Next Article in Journal
Examples of the Superiority of Ionic Liquids and Deep Eutectic Solvents over Aqueous Solutions in Electrodeposition Processes
Previous Article in Journal
Novel Continuous Laboratory Reactor with Selectively Tempered Capillary for Hyaluronic Acid Depolymerization: Design, Optimization, and Proof-of-Concept
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Pyrolysis-GCMS of Plastic and Paper Waste as Alternative Blast Furnace Reductants

1
CEMEG, Faculty of Science and Engineering, Swansea University, Swansea SA1 8EN, UK
2
School of Engineering, Cardiff University, Cardiff CF24 3AA, UK
*
Author to whom correspondence should be addressed.
ChemEngineering 2025, 9(1), 15; https://doi.org/10.3390/chemengineering9010015
Submission received: 20 November 2024 / Revised: 22 January 2025 / Accepted: 31 January 2025 / Published: 10 February 2025

Abstract

:
This paper reports studies on the thermal chemistry of the flash pyrolysis (heating rate of 20,000 °C/s up to 800 °C) of non-fossil fuel carbon (NFF-C) waste (or refuse-derived fuel, RDF) in the context of using this as an alternative reductant for blast furnace ironmaking. Gas chromatography–mass spectrometry (GCMS) analysis linked to the pyrolyser was used to simulate the thermal processes that take place during injection in the blast furnace raceway, where material experiences extreme temperature (ca. 1000 °C) over very short residence times (<300 ms). Species identification and qualitative analysis of evolved species generated are reported. Whilst the pyrolyser uses flash heating of a static sample, a drop tube furnace was also employed to study a sample moving rapidly through a pre-heated furnace held at 1000 °C to enable reductant burnout rates to be measured. The overarching aim of this piece of work is to study the suitability of replacing fossil fuel with non-recyclable plastic and paper as blast furnace reductants.

1. Introduction

Ironmaking uses large amounts of coal and natural gas fossil fuel to reduce Fe2O3 along with coke and injected coal in the blast furnace [1]. While commercial and environmental drivers exist to reduce fossil fuel use [2], significant technical challenges surround its displacement in blast furnaces because multiple chemistries occur across wide-ranging redox conditions, time and length scales. Blast furnaces are also pressurised counter-current reactors where solid raw materials are added at the top and descend slowly (5 h), whilst hot gas and coal dust are injected at the base and pass up through the blast furnace burden much more rapidly (e.g., in 30 s). The high fixed carbon (i.e., >80%) of coke and coal raw materials makes their displacement with non-fossil fuel carbon (NFF-C), biomass and non-recyclable plastic/paper waste challenging. Hence, understanding the thermal behaviour of NFF-C under the extreme conditions typical of a blast furnace is the subject of this paper.
The sheer scale of the steel industry presents further decarbonisation challenges with steelmaking being amongst the largest single-point emitters of greenhouse gases in some countries [3]. OECD stated that steel production was 1864 million metric tons of crude steel in 2020 [4] with blast furnace ironmaking generating ca. 1.8 t of CO2 per tonne of hot metal [5], so the steel industry accounts for ca. 8% of global CO2 emissions. The Paris Agreement limits global temperature increase to 1.5 °C [6], which requires drastic CO2 emissions reduction. Currently, H2 direct injection (H-Dr) [7], solid biomass and refuse-derived fuel (RDF) substitution [8], zero-C electricity substitution [9] and carbon capture and storage [10] are decarbonisation technologies being studied to reduce CO2 emissions against the backdrop of an expected doubling of steel production from 2012 to 2050 [11].
Capital expenditure is also important for decarbonisation. Therefore, this paper studies alternative refuse-derived fuel reductants that can lower CO2 emissions using existing infrastructure. In this context, the injection of carbon via blast furnace tuyeres can decrease [12] coke usage and, consequently, reduce raw material costs and emissions from coke-making plants using injectants such as renewable-derived H2; renewable-sourced biomass; or non-recyclable feedstocks (e.g., refuse-derived paper and plastic waste—such as Subcoal®, Nieuw-Bergen, The Netherlands) [13,14]. The blast furnace raceway is a fuel-rich environment, characterised by short residence times and a rapidly diminishing availability of O2 in the hot blast, and the injection of these alternative reductants (containing C and H2) in this environment can form CO and H2 [15]. Introducing new NFF-C reductants requires detailed studies to maintain blast furnace process efficiency.
Four main coal-based inputs that can be substituted by NFF-C are used in ironmaking [16]. These are (i) coking coal used in the coke ovens, (ii) coke breeze used at the sinter plant, (iii) nut coke charged at the top of the blast furnace and (iv) pulverised coal injected via tuyeres into the blast furnace. Biochar from food waste [17], biomass [18,19,20] or refuse-derived fuel have been considered as NFF-C in steelmaking [21]. For example, it has been reported that adding waste plastics to coking coal blends reduces energy consumption and blast furnace CO2 emissions, with 2 wt.% waste plastic addition to coke mitigating blast furnace CO2 emissions by 2% [22]. However, the lower fixed carbon and higher oxygen content of biomass-derived material [23] suggest that as-received material can require thermal processing before becoming valorised and suitable to displace fossil fuel-based carbon. Previous studies in this area have included studies of the pyrolysis of refuse-derived fuel [24]. However, most of the previous work has focused on relatively slow pyrolysis (i.e., over 1–2 h) [25,26] or over minutes at lower temperatures (i.e., <500 °C) [27]. These previous studies have also focused on mass loss and the kinetics of thermally driven reactions rather than studying the evolved volatile matter. Torrefaction has also been studied [28] to upgrade fuel properties, in terms of heating value, energy density and material hydrophobicity. It has been reported that it is mainly ligno-cellulosic and hemi-cellulosic biomass present in refuse-derived fuel that decomposes during torrefaction. Previous torrefaction studies have also reported limited success in torrefying refuse-derived fuel [28], with calorific values often remaining unchanged with the main effect being a drop in moisture content from >20% to 1–2%. Although some improvement has been observed with a relatively small amount of plastic (~13%)-based biomass pellets, enrichment during torrefaction of the high ash content samples seemed to decrease the combustion characteristics of the torrefied samples of the refuse-derived fuel and pure biomass (decreasing the excessive combustion characteristics of pure biomass) but improved the water repellence of the samples [29]. Early large-scale trials of injected waste plastic in blast furnaces have been reported previously [14] but have not focused on the different volatile species generated from differing NFF-C feedstocks versus coal.
This paper reports the flash pyrolysis and burnout of refuse-derived fuel (Subcoal®) and typical blast furnace coals to compare their thermal decomposition. This is because the high blast gas velocity and the limited residence time of injectants directed into the blast furnace give them limited time to combust, so NFF-C injectants that are not consumed will undergo thermal decomposition outside the raceway region in the absence of O2. To study this, chemical, calorific and morphological analyses were carried out and the volatile matter produced during pyrolysis was identified using gas chromatography–mass spectrometry (GCMS) and compared with blends of Subcoal® and coal to simulate the effects of fuel switching. Torrefied samples of Subcoal® were studied to understand the potential to upgrade refuse-derived fuel for blast furnace use.

