In__r

PRO[E55Inri~~~L.~EVIE;

Int. J. Miner. Process. 41 (1994) 215-225

Role of pH and dissolved mineral species inPittsburgh No.8 coal flotation system -

. Separation of pyrite and non-pyritic mineralsfrom coal

I

D. Liu, P. Somasundaran*, T.V. Vasudevan, C.C. HarrisHenry Krumb School of Mines, Columbia University. New York, NY 10027, USA

(Received 15 May 1992; accepted after revision 21 October 1993)

The role of pH and of dissolved mineral species on the separation of pyrite and non-pyritic minerals from Pittsburgh No.8 coal by flotation was investigated. In the pH rangeof 4 to 8, enhanced pyrite rejection was obtained at higher pH. However, pH itself had nomeasurable effect on the separation of non-pyritic minerals from coal in the pH range of 4to 10. The precipitation/adsorption of the dissolved mineral ions resulted in a measurabledecrease in the selectivity for non-pyritic minerals separation. Increased pyrite rejectionwith pH increase was shown, by induction time measurements, to be mainly due to thedecrease of the pyrite hydrophobicity. The adsorption ofCa ions in the coal flotation sys-tems had a positive effect on the pyrite separation, although it did not affect coal recoveryto a measurable extent. This could be due to the selective precipitation/adsorption of Caspecies on the pyrite surface. The precipitation/adsorption of Fe $pecies had no measura-ble effect on the selectivity, even though it depressed coal flotation considerably. This isattributed to the non-selective precipitation of Fe species between the surfaces of coal and

pyrite.

1. Introduction

Use of high sulfur coals as an environmentally acceptable fuel requires nearcomplete removal of pyritic sulfur. Froth flotation that exploits the surface prop-erties of minerals for separation is a promising technique for efficient and eco---*To whom correspondence should be addressed.

0301-7516/94/$07.00 @ 1994ElsevierScienceB.V. AllrighureservedSSDI 0301- 7516(93) EO064-R

iu et aI. / Int. J. Miner. Process. 41 (1994) 115-225

nomic removal of pyrite and other ash forming minerals from coal. However, theflotation process currently used in coal preparation plants is a relatively crudeprocess (Osborne, 1988), not being capable of rejecting pyrite to the desirablelevels. The separation efficiency of pyrite from coal by flotation is basically de-termined by the relative hydrophobicities of the coal and the pyrite. To enhancethe separation, the surfaces need to be modified appropriately so that the coalsurface is more hydrophobic and/or the pyrite surface more hydrophilic. Exten-si ve research work aimed to improve the efficiency of pyrite separation has beendone during the past several decades. A number of organic and inorganic reagentshave been reported to be good surface modifiers for either coal or pyrite (Chou-dhry and Aplan, 1992; Klimpel, 1992), but the separation efficiency has notreached to the desirable levels.

pH is a parameter commonly used in the study of the surface characteristics ofcoal and the relative minerals during flotation. The floatability of coal has beenfound to depend on both the pH and the precipitation of dissolved inorganicspecies Celik and Somasundaran, 1986; Liu et al., 1994a). Also, the natural hy-drophobicity of pyrite has been shown to be very much dependent on pH Koca-bag et al., 1990). Zimmerman's work (1948) has demonstrated the significanteffect of pH on the separation of pyrite from coal. However, earlier investigationsinvolved the study of the effect of only added salts, and the effect of inorganicspecies dissolved from the minerals present in coal was not studied in detail. Thiseffect is important since all the high sulfur coals contain high percentage of ash-forming minerals also. Further, in the earlier studies the mechanism of the effectof pH and of its interaction with dissolved mineral species on the separation wasnot elucidated.

In the present study, the role of pH and the dissolved mineral species on theflotation separation of pyrite and other ash forming minerals from a bituminouscoal sample was investigated. The experimental procedure allowed the effect ofpH to be isolated from that of the precipitation of dissolved mineral species.Mechanisms of the effect of interactions between pH and the dissolved mineralspecies on the separation of pyrite and non'-pyritic minerals from coal arediscussed.

