Reducing loss of energy for Lasers

[Josiah]

TL;DR

Reducing loss of energy

Hi
I was told by someone that if you place the laser in a vacuum, then if would not loose energy. Is this true?

 

[anuttarasammyak]

Not only laser but light in general do not lose energy in vacuum.
Laser beam go in one direction. A beam receiver get the energy without the loss. Light, for an example sun light, goes all the direction whose energy density drops by r^-2 law. Total r^2-surface surrounding the light source get generated energy without loss.

 

[berkeman]

@Josiah -- Please take care not to try to reopen a closed thread. Your previous thread on this question was closed because you veered off into non-physical territory with your laser power questions. I will let this thread here proceed, but only if you stick to known Physics.

Here is how your previous thread was tied off:

The OP question has been sufficiently addressed. Thread closed. Thanks to all who participated.

 

[PeterDonis]

I was told by someone

This is not a valid reference. Can you find a valid reference (textbook or peer-reviewed paper would be best) that says what you're asking about? If not, we can't discuss it here.

 

[DaveE]

TL;DR: Reducing loss of energy

Hi
I was told by someone that if you place the laser in a vacuum, then if would not loose energy. Is this true?

Do you mean the laser beam, or the entire laser? What's in the vacuum?

There are some laser applications that require a vacuum, but they are pretty unusual. One is very, very short wavelengths, often called "vacuum uv" because of the absorption by O₂ and N₂. Another application is in vacuum spatial filters, where the beam is focused so small that the energy breaks down (ionizes) the air molecules. This can also be a problem with ultra fast laser pulses.

Gas lasers aren't vacuums in the gain media (laser tube), but they are quite low pressure.

But basically, no. Putting a laser in a vacuum won't make it better. In the above examples, it's just part or all of the beam that needs a vacuum, not the entire machine. You're likely to break it because of some reliance on air convection for cooling of various parts. I recall having to significantly overdesign a 5KW forced air cooled power supply so we could sell it in Denver et. al.

As in my reply to your previous post, you can't easily make a laser more powerful with simple tricks. They have already made it as powerful as they can given constraints like how much money you can spend. From laser pointers to fusion research, they are all optimized by people that know more than us.

PS: "not losing energy" is laughable in the laser world. They are all very inefficient devices, every single one.

 

[berkeman]

Another application is in vacuum spatial filters, where the beam is focused so small that the energy breaks down (ionizes) the air molecules.

I was at a photonics trade show many years ago, and as I walked down one of the aisles, I heard a "snap, snap" noise every few seconds. As I got closer, I saw that it was a large laser (about the size of a small car, including the power source), and they had it set up to focus the pulsing beam in the air behind some clear shields. You could see the ionization at the focus point with each laser pulse. Pretty impressive.

 

[Matterwave]

https://www.edmundoptics.com/knowle...es/lasers/gaussian-beam-propagation/?hl=en-US

I am no expert in laser technology, but I found the above article helpful!

A laser in a vacuum (I'm thinking here, laser light traveling in a vacuum) does not lose energy since there's nothing to lose energy to (with the exception of the cosmological red shift effect), but no laser is perfectly collimated. Diffraction effects at the edges will always spread the laser out and the spread for a perfect Gaussian beam (which no real laser achieves, but can be used as an estimate based on the above article) is given by eqn 3 in the article:

$$
\theta=\frac{\lambda}{\pi w_0}
$$

Where $w_0$ is the initial beam width (see fig. 1).

If you plug in some reasonable numbers for that you'll see how your beam spreads out. Once you see how your beam spreads out you can calculate how much energy a certain observer some distance away would receive from your laser.

I was thinking you're interested in finding out if you can send a light signal very far away. But say you shine a laser at Alpha Centauri, 4 light years away, you'll find that even if you had a gigantic laser (of order meters wide in lens shape) the laser would still get spread apart quite a bit by the time it goes 4ly.

 

[DaveE]

...a perfect Gaussian beam (which no real laser achieves...

Yes, they do. Single frequency TEM₀₀ propagation is a thing people have been using for nearly 50 years. About as perfect as anything in modern technology. That's how the computer you're looking at was made. Also, for many lasers it's where most of the energy is in the far field even if they have higher order modes too.

 

[Matterwave]

Yes, they do. Single frequency TEM₀₀ propagation is a thing people have been using for nearly 50 years. About as perfect as anything in modern technology. That's how the computer you're looking at was made. Also, for many lasers it's where most of the energy is in the far field even if they have higher order modes too.