2. Materials and Methods

Coal samples were supplied by Tata Steel, Port Talbot, UK and had already been milled to particle size <0.5 mm. Coal was therefore used as received after drying. Subcoal® (supplied by N+P Group) consisting of non-recyclable paper and plastic waste pellets, 6 mm diameter and ca. 2 cm length, were used as supplied or were milled (Figure 1) to a fine homogenous consistency (typically ca. particle size <0.5 mm) before use. Milling typically involved sequentially passing through a Retsch UK SM300 cutting mill and a Retsch PM100 Planetary Ball Mill (200 rpm for 5–10 min using 2 cm diameter stainless steel ball bearings), followed by a Retsch ZM200 ultra-centrifuge. The fine, homogenous powders were then sieved using a Retsch AS400 sieve shaker, typically attaining particle size distribution <0.5 mm.
Pyrolysis-GCMS measurements were performed with a CDS instrument 5500 Pyrolizer (Py). Samples (ca. 20 mg) were loaded into silica quartz tubes (2 mm diameter) with silica glass wool loosely packed into each end of the tube to hold the sample in place and to prevent any particulate matter from entering the gas lines. This tube was placed inside a Pt filament at the end of the pyroliser gun. Samples were ultra-fast heated at 40 ms to 800 °C in 99.999% He as the carrier gas (flow rate = 100 mL/min). The emitted volatiles were first captured onto a sorbent trap and then re-volatilised by flash heating and eluted in a flow of 99.999% He gas (BOC) with a flow rate of 100 mL/min. This carrier gas was also used for the gas chromatograph (GC) and all the hyphenated pyrolysis-GCMS systems. The process of trapping the emitted volatiles and flash desorbing them together significantly improved the chromatography for early eluting compounds, producing improved peak shapes and peak resolutions. The pyroliser was interfaced to the injector of an Agilent 7890B (GC) using a needle through the injector septum attached to a heated transfer line held at 300 °C. The GC was equipped with a Restek Rxi-624Sil MS (fused silica, mid-polarity Crossbond phase) 60 m column. The GC oven was heated with ramp rate of 10 °C/min from 25 °C to a set temperature of 250 °C for the samples. The outlet of the GC was interfaced directly with a LECO Pegasus BT+ time-of-flight mass spectrometer (ToF-MS). GCMS chromatogram data were reported as Total Ion Count (TIC) versus retention time (RT). The mass spectral data were searched against the NIST database on the Leco GCMS ChromaTOF Version 5.5 software. The software identified the most likely compound matches with a probability of accuracy. Prominent functional groups were annotated onto the chromatograms for ease of comparison.
Torrefaction was undertaken in a horizontal tube furnace in a dry nitrogen (99.99%, 100 mL/min BOC) stream environment at two temperatures (250 °C and 300 °C) each for a duration of 1 h.
Calorific values (CVs) of the materials were measured using an Anton parr 6100 compensated jacket instrument using 99.99% oxygen (BOC). The instrument measured gross heating values (GHVs) in MJ/Kg and was calibrated using benzoic acid as the calibration standard.
Coke replacement ratio calculations were originally developed using computer models, e.g., Hutny et al. [30], and later adapted by several others [31]. They derived a relationship between the calorific value of injected coal and the replacement ratio (RR) using Equation (1):
R R = 0.6395 + 0.04 × S E
where RR is fractional replacement rate and SE is the Specific Energy MJ/kg dry ash free.
Recent RR methods include the HMB blast furnace model, which is asserted to be the most accurate in large-scale settings. This “HMB” method involves plotting the combined values of the carbon and hydrogen vs. measured coke replacement rates (ESI Table S1). This generates a curve, whose fitted line allows one to calculate the expected RR (ESI Table S2).
A drop tube furnace (DTF), manufactured by Severn Thermal Solutions, Dursley, Gloucester, UK, was used to measure the combustion burnout of coals or coal/Subcoal blends in the DTF and produce partially burnt chars for TGA analysis. The furnace is a vertical tube and has a laminar flow to allow the characterisation of devolatilisation and burnout of coal samples at 1100 °C in air at a residence time of 100 ms and prepare chars for analysis using the TGA. A different DTF dwell temperature was used compared to pyrolysis because the heating processes were different, and the work measured different aspects of the related samples. So, flash pyrolysis heats a static sample (from ambient to 800 °C). The released VM is trapped and analysed by GCMS. Running to higher temperatures does not yield new compounds. By comparison, the DTF measures the material remaining after burnout of a sample that has fallen through a furnace held at 1100 °C for 100 ms. The high heating rate and short residence times of the DTF have characteristics similar to those when coal is injected into the blast air of the blast furnace raceway [32]. Samples were fed into the top and entrained in a laminar air flow at 20 L/min and collected at the bottom by means of a cyclone collector. The ash tracer method was used to calculate the burnout of the coals [33].

