, e.g. for the production of electricity. This approach is very useful for third world countries, especially when small production units are used. However, depending on the type of feed and operating conditions, the process can give rise to significant amounts of unpleasant tars. These can contain mutagenic or carcinogenic organic compounds and cause severe pol- lution of the environment. Reduction of tars can be achieved by adding inorganic additives, or the tars can be catalytically or thermally cracked whilst still in the vapour phase in the raw producer gas.

For the optimization of the process a qualitative and semi-quantitative analysis of the tars is a pre-requisite. It is then possible to examine the amounts formed of different compounds (especially those that are very harmful) under varying conditions.

In the past, formation of tar compounds by gasification of wood has been studied [2,3]; however, there is a tendency to shift to the use of agricultural waste as the feedstock (e.g. in Mali, Ecuador and South East Asia). This paper describes the analysis of compounds formed by pyrolysis of rice husks in laboratory experiments, with the emphasis on the qualitative aspects.

0378-4347/91/$03.50 © 1991 Elsevier Science Publishers B.V.

532 V.A.C. HAANAPPEL e t al.

EXPERIMENTAL

Pyrolysis The experiments described are based on pyrolysis of dried rice husks (normally

ca. 6 g) in a stream of helium gas at 500°C in a Gray-King-type quartz retort. In the first series of experiments husks were used untreated (experiment 1), prepared by dry m!xing with 10.7% (w/w) K2CO3 (2), 10.5% Na2Co3 (3), 9.0% CaCO3 (4) and 1.6% KC1 (5), and prepared by impregnating with 5.6% (w/w, dry) K2CO3 (6), 3.7% Na2CO3 (7), 11.6% KC1 (8) and 10.5% NaC1 (9) solutions in water. The gases produced were led into a U-shaped cold-trap (0°C) provided with glass-wool plugs at both ends.

In the second series the retort was linked with a second reactor; in this reactor an additional flow of helium was used to vary the residence time of the volatiles. The reactor was connected to the cold-trap. Pyrolysis experiments were carried out with an empty reactor at 600°C/2.4 s residence time (10), 700°C/2.2 s (11) and 0.6 s (12), 900°C/1.8 s (13) and 0.5 s (14), and 100°C/1.7 s (15). Finally the reactor was filled with dolomite as a cracking catalyst (600°C/2 s residence time) (16).

The tars in the trap were extracted with methanol-acetone, the eluents were evaporated, and the "dry" weight was determined. Tars were dissolved in diethyl- ether and analysed by gas chromatography (GC). A sample resulting from experiment 9 was used for qualitative analysis by gas chromatography-mass spectrometry (GC-MS), and a sample from experiment 14 for a separate qualitative analysis based on polyaromatic hydrocarbon retention times.

Gas chromatography-mass spectrometry GC-MS analysis was performed using a Finnigan MAT 90 double-focusing

mass spectrometer (electron impact mode) coupled to a Varian 3400 gas chromatograph.

GC conditions were: column, fused-silica SGE (Scientific Glass Engineering, Ringwood, Australia) BP5 (25 m × 0.22 mm I.D.; film thickness 0.25 #m); temperature, l min at 40°C then 7°C/min to 250°C then 1 min at 250°C; carrier gas, helium at 1.5 ml/min.

Mass spectrometer parameters were: ionization voltage 70 eV; emission current, 1 mA; mass scan, from 40 to 400 dalton; resolution, 1430 (10% valley definition); sampling frequency, 26 500 kHz; cycle time, 1.2 s; samples per peak, 8. About 1250 mass spectra were recorded.

GC analysis was performed on a Varian 3400 gas chromatograph. GC conditions were: column, fused-silica SE 54 (J&W Scientific, Cordova, CA, U.S.A.) (15 m × 0.31 mm I.D.; film thickness 0.25 pro); temperature, 1 min at 40°C then 5°/min to 290°C then 10 min at 290°C; carrier gas, helium at 3 ml/min; injection, on column, 50°C at 100°C/rain to 300°C; flame ionization detection at 300°C, connected to data system Varian Vista CDS 401.

The mass spectra were identified by an automatic library search (National

GC-MS OF TAR COMPOUNDS 533

Bureau of Standards Library, 55 000 spectra). In those cases where no significant answer was obtained, the spectra were compared with the Eight Peak Index of Mass Spectra [4] or interpreted by the authors.

