The Influence of Addition of Expired Pharmaceuticals on Thermal Behaviour of Selected Types of Biomass

Abstract

The influence of 2 wt.% additives of expired paracetamol and naproxen on the thermal behaviour of densified samples of pea husks (PH), corncobs (CC), and sunflower inflorescences (SI) was studied using an analytical TG/FTIR unit. Gaseous, liquid, and solid pyrolysis products were evaluated using XRD, SEM, and EDX techniques along with FT-IR, ATR, and UV spectroscopies. It was found that the additives changed the yield and composition of pyrolysis products differently. The addition of paracetamol increases the contribution of guaiacyl rings in the condensed material of all samples, and the addition of naproxen—that of chromophores originating from the decomposition of lignin. The additives diversely affected the contribution of hydrocarbons in the composition of volatile products of pyrolysis: they decreased this contribution in PH samples, increased it in SI samples, and did not change in CC samples. The additives used changed the morphology and composition of organic and inorganic parts of pyrolyzed biomass. These changes in inorganics caused the changes in the composition of pyrolysis products. The conducted research proves not only the possibility of the utilization of expired pharmaceuticals during their pyrolysis with densified PH and CC samples but also the ability to reduce the undesirable hydrocarbons in the obtained volatile products.

1. Introduction

The accumulation of various types of agricultural biomass waste in the environment and the formation of larger amounts of organic wastes such as plastic waste, various types of sediments, and unused, expired pharmaceuticals make it necessary to develop methods of their effective commercial processing or rational disposal. The co-combustion of phytoremediation biomass of Sedum alfredii Hance and textile dyeing sludge [1] can be an example of the adjustment of waste disposal technology to a circular economy. Such a process of co-combustion made it possible to delay the peaks of release of SO₂, CS₂, and H₂S.

The literature on the subject contains lots of information about the thermal behavior of the blends of biomass waste or its components with plastics. The results of the conducted research showed that the additives from this waste cause changes in yield [2,3,4] and composition [4,5,6,7] of pyrolysis products. Moreover, the synergistic interaction between the components of blends affects the mechanism of reactions taking place during co-pyrolysis [8,9,10,11]. The thermogravimetric analysis data showed that a separate decomposition of the materials of two types in the composition of blends is observed during the co-pyrolysis of biomass with plastics [2,12]. Plastics undergo active thermal decomposition at temperatures higher than those of biomass. The products of their decomposition may not undergo secondary reactions with the products of biomass decomposition, which can result in the pollution of the environment with dangerous compounds.

Altogether, the aforementioned suggests that other organic wastes can also affect the thermal behaviour of biomass. Expired pharmaceuticals belong to such wastes. As Radioklinika reported [13], there are over 500 million prescriptions for non-steroidal anti-inflammatory drugs (NSAIDs) written annually worldwide, and 50 million tons of such pharmaceuticals are produced. The global NSAIDs market is expected to reach 24.35 billion USD by 2027 [14]. A revenue forecast for 2030 expects 31,287.73 million USD [15]. Most NSAIDs are available in Poland without any prescriptions. In its turn, the Medonet portal reports that Poland placed second in the EU in terms of medicine consumption without prescription [16]. Poles consume over 2 billion medicinal products every year, of which 90% are drugs that were not prescribed by a doctor. The possibility of free purchase without a prescription often results in long storage and uncontrolled accumulation of expired pharmaceuticals in households. There is a system for collecting expired medicines in pharmacies, from where medical waste is sent for thermal disposal. In 2021, the cost of collection and thermal disposal of 1 kg of cytotoxic and cytostatic drugs was PLN 4–5 (circa USD 1), whereas in 2022, it increased to almost PLN 30 (USD 7.50). The costs of the thermal disposal of psychotropic drugs are many times higher. This shows the need for the development of other, more economical methods of disposal of pharmaceuticals than their combustion.

Pharmaceuticals such as paracetamol and naproxen undergo active thermal decomposition at a temperature lower than that of the active decomposition of biomass [17]. Hence, the products of their decomposition can directly influence the course of biomass decomposition and, in this way, actively change such course. Therefore, the aim of this research is to explain the influence of two expired non-steroidal anti-inflammatory drugs containing paracetamol and naproxen on the thermal behaviour of three types of wastes of agricultural biomass—pea husks, corncobs, and sunflower inflorescences—in terms of an explanation of their possible disposal during pyrolysis with biomass.

