CA1056558A

CA1056558A - Strenghtening of mechanical pulp by chemical treatment

Info

Publication number: CA1056558A
Application number: CA228,090A
Authority: CA
Inventors: Thomas J.M. Mcdonough
Original assignee: Canadian Industries Ltd
Current assignee: PPG Architectural Coatings Canada Inc
Priority date: 1974-06-11
Filing date: 1975-05-30
Publication date: 1979-06-19
Also published as: NO752026L; NO145141B; BR7503681A; ES438434A1; SE7506647L; NO145141C; FI66031B; JPS517201A; FI751733A

Abstract

ABSTRACT OF THE DISCLOSURE

A process for strengthening mechanically prepared fibrous lignocellulosic material such as groundwood pulp, thermo-mechanical pulp, chemi-mechanical pulp or semi-chemical pulp. The lignocellulosic material is treated first with a chlorine-containing reagent such as chlorine dioxide or chlorine, washed, and then treated with an alkali metal hydroxide in admixture with a peroxide such as hydrogen peroxide or sodium peroxide. The process provides a mechanical pulp having increased paper-making strength.

Description

;6558 This invention relates to a process for strengthening mechanically prepared lignocellulosic pulp by chemical treatment.
Lignocellulosic pulp prepared by mechanical means, for example, grinding, costs less to produce than ligno-cellulosic pulp prepared by chemical means such as the sulphate process. A mechanical process also gives a higher yield of pulp than a chemical process. Because of these advantages it is desirable to use mechanical pulps to the fullest possible extent in paper making, However, the use of mechanical pulps is limited by their low papermaking strength.
We have now found that lignocellulosic mechanical pulp can be strengthened by treating it initially with a chlorine-containing reagent followed after washing by treatment with caustic with or without admixture with peroxide, The treatment provides a marked improvement in the physical properties of the pulp as shown by burst factor, tear factor (for deciduous woods), folding endurance and breaking length measurements, Thus the principal object of the invention is to provide a means for improving the strength of fibrous lignocellulosic material prepared by mechanical methods.
Additional ob~ects will appear hereinafter, The process of this invention comprises the treatment of fibrous lignocellulosic material prepared at least partially by mechanical means by the steps of 1) treating the material with a chlorine-containing reagent, ~a~56~51~3

2) washing the material with water,

3) treating the washed material with alkali metal hydroxide with or without admixture with hydrogen peroxide or alkali peroxide, and

4) washing the material with water The lignocellulosic material for which the process of this application is applicable includes pulp prepared by ~echanical disintegration of wood in the presence of water, thermo-mechanical pulp, chemi-mechanical pulp and semi-chemical pulp. The disinte~ration may be carried out by stone grinders or by refiners such as the Bauer double-disc refiner.

The preferred chlorine-containing reagents are chlorine and chlorine dioxide. However, alkali metal chlorates, chlor-ites or hypochlorites may be employed. The alkali metal chlorates and chlorites may be employed in acidic medium by admixture with an acid such as hydrochloric. A mixture of alkali metal chlorate, hydrochloric acid and a catalytic amount of vanadium pentoxide may be used. Also a mixture of alkali metal chlorite and chlorine has been found effective.

During the first step of a preferred embodiment of the process,aqueous pulp at a consistency of 2,0% to 35,0%
by weight, preferably 3,0% to 6.0% by weight, is treated with 0.5% to 8,0% by weight, preferably 0,5% to 4,0% by weight, of chlorine dioxide at 25C to 90C, preferably 40C to 70C, for a period of 1/2 hour to 5 hours, preferably 1 hour to 2 hours, The treated pulp is then washed with water, ~os~ss~

During the third st~p of a pr~ferred embodiment of the process,aqueous pulp at a consistency of 6,0% to 35,0%
~y weight, preferably 12,0% to 18,0% by weight, is treated 1~ with 2% to 16% by weight, preferably 4% to 8% by weight of i alkali metal hydroxide and 0,2% to 4,0% by weight, pre~er-ably 0,5% to 2,0% by weight of hydrogen peroxide and option-ally up to 5.0% by weight of alkali metal silicate and/or up to 1.0% by weight of a magnesium salt at 20C. to 80C., preferably at 20C. to 40C. for a period o~ 1 to 72 hours, pre~erably 2 to 4 hours. The pulp is then washed with water.

