BMW Motorrad’s electric C evolution
California awarding $46.6M for 28 new hydrogen refueling stations and 1 mobile refueler

Cambridge study of near-term alternative London bus technologies finds lean-burn CNG most costly with greater climate impact than diesel

Annual GHG emissions relative to BASE Scenario (BASE Scenario = 949 kt CO2e on a 100-year GWP basis) for differing London bus fleet technology options. Lean-burn CNG is disadvantaged by unburned HC and natural gas pipeline leaks. Credit: ACS, Chong et al. Click to enlarge.

Researchers from the University of Cambridge have conducted a comprehensive environmental cost–benefit analysis of near-term alternative bus technologies. The study considered emissions of non-methane hydrocarbons (NMHC), CO, NOx, PM2.5, sulfur dioxide (SO2), and ammonia (NH3), as well as the lifecycle climate impact of CO2 and non-CO2 greenhouse gases (GHG) on a CO2-equivalent basis.

Their findings indicated that emission control strategy retrofits are the least costly near-term intervention to reduce urban air pollution. Although hybrid buses provide net GHG reductions and air quality improvements, associated costs are higher and more uncertain than emission retrofits. Lean-burn (spark ignition) compressed natural gas (LB-CNG) delivers the lowest health impacts due to the significant reduction of PM2.5, but has relatively high associated CO2e emissions that negate the health benefits, they found. As a result, current LB-CNG vehicles are the most costly of all of the modeled technologies, they concluded. Their study appears in the ACS journal Environmental Science & Technology.

The study accounted for CO2 and non-CO2 greenhouse gas (GHG) emissions, such as methane (CH4), that impact climate change and included lifecycle emissions on a CO2-equivalent basis relative to the base diesel case. For the lean-burn CNG scenario, this meant considering exhaust and pipeline supply leaks. CH4 exhaust emissions are a result of unburned HC, which happens during lean misfire when flame initiation and propagation is slow and when erratic combustion leads to flame quenching. Pipeline supply of natural gas is also a source of methane leaks, especially at joints. The researchers took an estimated range of this to be 0.5–2% of total fuel supply based on measurements made on EU and UK pipelines.

For the study, the team, led by Dr. Adam Boies, created a bus traffic model to simulate the Greater London bus network spatially, then defined a baseline scenario along with four near-term future technology adoption scenarios.

  • The baseline scenario (BASE) represents the 2010 composition of the Greater London bus fleet (8,624 buses), which comprised EURO II to EURO V buses that have particle filters to meet EURO IV PM limits.

  • The SCRT scenario considered the adoption of a combination of SCR (selective catalytic reduction) and CRT (continuously regenerating traps).

  • The EGRD Scenario considered the adoption of EGR technology with a diesel particle filter (DPF) for EURO II and EURO III buses in order to meet EURO IV PM limits.

  • The HYBR (hybrid bus) and CNGL (lean-burn CNG) Scenarios envision replacing the entire fleet with those technologies, respectively.

In addition to the emissions analysis, they conducted a cost–benefit analysis was conducted on the basis of (i) a monetization of the health impact from air quality changes; (ii) the social damages from climate change; and (iii) the capital and operating costs for implementing each scenario.

Percent change relative to baseline scenario bus emissions using different drivetrain and emission control strategy scenarios. Lean burn CNG is disadvantaged in part due to the difficulty in controlling the air-fuel ratio. Credit: ACS, Chong et al. Click to enlarge.

Emissions.The primary emissions reductions across all scenarios occur due to modification or replacement of EURO II and EURO III buses, which are responsible for most of the fleet emissions; although EURO II and EURO III buses comprise 72% of the baseline fleet, they emit 84% of the fleet total NOx. Among the findings for the emissions analysis were:

  • The SCRT Scenario delivered significant mean reductions in NMHC (63%), CO (47%), NOx (68%), and PM (16%).

  • EGRD results in a more modest reduction in NOx (37%).

  • In the HYBR Scenario, more efficient engine operation and reduced running time decreased emissions of NMHC, CO, NOx, and PM relative to the BASE Scenario on average by 68%, 77%, 51%, and 42%, respectively. Fuel consumption changes impacted SOx emissions and resulted in a 14% reduction.

  • The CNGL Scenario resulted in no change in total NOx (<1% relative to baseline) but an overall decrease in NO2 because no after-treatment catalyst exists to oxidize NO to NO2.

  • CNGL sees a more than 10x increase in NMHC (from 43 to 462 t/year from the BASE to CNGL Scenarios), due in part to the difficulty in controlling the air-fuel ratio (AFR) in the LB-CNG engine. NMHC also increased between the BASE and CNGL Scenarios because of the higher efficacy of CRTs in oxidizing NMHC from diesel buses when compared to the CNGL Scenario.

    However, although the CNGL NMHC emissions are large relative to the baseline, the predicted NMHC emissions value for a hypothetical LB-CNG fleet were still below European emission standards.