2. Materials and Methods

2.1. Research Objects

The objects of research were the samples of three types of wastes of agricultural biomass—pea husks (PH), corncobs (CC), and sunflower inflorescences (SI)—along with a blend of 2 wt.% of expired pharmaceuticals (Apo-Napro produced by APOTEX EUROPE B.V., Leiden, Netherlands and APAP produced by US Pharmacia Sp. Z o.o., Wroclaw, Poland). Paracetamol (called N-(4-hydroxyphenyl) acetamide according to the IUPAC nomenclature) was the pharmacologically active substance in APAP tablets, whereas naproxen (called 2-(6-methoxynaphthalen-2-yl) propanoic acid according to the IUPAC nomenclature) in the Apo-Napro tablets. In the text, the Apo-Napro additives are shortened to NP, and the APAP additives are referred to as PR.

The pharmaceuticals used in the research undergo active thermal decomposition at the temperature of about 250 °C [17]. The structures of NP and PR contain rings and functional groups (Figure S1) similar to those occurring in biomass structure and in products of its decomposition. Moreover, paracetamol has a mobile hydrogen atom at the nitrogen atom in its structure. This hydrogen atom can take part in the reactions of redistribution of hydrogen with biomass destruction products whereas, as a result of the decomposition of naproxen, a naphthalene radical anion can be formed, which initiates the polymerization reaction involving electron transfer.

Table 1. Characteristics of the studied biomass.

Main Characteristics	PH	CC	SI
Cd [%]	40.99 ± 0.21	44.75 ± 0.20	39.46 ± 0.12
Hd [%]	5.77 ± 0.24	6.12 ± 0.01	5.59 ± 0.06
Nd [%]	0.03 ± 0.00	0	0.02 ± 0.00
Sd [%]	0	0	0
Ad [%]	9.59 ± 0.46	4.89 ± 0.73	13.9 ± 0.27
Oad [%]	43.62 ± 0.23	44.24 ± 0.31	41.03 ± 0.11
HHVb [MJ·kg−1]	16.40 ± 0.16	18.15 ± 0.31	19.49 ± 0.11
Si [mg·kg−1]	4210 ± 233	5521 ± 219	4009 ± 345
P [mg·kg−1]	233 ± 94	189 ± 79	1395 ± 172
S [mg·kg−1]	1628 ± 75	1172 ± 61	3292 ± 133
Cl [mg·kg−1]	475 ± 27	3176 ± 41	5995 ± 80
K [mg·kg−1]	475 ± 27	6684 ± 99	52,218 ± 325
Ca [mg·kg−1]	15,052 ± 331	6684 ± 99	18,636 ± 486

^(a) Calculated by difference, O [%] = 100 − C^d − H^d − N^d − S^d − A^d; ^(b) calculated by HHV [MJ·kg⁻¹] = 0.3491 × C^d + 1.1783 × H^d + 0.1005 × S^d − 0.0151 × N^d − 0.1034 × O^a − 0.0211 × A^{d (d)} dry basis.

presents the elemental composition and content of selected inorganic components in studied biomass. An Elementar Vario Micro Cube CHNS analyzer (Elementar Analysensysteme GmbH, Langenselbold, Germany) was used for the CHNS analysis. The contents of inorganic elements were determined by the ED-XRF technique. A Thermo Scientific Niton Goldd+ analyzer (Thermo Fisher Scientific Inc., Tewksbury, MA, USA) was used for this purpose.

The greatest content of C atoms is characteristic of CC sample that has the lowest value of A^d parameter. The greatest value of A^d parameter is characteristic of SI sample that has the lowest content of C atoms.

Before blending with biomass, the tablets of pharmaceuticals were ground to <0.2 mm. Before testing, the biomass samples were washed, dried first at room temperature, then placed in a dryer at a temperature of 105 °C, and then ground to a grain size of <0.2 mm. The tablets were formed from biomass and blended with the additives of expired pharmaceuticals; their densities are presented in

Table 2. Density of formed tablets.

Sample	PH	CC	SI
without additives	1.121 ± 0.019	1.016 ± 0.022	1.094 ± 0.015
with 2wt.% NP	1.068 ± 0.021	0.968 ± 0.033	1.042 ± 0.028
with 2wt.% PR	1.057 ± 0.024	0.959 ± 0.025	1.033 ± 0.031

.

