In the above image considering V2 as a constant source of 5 V and V1 as the input which can be taken to be 0 V or 5 V, we get the voltage difference (vout2-vout1) as 5 V and 0 V respectively which works like NOT gate. So, I have two questions:
Logic functions should be cascade-able so that one's output can feed another's input in a logical string. The OP's proposed NOT construction fails to be cascade-able because reference point for input differ from the reference point for output.
However, such a construction is possible at the beginning or end of a cascade, where no other logic need be driven. At the end of a logic cascade, we have freedom to choose the reference point. For cascaded logic strings, the reference point is always the same (most often GND).
An example:
Suppose we wish to drive an indicator like a LED. We have a choice to reference the LED to logic low or to logic high...that is, to GND or to Vdd. The choice of reference point allows us to invert the LED's indicator function:
simulate this circuit – Schematic created using CircuitLab
This LED example works because light is only produced when current flows - with zero voltage or reverse voltage you get no light. I would suggest that the freedom to choose its function is a result of the freedom of choosing its reference point. The LED diode itself cannot do the "invert" function if you do not have the freedom of choosing its reference point.
It isn't a NOT gate.
Your circuit can produce an inverting logic function. So it has the 'NOT' part - but not the 'logic gate' part. It's not a logic gate.
A logic gate includes the following properties:
Is a standard block than can be cascaded with identical copies.
A current gain of at least 2 (typically much higher). Its output must be able to drive multiple circuits of its own type.
Your circuit fails on both these.
To your questions...
(1) Why do we not consider this circuit and instead use transistors in practical NOT gate ICs?
We use a transistor circuit to get:
(2) Why can we not make NOT gates using diodes and resistors just like we do in OR and AND gates using Diode-Resistor Logic?
OR and AND gates don't invert the input signal's polarity. NOT gates do only that. Diode-resistor logic can't do inversions, so can't make NOT, or NAND, NOR, XOR gates either.
You can just switch to negative logic and then you don't have to do anything, the inverter is not required (effectively what you are doing, you can't actually invert without an active element like a transistor). And at least an inverter and a gate or an inverting gate is required to create arbitrary logic blocks by De Morgan's laws.
Practical general-purpose logic elements are more complex than diodes and resistors because we want to be able to chain them and have reasonable (like 10) fanout (so one gate output can drive many gate inputs). The inputs should have a consistent definition (eg. '1' is at > 2.4V and '0' is less than 0.8V wrt ground). And the outputs should be able to drive 'many' inputs and provide those voltages plus a reasonable noise margin.
Simplified textbook examples often don't meet that requirement, so they're useful for learning purposes but only practically useful in very limited circumstances, not as general-purpose logic building blocks.
Here is an example where switching to negative logic is sensible- we want an AND-OR logic block, but we have only NAND gates:
simulate this circuit – Schematic created using CircuitLab
The outputs of gates 1, 2, 3 are inverted compared to an AND gate, but feeding inverted inputs into a NAND gate gives a non-inverted OR output.
For this to work, you need the next gate to have an input which is measuring the difference between two voltages.
It has to return true if vout1-vout2 is not zero.
Fortunately one of the voltages, vout2, is tidied to the positive rail.
So it has to return true if vout1 is low and false if vout1 is high.
There is a logic gate that does this; it is called a NOT gate.
So you still need a real NOT gate to convert its output into something that can drive any downstream logic, so it can hardly be called a NOT gate itself.
Your question is very interesting. It shows a tendency towards deep thought rather than the mechanical application of common textbook clichés.
It is indeed a NOT gate if the output voltage (the voltage drop across resistor R1) is measured with respect to the positive rail of V1. In fact, this voltage is the complement of V1 to V2, which in this digital circuit represents its negation. Let's draw the circuit in its conventional form and analyze its behavior for both possible input voltage values (0 V and 5 V). Your R1 resistor is represented by a real voltmeter with 1 kΩ internal resistance acting as a load; hence the odd name "RL=1kΩ".