In the first step the reagent employed may be chlorine, When employing chlorine the pulp at a consis-tency~ of 2.0% to 12,0% by weight is treated with 1.0% to 8,0% chlorine at 10C to 60C for a period of 5 to 120 minutes.
In the third step the hydrogen peroxide reagent may be omitted, the treatment being with alkali metal hydroxide only.
The preferred alkali metal hydroxide reagent of the third step is sodium hydroxide, The process of this invention provides a mechanical pulp having increased papermaking strength.
The invention is illustrated additionally by the following Examples but its scope is not limited to the e~bodiments shown therein, The physical properties shown were measured by the following procedures, 1~)51~55~3 ~Iandsheet Formation TAPPI T 205 m-58 (hot disintegration) Brightness TAPPI T 217 m-47 (Elrepho, pulp soured with SO2) Opacity CPPA E-2P
j Freeness TAPPI T 227 m-58 ¦ sulk TAPPI T 220 m-60 ¦ Tear Factor TAPPI T 220 m-60 Burst Factor TAPPI T 220 m-60 ~reaking Length TAPPI T 220 m-60 Elongation TAPPI T 220 m-6~
~olding Endurance TAPPI T 220 m-60 ~0.5 Kg tension) ` 30 grams (oven dry basis) of a hardwood groundwood pulp composed mainly of cottonwood species were placed in a covered glass vessel equipped with a stirrer, Water was added to give a consistency of 3,3% by weight and stirring ! was commenced. An aqueous solution of chlorine dioxide - was then introduced below the surface of the pulp, the amount being sufficient to gi~e a charge of chlorine dioxide equivalent to 1% of the dry weight of the pulp and a final consistency of 3,0% by weight. After treatment for 60 minutes at 40C the pulp suspension was filtered to remove most of the liquor and resuspended in 3 litres of water and filtered again~ The partially treated pulp was then transferred to a plastic bag and an aqueous solution of sodium hydroxide and hydrogen peroxideand magnesium sulphate was added to give charges of the three chemicals of 8.0%, 1.0% and 0.05% by weight respectively, the final consis-tency being 9% by weight. The reac-tion was continued at 40C.
for 3 hours, following which the pulp was filtered and washed twice with water.
The resulting pulp was evaluated for yield, brightness and mechanical properties, and its characteristics were compared with those of the untreated pulp as shown in the accompanying Table I.
;

40 grams ~oven dry basis) of a softwood groundwood pulp composed mainly of spruce and balsam species were treated in a manner similar to that described in Example 1 except that the chlorine dioxide step was carried out for 1 hour at 40C and the charges of chlorine dioxide, sodium hydroxide and hydrogen peroxide were respectively 2,0%, 4,0% and 1.0% by weight. This experiment was carried out six times and the results averaged. These results are compared in the accompanying Table I with the corresponding - data for the untreated pulp, The procedure of Example 2 was repeated except that the charges of chlorine dioxide, sodium hydroxide and hydroyen peroxide were doubled to 4,0%, 8,0% and 2.0% by weight respect- -ively, The results are shown in the accompanying Table I.

30 grams (oven dry basis) of a hardwood groundwood pulp composed mainly of cottonwood species were placed in a covered glass vessel equipped with a stirrer, Water was added to give a consistency of 5,0~/O by weight and stirring was commenced, An aqueous solution of chlorine was ~hen introduced

- 5 - , ~C~56558 below the surface o~ the pulp, the amount being sufficient to give a charge of chlorine equivalent to 8.0% of the dry weight of the pulp and a final consistency of 3.0% by weight. After treatment for 30 minutes at 25C the pulp suspension was filtered to remove most of the liquor and resuspended in 3 litres of water and filtered again. The partially treated pulp was then transferred to a plastic bag and an aqueous solution of sodium hydroxide and hydrogen peroxide and magnesium sulphate was added to give charges o~ the three chemicals of 8.0%, 0.5% and 0.05% by weight respectively, the final consis-tency being 12.0% by weight. The reaction was continued at 40C. for 3 hours, following which the pulp was filtered and washed twice with water.
The resulting pulp was evaluated for yield, brightness and mechanical properties and its characteristics were compared with those of the untreated pulp as shown in the accompanying Table I.

