Semi-continuous recovery of chromium from waste water

Green Context

The use of chromium in oxidation processes and in e.g. tanning applications is still widespread, and for many applications, suitable green alternatives are not likely to be ready for several years. The recovery of chromium is thus a current and significant problem. Precipitation using magnesium oxide can be used, but has limitations. For example it cannot be used on a very large scale, and if the waste water is contaminated with oils (used in tanning to produce soft leather) it is of limited efficiency. The process also produces a lot of magnesium sulfate. This article describes the development of a method based on the use of sodium carbonate, which is more tolerant of organic contamination. Furthermore, the process has been designed to run semi-continuously. This is difficult to do with the magnesium oxide process for a number of reasons: the particle morphology obtained with sodium carbonate is better; the precipitated chromium from the new process does not contain as much oil as in the original process; the rate of neutralisation is much faster and sodium does not form ‘soaps’ with the oils used in the tanning process.

DJM

Summary

The use of chromium(III) salts is under review by global industries. One of the options for better management of chromium is through recovery and reuse. Batch type processes for the recovery of chromium and subsequent use in tanning has already gained importance in the tanning industry. Here we describe a novel semi-continuous method for the recovery of chromium(III) from tannery wastewaters containing the metal ion. In this system, sodium carbonate is used as the alkali to precipitate chromium(III) as chromium(III) hydroxide, instead of magnesium oxide, which is preferred for batch type systems. The hydrostatic pressure build-up and turbulence within the reactor is modulated to achieve online separation of chromium(III) hydroxide. Under steady state conditions, the inflow into the reactor is matched by the outflow. The outflow from the reactor can be discharged without causing ground-water hardness.

Introduction

The manufacturing processes of leather tanning requires considerable quantities of water and discharges nearly 30–35 L of water for every kilogram of leather processed.¹ Nearly 90% of the leathers manufactured and currently used contain at least a minor quantity of chromium either as a tanning material or in dyes. The biotoxicity of chromium has been a subject of active discussion. The biological implications of chromium are known to vary with the oxidation state of the metal ion.^2,3 Oxyanions like chromates are well-established human carcinogens.⁴ Numerous studies have attempted to determine the intracellular status of chromium as well as the chemical nature of the ultimate carcinogen or mutagen.⁵ It has been reported that, once inside the cell, Cr^III species can be mutagenic and genotoxic.⁶ The ability of Cr^III to (a) crosslink DNA and proteins, (b) participate in non-enzymatic phosphorylation and (c) influence calcium transport channels has been discussed.^7–9

In view of the potential toxicity of some forms of chromium, the environmental regulatory norms stipulate that the levels of chromium in waste waters are controlled. The discharge norms for industrial waste waters in different countries specify permissible concentrations of chromium in the range of <0.3–2 ppm.¹⁰ Such strict norms demand technological interventions for abating chromium pollution.

Chromium(III) recovery–reuse processes have now been accepted as the commercially most attractive.¹¹ It is now well established that the nature of alkali used in precipitating chromium(III) hydroxide influences the settling characteristics of chromium(III) hydroxide.¹² The characteristics of colloidal chromium(III) hydroxide particles are closely related to the zeta potential. Distribution of particle sizes, extent and degree of aggregation as well as settling behaviour and morphological properties of chromium(III) hydroxide vary markedly with the nature of alkali employed.¹²

Although chromium(III) hydroxide can be generated from aqueous solutions of Cr^III salts with almost any alkali at pH 7–10; the use of magnesium oxide is most extensive.¹³ This is because of its high bulk density and good settling of particulate matter. The particle characteristics enable easy handling of chromium(III) hydroxide generated using MgO from solutions of basic chromium sulfate. The process is batch type and is simple.¹⁴ The batch type process enjoys the advantages of (a) minimum process change in tanneries, (b) repeated recyclability of recovered chromium, (c) acceptable quality of leather after reuse of chromium and (d) a relatively short cost-recovery period of the initial installation investment.

MgO being a partially soluble alkali (0.00062 g/100 ml water), 2–3 h is required for raising the pH to desired levels of 8.0–9.0. Owing to the poor solubility of MgO, rapid addition of alkali may lead to the use of higher amounts of MgO rather than that required stoichiometrically. Commercially available magnesium oxide samples contain significant amounts of calcium salts as impurities. This could result in the generation of calcium sulfate, which is known to co-precipitate along with chromium(III) hydroxide. If inferior grades of MgO are employed in chrome recovery, special steps would be required to remove calcium sulfate from the recovered chromium.¹⁵ Further, the magnesium sulfate generated during the process is expected to render the ground water harder. At certain concentration levels the magnesium ion is a soil micronutrient. It is necessary to assess the carrying capacity of the region and balance the amounts of magnesium salts being added to the soil and ground water prior to the wide scale adoption of chromium(III) hydroxide recovery process in tannery clusters.