Ah gotcha. I don't know this field that well, I was simply parroting back the article:

In many laser optics applications, the laser beam is assumed to be Gaussian with an irradiance profile that follows an ideal Gaussian distribution. All actual laser beams will have some deviation from ideal Gaussian behavior.

 

[berkeman]

That's how the computer you're looking at was made.

Can you elaborate? Do you mean photolithography used in IC fabrication?

 

[DaveE]

Can you elaborate? Do you mean photolithography used in IC fabrication?

Yes. Lasers are used at many steps, each tailored to their very specific task. There ate many applications; mask creation and inspection, wafer inspection before & after, photoresist exposure, annealing, dicing, via drilling... I'm definitely no expert, but I'd put my bet on mask creation as the most sensitive to beam quality.

 

[gentzen]

Yes. Lasers are used at many steps, each tailored to their very specific task. There ate many applications; mask creation and inspection,

No, high end masks are written by e-beam lithography, either by VSB (variable shaped beam) machines, or by electron multi-beam (grey scale pixel based) mask writers. Inspection can use lasers for OCD (optical scatterometry), but defect review and repair is done again by e-beam (or ion-beam).

wafer inspection before & after,

wafer inspection is „horror“, because features are too small for optical inspection, and e-beam inspection is still too slow. So OCD is always used, but alone it is insufficient. During process development and pre-production, e-beam inspection is used in addition, but it has its own challenges, in addition to being slow.

photoresist exposure, annealing, dicing, via drilling...

Well, laser drilling of silicon vias would be too slow and error prone. Reactive ion etching is your friend here. No idea whether PCB vias are drilled by laser.

I'm definitely no expert, but I'd put my bet on mask creation as the most sensitive to beam quality.

I do work in „semiconductor manufacturing“, decide for yourself whether that makes me an expert.
Many of those decisions what to use for which task are based on engineering considerations, like speed vs robustness vs accuracy vs costs. With e-beam VSB mask writers, mask inspection and repair is a fixed part of the process. And sometimes the mask has to be written a second time, because even repair could not fix it (in less than a day). Such a process is totally OK for mask writing, but a complete no-go for writing some layer of a chip.

 

[DaveE]

No, high end masks are written by e-beam lithography

I guess I'm not surprised I'm out of date here. I've been out of the tech game for a while now. We sold Ion lasers to Etec Systems* for many years for mask creation. Also to KLAC for inspection.

* Now part of AMAT

 

[gentzen]

I guess I'm not surprised I'm out of date here. I've been out of the tech game for a while now. We sold Ion lasers to Etec Systems* for many years for mask creation. Also to KLAC for inspection.

I always associate the MEBES e-beam mask writers (and the corresponding file format) with Etec. But they also had the ALTA advanced scanned-laser mask lithography tool product line. Today‘s laser-based mask writers mostly come from Mycronic (https://www.mycronic.com/product-areas/photomask-equipment/Products/slx-series/#description). I think they use UV laser diodes, with a wavelength somewhere around 350 nm. Probably ALTA used similar wavelengths, to avoid the hassle with chemically amplified resists.

 

[DaveE]

I always associate the MEBES e-beam mask writers (and the corresponding file format) with Etec. But they also had the ALTA advanced scanned-laser mask lithography tool product line. Today‘s laser-based mask writers mostly come from Mycronic (https://www.mycronic.com/product-areas/photomask-equipment/Products/slx-series/#description). I think they use UV laser diodes, with a wavelength somewhere around 350 nm. Probably ALTA used similar wavelengths, to avoid the hassle with chemically amplified resists.

Alta used either 257nm or 244nm from a large frame ion laser, I don't recall. Diodes have a tough time with the shorter wavelengths. That company was sold to AMAT at their peak and spectacularly imploded, gone in 5 years or so. A multi billion $ mistaken purchase.

AFAIK, KLAC still buys small frame ion lasers for some of their inspection machines.

PS: Those ion lasers are a huge PITA, it makes lots of sense to replace them with solid state lasers, if you can. It's a slowly dying cash cow for the manufacturers now. I used to say "no one wants to buy an ion laser, they have no other choice."

edit: Oops, I think I got my lasers confused. I'm pretty sure Etec used the UV lines at either 351nm or 364nm. KLA uses the shorter frequency doubled versions.