Development of a high-power blue laser
(445 nm) for material processing
HONGZE WANG,1,*YOSUKE KAWAHITO,1,4 RYOHEI YOSHIDA,2YUYA NAKASHIMA,3AND KUNIO SHIOKAWA3
1Joining and Welding Research Institute (JWRI), Osaka University, 11-1 Mihogaoka, Ibaraki, Osaka 567-0047, Japan
2Graduate School of Engineering, Osaka University, Suita, Osaka 567-0871, Japan
3Japan Fuji Electric Corporation, Tokyo 191-8502, Japan
*Corresponding author: email@example.com‑u.ac.jp
Received 20 April 2017; revised 12 May 2017; accepted 12 May 2017; posted 15 May 2017 (Doc. ID 293204); published 6 June 2017
A blue diode laser has a higher absorption rate than a tradi-
tional laser, while the maximum power is limited. We re-
port the structure and laser beam profile of a 250 W
high-power blue laser (445 nm) for material processing.
The absorption rate of the blue laser system for the steel
was 2.75 times that of a single-mode fiber laser system
(1070 nm). The characteristics of the steel after laser irra-
diation were determined, validating the potential of this
high-power blue laser for material processing, such as heat
treatment and cladding. The cost of the developed laser sys-
tem was lower than that of the existing one. To the best of
our knowledge, this is the first blue laser with a power as
high as 250 W. © 2017 Optical Society of America
OCIS codes: (140.2020) Diode lasers; (350.3390) Laser materials
processing; (300.1030) Absorption.
Lasers have been widely used in material processing (e.g., weld-
ing, cutting, cladding, and heat treatment) because of their high
stability, quality, and processing speed [1–4]. Various kinds of
laser systems for processing have been developed since 1960
[5–11], including CO2lasers (wavelength = 10640 nm),
Nd: YAG rod lasers (1064 nm), Yb: fiber lasers (1070 nm),
Nd:YAG disk lasers (1030 nm), and infrared diode lasers
(808–980 nm). However, the high cost and low energy conser-
vation efficiency of these laser systems have restricted their
development scale and speed.
The laser absorption of a material determines the energy
conversion efficiency from light and has been investigated ex-
tensively [12–15]. When a material is irradiated with a laser,
part of the laser energy is absorbed by the electrons. The ab-
sorption rate of the material is determined by the interaction
between the laser and the material, where the wavelength of the
laser, the surface status and properties of the material, and the
processing parameters can affect the absorption rate. For laser
processing at a high power density (generally higher than
106W∕cm2), a keyhole is formed, the laser beam penetrates
the keyhole, the keyhole walls produce multiple reflections,
and a high absorption rate is achieved independently of the
laser wavelength . For laser processing at a low power den-
sity, the absorption rate is strongly dependent on the laser wave-
length . Thus, a laser system with a wavelength in a
particular range could improve the absorption rate considerably
for a material at a low power density.
Fresnel’s equation is used to calculate the absorption rate of
materials [17,18]. The absorption rate of the ideal flat surface is
calculated from the incidence angle and the optical parameters
related to the material and wavelength of the laser. The calcu-
lated result showed that the absorption rate of iron decreased as
the laser wavelength increased [14,19]. Though there was a
difference between the calculated absorption rate and the actual
value because the effect of surface roughness was ignored, the
obtained variation tendency would be acceptable. Thus, a laser
system with a shorter wavelength (e.g., a diode laser) would
have a higher absorption rate for iron.
Diode lasers are increasingly used because of their relatively
low cost and high energy conservation efficiency [20,21].
Various materials have been used for diode lasers, and the wave-
length of the laser depends on the material, for example,
InGaAs (940–990 nm), AlGaAs (720–880 nm), AlGaInP
(630–690 nm), and GaN (∼445 nm)[20,22,23]. In his
Nobel lecture, Nakamura  predicted that GaN-based laser
diodes may be used for the next generation of solid-state light-
ing because of their high electrical-to-optical energy conversion
efficiency and low cost. Since GaN-based laser diodes were first
reported in 1994 [22,25], they have been widely used in ap-
plications, including lighting and optical storage. GaN laser
diodes with a peak wavelength in the range of blue light
(400–480 nm) are called blue lasers . Generally, the maxi-
mum power of a single GaN laser diode is small (e.g., 4 W), and
is not suitable for applications in welding and heat treatment.
