Electron tomography is now able to resolve atomic scale structure in three dimensions. The novelty of yesterday’s preprint lies in absence of a requirement for a priori information about the structure. Further, while conventional electron tomography achieves ~ 1 nm resolution, this paper demonstrates 2.4 Å resolution.
| arXiv:1108.5350v1 [cond-mat.mtrl-sci] |
Ricardo Ruiz of Hitachi GST gave a short lecture on the status of patterned media development, to the local section of the IEEE Magnetics Society yesterday. I summarize by pointing to four publications listed below. R Ruiz, et al. (Hitachi GST, U Wisc, Hitachi) gives a compact, readable overview of patterned media development circa 2008. Figure 2 is particularly noteworthy; in order to become commercially interesting, one needs a bit density above 1Tb/in^2. Since the read/write head design constrains one to a 2:1 or (better) 4:1 aspect ratio, the dot pitch has to be below 20 nm. [SST 51(9)(2008)]
For reliable operation, every bit on a hard disk should switch in the same way. Happily, electron-beam directed assembly yields tighter switching field distributions than either e-beam lithography or block copolymer self-assembly alone. [Appl. Phys. Lett. 96, 052511 (2010), doi:10.1063/1.3293301]
Recently, Hitachi GST has also reported on an integrated thermally-assisted recording head and its operation on patterned media. Going under the moniker of “BP-TAR”, the authors have achieved 1Tb/in^2 on a 24 nm track pitch. This represents a solid advance over writing on continuous media, even with thermal assist. [Nature Photonics 4, 484 - 488 (2010), doi:10.1038/nphoton.2010.90]
By means of orthogonal double patterning, Hitachi is able to show 27 nm full pitch dots of arbitrary aspect ratio. This is achieved by using e-beam directed self-assembly in the radial direction of the media, and then cutting the resulting long (radial) lines into tracks with EBL. Directed self-assembly can accomodate small (~5%) variations from the block-copolymer’s preferred lattice period, thereby making equi-angular spacing feasible. [ACS Nano, 2011, 5 (1), pp 79–84, doi:10.1021/nn101561p]
In summary, patterned media development appears to be on track. Like anything complex, a lot of engineering and learning remains to be done.
Apart from the buzz generated by IBM’s paper in Science , graphene research continues apace. Here I will bring your attention to two sub-topics: graphene on boron nitride, and patterning of graphene.
Experimental research led by the LeRoy group at the University of Arizona has shown that graphene lays down nearly flat on boron nitride substrate. This is in contrast to the wrinkling that occurs when graphene is placed onto silicon dioxide. Consequently the electronic properties of graphene on BN are much closer to those properties for free-standing graphene, but the material is much more robust and therefore more accessible to experiment. [Nature Materials, also found at arXiv:1102.2642]
From the Kavli Institute of Nanoscience in Delft comes a report that multi-layer graphene can be sculpted by high-energy e-beam without defects. If one irradiates graphene at room temperature, one finds that the graphene rapidly converts to amorphous carbon. The Kavli Institute has discovered that raising the temperature to 600C allows the graphene to heal while being irradiated, so that patterning is possible. [arXiv:1102.0971]
Finally, work at the Cavendish Laboratory, Cambridge, shows that apparent scanning probe nanolithography on graphene may be deceptive. A voltage is applied between tip and substrate during contact mode scanning. One then inspects the sample in tapping mode, and finds indentations wherever the tip was dragged. However, it seems that current flow through the probe tip causes oxidation of the graphene substrate only if the current flowing through the tip drops to (near) zero. This current drop depends on setting the tip voltage high enough to drive the oxidation to completion. The authors do not report what atomic state the “pseudo-cut” (that is, indented) graphene remains in. [arXiv: 1102.2781]
In March I wrote about Point Projection Microscopy (PPM), in which electrons are emitted from a single-atom tip and accelerated to (and through) a target with very high coherence. The spatial resolution is quite high. I found that the same research group had earlier calculated that PPM should be able to distinguish atoms of differing atomic number one from another. This would imply analytical applications of the technique in addition to image formation. See arXiv:1101.5135.