2. Experimental

Coal used in this study is a Pittsburgh No.8 sample supplied by R and F CoalCo., Ohio. The 2-4 inch sample supplied was crushed to a nominal 1/4 inch sizein a laboratory roll crusher under argon atmosphere and the product was storedunder argon. The flotation feed was prepared by wet grinding (either in distiltedwater or in caustic solution) the 1/4 inch sample to 95% passing 200 mesh in arod milt. Details of the procedures used for the preparation of feed, batch flota-tion tests and dissolved species analysis are given in Part I (Liuet aI., 1994a).Coal-pyrite was obtained from a sample rejected from Pocahontas No.3 coal byDeister tables, supplied by Island Creek Coal Company, Oakwood, VA. The coal

;

,~:

D. Liu et al. / Int. J. Miner. Process. 41 (1994) 115-215

refuse supplied has a high content of pyrite and was concentrated further bywashing, sieving, shaking, and magnetic separation. The 60x 100 mesh size frac-tion of the fmal concentrate contained about 90% by weight of pyrite. This sam-ple was stored under argon atmosphere in a freezer. Before use, the stored samplewas washed three times with triply distilled water (TOW) to remove the sus-pected oxidation products and was ground in TOW. A desirable size fraction ob-tained from the ground sample was used as coal-pyrite sample for zeta potentialand induction time measurements.

Hydrophobicity of coal and coal-pyrite were determined by measuring theirinduction times using an APT-I 00 type induction timer. The important deviceparameters that could have affected the measured induction time were the volt-age which controlled the speed of vertical movement, bubble size (fixed at about1.5 mm diameter), the distance travelled by the bubble (0.2 to 0.3 mm), and thedistance between the bubble and the bed of particles (about 0.1 mm). The "in-duction time" was taken as the contact time for which fifty percent of the contactswere fruitful. A minimum of fifteen measurements (contacts) for each selectedcontact time was attempted.

The 100 X 140 mesh fractions of the samples of coal, coal-pyrite were used inthe induction time measurements. The samples were first washed three times withTOW and then conditioned for ten minutes in a solution at a solid concentrationof 10 %. pH was controlled throughout the conditioning time. The conditionedslurry was transferred to a special sample cell for induction time measurements.

Zeta potential of coal and coal-pyrite particles was determined by microelec-trophoresis technique using a Pen Kem 501 Lazer Zee Meter. Wet ground minus200 mesh coal samples and freshly crushed minus 400 mesh coal-pyrite sampleswere conditioned for ten minutes in either triply distilled water or coal superna-tant at a solid concentration of 6.5 % by weight. pH was adjusted using the sameprocedure as for flotation tests. A fraction containing about 0.05% by weight ofsolids of the conditioned slurry was used for zeta potential measurements. In ad-dition to the zeta potential measurements, the electrolyte conductivity of the slurrywas also determined using the same equipment. The latter was used in the eval-uation of the effect of ionic strength on zeta potential behavior of particles.

3. Results and discussion

In this paper, the results of the flotation tests are presented in the form of selec-tivity curves, which are plots of the percent pyrite or non-pyritic minerals rejec-tion versus the percent coal substance recovery. Increase in coal substance recov-ery along the selectivity curve is obtained by increasing the frother dosage, whilemaintaining the collector dosage and other parameters, such as the pH, the pointof addition of sodium hydroxide and the type of coal (washed or unwashed) thesame. Increased selecti vity is indicated by the shift of the curve towards the upperright hand comer.

3.1. Effect of pH and dissolved ions on pyrite separation

The effect of pH on the efficiency of separation of pyrite from coal under de-creasing pH (non-precipitation) and increasing pH (precipitation) conditions isillustrated in Figs. I and 2, respectively. It can be seen that between pH 8 and 10(or 9.2 under non-precipitation conditions), there is no measurable effect of pHon the selectivity. Between pH 4 and 8, on the other hand, enhanced selectivity isobtained at higher pH values under both the precipitation and the non-precipi-tation conditions. Comparison of the selectivities obtained under the two condi-tions suggests that for a given coal recovery a slightly higher pyrite rejection canbe obtained under the non-precipitation conditions when alkali is added in to themill (Fig. 3).