iO5~558 . _ Q~
U
~ C
h Ir) CJ~ ~
~I Ll'l O

~ O O ~ ~ ~ ~ o o~
O rl~
~1 ~
~ 0 _. .
~1 ~ ~ ~ a~
d' CO ~`1 ~1 ~`1 O
0 ~ ~
~ XN Lfl~ Li~ 1` ~ ~9 m _____ _ ___ __ o ~ ~ O
u) I' , , , , , U I` O ~ ~ CO t- ~
::5 la ~ N
m ~
.. ,. , ,_ h O
0 ~
~ U ~r) d' u~ u~ ~ ~ U) E~
~ , ~ _ HrX ~~D ~ ~ d' I` ~D ~
~1 ~1 \ C~ P
~ ~U .. .. . ,, P3 ~ u ~ ~ ~1 ~ ~`
E~ , . .
O Iq ~ u~, a) u) ~n ~o o oo u h O E~ ~O CO 1` Ln E'~ ~ ~

In ~ _I O ~ ~D O
~q . . , . . , .
~1 0 ~1 ~1 -1 L)~
h ~: ~ 1` ~ s) u~
m . _ __ _____ ___ a -rl U~ ~ ~ ~ ~ ~
~ la 5 ~n ~ ~ C~) m P~
a~ ~s 4~ ~ (a o ~ ~ ~ ~ ~ ~ a) ~
~ ~ ~ h ~a U~ ~ a) ~ o o ~ a~
~ h h h . . . . , X ~ ~ r~ ~
W
_ 1(~56558 ¦ EXAMPLE 5 40 grams (oven dry basis)` of the sof-twood groundwood of Example 2 were placed in a covered glass vessel equipped with a stirrer. Water was added to give a consistency of approxi-mately 4% by weight and stirring was commenced. An aqueous solution of chlorine was then introduced below the surface of the pulp, the amount being sufficient to give a charge of chlorine equivalent to 2% o~ the dry weight of the pulp and a final consistency of 3.0% by weight. After treatment ~or 30 minutes at 2SC. the pulp suspension was filtered to remove most of the liquor, resuspended in 3 litres o~ water and filtered again. The partially treated pulp was then transferred to a plastic bag and an aqueous solution o~ sodium hydroxide, hydrogen peroxide and 41 Bé sodium silicate and magnesium sulphate was added to give charges of the four chemicals of 4%, 1%, 5% and 0.05% by weight respectively, the final consis-tency being 12.0% by weight. The reaction was continued at 40C. for 3 hours following which the pulp was filtered and washed twice with water.
The resulting pulp was evaluated for yield, brightness and mechanical properties with the results shown in Table II.

The procedure of Example 1 was repeated except that the charges of chlorine dioxide, sodium hydroxide and hydrogen peroxide were 2.0%, 4.8% and 2.0% respectively. The properties of the resulting pulp are shown in Table II. The brightness obtained was 2 points higher than could be obtained by bleaching the same pulp in the conventional manner with 2.0% hydrogen peroxide, 1.9% sodium hydroxide, 5% 41 Bé sodium silicate and 0.05% magnesium sulphate.

` ~565S8 EX~MPLE 7 40 grams (oven dry basis) of the softwood groundwood of Example 2 was placed in a glass jar. Water was added to give a consistency of approximately 10% and the pulp was broken up by stirring. An aqueous solution of chlorine dioxide was then introduced below the surface of the pulp, the amount being sufficien-t to give a charge of chlorine dioxide equivalent to 4% of the weight of the pulp and a final consistency of

6.0% by weight. The pulp suspension was mixed by vigorous shaking and the reaction was allowed to continue for 3 hours at room temperature, following which the pulp was filtered, resuspended in 3 litres of water and filtered again. The partially treated pulp was then transferred to a plastic bag and an aqueous solution of sodium hydroxide was added to give a charge of sodium hydroxide equivalent to 8% of the wei~ht of the pulp, the final consistency being 12.0% by weight.
The reaction was continued at 80C. for 3 hours, folLowing which the pulp was filtered and washed twice with water. Its properties are shown in Table II.
2 0 EX~MPLE 8 The procedure of Example 1 was repeated except that hydrogen peroxide and magnesium sulphate were omitted from the liquor used in the alkaline stage. The properties of the - resulting pulp are shown in Table II.
EX~MPLE 9 The procedure of Example 5 was repeated except that 8%
chlorine was used in the first stage, and both hydrogen peroxide, sodium silicate and magnesium sulphate were omitted from the liquor used in the alkaline stage, which ~as carried out at 60C. for 2 hours. The results are shown in Table II.