2.2. Pyrolysis of Samples in a Tube Furnace

The studied samples were pyrolyzed in a tube furnace, PRC 70 × 708/110 M, manufactured by Czylok company (Jastrzębie-Zdrój, Poland) in a high-purity nitrogen flow with a flow rate of 330 mL·min⁻¹. The samples were heated to the temperature of 450 °C with a heating rate of 10 °C·min⁻¹. The samples were kept at this temperature for 30 min, and after turning off the heating, they were cooled under a nitrogen atmosphere. The formed volatile products of pyrolysis were passed through a layer of ice-cooled methanol in order to detain condensable products. Methanol was distilled after the heating was finished. The condensed material was kept at an ambient temperature in a VDL23 vacuum drier manufactured by Binder GmbH (Tuttlingen, Germany) for 48 h in order to obtain constant weight.

2.3. Spectroscopic Studies of Condensable Pyrolysis Products

The condensates were studied with the FT-IR (ATR) and UV spectroscopies according to the methodology described earlier [18]. For this purpose, a Nicolet iS10 spectrometer (Thermo Fisher Scientific Inc., Waltham, MA, USA) with a Smart MIRacle module (the ATR technique) was employed. A monocrystal ZnSe was used in the research. The spectra were registered in the wavenumber range of 4000–600 cm⁻¹. The height of the C=C band near 600 cm⁻¹ was selected as the reference one in all spectra. The baseline was corrected to eliminate the non-specifical background. The spectra were elaborated by OMNIC 9.2.86 software.

The solutions of condensates in acetonitrile were studied with a double-beam V-630 spectrophotometer manufactured by Jasco Corporation (Tokyo, Japan). The spectra were registered in the wavelength range of 195–350 nm and normalized at the wavelength of 195 nm. The spectra were elaborated on using Jasco Spectra Manager II software.

2.4. TG/FT-IR Test

The studied biomass samples without additives in loose and densified forms, along with the blends of biomass with additives in densified form, were heated in a Q50 thermobalance in a high-purity hydrogen flow with a heating rate of 10 °C·min⁻¹ up to 750 °C. The volatile pyrolysis products were directed by an interface to a Nicolet iS10 spectrometer in order to register their FT-IR spectra in the wavenumber range of 4000–600 cm⁻¹. All FT-IR spectra of volatile products were normalized with respect to the height of the CO₂ band in the range of 2400–2250 cm⁻¹. The DTG curves of densified PH, CC, and SI samples were deconvoluted according to the technique discussed in the work [18]. The deconvolution of DTG curves was conducted with regard to the temperatures of decomposition of biomass components presented in the works [19,20]. For this purpose, ONMIC 9 software was used, and the Gaussian/Lorentzian function was applied. The fixed baseline, minimum width of subpeak 5, and average level of noise equal to 5 were taken into account.

2.5. XRD Tests

The chars pyrolyzed at the temperature of 450 °C were studied by X-ray diffraction (XRD). To make a comparison of the ordering degree of C atoms in the structure of obtained chars easier, they were studied using the XRD technique with the use of an internal standard. NaF added to the studied samples in the amount of 10 wt.% was used as the internal standard. Reflex (002) from NaF in the diffractograms is observed near reflex (002) from the material of chars of samples. To make comparison easier, the diffractograms were normalized with respect to the reflex from NaF. An X’Pert Pro MPD diffractometer (PANalytical) equipped with a Cu-anode 1.8 kW X-ray tube with a linear exit window was employed. The diffractograms were registered in 2Θ angular range from 10 to 45 deg. The registration mode was U = 45 kV, I = 40 mA.

2.6. SEM Visualization

The chars of samples obtained at the temperature of 750 °C were studied with a scanning electron microscope (SEM) Quanta 3D FEG manufactured by FEI company (Hillsboro, OR, USA) at different magnifications. The accelerating voltage was 10 keV. Selected fragments of samples were subjected to an EDX microanalysis.

3. Results and Discussion

3.1. TG/FT-IR Test

In the FT-IR spectra of the volatile products of pyrolysis of samples of loose and densified biomass, the surfaces of bands from saturated and unsaturated hydrocarbons (the wavenumber range of 3200–2650 cm⁻¹), compounds with carbonyl groups (the range of 1900–1600 cm⁻¹), alcohols, phenols, and esters (the range of 1250–950 cm⁻¹) were measured. The ratios of the surfaces of these bands to the surface of the CO₂ band were calculated for different pyrolysis temperatures. Figure 1 presents the curves of changes in values of calculated ratios of surfaces of bands depending on temperature.

