[Home] [Journal papers]

peer reviewed logo


BLACK LIQUOR COMPONENTS AS POTENTIAL RAW MATERIALS

Klaus Niemelä, Tarja Tamminen, Taina Ohra-aho
Presented at 14th ISWFPC - Durban, South Africa

Key words: Kraft pulping, pine, black liquor, composition, lignin, hemicellulose, by-product, hydroxy acid, tall oil

ABSTRACT
Pine wood chips were cooked under different kraft pulping conditions which represented three different delignification rates: 1) typical, 2), lower rate (lower sulphidity), and 3) higher rate (higher temperature). The effect of the cooking conditions on black liquor composition was monitored during the course of pulping. Representative hemicellulose and lignin fractions were also isolated for deeper characterization.

The cooking conditions had varying effects on the amounts and structure of all main black liquor components, lignin, hemicelluloses and their degradation products, as well as extractives. The highest amount of carbohydrates was found in black liquor after the high-delignification rate cooking; in the bulk delignification stage their amount corresponded to 4.5% of wood; xylan was the dominating polysaccharide. In the initial cooking stages, more substantial amounts of arabinan and galactan were present in the black liquors. The cooking conditions also affected the yield of carbohydrates that could be isolated from black liquors with dioxane and acetic acid – the yield varied from 10 to 93%. The isolated hemicelluloses contained some lignin, but the presence of non-lignin type aromatic constituents was also evident: pyrolysis GC/MS indicated that 10–30% of the phenolic degradation products from the hemicellulose fractions are of non-guaiacyl type.

The cooking conditions had certain effects on the molar mass and sulphur and carbohydrate contents of kraft lignins. Depending on the conditions, the dominating carbohydrate was either galactan or xylan. The sulphur content varied between 0.9 and 2.7, being lowest after the low sulphidity cooking. The formation of acetic acid, formic acid, and various hydroxy acids during cooking was also monitored. Their total amount, more than 10% of wood, was not dramatically affected by different cooking conditions.

1. INTRODUCTION
Currently, the main by-products from kraft pulping include tall oil1,2 and turpentine3,4, and the annual global production figures for these are 1.5 million and 0.1–0.2 million tonnes, respectively. In addition to these by-products, small amounts (less than 0.1 million tonnes) of kraft lignin are also isolated from black liquors for chemical applications5. The carbohydrate-derived fractions6, polysaccharides and aliphatic carboxylic acids, are not currently isolated from black liquor and used as by-products. Their exploitation has been reduced by scattered information on their chemical nature or composition, isolation and purification problems, as well as limited markets.

In this work, we have now isolated hemicelluloses and lignin from pine kraft black liquors (during the course of pulping) for further characterization, and we have also analyzed the main aliphatic carboxylic acids. For deeper information on their formation and factors controlling them, our study was based on three slightly different cooks: normal (reference) cook and other cooks with significantly higher and lower delignification rates. A selection of the main results are now reported and briefly discussed, mainly from an analytical point of view.

2. EXPERIMENTAL
2.1 Cooking experiments
Three sets of kraft cooks were conducted with the chips of Scots pine (Pinus sylvestris) in 1-litre digesters (Table 1). In each case, the impregnation at 130 °C took 60 min. The liquor-to -wood ratios were 4 L/kg during impregnation and 5 L/kg during final cooking.

Code

Cook

Alkali charge*

Sulfidity**, %

Temp., °C

Time, min

R

Reference

3.5 + 1.5

40; 40

160

270

HA

High alkalinity

3.8 + 6.0

48; 0

180

30

LS

Low sulfidity

3.5 + 1.5

11; 0

160

420



Table 2.1. The applied cooking conditions.
*Effective alkali charge (mol/kg in the liquors) in the impregnation and cooking stages.
**Sulphidity of the impregnation liquor and added cooking liquor.
Each of the above cooks was conducted in six autoclaves, withdrawn at different cooking stages (yield of pulp c. 85 45%) for the sampling of black liquors. Appendix 6.1 lists more data on the cooks and black liquor samples, as a function of the H factors.


After cooking, all the liquors were analyzed for total organic carbon (TOC), total inorganic carbon (TIC), lignin by UV, and total carbohydrates by hydrolysis and HPLC7. The kraft lignins were also directly characterized8 (without isolation) for molar mass distribution and phenolic hydroxyl groups. Other analytical procedures are described in Sections 2.2–2.3.

