Dataset of Chilean Oak micropyrolysis over Zn and Ga supported on natural zeolite catalyst in oxygen-depleted (He) and reductive (H₂) atmospheres

Abstract

Biomass pyrolysis is a promising method for recovering value from waste; however, the high oxygen content limits the stability of the product. While pretreatment, acid-metal catalysts, and reducing atmospheres can reduce oxygenates and promote the formation of hydrocarbons, it remains unclear how torrefaction affects condensable gases during pyrolysis when using natural Chilean zeolite doped with Zn and Ga. Accordingly, we provide this dataset obtained from analytical pyrolysis (Py-GC/MS) experiments in a non-oxidizing and reactive atmosphere over torrefied Chilean Oak using natural Chilean zeolite catalysts loaded with Zn or Ga at 2 wt.% and 5 wt.%. Py-GC/MS analysis provides comprehensive information that can be correlated with the activation energy values obtained by thermogravimetric analysis (TGA) and the initial properties of raw materials, as determined by elemental, proximate, and fibre analysis of the biomass. In a Py-GC/MS experiment, a small biomass sample was put in a reactor and then pyrolyzed. An online gas chromatography system linked to mass spectrometry is then used to identify the reaction products and their abundance. Similarly, given the physicochemical differences generated in the catalysts after metal loading, characterization data such as N₂ physisorption, ammonia temperature-programmed desorption (TPD-NH₃), X-ray diffraction (XRD), transmission electron microscopy (TEM), scanning electron microscopy with X-ray energy dispersive detection (SEM-EDX), X-ray photoelectron spectroscopy (XPS), Pyridine adsorption-desorption followed by Fourier transform infrared spectroscopy (Pyr-FTIR) are provided. The metallic percentages were confirmed using inductively coupled plasma atomic emission spectrometry analysis (ICP). Finally, the data used to create this dataset is freely available at the Mendeley data repository. The information in this dataset includes input characteristics of raw materials, experimental conditions, observed behaviors during the process, and various response variables. It can be extremely valuable for multidimensional analyses as well as machine or deep learning applications. It allows for the prediction of process behaviors based on reaction mechanisms and facilitates the extrapolation of results to industrial scales. Additionally, with the known properties of the products, it becomes possible to assess their dynamic and thermal flow behaviors, thereby closing the gaps for potential real-world applications.

Specifications Table

Subject	Engineering & Materials science
Specific subject area	Thermochemical and catalytic transformation of woody residues into platform compounds.
Type of data	Table, Image, Figures, cured data from Py-GC/MS
Data collection	Thermogravimetric experiments were conducted in a PerkinElmer thermobalance model STA 6000 (323 K to 1073 K at 5, 10, 20, 40 K·min−1), using 10 – 15 mg of samples. Pyrolysis experiments were performed in a CDS Pyroprobe 5200HPR pyrolizer (CDS Analytical) coupled to a Perkin Elmer Clarus 690 chromatograph, connected to a Perkin Elmer Clarus SQ-8T MS Detector. Micropyrolysis experiments were performed for all biomass at 673 K, 723 K, 773 K, 823 K under 150 psi, 50 mL min−1 of non-oxidizing (He) or reductive (H2) gas flow of sample.
Data source location	Laboratory of Gas Chromatography and Analytical Pyrolysis, Wood Engineering Department, Universidad del Bío-Bío, Concepción 4030,000, Chile
Data accessibility	Repository name: Dataset of Chilean Oak micropyrolysis over Zn and Ga supported on natural zeolite catalyst in oxygen-depleted (He) and reductive (H₂) atmospheres. Data identification number: 10.17632/gkhjh4v8tg.2 Direct URL to data: https://data.mendeley.com/datasets/gkhjh4v8tg/2
Related research article	K.J. Fernández-Andrade, B.F. Rivadeneira-Mendoza, J.M. Rodríguez-Díaz, S. Alejandro-Martin, Enhanced Monoaromatic Formation via Hydropyrolysis of Torrefied Chilean Oak over Metal (Ga, Zn) Supported on Modified Natural Zeolite, Energy Fuels 39 (2025) 14,833–14,849. 10.1021/acs.energyfuels.5c02905.

1. Value of the Data

• •This dataset helps understand the compositional behaviour of condensable gases obtained from pyrolysis in both non-oxidizing and reductive atmospheres, using natural zeolite-based catalysts.
• •The catalytic and non-catalytic micropyrolysis experiments on torrefied and non-torrefied biomass provide relevant information on the thermal decomposition pathways of biomass. They also provide information on the possible involvement of the active sites of Chilean natural zeolite and the metals supporting it, without losing data due to the rapid consumption of unstable compounds.
• •Researchers interested in describing pyrolysis reaction mechanisms through data mining, computational chemistry, machine learning, or deep learning, or those associated with industrial simulation, can take advantage of these data.
• •The dataset provides detailed analytical pyrolysis tests, along with the respective characterisations of the biomass and catalysts. This data can be subjected to multivariate statistical tests to establish deep correlations in the process of behaviour.
• •Through the biomass compositions provided, it is possible to determine the interactions between the compounds formed during pyrolysis. This information allows us to establish the possible reaction mechanism governing the chemical reactions, both for the decomposition of biomass and for the formation of value-added chemical compounds.
• •The dataset includes >130 experimental runs, with processed data comprising compound identification, quantification (percentage area), compound family grouping, and atomic balance (carbon, hydrogen, oxygen, and nitrogen) for various pyrolysis, hydropyrolysis, co-pyrolysis, and co-hydropyrolysis conditions.