GC retention times were converted into retention indices by linear interpolation between retention times of a series of n-alkanes [5]. A series of compounds were identified independently by injection of a reference sample. For compounds identified by MS and entered in the Sadtler Retention Index Library [6] the retention indices were compared to the reference values. Polyaromatic hydrocar- bons (PAHs) were verified by injection of a series of nine reference PAHs (i.e. phenol, 1-naphthol, 2-naphthol, naphthalene, phenanthrene, anthracene, fluor- anthene, pyrene and chrysene) and interpolation of the other PAHs using the PAH retention index scale [7].

The total weight of material removed from the cold-trap was determined after evaporation of the solvents. The total amount of material (tar) dissolved in diethyl ether was determined by GC, using an average calibration factor to convert peak areas into masses. For two characteristic groups of compounds, viz. the 22 most intense hydroxyaromatics and the 20 most intense PAHs, the sum of the peak areas was calculated and converted into absolute amounts, again using an average calibration factor. The same was done for one of the most suspected compounds, i.e. benz[a]pyrene.

RESULTS AND DISCUSSION

The reconstructed ion current (RIC) chromatogram of experiment 9 as a function of the mass spectrum number is shown in Fig. 1. The results of the qualitative analysis based on the mass spectra are summarized in Table I. The numbers in the column Sp.No. refer to the mass spectrum numbers. The GC retention times are those obtained from the independent gas chromatogram. This chromatogram is shown in Fig. 2. Since concentrations of higher-boiling compounds were very low, comparison of peaks with retention times above ca. 19 min was not unambiguous; these times have been left out.

The results of the comparison of the separate run of experiment 14 with the set of PAH reference values are presented in Table II. This table contains almost exclusively PAHs. The chromatogram is given in Fig. 3. (Retention times in Table II are the average of two determinations, which explains the slight differences between Table II and Fig. 3). Compounds identified by independent injection of authentic standards are marked with an asterisk in both tables.

In Table III the total mass of material recovered for each experiment is presented, and the absolute amount of tar dissolved in diethyl ether. In the following two columns the total masses of the 22 most important compounds of Table I (mainly aromatic hydroxyl compounds) and of the 20 most important compounds (almost exclusively PAHs) of Table II are given, and in the last column the mass of benz[a]pyrene.

534 V . A . C . H A A N A P P E L et al.

CHRO: H57967 uer 1 on u l c ~ 0 2 0 B 2

R I C 4 9 7 0 1 9 6

1 : 0 0 i : 3 6 4 : 0 1 ×

13 -DEC*S9 E l a p s e : 0 0 : 0 0 : 1 0 . 2 3 9 9

B 4 3 4 5

7 : 0 4 ~E+08 [ . 4 0 5

11 1 , i

1 4 0 320 36 2 :52 18 6 :33 , .

t o 0 ' 3 : i . ,

~ , . . . .

. . . . 3~ . . . . ' " i ' " i . . . . . . ' ~ ; ,

CHRO: M S 7 9 6 7 v e t I on UIC ~02002 .

9 : 1 9 l a O q

I 5 6 5

8 0 ~ 1 1 : 3 .

5 4 2 11106 591

1 4 : 5 4

4o-I z76 22 4 9 0 679 1 5 : 5 3

: 3 6 1a :82 6 4 0 ' i 4 7 2 1 3 : 0 1 3 : 1 5 4

20-1 9 : 4 0 ' ' 1 ~ 08 ' : 21

¢ e o ~-5G 5 0 f l 550 6 0 0 6 5 0 7 0 0 7 5 g 0 0 0

1 3 - 0 E C - 8 9 E l a p s e : G 0 : 0 0 : 0 1 , 6 1

w E + e e i , 0 2 5

653 1 3 : 2 2 I

CHRO: M$7967 v e t I on UIC ~ 0 2 0 0 2

RZC

× 1 o o

Bg4

8@ 1 8 : 1 8

8 5 5 9 9 1 1 0 6 3 4 0 1 7 . , 3 0 , , 2 0 : 1 1 7 2 1 : , 4 6

, ' s : , . z z ; 2 0

e l . . . . i . . . . , . . . . i . . . . i . . . . , . . . . i . . . . i . . . . , . , B 0 0 B50 9 0 0 9 5 ~ 1 0 0 0 1 0 5 g 1 1 0 0 1 1 5 0 t 1 0 0 ' 1 ' 2 5 0

13-0EC-89 E l a p s e : 8 @ : 0 @ : @ 1 , 6 1

~E÷BB

Fig. 1. Reconstructed ion current chromatogram as a function of mass spectrum numbers of pyrolysis

products formed after impregnation with NaCI (experiment 9).

