- Email: [email protected]
The removal of sulphuric acid from natural and industrial waste waters
431
Desalination, 70 (1988) 431-442 Elsevier Science Publishers B.V., Amsterdam - Printed in The Netherlands
THE REMOVAL
OF SULPHURIC
ACID FROM NATURAL
AND INDUSTRIAL
WASTE
WATERS
A.E. SIMPSON and C.A. BUCKLEY Pollution Research Group, Department George V Avenue, Durban, 4001.
of Chemical Engineering,
University
of Natal, King
SUMMARY A process is described which enables the removal of sulphuric acid from effluents without the addition of chemicals to the effluent. The technique employs an anion selective membrane which separates the acidic effluent from a lime solution. In the case where no current is applied, facilitated transport occurs and the driving force for the demineralisation process is the difference in the chemical potential of the species on either side of the membrane. The equilibrium position may be shifted by the passage of a current through the membrane, in which instance the driving force becomes the applied electric potential. The sulphates are removed as calcium sulphate, which precipitates in the lime solution. The specific membrane area reauirements are a function of the desired dearee of acid reduction. The application of -low current densities (500 A/ms) reduces -the specific membrane area requirements for a particular duty by two orders of magnitude compared to facilitated transport. Results of laboratory investigations are given and indicate that the sulphuric acid component can be removed from waste waters without an undesirable build-up of additional dissolved solids. INTRODUCTION Because of its relative cheapness and availability, sulphuric acid is used widely in industry where acidic or anhydrous conditions are required. In addition, the weathering of sulphide bearing rock and the extraction and benefication of sulphide containing ores, leads to the production of sulphuric acid in waste streams by biological and chemical means. The cheap nature of sulphuric acid generally precludes the economic viability of recycling effluents containing the acid. Lime neutralisation is the usual treatment method, resulting in the liquid effluent being super-saturated with calcium sulphate. The interference of calcium sulphate precipitation is a problem in the recovery of high grade water from such wastes by techniques including soda lime softening, electrodialysis and seeded slurry evaporation and hyperfiltration. A process has been developed (ref. 1) by the Pollution Research Group, which enables the neutralisation of sulphuric acid and the removal of sulphates without the addition of further chemicals to the effluent stream. This process is suitable for specialised applications, where it is not economically viable to recover the sulphuric acid, and where the discharge of these effluents, acidic or limed, into the environment is prohibited. The process demineralises the sulphuric acid stream and immobilises the ions, thus preventing their release into the environment. This paper describes the laboratory evaluation of the process for the treatment of: a sulphuric acid containing waste stream from a chemicals and explosives (i) manufacturing plant, the sulphuric acid component being mainly due to the discharge of spent ion-exchange regenerant wastes. Environmental regulations govern the discharge of such an effluent since its release into the water system prejudices the potential for reuse of the OOll-9164/88/$03.50
0 1988 Elsevier Science Publishers B.V.
432 receiving water. At the factory concerned, the sulphuric acid effluent is combined with the effluents from explosive manufacture, limed and then discharged by spray disposal onto vast tracts of land. (ii) acid mine waters, in which the sulphuric acid component results from the oxidation This weathering is promoted in of pyritic minerals in rocks during natural weathering. various areas as a result of mining operations and the resultant mine drainage water becomes more acidic. Restrictive costs and limited availability of fresh water in certain mines, make it necessary to reuse mine waters wherever practical. The recycling of mine waters, however, results in a serious deterioration in the quality of the water, which increases the problems of corrosion, erosion and scale formation, often with severe practical and economic consequences (ref. 2). Corrective action needs to be taken, for example, by improving water quality management, by the prevention of pollution at source and by treating polluted water at source to an adequate quality, appropriate to its particular use (ref. 3). Typical compositions of streams (i) and (ii) are summarised in Table 1. Analyses are given for both the final effluent discharge from the chemical plant by spray disposal (ref. 4) and the acid ion-exchange regeneration stream. The ion-exchange regeneration stream referred to, contains small amounts of effluent from a detonator plant. This mixed stream contributes almost totally to the sulphate concentration of the final effluent, but minimally to the nitrogen concentration. The high calcium content of the final effluent results from liming prior to disposal. Liming destroys residual nitroglycerine in the final effluent. Two analyses are also given for mine drainage water (ref. 3). in both the worked and unworked areas. TABLE 1 Effluent characterisation Parameter
Chemical plant final waste
PH Total dissolved solids Ammonia-N Nitrate-N Sulphate Sodium Calcium Flow (Me/day)
4 to 8 1 550 1 450 2 293 249 807
acidic reg. stream 2.7 4 440 320 430 1 600 380 200 0.2
Mine drainage water worked areas
unworked areas
2.2 3 475
3.0 400
I 850
615
50 0.5
60 1.3
Note : All units in mg/L except pH and flow. PROCESS DESCRIPTION The process requires the use of a membrane stack of the type typically used in the chlor-alkali industry. The membrane is anion selective and separates the two electrolyte compartments. The sulphuric acid waste stream is passed through one compartment and an alkali solution is recirculated through the other compartment. The alkali solution should contain cations capable of forming an insoluble sulphate, calcium hydroxide being an obvious example. Two process options have been developed. These options are :
433 facilitated transport (ii) electrolysis. The option most suited to a particular application will venting facilities at the particular site of installation. electrolysis, hydrogen is evolved. If the process is to neutralisation and demineralisation of mine drainage water, the venting or the catalytic combustion of this gas.