Climate. The emission control strategy scenarios (EGRD and SCRT) resulted in a marginal change in lifecycle CO2e emissions relative to the baseline Scenario, with a mean increase in total GHGs of 2.6% and 0% for the EGRD and SCRT scenarios, respectively.

CO2e emissions for the HYBR Scenario—including the embodied energy required to produce hybrid bus batteries—decreased on average 1.20 × 108 kg/year relative to the baseline due to the fuel efficiency improvements due to hybrid operation.

However, the researchers determined that the CO2e emissions for CNGL scenario were more than 30% higher than the baseline, despite CNG by itself being some 13% less carbon intense than diesel. Contributing factors to the poor lifecycle GHG performance were:

  • LB-CNG spark ignition operation resulted in a 47% increase in energy consumption. The increase in fuel consumption was due to losses in thermal and pumping efficiency of spark ignition operation (compared to compression ignition diesel operation). There was also a loss in volumetric efficiency because the volumetric energy density of CH4 is lower than diesel.

  • On the basis of CH4 exhaust emissions factors, the CNGL Scenario results showed that fully 3% of LB-CNG bus fuel combustion (by mass) is incomplete when averaged over all London buses.

  • CH4 from pipeline supply leaks accounted for 5% of CO2e for annualized operation.

The results presented do not account for newer CNG technology being developed for stoichiometric combustion (as opposed to lean-burn). Stoichiometric operation would allow for use of a three-way catalyst (TWC) that would result in lower NMHC, unburned HC (including CH4), and NOx.(75) Data on S-CNG bus emissions were too sparse to allow for inclusion of the technology within this study, but further discussion is given in Section 4 of the Supporting Information, which indicates that future CNG technologies may result in decreased GHG emission relative to the lean-burn results.

—Chong et al.

Cost-benefit. On the health side, the model estimated that the baseline Scenario bus emissions cause 5 premature mortalities in Greater London every year—1% of the premature mortalities in Greater London due to the emissions of road transport. The SCRT and HYBR Scenarios were estimated to cause 4 premature mortalities; the CNGL scenario avoided more than 80% of the premature deaths compared to the baseline scenario; and the EGRD scenario had no impact.

They found that the mean total costs (purchase cost, fuel cost, health costs, and the social cost of carbon) of all scenarios were positive (i.e., a net reduction of societal wealth), but the uncertainty bands indicate that the emission control strategy scenarios have only nominal additional costs.

Monetized environmental and investment costs relative to the baseline gave estimated net present cost of LB-CNG or HEB conversion to be $187 million ($73 million to $301 million) or $36 million ($–25 million to $102 million), respectively, while EGR or SCRT estimated net present costs were $19 million ($7 million to $32 million) or $15 million ($8 million to $23 million), respectively.

Only the CNGL Scenario resulted in net costs that are significantly greater than zero for all scenarios. The high costs associated with LB-CNG vehicles are a result of low LB-CNG engine energy efficiency and high CH4 emissions.

The team is currently measuring emissions from dual-fuel (CNG-diesel) heavy goods vehicles.


  • Uven Chong, Steve H. L. Yim, Steven R. H. Barrett, and Adam M. Boies (2014) “Air Quality and Climate Impacts of Alternative Bus Technologies in Greater London,” Environmental Science & Technology 48 (8), 4613-4622 doi: 10.1021/es4055274



This will be a surprise for many, specially for posters promoting the use of CNG vehicles.

Steve Reynolds

Looks to be presented in a biased manner against CNG to me.
Seems they took as their CNG example an engine that did not work well.

I also wonder what time frame was used to compare the GHG impact of the CH4 emissions to CO2. It is easy to get a large CH4 impact if you use a short time frame.


Cambridge London Bus
The whole scope of the study is outdated. Euro II and III buses are not sold any more in EU; the same with Euro IV and V. Since January 1, 2014, all new vehicles must fulfil Euro VI. Retrofitting (or conversion to CNG) of relatively new vehicles (let’s say, Euro IV and V) might make some sense but not such old vehicles as Euro II. Such vehicles should not be operated in densely populated areas but preferably, they should be scrapped. If they cannot consider scrapping old vehicles, how could they even consider HEV and CNG vehicles that are generally much more expensive than new Euro VI vehicles? Studies like this one tends to be like historical research, e.g. as: “what were the reasons why the Roman Empire fell”. I have done quite some work myself in this field with many of the conditions similar to this study but this was 15 years – or more – ago, not today. If I would conduct a study today, I would certainly update my conditions.

Besides the flaws mentioned above, I could agree on the conclusion that CNG is very expensive. I came to the same conclusions myself ~15 years ago. However, this is also valid for new vehicles, so not much has changed in this respect.

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.


Post a comment

Your Information

(Name is required. Email address will not be displayed with the comment.)