We report on the search for a CPT and Lorentz invariance violating coupling of the 3 He and 129 Xe nuclear spins (each largely determined by a valence neutron) to background tensor fields which permeate the universe. Our experimental approach is to measure the free precession of nuclear spin polarized 3 He and 129 Xe atoms in a homogeneous magnetic guiding field of about 400 nT using LTC SQUIDs as low-noise magnetic flux detectors. As the laboratory reference frame rotates with respect to distant stars, we look for a sidereal modulation of the Larmor frequencies of the co-located spin samples. As a result we obtain an upper limit on the equatorial component of the background field interacting with the spin of the bound neutronb n ⊥ < 6.7 · 10 −34 GeV (68% C.L.). Our result improves our previous limit (data measured in 2009) by a factor of 30 and the world's best limit by a factor of 5. [4,5] test the isotropy of the interactions of matter itself. Searches for an anomalous spin coupling to a relic background field which permeates the universe have been performed with electron and nuclear spins with increasing sensitivity [6][7][8][9][10][11][12][13][14][15][16][17][18]. The theoretical framework presented by A. Kostelecký and colleagues parametrizes the general treatment of CPT and Lorentz invariance violating (LV) effects in a Standard Model Extension (SME) [19][20][21]. The SME was conceived to facilitate experimental investigations of Lorentz and CPT symmetry, given the theoretical motivation for violation of these symmetries. Although Lorentz-breaking interactions are motivated by models such as string theory [21,22], loop quantum gravity [23][24][25][26], etc., the low-energy effective action appearing in the SME is independent of the underlying theory. The SME contains a number of possible terms that couple to the spins of fundamental Standard Model particles like the electron, or composite particles like the proton and (bound) neutron. These terms are small due to Planckscale suppression (M p ), and in principle are measurable in experiments by detecting tiny energy shifts of order ∆E (n) ∼ ( mw Mp ) n · m w , where the low energy scale is set by the mass m w of the particle. Since n = 1 is largely ruled out by present experiments [27], tuning the measurement sensitivity to second order effects (n = 2) in Planck scale suppression is the current challenge 1 . To de- * Corresponding author: allmendinger@physi.uni-heidelberg.de 1 For the neutron (mn = 939 MeV) this is ∆E (2) ≈ 10 −38 GeV.termine the leading-order effects of a LV potential V , it suffices to use a non-relativistic description for the particles involved given bywhich can be interpreted as a coupling of the electron, proton or neutron spin σ w J to a hypothetical background fieldb w J . The most sensitive tests so far were performed on the bound neutron using a 3 He-129 Xe Zeeman maser [12, 13], a 3 He-129 Xe co-magnetometer [28] based on free spin precession, and a K-3 He co-magnetometer [7]. The latter one so far gave the highest energy resol...