TA

BL

E I

CO

MP

OU

ND

S I

DE

NT

IFIE

D B

Y M

AS

S S

PE

CT

RA

(G

C-M

S)

No.

t R

Sp

.No.

N

ame

For

mul

a M

W

(min

)

1 49

A

ceto

ne

C3H

60

58*

2 53

A

ceti

c ac

id m

ethy

l es

ter

C3H

602

74*

3 68

A

ceti

c ac

id

C2H

402

60*

4 73

A

ceti

c ac

id

C2H

402

60*

5 0.

908

77

3-M

etho

xy-3

-met

hyl-

2-bu

tano

ne

C6H

1202

11

6

6 1.

057

85

2-M

ethy

lhex

anoi

c ac

id

CvH

1402

13

0 2,

3-E

poxy

prop

anol

C

3H60

2 74

7

1.33

5 99

H

ydro

xyac

etic

aci

d m

ethy

l es

ter

C3H

603

90

8 1.

471

106

Pro

pano

ic a

cid

C~H

602

74*

9 2.

053

136

2-T

etra

hydr

ofur

anol

C

4H80

2 88

10

2.13

6 14

0 2-

Met

hylp

ropa

noic

aci

d C

4H80

2

88

11

2.46

1 15

7 3-

Hex

en-2

-one

C

6Hlo

O

98

2,5-

Dih

ydro

fura

n C

4H60

70

12

2.

673

187

2,5-

Dim

ethy

lfur

an

C6H

sO

96

13

2.72

9 19

5 4-

Hyd

roxy

-4-m

ethy

l-2-

pent

anon

e C

6H12

02

116

14

2.89

3 20

9 l-

(2-F

uryl

)-2-

met

hyl-

1,2

-but

aned

iol

C9H

1403

17

0 15

3.

028

222

2-P

ropo

xybu

tane

C

7H16

0 11

6 1 -

(1 -M

ethy

leth

oxy)

-2-p

ropa

none

C

6H 1

202

116

16

3.27

2 23

0 T

etra

hydr

o-2,

5-di

met

hoxy

fura

n C

6H12

0 3

132

17

3.45

5 24

4 T

etra

hydr

o-2,

5-di

met

hoxy

fura

n C6

H12

03

132

18

3.55

4 26

7 2-

Met

hyl-

2-bu

tena

l C

sHsO

84

19

3.

639

271

2-(3

H)-

Fur

anon

e C

4H40

2

84

20

4.48

9 31

5 5-

Met

hyl-

2- fu

ranc

arbo

xald

ehyd

e C

6H60

2 11

0 21

4.

639

319

3-M

ethy

l-2-

cycl

open

ten-

1 -on

e C

6H80

96

2,

4-D

imet

hylf

uran

C

6H 8

0 96

(Con

tinu

ed o

n p.

536

)

I g.

©

,-.]

> ©

©

Z

TA

BL

E I

(co

ntin

ued)

No.

t R

Sp

.No.

N

ame

For

mul

a M

W

(min

)

22

5.44

7 34

4 P

heno

l C

6H60

94

23

5.

882

359

2-M

etho

xy-6

-met

hyl-

4 //-

i~yr

an-4

-one

C

7H80

3 14

0 24

5.

882

359

4-M

etho

xy-6

-met

hyl-

2 t/

pyra

n-2-

one

C7H

80 3

14

0 25

6.

150

364

Tet

rahy

dro-

2-m

ethy

l-2-

fura

nol

C9H

1 s O

14

2 26

6.

325

391

2-H

ydro

xy-3

-met

hyl-

2-cy

clop

ente

n- l-

one

C6H

sO z

112

27

6.32

5 39

1 3-

Met

hyl-

1,2

,-cy

clop

enta

nedi

one

C6H

80 2

11

2 28

6.

558

339

Spir

o[2,

4]he

ptan

-4-o

ne

C~H

loO

11

0 29

6.

558

399

1,3,

5-T

rim

ethy

l- 1

H-p

yraz

ole

C6H

loN

2 ll

0

30

402

Unk

now

n 3

l 6.

926

4I I

3-M

ethy

l-2-

(5H

)- fu

rano

ne

C5H

602

98

4-H

exen

-3-o

ne

C6H

1 ~O

99

32

7.

151

421

2-M

ethy

lphe

nol

CvH

sO

108"

33

7.

