Thanks David, definitely learned something from your comments, eg. the Aol vs Supply.
Most of this is from the bias current generator inside the 741 producing a proportionally lower current at lower supply voltages. I used the 301A as an example where the bias current generator operates as a constant current source making its characteristics more independent of supply voltage and that is common with later designs. Additional slowdown comes from increased junction capacitances at lower voltages but there is not anything to be done about that other than using a different part.
Had no idea 741 would do anything at such low voltages. And looking at internal bias
I see max of 2 Vbes + some amount of Vcb to keep devices active region worst case turnoff........
Exactly, adding up the Vbe drops and Vce drops to stay out of saturation gives a good idea of the input and output voltage limits. Conveniently for the 741, the low bias currents produced by low supply voltages also lower the saturation voltages.
As danadak pointed out, there are much better parts for low voltage operation but the 741 will certainly work below 9 volts. In the past, a lot of applications used the 741 or 301A with a -3 to -5 volt negative supply and +12 to +24 or even higher positive supply for easy operation down to ground. For 9 volt battery applications with a rail splitter to make +/-4.5 volts, an operational amplifier other than the 741 or 301A would often be used for lower current operation to preserve battery life. An example of a part like this from the time of the 741 is the 776 which might be considered an improved 741 specified to operate down to +/-1.2 volts with a programmable supply current. Whereas a 741 would draw 500uA at +/-5 volts, a 776 could be configured to draw 1uA or less at any supply voltage. A modern part like that would be something like an LT1006 which also happens to be a single supply operational amplifier.