LIGNIN STUDIES

I. Investigations concerning lignin determination
Maija-Liisa Salo

Department of Animal Nutrition and Husbandry, University of Helsinki.

Received May 10, 1957

Lignin is a cell wall substance with a molecular structure still in many respects
hypothetical. Evidently it contains only C, H, and O. Most researchers consider
the N found in lignin preparations an impurity originated from the proteins (4, 12).
However, some authors (3) are of the opinion that the lignin molecule may contain
even N. It is generally agreed that lignin is a macromolecular substance composed
of phenylpropane derivatives. By a cautious oxidation both vanillin and syringalde-
hyde are obtained from lignin (13), and according to some investigators also p-
hydroxybenzaldehyde (14). Alcohols having the same cyclic structure, coniferyl and
syringoyl alcohol, are according to Freudenberg(9) important intermediary products
in the lignin formation in plants. All lignin preparations seem to contain methoxyl
groups.

The scanty knowledge concerning the structure of lignin makes the determi-
nation difficult. No appropriate components or reactions of lignin are known
according to which it would be possible even to prove the presence of lignin in a
certain plant product, let alone using them in a quantitative determination. Thus
it cannot with certainty be established whether a preparation is pure lignin. It is
also impossible exactly to decide upon the eventual solubility of lignin in a certain
procedure.

Different methods have been suggested for the determination of lignin. The
sulphuric acid methods are the most commonly used and are based on Pa yen’s
discovery (23) that the carbohydrates in wood are hydrolysed by strong sulphuric
acid at room temperature. Flechsig (8) was the first to use 72 per cent sulphuric
acid in cellulose hydrolysis. This fundamental principle has later been completed
with various additional treaments so that the acid methods at present used are
rather complicated and laborious.

Review of sulphuric acid methods

In the earliest methods used for the determination of lignin, extraction with
organic solvents was the only pretreatment. The material was, however, usually
wood, which makes the smallest demands on the lignin determination. After the

https://www.c-info.fi/en/info/?token=eeHDzDihS-KXO-F7.pPyNlFyMhpM_8O8689IFAQ.4oSZiWV2Ti_SyrtNAqSYBYtr4PUebNKLJK_q9L935ThkgP7oAtS06Lm5XeMwYjyi3mfJ5g40js6TGutWUpvU1SmXPxA6OOQX0X2eOn5xiZvVxCQkCnONw4ZWGxGl1qVrJUze-PKGC5fu5U8LrIIOY8gCnPN8GDa1d7JZN9SN


186

investigations were extended to herbaceous materials it became evident that a
more complicated pretreatment was needed. In the present methods there are
commonly three pre-extractions: 1) extraction with organic solvents, 2) extraction
with hot water or enzyme digestion, 3) prehydrolysis with dilute acid.

Ethanol-benzene (1 : 2) is commonly used for the extraction of lipid substances.
The length of extraction time differs in various methods. So e. g. Armitage et al.
(1) extract for 2 y 2 hours, Ellis et al. (7) for 4 hours, Norman and Jenkins (17)
and Goss and Phillips (10) for 30 hours. Thomas and Armstrong (24) found
that the extension of the extraction time from 8 to 16 and further to 30 hours
resulted in a considerable increase in the extractives and also in somewhat lower
lignin values. Moon and Abou-Raya (15) mention that a 4—6 hour extraction
removes the so called readily soluble lipid fraction, which is not completely hydro-
lysed with 72 per cent sulphuric acid. By prolonged extraction, however, further
extractives are obtained, but this fraction is soluble in 72 per cent sulphuric acid.

The pretreatment with hot diluted acid is necessary in particular for the removal
of readily soluble carbohydrates and proteins. Paloheimo (20) drew attention to
the fact that fructose and sucrose, when treated with supersaturated hydrochloric
acid, gave an insoluble residue up to 25 per cent which, in lignin determination,
would appear as lignin. Hilpert and Littman (11) and Norman and Jenkins (17)
have pointed out that in addition to those substances also pentosans may cause
a serious error, if not removed on prehydrolysis.

