Answers are grounded in this paper and cite their sources.
We investigate the behavior under Lorentz transformations of perturbative coefficient functions in a collinear twist-3 formalism relevant for high-energy observables including transverse polarization of hadrons. We argue that those perturbative coefficient functions can, a priori, acquire quite different yet Lorentz-invariant forms in various frames. This somewhat surprising difference can be traced back to a general dependence of the perturbative coefficient functions on light cone vectors which are introduced by the twist-3 factorization formulas and which are frame-dependent. One can remove this spurious frame dependence by invoking so-called Lorentz invariance relations (LIRs) between twist-3 parton correlation functions. Some of those relations for twist-3 distribution functions were discussed in the literature before. In this paper we derive the corresponding LIRs for twist-3 fragmentation functions. We explicitly demonstrate that these LIRs remove the light cone vector dependence by considering transverse spin observables in the single-inclusive production of hadrons in lepton-nucleon collisions, N→hX. With the LIRs in hand, we also show that twist-3 observables in general can be written solely in terms of three-parton correlation functions.
No takes yet. Share an insight, caveat, or question.
Kanazawa et al. (2016) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: