Laser Technology for Lighting Applications: A Review and Analysis of a Promising Technology
LpR 68 Article, page 44: LEDs are currently the dominating light source: efficient and cost effective. But LEDs also have some drawbacks. Another interesting technology developed slowly in the shadow of LEDs and has become an interesting solution for some specific applications: GaN based blue solid state laser devices. Although this technology offers some very interesting advantages, it also has challenges. Nicola Trivellin, Matteo Buffolo, Carlo De Santi, Gaudenzio Meneghesso, Enrico Zanoni and Matteo Meneghini from the University of Padova and its spin-off LightCube have been working toward the development of experimental systems and demonstrators and disclose their findings of the comparison between LED and LD systems.
In the last two decades the technology at the basis of the lighting field has seen a profound renovation: Light Emitting Diodes evolved from technological exotics to wellestablished products allowing for high efficiency, reliable and digital capabilities light sources. With the advent of luminous efficiencies up to 300 lm/W, mechanical standardization and reduced costs, Gallium Nitride based LEDs are now the standard light sources for home, industrial and automotive applications. Several research groups have been working toward the identification and the improvement of some still-present limitations of the LED technology, the most famous is known as efficiency droop, which causes a gradual decrease of light emission efficiency as the operating current density of the device increases. Efficiency droop not only has an effect on the maximum achievable efficiency at higher currents, but strongly affects the maximum light density that can be emitted from an LED chip. The direct effect of this limitation is the intrinsic need of larger optics, or multiple packages to control big Light Emitting surfaces in order to achieve high lumen output solutions. Several solutions have been proposed to improve the performance of Light Emitting Diodes, in particular semi-polar and nonpolar crystal growth directions are the most promising, but faces difficulties in growth stability and yield. Another emerging technology to achieve high flux density and to solve the droop issue is based on semiconductor laser light. This technology approach shall here be reviewed.
Introduction and the Current Status of Laser Lighting Technology
The development of high power GaN based blue laser devices [1] allows the development of remote phosphor converted laser based light source, where blue radiation emitted from a laser diode (or laser diodes array) is optically collimated (or focused according to the specific application) and excites a phosphor layer deposited over a transparent or reflective substrate, these systems are also known as LARP (Laser Activated Remote Phosphors). The combination of visible blue (450 nm) light and remote phosphor is a technology well known for LEDs, but finds application also for laser diodes lighting systems.
In the latter case, the radiation is generated from the laser facet with a size of approximately a few hundredths of μm², while for an LED similar optical power is generated typically from an active region of 1 mm². Moreover, the stimulated emission, typical of a Laser Diode, makes it possible to instantly recombine all the charges injected into the quantum region, thus not suffering from droop effects.
The laser radiation then hits the phosphor with a much higher irradiance, thus allowing much higher luminance, but also locally increasing the temperature of the phosphor due to Stokes shift losses, resulting in less than unity efficiency. The structure of a laser lighting system is dependent on whether the radiation passes through the phosphor deposited on a transparent substrate (similar to LED mixing chamber solution) or is reflected from the phosphor itself deposited into a mirrored substrate.
Figure 1: Commercial binderfree Phosphor Photoluminescence as a function of irradiance and temperature[I]
Experimental systems and demonstrators were developed [2] to analyze the state of the art technology and to study the advantages and limitation of LARP systems in comparison to standard LED based solutions. In the following, the major results of this work will be summarized and the characterization will be reported.
The comparison concerns:
• Binder-free Phosphors,
• Diffusive LARP based setups,
• Transmissive narrow beam LARP setups,
• Reflective narrow beam LARP setups
Diffusive LARP Based Setup
To characterize the efficiency of laser based white light systems, the luminous and chromatic performances of two identical prototypes with different light sources were compared.
The compared prototypes are:
• A commercially available GaN royal blue (455 nm) LED with an active area
of 1 mm² and a maximum driving current of 1 A (typical emitted power of
550 mW at 350 mA, 25°C)
• A high power GaN multimode Laser Diode in TO56 package with a
maximum optical output power of 1.6 W at a maximum drive current
of 1.5 A, 25°C
The prototypes were completed by a 3D formed commercial remote phosphor candle shaped structure (nominal CCT= 3000 K, CRI = 90, diameter 16.9 mm, height 21.2 mm). The output light source is of the diffused type, since the 3-D phosphor act as a light diffuser.
Figure 2: Comparison of luminous flux and efficacy of LD vs. LED
Results from the comparison of the absolute lumen output of the LED and LARP systems are reported in figure 2. Results indicate that, once the Laser diode has overcome its threshold current, the LARP system is able to achieve a flux in excess of 360 lm at 1.5 A, as opposed to a flux of approximately 260 lm at the same current for the LED based system. This behavior is opposed to the efficiency/current characteristic of the LED based system that, although higher at lower currents, drastically decreases when the driving current is increased, due to efficiency droop. The efficiency of the laser system overcomes that of the LED system at a current of 1.4 A for the tested devices. This comparison has been specifically designed to study the droop behavior and the LED is driven above its maximum absolute current. The low efficiencies are caused by the choice of a high CRI phosphor material and a sub-optimal mixing chamber for the setup, which is identical between the LED and the laser source and therefore far from ideal.
Figure 3: Comparison of the spectra of LD vs. LED
Figure 4: Comparison of CCT and CRI of LD vs. LED
The spectrum (Figure 3) of the blue peak of the LARP prototype is much narrower and thus much more intense as opposed to the LED prototype. The blue peak for the LARP setup is approximately one order of magnitude above the blue LED emission peak. Correlated color temperature (CCT) has a value of 3025 and 2950 K for the LARP system and LED prototype respectively; Color Rendering Index has an average value of 86 and 92 respectively. CCT and CRI do not show any significant variation with the driving current as presented in figure 4, thus indicating a good stability of the light chromaticity at different driving conditions.