The volume of leather processing activity using chromium(III) tanning is increasing. Batch sizes of chromium(III) tanning activities are also on the increase. Large batch sizes of spent chromium(III) tanning liquors are becoming available for chromium(III) hydroxide recovery. Even when small production systems are involved, waste water management processes adopted are based on common effluent treatment plants. Logical support to common waste water treatment concepts demands common chromium recovery plants. Newer trends in high exhaust chromium(III) tanning as well as stringent regulations on chromium discharge would result in large quantities of spent liquors bearing lower concentrations of chromium. There is now a need to develop continuous chromium(III) hydroxide recovery concepts.

It is now a common practice to use oils and fats during the chromium(III) tanning process to obtain soft leathers. In these cases, magnesium oxide based recovery systems could result in the formation of magnesium soaps, which adversely influence the recovery and reuse processes.¹⁶

The higher bulk density of magnesium oxide based chromium(III) hydroxide favors its use over sodium carbonate or sodium hydroxide based systems. Since MgO is a low soluble alkali, the equilibrated pH after addition of MgO to the reaction media is not immediately established. Continuous chromium(III) hydroxide recovery systems demand the use of an alkali for which the equilibrated pH is established more or less immediately.

Experimental

Spent chromium liquor was generated from a typical chromium(III) tanning process. The solution was filtered through a Whatmann filter paper No. 1 and the concentration of chromium estimated spectrophotometrically as per standard methods.¹⁷ The solution was then diluted to have an effective concentration of 1000 ppm Cr.

A semi-continuous chromium recovery process would need two vessels, one for the neutralization of the chromium liquor and the other for settling. While a cylindrical vessel was used for neutralization, a hopper bottom geometry was used for the settler. For efficient design of the equipment, the inflow of chromium bearing effluent into the settler and the outflow of chromium free supernatant need to be matched. The ratio of the volume of settler and that of spent chromium liquor was chosen to be one third of the total volume of spent chromium liquor being treated. A secondary settler of one-third the volume of the primary settler was included in the design as earlier investigations have reported that 30% of the chromium(III) hydroxide particles are fine (0.1–10 μm).¹² The secondary settler was included to trap the particles escaping from the primary settling tank without sedimentation. The geometry of the primary settler took into consideration the hydraulic pressure head, turbulence in a flowing reaction medium and buoyancy forces. The ratio of the height of cylindrical to the cone regions of the primary settler was 0.58.

A typical amount of spent chromium(III) liquor used for chromium(III) hydroxide recovery using the semi-continuous process was 3 L. A cylindrical neutralization tank (150 ml) was placed over a magnetic stirrer with an outlet at 1.5 cm from the bottom of the vessel. The height of the neutralization tank was set at 14 cm. A pressure head of 5 cm over the outflow tube of the neutralization vessel was employed. The primary settler had a cylindrical top and a hopper bottom. The height of the cylindrical portion was set at 17.5 cm. The bottom portion of the primary settler was the shape of a frustum of a cone with a 3 cm diameter at the bottom of the settler. The fluid receiving mechanism of the primary settler includes a well that modulated flow turbulence. The design of the primary and secondary settler is given in Fig. 1. A schematic drawing of the reactor setup is shown in Fig. 2. The chromium(III) bearing solution and alkali were pumped at pre-determined velocities into the neutralization tank (4 and 0.6 ml min⁻¹ respectively) into the neutralization tank and the reaction stream flowed through a pipeline with surface friction and under hydrostatic pressure. The suspension was passed through a primary settler. The reactor was designed to modulate the forces of buoyancy which could influence the settling of the finer particles of chromium(III) hydroxide. Under steady state conditions, the flow rates of fluids into and out of the settler need to be equal. The total volume of fluid which could be handled by the settler is correlated to the bulk volume and settling behavior of the precipitate being obtained.

Fig. 1 The semi-continuous chromium(III) hydroxide recovery system. (a) Primary settler and (b) secondary settler.

Fig. 2 A semi-continuous chromium(III) hydroxide recovery plant.