2. Background

Biomass is one of the most abundant carbonaceous materials in the world; Chile alone generates over 2 million tons per year [1], with 13 % of the residue coming from thinning for native forest management [[2], [3]]. Pyrolysis is an alternative to the utilization of this waste, but it has important disadvantages, including the low chemical stability of the liquid product due to the high presence of oxygenated compounds. Strategies have been employed to remove oxygen-containing compounds from the liquid fraction, such as using metal-added acid catalysts in reductive environments [4]. However, synthetic acid catalysts present problems with reproducibility in the synthesis, which limits their use at the industrial level and their economic feasibility [5]. Alternatively, this dataset provides detailed information on different Py-GC/MS experiments of torrefied and non-torrefied biomass using Chilean natural zeolite-based catalysts [6]. It also presents the detailed composition of the condensable gases obtained in each pyrolysis (He) and hydropyrolysis (H₂) experiment. To gain an in-depth understanding of the characteristics of biomass and its thermochemical decomposition, information is provided on its proximate, elemental, and fibre composition, as well as FTIR analysis and thermogravimetry (TGA) experiments conducted at varying heating rates. Furthermore, a detailed characterisation of the catalysts is provided, which sheds light on their catalytic properties, structures, and crystalline conformations.

Machine learning, kinetic fitting, and reaction network reconstruction can directly use data organized on multiple scales, adding scientific value beyond comparative analysis. Combining biomass properties (proximate, elemental, and fibre analysis), catalyst properties (total acidity, Brønsted/Lewis strength, metallic state, crystalline phases, particle size, surface area, pore volume, and metal percentage), TGA kinetics, and pyrolysis/hydro-pyrolysis results enables predictive models (regression/classification). The use of isoconversional approaches or global models to change apparent kinetic parameters and compare how torrefaction severity affects them is supported by TGA curves at different speeds. The implicit factorial design of the study of five biomasses (one dry and four torrefied at different severity indices) and six catalysts (NZ, H2NHZ, Zn, and Ga at 2 and 5 wt. % supported on H2NHZ) generates enough contrasts to infer and validate transformation networks (lumping by families or species) under inert and hydrogenating atmospheres, providing a quantitative framework for linking roasting-induced chemical changes and catalytic functionality.

The data presented here is related to research published in parts recently. The first part explains how torrefaction generates chemical changes in biomass and how these are reflected in its pyrolysis [7]. The second part explains how the distribution of compounds produced by pyrolysis of torrefied biomass can be changed using catalysts and what properties of these catalysts enable this change [8]. After torrefaction, biomass forms lignin-carbohydrate complexes. These complexes promote the production of phenols, which, thanks to Lewis acid sites in small spaces, can be converted into monoaromatic hydrocarbons. This article provides the raw data that supports the conclusions of the published studies, allowing readers to reprocess the data, verify our findings, draw their own conclusions, or simulate the process on a real scale. Unlike the other articles, this one presents the methodology in much greater detail, making it easier to reproduce the experiments and experimental data.

3. Data Description

To facilitate understanding of the terminology used in the article and in the dataset, Table 1 presents a brief glossary of terms with their respective definitions.

Table 1.

Glossary of terms.

BAS	Brønsted Acid Sites
DMPY	2,6-Dimethylpyridine
DRIFT	Diffuse Reflectance Fourier-Transform Infrared Spectroscopy
DTG	Derivative Thermogravimetry
EDX	Energy Dispersive X-ray
FTIR	Fourier-Transform Infrared Spectroscopy
Ga2-H2NHZ	2 wt. % of Gallium Supported on H2NHZ
Ga5-H2NHZ	5 wt. % of Gallium Supported on H2NHZ
H/Ceff	Effective Hydrogen-Carbon ratio
H2NHZ	Natural Protonated Zeolite
HC	Hydrocarbon
Hpc	Catalytic hydropyrolysis
Hyp	Hydropyrolysis
ICP-OES	Inductively Coupled Plasma-Atomic Optic Emission Spectrometer
LAS	Lewis Acid Sites
NIST	National Institute of Standards and Technology
NZ	Natural Zeolite
Non-cat	Non-catalytic
Oak	Dried Biomass (353 K, 24 h)
Py-GC/MS	Analytical MicroPyrolysis
Py	Conventional Pyrolysis
Pyc	Catalytic pyrolysis
Pyr	Pyridine
SEM	Scanning Electron Microscope
T1	Heat rate 1 (1000 K·s−1)
T2	Heat rate 2 (10,000 K·s−1)
TEM	Transmission Electron Microscope
TG	Thermogravimetry
TGA	Thermogravimetric Analysis
Torr1	Torrefied Biomass (523 K, 30 min)
Torr2	Torrefied Biomass (573 K, 15 min)
Torr3	Torrefied Biomass (573 K, 30 min)
Torr4	Torrefied Biomass (523 K, 15 min)
TPD-NH3	Ammonia Temperature-Programmed Desorption
XPS	X-ray Photoelectron Spectrometer
XRD	X-ray Diffraction
Zn2-H2NHZ	2 wt. % of Zinc Supported on H2NHZ
Zn5-H2NHZ	5 wt. % of Zinc Supported on H2NHZ

The dataset presents a comprehensive physicochemical characterization of both torrefied and non-torrefied biomass samples, including proximate analysis, elemental analysis, fibre analysis, thermogravimetric analysis, and FTIR analysis. In addition, it includes a detailed characterization of the catalysts prepared by subjecting pristine zeolite to sequential ion exchange treatments, initially removing the counter cations and subsequently loading zinc and gallium at a ratio of 2 wt. % and 5 wt. %. Thermogravimetric analysis (TGA) data at four heating rates (5, 10, 20, and 40 K·min⁻¹) are also included, which were used to determine the kinetic and thermodynamic behaviour of the decomposition of torrefied and non-torrefied biomass. The TGA data are accompanied by a series of analytical pyrolysis experiments that enable inferences to be made about the decomposition events associated with the formation of specific compounds or families of compounds. The analytical pyrolysis experiments of torrefied and non-torrefied biomass include non-catalytic pyrolysis at 673 K, 723 K, 773 K, and 823 K; non-catalytic hydropyrolysis at 823 K; catalytic pyrolysis at 823 K; catalytic hydropyrolysis at 823 K; and catalytic hydropyrolysis of Torr3 at 823 K at different biomass/catalyst ratios (1/3, 1/5, 1/7, 1/10, and 1/13). The data is freely accessible from the Mendeley Data repository by following the link provided in the Data Accessibility section.

