Antireflection Optical Coatings for High Power Fiber Lasers
Antireflection Optical Coatings for High Power Fiber Lasers
Gherorghe Honciuc, Emiliano Ioffe, Ophir Optics Group, MKS Instruments Inc.

High power kilowatt-level fiber lasers are used across various macro laser material processing applications, and particularly for metal cutting and welding machines. In such applications, various optical components such as lenses, mirrors, and windows, are used to deliver the laser beam to the processed material. As Continuous Wave (CW) fiber lasers reach power levels that exceed 10kW with diffraction-limited beam quality, the absorption in the optical components' coatings is one of the main limitations of the power-handling capabilities of the optical components. Absorption of the laser radiation in the coatings results in a temperature increase of the optical components due to heat diffusion which leads to risk of damage to the coatings and optics. In addition, absorption generates heat, which modifies the refractive index of glass and therefore modifying the properties of the propagated laser beam in an unfavorable manner. To avoid such effects, coatings and substrate materials with very low absorption rates should be used.

The most common substrate material is fused silica due to its optical properties and low absorption at the typical 1-micron wavelength of the fiber laser. The higher the power of the fiber laser, the lower the absorption of the fused silica needs to be.

Coating extended lifetime is the most important requirement for end users, and absorption is its limiting factor. Replacing optical components means downtime of the cutting machine, which can be reduced by selecting the right optical components.

In this paper we review the materials and coating technologies as well as measurement techniques used at Ophir Optics Group for high performance low absorption fiber lasers optics manufacturing.

Materials and coating technologies for high power lasers
When looking at the absorption properties of fused silica, there are two categories:

  1. Fused silica with 1-2ppm/cm absorption, e.g. Corning 7980.
  2. Fused silica with absorption less than 1ppm/cm level, e.g. Suprasil 3001¹.


Decisions regarding which type of fused silica to use are determined by a variety of factors, including the application, laser power, optical coatings, and the price of the optical components, which is a key factor mainly in high-volume production. In this vein, it is important to note that fused silica from category (b) is much more expensive than fused silica from category (a).

For fused silica from category (a), e-beam gun coatings, with or without Ion Assisted Deposition (IAD), have an absorption level of 2-12ppm. Glass absorption should always be correlated with the absorption of the coating for best results, as the overall lifetime of an optical component is limited by the overall optics absorption. The problem with the fused silica from category (a) is that glass producers don't guarantee any specific level of absorption, and there could be significant differences between glass batches. Therefore, measuring the optical component absorption before coating is a very important step in the production process.

The major requirements for optical coatings is to have low losses. For this purpose, and in addition to the low absorption requirement, we need the substrate to have also low roughness with minimal defects on the polished surface, as well as low reflection for anti-reflection (AR) coatings.

1. Low roughness considerations
At Ophir Optics, the standard roughness for fiber lasers is less than 1nm for plane and spherical polished surfaces, and less than 5nm for aspheric polished surfaces. This roughness must be controlled permanently in the production process. Fig. 1 represents the measurement of a plane polished surface.

Fig. 1 Roughness measurement of a plane polished surface

Most manufactures perform optical component cleaning manually. However, when the number of optical components is large, an automatic cleaning line is very useful. Any automatic cleaning process must have very strict controls, with a final manual visual inspection of each cleaned surface before coating. The second surface of the coating must always be cleaned manually, without affecting the coated surface.

2. AR coating considerations
Numerous papers describe how to obtain coatings with very low absorption or losses. The methods include e-beam gun evaporation with or without IAD)[2], Ion Beam Sputtering (IBS)[3], plasma-assisted reactive magnetron sputtering (PARMS)[4], sol-gel technique[5] Atomic Layer Deposition[6] , Molecular Beam Epitaxy (MBE) crystalline coatings[7], and anti-reflection (AR) microstructures (ARMs)[8].

This section mainly focuses on the e-beam gun evaporation method, and the evaporation materials HfO2, Ta2O5, and SiO2. The vacuum systems used for the coatings have e-beam guns, ion source for cleaning and for IAD, reactive evaporation, quartz crystal and optical monitoring, heating from above and below, and Meissner trap. The coating area is a class 1000 cleanroom. Cleaning, inspection, and packing are done in a class 100 environment.

The simplest AR coating is the V type, characterized by the "V" shape reflectance graph shown as a red line in Fig. 2. This type of coating has the thinner thickness for a high index material, which results in the lowest absorption. It is important to note that usually, before the 1-micron fiber laser beam is introduced, a red (visible) aiming beam is used to point at the site of processing area and, therefore, the AR coating must also have low reflectance in this spectral range. In Fig. 2, the blue line shows the reflectance of this coating (which is also a thicker double layer AR). If this type of coating is optimized for one laser wavelength (between 1030nm and 1070nm), then it's possible to guarantee a residual reflection of less than 0.1% for the defined wavelength. If these coatings are used for all wavelengths in the range of 1030-1070nm, it's possible to guarantee a residual reflection of less than 0.2% for this spectral range.