3. Results

Visual inspection of the as-received coal and Subcoal® samples (Figure 1a–d) shows a homogeneous dark grey colour for the coal sample. By comparison, the Subcoal® pellets were mainly a dark green/grey colour. However, the pellets were non-homogenous and contained particles of other colours (e.g., white, blue, red, etc., Figure 1a). The milled Subcoal® sample (Figure 1b) shows that the pellets are made up of different materials in line with its provenance as a refuse-derived fuel. As such, fibrous material can be seen, which appears to be cellulosic from paper waste and different coloured particles that appear to be plastic-based waste.
Fuel switching in ironmaking is not a new concept and pulverised coal injection through the blast furnace tuyères to displace more expensive coke additions from the top of the furnace is now widespread. However, the choice and quality of coal for use as the injected fuel significantly impacts the cost benefit that can be obtained by pulverised coal injection [34,35]. The primary factor that influences the cost and benefit of pulverised fuel injection is the amount of coke that can be replaced by the injected coal.
Table 1 shows energy density and calculated replacement ratios for different potential blast furnace feedstocks. These data are comparative rather than absolute, but they show that coal, Subcoal® and co-milled coal plus Subcoal® (90:10 w/w) all show similar energy values, whilst PET shows a slightly reduced heating value. Interestingly, the torrefaction of Subcoal® for 1 h at 300 °C shows an improvement in the energy value. The difference between calorific values for the two torrefied Subcoal® samples is expected to be due to the heterogeneous nature of the pellet compositions.
To compare the benefits of fuel switching and co-firing, we need to understand the Subcoal® material in coal standardised tests such as ultimate and proximate analyses (Table 2). The ultimate analysis shows a much higher %C for coal relative to Subcoal® but slightly less %H, which is in line with the higher fixed carbon in the proximate analysis. It should be noted that the fixed carbon in the proximate analysis is a different value than the total carbon from the ultimate analysis. The difference can be explained because the total carbon includes some organic carbon that escapes as gaseous volatile matter during combustion. The ultimate analysis shows slightly more nitrogen in the coal than Subcoal® and similar sulphur levels but substantially more oxygen in Subcoal® than coal. This reflects the fact that Subcoal® is a mixture of non-recyclable paper and plastic. Paper will contain cellulose [36] and lignin [37], which include substantial amounts of oxygen within their structures, and the widely used polyethylene terephthalate (PET) plastic is also likely to be present, which also contains oxygen. Previous reports have suggested that increased oxygen in blended feedstock can increase the char combustion rate versus coal alone [38]. Hence, coal and Subcoal® blends were studied in this paper. The data show that coal and Subcoal® have similar levels of post-combustion ash (8.4 to 10.0%) and similar moisture levels (2.4–4.0%). It is worth noting that the composition of inorganic minerals contained in the ash can impact the reactivity of the reductants and slag acidity/basicity in the furnace environment. Fillers and additives contained in plastics and paper contribute to the ash content. The proximate analysis of Subcoal® indicates an ash content in the same region as coal. Unless the feedstock is pre-dried, this water will be evaporated in the blast furnace, which requires energy and releases water into the burden. Finally, the data also show higher energy values (GHV and NHV) for coal over Subcoal® whilst the bomb calorimeter data (Table 1) show no discernible differences. The data in Table 2 were measured using a standardised method and corrected for sulphur content.
We studied the ultra-fast (20,000 °C/s) or flash pyrolysis of different carbon feedstocks under inert, oxygen-free helium gas (Figure 2). Samples were heated up to 800 °C in ca. 40 ms. This was not to completely volatilise the samples but rather to study the initial volatile matter released, so the experimental design differs slightly from the proximate analysis, which is carried out at 900 °C. Instead, this was to mimic the conditions the feedstocks would experience if injected through a tuyère into the raceway at the bottom of a blast furnace. However, under normal operating conditions, tuyere-injected carbon feedstock would also experience the oxygen-rich hot blast [1]. Hence, it is logical that, in a blast furnace raceway, most of any volatile matter will be rapidly converted to CO2, water, and potentially other fixed gases (CO, H2 etc.). This study aims to identify the volatile species and how they change for different feedstock materials. Hence, we excluded oxygen to prevent combustion. It should also be noted that the ultra-fast heating rate combined with the helium gas flow will mean that the system is under kinetic rather than equilibrium control because the inert gas will drive off the volatiles onto a sorbent trap so that post-pyrolysis reactions are expected to be limited. This is important because pyrolysis is a generic term that can equally be applied to heat-driven reactions taking place over hours or even days, where released volatiles remain within the system and are able to react with each other and/or with any remaining char [25,26,27].
In this context, Figure 3 shows GCMS chromatograms arising from the ultra-fast pyrolysis of the as-received blast furnace coal, Subcoal® and polyethylene terephthalate plastic (PET), respectively. The data shows that for coal (Figure 3a), the most abundant volatiles observed are single aromatic rings (e.g., benzene) along with several polycyclic aromatics, e.g., indene, which is a benzene ring fused with cyclopentadiene, and naphthalene, which is two benzene rings fused together. In addition, there are alkyl-branched examples (e.g., toluene is methyl benzene) along with two isomers of methyl-naphthalene (Table 3). The abundance of aromatic compounds when compared to the absence of non-aromatic aliphatic or heteroatomic organic compounds is interesting, particularly so when considering the prior work of Ming Sun et al. [39], who first distilled coal tar from coal samples before analysis by pyrolysis-GCMS. Their data does show a predominance of aromatic compounds, but they also observe aliphatic and heteroatom compounds (mainly oxygenated and nitrogen-containing compounds). This suggests that these other compounds are likely to be present; this is backed up by the ultimate analysis (Table 2), which shows that 1.8% nitrogen, 0.3% sulphur and 3.8% oxygen are present by weight.
Given that heteroatomic organic compounds often only contain a single heteroatom, if these values were translated into volatile matter, this should produce measurable quantities of heteroatom compounds. Thus, these data suggest that these are not the first compounds emitted from the heated coal, which in turn means that they are likely to be retained in the char (Figure 2).
The similarities in the aromatic compounds between the coal and Subcoal® (Figure 3a,b, respectively) are perhaps not surprising when it is considered that coal is formed from ancient biomass and contains the products of the conversion of highly aromatised lignin components [40]. Subcoal® does contain cellulosic material from the paper present, but the paper will most likely also contain some lignin. Subcoal® also contains waste plastics and, given the prevalence of polyethylene terephthalate and polystyrene in everyday use, these are also likely to pyrolyse into aromatic compounds. Indeed, the heterogeneity of Subcoal® is reflected in the larger number of peaks observed in the chromatogram (Figure 3b). Some of these new peaks, given that the refuse-derived fuel is a heterogeneous material, have an inherent variability in the components due to the plastic and paper used in each pellet. Interestingly, analysis of the extra peaks particularly prevalent at shorter retention times (i.e., <600 s) with cut-offs of peak area >1 × 109 and match probability’s > 0.50 (Table S1) shows that they include examples of unsaturated aliphatics (e.g., 1,3–butadiene, 1,3-cyclopentadiene, 1-buten-3-yne, 3-methyl-cyclopentene and 1,5-hexadiene) along with heteroatom compounds. These include the highest number of examples for oxygen (e.g., 2-propenal, propanedioic acid, acetone, cyclopentanone, benzoic acid, furfural and propargyl alcohol), fewer for nitrogen (e.g., benzonitrile, aniline, pyridine, 2-propenenitrile, aminoacetonitrile and pyrrole) and one confirmed example for chlorine (i.e., chloromethane). The most likely source of chloromethane is from sub-components of polyvinylchloride (PVC) plastic waste within Subcoal®, which is quoted as typically containing 0.8% Cl by weight [41]. Because the presence of chlorine (through the raw material input) can lead to the formation of HCl in a blast furnace, the input of materials containing this is limited by operators according to the specific maximum chlorine limit of their furnace.
The higher proportion of heteroatom compounds for Subcoal® is not surprising considering the much higher oxygen content (26.9%) shown in the ultimate analysis (Table 2). In addition, although these analyses are qualitative, the much higher total ion current in the Subcoal® chromatogram (full-scale TIC in Figure 3b is 3 × 109) is in line with the higher proportion of volatile matter in Subcoal® (73.8%) compared to coal (24.0% and a full-scale TIC in Figure 3a is 6 × 107).
Looking at the chromatogram in Figure 3c for virgin polyethylene terephthalate (PET), the data look increasingly like that of coal (Figure 3a), with very similar amounts of benzene, toluene and naphthalene all clearly present as the most abundant compounds but, again, very few peaks observed at retention times <600 s. However, one key difference is the much larger peak area in the region of the naphthalene peak for PET (retention time ca. 1360 s), which dominates in terms of peak areas for this sample. A second compound, beta-1,5-O-dibenzoyl-ribofuranose, is identified here by the GCMS. This compound contains benzyl ester groups linked to an oxygen-rich furanose ring, so this is not unsurprising given this arises from the PET polymeric structure. This peak also appears to co-elute with naphthalene so there may be an overlap between several species in this region owing to the high naphthalene content. From these data, plastics can be considered like lower-rank coal, containing high content of volatiles, which can be easily decomposed and released in the form of tar and gas during pyrolysis. The main extra compounds PET produces compared to raw coal can be grouped into O-containing molecules (benzaldehyde, 1-(2-methylphenyl)ethanone, benzoic acid and xanthene) along with the aromatic compounds styrene and much higher levels of naphthalenes. Based on the peak areas, the O-compounds make up 28.3% of the total for PET, and the styrene and naphthalene make up 51.5%. In terms of their subsequent reduction chemistries in the blast furnace, the O-compounds are likely to form CO whilst the aromatic compounds are likely to take part in the Boudouard reaction.

