Sensitive terahertz (THz)-wave sensor at room temperature is crucial for most applications such as 2-dimensional realtime
imaging and nonlinear phenomena in semiconductors caused by multi-photon absorption, light-induced ionization,
and saturated absorption. LiNbO3 is a promising material for frequency up- and down-conversion because of its high
nonlinearity and high resistance to optical damage. In this report, we propose a slant-stripe-type periodically poled Mg
doped LiNbO3 (PPMgLN) crystal for the construction of a practical THz detector. The PPMgLN solves compromised
optical design and low coupling efficiency between THz and infrared (IR) pump beam due to imperfect dichroic coupler.
The effective coupling of both pump beam and THz-wave into identical interaction region of up-conversion device
promotes the THz detector in practical use. The phase-matched-condition in slant-stripe-type PPMgLN was designed to
offer collinear propagation of two optical waves, the pump and up-conversion signal beams, because of efficient
frequency conversion. The phase-mached-condition was calculated and a slant-stripe-type PPMgLN with an angle (α) of
20° and a grating period (Λ) of 29.0 μm was used in this experiment. A minimum detectable energy of 0.3 pJ/pulse at the frequency of 1.6 THz was achieved with the pump energy of 1.8 mJ/pulse in room temperature. The dynamic range of the incident THz-wave energy of 60 dB was demonstrated. Further improving for the sensitivity using longer interaction
length in a PPMgLN crystal was also investigated.
Terahertz imaging has attracted a lot of interests for more than 10 years. But real time, high sensitive, low cost THz imaging in room temperature, which is widely needed by fields such as biology, biomedicine and homeland security, has not been fully developed yet. A lot of approaches have been reported on electro-optic (E-O) imaging and THz focal plane arrays with photoconductive antenna or micro-bolometer integrated. In this paper, we report high sensitive realtime THz image at 60 frames per second (fps) employing a commercial infrared camera, using nonlinear optical frequency up-conversion technology. In this system, a flash-lamp pumped nanosecond pulse green laser is used to pump two optical parametric oscillator systems with potassium titanyl phosphate crystals (KTP-OPO). One system with dual KTP crystals is used to generate infrared laser for the pumping of THz difference frequency generation (DFG) in a 4- Dimethylamino-N-Methyl-4-Stilbazolium Tosylate (DAST) crystal. The other one is for generation of pumping laser for THz frequency up-conversion in a second DAST crystal. The THz frequency can be tuned continuously from a few THz to less than 30 THz by controlling the angle of KTP crystals. The frequency up-converted image in infrared region is recorded by a commercial infrared camera working at 60 Hz. Images and videos are presented to show the feasibility of this technique and the real-time ability. Comparison with a general micro-bolometer THz camera shows the high sensitivity of this technique.
The dynamic properties of the laser emission are very important in studying the characteristics of the laser and may
reveal the underlying operating mechanism. Here we report a more precise measurement of the build-up time of random
laser pumped by picosecond pulse laser. The build-up time is defined as the time delay from the peak of the pumping
pulse to that of the emission. The random laser is R6G dye solutions with nanometer size TiO 2 as the scatterer. Various
dye concentrations and scatterer density are tried and measured. A specially customized fiber and a streak camera with a
spectrometer are employed to make the simultaneous measurement. The fiber has two branches and the lengths of both
branches are made equal with a difference of much less than 1 mm. The dispersion of the fiber, which introduces much
error in the results, is also measured and later compensated in the following data processing. The streak camera with
spectrometer can catch the random laser pulse and the pumping pulse signal in one shot with a resolution of less than 2
picoseconds. The results show that the build-up time changes evidently with the dye concentration, while it changes a
little along with the scatterer density. The pulse width almost remains the same in our experiment considering the errors.
We present a theoretical model to maximize the illumination and collection efficiency in designing fiber optic probes for
biomedical spectroscopy measurement. This model is in general applicable to probes with single or multiple fibers. We
investigated a number of probe configurations and find that contact measurement is very inefficient for multi-fiber
probes. Contact measurement is a good choice for single-fiber probes, but for multi-fiber probes, there is an optimal
probe distance. By carefully choosing the probe and sample distance, the signal can be enhanced by 5-10 fold.
Experiments demonstrate the distance dependence of collection efficiency in multi-fiber probes.
We experimentally measured the spectral and the temporal properties of various emissions from the Rhdamine 6G dye
solution pumped by picosecond laser pulses. These emissions involve the transverse amplified spontaneous emission
(ASE), the longitudinal ASE, the longitudinal ASE coupled with partial and stronger coherent feedback, and the random
laser. The random laser is made by adding TiO2 nanopowder to the dye solution as scatterer. The spectral and the
temporal shapes and widths of these emissions are measured using a spectrometer and a streak camera combined with a
CCD. The transverse ASE and the longitudinal ASE have similar spectral widths, while the longitudinal ASEs coupled
with the coherent feedback show narrowed spectral widths. The pulse shapes and widths of the longitudinal ASE with
coherent feedback show how the coherent feedback comes to existence. The experimental results show clearly the effect
of the degree of the coherent feedback on the spectral narrowing. The spectra and the pulses of the random laser of some
scatterer densities are presented, showing that the random laser without sharp peaks also has partial coherent feedback in
it at certain scatterer densities in the weak scattering regime.
We present time-resolved measurements of pulse transmission at wavelength 532 nm (60 ps pulse width, 10 Hz
repetition rate) on samples of titanium powders suspended in methanol. The average particle diameter of the powders is
80 nm. We used a streak camera with 2 ps time revolution to record the transmitted signals. When the particle density is
low, the results agreed with the diffusion theory and we obtained the time-independent diffusion constants. By adding
the titanium powders gradually in methanol, we obtained the relationship between the diffusion constant and the particle
density of TiO2 in the suspended solution. When using the TiO2 powders as the sample with a particle density of
1.36x1015 cm-3, the experimental result showed a little deviation from the diffusion theory, which may be the signature of
localization in the random media.
We studied the spectral shift of random lasing in the Rhodamine 6G dye solution with TiO2 nanoscatterers under
picosecond pulses pumping. The red shift, resulting from the re-absorption and re-emission of the dye, indicates a longer
optical path length of the emitted laser traveling inside the medium. Thus the optical paths of the random laser in the
solution can be estimated using the values of red shifts for different dye concentrations and scatterer densities. The
diffusion theory is provided and the theoretical results agree very well with that calculated from the red shifts before the
inflection points appear for increasing scatterer density. The followed increasing scatterer density results in the lights
staying longer in the medium, in contrast to that predicted by the diffusion theory. So it is clear that the inflection point
shows that the system is changing from a diffusion system to a weakly localized one in which the light stay longer
because of the localization.