Ultra-Cold Neutrons

R. Golub, D Richardson, S.K Lamoreaux

• Publisher: CRC Press
• Year: 2017
• ISBN:
  9781351406390
• Edition: 1
• Topic: Sciences physiques
• Subtopic: Physique nucléaire

Ultra-Cold Neutrons is a complete, self-contained introduction and review of the field of ultra-cold neutron (UCN) physics. Over the last two decades, developments in UCN technology include the storage of UCN in material and magnetic bottles for time periods limited only by the beta decay rate of the free neutron. This capability has opened up the possibility of a wide range of applications in the fields of both fundamental and condensed state physics. The book explores some of these applications, such as the search for the electric dipole moment of the neutron that constitutes the most sensitive test of time reversal invariance yet devised.The book is suitable as an introduction to the field for research students, as a useful compendium of results and techniques for researchers, and is of general interest to nonspecialists in other areas of physics such as neutron, atomic, and fundamental physics and neutron scattering.

1

Research with ultra-cold neutrons

Following their discovery in 1932 by Chadwick, free neutrons could only be studied under conditions where they spent no more than a brief moment within the experimental apparatus. Even ‘long-wavelength’ neutrons (long from the point of view of classical neutron scattering) produced in a cold source (see Chapter 3) with λ ∼ 10 Å, v = 400 m s^–1 take only 2.5 ms to travel 1 m. When diffusing through matter neutrons have average lifetimes of, for example, 0.2 ms (H₂O) or 130 ms (D₂O). However, the development, in the last two decades, of the technology of ultra-cold neutrons, has now reached the point where neutrons can be stored in material and magnetic bottles, for times which are essentially limited only by the neutron’s β-decay lifetime (τ_β ∼ 900 s). This has opened up the possibility of a wide range of new applications, some of which have reached a comparatively advanced stage of development while others are only taking their first tentative steps.

It was Fermi who first realized that the coherent scattering of slow neutrons would result in an index of refraction, or effective interaction potential V for slow neutrons travelling through matter, and that this potential would be positive (index of refraction n < 1) for most materials (see Chapter 2). This effective potential is crucial for many of the effects grouped together under the topic of Neutron Optics. Sears (1989) provides a comprehensive discussion of this field in a book which in many ways is complementary to this one.

Fermi also realized that the result of this was that those neutrons (with energy E) incident on a surface at a glancing angle θ which satisfied

Esin2θ≤Vsinθ≤sinθc=(VE)1/2⁢(1.1)

would be totally reflected just as light can be totally reflected on aproach-ing a glass-air boundary from the glass side. Fermi and Zinn (1946) and Fermi and Marshall (1947) performed the first experimental demonstration of this effect. Total reflection of neutrons has provided the basis for the highly successful technique of neutron guide tubes, in which neutrons whose angles satisfy (1.1) can be transported large distances through guides whose surfaces are smooth enough so that non-specular reflections (reflections for which the angle of incidence is not equal to the angle of reflection) are negligible, as first suggested by Maier-Leibnitz and Springer (1963). The neutron guide technique has virtually transformed slow neutron scattering from a somewhat esoteric technique to one of much wider applications. Bée (1988) and ILL (1988) give a picture of the guide tube installations at the Institut Laue-Langevin in Grenoble while Arif et al (1989) describe the installations at the NIST in Maryland. See also Serebrov (1989) for a description of the impressive facility under construction at Gat china, south of Leningrad. There are many other installations of which we mention the Rutherford laboratory, Saclay, Jülich and Geesthacht.

The observation of the total reflection of neutrons led to the speculation that if neutrons with energies

E≲V⁢(1.2)

could be obtained—this is not obvious as typical materials have V ∼ 10⁻⁷ eV while thermal neutrons have energies of 2.5 × 10^–2 eV—they would undergo total reflection at any angle of incidence and hence could be stored in closed vessels. We refer to such neutrons as Ultra-Cold Neutrons (UCN). Although this book is devoted to UCN we will, from time to time, discuss work with faster neutrons. While many workers in the field of neutron physics attribute the idea of neutron storage to Fermi, the first person to take the idea seriously enough to put it into print was Zeldovich (1959).

He pointed out that, although the lifetime of a neutron in, for example, graphite is only 10^–2 s (independent of velocity, see Chapter 2), because of the small penetration depth of a UCN during total reflection (∼10² Å = 10⁻⁶ cm) the fraction of the time that stored UCN would in fact spend in contact with the walls is quite small (∼10⁻⁷) and so one could expect an absorption time of approximately 10⁵ s for stored UCN. This is in good agreement with more detailed calculations (Chapter 4). Zeldovich also estimated that a thermal flux of 10¹² n cm⁻² s⁻¹ cooled to 3 K in liquid helium would produce a UCN density of 50 cm⁻³. It is interesting to note that such densities have now been achieved at the Institut Laue-Langevin, Grenoble, using a reactor with a thermal flux of 10¹⁵ n cm⁻² s⁻¹ cooled to 20 K in a deuterium-filled cold source (Chapter 3). Zeldovich suggested it would be interesting to study the interactions of the stored UCN with substances introduced into the cavity e.g. (11,7) absorbers.