(i)
depend on the availability of gas In the second option, involving be applied underground in the then provision must be made for
Facilitated transoort Fig. 1 shows the first option, facilitated transport, using lime as the alkali solution. The pH differential across the membrane causes anions from the lime solution to pass through the membrane and into the compartment containing the sulphuric acid, thus neutralising the acid. At the same time, charge balance considerations require that the sulphate ions pass in the opposite direction, thus deionising the sulphuric acid steam. The sulphate ions will precipitate with the calcium in the alkali solution and may be continuously removed by filtration. Lime is continuously added to this recirculating liquor to replenish the hydroxide and calcium supplies and to maintain the pH differential across the membrane, thus providing the driving force for the demineralisation and neutralisation of the sulphuric acid stream.
calcium
demineralised neutral
sulphate
solution
M
I
I---+
filtratio
g alkali
I
recycle loop
sulphuric
acid
anion
selective
membrane waste
stream
saturated of lime calcium
Fig. I. Facilitated transport
solution and sulphate
IA -
make-up lime
slurry
434
Where the effluent contains mainly sulphuric acid, neutralisation and almost complete sulphate removal may be achieved. The final pH and sulphate content will be a function of the position and degree of attainment of the equilibrium. Where the effluent contains sulphates, associated with cations other than hydrogen, neutralisation of the effluent will not remove all the sulphates. The higher the pH value of the recirculating alkali the greater will be the possibility of removing the sulphates. The potential exists for the effluent stream to become alkaline. Electrolvsis In the second option, the equilibrium position is displaced by passing on electrical direct current through the cell such that the sulphuric acid compartment is at a negative potential relative to the alkali compartment. The system is illustrated in Fig. 2 for the case in which lime is the alkali.
ti anode -+
cathode -
filtratlo
-
alkali recycle loop
sulphuric waste
acid
stream
anion selective membrane
saturated solution of lime and calcium sulphate
%Ly lime slurry
Fig. 2. Electrolysis The application of the potential across the electrodes results in: the reduction of water at the cathode with the evolution of hydrogen gas 6) 2HzO+2e+20H-+H, (ii)
E”- -0.84V
(1)
the oxidation of water at the anode with the evolution of oxygen gas
H,O-2e+;0,+2H*
E” - +0.99V
(2)
435 the transport of current carrying anions through the membrane from the catholyte (iii) or sulphuric acid stream toward the positively charged anode. The cathodic reaction forms products which assist in the neutralisation of the sulphuric acid electrolyte. Unfortunately the anodic reaction forms products which neutralise the lime, thus necessitating the extra addition of lime. Insoluble sulphate salts, which precipitate after the sulphate ions have migrated into the alkaline solution, are removed by filtration of the alkaline solution. As with the facilitated transport option, the final composition of the effluent will be a function of the purity of the initial sulphuric acid waste and also of the extent of electrolysis. In theory, the passage of each Faraday of electricity will result in: the transfer of a mole equivalent of anions (i.e. 0.5 mole SO& (i) the evolution of 0.5 mole hydrogen gas and 0.25 mole oxygen gas. (ii) the neutralisation of one mole of hydrogen ions in the sulphuric acid waste and (iii) one mole of hydroxide ions in the alkaline solution. In general, the electrode reactions will proceed at 100% current efficiency. Current inefficiencies for sulphate removal will result in the system if: anions, other than the sulphate ion; transport the current through the membrane. (i) This phenomenon is likely to occur only in two instances during the electrolysis of relatively pure sulphuric acid streams: in the cases where the pH of the effluent is allowed to increase above neutral or where polarisation of water occurs at the membrane surface. In both these instances, hydroxide ions, will be present, which will transport part of the current in place of sulphate ions. Polarisation may be prevented by ensuring that the limiting current density of a particular system is not exceeded and that the flux of sulphate ions through the solution to the membrane surface is sufficient. If the sulphuric acid stream is contaminated by other anions, such as nitrates or chlorides, some current will be transported by these other ions. The portion of the current transported by the other anions will be a function of their concentration and mobility and the membrane selectivity. (ii) there is back-migration of sulphate ions from the lime solution to the sulphuric acid stream. This results in the transport of current in the opposite direction to that which is desired. (iii) there is a movement of calcium ions from the lime solution to the sulphuric acid solution. The amount of movement of cations through the membrane is a function of the membrane selectivity. Process drivine force In the facilitated transport option, the driving force for the ion transport is the difference in the chemical potential of the species on either side of the membrane. The driving force enables the sulphate ions, from the acid stream and hydroxide ions, from the lime stream, to The driving force decreases as equilibrium is be redistributed within the electrolytes. approached. In the electrolysis option, the total driving force of the process is the sum of the chemical and electrical driving forces. The chemical and electrical potentials exhibit sympathetic behaviour until equilibrium point is reached and it is possible that observed current efficiencies could be greater than 100%. As the chemical equilibrium is approached and displaced, the chemical and electrical potentials exhibit opposing behaviour. A portion of the applied electrical potential is consumed in overcoming the chemical tendency towards reaction reversal and current efficiencies may be expected to drop.
