The London congestion charge has done little to improve air quality in the capital, researchers said.
The toll, introduced in central London in 2003 and expanded in both price and area since then, was designed to cut traffic and congestion in the city centre.
Car numbers have fallen by a fifth, but research showed there has been little change in pollutants such as smog, diesel soot and carbon monoxide, New Scientist magazine reported.
Nitrogen oxides levels even increased slightly, according to Kings College London researcher Frank Kelly who led the research.
Prof Kelly said the rise in nitrogen oxides may be due to the filters on buses - which have increased in number by 25% since the congestion charge was introduced - which trap soot but let gases out into the atmosphere.
The researcher said the scheme did not have air quality or people's health in mind, but he conducted the experiment to see if it could have an affect on the air in London.
His team collected air quality measurements over two years before and after the charge - which was originally £5 before rising to £8 - was brought in. There was little change in pollutants after the charge was introduced, but Prof Kelly suggested expanding controls to the rest of London could cut pollution.
"If one enlarged that area, then you would be able to have a small but important impact on air quality and presumably health."
Transport for London (TfL) said the primary aim of the scheme was to reduce congestion, not improve air quality. The Low Emission Zone (LEZ) introduced in February aims to address air quality, a spokeswoman said.
It currently applies to diesel lorries over 12 tonnes, which are forced to meet European standards on particulates or pay a daily charge of £200. She said the congestion charge had cut emissions of nitrogen oxides by 8%, particulate matter by 6% and CO2 levels by 16% since 2003.