Optical waveguides – With optical coupler – Input/output coupler
Reexamination Certificate
2002-06-07
2004-08-17
Lee, John D. (Department: 2874)
Optical waveguides
With optical coupler
Input/output coupler
C385S033000, C359S642000
Reexamination Certificate
active
06778732
ABSTRACT:
FILED OF THE INVENTION
The present invention relates to diode lasers and more particularly, a configuration of and a method for optical beam-shaping of diode laser bars to produce an optical beam of high power and brightness, allowing for efficient coupling of the optical diode laser bar output into an optical fiber.
DESCRIPTION OF THE PRIOR ART
High power solid state lasers and particularly high power fiber lasers generally depend on the availability of optical pump beams with high optical power and brightness, i.e., the required pump power needs to be made available in as small a space as possible. Semiconductor diode laser arrays are firmly established as the main source of high power optical beams, i.e., 1 watt and greater including ultra-high powers of greater than 10 watts. With current semiconductor technology, a power approaching 40 kW can be produced by a stacked diode array measuring only 5×10 cm in size (R. J. Beach et al., Laser Focus World, December 2001).
The stacked diode array typically consists of individual diode bars of around 1 cm in-length, which in turn incorporate 10-50 individual 100-200 &mgr;m long emitters spaced 0.1-1.0 mm apart. Generally, the total length of the individual emitters comprises only a fraction of the total bar length. The ratio N*l
i
/L, where N is the number of individual emitters, each of length l
i
, and L is the total length of the diode bar, is referred to as fill factor. In a diode stack the diode bars are typically separated by 1-2 mm and arranged in tiles of 10-20. Each bar typically produces a cw (continuous wave) power up to 100 W. Thus for 40 kW of power, 400 bars are needed.
A laser beam originating from one individual emitter typically has a divergence of 10°×50°. The small divergence beam is in the plane of the x axis and the high divergence beam is in the plane of the y axis. These axes are sometimes referred to as slow and fast axis respectively. The fast axis beam is generally diffraction limited and can be collimated to within a fraction of the diffraction limit with a cylindrical lens aligned parallel to the slow axis of the single emitter. The slow axis beam is typically 10-20 times diffraction limited.
The brightness B of the optical beam from an individual emitter can be calculated as B=power/(emitting area×angular divergence
x
×angular divergence
y
), where angular divergence
x,y
represents the angular divergence along the x and y axes respectively. For an individual emitter of dimensions 1×100 &mgr;m operating at a power of 4 W a brightness of B=27 MW/cm
2
is obtained.
In contrast, the brightness of a diode bar with dimensions 1 &mgr;m×1 cm operating at a power level of 100 W is only about 6.6 MW/cm
2
, whereas the brightness of a diode stack as described above operating at a power of 40 kW is only of the order of 5 kW/cm
2
. High brightness diode bars are generally manufactured by maximizing the fill factor; fill factors as high as 80-90% are commonly used to reach bar power levels of 100 W or more.
For many applications, such as fiber coupling, focusing of the diode beams into as small a spot as possible is required, where ideally the dimension of the focused spot should be the same along both axes of the diode beam. The diode bar from the example above emitting a power P=100 W can for example be coupled into a fiber with a diameter of 4 mm and a numerical aperture NA=0.22. However, the fiber coupled brightness B thus reduces to 4 kW/cm
2
, where the brightness B obtained from a fiber with core area A is calculated as B≈P/(A*&pgr;NA
2
).
The brightness of diode laser beams has traditionally been increased by the implementation of beam-shaping optics, ie., by optically combining the individual emitter beams from the diode array to generate a single optical beam which can be efficiently coupled into an optical fiber (See, e.g. U.S. Pat. No. 5,168,401 of Endriz, hereinafter Endriz '401).
Optical beam-shaping is possible by a variety of means. A first class of methods uses optical beam rotation of each emitter to rotate the beam by 90°, where the direction of the emitter beam is further deflected by around 90°after reflection from at least two reflecting surfaces. The Endriz '401 patent describes such an example. Further examples of such a method are described in U.S. Pat. No. 5,418,880 of Lewis et al. and U.S. Pat. No. 6,044,096 of Wolak, et al.
A second class of methods is based on optical beam rotation using beam-rotating prisms such as the Abbe-Konig prism as disclosed in U.S. Pat. No. 5,243,619 of Albers et al. The advantage of this design is that it avoids a 90° deflection of the beam direction such that only a small displacement in the propagation direction results.
A third class of methods is based on beam deflection in a set of multi-facetted mirrors or prisms. In these methods a first multi-facetted optical structure deflects the beam to obtain some beam spacing in the y direction, while a second multi-facetted optical structure deflects the beams to overlay the beams along the x direction (see, e.g., U.S. Pat. No. 5,887,096 of Du et al., U.S. Pat. No. 6,151,168 of Goering et al. and U.S. Pat. No. 5,987,794 to Ullmann et al.). Such systems do not require beam rotation optics, but generally employ a beam deflection along the propagation direction, relying on the manufacturing of expensive high precision multi-facet bulk optics. Though the beam-shaping device described by Ulmann et al. can work for high fill factor bars and diode arrays, the optical path lengths of individual beams through the beam-shaping optic are generally different, thereby limiting the focussability of the resulting beam. Moreover, it is difficult to implement this method with bars tightly stacked in one dimension.
The function of beam deflection and overlay can also be accomplished in one single optical element as disclosed by U.S. Pat. No. 5,825,551 of Neilson, et al. A limitation of the approach taken by Neilson et al. is the variation between the optical path lengths of each individual emitter beam through the beam-shaping optic, which in-turn limits the focussability of the resulting beam. Moreover, the adaptation of the technique by Neilson et al. to fiber coupling of diode bar stacks requires two such sets of beam-shaping optics, which is expensive to implement. Note that the first three classes of beam-shaping optics use non-focussing optics.
A fourth class of beam-shaping methods is based on beam rotation in an array of imaging transmissive optical elements that comprises at least one spherical cylindrical surface or a toroidal surface with two different curvatures along two orthogonal axes. (Such an array is described by Lissotschenko et al., German Patent No. DE 19920293 and U.S. Patent Publication No. 2002/0015558) as depicted in FIG.
1
. As shown, beam-rotation assembly
100
comprises four pairs of spherical cylindrical lenses
101
-
104
, which beam-rotate four individual beamlets
105
-
108
emitted from a diode bar with a slow axis represented by line
109
.
FIG. 1
represents only the front surfaces of the beam-rotating cylindrical lens pairs; the back surfaces of the beam-rotating cylindrical lens pairs are positioned directly behind the front surfaces and are not visible. In
FIG. 1
the individual beamlets
105
-
108
are depicted as squares to represent the approximate extension of the beamlets after collimation of the fast axis. The axes of the cylindrical lenses are aligned at an angle of 45° with respect to the slow axis
109
of the diode bar. To avoid beam-clipping in the cylindrical lenses, the beamlets need to be located centrally within the cylindrical lenses. It can be seen from
FIG. 1
that the separation of the individual emitters along the slow axis must be twice their length (for square beamlets) to fit the beam through the beam-shaping optical element. Hence the theoretical maximum allowable diode fill factor is 50%, practical implementations resulting in an actual maximum fill factor of about 40% because of beam-div
Boston Laser Inc.
Fulbright & Jaworski LLP
Lee John D.
Lin Tina M
LandOfFree
Generation of high-power, high brightness optical beams by... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Generation of high-power, high brightness optical beams by..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Generation of high-power, high brightness optical beams by... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3348888