Skip to main content
Printable page generated Monday, 8 Aug 2022, 03:16
Use 'Print preview' to check the number of pages and printer settings.
Print functionality varies between browsers.
Unless otherwise stated, copyright © 2022 The Open University, all rights reserved.
Printable page generated Monday, 8 Aug 2022, 03:16

The three-way catalytic converter


Ensuring good quality air is essential for the protection of public health. Governments worldwide have adopted a range of increasingly demanding measures to curb air pollution with a particular focus on the emissions from motor vehicles. An important part of this strategy has been the development of the three-way catalytic converter to remove exhaust pollutants such as carbon monoxide, unburnt hydrocarbons and nitrogen oxides. This course takes an in-depth look at the construction of this converter for petrol-driven vehicles and investigates the catalytic chemistry taking place at the molecular level. It is assumed that you already have a scientific background.

This OpenLearn course provides a sample of Level 3 study in Science.

Learning outcomes

After studying this course, you should be able to:

  • discuss how the gas mixture expelled from the engine, and the conversion performance of the three-way catalytic converter, depend on the air/fuel (A/F) ratio

  • list the chemical reactions whereby the three-way catalyst removes carbon monoxide (CO), hydrocarbons and oxides of nitrogen (NOx) from petrol vehicle exhausts

  • interpret the results of experimental studies (involving activity tests, kinetic measurements, adsorption studies and/or various surface science techniques) of the three-way catalyst and appropriate model systems

  • discuss possible mechanisms for the catalytic reactions removing CO, hydrocarbons and NOx from vehicle exhausts

  • outline the modes of deterioration of the three-way catalyst, and comment on the strategies that could be used to reduce H2S emissions.

1. Overview

4.1 Exhaust pollutants

The most important chemical reaction in a petrol engine – that is, the one that provides the energy to drive the vehicle – is the combustion of fuel in air. In an ‘ideal’ system, combustion would be complete so that the only exhaust products would be carbon dioxide and steam. In practice, the complete oxidation of the fuel depends on a number of factors: first, there must be sufficient oxygen present; second, there must be adequate mixing of the petrol and air; and finally, there must be sufficient time for the mixture to react at high temperature before the gases are cooled. In internal combustion engines, the time available for combustion is limited by the engine’s cycle to just a few milliseconds. There is incomplete combustion of the fuel and this leads to emissions of the partial oxidation product, carbon monoxide (CO), and a wide range of volatile organic compounds (VOC), including hydrocarbons (HC), aromatics and oxygenated species. These emissions are particularly high during both idling and deceleration, when insufficient air is taken in for complete combustion to occur.

Another important result of the combustion process, particularly during acceleration, is the production of the oxides of nitrogen – nitric oxide (nitrogen monoxide, NO) and nitrogen dioxide (NO2). Conventionally, these two oxides of nitrogen are considered together and represented as NOx. At the high temperatures involved (in excess of 1 500 °C) nitrogen and oxygen in the air drawn in with the fuel may combine together to form NO. On leaving the engine, this monoxide cools down and is oxidized by oxidants in the atmosphere to form the dioxide. Although the ‘fixing’ of nitrogen from the air is the major source of NOx, it may also arise from the oxidation of any nitrogeneous components in the fuel.

Primary pollutants are defined as those gases emitted directly from the exhaust of a vehicle. None of these is a desirable addition to the atmosphere, but perhaps the most notorious consequence of exhaust emissions is their role in the formation of photochemical smog – a mixture of ozone, nitrogen dioxide, other secondary products and small particulates. These secondary pollutants can cause severe damage to human health.

The role of an emission control catalyst is to simultaneously remove the primary pollutants CO, VOCs and NOx by catalyzing their conversion to carbon dioxide (CO2), steam (H2O) and nitrogen (N2).

4.2 The three-way catalytic converter

4.2.1 Composition

The current three-way catalyst, shown schematically in Figure 1, is generally a multicomponent material, containing the precious metals rhodium, platinum and (to a lesser extent) palladium, ceria (CeO2), γ-alumina (Al2O3), and other metal oxides. It typically consists of a ceramic monolith of cordierite (2Mg.2Al2O3. 5SiO2) with strong porous walls enclosing an array of parallel channels. A typical monolith has 64 channel openings per cm2 (400 per in2), This design allows a high rate of flow of exhaust gases Cordierite is used because it can withstand the high temperatures in the exhaust, and the high rate of thermal expansion encountered when the engine first starts – typically, the exhaust gas temperature can reach several hundred degrees in less than a minute. Metallic monoliths are also used, particularly for small converters, but these are more expensive.

Figure 1
Figure Schematic diagram of the three-way catalytic converter. The catalytic converter, in a metal canister, is placed in the exhaust system of the vehicle. As the exhaust gases pass through it, they flow through the channels in the ceramic monolith, where they encounter the particles of alumina impregnated with the metal catalysts.
Figure 2
Figure 2 Electron micrograph of a cross section of a ceramic monolith coated with an alumina washcoat

To achieve a large surface area for catalysis, the internal surfaces of the monolith are covered with a thin coating (30–50 μm) of a highly porous material, known as the washcoat (Figure 2). The total surface area is now equivalent to that of about two or three football pitches. The washcoat generally consists of alumina (70–85%) with a large surface area, with oxides, such as BaO, added as structural promoters (stabilisers to maintain surface area) and others, for example CeO2, as chemical promoters. This system becomes the support for the precious metal components (Pt, Pd and Rh). These metals constitute only a small fraction (1–2%) of the total mass of the washcoat, but they are present in a highly dispersed form. They are generally applied by deposition from solution, although they may instead be introduced during formation of the washcoat itself. Exact catalyst formulations are, as one might expect, closely guarded secrets. Some compositions use all three metals; others use Rh together with only one of the other two, typically Pt, as in the present generation of Pt-Rh converters used in the UK, in which Pt constitutes 80–90% of the total precious metal mass.

4.2.2 Catalyst performance

Figure 3 shows the difference in the emission levels for CO, VOC and NOx for a vehicle, with and without a three-way catalytic converter. It is evident that the catalytic converter reduces the emissions of all three classes of pollutants quite dramatically over a wide range of speeds. Before we discuss the data in any detail, a few words about how they were obtained are in order.

Figure 3
Figure 3 Emission levels for CO, VOC and NOx for petrol-engined vehicles as a function of speed, with (green) and without (black) a three-way catalytic converter.

Federal and European Test Procedures are used to test emissions from a complete ‘finished’ converter and engine together, to ensure that a new car model, for instance, will meet the current emissions legislation. Some sort of smaller-scale testing is obviously required in the laboratory. In the research and development of automotive catalysts, activity testing fulfils the function of screening and comparing novel and modified catalysts, and examining their performance under different conditions. The process of screening must provide a reliable means of identifying materials that will perform as active, selective and durable catalysts under automotive conditions. The approach usually taken is to measure conversion of the pollutants as a function of temperature, using a simulated exhaust-gas mixture flowing through a bed of powdered catalyst: the flow-rate has to be high enough to mimic the ‘through-put’ or space velocity of a catalytic converter (typically a contact time for the gases with the catalyst of 72 milliseconds is used). The test is then repeated using a different simulated exhaust-gas to represent a different engine mode. Ageing studies are performed by exposing the catalyst to different, and often extreme conditions, for varying lengths of time.

Figure 4 shows a typical graph of catalytic performance over the normal range of operating temperature, 100–600 °C. Until the incoming gases have heated the catalyst to around 250–300 °C, the activity of the catalyst is low. This temperature, at which the efficiency of the catalyst rapidly increases, is known as the light-off temperature. Until this temperature is reached, the catalyst is not working at full efficiency, and so CO, NOx and hydrocarbons will all be emitted from the exhaust pipe in significant amounts. This problem is known as cold start. Ideally the light-off temperature should be as low as possible.

Figure 4
Figure 4 Activity of a three-way catalyst for the simultaneous conversion of CO (black), NOx (solid green) and the hydrocarbon propene (C3H6) (dotted green)

4.3 Exhaust emission characteristics

Before we consider how the three-way catalyst functions in any detail, it is important to understand how the emissions of CO, HC and NOx, from the engine depend on the ratio of air (A) to fuel (F) – the air/fuel ratio (or A/F ratio). The significance of this will become clear when we see that the ratio at which the three-way catalytic converter operates is crucial for its success.