436 Exnerimental Laboratory evaluation of the process has been conducted in a two compartment batch cell The compartments of the cell were separated by a sheet of operated at ambient temperatures. anion exchange electrodialysis membrane (Ionics 103 QZL 386) with an exposed area of 8.1 x 10-s ms. The electrolytes were stirred continuously to promote turbulence at the membrane surface. The compositions of each electrolyte were monitored for the duration of each test. For the second process option, electrolysis, a potential was applied across two platinised titanium mesh electrodes which were placed on either side of the membrane. Relatively low current densities (60 to 400 A/ms) were employed so as to extend the duration of the test to enable the collection of comprehensive data. RESULTS AND DISCUSSION Typical results for the facilitated transport of acidic ion-exchange mine drainage water and for the electrolysis of acidic ion-exchange given in Tables 2, 3 and 4.
regeneration regeneration
liquor and liquor are
TABLE 2 Facilitated transport of acidic ion-exchange regeneration liquor Alkali solution : initially saturated lime, no make-up Electrolyte volumes : 820 mL Time
Effluent PH
(h) 0.0 0.7 2.5 4.5 7.0 22.5
2.8 3.4 4.4 5.4 7.8 8.8
Lime
Cond mS/cm
so*=
Na+
Ca++
mglt
mg/L
mgll
6.7 6.2 6.0 5.9 5.5 5.1
1 280 1 210 1 140 1 140 1 050 900
322 307 293 307 297 307
278 277 312 266 252 268
PH
Cond mS/cm
12.1 11.9 12.0 11.8 12.0 10.9
SO4=
Na+
mgle
mg/e
20 170 370
9 22 24 24 24 27
7.0 6.3 5.9 5.0 4.2 1.7
TABLE 3 Facilitated transport of mine drainage water, worked areas Alkali solution : Saturated lime and calcium sulphate, continuous lime addition Electrolyte volumes : 1 000 mL Time
Effluent PH
(h) 0.0
2.4
Lime
Cond mS/cm
S04= mg/L
Clmg/e
Na+ mg/L
mglt
7.1
1 580
590
105
525
Ca++
pH
Cond mS/cm
SO4= mg/L
Clmg/e
Na+ mg/L
12.0 11.9
2.4 -
1 450 -
16 -
0 -
12.0 11.9 11.9 12.1 12.1
;:;
1
:
I
2.9 -
I 1 625
I 140
I 0
437 -l-ABLE 4 Electrolysis of acidic ion-exchange regeneration liquor : initially saturated lime, no make-up Alkali solution Electrolyte volumes : 820 mL Current density : 60 A/ms Faraday consumed mF
pH
0 9 19 28 39 50
2.8 8.5 9.6 10.2 11.6 11.7
Incremental
Lime
Acid SO1=
NOs-
Na+
mgll
mgll
mglt
1 320 1180 850 600 580 380
28 43
340 360 -’ 380
pH 12.1 12.0 12.0 11.1 1.8 7.6
SO4=
mgll
2.5 100 500 580 830
10 10
Overall sulphate current efficiency
sulphate current efficiency (96)
Na+
mgP
17 60 45 30 41
(%)
Facilitated transnort Fig. 3 shows the typical change which occurs in the electrolytes, with respect to pH and sulphate concentration, during the facilitated transport of acidic ion-exchange regenerant waste.
1300
900 5 800 !
700
; 600 c 5 500 cn 400
0
4
8
12 Time (h)
Fig. 3. Facilitated transport of acidic ion-exchange
16
regeneration
20
liquor
24
438 Initially, demineralisation and neutralisation of the effluent proceeded rapidly. As equilibrium was approached, the chemical driving force was diminished and the rate at which ions were transported through the membrane was reduced (see Fig. 4). The rate of change in the hydrogen ion concentration of the effluent has been calculated from pH measurements, since the effluent did not possess any buffering capacity. 40 0
30
0 0
20
0 0
2 E
0
lo 0
2: 2 -10
+ +
g-20
+
+ .
.
g -30 9 G ” -40 ‘;; -50 9 ’ -60 cc: -70 -80 0
Time Fig. 4. Facilitated transport of ion-exchange
12 (h)
regeneration
liquor
Facilitated transport of mine drainage water produced similar trends with regard to pH change (Figs. 5 and 6). The rate of change of the sulphate ion concentrations in the electrolytes could not be determined accurately since the total change was very small. During facilitated transport, neutralisation of both effluents occurred with a resulting decrease of 10 to 20% in sulphate concentration. The estimated membrane duty (in C of effluent treated per sq.m of membrane per hour) for the neutralisation of the two effluents to various pH values is given in Fig. 7. The total membrane requirements for a typical effluent flow are very high, and increase rapidly where pH values near or above neutral are required. For example, to neutralise the acidic ion-exchange regeneration liquor and the mine drainage water (worked areas), at the flows specified in Table 1, necessitates 540 and 8 300 ms of anion exchange membrane respectively. Electrolvsis The equilibrium reached during facilitated transport was displaced in the desired direction of sulphate removal by the application of an electric potential across the membrane. Fig. 8 shows the typical change in sulphate concentration and pH for each electrolyte during the electrolysis of acidic ion-exchange regenerant waste.
