https://en.wikipedia.org/wiki/Sigma_Orionis wrote:
<<σ Orionis is a naked eye star at the eastern end of Orion's Belt, and has been known since antiquity, but it was not included in Ptolemy's Almagest. It was referred to by Al Sufi, but not formally listed in his catalogue. In more modern times, it was measured by Tycho Brahe and included in his catalogue. In Kepler's extension it is described as "Quae ultimam baltei praecedit ad austr." (preceding the outermost of the belt, to the south). It was then recorded by Johann Bayer in his Uranometria as a single star with the Greek letter σ (sigma). He described it as "in enſe, prima" (in the sword, first). It was also given the Flamsteed designation 48.
In infrared images, a prominent arc is visible centred on σ Ori AB. It is about 50" away from the class O star, around 0.1 parsecs at its distance. It is directed towards IC434, the Horesehead Nebula, in line with the space motion of the star. The appearance is similar to a bowshock, but the type of radiation shows that it is not a bowshock. The observed infrared emission, peaking at around 45 microns, can be modelled by two approximately black-body components, one at 68K and one at 197 K. These are thought to be produced by two different sizes of dust grains.
The material of the arc is theorised to be produced by photoevaporation from the molecular cloud around the Horsehead Nebula. The dust becomes decoupled from the gas that carried it away from the molecular cloud by radiation pressure from the hot stars at the centre of the σ Ori cluster. The dust accumulates into a denser region that is heated and forms the visible infrared shape.
The term "dust wave" is applied when the dust piles up but the gas is largely unaffected, as opposed to a "bow wave" where both dust and gas are stopped. Dust waves occur when the interstellar medium is sufficiently dense and the stellar wind sufficiently weak that the dust stand-off distance is larger than the stand-off distance of a bow shock. This would clearly be more likely for slow-moving stars, but slow-moving luminous stars may not have lifetimes long enough to produce a bow wave. Low luminosity late class O stars should commonly produce bow waves if this model is correct.>>