Taking octane (C8H18) to be the only constituent of fuel, and assuming that air is 20% O2 by volume, estimate the stoichiometric A/F ratio (mass ratio) required for total combustion to occur. At this stage neglect the effect of NO as an oxidant. Comment on the difference between the value you obtain and the experimental value of 14.7:1 (Use the following relative atomic masses: C, 12.01; H, 1.01; O, 16.00; N, 14.01.)


The stoichiometric equation for the complete combustion of octane can be written as follows:

C8H18 + 121/2O2 = 8CO2 + 9H2O

So combustion of 1 mol of octane will require 12.5 mol of oxygen.

Assuming air to be approximately 20% O2 and 80% N2 (by volume), the mass of air required will be 12.5 (32.00 + 4 × 28.02) g = 1 801 g.

The mass of 1 mol of octane is (8 × 12.01 + 18 × 1.01) g = 114.26 g.

Thus, the A/F mass ratio for complete combustion is:

A/F = 1 801/114.26

= 15.8:1

This is as close as you would expect to the experimental value of 14.7:1, because we have used a very simplified system. We had not included NO as an oxidant or the other hydrocarbons, CO or H2 as reductants, and we have used octane, not the real mix of hydrocarbons in petrol.

Figure 5
Figure 5 The effect of changing air/fuel ratio on the levels of NOx (solid green), CO (black) and HC (dotted green) produced in the engine. The diagram also shows qualitatively how the engine power output changes with the A/F ratio.

A general relationship between levels of CO, HC and NOx released from the engine and the A/F ratio is shown in Figure 5. At A/F ratios somewhat above stoichiometric (14.7:1) – that is, when the engine is operating under fuel-lean, net oxidising conditions – low levels of HC and CO are produced in the engine, and there is a peak in NOx concentration. At higher A/F values, NOx falls, but the hydrocarbon concentration increases as the engine begins to misfire.

Activity 1

Why do you think the CO and HC levels in Figure 5 increase under fuel-rich conditions; that is, at low A/F ratios?


The levels released in the engine increase because, below the stoichiometric ratio, there is insufficient oxygen present for total combustion

When the exhaust gas is close to its stoichiometrically balanced composition, at an A/F ratio of about 14.7:1, the concentrations of oxidising gases (NO and O2) and reducing gases (HC and CO) are matched; in theory, it should then be possible to achieve complete conversion to produce only CO2, H2O and N2. This is, of course, the objective of the three-way catalytic converter, and so, ideally, it should be operated in a narrow band, or window, close to the stoichiometric ratio, within which it will promote simultaneously the nearly complete reduction of NOx to N2 and the nearly complete oxidation of CO and HC to CO2 and H2O. Figure 6 shows the catalyst conversion efficiency for all three classes of pollutants as a function of A/F ratio, with the dotted lines defining the window for conversions of 80% and above.

Figure 6
Figure 6 Activity of a three-way catalyst for the simultaneous conversion of NOx (solid green), CO (black) and HC (dotted green) as a function of the air/fuel ratio. The shaded area defines the window for conversions of 80% and above for all three pollutants. (Note that, for clarity, the A/F ratios are expressed as the amount of air per unit of fuel, e.g. 14.7 instead of 14.7 : 1. We shall use this notation for the rest of this course.)

Activity 2

Ideally, would we want this window to be as wide or as narrow as possible?


A wide window is desirable for catalytic emission control as it lessens the need to tighten the A/F control of the engine.

This window also happens to correspond closely to the optimum range for high performance of the vehicle (see engine power in Figure 5), which was also of growing importance at the time of development (1970s).

Study comment

You should approach the following SAQ by thinking about how the mixture expelled from the engine will vary depending on the A/F ratio, and the effect that this will have on the balance of reductants/oxidants present. You should consider how efficient the catalyst will be in converting this mixture, and hence, how this will affect the gases finally emitted from the exhaust (that is, leaving the catalyst).


Using the information given in Figures 5 and 6, explain the changes in conversion efficiency seen for all three pollutants when the A/F value is (a) greater than the window for optimum conversion, and (b) less than the window for optimum conversion.


(a) Figure 5 shows that over the narrow range of A/F ratios covered in Figure 6 the amounts of CO and HC emitted form the engine decrease as A/F increases. As there is a simultaneous increase in the total amount of oxidants (air + NOx), the overall conversion of CO and HC increases to approach effective completion at the stoichiometric ratio, and then remains constant in the net oxidising conditions beyond that point. The sharp fall in NOx conversion for A/F values approaching and above stoichiometric is understandable in terms of the virtual elimination of reductants in this region. Because the system is unable to remove all of the NOx, we would expect to see an increase in NOx emissions from the exhaust. The three-way catalytic converter is therefore unsuitable for engines that run lean.

(b) At A/F ratios below the window value there is less NOx and more HC and CO present in the mixture expelled from the engine (Figure 5). All the NOx present will react over the catalyst, so the NOx conversion will still be high (as seen in Figure 6). However, we see a decrease in catalyst efficiency for destroying HC and CO, as there are insufficient oxidants present for complete conversion. We would therefore expect to see an increase in the HC and CO levels emitted from the exhaust.

Obviously in both cases, in the absence of the three-way converter the levels of CO, HC and NOx emitted for the exhaust would be much higher.

Engine control systems have been developed to include an oxygen sensor (or lambda, λ, sensor as it is sometimes called), and an electronic module to regulate the A/F ratio, so that the exhaust composition is kept within the window for optimum conversion. However, because there are time delays in the A/F correction, the ratio cycles very rapidly between slightly fuel-rich and slightly fuel-lean, oscillating about the stoichiometrically balanced composition (14.7 ± 0.3) at a typical frequency of 1 cycle per second. Minimising the amplitude of the oscillation increases the effectiveness of the converter.

4.4 The chemical reactions

4.4.1 Introduction

Since its development, the three-way catalyst has been exposed to the full spectrum of techniques available for the characterisation of catalytic materials. The data provided can be correlated with the results of activity tests and kinetic measurements, which provide information on the performance of the catalyst. This reveals that although the catalyst functions as a composite material, it can be divided into distinct groups of catalytic centres that provide several different types of site, active for one or more of the many different reactions. The participation of a particular type of site at any given moment will depend on the conditions experienced by the catalyst; for example, whether the gases are a net reducing, stoichiometric, or oxidising mixture.

Measurements of intrinsic kinetics are usually carried out on simple gas mixtures to allow activation energies and reaction orders to be calculated for specific reactions. The data can often contribute to an understanding of the mechanisms by which the surface reactions occur. They are also used to create reaction models that will predict the performance of the catalyst under various anticipated conditions.

The overall reaction scheme is complicated, with many contributing processes. The strategy of the three-way catalyst is to simultaneously remove CO, HC and NOx, and our treatment will accordingly be divided into three subsections. The desired reactions can be expressed in simple terms as follows:

Removal of CO

CO oxidation:

Water-gas shift (WGS) reaction:

Removal of hydrocarbons

Hydrocarbon oxidation:

for example:

Steam reforming:

Removal of NO (plus CO or HC (not shown))

CO + NO redox reaction:

or with hydrogen:

Any number of these reactions may be occurring simultaneously as the A/F ratio goes through its cycle about the stoichiometric composition. The following subsections will look more closely at the removal of each of the pollutants under various conditions, and will also examine the role of the catalyst components.

The supported commercial catalyst is the one most difficult to study because of its complexity, with a large number of different components – Pt, Rh, Al2O3, CeO2, BaO, etc. – present in one catalyst. It is therefore often simpler to study model systems, such as Pt/Al2O3 or Rh/CeO2, and if certain surface-science techniques are to be used, the ‘catalyst’ under study has to be even simpler – a particular face of a metal single crystal. These studies, often performed under ultrahigh vacuum (UHV), are far removed from the real catalyst system and the conditions it experiences. Hence, it cannot be assumed automatically that the results will be directly relevant to what is actually happening in a converter fitted to an operational vehicle.

4.3.2 Removal of CO

Under fuel-lean conditions (excess O2), the oxidation of CO has been studied over a very large range of single crystals and model noble metal catalysts, one of the most intensively investigated examples being the Pd(111) surface. Although this metal is not a component of the current three-way catalyst used in the UK, it is worth considering the results in some detail for a number of reasons. The reaction on metals such as Pt is in many ways similar to that on Pd and, in any case, palladium is already being incorporated into future generations of catalytic converter, particularly for the US market. Most notable, however, is the fact that this is one of the few cases in which surface-science techniques have successfully revealed the details of a ‘real-world’ catalytic mechanism. Specifically, we will see how surface studies of the adsorption of CO and oxygen on Pd(111) – both individually and together – have led to the current understanding of the mechanism of CO oxidation.