439
13
12-- +
+
+
+
+
+
ll10 9X a
87n
6-
. .
5-
.
43-
.
. 3 0
10
20
30
Time
40
50
(h)
Fig. 5. Facilitated transport of mine drainage water : Electrolyte pH change 0
.
. n
.
-60
1 0
. 1
I 10
I
I 20 Time
Fig. 6. Facilitated concentration
I
I 30
I
I 40
(h)
transport of mine drainage water : Rate of change in the acid hydrogen ion
440
60
,
.
2
Ei $50
-
> 9 2 40,” al 2 c
30
.
-
x ‘; n 20ae
.
m 2 lo-
.
+
E 2
+ I
0 0 Fig. 7. Estimated
1300
I 2
I
, 4
I 6
/
/ 8
1
I
10
Final effluent pH membrane duty for facilitated transport
*
1200 i’
13 P
&
0
1100-
A 0
1000 s
+
I
900 -
-11 -10
0 .
0
- 9
g 800 -
A
2 700 2 600
.
$ 500 UY400 -
I
+
-
8
i - 7 - 6s” - 5 ,- 4
300 7,
-
200 100 0 0
-12
3
-2
-
+ I 9
1 0
19 Electricity
Fig. 8. Electrolysis of acidic ion-exchange
28 passed
regeneration
(mF) liquor
39
50
441
Approximately 70% of the sulphate was removed from the effluent. The final effluent was alkaline. Low current densities were used in the trials so that the reactions would In larger proceed sufficiently slowly to enable meaningful experimental data to be collected. applications, the applied current may be increased to the limiting value, this value being a function of the flow and the ionic characteristics of the electrolytes. Fig. 9 shows the change in the incremental sulphate current efficiency as a function of the difference in the total hydrogen and sulphate concentrations in the acid and lime solutions. At constant applied electric potential, the incremental current efficiency decreases as the chemical driving force decreases and electrical energy is consumed in opposing the reverse movement toward equilibrium.
-6
-4
-2
0
2
4
6
8
10
acid([hydrogen]+[sulphate])-lime([hydrogen]~[sulphate])
(mmol)
Fig. 9. Electrolysis efficiencies
of acidic ion-exchange
regeneration
liquor : Incremental
sulphate current
The estimated membrane duty (in kL of effluent treated per sq.m of membrane per hour) for the neutralisation and demineralisation of this effluent is shown in Fig. 10. An operating current density of 500 A/ms and an average current efficiency of 60% have been assumed. Thus the application of small potentials across the membrane can substantially increase the membrane duty. In the present example, membrane duty is increased by two orders of magnitude above the membrane duty observed under facilitated transport applications. In order to neutralise the effluent (with a 20% sulphate removal) the anion exchange membrane area requirements are estimated from Fig. 10 to be 5.5 ms. Higher sulphate removal would necessitate larger membrane areas. Membrane area requirements, suitable for typical effluent flows are therefore small in the case of electrolysis.
442 CONCLUSIONS Laboratory results have indicated that the proposed membrane process is effective in removing the sulphuric acid from industrial and mining waste streams containing the acid. The proposed method has the advantage over conventional treatment methods in that the acid stream is neutralised and demineralised without the addition of further salts to the main effluent stream. The sulphate ions are removed from the effluent and immobilised by precipitation as calcium sulphate in a low volume recirculating saturated lime solution. Where suitable facilities exist for handling oxygen and, in particularly hydrogen gas streams, the use of the electrolysis option is advantageous over the facilitated transport option, since membrane area requirements may be reduced by at least two orders of magnitude. FUTURE DEVELOPMENTS A pilot plant study is to be conducted to assess the feasibility of the process to large scale applications and to obtain suitable data for the design of a full scale plant.
0
2
4
Final Effluent
6
pH
a
10
Fig. JO. Estimated membrane duty for electrolysis : (500 A/sq.m and 60% current efficiency) ACKNOWLEDGEMENTS This work was performed patent has been assigned.
under contract to the Water Research Commission,
to whom the
REFERENCES J C.A. Buckley and A.E. Simpson, The removal of sulphuric acid from an aqueous medium containing the acid, South African Patent No. 81/5749, assigned to the Water Research Commission, 4 August 1987. 2. D.J. Bosman, Water quality problems in the mining industry, IWPC Johannesburg, South Africa, Aprd 1987. 3. J.G. Mackay and H.N.S. Withers, Simulation of water quality in underground gold mine service water circuits. Third International Mine Water Congress, Melbourne, Australia, October 1988. 4. J. Bosman, A.A. Eberhard and C.I. Baskir, Denitrification of a concentrated nitrogenous industrial effluent using packed column and fluidised bed reactors, Prog. Wat. Tech., JO (5/6) (1978) 297-308.