The above results suggest that both pH and the precipitation/adsorption ofdissolved ions has an effect on the pyrite separation. The favorable effect of pHincrease on the selectivity could, at the moment, be attributed to the chemicaleffects of hydroxyl ions on the hydrophobicity of coal and pyrite and will be dis-cussed later. The effect of precipitation/adsorption of dissolved ions on the selec-tivity is quite complex in the natural coal-minerals system, although Fig. 3 showsthat the selectivity obtained under the non-precipitation conditions is higher thanthat obtained under the precipitation conditions. The results presented in thisfigure suggest that both the precipitation/adsorption of dissolved ions and thedifferent pH conditioning times (10 minutes for precipitation conditions, about100 minutes for non-precipitation conditions) could have an effect on theselectivity.

90

-~~~

,r:~- . ;, i!-,'"

M

Z0t=Co)&oJ-.&oJa:

a::)i-.-':)II)

Co).:::~>-~

80

75

70

65~

0 pH 4.0. pH 4.aV pH 5.2. pH 7.50 pH 8.0. pfI 9.2

NaOH added in mill0.9 g/kg dodKone0.07 to 0.54 g/kg MIeC

60 . , . .40 50 60 70 80

COAL RECOVERY. ~Fig. I. Effect of pH on the separation of pyrite from coal under decreasing pH conditions (additionof sodium hydroxide in the mill).

. ,

90 100

D. LiuftaL /lnLJ. MiMr. ~ 41(1994) 215-125 219

~

z0~(.)'"','"'~

~~"--I~(f)

(.)~«>-a..

COAL RECOVERY. ~

rig. 2. Effect of pH on the separation of pyrite from coal under increasing pH conditions (addition ofMl<iium bydroxide in the ceO ).

~

z0~UL6I

L6I

«

~

~

Lo.

-I

~

~

u

i=

~

~

Fig. 3. Comparison of pyrite rejections obtained under decreasing pH and increasing pH conditions.

220 D. Lill ft III 1 lilt. J. A(iMr. Procas. 41 (1994) 115-215

3.2. Effect of pH and dissolved ions on non-pyritic mineral separation

Selectivity curves obtained for the separation of non-pyritic minerals from coalunder decreasing pH (non-precipitation) and increasing pH (precipitation) con-ditions are shown in Figs. 4 and 5, respectively. It can be seen that under the non-precipitation conditions there is no effect of pH on selectivity. However, underthe precipitation conditions, selectivity decreases with pH increase in the rangeof pH 4 to 8. This effect is opposite to that observed for selectivity in terms ofpyrite rejection.

These results suggest that pH itself has no measurable effect on the selectivityfor non-pyritic minerals separation and that the precipitation/adsorption of dis-solved ions has a negative effect on the selectivity (Fig. 5). This was attributedto the decrease in the hydrophobicity of the coal due to both the adsorption ofthe precipitates of metal-hydroxides and the precipitation of the dissolved ionson the coal surfaces, as discussed in Part I.

3.3. Electrokinetic behayior of pyrite

Zeta potential behavior of the coal-pyrite in triply distilled water and in coalsupernatant under both increasing pH and decreasing pH conditions is shown asa function of pH in Fig. 6. In the pH range of 5 to 1O, the zeta potential of coal-pyrite measured in triply distilled water is the most negative followed by that inthe coal supernatant under decreasing pH conditions. The zeta potential of pyritemeasured in the coal supernatant under increasing pH conditions is the least neg-ative. Zeta potential of the precipitate of the coal supernatant (obtained in the

85 r

~

z0j::(.JLoJ-.LoJ~

x:V)-c

(.Jj::~>-Q.I

Z0Z

80

75

70

15

60

55~O _J 60 70 80 90 100

COAL RECOVERY. %

Fig. 4. Effect of pH on the separation of non-pyritic minerals from Q)8) under d«reQSing pH conditions.

50

D. Liu et aL / Int. J. MinD'. Pr«ess. 41 (1994) 115-115 221

~

z0~U&01-.&01~

:I:II)-c

u~~>-a-t

z0z

rJ&, s. Effect of pH and precipitation on the separation of non-pyritic minerals from coal under in-creasing pH conditions.