3.1. Co-Milling of Coal with Subcoal®

Co-firing or co-milling is typically used to modify the coal rank score, usually to improve the thermal characteristics of lower-rank coal [38]. Coal ignition and burnout characteristics are also modified. Coal ignition is clearly an important factor for combustion, boiler design and blast furnace injection. Different experimental equipment can be used to study this process (e.g., thermogravimetric analysers or entrained flow reactors). The latter are more suitable for studying the ignition in pulverised boilers, as they simulate combustion conditions more closely (i.e., particle residence time, heating rate). However, the combustion conditions in thermogravimetric tests (i.e., temperature and kinetic regime) differ greatly from those encountered in a pulverised combustor. Other researchers have studied the reactivity of coal chars in entrained flow reactors under rich oxygen conditions and high temperatures to obtain kinetic parameters [42,43]. These studies have shown that the char combustion rate increases with oxygen concentration as might be expected.
However, here, we again used pyrolysis-GCMS under inert gas to identify the components of volatile matter evolved from co-milled coal and Subcoal® samples. In these experiments, coal and Subcoal® were intermixed at the first stage of milling (i.e., at the cutting mill stage). The proportions of coal to Subcoal® were chosen to be 90:10, 75:25 and 50:50 (coal–Subcoal® by weight). These samples were then passed through the ball mill and ultra-centrifuge milling processes until fine, homogeneous powders of each sample were obtained. These co-milled samples were analysed by pyrolysis-GCMS, and the data are shown in Figure 4 and Table 4.
The chromatograms (Figure 4) show a general agreement being mostly largely dominated by benzene, toluene and naphthalene peaks. Interestingly, as Subcoal® loading is increased, chlorobenzene is detected in the evolved gas. Chlorine is known to be an element of concern inside the blast furnace [44] as chlorine compounds in the blast furnace gas lead to the corrosion of pipelines, air heaters and tuyères (which are made of copper and thus especially susceptible to attack from chlorine compounds. Hence, it is important to control the input levels of chlorine in the furnace. Here, we detect a peak area of 1,067,935,306 or ~13.7% that of the benzene peak for the chlorobenzene signal for the 25% Subcoal® sample. It is important to note that these analyses are qualitative and that we used a time-of-flight GCMS to analyse these samples with a limit of detection (LOD) in the order of 20 femtograms [45]. Hence, detecting chlorinated organics does not necessarily discount the use of Subcoal® in a blast furnace. Instead, it does indicate the type of volatile organic compounds that would be released in this scenario. In practice, these would likely be rapidly decomposed into HCl(g). As stated previously for neat Subcoal®, the most likely source of chlorobenzene is from PVC within the plastic waste within this refuse-derived fuel, which typically contains 0.8% Cl by weight [41]. However, here, a different chlorinated organic is observed. This might suggest that chlorine radicals are produced during pyrolysis, and these chlorinate the most labile organic species. For neat Subcoal®, that appears to be methyl radicals, whereas, here, it is benzyl radicals or benzene (heat of combustion 3268 kJ mol22121) [46]. But in terms of a blast furnace application for either neat or blended Subcoal®, these data confirm that chlorinated organics are produced and so the Cl level in the feedstock would need to be monitored (as it would be for any other feedstock). A large increase in the naphthalene peak (heat of combustion 5150 kJ mol−1) [47] is also observed as Subcoal® loading increases. It has been reported that compounds rich in hydrogen (e.g., plastics) are more favourably cracked to generate radicals that can combine with phenolic precursors in biomass-coal blends when compared to more carbon-rich (or more hydrogen-deficient) materials [48]. This was found to effectively increase the content of phenols in the resultant coal tars. Hence, it could be that a related mechanism occurs here, which results in elevated naphthalene levels.
This suggests that increasing the level of more hydrogen-rich feedstocks (e.g., plastic) compared to more carbon-rich materials (e.g., coal or coke) in a blended situation does vary the generated species. It could also be that injecting more H2-rich materials may also enhance direct iron reduction, with more H2 available from the waste plastics contributing to the reduction process and steam (H2O) as the gaseous reduction product. It has been reported that this could lower the amount of CO2 generated by ca. 30% in comparison with the use of coke and coal alone [14].

3.2. Monitoring Valorisation of Refuse-Derived Fuel

The torrefaction of Subcoal® was also studied at 250 °C and 300 °C. In both cases, the samples were heated for 1 h under nitrogen (100 mL/min); thus, effectively, the material experienced a low-temperature, slow pyrolysis step before then undergoing ultra-fast pyrolysis up to 800 °C to again simulate the potential to upgrade the feedstock value of Subcoal® before injecting it into a blast furnace. It should be noted that thermal pre-treatment at higher temperatures and for long periods will lead to carbon loss. Table 1 shows that an increase of ca. 20% is measured for the gross energy for the torrefied material at 300 °C. Clearly, some of this typically takes place at the expense of mass loss in the form of volatile components (~17% ESI Figures S2 and S3). Interestingly, it has been reported previously that typically, as torrefaction temperature and duration are increased, material degassing increases, resulting in an increase in ash content, which can unfavourably influence the higher heating value (HHV) or gross energy [49]. Our data suggest that 1 h at 300 °C is not a high enough temperature and/or a long enough residence time to detrimentally affect the gross energy of torrefied Subcoal®.
Pyrolysis-GCMS data for the torrefied samples (Figure 5 and ESI Table S3) show there is a clear change in the pattern of the retention times of the peaks. Thus, Subcoal®, which was not torrefied (Figure 3b), shows many peaks at retention times < 600 s. These compounds generally have the smallest molecular weights and are the most volatile, so they pass through the GC column first. The number and intensity of the peaks at retention times < 600 s decrease after torrefaction at 250 °C to a single peak after torrefaction at 300 °C. This is interesting because Subcoal® became more coal-like in its ultra-fast pyrolysis behaviour (see Figure 3a) and its gross energy value increased. At the same time, there is a general increase in the larger MW compounds at longer retention times (e.g., benzene, toluene, styrene) with a large increase in the naphthalene component for the 300 °C treated sample. In this context, Nobre et al. previously studied the torrefaction of refuse-derived fuel and reported that refuse-derived fuel char has a fuel profile similar to that of lignite coal, which is in line with our data [48]. Interestingly, in light of the earlier data for chlorine-containing organics in non-torrefied Subcoal®, we do not observe any chlorinated organics in the torrefied samples, which suggests they are removed in this process. However, Nobre et al. also reported that nearly all chlorine compounds in their samples could be removed by water washing at room temperature [50].

3.3. Subcoal® Burnout in a Drop Tube Furnace Rig

If adopted as an alternative reductant, Subcoal® is well suited to injection into the raceway of the furnace. This balloon-like void formed by the high blast velocity is a very dynamic region with much variation in dimensions, temperature and gas composition. It is characterised by short residence times [51] and a rapidly diminishing availability of oxygen for combustion [52]. For this reason, if the reductants are not completely consumed in the raceway, the coal chars formed by the partial combustion will ascend the shaft, and their properties and reactivity play a role in the blast furnace’s thermal and process stability. To assess Subcoal’s® effect on the burnouts and its suitability for blending, samples were run through a drop tube furnace to measure and compare the burnout at different Subcoal® loadings (Table 2 and Table 5).
The drop tube furnace results (Figure 6) clearly show a trend with the addition of Subcoal® to the lower volatile matter (VM) content coal DTF 1. The burnout rate steadily increased with the increasing addition of Subcoal®, which will benefit the increased utilisation in this region of the furnace. This is ascribed to the fact that Subcoal contains 73.8%VM (Table 2), so adding 5%, 10% or 20% Subcoal to coal with no VM will increase the total VM by 3.7%, 7.4% or 14.8%. Figure 6 shows that the burnout for the low VM coal (DTF1) increases linearly in line with the increase in VM from Subcoal; thus, the low VM/high fixed carbon coal benefits from the high VM Subcoal® addition, as particle temperatures are raised due to the volatile combustion and promote the increased burnout of the fixed carbon. By comparison, there is a less clear relationship observed when the volatile content is higher. This is because the VM in the coal is being displaced by VM from Subcoal® so the addition of Subcoal to medium and high VM coals has less effect on the burnout rates. As such, the combined volatile matter content of both Subcoal® and coal varies less with Subcoal loading and a limit is reached in the combustion due to the balance between devolatilisation and oxygen availability for combustion. As VM increases, there must be sufficient O2 to ensure burnout [53,54]. Comparing coal with Subcoal® (Table 2), coal burnout follows the equation C76.8H5.4O3.8 + 76.25O2 → 76.8CO2 + 2.7H2O, which, for 12.7% VM, requires 9.68 moles of O2. By comparison, Subcoal® burnout follows the equation C48.8H6.6O26.9 + 37O2 → 48.8CO2 + 3.3H2O, which, for 73.8% VM, requires 27.3 moles of O2. So, if coal were completely substituted by Subcoal, this would require O2 enrichment up to 59.2% O2 or a 2.82 times increase in air flow rate to achieve complete burnout. However, if Subcoal® displaces 20% coal, this would require 13.2 moles of O2, which would require either an O2 enrichment from 21% to 28.6% or a 1.36 times increase in flow rate. Overall, the results suggest that Subcoal® can be incorporated at levels up to 20% by mass but there is potential for some burnout variability. This is supported by the data in Table 2, which show that when comparing the CxHyOz stoichiometries of coal C76.8H5.4O3.8 and Subcoal C48.8H6.6O26.9, increasing Subcoal loading relative to coal decreased the carbon but increases the oxygen present. This suggests that only partial displacement of coal is possible (e.g., up to 20%) in a blast furnace.