Two different alkalies viz. sodium carbonate and sodium hydroxide were used in this study. Either 0.1 M NaOH or 0.05 M Na₂CO₃ was employed for pH adjustment. The density and particle size of chromium(III) hydroxide were measured as per procedures reported earlier.^12,18 The settling velocity was calculated using Stokes law¹⁹ after determining the density, viscosity and temperature.

The capacity of the settler and the rate of fluid inflow determine the available residence time of chromium(III) hydroxide for settling. The flow rate of chromium(III) bearing effluent into the mixing tank was varied from 2 to 10 ml min⁻¹.

The efficiency of the reactor design has been studied over a range of chromium(III) concentrations of 400–2000 ppm. The settled volume of chromium(III) hydroxide and chromium(III) content in the supernatant was studied.

The influence of anionic polyelectrolytes on the settling characteristics of chromium(III) hydroxide was studied using a commercial polyelectrolyte (Synfloc 110 A of Chemical Specialties India (P) Ltd.) and varying its concentration from 0.2 mg L⁻¹ of chromium liquor treated to 1.0 mg L⁻¹. The settled volume was measured and the polyelectrolyte concentration optimized.

To the spent liquor, 1% acid stable anionic fatliquor (Lipoderm Liq SA of BASF India) was added. The resulting mixture was treated with alkali in the neutralization tank and chromium(III) hydroxide settled. The sludge obtained was analyzed for oils and fat content by standard procedures.²⁰ A comparative assessment of the oil and fat content was made with a standard batch type system using magnesium oxide as the precipitant alkali.

The aggregation characteristics of chromium(III) hydroxide were studied in a flow tube (Fig. 3). The tube had sampling outlets at constant intervals. Chromium(III) solution and alkali were flown along the two arms of the Y tube. Aggregation and formation of particles of chromium(III) hydroxide along the flow path were monitored by turbidometric techniques and a UV–VIS spectrophotometer. The flow rate of chromium liquor ([Cu^III] = 200 ppm) was maintained at 6 ml min⁻¹, while that of the alkali varied and adjusted the pH to 7.5, 8.5 and 9.5. Samples were drawn through the outlet positioned at various distances from the point of mixing of chromium solution and alkali. Optical density was measured at 400 nm after the solutions had been allowed to flow for 45 min. The rate of particulation of chromium(III) hydroxide was considered to be first order and the rate constant for the particulation of chromium(III) hydroxide can be given as k = 2.303t⁻¹ [log(Δturbidity)], where ‘t’ is the time taken to travel a distance ‘d’. The rate constant was estimated by plotting the logarithmic values of changes in absorbance or turbidity against the time taken to travel a distance d, by flowing suspension at a given flow rate. The data were analyzed using a non-linear least squares fit program.

Fig. 3 Continuous flow tube configuration adapted to investigate the rate of particulation of chromium(III) hydroxide.

Results and discussion

The particle size distribution and settled volume of chromium(III) hydroxide generated using NaOH and Na₂CO₃ under the conditions used in the study are presented in Table 1. One important observation is that when the sludge generated settles under gravity overnight in the primary settler, the sludge volume was reduced to 11 and 12.5% with respect to NaOH and Na₂CO₃. This volume can be compared with 9–10% obtained with magnesium oxide based batch processes.¹⁵ The velocity of settling calculated using Stokes’ law was found to be 381 and 314 × 10⁻³ m⁻¹ respectively, indicating that settling under the semi-continuous system is independent of the alkali employed.

Table 1 Influence of the nature of the alkali on the settled volume of chromium(III) hydroxide

Particle size distribution/μm
Alkali	Settled volume (%)	Density/ kg m^−3	<5	5.1–10	10.1–15	>15
Flow rate 4 ml min^−1. [Cr^III] = 1000 ppm
NaOH	22	2220	28	56	14	2
Na2CO3	23	2130	22	54	22	2

---

The settled volume under dynamic conditions will be influenced by the time available for settling. Hence, in principle the optimum inflow rate of solution will be limited by the settling characteristics of the chromium(III) hydroxide precipitate and the settler volume. Employing a chromium concentration of 1000 ppm and maintaining a pH of 8.0–8.5 using 0.05 M Na₂CO₃ the flow rate was varied from 2 to 10 ml min⁻¹. The settled volume was found to increase from 21% for 2 ml min⁻¹ to 34% for 10 ml min⁻¹, while the chromium content in the supernatant increased from 1.3 to 4.0 ppm. The results are presented in Table 2. Factors like turbulence and forces of buoyancy may be responsible for the relatively higher volumes of chromium(III) hydroxide being precipitated at higher flow rates. From the experiments conducted in this study a flow rate of 6 ml min⁻¹ could well be employed, resulting in a sludge volume of 25%. Thus for a reactor capacity of 1.2 L and a continuous operation the reactor can treat approximately 7 times its volume per day.