The data in Figure 1 imply that the densification of samples causes an increase in the relative amount of hydrocarbons emitted from the PH sample during pyrolysis and a decrease in that from SI and CC samples. Therefore, the densification of studied samples affects the emission of hydrocarbons during pyrolysis in distinct ways.

Figure 2a shows the TGA curves of densified samples were compiled, and Figure 2b shows their DTG curves.

The shape of the TGA curves of the studied samples (Figure 2a) points to their different thermal stability. The active decomposition of densified SI samples starts at temperatures lower than those of other samples. On the one hand, this may be facilitated by a high content of K atoms (Table 1) that, according to Eom et al. [21], should lower the temperature of biomass decomposition. On the other hand, the shape of the DTG curve of this sample (Figure 2b) suggests that its components undergo decomposition at different temperature ranges during pyrolysis. In turn, the CC sample has greater thermal stability than other samples. The weight loss in the CC sample to the temperature of about 300 °C is caused by the removal of moisture and LVS (light volatile substances). At temperatures of 300–375 °C, the CC sample loses weight much faster than other samples (Figure 2a), and there are two characteristic maxima in the DTG curve in this temperature range. The deconvolution of DTG curves was made (Figure 2c) on the basis of the data on the temperatures of decomposition of basic biomass components published in works [19,20]. It follows from the diagrams in Figure 2c that the studied samples differ substantially by composition. In the composition of the PH and SI samples, there are diverse amounts of moisture and LVS, extractives, hemicellulose, cellulose, and lignin, whereas in the composition of the CC sample, there is a high content of cellulose, probably of two types (amorphous and crystalline) [18] and much less lignin. A high content of extractives (20.9%) and hemicellulose (19.8%) in the SI sample substantially decreases the temperature of its decomposition and causes a greater weight loss below the temperature of 300 °C.

Figure 3 presents the results of a thermogravimetric analysis of the pyrolysis of the studied samples with NP and PR additives.

In Figure 3a, there is a noticeable tendency for an increase in weight loss in the blends of the PH sample densified with NP and PR at temperatures above 500 °C. Such behaviour of the sample can be caused by the interaction of volatile pyrolysis products with the surface of formed char, which was underlined by the authors of works [22,23,24]. Respectively, the shape of TGA curves in Figure 3c,e implies a lack of substantial influence of the NP and PR additives on the course of changes in the weight of CC and SI samples. As it follows from the shape of the curves in Figure 3b, the thermal stability of hemicellulose and cellulose (the height of the peaks near 250 °C and 325 °C increases) tends to decrease under the influence of the NP addition to the PH sample whereas the thermal stability of cellulose in PH sample under the influence of PR additive rises. In the blend of the CC sample with NP, the thermal stability of amorphous cellulose tends to decrease (Figure 3d). During the pyrolysis of the blends of SI samples with additives, there is a noticeable tendency of a decrease in weight loss rate at temperatures above 550 °C (Figure 3f). The aforementioned implies that the additives affect the course of weight loss of components in the studied samples in different ways.

Figure 4 presents the normalized FT-IR spectra of volatile products of pyrolysis of the PH sample and its blends with pharmaceuticals.

The shape of the spectra of volatiles proves that the additives used in blends cause a decrease in surfaces of the bands in the ranges of wavenumbers of 3250–2800 cm⁻¹, 1850–1550 cm⁻¹, and 1250–950 cm⁻¹. This consequently indicates a decrease in the contribution of saturated and unsaturated hydrocarbons, compounds with carbonyl groups, phenols, alcohols, and esters in the composition of volatile products of pyrolysis of blends with additives in the whole temperature range.

In Figure 5, the FT-IR spectra of volatiles of the CC sample and its blends with additives look completely different.

The lack of any changes in the shape and height of bands in the normalized FT-IR spectra of volatile products (Figure 5) implies that the additives used do not affect the composition of volatile pyrolysis products. This gives grounds to presume that the additives do not change the overall amount of hydrocarbons in the composition of volatile pyrolysis products in blends with the CC sample. The aforementioned suggests the possibility of the disposal of expired pharmaceuticals in blends with CC the sample without increasing the emission of environmentally harmful compounds during their combustion in densified form. In the case of the PH samples, the addition of 2 wt.% of expired PR and NP is strongly recommended in order to reduce the emission of dangerous compounds into the atmosphere during their combustion.