Diode arrays containing many single laser diodes are used to
achieve high output powers [26,27]. Although high-power laser
systems with wavelengths longer than 630 nm have been widely
used for production, high-power blue laser systems (∼445 nm),
which would have a higher light absorption rate, have never
We developed a high-power blue laser system with a maxi-
mum output power of 250 W. The absorption rate of the blue
Letter Vol. 42, No. 12 / June 15 2017 / Optics Letters 2251
0146-9592/17/122251-04 Journal © 2017 Optical Society of America
laser system was measured experimentally. The characteristics
of the steel after laser irradiation were determined, and we dem-
onstrated the potential of our high-power blue laser for material
Figure 1shows our blue laser system. Figure 1(a) shows a
schematic of the light path. Two laser diode clusters were
placed vertically, and the light from these two clusters was com-
bined by a mirror and focused by a convex lens. The mirror
could be divided into eight sub-areas; four areas were covered
with anti-reflection coatings, and the other four areas were
covered with high-reflection coatings. A blue laser from the
LD cluster on the upper side rayed on the areas covered with
a high-reflection coating; then it was reflected to the convex
lens. While a blue laser from the LD cluster on the left side
rayed on the areas covered with anti-reflection coatings, then
penetrated the mirror, raying to the convex lens. Finally, light
rayed to the convex lens focused at the focus point. A schematic
of the laser diode cluster is shown in Fig. 1(b). Each cluster was
made up of four groups (G1, G2, G3, and G4), and each group
was made up of eight laser diodes. The power of the single laser
diode was 4 W, and the wavelength was 445 nm. Thus, the
blue laser system had a maximum designed power of
256 W. Figure 1(c) shows a photograph of the blue laser equip-
ment. Cooling water was used to remove the large amount of
heat produced during the laser operation. Figure 1(d) shows a
photograph of the blue laser passing through water. Sixty-four
laser beams, each from a laser diode, were clearly observed in
the water, and the focusing of these laser beams could be clearly
viewed. The diameter of laser beam decreased to the minimum
value at the focus position, and then increased. A high power
density was achieved at the focus position, or at positions with a
small defocus distance.
Figure 2shows the designed power density distribution of
the blue laser system at the focus position. Figure 2(a) shows
the overall distribution of the power density. The blue laser
focused on an approximately rectangular area, with a high
power density that decreased quickly to zero outside the area.
Figures 2(b) and 2(c) show the power density distributions
along the Xand Yaxes, respectively. According to the defini-
tion of the diameter of a circular laser beam , the dimension
of the blue laser focused zone is the zone where the laser power
density is higher than 1∕e2of the peak. The length of the
focused area along the Xaxis was 0.50 mm, and that along
the Yaxis was 1.33 mm.
The maximum power of the developed blue laser system was
measured by L300W-LP1-50 laser measurement sensor (Ophir
Optronics Solutions Ltd), and the value was 250 W. As the
total power of 64 laser diodes was 256 W, it meant that the
part of light with the power of 6 W was lost in the system,
and the output efficiency of this system was 97.7%. The profile
of the actual laser beam was measured by SP620U CCD cam-
era (Ophir Optronics Solutions Ltd). Figure 3shows the mea-
sured profile of a laser beam at the focus position. The color in
the figure represents the laser intensity, and the color bar is
shown on the right. The actual profile in Xdirection was wider
than the designed one, and two laser intensity peaks appeared
in the profile, which were caused by the position offset of two
LD clusters. In the defocus situation, the dimension of the laser
beam profile would be larger, and these two laser intensity
peaks would be more obvious.
