TCT -Transient Current Technique

As the sensor is traversed by a charged particle, it creates/ionizes free charge carriers, electron-hole
(e-h) pairs, which separate under the influence of an externally applied electric field and travel
towards their respective electrodes (see Figure 1). Each charge carrier induces current in the pad
electrodes (pad size >> thickness):

𝐼 = 𝑒0 𝑣/𝐷 = 𝑒0 𝜇𝐸/𝐷, (1)

where e0 –elementary charge, v – velocity of electrons or holes, µ - mobility of charge carriers, E –

electric field, D – sensor thickness. This equation shows that it is possible to determine different sensor
properties from the shape of the induced current: e.g. the mobility of charge carriers and the electric
field profile. Minimum ionizing particles are ionizing free charge carriers along its whole path and
therefore it is difficult to separate the contributions of electrons and holes to the induced current.
Since silicon has a very narrow energy gap of merely 1.12 eV, it is possible to generate free charge
carriers also using light. The penetration depth is wavelength dependent and it’s only a few μm for
red light (660 nm, 1.9 eV). If the detector surface is illuminated, charge is generated near the
electrode. Therefore, charge carriers of one type end their drift at the nearby electrode, while the
other travel through the whole detector thickness and induce current (Figure 1).

Figure 1: Schematic representation of e-h pair generation when observing induced signals from electrons (a) and
holes (b). Electrons are travelling from the high field region towards the small field region, while for holes it is
the other way around.
By observing the induced current, it is possible to determine the main detector properties such as
effective space charge (hence electric field), full depletion voltage and also material properties such
as e.g. mobility of charge carriers, saturation velocity or minority carrier lifetime in un-depleted bulk.

As follows from eq.1, the current is proportional to the velocity of the charge carriers and thereby the
electric field. Hence, the increase in current with time signifies that the charge is moving in the
increasing electric field and vice-versa. The steepness of the rise/decrease depends on the field
magnitude. In very high fields, the velocity saturates and there is virtually no change in the induced
current (signal). At bias voltages higher than the full depletion voltage, the signals get narrower. The
passage of generated carriers from one electrode to the other depends on the illumination side and
the difference can be clearly observed. If the full depletion voltage of the detector is small (very low
Neff, highly resistive silicon), an almost homogenous electric field can be obtained already at
moderate voltages. With the detector thickness known, as well as the travel time of generated charge
carriers, it is possible to calculate the velocity of these charge carriers and by that also their mobility
as a function of the electric field strength.

Tasks:
1) Observing the induced current signals in the detector
a. How should the detector be connected in order to be reversely biased? How can you
make sure if you connected it properly?
b. Explain the influence/contribution of both electrons and holes to the collected charge
(current integral in time). Explain its dependence on the bias voltage.
2) Determine the dependence of electrons and holes mobility on the electric field
a. Which diode is appropriate for determining the mobility?
b. What is approximate saturation velocity of charge carriers in silicon?
3) Determine the dependence of the saturation velocity of electrons and holes on the
temperature (if time permits).

Apparatus:
Pad detectors will be used shown in Figure 2. These detectors are appropriate for studying silicon
properties, as they are small and only have two electrodes. On the top side they have an opening in
the metallization layer which allows laser illumination. The whole bottom side is for this purpose
metalized in form of a net. The active part of the detector is 5x5 mm2 in size.

In this exercise, the sample is mounted in a way so that it is possible to be cooled down using the
Peltier element.
Figure 2: Schematic view of a diode used in the study. The guard ring (Al) surrounds the electrode at the top
side. Inside the electrode there is an opening that allows for charge generation using light.

The measurement system for determining the induced current signals

The schematic view of the setup/measurement system is shown in Figure 3, while the photo of the
whole setup is shown in Figure 4. The measurement system is composed of: a laser, amplifier, Bias-T
(decoupling circuit) and HV filter. The photo also shows how these components are connected
between each other. Laser positioning, its frequency and pulse width is performed manually.

Figure 3: Schematic view of the Transition Current Technique (TCT) measuring system. The laser pulse generates
charge carriers at the detector surface, which then travel under the influence of the electric field inside the
detector and induce current in the readout electrodes. This current signal is amplified and shown on the
oscilloscope. The Bias-T circuit is used for decoupling high voltage from the amplifier input.

