X-rays induce a current in many electronic devices. Very much like visible photons in photovoltaic solar cells. The signal is called x-ray beam induced current.
In other words, the test devices are operated as x-ray detectors and XBIC yields the local device performance. XBIC combines the high special resolution of electron beam-induced current with a high penetration depth of laser beam induced current. This combination yields the local performance even in varied structures like in encapsulated solar cells with high resolution.
From the XBIC signal, we can determine the spatially resolved charge collection efficiency, which is critical for the electrical performance of semi-conductor devices. So in principle, XBIC measurements can be performed on all systems that show electrical response on its space, such as solar cells, x-ray detectors, on nano wires from semi-conductors. Taking XBIC measurements is actually surprisingly simple if you follow the signal path from the device to the amplifiers and the data acquisition.
Begin by designing a sample holder to provide maximum freedom to the placement of different detectors in close proximity. Set the sample holder onto a kinematic base to allow easy repositioning of samples with micrometer position. Use a printed circuit board that has been designed so that it can be used as a mount for the electronic device for XBIC measurements.
Next, glue the electronic device to be tested onto the printed circuit board. Pay attention to avoid short circuiting by using polyimide tape. Fixate the contact wires also with tape.
Connect the upstream contact that is facing the incident x-ray beam with the shield of the coaxial cable. Then, connect the downstream contact with the core of the coaxial cable. Next, mount the printed circuit board into the sample holder.
Then mount the sample holder on the sample stage. Connect the sample through the BNC connector on the sample mount. Position the wiring such that no mounting part or wiring blocks the incident x-ray beam or any detector.
Ensure that the sample wiring is strain-relieved so that it will not restrict the sample movements. Check that the sample is well-grounded. Now rotate the stage such that the plane of interest is perpendicular to the incident beam.
This will minimize the beam footprint and maximize the spatial resolution. If you will be performing multi-modal measurements, place the detectors around the sample, for example for x-ray fluorescence measurements. Next, measure the signal amplitude of the test device to test the signal's range under different conditions.
Place a pre-amplifier in proximity of the sample and connect it to a control unit outside of the hutch. This will enable remote setting modifications without needing to re-enter the hutch and will automatically save the amplification settings. Connect the pre-amplifier to a clean power circuit and power it on.
Make sure that the signal amplitude of the test device matches the input range of the pre-amplifier. It is good practice to keep the amplification of the pre-amplifier at the minimum sensitivity whenever no measurement is going on to avoid accidental over saturation. Now connect the test device to the pre-amplifier.
Given the small signal amplitude, it is critical to keep the wiring short and at a distance from noise sources. Next split the pre-amplified signal into three parallel signal branches. These are used to separately record the positive and negative DC values, along with the modulated AC components.
Connect the lock-in amplifier to a control unit outside of the hutch. Power it from a clean power circuit. Make sure the output of the pre-amplifier matches the input of the lock-in amplifier under all conditions.
Here the maximum output of the pre-amplifier is 10 volt, but the maximum input range of the lock-in amplifier is 1.5 volt. Therefore, test the signal amplitude after the pre-amplifier and make sure that the input range of the lock-in amplifier is at its maximum. Next connect the output of the pre-amplifier to the input of the lock-in amplifier.
Mount the x-ray chopper onto a motorized stage with the ability to move in and out of the x-ray beam and power it via chopper controller. Connect the chopper to the control unit, in this case via the lock-in amplifier. Then drive the optical chopper with the demodulation frequency of the lock-in amplifier.
Next connect the output of the lock-in amplifier to a voltage-to-frequency converter. Then output the root-mean-squared amplitude R of the lock-in amplified signal as the analog AC signal of the device. Make sure that the device under test is shielded from all of the lights in the hutch.
Search the hutch. Please leave the area. Attention, please, notice switch on.
And turn on the X-ray beam. If everything is set up correctly, and the X-ray beam hits the sample, a modulated XBIC signal will be visible. Adapt the amplification of the pre-amplifier and the input range of the lock-in amplifier so that they match.
Make sure that the response of the pre-amplifier is fast enough for the chosen chopper frequency. A rectangular XBIC signal should be observed. If a strong delay is visible, the chopper frequency needs to be reduced or the filter rise time of the pre-amplifier needs to be adjusted.
Set the low pass filter frequency of the lock-in amplifier to the minimum that is compatible with the scanning speed. Then, maximize the amplified signal with respect to the ratio beam on and beam off and with respect to the signal to noise ratio. The setup is now ready for XBIC measurements.
Go to a pristine spot on the sample and start the measurement. The key advantage of using lock-in amplification for XBIC measurements is the dramatic increase of the signal to noise ratio as compared to measurements with standard amplification. Here, the pre-amplified device under test response is shown as measured by a scope without and with a bias light turned on.
Despite the presence of strong noise or deceit components induced by bias light or voltage, it is possible to extract the modulated X-ray beam induced current signal from the background signal, even if it is orders of magnitude smaller. Comparing these two images, note an offset signal on the order of eight millivolts that is shifted to minus 65 millivolts by turning on the bias light from fluorescent tubes. Furthermore, the signal variation on short time scales is significantly enhanced by the bias light.
With appropriate settings, both the offset and the high frequency modulation can be mitigated. Nevertheless, all sources of unintentional bias such as ambient lighting and electromagnetic noise should be eliminated for highest signal to noise ratio. These graphs highlight the effect of bias light and different low pass filter settings on the lock-in amplified RMS amplitude.
For high scanning frequency, the low pass filter cut off frequency should be as high as possible but highest signal to noises obtained with low cut off frequencies. In this case, a low pass filter with a cut off frequency equal to 10.27 Hertz offered the best compromise for scanning at moderate two Hertz. Here, you can see the impact of lock-in amplification on the signal to noise ratio in X-ray beam induced current measurements.
The noisiness of the direct signal is apparent and the lock-in amplified signal shows fine features in good detail. For quantitative analysis, the shape of the modulated XBIC signal should represent the shape of the modulated X-ray intensity. Therefore, it is important to optimize the chopper frequency and low pass filters with respect to that.
Lock-in amplification allow us to measure devices under different conditions. For example, we can apply bias voltage or bias light. Ultimately, this allow us to measure entire IV curve with high spatial resolution at the nanoscape.
XBIC is particularly useful when we combine it with other techniques. For example, with X-ray fluorescence diffraction, tachography, or X-ray excited optical luminescence. If we combine all of that, we can resolve and deconvolute the composition structure and performance.
Apart from general precautions to be taken when dealing with electrical power and intense X-rays there's no specific risk in performing XBIC measurements for the operate at least the sample, however, may die because of radiation damage. With diffraction limited sources, such as petra four, the nanofocus X-ray flux will increase by orders of magnitude. This will boost measurement speed signal to noise ratio and enable entirely new experiments in situ and operando.
