Page 490 - Tunable Lasers Handbook

450 Stephen Vincent Benson

2. METHODS OF WAVELENGTH TUNING
Obviously one can vary the resonant wavelength defined in Eq. (1) by vary-
ing any of the four parameters on the right-hand side: the electron energy, the
wiggler magnetic field, the harmonic number, or the wiggler wavelength. The
last two of these are not continuously variable. so they are more useful for
changing wavelength ranges rather than continuous wavelength tuning. There
are good reasons for using these parameters to extend the wavelength tuning
range, as will be shown later. I will discuss the advantages and disadvantages of
each method of wavelength tuning. One should remember that the methods are
not mutually exclusive but can all be used in one facility.
There are a few other means to tune the wavelength of the laser that I will
not discuss in detail. One is changing the average angle of the beam. This
method is usually not feasible because the gain degrades too strongly with the
electron-beam angular spread. The second is gas-loaded operation [lo]. This has
been demonstrated on both the Mark I11 and SCAFEL lasers at Stanford Univer-
sity but is still very technically challenging to implement and has not yet
achieved broadband tuning. Harmonic generation outside the laser has been
demonstrated using conventional second harmonic generation techniques [ 1 11.
In principle. it is possible to drive an optical parametric oscillator or amplifier as
well. These methods are quite useful when the wavelength range is limited by
the design of the laser, but more power can usually be obtained by operating the
laser at the desired wavelength.

2.1 Energy Tuning
The first demonstrated method of wavelength tuning was to change the elec-
tron-beam energy. This was done on the first EL at Stanford [12] but the tuning
range was limited to +lo% by the rather narrow reflectivity band of the res-
onator mirrors.
The group at Los Alamos National Laboratory (LANL) used copper mirrors
with hole output coupling to change the laser wavelength from 9 to 35 pm by
varying the electron-beam energy by a factor of approximately 3 [13]. The evi-
dence for lasing at the longer wavelengths was indirect however (the output win-
dow was opaque to the laser radiation) so it was not known with certainty
whether fundamental lasing was achieved over this range. In later work [ 141 the
LANL team demonstrated lasing over a range of 9 to 35 pm with direct observa-
tion of the laser light.
The far-infrared laser at the University of California at Santa Barbara
(UCSB) demonstrated operation at wavelengths covering the range of 200 to
SO0 pm [Is]. Tuning via energy change was continuous only over a very small
energy range due to the necessity of maintaining good energy recovery in the