Cosmic Microwave Background prediction

[iantresman]

I'm trying to clarify the following

1. When the Cosmic Microwave Background (CMB) was first predicted
2. Whether this was in connection with the Big Bang
3. How this differs from earlier predictions of the (non-CMB) background
   temperature.
4. When the term Cosmic Microwave Background (CMB) was first used in print

Wikipedia and Edward Wright suggest that the CMB was predicted by George Gamow, Ralph Alpher, and Robert Hermann in the 1940s. But I can't find any indication that they (a) used the terms blackbody or CMB, and (b) they were discussing the Big Bang.

Apparently Guillaume suggested a temperature for space in 1896 of about 5-6K, and Arthur Eddington of 3.18K in 1926 (See History of the 2.7 K Temperature Prior to Penzias and Wilson, PDF). But Wright disputes Eddington's estimate [ref], because he did not mention CMB. But did Gamow, Alpher Hermann in the 1940s?

Of course Eddington did not mention CMB because the concept wasn't around. Anyone know when CMB was first used in print?

Regards,
Ian Tresman

[Tim Thompson]

I do not know when the phrase "cosmic microwave background" first appeared. However, I will refer you to 2 exceptionally informative books, which include considerable discussion of the history of big bang cosmology.

• Genesis of the Big Bang, Ralph A. Alpher & Robert Herman, Oxford University Press, 2001
• Cosmology - The Science of the Universe, Edward Harrison, Cambridge University Press, 2000 (2nd edition, much better than the 1981 1st edition)

I will opine that it is a virtual certainty that such a specific phrase was in connection with big bang cosmology. There is little doubt that Alpher, Gamow & Herman were the principal technical architects of big bang cosmology, but the idea is usually credited to the Belgian Jesuit priest Georges Lemaitre and his "Primeval Atom" (although the idea might be traced as far back as 1922 and the Russian mathematician Alexander Friedmann).

As for Guillame, I've read that all a long time ago, and he does not know what he is talking about. Eddington never calculated a background temperature, he calculated an effective temperature, and was very explicit about that. The problem is that too many people don't appreciate the difference between the Stefan-Boltzmann law & the Planck Law.

Eddington calculated the total energy from stellar emission, and assigned that energy an arbitrary equivalent temperature by sticking it into the Stefan-Boltzmann equation. But all that tells you is that a true blackbody at the given temperature would emit that total amount of energy, without telling how the energy is spectrally distributed. Eddington knew this, and says in his book (The Internal Constitution of the Stars, 1926 & 1930) that the spectral distribution of the energy would not be in acordance with the Planck Law.

The Planck Law, when integrated over all frequencies, returns the Stefan-Boltzmann law. If the energy in question is a blackbody, it must have a spectral Planck shape, and no other. One of the fundamentals of big bang cosmology is that the background radiation must have a Planck Law spectrum, it must be blackbody. Any other spectral shape would seriously damage big bang cosmology, if not simply falsify it at once. Measuring that spectral shape was the primary goal of the COBE mission. The spectral shape was measured by COBE's FIRAS instrument and found to have a Planck shape, as required by big bang cosmology.

So far as I know, nobody, Eddington included, predicted a thermal (i.e., Planck Law) background radiation until it was predicted in theory by early big bang cosmologists, probably by Gamow specifically. But they could not make much of a "precise" prediction, as they knew virtually nothing about the specifics of the early universe (we know more now, relatively speaking, about what the early universe should look like in a big bang scenario, but doubtless still "don't know" more than we "do know"). So the early predictions were, in a manner of speaking, all over the map, varying by an order of magntiude or more.

[Zahl]

But Wright disputes Eddington's estimate [ref], because he did not mention CMB. But did Gamow, Alpher Hermann in the 1940s?

Wright doesn't dispute it because the CMB is not mentioned, but because the spectrum is wrong (not Planckian, Eddington fully acknowledges this in the quote) and the frequency range is wrong. Starlight is not very isotropic either. There are many temperature components of "space" - the well known starlight component, the interstellar dust component, the cosmic X-Ray background (CXB), the CMB and there are others. The greatest thing about the CMB prediction was not the name used or the temperature, but that a previously unknown isotropic temperature component with a Planck spectrum was predicted to exist in the first place in a frequency range that was supposedly pitch black.

