Comment on “submicron Imaging with a Planar Silver Lens” [appl. Phys. Lett. 84, 4403 (2004)]

S Durant,N Fang,X Zhang
DOI: https://doi.org/10.1063/1.1881780
IF: 4
2005-01-01
Applied Physics Letters
Abstract:Melville, Blaikie, and Wolf reported a submicron imaging using a 120-nm-thick silver slab. Diffraction-limited features as small as 350 nm sat a 700 nm period d were imaged onto the photosensitive material with the silver slab. The image resolution is said to be improved compared to a control experiment with 120 nm of poly smethylmethacrylate d sPMMAd between the mask and the photoresist wi hout silver filmd. Despite an extensive and intended connection to the superlens theory 2 throughout the letter and the following news thereof, 3 this experiment in our opinion is irrelevant to the referred superlens theory. This comment clarifies the interpretation of the experiment reported by Melville et al. Regarding the work reported in Ref. 1, we believe that the 120 nm silver film does not help to record an image of the mask. As we will show, the transmission behavior through this thick silver film shows that incident waves are strongly attenuated for all near field features including propagating and evanescent waves, thus the system does not enhance any evanescent waves as a superlens does. The blurred image recorded by the control experiment is in our opinion likely due to an excessive exposure. The basic theoretical idea suggested by Pendry 2 is based on the potential of a silver film to transmit and strongly enhance evanescent waves which carry subwavelength informationsoriginally scattered by the mask d. This so-called superlensing effect may be used to rebuild a subwavelength image. 4 But because the minimum feature size of the mask in this experiment is about the incident wavelength, the imaging reconstruction of the mask does not require a subwavelength imaging method, in other words, is not relevant to a superlens that significantly enhances evanescent waves scattered by the object. In Fig. 1, we calculate the transmission transfer function of TM wavesslike in the experiment Fig. 1 of Ref. 1 d for a 120 nm silver at 365 nm wavelength, which is the amplitude ratio of the transmitted H field over the incident H field as a function of the transverse wave number. It can be seen that the 120 nm silver film actually acts like an attenuator. For instance, for the 700 nm period object, the corresponding incident propagating waves have only less than 1% of the transmitted intensity. More importantly, this 120 nm silver film does not enhance evanescent waves. Since a broadband wavelength of 300–450 nm is used in Ref. 1 as described in detail in Ref. 5, we further found that the transmission for 300–450 nm has very similar behavior as 365 nm. A superlens effect is expected only for wavelength longer than 400 nm for the system used in Ref. 1 for a very narrow band of transverse wave number corresponding to evanescent waves that cannot be generated by the mask in the experiment of Ref. 1. A blurred optical image shown Fig. 2 sad of Ref. 1 is obtained by a control experiment using the same thickness of PMMA s120 nmd but without silver film. In contrast, the image obtained using a silver film sandwiched between two 60 nm layers of PMMA is in agreement with the mask topography. According to Melvilleet al., the image is blurred because of diffraction effect in the PMMA. In our opinion, the blurred image is likely due to an overexposure. We note that a comparable exposure time of 2 min is used in both the experiment with silver and the control experiment. In the control experiment, the transmission will be significantly higher than the experiment with silver where less than 1% of the energy is transmitted through this silver film. We think that the appropriate exposure time used in the control experiment should be drastically reduced to obtain a good image. Therefore, the blurred image in the control experiment cannot serve as a trustworthy verification or confirmation in Ref. 1. Despite the very low transmission factor of waves through this thicks120 nmd silver, the recorded image shown in Fig. 2sbd of Ref. 1 is in agreement with the mask topog-
What problem does this paper attempt to address?