A precious resource

Catalysts containing precious metals are worth recycling carefully

When catalysts based on platinum, palladium and other precious metals lose their effectiveness, their remaining value will still affect your bottom line. Here are some tips for getting the best from catalyst recycling.

ROBERT T. JACOBSEN
Catalysts containing precious metals are used in many pharmaceutical and chemical processes, such as in hydrogenation and to remove volatile substances from exhaust gases. Most catalysts lose their effectiveness after a maximum of five or six years in service, at which point it is worth disposing of them with care. With platinum currently worth around $800 per troy ounce (31 g), spent catalysts can represent a serious investment. An increasing number of catalyst users are making a determined effort to cut their costs by setting up independent asset recovery programs or departments that function as profit centers for the recovery of precious metals.

For many catalyst users, working with a company that specializes in refining precious metals is the most profitable approach. Your choice of refiner can make a significant difference to your bottom line, so it is in your interest to learn as much as possible about any company you are considering. How does the organization recover spent catalysts? What are its materials sampling and assaying methods, its environmental protection policies and systems? Perhaps most important, what is the refiner's reputation with its long-term customers? This article provides some background on both the technical and the financial issues.

Precious-metal catalysts contain one or more of the six platinum group metals (PGMs): platinum, palladium, ruthenium, rhodium, iridium and osmium. Most catalysts take the form of beads, pellets, powders or extrudates of carbon or ceramic, though liquid-phase (homogeneous) catalysts are also used. Precious-metal recovery and refining includes many steps — materials documentation, contamination removal, sampling, assaying, refining and environmental considerations — that influence the value of the materials recovered and the time this takes. Of these steps, sampling the spent catalyst is perhaps the most critical in maximizing the value of metal recovered.

The importance of sampling

The basic problem in sampling is that catalysts, especially after spending years in a harsh process environment, are far from homogeneous. They contain small quantities of precious metals dispersed in large quantities of insoluble and refractory carrier materials, and they are also likely to be contaminated with sulfur, carbon, solvents and with water.

Melting and dissolution, two methods sometimes used to sample spent materials containing precious metals, are not usually appropriate for catalysts. As a result, dry sampling is the usual method for catalysts. The aim is to take a large quantity — up to several tons — of spent catalyst and from it extract a sample of around 150 g in which the amount of precious metal is truly representative of that in the larger quantity. Because spent catalysts are difficult to homogenize, dry sampling is a complex process that requires good judgment. Customers need to know not only how much precious metal their spent catalysts contain, but also how the refiner arrived at this estimate.

In a typical precious metal refinery, incoming catalyst materials are inspected, weighed, assigned tracking numbers and stored. Each lot is tracked and handled individually at all times. The next step is to remove contaminants that might affect either the sampling process or the ultimate recovery of precious metals. Sulfur, carbon and other contaminants are removed in a rotary kiln, a multiple-hearth furnace or a fluidized-bed furnace. This "pre-burning" is best done at the refiner's own plant, rather than at a third-party "regenerator", to reduce both the risk of sample identification errors and the cost of transport. Another critical issue here concerns the time involved for transporting the material back and forth to an off-site facility: this added time could easily consume a month or more thus locking up the metal value unnecessarily.

After the pre-burned material has been crushed and screened to remove agglomerates and fines, an auto-sampling system removes two large samples, each representing ten percent of the entire lot. Next, these large samples are again divided to create two smaller samples, each representing one percent of the lot. From one of these samples are taken further sub-samples weighing around 0.5 kg each. In the refiner's laboratory these samples are burned under controlled conditions to determine the amount of volatile material remaining, known as the loss on ignition (LOI).

The other one percent sample, which may weigh as much as several hundred kg, is ground in a ball mill, screened to a particle size below 40 mesh (0.3-0.4 mm) and then divided into smaller samples. This grinding and screening step is critical, because the particle size distribution at this point affects the accuracy of the entire analysis.

Independent assay

Next, a rotary carousel screener divides the material into 24 individual bins. A 1 kg quantity from one of these bins is ground in a ring-and-puck mill to below 100 mesh (0.15 mm), then split into eight 150 g samples using a small rotary sampler. These samples are sealed and distributed to various locations. One sample is sent to the customer, who arranges an independent assay to determine its precious-metal content, while another sample is assayed by the refiner.

Throughout the sampling procedure, the refiner should adhere to all the applicable environmental codes and laws. The sampling system should ideally be enclosed, with its own air extraction system and baghouse to stop toxic materials being discharged to the atmosphere. There is an economic justification too: dust collected during sampling should be recovered, sampled and its precious-metal value returned to the customer. The precious-metal content of the dust can be very different to that of the bulk material.

If the refiner's and customer's assays agree within predetermined limits, they are averaged to arrive at a final figure. If they do not agree, one of the remaining sealed samples is sent to an independent laboratory which acts as an umpire. The remaining samples, which may carry the seals of both the refiner and the customer, can be used to settle any further dispute over the assay results. The assay itself uses a variety of laboratory techniques to discover not just the precious-metal content of the material but also other information that will help maximize the yield. Procedures include X-ray fluorescence spectroscopy (XRF), atomic absorption spectroscopy (AA) and inductively coupled plasma (ICP) emission spectroscopy, as well as classic volumetric, gravimetric and fire assay techniques. Once sampling and assay are complete, and the value of the entire spent catalyst lot has been agreed on, it is time to recover the precious metal. This is done by transferring the spent catalyst to an electric arc furnace (EAF), with the addition of a flux and a carrier metal, usually copper or iron, in proportions calculated from the assay. The EAF yields two products: molten metal and slag. The molten metal, which contains most of the precious-metal content, is cast into ingots and subsequently refined further. The slag contains trace amounts of precious metals, which are also recovered by further refining.

Understanding the market

Catalyst users anxious to keep their costs under control understandably want accurate analyses and high rates of precious metal recovery. The influence of the precious metal market and the speed of catalyst reprocessing are just as important, but less obvious, in controlling costs. The costs of PGMs have increased greatly in the last few years, and they also fluctuate wildly. Prices at one point reached $1,100 an ounce for palladium and over $900 an ounce for platinum: the current rates are around $230 for palladium and $825 for platinum. Such fluctuations are obviously risky for catalyst users. Precious metals used in catalytic processes — especially platinum and palladium — are typically not purchased outright by their users. Instead, they are leased from a physical stockpile of precious metal known as a "pool account". Leasing metal from a pool is a lot like borrowing money from the bank, except that the costs are less predictable: metal leasing rates have been known to vary over the range 3-200 percent per year. These fluctuations are caused by shortages and surpluses of metal, both at the mine (primary production) and in stockpiles and distribution centers. Some catalyst users buy their own precious metals, especially when lease rates are high, or expected to rise. Many companies, however, prefer leasing because they do not want their precious metals to appear on the balance sheet, as either inventory or fixed assets. For this advantage they are willing to incur the extra expense of leasing. Companies who own more precious metal than they currently need can "bank" or lend it to institutions or other businesses in return for interest charges. Banking is typical of the precious metals market as a whole, but rare in the catalyst business. At a typical leasing rate of 8% and platinum price of $770/oz, it costs over $4,200 a week to lease the metal contained in a 40,000 lb catalyst charge containing 0.6 percent platinum. The metal itself is worth nearly $2.7 million. It is therefore in the user's interest that spent catalyst should be refined quickly so that its precious-metal content can be returned to productive use. The longer the turnaround time, the higher the costs of leasing extra precious metals to cover the down-time.