Marvin Cohen and Alex Zettl have published a small review of boron nitride nanotubes [Physics Today 63 (11), 34-38 (2010)]. Two potential applications loom. The authors point out that these nanotubes (abbreviated BNNT) show field emission that is unusually stable compared to carbon nanotube field emission. Also, BNNT field emission closely follows the Fowler-Nordheim law. The implications to cold field emission device construction are obvious.
Furthermore, the BNNT bandgap can be tuned by application of an electric field transverse to the tube axis. At zero applied voltage, the gap is around 5 eV. At sufficiently large applied voltage, the band gap collapses to 0 eV. Perhaps one may use a BNNT as a transistor channel based on this giant Stark effect.
Other electronic applications may arise from the peculiar spatial localization of the lowest conduction band. It is physically situated in the tube, leading lightly doped BNNTs to act like plumbing for nearly a free electron gas.
Mutus et al. at the University of Alberta have recently examined a suspended graphene film by low energy electron point projection microscopy (LEEPPM). [arXiv:1102.1758] Even though the electron energy was only 100-200 eV, approximately 75% of the electrons get through the film. With an effective source size of < 5Å, imaging resolution is quite good. The authors claim that the low beam energy does not result in sample contamination in the way higher voltage techniques such as STEM do. (I find this claim puzzling, since it is the secondary electrons which are largely responsible for chemical reactions induced by high voltage e-beams.) They speculate that a graphene film could be used as the ultimate microscope slide for imaging thin objects by PPM.
The SPIE Advanced Lithography Conference was dominated this year by incremental progress. I paid most attention to the e-beam patterning sessions, running off for a few other interesting papers.
Possibly the most exciting development was Paper 7970-14, “Self-assembly patterning for sub-15 nm half-pitch”, reporting joint work from Applied Materials and IBM Almaden. Chris Bencher showed a directed self-assembly process which resulted in etched patterning on silicon wafers. Using polystyrene-block-polymethylmethacrylate pinned by patterned BARC, Mr Bencher showed chip-area, periodic patterns at very high resolution. Defectivity is acceptable for this point in the R&D effort, comparable to the early days of 193i R&D. Interestingly, the defects are mostly particles, not self-assembly errors (except adjoining the streets).
In e-beam, the REBL project appeared much more convincing this time around. [Paper 7970-43, "New advances with REBL for maskless, high-throughput EBDW lithography"] Paul Petric (KLA-Tencor) reported that beam down on column design #2 was due during the week of the conference. He reported some difficulties with getting a functional modulator assembly, but sounded optimistic that the difficulties would be resolved shortly. An architectural overview may be found in [J. Vac. Sci. Technol. B 28, C6C6 (2010); doi:10.1116/1.3511436].
In EUV lithography, carbon deposition on the optics seriously degrades performance. A nice experiment at the University of Hyogo has shown that bleeding oxygen or ozone into the optics while irradiating with EUV can remove the deposits. [J. Vac. Sci. Technol. B 29, 011030 (2011); doi:10.1116/1.3533945]
An old invention by Somekh comes to mind; he discovered the same trick for electron beam systems in 1999. [US 6394109]
To my friends and colleagues in Japan, I extend my deep sympathy to all those who suffer in the aftermath of this great earthquake.
- Richard L Lozes
Reading this article in the LA Times today, I could not help but think of all the hollowed out semiconductor equipment companies in Silicon Valley. The SEMI heavyweights have been outsourcing for years, shedding ever more employees and expertise. Is it any wonder that there are no leading-edge vendors of lithography equipment in the U.S. anymore?
In a related vein, Gio Wiederhold (Stanford Univ.) details the process by which U.S. companies set up foreign subsidiaries strictly for the purpose of holding the company I.P. (patents, for example). Why? Simply to fluff up the finances and avoid U.S. taxes. (The paper is also available from Prof. Wiederhold’s home page.)
We can rest assured that all the hot air about job creation will come to naught until the rules of the game change.
Serendipitous update: See the just published article by Dominic Barton in Harvard Business Review on “Capitalism for the Long Term”.
Nanolithography in the Large is proudly powered by
WordPress. Entries (RSS) and Comments (RSS).
Lozes Technology Consulting © 2008-2012 Richard L. Lozes, unless otherwise specified. Written permission is required to reproduce any material contained in this website. Valid XHTML and CSS. Website by Danielle Signor Design.