The first use of the "Cosmic Microwave Background" that I am aware of is "Peculiar Velocity of the Sun and its relation to the Cosmic Microwave Background" by Stewart and Sciama, Nature (1967). There is some talk that Dicke's team at Princeton used it even earlier. Some articles don't mention "microwave" because they are talking about the primordial spectrum that was not redshifted there yet.

Below is the short Nature article by Alpher & Herman from 1948 where they correct Gamow's earlier temperature prediction and first give their 5 K figure. It is obvious from the text that they are talking about in the context of an expanding universe that has cooled down from a hotter primordial state. I don't have the long 1949 Physical Review paper where they predict the Planck spectrum. Maybe Tim can quote the relevant parts if he has it?

---
Evolution of the Universe

In checking the results Presented by Gamow in his recent article on "The
E-volution of the Universe" [Nature, of October 30, p. 680], we found that his
Expression for matter-density suffers from the following errors : (1) an error
of not taking into account the magnetic moments in Eq. (7) for the capture
cross-section, (2) an error in estimating the value of \alpha by integrating the
equations for deuteron formation (the use of an electronic analogue computer
leads to \alpha = 1), and (3) an arithmetical error in evaluating \pho-sub-zero
from Eq. (9). In addition, the coefficient in Eq . (3) is 1.52 rather than
2.14. Correcting for these errors, we find

\rho-sub-mat = 4.83E-4/t^(3/2)

The condensation-mass obtained from this corrected density comes
out not much different from Ganow's original estimate. However,
the intersection point \rho-mat=\pho-rad occurs at t = 8.6^17 sec.
approx 3^10 years (that is, about ten times the present age of the universe).
This indicates that, in finding the intersection, one should not neglect the
curvature term in the general equation of the expanding universe.
In other words, the formation of condensations must have taken place
when the expansion was becoming linear -with time.
Accordingly, we have integrated analytically the exact expression:

dl/dt = [....]

T approx 1/l and R-sub-zero= 1-9 X 10i light-years [Note they are doing
analytic integration so the i=square root of negative 1 is a mathematical
convenience] .
The integrated values of \rho-mat and \pho-rad intersect at a reasonable time,

namely, 3.5 x 10^14 sec approx 1^7 years, and the masses and radii of
condensations at this time become, according to the Jeans' criterion,
M-sub-c = 3.8 x 10^7, sun masses, and R-sub-c = 1. I X 10^3 light years.
The temperature of the gas at the time of condensation was 600 K., and
the temperature in the universe at the present time is found to be about 5 K.
We hope to publish the details of these calculations in the near future.
Our thanks are due to Dr. G. Gamow for the proposal of the topic and his
constant encouragement during the process of error-hunting. We wish also
to thank Dr. J. W. Follin for his kindness in performing the integrations
required for the determination of \alpha on a Reeves Analogue Computer.
The work described in this letter was supported by the United States Navy,
Bureau of Ordnance, under Contract NOrd-7386.

RALPH A. ALPHER
ROBERT HERMAN

Applied Physics Laboratory,
Johns Hopkins University,
Silver Spring, Maryland
Oct. 25.

1 Gamow, G., Phys. Rev., 70, 572 (1946).

[iantresman]

Originally posted by Zahl@Jun 17 2005, 12:04 AM
Below is the short Nature article by Alpher & Herman from 1948 where they correct Gamow's earlier temperature prediction and first give their 5 K figure. It is obvious from the text that they are talking about in the context of an expanding universe that has cooled down from a hotter primordial state.

But they don't mention that this 5K temperature is black body, microwave, or planckian, so couldn't they have been talking about an effective temperature?

Regards,
Ian Tresman

[iantresman]

Originally posted by Tim Thompson@Jun 11 2005, 12:44 AM
Eddington calculated the total energy from stellar emission, and assigned that energy an arbitrary equivalent temperature by sticking it into the Stefan-Boltzmann equation. But all that tells you is that a true blackbody at the given temperature would emit that total amount of energy, without telling how the energy is spectrally distributed. Eddington knew this, and says in his book (The Internal Constitution of the Stars, 1926 & 1930) that the spectral distribution of the energy would not be in acordance with the Planck Law.