2.2 Isolation and characterization of hemicelluloses and lignin
For deeper characterization, hemicellulose and lignin fractions were isolated from five selected black liquors, representing bulk delignification stages after R (H1000), HA (H350), and LS (H3000) cooks, and stages before (H30) and after (H2500) R bulk delignification. The hemicellulose fractions were first precipitated from 20-mL black liquor samples by adding dioxane and acetic acid9,10, leaving lignin in the liquor phase. Lignin was then precipitated11 by lowering pH to 2.5. The isolated fractions were analyzed for the content and composition of carbohydrates7, and for the aromatic structural units by pyrolysis GC/MS12. In addition, the isolated kraft lignins were analyzed for elemental composition (C, H, O, N, S), phenolic hydroxyl groups by ionization difference UV8, extractives by GC, and molar masses by size exclusion chromatography8,13.

2.3 Analysis of aliphatic carboxylic acids
Three types of aliphatic carboxylic acids were analyzed in most of the black liquors: hydroxy monocarboxylic acids and dicarboxylic acids by GC/MS as their trimethylsilyl (TMS) derivatives14, formic and acetic acids (and glycolic and lactic acids) by capillary electro-phoresis15, and tall oil acids by Saltsman–Kuiken method16, followed by GC/MS of the TMS derivatives.

3. RESULTS AND DISCUSSION
3.1 Carbohydrates in black liquors
The total amount of polysaccharides in the black liquors after the bulk delignification phase varied from 3.1 to 9.6 g/L (corresponding to 1.5–4.5% of wood), depending on the applied cooking process (Appendix 6.1). These total figures are in a good agreement with the pioneering data reported by Simonson17. The highest figure was found after the high alkalinity cooking at 180 °C.

The main monosaccharide units in the polysaccharides included arabinose, galactose, and xylose (Table 3.1), with only some glucose and mannose. This indicates that the main hemicelluloses are xylan, arabinan and galactan (cf. ref. 11). It clearly appears that the hemicelluloses dissolved at an early pulping stage contain high proportions (up to 50%) of galactose. The proportion of xylose typically varied from 30 to 65%, being highest for the high alkalinity black liquors. This is a reasonable figure for a softwood kraft black liquor.
 

Black liquor

Arabinose

Galactose

Glucose

Xylose

Mannose

R-H1

26.4

50.4

7.7

15.6

+

R-H30*

30.3

47.7

4.6

17.3

+

R-H200

26.3

26.0

3.2

44.5

+

R-H1000*

25.5

20.6

2.8

51.0

+

R-H1500

26.7

26.7

2.9

43.7

+

R-H2500*

28.0

36.9

4.3

30.8

+

HA-H30

32.3

47.4

4.1

16.2

+

HA-H350*

15.7

10.5

2.2

67.0

4.6

HA-H750

14.9

13.9

2.6

64.5

4.2

LS-H30

30.9

49.3

5.0

14.8

+

LS-H3000*

26.4

35.0

3.5

35.0

+

LS-H6000

27.9

48.0

6.1

17.9

+



Table 3.1. Relative monosaccharide composition (%) of hemicelluloses in the black liquors.
*Hemicelluloses isolated for further characterization.
 
The hemicelluloses were isolated with dioxane and acetic acid from five black liquors (Table 3.1), with widely varying yields (calculated from their total amounts). The yield was 93% for HA-H350 and R-H1000, 59% for R-H30, 34% for R-H2500, but only 10% for LS-H3000. Their monosaccharide compositions were analyzed by pyrolysis GC/MS (all samples) and acid hydrolysis – HPLC method (R-H1000 and HA-H350). It thus became evident that the composition of the total carbohydrates and those recovered by organic solvents may differ a lot, particularly if the isolation yields remain low. Typically, the isolated hemicelluloses contained somewhat more xylose and glucose, and less arabinose and galactose, than the bulk hemicelluloses present in the black liquors.

The hemicellulose preparations isolated with dioxane and acetic acid are known to contain circa 5% of lignin as the main impurity9,10. The lignin impurities were characterized by pyrolysis GC/MS, and compared with the corresponding kraft lignins. The comparison revealed a number of interesting features. It also indicated the presence of aromatic structures that are not characteristic of lignin.

Relative distribution of the guaiacyl-type degradation products from the isolated hemi-celluloses is compared with the products from the corresponding kraft lignin in Fig. 3.1. The most striking differences include more pronounced formation of guaiacol from the hemicellulose fractions, whereas the amount of 4-methylguaiacol is clearly decreased. It is likely that these changes are associated with the presence of lignin-carbohydrate bonds, although more detailed structural speculations are not currently possible.