LEED results for the adsorption of CO on Pd(111), obtained at room temperature and below, have been interpreted in terms of the structural models shown in Figure 7. One of the significant observations from this work is the readiness with which one arrangement of CO on the surface evolves into another. Thus at a surface fractional coverage of θ = 1/3 (Figure 7a), the CO occupies a hollow site where it can bind to three Pd atoms. As θ is increased to 1/2 (Figure 7b), CO moves out of the hollow to a bridging site, where it binds to two Pd atoms. Finally, at θ = 2/3 (Figure 7c), a hexagonal structure forms, in which half of the CO molecules reoccupy hollow sites, while the remainder bind to single Pd atoms at terminal sites. The readiness with which the CO molecules can reposition themselves suggests that the activation energy for surface migration in the chemisorbed state is low, and that CO is a highly mobile species under catalytic conditions.

Oxygen adsorbs dissociatively on Pd(111), and the O atoms are found to be less mobile on the surface than CO molecules. The structure of the chemisorbed layer at maximum coverage is shown in Figure 8.

Figure 7
Figure 7 Structural models for the adsorption of CO on Pd(111) at a surface fractional coverage of (a) θ=1/3; (b) θ=1/2; and (c) θ=2/3.

Identify the adsorbate structure shown in Figure 8 in terms of the (m×n) notation, and determine the fractional surface coverage, θ, of oxygen atoms.

Figure 8
Figure 8 The surface structure of O atoms adsorbed on Pd(111) at maximum surface coverage

The unit meshes of the substrate (1 × 1) structure and of the adsorbate structure are shown in Figure 8A. Evidently, the latter is (2 × 2) in the (m × n) notation, so the full description of this structure should be Pd(111)(2 × 2)–O.

For adsorption on a single crystal surface, the fractional surface coverage is given by

θ = x/(m × n)

where x is the number of adsorbate species within the (m × n) adsorbate unit mesh. In this case (Figure 8A), O atoms occur only at the corners of the mesh, so the latter contains a total of (4 × 1/4) = 1 atom. Hence

θ = 1/(2 × 2) = 1/4

Figure 8A
Figure 8A The surface structure of O atoms adsorbed on Pd(111) at maximum surface coverage, showing the substrate unit mesh and the adsorbate unit mesh.

We might now assume that when CO and O2 are adsorbed together during the oxidation reaction, the properties of the system will be a simple combination of those of the two molecules adsorbed separately. The surface layer would then consist of mobile CO (maximum coverage, θ = 2/3) within a fixed lattice of O atoms (maximum coverage, θ = 1/4). The fact that this is not the case, as we shall see below, demonstrates an important point. Because of mutual interactions, the behaviour of two (or more) co-adsorbed species very often differs from their behaviour when adsorbed separately.

In the case of CO and O2, the order in which adsorption is carried out is significant. If CO is adsorbed first, to a coverage greater than one-third of a monolayer (θ = 1/3), subsequent oxygen adsorption is completely blocked. With lower coverages of CO, dissociative oxygen adsorption does occur; but the two species form separate domains on the surface (Figure 9a). Oxidation will then take place only at the boundaries between domains, and so it will be relatively slow.

Figure 9
Figure 9 Schematic representation of domains of CO(ad) and O(ad) on Pd(111). (a) Separate domains (CO adsorbed first); (b) mixed domains (O2 adsorbed first).

When oxygen is adsorbed first to its maximum coverage, θ = 1/4, subsequent CO adsorption occurs readily and compresses the O atoms into domains in which the local coverage reaches θ = 1/3. At first, the adsorbed CO is found in separate areas (as when CO is adsorbed first), but as more is added mixed domains form, containing both CO and O, each at a local coverage of θ = 1/2 (Figure 9b).

These mixed domains bring CO(ad) and O(ad) into intimate contact, with the O atoms at twice the surface concentration possible in the absence of CO. Thus, the stoichiometry is now that required for the oxidation reaction. Moreover, an electronegative O atom will withdraw charge from the surface. In turn, the surface will withdraw charge from neighbouring Pd–CO bonds, weakening them and so making the CO more readily available for reaction. The net result is that the mixed domain is highly reactive and generates CO2 at temperatures far below room temperature.

Having thus established that, on Pd(111), rapid CO oxidation can occur by way of a Langmuir–Hinshelwood type process (a surface reaction between two adsorbed species), we are almost in a position to propose a detailed mechanism. First, however, we must consider the possibility of an alternative Eley–Rideal type mechanism, in which the rate-limiting step involves reaction between an adsorbed species and a molecule in the gas phase. In this case, there are two such possibilities:

Activity 3

Experiments show that exposure of a surface saturated with adsorbed CO to gas-phase O2 does not lead to CO2 formation. How does this simplify consideration of the above reactions?


This observation rules out the Eley–Rideal reaction between CO(ad) and O2(g), reaction 8.

We can make a decision about reaction 9 on the basis of the information provided in Figure 10. A Pd(111) surface presaturated with a quarter of a monolayer of O atoms was exposed to a beam of gaseous CO, and the surface coverages and the oxidation rate were monitored with time. Figure 10 shows that the rate became significant only after a population of CO had built up on the surface, and it reached a maximum when the coverages of O and CO were approximately equal. This is clear evidence for a Langmuir–Hinshelwood process. If reaction 9 had been operative, the rate would have been high initially, and would have fallen continuously as the oxygen layer was consumed by the reaction.

Figure 10
Figure 10 Changes in the rate of CO2 formation from CO and in the surface density of oxygen and CO on Pd(111). The surface was precovered with a quarter of a monolayer of oxygen atoms at time zero, and then exposed to a constant stream of CO at a pressure of 7.9×10−11 atm.

Given the research effort that was involved, the mechanism finally proposed for CO oxidation on Pd(111) is deceptively simple:

Although the exact nature of the surface intermediates is still not known, the depth of understanding of the catalytic mechanism is quite an accomplishment. But just how relevant are these surface studies to the practical catalysis taking place in the converter?

Activity 4

How do the observations outlined below lead to the conclusion that studies of CO oxidation, such as those for Pd(111) above, using surface-science techniques such as LEED, can be of direct relevance to the catalysis taking place in the converter?

  • During the reaction, surface contaminants such as carbon or sulfur are burnt off the Pd catalyst, so that the surface is clean, with the exception of the adsorbed reactants;

  • The catalytic activity does not depend on the crystallographic structure of the Pd metal surface. This is illustrated in Figure 11 where the rate of CO oxidation, over a wide temperature range, is shown to be virtually identical on five different Pd surfaces;

  • The rate of CO oxidation is independent of pressure, depending only on the ratio of the partial pressures of CO and O2, at least over a restricted range of conditions.

Figure 11
Figure 11 Rates for the catalytic oxidation of CO over a variety of different Pd surfaces.

Techniques such as LEED examine adsorption at very low pressures onto clean single crystal surfaces, but the real catalytic system is more complicated than this. We are not dealing with a single crystal or a defined crystal plane, and the pressures involved are much higher. However, according to the factors listed above, CO oxidation has been shown to be independent of the Pd surface exposed, and independent of the total pressure of the reactants. In addition, the first observation suggests that under real conditions, the catalyst surface is clean; hence the effects of the adsorbed CO and O are likely to be the only relevant factors. Thus, the conclusions drawn about mechanism from surface studies may not be too far removed from those that apply in the case of the ‘real’ catalyst.

Indeed, Figure 12 shows that in the case of rhodium there is excellent agreement between the rates of CO oxidation over a Rh(111) single crystal surface and over a Rh/Al2O3 catalyst.

Figure 12
Figure 12 Comparison of the rates for CO oxidation measured over Rh(111) (black) and over 0.01 mass % Rh/Al2O3 (green) at p(CO)=p(O2)=0.01 atm, as a function of temperature.

Although the sequence of elementary steps is quite simple, the overall kinetics of the CO oxidation reaction is not. The non-uniformity of the surface, and the segregation of the reactants in surface domains, complicates the detailed modelling of the kinetics. The exception is the special case of low surface coverages of CO and O atoms, when they are found to be randomly distributed over the surface and so satisfy one of the criteria for applicability of the Langmuir isotherm. Under these circumstances, Langmuir–Hinshelwood kinetics can be applied.

