Fe-modi ﬁ ed activated carbon obtained from biomass as a catalyst for α -pinene autoxidation

The presented work describes the autoxidation of alpha-pinene for the ﬁ rst time using a catalyst based on activated carbon from biomass with introduced Fe. The raw material for the preparation of the carbon material was waste orange peel, which was activated with a KOH solution. The following instrumental methods characterized the obtained catalyst (Fe/O_AC):N 2 adsorption at 77 K, XRD, UV, SEM, TEM, X-ray microanalysis, and catalytic studies. It was shown that the Fe/O_AC catalyst was very active in the autoxidation of alpha-pinene. The main reaction products were: alpha-pinene oxide, verbenone, verbenol, and campholenic aldehyde. .


INTRODUCTION
The preparation of carbon materials from inexpensive and widely available precursors, such as waste biomass from the food industry, has become the main topic of interest in new material research 1-3 . The development of a method of synthesizing carbon materials from biomass enables the use and management of waste, the disposal of which and environmentally safe storage for it has been problematic 4 .
The cultivation of citrus trees is considered to be the most developing agricultural sector in recent years. The citrus fruit grown on the most signifi cant scale is the sweet orange (Citrus sinensis), which accounts for up to 70% of citrus fruits' total production and consumption 32 . These fruits are processed on an industrial scale into juices, jams, and marmalades 33 . It is estimated that the citrus fruit processing industry generates nearly 40 tons of waste annually, a signifi cant proportion (~65%) of orange peels 34 . The high content of organic matter in the orange peel makes it challenging to dispose of it quickly because it affects the bacterial fl ora of the soil. The storage of such waste requires neutralizing the acidic pH of this biomass 35 . The demand for fresh orange juices and citrus preserves is constantly growing, so it is so important to minimize waste during production. Converting waste biomass to carbonaceous materials helps lower the production costs of activated carbons and is a potentially more economical alternative to available commercial carbons 36 . However, before turning waste orange peels into carbonaceous materials, they can be used to extract a very valuable terpene compound: limonene. Limonene itself and its oxidation as well isomerization products have many applications in organic synthesis and the cosmetics, food, perfumery, and cosmetics industries 37- 43 .
The preparation of activated carbons from orange peels on a laboratory scale is of interest to many researchers. Chen et al. 44 described the preparation of activated carbons from orange peels by pyrolysis at various temperatures and with limited oxygen supply. The carbons obtained in this way can be used as adsorbents for removing toxic compounds from the environment, e.g., naphthalene and 1-naphthol. Giraldo et al. 45 reported the preparation of catalysts prepared by coating with a layer of TiO 2 activated carbon obtained from orange peel. In the biomass impregnation cell, KOH was used as a chemical activator, and the pyrolysis of the carbonaceous substrate was carried out in a vertical tubular reactor at 550 o C. The resulting material was then coated with titanium dioxide and used in the NO X decomposition. Dhorabe 46 described the synthesis of activated carbons from orange peels, which was a waste product from a local marketplace. The researchers used orthophosphoric acid as a chemical activator and then heated the impregnated biomass in a muffl e furnace at 350 o C. Another team of scientists described the pyrolysis of orange peel in a carbon dioxide environment 47 . Biomass pyrolysis was carried out in a tubular reactor, the fl ow of N 2 and CO 2 was 300 ml/min, and the temperature range was from 200 o C to 700 o C. Hui Pan et al. 48 reported the preparation of sulfonated carbons from fresh orange peels by partial hydrothermal carbonization. Concentrated sulfuric acid(VI) was used for chemical activation. The materials obtained in this way were active in the esterifi cation of oleic and citric acids.
Alpha-pinene is a cyclic compound from the group of monoterpenes. This compound can be obtained from turpentine obtained from pine tree resin 49. Alpha-pinene occurs in nature as a mixture of two enantiomers, the content of which in the mixture may vary depending on the species, age of the tree, and the tissue from which it was obtained (e.g., a needle or xylem) 50, 51 . This compound is used extensively in the perfumery, cosmetics, food, and pharmaceutical industries 52 . This compound is used extensively in the perfumery, cosmetics, food, and pharmaceutical industries. In addition to industrial applications, this compound also exhibits anti-infl ammatory, antibacterial, antiseptic 53 , and anti-cancer properties 54 . Alpha-pinene has also been used as a valuable substrate in organic syntheses (oxidation (including autoxidation) and isomerization reactions) 49, 55-60 .
There are two ways of conducting the alpha-pinene autoxidation process: autoxidation without the presence of catalysts 61 or with the use of catalysts 62 . In the fi rst step, the reactions were carried out at 100 o C using oxygen and without the presence of light. The main products of this reaction were oxygen derivatives such as alpha-foam, verbenone, and myrtenal 61 . The autoxidation of alphapinene, which required the use of catalysts, was carried out in the temperature range of 65-90 o C. Metal-modifi ed catalysts (Cr, Co, Cu) were used in the reaction. The main products that were obtained during this reaction were alpha-pinene oxide, verbenol, and verbenone 62 .
Currently, conducting catalytic reactions in the presence of catalysts is at the center of interest of scientists dealing with catalytic processes. CoCl 2 in the presence of oxygen, acetic acid, and acetonitrile as a solvent was used to carry out the catalytic autoxidation of alphapinene. The main products were alpha-pinene oxide, alpha-campfolaldehyde, and verbenol and transverbenol. The highest selectivity was recorded for alpha-pinene oxide, and it was 31%, and transverbenol 21%, and also for verbenone 26%. 55% alpha-pinene conversion was achieved 63 . In another work, Mao et al. 64 used the Fe(III)/SiO 2 composite as the catalyst. The reaction was carried out in the presence of pure oxygen and using acetonitrile as the solvent. The main products were α-pinene oxide (selectivity was 21%), verbenol (selectivity 45%), and verbenone (selectivity 18%), and the α-pinene conversion was 73%. Gomes et al. 65 carried out alpha-pinene autoxygenation reactions using a Co(OAc) 2 / bromide catalyst. The main reaction products were cisverbenyl, verbenone, and myrtenal. The selectivities of the main products were 30%, 20%, and 10% respectively. Wróblewska 49 carried out autoxidation of alpha-pinene on TS-1 catalysts with different Ti content. The main products of autoxidation were alpha-pinene oxide, verbenone, and verbenol. It was found that under the most favorable reaction conditions (80 o C, catalyst content 1 wt.%, reaction time 24 h) for the TS-1 20:1 catalyst, the selectivity of alpha-pinene oxide was 30.25 mol%, and the conversion of alpha-pinene reached 41.81 mol%.
The aim of the work is research on the autoxidation of alpha-pinene on Fe/O_AC catalyst. According to the current knowledge, catalyst based on activated carbon obtained from biomass doped with Fe has not been described for this. It has not been used in the process of alpha-pinene autoxidation. The work presented by us illustrates the possibility of utilizing waste such as orange peels from food processing to produce activated carbons. Then, the synthesis of the catalyst based on activated carbon and the use of the catalyst obtained in this way in the catalytic autoxidation of alpha-pinene.