It seems to me that there are several characteristics of the CMB which are part of the Big Bang prediction:

1. The temperature
2. That it is blackbody
3. That it has a Planckian spectrum shape according to Planck's Law (1901)
4. The temperature has a frequency in the microwave range

It also seems to me that some people did predict some of these characteristics long before Gamow, et al. So although Eddington suggested an effective temperature of 3.18K (black body), he was wrong on it corresponding to Planck's law, and never mentioned the frequency.

Would it also be fair to say that anyone who estimates temperature without mentioning the other three characteristics is also "incorrect"?

Are there any other fundamental charactistics that should be included in this list?

Regards,
Ian Tresman

[Tim Thompson]

Originally posted by iantresman@Jun 17 2005, 06:43 PM
But they don't mention that this 5K temperature is black body, microwave, or planckian, so couldn't they have been talking about an effective temperature?

Not a chance. This is a professional paper, you know, in which case the word "temperature" actually means "temperature", unless explicitly modified to the contrary. And the word "temperature", like most scientific terms, has a precise, but context dependent definition. In the context of radiation, it specifically means "Planckian".

So the use of the unmodified word "temperature" automatically implies a Planck Law spectrum. "Planckian" and "black body" are synonomous, so the word "temperature" automatically implies both "Planckian" and "black body".

By fixing the temperature (in this case 5 Kelvins), they also automatically fix the frequency, because they have fixed the spectral shape (Planck Law). So explicitly saying the word "microwave" would be a redundant waste of space. They already said "microwave" when they said "5 Kelvins".

[Tim Thompson]

Originally posted by iantresman@Jun 17 2005, 07:46 PM
It seems to me that there are several characteristics of the CMB which are part of the Big Bang prediction:

1. The temperature
   
2. That it is blackbody
   
3. That it has a Planckian spectrum shape according to Planck's Law (1901)
   
4. The temperature has a frequency in the microwave range
   

It also seems to me that some people did predict some of these characteristics long before Gamow, et al. So although Eddington suggested an effective temperature of 3.18K (black body), he was wrong on it corresponding to Planck's law, and never mentioned the frequency.

Would it also be fair to say that anyone who estimates temperature without mentioning the other three characteristics is also "incorrect"?

Are there any other fundamental charactistics that should be included in this list?

Regards,
Ian Tresman

On your list, 2 & 3 are synonymous; "blackbody" and "Planckian" mean exactly the same thing, so they only count as 1. Furthermore, 1 & 4 are not independent; if you fix the temperature, you automatically fix the frequency, and vice versa. In fact, I would say that your list of 4 really boils down to a list of 1: "Planckian", or in the more commonly used jargon, "thermal". Once a radiation field is described as "thermal", then that word includes within it both "Planckian" and "blackbody". Furthermore, once a number is associated with it (the temperature), then the frequency range is also automatically determined. So, if the temperature is specified (i.e., 5 Kelvins), then "blackbody", "Planckian" and "microwave" are automatically determined (a different numerical value for the temperature would fix a different frequency range).

In light of that, ...

Would it also be fair to say that anyone who estimates temperature without mentioning the other three characteristics is also "incorrect"?

No, simply because once the word "temperature" is properly used, then all of the other characteristics are instantly defined & determined, so mentioning them, without a specific reason to do so, is a wase of time. But do keep in mind that, prior to the determination of Planck's Law, the word "temperature" obviously does not have the same rigorous definition. Hence, comparing predictions made pre-1900 to predisctions made post-1900 may not be so easy, since the root concept of the modern meaning of "temperature" did not yet exist.

I said:

Eddington knew this, and says in his book (The Internal Constitution of the Stars, 1926 & 1930) that the spectral distribution of the energy would not be in acordance with the Planck Law.

And Ian says:

So although Eddington suggested an effective temperature of 3.18K (black body), he was wrong on it corresponding to Planck's law, and never mentioned the frequency.

