# End to the Cosmic-Ray Spectrum?

> An annotated edition of this paper on Fermat's Library, where scientists and readers explain its ideas, equations and history in the margins. Annotated edition: https://fermatslibrary.com/s/end-to-the-cosmic-ray-spectrum

- **Authors:** Kenneth Greisen
- **Published:** 1966
- **Topic:** Physics
- **Original:** http://physics.princeton.edu/~mcdonald/examples/EP/greisen_prl_16_748_66.pdf
- **Pages:** 3
- **Annotations:** 8 comments in 7 threads
- **Featured:** Fermat's Library Journal Club
- **Annotated edition:** https://fermatslibrary.com/s/end-to-the-cosmic-ray-spectrum

## Annotations

### Annotation 1 (page 1)

Beside the passage: “The primary cosmic-ray spectrum has been measured up to an energy of 10' eV, ' and several groups have described projects under development or in mind' to investigate the spectrum further, into the energy range 10"-10" eV. This note predicts that above 10' eV the primary spectrum will steepen abruptly, and the …”

**Annotation** by Micael Oliveira (2016-01-05):

> Particles having energies above $10^{18}$ eV are the so-called ultra-high-energy cosmic rays (UHECR). The UHECR are the most energetic particles known in the Universe.

**Reply** by Micael Oliveira (2016-01-05):

> Spectrum of cosmic rays as a function of their energy given by several different experiments.
> ![Cosmic Ray Spectrum](http://www.iexp.uni-hamburg.de/groups/astroparticle/score/images/spectrum.png)

### Annotation 2 (page 1)

Beside the passage: “The primary cosmic-ray spectrum has been measured up to an energy of 10' eV, ' and several groups have described projects under development or in mind' to investigate the spectrum further, into the energy range 10"-10" eV. This note predicts that above 10' eV the primary spectrum will steepen abruptly, and the …”

**Annotation** by Micael Oliveira (2016-01-05):

> The Cosmic Microwave Background (CMB), discovered in 1966 by A. Penzias and R. Wilson, corresponds to the remnant photons of the Big Bang and acts like a black body spectrum with temperature T$_{\mathrm{CMB}} \simeq$ 2.73 K. At the highest energies, the Universe is no longer transparent to Cosmic Rays, that start interacting with the CMB photons.
>
> ![CMB](http://map.gsfc.nasa.gov/media/ContentMedia/990015b.jpg)

### Annotation 3 (page 2)

Beside the passage: “give strong assurance that the radiation is indeed universal. The equivalent black-body temperature has been reported as 3. 1+ 1'K' and 3. 0+0. 5'K. ' For our discussion, we shall consider T =3. 0, at which temperature the photon density is 548 cm ' and the mean photon energy 7. 0x10 4 eV. Although at this temperature …”

**Annotation** by Micael Oliveira (2016-01-05):

> The main reactions causing energy losses while the CR propagation are:
> * Compton interactions of nuclei
> * Pair production
> * Photodisintegration of the nucleus
> * Photopion production
>
> In the case of UHECR the reaction that has the most influence on the spectrum is photopion production given by the following interactions:
> \begin{eqnarray*}
> p +\gamma_{\text{CMB}}  &\longrightarrow& \Delta^+ \longrightarrow p + \pi_0\\
> p +\gamma_{\text{CMB}}  &\longrightarrow& \Delta^+ \longrightarrow n + \pi^+\\
> p +\gamma_{\text{CMB}}  &\longrightarrow& \Delta^{++} +\pi^- \longrightarrow p + \pi^+ + \pi^-\,\,
> \end{eqnarray*}where $p$ is the UHECR and $\gamma_{\text{CMB}}$ are the CMB photons.

### Annotation 4 (page 2)

Beside the passage: “give strong assurance that the radiation is indeed universal. The equivalent black-body temperature has been reported as 3. 1+ 1'K' and 3. 0+0. 5'K. ' For our discussion, we shall consider T =3. 0, at which temperature the photon density is 548 cm ' and the mean photon energy 7. 0x10 4 eV. Although at this temperature …”