Some wood chemists (5, 2) deny the use of prehydrolysis because the boiling
with dilute acids removes from hardwoods substances having methoxyl contents
very similar to those of lignin. According to Moon and Abou-Raya (15) prehydro-
lysis removes lignin also from herbaceous plants. Brauns (4), however, considers it
doubtful whether a treatment with diluted acid will remove any true lignin.

There are small differences in various methods concerning the length of pre-
hydrolysis and the concentration of the acid. So e. g. Norman and Jenkins (17)
and Ellis et al. (7) reflux for 1 hour with 5 per cent sulphuric acid, Goss and Phil-
lips (10) for 1 hour with 1 per cent hydrochloric acid and Armitage et al. (1) for
1 hour with 5 per cent hydrochloric acid. Which concentration and length of time
would lead to most accurate results is as yet not settled.

Prehydrolysis removes also the main part of proteins. However, lignin prepa-
rations, separated from materials with high protein contents, always contain nitrogen.
This fact was first recognized by Paloheimo (19). He assumed that the nitrogen
in the lignin was present as protein and suggested that all lignin values obtained
from nitrogen-containing materials should be corrected by subtracting an amount
equal to the nitrogen content of the lignin multiplied by 6.25. A similar correction
is applied by some other investigators (25, 10). The removal of protein has been
accomplished also with enzymic digestion. One of the most frequently used enzy-
mes is pepsin (7, 15). Trypsin is also used (1), and in yet another method three
proteolytic enzymes successively (6). The enzymic digestion removes from plant
materials more nitrogen than the acid pretreatment, nevertheless the lignin pre-
parations isolated by methods having an enzymic pretreatment arc far from nitrogen-
free.



187

The hydrolysis of the cellulose is accomplished usually by a treatment with
72 per cent sulphuric acid for 2 hours at 20° C with occasional stirring. The mixture
is then diluted with water, generally to a concentration of 3 per cent, and the diluted
acid solution refluxed for 2—4 hours (17, 6). The hydrolyzed residue is filtered on
an asbestos-matted Gooch crucible, washed free of acid with hot water, and dried.
The lignin is determined by the loss of weight on ignition.

Experiments and results

The lignin determination method used by the writer is in the main based on
the principles described above. Special attention is called to the fact that the lignin
determined by the method should correspond with the indigestible lignin in the
matter to be analysed. This fact is important because in our department lignin
is used as a tracer in digestibility determinations as well as in investigations con-
cerning the time of the retention of food residues in the alimentary tract (21, 16).

The first modification of the method (method A) is carried out as follows:
The material is dried in vacuo at 65° C and ground to pass an 1 mm. screen.

2 samples of 2 g. weight are refluxed with 100 ml. distilled water for y 2 hour, filtered
into a filter paper and washed with hot water. The residue is transferred into an
Erlenmeyer flask with 100 ml. 2 N hydrochloric acid, refluxed for one hour and
filtered into a filter paper, washed with hot water and three times with 94 per cent
ethanol. The slightly dried precipitate, together with the filter paper, is introduced
into a Soxhlet apparatus and extracted for 8 hours with ethanol-benzene. The
dried residue is transferred into a 100 ml. beaker, moistened with 5 ml. water and
mixed with 45 ml. 77 per cent sulphuric acid. The concentration of the final solution
is 72 per cent. Water and acid are kept in a refrigerator to avoid the temperature
exceeding 20° Cby the adding of acid. The beaker is kept for 3 hours in an 18—20° C
water bath with occasional stirring. After this the contents of the beaker are trans-
ferred to a 600 ml. beaker and diluted with water up to a volume of 350 ml. The
solution is heated to boiling point and one of the samples is filtered on an asbestos-
matted Gooch crucible and the other on a filter paper. The residues are washed
with hot water until free of acid. The ignition loss is determined from the precipitate
on the Gooch crucible. This is the crude lignin. The crude protein is determined
from the precipitate on the filter paper. The quantity of lignin is obtained by
subtracting the protein correction from the crude lignin value.

Method B differs from method A in so far that a pepsin digestion is subs-
tituted for refluxing with water. A sample is placed into a 100 ml. glass-topped
Erlenmeyer flask, which is filled with 0.1 N hydrochloric acid solution containing
1 per cent pepsin (1: 900). The flasks are then kept for 20 hours at 40° C with
occasional shaking.