Figure 13 shows the comparative performance of single-metal catalysts for the oxidation of CO at a fixed temperature. Evidently, all three of the platinum group metals present in automotive catalysts are active for CO oxidation. In addition, results have shown that Rh may improve low-temperature activity. In the current (1996) three-way catalyst used in the UK, in which Pt constitutes 80–90% of the noble metal composition and Rh the remainder, it is the Pt that is mainly responsible for CO oxidation.

Figure 13
Figure 13 Comparison of catalytic activity for CO oxidation at 400 °C for Pt, Pd and Rh at different A/F ratios

Under stoichiometric or slightly fuel-rich (reducing) conditions, where there is insufficient oxygen present to oxidise all of the CO, conversion can also occur by one of the following routes:

  • via the CO + NO redox reaction (reaction 6). This will be discussed in detail in section 4.4,

  • via the water-gas shift reaction (equation 2), because H2O is present in the exhaust gases as a product of combustion:

The water-gas shift reaction is catalysed by Pt and/or Rh, with ceria acting as an excellent promoter. Pt/CeO2–Al2O3 and Pt-Rh/CeO2–Al2O3 are particularly active combinations for the removal of CO under slightly fuel-rich conditions. The hydrogen produced in this reaction will react, in preference to CO, with any oxygen present. Hence, although the water-gas shift reaction removes CO, it also inhibits CO oxidation by producing hydrogen, which will remove any O2 present:

4.3.3 Removal of hydrocarbons

Figure 14 shows a comparative study for hydrocarbon oxidation over single-metal catalysts: it can be seen that Rh, Pd and Pt all give high conversions for A/F ratios at and above stoichiometry. Again (as in the case of CO), in the current (1996) UK three-way catalytic converter, Pt is the main component responsible for oxidation of the hydrocarbons. On noble metal surfaces, alkane adsorption is dissociative, whereas unsaturated and aromatic hydrocarbons adsorb either dissociatively or associatively as -complexes. The subsequent oxidation process is thought to be considerably more complicated than the oxidation of CO, and we shall not consider it in any detail.

Figure 14
Figure 14 Comparison of catalytic activity for HC oxidation at 400 °C for Pt, Pd and Rh at different A/F ratios

When the engine exhaust gas composition is reducing (fuel-rich), hydrocarbons compete effectively with CO for oxygen, and they can also react with water vapour to produce CO and H2 – a reaction known as steam reforming:

This is catalysed by Rh and/or Pt with ceria and, as in the case of the water-gas shift reaction, the combination Pt–Rh/CeO2–Al2O3 is particularly active. As we noted earlier, the H2 produced may react preferentially with any O2 present, thus reducing the amount of oxygen available to react with hydrocarbons and CO. In addition, the CO produced adds to the burden of carbon monoxide to be removed.

4.3.4 Removal of NO

Laboratory experiments have shown that, under the conditions in the catalytic converter, the decomposition of NO to O2 and N2 over noble metal catalysts is too slow to be significant. When the A/F ratio is stoichiometric (or below stoichiometry), NO can be removed by reduction with CO and/or hydrocarbons. For simplicity we shall consider only reduction with CO, as with the oxidation reaction, the situation with hydrocarbons is considerably more complicated.

In principle, a variety of products can be formed, specifically:

In addition, H2 produced from the water-gas shift or steam reforming reactions can reduce NO to N2, N2O or NH3:

Activity 5

Which reactions should a catalyst ideally promote?


The aim of the catalyst is to selectively promote reactions 6 and 7 to produce N2, rather than N2O (a greenhouse gas) or NH3 (a potentially serious general pollutant).

The NOx activities of Rh, Pt and Pd are shown in Figure 15. It is evident that Rh has the highest activity, particularly under net reducing conditions (low A/F). So why is Rh superior? To answer this question, we need to consider the mechanism of the reaction.

Figure 15
Figure 15 Comparison of catalytic activity for NOx reduction at 400 °C over Rh, Pt and Pd at different A/F ratios.

The catalytic reduction of NO by CO and/or H2 over a variety of surfaces has been the subject of a great deal of research. Application of various surface-science techniques has provided some understanding of the elementary steps involved, but the exact mechanism is still controversial. One view is that the first step is the dissociative chemisorption of NO. (You should note that although we describe this as dissociative chemisorption strictly it does not meet the definition, as NO is in fact first adsorbed associatively and then dissociates on the surface.) The O atoms produced are then removed by the reducing agents CO or H2. The N atoms can combine to give N2, react with chemisorbed NO to give N2O (particularly important at low temperatures), or react with chemisorbed H atoms to form NH3. These and other processes that may be involved are listed below.



NO(ad) N(ad) + O(ad)

In the following steps we have assumed, for simplicity, that all products are desorbed as quickly as they are produced. You should recognise, however, that adsorbed species, no matter how transient, will be formed initially.

Surface reactions and desorption

Reactions with hydrogen

Figure 16 compares the rate of the NO–CO reaction over a Rh(111) single crystal with that over a Rh/Al2O3-supported catalyst.

Figure 16
Figure 16 Comparison of the rates for the NO–CO reaction measured over Rh(111) (black) and over 0.01 mass % Rh/Al2O3 (green) at p(CO)=p(NO)=0.01 atm, as a function of temperature
Activity 6

After the reaction on Rh(111), the surface was found to have a high coverage of N atoms. What does this suggest about the mechanism?


It suggests that N atom combination (reaction 21) may be the rate-limiting step in the overall NO–CO reaction on Rh(111).

This seems to be the case, particularly at higher temperatures. At lower temperatures, the concentration of NO(ad) increases and reactions 22 and 23 would then be expected to contribute to N atom removal.

The elementary steps 21–23 all require surface mobility of N atoms (to encounter either NO(ad) or other adsorbed N atoms). Although this process may occur on surfaces that are extensive on the atomic scale, such as those of single crystals or large supported crystallites, it has been argued that such mobility will be insignificant on or between the small highly dispersed particles of Rh in the automotive catalyst. Therefore, we might expect the rate-limiting steps and the observed kinetics in the cases of Rh(111) and supported Rh to be different. The Arrhenius-type plots in Figure 16 confirm that this is so: over Rh/Al2O3, the reaction has a higher activation energy (the plot in Figure 16 has a larger gradient) and a lower rate (at a given temperature) than over Rh(111).

What then is the rate-limiting step with the supported catalyst? One suggestion is NO dissociation (reaction 19) but there is a more radical alternative, involving a different overall mechanism. Infrared spectra for NO adsorbed on Rh/Al2O3 (Figure 17) show bands at 1 743 cm−1 and 1 825 cm−1, which have been taken as evidence of a dinitrosyl species, O=N–N=O, formed by reaction 26:

Figure 17
Figure 17 Infrared spectra, recorded at 300 K, for NO adsorbed on Rh/Al2O3 as a function of NO coverage, increasing from spectrum A to spectrum D. The bands at 1 743 cm−1 and 1 825 cm−1 have been assigned to the dinitrosyl species O=N–N=O.

This step provides a means, other than diffusion of N(ad), of accomplishing the most important task in the reduction of NO, namely the pairing of two nitrogen atoms on the surface. Once formed, the dinitrosyl species is thought to lose its two oxygen atoms by way of an N2O intermediate; for example:

To summarise Whichever mechanism is correct – NO pairing to form a dinitrosyl species, or NO dissociation followed by N-atom combination and N2 desorption – both require catalytic sites that can not only bind NO but also donate charge to the adsorbate. In the first case, this charge would be used to coordinate the two NO molecules. In the second case, it would be transferred into the partially vacant 2* antibonding orbital of NO (Figure 18), weakening the N–O bond and hence facilitating dissociation.

On examining the electronic structures of the noble metals, that of rhodium is found to be particularly suitable for facilitating charge transfer to adsorbed NO, with the uppermost occupied electron levels of the metal at higher energy than the partially vacant 2* antibonding orbital of NO (Figure 18). For Pt (and also Pd), however, the situation is reversed. Vacant metal levels lie at energies below that of NO 2*, so charge will drain from this orbital to the surface, strengthening the N–O bond.