In addition to the guaiacyl-type compounds, other phenolic compounds were also liberated from the hemicellulose and lignin fractions during pyrolysis (and identified by GC/MS). Their yield was 3–5% (of all aromatic products) from the lignin fractions, but up to 10–30% from the hemicellulose fractions. It also shows that their relative amount in the hemicellulose fraction increases during the course of pulping. This was clearly revealed by the reference cook series: the share of the other aromatic degradation products from hemicelluloses was 10% at H30, 13% at H1000, and 30% at H2500. In the bulk delignification stages of the high alkalinity and low sulphidity cooks, the corresponding figures were 15 and 10%, respectively.

The nature of the non-guaiacyl aromatic pyrolysis products appears in Fig. 3.2, showing distinct differences between the hemicellulose and lignin fractions. Some products (hydroquinone and 4 -hydroxybenzaldehyde) were liberated only from the hemicellulose fractions, whereas more substantial amounts of catechol and 4-methylphenol were derived from the lignin fractions than from the hemicelluloses.

It is reasonable to assume that certain non-guaiacyl-type pyrolysis products (particularly hydroquinone) derived from hemicelluloses originate from specific aromatic structures, formed into the polysaccharide chains by aromatization reactions during cooking.

 tech_black_1
Fig. 3.1. Distribution of the guaiacyl-type degradation products derived from the isolated hemicellulose and lignin fractions, as analyzed by pyrolysis GC/MS.

tech_black_2
Fig. 3.2. Distribution of non-guaiacyl-type degradation products from the analytical pyrolysis of the isolated hemicellulose and lignin fractions.

 
3.2 Kraft lignins
All the black liquors were analyzed for the lignin contents (Appendix 6.1) and characterized for lignin phenolic groups and molar mass distributions. The main focus is now given, however, on the selected properties of lignins isolated from five black liquors for further characterization (Table 3.2).

All the isolated kraft lignins contained some carbohydrates, varying from 1 to 9%. The highest amount of them were present in lignin from the high alkalinity cook, in agreement with the total carbohydrate contents of the black liquors (Appendix 6.1).

Property

R-H30

R-H1000

R-H2500

HA-H350

LS-H3000

Carbohydrates, %

1.0

4.0

2.5

9.2

2.7

Extractives, %

13.5

2.7

2.2

2.4

3.8

Sulfur, %

2.5

1.7

1.7

1.8

0.9

Phenolic OH groups, mmol/g

2.3

2.8

2.9

2.9

2.8

Conjugated phenolic groups, %*

17

13

11

14

10

Molar mass (Mw), mol/g

4460

6580

5980

9860

7070

Table 3.2. Selected properties of isolated pine kraft lignins.
*Percent of total phenolic OH groups.


Xylose and galactose were the dominating monosaccharide units in the lignin-bound carbohydrates (Fig. 3.3). The amount of galactose was especially high at the end of the reference cook, and in the bulk delignification stage of the low sulphidity cook.

tech_black_3
Fig. 3.3. Relative monosaccharide composition (%) of lignin-bound carbohydrates.
 
The amount of extractives in the isolated lignins varied from circa 2 to 4%, apart from the lignin fraction isolated at an early stage of cooking (R-H30, 13.5% of extractives). The GC analysis indicated that in each case fatty and resin acids were the main constituents, although some sterols and other compounds were also present. An important impurity of the isolated kraft lignins is organically bound sulphur. Its amount was expectedly the lowest (only 0.9%) after the low sulphidity cooking. The results from the pyrolysis GC/MS studies were already discussed in Section 3.1.

The amounts of the phenolic hydroxyl groups were little affected by the cooking conditions (or cooking stages). The molar mass of kraft lignin was clearly highest after the high alkalinity cooking.

3.3 Aliphatic carboxylic acids
The total amount of low-molecular-weight carboxylic acids (after the bulk delignification stages) was more than 20 g/L (Table 3.3), corresponding to c. 10% of wood. However, it appears that the amount of the hydroxy acids was now somewhat lower than could perhaps be expected on the basis of previous studies (e.g. ref. 18). In any case, these types of compounds are formed in substantial amounts during cooking, as a result of the degradation of polysaccharides (mainly hemicelluloses).

Most of acetic acid was formed (from the acetyl groups of glucomannan) during the initial stage of cooking, whereas the formation of the other carboxylic acids proceeded more or less constantly during the entire cooking.

Carboxylic acid

R-H1000

LS-H3000

HA-H350

Formic acid

6.3

8.0

6.3

Acetic acid

4.2

3.7

3.5

Hydroxy mono- and dicarboxylic acids

9.8

11.0

10.8

Total

20.3

22.7

20.6

Table 3.3. Concentrations (g/L) of low-molecular-weight carboxylic acids in the black liquors.