Method A has been proved to be very useful in tracer experiments. By method
B some lower and obviously more exact lignin values have been obtained fromplant
materials containing greater amounts of protein. Further, because the filtration
is more rapid after the pepsin digestion than after the refluxing with water, method
B saves time.



188

Table 1. The effect of the degree of fineness of the material on the results of the analyses.
A = 1 mm. screen, B = 0. 25 mm screen.

Membrane
substances ■)

Lignin
% of dry matter

Crude protein in
crude lignin

% of crude lignin% of dry matter % of crude lignin
A BA BA B

Timothy, early head stage 43.4 41.5 4.1 3.9 19.9 20.2
* in bloom 61.1 57.5 8.0 7.7 12.6 12.4
» after ripening of seed . . 62.6 61.9 9.1 9.1 5.2 4.7

Red clover, in bud 30.0 28.2 4.8 4.6 32.3 32.3
» in bloom 38.2 38.1 6.6 6.6 26.3 27.0
» after blooming 52.4 52.4 10.5 10.0 20.8 20.3

Birch wood 91.6 91.2 12.7 12.4 1.3 1.3

>) Determined by the method of Paloheimo (22).

Investigations concerning the significance and the necessity of the different
details in the methods have been carried out. As has been stated earlier, the materials
to be analysed have been ground in a Wiley mill to pass an 1 mm. screen. In des-
criptions of lignin determination methods various demands concerning the grade
of fineness are prescribed. However, these demands are not experimentally grounded.
The writer has made a comparative study of this question. One sample was ground
to pass the conventional 1 mm. screen the other to pass a 60 mesh (0.25 mm) screen.
The results presented in Table 1 show that the use of the finer screen resulted in
only very slightly lower yields.

The extraction with ethanol-benzene is performed after the prehydrolysis
treatments as this has been found to have some practical advantages. In compara-
tive analyses made to study the influence of various locations of the lipid extraction,
different materials behaved in different ways. So e. g. the lignin values obtained
from young timothy hay and young red clover were lower when the prehydrolysis
was made before lipid extraction, but mature hay and clover gave lower results
with the pretreatments in the opposite order. The lignin yields obtained from cow
faeces were similar with both modifications of the method.

The sufficiency of an 8 hour period for the extraction was also studied by
extending the length of the time from 8 to 16 hours. The amounts of extractives
obtained by an additional 8 hour extraction were e. g. in full-bloom timothy and
red clover only 0.3—0.4 per cent of the original dry matter. The prolonged extraction
diminished, however, to some extent the lignin values, as did the additional treat-
ments in general.

The acid prehydrolysis is made with 2 N hydrochloric acid. Thus the treatment
is more drastic than in the other lignin determination methods. A refluxing with
water precedes this treatment since experiments with various materials have shown
that boiling with water for y 2 hour diminishes the lignin values by o—lo0—10 per cent, in
a certain sample of young grass as much as 23 per cent. One reason for this is probably
the formation of insoluble substances from fructose and fructosans by boiling even



in diluted hydrochloric acid. The amount of those substances does not, however,
correspond to the decrease in the lignin yield by a pretreatment with water. When
e. g. fructose is refluxed with 2 N hydrochloric acid for 1 hour, the formed insoluble
residue is about 2 per cent of the fructose used, and after a subsequent extraction
with ethanol-benzene, which removes the greater part of the residue, only about
0.5 per cent is left. Some other water-soluble substances may also interfere with
the later treatments.

Metod A contains no enzyme digestion, since it was considered unnecessary
because of the use of the protein correction. However, in the later investigations it
appeared that the treatment with pepsin, besides the reduction of the protein
content of the preparate, had yet another effect on the course of the analysis. This
became evident in experiments arranged for checking the competence of the factor
6.25 in the protein correction. In these experiments the samples of different materials
were treated with pepsin in the manner described above (pretreatment B) and the
controls were treated in the same manner, but without the addition of the enzyme
(pretreatment A). Table 2 shows the crude lignin and the lignin percentages obtained
after the two pretreatments. It appears that the crude lignin values pertaining to
pretreatment A are greater than the corresponding values pertaining to pretreat-
ment B. The nitrogen percentages of these differences were calculated. As can be
seen these percentages are appreciably below the figure 16 which corresponds to
the factor 6.25. However, it is hardly possible that the N-percentage of the protein
fraction in the crude lignin after pretreatment A could be as low as the figures in the
last column of Table 2 indicate. Therefore one must conclude that by digesting the
proteins the pepsin facilitates the dissolution of some N-free substances which
otherwise would become a part of the crude lignin in full or in part. As to the
validity of the protein factor 6.25 in the crude lignin correction the experiments
could not give any solution.

