Field evaporation phenomena, the removal of atoms from positively charged surfaces at strong field of tens of volts per nanometer, are of tremendous scientific and technological importance. The technique of field-ion microscopy (FIM) has been developed based on this phenomenon, and is historically the first technique to have achieved atomic resolution. The atom probe, a variation of FIM, came later and is generally used to identify the species of atoms that are being individually evaporated from the surface of a FIM tip. Recently, atom-probe tomography (APT) has been developed as a powerful analytical characterization technique that ties compositional information to structure. This is the only approach that is capable of determining the 3D location and elemental identity of atoms in a sample with near atomic precision. In principle, the field evaporation phenomenon can be utilized not only for materials characterization, but also for materials processing and morphology control with extremely high precision because of its unique atom-by-atom removal capability. However, FIM and atom probe investigations only provide access to structural information on the tip surface or the evaporated segments of the specimen, and it is not possible to directly observe the structural evolution of the sample. This limitation has greatly restricted the potential applications of field evaporation as a materials-processing tool. Here, we have investigated the positive field evaporation of nanomaterials by transmission electron microscopy (TEM). Direct TEM observations of the details of the structural evolution during field evaporation have been obtained for the first time. In previous work, the shortening of individual positively biased carbon nanotubes (CNTs) has been observed by scanning electron microscopy (SEM), but the observed phenomenon may perhaps arise from the electron ablation effect because of electrons emitted from the counter CNT electrodes. Detailed information about the end caps of CNTs is difficult to obtain because of the limited spatial resolution of SEM. On the other hand, structural changes in individual CNTs have also been observed in previous in situ field-emission measurements. However, it is likely that the observed structural changes do not arise from “pure” field evaporation phenomena. The reason for this is that when CNTs are negatively charged, the large field-emission current can induce very irregular tip structures and can even damage the emitters. In contrast, as will be demonstrated here, positive field evaporation is much more controllable. Indeed, we demonstrate that positive field evaporation in combination with in situ TEM may actually provide a very simple and effective means for controlling the morphology of nanomaterials with atomic precision. All the in situ experiments have been carried out using a 200 kV Tecnai G20 transmission electron microscope with a vacuum level of about 1×10 Pa. The manipulations and measurements have been made using a Nanofactory single-tilt sample holder. All CNTs used in this work are multiwalled CNTs (MWCNTs) prepared via chemical vapor deposition. In our experiment, a single CNT protruding from the edge of a Pt wire has been selected using a W tip. By running a high current, the CNT is broken into two segments with the bottom section attached to the W tip. As shown in Figure 1a, the W tip is then moved to a different location where the counter (top) Pt electrode is microscopically flat. The distance between the CNT tip and the counter electrode is usually several tens to hundreds of nanometers. We have used electronbeam-induced deposition to deposit amorphous carbon (a-carbon) on the contact area, which effectively prevents the removal of the CNT from the W tip even under extremely strong fields. A maximum voltage of ±140 V can be applied between the Pt electrode and the grounded counter W tip. The attached CNT can be positively charged for field evaporation experiments by applying a negative voltage to the Pt electrode, or negatively charged for field-emission experiments by applying a positive bias to the Pt electrode. During the field-emission or field evaporation measurements, the electron-beam (e-beam) has been blanked out. The CNT has been positively charged with a constant bias of 140 V, which may produce enough of a local field to cause field evaporation. As the CNT end is progressively shortened and becomes flat, field evaporation slows down and eventually stops. Figure 1b and c show the end cap of the CNT before and after the evaporation experiment, respectively; the corresponding field-emission current–voltage (I–V) curves are shown in Figure 1d. Here, field-emission measurements of the CNT have been conducted to provide quantitative estimates of the field conversion factor and threshold field for positive field evaporation. C O M M U N IC A TI O N
Read full abstract