644

445

3-M

ethy

lphe

nol

CT

HsO

10

8"

4-M

ethy

lphe

nol

CT

HsO

10

8"

34

7.93

4 45

4 2-

Pro

pylo

xeta

ne

C6H

120

100

35

7.93

4 45

4 3,

4,5-

Tr~

met

hyl-

2-cy

c[op

ente

n-

1 -on

e C

aH~2

0 12

4 36

7.

934

454

3,4,

4-T

rim

ethy

l-2-

cycl

open

ten-

1 -on

e C

sHlz

O

124

37

7.93

4 45

4 3,

5,5-

Tri

met

hyl-

2-cy

clop

ente

n- 1-

one

CsH

a20

124

38

8.46

7 48

9 2-

Met

hyl-

3-hy

drox

y-4-

pyra

none

C

6H60

3 12

6 39

8.

794

504

Pen

tane

dioi

c ac

id d

imet

hyl

este

r C

7H12

0 4

160

40

8.79

4 50

4 P

enta

nedi

oic

acid

eth

yl m

ethy

l es

ter

CsH

~,O

, 17

4 41

9.

062

510

2-E

thyl

phen

ol

CsH

xoO

12

2"

42

9.45

7 52

0 2,

4-D

imet

hylp

heno

l C

sHao

O

122"

43

9.

765

536

4-H

ydro

xybe

nzen

eeth

anol

C

s/-I

loO

2 13

8 44

10

.018

.

542

3-E

thyl

phen

ol

CsH

IoO

12

2"

45

10.1

12

549

1,4-

Dim

etho

xybe

nzen

e C

sH~o

O z

138

46

10.1

52

552

2,3-

Dim

ethy

lphe

nol

CsH

toO

12

2 47

10

.219

55

7 2-

Met

hoxy

-4-m

ethy

lphe

nol

C8H

loO

2

138

48

10,3

26

565

4,5-

Dim

ethy

l- 1

,3-b

enze

nedi

ol

CsH

aoO

z 13

8

.<

.>

1>

;>

Z

g~

49

5O

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

8I

10.6

59

10.9

05

10.9

05

11.1

80

11.3

20

11.4

60

11.6

19

11.7

80

11.9

82

11.9

82

11.9

99

12.0

72

12.2

28

12.5

54

12.7

07

12.8

73

13.1

21

13.2

25

13.4

69

13.7

20

14.3

12

14.4

65

14.5

69

14.9

06

15.3

29

15.6

08

15.7

67

15.7

67

16.7

11

17.1

31

17.6

72

17.9

95

18.2

II

568

571

571

575

591

599

6O4

6O8

613

613

619

640

646

652

658

662

668

671

678

688

727

732

736

749

764

775

778

778

817

854

87O

87

9 88

5

3,4-

Dir

neth

ylph

enol

2,5-

Dih

ydro

-2,5

-dim

etho

xy-2

-fur

yl

met

hyl

keto

ne

2,6-

Dim

ethy

l-4-

prop

yl-4

-hep

tano

l 1,

3-B

enze

nedi

ol

1,2-

Ben

zene

diol

2,3-

Dih

ydro

benz

ofur

an

p-Is

opro

pylp

heno

l 2-

Met

hylp

ropa

noic

aci

d is

open

tyl

este

r p-

Isop

ropy

lphe

nol

5-M

etho

xy-2

,3-d

imet

hylp

heno

l

5-(H

ydro

xym

ethy

l)-2

- fu

ranc

arbo

xald

ehyd

e 3-

Met

hoxy

-1,2

-ben

zene

diol

3-

Met

hyl-

1,2-

benz

ened

iol

4-E

thyl

-2-m

etho

xyph

enol

1-

Inda

none

3-Is

opro

pylp

enta

nedi

one

2-M

ethy

lnap

htha

lene

U

nkno

wn

2-M

ethy

l- 1,

4-be

nzen

edio

l 2-

(l,l

-Dim

ethy

leth

yl)p

heno

l

3,4-

Dim

etho

xyph

enol

5-

Met

hylq

uino

line

6-

Met

hylq

uino

line

3-

Hyd

roxy

benz

alde

hyde

4-

Eth

yl- 1

,3 -b

enze

nedi

ol

4-H

ydro

xy-3

-met

hoxy

benz

alde

hyde

2-

Met

hoxy

-6-(

1-pr

open

yl)-

phen

ol

2,3,

5,6-

Tet

ram

ethy

l-2,

5-cy

cloh

exad

iene

- 1,

4-di

one

2-M

etho

xy-4

-(l-

prop

enyl

)phe

nol

4-H

ydro

xy-3

-met

hoxy

-ace

toph

enon

e 2,

6-B

is(l

,l-di

met

hyle

thyl

)-4-

met

hylp

heno

l 4-

Hyd

roxy

-3-m

etho

xybe

nzoi

c ac

id m

ethy

l es

ter

2,2'