Fig. 2 Reflectance for: Red – AR V-type coating; Blue – double layer AR coating with thick high-index material; Green – four layer AR coating with low reflectance (R < 0.1%) in spectral range 1020-1090nm.

In addition to low absorption characteristics, the coatings should also have low reflectivity characteristics. In this vein, two factors contribute to the increase of the reflectance: (a) the coatings thickness on curved surfaces is not perfectly uniformed, resulting in shifted spectral reflectance, which increases the overall optics reflectance; (b) the non-uniformity on spherical calotte as well as the wide AOI range, increases the reflectance. Therefore, low reflectance must be guaranteed on a broader spectral range (especially for longer wavelengths). In these circumstances, we design multilayer coatings to obtain low reflectance in the spectral range of 1030–1090nm, as illustrated by the green line in Fig. 2.

As laser power increases, scattered and reflected light become important, as light is transformed into heat when hitting the inner surface of the cutting head. To enable uncooled cutting heads, the optical components absorption is even more significant. For low reflectance, there must be good coating uniformity over the component's surface, and the coating must ensure low reflectance for all relevant incidence angles.

For high power lasers (> 12 KW), we produce a coating with very low reflectance (R < 0.05%) for the spectral range of 1030-1140nm. This reflectance is represented in Fig. 3.

Fig. 3 Reflectance for: Red – multilayer AR coating with R < 0.05% in spectral range 1030-1140nm; Blue – double layer AR coating with Ta2O5 thick high-index material for 650nm and 1030nm; Green – double layer AR coating with HfO2 thick high index material for 650nm and 1050nm.

Table 1 below summarizes the properties of AR coatings for 1-micron high-power lasers based on Fig. 2 and 3.

To the best of our knowledge, the absorption values in the table are the best that can be achieved with IAD technology. Ion Beam Sputtering (IBS) is the only way to obtain lower absorption rates, but with little difference in coating lifetime. Price-performance plays a major role in this case.

Table 1: AR coatings comparison for 1-micron, high-power, fiber lasers

Coating 5011
Fig. 2 Red
Fig. 3 Green
Fig. 3 Blue
Fig. 2 Blue
Fig. 2 Green
Fig. 3 Red
Transmittance T1064 >99.8%
T650 > 60%
T1064 >99.6%
T650 > 90%
T1064 >99.8%
T650 > 60%
T1064 >99.6%
T650 > 85%
T1030-1080 >99.8%
T650 > 80%
T1030-1140 >99.9%
T650 > 80%
Reflectance at 1064nm R < 0.1% R < 0.2% R < 0.1% R < 0.2% R < 0.1% for 1030-1080nm R < 0.05% for 1030-1140nm
Absorption (*) < 8ppm < 10ppm < 10ppm < 10ppm 12ppm 12ppm
Hardness Good Good Very Good Very Good Very Good Very Good
CW Maximum power (KW) 10 10 10 10 15 17
(*) Values obtained with CPI absorption measurements on optical components.

3. Absorption measurements
Controlling the absorption of every optical component is the key to producing AR coatings for high power fiber lasers - this includes controlling the absorption in coatings, polished surfaces, and glass (substrate). From each coating batch produced, absorption is measured at several samples.

Absorption quality control during production is performed with a calorimetric OQM system. The opto-mechanical set-up is represented in Fig. 4. The pyro detector works in the spectral range of 8-12μm, measuring the surface temperature of the upper surface (fused silica is completely absorbent in this spectral range). To accurately represent the integral effects of absorption, the irradiation time is 3 minutes using a 500W CW laser at 1030nm.

It is quite problematic to calculate absolute absorption from the OQM measurement data, and it is not possible to compare data measured at samples with different shapes or sizes. Therefore, absorption measurements are also made with photothermal Common Path Interferometry[ 9] (CPI). With this method, absolute absorption can be measured, and the results can be used for calibrating the data from the OQM.

Fig. 4 Schematic illustration of the OQM opto-mechanic set-up for absorption measurement.

As was mentioned above, the glass manufacturer doesn't guarantee any specific level of absorption for the glass. In addition, the polishing process can affect the absorption. It is therefore crucial to check the optical components before coating. Every measurement referred to in the data below was taken on the same type of optical component: a 50mm diameter window with 5mm thickness, made from Corning 7980. Figs. 5(a) and 5(c) represent the absorption measurement of the same un-coated optical component. Figs. 5(b) and 5(d) represent the absorption measurements of the same optical component, coated with 5011 coating. The maximum value in each plot corresponds to the absorption value of the component with or without coating.

Fig. 5(a) CPI absorption measurement for un-coated optical component from Corning 7980.

