**1. Introduction**

The use of biomass resources for energy generation has been of considerable importance in recent years [1]. The global increase in energy demand has been one of the main reasons for their use. Added to this situation, there is also a need for dealing with certain problems, such as the depletion of fossil fuel reserves and the increase in environmental pollution from the use of these energy sources [2].

In this context, biomass has the advantage of being the only renewable resource that can be used in solid, liquid and gaseous forms [3]. Furthermore, biomass has the great capacity of producing by-products of high interest, such as catalytic carbons [4] and bioplastics [5]. However, biomass has a number of features that make it difficult to use, including its moisture content, low-energy density and complex structure. Lignocellulosic biomass is made up of a structure that includes mainly cellulose, hemicellulose and lignin [6,7]. The proportions and distribution of these components in the biomass physical structure is complex and depends on the type of species. The knowledge of this composition is very important for its use in different industrial applications.

Up to now, the determination of biomass composition, in terms of hemicellulose, cellulose and lignin contents, has been made by the chemical method. However, it is a slow and expensive methodology, which requires complex techniques and the use of multiple chemical reagents [8]. This means that it is not a suitable method for use in industrial applications.

Thermogravimetric analysis and, especially, the derivative thermogravimetric (DTG) curve is often used for the preliminary study of various thermochemical processes with biomass, since it allows the determination of the different stages of biomass devolatilization. In general, the process of devolatilization of the biomass in the absence of oxygen usually differentiates four stages corresponding to the loss of moisture and the three lignocellulosic components (hemicellulose, cellulose and lignin) [3]. Numerous articles have been published in which the thermal decomposition intervals of these lignocellulosic components are presented based on the deconvolution of the DTG curves [9–15]. It has been observed that, after moisture removal that takes place up to 150 ◦C, the decomposition of the three biomass lignocellulosic components takes place: hemicellulose is the first component to decompose between 200–300 ◦C, followed by cellulose between 250–380 ◦C. Regarding the thermal decomposition of lignin, it is the component with the most complex structure, and its decomposition range is the widest [16], occurring from 200 ◦C up to high temperatures such as 1000 ◦C [17,18].

There are different studies based on determining the lignocellulosic composition by analyzing DTG curves. However, most of these studies are based exclusively on applying deconvolution methods without taking into account their kinetic interpretation of the process [3,19].

On the other hand, kinetic studies on the thermal decomposition of biomass are extensive, in which the use of different kinetic models is analyzed [20,21], providing the kinetic parameters that best fit the experimental data. However, these studies do not focus on finding a method that allows the quantification of the three main lignocellulosic fractions of the biomass.

The use of kinetic analysis to the quantification of the main lignocellulosic fractions allows to include restrictions for a more precise quantification, while a physical interpretation is added to the deconvolution process.

The main objective of this work is to provide a new efficient, low-cost and fast method for the determination of the hemicellulose, cellulose and lignin contents of different types of biomasses, from agricultural by-products to wood. The method is based on applying deconvolution techniques on DTG pyrolysis curves based on a kinetic analysis of the process, and the kinetic model used is based on the assumption that the degradation of each lignocellulosic fraction can be represented by the evolution of a certain number of pseudocomponents.

#### **2. Materials and Methods**

#### *2.1. Biomass Samples*

Five raw materials representing different types of biomass have been selected, including agricultural biomass (wheat straw) and forest biomass, both as softwood barks (spruce bark and pine bark) and hardwoods (poplar and willow).

The pine bark originated from Sweden, while the other biomasses (wheat straw, poplar, spruce bark and willow) came from the South of France.

#### *2.2. Experimental Method*

Each sample was crushed in a mill (Model A 10 basic, IKA-Werke GmbH & Co. KG, Staufen, Germany) and then sieved. The sample sizes were all less than 100 μm in order to minimize the heat transfer resistances and mass transfer diffusion effects.

The TG (thermogravimetric) analysis was performed on a TG-DTA analyzer (Model DTG-60H, SHIMADZU Co. Ltd., Kyoto, Japan). The analyses were carried out using a nitrogen atmosphere with a flow rate of 50 mL min<sup>−</sup>1. The heating rate used was 5 ◦C min<sup>−</sup>1, from room temperature to a final temperature of 1000 ◦C. The sample weight was c.a. 10 mg.

To reduce temperature-related errors, the equipment used was calibrated across the entire temperature range. In addition, the actual sample temperature was used directly to solve the kinetic equations and to calculate the actual sample heating rate [22].

The information obtained in these analyses was the weight loss as the temperature and time of analysis increase (TG curve).