As previously mentioned, the pretreated sample is moistened with water
before the adding of the strong acid. In that way it is possible to soften and disperse

Table 2. The influence of pepsin on the lignin values.

Crude lignin % of dry
matter Lignin % of dry matter N-% of the

difference

Incubation Incubation Incubation Incubation a 6
without with pepsin without with pepsin
pepsin pepsin
abed

Blades of rye 3.2 2.4 2.1 1.9 12.8
Red clover, in bud

.... 6.5 4.5 4.4 3.6 9.9
Ked clover, in bloom . . 8.4 6.7 6.5 5.9 9.8
Wheat bran 5.9 5.3 4.9 4.7 9.0
Cotton seed cakes ....

7.3 6.6 5.1 4.9 10.6
Sunflower cakes 7.6 5.8 5.1 4.6 12.0

189



190

Table
3.

Comparison
of
the
lignin
values
obtained
by
methods
A
and
B
and
by
the
method
of
Ellis,
Matrone,
and
Maynard.

Lignin
%

of
dry
matter

Lignin
methoxyl

Crude
protein
in

crude
lignin
%

of

Method
Method

EMM-
EMM-
%

G
f

lignin
EMM-
%

of
original
dry
matter

crude
lignin

A

B

method
method

method
~

,

with
pro-

Me

"
Me

" wl
th
pro-

Me"Me'EMM-
Me

_
Me

_
j.'yjyj.

tein
cor-

thod
thod
.

.

"

thod
thod
method
,

,
,

,
iciu
cor

tein
cor-
.„
thod
thod
me-

rection

rection

A

B

thod

Timothy,
leaf
stage

I

3.1

3.2

3.3

2.9

9.5

8.4

7.6

0.29
0.27
0.22

26.3
12.5
12.1

*

early
head
stage

|

4.1

4.0

4.3

3.9

10.8
11.0

9.9

0.44
0.44
0.39

19.9
9.7

9.1

»

in
bloom

I

8.0

7.6

7.8

7.4

12.9
13.2
12.9
1.03
1.00

0.95

12.6
4.3

5.5

»

after
ripening
of
seed
.

.

j

9.4

9.8

9.9

9.5

15.5
15.0
15.6

1.46
1.47
1.48

5.1

1.8

3.9

Red
clover,
in
bud

4.8

3.6

4.5

3.8

9.4
11.4
10.8

0.45
0.41
0.41

32.3
19.6
15.7

*

in
bloom

6.6

5.9

6.3

5.7

11.1
12.7
13.3

0.73
0.75
0.76

26.3
12.9
10.6

*

after
blooming

10.5

10.7
11.8

10.3

12.5
11.9

12.9

1.31
1.27
1.33

20.8
10.5
12.2

Sugar
beet
pulp

2.1

1.9

4.6

3.9

10.2
10.4

5.1

0.21
0.20
0.20

33.3
22.8

15.3

Wheat
bran

i

4.7

4.7

5.1

4.7

4.3
4.2

2.6

0.20
0.20
0.12

15.9
11.6

8.1

Horsetail
(E.

pratense)

I

2.5

2.4

3.1

2.6

2.1

2.0

1.6

0.05
0.05
0.04
j

36.5
30.7
15.6

Cow
faeces

I
19,1

21.8

19.0

11.5

11.7
2.20

2.22

14.0

13.2



191

the material, which, after the extraction with ethanol-benzene, in some cases dries
into solid pieces. 2—3 hours are required to complete the hydrolysis at 18—20° C.
The experiments showed that the extension of the hydrolysis time to at least 6 hours
had no influence on the lignin yield. Neither did the use of a whisking apparatus
instead of the occasional stirring in the beaker have any influence on the result
of the analysis. It further appeared that the preceding cooling of the water and the
77 per cent acid were not necessary.

