Platinum-group elements: quantification in collected exhaust fumes and studies of catalyst surfaces

Abstract

Automotive catalytic converters, in which Pt, Pd and Rh (platinum-group elements; PGEs) are the active components for eliminating several noxious components from exhaust fumes, have become the main source of environmental urban pollution by PGEs. This work reports on the catalyst morphology through changes in catalyst surface by scanning electron microscopy/energy dispersive X-ray spectroscopy (SEM/EDX) and laser-induced breakdown spectrometry (LIBS) from fresh to aged catalytic converters. The distribution of these elements in the fresh catalysts analysed (Pt–Pd–Rh gasoline catalyst) is not uniform and occurs mainly in a longitudinal direction. This heterogeneity seems to be greater for Pt and Pd. PGEs released by the catalysts, fresh and aged 30 000 km, were studied in parallel. Whole raw exhaust fumes from four catalysts of three different types were also examined. Two of these were gasoline catalysts (Pt–Pd–Rh and Pd–Rh) and the other two were diesel catalysts (Pt). Samples were collected following the 91441 EUDC driving cycle for light-duty vehicle testing. The results show that at 0 km the samples collected first have the highest content of particulate PGEs and although the general tendency is for the release to decrease with increasing number of samples taken, exceptions are frequent. At 30 000 km the released PGEs in gasoline and diesel catalysts decreased significantly. For fresh gasoline catalysts the mean of the total amount released was approximately 100, 250 and 50 ng km⁻¹ for Pt, Pd and Rh, respectively. In diesel catalysts the Pt release varied in the range 400–800 ng km⁻¹. After ageing the catalysts up to 30 000 km, the gasoline catalysts released amounts of Pt between 6 and 8 ng km⁻¹, Pd between 12 and 16 ng km⁻¹ and Rh between 3 and 12 ng km⁻¹. In diesel catalysts the Pt release varied in the range 108–150 ng km⁻¹. The soluble portion of PGEs in the HNO₃ collector solution represented less than 5% of the total amount for fresh catalysts. For 30 000 km the total amount of soluble PGEs released was similar or slightly higher than for 0 km.

Introduction

Environmental contamination by Pt, Pd and Rh, which belong to the platinum-group elements (PGEs), is mainly related to the increasing use of noble metal-based automotive catalytic converters for the reduction of emissions of hydrocarbons, CO and NO_x below the present legislated level. After the introduction of automotive catalysts, research into Pt, Pd and Rh emission has focused mainly on two objectives: the first is to realistically assess the risk that the release of these elements from the catalyst surface represent for man and other living organisms; and the second is to enable the automobile industry to determine how effectively these catalysts remove the above pollutants, in order to make improvements in catalyst technology. The extent of PGE release has not been fully investigated, but an exponential rise in environmental concentrations has been observed as the number of cars fitted with these catalytic converters has grown (Hodge and Stallard, 1986, Schäfer et al., 1999).

The catalytic converters installed in most cars consist of a honeycomb type monolithic support made of a high melting point ceramic material, normally cordierite (2MgO·2Al₂O₃·5SiO₂). The cordierite is coated with a highly porous washcoat that consists of approximately 90% γ-Al₂O₃ where PGEs are finely dispersed in a metallic form.

Today, there is a wide range of possible combinations and concentrations of Pt, Pd and Rh that can be used to achieve the different performance features required by car manufacturers. In the various Pd–Rh, Pt–Rh, Pt–Pd, Pt-only or Pt–Pd–Rh catalysts, the percentage of PGEs with respect to the bulk material is <0.1 wt.%. Other elements present in the catalyst, such as Ce, Zr and rare earths are used as components of the washcoat to increase defined properties such as catalyst PGE impregnation, oxygen storage capability and chemical inertness (Burtin et al., 1987, Cuif et al., 1997).

The loss of catalyst performance during on-road lifetime can be attributed to a number of mechanisms namely chemical, thermal, fouling and mechanical processes (Carol et al., 1989, Koltasakis and Stamatelos, 1997).

From the environmental point of view, the mobility of PGEs on/from the catalyst surface is the most important factor for the increase in their environmental background level and can be mainly attributed to thermal sintering, evaporation and mechanical or thermal erosion (Stenbom, 1994).

Sintering occurs when small crystallites migrate towards larger crystallites or when single atoms emitted from small crystallites are captured by larger ones. During this migration some particles or atoms can leave the catalyst surface (Flynn and Wanke, 1975, Wynblatt and Gjostein, 1975, Ruckenstein and Dadyburjor, 1977). Evaporation of the precious metals results in a loss of catalytic active surface. Although PGEs are quite thermally stable (boiling points between 3000 and 4000°C), losses of Pt and Rh have been observed under oxidizing conditions and at temperatures above 900°C (Raub and Plate, 1957, Schäfer and Tebben, 1960). This temperature and oxidizing conditions can be easily reached on the catalyst surface during automobile operation. Mechanical or thermal erosion are the main causes of release to the environment although this is probably not a significant problem for the efficiency of monolithic catalysts.

