# Comparison of beam divergence between laser diode and DPSS lasers

I'm currently comparing a pulsed laser diode and a pulsed DPSS laser. For the DPSS laser, the beam divergence is listed as "beam divergence (80% of total energy): $$<4 \text{mr}$$". The laser diode is said to already be collimated, and has a beam divergence angle of "$$\theta_\perp$$: $$\le 10^\circ$$" and "$$\theta_\parallel \le 12^\circ$$". I know that plain laser diodes have more divergent beams than DPSS lasers, but I want to understand the differences here and whether I can lower the beam divergence of the laser diode to be closer to the DPSS laser. What is meant by "beam divergence (80% of total energy)"? And what is meant by a beam divergence angle of "$$\theta_\perp$$: $$\le 10^\circ$$" and "$$\theta_\parallel \le 12^\circ$$"? Furthermore, how do I compare "beam divergence (80% of total energy): $$<4 \text{mr}$$" to a divergence angle of "$$\theta_\perp$$: $$\le 10^\circ$$" and "$$\theta_\parallel \le 12^\circ$$"? And, lastly, how can I decrease the beam divergence of the LD so that it is closer to the DPSSL? The experiment is over relatively short distances (varying between 10-100cm).

## EDIT

This datasheet seems to have the same parameters.

• I noticed you posted a lot of laser related questions, but without "laser" as tags. Just saying I would have missed it as 80% of the time I browse only laser related topics, and this question is for laser experts. Commented Jul 21, 2021 at 19:21
• Hi @JoséAndrade. Based on the tag description for "laser", I thought it seemed more-so for questions about light amplification by stimulated emission (more pure laser questions), rather than photonics (lasers and laser electronics), so I left it out. Perhaps one of the moderators can clarify what the correct course of action is. Do you happen to know the answer to my question here? It doesn't seem like I'm getting a response. Commented Jul 21, 2021 at 19:23
• writing it. And, yes, you are right, photonics is closer, but I rarely think of my job as photonics, so, it might be that. Commented Jul 21, 2021 at 19:24
• @JoséAndrade oh, ok, I see; thanks for telling me. I might post a question about this to meta.physics.stackexchange, so that one of the moderators can clarify. Commented Jul 21, 2021 at 19:26
• @JoséAndrade By the way, just to add some additional information, I am currently looking at aspheric lenses on Thorlabs. Not totally sure how suitable they are in this specific case (for the laser diode). Commented Jul 21, 2021 at 19:28

DPSS lasers typically have much better $$M^2$$ than laser diodes, hence the findings you have (0.2° for the DPSS laser and 10/12° for the laser diode).

If you expand the beam, then you will intrinsically have lower divergence, so in principle yes, it is possible to have a laser diode have a divergence <0.2°, but at the cost of fluence, ie, your beam will be extremely big.

The 80% of energy is kinda close enough to the ISO standard of beam divergence (but not exactly), all it means is that 80% of the energy of the beam will always be contained in a cone with a half-angle of 4mrad.

I would imagine the values for the diodes are either given in FWHM or ISO standard values. You need 2 because diodes have 2 emission axis and due to the confinement in the gain region, each has its own divergence. I point you to this website to read more and understand what beam divergence is and how to compare the two.

For an experiment with less than a meter I don't think that the divergence will be your problem. As I seen you want to do interferometry, maybe your limiting factor would be coherence length instead. In any case, all parameters will probably be better with a DPSS laser.

• I think that $M^2$ should be unit-less and a source or beam of laser light can easily have a divergence of 10° - just put a high quality short focal length lens (e.g. f ~ dozen beam diameters) in front of an excellent DPSS laser and you can get a big divergence without hurting $M^2$. I think that the laser diode might have a large divergence simply because the size of the cavity mode is small, perhaps a dozen wavelengths. For semiconductor diode lasers to be efficient they need high current densities and so gain regions with small transverse size.
– uhoh
Commented Apr 6, 2022 at 2:03
• I also think that the OP is wrong that the laser diode is collimated, and so you've found yourself explaining something that wasn't true in the first place.
– uhoh
Commented Apr 6, 2022 at 2:03
• If you read my answer, you will see that indeed I said that expanding the diode beam, even with an M² in excess of 100 can give you a divergence of 0.2° at the cost of fluence, ie, your beam is probably 1m in diameter. I do believe the claim that the diode is collimated, it's well known that single emitter diode lasers have 2 axis with one of them with a really high M² and hence high divergence. Of course that the collimated beam diameter also plays a role, but for both the typical stated sources they usually have very similar diameters (~1mm). Commented Apr 7, 2022 at 8:19
• Maybe I should have not simplified it too much. The comparison in my answer to M²=1 was meant to illustrate a low divergence. Of course that 2 beams with the same diameter and different M² will display different divergences, as that is, per definition, the M². I will edit my post go reflect that. Commented Apr 7, 2022 at 8:26
• I don't feel comfortable seeing $M^2$ and divergence mixed like that. An $M^2=1$ beam can have huge or tiny divergence, in fact if $M^2$ deviates significantly from unity it becomes hard to even define convergence in a reasonable way. If you've got strong higher order transverse modes the shape of the beam could be quite ugly.
– uhoh
Commented Apr 7, 2022 at 9:52