Dear Sandeep Kumar

These articles might be asset, have a look:

1. Chapter Impact of Temperature on Circuit Metrics of Various Full Adders

2. https://ieeexplore.ieee.org/document/6719819

3. Conference Paper Effect of temperature dependence on performance of Digital C...

Kind Regards

Qamar Ul Islam

The power dissipation in semiconductor devices always increases with temperature in the real devices. This happens because of the thermal runaway. The number of free electrons and holes increases with temperature (because of thermal generation), the current through the device increases and thus the dissipated power increases

Dear Dr. Sandeep,

This is an important question, whose answer depends on the specific category of circuits you are simulating as well as the method of implementation of such circuits. Therefore, you may find hundreds of articles for the temperature effect on each circuit category.

As a matter of fact, temperature affects the performance of both MOSFET and bipolar devices as well as their circuit-model (SPICE) parameters. These effects can be found in the datasheets of such devices and their hitkits (provided by vendors of EDA Tools). In addition, temperature do affect the value of passive elements in the circuit, specifically the resistor value R(T), whether they are discrete or integrated.

Generally speaking, there is no rule of thumb and each circuit should be standalone studied for the temperature effects. However, there exist many techniques to stabilize circuit performance against temperature variations. Among these, I dully note the band-gap reference sub-circuits, which are very important to stabilize voltage and current sources in many analog/mixed signal and pure digital applications.

Thank you Valentin Bogatu sir. But increase in temperature also decreases carrier mobility, hence results in decrease in current.

Thank you Muhammad El-Saba sir for adding your answer.

If power is decreasing with temperature then what could be the possible reasons?

Dear Sandeep Kumar ,

Adding to the answers of my colleagues above, if one assumes a constant power supply voltage and then vary the environmental temperature of the device, the if the circuit is built from FET transistors then their operating current will decrease by increasing temperature as the MOST temperature increases. Consequently the power dissipation will decrease up to certain temperature.

If the devices are bipolar transistors then their current will increase with temperature and their power dissipation will increase by the temperature.

Best wishes

Thank you @Abdelhalim abdelnaby Zekry sir.

In most of the literatures for radiation hardened latch designs, PVT section has stated that " Generally increase in temperature increases the power dissipation, since junction temperature is highly increases".

Dear Sandeep Kumar ,

Can you give the citation of the literature?

Best wishes

Sure sir, sent you a message.

Dera Sandeep Kumar ,

The above answers of the respectable researchers are all worth considering. You may assume they are saying different things about some points, however all are giving information based on real phenomena, but they may concentrate on some different aspects, because it is difficult to give a general answer to your question.

Let me explain:

Several parameters which have dominant effect on the current of a device may be incremental or decremental. On the other hand, these parameters may depend on temperature with a positive or negative temperature coefficient. Besides, many of these parameter and temperature effects are usually nonlinear. Furthermore, depending on the types of the devices (e.g. FET or BJT), the way the devices are operated in the circuit (e.g. their operating region(s) and the technique use to bias or switch them) and the types of the circuit blocks used in the circuit, the total current drawn from the power supply may have a different nature and a different dependency on temperature.

Nevertheless, if it is possible to specify the conditions special to your case (e.g. a circuit employing CMOS latches of a specific design only), then a specific answer can be given.

In view of above comments, I can say it is possible to have power dissipation increasing for some designs and decreasing for some other different designs. Besides, if you shift the temperature range to some other range, it may also be possible for some circuits to invert their temperature dependency (regarding power dissipation) because of nonlinear behavior.

Best regards...

Sir Sandeep:

You could benefit from this article about your topic:

"Analysis of the Effect of Temperature and Vdd on Leakage Current in Conventional 6T-SRAM Bit-Cell at 90nm and 65nm Technology"

By: Neeraj Kr.Shukla, Shilpi Birla, Kapil Rathi, R.K.Singh, Manisha Pattanaik.

I have attached the pdf file ...

With my best wishes ...

Dear Sandeep:

In addition to the opinions of my colleagues, I would like to add the following, which I hope it will be helpful:

--+++++++------+++++++

Typical transistor parameters affected strongly by temperature are turn on voltage (Vbe-on for bipolar, Vt for MOS), turn-off leakage current, current gain.

For bipolar transistors with increasing temperature:

A- Vbe decreases

B- Turn-off leakage increases (with a constant voltage below Vbe-on; this is related to transconductance below)

c- Current gain ẞ increases.

For MOS transistors with increasing temperature:

A- Vt decrease.

B- Turn-off leakage increases (with a constant voltage below Vt; also related to transconductance below).

C- Idss decreases.

These parameters affect how one may design circuit with a wide temperature range. These variation in parameters can typically be mitigated with proper circuit design.

However there are couple of important parameters that basically limit the ultimate operating speed and operating frequency of the transistor; and there is nothing we can do about it. These parameters are not often talked about.

A- Parasitic resistances (base, collector and emitter) for bipolar and channel resistance (related to Idss above) for MOS increase with increase in temperature.

B- For both transistor types, transconductance (dIout/dVin) decreases with increasing temperature.

++++++---------- Also----------++++++++

we care about these parameters about a circuit: Noise, Speed, Power Consumption, Reliability.

All these parameters degrade with increase in temperature.

A-- Noise: With increase in temperature thermal energy (kT, k-Boltzmann Constant, T-Temperature in Kelvin) of the charge carriers (electrons) increases. This increase in thermal energy will lead to more collisions between the electrons and increases noise.

B-- Power Consumption:

To achieve same performance from a circuit at higher temperature, it consumes more power compared to lower temperature.

C-- Leakage currents (current drawn by off switches from supply) increases exponentially with temperature. So power consumption of an idle circuit increases with temperature.

D-- Speed: speed of a circuit reduces with increase in temperature.

E-- Reliability: Circuit reliability degrades if its operated at higher temperatures.

------++++++++ Also++++++------+++

Temperature and germanium diode:

A germanium diode responds in the same way any other diode responds. As you increase the temperature, the intrinsic carrier concentration increases. This pushes the fermi level closer to the intrinsic fermi level (the middle of the band gap). Since the built-in potential of a diode is determined by the difference in fermi-levels in the p-type and n-type regions, the fermi level in each region moves closer to the middle of the gap, and the built-on potential is decreased.

This is also a major contributor to why the performance of solar cells decreases at high temperatures. The operating voltage of a solar cell is related to the diode's built-in potential. As a solar cell gets hot, the voltage is reduced, and therefore the power output is reduced.

Another effect that can be considered: as the temperature of any semiconductor is increased, its band gap is reduced. This means that a solar cell will produce slightly more current. However, this is not enough to compensate for the decrease in voltage.

None of this is specific to germanium diodes. The voltage reduction, and band gap reduction will happen in all semiconductor diodes. The only difference is that germanium has a relatively small band gap, compared to Si and GaAs. This means that the intrinsic carrier concentration is much more rapidly affected by the temperature increase, and therefore there will be a more rapid decrease in diode voltage

++++++-----

Indeed you could benefit from the lecture, which i have attached: