Hi Mohammadreza,

It probably would be useful to use a microscope objective. The laser light entering by the back of the objective and placing the fiber just in the focus of the objective. It would be helpful too using an x,y positioner- better with tilt correction- for the fiber.

Hope it helps!

Manuel

Hi dear Manuel

Thanks to you for your comment. The microscope is massive and have complex system! I need to a common system that I can use it for industrial applications, not laboratory applications.

Thanks so much

Mohammadreza

Mohammadreza Maleki: by "without dissipation" I assume you mean with minimal coupling losses.

Manuel Gómez' answer is appropriate if you have a well-collimated beam from your laser diode source. Note that he suggested using a microscope objective lens, not an entire compound microscope instrument.

Having said that, if the core diameter of your fibre is 1.5 mm, and the numerical aperture reasonably large (say 0.2 or greater), I see little need for tilt adjustment on the fibre mount. For industrial applications, a singlet lens is probably good enough. An aspheric lens, or a plano-convex lens with the planar side towards the fibre, and the convex side facing the collimated beam, will have lower aberrations than a bi-convex lens.

On the other hand, if you are attempting to couple light directly from the exit facet of a laser diode into the fibre, you need to consider the size of the active area and the far-field radiation pattern.

What is the numerical aperture of your fibre? Does it have a graded refractive index profile or (more likely) a step refractive index profile?

For efficiency reasons, the active volume of a diode laser is usually a thin stripe, with a far-field radiation pattern which is highly divergent out of the plane of the stripe. https://www.newport.com/t/laser-diode-technology

It is not uncommon for the divergence angle (half-width at 1/e2 intensity) to exceed 30°, with significant power at angles as high as 50 degrees.

Article Asymmetric broad waveguide diode lasers (λ = 980 nm) of larg...

If this is the case, a high numerical aperture lens is essential. A decent quality aspherical lens may be more cost-effective than a microscope objective, and could offer comparable efficiency. Precision machined and polished aspheres offer near diffraction-limited performance, but a moulded lens is probably good-enough to couple to your a 1.5 mm core diameter fibre.

https://www.edmundoptics.co.uk/c/precision-glass-molded/1233/#27929=27929_s%3AMC42NA2

In the case of a broad stripe laser such as that described in the paper by Lee, Mawst, & Botez linked above, the laser cannot be treated as a point source, and there will be some increase in aberrations at the outer edges of the active area of the laser facet. This will be more pronounced for shorter focal length lenses, but I doubt this will have a significant impact with focal lengths of 8 mm or larger. Shorter focal lengths could be acceptable, but I am guessing here as I don't have direct experience.

You appear to be aiming for 100% efficiency which will not be possible. 99% will be very difficult to achieve, with perhaps 85%-95% more likely. At normal incidence you lose 4% of light from each uncoated air-glass interface, so anti-reflection coatings with good performance at the laser wavelength are essential. Are you able to AR coat the fibre entrance and exit facets to avoid Fresnel losses here?

You can improve the collection of high angle light by combining a high NA aspheric with a longer focal length best form, plano-convex or achromat secondary lens. Mount with each lens' with most strongly curved surface facing the other lens. With a 100 μm stripe laser, I suggest a secondary lens directing light to the fibre with a focal length 5x to 10x larger than that of the asphere collecting light from the laser.

https://www.thorlabs.com/thorproduct.cfm?partnumber=ACL12708U-B

https://www.thorlabs.com/thorproduct.cfm?partnumber=LA1304-B

Magnification = focal length of secondary / focal length of primary asphere

Choose magnification > Aspheric lens NA / Fibre NA

Choose magnification < Fibre core diameter / laser stripe width

A ray tracing package such as Zemax will allow more precise optimisation if the highest possible efficiency is essential.

Dear Alan Robinson

Thank you very much for your help. Your advice helped me a lot.

My fiber NA is 0.47. According to your advices, I think the best option for me is to use of a Plano-concave lens with 5 numerical aperture.

I can deposition the lens surface and the fiber head. In your opinion, what material do I use to reduce the reflection at 650-700 nm?

Mohammadreza Maleki You said: "My fiber NA is 0.47. According to your advices, I think the best option for me is to use of a Plano-concave lens with 5 numerical aperture."

How does a diverging plano-concave lens help? This is not what I suggested.

Do you mean a plano-convex lens with numerical aperture 5.0 ? A numerical aperture greater than 1.0 is impossible for a lens in air, so I do not understand what you propose. See: https://en.wikipedia.org/wiki/Numerical_aperture

Can you clarify what you are trying to do, and in particular what are the properties of your laser diode source? What is the diameter of the exit beam and its divergence? What focal length and diameter do you propose for the lens? How far from the lens will you position the fibre - this and the focal length determines the size and numerical aperture of the spot focused onto the fibre core.

If your source is a reasonably well-collimated beam whose waist diameter is a few millimetres or less, then coupling into a 1.5 mm diameter 0.47 NA fibre core should be straightforward with any decent converging lens whose aperture is large enough to capture the full input beam width, and whose focal length is somewhat larger than the lens aperture diameter. With such a large fibre core, lens aberrations are unlikely to be a concern, but we cannot be sure without more information.

What are the properties of the double convex lens you have used previously? What coupling efficiency did you achieve? If this was ineffective, it suggests either a poor choice of lens, or that your source does not resemble a collimated Gaussian beam with a beam propagation (M2) factor near unity which is easily focused to a small spot.

https://www.rp-photonics.com/gaussian_beams.html

https://www.rp-photonics.com/m2_factor.html

If instead you have a laser chip with a highly divergent beam, then a single plano-convex spherical lens may well have too small a numerical aperture to collect all of the available light.