In this method the mixture obtained by the dilution with water is not refluxed
but only heated to boiling point. Evidently the effectiveness of the acid prehydro-
lysis (2 N hydrochloric acid) used makes a longer boiling unnecessary. Only some
materials poor in membrane substances seem to demand prolonged boiling. If the
generally recommended prehydrolyses either with 2 per cent (0.56 N) hydrochloric
acid or with 5 per cent (1.05 N) sulphuric acid are used, and the refluxing of the
diluted mixture is omitted, higher lignin values are obtained and the methoxyl
content of the lignin is low.

In Table 3 the lignin values obtained by methods A and B are compared with
those obtained by the method of Ellis, Matrone, and Maynard and with the protein
corrected values obtained by the same method. The methoxyl contents of the
lignin preparations are also given. The methoxyl determinations have been made
by a modification of the Zeisel method (18).

A glance at Table 3 shows that with some materials method B gives somewhat
lower lignin values than method A. It is the result of the influence of pepsin as
previously shown. There is no significant difference in materials poor in protein.
On the nitrogen content of faeces lignin the use of pepsin had no effect. The Ellis
et al. method completed with protein correction gave values of the same kind as
those obtained by method B. Sugar beet pulp was the only exception. However,
from the methoxyl contents of the lignin preparations obtained from this material
the conclusion can be drawn that the lignin values obtained by methods A and B
are more reliable. Altogether it can be concluded that the lignin values obtained
by the three methods in question correspond to each other, a fact, which further
indicates that the acid methods are not very conventional, provided that certain
fundamental principles are followed.

Summary

In this investigation a sulphuric acid method for the determination of lignin
is presented and the influence of various details critically studied.

The samples ground to pass through a6O mesh (0.25 mm.) screen gave o—s0—5 %

smaller results than those ground through an 1 mm. screen.
Locating the extraction with ethanol-benzene before or after the prehydrolysis

had no significant influence on the lignin values.
The refluxing with water before the acid prehydrolysis lowered the lignin values

(N-free) to some extent. The pepsin pretreatment lowered them still further. The
nitrogen content of the part of crude lignin which was digested by pepsin was



192

9—13 %. On the nitrogen content of crude lignin obtained from faeces the use of
pepsin had no influence.

The hydrolysis procedure with 72 per cent sulphuric acid did not appear to
be conventional as to the extension of the hydrolysis above the time needed, the
effeciency of the stirring of the mixture or the stability of the temperature.

The refluxing of the dilution after 72 per cent acid treatment was not necessary
if the acid prehydrolysis was sufficiently effective.

The lignin values obtained by different acid methods correspond to each other
provided that certain fundamental principles are followed.

Acknowledgement. The author wishes to express her gratitude to Prof. Lauri
Paloheimo for the valuable help given at all stages of this work.

REFERENCES

(1) Armitage, E. R. & Ashworth, R. D. & Ferguson, W. S. 1948. The determination of lignin in
plant material of high protein content. J. Soc. Chem. Ind. 67: 241—243.

(2) Bamford, K. F. & Campbell, W. G. 1936. The determination of lignin in the analysis of woods.
Biochem. J. 30: 419—428.

(3) Bondi, A. & Meyer, H. 1948. Lignins in young plants. Biochem. J. 43: 248—256.
(4) Brauns, F. E. 1952. The chemistry of lignin. 808 p. New York.
(5) Cohen, W. E. & Harris, E. E. 1937. Pretreatment of wood with hot dilute acid. Effect on lignin

values. Ind. Engin. Chem. Anal. Ed. 9; 234—235.
(6) Davis, R. E. & Miller, C. O. 1939. Partition of the less easily digested carbohydrate complex

of forages. Ibid. 11; 651—652.
(7) Ellis, G. H. & Matrone, G. & Maynard, L. A. 1946, A 72 percent H 2 SO, method for the de-

termination of lignin and its use in animal nutrition studies. J. Anim. Sei. 5: 285—305.
(8) Flechsig, E. 1883. lieber Darstellung und chemische Natur des Cellulosezuckers. Z. physiol.