Shortly afterwards Vladimirskii (1961), suggested the use of magnetic field gradients to produce focused beams of polarized neutrons and ‘magnetic mirrors’ to confine UCN. For a reactor with a thermal flux of 2 ÷ 10¹³ n cm⁻² s⁻¹ he estimates a UCN density of 10⁻² cm⁻³ for the small effective potential V_e ∼ 3 × l0⁻⁸eV for B = 5 × 10³ G. Vladimirskii also suggested extraction of the UCN from the reactor by means of a vertical channel with the magnetic bottle located above the reactor, and pointed out the widening in solid angle as the neutrons travel up the guide (see Chapter 3). He emphasized the importance of the potential in the moderator material and that its effects can be countered by vertical extraction.

This was followed by Doroshkevich (1963) who suggested beryllium as a material for a storage vessel and estimated the temperature dependence of the loss rate due to inelastic scattering of the UCN by the walls. He estimated the loss rate due to wall vibrations as less than 10^–7 s⁻¹.

Foldy (1966) published some speculations concerning the storage of UCN. He suggested a bottle whose walls were coated with liquid helium (V = 1.1 ÷ 10⁻⁸ eV). A degenerate Fermi gas of neutrons up to this potential would have a staggering density of 10¹⁴ UCN/cm³ but he made no suggestion as to how such densities could be achieved.

In 1968, Shapiro published a review article on the electric dipole moment (EDM) of elementary particles. In this article he pointed out the advantages of UCN for the search for a neutron EDM (see Chapter 7), especially the greatly increased observation time and the reduction of the ^lv ÷ E’ effect (a magnetic field, produced in the frame of the moving neutron by the applied electric field, interacting with the neutron’s magnetic moment and mimicking an EDM). See also Golub and Pendlebury (1972) for a more detailed discussion of this point.

Given the fact that the energy V (1.2) is some 10⁵ times smaller than the thermal energy of neutrons in the reactor moderator and that the Maxwellian energy spectrum for neutron flux is proportional to E for low energies, it is remarkable that two groups independently had the courage to invest the time and effort to construct the necessary installations on the chance that neutrons so far from the peak of the Maxwell distribution did indeed exist inside the reactor, and that they could be extracted without crippling losses of intensity. That both groups were successful almost simultaneously is one of those coincidences which seem to be so common in the history of physics.

The Dubna group under F L Shapiro (Luschikov et al 1968, 1969) extracted UCN from a very low power pulsed reactor by means of a curved horizontal channnel, 9.4 cm ID, 10.5 m long. Counting rates of 0.8 counts/10² s (background ∼ 0.4 counts/10² s) were obtained. By admitting helium gas to the extraction pipe the authors attempted to estimate the storage time in the pipe. The idea is that when the average lifetime of a neutron for collisions with the helium

τHe=[NHeσHev¯He]-1⁢(1.3)

is equal to the average lifetime for wall losses the counting rate should be reduced by a half with respect to that in the absence of helium. This first attempt to measure storage times gave a result of 200 s which is to be compared with the 12 s obtained from later more detailed measurements (Groshev et al 1971), the discrepancy being attributed to possible impurities in the helium.

Working at Munich, Steyerl (1969) obtained UCN by vertical extraction from a steady state reactor. The beam was pulsed by a rotating chopper constructed out of 13 boron silicate glass plates located deep within the reactor swimming pool 2 m above the core, allowing time of flight measurements of neutron spectra. The counting rate showed a steep drop below 10 m s^–1, probably due to absorption in the aluminium windows, to reflection losses and the limited acceptance angle of the detector. However total cross sections were measured for neutron velocities down to 7 m s⁻¹ for gold and 5ms⁻¹ for aluminium (Chapters 6 and 8).

It is noteworthy that both these initial attempts were made at relatively low intensity sources, an average thermal flux of 1.6 × 10¹⁰ n cm⁻² s^–1 in the Dubna experiment and 10¹³ n cm⁻² s^–1 in the Munich experiment, thus demonstrating the ability to carry out really new and important innovations at weak sources.

Following these first experiments Okun (1969) called attention to Shapiro’s point that UCN offered a promising method for improving the sensitivity of the search for a neutron EDM emphasizing the potential improvement in observation times—10³ s for UCN compared with 10⁻² s in a typical beam experiment. He also mentioned the attraction of UCN for measuring the neutron lifetime.

The ne...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Table of Contents
• Preface
• 1 Research with Ultra-Cold Neutrons
• 2 Interaction of UCN with Matter
• 3 Production of UCN
• 4 UCN Gas
• 5 UCN Storage Measurements
• 6 UCN Spectrometers, Microscopes and Monochromators
• 7 Applications of UCN to Fundamental Physical Tests and Measurements
• 8 UCN Scattering
• Appendices
• References
• Index