The repository contains detailed information about the data provided, its organization, paths, and more. Upon entering the repository, three folders can be seen; two of them contain the database in two different formats, XLS and CSV (folders named "XLS files" and "CSV files", respectively), and another contains the descriptive files of the database for machine learning, described as a machine-readable metadata file. These files are metadata_files.csv, which contains the names of the files and conditions for each experiment; data_dictionary.csv, which includes the column headers and units for each report; and compounds.csv, which contains all the identified compounds with their respective formulas and synonyms. Fig. 1 shows the hierarchical organization within the folders containing the data (the organization is the same for XLS files and CSV files) to facilitate navigation within the dataset. The dataset also includes different forms of visualization, which can be accessed through the preview provided by the Mendeley Data tool or by downloading the entire repository, which contains accessible data files.

Fig. 1.

Organization scheme for the data contained in the Mendeley Data repository.

Folder 1. This folder contains all the information related to the characterization of both torrefied and non-torrefied biomass samples. Here, data from proximal, ultimate, fibre, and FTIR analyses are presented in data tables. More information about the findings in this data was presented in a paper previously published by this research group [7]. The main conclusions derived from this dataset indicated that torrefaction selectively reduces the hemicellulose content in biomass while increasing insoluble lignin through lignin-carbohydrate complexes, which reduces acid formation during pyrolysis.

Folder 2. Data on catalyst characterization includes SEM and TEM images in PNG format, chemical element mapping, elemental composition, and metal particle size distribution. The dataset also has N₂ physisorption graphs, TPD-NH₃ profiles, FTIR spectra related to pyridine bonding, XRD patterns, and XPS spectra. The article associated with this dataset has more details [8].

Folder 3. This folder contains the TGA profiles of the non-torrefied sample (Chilean Oak), and the samples torrefied under four different conditions, which are Torr1 (523 K – 30 min), Torr2 (573 K – 15 min), Torr3 (573 K – 30 min), and Torr4 (523 K – 30 min). With this data, it was possible to calculate characteristic parameters of biomass thermal decomposition, such as conversion and differential conversion concerning time and temperature (Fig. 2). These parameters allow the calculation of activation energy, kinetic constant, Gibbs free energy variation, enthalpy variation, and entropy variation at each degree of conversion. These parameters enabled the kinetic and thermodynamic description of the thermochemical decomposition of biomass. The other publication associated with this dataset contains more details [7].

Fig. 2.

A) Conversion and B) dα/dT profile obtained from TGA of non-torrefied biomass (Oak).

Folder 4. This folder contains data related to non-catalytic pyrolysis tests in either a non-oxidizing atmosphere (He) or a reductive atmosphere (hydropyrolysis, H₂). The pyrolysis experiments in He were conducted at two heating rates (1000 and 10,000 K·s⁻¹) and at four temperatures (673, 723, 773, and 823 K) for all biomass samples and are presented in folder 4.1. The folder includes a document called “Chemical compounds grouped by families”, which contains a summary of the distribution of compounds by families ( % area) obtained from the experiments in that folder (Table 2). This document also includes the overall H/C_eff, H/C, and O/C molar ratios for each experiment. The same folder also contains the individual documents for each experiment (Table 3), which presents the list of identified compounds with their respective abundance (area and % area), as well as the elemental balance (C, H, O, N). Compounds were identified by comparing mass spectra using TurboMass 2.0 software. The library employed was sourced from the National Institute of Standards and Technology (NIST 14) for mass spectrometry, with compounds selected based on a match score of 750 or higher. The percentage area calculation is performed based on the assumption that each compound has a response factor of 1. Consequently, the results ought to be analysed in terms of families rather than as isolated compounds. The dataset presents the areas of each compound, enabling readers to apply any appropriate response factor they believe to be suitable. In individual experiment reports, compounds classified as permanent or non-condensable gases (CO, CO₂, hydrocarbons <C5, NOx, SOx), high molecular weight compounds (>C20), or water will not be included. This implication results in an incomplete mass balance; therefore, complementary data, such as TGA data, must be incorporated to achieve a complete balance. The names of the individual files contain relevant information about the conditions under which each experiment was conducted, and they follow the following formula:

$$\text{KF}10\_\text{sam}\_\text{py}450\_\mathrm{t}1\_\text{nocat}\_150\_\text{Oak}\_\text{Torr}4$$

Table 2.

Distribution of compounds grouped by families obtained from non-catalytic pyrolysis of biomass. Heating rate: 1000 K·s⁻¹, pressure: 150 psi, gas: He, pyrolysis time: 15 s.

