Exploring the Relationship Between Color and Temperature in Thermal Measurements

[Beanyboy]

There's a passage in Kumar's (excellent) book, "Quantum", which has me confused.

He notes that Planck is aware that as the temperature of a heated poker rises, the colours change from red through to bluish white. Later in the passage, he refers to Herschel's earlier work on the relationship between heat and light, noting that "... as moved he moved the thermometer across the bands of different colours from violet to red, the temperature rose (my italics).

The Planck observation suggest "red" is associated with lower temperatures, while Herschel's work says the "red" is associated with higher temperatures. Can you help me please?

 

[davenn]

The Planck observation suggest "red" is associated with lower temperatures, while Herschel's work says the "red" is associated with higher temperatures. Can you help me please?

well in general every day observations red is lower white is much higher

take a piece of hot metal …. as it warms up, it starts glowing red before going through orange … yellow etc on it's way to blue/white
it's the same for telling which stars in the sky are hotter or cooler

I have never read any of Herschel's work on this, so I don't know if Herschel is in error or if he has been misquoted ??

Dave

 

[gleem]

See this link for Herschel's experiment.

http://practicalphysics.org/william-herschel-and-discovery-infra-red-radiation.html

 

[sophiecentaur]

while Herschel's work says the "red" is associated with higher temperatures.

Is it really worth bothering with what was probably a misunderstanding that occurred over a century ago?
Looking at the spectrum of light (and invisible em, too) that's associated with the temperature of a black body, shows that the peak spectral component, which dominates our perception of colour, correlates with the temperature. Google Colour Temperature and black body spectrum to get hundreds of hits and images.

as moved he moved the thermometer across the bands of different colours from violet to red, the temperature

I just re-read this. The experiment was 'the other way round' to an experiment that relates the temperature of a hot object and its perceived colour. The experiment involved producing a spectrum of the incident light with the different wavelengths of light going in different directions and putting a thermometer in different portions of the spectrum. The reason that the temperature of his thermal detector rose in the Red region would have been that there was more power in that region. There is less power at the blue end of the spectrum of the sunlight that actually arrives on the Earth's surface, compared with that in the a red region AND the spectrum slopes off faster in the blue region. Also, his prism may have contributed to the effect by absorbing some more of the blue end than the red and the size of his thermometer may have meant that it overlapped into some of the UV region, which is even more absorbed by glass. This would have heated the thermometer more in the red region. So his result was not surprising and not controversial. A lesson on the need to understand exactly what an experiment is expected to deliver.

 

[Beanyboy]

Also, his prism may have contributed to the effect by absorbing some more of the blue end than the red and the size of his thermometer may have meant that it overlapped into some of the UV region, which is even more absorbed by glass. This would have heated the thermometer more in the red region. So his result was not surprising and not controversial. A lesson on the need to understand exactly what an experiment is expected to deliver.

So, are you saying that we could expect different results if we used a laser thermometer and a contemporary prism?

 

[sophiecentaur]

So, are you saying that we could expect different results if we used a laser thermometer and a contemporary prism?

When Hershel did his measurements, he only had thermal methods for measuring Energy flux. I wouldn't plan to do it his way (nor would anyone else, It I think. I would use a solid state light sensor or, a few years ago, a photomultiplier. The 'temperature' Herschel was measuring just used the heating effect on a small mass from the spectral components hitting it.
A thermometer of any type would not be as sensitive or accurate. as an electronic photon detector.
I see you have edited your last reference to temperature of colours. Yes the only 'temperature' of a colour is the heating effect of a narrow range of spectral components.
But your original apparent paradoxical result is due to the fact that he was not looking at the relationship between apparent colour and temperature. I think we have dealt with that. The spectrum of light emitted by the sun or any other hot body is not flat but has a peak which can be in the far Infrared, in the optical region or even in the UV region. Sunlight peaks around green wavelengths and Herschel would have found that.
Note, as far as I know, there are no stars that look 'green'. They vary from 'reddish', through 'Orangeish' to 'whitish' and bluish/ white. That is because of the way we interpret the spectra of light and name them as 'colours'.

 

[Beanyboy]

When Hershel did his measurements, he only had thermal methods for measuring Energy flux. I wouldn't plan to do it his way (nor would anyone else, It I think. I would use a solid state light sensor or, a few years ago, a photomultiplier. The 'temperature' Herschel was measuring just used the heating effect on a small mass from the spectral components hitting it.
A thermometer of any type would not be as sensitive or accurate. as an electronic photon detector.
I see you have edited your last reference to temperature of colours. Yes the only 'temperature' of a colour is the heating effect of a narrow range of spectral components.
But your original apparent paradoxical result is due to the fact that he was not looking at the relationship between apparent colour and temperature. I think we have dealt with that. The spectrum of light emitted by the sun or any other hot body is not flat but has a peak which can be in the far Infrared, in the optical region or even in the UV region. Sunlight peaks around green wavelengths and Herschel would have found that.

I'm much obliged to you for helping me out here. It'll come as no surprise to you that I'm struggling to keep up with the technicalities here. You've given me much to think about and research. Thanks again.