This picture provides an appealingly simple explanation for the observation that rhodium is the most active of the noble metals for NOx removal, particularly under-net reducing conditions (Figure 15). In fact, not only do Pt and Pd show lower activities, they are also less selective, with Pt, for example, promoting ammonia formation (reaction 17). Rhodium is therefore the essential ingredient in the automotive catalyst for NOx control. Because it is so active, the amount required is small – about 0.1 mass % of the catalyst, or 0.1–0.2 g, highly dispersed over the surface – but even so, catalytic converters account for around 90% of world rhodium demand (see Box 1).

Figure 18
Figure 18 (a) The partial orbital energy-level diagram of molecular NO. (b) Schematic energy-level diagram showing the highest occupied level of Rh and the lowest vacant level of Pt, in relation to the 2π* molecular orbital of NO.
Box 1 Supply and demand of precious metals

Pt, Pd and Rh form part of the platinum group metals, along with Ru, Os and Ir; they are closely related in terms of chemical and physical properties, and occur together in nature. The metals are all quite rare, and are found in only a few comparatively concentrated deposits – principally in South Africa, the former USSR, Canada and the USA. In South Africa, deposits are mined specifically for the platinum group metals, whereas in Canada and the former USSR, they are produced as byproducts from nickel and copper mining.

Separation of the metals is a long and difficult process because they, and their compounds, are so chemically similar in nature. This difficulty in separation, together with the costs incurred in refining the metals to the high purity required, makes their production an expensive business. In addition, huge amounts of the ore have to be mined and processed; for example, it takes approximately 350 kg of South African ore to yield just 1 g of platinum.

Rhodium is essentially a by-product of platinum mining, and is by far the most expensive metal in the platinum group, with a price of approximately $18 250 per kg in July 1995 (compared with Pt ($15 300) and Pd ($5 500)). Because the ratio of Rh to Pt used in automotive catalysts is richer than the mine ratio, the shortfall in Rh came, until recently, from Pt mined for other uses, for example, jewellery. Automotive catalysts are a major application for platinum group metals, the percentage of total demand being 87% for Rh and 34% for Pt (1992). Up to date prices of precious metals are always available on the world wide web.

4.3.5 Postscript

It is important to remember that the reactions discussed in sections 4.2–4.4, and often studied separately, occur simultaneously in the presence of all the other exhaust constituents. This may have an effect on the efficiency of the individual reactions. For example, kinetic experiments have demonstrated that NO has a strong inhibiting effect on the rate of the reaction of CO with O2: Figure 19 illustrates the point for a Rh/Al2O3 catalyst.

Figure 19
Figure 19 The temperature dependence of CO conversion rates over 0.01 mass % Rh/Al2O3 in CO–O2, CO–NO–O2 and CO–NO mixtures The concentration of each of the reactants was 0.5 vol % in all cases.
Activity 7

With reference to Figure 19, compare the temperatures at which each reaction reaches a rate of 1×10−6 mol CO (g cat)−1 s−1 (the dotted line)


The green line labelled CO–O2 in Figure 19 shows the CO conversion rate for the CO–O2 reaction in the absence of NO. This approaches a value of 1×10−6 mol CO (g cat)−1 s−1 at about 230 °C. When NO is present (the CO–NO–O2 mixture) the overall CO conversion rate decreases, and approaches a value of 1×10−6 mol CO (g cat)−1 s−1 only at about 290 °C. In fact, this is close to the temperature at which the CO–NO reaction, shown as the dashed line in Figure 19, becomes significant. Therefore, the inhibition of the CO–O2 reaction is believed to be due to blocking of the active sites by adsorbed NO.

4.3.6 The role of CeO2

Figure 20 shows the effect on performance of adding CeO2 to a Pt catalyst for three-way catalytic conversion.

Figure 20
Figure 20 The effects of adding CeO2 to a Pt catalyst in terms of the simultaneous conversion of CO, HC and NOx. The green lines show the conversions for the Pt catalyst with CeO2 and the black lines those without.
Activity 8

What is the effect of CeO2 on the conversion efficiencies for CO, hydrocarbons and NOx?


Substantial improvements in all three conversion efficiencies are seen, particularly at A/F ratios just below stoichiometry.

Thus, ceria, which is added with the alumina in the washcoat, is an essential ingredient of the three-way catalyst. It plays a number of roles:

1: Ceria is a structural promoter, stabilising the precious metals and alumina against sintering and particle growth. Figure 21 emphasises this point.

Figure 21
Figure 21 The variation of dispersion stability as a function of ageing temperature, with (green) and without (black) the addition of CeO2 to 0.9 mass % Pt/Al2O3
Activity 9

How does the addition of ceria to the catalyst (the coloured line in Figure 21) affect the Pt dispersion?


Clearly, in the absence of ceria, substantial sintering of Pt occurs between 500 °C and 600 °C, causing a sharp reduction in dispersion. Addition of CeO2 results in a significant stabilization of the Pt metal dispersion up to 700–800 °C.

Ceria also stabilises the γ-Al2O3 used in the support, inhibiting a phase change to -Al2O3, which has a lower surface area. (Lanthanum oxide and/or barium oxide are also often added as stabilisers to help maintain the surface area of γ-Al2O3.)

2: Ceria is known to be able to pick up and store oxygen from the gas phase under fuel-lean operating conditions (excess oxygen) – thus promoting the reduction of NO to N2 – and to release it under fuel-rich conditions (excess fuel), for reaction with CO, H2 or hydrocarbons. Thus, it effectively dampens the variations in the A/F ratio as the exhaust gas mixture cycles about stoichiometry, thereby helping to keep operation within the desired window for optimum conversion over the catalyst.

3: As we have seen in section 4.2, the ceria also enhances the water-gas shift activity of Pt–Rh three-way catalysts, and hence promotes CO removal via the following reaction under fuel-rich conditions:

The point is illustrated by the results shown in Table 1 for CO conversion under fuel-rich conditions. Increasing the ceria content of the catalyst in the absence of water has no effect, but when water is present the water-gas shift reaction becomes increasingly important. (Addition of ceria also leads to an improvement in activity for steam reforming.)

Table 1 CO conversion for a 1.08 mass % Pt-Rha catalyst on γ-Al2O3 with 1.5, 4.0 and 8.0 mass % Ce levels, under fuel-rich conditions, with and without water present. (aPt 0.9 mass % and Rh 0.18 mass %.)
CO conversion/%
With H2OWithout H2O
1.5 mass % Ce5449
4.0 mass % Ce6449
8.0 mass % Ce7049

4: Enhanced conversions of CO, C3H6 and NO at low temperatures have also been observed for Pt/CeO2 catalysts that have undergone a reducing pretreatment. This is believed to be due to an interaction between Pt and CeO2 induced by the reduction, causing an increase in the number and activity of the active sites.

4.3.7 Chemical reactions: summary

  1. Surface studies of the adsorption of CO and O2 on single crystals and model catalysts have led to the development of a possible mechanism for the oxidation of CO. Dissociatively adsorbed O atoms undergo a surface reaction with adsorbed CO, to form CO2.

  2. Under slightly fuel-rich conditions, where there is insufficient oxygen present for complete oxidation, CO can be removed by the water-gas shift reaction, using water produced in the combustion process in the engine. This is promoted by ceria.

  3. Hydrocarbons can be removed by oxidation or by reaction with water (a process known as steam reforming).

  4. Both CO and hydrocarbons can be removed by reaction with NO under stoichiometric or fuel-rich conditions. The NO-CO redox reaction is believed to proceed either by dissociation of NO(ad) followed by N atom combination, or by pairing of NO(ad) to give a dinitrosyl species, followed by dissociation. Whatever the detailed mechanism, Rh is particularly active for this reaction, and as such is currently an essential ingredient of the three-way catalyst.

  5. Ceria plays a number of important roles in the three-way catalyst: it is a structural promoter, stabilising both the noble metals and the support against particle growth and sintering; it is an oxygen-storage component, storing oxygen under fuel-lean conditions, and releasing it under fuel-rich conditions; it is a promoter for the water-gas shift and steam reforming reactions and it can enhance the low-temperature activity of the catalyst after certain types of pretreatment.

Study comment

The following SAQs invite you to collect together the different roles proposed for each component of the three-way catalyst. It would be a good plan to try them, and check your answers, before moving on.


What are the major roles proposed for the different components of the Pt-Rh/CeO2–Al2O3 three-way catalyst?