A	B	C	D	E	F	G	H	I	J	K	L	M	N	O	P	Q	R	S	T
KF1	673	Oak	43.8	0.4	10.3	0.0	4.0	0.6	1.1	13.4	0.0	0.0	16.8	2.4	0.0	7.2	100	0.23	0.65
KF2	673	Oak_Torr1	36.2	0.4	10.7	0.0	3.7	0.9	0.8	14.4	0.0	0.0	20.9	3.8	0.0	8.2	100	0.28	0.62
KF3	673	Oak_Torr2	36.5	0.7	12.4	0.0	3.2	0.6	1.4	15.3	0.0	0.0	19.0	2.4	0.0	8.4	100	0.29	0.63
KF4	673	Oak_Torr3	24.2	0.8	14.6	0.0	2.3	1.3	0.8	17.3	0.0	0.2	25.1	3.4	0.0	9.9	100	0.34	0.57
KF5	673	Oak_Torr4	41.8	0.0	11.5	0.0	2.9	0.6	0.0	14.6	0.0	0.0	17.0	1.8	0.0	9.9	100	0.21	0.67
KF6	723	Oak	35.2	0.8	12.0	0.0	2.1	0.3	0.9	13.1	0.0	0.2	23.2	3.2	0.0	9.0	100	0.31	0.61
KF7	723	Oak_Torr1	36.7	1.0	11.8	0.0	1.6	0.7	0.8	12.8	0.0	1.3	22.6	2.9	0.0	7.9	100	0.31	0.62
KF8	723	Oak_Torr2	27.2	0.0	16.2	0.0	1.5	0.4	0.4	14.1	0.0	0.3	28.9	4.5	0.0	6.6	100	0.31	0.59
KF9	723	Oak_Torr3	22.5	0.8	17.1	0.0	2.2	0.5	0.2	16.5	0.0	0.0	28.7	5.4	0.0	6.2	100	0.34	0.55
KF10	723	Oak_Torr4	36.3	1.4	14.8	0.0	3.8	0.0	0.6	12.4	0.0	0.0	22.2	3.2	0.0	5.3	100	0.27	0.63
KF11	773	Oak	30.5	1.6	15.5	0.0	3.5	0.2	0.4	12.2	0.0	0.5	24.8	4.5	0.0	6.4	100	0.33	0.59
KF12	773	Oak_Torr1	30.7	0.9	14.1	0.0	3.2	0.2	1.0	12.3	0.0	0.5	23.0	3.8	0.0	10.3	100	0.32	0.59
KF13	773	Oak_Torr2	26.1	1.0	17.3	0.0	3.0	0.3	0.3	11.3	0.0	0.5	24.4	7.6	0.0	8.2	100	0.36	0.56
KF14	773	Oak_Torr3	18.7	1.9	16.6	0.0	2.9	0.5	0.0	13.2	0.0	2.1	31.2	5.7	0.0	7.2	100	0.42	0.51
KF15	773	Oak_Torr4	28.1	2.0	17.6	0.0	3.9	0.2	0.5	11.5	0.0	0.0	26.3	3.7	0.0	6.2	100	0.34	0.59
KF16	823	Oak	23.3	5.6	20.1	0.0	0.7	11.9	2.2	6.8	0.0	2.4	6.4	14.9	0.0	5.7	100	0.59	0.48
KF17	823	Oak_Torr1	23.5	3.5	13.4	0.0	9.0	4.2	0.4	2.5	0.0	0.0	8.0	26.0	0.0	9.4	100	0.52	0.45
KF18	823	Oak_Torr2	11.6	5.6	15.0	0.0	1.5	3.1	2.0	4.1	0.0	0.4	9.6	39.1	0.0	7.9	100	0.57	0.40
KF19	823	Oak_Torr3	4.9	3.9	10.4	0.0	4.2	7.2	4.7	5.6	0.0	0.0	19.1	34.5	0.0	5.8	100	0.68	0.38
KF20	823	Oak_Torr4	19.0	5.0	14.3	0.0	12.1	4.6	0.2	5.4	0.0	0.3	11.4	17.9	0.0	9.7	100	0.42	0.49

A: Experiment, B: Temperature ( °C), C: Biomass Type, D: Acids ( % area), E: Alcohols ( % area), F: Aldehydes ( % area), G: Amides ( % area), H: Amines ( % area), I: Esters ( % area), J: Ethers ( % area), K: Furans ( % area), L: HC_ Aromatics ( % area), M: Hydrocarbons ( % area), N: Ketones ( % area), O: Phenols ( % area), P: Sugars ( % area), Q: Unknown ( % area), R: Total ( % area), S: H/Ceff, T: O/C.

Table 3.

Compounds identified for individual biomass catalytic hydropyrolysis experiments. Catalyst: Zn2-H2NHZ, biomass/catalyst ratio: 1/13, biomass: Oak_Torr3, temperature: 823 K, heating rate: 1000 K·s⁻¹, pressure: 150 psi, pyrolysis time: 15 s.