The water calorimetric method  was used to measure the
absorption rate of the blue laser. The upper surface of the steel
was irradiated with the laser, and the lower surface of the
steel was cooled with a flow of water. The temperatures of
the water at the entrance and the exit were measured, and
the mass of the water flow was recorded. The absorption rate
of the single-mode fiber laser with the maximum power of
500 W was also measured as a control. The wavelength of this
fiber laser was 1070 nm, and the beam parameter product was
1.04mm mrad. The defocus distance of the fiber laser was set
to be −5.0 mm, ensuring that the heat conduction mode was
the same for the fiber laser as for the blue laser. The power of
Fig. 1. Blue laser system: (a) schematic of the light path, (b) sche-
matic of the laser diode cluster, (c) photograph of the blue laser equip-
ment, and (d) photograph of the blue laser passing through water. LD:
Fig. 2. Designed power density distribution of the blue laser system at the focus position: (a) overall view, (b) along the Xaxis, and (c) along the
2252 Vol. 42, No. 12 / June 15 2017 / Optics Letters Letter
the fiber laser was set to be 250 W. The intersection angle be-
tween two kinds of laser beams and the normal line of the plane
was 5°, the moving speed of the specimen was 5 mm/s, and the
laser irradiation time was 6 s. For each laser, the experiments
were repeated three times, and the average value was used as the
absorption rate. Figure 4(a) shows the variation in water tem-
perature during the experiment. The variation of water temper-
ature in the blue laser experiment was larger than that in the
fiber laser experiment. In addition, it took a longer time to cool
the specimen in the blue laser experiment than in the fiber laser
experiment. The reason for these phenomena was that more
heat was absorbed by the steel in the blue laser experiment.
The measured flow speeds of water in the blue laser experiment
and the fiber laser experiment were 7.8 and 4.8 g/s, respectively.
Then the absorption rate could be calculated based on the mea-
sured water temperature variation and the flow speed of water.
Figure 4(b) shows that the absorption rate of the blue laser was
2.75 times that of the fiber laser, which confirmed that the
blue laser system had a considerably higher absorption rate
for the steel, compared with the single-mode fiber laser system
The suitability of the blue laser for heat treatment was va-
lidated by using stainless steel SUS420J1, which initially con-
sists of the austenite phase. Laser heating converts the austenite
phase to the martensite phase, and increases the hardness.
During the laser heat treatment, the power density of the laser
was reduced by setting the defocus distance to −15 mm to
avoid melting the surface. The power was set as 250 W; the
speed was set as 1 mm/s; and the intersection angle between
the laser beam and the normal line of the plane was set as
5°. The profile of the laser beam at the defocus status was mea-
sured with a test paper. The test paper was irradiated with the
blue laser at the power of 30 W for 1 s; the area of the paper
irradiated with the laser was burned away, recording the laser
profile. Figure 5(a) shows that the blue laser system at this de-
focus distance had two laser intensity peaks, and that the laser
intensity between the two peaks was low. Thus, the direction of
the heat treatment should affect the properties of the treated
zone substantially. Two typical laser treatment directions were
examined. The first direction was parallel to the line passing
through the centers of two peaks [(i) in Fig. 5(a)], and the char-
acteristics of the treated section are shown in Fig. 5(b). The
second direction was perpendicular to the line passing through
the centers of the two peaks [(ii) in Fig. 5(a)], and the character-
istics of the treated section are shown in Fig. 5(c).
In the first treatment, the metal was treated with two peaks
sequentially, and heat accumulated in the zone corresponding to
the center of the peak. A treated depth of 0.76 mm was obtained
because of the heat accumulation, which was larger than that of
the second treatment. In the second treatment, the two laser in-
tensity peaks were moved in parallel, and two treated regions were
Fig. 3. Measured laser beam profile of the developed blue laser sys-
tem at the focus position.
Fig. 4. Absorption rate experiment: (a) variation in water temper-
ature and (b) measured absorption rate.
Fig. 5. Laser hardening experiment: (a) profile of the defocused laser beam and laser hardening directions; (b) section showing the characteristics
of the specimen hardened in the (i) direction; (c) section showing the characteristics of the specimen hardened in the (ii) direction.
Letter Vol. 42, No. 12 / June 15 2017 / Optics Letters 2253
obtained with a greater overall width of 5.06 mm, compared with
the first treatment. Thus, the first heat treatment direction pro-
duced a deeper heat-treated region, whereas the second heat treat-
ment direction produced a wider heat-treated region.