Fig. 5(b) CPI absorption measurement for optical component after coating on both sides with 5011 coating

Fig. 5(c) Temperature variation of un-coated optical component from Corning 7980

Fig. 5(d) Temperature variation of optical component after coating on both sides with 5011 coating.

From a user's point of view, the OQM data is more useful than the CPI data because it provides direct information about the temperature increase during irradiation with a high-power laser beam. On the other hand, a major disadvantage of the OQM data is that it does not provide absolute absorption values, meaning that values from samples with different dimensions cannot be compared directly. Therefore, reference data and tolerances need to be defined separately for each type of lens/window.

One major problem in the production of coatings with low absorption is the uniformity of the layer thickness and optical constants (n and k) on all the elements in a coating batch (on the full calotte), especially when we use the ion source. The parameters of the evaporation process must be chosen to ensure minimal variation of absorption values of all components coated within a batch. A variation of 15% in absorption across the full calotte of a 760mm chamber is acceptable, but it must remain within the requested tolerances.

High power fiber lasers, working in the spectral range of 1.0–1.1μm, are largely used for macro-material processing in industry, and are gradually replacing CO2 lasers. The optics that deliver the beam from the laser to the processed material must withstand the high levels of thermal stress that occur as a result of the laser's high power. Ophir Optics develops a wide array of antireflection coatings from which customers can choose the best performance/cost ratio solutions for their applications:

  • Low to high power lasers: low absorption, high resistance
  • Residual reflectance less than 0.2% to 0.05%
  • Single wavelength to broader spectral range for curved surfaces
  • Large production volume at a short delivery time
  • Optical elements: windows, spherical and aspherical lenses

The production process is strictly controlled to achieve the best quality of our products by:

  • Pre-production substrate absorption tests
  • Roughness control after-polishing
  • Post-production tests, including spectral perfor- mance, absorption, and durability

Ophir Fiber Laser Coatings are used in a variety of applications with proven success in 15kW power lasers and higher.



  1. F. Nürnberg, M. Altwein, B. Kühn, A.Langner, G. Schötz, R. Takke, S. Thomas, J. Vydra, "Bulk damage and absorption in fused silica due to high-power laser applications" Proc. SPIE 9632, Laser-Induced Damage in Optical Materials: 2015, 96321R (23 November 2015); doi: 10.1117/12.2194289
  2. S. M. JAVED AKHTAR*, D.RISTAU, J. EBERT, H. WELLING, High damage threshold single and double layer antireflection (AR) coatings for Nd:YAG Laser: conventional systems", JOURNAL OF OPTOELECTRONICS AND ADVANCED MATERIALS Vol. 9, No. 8, August 2007, p. 2391 – 2399".
  3. Derrick Carpenter, Chris Wood, Ove Lyngnes and Nick Traggis, "Ultra-low Absorption Glasses and Optical Coatings for Reduced Thermal Focus Shift in High Power Optics"
  4. Dam Bé L. Douti, Laurent Gallais, Christophe Hecquet, Thomas Bégou, Julien Lumeau, et al. "Analysis of energy deposition and damage mechanisms in single layer optical thin films irradiated by IR an UV femtosecond pulses" Proc. SPIE 9627, Optical Systems Design 2015: Advances in Optical Thin Films V, 962718 (2 October 2015); doi: 10.1117/12.2192983
  5. Herve G. Floch, Philippe F. Belleville, "Scratch-resistant single-layer antireflective coating by a low-temperature sol-gel route", Proc. SPIE 1758, Sol-Gel Optics II, (7 December 1992); doi: 10.1117/12.132008.
  6. H. Liu, L. Jensen, H. Mädebach, P. Ma, and D. Ristau "Comparison of ALD and IBS HfO2/Al2O3 anti-reflection coatings for the harmonics of the Nd: YAG laser", Proc. SPIE 10691, Advances in Optical Thin Films VI, 106911B (5 June 2018);
  7. Garrett D. Colle, Wei Zhang, Bryce J. Bjork, …, High-Performance near- and mid-infrared crystalline coatings, OPTICA, Vol.3, No. 6, 2016, 647
  8. Douglas S. Hobbs*, Bruce D. MacLeod, Ernest Sabatino III, "Contamination resistant antireflection nano- textures in fused silica for laser optics". Laser-Induced Damage in Optical Materials: 2013, edited by Gregory J. Exarhos, Vitaly E. Gruzdev, Joseph A. Menapace, Detlev Ristau, MJ Soileau, Proc. of SPIE Vol. 8885, 88850J • © 2013 SPIE • CCC code: 0277-786X/13/$18 • doi: 10.1117/12.2042888.
  9. Ashot Markosyan, Roger Route, Helena Armandula, Martin Fejer, …" PCI Technique for Thermal Absorption Measurements", G080315/000/G080315-00.pdf.

Read TutorialRead as PDF