Catalyst characterization and decrease in catalyst conversion efficiency for field-aged catalytic converters have been observed through techniques based on the interaction of ions with solids such as secondary-ion mass spectrometry (SIMS) and low energy ion spectrometry (LEIS) or the more widely applied X-ray photoelectron spectroscopy (XPS) or X-ray diffraction (XRD) (Delmon, 1986, Nimantsverdriet, 1993, Unger and Gross, 1998). Recently, techniques based on the interaction of laser radiation with solids such as laser-induced breakdown spectrometry (LIBS) have proved capable of mapping the distribution of atomic species on the surface (Romero and Laserna, 1997). Mapping of Pt, Pd and Rh in fresh and aged catalysts is one of the objectives of this study and experimental conditions of mapping have been demonstrated elsewhere (Lucena et al., 1999).

The total amount of PGEs released into the environment by catalysts can be evaluated by directly determining their content in car exhaust fumes or by quantifying the anthropogenic PGEs in environmental materials such as soil, airborne particles, sludge, water, road dust, etc. (Hees et al., 1998), and modelling these data together with traffic statistics. Direct determination requires an effective sampling procedure where representative amounts of the released PGEs are collected. Table 1 shows the results obtained by authors following either strategy. Comparison of the data for Pt, which is the element most frequently analysed, reveal an evident discrepancy, with indirect determinations giving generally higher values.

Other sources of Pt in the environment are industrial processes like polymerization of olefins, refining of petroleum, ammonium oxidation, electronic industries, glass and jewellery manufacturing, etc. (Johnson Matthey, 1996). However, little data are available about the contribution of these sources (WHO, 1991). Hospital effluents containing Pt drugs are also an important source for the emission of Pt into wastewater and sewage sludge (Kümmerer and Helmers, 1997). Kümmerer et al. (1999) reported a total emission of 14.2 kg of Pt in 1996 in Germany corresponding to this source vs. 187.2 kg of Pt from cars.

As with Pt, the concentrations of Pd and Rh in the environment are also increasing, but this concentration seems to be higher than would be expected if the catalysts were the only source. Other possibilities have been pointed out: (i) Pd and Rh come from sources other than automobiles (Helmers et al., 1998). The growing industry of both Au–Pd and Pd-based alloys could explain the relatively high Pd concentration found in wastewater effluents and sewage sludge. However, this explanation is not acceptable for road dust, plants growing near roads and also for Rh that is not used for these purposes; (ii) Pd and Rh come from automobile sources other than the catalyst, such as fuel, oil, spark plugs or engine abrasion. However, this explanation would not apply to Rh because of the high cost of this element; and (iii) these elements also come mainly from catalysts but their mobilities are higher than that of Pt.

An additional problem for the ultratrace environmental determination of Pd and Rh is that the analytical methodology for their quantification is subject to positive interferences that cannot be easily controlled (Krachler et al., 1998, Gómez et al., 2000). The most important analytical technique for environmental concentrations of these elements is inductively coupled plasma mass spectrometry (ICP-MS), which has significant mass interferences from ArCu⁺, oxides such as YO⁺ and SrO⁺, or double ions such as Pb²⁺ with charge/mass ratios very close to that of the measured isotope of the noble metal element. All these interferents are present in environmental matrices.

The studies reported here belong to a multi-partner project (CEPLACA, 1997) whose final objective is to assess the health and ecosystem risks based on the emission of the three autocatalyst PGEs. The main problem in assessing the risk for humans following Pt, Pd and Rh exposures is to correctly estimate the magnitude and nature of exposure. We have limited knowledge about the compounds formed in the catalyst, the route of administration and the dose reaching critical body organs. Occupational exposure to platinum salts such as hexachloroplatinic (IV) acid has been associated with respiratory allergic reactions and skin reactions (Lindell, 1997). Platinum compounds (e.g. cisplatin and carboplatin) are used in the treatment of human cancer (Parsons et al., 1987) and such anticancer drugs have been shown to cause extensive DNA damage when delivered in high doses. The health effects caused by exposure to Pd and Rh compounds are less well known than for Pt, although the acute toxicity of PGE compounds in general seems to be relatively low. As a first step in addressing this general task, the aims of this present study are to: (i) study changes in catalyst surface due to ageing by LIBS and SEM/EDX; (ii) monitor PGEs in exhaust fumes from fresh to medium-aged catalysts by ICP-MS and cathodic stripping voltammetry (CSV); and (iii) establish a quality control scheme for sampling and analysis.