20 r ." '. . .'; .riF..~~,. ,'1

10

0

.10 ~--ct- -20wN

-30

.10 .. .2

r

rJl. 6. Zeta potential of coal-pyrite as a function of pH under various conditions

absence of solids), which is positive at all pH values studied, is also shown in thesame figure. Similar to the zeta potential behavior of coal (Fig. S in Part I), theless negative surface of pyrite under the increasing pH conditions can be attrib-uted to the precipitation/adsorption of the species of Fe, Ca and Mg.

>E-1..c~zwt-oGo

Ionic strength; 0.03 kmol/mJ

0 pyrite in diltined weier - Increoslng pH. 0 pyrite in cool s--t - in_ling pH

. PyrIIe in ~I ~ - Decreasing pHA PrKlpllote from coal ~t

-40 . . . . "

23~56789 !

EQUIUBRIUM oH

222 D. Liu et al. / Int. J. Miner. Process. 41 (1994) 215-225

The results of zeta potential measurements for pyrite and coal (shown in Fig.9 and Fig. 5 in Part I, respectively) indicate that coating of the negative Particlesurfaces with the positively charged precipitates and/or the metal ions or theirrelated ion hydroxide complexes which are expected to occur under the precipi-tation conditions is not selective between the surface of coal and that of pyrite.Therefore, the reduction of the selectivity for pyrite separation obtained underprecipitation conditions in comparison to that for non-precipitation conditions(Fig. 3) can be attributed to the effect of pH conditioning time, and not to theselective precipitation/adsorption of the dissolved ions on the surfaces of coalover those of pyrite. This hypothesis is examined below.

3.4. Isolation of effects of pH and different ions on the separation ofpyrlteandnon-pyritic minerals from coal

Flotation tests to determine the effects of pH and the pH conditioning time onpyrite separation were conducted using washed coal samples to eliminate the ef-fect of dissolved ions, and the results are shown in Fig. 7. As expected, both pHand the pH conditioning time show an effect on the selectivity. Increasing the pHfrom 4 to 8 results in a substantial increase in pyrite rejection while increase inthe pH conditioning time from 10 minutes to 60 minutes increases the effective-ness of the pH modification on pyrite separation. This time effect will be furtherdiscussed later. Fig. 8 shows the results of the effects of dissolved Fe2+ and Ca2+ions on pyrite separation for washed coal under increasing pH conditions (at pH8). It can be seen that calcium chloride addition or Ca2+ ion adsorption results

~

z0i=(.)&oJ-.&oJ~

~~'"--'~III

(.)i=~>-Q..

40 50 60 70 80 90 100

COAL RECOVERY. %Fig. 7. Effects of pH and pH conditioning time on the separation of pyrite from coal obtained forwashed coal under increasing pH conditions.

D. LiliftaL / Int. J. Minlr.~. 41 (1994)115-115 223

.-.'t. . .

0.9 gjkg dodecan.. 0.13 gjkg Mlac

z0~uw-.w~

~~

~~If)

u~~>-a..

..

JtOO

1+ 1+. Me/*- r. nor Ca/ J 1+0 O.OO~ kg alamo M r.

J 1+0 0.0024 kg a-./W CaWQ8h8d Goal.

reCIl ar CaCtI added M eel

NaOH added In ceH~ of pit . ,« 10 -L- .. . .

50 60 70 80

COAL RECOVERY. ~

-L-

to40

Fig. 8. Effects of ferrous chloride and calcium chloride addition on the separation of pyrite fromwashed ooaI obtained at pH 8 and under increasing pH conditions.

in a positive effect on the selectivity, although it does not affect coal recovery toa measurable extent in the Ca2+ concentration range tested. Addition of ferrouschloride, or precipitation/ adsorption of dissolved Fe ions, does not affect theselectivity, even though it does depress coal flotation significantly.

3.5. Effect of pH on the hydrophobicity of coal and pyrite

Induction time measurements were carried out to determine the effect of pHon coal and pyrite hydrophobicities. and the results are shown in Fig. 9. In thecase of coal, the induction time (which is inversely related to hydrophobicity)decreases slightly with pH increase in the pH range of 4 to 7. Above pH 8, thetrend is reversed with the induction time increasing measurably with pH in-crease. Decrease in coal hydrophobicity above pH 8 can be attributed to the ion-ization of its surface carboxylic groups (Hamieh and Siffert, 1991).

