Cathodoluminescence for in situ plasmonic sensing of beam effects

Abstract

In transmission electron microscopy (TEM), various in situ measurement in different environmental conditions, such as high temperature, gas atmosphere and aqueous solution, have become more popular. However, electron beam damage complicates the measurement. The true dynamics, which is to be observed, is no longer distinguishable from the continuously increasing damage caused by electron beam irradiation. To properly extract the real phenomenon, excluding the electron beam effect, it is important to know the electron beam damage quantitatively and consider the possible influences. Quantitative evaluation of the electron beam damage is also necessary to find the best measurement condition, such as beam dose and acceleration. Although the electron beam damage has been estimated either empirically or theoretically, experimental quantitative analysis has not been much performed due to the lack of local measurement methods in such a small scale as well as due to the limited accessibility in the TEM objective lens. Here in this research, we propose to measure the electron beam damage effect using nanoplasmonic sensors. In plasmonic sensing the optical properties of metal nanostructures are utilized to sense, locally, changes occurring at the nanoscale either to the metal nanostructure itself or to its surrounding environment. We apply these nanosensors to monitor the electron beam induced environmental change and quantitatively evaluate the electron beam effect. We take advantage of cathodoluminescence technique to simultaneously measure the plasmonic response while the electron beam is irradiated on the sample. One structure we introduce to measure the environmental change is a nanosized water container that we call the nanocuvette. The structure consists of a plasmonically active nanohole gold film caped by thin carbon films, see Figure 1. This structure allows for the inclusion of the system of interest into the nanoholes which are then sealed by the carbon layers. The system can successively be studied inside the TEM and, because of the plasmonically active gold film, it is also possible to detect changes happening to the specimen contained in the hole by following the plasmonic signal. Inside the TEM this would be achieved by cathodoluminescence. Accelerated electrons excite plasmons through transition radiation and the light radiation by the plasmon resonance can be simultaneously detected. In the work presented here we show the production of the proposed structure, and verify its plasmonic properties through ex situ measurements, combined with modelling. In particular we study the possibility to use the structures to optically detect temperature changes to the sample. This is, in this first step, done ex situ by heating a sensor structure inside a vacuum cell. The plasmonic response upon this heat treatment is studied by recording the optical transmission of the sample. We find that it is indeed possible to detect temperature changes to the sensor structure by studying its plasmon resonances. Another structure studied is a plasmonic nanoparticle. Compared to nanopore/hole structures, which are based on continuous metal films with high thermal conductivity, the temperature increase by electron beam irradiation should be more localized inside the particle. The cathodoluminescence signals of some gold particles are shown in Figure 2. As is evident from the figure, the signal vary greatly from particle to particle. It is therefore necessary to tune the particle size and structure in order to achieve enough sensitivity and signal to noise.