Salut Tout le Monde,
Bianca and I discussed the pump-probe setup for THz measurements and I feel more confident explaining what I know.
The equipment used to acquire THz signals can have response times on the order of milliseconds, microseconds at best, but certainly not picoseconds. The THz signals themselves are of a couple picoseconds duration. Thus collecting these waveforms is a tricky matter; but with the introduction of a probe beam path with an optical delay line the system can have enough time to take data.
Each individual femtosecond pulse centered at 800 nm is split by a beam splitter and the resulting pulses travel the pump and probe paths respectively. It is essential that these pulses originate from the same pulse for temporal and spatial coherence's sake. If the pulses used were from different lasers, for example, the THz signal would not be detected. The pulses along the pump line interact with a photoconductive transmitter to produce THz radiation that then is either transmitted or reflected by a sample and finally arrive at a receiver similar to the transmitter. The pulses along the probe line can be delayed, travel a longer path than the pump line, and arrive at the receiver. (Note that the same can be true for the pump line. Just a difference in length between the two paths is needed.) The probe pulses reach mirrors on a mechanical scanner, which translates in the direction of pulse propagation. The probe pulse is used to inject carriers into the receiver and the THz radiation from the sample supplies a transient electric field (i.e., bias) needed to produce a current across the receiver's electrodes that is then detected by the measuring equipment.
The way in which the gate sampling/delay line works is that it allows for an incremental measurement of the THz radiation waveform. Looking at the figure above: The probe path is increased in step sizes equal to a. At each step the mechanical scanner pauses for however long it takes for your measuring equipment to respond to a current change in the detector (e.g. , 30 ms). Distance a corresponds to a position along the transmitted waveform. So the probe pulses will strike the receiver giving the equipment enough time to acquire one amplitude and time value off of the waveform. For the case of a 5 ps transmitted pulse and steps of a = 150μm, corresponding to a 1ps delay (~2a= c*Δt), you can acquire the waveform in 5 steps. In this way the THz signal profile can be pieced together. For finer resolution, a can be much shorter. The total sampling window size is determined by l. (There are limitations to the mechanical scanner so if resolution is important, it is probably best to make your sampling window shorter, collecting just the waveform, and then minimize your step size.) One thing to note is that the pauses and system response are much larger than the pulse duration, but the transmitted THz pulses strike the detector every 12.5 ns (80 MHz repetition rate). Each time a measurement is made, it is essentially of a different THz pulse. However, the waveform measurements are done repeatedly and each collected point is an average of multiple readings.
Concerning the types of scans:
Rapid scan allows for shorter time intervals between measurements, (this is dictated by system response time) but does not have good signal to noise ratio, might cut out some signal frequencies, and is not very stable (e.g., can introduce vibration to the setup). Its speed is beneficial in large sample imaging because it reduces the time you would have to sit around and wait.
When it comes to accuracy, stability, resolution, and bandwidth (which are essential in spectroscopy) a slower scan proves more useful.
Bianca and I discussed the pump-probe setup for THz measurements and I feel more confident explaining what I know.
The equipment used to acquire THz signals can have response times on the order of milliseconds, microseconds at best, but certainly not picoseconds. The THz signals themselves are of a couple picoseconds duration. Thus collecting these waveforms is a tricky matter; but with the introduction of a probe beam path with an optical delay line the system can have enough time to take data.
Each individual femtosecond pulse centered at 800 nm is split by a beam splitter and the resulting pulses travel the pump and probe paths respectively. It is essential that these pulses originate from the same pulse for temporal and spatial coherence's sake. If the pulses used were from different lasers, for example, the THz signal would not be detected. The pulses along the pump line interact with a photoconductive transmitter to produce THz radiation that then is either transmitted or reflected by a sample and finally arrive at a receiver similar to the transmitter. The pulses along the probe line can be delayed, travel a longer path than the pump line, and arrive at the receiver. (Note that the same can be true for the pump line. Just a difference in length between the two paths is needed.) The probe pulses reach mirrors on a mechanical scanner, which translates in the direction of pulse propagation. The probe pulse is used to inject carriers into the receiver and the THz radiation from the sample supplies a transient electric field (i.e., bias) needed to produce a current across the receiver's electrodes that is then detected by the measuring equipment.
The way in which the gate sampling/delay line works is that it allows for an incremental measurement of the THz radiation waveform. Looking at the figure above: The probe path is increased in step sizes equal to a. At each step the mechanical scanner pauses for however long it takes for your measuring equipment to respond to a current change in the detector (e.g. , 30 ms). Distance a corresponds to a position along the transmitted waveform. So the probe pulses will strike the receiver giving the equipment enough time to acquire one amplitude and time value off of the waveform. For the case of a 5 ps transmitted pulse and steps of a = 150μm, corresponding to a 1ps delay (~2a= c*Δt), you can acquire the waveform in 5 steps. In this way the THz signal profile can be pieced together. For finer resolution, a can be much shorter. The total sampling window size is determined by l. (There are limitations to the mechanical scanner so if resolution is important, it is probably best to make your sampling window shorter, collecting just the waveform, and then minimize your step size.) One thing to note is that the pauses and system response are much larger than the pulse duration, but the transmitted THz pulses strike the detector every 12.5 ns (80 MHz repetition rate). Each time a measurement is made, it is essentially of a different THz pulse. However, the waveform measurements are done repeatedly and each collected point is an average of multiple readings.
Concerning the types of scans:
Rapid scan allows for shorter time intervals between measurements, (this is dictated by system response time) but does not have good signal to noise ratio, might cut out some signal frequencies, and is not very stable (e.g., can introduce vibration to the setup). Its speed is beneficial in large sample imaging because it reduces the time you would have to sit around and wait.
When it comes to accuracy, stability, resolution, and bandwidth (which are essential in spectroscopy) a slower scan proves more useful.
No comments:
Post a Comment