In the case of pyrite, the induction time increases with pH throughout the en-tire pH range tested (4 to 10) with the steepest increase occurring between pH 6and 8. Thus, the results of induction time measurements (Fig. 9) suggest that thedecrease in pyrite hydrophobicity with pH increase can contribute significantlyto the improved pyrite rejection with such pH increase, in the pH range of 4 to 8.The sharp increase in the hydrophobicity of pyrite around pH 6 is attributed to--..~te hydrophobicity was measured in the supernatant of the flotation feed to attempt to maintainStmilar solution condition~

D. Liu n aJ. / Int. J. Miner. Proce.u. 41 (1994) 115-215

."0C0u..

S'Ew~~%0~uj0~

pH

FI&- 9. Effect of pH on indUction times of coal and pyrite particles measured in the supernatant of theflotation feed under decreasing pH conditions.

the change of pyrite surface species from partially hydrophobic FeS2 to hydro-philic Fe (OH) 3 with pH increase, as discussed elsewhere (Liu et aI., 1994b).

4. Conclusions

2

3.

Both pH and precipitation/adsorption of the dissolved mineral ions werefound to have an effect on the pyrite separation.Increase of the pH, from 4 to 8, resulted in enhanced pyrite rejection. Thiswas shown, by induction time measurements, to be mainly due to a decreaseof pyrite hydrophobicity with pH increase.The adsorption of Ca ions in the coal flotation systems showed a positiveeffect on the pyrite separation, although it did not affect coal recovery to ameasurable extent. This could be due to the selective precipitation/ adsorp-tion of Ca species on the pyrite surface. The precipitation/adsorption of Fespecies had no measurable effect on the selectivity, even though it depressedcoal flotation considerably. This is attributed to the non-selective precipita-tion of Fe species between the surfaces of coal and pyrite.pH itself has no measurable effect on the separation of non-pyritic mineralsfrom Pittsburgh No.8 coal in the pH range of 4 to 10. The precipitation!adsorption of the dissolved mineral ions results in a measurable decrease inthe selectivity for non-pyritic minerals separation.

4

D. Liu et aL / Int. J. Miner. Process. 41 (1994) 215-225 22S

Acknowledgement

The authors acknowledge the United States Department of Energy for the fi-nancial support (DE.AC22-88PC88878) and Prof. D. W. Fuerstenau ofUniver-sity of California at Berkeley for his helpful suggestions.

References

Celik, M.S. and Somasundaran, P., 1986. The effect of multivalent ions on the floatation of coal. Scp.Sci. Technol., 21 (10): 393-402.

Choudhry, V. and Aplan, F.F., 1992. Pyrite depression during coal flotation Part I - inorganic ions.Miner. Metall. Process., 9(5): 51-56.

Hamieh, T. and Siffert, B., 1991. Determination of point of zero charge and acid-base superficial coalgroups in water. Colloids Surf., 61: 83-96.

Klimpel, R.R., 1992. Some results of various new chemical reagents for modifying coal flotation per-fonnance. Coal Prep., 10: 159-175.

Kocabag, D., Shergold, H.L and Kelsall, G.H., 1990. Natural oleophilicity /hydrophobicity of sul-phide minerals, II. Pyrite. Int. J. Miner. Process.,. 29: 211-219.

Liu, D., Somasundaran, P., Vasudevan, T. V. and Harris, C.C., 1994a. Role of pH and dissolved min-eral species in Pittsburgh No.8 coal flotation system - 1. The floatability of coal. Int. J. Miner.Process., 41: 201-214.

Liu, D., Somasundaran, P. and Duby, P.F., 1994b. Electrochemical equilibria in coal flotation sys-tems and their role in determining the interfacial properties of pyrite and its separation from coal.In: B.K. Parekh (Editor), Proc. 5st Int. Conf. on Processing and Utilization of High Sulfur Coals.Elsevier, New York, pp. 47-53.

Osborne, D.G., 1988. Coal Preparation Technology. Graham and Trotman, London, Vol. I.Zimmennann, R.E., 1948. Flotation of bituminous coal Trans. AIME, 117: 338-356.