 
The main hydroxy monocarboxylic acids included glycolic, lactic, 2-hydroxybutanoic, 2,5-dihydroxypentanoic, xyloisosaccharinic, and isomeric glucoisosaccharinic acids, although a large number of minor carboxylic acids were also identified. Their relative amounts were now generally in a good agreement with the previous studies6,18 and are not listed in more detail. In addition, small amounts of aromatic hydroxy carboxylic acids,  -guaiacyl-2-hydroxyalkanoic acids19, could also be identified. Their formation requires condensation reactions between lignin-derived and carbohydrate-derived fragmentation intermediates.

As a whole, it appears that the different cooking conditions did not have dramatic effects on the formation of various low-molecular weight carboxylic acids. However, the results suggest that the relatively slow low sulphidity cooking has slightly favoured their formation.

Extractives (crude tall oil) were also isolated from the black liquors after the bulk delignification stages, and analyzed for individual compounds. The only striking feature worth mentioning is that the high alkalinity cook (at the temperature of 180 °C) had resulted in some losses of unsaturated fatty acids, especially linoleic acid.

4. CONCLUSIONS
The present results demonstrate that different pulping conditions can result in various changes in the composition and structure of the main black liquor compounds, although such differences are seldom very distinct. This type of information can be used to search for pulping conditions suitable for the production and isolation of potential raw materials (by-products) for various applications. Naturally, the pulp quality should not be compromised.

It became evident that the total amount of carbohydrates, their relative composition, and their recovery with organic solvents (at least with dioxane – acetic acid) can depend, to a large extent, on the cooking conditions. Typically, however, hardwood kraft black liquors are expected to contain more substantial amounts of hemicelluloses (especially xylan) than softwood black liquors, providing other opportunities for their recovery. It was also found that the extent of carbohydrate bonding with kraft lignin, to hamper their isolation, may be affected by the cooking conditions.

Several important kraft lignin properties (such as sulphur content and molar mass distribution) may also depend on the cooking conditions, although certain other properties (such as the amount of reactive phenolic hydroxyl groups) seem to be more constant.

The aliphatic carboxylic acids form an interesting but little studied fraction of potential by-products. Their separation has attracted some interest in the past20,21 but has not yet been realized in industrial scale. The main hydroxy carboxylic acids include products (such as lactic and glycolic acids) with well-established applications, whereas the properties and uses of some others (especially isosaccharinic acids) would still require further studies.
 
