Note: Descriptions are shown in the official language in which they were submitted.
`` ~205~3~
-
r~he present invention relates to a method for
the purification o~ condensates from the sulfa~e process.
In the production of cellulose according to the
sulfate process the fibrous raw material (generally wood chips)
i~ treated under high pressure and high temperature with so
called white liquor~ the 'active chemicals of which are so-
dium hydroxide and sodium sulfide (l,Z). Additionally, some
manufacturers use other chemicals, e.g. polysulfide.
lQ
After digestion the' fibrous materi~l (sulfate
cellulose or kraft pulp) is separated from the dissolved
organi~ material and the inorganic chemicals. Because of
its colour the liquid phase is now called black liquor.
Said black liquor is now concentxated to such a high solids
content that it can ~e burnt. Heat is, thus, released
(possibly electric power can be obtained), and inorganic
chemicals are recovered and may be converted to white
liquor chemicals, which is necessary ~o make the process
economical. Some volatile compounds (turpentine methanol
and others) and air are 'removed separately from the digester
or are separated from the black liquor when it is subjected
to reduced pressure (as compared to the digestion pressure).
Aftex cooling, certain of said components will leave with the
water steam to ~e condensed. The water insoluble phase is
generally called sulfate turpentine. The condensates from
this process are generally called digester condensates.
Concen'tration of the black liquor, including the '
3Q washing water from the'displacement of black liquor from the
cellulose,'is generally carried out in a multi-step evaporation
plant, partly in a vacuum. Evaporated water is discharged
as con-
~Z05~3~ -
, . .
densates. To~ether with any water from the vacuum pump of
the plant, the wa-ter dlscharged at this sta~e is called
evapora-tion condensa-tes. Apart from the portion origina-ting
from indirect heatin~ with live steam, said evaporation
condensates are also more or less polluted
In some plants part of the s-team from the digèstion process
is used for pre-evaporation of the black liquor. Such pre-
evaporation results in condensates considered as something
between digester and evaporation condensatesO rrhe different
kinds of stripping steam and condensates can be lin~ed to-;
s ~ ~o~ , A.'s
gether in various manh~Es in the process~ ~L~Y:, th~t is~
of no concern in this connection~
A soap (sulfate soap) is separated from black liquor. It
consists mainly of saponified fatty acids and rosin acids
(abietic acids), but it also contains unsaponified com-
pounds. Said soap is generally recovered and typically con-
verted to so called tall oil. The evaporation condensates
will also contain more or less finely divided "oil", i.e. a
water insoluble phase that is disadvantageous to any fur-
ther treatment of the condensates.
It is common to all the above mentioned condensates that
they contain various volatile substances, some of which are
lower sulfur containing compounds. Especially during evapo-
ration foul and toxic gases, e.g. hydro~en sulfide (H2S),
methyl mercap~an (C~3S~I) and the like, escape from the
black liquor, and quite a significant portion of these wiil
be found in the condensates. Thus, said condensates have
an obnoxious smell and are toxic.
Their content of methanol and the ]ike makes condensate
effluents a pollution load on the recipient because dis-
solved organic material will cause a chemical oxygen demand(COD) and a biochemical oxygen demand (BOV). These three
factors, namelythe obnoxious smell, the toxicity to aquatic
life, and the organlc~` load cause the c~nden,~tes from the
. ' ~
. :
.
lZ~9;~
kraft process to represen-t a serious environmental hazard.
The possibili-ties of reusing unpurified condensates as a
- source of li~uor in the process are also limited because
they release toxic and foul gases and contain finely divided
oils.
The methods in use today for puri:Fying the condensates from
the kraft process are primarily based on the removal of
volatile compounds (among others methanol and substances
having an obnoxious smell) by the`aid of steam and air
(stripping) (2a,2b). The removed gases/vapours are typical-
ly destroyed by burning together with any non-condensable
gases from the digestion and recovery process.
A process of purification based on decomposition by the aid
of microorganisms is also used (3).
.
It is an object of the`present inven-tion to remove the foul
substances in the condensates by combined oxidation and ad-
sorption after a preceding removal of finely divided oil,
inter alia by coalescence when that is necessary. A smallpilot plant has been tested for a few years in a Norwegian
kraft mill for examination of various types of condensates,
catalysts, life times and possib~e means of regeneration of
the catalysts and so on.