Figure 6 presents the normalized FT-IR spectra of the volatile products of pyrolysis of the densified SI sample and its blends with additives of the expired pharmaceuticals.

At the pyrolysis temperatures above 320 °C, there is a tendency for an increase of surfaces of the bands in the ranges of 1850–1600 cm⁻¹ and 1150–950 cm⁻¹ observed in the normalized FT-IR spectra. At higher temperatures, the surfaces of bands in the ranges of 3250–1800 cm⁻¹, 1850–1600 cm⁻¹, and 1150–950 cm⁻¹ increase. The noticed changes in the surfaces of bands imply that there is an increase in the contribution of saturated and unsaturated hydrocarbons, compounds with carbonyl groups, alcohols, phenols, and esters in the composition of volatile pyrolysis products from the blends of SI samples with expired pharmaceuticals. Therefore, the shape of FT-IR spectra in Figure 6 evidently suggests a negative influence of PR and NP additives on the composition of emitted volatile products of pyrolysis of blends and, hence, the impossibility of disposal of these expired pharmaceuticals during combustion of densified SI samples.

3.2. The Analysis of Influence of Additives on the Structural-Chemical Parameters of Obtained Condensates

The literature on the subject presents lots of information about the decomposition pathways of basic biomass components. During the primary reactions of biomass pyrolysis, anhydrosugars, light oxygenates compounds, pyran, and furan compounds emerge that further undergo secondary decomposition reactions with the formation of aliphatics, phenols, alcohols, esters, and other secondary compounds capable of secondary reactions [25,26]. The products of decomposition of the NP and PR additives can affect the thermal stability of the biomass components and interact with their biomass decomposition products. Such interactions can lead to changes in the composition of volatile pyrolysis products and condensable products.

The results of the analysis of the material from volatile pyrolysis products condensed in methanol are presented in Figure 7, Figure 8, Figure 9 and Figure 10. Particular bands in the ATR spectra were assigned the same way as in other works [18,27,28].

It follows from a comparison of the heights of bands and profiles of normalized ATR spectra of the PH sample and its blends with additives (Figure 7) that both additives increase the contribution of compounds with groups of C_al-H type (the range of 3000–2850 cm⁻¹) and fragments containing C-O stretching originating from cellulose (band 8). However, the PR additive increases and the NP additive decreases the contribution of H-bonds in the formation of the band of condensate in the wavenumber range of 3680–2400 cm⁻¹. This implies that the additives change the composition and contribution of polar compounds, which are able to form hydrogen bonds in different ways. The additives influence the height of other bands differently: the PR additive increases and the NP additive decreases the contribution of bands that originate from quinines and quinine methides (band 2), fragments of the destruction of aromatic skeletal vibrations of guaiacyl rings (band 4). The PR additive decreases, and the NP additive increases the height of band 1, to which the C=O vibrations of esters, ketones, and aldehydes are assigned.

Figure 8 presents the ATR spectra of products condensed in methanol that emerged during the pyrolysis of the CC sample and its blends with additives. The height of bands and shape of normalized spectra imply that the contribution of groups of C_al-H type in the range of 3000–2850 cm⁻¹ and polar compounds able to form hydrogen bonds decrease in the condensates from blends under the influence of additives. In the ATR spectra of condensates from blends, the height of bands 1, 6, 7, and 10 diminishes what corresponds to a decrease in the contribution of C=O vibrations of esters, ketones, and aldehydes, CH vibrations in fragments from lignin, guaiacyl ring along with compounds with 1–2-disubstituted aromatic rings. This suggests that the compounds originating from the decomposition of additives slow down the decomposition of cellulose and lignin or participate in secondary reactions with the products of the decomposition of cellulose and lignin. The activity of the NP additive in these processes is somehow higher than that of the PR additive.

Figure 9 presents a comparison of the shape and height of bands in the ATR spectra of condensates of the SI sample and its blends with expired pharmaceuticals.

Under the influence of PR and NP additives, the contribution of hydrogen bonds did not change in the normalized ATR spectra of condensate of the SI sample, whereas the contribution of compounds of C_al-H type in the range of 3000–2850 cm⁻¹ and the height of band corresponding to C-O stretching fragments originating from cellulose (band 8) decreased. The PR additive changed the heights of bands in the ATR spectra to some greater extent: the contribution of bonds of quinines and quinines methides (band 2), aromatic skeletal vibrations of guaiacyl rings (band 4), fragments containing a guaiacyl ring, C-O stretch in lignin, and C-O linkage in guaiacyl aromatic methoxyl groups (band 7) rose slightly.