Anti-Reflection Coating: You said "I can deposition the lens surface and the fiber head. In your opinion, what material do I use to reduce the reflection at 650-700 nm?"

The simplest anti-reflection coating is a quarter-wavelength thickness of a material whose refractive index is the square root of the product of the refractive indices of the materials on each side of the surface to be coated. For a BK7 borosilicate crown glass lens in air at 670 nm, the refractive index of the glass is 1.5139, so the optimum coating index is 1.23, and the optimum thickness 136 nm.

https://refractiveindex.info/?shelf=glass&book=BK7&page=SCHOTT

https://refractiveindex.info/?shelf=main&book=MgF2&page=Dodge-o

https://en.wikipedia.org/wiki/Magnesium_fluoride

A widely used material for single layer coatings is Magnesium Fluoride, refractive index 1.3765. This is higher than our preferred value, but there are relatively few materials with suitable physical properties and a lower refractive index. Sodium fluoride, for example, has index 1.324, but is soluble in water, so would be very easily damaged.

Minimum reflectance for a single quarter-wave MgF2 coating on BK7 is around 1.3%, however much better performance is possible with multi-layer broadband or narrow band coatings.

https://www.edmundoptics.co.uk/resources/application-notes/optics/anti-reflection-coatings/

https://www.thorlabs.com/newgrouppage9.cfm?objectgroup_id=5840

https://www.thorlabs.com/newgrouppage9.cfm?objectgroup_id=5383

https://www.thorlabs.com/images/TabImages/Standard_BBARC_B_Graph_780.gif

From Thorlabs, the additional cost for an AR coated lens is typically £8 (or around $10) for a single lens, and I suggest you purchase suitable coated lenses rather than coat them yourself. There are many alternative suppliers, and you will almost certainly get a better deal if you need larger volumes.

A quarter wave coating on the silica fibre will be beneficial, provided the coating quality and thickness are within acceptable limits. Whether this is worth doing is for you to decide, but I would be inclined to address other aspects of coupling first.

Hope this helps.

Alan.

Mohammadreza Maleki: by "without dissipation", like Alan Robinson: " I assume you mean with minimal coupling losses", and you can minimize them as he clearly explains. Furthermore, it is important to know how good it is the fiber input surface. I mean, may be you have cut it yourself, and may be the cut is not flat, or it is not perpendicular to the fiber axis which could cause losses and a bad coupling, even with surface coatings. Probably it is not the case, but it is something easy to check and you could correct it readily.

Hope this helps.

Alan Robinson

Thanks a lot for your guidance.

I typed incorrectly plano-concave. I mean plano-convex lens with 0.5NA.

About “Can you clarify what you are trying to do, and in particular what are the properties of your laser diode source?“ I want to transfer a high-intensity laser light in to the body through the fiber optic.

About “ What is the diameter of the exit beam and its divergence? “ I want to transfer a high-intensity laser light in to the body through the fiber optic. θ//=10.5 & θ⊥=23.

Using your guidance, I have found the right lens for this.

Thanks a lot for your valuable guidance

Mohammadreza

Manuel Gómez Thanks a lot for your guidances. I mean with minimal coupling losses. I have cuted fiber. But, The cut is good.

Thanks a lot for your valuable guidance

https://www.youtube.com/watch?v=3JKRlEP7TGg

As mentioned Alan Robinson, after cutting fiber with zero angle and before coating that, it should be Polished enough also.

Calculating laser diode Focal depth, beem waste, apartur cone angle and choosing best lens regarding calculating and wave length are basics as you know.

Dear Alan Robinson

I need to use these equations in my paper:

Magnification = focal length of secondary / focal length of primary asphere

Choose magnification > Aspheric lens NA / Fibre NA

Choose magnification < Fibre core diameter / laser stripe width

Do you have any references? I need references for my paper.

Mohammadreza Maleki : I do not have a direct reference to the published literature.

The optical relay I suggested images the exit facet of the laser onto the core of the fibre. To minimise losses, you require that:

1. The image of the laser active area falls entirely within the core of the fibre.

2. Rays incident on the fibre are within the acceptance cone of the fibre.

To satisfy (1), the magnification must be less than the ratio of fibre diameter to laser stripe width.

The numerical aperture of the illumination reaching the fibre is given by the smaller of the NA of the secondary lens, or (the smaller of the NA of the aspheric lens, or the sine of the maximum radiation angle from the laser) divided by the magnification. In a well-corrected optical system, the dependence on magnification follows from the Abbe sine condition, or the conservation of Etendue.

https://en.wikipedia.org/wiki/Numerical_aperture

https://en.wikipedia.org/wiki/Abbe_sine_condition

https://en.wikipedia.org/wiki/Etendue

The simple relationship between focal length and magnification assumes that the laser is located at the focus of the aspheric, so that rays from the laser form a collimated beam which is substantially parallel to the axis as it reaches the secondary lens, which then forms an image of the exit facet at the focus of the secondary lens. The aspheric (and ideally the secondary lens also) are then operating under conditions for which they are optimised and aberrations are low.

This all follows from basic properties of lenses. If you need a traditionally published reference, you could use an undergraduate text book such as Jenkins & White, "Fundamentals of Optics" 4th ed. (1981), or a more recent work if you prefer.