Doing the right thing: Authors retract lubricant paper whose findings they can’t reproduce

The journal Wear — an Elsevier title, not a Condé Nast fashion magazine — has retracted a paper by a pair of Chinese physicists after the researchers were unable to replicate their findings.

The 2009 article, “Microstructure and tribological characterizations of Ni based self-lubricating coating,” was written by authors from the MOE Key Laboratory for Nonequilibrium Synthesis and Modulation of Condensed Matter and the MOE Key Laboratory for Strength and Vibration at Jiaotong University, in Xi’an. It purported to find that:

In this study, HFIS304 coating comprising NiCr (80/20) matrix (60 wt%) combined with Cr2O3 (20 wt%), Ag (10 wt%) and eutectic BaF2/CaF2 (10 wt%) as solid self-lubricants additives has been successfully prepared by cold isostatic pressing and high-frequency induction sintering (HFIS) on a Ni-based superalloy rod. The microstructure of HFIS304 was fine and dense, the size of Cr2O3, BaF2/CaF2 particles was below 1 μm, and that of Ag particles about 1 μm. The tribological properties of HFIS304 composites against Inconel X-750 were examined. The wear resistance and friction coefficient of the HFIS304 were superior to that of the PS304 in the temperature range from room temperature to 600 °C. The improvement in wear resistance of the HFIS304 was discussed in the terms of its microstructural characteristics. Additionally, the model of high temperature self-lubricating particles was proposed.

But things after that evidently did not go so smoothly. According to the notice:

This article has been retracted from Wear upon the authors׳ request. It was discovered after publication that due to power supply stability and metering problem, the sintering conditions stated in the paper to produce the materials that were described in the paper could not be replicated in later trials. In view of these uncertainties in procedure, and with the authors׳ approval, this paper has been retracted and should not be cited in future publications.

The paper has been cited twice, according to Thomson Scientific’s Web of Knowledge.