_ g 10565S~

.~

~ t~ ~ LO ~ 1--o , ~ s ,, ~ o ~ ~
X ~ ~ a~ d' ~ Lr~ I`
~ ~ *
m __ ...
~ $ u~
~, o ~9 o~
~1 m ~ o ~9 ~ ~ ~ 0 a~ u u~ ~ d' E~

~ ~ 9 U7 H .Y ~~I ~ CO 1` t~l H ~ . . . .
~1 m u . o ~ ~q d' In o ~D CO

~1 In ~ u~ o ~ ~ ~ ~ ~ ~9 a) o~ ~ ~ ~ o ., ~S

~u~

~ ~ In ~ I~
- - --~5~558 EX~MPLE 10 150 grams of bisulphite chemi-mechanical pulp, prepared in 88% yield from softwood chips composed mainly of spruce and balsam species in the laboratory, were treated as described in Example 2. Subsamples of the resulting pulp were beaten to varying degrees in a laboratory PFI mill and their properties determined as a function of the degree of beating. In Table ~II
the resulting data are summarized and compared with corres-ponding data for the untreated pulp.

150 grams of busulphite chemi-mechanical pulp, prepared - in 82% yield from hardwood chips composed chiefly of aspen, were treated as described in Example 10, wi-th the results shown in Table III.

150 grams of a softwood cold soda pulp prepared in the laboratory from chips composed mainly of spruce and balsam species were treated as described in Example 10. With increasing ~ degree of beating, the tear factor of the treated pulp increased, `- 20 whereas that of the untreated pulp decreased. At a tear factor of 53, the breaking length of the treated pulp was 3.1 km., while that of the untreated pulp was I.9 km.

150 grams of an 85% yield neutral sulphite chemi-mechanical pulp prepared in the laboratory from chips composed chiefly of aspen wood were treated as described in Example 10. Some proper-ties of the treated pulp are compared with those of the untreated pulp in Table III.

150 grams of a 79% yield alkaline sulphi-te semi-chemical pulp prepared in the laboratory from softwood chips composed mainly of spruce and balsam species were treated as described in Example 10. Properties of the treated and untreated pulps are compared in Table III.

150 grams of a commercially produced, coarse softwood thermo-mechanical pulp were treated as described in Example 10.
Properties of the treated and untreated pulps are compared in Table III.

50 grams of commercial screened softwood thermo-mechanical pulp were treated in a manner similar to that described in Example 1 except that the chlorine dioxide stage was carried out ' ' for 1 hour at 40C. and the charges of chlorine dioxide, sodium hydroxide and hydrogen peroxide were respectively 2.0%, 8.0%, and 2.0% by weight. Upon completion of the a~kaline stage the ' pulp was diluted to 2% consistency and the pH adjusted to 5.0 by addition of an aqueous solution of sulphur dioxide. Following, - this, it was filtered and washed with water. The properties of the resulting pulp are compared with those of the untreated pulp in Table IV.

~ ' -/

~ [3565~8 _ . . .
U~
I` ~ u~ ~ o In o ~
u~ D ~ ~CO O
.~ Lt) ~D ~3 m .
~ ~ . . . . . ~
,~ ~ ~ ta u~
-~ ~ ~ c~ ~:
~ ~ ~ a~
a) _ _ ...
E~ ~ 0 0 f~ O 0 ~ 0 U~ ~ o o o o O ~ n .~ - -a~ ~ c~
.,, ~ ~ ~ 1 ~0 O ~ ~ g ~ 0 m . ~,~
~ ~ . ~
~: h ~ U~ `I ~1 ~:
U ~ ~ d~ S) 0 ~ S E~ ~ _l ~ ~ ~
m u' : . . ~ ~ ~
~ ~ .C . ~ ~ ~:
~ ~ ,~ U) ~
h ~ ,~ I` I` 1` r~d' d' OD OD
H E~ h ~:
H m ~
~n ~: ~ O
~:: h 00 00 0 0 0 0 00 i-1 ~a Z
E~ ~ O O O O o o o o o o H ~ O r~ n O O t` O Od' d' 1~ 1 0 11 0 Ir)1` 1~ 0 (~Il') 1` 1` d' r~ ~ S:
O
_ P~ ~
0~ *
O
3 ~D ~I ~ u) c5 a~ 0 ~ Q~
~ O
.~