Table 2 Influence of flow rate and flocculent on the settling characteristics of chromium(III) hydroxide

Flow rate/ml min^−1	Flocculent concentration/ ppm	Settled volume (%)	Cr content in supernatant/ ppm
Inflow [Cr^III] = 1000 ppm. Alkali employed = 0.05 M Na2CO3. Flow rate of chrome liquor = 4 ml min^−1
2	0	21	1.3
4	0	23	1.8
0.2	21	1.9
0.4	19	1.9
0.6	18	1.8
0.8	18	1.8
1.0	20	1.9
6	0	25	2.3
8	0	29	3.2
10	0	34	4.0

---

The settled volume of chromium(III) hydroxide and the chromium(III) content in the supernatant for varying concentrations of spent chromium(III) solutions are presented in Table 3. The settled volume was found to be dependent on inflow chromium(III) concentration and varied from 14–40%. However, when the chromium(III) hydroxide was dewatered using a basket centrifuge, the differences in the settled volume of chromium(III) hydroxide was only marginal. The reactor design, for operational convenience, has not incorporated the roughness parameter. In commercial practice, the use of fibre reinforced plastic would bring in such parameters and this would reduce the chromium(III) content in the supernatant to stipulated levels.

Table 3 Influence of inflow chromium(III) concentration on the settling characteristics of chromium(III) hydroxide

Settled volume of chromium(III) hydroxide(%)
Inflow chromium(III) concentration/ppm	Before dewatering	After dewatering	Cr content in supernatant/ppm
Alkali employed = 0.05 M Na2CO3. Flow rate of chrome liquor = 4 ml min^−1
400	14	9	1.4
750	21	15	1.9
1250	30	20	2.3
1500	35	22	2.5
2000	40	24	3.1

---

The effect of a commercial anionic flocculent on settling characteristics of chromium(III) hydroxide has been studied. Data on settled volume of chromium(III) hydroxide as a function of flocculent concentration is presented in Table 2. It has been observed that the reduction in sludge volume is only marginal when the flocculent concentration is increased above 0.4 ppm. A flocculent concentration of 0.4 ppm appears optimum. When the flocculent concentration was raised to 1.0 ppm, an increase in the sludge volume was observed. A mechanism suggested for this action of an anionic polyelectrolyte on the settling behavior of a surface active colloid involves desolvation, favorable ionic environment for particulate aggregation and bridging of particulate matter by a polymeric flocculating aid. When a higher than optimum dose of polyelectrolyte was employed, surface loading of the colloid by the polymer could well increase the ionic charge of the colloidal particles and promote solvation. In other words, restabilization of colloid by polymer is likely.

The influence of oils and fats present in the spent chromium liquor on chromium(III) hydroxide recovery has been examined by analysis of oils and fats contained in the chromium(III) hydroxide sludge after precipitation of chromium(III) hydroxide. The oils and fats content in the chromium(III) hydroxide sludge obtained by the batch process was 47% as against 22% in the case of the sodium carbonate based continuous process. The oils and fat content in the chromium(III) hydroxide sludge was further reduced to 9.8% after dewatering, i.e. a 79% decrease when compared to the MgO based batch type process. This indicates that the continuous chromium recovery system based on sodium carbonate as the alkali for precipitation is a viable option for treatment of spent chromium(III) liquors from wool-on leather industries. Dewatering does not seem to influence the oils and fat content of chromium(III) hydroxide sludge generated by employing MgO. Since chromium(III) hydroxide sludge generated using Na₂CO₃ affords a fluffy precipitate, dewatering using a basket centrifuge reduces the oils and fats content more significantly.

The rate constants for particulation of chromium(III) hydroxide for varying pH conditions are reported in Table 4. It can be seen from the data given in the table that under the conditions employed in this study a pH of 8.5 is marginally preferable for achieving higher rates of particulation of chromium(III) hydroxide.

