Letters from Readers

Jun 1st, 2002

Diamond deposition article sparks heated debate
With regard to the cover story, "Depositing Diamond in Open Air" (March, p. 15), there's only one problem. All of this work, and much more, was accomplished between 1986 and 1991 at McDonnell Douglas (now Boeing) in St. Louis, MO, and the methodology was coined "Laser Absorption Wave Deposition" (LAWD). Ironclad legal and technical substantiation of this claim can be found in the references and patents that carry my name as author/inventor.1-10

These references can also be seen at www.imagination-engines.com/diamond/diamond.htm.

Hey guys, maybe next time you should check the novelty of the work you report before enshrining it.

Stephen L. Thaler
President and CEO,
Imagination Engines Inc.

1. U.S. Patent 4,981,717, Diamond-Like Coating and Method of Forming, issued Jan. 1, 1991.

2. "Fastest Diamond Deposition Rate Claimed," Diamond Depositions Science & Technology, (October 1992).

3. "Diamond Deposition at 1 Micron per Second," High-Tech Materials Alert, (December 1992).

4. "Laser Absorption Waves as a CVD Plasma Source for Diamond," Fourth Annual Diamond Technology Workshop, Madison, Wisconsin (March 1993).

5. "Laser Absorption Wave Creation of Steel-Diamond Composites," Laserion '93, Munich, Germany (June, 1993).

6. "Neural Net Predicted Raman Spectra of the Graphite to Diamond Transition," Proceedings of the Third International Symposium on Diamond Materials, Honolulu (May 16-21, 1993).

7. "Flash Laser Hardening of Drill Bits," ICMCTF'94, San Diego, California (April 1994).

8. U.S. Patent 5,547,716, Laser Absorption Wave Deposition Process and Apparatus, issued Aug. 20, 1996.

9. U.S. Patent 5,612,099, Method and Apparatus for Coating a Substrate, issued March 18, 1997.

10. U.S. Patent 5,814,152, Apparatus for Coating a Substrate, issued Sept. 28, 1998.

The researcher responds. . .
The only feature common to both methods is the use of lasers to optically induce plasma as an energy source for the deposition process. But the plasma parameters and the deposition mechanisms in the two cases are completely different. In the photon plasmatron we use a stationary optical gas discharge in continuous-wave mode, whereas in Dr. Thaler's method the so-called "pulsed optical gas breakdown" (also known as "laser absorption wave"), a detonation mode is used.1 The latter process was extensively investigated in the 1970s by some of the developers of the plasmatron method.2, 3 It has also been applied among other things for thin-film deposition.4, 5, 6

Simeon Metev
Bremen Institute of Applied Beam Technology

1. Yu. P. Raizer, "Laser spark and spreading of discharges," Nauka, Moskow, 1974.

2. V. Konov, S. Metev et al., "Pulsed CO2 laser-induced gas breakdown near solid targets," Izwestiya VUZ, Ser. Fizika, 11, 34, (1977).

3. V. Konov, S. Metev: "Low-threshold gas breakdown initiated by pulsed-periodical CO2 laser radiation," Sov. Tech. Phys. Lett. 3, 1291, (1977.

4. A. Prokhorov, V. Konov et al., "Laser radiation interaction with metals," Editura Akademiei, Bucharest, 283 (1988).

5. I. N. Mihailescu, Konov et al, "Pulsed laser plasmotrons," Proc. Int.Conf. on Laser Advanced Material Processing (LAMP), 1, 465, Osaka, Japan, (1987).

6. V. Konov, P. Nikitin et al., "Thin film deposition by gas phase pirolisis of Fe(CO)5 and SiH4," Izv. Akad. Nauk, SSSR, Ser. Fiz. 55, 1448 (1990).

and Thaler replies . . .
The family of patents I developed at McDonnell Douglas not only teach the Laser Absorption Wave Deposition process (LAWD), but a whole gamut of laser-driven deposition techniques that include a process hauntingly similar to that described in the March 2002 issue of Laser Focus World.

My gripe is primarily with LFW for the very misleading theme of its March 2002 cover story, tempting its readership to conclude that some revolutionary breakthrough had occurred—allowing diamond phase to be grown in open air via laser. The fact of the matter is that a laser plasma technique, in open air, produced a continuous, "lily-white" diamond coating, without the slightest trace of graphitisation in 1991, at McDonnell Douglas.