Desalination, 70 (1988) 431-442 Elsevier Science Publishers B.V., Amsterdam - Printed in The Netherlands
THE REMOVAL
OF SULPHURIC
ACID FROM NATURAL
AND INDUSTRIAL
WASTE
WATERS
A.E. SIMPSON and C.A. BUCKLEY Pollution Research Group, Department George V Avenue, Durban, 4001.
of Chemical Engineering,
University
of Natal, King
SUMMARY A process is described which enables the removal of sulphuric acid from effluents without the addition of chemicals to the effluent. The technique employs an anion selective membrane which separates the acidic effluent from a lime solution. In the case where no current is applied, facilitated transport occurs and the driving force for the demineralisation process is the difference in the chemical potential of the species on either side of the membrane. The equilibrium position may be shifted by the passage of a current through the membrane, in which instance the driving force becomes the applied electric potential. The sulphates are removed as calcium sulphate, which precipitates in the lime solution. The specific membrane area reauirements are a function of the desired dearee of acid reduction. The application of -low current densities (500 A/ms) reduces -the specific membrane area requirements for a particular duty by two orders of magnitude compared to facilitated transport. Results of laboratory investigations are given and indicate that the sulphuric acid component can be removed from waste waters without an undesirable build-up of additional dissolved solids. INTRODUCTION Because of its relative cheapness and availability, sulphuric acid is used widely in industry where acidic or anhydrous conditions are required. In addition, the weathering of sulphide bearing rock and the extraction and benefication of sulphide containing ores, leads to the production of sulphuric acid in waste streams by biological and chemical means. The cheap nature of sulphuric acid generally precludes the economic viability of recycling effluents containing the acid. Lime neutralisation is the usual treatment method, resulting in the liquid effluent being super-saturated with calcium sulphate. The interference of calcium sulphate precipitation is a problem in the recovery of high grade water from such wastes by techniques including soda lime softening, electrodialysis and seeded slurry evaporation and hyperfiltration. A process has been developed (ref. 1) by the Pollution Research Group, which enables the neutralisation of sulphuric acid and the removal of sulphates without the addition of further chemicals to the effluent stream. This process is suitable for specialised applications, where it is not economically viable to recover the sulphuric acid, and where the discharge of these effluents, acidic or limed, into the environment is prohibited. The process demineralises the sulphuric acid stream and immobilises the ions, thus preventing their release into the environment. This paper describes the laboratory evaluation of the process for the treatment of: a sulphuric acid containing waste stream from a chemicals and explosives (i) manufacturing plant, the sulphuric acid component being mainly due to the discharge of spent ion-exchange regenerant wastes. Environmental regulations govern the discharge of such an effluent since its release into the water system prejudices the potential for reuse of the OOll-9164/88/$03.50
0 1988 Elsevier Science Publishers B.V.
432 receiving water. At the factory concerned, the sulphuric acid effluent is combined with the effluents from explosive manufacture, limed and then discharged by spray disposal onto vast tracts of land. (ii) acid mine waters, in which the sulphuric acid component results from the oxidation This weathering is promoted in of pyritic minerals in rocks during natural weathering. various areas as a result of mining operations and the resultant mine drainage water becomes more acidic. Restrictive costs and limited availability of fresh water in certain mines, make it necessary to reuse mine waters wherever practical. The recycling of mine waters, however, results in a serious deterioration in the quality of the water, which increases the problems of corrosion, erosion and scale formation, often with severe practical and economic consequences (ref. 2). Corrective action needs to be taken, for example, by improving water quality management, by the prevention of pollution at source and by treating polluted water at source to an adequate quality, appropriate to its particular use (ref. 3). Typical compositions of streams (i) and (ii) are summarised in Table 1. Analyses are given for both the final effluent discharge from the chemical plant by spray disposal (ref. 4) and the acid ion-exchange regeneration stream. The ion-exchange regeneration stream referred to, contains small amounts of effluent from a detonator plant. This mixed stream contributes almost totally to the sulphate concentration of the final effluent, but minimally to the nitrogen concentration. The high calcium content of the final effluent results from liming prior to disposal. Liming destroys residual nitroglycerine in the final effluent. Two analyses are also given for mine drainage water (ref. 3). in both the worked and unworked areas. TABLE 1 Effluent characterisation Parameter
Chemical plant final waste
PH Total dissolved solids Ammonia-N Nitrate-N Sulphate Sodium Calcium Flow (Me/day)
4 to 8 1 550 1 450 2 293 249 807
acidic reg. stream 2.7 4 440 320 430 1 600 380 200 0.2
Mine drainage water worked areas
unworked areas
2.2 3 475
3.0 400
I 850
615
50 0.5
60 1.3
Note : All units in mg/L except pH and flow. PROCESS DESCRIPTION The process requires the use of a membrane stack of the type typically used in the chlor-alkali industry. The membrane is anion selective and separates the two electrolyte compartments. The sulphuric acid waste stream is passed through one compartment and an alkali solution is recirculated through the other compartment. The alkali solution should contain cations capable of forming an insoluble sulphate, calcium hydroxide being an obvious example. Two process options have been developed. These options are :
433 facilitated transport (ii) electrolysis. The option most suited to a particular application will venting facilities at the particular site of installation. electrolysis, hydrogen is evolved. If the process is to neutralisation and demineralisation of mine drainage water, the venting or the catalytic combustion of this gas.