**Annotation** by Micael Oliveira (2016-01-05):

> We can easily determine the distance scale for energy loss given by:\begin{eqnarray*}\lambda_{E}=\frac{1}{\kappa \rho_{\gamma} \sigma_{\text{p}\gamma}}\end{eqnarray*}where $\kappa$ is the inelasticity (fraction of the initial energy available for the production of new particles), $\rho_\gamma$ is the CMB photon density, and $\sigma_{\text{p}\gamma}$ is the mean p-$\gamma$ interaction cross section. Using these values we find that $\lambda_{E}\simeq 10^{25}\,\textrm{cm}\simeq 3.24\, \textrm{Mpc}$.
>
> For each distance $\lambda_{E}$ the CR will loose 22% of its energy due to interactions with the CMB. The distance obtained corresponds to a characteristic time scale for energy loss of the order of $10^{15}$ s which is several orders of magnitude smaller than the age of the Universe ($4.3\times10^{17}$ s). This result could indicate that super GZK CRs:
> * There are near sources that we have not identified yet. Protons arriving at Earth with super-GZK energies must come from sources situated at less than $\sim 100\,Mpc$ away.
>
> * At the cutoff, the “visible” universe shrinks to a sphere of a few tens of Mpc of radius. This feature should be reflected in the energy spectrum of cosmic rays as a sharp drop.

### Annotation 5 (page 2)

Beside the passage: “give strong assurance that the radiation is indeed universal. The equivalent black-body temperature has been reported as 3. 1+ 1'K' and 3. 0+0. 5'K. ' For our discussion, we shall consider T =3. 0, at which temperature the photon density is 548 cm ' and the mean photon energy 7. 0x10 4 eV. Although at this temperature …”

**Annotation** by Micael Oliveira (2016-01-05):

> As a simple exercise, one can calculate the threshold energy of protons (we consider the UHECR to be a proton) for the photopion production. It is easy to obtain the mean energy for the CMB. We know that the CMB has a temperature of $T_{\gamma} \simeq \, 2.73$ and a mean wavelength of $\lambda_{\gamma}= 1.96\,mm$, from which we can compute the mean energy as $E_{\gamma} = hc/\lambda_{\gamma} \simeq 6.34 \times 10^{-4} \, eV$.
>
> Considering the proton and photon 4-momenta we have that the energy in the center of mass (CM) is given by:
> \begin{eqnarray*}
> \label{gzk1}
> \left(  p_{\text{p}} + p_{\gamma}  \right)^{\mu}\left(  p_{\text{p}} + p_{\gamma}  \right)_{\mu} &=&  m_{\text{p}}^2c^2 + 2\frac{E_{\text{p}}E_{\gamma}}{c^2}-2\vec{p}_{\text{p}}\cdot\vec{p}_{\gamma}
> \end{eqnarray*}where $E_{\text{p}}$ is the energy, $m_{\text{p}}$ is the mass, and $\vec{p}_{\text{p}}$ is the three-dimensional momentum of the proton in the laboratory frame, and similarly for the photon. We can write the equation:\begin{eqnarray*}
> \label{gzk2}
> \left(  p_{\text{p}} + p_{\gamma}  \right)^{\mu}\left(  p_{\text{p}} + p_{\gamma}  \right)_{\mu} &=&  m_{\text{p}}^2c^2 + 2\frac{E_{\text{p}}E_{\gamma}}{c^2} \left(  1- \cos{\theta} \right)
> \end{eqnarray*}where we have used the fact that protons are ultrarelativistic so $\beta \approx 1$. To find the threshold energy of this reaction  we use the above equations and we assume a frontal collision $\theta=\pi$. We know that the threshold energy in the center of mass for this reaction is given by $\sqrt{s}=\left(m_{\text{p}}+m_{\pi^0}\right)c$. Finally we have that the threshold energy for the protons is given by:\begin{eqnarray*}
> \label{gzk3}
> 4\frac{E_{\text{p}}E_{\gamma}}{c^2} + m_{\text{p}}^2c^2  &=&  \left(  m_{\text{p}} + m_{\pi^0} \right)^2c^2
> \end{eqnarray*}and the solution to of this equation for $E_{\text{p}}$ is:\begin{eqnarray*}
> \label{gzk4}
> E_{\text{p}} &=&\frac{m_{\pi^0}\left( 2m_{\text{p}} + m_{\pi^0} \right)c^4}{4E_{\gamma}}\,\,.
> \end{eqnarray*}Using the mean energy for the CMB we find that $E_{\text{p}}\simeq 1.07\times 10^{20}$ eV.
>
>
> We have found that protons with energies equal or higher than $10^{20}$ eV will interact with the CMB thus losing part of their energy.