Preparation of activated carbon
The pulp residues were removed from the harvested orange peels and then air-dried at room temperature for 24 hours. Then the material was placed in a dryer at 70 o C for 24 hours. The dried biomass was grounded and chemical activated with saturated KOH solution for 3 hours at ambient temperature. The carbon substrate was then dried for 19 hours at 200 o C and carbonized at 800 o C for 1 hour under a nitrogen fl ow of 18 l/h. After the carbonization process, the carbon sample was rinsed with distilled water until the fi ltrate reached neutral pH. Then it was fl ooded with 1M hydrochloric acid and left for 19 hours. The sample was rinsed again with distilled water until the fi ltrate pH was approx. 7. The carbon material thus obtained was dried for 19 hours at a temperature of 200 o C. The obtained activated carbon was designated as O_AC.

Preparation Fe/O_AC
To apply the iron particles on the carbon surface, 1 g of activated carbon (O_AC) was weighed and suspended in 400 ml of iron salt solution (7.8 g FeCl 3 x 6H 2 O and 3.8 g FeSO 4 x7H 2 O) and then placed on a magnetic stirrer. Then the carbon solution was heated on a magnetic stirrer (700 rpm) to 80 o C and kept at this temperature for 90 minutes. After this time, 5M NaOH solution was added dropwise to the material until the pH was about 11, and then it was left on a magnetic stirrer at 80 o C, 700 rpm for 1 hour. After impregnation was completed, the sample was removed from the magnetic stirrer and left in the fl ask for 24 hours at ambient temperature. The material was poured with distilled water until the fi ltrate pH was constant -about 7. After the fi ltrate was neutral, the excess water was evaporated from the sample by heating it on an electric cooker. Then the material was dried in an oven at 100 o C for 24 hours. The obtained catalyst was designated Fe / O_AC

Alpha-pinene autoxidation
The reaction of alpha-pinene autoxidation was carried out in a glass reactor with a capacity of 5 cm 3 , equipped with a refl ux condenser and a stirrer. The fl ask was placed in an oil bath set on a magnetic stirrer with a heating function. 2 g of alpha-pinene (98%, Aldrich) was used for the autoxidation studies. The catalyst activity was tested under the following conditions: the reaction temperature range was 70-120 o C, the amount of the catalyst was 0.5-5 wt.%. Samples were taken after 2 hours and 24 hours. The most favorable conditions were determined Figure 1. Main products of alpha-pinene autoxidation based on the conversion and selectivity of alpha-pinene oxide. Other products determined during the tests were: verbenone, verbenol, and campholenic aldehyde (Fig. 1).
To perform the quantitative analysis, the reaction mixture was centrifuged and dissolved in acetone in a weight ratio of 1:10. The quantitative analysis was performed with a Thermo Electron FOCUS chromatograph equipped with an FID detector and a ZB-1701 column (30m x 0.53mm x 1um). The operating parameters of the chromatograph were as follows: helium fl ow 1.2 ml/ min, injector temperature 220 o C, detector temperature 250 o C, furnace temperature isothermally for 2 minutes at 50 o C, increase at a rate of 6 o C/min to 120 o C, at 120 o C isothermally for 4 min, then rising at 15 o C/min to 240 o C.
To determine the composition of post-reaction mixtures, the method of internal normalization was used.

