Two engine oil additives with graphene were tested in diesel engines. The first was a graphene oxide (GO)-based, commercially available additive that the supplier recommends using at a 3% v/v concentration in engine oil. The second was a graphene nanoplatelet (GNP)-based additive that is under development, which is more concentrated and allows for the addition of a much smaller amount of additive. Using the GO additive results in a reduction of brake-specific fuel consumption from 0.2% to 0.7%, depending on the engine load, and a 2% reduction in fuel consumption when the engine is run without load. The use of 0.1% wt of GNPs led to 0.4% of fuel savings on an ESC emission cycle. Increasing the GNP concentration to 0.2% did not further reduce fuel consumption.

The effect of two graphene additives to engine oil on diesel engine efficiency was studied. The first additive was a commercially available additive based on graphene oxide (GO). The additive was tested on a small automotive diesel engine. The use of the additive concentration recommended by the manufacturer at 3% in the engine oil resulted in a reduction of the specific fuel consumption from 0.2% to 0.7%, depending on the engine operating conditions. The second additive, currently under development, was based on graphene nanoplatelets (GNP). The additive was tested on a medium-sized diesel engine in a truck. The use of the equivalent GNP concentration of 0.1% resulted in a reduction of fuel consumption in the ESC test by 0.4%. Increasing the concentration of this additive to 0.2% GNP did not result in a further reduction in fuel consumption. Because the engine efficiency benefits resulting from the use of improved oils were close to the measurement uncertainties, the applicability of machine learning using engine on-board diagnostics (OBD) readings to analyze the impact of lubricant additives was investigated. The use of Random Forest, machine learning digital twins, was able to reproduce the OBD instantaneous fuel consumption with excellent accuracy. Further analysis with SHAPLEY values helped to identify the more important engine parameters that affected instantaneous fuel consumption.