Raman spectroscopy investigation of the H content of heated hard amorphous carbon layers

Abstract

We revisit here how Raman spectroscopy can be used to estimate the H content in hard hydrogenated amorphous carbon layers. The H content was varied from 2 at.% to 30 at.%, using heat treatments of a a-C:H, from room temperature to 1300 K and was determined independently using ion beam analysis. We examine the correlation of various Raman parameters and the consistency of their thermal evolution with thermo-desorption results. We identify a weak band at 860 cm -1 attributed to H bonded to C(sp 2 ). We show that the H D /H G parameter (Height ratio between the D and G bands) is quasi-linear in the full range of H content and can thus be used to estimate the H content. Conversely, we show that the m/H G parameter (ratio between the photoluminescence background, m, and the height of the G band), often used to estimate the H content, should be used with care, first because it is sensitive to various photoluminescence quenching processes and second because it is not sensitive to H bonded to C(sp 2 ).

Long-term H-release of hard and intermediate between hard and soft amorphous carbon evidenced by in situ Raman microscopy under isothermal heating

“…In the case of heat-treated a-C:H films for which the H content varied from 2 to 30 at. %, we have recently obtained a linear relation between log 10 (m/H G ) and the H content (H/H+C = 25 + 9 log 10 (m/H G )) [4], similarly to what was previously obtained for as-deposited samples for which the H content varied from 20 to 47 at. % [5,6].…”

“…Conversely, figures 2.c and 2.d show that for m/H G and H D /H G this rapid evolution is followed by a slow and roughly linear change (decrease for m/H G and increase for H D /H G ). H D /H G and m/H G are both correlated with the H content [4]. We thus interpret these evolutions as due to the isothermal H release, with a first phase occurring rapidly together with structural changes (within about 100 minutes), followed by a second phase occurring on a longer time scale.…”

“…When the sample is heated, the silicon signature diminishes because the a-C:H becomes more absorbent [11,12]. The photoluminescence background significantly increases, due to the defect passivation (and the quenching of non-radiative relaxation processes [4,6]) before it starts decreasing as hydrogen is released. The G band narrows and slightly blueshifts, as previously observed in [4,13].…”

“…The photoluminescence background significantly increases, due to the defect passivation (and the quenching of non-radiative relaxation processes [4,6]) before it starts decreasing as hydrogen is released. The G band narrows and slightly blueshifts, as previously observed in [4,13]. The D band also narrows and its height increases.…”

“…H D /H G probes the full H content, and its different final values suggest that not all the H bonded to C(sp 2 ) can be released in this temperature range, even with infinite time. Applying the relation between the H content and H D /H G (see [4]), the H contents after heating can be estimated at 12 and 8 % at 500 °C and 600 °C, respectively. At 450°C, 6000 minutes are not enough to reach a plateau.…”