This page is part of the [[optical:spy1|Spy1 construction tutorial]].

====== Build your own Yb:KYW femtosecond laser ======

Contributed by [[dejan@salk.edu|Dejan Vučinić]]

Biological multiphoton imaging has been dominated by
femtosecond Ti:sapphire lasers since its inception in
1990.((Denk, W., Strickler, J.H., Webb, W.W. "Two-photon laser scanning fluorescence microscopy",
[[http://www.sciencemag.org/cgi/content/abstract/248/4951/73|Science 248, 73 (1990)]].))
These lasers are complex and expensive, and most of
them have difficulty tuning to wavelengths beyond the
region of high absorption around 950nm.

Many other laser materials exist, but few have been
applied to biological imaging.  Ytterbium is particularly
well suited for lasing near 1000nm, a region particularly
interesting for the excitation of several popular
voltage-sensitive dyes.  When used within
a KY(WO<sub>2</sub>)<sub>4</sub>  (aka KYW)
host crystal, Ytterbium has a high stimulated emission cross-section
(3x10<sup>-20</sup> cm<sup>2</sup>) and a very long fluorescence
lifetime (0.6 ms), which makes it a relatively forgiving
material when designing a femtosecond laser.

This page contains instructions for building your own
femtosecond Yb:KYW laser from scratch.  Before embarking on this
adventure, understand that building a laser is not
very difficult, but it is a very **time-consuming**
hobby.  Tuning a resonator is just like tuning
a musical instrument: at first it will seem impenetrable,
but with a little practice it becomes quite easy.  The
best way to learn this skill is to have someone show
it to you, so if you have access to a laser lab by
all means try to get some hands-on instruction.  If not,
do not despair; the laser described here was my first
attempt, and I did it in a physiology
lab that was not equipped for optics research.
This crude first attempt has been in daily use for
several months now.


===== Safety! =====

**This laser will produce light at intensities
that can irreversibly damage your eye in an instant.**

Play it safe: always wear goggles, such as
[[http://www.thorlabs.com/thorProduct.cfm?partNumber=LG1|these]] from
[[http://www.thorlabs.com/|Thorlabs]], when working on it.

Here are a couple of signs you can print and display
near the area where the construction will take place:

{{cl4dps1.gif?100}} {{scarylaser.gif?70}}

===== Education =====

You will need a modicum of knowledge about lasers in
general to be successful at building this project.
The book [[http://www.amazon.com/o/ASIN/3540650644/ref=s9_asin_image_1/103-1322637-2658232?pf_rd_m=ATVPDKIKX0DER&pf_rd_s=center-2&pf_rd_r=0P5CE9SEYSZMHQWW2D7G&pf_rd_t=101&pf_rd_p=279530701&pf_rd_i=507846|Solid-state  
Laser Engineering]] by Walter Koechner is a good starting point.

You should read all the published literature
((G. Paunescu, J. Hein, R. Sauerbrey, [[http://www.springerlink.com/content/pnc2t436hr29qjq0/|Appl. Phys. B79, 555 (2004)]].))<sup>,</sup>((A. A. Lagatsky et al., [[http://scitation.aip.org/getabs/servlet/GetabsServlet?prog=normal&id=ELLEAK000039000015001108000001&idtype=cvips&gifs=Yes|Elec. Lett. 39, 1108 (2003)]].))<sup>,</sup>((P. Klopp et al., [[http://www.opticsexpress.org/abstract.cfm?id=67737|Opt. Exp. 10, 108 (2002)]].))<sup>,</sup>((H. Liu, J. Nees, G. Mourou, [[http://ol.osa.org/abstract.cfm?id=65549|Opt. Lett. 26, 1723 (2001)]].))<sup>,</sup>((F. Brunner et al., [[http://ol.osa.org/abstract.cfm?id=62097|Opt. Lett. 25, 1119 (2000)]].)) you
can find on Yb:KYW and
Yb:KGW lasers to get an idea what has been tried and what
works.  These papers will also give you a feel for which
design parameters affect what aspects of lasing.

When the rubber hits the road, however, there is no
substitute for modeling your cavity to develop
intuition for where in parameter space you are at
the moment.  The modeling [[#Software|software]] allows
you to do this without having to do calculations by hand.


===== Tools =====

Unless you have access to a well-equipped optics lab,
you'll need to buy or borrow a variety of tools.  An
infrared viewer and a power meter are absolutely
essential, any two-photon lab should have these.
A beam profiler and a spectrometer are normally used
only during construction and tuning, so if you're on
a budget try to borrow these.  Depending on your
choice of pump light source you may not need a beam
profiler at all.