Retention Time [min]	Compound	Peak Area [u.a]	Relative Area [ %]	Formula	%C	%H	%O	%N
1.973	Carbon dioxide	2013,985	0.23	CO2	0.08	0.00	0.15	0.00
2.294	2-Butene, (E)-	1888,848	0.21	C4H8	0.07	0.14	0.00	0.00
2.535	Cyclopropane, 1,2-dimethyl-, trans-	25,703,068	2.91	C5H10	0.97	1.94	0.00	0.00
2.569	Furan	27,645,596	3.13	C4H4O	1.39	1.39	0.35	0.00
2.713	Acetone	65,922,708	7.47	C3H6O	2.24	4.48	0.75	0.00
3.048	Furan, 2-methyl-	39,314,784	4.46	C5H6O	1.86	2.23	0.37	0.00
3.334	Methyl glyoxal	57,606,524	6.53	C3H4O2	2.18	2.90	1.45	0.00
3.615	Benzene	22,931,820	2.60	C6H6	1.30	1.30	0.00	0.00
4.017	Furan, 2,5-dimethyl-	12,186,385	1.38	C6H8O	0.55	0.74	0.09	0.00
4.165	3-Penten-2-one	12,378,549	1.40	C5H8O	0.50	0.80	0.10	0.00
4.304	Acetic acid	31,153,472	3.53	C2H4O2	0.88	1.77	0.88	0.00
4.633	2,3-Pentanedione	5871,166	0.67	C5H8O2	0.22	0.35	0.09	0.00
4.996	2-Propanone, 1‑hydroxy-	17,371,986	1.97	C3H6O2	0.54	1.07	0.36	0.00
5.186	Toluene	37,453,696	4.24	C7H8	1.98	2.26	0.00	0.00
5.625	Hepten-2-yl angelate, 6-methyl-5-	1842,027	0.21	C13H22O2	0.07	0.12	0.01	0.00
5.812	3-Penten-2-one	4164,639	0.47	C5H8O	0.17	0.27	0.03	0.00
6.308	Propanoic acid	3329,951	0.38	C3H6O2	0.10	0.21	0.07	0.00
7.179	Cyclopentanone	4245,632	0.48	C5H8O	0.17	0.27	0.03	0.00
7.448	o-Xylene	12,589,821	1.43	C8H10	0.63	0.79	0.00	0.00
7.661	m-Xylene	17,219,182	1.95	C8H10	0.87	1.08	0.00	0.00
8.287	3-Furfural	1984,569	0.22	C5H4O2	0.10	0.08	0.04	0.00
8.509	o-Xylene	6170,522	0.70	C8H10	0.31	0.39	0.00	0.00
8.713	Acetyl ether	6735,165	0.76	C4H6O3	0.23	0.35	0.18	0.00
9.13	Furfural	34,626,880	3.92	C5H4O2	1.78	1.43	0.71	0.00
10.516	Benzene, (1-methylethyl)-	4507,817	0.51	C9H12	0.22	0.29	0.00	0.00
10.902	2,3-Butanedione	3077,074	0.35	C4H6O2	0.12	0.17	0.06	0.00
10.953	2-Cyclopenten-1-one, 2-methyl-	5500,892	0.62	C6H8O	0.25	0.33	0.04	0.00
11.457	Acetylfuran	3103,115	0.35	C6H6O2	0.15	0.15	0.05	0.00
11.73	Benzene, 1,3,5-trimethyl-	4993,038	0.57	C9H12	0.24	0.32	0.00	0.00
12.711	2-Cyclopenten-1-one, 2‑hydroxy-	2281,979	0.26	C5H6O2	0.10	0.12	0.04	0.00
12.903	Benzofuran	3831,453	0.43	C8H6O	0.23	0.17	0.03	0.00
13.765	Furfural, 5-methyl-	5268,573	0.60	C6H6O2	0.26	0.26	0.09	0.00
13.978	Propanoic acid, ethenyl ester	2474,297	0.28	C5H8O2	0.09	0.15	0.04	0.00
14.074	Unknown	2705,444	0.31	——–	0.00	0.00	0.00	0.00
14.236	2-Cyclopenten-1-one, 3-methyl-	3947,005	0.45	C6H8O	0.18	0.24	0.03	0.00
14.284	Benzene, 1-ethynyl-4-methyl-	1800,479	0.20	C9H8	0.11	0.10	0.00	0.00
14.386	Butanoic acid, 4‑hydroxy-	2385,196	0.27	C4H8O3	0.07	0.14	0.05	0.00
14.71	2(5H)-Furanone	3315,886	0.38	C4H4O2	0.15	0.15	0.08	0.00
16.201	3-Methylcyclopentane-1,2-dione	13,004,299	1.47	C6H8O2	0.55	0.74	0.18	0.00
16.312	2(5H)-Furanone, 3-methyl-	2784,805	0.32	C5H6O2	0.12	0.15	0.05	0.00
16.748	2-Furanone, 2,5-dihydro-3,5-dimethyl	2293,593	0.26	C6H8O2	0.10	0.13	0.03	0.00
16.839	Benzofuran, 2-methyl-	3098,562	0.35	C9H8O	0.18	0.16	0.02	0.00
17.534	Phenol	54,472,728	6.17	C6H6O	2.85	2.85	0.47	0.00
17.959	Phenol, 2‑methoxy-	25,364,774	2.87	C7H8O2	1.18	1.35	0.34	0.00
19.209	Phenol, 2-methyl-	27,613,302	3.13	C7H8O	1.37	1.56	0.20	0.00
19.652	Phenol, 3,4-dimethyl-	1961,616	0.22	C8H10O	0.09	0.12	0.01	0.00
19.839	Naphthalene	6202,118	0.70	C10H8	0.39	0.31	0.00	0.00
20.131	Unknown	1840,713	0.21	——–	0.00	0.00	0.00	0.00
20.42	p-Cresol	15,358,391	1.74	C7H8O	0.76	0.87	0.11	0.00
20.494	Phenol, 3-methyl-	41,012,256	4.65	C7H8O	2.03	2.32	0.29	0.00
21.14	Levoglucosenone	3642,057	0.41	C6H6O3	0.17	0.17	0.08	0.00
21.503	Creosol	16,733,741	1.90	C8H10O2	0.76	0.95	0.19	0.00
22.016	Phenol, 3,4-dimethyl-	13,536,221	1.53	C8H10O	0.65	0.81	0.08	0.00
23.193	Phenol, 2,4-dimethyl-	2515,055	0.29	C8H10O	0.12	0.15	0.02	0.00
23.261	Phenol, 3,5-dimethyl-	3088,802	0.35	C8H10O	0.15	0.18	0.02	0.00
23.44	Phenol, 2-ethyl-	2102,918	0.24	C8H10O	0.10	0.13	0.01	0.00
23.559	Naphthalene, 2-methyl-	8982,677	1.02	C11H10	0.53	0.48	0.00	0.00
24.143	Naphthalene, 2-methyl-	7703,187	0.87	C11H10	0.46	0.42	0.00	0.00
24.296	Phenol, 4-ethyl-2‑methoxy-	3992,069	0.45	C9H12O2	0.18	0.24	0.04	0.00
27.798	Phenol, 2,6-dimethoxy-	47,333,132	5.36	C8H10O3	2.04	2.55	0.77	0.00
30.619	3,5-Dimethoxy-4-hydroxytoluene	29,306,878	3.32	C9H12O3	1.25	1.66	0.42	0.00
32.794	5‑tert-Butylpyrogallol	3823,645	0.43	C10H14O3	0.16	0.22	0.05	0.00
——	Unknown	43,004,221	4.87	——–	0.00	0.00	0.00	0.00

