A new membrane based on polymer nanofibers makes it possible to extract precious metals from spent automotive catalytic converters — devices for purifying vehicle exhaust gases. In experiments with a solution obtained after processing such a converter, the overall extraction efficiency reached 82.3% for palladium, 71.5% for platinum, and 75.9% for rhodium. The membrane was created by scientists from Nelson Mandela University, the University of the Western Cape, and Stellenbosch University in South Africa, together with researchers from the Joint Institute for Nuclear Research (JINR) in Dubna.
Platinum, palladium, and rhodium are in demand for the production of catalysts, fuel cell electrodes, and electronics. However, their natural resources are scarce and unevenly spread across the globe, making ore extraction expensive. Automotive catalytic converters are one additional source of these metals, although not the richest: the content of platinum group metals in them is only about 0.2%. It is this low concentration that significantly complicates their extraction.
To solve this problem, the researchers used a track membrane, which is a thin film of polyethylene terephthalate with many identical pores 0.2‑0.3 µm in diameter. Such membranes alone are poorly suited for selective metal extraction, so the base was coated with a thin layer of titanium, and on top of that a grid of polymer nanofibers treated with triethylenetetramine was applied. This compound contains active groups capable of chemically binding metal ions. As a result, the solution passes through the membrane, while the metals it contains can be retained on the surface of the nanofibers.
Three variants of this membrane were produced, specifically, with nanofibers made of polyacrylonitrile, polystyrene, and polysulfone. The best performance was shown by the material based on polyacrylonitrile. It was possible to attach 2.081 mmol of the active compound per gram of its nanofibers, compared to 0.601 mmol for polysulfone and about 0.43 mmol for polystyrene. In addition, the surface of the polyacrylonitrile material was completely wetted by the acidic aqueous solution, allowing the liquid to pass through the membrane more easily.
In the initial experiments, the membrane was placed in model solutions of each metal individually. It absorbed platinum best: in about five minutes, the material extracted about 90% of this metal. For palladium, the maximum value was only 27%, for rhodium — 35%, and the process was largely completed within 15 minutes.
When several metals were present simultaneously, the results changed. In a model solution with equal concentrations of platinum, palladium, and rhodium, the membrane extracted about 60% of platinum, 50% of palladium, and 30% of rhodium. In a real dilute solution obtained after treating ground automotive catalytic converter with a mixture of hydrochloric acid, sodium chloride, and hydrogen peroxide, it absorbed about 90% of platinum, while the proportions of palladium and rhodium were less than 10%. According to the scientists, the performance of the material is affected by the complex composition of the real solution and the competition of various metal compounds for the active sites on the membrane.
After the static experiments, the membrane was tested under continuous filtration conditions using a more concentrated solution obtained after processing the converter. It was passed through the membrane at a flow rate of 3 ml per minute. After passing 70 ml of solution, the membrane capacity was 1.188 mg per gram of material for palladium, 1.129 mg/g for platinum, and 0.321 mg/g for rhodium. During the flow experiment, rhodium was practically not retained by the membrane and mostly remained in the solution that passed through it. This made it possible to separate it from platinum and palladium already at the initial stage. Platinum and palladium, on the contrary, accumulated on the membrane, after which they could be removed from it and obtained in a separate solution.
For this, the researchers used two reagents: first sodium perchlorate, intended primarily for extracting platinum, and then thiourea for palladium. However, clear separation of the two metals has not yet been achieved: both platinum and palladium partially passed into solution at both stages. Overall, the total extraction efficiency for the entire process was 82.3% for palladium, 71.5% for platinum, and 75.9% for rhodium. The membrane also withstood three consecutive cycles without any noticeable loss of metal absorption capacity.
The main unresolved problem remains the separation of platinum and palladium, so the next step should be the selection of more selective reagents. If this problem can be solved, such membranes could be used to extract valuable materials from a wide range of secondary raw materials in continuous flow mode.
