Do No Harm

How efficiency does matter

In our last newsletter we showed how the energy efficiency of vehicles is not the only consideration in decarbonisation.  We should also think about efficiency of resource allocation.  Resources are scarce, and the job of decarbonisation is big, so we need to invest our money carefully.  One way to judge this is to consider an alternative measure of optimality: Pareto efficiency.  Vilfredo Pareto was a nineteenth century Italian economist, who described a system being Pareto efficient when is not possible to change the allocation of goods without harming at least one person.  Interpreting that for transport decarbonisation, we can say that a change of powertrain can be a Pareto improvement if it is possible to improve all relevant aspects without disadvantage to someone.  For clarity, following Pareto improvements does not necessarily lead us to the globally best outcome, but offers relatively easy ‘wins-wins’, opportunities which are scarce in complex modern economies.  What can this tell us if we apply it to some real-world test data?

To study this, Emissions Analytics performed a comparison test between a Tesla Model Y and a Kia Niro full hybrid electric vehicle (FHEV), on its real-world EQUA test cycle conducted in the UK in March 2023.  The vehicles were less than one year old, equipped with the original tyres, with similar tread depths, and closely matching odometer readings.  Both are standard size sports utility vehicles, with the same number of seats, while the Tesla is physically slightly larger.  Importantly, due primarily to the 78.1 kWh battery, the empty weight of the Tesla was 489 kg greater.  From a driveability point of view, this is offset by almost treble the power and double the maximum torque.  The detailed specifications are shown in the table below.

Selecting the appropriate pair of vehicles presented various dilemmas, and the conclusion was that there is no perfect answer.  Perhaps the most obvious approach would have been to compare the hybrid to a Kia Niro EV.  From an engineering point of view this would have been a good like-for-like comparison.  However, Emissions Analytics is committed to understanding what happens in the real world and, therefore, it would be unwise to the neglect the biggest selling battery electric vehicle (BEV) by revenue in the world in 2022, the Tesla Model Y.  The pairing of this with the Niro hybrid reflects a common choice that real customers are making, even if that means a technically less perfect comparison.  As evidence of this being a good comparison, the table include effective monthly lease costs, which shows there is less than 10% difference.  These figures assume the buyer is a company car driver paying the higher rate of income tax in the UK, which reflects the most common type of buyer; the difference is caused by the low benefit-in-kind tax on the BEV. The mix of incentives varies around the world, some more generous, some less so.
 
With the background of climate change and Dieselgate, there is a strong narrative that we need to “stop burning stuff” to solve our environmental problems.  Non-combustion sources of emissions, such as from tyres, typically get dismissed as just big chunks/having no air quality effect/not very toxic/inconvenient.  At the same time, some BEV owners report that they see lower tyre wear, and other higher wear compared their previous internal combustion engine (ICE) vehicle.  However, it has been unclear whether there has been simultaneous behaviour change, in terms of driving style or routes, or the measurements are unreliable.

For this test, the vehicles travelled in convoy to eliminate any effects of driving style or climatic conditions.  The Kia was instrumented with a tailpipe Portable Emissions Measurement System (PEMS) and the Tesla with an equivalent mass.  The test cycle was made up of five repeats of the EQUA cycle, totalling 741 km.  At regular points through the test, all the wheels were dismounted, cleaned, weighed and remounted to calculate the mass loss.  The results, compared to the relevant regulatory limits, are shown in the table below.

The mass loss data points and trend over the total distance are shown in the chart below.

The first striking result is that there are almost no tailpipe pollutants measured from the Kia except carbon dioxide (CO2).  The gasoline particulate filter reduces particle mass to almost zero, and particle number to 97% below the Euro 6 regulatory limit.  Every air pollutant is comfortably more than 90% its limit.  While there is no limit value for CO2, compared to the average of all the current-generation gasoline vehicles we have tested, the Kia is 38% lower at 113.4 g/km.  The Tesla is, of course, zero on all these measures; we are ignoring upstream emissions in this test.

Turning to the non-exhaust, the second striking finding is that tyre wear mass emissions are five orders of magnitude greater than particle mass from the tailpipe.  This is two orders of magnitude greater than in previous tests by Emissions Analytics on different vehicles.  Further, tyre wear emissions were 26% greater from the Tesla, due to the extra weight and torque, despite their being specifically-designed ‘green tyres’ for electric vehicles.  The gap may have been greater had the car not been equipped with special EV tyres.  In absolute terms, the increased tyre wear was 11 mg/km, which is 2.4 times the maximum permissible tailpipe particle mass emissions.