_ _ _ _ _ ~
O ~ ~J ~1 O
~ ~QJ a) o a~
.,~ ~ u u ~
S ~rl rl ~ ~ ~ ~ r~ S
~ 1~ ~ s 0 --I ~l QJ Q,h S h S 1~ ~ ~15 ~ 115 n~ ,1 ~ I ~ ~ ~q ~
a) tq ~ O u ~ 3 ~ m m h mm ~~ o o . . . .~ a~
~ ~ ~ . ~:S ~3 3 h _1 3 3 ~ ~ 3 3~ ~ ~ O 4-1 ~
::~ ~ ~ 3 3 '1~ O
P. ~ ~ ~ ~ P~
u~C o . , ,. . - ~ ~
. ~ O (O
-- ~ h ~ h h h ~ h x o 'I ~ ~r In .lC
Z
_ _ _ .

-~Lo~ss~
- - - ~

O ~
~ - ~

r~

s~ ~ ~ Ln . . ___ N ~r O ~ ~
~1 la r-l ~) m ~ .
5~0 o CO
, ~ _ .
OD ~
H m_ ~ . . ~

o s~ ~ ~ ~
_ o ~..

~ O U) ~D
- ~

~ ~ ~ o ~ . i.
'~ ~ _l ~

a) ~ ~
o ~

~05655~
EX~MPLE 17 50 grams of a commercial soEtwood refiner groundwood were treated according to -the procedure described in Example 2. The experiment was then repeated three times, each time replacing the sodiunl hydroxide used in the alkaline s-tage with an equivalent amount of an alternative base. The amounts and bases used were 3.5% NH40H, 6.2% Na4C03.H20 and 3.7% Ca(OH)2. The properties I of the four resulting pulps are compared with those of -the j original refiner groundwood in l'able V.

50 grams of a commercial softwood refiner goundwood were treated according to the procedure describe~ in Example 2, e~cept that 4% chlorine dioxide was used in the first stage and 8% ~odiu~ hydroxide was used in the second stage~ The exper-iment was then repeated four times, each time replacing the chlorine dioxide used in the first stage with an equivalent amount of an alternative oxidant. The amounts and oxidants used were as follows: (a) 6.7% NaC10~ with 3.2% HCl, ¦ (b) 5.4% NaCl02 with 2.1% C12, (c~ 6.4% NaCl03 with 4% HCl and 0.032% V205, and ~d) 10% NaOCl. The properties of the five resulting pulps are compar~ with those of the original refiner groundwood in Table VI.
- .

~05tiSS8 . ._ ~ U
O
0~ ~ O ~ O d E4 ~
O
t~ S~ ~ O
O ,~
E~
~ S
.Y ~ ~ o ~ I
.Y
S~ ~
m _ _ _ .
O QO
S~ U ~ ~ ~I` ~D

Q) U

e .Y ~
~S U ~ ~ ~ I t`J
~ U
.. _.
~q a O o U~ O U~
~ ~ o _1o ~1 i l ~4 ~ 1 g ,1 O ~ ~ I`
U ~
~ ~`~`3 Ln~ Lr) _-,~ u~
S~ ~ Ln~D d'~D
m ~ ~ _ o ~, ~ ~I o ~i~ ~ ~
Q) ~ ~ ~ O ~ ~a~ o .,, t.
~m ~: N
~1 ~ ,_ ~_, m o m ~:: oC~ o ,Y tJ) Q ~ N --s~
~ ~ Z Z ~ C~
m r, 4~ ~ s~
O ~ ~E~
~ P~ g . _ .