4. Discussion

Displacing carbon in blast furnaces is not a new concept. For example, pulverised coal with high (ca. 75%) fixed carbon is routinely injected through tuyères to displace more expensive coke. At the other end of the volatile matter scale, gases such as hydrogen or natural gas can also be injected. Interestingly, Subcoal® (the subject of this paper) is an example of a commercial refuse-derived fuel which, whilst solid at room temperature, contains ca. 75% volatile matter (VM). Hence, it is the opposite of coal in VM terms. Considering this in the context of blast furnace operation, analysis of the volatile matter from Subcoal® shows it contains many aromatic compounds similar to coal but also a range of aliphatic and heteroatom compounds not seen from coal pyrolysis. However, the aliphatic compounds do not differ drastically from natural gas; the oxygen and nitrogen heteroatom compounds would not necessarily be expected to combust in the same way as other hydrocarbons. Although few chlorinated organics were observed at low levels, chlorobenzene was observed coming from the PVC, which we have reported previously [55]. We have shown that PVC unzips and loses HCl but the unsaturated C=C backbone can then cyclise to form benzene. If either chlorine remains in the structure or the abundant Cl present recombines with the aromatic ring, it can produce chlorobenzene [49]. This does require monitoring to prevent blast furnace operational problems. Simulations could be used to predict the specific chemical pathways leading to the formation of pyrolysis products to compare with the experimental data in this paper. However, Subcoal® contains many sub-components and individual plastics or paper, each producing hundreds of pyrolysis products. This paper supports this with the multicomponent Subcoal pyrolysis producing >1000 individual molecules, which would make such simulations challenging.
Creating blends of coal with Subcoal® has also been studied because expecting blast furnace operators to transition from years of expertise and optimisation using fossil fuel to NFF-C feedstocks in one step is unrealistic. The data show that carbon switching from a 10% refuse-derived fuel to a 90% coal mix demonstrates that most of the volatiles produced remained the same as 100% coal, with the main changes observed at a 50:50 refuse-derived fuel loading, which generated more lower molecular weight compounds. Whilst this has the potential to improve char combustion, it could increase blast furnace pressure and thus would need to be monitored accordingly. However, the findings show that most issues can be addressed by torrefaction of the refuse-derived fuel; this removes most low molecular weight volatile matter and chlorinated organics and increases the gross energy at an increased processing cost due to the extra treatment step.
The energy is important because ironmakers minimise cost by matching the highest practical fuel injection rates with the lowest blast O2 enrichment level that will support that amount of injected fuel [30]. However, O2 enrichment allows operators to displace coke with pulverised coal and/or natural gas, which lowers the coke rate and can reduce the production of coke breeze.

5. Conclusions

This work suggests that refuse-derived fuel could be used to improve the combustibility of coal feedstocks, as, typically, lower volatile coals require higher O2 enrichment, adding to the overall cost. In addition, the fact that Subcoal is produced from waste raises issues about the consistency of the composition between batches. Subcoal is a commercial material made by combining wastes from Belgium, Germany, the Netherlands and the UK [13]. The plastic inputs reflect the plastic production and recycling in those countries. N+P states that the sophisticated waste processing systems in these countries keep the raw material inputs into Subcoal consistent. Plastic production can be broadly grouped into aliphatic-based polymers (PE, PP—45.2%) and aromatic-/polyester-based polymers (PS, PET, polyester fibres–28.7%) with PVC—9.3% [56]. Subcoal for blast furnace use contains ca. 50%:50% paper to plastic, low moisture (4% in Table 2) and low chlorine (<0.8%), reflecting low PVC. Hence, aliphatic to aromatic plastic in Subcoal is estimated to be ca. 60:40. This paper also shows that pyrolysis-GCMS is a powerful technique for looking at the individual species present during de-volatilisation at high temperatures ca.>1000 °C. The nature of the pyrolysis-GCMS allows any matrix to be quickly analysed, with no prior treatment or the need for solvents (avoiding solvent incompatibility issues with both sample and GCMS column), and makes it ideal for investigating high-temperature coal/plastic/biomass chemistries. It should be noted that differing minerals and acidic/basic ratios of the resulting ash chemistries post-combustion, with resulting catalytic or inhibiting effects, should also be further investigated. However, in this current pyrolysis-GCMS setup, very few volatiles would remain post-pyrolysis/combustion to analyse via condensed gases using the pyrolysis-GCMS.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/chemengineering9010015/s1, Figure S1: Graph of HMB model of predicted RR vs. C + H content of various material set, Figure S2: TGA data for Subcoal® showing 88% mass loss to 2.4 mg, Figure S3: TGA graph of Subcoal® which has been torrefied at 300 °C for 1h under N2 showing 71.2% mass loss to 5.7 mg.; Table S1: Subcoal Peak Match Database matches, using NIST GCMS database, Table S2: HMB Blast furnace balance replacement ratio (RR) model data used, Table S3: Volatile matter identified during pyrolysis of torrefied Subcoal® at 800 °C

Author Contributions

Conceptualisation, E.W.J., F.O. and P.J.H.; methodology E.W.J. and F.O.; validation, P.J.H., J.S. and R.M.; formal analysis, E.W.J. and F.O.; investigation, E.W.J. and F.O.; resources, Sustain and Steel and Metals Institute Swansea University; data curation, E.W.J., F.O., P.J.H. and J.S; writing—original draft preparation, E.W.J. and F.O.; writing—review and editing, P.J.H., F.O. and J.S.; visualisation, E.W.J.; supervision, P.J.H.; funding acquisition, P.J.H. and J.S. All authors have read and agreed to the published version of the manuscript.

Funding

This work is supported by EPSRC’s Future steel Manufacturing research hub SUSTAIN programme EP/S018107/1 (E.W.J., P.J.H.), EPSRC SUSTAIN Research Hub—Feasibility Study SFSC1 002 (J.S.), EPSRC/Tata Steel iCASE (F.O.), HEFCW for funding capital grant for the pyrolysis-GCMS and STA and Welsh Govt Circular Economy Capital Fund (Grant # 243) for funding the bomb calorimeter.

Institutional Review Board Statement

Not applicable.

Data Availability Statement

Data in ESI or available on request due to privacy restrictions.

Acknowledgments

The authors thank Tata Steel UK Ltd., British Steel Ltd. and CPL Industries for providing samples and the Steel and Metals Institute at Swansea University for assistance.