So, as tailpipe air pollutants were tending to zero from the Kia, it is a fair summary to say that a consumer choosing between switching from a traditional gasoline internal combustion engine (ICE) vehicle to either a Kia Niro hybrid or a Tesla Model Y, is weighing up an extra 62% point reduction in CO2 against a 26% increase in particle emissions.  However, if full lifecycle CO2 emissions are taken into account, BEVs currently offer around 50% CO2 reduction on average, so in reality the decision to opt for the Tesla is between 12% points of extra CO2 reduction compared to the ICE baseline but 26% more particles.  

How robust is this conclusion?  In other words, would we get a very different answer if we had chosen a different FHEV/BEV pair of vehicles?  From Emissions Analytics’ wider EQUA programme, we have tested many different vehicles for tailpipe emissions and tyres for wear rates.  From this we conclude that the key factors in the relative emissions at the whole-vehicle level are the vehicle mass and torque, the model year and the tyres the vehicle is equipped with.  Therefore, it is of little significance that the Kia Niro is physically smaller than the Tesla Model Y, as the differences in mass and torque are representative of the typical of choices that consumers are making in today’s market.  What this shows is that the vehicle manufacturer’s choice of tyres is increasingly important in overall emissions, both for tyre wear and CO2 via rolling resistance.

What does this mean in terms of the concept of Pareto efficiency set out at the start?  Moving from a gasoline ICE to the hybrid would be a Pareto improvement: it is better on every measure.  But moving from the gasoline ICE to the BEV is not such an improvement, due to the increase in tyre particles.  So, given gasoline ICEs are predominant at the moment, the optimal and most efficient move in terms of scarce resource allocation is to move to FHEVs, not BEVs.

But it doesn’t end there.  Tyres don’t just affect the environment in terms of a mass of small particles.  Also important are the number of particles emitted and chemicals that leach out of them over time as they settle on soil or in water, as well as the vexed problem of handling end-of-life tyres.  A neglected further effect is volatile organic compounds (VOCs) that ‘off-gas’, or evaporate, from the surface of the tyre all the time.  Similar compounds are also released from the tailpipe, although they are regulated only as part of a ‘total hydrocarbons’ measure.  Which is greater – VOCs from the tyres or the tailpipe?

During the same test above, we measured tailpipe VOCs using Emissions Analytics’ proprietary sampling equipment that allows a full speciation using two-dimensional gas chromatography and time-of-flight mass spectrometry.  These VOCs matter as they are a precursor to smog formation and contribute to the secondary formation of particles in the air, as they react chemically.  These effects are in addition to the direct health effects, especially for aromatic compounds, which are often carcinogenic in certain concentrations and exposures.  Within that group, polycyclic aromatic hydrocarbons (PAHs) and nitro-aromatics are typically the worst.

Large samples from one tyre on each vehicle were also taken and placed in a ‘microchamber’ heated to 20 degrees Celsius, around the temperature of a vehicle certification test, and held at that level for the same duration of the on-road EQUA test – around three-and-a-half hours.  The off-gassed VOCs were analysed and quantified, and then scaled up by the relative surface area of the sample to that of all four tyres on the vehicle.  The results are shown in the table below.

This shows that VOCs off-gassed into the air from the tyres are about two orders of magnitude greater than those from the tailpipe of the Kia.  Adding the tailpipe and tyre sources, we see that the Kia had total emissions less than half of the Tesla’s.  This result is driven by the larger diameter and width of the Tesla tyres, despite their being lower profile.  These results will be unpacked further in a future newsletter, but for now we can see that they are consistent with the pattern of the regulated pollutants: there is very little coming from the tailpipe relative to tyres.  

In this context, the concept of Pareto improvement is reminiscent of the no-harm principle in medicine: primum non nocere. ‘Do no harm’ means taking a step back from an intervention to look at the broader context and mitigate potential negative effects on the social fabric, the economy and the environment.  By switching to FHEVs we can create a ‘no harm’ intervention, unlike BEVs.  That is not to say that FHEVs emit literally zero, but no additional harm is done and, in fact, improve all aspects.  That is also not to say that BEVs and their associated tyres might not improve – they very likely will.  At that point, it would then be right to change policy.  To the trade-off described earlier – between 26% more particles as the price of 12% points of extra CO2 reduction – no verdict is passed as to whether this is a good trade-off from a policy point of view, but a trade-off it is.

In the meantime, FHEVs are the win-win, do-no-harm option, while BEVs are the win-lose, the vexatious trade-off in a situation of significant technology uncertainty.  On this basis, and of Pareto efficiency, until BEVs reach certain performance characteristics, government and industry support should be switched immediately from BEVs to FHEVs to create maximum welfare.