It is known that hydrogen sulfide gas in air is oxidized on
a suitable catalyst, e.g. activated carbon. In said gas
reaction H2S is converted to elemental sulfur. Said process
may be extended to sulfur diox~de (S02), which is also (to-
gether with some S03) the final product when H2S, sulfur
and many other sulfur compounds are burned~ I-t has also been
sugges-ted to purify foul gases by the aid of alkali and
activa-ted carbon.
Hydrogen sulfide is soluble in alkali, e.g. caustic soda
with formation of sulfides(NaHS and /or Na2S in NaOH). In
~- the very stron~ly basic white liquor from the sulfate pro-
- ~205~30
,, ~ . ,
cess sodium hydroxide and sodium sulfide are as mentioned
the active digestion chemieals. Such a white liquor is
slowly oxidized by air, and the oxidation pr~ducts will be
sodium sulfite~ sodium thiosulfate and sodium sul~ate. In
the presence of a suitable oxygen converter or eatalyst the
oxidation time may be eonsiderably reduced, and sodium poly-
sulfide and sodium thiosulfide are formed. By way of example
admixture of a eertain portion of blaek liquor (PFI-patent)
or use of a speeial eatalyst mass (Meads Moxy proeess) (4,5)
1~ may be mentioned.
Activated carbon is generally known in the art
of purification for the removal of small amoun-ts of eontamin-
ants. This is based on the large internal surface area of
the substanee~ The surfaee will be blocked by the adsorbed
substance, and some kind of regeneration will be neeessary
in eases of larger amounts of eontaminants.
According to the present invention there is provided a method
for the purification o~ condensates from the sulfate proeess which com-
prises passiny the condensates eoncurrelltly with an oxygen-containing
gas through a column containing activated earbon.
~,
EXPERIMENTS
Polluted and fou`l condensates from a kraft mill
are passed with air through a eolumn filled with aetivated
earbon.
Preliminary tests showed that
1) The obnoxious smell was removed by said treatment.
2) EIydrogen sulfide and methyl mercaptan were removed (de-
monstrated by potensiometrie titrations).
.
3) The smell, hydrogen sulfide and methyl mereaptan were not
~rt
~- 4 -
lZ~93(;~
. .
removed, unless sufficient amounts of air were passed
with the liquid.
4~ The column lost its efficency after some time.
In addition to the fact that it was possible to
remoye the obnoxious smell, hydrogen sulfide, and methyl
mercaptan ~y this method, the tests disclosed that what
had taken place had to be an air oxidation (items 1, ~, 3)
1~ and that the
. .
.
~ - 4a -
., ~
: ~Z~S~3(~
column material had been active (item 4), i.e. that a cata-
lytic oxidation had occurred.
It was assumed that hydrogen sulfide was oxidized to elemen-
tary sulfur, which remained on the column material. This
assumption was corroborated by the fact that hot white li-;
quor passed through a column was enriched with
polysulfide. (Colour yellow/orange/red, dependent on the
concentration. Determined quantitatively by potensiometriG
titration after conversion to an equivalent amount of thio-
~- sulfate by sodium sulfite.~ It is known that elementary sulfur is dissolved in sulfide solutions, e.g~ white and
green liquor. The test, thus, also showed -thai th~ sulfate
process itself provides chemicals that may be used to libe-
- 15 rate the column material from the deposited sul~ur.
.
The untreated condensate contained oil-li~e droplets of
varying amounts and droplet sizes. After havin~ passed
through said column the condensate, however, was perfectly
- 20 clear. Such oil deposition assumedly must reduce the active
surface of the column material. This was corroborated by
the fact that the column after losing its efficiency was
made efficient again after treatment with steam at approxi-
mately l~O til 190C.
Observations of the condensates for some time showed that
the amount of oil could var~ within wide limits . Efforts
were, thus, made to remove the oil before the condensate
reached ~he column. An efficient oil removal was achieved
by first removing large droplets that were allowed to float
on top of the aquous phase and then removing the finely
divided floating oil droplets from the water phase by coa-
lescence.