%C, %H, %O, %N represents the Relative elemental contribution weighted by abundance ( %).

Where “KF” and “sam” are internals laboratory codes, “10″ is the experiment identification number, “py” indicates that the pyrolysis is non-catalytic and was performed in He (“hyp” for hydropyrolysis), “450″ is the temperature in degrees Celsius, “t1” indicates that the heating rate was 1000 K·s⁻¹ (“t2” for 10,000 K·s⁻¹), “150″ indicates the pressure in psi at which the experiment was performed, and “Oak_Torr4” indicates the biomass sample used (for non-torrefied biomass, only “Oak”). Based on the results in He, the non-catalytic hydropyrolysis experiments were performed only at 823 K and are found in folder 4.2. Fig. 3 shows the most representative chromatograms of the catalytic and non-catalytic pyrolysis and hydrothermal pyrolysis of Torr3 torrefied biomass.

Fig. 3.

Typical chromatograms for Py-GC/MS experiments under different reaction conditions.

Folder 5. This folder includes catalytic pyrolysis and hydropyrolysis experiments, which are in folders 5.1 and 5.2, respectively. In both cases, the experiments were carried out at 823 K and 1000 K s⁻¹; however, in the case of pyrolysis, only the 1/7 biomass/catalyst ratio (also presented as 1–7) was tested. Meanwhile, in the catalytic hydropyrolysis experiments, different biomass/catalyst ratios were tested (1/3, 1/5, 1/7, 1/10, and 1/13) and grouped in folders 5.2.1 to 5.2.5, named with the corresponding ratio. The names of the individual files follow a similar formula to the documents contained in folder 4, with minor variations shown below:

$$\text{KF}103\_\text{sam}\_\text{hpc}550\_\mathrm{t}1\_\text{Zn}2-\mathrm{H}2\text{NHZ}\_1-7\_150\_\text{Oak}\_\text{Torr}3$$

Where “hpc” is the type of pyrolysis (catalytic hydropyrolysis, “pyc” for catalytic pyrolysis), “Zn2-H2NHZ” is the catalyst used (NZ, H2NHZ, Zn2-H2NHZ, Zn5-H2NHZ, Ga2-H2NHZ, Ga5-H2NHZ). Finally, “1–7” represents the biomass/catalyst ratio used in the experiment (1–3, 1–5, 1–7, 1–10, 1–13).

4. Experimental Design, Materials and Methods

4.1. Raw materials pre-treatment and characterization

Chilean Oak (Nothofagus obliqua) was the biomass selected. First, using a rotary knife cutter and sieves, the particle size was reduced to a fraction between 75 and 125 μm. Then, four different torrefaction treatments modified the biomass properties by changing the temperature (523 K and 573 K) and time (15 and 30 min). The torrefied biomass was labelled as follows: Oak_Torr1 (523 K, 30 min), Oak_Torr2 (573 K, 15 min), Oak_Torr3 (573 K, 30 min), and Oak_Torr4 (523 K, 15 min). The nitrogen flow rate (100 mL min⁻¹) and heating rate (10 K·min⁻¹) were constant throughout the experiments. The torrefaction temperature and time were determined by analysing the DTG and D2TG curves of Oak decomposition, identifying the onset and endpoint of hemicellulose decomposition, which represents the initial event of moisture-free mass loss. For more details on biomass pretreatment and characterization, please refer to the publication associated with this dataset [7].

All biomass samples were characterized by proximate and thermogravimetric analysis (PerkinElmer thermobalance model STA 6000), elemental analysis (Thermo Scientific Flash Smart elemental analyser), and higher heating value (Parr Model Parr 6050 calorimeter). Proximal analysis was performed following the methodology previously described by Azocar et al. [9]. For thermogravimetric analysis, 10–15 mg of biomass was placed in an alumina crucible. The sample was heated from room temperature to 323 K at a heating rate of 30 K·min⁻¹ in a nitrogen flow of 20 mL min⁻¹. The temperature was maintained for 20 min, and subsequently the sample was heated to 1073 K at different heating rates (5, 10, 20, and 40 K·min⁻¹).

Fiber analysis followed a methodology previously reported by Aguayo et al. (2010) and Mendonça et al. (2008) [10,11]. This process involves isolating carbohydrates and soluble lignin from insoluble lignin through acid thermohydrolysis using H₂SO₄. The resulting solid was characterised as insoluble lignin, while the supernatant liquid was subjected to two complementary techniques. First, it was taken to a spectrophotometer to measure its absorbance at 250 nm and determine the soluble lignin content. Then, it was subjected to high-performance liquid chromatography analysis with an infrared detector to determine the sugars present, associated with the hemicellulose and cellulose content. The article about the dataset provides additional details [7].

