Edge-emitting semiconductor laser arrays are also known as laser bars. These devices are most certainly the most widely known and widespread architecture in high-power diode lasers (HPDLs).
When employed under an electron-pumping scheme, these structures are currently able to generate up to 500 W of CW optical power while maintaining an overall active material volume of less than 0.01 mm3.
While the electro-optical efficiency of these devices can comfortably surpass 50% - particularly with GaAlAs- and InGaAs-based diode lasers - the portion of energy that is not converted into light is almost as high.
This translates into a significant amount of energy that has to be dissipated from the laser device as heat; otherwise, the active medium would likely melt in a matter of microseconds.
Heat dissipation is the most pressing concern in HPDL mounting technology. Any heat generated is initially transmitted to the surrounding substrate volume via conduction. This substrate’s volume is normally between 1 and 4 mm3.
The heat rate is too high for such a limited volume, however, meaning that further dissipating steps are required in order to dissipate the heat via a larger volume of material before this is ultimately removed by the environment – generally via forced convection towards water or air.
This process must be rapid enough to avoid extreme temperature rise in the active medium. Copper is a popular choice for this type of application due to its excellent electron conductivity.
Copper also shows the second largest thermal conductivity among all metals (k~385 W·m-1·K-1), only slightly surpassed by silver (k~405 W·m-1·K-1), which is generally not a viable option due to cost implications.
Copper has been used as the preferred sub-mount material since semiconductor laser technology was first developed. A new technical challenge has since appeared, however: establishing a proper interface between the laser bar and the copper heat sink.
This was initially achieved via soldering and this is still the most commonly adopted technology. Soldering poses several inherent problems, however.
The interface material selected for soldering purposes requires a melting temperature that is comfortably below the melting points of both the laser bar and the heat sink.
It should also be high enough to ensure thermal and mechanical stability over the laser diode’s operating temperature - generally between 15ºC and 80ºC. The first material employed for this purpose was Indium - Tmelt ~157ºC.
The goal of the soldering process is to create a solid joint between the laser bar and the copper heat sink, achieving this around the indium melting temperature.
Because each of the three materials present exhibit different coefficients of thermal expansion (CTE), the device will always experience a degree of residual stress when it cools to room temperature.1
A key implication of this is the appearance of what is known as the smile phenomena, whereby the bar suffers a curvature that results in emitters within the array not lasing completely parallel to the horizontal axis.
Rather, a separation in height ranging from 2 μm to 5 μm will often appear from the bottom to the top emitter (Figure 1).
This distance is usually the best measure of smile magnitude. Smile in laser bars is an essential consideration where external resonator configurations are necessary or where if maximum brightness is required of the fast axis.
Figure 1. Theoretical emission intensity pattern after fast axis collimation and slow axis imaging of a 19-emitter laser bar. Top image corresponds to a smile-free laser bar. The two bottom images correspond to two different kind of smile effect. Image Credit: Monocrom
External Resonator Configurations
Constructing an external resonator with a laser bar is a common approach to increasing the spectral brightness or power brightness of the laser bar itself. High spatial brightness (W·cm-2·sr) is an important factor in high-power diode lasers used in industrial applications like metal cutting, soldering or drilling.
High spectral brightness (W·cm-2·sr·nm-1) and low wavelength thermal shift (nm·K-1) are also commonly related to solid-state laser pumping applications.
Power brightness’s external feedback is generated using a reflective diffraction Bragg grating (Figure 2). This approach results in effective spatial superimposition of all the laser beams from the laser bar, in the same way as if intensity was generated from a single emitter within the laser bar.
Figure 2. Basic scheme illustrating the spectral beam combining principle. Spatially separated emitters emitting at slightly different wavelengths impinge the diffraction grating at different incident angles. However, the diffracted angle is common to all of them (different colours are used here just to illustrate the difference in wavelength but they are not representative of the wavelength itself). Image Credit: Monocrom
Consequently, spatial brightness is increased by an order of magnitude.
This can also be achieved by enlarging the emission bandwidth (resulting in lower spectral brightness) whilst accepting a certain percentage of power and optical losses; for example, an overall 20% to 40% reduction in power w.r.t. free beam operation.
These unavoidable, inherent optical losses are linked to the diffraction grating’s efficiency and the lenses’ transmission.
Figure 3. The emitters of a laser bar with no smile are contained within the plane defined by the fast axis and the optical axis of the system. As a consequence, the emitted laser beams and its partially-reflected counterpart (feedback) are spatially coincident in an external resonator configuration. Image Credit: Monocrom
Figure 4. Most of the emitters in a laser bar with smile are partially or totally out of the plane defined by the fast axis and the optical axis of the system. This results in a partial lack of optical feedback in an external resonator configuration (most of the emitted and feedback beams are partly or totally non-coincident). Image Credit: Monocrom
Crucially, optical losses are also impacted by the way the laser beam returns to the emitter after part of its intensity is bounced back at the outcoupling mirror. This is essentially feedback emission from the external resonator, with larger smile effects resulting in higher losses (Figures 3 and 4).
Volume Bragg gratings (VBGs) are typically placed in front of a fast axis collimated laser bar to improve spectral brightness. The Bragg grating requires external feedback, in this instance serving to narrow and ‘lock’ the emission wavelength of the whole laser bar.