(i)
depend on the availability of gas In the second option, involving be applied underground in the then provision must be made for
Facilitated transoort Fig. 1 shows the first option, facilitated transport, using lime as the alkali solution. The pH differential across the membrane causes anions from the lime solution to pass through the membrane and into the compartment containing the sulphuric acid, thus neutralising the acid. At the same time, charge balance considerations require that the sulphate ions pass in the opposite direction, thus deionising the sulphuric acid steam. The sulphate ions will precipitate with the calcium in the alkali solution and may be continuously removed by filtration. Lime is continuously added to this recirculating liquor to replenish the hydroxide and calcium supplies and to maintain the pH differential across the membrane, thus providing the driving force for the demineralisation and neutralisation of the sulphuric acid stream.
calcium
demineralised neutral
sulphate
solution
M
I
I---+
filtratio
g alkali
I
recycle loop
sulphuric
acid
anion
selective
membrane waste
stream
saturated of lime calcium
Fig. I. Facilitated transport
solution and sulphate
IA -
make-up lime
slurry
434
Where the effluent contains mainly sulphuric acid, neutralisation and almost complete sulphate removal may be achieved. The final pH and sulphate content will be a function of the position and degree of attainment of the equilibrium. Where the effluent contains sulphates, associated with cations other than hydrogen, neutralisation of the effluent will not remove all the sulphates. The higher the pH value of the recirculating alkali the greater will be the possibility of removing the sulphates. The potential exists for the effluent stream to become alkaline. Electrolvsis In the second option, the equilibrium position is displaced by passing on electrical direct current through the cell such that the sulphuric acid compartment is at a negative potential relative to the alkali compartment. The system is illustrated in Fig. 2 for the case in which lime is the alkali.
ti anode -+
cathode -
filtratlo
-
alkali recycle loop
sulphuric waste
acid
stream
anion selective membrane
saturated solution of lime and calcium sulphate
%Ly lime slurry
Fig. 2. Electrolysis The application of the potential across the electrodes results in: the reduction of water at the cathode with the evolution of hydrogen gas 6) 2HzO+2e+20H-+H, (ii)
E”- -0.84V
(1)
the oxidation of water at the anode with the evolution of oxygen gas
H,O-2e+;0,+2H*
E” - +0.99V
(2)
435 the transport of current carrying anions through the membrane from the catholyte (iii) or sulphuric acid stream toward the positively charged anode. The cathodic reaction forms products which assist in the neutralisation of the sulphuric acid electrolyte. Unfortunately the anodic reaction forms products which neutralise the lime, thus necessitating the extra addition of lime. Insoluble sulphate salts, which precipitate after the sulphate ions have migrated into the alkaline solution, are removed by filtration of the alkaline solution. As with the facilitated transport option, the final composition of the effluent will be a function of the purity of the initial sulphuric acid waste and also of the extent of electrolysis. In theory, the passage of each Faraday of electricity will result in: the transfer of a mole equivalent of anions (i.e. 0.5 mole SO& (i) the evolution of 0.5 mole hydrogen gas and 0.25 mole oxygen gas. (ii) the neutralisation of one mole of hydrogen ions in the sulphuric acid waste and (iii) one mole of hydroxide ions in the alkaline solution. In general, the electrode reactions will proceed at 100% current efficiency. Current inefficiencies for sulphate removal will result in the system if: anions, other than the sulphate ion; transport the current through the membrane. (i) This phenomenon is likely to occur only in two instances during the electrolysis of relatively pure sulphuric acid streams: in the cases where the pH of the effluent is allowed to increase above neutral or where polarisation of water occurs at the membrane surface. In both these instances, hydroxide ions, will be present, which will transport part of the current in place of sulphate ions. Polarisation may be prevented by ensuring that the limiting current density of a particular system is not exceeded and that the flux of sulphate ions through the solution to the membrane surface is sufficient. If the sulphuric acid stream is contaminated by other anions, such as nitrates or chlorides, some current will be transported by these other ions. The portion of the current transported by the other anions will be a function of their concentration and mobility and the membrane selectivity. (ii) there is back-migration of sulphate ions from the lime solution to the sulphuric acid stream. This results in the transport of current in the opposite direction to that which is desired. (iii) there is a movement of calcium ions from the lime solution to the sulphuric acid solution. The amount of movement of cations through the membrane is a function of the membrane selectivity. Process drivine force In the facilitated transport option, the driving force for the ion transport is the difference in the chemical potential of the species on either side of the membrane. The driving force enables the sulphate ions, from the acid stream and hydroxide ions, from the lime stream, to The driving force decreases as equilibrium is be redistributed within the electrolytes. approached. In the electrolysis option, the total driving force of the process is the sum of the chemical and electrical driving forces. The chemical and electrical potentials exhibit sympathetic behaviour until equilibrium point is reached and it is possible that observed current efficiencies could be greater than 100%. As the chemical equilibrium is approached and displaced, the chemical and electrical potentials exhibit opposing behaviour. A portion of the applied electrical potential is consumed in overcoming the chemical tendency towards reaction reversal and current efficiencies may be expected to drop.