4.2. Catalyst preparation and characterization

Chilean natural zeolite (NZ) obtained from the Lloimavida deposit N° 9145, located in the Maule Region, Chile (−36.30921, −71.60838 WGS84 decimal degrees) was the parent material. A ball mill with tungsten spheres reduced the zeolite particle size to 75–106 µm. Washing with demineralized water (pH 6.1) using a metal sieve (No. 200) removed the fine fraction (<75 μm) remaining in the material. After, NZ dried at 398 K for 24 h. Following a reported methodology [12], an ion exchange with 1 M ammonium sulphate solution (pH 5.3) at 363 K, at a ratio of 1:10 grammes of solid to mL of solution, for 2 h eliminated the NZ compensating cations (Fig. 4). The resulting NZ was washed with plenty of demineralized water for 4 h at 363 K with water changes at 2 h. After the first exchange with ammonium sulfate and washing, the procedure was repeated once more to obtain double-exchanged zeolite. The remaining ammonia was then removed using a degassing process at a flow rate of 100 mL min⁻¹ of N₂ and a temperature of 623 K for 1 hour, resulting in a surface of NZ loaded with H⁺ groups (H2NHZ). Finally, an ion exchange with zinc nitrate hexahydrate (Zn(NO₃)₂·6H₂O) and gallium nitrate hydrate (Ga(NO₃)₃·xH₂O) at 363 K loaded the metals onto the surface of the catalyst (Zn and Ga, respectively). To prevent metal precipitation, nitric acid was used to adjust the pH of the solutions to 5 for Zn and 3 for Ga. The resulting samples were called MX-H2NHZ, where M is the metal (Zn or Ga) and X the percentage of metallic load (2 wt. % and 5 wt. %). The samples were dried at 333 K for 48 h and subsequently calcined in a muffle furnace at 823 K for 4 h with a heating rate of 1 K·min⁻¹ in the presence of atmospheric air [8].

Fig. 4.

Representation of the catalysts preparation process.

The catalysts were characterised by N₂ physisorption in Micromeritics Gemini VII series 2390. A Rigaku X-ray diffractometer (Smartlab) was used to obtain the X-ray diffraction (XRD) patterns of both the support and the metal catalysts. A Hitachi SU3500 scanning electron microscope with Bruker XFlash 610 M X-ray Energy Dispersive Detector (SEM-EDX) determined the elemental composition of the catalysts and their surface morphology. A JEOL JEM 1200 EXII transmission electron microscope (TEM) allowed observing the morphology of the metal nanoparticles supported on the zeolite. A Thermo Scientific NEXsa X-ray photoelectron spectrometer (XPS) analyzed the photoelectron properties of the chemical elements composing the zeolites and the supported Zn and Ga, using Al Kα radiation (1486.6 eV) at 6 mA and 12 kV.

The nature and strength of the acid sites of the catalysts were determined using three methods. The first method was temperature-programmed desorption of ammonia (TPD-NH₃), which follows a previously reported methodology [13]. This method involves placing 100 mg of catalyst in a glass tubular reactor, which is heated from room temperature to 823 K at a rate of 10 K·min⁻¹, where it is held for 1 hour to degas the sample. Subsequently, the sample is cooled to 398 K, and a stream of NH₃ vaporised from a commercial NH₄OH solution is passed through for 1 hour to saturate the acid sites. After the time, the sample was desorbed with a temperature ramp from 398 K to 823 K at 10 K·min⁻¹, and a thermal conductivity detector (TCD) captured the released NH₃ signal. The signals collected were normalised based on the exact mass of the catalyst used, and using a deconvolution method, the relative amounts of weak, strong, and medium acid sites were determined.

The previously reported pyridine and 2,6-dimethylpyridine (DMPY) pulse titration method was used to quantify the amount of acid sites and to determine their nature [14,15]. This method involves putting 30 mg of catalyst into a small reactor inside the oven of a Hewlett-Packard 5890 Series II gas chromatograph, which is linked to the injector and chromatographic column. The oven is heated to 573 K in N₂ flow to degas the sample. Then, 2 μL pulses of a 1 M pyridine solution are injected, and the chromatograph detector quantifies the amount of unadsorbed pyridine. The titration ends when the two peaks of the chromatograph are of equal size, indicating that the surface is saturated with pyridine. The procedure is repeated by increasing the injection to 3, 4, and 5 μL; thus, the total acidity can be determined. Then, the used catalyst is replaced with a new one, and the exact steps are repeated using DMPY as the probe molecule, which helps determine the number of Bronsted acid sites (BAS). The determination of Lewis acid sites (LAS) was based on the difference between the total acidity and BAS.

Finally, pyridine TPD was followed by diffuse reflectance infrared spectroscopy (DRIFTS). The catalyst is placed in the cell, which is hermetically sealed and heated up to 723 K and held for one hour to degas the catalyst. The cell was then cooled to 313 K to saturate the catalyst with pyridine, and a focused spectrum was taken between 1400 and 1600 cm⁻¹, where the pyridine-coordinated bonds appear. Subsequently, spectra were taken at 423 K, 573 K, and 723 K to observe the strength of the BAS and LAS. The molar extinction coefficients proposed by Emeis [16] were used to determine the LAS/BAS ratio and compare with the ratio determined by pulse titration.

4.3. Micropyrolysis assay

The micropyrolysis tests were carried out in a CDS Pyroprobe 5200 HPR Micro-pyrolizer, analyzing the evolved gases using a Perkin Elmer Clarus 690 Gas Chromatograph connected to a Perkin Elmer Clarus SQ-8T MS Detector Mass Spectrometer (Fig. 5).

