Erratum: Discovery of New Coronal Lines at 2.843 and 2.853 μm (Astrophysical Journal Letters (2018) 856 (L29) DOI: 10.3847/2041-8213/aab434)

In the process of re-aligning the Airborne Infrared Spectrometer (AIR-Spec) for its 2019 eclipse flight, we discovered that internal reflections in the sapphire slit-jaw substrate are producing multiple images of the slit on the detector. We believe that the 2.843 μm line that we identified as the 3p⁵3d ³F₃° → 3p⁵3d ³F₄° transition in Fe IX is not a real detection, but actually a reflection of the Si X 1.430 μm line. The 2.853 μm line stands as a real detection. Its χ-squared goodness of fit parameter indicates that it is likely due to the 3p⁵3d ³F₃° → 3p⁵3d ³F₄° transition in Fe IX. This line was near the AIR-Spec detection limit in 2017, so we will wait for the more sensitive 2019 observations to confirm or reject this detection. If the 2.853 μm line is confirmed as Fe IX, then the 2017 eclipse experiment using AIR-Spec will have observed all five, and only five, of its target lines. The non-detection of additional lines suggests that our initial optimism in finding new M1 lines bright enough for magnetometry, for example with the Daniel K. Inouye Solar Telescope (DKIST), was unwarranted. Figure 1 shows the refraction and reflection at the slit-jaw interfaces responsible for producing multiple images on the detector. This artifact is best observed by removing the dispersive element from the instrument so that the slit forms an image on the detector; the resulting image is shown in Figure 2. The reflections are spaced about 76 pixels apart to the right of the main image, and the first reflection is about 5% as bright as the main image. This is exactly the intensity and distance of the 2.843 μm line relative to the 1.430 μm Si X line. The 5% ratio of the 2.843-1.430 μm lines is consistent throughout the corona (Figure 3), while extreme ultraviolet images show that the spatial structure of emission from Si X is very different from that of Fe IX on the day of the eclipse. There are only a few places where this instrumental artifact shows up in the 2017 AIR-Spec data: Si X 1.430 μm, and three H I lines in the third contact chromospheric spectrum. In each case, only the first reflection is detected. For the other three coronal lines detected by AIR-Spec, all reflections are below the detection limit. For the bright 4 → 3 transition in hydrogen, all reflections are imaged beyond the edge of the detector.