Conflicts of Interest

The authors declare no conflicts of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

References

  1. Geerdes, M.; Toxopeus, H.; van der Vliet, C. Modern Blast Making an Introduction; Delft University Press: Delft, The Netherlands; IOS Press: Amsterdam, The Netherlands, 2009. [Google Scholar]
  2. United Nations. The Paris Agreement (FCCC/CP/2015/10/Add.1); United Nations Climate Change (UNCC): Paris, France, 2015. [Google Scholar]
  3. Peplow Mark. Can Industry Decarbonize Steelmaking? Chem. ACS Stud. Mag. 2021, 99, 22–33. [Google Scholar]
  4. OECD. Latest Developments in Steelmaking Capacity, January 2024. Available online: https://one.oecd.org/document/DSTI/SC(2023)3/FINAL/en/pdf (accessed on 30 January 2024).
  5. Pinto, R.G.D.; Szklo, A.S.; Rathmann, R. CO2 emissions mitigation strategy in the Brazilian iron and steel sector–From structural to intensity effects. Energy Policy 2018, 114, 380–393. [Google Scholar] [CrossRef]
  6. Höhne, N.; Kuramochi, T.; Warnecke, C.; Röser, F.; Fekete, H.; Hagemann, M.; Day, T.; Tewari, R.; Kurdziel, M.; Sterl, S.; et al. The Paris Agreement: Resolving the inconsistency between global goals and national contributions. Climate Policy 2016, 17, 16–32. [Google Scholar] [CrossRef]
  7. Vogl, V.; Åhman, M.; Nilsson, L.J. Assessment of hydrogen direct reduction for fossil-free steelmaking. J. Clean Prod. 2018, 203, 736–745. [Google Scholar] [CrossRef]
  8. Mandova, H.; Leduc, S.; Wang, C.; Wetterlund, E.; Patrizio, P.; Gale, W.; Kraxner, F. Possibilities for CO2 emission reduction using biomass in European integrated steel plants. Biomass Bioenergy 2018, 115, 231–243. [Google Scholar] [CrossRef]
  9. Fan, Z.; Friedmann, S.J. Low-carbon production of iron and steel: Technology options, economic assessment, and policy. Joule 2021, 5, 829–862. [Google Scholar] [CrossRef]
  10. Toktarova, A.; Karlsson, I.; Rootzén, J.; Göransson, L.; Odenberger, M.; Johnsson, F. Pathways for Low-Carbon Transition of the Steel Industry—A Swedish Case Study. Energies 2020, 13, 3840. [Google Scholar] [CrossRef]
  11. Pauliuk, S.; Milford, R.L.; Müller, D.B.; Allwood, J.M. The Steel Scrap Age. Environ. Sci. Technol. 2013, 47, 3448–3454. [Google Scholar] [CrossRef]
  12. Alexander, B. Blast furnace injection for minimizing the coke rate and CO2 emissions. Ironmak. Steelmak. Process. Prod. Appl. 2021, 48, 728–741. [Google Scholar]
  13. N+P. Subcoal-International n.d. Available online: www.npgroup.com/trend-report-myths-facts-alternative-fuels? (accessed on 10 November 2024).
  14. Al-Moftah, A.M.S.H.; Marsh, R.; Steer, J. Thermal Decomposition Kinetic Study of Non-Recyclable Paper and Plastic Waste by Thermogravimetric Analysis. ChemEngineering 2021, 5, 54. [Google Scholar] [CrossRef]
  15. Suopajärvi, H.; Fabritius, T. Towards More Sustainable Ironmaking—An Analysis of Energy Wood Availability in Finland and the Economics of Charcoal Production. Sustainability 2013, 5, 1188–1207. [Google Scholar] [CrossRef]
  16. Jenkins, B.M.; Baxter, L.L.; Miles, T.R., Jr.; Miles, T.R. Combustion properties of biomass. Fuel Process. Technol. 1998, 54, 17–46. [Google Scholar] [CrossRef]
  17. Munir, M.T.; Ul Saqib, N.; Li, B.; Naqvi, M. Food waste hydrochar: An alternate clean fuel for steel industry. Fuel 2023, 346, 128395. [Google Scholar] [CrossRef]
  18. Gan, M.; Li, J.; Ji, Z.; Fan, X.; Yu, D.; Liu, L.; Sun, Z. Preparation of high-quality biomass char and gas as fuels for ferrous metallurgy by integrated water leaching and pyrolysis: A case study using maize straw. Biomass Bioenergy 2024, 184, 107200. [Google Scholar] [CrossRef]
  19. Reis, S.; Holliman, P.J.; Martin, C.; Jones, E. Biomass–Coal Hybrid Fuel: A Route to Net-Zero Iron Ore Sintering. Sustainability 2023, 15, 5495. [Google Scholar] [CrossRef]
  20. Reis, S.; Holliman, P.J.; Martin, C. Evaluation of the effect of a biomass fuel source on the thermal properties of iron ore sinter. Fuel 2025, 381 Pt B, 133172. [Google Scholar] [CrossRef]
  21. Han, J.; Huang, Z.; Qin, L.; Chen, W.; Zhao, B.; Xing, F. Refused derived fuel from municipal solid waste used as an alternative fuel during the iron ore sinter process. J. Clean. Prod. 2021, 278, 123594. [Google Scholar] [CrossRef]
  22. Dou, B.; Park, S.; Lim, S.; Yu, T.-U.; Hwang, J. Pyrolysis Characteristics of Refuse Derived Fuel in a Pilot-Scale Unit. Energy Fuels 2007, 21, 3730–3734. [Google Scholar] [CrossRef]
  23. Hanrot, F.; Sert, D.; Delinchant, J.; Pietruck, R.; Bürgler, T.; Babich, A.; Fernández, M.; Alvarez, R.; Diez, M.A. CO2 mitigation for steelmaking using charcoal and plastic wastes as reducing agents and secondary raw materials. In Proceedings of the 1st Spanish National Conference on Advances in Materials Recycling and Eco-Energy, Madrid, Spain, 12–13 November 2009; p. 4. [Google Scholar]
  24. Rajca, P.; Poskart, A.; Chrubasik, M.; Sajdak, M.; Zajemska, M.; Skibiński, A.; Korombel, A. Technological and economic aspect of Refuse Derived Fuel pyrolysis. Renew. Energy 2020, 161, 482–494. [Google Scholar] [CrossRef]
  25. Silva, R.B.; Martins-Dias, S.; Arnal, C.; Alzueta, M.U. Pyrolysis and Char Characterization of Refuse-Derived Fuel Components. Energy Fuels 2015, 29, 1997–2005. [Google Scholar] [CrossRef]
  26. Świechowski, K.; Syguła, E.; Koziel, J.A.; Stępién, P.; Kugler, S.; Manczarski, P.; Białowiec, A. Low-Temperature Pyrolysis of Municipal Solid Waste Components and Refuse-Derived Fuel—Process Efficiency and Fuel Properties of Carbonized Solid Fuel. Data 2020, 5, 48–55. [Google Scholar] [CrossRef]
  27. Białowiec, A.; Pulka, J.; Stępień, P.; Manczarski, P.; Gołaszewski, J. The RDF/SRF torrefaction: An effect of temperature on characterization of the product—Carbonized Refuse Derived Fuel. Waste Manag. 2017, 70, 91–100. [Google Scholar] [CrossRef] [PubMed]
  28. Haykiri-Acma, H.; Yaman, S. Effects of torrefaction after pelleting (TAP) process on strength and fuel characteristics of binderless bio-pellets. Biomass Convers. Biorefin. 2022, 14, 3489–3500. [Google Scholar] [CrossRef]
  29. Carpenter, A. Injection of Coal and Waste Plastics in Blast Furnaces; IEA Clean Coal Centre: London, UK, 2010; ISBN 978-92-9029-486-3. [Google Scholar]
