Open Access

Compact solid discharge products enable energy storage devices with high gravimetric and volumetric energy densities, but solid deposits on active surfaces can disturb charge transport and induce mechanical stress. In this Letter, we develop a nanoscale continuum model for the growth of Li2O2 crystals in lithium–oxygen batteries with organic electrolytes, based on a theory of electrochemical nonequilibrium thermodynamics originally applied to Li-ion batteries. As in the case of lithium insertion in phase-separating LiFePO4 nanoparticles, the theory predicts a transition from complex to uniform morphologies of Li2O2 with increasing current. Discrete particle growth at low discharge rates becomes suppressed at high rates, resulting in a film of electronically insulating Li2O2 that limits cell performance. We predict that the transition between these surface growth modes occurs at current densities close to the exchange current density of the cathode reaction, consistent with experimental observations.

Raman spectra from three different binary gasoline-ethanol blends (with ratios 95:5, 90:10, and 85:15) have been obtained by using a low-cost, frequency precise Fourier-transform Raman spectrometer (FT-Raman) prototype. The spectral information is presented in the range of 0 to 3500 cm-1 with a resolution of 1.66 cm-1, which is greater than the required for most liquid and solid chemical samples. This set-up delivers spectral information about the sample with a reduced spectral deviation compared to theoretical values (less than 0.4 cm-1 without compensation for instrumental response). The robust and highly fexible FT-Raman prototype presented for the spectral analysis, consisting mainly of a Michelson interferometer and a self-designed photon counter, is able to deliver high resolution and frequency precise Raman spectra from the gasoline-ethanol blends comparable to the obtained by using commercial devices. This FT-Raman set-up does not need additional complex hardware or software control and relies on re-sampling and interpolation algorithms. The qualitative spectral information obtained has been used to calculate the proportion of gasoline and ethanol present in the used chemical samples without using extra calibrations methods or chemical markers.