436 Exnerimental Laboratory evaluation of the process has been conducted in a two compartment batch cell The compartments of the cell were separated by a sheet of operated at ambient temperatures. anion exchange electrodialysis membrane (Ionics 103 QZL 386) with an exposed area of 8.1 x 10-s ms. The electrolytes were stirred continuously to promote turbulence at the membrane surface. The compositions of each electrolyte were monitored for the duration of each test. For the second process option, electrolysis, a potential was applied across two platinised titanium mesh electrodes which were placed on either side of the membrane. Relatively low current densities (60 to 400 A/ms) were employed so as to extend the duration of the test to enable the collection of comprehensive data. RESULTS AND DISCUSSION Typical results for the facilitated transport of acidic ion-exchange mine drainage water and for the electrolysis of acidic ion-exchange given in Tables 2, 3 and 4.
regeneration regeneration
liquor and liquor are
TABLE 2 Facilitated transport of acidic ion-exchange regeneration liquor Alkali solution : initially saturated lime, no make-up Electrolyte volumes : 820 mL Time
Effluent PH
(h) 0.0 0.7 2.5 4.5 7.0 22.5
2.8 3.4 4.4 5.4 7.8 8.8
Lime
Cond mS/cm
so*=
Na+
Ca++
mglt
mg/L
mgll
6.7 6.2 6.0 5.9 5.5 5.1
1 280 1 210 1 140 1 140 1 050 900
322 307 293 307 297 307
278 277 312 266 252 268
PH
Cond mS/cm
12.1 11.9 12.0 11.8 12.0 10.9
SO4=
Na+
mgle
mg/e
20 170 370
9 22 24 24 24 27
7.0 6.3 5.9 5.0 4.2 1.7
TABLE 3 Facilitated transport of mine drainage water, worked areas Alkali solution : Saturated lime and calcium sulphate, continuous lime addition Electrolyte volumes : 1 000 mL Time
Effluent PH
(h) 0.0
2.4
Lime
Cond mS/cm
S04= mg/L
Clmg/e
Na+ mg/L
mglt
7.1
1 580
590
105
525
Ca++
pH
Cond mS/cm
SO4= mg/L
Clmg/e
Na+ mg/L
12.0 11.9
2.4 -
1 450 -
16 -
0 -
12.0 11.9 11.9 12.1 12.1
;:;
1
:
I
2.9 -
I 1 625
I 140
I 0
437 -l-ABLE 4 Electrolysis of acidic ion-exchange regeneration liquor : initially saturated lime, no make-up Alkali solution Electrolyte volumes : 820 mL Current density : 60 A/ms Faraday consumed mF
pH
0 9 19 28 39 50
2.8 8.5 9.6 10.2 11.6 11.7
Incremental
Lime
Acid SO1=
NOs-
Na+
mgll
mgll
mglt
1 320 1180 850 600 580 380
28 43
340 360 -’ 380
pH 12.1 12.0 12.0 11.1 1.8 7.6
SO4=
mgll
2.5 100 500 580 830
10 10
Overall sulphate current efficiency
sulphate current efficiency (96)
Na+
mgP
17 60 45 30 41
(%)
Facilitated transnort Fig. 3 shows the typical change which occurs in the electrolytes, with respect to pH and sulphate concentration, during the facilitated transport of acidic ion-exchange regenerant waste.
1300
900 5 800 !
700
; 600 c 5 500 cn 400
0
4
8
12 Time (h)
Fig. 3. Facilitated transport of acidic ion-exchange
16
regeneration
20
liquor
24
438 Initially, demineralisation and neutralisation of the effluent proceeded rapidly. As equilibrium was approached, the chemical driving force was diminished and the rate at which ions were transported through the membrane was reduced (see Fig. 4). The rate of change in the hydrogen ion concentration of the effluent has been calculated from pH measurements, since the effluent did not possess any buffering capacity. 40 0
30
0 0
20
0 0
2 E
0
lo 0
2: 2 -10
+ +
g-20
+
+ .
.
g -30 9 G ” -40 ‘;; -50 9 ’ -60 cc: -70 -80 0
Time Fig. 4. Facilitated transport of ion-exchange
12 (h)
regeneration
liquor
Facilitated transport of mine drainage water produced similar trends with regard to pH change (Figs. 5 and 6). The rate of change of the sulphate ion concentrations in the electrolytes could not be determined accurately since the total change was very small. During facilitated transport, neutralisation of both effluents occurred with a resulting decrease of 10 to 20% in sulphate concentration. The estimated membrane duty (in C of effluent treated per sq.m of membrane per hour) for the neutralisation of the two effluents to various pH values is given in Fig. 7. The total membrane requirements for a typical effluent flow are very high, and increase rapidly where pH values near or above neutral are required. For example, to neutralise the acidic ion-exchange regeneration liquor and the mine drainage water (worked areas), at the flows specified in Table 1, necessitates 540 and 8 300 ms of anion exchange membrane respectively. Electrolvsis The equilibrium reached during facilitated transport was displaced in the desired direction of sulphate removal by the application of an electric potential across the membrane. Fig. 8 shows the typical change in sulphate concentration and pH for each electrolyte during the electrolysis of acidic ion-exchange regenerant waste.