  30. Hutny, W.P.; Price, J.T.; Gransden, J.F. Evaluation of coals for blast furnace injection using a computer model. In Proceedings of the 49th Ironmaking Conference Proceedings, Detroit, MI, USA, 25–28 March 1990; Iron and Steel Society: Warrendale, PA, USA, 1990; pp. 323–330. [Google Scholar]
  31. Campos, A.M.A.; João, P.E.B.; Paulo, S.A. Blast Furnace’s Replacement Rate Calculation for Biomasses based on chemical and Thermal Properties. Int. J. Adv. Eng. Res. Sci. (IJAERS) 2021, 8, 125–133. [Google Scholar] [CrossRef]
  32. Steer, J.; Greenslade, M.; Marsh, R. A Comparison of Laboratory Coal Testing with the Blast Furnace Process and Coal Injection. Metals 2021, 11, 1476. [Google Scholar] [CrossRef]
  33. Ballantyne, T.R.; Ashman, P.J.; Mullinger, P.J. A new method for determining the conversion of low-ash coals using synthetic ash as a tracer. Fuel 2005, 84, 1980–1985. [Google Scholar] [CrossRef]
  34. Lyalyuk, V.P.; Tarakanov, A.K.; Kassim, D.A.; Otorvin, P.I.; Pinchuk, D.V. Blast-Furnace Operation with Pulverized-Coal Injection and with Chunk Anthracite. Steel Transl. 2017, 47, 469–472. [Google Scholar] [CrossRef]
  35. Shen, F.; Sundelin, B.; Paulsson, K.; Kapilashrami, E.; Wei, G.; Chu, M.; Shen, Y. Industrial Practice of BiPCI Process of Pulverized Coal Injection for Blast Furnace Ironmaking at SSAB. Steel Res. Int. 2008, 78, 11–16. [Google Scholar] [CrossRef]
  36. Gupta, P.K.; Raghunath, S.S.; Deepali, P.D.V.; Venkat, P.; Shree, V.; Chithananthan, C.; Choudhary, S.; Surender, K.; Geetha, K. An Update on Overview of Cellulose, Its Structure and Applications, in Cellulose; IntechOpen: Rijeka, Croatia, 2019. [Google Scholar] [CrossRef]
  37. Lu, Y.; Lu, Y.-C.; Hu, H.-Q.; Xie, F.-J.; Wei, X.-Y.; Fan, X. Structural Characterization of Lignin and Its Degradation Products with Spectroscopic Methods. J. Spectrosc. 2017, 2017, 8951658. [Google Scholar] [CrossRef]
  38. Sami, M.; Annamalai, K.; Wooldridge, M. Co-firing of coal and biomass fuel blends. Prog. Energy Combust. Sci. 2001, 27, 171–214. [Google Scholar] [CrossRef]
  39. Sun, M.; Zhang, D.; Yao, Q.; Liu, Y.; Su, X.; Charles, Q.J.; Hao, Q.; Ma, X. Separation and Composition Analysis of GC/MS Analyzable and Unanalyzable Parts from Coal Tar. Energy Fuels 2018, 32, 7404–7411. [Google Scholar] [CrossRef]
  40. Hatcher, P.G.; Clifford, D.J. The organic geochemistry of coal: From plant materials to coal. Org. Geochem. 1997, 27, 251–274. [Google Scholar] [CrossRef]
  41. Alternative Fuel Company N+P Europe’s Market Leader. Available online: www.subcoal-international.com/subcoal/specifications (accessed on 10 November 2024).
  42. Murphy, J.J.; Shaddix, C.R. Combustion kinetics of coal chars in oxygen-enriched environments. Combust. Flame 2006, 10, 710–729. [Google Scholar] [CrossRef]
  43. Saito, M.; Sakadata, M.; Sato, M.; Soutome, T.; Murata, H. Combustion rates of pulverized coal particles in high-temperature/high-oxygen concentration atmosphere. Combust. Flame 1991, 87, 1–12. [Google Scholar] [CrossRef]
  44. Murav’eva, I.V.; Bebeshko, G.I. Determining the chlorine content in the blast-furnace system. Steel Transl. 2017, 47, 287–290. [Google Scholar] [CrossRef]
  45. Pegasus BT Gas Chromatography Time-of-Flight Mass Spectrometer. Available online: www.leco.com/product/pegasus-bt (accessed on 10 November 2024).
  46. Good, W.D.; Smith, N.K. Enthalpies of combustion of toluene, benzene, cyclohexane, cyclohexene, methylcyclopentane, 1-methylcyclopentene, and n-hexane. J. Chem. Eng. Data 1969, 14, 102–106. [Google Scholar] [CrossRef]
  47. Balcan, M.; Arzik, S.; Altunata, T. The determination of the heats of combustion and the resonance energies of some substituted naphthalenes. Thermochim. Acta 1996, 278, 49–56. [Google Scholar] [CrossRef]
  48. Wang, M.; Wang, Q.; Li, T.; Kong, J.; Shen, Y.; Chang, L.; Xie, W.; Bao, W. Catalytic Upgrading of Coal Pyrolysis Volatiles by Porous Carbon Materials Derived from the Blend of Biochar and Coal. ACS Omega 2021, 6, 3800–3808. [Google Scholar] [CrossRef]
  49. Paweł, S.; Andrzej, B. Kinetic parameters of torrefaction process of alternative fuel produced from municipal solid waste and characteristic of carbonized refuse derived fuel. Multidiscip. J. Waste Resour. Residues 2018, 3, 75–83. [Google Scholar] [CrossRef]
  50. Nobre, C.; Gonçalves, M.M.; Vilarinho, C.; Mendes, B.S. Upgrading of Refuse Derived Fuel through torrefaction: Evaluation of RDF char fuel properties. In Proceedings of the of ECOS 2018—31st International Conference of Efficiency, Cost, Optimization and Environmental Impact of Energy Systems, Guimaraes, Portugal, 17–22 June 2018; Available online: https://www.researchgate.net/publication/333092140 (accessed on 20 January 2025).
  51. Zhang, S.; Bai, C.; Wen, L.; Qiu, G.; Lü, X. Gas-Particle Flow and Combustion Characteristics of Pulverized Coal Injection in Blast Furnace Raceway. J. Iron Steel Res. Int. 2010, 17, 8–12. [Google Scholar] [CrossRef]
  52. Cavaliere, P. Clean Ironmaking and Steelmaking Processes; Springer: Berlin, Germany, 2019. [Google Scholar]
  53. Miao, M.; Deng, B.; Kong, H.; Yang, H.; Lyu, J.; Jiang, X.; Zhang, M. Effects of volatile matter and oxygen concentration on combustion characteristics of coal in an oxygen-enriched fluidized bed. Energy 2021, 220, 119778. [Google Scholar] [CrossRef]
  54. Zhou, Z.; Wang, R.; Yi, Q.; Wang, G.; Ma, C. Combustion Enhancement of Pulverized Coal with Targeted Oxygen-Enrichment in an Ironmaking Blast Furnace. Processes 2021, 9, 440. [Google Scholar] [CrossRef]
  55. Sullivan, G.L.; Delgado Gallardo, J.; Jones, E.W.; Holliman, P.J.; Watson, T.M.; Sarp, S. Detection of trace sub-micron (nano) plastics in water samples using pyrolysis-gas chromatography time of flight mass spectrometry (PY-GCToF). Chemosphere 2020, 249, 126179. [Google Scholar] [CrossRef] [PubMed]
  56. Geyer, R.; Jambeck, J.R.; Law, K.L. Production, use, and fate of all plastics ever made. Sci. Adv. 2017, 3, e1700782. [Google Scholar] [CrossRef]
Figure 1. Images of (a) Subcoal® pellets, (b) milled Subcoal®, (c) milled coal and (d) mixture of 90% coal: 10% Subcoal® sample. Ruler showing markings in centimetres.