### Annotation 6 (page 2)

Beside the passage: “give strong assurance that the radiation is indeed universal. The equivalent black-body temperature has been reported as 3. 1+ 1'K' and 3. 0+0. 5'K. ' For our discussion, we shall consider T =3. 0, at which temperature the photon density is 548 cm ' and the mean photon energy 7. 0x10 4 eV. Although at this temperature …”

**Annotation** by Micael Oliveira (2016-01-05):

> Spectrum of UHECR as observed by HiRes, TA and Auger experiments (as labeled). One can clearly observe a drop in the spectrum for Cosmic Ray energies above $10^{20}$ eV. K. Greisen's prediction was experimentally observed by three different experiments.
>
> ![GZK cutoff](https://ned.ipac.caltech.edu/level5/March14/Aloisio/Figures/figure5.jpg)

### Annotation 7 (page 3)

Beside the passage: “Ordinary optical interstellar radiation can also produce y-p photopions and heavy nucleus disintegrations, at energies 1000 times less than those discussed above; but the intergalactic optical photon density is smaller than that of the 3' radiation by a factor of about 5x10, and the mean paths are correspondingly …”

**Annotation** by Micael Oliveira (2016-01-05):

> Today the upper limit on the energy of Cosmic Rays is called **Greisen–Zatsepin–Kuzmin limit (GZK limit)**. Several months later G. Zatsepin and V. A. Kuzmin published
> [Upper Limit of the Spectrum of Cosmic Rays](http://www.jetpletters.ac.ru/ps/1624/article_24846.pdf).

## Full text

Extracted from the PDF. [Annotation N] marks where each annotation above sits.

#### Page 1

VOLUME 16, NUMBER 17 PHYSICAL REVIEW LETTERS 25 APRiL 1966

tons. When the complete potential is deformed, the optical parameters (Table I) found so far to give the best fit to the elastic polarization also produce the best prediction of inelastic asymmetry.

All the curves in Fig. 2 use a central-well deformation parameter of P, =0. 39 for ~si and P, = 0. 22 for ~Ni. The deformation parameter of the spin-orbit term is 1. 5 times the centralwell value, which produces slightly better agreement with the asymmetry data for "Si. Both real and imaginary parts of the spin-orbit interaction are included, but since i W~I «VS, the imaginary part makes little difference. The curves also include Coulomb-excitation amplitudes, ' which make little difference in either the asymmetry or the cross section. We find that for all of the calculations made, the predictions of inelastic asymmetry and inelastic polarization are very nearly identical.

In summary we find that, provided the imaginary and spin-orbit terms are included, the collective-model generalization of the optical potential gives a good account of the present inelastic asymmetry data at all but the most forward angles. It is quite possible that a more comprehensive treatment of the spin-dependent interaction will improve matters in this region, and such calculations are in progress.

It is a pleasure to acknowledge many useful

conversations concerning this work with G. R. Satchler and N. M. Hi.ntz. We are much indebted to the indefatigable ORIC cyclotron operators, and to M. B. Marshall, W. H. White, I. B. Schneider, and A. W. Riikola of the ORIC staff, for their essential contributions to the experimental effort.

*Research sponsored by the U. S.Atomic Energy Commission under contract with the Union Carbide Corporation.

)Oak Ridge Graduate Fellow from the University of Minnesota under appointment from Oak Ridge Associated Universities. f. Graduate Fellow from the University of Tennessee under appointment from the National Science Foundation.

~R. H. Bassel, G. R. Satchler, R. M. Drisko, and E. Rost, Phys. Rev. 128, 2693 (1962). The present DW calculations include the spin-orbit term in the elag+ic distortion.

M. P. Fricke and G. R. Satchler, Phys. Rev. 139, B567 (1965). 3T. Stovall and N. M. Hintz, Phys. Rev. 135, B330 (1964).

See, for example, W. S. Gray, R. A. Kenefick, J. J. Kraushaar, and G. R. Satchler, Phys. Rev. 142, 735 (1966); G. R. Satchler, to be published. 5E. R. Flynn and R. H. Bassel, Phys. Rev. Letters 15, 168 (1965), and other references given there. 6L. N. Blumberg, E. E. Gross, A. van der Woude, and A. Zucker, Nucl. Instr. 5 Methods ~39 125 (1966); L. N. Blumberg, E. E. Gross, A. van der Woude, A. Zucker, and R. H. Bassel, to be published.

END TO THE COSMIC-RAY SPECTRUM'~

Kenneth Greisen Cornell University, Ithaca, New York (Received 1 April 1966)

The primary cosmic-ray spectrum has been measured up to an energy of 10' eV, ' and sev- [Annotation 1] eral groups have described projects under development or in mind' to investigate the spectrum further, into the energy range 10"-10" eV. This note predicts that above 10' eV the primary spectrum will steepen abruptly, and the [Annotation 2] experiments in preparation will at last observe it to have a cosmologically meaningful termi. - nation.

