[time-nuts] lasers (was: Microsemi up for sale?)

[Attila Kinali]

On Sat, 3 Mar 2018 19:04:57 -0800
"Richard (Rick) Karlquist" <richard at karlquist.com> wrote:

> On 3/3/2018 6:18 PM, Hal Murray wrote:
> > Thanks.
> > 
> > Is there something fundamental in there, or is the lack of products because
> > nobody has made the big investment required to figure out how to do it.
> > 
> > What is the bandwidth of the laser?  What happens if it drifts slightly?  Can

Commercial laser diodes are available with line widths from 10MHz up
to a few 100MHz. The problem with this is, that the center frequency
varies with temperature and current. As ballpark figures you can
use 10GHz per °C and mA.

For laser pumping of atomic states, you need to stabilize the laser diode
with additional feedback, a so called extended cavity diode laser (ECDL).
Most commonly these days, a grating[1] or a distributed feedback (DFB, 
DFB external to the diode laser cavity is called "volume grating")
is used. The grating can be thought of as a mirror with wavelength/frequency
dependent reflection angle. A DFB is basically a wavelength selective
filter. Both, grating and DFB are put in the laser path, such that
they become part of the cavity, ie add another frequency selective element
to the laser "oscillator". With this, one can get the linewidht down
to 1kHz to 1MHz, mostly dependent on the strength of the feedback and
on how much temperature and current fluctuations affect the center
frequency.

A good summary on how different factors add up in an ECDL can be found
in [3]. The authors also presented a neat and cheap way how to build
an ECDL in [4]. An alternative, a bit more complicated, but also more
complete description can be found in [5].

Under the assumption you have selected the right diode and designed
the feedback system well, you can get system that will be first order
indepented of the laser diodes behaviour. I.e. the external cavity
will be the main wavelength determining element. 

The problems for long term use in commercial standards seem to be
the second order effects. Mainly the position of mode jumps (sudden,
large jumps of frequency when applying only a small change of tuning
to the system). While it is easy to tune the ECDL such, that these
mode jumps do not happen withing the region of interest, it is not
so easy to ensure that aging does not bring one of these mode jumps
right to the point where one would like to keep the laser. Detecting
of these mode jumps is also not easy, the straight forward way requires
a second laser to measure the difference frequency. A second way of
using the amplitude varation during mode jumps seems not to be as reliable.

But! For 1.5µm communication laser systems, all these problems have
been solved and we have very reliable lasers that stay on frequency
with a very high stability (modern laser communication systems use
a channel spacing as low as 1GHz @ 1.5µm). So there is hope that
these techniques will spill over to wavelength ranges more interesting
to atomic standards. Most likekly the first one would be 850nm
(close to the Ceasium D2 line), which is being used for short range,
cheap optical networks.


> When I was working on fiber optic communication test,
> I remember hearing about lasers that were "tuned" with
> variable Peltier coolers.  Power consumption is critical in
> a cesium standard that can run on batteries.  Maybe
> the power consumption of the coolers is a deal breaker.

If all you have is just a laser diode with at most DFB, but no
external cavity (as is usualy with communication lasers), then
the only two ways to tune are temperature and current.
But current will also modulate intensitiy, so people tend
to try to keep current more or less constant, while doing
the tuning with temperature. For a system with external
cavity, it's enough to keep current and temperature stable,
which makes the power consumption less pronounced. The gotcha
here is, that one might need to tune temperature a bit to
move a mode jump out of the way.

But even if you dont have an ECDL, it's only a laser you
have to cool, which is a very low mass, and the cooling power needed
is just what you pump into it (usually a 5mW laser is enough
for atomic spectroscopy, which translates to <100mW electrical
power needed).


On Sun, 4 Mar 2018 21:41:15 +1300 (NZDT)
Bruce Griffiths <bruce.griffiths at xtra.co.nz> wrote:

> External cavity lasers with piezoelectric tuning (usually varies cavity length or tilts frequency selective grating) are usually used for such applications.
> 
> Sacher laser do some compact units with integrated Peltier cooler etc:
> 
> https://www.sacher-laser.com/home/industrial-lasers/point_and_line_laser_module/industrial_laser_modules/micron_laser.html
> 
> To get pricing registration is required.

The Sacher module is a neat little device, that integrates all of the
components needed into a tiny little package. More information on its
construction and performance can be found in [6] (note: their calculation
of linewidth is wrong, it should be 36kHz/sqrt(2) = 25kHz).

The price was, when I looked at it 2 years ago, at 2500€.


			Attila Kinali


[1] https://en.wikipedia.org/wiki/Diffraction_grating

[2] https://www.youtube.com/watch?v=EuWI1WPrZ2I

[3] "Mode stability of external cavity diode lasers", by Saliba,
Junker, Turner, Scholten, 2009
http://dx.doi.org/10.1364/AO.48.006692
https://cpb-ap-southeast-2-juc1ugur1qwqqqo4.stackpathdns.com/blogs.unimelb.edu.au/dist/5/118/files/2015/11/Saliba-et-al.-2009-Applied-Optics-Mode-stability-of-external-cavity-diode-lasers-1buob6t.pdf

[4] "Littrow configuration tunable external cavity diode laser with
fixed direction output", by Hawthorn, Weber, Scholten, 2001
https://doi.org/10.1063/1.1419217
https://www.thevespiary.org/library/Files_Uploaded_by_Users/no1uno/pdf/Instrumentation/Laser.Design/Hawthorn.Weber.Scholten.Littrow.Configuration.Tuneable.External.Cavity.Diode.Laser.with.Fixed.Direction.Output.Beam.pdf

[5] "A narrow-band tunable diode laser system with grating feedback,
and a saturated absorption spectrometer for Cs and Rb", by MacAdam,
Steinbach, Wieman, 1992
https://doi.org/10.1119/1.16955
https://www.thevespiary.org/library/Files_Uploaded_by_Users/no1uno/pdf/Instrumentation/Laser.Design/Laser.Locking/MacAdam.Steinchach.Wieman.A.Narrow.Band.Tunable.Diode.Laser.System.with.Grating.Feedback.and.a.Saturated.Absorption.Spectrometer.for.Cs.and.Rb.pdf


[6] "Compact Bragg Grating Stabilised Ridge Waveguide Laser Module
with a power of 380mW at 780nm", by Rauch and Sacher, 2015
https://doi.org/10.1109/LPT.2015.2438545
https://www.sacher-laser.com/downloads/publications/5r1249_en.pdf


-- 
<JaberWorky>	The bad part of Zurich is where the degenerates
                throw DARK chocolate at you.