439
13
12-- +
+
+
+
+
+
ll10 9X a
87n
6-
. .
5-
.
43-
.
. 3 0
10
20
30
Time
40
50
(h)
Fig. 5. Facilitated transport of mine drainage water : Electrolyte pH change 0
.
. n
.
-60
1 0
. 1
I 10
I
I 20 Time
Fig. 6. Facilitated concentration
I
I 30
I
I 40
(h)
transport of mine drainage water : Rate of change in the acid hydrogen ion
440
60
,
.
2
Ei $50
-
> 9 2 40,” al 2 c
30
.
-
x ‘; n 20ae
.
m 2 lo-
.
+
E 2
+ I
0 0 Fig. 7. Estimated
1300
I 2
I
, 4
I 6
/
/ 8
1
I
10
Final effluent pH membrane duty for facilitated transport
*
1200 i’
13 P
&
0
1100-
A 0
1000 s
+
I
900 -
-11 -10
0 .
0
- 9
g 800 -
A
2 700 2 600
.
$ 500 UY400 -
I
+
-
8
i - 7 - 6s” - 5 ,- 4
300 7,
-
200 100 0 0
-12
3
-2
-
+ I 9
1 0
19 Electricity
Fig. 8. Electrolysis of acidic ion-exchange
28 passed
regeneration
(mF) liquor
39
50
441
Approximately 70% of the sulphate was removed from the effluent. The final effluent was alkaline. Low current densities were used in the trials so that the reactions would In larger proceed sufficiently slowly to enable meaningful experimental data to be collected. applications, the applied current may be increased to the limiting value, this value being a function of the flow and the ionic characteristics of the electrolytes. Fig. 9 shows the change in the incremental sulphate current efficiency as a function of the difference in the total hydrogen and sulphate concentrations in the acid and lime solutions. At constant applied electric potential, the incremental current efficiency decreases as the chemical driving force decreases and electrical energy is consumed in opposing the reverse movement toward equilibrium.
-6
-4
-2
0
2
4
6
8
10
acid([hydrogen]+[sulphate])-lime([hydrogen]~[sulphate])
(mmol)
Fig. 9. Electrolysis efficiencies
of acidic ion-exchange
regeneration
liquor : Incremental
sulphate current
The estimated membrane duty (in kL of effluent treated per sq.m of membrane per hour) for the neutralisation and demineralisation of this effluent is shown in Fig. 10. An operating current density of 500 A/ms and an average current efficiency of 60% have been assumed. Thus the application of small potentials across the membrane can substantially increase the membrane duty. In the present example, membrane duty is increased by two orders of magnitude above the membrane duty observed under facilitated transport applications. In order to neutralise the effluent (with a 20% sulphate removal) the anion exchange membrane area requirements are estimated from Fig. 10 to be 5.5 ms. Higher sulphate removal would necessitate larger membrane areas. Membrane area requirements, suitable for typical effluent flows are therefore small in the case of electrolysis.
442 CONCLUSIONS Laboratory results have indicated that the proposed membrane process is effective in removing the sulphuric acid from industrial and mining waste streams containing the acid. The proposed method has the advantage over conventional treatment methods in that the acid stream is neutralised and demineralised without the addition of further salts to the main effluent stream. The sulphate ions are removed from the effluent and immobilised by precipitation as calcium sulphate in a low volume recirculating saturated lime solution. Where suitable facilities exist for handling oxygen and, in particularly hydrogen gas streams, the use of the electrolysis option is advantageous over the facilitated transport option, since membrane area requirements may be reduced by at least two orders of magnitude. FUTURE DEVELOPMENTS A pilot plant study is to be conducted to assess the feasibility of the process to large scale applications and to obtain suitable data for the design of a full scale plant.
0
2
4
Final Effluent
6
pH
a
10
Fig. JO. Estimated membrane duty for electrolysis : (500 A/sq.m and 60% current efficiency) ACKNOWLEDGEMENTS This work was performed patent has been assigned.
under contract to the Water Research Commission,
to whom the
REFERENCES J C.A. Buckley and A.E. Simpson, The removal of sulphuric acid from an aqueous medium containing the acid, South African Patent No. 81/5749, assigned to the Water Research Commission, 4 August 1987. 2. D.J. Bosman, Water quality problems in the mining industry, IWPC Johannesburg, South Africa, Aprd 1987. 3. J.G. Mackay and H.N.S. Withers, Simulation of water quality in underground gold mine service water circuits. Third International Mine Water Congress, Melbourne, Australia, October 1988. 4. J. Bosman, A.A. Eberhard and C.I. Baskir, Denitrification of a concentrated nitrogenous industrial effluent using packed column and fluidised bed reactors, Prog. Wat. Tech., JO (5/6) (1978) 297-308.