The major roles proposed for the different components of the Pt-Rh/CeO2–Al2O3 three-way catalyst can be summarised as follows:


  • CO oxidation

  • HC oxidation

  • steam reforming and water-gas shift reactions (with CeO2)


  • CO–NO redox reaction

  • steam reforming and water-gas shift reactions (with CeO2)


  • oxgen storage

  • structural promoter maintaining both metal and support surface areas

  • chemical promoter for the water-gas shift and steam reforming reactions

  • enhancement in low-temperature activity of Pt after reducing pretreatment


In Figures 13, 14 and 15, Pd in its fresh state is seen to be superior to Pt for the conversion of all three pollutants. Considering the properties we require of a catalyst, what possible reason can you suggest for the current widespread use of Pt in three-way catalysts, rather than Pd?


Pd is in fact cheaper than Pt (Box 1), so the reason for using Pt is not an economic one. The principal properties we require of a catalyst are activity, selectivity and stability. Because we have seen that the first two of these are at least as good for Pd as for Pt, this should lead us to consider the third, stability. In fact, as we shall discover in section 5.3, Pd has a high susceptibility to poisons, especially lead and sulfur. In addition, it sinters in reducing atmospheres, and can also form an alloy with Rh, reducing the activity of the latter.

4.5 Catalyst deterioration

4.5.1 Introduction

In the UK, three-way catalysts must currently (1996) meet emission standards for a life of 50,000 miles; however, research efforts and legislation are set to double this requirement in the very near future to the current US standard of 100,000 miles. The catalysts do deactivate with use. Indeed the ability to withstand mild deactivation is built into the design of the catalyst, and into the entire emission control system in the vehicle. This is done by setting up vehicles at efficiencies well above the legal requirements at low mileage, so that as the catalyst slowly deactivates, it will still meet the emission standards.

However, the catalyst may be exposed to conditions that result in more severe deactivation above and beyond that which is ‘allowed for’ in its lifetime. The major causes of deterioration are thermal damage (due to exposure of the catalyst to high temperatures), and poisoning by contaminants in the exhaust (notably phosphorus, lead and sulfur). Research aimed at detecting deterioration, and trying to understand its nature, has included post-mortem examinations of used catalysts, and simulated ageing studies, in which the catalyst is exposed to high temperatures or catalyst poisons.

4.5.2 Thermal effects

Exhaust catalysts usually operate in the temperature range 150–600 °C, but they can experience temperatures of up to 1000 °C. The conditions that can produce such high temperatures include repeated misfire (resulting in the oxidation of large amounts of unburned fuel over the catalyst) and high driving speeds. In addition, very high temperatures may be experienced if the catalyst is ‘close coupled’ to the engine, which is one of the possible solutions to the cold-start problem. Although commercial catalysts are designed to withstand occasional high-temperature operation prolonged and repeated exposure to temperatures in excess of 800 °C, especially under oxidising conditions, have a number of serious effects.

High temperatures may affect all the components of the catalyst. The noble metal particles may sinter (recall Figure 21 in section 4.6), resulting in a decrease in the fraction of the metal available for catalytic reactions. Such sintering particularly affects the low-temperature activity of the catalyst. It can be countered to some extent by the addition of ceria as a structural promoter, which also stabilises the alumina support against sintering. However, the ceria may itself undergo crystallite growth at elevated temperatures. This can be inhibited by the addition of barium and zirconium, as shown by the comparison in Figure 22: it is apparent that barium and zirconium help to stabilise the catalytic activity.

Figure 22 Comparison of catalytic activity for a Pt–Rh/CeO2 catalyst with (green) and without (black) the addition of Ba and Zr, after ageing at A/F=16.5 at 950 °C for 40 hours.

At very high temperatures the support may itself sinter or even undergo a phase change, affecting the total surface area. Mechanical loss of catalyst support material may also occur in cases where the washcoat shrinks and cracks, causing it to separate from the monolith (Figure 23a). Again, the problem can be overcome by incorporating so called ‘phase stabilisers’ – examples include barium and lanthanum – into the washcoat (Figure 23b). At excessively high temperatures the ceramic monolith may even melt, forming additional channels that may allow the exhaust gases to pass through the converter without contacting the catalyst.

Figure 23
Figure 23 Optical micrographs of a washcoat after exposure to high temperatures. In (a) sintering has caused severe shrinkage, whereas in (b) the addition of ‘phase stabilisers’ has prevented shrinkage cracking.

High temperatures can also promote damaging interactions between the noble metals, resulting in the formation of a less active alloy, or between a noble metal and base metals in the washcoat support. In particular, it has been established that Rh begins to penetrate the surface of γ-Al2O3 at temperatures greater than 600° C by a solid state reaction between Rh2O3 and γ-Al2O3. This subsurface penetration and loss of active Rh can be slowed down if the reactivity of the support is minimised by first supporting the Rh on zirconia, ZrO2, and then incorporating the resulting powder into the γ-Al2O3 washcoat. Figure 24 shows the dramatic effect this can have on the catalytic activity of a Rh/γ-Al2O3 catalyst. Unfortunately, the incorporation of Rh/ZrO2 into three-way catalysts requires complex manufacturing methods, which are not yet suitable for high-speed production. An alternative approach has been suggested by work that indicates that the Rh/Al2O3 interaction may occur preferentially at the grain boundaries of the support. Ceria can be incorporated as a stabilizer into the alumina in an attempt to preferentially block this interaction.

Figure 24
Figure 24 Schematic comparison of the catalytic activity of (a) Rh/γ-Al2O3 and (b) [Rh/ZrO2]/γ-Al2O3 after thermal treatment at 1 100 °C in air for 1 h. The activity of the ZrO2-supported catalyst (b) remains virtually intact after the calcination, whereas that of the standard Rh/γ-Al2O3 catalyst (a) is seriously diminished.

4.5.3 The effect of poisons

The use of catalytic converters was one of the major contributors to the phasing-in of unleaded petrol. Lead in petrol is a severe poison for the catalyst, and there have been many stories, particularly in the early days of the converter, of people disabling the catalyst by misfuelling. Figure 25 shows how the activity of a typical three-way catalyst is impaired during, and following, intermittent operation with leaded fuel (0.26 g l−1) during 15,000 miles of vehicle operation. The efficiency of a control vehicle (unleaded petrol only) was virtually unchanged at 94% for hydrocarbons, 95% for CO and 66% for NOx. Following the misfuelling, the CO-conversion efficiency (Figure 25a) decreased, but subsequently recovered to an acceptable level. By contrast, the conversion efficiencies for the hydrocarbons (Figure 25b), and especially NOx (Figure 25c), did not recover to passable values, and hence did not meet the emission regulations current at the time (1986).

Figure 25
Figure 25 Effect of intermittent use of leaded petrol (black points) on catalyst efficiency (for a typical three-way catalyst) for the conversion of (a) CO (b) HC and (c) NOx.

Of the various noble metal components, Pd is the most sensitive to lead poisoning. Its activity decreases when there are just trace amounts in the fuel. Rh is slightly less susceptible, and Pt is by far the most resistant. Clues to the mechanism of lead poisoning have come from model systems, which are amenable to detailed surface analysis.

Activity 10

Figure 26 shows electron probe elemental maps of Pt/γ-Al2O3 after exposure to a simulated exhaust gas mixture containing 0.33 g l−1 of Pb. With which element, Pt or Al, is Pb associated?


Pb is associated with Pt, because the Pt and Pb maps are exactly superimposable. Similar results are obtained whether the metal is Pt, Pd or Rh.

Figure 26
Figure 26 Electron probe elemental maps of Pt/γ-Al2O3 after exposure to Pb for 24 hours at 700 °C. Electron probe microanalysis (EPMA) is based on the emission of ‘characteristic’ X-ray photons following the ionisation of atoms by high-energy electrons. EPMA allows quantitative determination of the composition of solids with a resolution of the order of 1 μm. The whiter the area on the map, the higher the concentration of the element under study. The X-ray spectrophotometer is a common attachment on scanning and transmission electron microscopes.

This deposition of lead specifically onto the noble metal is believed to occur because the molecules that ‘carry’ the lead out of the engine, probably halides or oxyhalides, decompose on the noble metal, leaving the lead on the surface.

The fact that Pt is more resistant to lead poisoning than Rh or Pd is largely due to an indirect effect. The small amount of sulfur also present in fuel can act as a scavenger for lead. Provided that the sulfur is in its hexavalent oxidation state (SVI), in the form of SO3, it can combine with lead oxide to form a stable lead sulfate, which, although a poison itself, is not site-specific. Only Pt, however, is a good catalyst for the oxidation of SO2 (produced from sulfur in the combustion reaction, and present in the exhaust mixture) to SO3: indeed, it is used for this purpose in the industrial production of sulfuric acid.