The UV spectra of condensates originating from the densified samples and their blends with expired pharmaceuticals are shown in Figure 10.

The changes in absorbance in spectra presented in Figure 10 imply that the additives influence the contribution of compounds with chromophore groups in the composition of studied condensates in different ways. In the wavenumber length of 190–210 nm, the PR additive increases the absorbance in condensate from the PH sample but does not change the absorbance value in condensate from volatile products of CC and SI samples. Under the influence of PR additive, the absorbance in the range of wavenumbers above 220 nm grows for all condensates, which points to an increase in the contribution of compounds with chromophore groups. In the UV spectra of condensates with the NP additive, there are maxima near 200 and 260 nm. The maximum absorbance near 260 nm suggests that there is a quite high concentration of cyclic compounds with chromophore groups. Taking into account the results published in works [29,30], it can be suggested that, under the influence of the NP additive, there occurred a greater concentration of chromophores in the products of decomposition of lignin. The addition of PR to biomass samples causes not the formation of such high concentrations of chromophores but the formation of guaiacyl rings.

Thus, the FT-IR spectra presented in Figure 4 and Figure 6 show that the addition of expired pharmaceuticals can change the overall yield of hydrocarbons of various types in the composition of emitted volatile pyrolysis products. However, the data in Figure 7, Figure 8 and Figure 9 point out the occurrence of the same groups of atoms and functional groups in the composition of compounds condensed in methanol, but their contribution ratio in the formation of ATR spectra is changing.

The analysis of spectra in Figure 7, Figure 8 and Figure 9 shows that, under the influence of the PR additive, the height of band 4 increases in the ATR spectra of all samples, which implies the occurrence of aromatic skeletal vibrations of guaiacyl rings. A greater contribution of aromatic skeletal vibrations of guaiacyl rings in the composition of condensed material suggests that PR additive intensifies the decomposition of lignin. The analysis of the data presented above implies that there should be similar and distinct effects of the influence of additives on the composition of condensable and non-condensable volatile products of pyrolysis of studied biomass. However, such differences in the composition of pyrolysis products cannot be explained by distinct biomass composition only.

3.3. The Investigation of Chars of PH, CC, and SI Samples and Their Blends with Additives

The diffractograms of chars that were obtained at the temperature of 450 °C are compiled in Figure 11. At this temperature, the weight loss of samples varies from 30 to 35%.

It follows from the analysis of diffractograms in Figure 11a,c that, under the influence of additives, the parameter of surface ratio of (002) reflexes of chars and NaF (A_{002 char}/A₀₀₂ NaF) from blends of PH and SI samples with additives decreases whereas that one of blends of CC sample with additives increases (Figure 11b). The changes in the value of this ratio point to the changes in the amount of C atoms, causing a coherent X-ray scattering. Under the influence of additives, the reflexes from Mg(OH)₂ and sharp peaks from SiO₂ disappear (Figure 11a). In a similar way, the modification of inorganics takes place under the influence of additives from plastic waste [12].

As a result of weight loss with the rise of pyrolysis temperature, the contribution ratio of inorganics in chars should increase, and they can be more visible. The visualisation of surfaces of chars that were obtained at the temperature of 750 °C and the results of microanalysis of atoms present in chars are shown in Figure 12, Figure 13 and Figure 14. The SEM images of chars from the PH sample and its blends with additives are compiled in Figure 12.

The fragments of surfaces of the pyrolyzed samples having characteristic relief elements were selected for analysis. The visualization of surface morphology (namely ‘star-like’ elements in Figure 12a,f,k, and ‘bricks’ in Figure 12b,g,l, as well as the surface and interior of fibres in Figure 12c–e,h–j,m–o) was presented in order to specify the appearance and composition of sediment formed on them. No sediment was observed on the surface of a ‘star’ in the char of the PH sample without additives (Figure 12a) and ‘bricks’ (Figure 12b). On the surface of a ‘star’ of char from the PH sample with the NP additive (Figure 12f), there is some visible sediment in the form of small flat plates; ‘bricks’ are covered with a great amount of sediment (Figure 12g). Under the influence of PR additive, large amounts of drops of material probably condensed from the gas phase are formed on the surface of ‘star’ (Figure 12k), whereas no condensates are visible on the surface of ‘bricks’ (Figure 12l). The data of EDX microanalysis of the surfaces of ‘stars’ imply the changes in the composition of atoms under the influence of additives: the NP additive removes P atoms from the surface of ‘star’ (Figure 12f), whereas the PR additive removes Cl atoms (Figure 12k). Ca atoms prevail on the surfaces of ‘bricks’ in the chars with PR additive and without additives, and K, O, C, and Cl atoms are present in the char without additives.