Our blue laser system can be used in laser cladding, heat
treatment, and other applications that require no keyhole.
The high absorption rate makes this blue laser promising for
saving energy in material processing. After mass production
is achieved, the cost of the developed blue laser system may
be as low as $10,000 per kilowatt, which is significantly cheaper
than the existing laser system. In the future, a blue laser system
with higher power will be produced with the structure intro-
duced in this Letter, which can be used in laser welding, and
heat treatment with a greater depth requirement. However, the
difficulty to control the position offset of LD clusters will
increase with the laser diode number.
Funding. Advanced Research Program for Energy and
Environmental Technologies, New Energy and Industrial
Technology Development Organization (NEDO).
1. H. Cline and T. Anthony, J. Appl. Phys. 48, 3895 (1977).
2. E. Schubert, M. Klassen, I. Zerner, C. Walz, and G. Sepold, J. Mater.
Process. Technol. 115, 2 (2001).
3. J.-P. Kruth, G. Levy, F. Klocke, and T. Childs, CIRP Ann. 56, 730
4. A. K. Dubey and V. Yadava, Int. J. Mach. Tools Manuf. 48, 609 (2008).
5. T. H. Maiman, The Laser Odyssey (Laser, 2000).
6. J. D. Crow, L. D. Comerford, R. A. Laff, M. J. Brady, and J. S. Harper,
Opt. Lett. 1, 40 (1977).
7. J. Hecht, Opt. Eng. 49, 091002 (2010).
8. C. K. N. Patel, Phys. Rev. 136, A1187 (1964).
9. K. Yin, R. Zhu, B. Zhang, G. Liu, P. Zhou, and J. Hou, Opt. Express 24,
10. X. Zhu, W. Shi, J. Zong, D. Nguyen, R. A. Norwood, A. Chavez-Pirson,
and N. Peyghambarian, Opt. Lett. 37, 4167 (2012).
11. T. Du, Z. Luo, R. Yang, Y. Huang, Q. Ruan, Z. Cai, and H. Xu, Opt.
Lett. 42, 462 (2017).
12. Y. Kawahito, N. Matsumoto, Y. Abe, and S. Katayama, Weld. Int. 27,
13. D. Bergström, “The absorption of laser light by rough metal surfaces,”
Ph.D. thesis (Luleå University of Technology, 2008).
14. F. Dausinger and J. Shen, ISIJ Int. 33, 925 (1993).
15. Q. Nguyen and C.-Y. Yang, Int. J. Heat Mass Transfer 102, 1222
16. Y. Kawahito, N. Matsumoto, Y. Abe, and S. Katayama, J. Mater.
Process. Technol. 211, 1563 (2011).
17. M. A. Ordal, L. L. Long, R. J. Bell, S. E. Bell, R. R. Bell, R. W.
Alexander, and C. A. Ward, Appl. Opt. 22, 1099 (1983).
18. A. M. Prochorov, Laser Heating of Metals (Hilger, 1990).
19. J. H. Weaver, C. Krafka, D. W. Lynch, and E. E. Koch, Phys. Data 18,
20. L. Li, Opt. Lasers Eng. 34, 231 (2000).
21. F. Bachmann, Appl. Surf. Sci. 208, 125 (2003).
22. S. Nakamura, T. Mukai, and M. Senoh, Appl. Phys. Lett. 64, 1687
23. T. Suhara, Semiconductor Laser Fundamentals (CRC Press, 2004).
24. S. Nakamura, Ann. Phys. 527, 335 (2015).
25. S. Nakamura, T. Mukai, and M. Senoh, J. Appl. Phys. 76, 8189 (1994).
26. J. Endriz, M. Vakili, and G. Browder, IEEE J. Quantum Electron. 28,
27. Z. Zhu, M. Jiang, H. Yu, Y. Hui, H. Lei, and Q. Li, Opt. Lett. 41, 4712
28. Y. Kawahito, N. Matsumoto, Y. Abe, and S. Katayama, Weld. Int. 26,
2254 Vol. 42, No. 12 / June 15 2017 / Optics Letters Letter