Adsorption of heavy metals by exhausted coffee grounds as a potential treatment method for waste waters

The adsorption of the heavy metal ions Cu, Zn, Cd and Pb from aqueous solution by used coffee grounds has been investigated as a potential low-cost treatment method for heavy metal-containing waste waters that is based on a readily available natural by-product. The results show that metal ion adsorption is efficient over a fairly wide pH range and adsorbed metals are reversibly leached from the exhausted coffee by dilute acid without significant loss of the adsorptive capacity for subsequent re-use. [DOI: 10.1380/ejssnt.2006.504]


I. INTRODUCTION
The re-use of natural waste materials that arise through various industrial processes for additional purposes, rather than simple disposal, makes both environmental and commercial sense [1]. Thus there has been a recent focus on agricultural and food industry wastes such as tea, coffee grounds and rice hull [2] as alternatives to synthetic ion-exchange resins or activated carbon for treating metal-containing waste streams. Tannincontaining materials such as exhausted coffee contain metal-binding polyhydroxy polyphenol functional groups [1] and are available in large quantities from the manufacture of instant coffee [3]. While there have been several studies of metal ion adsorption by tea and coffee [4,5] the detailed chemistry causing their affinity for different metal ions is not yet well-known.
Here we report the use of exhausted coffee grounds as an adsorbent for the heavy metal ions Cu 2+ , Zn 2+ , Cd 2+ and Pb 2+ from aqueous solution. We present results describing adsorption isotherms for the metal ions, and the effects of adsorbent concentration, pH and metal ion concentration. In addition, we present results for the use of the adsorbent in a flow-through column.

II. METHODS
Commercial ground coffee (Blue Mountain variety) was sieved through an ASTM 18 stainless steel sieve (1 mm mesh size). It was then repeatedly leached with excess 0.01 M NaOH solution (30 g coffee in 800 mL solution) at ∼ 60 • C for 10-30 min with decanting and replacement of the NaOH until the yellow colour of the solution was no longer observed (4-6 solution changes). After this the coffee grounds were suspended in 0.01M HNO 3 and poured through a filter funnel containing a Whatman 114 filter and then washed repeatedly with deionized water (Millipore Milli-Q system) until a pH close to 6.0 was achieved. Finally the grounds were oven dried at 100 • C for 2 h, cooled to room temperature and sealed with parafilm. Earlier experiments [6] employed coffee leached with deionized water only, and our subsequent * Corresponding author: han@alkali.otago.ac.nz work showed that this adsorbent continued to release soluble coffee material on subsequent exposure to aqueous solution. This complicates adsorption by competing for the metal cations. However, the alkali-leached material released very little coloured material.
Batch experiments measuring adsorption of Cu 2+ , Zn 2+ , Cd 2+ and Pb 2+ were conducted by adding known concentrations of each metal ion to a suspension of coffee adsorbent in water and equilibrating overnight at room temperature with occasional shaking. Typical conditions were 3 g·L −1 of coffee grounds and 200 µM of metal ion. The pH of the suspension was then adjusted to the required value by addition of dilute NaOH or HNO 3 as required. pH buffers were avoided since these could compete as ligands for the metal cations. After equilibration, the suspensions were filtered through a Whatman 114 filter, the filtrate was analyzed for metal ions by atomic absorption spectrometry and the amount adsorbed calculated by difference. The effects of pH and the concentrations of adsorbent and metal ion were studied.
Column adsorption experiments were also conducted in which solutions of metal ion concentration were pumped through a column containing 0.5 g of exhausted coffee adsorbent using a peristaltic pump, with aliquots of the effluent collected for metal ion analysis. The efficiency of recycling was investigated by repeated leaching of metal ions from the column using 0.1 M HCl and then readsorption of the same metal ions.

III. RESULTS AND DISCUSSION
All of the metal ions exhibited a strong affinity for exhausted coffee grounds as an adsorbent. Figure 1 shows how the percentage of metal ion adsorbed at pH ∼5 increases with the concentration of coffee adsorbent, reaching a plateau (75-90% adsorbed) at ∼20 g coffee L −1 .
The adsorption equilibria were investigated using pH 5.0 solutions containing 3.0 g·L −1 of coffee grounds and metal ion concentrations in the range 20-200 µM. The results were then fitted to a Langmuir isotherm [7] as follows: where X is the amount of metal ion adsorbed per g of adsorbent, X m is the maximum amount of metal ion that can be adsorbed per g of adsorbent, c e is the final equilibrium concentration of the metal ion in solution (M) and b is effectively the equilibrium constant for adsorption. Figure 2 shows the Langmuir adsorption plot for Zn 2+ adsorption. Linear least-squares regression was used to calculate b and X m from the best-fit line. Table I presents the resultant values for all of the metal ions studied. Figure 3 shows the pH dependence of metal ion adsorp-  tion at fixed concentrations of metal ions and coffee adsorbent. The extent of adsorption was largely independent of pH over a wide pH range, but was significantly reduced at very low (pH< 4) and to a lesser extent at high pH (pH> 10). Good adsorption efficiency is observed at intermediate pH values likely to be encountered with many waste waters. Figure 4 shows the percentage of metal ion adsorbed at pH 5.0 from a suspension containing 10 g·L −1 of coffee as a function of the metal ion concentrations. In each case, the fraction adsorbed was fairly constant at metal ion concentrations below about 10 mg·L −1 but at levels above 100 mg·L −1 the adsorptive capacity of the coffee   was exceeded. Table II shows the theoretical adsorption capacity for this concentration of coffee adsorbent calculated from the X m values presented in Table I, from which it is clear that Pb(II) should exhibit the greatest limit for adsorption and Cu(II) the least. This is approximately in agreement with the results shown in Figure 4.
Column experiments indicated almost 100% initial removal of metal ion from solution, with breakthrough occurring after the adsorption capacity of the adsorbent was exceeded, depending on pH, flow rate, amount of column adsorbent and metal ion concentration. This is illustrated by Figure 5 for Cu(II). Using the X m value reported in Table I for this cation, it was calculated that breakthrough should occur after 97 mL of 10 mg·L −1 Cu(II) solution had been passed through the column containing 0.5 g of coffee, which agrees very well with the breakthrough at ∼100 mL actually observed in this case. Table III shows the result of repeated adsorptionelution experiments for Cd(II) and Zn(II). In each case, 100 µg of metal ion comprising 100 mL of 1.0 mg L −1 solution was passed through a column containing 0.5 g of coffee. After rinsing with 50 mL of Milli-Q water, the column was eluted with 100 mL of 0.1M HCl which was retained for analysis. The results show that after the first adsorption-elution cycle, both metal ions are quantitatively adsorbed by, and then eluted from, the column. However, there is some initial loss on the first cycle.

IV. CONCLUSION
Our preliminary results indicate that exhausted coffee grounds offer considerable promise as a low-cost natural medium for waste water treatment. The adsorption is efficient over a fairly wide pH range and adsorbed metals are reversibly leached from the exhausted coffee by dilute acid without significant loss of the adsorptive capacity for subsequent re-use.