Laser and

Amplifier PS

HV Bias, Fan bias,

Peltier cooling bias

Termometer

DRS
HV filter
oscilloscope
 FAN

sample

LASER

Figure 4: A photo of a TCT measuring system with appropriate components labeled.

Literature

www-f9.ijs.si/~gregor/papers/dok_eng.pdf
http://www.ioffe.ru/SVA/NSM/Semicond/Si/electric.html

Data analysis

Baseline correction

The signal before the laser beam/illumination (a ns before the pulse) is used for determining the
oscilloscope baseline/pedestal (zero) level. This DC level needs to be determined and subtracted for
every TCT analysis, in particular when determining the collected charge (current integral in a specified
time frame).

Induced charge

When using particle detectors, usually the integral of induced current is measured. The time frame
used for integrating the current is usually chosen to be about a ns before the observed increase of the
induced current and a few ns after it’s gone. Since the number of photons in the laser beam is
unknown, it is not possible to determine how much charge is generated inside the detector. This can
be roughly checked if an α-radioactive particle source is used instead of laser light. If 241Am is used,
emitted particles have energy of 5.8 MeV. Their energy losses are around 1 MeV/cm in air. Therefore,
in this setup, the α-particles approach the Si detector with around 0.5 MeV less (since the distance
between the radioactive source and the detector is around 5 mm) and then they lose some energy in
the detector metallization and SiO2 layer, which covers the silicon. Finally, around 4.8 MeV of energy
is released in Si, which corresponds to around 1.3 – 1.4 · 106 pairs of electron-holes. In this way, the
scale may be calibrated in units of elementary charge.

Measurements with LGAD (“Low Gain Avalanche Detectors”)

If a thin layer of highly doped p-type semiconductor is inserted between n and p layers of the n+-p
detector, an n++-p+-p (+ indicates high doping in the order of 1016 cm-3, while ++ stands for extremely
high doping in the order of 1019 cm-3) structure is obtained. Figure 5 gives a schematic representation
(not to scale) of such processed detector, with its expected electric field profile shown on the left.

Figure 5: Schematic representation of an LGAD detector (not to scale; the n++-p+ layers are typically
ca. 1-3 µm thick); established electric field profile shown on the left.

Inside this thin layer, the electric field may reach very high magnitudes, causing electrons to gain
sufficient energy (Eg = 1.12 eV) within their free path and multiply (charge multiplication), inducing
secondary ionization/e-h pair generation. The process is similar to the one encountered in gas
proportional detectors (GPDs). The required electric fields for the onset of this multiplication process
are of the order of 15-20 V/µm. These detectors are able to achieve better Signal-to-noise ratio and
hence detection efficiency, spatial resolution and time resolution as well, because the signal is big and
fast enough for application even in very thin detectors (ca. 50 µm).

The induced signal shape in such a detector is different from standard detectors, because at high
enough multiplication most of the signal comes from the long drift of holes created in the
multiplication layer, traveling towards the back plane of the detector. A typical pulse resulting from
back side illumination of a fully depleted detector (this is how the drift of electrons towards the upper
electrode is observed in n+-p detectors) is shown in Figure 6.
Figure 6: A typical pulse from an n+-p detector, resulting from back side illumination.

Five different stages of signal development can be observed: 1) drift of electrons, traveling from the
back plane of the detector (where they were created by laser illumination) towards the front, 2)
multiplication of electrons, 3) end of multiplication, 4) drift of holes towards the back plane of the
detector, 5) end of drift.

The magnitude of multiplication can be easily estimated from the shape of the pulse, if the integral of
induced current is divided into two parts: a part that comes from the drift of electrons Qe (current
integral between 0 and 9 ns in Figure 6) and a part which comes from the drift of holes Qh (9 – 32 ns).
Charge multiplication factor is then simply calculated as:
𝑄𝑒 + 𝑄ℎ 𝑄ℎ
𝐺= =1+
𝑄𝑒 𝑄𝑒
Tasks:
1) Determine the multiplication factor for the device under test (DUT) depending on the applied
bias voltage.
2) What is the full depletion voltage of the DUT?