Table 4 Influence of pH on the rate of particulation of chromium(III) hydroxide

pH	Rate constant for particulation/s^−1
Flow rate = 6 ml min^−1. [Cr^III] = 200 ppm.
7.5	2.67 ± 0.5 × 10^−3
8.5	4.50 ± 0.9 × 10^−3
9.5	2.50 ± 0.4 × 10^−3

---

Conclusions

The present study demonstrates a feasible semi-continuous chromium recovery system based on sodium carbonate or sodium hydroxide as alkalies in preference to an MgO based batch type system. Sodium carbonate or sodium hydroxide provides a fluffy precipitate free of contamination but requires dewatering using appropriate mechanical devices such as a continuous centrifuge or filter press. The present investigation provides a viable design for a semi-continuous chromium recovery process based on the use of relatively more cost effective alkalies like Na₂CO₃ or NaOH. The salient features of the present investigation are the findings that (a) the settled volume of chromium(III) hydroxide generated using NaOH–Na₂CO₃ can be effectively reduced to match that obtained in MgO based process through equipment modifications, (b) cleaner chromium(III) hydroxide sludge can be continuously generated and dewatered using appropriate devices and (c) a supernatant which could be discharged without causing ground water hardness.

Acknowledgements

K. J. S. thanks the CSIR-India for a research fellowship. The efforts of Mr V. K. Rangaswamy and his colleagues at the glass blowing unit, CLRI, are acknowledged.

References

1. T. Ramasami, J. R. Rao, N. K. Chandrababu, K. Parthasarathi, P. G. Rao, P. Saravanan, R. Gayatri and K. J. Sreeram, J. Soc. Leather Technol. Chem., 1999, 83, 39.
2. W. Mertz, J. Am. Coll. Nutr., 1998, 17, 544.
3. S. A. Katz, Environ. Health. Perspect., 1991, 92, 13.
4. M. Costa, Environ. Health Perspect., 1991, 92, 45.
5. S. Langard, Am. J. Ind. Med., 1990, 17, 189.
6. K. W. Jennette, Biol. Trace Elem. Res., 1979, 1, 55.
7. K. Salnikow, A. Zhitkovich and M. Costa, Carcinogenesis, 1992, 13, 2341.
8. T. C. Tsou, R. J. Lin and J. L. Yang, Chem. Res. Toxicol., 1997, 10, 962.
9. K. Balamurugan, C. Vasant, R. Rajaram and T. Ramasami, Biochim. Biophys. Acta, 1999, 1427, 357.
10. J. Buljan, World Leather, 1996, November, p. 65..
11. J. S. A. Langerwerf, J. W. van Groenstijn, A. De Vriet, T. Ramasami, S. Rajamani, R. Ravindranath, N. K. Chandrababu, J. R. Rao, A. D. Covington, A. Long, K. M. Nair and K. G. K. Warrier, Leather Manuf., 1998, 116, 12.
12. K. J. Sreeram, J. R. Rao, R. Venba, B. U. Nair and T. Ramasami, J. Soc. Leather Technol. Chem., 1999, 83, 111.
13. J. R. Rao, N. K. Chandrababu, P. S. Rao, R. Ramesh, R. Suthanthararajan, B. U. Nair, K. M. Nair, K. G. K. Warrier, S. Rajamani, T. Ramasami and J. S. A. Langerwerf, Science and Technology for Leather into the Next Millenium, Tata McGraw-Hill Publishing Company, New Delhi, 1999, p. 295..
14. A. D. Covington, R. L. Sykes, J. R. Barlow and E. T. White, J. Soc. Leather Technol. Chem., 1983, 67, 5.
15. J. R. Rao, R. Venba, B. U. Nair, R. Suthanthararajan, S. Rajamani, T. Ramasami and J. S. A. Langerwerf, Proc. 30th Leather Research Industry Get-together, CLRI, Chennai, 1995, p. 37..
16. E. Heidemann, J. Am. Leather Chem. Assoc., 1991, 86, 331.
17. G. H. Jeffery, J. Basset, J. Mendham and R. C. Denney, Vogel’s Textbook of Quantitative Chemical Analysis, Longman Scientific and Technical Publication, UK, 1994, p. 643..
18. R. R. Irani and C. F. Callig, Particle Size Measurements, Interpretation and Applications, John Wiley, New York, 1963..
19. J. M. Coulson, J. F. Richardson, J. R. Backhurst and J. H. Harker, Chemical Engineering, ed. J. M. Coulson and J. F. Richardson, Pergamon Press, Oxford, 1983, vol. 2, p. 172..
20. P. K. Sarkar, Analytical Chemistry of Leather Manufacture, ILTA Publication, Calcutta, India, 1984..

---