Samples of the cleared condensates after coalescence fil-tra-
tion were extracted with organic solvents and -the extracts
were examined by gas chromatography. The analyses showed
that a large amoun-t of polar and non-polar orqanic com-
~ ' ' .
12~ 306
pounds were present. Corresponding analyses of samples
taken aEter the condensates had passed through -the column
showec1 that said compounds were almost cornpletely removed
by -the treatment. The compounds detec-ted by said method oE~
analysis were assumed to be higher organic sulfur compounds.
Since these compounds as well as hydrogen sulfide and lower
organic sulfur compounds are expected to have an acute
toxic effect on the life in the recipient, condensate
samples were used for mortality tests with salmon fry. I-t
was found that the untrea-ted condensates were 3 to 35 times
more toxic than the treated on~s.
.
The toxicity tests ~96 hours - LC50) were carried out at
the Norwegian Institute of Water Research (NIV~). The gas
chromatographic investigations were carried out at the
Norwegian Central Institute for Indus-trial Research (SI),
among others by the aid of a ~lass capillary column and a `
flame ionization detector. The content of low boilin~ orga-
nic compounds, e.g. methanol, was studied at -the Norwegian
Pulp and Paper Research Institute (PFI) ~y the aid of gas
chromatography with a polyalkylene glycol column and a hot
wire detector. In addi-tion to methanol there were ound
traces of another compound (assessed at less than 1,5% of
the amount of methanol). This indicatès that methanol could
in pratice be enriched from the purified condensates with-
out too great demands on fractionated separa-tion. Examina-
tions oE volatiles by head-space-technique showed a re-
duction of the num~er and the total content of escaping
odourants at 40C (SI).
The pilot plant used is shown in a flow sheet of the accom-
panying drawings. Its construction is as follows:
By the aid of valves (1,2) and pumps (Pl,P2) a portion of
the flow was taken from the condensate line ~in the embodi-
ment shown in the drawing it is taken from the outlet of
the evaporation stage). Any solids and larger oil droplets
are separated in a preliminary filter (3) having an over-
. ~ , .
. .
IZ~ 3t~
flow (4). The coalescence uni-t consists of two parallel
coalescence filter cartridges (5~ of the Bdls-ton type. The
oxidation is carried out in a 4 m long steel column contain-
ing 8 l of catalyst mass. The column receives condensate
(alternatively white liquor for the removal of deposited
sulfur) via pump (P3) and a flowmeter (9), it receives
compressed air via a meter (l8) and steam for regeneration.
The efficiency of the column is monitored by a H2S detector
mounted below the cover of the collecting tank and additio-
nal control is achieved by sampling for assessment of theodour and for potentiometric titration of H2S and CH3SH.
.
-Examination of the Efficiency of the Oxidation Column:
~ `
15 Column material: Activated carbon, 8 litres, ~rain size 0~5-
2 r5 mm
Type Lurgi Hydraffin LS supra
Note: Preliminary tests carried out with
inter alia ordinary activated carbon
lacking technical specifications and
bought in a shop in Oslo yielded
corresponding results às regards
odour, H2S and CH2SH.
The column was filled on Oc-t. 2nd/ 1979.
Experiments were carried out from Oct. 2nd to Oct. 4th, 1979.
Incoming condensate: EVAPORATION CONDENSATES:
H2S Orl - 0~9 g S/l
CH3SH o 0 3
Ei2S + CH3SH O~l - lr2 g S/l
pH 7,~6 - 8~0
Outgoing condensate:
H2S + CH3SH O g S/l
pH approx. 9~7
Amount of condensate 15 l/h, totally 720 l
Amount of air approx. 40 l/h,totally 1800 l
Totally removed H2S ~ CH3SH corresponding to 360 ~ S
Regenerated with approx. 5 kg steam (190C, 13 kg/cm
Continued Oct. 15 - Oct. 17.
~ , .