On the surfaces of ‘bricks’ in the char from the PH sample with the NP additive, there is a large amount of powder-like sediment. There is a much greater variety of atoms in the sediment—Ca, K, Cl, Si, Mg, O, and C, but no S atoms. The powdered sediment was probably formed on the ‘bricks’ as a result of condensation of inorganics from the volatile pyrolysis products or also as a result of migration of atoms in the pyrolyzed material [31]. The visualisations of the surfaces and the porous interiors of fibres differ in the chars obtained with PR and NP additives or without any. On the flat surface of fibres in the char without additives (Figure 12c), there are small amounts of sediment that increase in case of the presence of Cl atoms in its composition (Figure 12d,e).

Under the influence of the NP additive, the interior of fibres undergoes a substantial weight loss (Figure 12h); there appears a specific fuzz made of numerous small scales (Figure 12i), which suggests the occurrence of a crystalline-like state in inorganics. In the composition of scales, there is a wide range of atoms present: C, O, Na, Mg, Si, P, S, Cl, K, and Ca. The occurrence of sodium atoms in the char of the PH sample with NP additive can result from the presence of sodium lauryl sulphate in Apo-Napro tablets because there are no Na atoms in the char of the PH sample. There are no visible traces of condensed sediment on the flat surface of the fibres (Figure 12j). However, in the composition of the material from char with the NP additive itself, there are Mg, S, K, and Ca atoms (Figure 12j).

Under the influence of PR additive, the drops of sediment are condensed on the surface and in the pores of fibres (visible at a magnification of M25k) (Figure 12m) and also in small pores on the flat surface (visible at a magnification of M100k) (Figure 12n,o). The aforementioned shows that PR and NP additives change the surface relief of the char of the PH sample and the composition and state of inorganics condensed on it (powder-like or crystalline-like) in different ways.

Figure 13 presents the images of chars of the CC sample and its blends with additives.

In the chars of the CC sample and its blends with additives, there are visible characteristic ‘plaques’ (Figure 13a,f,k,l). The spot microanalysis of the ‘plaques’ shows their similar composition: large amounts of the Si and O atoms and small amounts of the C and K atoms. The ‘plaques’ are located on the surface of the material (let’s call it matrix), the morphology of which changes under the influence of additives. The microanalysis of the area of a matrix with ‘plaques’ indicates a greater presence of other atoms, the type of which changes under the influence of additives. The char of the matrix without additives is porous (Figure 13a). The material containing C, O, Na, Si, and K atoms ‘hatches’ from the char of matrix with NP (Figure 13g,h). The char of a matrix with PR is plump (Figure 13k); its morphology differs from the chars of matrixes with NP additives and without additives. Unlike the char of matrix with NP, there are no Na atoms present in its composition. The analysis of the composition of inorganics in the chars of matrixes of the CC sample implies that a greater weight loss of organic material from the char without additives can be initiated by Mg, Cl, and P atoms. Aggregations of different shapes of inorganic objects formed (Figure 13c) as a result of the migration of inorganic atoms with volatiles on the surface of the char without additives (Figure 13b).

The presence of the same atoms is characteristic of the porous material of fibre walls of chars in Figure 13d,i,m. However, under the influence of additives, there are some round grains containing great amounts of K and Cl atoms (Figure 13i,m). Such grains were not found in the porous char of the CC sample without additives (Figure 13d).

Figure 13e,j,n visualise the interior of pyrolyzed fibres of the CC sample. The composition of the porous material of the interior of chars with additives (Figure 13j,n) differs from that of the char fibres without additives (Figure 13e): there are S and Na atoms present in the case of additives, and no such atoms were determined in the interior of char fibres without additives. Moreover, the char material of the CC sample with additives has an almost 2.5–3-fold content of K atoms. This gives a reason to presume that the used additives initiate the processes of migration of K atoms [32,33] in biomass samples during pyrolysis. It cannot be excluded that besides the K atoms, other atoms may undergo migration with volatiles.

Figure 14a–j present the visualization of the morphology of the surfaces of chars of SI sample without additives.