5. LITERATURE CITED
1.Lofthouse R. The European pine chemicals industry, 1987-1997. For. Chem. Rev. 108 (1998):6, 11–14.
2.Ragaskaukas A.J., Nagy M., Kim D.H., Eckert C.A., Hallet J.P., Liotta C.L. From wood to fuels: integrating biofuels and pulp production. Ind. Biotechnol. 2 (2006):1, 55–65.
3.Sainte-Cluque P. Global overview of crude sulphate turpentine. For. Chem. Rev. 109 (1999):1, 8–10.
4.Hinson J.M. Worldwide turpentine outlook 2002: optimism or concern? For. Chem. Rev. 112 (2002):6, 12–15.
5.Pye E.K. Industrial lignin production and applications. In: Biorefineries – industrial process and products (Eds. B. Kamm, P.R. Gruber, M. Kamm), Wiley-VCH, Weinheim, Germany, Vol. 2, pp. 165–200.
6.Niemelä K., Alén R. Characterization of pulping liquors. In: Analytical methods in wood chemistry, pulping, and papermaking (Eds. E. Sjöström, R. Alén), Springer, Berlin, 1999, pp. 193–231.
7.Hausalo, T. Analysis of wood and pulp carbohydrates by anion exchange chromatography with pulsed amperometric detection. 8th International Symposium on Wood and Pulping Chemistry, Helsinki, Finland, June 6–9, 1995, Vol. III, 131–136.
8.Tamminen T., Hortling B. Isolation and characterization of residual lignin. In: Progress in lignocellulosics characterization (Ed. D. Argyropoulos), TAPPI Press, Atlanta, USA, 1999, pp. 1–42.
9.Engström N., Vikkula A., Teleman A., Vuorinen T. Structure of hemicelluloses in pine kraft cooking liquors. 8th International Symposium on Wood and Pulping Chemistry, Helsinki, Finland, June 6–9, 1995, Vol. III, pp. 195–200.
10.Vikkula A. Hemicelluloses in kraft cooking liquors. Licentiate Thesis, Helsinki University of Technology, Finland, 81 p.
11.Tamminen T., Vuorinen T., Tenkanen M., Hortling B. Analysis of lignin and lignin-carbohydrate complexes isolated from black liquor. 8th International Symposium on Wood and Pulping Chemistry, Helsinki, Finland, June 6–9, 1995, Vol. II, pp. 297–302.
12.Tamminen T., Ohra-aho T., Hortling B., Tenkanen M. Residual lignin in hydrogen peroxide-bleached softwood pulps. 12th International Symposium on Wood and Pulping Chemistry, Madison, WI, USA, June 9–12, 2003, Vol. I, pp. 69–72.
13.Hortling B., Turunen E., Kokkonen P. Molar mass and size distribution of lignins. In: Handbook of size exclusion chromatography and related techniques (Ed. C.-s. Wu), 2nd ed., Marcel Dekker Inc., 2003, pp. 355–384.
14.Alén R., Niemelä K., Sjöström E. Gas-liquid chromatographic separation of hydroxy monocarboxylic acids and dicarboxylic acids on a fused-silica capillary column. J. Chromatogr. 301 (1984), 273–276.
15.Tamminen T., Ranua M., Dufour B., Kokkonen R., Kauliomäki S. Filtrate analysis as tool to follow peroxide bleaching performance. Papel 68 (2007):2, 82–91.
16.Saltsman W., Kuiken K.A. Estimation of tall oil in sulphate black liquor. Tappi 59 (1959), 873–874.
17.Simonson R. The hemicellulose in the sulphate pulping process [inaugural dissertation]. Svensk Papperstidn. 68 (1971):21, 691–700.
18.Alén R., Lahtela M., Niemelä K., Sjöström, E. Formation of hydroxy carboxylic acids from softwood polysaccharides during alkaline pulping. Holzforschung 39 (1985):4, 235–238.
19.Gierer J., Wännström S. Formation of alkali-stable C-C-bonds between lignin and carbohydrate fragments during kraft pulping. Holzforschung 38 (1984), 181–184.
20.Alén R, Sjöström E. Isolation of hydroxy acids from pine kraft black liquor. Part 2. Purification by distillation. Pap. Puu 62 (1980):8, 469–471.
21.Alén R., Sjöström E., Suominen S. Application of ion-exclusion chromatography to alkaline pulping liquors; separation of hydroxy carboxylic acids from inorganic salts. J. Chem. Tech. Biotechnol. 51 (1990), 225–233.
 
6. APPENDICES
Appendix 6.1. List of the conducted cooks and the corresponding black liquor samples. The bold figures refer to the bulk delignification stages.

H
factor

Yield
%

Kappa
number

Delign.
degree, %

TOC
g/L

TIC
mg/L

Lignin
g/L

Carbo-hydr., g/L

Reference cook (R)

 

 

 

 

 

 

1

85.9

10.6

15

560

6.9

0.8

30

74.0

29.0

36

690

25.8

1.7

200

66.2

54.0

36

680

37.8

2.7

1000

50.9

50.6

88.3

53

670

56.8

4.3

1500

47.1

32.6

92.8

61

620

61.4

4.0

2500

45.0

21.8

95.9

63

720

63.3

3.2

High alkalinity cook (HA)

 

 

 

 

 

1

81.5

14.9

13

960

7.3

n.d.

30

71.1

33.1

35

650

26.5

1.9

130

57.4

61.9

42

2290

38.6

n.d

350

45.4

45.2

90.0

47

750

50.0

9.6

600

42.4

21.8

95.2

n.d.

n.d.

57.6

n.d.

750

41.6

19.3

96.2

64

770

65.4

10.0

Low sulfidity cook (LS)

 

 

 

 

 

1

83.5

10.7

15

600

6.0

n.d.

30

74.9

21.0

32

670

19.7

1.7

200

66.8

37.1

31

1310

25.0

n.d.

3000

49.0

49.8

88.2

57

630

56.3

3.1

4500

47.5

36.9

91.5

n.d.

n.d.

57.7

n.d.

6000

47.0

29.3

93.1

59

710

55.5

2.3



7. CONTACT DETAILS
Klaus Niemelä 1, Tarja Tamminen 1,2, Taina Ohra-aho 1
1 KCL, PO Box 70, FI-02151 Espoo, Finland (e-mail: klaus.niemela@kcl.fi)
2 Current address: VTT, PO Box 1000, FI-02044 VTT, Finland

[Home] [Journal papers]