-Dim

ethy

l- 1,

1 '-b

iphe

nyl

2,3'

-Dim

ethy

l- I,

1 '-b

iphe

nyl

CsH

loO

C8H

1204

C

12H

260

C6H

602

C6H

602

CsH

sO

C9H

120

C9H

1802

C

9H12

0 C

9H12

02

C6H

603

CT

HsO

3 C

7H80

2

C9H

12O

2

C9H

80

C8H

1402

C

11H

lo

C7H

802

C10

H14

0

CaH

loO

3 C

loH

9N

Clo

H9N

C

7H60

2 C

8H10

02

CsH

sO3

Clo

H12

02

Clo

H12

02

Clo

H12

02

C9H

loO

3 C

15H

240

C9H

loO

4 C

14H

14

C14

H14

122

172

186

110

110

120

136"

15

8 13

6 15

2

126

140

124

152

132

142

142"

124

150"

154

143"

14

3 12

2 13

8

152

164

164

164

166

220

182

182

182

9 ©

> 0 0 Z

(Con

tinu

ed o

n p.

538

)

TA

BL

E 1

(con

tinue

d)

No.

t,

Sp

.No.

N

ame

For

mul

a M

W

(rai

n)

82

18.4

72

893

83

18.4

72

893

84

18.7

01

900

85

927

86

930

87

936

88

963

89

963

90

991

91

991

92

1003

93

10

16

94

1062

95

10

74

96

1090

97

1090

98

11

30

99

1130

100

1193

101

1220

Prop

yl 2

-pro

pylc

yclo

pent

en-l

-eny

l ket

one

l-(4

-Hyd

roxy

-3-m

etho

xyph

enyl

)-2-

prop

anon

e M

ethy

l-cc

-o-g

luco

pyra

nosi

de

Met

hyl-

fl-D

-glu

copy

rano

side

2,

3-D

ihyd

ro-6

-met

hyl-

4H-b

enzo

pyra

n-4-

one

4-H

ydro

xy-3

-met

hoxy

benz

oic

acid

A

ceti

c ac

id e

thyl

-p-h

ydro

xyph

enyl

est

er

Prop

anoi

c ac

id m

ethy

l-p-

hydr

oxyp

heny

l es

ter

c~-P

heny

lben

zene

acet

alde

hyde

~-

Phen

ylbe

nzen

eace

tic

acid

2,4-

Dih

ydro

xy-3

,6-d

imet

hylb

enzo

ic a

cid

met

hyl

este

r 4-

Hyd

roxy

-3-m

etho

xybe

nzen

eace

tic

acid

4-

Hyd

roxy

-3,5

-dim

etho

xybe

nzal

dehy

de

2,4-

Dim

etho

xy-6

-met

hylb

enzo

ic a

cid

met

hyl

este

r

1-(4

-Hyd

roxy

-3,5

-dim

etho

xyph

enyl

)eth

anon

e 3-

(4-h

ydro

xyph

enyl

)-2-

prop

enoi

c ac

id m

ethy

l es

ter

N-A

min

ocar

bony

l-4-

hydr

oxy-

3-m

etho

xy

benz

enea

ceta

mid

e (2

,4,6

-Tri

hydr

oxy-

3-m

ethy

lphe

nyl)

- 1-

buta

none

1-

Met

hoxy

-3-p

heno

xybe

nzen

e 2,

2'-D

ieth

yl- 1

,1 '-

biph

enyl

14-M

ethy

lpen

tade

cano

ic

acid

met

hyl

este

r 13

-Met

hylp

enta

deca

noic

aci

d m

ethy

l es

ter

Hex

adec

anoi

c ac

id

C12

H2o

O

180

Clo

H12

03

180

C7H

1406

19

4 C

7H14

06

194

Clo

Hlo

O 2

162

CsH

80,,

168

Clo

H12

03

180

CLO

H12

03

180

c14H

120

196

C14

H12

02

212

Clo

H12

04

196

C9H

loO

4 18

2 C

9Hlo

O 4

182

Cxl

Hlg

O 4

210

Clo

Hlz

O4

196

Clo

Hlo

O 3

178

Clo

H12

04

224

Cll

H14

04

210

C13

H12

02

200

Ct6

Hla

21

0

C17

H34

02

270

C 1

7H34

02

270

C16

H32

02

256

< (3

e~

~t

L~

:i ii

~1 .11

ii Iii

N [lii ~

I

fi~

iflllltN

I [/~l-

~ lt~

II li

Fig.