4.5.4 Phosphorus

Phosphorus is recognised as a potential poison for automotive catalysts. The phosphorus level in fuel is generally very low (2×10−5 g l−1), but it is present in higher concentrations in engine oils (1.2 g l−1). Phosphorus derived from the engine oil is believed to react with the alumina support, and also to reduce the activity of the noble metal component. This deactivation is particularly important for Pd, with which phosphorus may form an alloy. At the time of writing, phosphorus levels in engine oil are becoming an issue, and oils with reduced levels are appearing on the market.

4.5.5 Sulfur


The presence of sulfur in the exhaust gas mixture causes a reduction in the activity of the three-way catalyst, particularly for the water-gas shift and steam reforming reactions – processes that are important mechanisms for the removal of CO and hydrocarbons under fuel-rich conditions. Sulfur also decreases the efficiency of NOx removal. The deleterious effect of exposure to SO2 on the catalytic activity of a commercial monolithic catalyst (Pt–Rh/CeO2–Al2O3) is evident in Figure 27. Notice, however, that the conversion efficiency recovers quite rapidly once the sulfur has been removed from the gas stream. This is significant because it suggests that a change in the sulfur content of fuel (average 208 mg l−1 in the UK), could achieve a reduction in emissions from vehicles currently in use.

Figure 27

This element is no longer supported and cannot be used.

The general resistance of Pt and Rh to sulfur poisoning, and their ability to recover from it, were two of the factors in the original decision in the 1970s to use noble metals rather than the less active, but cheaper, base-metal oxidation catalysts. (In terms of the more stringent legislation now in force, let alone that which will apply in the future, it is unlikely that base-metal activity (or durability) would be sufficient to meet emission control requirements, even if all the sulfur in fuel were to be removed.) Pd is more susceptible to long-term damage, and it is this susceptibility to poisoning that limits its use in the UK at the moment. Pd-only catalysts are, however, under development for the US market – where there are stricter controls over the contaminant levels in fuel.

Some understanding of the mechanism of catalyst deactivation by sulfur has been obtained by examining the effects of SO2 on the surface area of metals The physical adsorption of N2 can be used to determine the ‘total’ surface area of a catalyst. The capacity of a supported metal catalyst to chemisorb CO can be used as a measure of the free metal surface area. It is typically expressed as the ratio CO/M, where CO is the amount of CO chemisorbed by a fixed mass of catalyst (proportional to the number of surface metal atoms), and M is the metal content of the catalyst (proportional to the total number of metal atoms). The higher the value of CO/M, the higher the surface area of the metal. Table 2 gives values of the CO chemisorption capacity of a Pt–Rh/CeO2–Al2O3 model catalyst after ageing in a fuel-rich mixture, in the presence and absence of SO2.

Table 2 Effect of fuel-rich ageing on the metal-surface area (expressed as CO/M) for a Pt–Rh/CeO2–Al2O3 catalyst, in the presence and absence of SO2. The value of CO/M for a fresh sample is 0.79.
Ageing temperature/°CCO/M after ageing
In the presence of SO2In the absence of SO2
Activity 11

What is the effect of SO2 on the chemisorption capacity of the catalyst? What do you suggest is happening to the metal surface in the presence of SO2?


At a given ageing temperature, the CO chemisorption capacity of the catalyst is dramatically decreased in the presence of SO2. This suggests that some sort of sulfur species has been formed on the metal surface, blocking sites previously available to chemisorb CO, and possibly also inhibiting CO adsorption at neighbouring sites.

Because the effects are not permanent, this sulfur species is presumably relatively weakly adsorbed. As a result it is desorbed, and the activity of the catalyst is regenerated, when the gas-surface equilibrium is shifted in favour of the gas-phase by removal of SO2 from the gaseous mixture (Figure 27).

The production of hydrogen sulfide

Recently, there has been a great deal of interest in the interactions of sulfur with the three-way catalyst, not so much because of its impact on activity, but rather because of a smelly side-effect. Ever since catalytic converters were introduced, the odour of hydrogen sulfide (H2S), described as ‘smelling like rotten eggs’, has been an issue. This is particularly noticeable when a car in front accelerates after idling (at traffic lights or a roundabout, say) or decelerates sharply after cruising. After combustion in the engine, sulfur in the fuel is released to the exhaust gases as SO2. Under fuel-rich (reducing) conditions, this is converted into H2S over Pt, the suggested mechanism involving the formation of the metal sulfide as a reaction intermediate. It has been found, however, that the amounts of H2S generated when engine conditions become fuel-rich, following prolonged running under fuel-lean (or oxidising) conditions, are larger than expected. The catalyst can apparently ‘store’ sulfur under lean conditions and release it under rich conditions.

The noble metals do not retain any sulfur under these conditions. However, adsorption studies with both model catalysts and the commercial (fully formulated) catalyst have shown that sulfur storage under lean conditions can occur by interaction of SO2 with both CeO2 and Al2O3 in the support. Under typical conditions, CeO2 provides the preferred adsorption sites, and the sulfur storage by the catalyst has been found to increase with increasing Ce content. XPS studies (some results of which are collected in Table 3) have been used to study the nature of the species involved.

Table 3 XPS data recorded for various cerium and sulfur compounds.
SampleHistoryBinding energy/eV
Ce 3dS 2p
CeO2550 °C, air881.7
CeO220 °C, SO2881.6168.0
CeO2550 °C, SO2882.8168.8
Ce(NO3)3.6H2O 883.3
SVI compounds 168.5–168.9
SIV compounds 166.3–167.9
Activity 12

Comparing the S 2p binding energies with the typical values for SVI and SIV compounds included in Table 3, how would you assign the sulfur species present when CeO2 is exposed to SO2 at 20 °C and 550 °C?


The typical values suggest that at 550 °C, the sulfur exists as SVI. The value at 20 °C is slightly harder to assign, but it may be attributed to SIV associated with a highly charged cation such as CeIV or CeIII.

Activity 13

Comparing the Ce 3d binding energies with those for CeO2 in air and for Ce(NO3)3.6H2O, what do you conclude about the Ce species present when CeO2 is exposed to SO2 at 20 °C and 550 °C?


The Ce 3d binding energy of CeO2 in SO2 at 20 °C is similar to that of CeIV oxide (CeO2 in air). The value at 550 °C agrees reasonably well with that for CeIII in Ce(NO3)3.6H2O, suggesting that reduction of some of the surface sites to CeIII may have occurred.

Hence, interpretation of the XPS data suggests that, at room temperature, SO2 is adsorbed on CeO2, possibly to form a sulfite species (that is, a species containing SIV). At 550 °C the CeIV oxide, CeO2, appears to participate in a redox reaction with SO2; the sulfur is oxidised to SVI and the cerium is reduced to CeIII, possibly resulting in the formation of cerium(III) sulfate, Ce2(SO4)3.

The sulfate/sulfite species formed under fuel-lean conditions with the cerium and aluminium in the support decompose in fuel-rich conditions to release the SO2/SO3 species, which are then converted into H2S over the noble metal component of the catalyst. Catalysts that do not contain noble metals do not produce H2S.

Although it has been known for some time that automobile exhaust catalysts can produce H2S in fuel-rich exhaust streams, there has been an increase in the levels emitted in recent years. This is believed to be a consequence of the considerable improvements made in catalyst activity over the years. However, during one study into this effect, it was noted that H2S emissions from ‘engine-aged’ catalysts, that is, those that had been ‘on the road’ for 50,000 miles, were much lower than those from the fresh catalyst. In addition, the H2S ‘spikes’ when going from lean to rich conditions were found to be much smaller on the engine-aged catalyst.

Examination of an aged catalyst revealed traces of the usual poisons, including S and Pb from the fuel, and phosphorus (and zinc) from the engine oil. This suggested that at least one of these components can reduce the storage of sulfur by the Pt–Rh/CeO2–Al2O3 system. Phosphorus appears to be a likely candidate, because it has been found that phosphorus-doped Pt–Rh/CeO2–Al2O3 catalysts exhibit a lower capacity for adsorption of SO2 than an undoped reference catalyst. This suggests that the phosphorus is somehow ‘interfering’ with the component of the catalyst that adsorbs the SO2 – the ceria. Indeed, it has been proposed that a Ce–P–O species (possibly CePO4) is formed, which is more stable than, and hence inhibits the formation of, a Ce–S–O species (for example, Ce2(SO4)3). This would have the effect of reducing the storage capabilities of the ceria, and hence of reducing the size of the H2S spike on going from lean to rich conditions.

