George B Field Cyanogen and the CMBR

George Field is Senior Research Fellow at the Smithsonian Astrophysical Observatory, and was Director of the Observatory from 1973 to 1982. His current research interest is turbulence in astronomical settings.

My encounter with the microwave background began in 1955. I had come to Harvard as a postdoc, intending to search for intergalactic hydrogen by looking for 21-cm absorption in the spectrum of the radio source Cygnus A. While I was making the observations at the Harvard 28-footer, I studied the problem of the excitation of the upper level of the line, as that is crucial in calculating the absorption coefficient. I realized that it can be excited by fluorescence Lyman-a photons, by collisions with atoms or free electrons, or by absorption of 21-cm photons from whatever source. Ed Purcell and I calculated the collision cross section, I estimated the effect of Lyman-a radiation, and I proposed that we measure the continuum near 21 cm to get the radiation field that would excite the line. When I asked Doc Ewen how to measure the continuum, he said it could not be done at that time because it would require an absolute measurement whose zero point was known. So I extrapolated continuum maps at 21 cm to the coldest point and estimated 1K. Clearly that was a lower limit, because without a zero point, there was no way to know what the coldest point represented. I published the result in 1959. Of course we now know that the zero point is 3 K.

In 1957, I joined the faculty at Princeton, where I had taken my PhD in astronomy in 1955. I knew about interstellar molecules from Lyman Spitzer, who was studying optical interstellar lines at Mt. Wilson as part of his research at Princeton on the interstellar medium. In fact, in my first published paper, in 1955, Lyman and I mentioned an unidentified line that appeared in our tracing that later was identified as interstellar CH+. But I was particularly intrigued by a reference in Herzberg's (1950) book on diatomic molecules which stated that one of the lines of interstellar CN arose from a rotationally excited state (J = 1) in the ground electronic and vibrational state. The excitation temperature was estimated to be 2.3 K. This was unique in interstellar studies, and so with my experience with atomic hydrogen, I calculated the excitation to be expected from collisions and fluorescence radiation transitions. They failed by a large factor to account for the excitation. To calculate the effect of radiation at 2.6-mm wavelength, which might excite the molecule from the J = 0 to 1 levels, I needed two things: the permanent dipole moment of CN, in order to calculate the Einstein B coefficient, and the mean intensity of 2.6-mm radiation at the positions of the interstellar molecules.

The dipole moment had never been measured, so I estimated it from CO, for which it is 0.1 Debye, to be 0.05 Debye, enough to couple the excitation to the radiation field at 2.6 mm. Just as in the case of the 21-cm line, the mean radiation intensity, expressed as a radiation temperature, had not been measured either, but I convinced myself that from the CN observations themselves, 2.3 K was a good estimate. I wrote all this up, and concluded that there must be previously unrecognized source of radiation at 2.6 mm. I gave the paper to Lyman Spitzer to read. He thought it was too speculative to submit for publication, probably because the dipole moment, which determines the coupling to the radiation field, was only an estimate. All this took place before 1960. I recall that because I was then at the old Observatory on Prospect Street in Princeton, whereas we moved to a new building at that time.

One event that took place in the new building was a visit from Arno Pen-zias. I recall standing in the door of his office discussing his plans to observe 21-cm radiation emitted by atoms in intergalactic space. If the hydrogen is excited solely by the background radiation, no emission will be detected, as it is exactly cancelled by the absorption of the background. Thus we were led to think about the temperature of the background radiation. As I recall, Arno was not optimistic about the absolute measurements required.

The Dicke group was working on the Brans-Dicke (1961) theory of gravitation at Palmer Lab. I knew Dicke and Peebles, and recall attending a seminar there by Jim Peebles explaining his work on helium production in the big bang, of course in Brans-Dicke cosmology. I went up afterward and told Jim that colleagues of George Gamow, including Alpher and Herman, had done similar work. I think I knew at that time of the prediction of 5K for the background radiation by Alpher and Herman, but I don't recall mentioning it. Moreover, it did not occur to me to mention my work on CN either, because I had not made the connection with the big bang.

I also recall that while teaching a course in Palmer I noticed a microwave horn out of the window, pointing to the vertical. It must have been Roll and Wilkinson's experiment, but again I did not make the connection.

Fast forward to 1965, when the discovery was published in The New York Times (Sullivan 1965). I missed it, perhaps because I was packing to move to Berkeley that summer. However, I soon learned about it from a call from Bernie Burke that I got in my Berkeley office. When he said "3 K" I at once realized that it could be the source of radiation that I had predicted in my work on CN before 1960. Unfortunately, my manuscript on the subject at that time is either lost or in cold storage.

Nevertheless, I thought maybe the CN data would be useful. I knew that to make the case I needed to find a value for the CN dipole moment. By a strange coincidence, it was hiding in my wastebasket. At the time, I was writing an article for the Annual Reviews of Astronomy and Astrophysics, and the editor had sent me proofs from another article as a guide to marking my own proofs. The article was on The Spectra of Comets, by Claude Arpigny, in which he discusses how to predict the emission spectra of molecules - including CN - using rate equations for level populations. One of the parameters is the dipole moment of the ground electronic state, which he had adjusted to fit the data. His number, 1.2 Debye, was not far from a more recently measured laboratory value, 1.4 Debye. Arpigny's dipole moment enabled me to calculate the coupling of the J = 1 rotational level of CN to the radiation field at 2.6 mm. I found that the coupling to radiation is stronger than to collisions or fluorescence by a large factor. Much stronger, even, than I had concluded before 1960, by the square of the dipole moment, a factor of 200. I knew then that we had a radiation thermometer at 2.6 mm.

Another coincidence occurred the same day. When John Hitchcock, a graduate student working in the next office, heard what I was doing, he came in and told me that at that moment he was working on observations of the rotational excitation of interstellar CN. He was reducing data that he had taken from six plates that George Herbig had taken of the spectrum of the star Zeta Ophiuchi at the wavelength of the interstellar CN line. Suddenly we had new data to which to apply the theory of excitation. Together with George Herbig we wrote an abstract of a paper for the 120th meeting of the American Astronomical Society, which was meeting in Berkeley (another coincidence). At that meeting, held December 28-30, 1965, we presented evidence that the background radiation follows a blackbody spectrum over the 28-fold wavelength interval from 7.4 cm to 2.6 mm. Our value of the temperature was given as 2.7-3.4K (Field, Herbig and Hitchcock 1966).

John and I published two more papers on the subject (Field and Hitchcock 1966). One in Physical Review Letters gave a result of 2.7-3.6K for Zeta

Persei, a star on the other side of the sky from Zeta Oph, and 300 pc distant from it. Thus the hypothesis that the radiation is universal passed the test. In a later paper in The Astrophysical Journal we considered the possibility that the spectrum of radiation is not blackbody after all, but as suggested to us by Nick Woolf, dilute blackbody at a higher temperature. We were able to rule out this hypothesis with reasonable certainty. It is interesting that the peak of the blackbody curve in frequency units is 1.7 mm. With our measurements at 2.6 mm, we were climbing the peak.

Telescopes Mastery

Telescopes Mastery

Through this ebook, you are going to learn what you will need to know all about the telescopes that can provide a fun and rewarding hobby for you and your family!

Get My Free Ebook

Post a comment