.
l2~8sg~ `
Incoming condensate: -
S 0~0~7 ~ S/l
13SH 0~0,r3
H2S + HC3S11 0~.2 ~ 0 g S/l
pH approx~ 8
Outgoing condensateo
H2S + CH3SH O g S/l
pH approx. 9~.6
Amount of condensate 15 l/h, totally~ ~50 1
Amount of air ~ 40 r~ 1/ (31.0 1
Totally removed l12S -~ CH3SH corresponding to 2 O ~ S
Continued Oct. 22, - Oct. 25,
Amoun-t of condensate totally 600 1
Amount of air " 1630 1
~ pH (in) approx. 8
pH (out) " 9~5
Totally removed H2S + CH3SH corresponding to 310 g S
Amount of condensate after previous steam regeneration:
350 1 + 600 1 = 950 1
Amount of condensate totally: 720 1 + 950 1 = 1470 1
Removed 112S + CH3SH totally corresponding to 54Q g S
Regenerated wi h approx. 6 kg steam (190C, 12~ kg/cm2)
Regenerated with approx. 4~1 1 hot white liquor, 30 min.
Drained off Li~uor containing polysulfide. Washed with
approximately 40 l water.
Continued February 25th - 27th, 1~80
~mount of condensate totally 330 1
~ount of air " 1000 1
pH (in) approx. 7~8
pH (out) " 9~7
Totally removed H2S + CH3SH corresponding to 230 g/S
Continued March 4th
Rn~ount of condensate totally 470 i
Amount of air ~ " 1400 1
Totally removed H2S -~ CH3SH corresponding to 330 g/S
- 560 g/S
:
: ..
~205~30
The series of experiments was ended after the -treatment of
totall~ 2270 1 of non-purified evaporation condensate and
removal of ~2S ~ CH3SH corresponding to 1430 g sulfur.
Experiment carried out May 28th - Oct. 9th 1980: Di~ester
Condensate
Column ma-terial as above
Column filled May 21st
Incoming condensate:
~ H2S + CH3SH traces
pH 9~0 - 9
Outgoing condensate:
H2S + CH3SH O ~/
pH 9t`5 10 p
Amount of condensate: 15 l/h, totally approx. 5000 1
Amount o air ~0 " " " 13300 1
Experiment completed, column still efficient.
Gas Chromatographic Analyses (carried out by SI)
Extraction with cyclohexane involves non-polar compounds,
~ollowe~ by acidification and extraction with butyl acetate
and deri~ation with a trimethylsilyl-reagent to determine
the more polar compounds. The first mentioned were studied
in a ~lass capillary column, the last mentioned in a packed
column, both by the aid of a flame ionization detector.
Evaporation Condensate from Oct. 17th, 73
The amounts of both non-polar and polar compounds were very
much reduced after the condensate treatment in -the oxida-
tion ~olumn. The main component of the non-polar compounds
was reduced approximately ?0 000 times ~from 60 ppm to 3.10 3
ppm). A reduction of approximately 600 times was found for
the main component of polar compounds.
~205~30
Digester Condensate from early in June 1980
For the non-polar compounds a reduction of approx. 160
times was found. In the polar frac-tion no compounds were
found in the sample taken after the oxidation column, which
corresponds to a reduction of at least 200 times, based on;
the detection limit.
.
Odour Analyses by ~ead-space Technique (carried out by SI)
The evaporation condensate samples from upstreams and down-
streams the oxidation column were heated to ~0Cfbr 1 hour,
and the gas above the liquid (head-space) was extracted,
concentrated on activated carbon and analyzed by gas chro-
matography. Great difference of the samples was re~orted,approximately hàlf of the compounds from the sample up-
streams of the oxidation ~eing absent in the sample down-
streams of the oxidation. The total content is approximate-
ly 4 times higher in the untreated sample. A few new vola-
tile compounds seemed to have been produced, but these werepresent in very small amounts. The main compound present
seems to be dimethyl disulfide.
Destillation of Metanol and Determina-tion of its Purity
Methanol was destilled off (amounting to approx. 5 g/l) and
was analysed by gas chromatography at PFI. The chromato~rams
only showed traces of one compound in addition to methanol,
~ the amount of which was assessed to be less than 1~.5% of
the amount of metanol.
Coalescence
The efficiency of coalescence filtration was proven by the
fact that 1) "oil" accumulated on top of the unit and
that 2) the condensate was often mil~ upstreams of the
unit and clear downstreams. The oil content in the un-
treated condensates varies within very wide llmits.
~ ,
:
... ... . ,. .. ~.. .