11~56~5 . _ ~ ~, ~:
rl ~ ~ ~9 ~ ~ ~,~ U~
,~ ~ ~ n ~ ~ ~ In O ~ L~
h ~:

~ O ~ d~
O ,1~
,1 ~ ~1 ~1~i ~1 ~1 ~1 ~3 ~
~1 ~ ~ ~ ~ ~u7 n .Y ~ o ~ ~ o m O OD ~Du~ ~~7 , ~q ~
5~ ~ d~ ~
~ 0 ~, m ~
.
5~ ~
~9 d' 1` 0 0 ~ U
E~ ~
~ _ _ .Y ~ ~
P ~ ~ ~ o ~1 0 d' m u .
~n ~ ~ o u~ ~ ~1 o In aJ ~ ~ CO ~ o ~I
s~ ~ ~ ~
~1 ~1 _ u -m ~i 0 ~ tn ,, tO ~ ~ ~~ o h 0 m ~ _ I
o .~9 o . ~ ~
m N ,1 rl (D
O r-l N ~ (i~ 1~ t~l ~ ~,_1 p~
U) ~1 U~
_ ~
O ~I N
~) O ~
~ rl ~ N II ~I C) 111 U~ O N N C) ~\
,_1 oo .~ ns .,, ~ ~c~ ~ ~ æ z ~ C~ C~
c~. ~ (a ,~
. æ ~
.... I
q:~
~1 h S~ h 51 Si h O ~ ~ E~
, U~ , P

Claims

The embodiments of the invention in which an exclusive property or privilege is claimed are defined as follows:

1. A process for the strengthening of fibrous ligno-cellulosic pulp prepared at least partially by mechanical treatment which comprises the steps of 1) treating the lignocellulosic pulp with a chlorine-containing reagent, 2) washing the treated pulp with water, 3) treating the washed pulp with alkali metal hydroxide in admixture with a member of the group consisting of hydrogen peroxide and alkali metal peroxides, and 4) washing the hydroxide-peroxide treated pulp with water, the resulting product being characterized by increased strength compared with the initial pulp used in Step (1).

2. A process as claimed in Claim 1 wherein the fibrous lignocellulosic pulp prepared at least partially by mechanical treatment is groundwood pulp, thermo-mechanical pulp, chemi-mechanical pulp, or semi-chemical pulp.

3. A process as claimed in Claim 1 wherein the chlorine-containing reagent is chlorine or chlorine dioxide.

4. A process as claimed in Claim 1 wherein the chlorine containing reagent is an alkali metal chlorate or chlorite employed in acidic medium.

5. A process as claimed in Claim 1 wherein the chlorine-containing reagent is a mixture of alkali metal chlorite and chlorine.

6. A process as claimed in Claim 1 wherein the chlorine-containing reagent is an alkali metal hypochlorite.

7. A process as claimed in Claim 1 wherein the first step of the process defined therein comprises treating aqueous lignocellulosic pulp at a consistency of 2.0% to 35.0%
by weight with 0.5% to 8.0% by weight of chlorine dioxide at 25°C. to 90°C. for a period of 30 minutes to 5 hours.

8. A process as claimed in Claim 1 wherein the first step of the process defined therein comprises treating aqueous lignocellulosic pulp at a consistency of 3.0% to 6.0%
by weight with 0.5% to 4.0% by weight of chlorine dioxide at 40°C. to 70°C. for a period of 1 to 2 hours.

9. A process as claimed in Claim 1 wherein the first step of the process defined therein comprises treating aqueous lignocellulosic pulp at a consistency of 2.0% to 12.0%
by weight with 1.0% to 8.0% by weight of chlorine at 10°C. to 60°C. for a period of 5 to 120 minutes.

10. A process as claimed in Claim 1 wherein the third step of the process defined therein comprises treating the washed lignocellulosic pulp in aqueous medium at a con-sistency of 6.0% to 35.0% by weight with 2.0% to 16.0% by weight of alkali metal hydroxide and 0.2% to 4.0% by weight of hydrogen peroxide at 20°C. to 80°C. for a period of 1 to 72 hours.

11. A process as claimed in Claim 1 wherein the third step of the process defined therein comprises treating the washed lignocellulosic pulp in aqueous medium at a con-sistency of 12.0% to 18.0% by weight with 4.0% to 8.0% by weight of alkali metal hydroxide and 0.5% to 2.0% by weight of hydrogen peroxide at 20°C. to 40°C. for a period of 2 to 4 hours.

12. A process as claimed in Claim 10 or 11 wherein the alkali metal hydroxide reagent is sodium hydroxide.

13. A process as claimed in Claim 10 or 11 wherein there is employed as an additional reagent up to 5.0% by weight of an alkali metal silicate.

14. A process as claimed in Claim 10 or 11 wherein there is employed as an additional reagent up to 1.0% by weight of a magnesium salt.