The cause of the catastrophic cutoff is the intense isotropic radiation first detected by

Penzias and Wilson' at 4080 Mc/sec (7. 35 cm) and now confirmed as thermal in character by measurements of Roll and Wilkinson4 at 3. 2 cm wavelength. It is not essential to the present argument that the origin of this radiation conform exactly to the primeval-fireball model outlined by Dicke, Peebles, Roll, and Vfilkinson', what matters is only that the radiation exists and pervades the observable universe. The transparency of space at the pertinent wavelengths, and the consistency of intensity observations in numerous directions,

748

#### Page 2

VOLUME 16, NUMBER 17 PHYSICAL REVIEW LETTERS 25 APRIL 1966

give strong assurance that the radiation is indeed universal. The equivalent black-body temperature has been reported as 3. 1+ 1'K' and 3. 0+0. 5'K. ' For our discussion, we shall consider T =3. 0, at which temperature the photon density is 548 cm ' and the mean photon energy 7. 0x10 4 eV. Although at this temperature the number of photons in the spectral range of the measurements (A =- 3. 2 cm) is only 5 &10 ' of the total, the slope of the spec- [Annotation 3] trum is such that any reasonable extrapolation to shorter wavelengths would yield at least a substantial [Annotation 4] part of the 3' black-body photon density. Moreover, two indirect confirmations of the existence of the radiation have been reported: One lies in the slope of the isotropic part of the xand gamma-ray spectrum' and the other in the absence of muon-poor air showers above 10" eV. ' As the last statement implies, severa, l consequences of the existence of the thermal radiation have quickly been noted. One is to provide a source of x rays and gamma rays by inverse Compton interactions with cosmicray electrons. '» Another is to make the universe opaque to high-energy photons, above 2 x10'4 eV, because of positron-electron pair creation by photon-photon interactions. '~" A third effect is to deplete the density of energetic electrons by the energy losses in the inverse Compton interactions. '~' Hoyle' also considered the effect of the thermal radiation on cosmic-ray protons, but concluded that the time scale for energy degradation is greater than the expansion time of the universe for all protons up to 10 ' eV. This conclusion is wrong because he only considered the proton Compton effect and neglected two stronger processes, namely pair creation and photopion production, which we now wish to examine. The threshold energy for pion production by protons on photons of energy 7 x10 ~ eV (the mean [Annotation 5] energy of black-body radiation at 3'K) is 10' eV, and some pion production occurs at lesser proton energies because of the highfrequency tail of the photon spectrum. The cross section rises rapidly above the threshold p going through a peak exc ceding 400 p, b at the ~„~ resonance (2. 3 x10'0-eV proton energy on 7 x10 '-eV photon), and descending thereafter to about 200 p. b, about which minor wiggles occur owing to the [Annotation 6] superposition of higher resonances. With a mean cross section of 200 pb and a photon density of 550 cm

the mean path for interaction is (nv) ' =9 x10'4 cm. However, the distance scale for loss of energy is L =(E/AE)(nv) ', E being the initial proton energy and AE the energy loss per interaction. At the threshold for single-pion production, ~/E is only 0. 13, but it rises to an average value of 0. 22 at the -, ', -, resonance, and continues to rise thereafter as multiple pions are produced or more kinetic energy is given to a, single pion. L is therefore on the order of 4x10" cm, and the time sca. le for energy loss is 10" sec, which is several hundred times less than the expansion time of the universe. L is also more than an order of magnitude less than the distance to the nearest quasar. There is abundant evidence that above 10" eV, the cosmic rays are not confined to the galaxy; the local intensity is a sample of the flux in a much larger sphere. If the sources of very high-energy particles are uniformly distributed in space and time, the effect of interactions like those described here is to deplete the spectrum by a factor equal to the ratio of the time scale for energy loss to onethird the expansion time. If, on the other hand, the sources of such particles exist only far back in time or at great distances, the depletion is much stronger. It may also be noted that if the primeval-fireball model is correct, going back in time raises the mean photon energy as (1-t/T) ' and the photon density as (1 t/T), T bein— g the expansion time; thus the effect may be somewhat larger than our computations on a static model indicate. It should be noted that the cut in the spectrum due to photopion processes is rather sharp, because of the steepness of the high-frequency tail of the Planck distribution. Only 1% of the photons have energies exceeding 3 times the mean value; also, close to the threshold the cross section is smaller than 200 p, b and the fractional energy loss per interaction is a minimum. Therefore, below 3 &10" eV the process should have a completely negligible effect on the proton spectrum. As 10' eV is approached, the effect should rise rapidly', and above 2 x 1 0 eVp it should be a, factor of several hundred. At present the data above 10'9 eV are rather sparse, and the highest energy recorded is represented by a single event at 10'0 eV. ' A smooth representation and extrapolation of the spectrum gives an integra, l frequency of about one event on 100

