Laser Gain Medium Interactions

[stevenjones3.1]

I have a diode laser that emits multi-mode light. Two frequencies of this light are selected in an external cavity and reflected back into the laser diode. The laser diode's gain medium then resonate at these two frequencies and emits dual mode light. The laser operates in the 1550nm region.

When the two back-reflected frequencies have a difference of around 2THz a third average mode is also emitted. That is, for proper power balance the laser diode will then emit three modes of light simultaneously, the two expected modes and an additional mode with a frequency at the average of the other two.

I am looking for some sort of theory or explanation for this, likely a non-linear interaction but I have not idea what. It is not four-wave-mixing since the new light mode is at the average frequency and not 2*f1-f2 or 2*f2-f1 which would be outside of the other light modes.

Does anyone have any thoughts on what could be the cause of this?

Thanks in advance for any help!

 

[UltrafastPED]

Consider the laser cavity modes: http://en.wikipedia.org/wiki/Mode-locking#Laser_cavity_modes

The equal spaceing of your modes suggests that central longitudinal mode is sometimes being suppressed.

Do you have good spectroscopic studies of the uncontrolled multi-mode light from the diode laser, and also for the controlled light output?

 

[stevenjones3.1]

I'm a little unclear what you are suggesting.

This is an external cavity laser so the dominant modes should be those decided by the Bragg gratings. However is seems as though when the light is back reflected into the laser some sort of nonlinear interaction occurs where an average frequency is produced from the two back reflected frequencies. Almost as if each frequency is pulled towards the other and they settle on some intermediate frequency.

Both the uncontrolled light and controlled output is observed with a optical spectrum analyzer

 

[UltrafastPED]

Yes ... but you are asking detailed technical questions about "possibilities and theories" which could be affecting your laser, but you are not providing very much information.

I presume that for a commercial system you would call their tech support ... hence it must be something built in your lab. In that case you need to share more information; do you have a schematic for the system? Can you share the output spectrums for the controlled and uncontrolled results?

You can check "About Me" for my laser background ... it's mostly ultrafast lasers and non-linear optics.

 

[UltrafastPED]

Also, is this related to your question "vibrational modes of quartz crystal"?

Is the quartz crystal your external cavity?

 

[stevenjones3.1]

Yes it is.

The system is a multi-mode laser diode emitting light towards a quartz crystal being used as an acousto optic tunable filter (AOTF). The AOTF is driven by two external sinusoidal signals operating at 292MHz and 294MHz (or anywhere from 280-310MHz but that is where the third mode is first seen). The AOTF is meant to diffract two selected frequencies of light which are then reflected off of a mirror and retrace their path and finally return to the laser diode. Once the light has been back reflected into the laser diode the laser gain medium resonates at the two selected frequencies and emits two tone optical light around 1550nm.

When the frequency driving the AOTF is <= 2MHz (optical differential frequency <=2THz ish) three optical peaks are seen and strong interference occurs between both of the output freuqencies (i.e. changing the controlling variables of one frequency will affect the other). Once withing the interference region, 3, 2, or 1 optical peak can be observed depending on the differential frequency and the supplied power to each vibrational mode of the AOTF.

The optical output is measured with an optical spectrum analyzer at 10GHz resolution.

Current theories are that this third average peak is due to nonlinear interactions within the AOTF or the Laser diode's gain medium. A similar phenomena (although not identical) has been reported in dual mode DFB lasers with two bragg gratings where a compound cavity is created due to both gratings.

Attached are 4 output spectrum for approximately equal supplied power to both vibrational modes, if the power to one mode is significantly stronger than the other than the corresponding optical peak will remain WITH the new middle peak (their relative amplitudes will vary with power settings).

 

[stevenjones3.1]

hope this helps, thanks for all the support!

 

[UltrafastPED]

I'd like to focus on the quartz crystal ... what are the dimensions, what is the crystal cut, how is it mounted? If its a commercial unit, what is the make and model?

I've done some projects with ultrasound transducers and crystal oscillators ... there are many things that can go wrong ... but that is true for most everything! A picture of the crystal, its mount, and the excitation connections may ne useful.

 

[stevenjones3.1]

I do not have all of the information of the crystal at the moment but hopefully I will soon, but it is approximately 10mm x 5mm x 3mm and it is mounted by adhering the bottom to a surface. The transducers are attached along the long 10mm side and light also passes through the long side. I cannot get a picture of it at the moment but I have quickly made a fairly terrible illustration with paint, I hope it helps.

 

[stevenjones3.1]

The yellow squares are the electrical contacts, the thin black rectangle is the transducer and the thick black rectangle is the bulk AOTF crystal

 

[UltrafastPED]

OK ... I will work on this tomorrow!

 

[stevenjones3.1]

Thanks for the help!

Here is a slightly better illustration of what is going on. Nothing is to scale I was just going for emphasis on what is occurring

 

[UltrafastPED]

From my reading so far: the acoustic modulators should be attached to opposing, parallel surfaces of the rectangular prism. This was from the description of a high precision system.

More tomorrow ...

 

[stevenjones3.1]

My mistake, I got a reply from the manufacture the crystal material is TeO2 and it is [0 0 1] cut.

 

[Andy Resnick]

<snip>
Does anyone have any thoughts on what could be the cause of this?

Thanks in advance for any help!

UltrafastPED has some excellent ideas- I am wondering if the detector is responsible for the 'new' frequency- how are you measuring the spectral output?

 

[stevenjones3.1]

With on optical spectrum analyzer

 

[Cthugha]

Are you going through free space or do you use a fiber to couple into the osa?

 

[stevenjones3.1]

polarization maintaining optical fiber.