*3.1. Raw Material Characterization*

A comprehensive analysis of the raw material is needed to monitor the effect of steam refining. In the first step, the delivered corn stover was fractionated into its different components and existing impurities. The biggest fractions were leaves (44.4%) and stalks (38.6%). In nearly equal amounts, impurities (8.8%), mainly sand and corncobs (7.2%), occur. Further, minor amounts of

corn silks (0.5%) and corn kernels (0.4%) were contained. In contrast to the determined corn stover composition, other studies report varying results. For example, Pordesimo et al. [31] show for corn stover from Tennessee, USA, a composition of 50.9% stalks, 21% leaves, 15.2% corn cobs and 12.9% husks after excluding the grain fraction. Further, they report good accordance with previous studies. However, corn stover is a natural product and its composition depends on the used variety of maize, the environmental conditions, the harvest time of the raw material and the harvesting technology applied [32].

As described before, the main components of corn stover (leaves, stalks and corncobs) represent around 90% of the delivered material. Not surprising, there is only a small amount of corn kernels left after harvesting the corn with a combine harvester. Nevertheless, a significant amount of sand is included as an impurity. For further investigations, the fractions were separated and the steam refining was performed with the leave and stalk fraction.

Hereafter, the chemical compositions of the used raw material and of pure leaves and stalks were analyzed. The results for the carbohydrate distribution, expressed as monomers, the lignin content, ash and the amounts of extracts, are listed in Table 3.


**Table 3.** Chemical composition of the used raw material in % based on raw material.

<sup>1</sup> Mainly analogs to Klason–Lignin [33,34]. <sup>2</sup> Proteins, e.g., from leaves can partly be detected as well in the acid-insoluble residue after hydrolysis.

As mentioned, the main characteristics of the original raw material mix were analyzed by ASE, two-step acidic hydrolysis with the following borate–AEC and determination of the ash content (Table 3). Nonetheless, due to the measuring method, proteins from the leaves may partly be detected in the acid-insoluble hydrolysis residue. Therefore, they are overestimating the detected acid-insoluble lignin content. The insoluble amounts of the raw material might be as well indicating slightly high lignin contents.

However, the found chemical composition is in good accordance with reference values for the different chemical fractions presented in the literature [35–37].
