Optical: systems and elements – Optical frequency converter
Reexamination Certificate
2000-01-28
2003-04-01
Lee, John D. (Department: 2874)
Optical: systems and elements
Optical frequency converter
C359S332000, C385S122000
Reexamination Certificate
active
06542285
ABSTRACT:
FIELD OF THE INVENTION
This invention relates generally to light sources. More particularly, it relates to electric light sources.
BACKGROUND ART
Highly efficient and economical multi-watt red, green, and blue lasers are desirable for display applications. Unfortunately, existing semiconductor lasers are generally not available at power levels and wavelengths suitable for display applications. Displays for consumer applications require reliable multi-watt sources that can be manufactured in high volumes. Although many factors influence manufacturing costs, some of the most significant are the number of parts and alignment steps, tight tolerances, and the cost of components.
The wavelengths of currently available lasers can be converted to those required for displays using nonlinear optics. Nonlinear optics has been used to produce wavelengths throughout the visible spectrum, over a wide range of powers, with optical-to-optical efficiencies well in excess of 50%. However, the relatively low nonlinear coefficient of available materials requires resonant or mode-locked frequency conversion schemes that are incompatible with the economics of displays. Alternatively, nonlinear waveguides may be used, but these have limited power-handling capability. To meet the needs of display applications, a new class of nonlinear optical materials employing quasi-phase-matching is required.
To reduce the number of parts and alignment steps, a bulk single-pass configuration for second harmonic generation (SHG) using one infrared semiconductor laser and one nonlinear crystal, is preferred over a resonant design. Generally, the semiconductor laser and nonlinear crystal are the costliest components in this source. Semiconductor laser bar prices have declined at approximately 30%/year since 1985. Single-emitter multi-watt semiconductor lasers are expected to follow a price-volume relationship similar to that of laser bars. The cost of the nonlinear crystal is also strongly tied to volume and the stability of the fabrication technology. Consequently, it is more reasonable to modify existing nonlinear optical materials than to develop new materials. Lithium niobate (LiNbO
3
), often referred to as “the silicon of nonlinear optics,” is an excellent material for SHG for two reasons. First, LiNbO
3
is already produced at a volume of 40 tons per year for consumer applications (cellular phones and televisions) using a very stable fabrication technology. Second, LiNbO
3
is transparent from 350 nm to 5000 nm, providing low loss for both the fundamental and harmonic for visible light generation. Finally, LiNbO
3
has nonlinear coefficients for visible light generation among the highest of all inorganic materials.
While LiNbO
3
is an attractive material because of its status as a commodity material, the only component of its nonlinear tensor large enough to satisfy the requirements of display applications is d
33
, having a value of 25.2 pm/volt. While dispersion prevents direct access to the full d
33
coefficient, quasi-phase-matching (QPM) can provide up to 64% of the full nonlinearity, or 16 pm/volt, making LiNbO
3
a very strong candidate for display applications that use QPM.
Essentially QPM is a technique that compensates for the difference in phase velocity between the fundamental wave and its harmonic in a nonlinear crystal caused by natural dispersion. In QPM, two waves having different phase velocities shift &pgr; out of phase relative to one another over a distance called the coherence length. The sign of the nonlinear coefficient reverses every coherence length, causing the locally generated harmonic field to transfer power to the harmonic beam. By compensating for phase-velocity mismatch in this way, all elements of a crystal's nonlinear tensor can be accessed throughout the entire transparency range.
Two other potential materials in which QPM has been demonstrated for visible light generation are LiTaO
3
and KTiOPO
4
(KTP). LiTaO
3
has a normalized room temperature conversion efficiency of 0.83%/(watt-cm), below that required for bulk single-pass 1064 nm SHG. However, for 852 nm SHG, LiTaO
3
has a normalized conversion efficiency of 1.8%/(watt-cm) and would be suitable for that application. KTP has a normalized conversion efficiency for 1064 nm SHG of 1.7%/(watt-cm). To achieve 25% single-pass conversion efficiency of a one watt fundamental, a crystal of 3.6-cm length is required; however, the maximum crystal length in production is 3 cm. For 852 nm SHG, KTP's normalized conversion efficiency is 4.1%/(watt-cm), and would be a strong candidate for that application.
Various approaches have been studied to create QPM structures, including use of rotationally twinned crystals, stacking of alternately oriented thin plates, and growth of periodic domain structures in ferroelectrics. For waveguides where QPM is required only at the surface of the crystal, periodic annihilation of the nonlinear coefficient and periodic domain inversion by dopant indiffusion in ferroelectrics have been employed. Periodic domain structures can be formed in ferroelectrics by applying an electric field using lithographically defined periodic electrodes. Yamada, et al. and Fejer were the first to report a demonstration of this approach. This last technique is referred to as electric field periodic poling, and is now often referred to simply as periodic poling. (“Poling” refers to the process whereby the spontaneous polarization of a ferroelectric crystal can be reversed under the influence of a sufficiently large electric field. In this application, the term “electric field periodic poling” will be used to differentiate from other periodic poling techniques.)
In both electric field periodic poling and waveguides, lithographic techniques are used to assure the periodicity of the QPM structure. The fabrication of masks for lithography typically employs interferometric feedback control. This type of control can limit the positional error of any feature over the dimension of the mask to less than a quarter-wavelength of radiation, e.g. 0.16 &mgr;m for He-Ne. For a 5-cm-long
5-
&mgr;m-period grating, this amounts to a maximum period fluctuation of 6 parts in 1 million, resulting in a negligible reduction in conversion efficiency. The ability to define QPM structures with lithographic precision created an opportunity in nonlinear optics to fabricate devices with interaction lengths not possible using non-lithographic techniques.
Previous periodic poling techniques have produced
50
-mm-long, 0.5-mm-thick periodically poled LiNbO
3
(PPLN) with a 29.75-&mgr;m period. For visible light generation using PPLN, many applications would benefit from electric field periodic poling technology capable of producing domain periods below 15 &mgr;m in devices at least 1.0 cm long. Domain periods between 6 &mgr;m and 7 &mgr;m are required for green light generation, and blue light generation requires domain periods between 4 &mgr;m and 5 &mgr;m. The longest prior PPLN devices for visible light had a period of 4.6 &mgr;m, were 6 mm long, and a thickness of 200 &mgr;m. Domain pattern quality had decreased as the period was reduced.
Therefore, a need exists in the art for an electric field poling process that accounts for the dependence of domain quality on period and thickness to produce shorter domain periods.
OBJECTS AND ADVANTAGES
Accordingly, it is a primary object of the present invention to provide a model of the electric field periodic poling process in LiNbO
3
that predicts poling outcomes and is useful as a design tool. It is a further object to provide an optimized poling waveform. It is an additional object to provide a means for fabricating non-linear crystals that are quasi-phase-matched over their entire length.
SUMMARY
These objects and advantages are attained by a novel method for fabricating a periodically poled structure from a ferroelectric substrate having an electrode structure and an insulator structure. The method produces an electric field within the substrate by applying a voltage waveform to the electro
Batchko Robert G.
Byer Robert L.
Fejer Martin M.
Miller Gregory D.
Shur Vladimir
Lee John D.
Lumen Intellectual Property Services Inc.
The Board of Trustees of the Leland Stanford Junior University
LandOfFree
Backswitch poling method for domain patterning of... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Backswitch poling method for domain patterning of..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Backswitch poling method for domain patterning of... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3047768