Nuclear Magnetic Moment ofRb85: Resolving a Discrepancy

Abstract

The ratio of the nuclear gi factor to the atomic gj factor in 86 Rb has been measured by the atomic-beam magnetic-resonance technique to be g/(uncorr)/gj-= -1. 466478 (22) X10~4. This result is in excellent agreement with recent optical-pumping measurements but is lower than a previously reported atomic-beam result. R ECENT optical-pumping determinations of gi/gj for free 85 Rb atoms 1 are in considerable disagreement with an earlier atomic-beam measurement of the same ratio. 2 Because the discrepancy between the optical-pumping and atomic-beam results is nine times the larger experimental uncertainty, the question immediately arises whether this discrepancy is indeed real, or whether one of the measurements is in error. If the discrepancy were real, this would have had interesting implications for atomic theory.In an attempt to resolve this question, we have remeasured the nuclear magnetic moment of 85 Rb, using the atomic-beam magnetic-resonance technique, and have obtained a result in agreement with the opticalpumping measurement. In an attempt to vary our technique from that used in Ref. 2, we have used the triple-loop method 3 in our measurements and have observed the CF=3, MF= -1) <-> (3, -2) transition at its field-independent point at 2076.44 G. The frequency of this line is strongly dependent upon the value of gr (dv/dgr= -2905, where v is in MHz and gi is referred to the Bohr magneton), but depends less on the other parameters (dv/da= 9.58, dv/dgj= -0.43) in the BreitRabi equation; thus this transition is well suited to a direct measurement of gi.A special triple hairpin, shown in Fig. 1, was constructed to observe this transition. The two end hairpins were constructed of standard rigid coaxial line and were used to induce the (3, •-2)<-> (3, -3) transition. In use, the magnetic field was shimmed as necessary to make the transition frequency the same in both hairpins. The center hairpin, which was used to induce the (3, -1) <-» (3, -2) transition, was of a simple box-type construction with a center conductor shorted at the bottom of the hairpin. The center hairpin had a length of 4.5 in., resulting in a resonance linewidth of 5 kHz for this transition. This type of hairpin has been used and tested quite extensively by our group, and has yielded reliable results for frequencies up to about 700 MHz. A typical 85 Rb (3, -1) <-> (3, -2) resonance observed with this hairpin is shown in Fig. 2. The magnetic field was calibrated by measuring the frequency of the (3, --2) <-> (3, -3) transition with the center hairpin.The center frequency of the resonance was determined by measuring the frequencies of three pairs of points symmetrically placed about the center of the resonance. These six points were taken at random heights varying from one-third to near full height for each resonance. We were careful to record each pair of symmetrically placed points as quickly as possible, to avoid errors introduced by random fluctuations of resonance height. The average of the three pairs was taken to be the center frequency...

Determination ofg-Factor Ratios for FreeRb85andRbWhite1, Hughes2, Hayne3 et al. 1968Phys. Rev.612241View full textAdd to dashboardCite

“…The gj/gj (Rb 85 ) value is in agreement with the recent atomicbeam value. 5 The low value of magnetic field available in this work coupled with the observation of only AF = 0 Zeeman transitions precluded a severe test of the Breit-Rabi formula. Nevertheless, the agreement between calculated and observed frequencies (for transitions other than those used in gj/gj determinations) was in general ~ 5 parts in 10 9 .…”

Experimental determinations of the hyperfine structure in the alkali atoms

“…The agreement is satisfactory, and this is especially true between the two most accurate measurements. Penselin et al (1962) have also carried out a measurement of gz/gz, that has been found to be in considerable disagreement with later measurements by White et al (1968) and by Ehlers et al (1968). Nevertheless, the hfs measurement by Penselin et al (1962) seems to be reliable.…”

Study of NMR signals of rubidium in aqueous solutions and determination of the magnetic moments of Rb-85 and Rb-87 nuclei

“…Table 4 shows a comparison of the data on determining the rubidium nuclear magnetic moment from the papers [1][2][3][4][5] and the data of the present paper. The results of the papers [1][2][3][4][5] were later included in reference publications [6,7] after being corrected for the nuclear electron screening.…”

“…Data on the rubidium nuclear magnetic moments were previously obtained: by optical excitation method with radio frequency detection [1] (OP/RD method); by the nuclear magnetic resonance method [2] (NMR−H 2 O method); recording of frequency resonance on rubidium nuclei in an atomic beam [3] (method AB/D); methods of laser spectroscopy in atomic beams [4] (method ABLS) and NMR method at dissolution of rubidium in heavy water [5] (method NMR−D 2 O). However, after the required corrections for the electron screening of the nuclei [6,7], it turned out that the data [1][2][3][4][5] diverge in units of the fifth or quadruple sign, and when calculating the nuclear magnetic moment of rubidium isotopes µ( 85 Rb) and µ( 87 Rb) not all components of the uncertainty were correctly taken into account.…”