The results in a substantial reduction in the wavelength shift w.r.t. temperature, from 0.3 nm·K-1 to less than 0.08 nm·K-1, similar to that found in distributed feedback lasers that are either applied on the diode structure itself or connected using an optical fiber.
When working with a laser diode array, avoiding the presence of smile is an essential factor in ensuring uniform feedback on each emitter, particularly when considering that the optical elements are common to every emitter (Figures 3 and 4).
Fast Axis Brightness
Laser emission in the fast axis is almost always diffraction limited (M2~1). This offers a notable advantage in applications where maximum brightness is required along a line-shaped laser spot, for example, offset laser printing (computer-to-print machines).
The presence of the smile effect in a laser bar has the potential to reduce overall fast axis brightness between 50% and 80%. This is because the apparent height of a laser source in the fast axis is increased proportionally to the smile.
Placing a fast axis collimator (FAC) lens in front of the laser bar will either result in higher residual divergence or a larger focused spot (Figure 6).
Figure 5. Near-field representation of a 10-emitter laser bar with no smile (top) and with 3 μm smile effect (bottom). The apparent size (represented by the dashed frames below) is enlarged in the fast axis due to the smile. Image Credit: Monocrom
Figure 6. Different smile patterns (left) yield different fast axis intensity profiles under fast axis collimation (middle). The far field intensity profile along the fast axis is the result of superimposing as many line-shaped spots as emitters within the laser bar (right). Image Credit: Monocrom
Alternatives to Smile Suppression
Despite the challenges outlined above, the HPDL industry has discovered means of overcoming the smile effect. One prominent solution is known as ‘hard-soldering’, which involves using an AuSn alloy as the interface material and CuW as the heat sink (Figure 7, left).2
Hard soldering sees the CTE of the semiconductor, heat sink metal and soldering interface becoming much closer, resulting in an even more reliable joint than can be achieved using indium (Figure 7, center). This approach leads to the smile effect being minimized, but at a cost.
CuW displays a significant reduction in thermal conductivity when compared to copper (approximately 50% lower), as well as a much lower degree of machinability and far higher cost.
CuW is therefore only employed as an intermediate volume between the laser bar and the heat sink, which is generally manufactured from copper. This method adds additional thermal resistance jumps to the laser diode package, but the benefits of this approach are limited to a small number of applications.
One method stands out from approaches involving indium and hard soldering. The benefits of Clamping™ have been demonstrated by Monocrom for over two decades, and this method relies on the extremely simple principle of mechanical pressure (Figure 7, right).
Conceptual simplicity does not always imply straightforward engineering, and Monocrom remains the only company capable of bringing this kind of revolutionary approach to laser diode packaging.
Clamping™ technology primarily relies on a superior surface finish being present on the copper heat sink and the establishment of direct thermal and electrical contact with the laser bar. These factors are then enhanced by the application of mechanical force.
Soldered bars are contacted to the p-side by the heat sink (anode) and wire bonding is used to connect to the n-side (cathode). Clamped bars are ‘sandwiched’ from both sides using bulky heat sinks that function as the anode and cathode.
Figure 7. Comparison between the most common soldering approaches for laser bar packaging and ClampingTM technology. Image Credit: Monocrom
The advantages of Clamping™ are numerous:
- Cold process: No residual stress is caused by dissimilar CTE, meaning that smile is limited to the surface flatness achieved in the heat sink itself – generally μm over 1 cm2 (Figure 8).
- Minimum thermal resistance jumps: Heat is directly evacuated to the heat sink from the laser bar, so it is only necessary to account for the contact resistance between copper and the semiconductor material.
- Heat will be dissipated from both the p- and n-sides, providing an additional path for heat dissipation. In the case of CW and high duty cycle QCW operation, this allows the use of water/glycol cooling channels machined on both electrodes. This also reduces the need for micro-channel cooling - millimeter-sized channels can be utilized, resulting in reduced corrosion sensitivity and ensuring minimal maintenance requirements.
- Simplicity and cost reduction: No interface material is involved, and no soldering equipment is required. Mechanical force is applied by means of a stainless-steel screw, with the perfect surface finish of the copper heat sink being a key factor.
- Superior performance in pulsed mode: A lack of horizontal residual stress eliminates fatigue suffered by the joint and laser bar during successive on/off cycles, ultimately enhancing the device’s service life.
Figure 8. Typical intensity profile of the individual emitters (fast axis collimation plus slow axis imaging) within a soldered laser bar (left) and a clamped laser bar (right). Image Credit: Monocrom
Clamping™ can be understood as a smart solution to a challenging packaging issue, one which Monocrom has effectively implemented as part of the standard manufacturing process.
- CTEIn = 33 μm·m-1·K-1; CTECu = 17 μm·m-1·K-1; CTEGaAs = 5 μm·m-1·K-1
- CTEAu(80)Sn(20) = 16 μm·m-1·K-1; CTECuW = 5-9 μm·m-1·K-1
This information has been sourced, reviewed and adapted from materials provided by Monocrom.
For more information on this source, please visit Monocrom.