Chem. 7: 523—540.
(9) Freudenberg, K. 1952. The formation of lignin in plants. Mitt, österr. Ges. Holzforsch. 4: 40.

(Ref. Chem. Abstr. 46: 9172.)
(10) Goss, M. J. & Phillips, M. 1936. Studies on the quantitative estimation of lignin. I. Factors

affecting the determination by the fuming hydrochloric acid method. J. Assoc. Off. Agric.
Chem. 19: 341—350.

(11) Hilpert, R. S. & Littmann, E. 1934. Über die Verharzung der Zucker durch Säuren und ihre
Beziehung zur Lignin-Bestimmung. Ber. deut. chem. Ges. 67: 1551—1556.

(12) Hägglund, E. 1918. Beiträge zur Kenntnis des Lignins. Ark. kemi, miner, geol, 7,8; I—2o.
(13) Leger, F. & Hibbert, H. 1938. Studies on lignin and related compounds. J. Am. Chem. Soc.

60: 565—567.
(14) Leopold, B. & Malmström, 1.-L. 1952. Studies on lignin. IV. Investigation on the nitrobenzene

oxidation products of lignin from different woods by paper partition chromatography.
Acta chem. scand. 6: 49—54.

(15) Moon, F. E. & Abou-Raya, A. K. 1952. The lignin fraction of animal feeding stuffs. 111. The
determination of »total» lignin. J. Sei. Food Agric. 3; 595—608.

(16) Mäkelä, A. 1956. Studies on the question of bulk in the nutrition of farm animals with special
reference to cattle. Acta agr. fenn. 85: I—l3o.

(17) Norman, A. & Jenkins, S. H. 1934. The determination of lignin. I. Errors introduced by the
presence of certain carbohydrates. Biochem. J. 28; 2147—2159.



(18) Official methods of analysis of the association of official agricultural chemists. 910 p.Wisconsin.
(19) Paloheimo, L. 192? Die Verwendung der Säurehydrolyse zur Ligninbestimmung. Biochem.

Z. 165: 463—464.
(20) —>i—- 1926, Lignin determination by acid hydrolysis. 62 p. Helsinki.
(21) —*— & Mäkelä, A. & Salo, M.-L. 1955. Some quantitative data on the role of ruminant pro-

ventriculi in the digestion and absorption of nitrogen-free organic matter. J. Scient.
Agric. Soc. Finl. 27: 70—76.

(22) —» — & Paloheimo. I. 1949. On the estimation of the total of vegetable membrane substances.
Ibid. 21: I—ls.

(23) Payen, M, 1839. Extrait d’un memoire sur les ligneux. Compt. rend. 9: 149—153.
(24) Thomas, B. & Armstrong, D. G. 1949. A study of some methods at present used for the deter-

mination of lignin. J. Agric. Sei. 39: 335—345.
(25) Waksman, S. A. & Stevens, K. R. 1930. A system of proximate chemical analysis of plant ma-

terials. Ind. Engin. Chem. Anal. Ed. 2: 167—173.

SELOSTUS:

LIGNIINITUTKIMUKSIA

I. LIGNIININ MÄÄRITTÄMISTÄ KOSKEVIA TUTKIMUKSIA

Maija-Liisa Salo

Yliopiston kotieläintieteen laitos, Helsinki

Yliopiston kotieläintieteen laitoksessa on kehitetty ligniinin määrittäraistapa, jonka kulku on
pääpiirteissään seuraavanlainen. Hienoksi jauhettu näyte keitetään vedessä tai käsitellään pepsiinillä.
Jäännös keitetään 2 N suolahapossa sekä uutetaan etanoli-bentseenillä. Uutettu jäännös käsitellään
72-prosenttisella rikkihapolla, seos laimennetaan 6 kertaisella vesimäärällä, kuumennetaan kiehuvaksi
ja suodatetaan asbestin läpi. Hehkutuskevennyksenä saatuun raakaligniiniin tehdään toisesta näyt-
teestä määritetyn typpimäärän perusteella proteiinikorjaus.

Menetelmän eri vaiheiden merkitystä ja tarpeellisuutta on tutkittu ja kriitillisesti tarkasteltu.

193