#### Page 3

VOLUME 16, NUMBER 17 PHYSICAL REVIEW LETTERS 2$ APRIL 1966

km' in one year at energies above 2 x10 eV. If this number is cut by a factor of several hundred, owing to the y-p reaction, the rate will be far too low to be detected by any of the methods yet proposed; even the one event recorded at 10' eV appears surprising. One cannot save the day for superhigh-energy cosmic rays by calling on heavy nuclei. The threshold for photodisintegration against photons of 7 &10 ~ eV is only 5x10'8 eV/nucleon, and at 10" eV/nucleon most of the photons can excite the giant dipole resonance, for which the cross section is on the order of 10 " cm'. At this energy the mean path for photodisintegration is on the order of 2 x10 ' cm, much less than the size of the galaxy. Even nuclei 5 times less energetic would be decomposed in a time short compared with the expansion time of the universe, owing to the high-frequency tail of the black-body spectrum.

Ordinary optical interstellar radiation can also produce y-p photopions and heavy nucleus disintegrations, at energies 1000 times less than those discussed above; but the intergalac- [Annotation 7] tic optical photon density is smaller than that of the 3' radiation by a factor of about 5x10, and the mean paths are correspondingly longer. So the effect on the proton spectrum is negligible, but not the effect on the heavy nuclei: Above 10" eV/nucleon the mean time for photodisintegration is an order of magnitude less than the expansion time. Nuclei confined in the galaxy encounter a higher density of optical photons and are fragmented much faster. In addition to photopion interactions as a source of energy loss to high-energy protons, one should consider pair production by the thermal photons. The proton energy threshold for this reaction against photons of 7x10 eV is only 7 x10" eV. The energy loss in the laboratory system arises primarily from the small longitudinal momentum given to the proton in its rest system. At the threshold the fractional energy loss is 2m/M=10 ', where m and M are the electron and proton masses. At higher energies the energy loss depends on the relative velocity of the electron and positron and the transverse momentum given to the proton, but the average energy loss in

the laboratory is approximately constant, making the fractional energy loss f = 10 '/x, where x is the ratio of the proton energy to its threshold value. The cross section with no screening is approximately 1. 8X 10 27(lm -0. 5) cm'. Therefore, the scale length for energy loss is given by L = (nfl) '= 10 7 x (1ruc -0. 5) ' cm. The minimum value of L occurs at x =4. 5 or E =3x10' eV and is about half of the Hubble length. Thus, the effect on the primary spectrum is barely significant, creating a small depression (never exceeding a factor of about 3) in the interval 10"-10" eV.

Even this small depletion of the flux above 10" eV, however, followed above 5x10'9 eV by a stronger depression due to the photopion process, makes the observed' flattening of the primary spectrum in the range 10' -10' eV quite remarkable. The injection spectrum of the intergalactic flux must be much less steep than that of the galactic particles which dominate at lower energies. The author expresses thanks for the hospitality of the Physics Department of the University of Utah, where this Letter was written.

J. Linsley, Phys. Rev. Letters 10, 146 (1963); and in Proceedings of the International Conference on Cosmic Rays, Jaipur 1963, edited by R. R. Daniel et al. (Commercial Printing Press, Ltd. , Bombay, India, 1964-1965), Vol. IV, p, 77. C. B. A. McCusker and K, Greisen, in Proceedings of the International Conference on Cosmic Rays, London, September, 1965 (to be published); and S. Colgate, private communication. 3A. A. Penzias and R. W. Wilson, Astrophys. J. 142, 419 (1965). 4P. G. Roll and D. T. Wilkinson, Phys. Rev. Letters 16, 405 (1966). R. H. Dicke, P. J. E. Peebles, P. G. Roll, and D. T. Wilkinson, Astrophys. J. 142, 414 (1965). 8Private communication from Penzias and Wilson, reported in Ref. 4. J. E. Felten, Phys. Rev. Letters 15, 1003 (1965). R. J. Gould and G. Schrbder, Phys. Rev. Letters 16, 253 (1966). ~F. Hoyle, Phys. Rev. Letters 15, 131 (1965). R. J. Gould, Phys. Rev. Letters 15, 511 (1965)'. ~~J. V. Jelley, Phys. Rev. Letters 16, 479 (1966). R. R. Wilson, Phys. Rev. 110, 1212 (1958).

750

---

Source: [Fermat's Library](https://fermatslibrary.com/) — https://fermatslibrary.com/s/end-to-the-cosmic-ray-spectrum. More annotated papers: https://fermatslibrary.com/llms.txt