Transmissive Narrow Beam LARP Setup
The first iteration of the study for a focalized LARP solution is based on a transmissive structure. The laser is collimated on the phosphor template, which is a structured glass substrate with the phosphor material encapsulated onto a silicone layer. This commercial phosphor structure allows an optimal uniformity, but the thermal resistance is limited by the conductivity of the glass thus only sustaining reduced laser irradiance. The setup structure, reported in figure 5, is composed by the laser diode positioned over a heatsink, a double lens condenser, a phosphor template and a focalizing lens; all the optical elements of the system are 1” spherical lenses with different focal distances. The emitted beam from the optical transmissive structure has been projected over a white reference screen, placed at 1320 mm from the focalizing lens, where the intensity has been measured by means of a calibrated CCD camera. The total flux of the light source has been measured by enclosing the entire structure into a Labsphere LMS-650 sphere.
Figure 5: Sketch of the laser transmissive setup [I]
The system optical performance analysis indicates that upon an accurate focalization the source is able to achieve a narrow emitted beam by means of small size optics. Figure 6 demonstrate a 2° average divergence is achievable with 1 inch optics, with average color uniformity over the projected image, since some yellow ghosting is visible on the minor axis of the beam.
Figure 6: Projected beam
The clear drawbacks of the transmissive system are:
• Low efficiency due to high optical losses and bidirectional emission
of the phosphor template
• Limited maximum irradiance over the phosphor template due to
low thermal conductivity of the transparent substrate
Figure 7: Burning marks on silicone encapsulated phosphors over glass substrate when excited by a too high laser irradiance [I]
Figure 8: Sketch of the lens based reflective setup [I]
Figure 9: Sketch of the parabolic reflector based reflective setup [I]
Figure 9 shows the feed forward effects of efficiency reduction with self-heating temperature increase. The two drawbacks can be reduced by implementing a reflective phosphor structure described in the following.
Reflective Narrow Beam LARP Setups
As previously described a different approach with respect to the transmissive structure is related to the possibility of layering the phosphor over a reflective surface. As presented in figures 8 and 9, reflective phosphor surfaces have the clear advantage of (nearly) doubling the amount of light collected by the optical setup. Prototypes of reflective structures are built around a binder-free phosphor layer deposited over a glass based optical mirror (based on dielectric reflector).
Figure 8 reports on a setup based on a tilted phosphor template that is excited by a collimated laser beam. The emitted white light is then focalized by an optical structure based on two 2” spherical lenses and an engineered symmetrical glass diffuser to homogenize the emitted beam. Figure 10 reports on the shape, size and chromatic a distance of 330 mm from the last focalization lens. Results report good color uniformity over a beam of 6° divergence, although quite far from the ideal white spot.
Figures 10: Beam of the lens based reflective setup [I]
Figures 11: Beam of the parabolic reflector based reflective setup [I]
An alternative structure can be manufactured and based on a parabolic reflector presented in figure 9, where the laser beam is focalized through a hole in the reflector and thus exciting the phosphor template placed at the focal point of the parabolic reflector. The resulting beam size (Figure 11) has an average divergence of 8.5° and a good color uniformity.
Summary and Conclusions
The summarizing table 1 reports the major results of the three narrow beam laser tested solutions, where care should be taken of the fact that the phosphors are of different origins between the transmissive setup (commercial silicone encapsulated on glass) and reflective setup (custom drop casted on mirror). It is interesting to note that efficiency strongly improves on reflective setups, but also the divergence of the emitted beam. Of course, efficacies are still relatively low, but a significant improvement can be reached through laser diode and phosphor optimization.
Table 1: The major results of the three narrow beam laser tested solutions
In conclusion, the research shows that laser-based lighting, although still a growing technology, can push the limits of solid-state lighting in terms of efficiency at high currents thanks to low droop and optical management of the emitted light. The technological limits are still related to the laser diode performance and costs and the development of efficient cooling structures for the phosphor template.
Will LEDs or lasers win in the long run? It is very likely that both technologies will find wide application and create a real change in the lighting paradigm. The real winners will be the end users who will have access to two flexible and different technologies for lighting: LEDs and lasers. This will increase the degrees of freedom for designers, leading to an even bigger penetration of solid-state lighting in the application market.
Credits:
[I] Image/graph courtesy of MDPI Materials
References:
[1] H. König, A. Lell, B. Stojetz, M. Ali, C. Eichler, M. Peter, A. Löffler,
U. Strauss, “Blue 450nm high power semiconductor continuous wave
laser bars exceeding rollover output power of 80W”, SPIE Photonics
West 2018, San Francisco, Paper 10514-1 (2018)
[2] N. Trivellin, M. Yushchenko, M. Buffolo, C. De Santi, M. Meneghini,
G. Meneghesso, E. Zanoni, “Laser-Based Lighting:
Experimental Analysis and Perspectives”, Materials 11;10(10), 2017
[3] M. Dal Lago, M. Meneghini, N.Trivellin, G.Mura, M.Vanzi,
G.Meneghesso, E. Zanoni, “Phosphors for LED-based light sources:
Thermal properties and reliability issues” Microelectronics Reliability
Volume 52, Issues 9-10, Pages 2164-2167, September-October 2012
[4] A. Salimian, J. Silver,G. R Fern, H. Upadhyaya, A. Metcalfe,
T. G. Ireland, P. Harris, and R. Haghpanahan, “Investigating the
Emission Characteristics of Single Crystal YAG When Activated by
High Power Laser Beams”, ECS JOURNAL OF SOLID STATE
SCIENCE AND TECHNOLOGY, 5 (10), pp. R172 - R177 (6) (2016)