Yu. V. Larionov

Fabrication of fiber Bragg gratings with 267 nm femtosecond radiation

Strong high-quality fiber Bragg gratings with photoinduced refractive-index modulation of more than 10-(3) were written in a Corning SMF-28 fiber, a P(2)O5-doped-core fiber and a pure-silica-core fluorine-doped-cladding fiber by third-harmonic radiation (267 nm, 150 fs and 1.2-1.8x1011 W/cm(2)) of a femtosecond Ti:sapphire laser using a phase mask. We compare the 267-nm photosensitivity responses with the results of irradiation by 193-nm ArF and 157-nm F(2) excimer lasers. The dependence of the refractive-index change on the exposure dose and the annealing characteristics of the fabricated gratings are typical for Type-I UV-written fiber gratings.

Effect of sample contamination in SEMs on linear size measurements

It is shown on the basis of the published data that taking contaminations of nanoscaled length measures into account is important in the case where these length measures are repeatedly scanned in a scanning electron microscope (SEM). Examples of changes in the linear sizes of relief elements with contaminated surfaces are given. Sources of errors are discussed. Distortions of SEM images related to the contamination of relief elements can be an extra source of errors in addition to the distortion of the initial element profile caused by a contamination film. The main patterns of contamination-film growth and phenomena resulting in distortions of SEM images of elements are described to develop a better understanding of the mechanism for producing element measurement errors. Promising tools for cleaning sample surfaces are presented. The conclusion is reached that it is necessary to use a systematic approach to the registration and exclusion of measurement errors related to the formation of contaminations on nanoscaled objects.

Measurement of the SEM-beam diameter using a relief structure: Influence of contamination

Relief structures on plate surfaces are used to measure the diameter of the scanning electron microscope (SEM) beam. A contaminant film which alters the geometric parameters of the relief structure appears on its surface even in a high-vacuum SEM. It turns out that the film is deposited onto the surface nonuniformly, and there exist such regions of the film that remain almost unchanged after structure scanning for many hours. The discovered phenomenon is explained in this paper. The values of the beam diameter determined using these regions can be used to monitor the stability of a measuring SEM under conditions of relief-structure surface contamination. The beam diameter is determined using two models. In a model with an a priori unchanged beam form, the variation in the effective diameter value is ∼2 nm during long-term scanning. In a model with the recovery of the beam form, the diameter variation is larger; however, the character of the time dependence is similar to that observed for the effective value in the model with the unchanged beam form.

Estimation of the SEM beam diameter by video-signal curves from a trapezoidal structure: 1. Simulation experiment

An experiment is performed to determine the electron beam diameter using a comparison of a video signal curve from a relief structure and the one calculated based on a simplified model of curve formation. Rather than the use of an experimental curve for comparison, this model experiment suggests the use of a curve calculated from the data on electron scattering in substrate by the Monte Carlo method. The comparison results revealed the difference between the calculated diameter value and the value of diameter of a model incident beam. It appeared that the parameters of the relief structure and the electron beam affected the calculated diameter values; as a result, these values can vary by several hundred percent in some cases.

Estimation of the diameter of a SEM beam by video-signal curves from a trapezoidal Structure: 2. Experiment

The main sources of systematic error in measurements of SEM-beam diameter are revealed. The reasons for the occurrence of error in measurements using a relief structure with a trapezoidal profile are studied. One of the most important causes of the error is selection of the ratio between the height of the relief structure and the focal depth of the SEM; another cause is related to the modification of raw SEM images of the structure, measured in the form of a video-signal curve, necessary for the used method of measurement of the beam diameter. These can lead to systematic errors in the hundreds of (and even up to a thousand) percent. Most of the other causes of error occur at the stage of the formation of a beam of electrons emitted by the surface of the structure and forming the video signal. These errors arise due to imperfection of the calculation model of low-energy secondary electron emission near the edge of the structure and some significant quantities of edge variation due to poor technological reproducibility. The contribution of these sources to the relative error was approximately 100%. Ways of eliminating the systematic error are proposed.