Allow me also to note that LAWD diamond growth at the phenomenal deposition rate of in excess of 1 μm per second could not have been achieved without the use of highly advanced artificial intelligence techniques. This AI fine tuning both predicted and confirmed that generating a laser plasma, having the characteristics of those used in the photon plasmatron, would produce only a mixture of diamond and graphitic phases. This result was clearly reconfirmed in the research reported upon by the LFW article in which it is noted, "diamond amounts to only a fraction of the carbon phases produced. . . ." But, alas, mixtures of graphitic phases with diamond are commonplace among a wide range of PVD, CVD, and laser-ablation techniques.

The recurrent pattern among researchers, particularly in the early 1990s, was to scan the Raman exciter over their coating until the tell-tale signature of diamond, the zone-center optic phonon at 1332 cm-1, was observed. The resulting anecdotal Raman spectra provided the "Eureka slides" for conferences, but in reality represented only tiny, isolated islands of diamond phase within an inhomogeneous coating having dubious commercial value.

Stephen L. Thaler

Research did not demonstrate attosecond duration
The Krausz work discussed in your article, "Optical pulses reach attosecond length" (February, p. 17) was presented very deceptively. The title of the paper in Nature is "Attosecond metrology," implying that the group had actually measured an attosecond duration pulse. However, this is not the case. In the case of optical lasers, a pulse-duration measurement is made using an "autocorrelation" that is a more-or-less direct measurement of the physical length of a light pulse. In Krausz's work, there is no such measurement, and instead he relies mostly on comparison with theoretical models. The strength of this comparison is questionable in several very significant ways; thus, an honest summary of this work would have stated that "the data are indicative of generation of light confined to the most part in a single subfemtosecond pulse." This is far different from a conclusive, or even a persuasive, demonstration, and in fact some of the statements in the paper are unjustifiable.

However, this group does first-rate experiments, and much of the work presented is a significant addition to the literature. What is not in dispute is the fact that high-harmonic generation generates light as a series of attosecond-duration bursts. This has been well-known from the theory for several years.1 The recent experiments of P. M. Paul et al. made measurements that in fact verified, in a reasonably straightforward way, the attosecond structure of the emission.2 The Krausz work is another confirmation of this attosecond time structure, done in a different and complementary way. The data of Fig. 3 fits electron energy distributions to varying estimated x-ray burst durations, concluding that a best fit is to a ~0.5-fs duration.

Where the paper falters is in making the argument that the emission is in a single such burst. The measurement geometry used in this work cannot distinguish between a single pulse and a train of a small number of similar pulses. I have verified this fact with Paul Corkum, who originally proposed this technique. The paper states that the fact that the lack of structure in the spectrum of the x-ray emission, coupled with their models, demonstrates that they have generated a single attosecond pulse. This is unjustified. Many measurements during the past several years have demonstrated the generation of high-harmonic x-ray emission that lacks spectral structure. We have all speculated that this probably means we have been generating attosecond pulses for several years now. However, in some cases, this seems unlikely; more likely, effects such as ionization and extended propagation can smear out spectral structure even in a longer-duration pulse. In fact, in our research using temporally shaped visible-wavelength light pulses for "coherent control," we routinely make use of shaping techniques that give a train of pulses that still possesses a smooth spectral structure. Although one can learn a lot from comparing spectra between theory and experiment, determining anything definitive from a lack of spectral structure is a highly questionable proposition.

Another set of measurements in the paper is also cited as "verification" for the single-pulse structure of their emission. However, this is very much an approximate agreement between theory and experiment without any rigorous or conclusive analysis. Thus, although it very well may be possible that they (and others) have generated a single attosecond pulse, this has not been convincingly demonstrated. A conclusive demonstration would be an actual autocorrelation measurement demonstrating a width corresponding to a subfemtosecond duration—as is routine for measuring visible wavelength light pulses.

Nonetheless, it is equally important to point out that the measurement of an "isolated" attosecond pulse is secondary to the question of whether we can study the dynamics of matter on the fastest, attosecond, time scales. In fact, a number of recent papers, especially during 2000-2001 do show that "attosecond science" is now a reality. I believe that our work represents the first direct, time-domain measurements of the attosecond time-scale physics relating to the high-harmonic generation process.3, 4 In this work, we are using temporally shaped light pulses to probe the attosecond time-scale response of an atom to a very strong laser pulse.

Henry Kapteyn
University of Colorado, Boulder

1. I. P. Christov et al., Phys. Rev. A 57, R2285 (1998); Phys. Rev. Lett. 78, 1251 (1997).

2. P. M. Paul et al., Science 292, 1689 (2001).

3. R. Bartels et al., Nature 406, 164 (2000).

4. I. P. Christov et al., Phys. Rev. Lett. 86, 5458 (2001).

More in Research