So how could Eddington have been "wrong on it corresponding to Planck's law", when in fact, he said that it "would not be in acordance with the Planck Law"?

In fact, Eddington said:

Radiation in interstellar space is about as far from thermodynamical equilibrium as it is possible to imagine, and although its density corresponds to 3.18 degrees it is much richer in high-frequency constituents than equilibrium radiation at that temperature
The Internal Constitution of the Stars, 2nd edition (1930), as reprinted by Dover Publications (1959 and many times thereafter), page 371.

In this passage "thermodynamical equilibrium" is a reference to Planck's Law, which describes the spectral energy distribution of electromagnetic radiation in thermodynamical equilibrium. That is where Eddington explicitly says that the radiation he is talking about is non-thermal (non-Planckian). Alpher & Herman made no such concession in their paper, and so were necessarily talking about strictly thermal (Planckian) radiation. The "density" in Eddington's sentence is the volume energy density he computed for the radiation.

And I might point out that Eddington does mention the frequency. Specifically, in table 48 (page 372), and surrounding text, he points out that at each frequency, the "effective temperature" would be different, and gives the temperatures in the table (i.e., at a wavelength of 600 Angstroms, that "3.18 Kelvin" radiation field would look like 4707 Kelvins!). A frequency dependent "temperature" is another dead giveaway that "Planckian" is not part of the game.

Nobody could have predicted the same thing as Alpher & Herman (and Gamow & others) predicted, prior to about 1900 in any event, since the root concepts did not exist. Microwaves were certainly not known before the discovery of electromagnetic waves by Hertz, about 1885, so nobody would have used the word, or even understood the concept, prior to about 1865, when Maxwell theoretically predicted the existence of electromagnetic waves. And both "black body" and "Planck Law" did not have their modern meaning until about 1900.

[iantresman]

Tim, many many thanks for taking the time to explain all the details. Stuff like "blackbody" being synonymous with "Planckian" you just can't look-up anywhere. Likewise fixing the temperature, fixes the frequency, in this context.

Understanding it all, is another matter. For anyone else who is following this thread who would like more information, the following pages may help:

• Multiwavelength Milky Way: Radiation Laws
• Planck's law of black body radiation (Wikipedia)
• Blackbody Spectrum (Interactive Appelt)

Regards,
Ian Tresman

[StupendousMan]

The first use of the "Cosmic Microwave Background" that I am aware of is "Peculiar Velocity of the Sun and its relation to the Cosmic Microwave Background" by Stewart and Sciama, Nature (1967).

I found a (very) slightly earlier use of the term thanks to ADS. Kip Thorne's paper "Primordial Element Formation, Primordial Magnetic Fields, and the Isotropy of the Universe"

http://adsabs.harvard.edu/abs/1967ApJ...148...51T

was published in the April, 1967, volume of ApJ, seven months before the Stewart and Sciama paper hit the streets.

[Nereid]

[Moderator Note]

This old thread, and several others, was hijacked by a BAUT member, to promote an ATM idea.

The hijacking posts have been moved to a new thread in the ATM section.

[/Moderator Note]

[blueshift]

The thing I most remember concerning this subject was the insistence that the CMB should be the most perfect black body spectrum ever produced and I think it was Dicke that gave such a prediction.

Alex Fillipenko, in his first edition of the video series "Understanding the Universe: An Introduction to Astronomy" gave mention to the fact that quite often is the case that physics students will often try to make their data fit what is ideal in recreating Planck curves on paper. So they fudge it a little. However, when they took measurements of the CMB, they found that every piece of data wound up exactly on a perfect curve as predicted by Dicke. Such a situation would have had no ambient energies coming from any "outside" source of the universe and, hence, should give a precise curve AFAIK. Maybe others here heard a little different?

BTW, for those that are interested, Filippenko's second edition is more involved and he cross examines Arp more closely than anything I have read at this forum.

[KaiYeves]

I had a dream that I recived a message from a planet of super-intelligent dolphins that was, you guessed it, whisles and clicks. But because it sounded just like background radiation, I got a pink slip.