Fig. 5.

Experimental analytical pyrolysis system.

A small amount of sample (0.5 mg of torrefied or non-torrefied biomass) was placed in a quartz tube of 20 mm in length and 2 mm in diameter, with the ends sealed with quartz wool (non-catalytic experiments). In the catalytic experiments, the catalyst was placed at the ends of the reactor, as shown in Section A of Fig. 5, with an amount corresponding to the biomass/catalyst ratio established in the experiment design. Inside the reaction chamber, the quartz reactor remained surrounded by a resistive Pt filament (Section B, Fig. 5), which allowed heating at high heating rates (1000 or 10,000 K·s⁻¹) up to the pyrolysis temperature (673, 723, 773, 823 K). The reactor was held at the pyrolysis temperature for 15 s to decompose the biomass, and then rapidly cooled to 323 K to stop the pyrolysis process. With a flow rate of 40 mL min⁻¹ He, an oxygen-depleted atmosphere was guaranteed in the conventional pyrolysis experiments.

In contrast, the same flow rate for H₂ was used in the hydropyrolysis experiments, both cases being conducted at a total pressure of 150 psi. The pyrolizer outlet was coupled to a Tenax trap, which held the produced compounds before sending them to the gas chromatograph through a thermal sleeve at 573 K via a heating ramp (373 to 573 K). The current that entered the chromatograph was diluted at a 1:50 ratio with He to prevent detector saturation. The column (Elite 1701, 30 m × 0.25 mm × 0.25 μm) separated the compounds at 743 K during a 40-minute run.

Table 4 presents the factors and levels employed in the experimental development to generate this data set. The variables taken as factors were biomass, which referred to the non-torrefied biomass and the four torrefied samples. The atmosphere differentiated between conventional pyrolysis and hydropyrolysis. Two heating rates and four temperatures were evaluated to find the lowest acid formation and the highest furan and phenol formation. The catalyst variable included five different catalysts, as well as the unmodified Chilean natural zeolite, which were tested with various amounts of biomass compared to the catalyst.

Table 4.

Factors and levels used in experimental development for the data set record.

Factor	Level
Biomass	Chilean Oak, Oak_Torr1, Oak_Torr2, Oak_Torr3, Oak_Torr4
Atmosphere	Conventional pyrolysis (He), Hydropyrolysis (H2)
Heat rate	1000 K·s−1, 10,000 K·s−1
Temperature	673 K, 723 K, 773 K, 823 K
Catalyst	NZ, H2NHZ, Ga2-H2NHZ, Ga5-H2NHZ, Zn2-H2NHZ, Zn5-H2NHZ
Reaction chamber H2 filling time	2 min (3.7 mmol), 3 min (14.5 mmol)
Biomass/catalyst ratio	1/3, 1/5, 1/7, 1/10, 1/13

The effect of the variables presented in Table 4 was carried out in a systematic sequence to obtain the highest amount of furans, phenols, and monoaromatics hydrocarbons. In this sense, Fig. 6 presents the scheme of the sequential steps followed in the experimental development. In experimental series 1, the aim was to evaluate how the torrefied biomass, temperature, and heating rate affect conventional pyrolysis (He). Then, the best conditions to obtain the highest amount of furans and phenols were tested in hydropyrolysis (H₂, experimental series 2) with all torrefied and non-torrefied biomass. In experimental series 3, the effects of the catalyst, reaction atmosphere, and biomass on the process were analyzed, while keeping the heating rate, temperature, and amount of H₂ in the reaction chamber constant for hydropyrolysis. Finally, in experimental series 4, the Oak_Torr3 biomass was selected, and the impact of the biomass/catalyst ratio was studied for all the prepared catalysts. In this series, the time to fill the reaction chamber with H₂ was increased to 3 min to improve the catalytic activity.

Fig. 6.

Workflow diagram of experimental development.

Limitations

Pyrolysis data were obtained in two different ways: first, through thermogravimetric analysis at varying heating rates, and second, by analytical pyrolysis with modified operating conditions. The primary difference between the two types of data acquisition lies in the heating rate and sample setup employed during the experiment. Consequently, Py-GC/MS data cannot be used as direct correlations with TGA data, nor vice versa, since their results refer to different pyrolysis products (TGA = solid, Py-GC/MS = Gases). TGA is limited in its ability to infer the molecular composition of solids or decomposition gases, in contrast to Py-GC/MS, which can provide such information. Nonetheless, TGA can quantify the mass loss from the solid to the gaseous phase, a function not achievable by Py-GC/MS. For example, it is recommended to collect TGA data for biomass decomposition and Py-GC/MS data for the condensable gas product. Furthermore, for these data, the mass spectrometry detector was set to m/z ranging from 35 to 300, ensuring that no pyrolysis-specific gaseous compounds such as CO, H₂O, and partially CO₂ were detected.

Ethics Statement

The authors state that neither humans nor animals were involved in the acquisition of the dataset. These data were obtained through experiments, so we do not present data from social networks or public databases.

CRediT Author Statement

Kevin J. Fernández-Andrade: Conceptualization, Formal analysis, Data Curation, Writing - Original Draft. Konstanza Andrea Ortiz-Araya: Data Curation, Writing - Original Draft, Formal analysis, Visualization. Francisco Medina-Jofre: Methodology, Data Curation, Writing - Original Draft. Serguei Alejandro-Martín: Writing – review & editing, Supervision, Resources, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Conceptualization.

Data Availability

• Mendeley DataDataset of Chilean Oak micropyrolysis over Zn and Ga supported on natural zeolite catalyst in oxygen-depleted (He) and reductive (H₂) atmospheres (Original data)