To summarise In view of the discussion above, it would seem that several different strategies could be used to reduce H2S emissions: (i) decreasing the Ce surface area (however, this will have detrimental effects on the catalyst activity); (ii) improving the A/F control; (iii) reducing the sulfur content of the fuel; (iv) including an H2S scavenger in the catalyst (in the USA nickel is used, but in Europe this is prohibited because of concern that it could lead to the formation and release of carcinogenic nickel carbonyl).

4.5.6 Catalyst deterioration: summary

  1. An ability to withstand mild deactivation is built into the design of the catalytic converter. However, severe deactivation could prevent the system from meeting emissions legislation.

  2. The major causes of deactivation are thermal damage and poisoning.

  3. High temperatures may cause sintering of the metals and/or the support; this can be prevented to some extent by the addition of ceria as a structural promoter. Damaging interactions between the noble metals, or with the support, can also occur at high temperatures. The interaction between Rh2O3 and γ-Al2O3 can be slowed down by first supporting the Rh2O3 on ZrO2.

  4. Lead is a severe poison, particularly for Pd, and is believed to associate with the noble metal.

  5. Phosphorus from engine oil can contaminate the catalyst and cause deactivation.

  6. Sulfur present in fuel has two major undesirable effects. It can cause deactivation of the catalyst, and it also leads to generation of H2S.

  7. Sulfur is oxidised to SO2, which is believed to block sites on the metal surface, forming a weakly adsorbed species that is desorbed when the sulfur is removed from the gas stream.

  8. Sulfur can also be stored under fuel-lean conditions by Al2O3, and especially CeO2, in the support, and released as H2S under fuel-rich conditions. The sulfur is believed to be stored as Ce2(SO4)3. On going to fuel-rich conditions, this species decomposes, releasing SO2, which is converted into H2S over the noble metal. It has been found that adsorption of SO2, and hence the storage capabilities of a Pt–Rh/CeO2–Al2O3 catalyst, are reduced in the presence of phosphorus. The preferential formation of a Ce–P–O species (possibly CePO4) inhibiting formation of the Ce–S–O species (Ce2(SO4)3) has been proposed.


Refer back to the CO/M values listed in the final column of Table 2. How would you explain the variation in these CO/M values with increasing ageing temperature?


The decrease in the value of CO/M observed as the ageing temperature increases indicates that the free metal surface area of the catalyst is decreasing with increasing temperature. This is likely to be due to sintering of the noble-metal particles, and/or the support.


This free course provided an introduction to studying Science. It took you through a series of exercises designed to develop your approach to study and learning at a distance and helped to improve your confidence as an independent learner.


The content acknowledged below is Proprietary (see terms and conditions) and is used under licence.

Grateful acknowledgement is made to the following sources for permission to reproduce material in this course:

Course image: Ray T in Flickr made available under Creative Commons Attribution 2.0 Licence.

Figure 2 Nortier, P. and Soustelle, M. (1987) 'Alumina carriers for automotive pollution control', in Cruecq, A. and Frennet, A. (eds) Catalysis and Automotive Pollution Control, Elsevier Science Publishers;

Figure 3 Eggleston, H. S. et al. (1991) Corinair Working Group on Emmision Factors for Calculating 1990 Emissions from Road Traffic, Volume 1: Methodology and Emission Factors, Commission of the European Communities;

Figure 4 Courtesy of Dr S. Golunski, Johnson Matthey;

Figure 5 Acres, C. J. K., Thomas, J. M. and Zamaraev, K. I. (1991) Perspectives in Catalysis, Blackwell Science Ltd;

Figure 6 Reprinted from Catalysis and Automotive Pollution Control, Gandhi, H. S. and Shelef, M. (1987) 'The role of research in the development of new generation automotive catalysts', p. 200, with kind permission of Elsevier Science - NL, Sara Burgerhartstraat 25, 1055 KV, Amsterdam, The Netherlands;

Figure 7 and 9 Reprinted from Surface Science, 76, (2), 1978, Conrad, H. et al. 'Interactions between oxygen and carbon monoxide on a palladium(III) surface';

Figure 8, 11 and 8A Reprinted from The Chemical Physics of Solid Surfaces and Heterogeneous Catalysis, Volume 4, Engel,T. and Ertl, G. 'The role of research in the development of new generation automotive catalysts', 1982, with kind permission of Elsevier Science - NL, Sara Burgerhartstraat 25, 1055 KV, Amsterdam, The Netherlands;

Figure 10 Engel, T. and Ertl, G. 'A molecular beam investigation of the catalytic oxidation of CO on Pd(111)', Journal of Chemical Physics, 69, (3), August 1978, The American Institute of Physics;

Figure 12 and 16 Reprinted from Journal of Catalysis, 1986, 100, Oh, S. H. et al. 'Comparative kinetic studies of CO-O2 and CO-NO reactions over single crystal and supported rhodium catalysts', p. 360, with kind permission of Elsevier Science - NL, Sara Burgerhartstraat 25, 1055 KV, Amsterdam, The Netherlands;

Figure 13, 14, 15, 20 and 22 Reprinted from Catalysis Today, 1991, 10, Funabiki, M. et al. 'Auto exhaust catalysts', pp. 34 and 35, with kind permission of Elsevier Science - NL, Sara Burgerhartstraat 25, 1055 KV, Amsterdam, The Netherlands;

Figure 17 Liang, J. et al. 'FT-IR study of nitric oxide chemisorbed on Rh/Al2O3', Journal of Physical Chemistry, 89, 1985, American Chemical Society;

Figure 19 Reprinted from Journal of Catalysis, 1986, 101, Oh, S. H. and Carpenter, J. E. 'Role of nitric oxide in inhibiting carbon dioxide over alumina-supported rhodium', p. 114, with kind permission of Elsevier Science - NL, Sara Burgerhartstraat 25, 1055 KV, Amsterdam, The Netherlands;

Figure 21 Reprinted from Catalysis and Automotive Pollution Control II, 1991, Diwell, A. F. et al. 'The role of ceria in three-way catalysts', p. 142, with kind permission of Elsevier Science - NL, Sara Burgerhartstraat 25, 1055 KV, Amsterdam, The Netherlands;

Figure 23 Courtesy of Johnson Matthey;

Figure 24 Society of Automotive Engineers (1980) SAE Paper 800843, Copyright © 1980 Society of Automotive Engineers, Inc.;

Figure 25 McIntyre, B. R. and Faix, L. J. (1986) 'Lead detection in Catalytic emission systems and effects on emissions', SAE Paper 860488, Copyright © 1986 Society of Automotive Engineers, Inc.;

Figure 26 Shelef, M. (1987) 'The role of research in the development of new generation automotive catalysts', in Cruecq, A. and Frennet, A. (eds) Catalysis and Automotive Pollution Control, Elsevier Science Publishers;

Figure 27 Reprinted from Catalysis and Automotive Pollution Control II, 1991, Monroe, D. R. et al. 'The effect of sulfur on three-way catalysts', p. 612, with kind permission of Elsevier Science - NL, Sara Burgerhartstraat 25, 1055 KV, Amsterdam, The Netherlands;

Table 1 Reprinted from Catalysis and Automotive Pollution Control II, 1991, Diwell, A. F. et al. 'The role of ceria in the three-way catalysts', p. 145, with kind permission of Elsevier Science - NL, Sara Burgerhartstraat 25, 1055 KV, Amsterdam, The Netherlands;

Table 2 Ansell, G. P. et al. (1991) 'Effects of SO2 on the alkane activity of three-way catalysts', Catalysis Letters, 11, p. 187, © J. C. Baltzer A.G. Scientific Publishing Company;

Table 3 Diwell, F. et al. (1987) 'The impact of sulphar storage on emissions from three-way catalysts', SAE Paper 872163, Copyright © 1987 Society of Automotive Engineers, Inc.

We gratefully acknowledge the help of Dr Stan Golunski and Dr Andrew Walker in the preparation of the original material.

Don't miss out:

If reading this text has inspired you to learn more, you may be interested in joining the millions of people who discover our free learning resources and qualifications by visiting The Open University - openlearn/ free-courses