2. G

as c

hrom

atog

ram

of

pyro

lysi

s pro

duct

s (e

xper

imen

t 9).

? ©

-]

O

©

< Z

TA

BL

E I

I

CO

MP

OU

ND

S I

DE

NT

IFIE

D

BY

RE

TE

NT

ION

TIM

ES

No.

t R

N

ame

For

mul

a M

W

(rai

n)

1 1.

658

o-X

ylen

e C

sH to

10

6 2

2.48

1 p-

Xyl

ene

CsH

1 o

106

3 3.

100

Sty

rene

C

sH s

104

4 5.

026

Ben

zofu

ran

C8H

60

118

5 5.

448

Phe

nol

C6H

60

94*

6 6.

366

Ace

tic

acid

phe

nyl

este

r C

sH8

02

13

6

7 6.

917

m/p

-Cre

sol

C7

H8

0

108"

8

7.71

0 D

ihyd

rona

phta

lene

C

loH

lo

130

9 9.

856

Nap

htha

lene

C

loH

s 12

8"

10

12.2

42

Tol

uald

ehyd

e C

sH8

0

120

11

12,6

77

2-M

ethy

lnap

htal

ene

C 1

1H a o

14

2*

12

13.0

85

1-M

ethy

lnap

htha

lene

C

llH

lo

142

13

14.8

37

Bip

heny

l C

12H

lo

154

14

15,4

08

2,7-

Dim

ethy

lnap

htha

lene

C

12H

~z

156

15

15.5

71

1,7-

Dim

ethy

lnap

htha

lene

C

12 H

a 2

156

16

15.9

78

Un

kn

ow

n

C12

H s

152

17

16.3

85

Un

kn

ow

n

C12

H1o

15

4 l 8

17

.368

A

cena

phth

ylen

e C

t 2H

s 15

2

19

17.5

46

Ace

naph

then

e C

l 2 H

~ o

154

20

17,9

40

Dib

enzo

fura

n C

12H

sO

168

21

18.1

65

1-N

apht

hol

Clo

H8

0

144"