==== Infrared viewer ====

The wavelengths at which Yb:KYW is pumped and lases
are invisible to the naked eye, some kind of infrared
viewer is absolutely necessary.

[[http://www.findrscope.com/|FJW Optical Systems]] sell
a variety of hand-held infrared viewers.  Get one that
can be connected to a monitor.

Alternatively, you can remove the infrared filter from
a cheap webcam---silicon-based CCDs are sensitive to 1100nm
and work just fine for our purpose.
There are many sites that describe how to do this with
common webcams, see for example
[[http://www.kailashnadh.name/docs/ir_cam/ir_cam.html|here]]
or
[[http://www.hoagieshouse.com/IR/|here]].
This is my preferred method since a webcam is easily
attached to a tripod, freeing both hands for tuning.

==== Power measurement ====

To measure optical powers for this project you'll need
a power meter with a thermal head.
[[http://www.newport.com/|Newport]] and
[[http://www.thorlabs.com/|Thorlabs]] sell a variety of these.

If you opt for a low beam height it may be difficult
to fix the large thermal head close enough to the
breadboard for the detecting area to be in the beam path.
You may find it easier to build a custom detector using
a large photodiode (for example
[[http://www.digikey.com/scripts/DkSearch/dksus.dll?Criteria?Ref=165673&Site=US&Cat=34407245|PDB-C613-2]]
 or
[[http://www.digikey.com/scripts/DkSearch/dksus.dll?Criteria?Ref=166477&Site=US&Cat=34407245|PDB-C615-2]]
from [[http://www.advancedphotonix.com/|Advanced Photonix]])
with a neutral density filter
(for example NMC0120 from [[http://www.casix.com/|CASIX]])
affixed at the appropriate height to a mount that is easy to
clamp down onto the working surface.  Reverse-bias the
diode with a battery and connect to an oscilloscope for tuning.


==== Spectrometer ====

You will need a spectrometer sensitive in the 930--1050 nm range.
[[http://www.oceanoptics.com/|Ocean Optics]] 
[[http://www.oceanoptics.com/products/usb4000.asp|USB4000]] with grating #4 
is a good choice.



==== Beam profiler ====

One of the most challenging tasks in the construction of this
laser is the shaping of the pump beam to match the shape
of the fundamental resonator mode.  The difficulty comes from
the need to measure the shape of a ~50μm spot receiving several
Watts of power.  Profiling at lower powers is not
accurate since the beam shape of common pump diodes
changes with power output.

CCD-based beam profilers are not up to this
task because they can't absorb so much heat, and putting a
neutral-density filter in front may deform the measured beam shape.
[[http://www.thorlabs.com/|Thorlabs]] sell a
[[http://www.thorlabs.com/NewGroupPage9.cfm?ObjectGroup_ID=804&Visual_ID=2140|slit-scanning beam profiler]]
that can absorb high powers.

An alternative is to build your own.  Here
is a beam profiler built using two razor
blades glued to a disk attached to an electric motor:

{{beam_profiler.jpg?100}}

The ring of darker black you see in the photo are burn
marks from the pump beam.  See above regarding eye protection...

Glue the blades to the disk under a dissection stereomicroscope
to make a ~5μm slit.
The motor spins the slit across the beam, and a 
photodiode placed behind the disk reports the shape
of the beam to an oscilloscope.  For example, if the spinning rate
is adjusted so the circumference of the disk in
millimeters at slit position is made equal to the
full rotation time in milliseconds, then the transit
time in microseconds reads out the beam shape in
micrometers, as shown on this oscilloscope readout:

{{beam_profiling.jpg?100}}

The three spikes on the top trace (two are covered by cursors)
are the crossings of the slit across the beam every 188ms,
corresponding to the 188mm circumference of the disk at the
radius at which the beam is incident.  The bottom trace is
a zoom of one of the spikes showing that the spot is roughly
Gaussian-shaped with a full width at half maximum (FWHM) of 95μs,
i.e. 95μm.  To convert this into the more commonly used 1/e<sup>2</sup>
spot diameter multiply by 1.7.

==== Software ====

While one should understand the basics of Gaussian beam math
before attempting to design a laser, in this day and age
calculations are seldom done by hand.  There are
several cavity design programs available.  I used
[[http://www.st-andrews.ac.uk/~psst/|Psst!]] from
[[http://www.st-andrews.ac.uk/|University of St Andrews]]
which is available for download free of charge.


===== Construction =====

==== Design elements ====

==== Pump beam shaping ====

==== Resonator ====

{{laser_beams.jpg?100}}