Figure 1. Images of (a) Subcoal® pellets, (b) milled Subcoal®, (c) milled coal and (d) mixture of 90% coal: 10% Subcoal® sample. Ruler showing markings in centimetres.
Chemengineering 09 00015 g001
Figure 2. Schematic of ultra-fast heating and pyrolysis of carbon feedstocks with coupled GCMS analysis of evolved gases.
Figure 2. Schematic of ultra-fast heating and pyrolysis of carbon feedstocks with coupled GCMS analysis of evolved gases.
Chemengineering 09 00015 g002
Figure 3. Pyrolysis-GCMS data for (a) coal, (b) Subcoal® and (c) PET—all pyrolysed at 800 °C.
Figure 3. Pyrolysis-GCMS data for (a) coal, (b) Subcoal® and (c) PET—all pyrolysed at 800 °C.
Chemengineering 09 00015 g003
Figure 4. Pyrolysis-GCMS data for milled mixtures of coal–Subcoal® pyrolysed at 800 °C showing (a) 90:10 w/w, (b) 75:25 w/w and (c) 50:50 w/w coal: Subcoal®.
Figure 4. Pyrolysis-GCMS data for milled mixtures of coal–Subcoal® pyrolysed at 800 °C showing (a) 90:10 w/w, (b) 75:25 w/w and (c) 50:50 w/w coal: Subcoal®.
Chemengineering 09 00015 g004
Figure 5. Pyrolysis-GCMS data for Subcoal® torrefied for (a) 1 h at 250 °C and (b) 1 h at 300 °C and pyrolysed at 800 °C.
Figure 5. Pyrolysis-GCMS data for Subcoal® torrefied for (a) 1 h at 250 °C and (b) 1 h at 300 °C and pyrolysed at 800 °C.
Chemengineering 09 00015 g005
Figure 6. Sample burnout data for low, medium and higher volatile matter coals, with varying Subcoal® loadings (0 to 5 to 10 to 20% shown as blue, red, green and purple, respectively).
Figure 6. Sample burnout data for low, medium and higher volatile matter coals, with varying Subcoal® loadings (0 to 5 to 10 to 20% shown as blue, red, green and purple, respectively).
Chemengineering 09 00015 g006
Table 1. Measured gross energy values (GHVs) and replacement ratios calculated from their C+H values from the ultimate analysis and GHV value using the HMB blast furnace model (GHV data are the average of three samples).
Table 1. Measured gross energy values (GHVs) and replacement ratios calculated from their C+H values from the ultimate analysis and GHV value using the HMB blast furnace model (GHV data are the average of three samples).
SampleSample NotesGHV (MJ/Kg)
(+/− 0.10) [Avg. of 3 Runs]
Predicted RR Using C+H and Heat Dissociation
Coal—Tata 7Milled26.800.80
Subcoal® RDFMilled26.490.46
Polyethylene terephthalate (PET)Milled (mm fragments)23.230.61
Coal 90% + Subcoal® 10%Co-milled26.450.80
Torrefied Subcoal® 250 °C 1 h N223.220.46
Torrefied Subcoal® 300 °C1 h N228.250.45
Table 2. Ultimate and proximate analysis of reference coal and Subcoal® (average of 3 samples) along with gross heat value (GHV) and net heat value (NHV). GHV and NHV are quoted in their usual units of kcal/kg. The values in parentheses are MJ/kg to compare with data in Table 1.
Table 2. Ultimate and proximate analysis of reference coal and Subcoal® (average of 3 samples) along with gross heat value (GHV) and net heat value (NHV). GHV and NHV are quoted in their usual units of kcal/kg. The values in parentheses are MJ/kg to compare with data in Table 1.
Ultimate Analysis
FeedstockC %H %N %S %O %GHV
(kcal/kg)
NHV
(kcal/kg)
Coal76.85.41.80.33.88144
(34.07 MJ/Kg)
7870
(32.92 MJ/Kg)
Subcoal®48.86.61.10.426.96286
(26.30 MJ/Kg)
5946
(24.88 MJ/Kg)
FeedstockAsh (%)Proximate Analysis
Moisture (%)
Volatile Matter (%)Fixed Carbon (%)
Coal8.42.412.776.4
Subcoal®10.04.073.812.1
Torrefied Subcoal®
250 °C 60 min20.10.268.211.2
300 °C 60 min36.30.252.011.5
DTF 1 Coal8.8oven-dried13.777.5
DTF 2 Coal7.7oven-dried21.171.1
DTF 3 Coal4.7oven-dried33.461.8
Table 3. Most abundant compounds in the volatile matter evolved during pyrolysis of coal, Subcoal® or PET.
Table 3. Most abundant compounds in the volatile matter evolved during pyrolysis of coal, Subcoal® or PET.
SampleCompounds IdentifiedRetention Time (s)Peak Area
(×108 Ion Count)
Raw CoalBenzene6892.1
Toluene8361.9
1,3-dimethylbenzene10040.3
Indene12030.7
Naphthalene13661.3
1-Methyl Naphthalene14710.68
Subcoal®Benzene68913.0
Toluene8388.15
Styrene100637.6
Indene120237.8
Naphthalene136411.7
Acenaphthalene166248.3
PETBenzene69231.1
Toluene83131.4
Ethylbenzene95824.1
Phenylethyne98315.3
Styrene10035.89
Benzaldehyde112911.5
Indene11995.86
Acetophenone123521.6
Ethanone,1-(2-methylphenyl)1360 84.2
Naphthalene13638.28
All Naphthalene derivativesn/a261
Benzoic acid13867.9
Xanthene195125.9
Table 4. Volatile matter identified during pyrolysis of coal–Subcoal® mixtures at 800 °C.
Table 4. Volatile matter identified during pyrolysis of coal–Subcoal® mixtures at 800 °C.
Coal: Subcoal® Ratio (w/w)Compounds IdentifiedRetention Time (s)Peak Area
(×108 Ion Count)
90:10Benzene68843.5
Toluene83835.1
Styrene100324.7
Benzoic Acid135224.5
Naphthalene136326.0
2-methylnaphthalene147017.9
Bi-phenyl152716.9
75:25Benzene68873.9
Toluene83447.6
Chlorobenzene95710.7
Styrene100326.4
Benzoic acid135315.7
Naphthalene136431.8
50:50Butadiene3625.63
Benzene6851.39
Toluene8342.55
Styrene10071.50
3-phenyl-1-propyne12030.87
Naphthalene13631.84
Table 5. Ultimate analysis of drop tube furnace samples (oven-dried).
Table 5. Ultimate analysis of drop tube furnace samples (oven-dried).
DescriptionCarbon (%)Hydrogen (%)Nitrogen (%)Oxygen (%)
Coal
DTF 181.43.32.13.2
DTF 268.94.01.70.2
DTF 377.44.91.43.9
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Jones, E.W.; Steer, J.; Ojobowale, F.; Marsh, R.; Holliman, P.J. Pyrolysis-GCMS of Plastic and Paper Waste as Alternative Blast Furnace Reductants. ChemEngineering 2025, 9, 15. https://doi.org/10.3390/chemengineering9010015

AMA Style

Jones EW, Steer J, Ojobowale F, Marsh R, Holliman PJ. Pyrolysis-GCMS of Plastic and Paper Waste as Alternative Blast Furnace Reductants. ChemEngineering. 2025; 9(1):15. https://doi.org/10.3390/chemengineering9010015

Chicago/Turabian Style

Jones, Eurig Wyn, Julian Steer, Fawaz Ojobowale, Richard Marsh, and Peter J. Holliman. 2025. "Pyrolysis-GCMS of Plastic and Paper Waste as Alternative Blast Furnace Reductants" ChemEngineering 9, no. 1: 15. https://doi.org/10.3390/chemengineering9010015

APA Style

Jones, E. W., Steer, J., Ojobowale, F., Marsh, R., & Holliman, P. J. (2025). Pyrolysis-GCMS of Plastic and Paper Waste as Alternative Blast Furnace Reductants. ChemEngineering, 9(1), 15. https://doi.org/10.3390/chemengineering9010015

Article Metrics

Back to TopTop