22

18.6

10

Met

hylb

iphe

nyl

C~

sHaz

16

8 23

19

.014

1-

Met

hyla

cena

phty

lene

C

13 H

lo

166

24

19,4

28

Flu

oren

e C

~ 3 H

~ o

166*

25

19

.862

M

ethy

lace

naph

thyl

ene

C~

3H 1

o 16

6

26

20.3

20

M e

thyl

acen

apht

h yl

ene

C ~

3 H 1

o 16

6

27

20.5

83

Xan

then

e C

a3H

lo O

18

2 28

21

.920

D

ihyd

roph

enan

thre

ne

C 1

4 H 1

2 18

0

Z

;>

t'-

2

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

22.1

30

22.2

74

22.9

09

23.5

13

23.6

98

25.2

26

25.8

87

26.1

42

26.5

97

27.3

10

28.3

40

28.7

72

29.1

99

29.6

42

30.4

22

31.4

10

31.7

23

32.1

21

32.2

44

32.4

64

34.3

00

34.6

29

35.0

07

35.1

83

35.3

35

35.8

79

37.5

92

39.7

89

40.1

50

40.7

45

40.8

86

44.9

08

45.6

41

46.0

69

Unk

now

n C

I,H

lo

Met

hylf

luor

ene

C14

H1

z

Met

hylf

luor

ene

C14

Hlz

P

hena

nthr

ene

C~4

Hao

A

nthr

acen

e C

a4H

~o

1-P

heny

lnap

htha

lene

C

16H

12

Met

hylp

hena

nthr

ene

Ca s

H 1

z

4H-C

yclo

pent

a (d

eJ)p

hena

nthr

ene

C~

sH~

o M

ethy

lphe

nant

hren

e C

lsH

a2

2-P

heny

lnap

htha

lene

C

16H

a 2

Dih

ydro

pyre

ne

C16

H12

F

luor

anth

ene

Ca o

H~o

Ace

phen

anth

rene

C

16H

10

Pyr

ene

C16

H10

M

ethy

lflu

oran

then

e C

a 7H

12

Ben

zo[a

]flu

oren

e C

17 H

12

Ben

zo[b

]flu

oren

e C

x 7H

12

Met

hylp

yren

e C

a 7 H

~ 2

Met

hylp

yren

e C

17H

12

Eth

ylm

ethy

l-4H

-cyc

lope

ntap

hena

nthr

ene

Cas

Ha6

B

enz°

[gh0

flu°

rant

hene

C

18H

10

Ben

zo[c

]phe

nant

hren

e C

lsH

12

Cyc

lope

nta[

cd]p

yren

e C

18H

lo

Ben

z[a]

anth

race

ne

C a a

H 12

C

hrys

ene

Ca s

H~2

N

apht

hace

ne

C 18

H 1

2 1-

Phe

nylp

hena

nthr

ene

C2o

H14

Ben

zoL

/+ b

]flu

oran

then

e C2

oH12

B

enzo

[k]f

luor

anth

ene

C20

H12

B

enzo

[b]p

yren

e C

20H

12

Ben

zo[a

]pyr

ene

C2o

H 1

2 D

iben

zant

hrac

ene

C2 2

H 1

4

Pice

ne

C22

H14

D

iben

zoch

ryse

ne

Cz 2

H 1

4

178

180

180

178'

17

8"

204

192

190

192

204

204

202*

202

202*

21

6 21

6 21

6

216

216

232

226

228

226

228

228*

22

8 25

4

252

252

252

252

278

278

278

9 K

o > ~o

0 0 Z U

I~" ~:HA~T SPEED L+O CM/mlN ~TTEN : B ZERO: 10~i 1 IIIN/TICX

L~ | L%.,_~ .............................. ~.~

~c ow

= 2,479

1

o _

o ~-

e-, .-

U+ F

SIB97

6~i+66

8.2S~

i1,+;++

1~" +'3 l 1~ g62 -~I~

t2.6+~ ~-- L 3. ~49 t

~=:=~-,e..e.a- ~7 951

I9.~74

Zt'02:l 21.a24

Z4.+~S

~rm

<" 33 ~4 ~' 33+7t7

~4.3e9 2~

-- ---~- 35. 344

~ ~B:~,~ ~ .3~

~ 1.233

~Z.6t3 4~. 146 4~,S6)

+3 .gB~ 4.486 4.743

+6+B:'E ? 47.t61

......... g, dC~B

t6.3a5

~c ~ .... 26,L53

"1 t~ ~'~ q~tddVNVVH +D "V "A Elrg

G C - M S OF T A R C O M P O U N D S 543

In general, the product composition of tars from rice husks is in good agree- ment with the composition of coal tar as described by Novotny et al. [8], the main products being hydroxyaromatics and PAHs. However, only a small number of nitrogen-containing compounds were detected.

A striking mass spectrum is scan No. 900 (57-60-137-73-121). A probable fit is methyl-~-D-glucopyranoside or its fl-isomer (C7H1406) . This seems to be the first time that a sugar-type compound has been identified.

The interpretation of some mass spectra is not unambiguous. Some examples are: MS scan No. 222 (strongest peaks 43-86-116-42-73) , 2-propoxybutane or 1-(1-methylethoxy)-2-propanone; MS scan No. 319 (96-67-81-53-95), 3-methyl-2-cyclopenten-l-one or 2,4-dimethylfuran; MS scan No. 445 (108-107-77-71- 79); m- or p-cresol or (most probably) a mixture.

The retention indices of about 20 of the compounds mentioned in Table I were retrieved in the Sadtler Retention Index Library [6]. Since this library contains only values for 2°C and 8°C/min temperature increase the mean of these two values was used as a first approximation. In Fig. 4 these calculated R.I. values are plotted against the neat retention times of the independent gas chromatogram, for twelve compounds listed in Table I. In general the relationship is very good, but some very striking results were obtained. Compound 9 (MS scan 136) was

TABLE III

Q U A N T I T A T I V E ANALYSIS OF T A R SAMPLES F R O M PYROLYSIS OF RICE HUSKS

Exp. No. Total mass Tar Hydroxyaromatics Polyaromatics Benz[a]pyrene trapped (mg) (mg) (mg) (mg) (g)