Vin = 0 V ("0"), Vout = 5 V ("1")
simulate this circuit – Schematic created using CircuitLab
Note that since referenced to Vcc the output voltage across the load RL is negative.
Vin = 5 V ("1"), Vout = 0 V ("0")
Analogously, in an analog circuit like a common-emitter, the voltage across the collector resistor (in orange) is the complement of the collector voltage. We prefer to use the collector voltage since it is referenced to ground.
Why do we not consider this circuit and instead use transistors in practical NOT gate ICs?
However, in circuits, we prefer to refer signals to ground because this simplifies the connection between stages and the measurement of voltages. So, we have to feed the signal back to ground (e.g., with a PNP transistor) by "folding" the circuit.
Vin = 0 V ("0"), Vout = 5 V ("1")
Vin = 5 V ("1"), Vout = 0 V ("0")
For this purpose, voltage, rather than current, is used as the carrier of information in logic gates as well as in most analog circuits. This allows us to connect input-output ports in parallel (cascade) with respect to ground.
Conceptual circuit. Thus, logic gates deal with voltages, and the device with which voltages are controlled in the most general case is a voltage divider. It consists of a "pull-up" element (e.g., resistor R1, switch SW1, PNP transistor Q1, etc.) and a "pull-down" element (e.g., grounded resistor R2, switch SW2, NPN transistor Q2, etc.). The most general property of these elements is that they have resistance (no matter linear or non-linear). In the simplest case, they are represented as voltage-controlled switches with infinite resistance when open and zero resistance when closed. Let's explore this conceptual circuit by CircuitLab.
Vin = 5 V ("1"), R2 = 0, Vout = 0 V ("0")
Vin = 0 V ("0"), R2 = 100 kΩ, Vout = 5 V ("1")
BJT implementation: Let's now explore the classic BJT NOT gate implementation.
Vin = 5 V ("1"), Q on, Vout = 0 V ("0")
Vin = 0 V ("0"), Q off, Vout = 0 V ("1")
In order for the input and output voltages to be relative to ground, it is simplest to control the resistance of the lower element. If the dependence between the input voltage and the resistance is inverse (aka "enhancement type"), with an increase in voltage, the resistance will decrease and the voltage drop across it (the output voltage) will also decrease. As a result, the circuit will invert, i.e., it is a NOT gate.
Why can we not make NOT gates using diodes and resistors...
In principle, diodes are switches but 1-port; they have only two terminals that are both input and output (unlike transistors which are 2-port, with separated input and output). While conventional 2-port switches are connected in series with a constant voltage source (Vcc), diode switches are connected in series with the varying input voltage source. This does not allow a NOT gate to be made with them. Let's explore it by CircuitLab to see why.
Forward-connected diode: If the input voltage source is connected to the anode...
... and Vin = 0 V, the switch is off and Vout = 0 V.
If Vin = 5 V, the switch is on and Vout = 5 V. So, this diode circuit is not an inverter (NOT gate) but a follower.
Backward-connected diode: Then let's try to connect the input voltage source to the cathode ("pulling up" the anode by a resistor R to 5 V).
Now, if Vin = 0 V, the switch is on and Vout = 0 V. Note that the current enters the input voltage source (basic property of DTL and TTL gates).
If Vin = 5 V, the switch is off and Vout = 5 V. Again, this diode circuit is not an inverter (NOT gate) but a follower.
Thus, in both diode configurations, the diode switch directly connects the input source to the load, resulting in a non-inverted output voltage.
... just like we do in OR and AND gates using Diode-Resistor Logic?
However, these diode circuits are useful when there are multiple input sources (in OR and AND logic gates) because they isolate them. More specifically, in the first (OR) circuit, the diode allows the input source to connect to the load only when there is a positive input voltage, while in the second (AND) circuit, it allows the connection only when there is zero input voltage. Figuratively speaking, the diodes have made the two sources unidirectional - in the first case, the source can only provide (source) current, and in the second case, it can only receive (sink) current. This prevents conflicts between the input sources when they have different voltages.