A large amount of inorganic components is characteristic of the SI sample (Table 1). The C, O, Na, Mg, Si, P, S, Cl, K, and Ca atoms are present inside the char of the SI sample without additives (Figure 14a), whereas the presence of the C, O, Mg, P, S, Cl, and P atoms (Figure 14b) was registered on the surface (Figure 14b). Such differences imply that the heterogenous decomposition of inorganic atoms is characteristic of studied chars because the Ca and Si atoms do not undergo migration during the process of pyrolysis. However, some atoms can migrate with volatiles and, hence, form some irregular networks that merge nanotubes (Figure 14b–d), having about 82 × 210 nm in size (Figure 14e).

The flat surfaces are visible in the chars without additives, on which some objects with distinct shapes grew: needle-like (Figure 14f) and scaly-like (Figure 14g). Despite the differences in morphology, these objects have similar compositions. The surface of char between ‘scales’ at a magnification of M25k shows porosity that may have been caused by intense gasification under the influence of K, Ca, and Mg atoms (Figure 14h). The ‘scales’ presented in Figure 14i have the shapes of hexagonal plates that grow directly from the char.

Under the influence of the PR and NP additives, some changes in the material of chars appear that are connected with the formation of areas of melted inorganics (Figure 14k,p) in the composition of which great amounts of K and O are present. It cannot be excluded that this is K₂O, which did not undergo melting at the temperature of 750 °C. Under the influence of the NP additive, thin plates of inorganics were formed in the chars (Figure 14k–m). There are C, O, Mg, P, S, Cl, K, and Ca atoms present in the composition of these thin plates, whereas in the char of SI sample with PR additive, there were some thick ‘scales’ formed that contain, apart from the aforementioned atoms, some Si atoms (Figure 14o). If there are no Si atoms on the surface of the char of the sample with PR, then a ‘powder-like’ sediment is formed (Figure 14p). Such ‘powder-like’ sediment is presented in Figure 14q, but it also contains Na atoms. Attention should be drawn to the fact that Na atoms are present in the char of the SI sample without additives (Figure 14a), but the analysis of images in Figure 14 shows that Na atoms do not occur on all morphological elements of the structure. A greater content of Mg atoms is characteristic of loose sediment that is visible in Figure 14p,q. The chars of SI biomass contain porous fibres. It follows from the comparison of Figure 14j,n,r that the interior of fibres in chars undergoes significant gasification (Figure 14n). The results of microanalysis of the inner pore surface imply that this can result from an increased amount of K atoms (above 80 wt.%) in the sediment accumulated on the walls.

The SEM images of the chars of the studied samples presented in Figure 12, Figure 13 and Figure 14 point out the changes in the morphology of inorganics during their pyrolysis under the influence of PR and NP additives. This implies that the changes in inorganics can catalyze the reactions taking place in biomass material during pyrolysis in different ways. Similar effects of the changes in catalytic activity of the alkali and alkaline earth metals were observed by Huang et al. [34], who investigated the influence of acid pretreatment on the process of gasification of Pteris vittata L.

4. Conclusions

• The influence of the addition of 2 wt.% of expired paracetamol and naproxen on the thermal behaviour of the samples of pea husks, corncobs, and sunflower inflorescences was studied.
• The additives of expired pharmaceuticals affected the contribution of hydrocarbons in the composition of volatile products of pyrolysis; they decreased this contribution in pea-husk samples, increased it in sunflower inflorescence samples, and did not change in corncob samples.
• Under the influence of both additives, the contribution of the fragments originating from the decomposition of cellulose increased in the condensed material of pea-husk samples, whereas it decreased in corncob samples.
• The addition of paracetamol to every sample promoted the decomposition of lignin and increased the contribution of guaiacyl rings in the condensed material, whereas the addition of naproxen increased the concentration of chromophores in this material, which caused the formation of the peak of absorbance near λ = 260 nm.
• The additives of expired pharmaceuticals changed the morphology and composition of the organic and inorganic parts of the pyrolyzed biomass. These changes caused the alterations in interactions between char and volatiles and modified the course of secondary reactions between the volatile products of pyrolysis.
• The positive effect of the influence of additives on the composition of volatile products of pyrolysis of pea-husk samples and the lack of any effect on the composition of volatiles of corncob samples proved that the paracetamol and naproxen additives can be disposed of with these types of biomass. The exact contribution of additives in the biomass samples will be determined in further studies.