I 1.20 178 68 15 0.23 2 0.84 116 49 9 0.12 3 0.96 188 87 13 0.22 4 1.14 89 42 6 0.09

5 1.20 150 67 12 0.21

6 0.60 157 70 I0 0.33 7 0.72 247 78 15 0.38 8 1.08 161 78 11 0.16 9 1.08 252 121 21 0.21

10 0.66 149 83 10 0.18 11 0.36 73 28 8 0.15 12 0.42 I l l 40 8 0.13 13 0.10 31 4 13 0.32 14 0.12 32 6 14 0.24 15 0.06 50 13 15 0.37

16 0.04 56 26 I1 0.12

544 V . A . C . HAANAPPEL et al.

1500-

R.I

- / 1300-

- 4 2

1100-

g00-

7oo ~ ' $ ' ~ ' 1~ ' lg

--- t'R (min)

Fig. 4. Literature reference values of R.I. plotted as function of neat retention times (experiment 9, gas chromatogram of Fig. 2). The numbers refer to the compound numbers listed in Table I.

originally interpreted as 3-tetrahydrofuranol, reference index 835.40, experimental value c a . 785. A probable explanation is that this compound is 2- tetrahydrofuranol. For this compound no reference index is available but internal hydrogen bonding might very well lower the index by c a . 50 R.I. units. Com- pounds 46 and 49 (MS scans 552 and 568) were originally interpreted as 2,5- dimethylphenol and 3,5-dimethylphenol, reference indices 1149.59 and 1169.05, experimental values c a . 1175 and 1190. After reconsideration of the mass spectra, the compounds 2,3-dimethylphenol and 3,4-dimethylphenol, retention indices 1177.58 and 1193.38, appeared a much more probable fit.

These results indicate that the use of retention indices can yield essential additional information to GC-MS interpretation, provided that a sufficiently compre- hensive and reliable reference collection is available.

Qualitative analysis of tars and corresponding mixtures is often based on chemical class separation prior to the analysis. Parees and Kamzelski [9] have shown that the analysis of coal-derived liquids can be performed in one run without previous separation by using GC-MS with a fused-silica SE 54 GC column. This work supports this approach despite the fact that the interpretation of some mass spectra is slightly hindered by peak overlap.

From Table III it can be concluded that production of tar is reduced by c a .

50% when rice husks are impregnated with potassium carbonate, by c a . 90% on thermal cracking at higher temperatures and by c a . 97% on catalytic cracking. Formation of hydroxyaromatics is hardly influenced by the addition of catalysts, but thermal cracking reduces the amounts by 80-95%, especially at higher temperatures (experiments 13-15). On catalytic cracking a reduction of 70% is obtained. The influence on polyaromatics is somewhat smaller: addition of potassium carbonate or calcium carbonate and thermal cracking yield c a . 50% reduction and catalytic cracking c a . 30%. Finally, benz[a]pyrene is reduced from 0.23 mg to 0.12 mg ( c a . 50%) on catalytic cracking. In experiment 16 the total

GC-MS OF TAR COMPOUNDS 545

amount of material trapped was somewhat lower than the amount of tar determined: this is probably caused by the inaccuracy inherent.in the weighing proce- dure.

REFERENCES

1 A. Kaupp and R. J. Goss, Small Scale Gas Producer-Engine Systems, F. Vieweg, Wiesbaden, 1984. 2 0 . Beaumont and Y. Schwob, Ind. Eng. Chem. Chem. Process Des. Dev., 23 (1984) 637. 3 E.G. Baker, L. K. Mudge and D. H. Mitchell, Ind. Eng. Chem. Chem. Process Des. Dev., 23 (1984) 727. 4 Eight Peak Index of Mass Spectra, The Mass Spectrometry Data Centre, The Royal Society of Chem-

istry, Nottingham, 1983. 5 H. van den Dool and P. D. Kratz, J. Chromatogr., 11 (1963) 463. 6 The Sadtler Standard Gas Chromatography Retention Index Library, Sadtler Research Laboratories,

Philadelphia, PA, 1986. 7 M. L. Lee, D. L. Vassilaros, C. M. White and M. Novotny, Anal. Chem., 51 (1979) 768. 8 M. Novotny, J. W. Strand, S. L. Smith, D. Wiesler and F. J. Schwende, Fuel, 60 (1981) 213. 9 D. M. Parees and A. Z. Kamzelski, J. Chromatogr. Sci., 20 (1982) 441.

Gas chromatographic-mass spectrometric analysis of tar ...

Documents