The Spin Structure of the Neutron

Abstract

A description of SLAC experiment E154, a precision measurement of the neutron's longitudinal spin structure function g{sub 1}{sup n}, is presented. Deep inelastic electron scattering was used to measure the structure function in the kinematic range 0.014 < x < 0.7, and 1 < Q{sup 2} < 17 GeV{sup 2}. A measurement of the transverse spin structure function g{sub 2}{sup n} was also made, but with significantly lower statistical precision. Electrons with an average polarization of 82 {+-} 2% and an energy of 48.3 GeV were scattered off polarized {sup 3}He nuclei having an average polarization of 38%. Two independent magnetic spectrometers set at scattering angles of 2.75{sup o} and 5.5{sup o} were used to acquire about 100 million events during a two month run in late 1995. The data were analyzed to yield the integral over the measured region: {integral}{sub 0.014}{sup 0.7} g{sub 1}{sup n}(x)dx = -0.036 {+-} 0.004(stat) {+-} 0.005(syst), which is several standard deviations below the Ellis-Jaffe sum rule predictions. When these data were combined with the proton g{sub 1}{sup p} structure function data from the SMC and E143 experiments, the Bjorken sum rule over the measured x range was found to be within 10% of the predicted value. The integral of the g{sub 2}{sup n} data, dominated by the statistical uncertainty, was found to be {integral}{sub 0.014}{sup 0.7} g{sub 2}{sup n}(x)dx = 0.19 {+-} 0.17(stat) {+-} 0.02(syst), in agreement with the Burkhardt-Cottingham sum rule prediction. The g{sub 1}{sup n} structure function data at low x were found to be inconsistent with the traditional asymptotic forms, bringing into question the methods used in the past.