Characterization
For textural characterization was used ASAP Sorption Surface Area and Pore SizeAnalyzer. Nitrogen adsorption isotherm was measured at -196 o C. Before adsorption measurements sample was outgassed at 250 o C for 19 hours. The specifi c surface area was calculated using the Brunauer-Emmett-Teller (S BET ) equation from nitrogen adsorption isotherms. The total pore volume (V tot ) was evaluated from the nitrogen volume adsorbed at a relative pressure of ~0.98. The volumes of micropores were calculated by the DFT method based on nitrogen adsorption. The pore size distribution (V mic(N2) ) was obtained from the DFT model in the ASAP 2460 Version 3.01 software package based on the N 2 sorption isotherm. DFT used: N 2 at 77 K on carbon (slit N 2 -DFT Model adsorption).
The morphology of the sample was observed via scanning electron microscopy with cold emission (UHR FE-SEM Hitachi SU8020). The sample preparation for SEM involved the sprinkling of the powder sample on a double-sided carbon tape mounted on the SEM stub. Images were taken with a 5 kV accelerating voltage using a triple detector system.
The morphology of the sample in greater detail was examined using high-resolution transmission electron microscopy (HR-TEM, Tecnai F20). The catalyst was mounted on a copper grid.
The infrared spectra were acquired at room temperature with a Nicolet 380 ATR-FTIR spectrometer (Thermo Scientifi c, USA). Sixteen scans were averaged for each sample in the range of 4000-400 cm −1 .
The X-ray diffraction (XRD) patterns of the catalyst were recorded by an Empyrean PANalytical X-ray diffractometer using Cu K (λ = 0.154 nm) as the radiation source in the 2θ range 10-80 o with a step size of 0.026.

Characteristics of the catalyst
The analysis of nitrogen adsorption-desorption allows determining parameters of the carbon catalyst's porous structure: specifi c surface area, total pore volume, and micropores. The tested catalyst has surface area values of 452 m 2 /g and a total pore volume of 0.496 cm 3 /g. The micropore volume estimated based on N 2 adsorption was 0.139 cm 3 /g. Fe content in the tested material was at the level of 26.67%. The above results show that it was possible to obtain a carbon catalyst with a relatively high specifi c surface area with simultaneous high metal content in the catalyst. Figure 2 shows the adsorption-desorption isotherm for the Fe/O_AC catalyst. In the UPAC classifi cation, the isotherm corresponds to Type-II. Except for the pressure range p/p0 =0.2-0.8, in which it is approximately parallel to the x-axis, which makes it similar in this pressure range to the Type-I isotherm and proves a very small share of mesopores. The isotherm in this section is similar to the Type-I isotherm. Very weak adsorption can be seen here, up to the relative pressure p/po of 0.8. With such high pressures, the macropores are fi lled. Quantachrome Instruments is not used to measure macropores, but it can be concluded that the pore diameter corresponding to p/p0 1 is 784 nm. The visible hysteresis loop is defi ned as the H3 type, which suggests that there are slotted pores.  Figure 3 shows the pore size distribution (PSD) of the Fe/O_AC catalyst, determined by the DFT method based on N 2 adsorption at -196 o C. Based on Fig. 3, the presence of micropores and the absence of mesopores can be found, which is consistent with the conclusions drawn based on Fig. 2. The DFT method allows the calculation of the pore size up to 100 nm. Taking into account that the maximum pore diameter determined by this method is 784 nm, it can be concluded that only macropores larger than 100 nm are present in the sample. Based on Fig. 2, Fig. 3, and the values of textural parameters, it can be concluded that macropores dominate in the sample. There are also micropores and the proportion of mesopores is negligible. It is also possible to estimate the macropore volume up to a diameter of 784 nm by subtracting V mic from V tot . Approximate macropore volume up to 784 nm is 0.357 cm 3 /g.
According to JCPDS card 88-0315 (cubic, Fd-3m) and 79-0416 (cubic, F-43m), the XRD plot showed characteristic peaks of magnetite. Based on the location and the shape (bordering and intensity) of the peaks, the two fi rst brother signals were identifi ed as magnetite JCPDS card 79-0416. The rest of the signals were assigned to magnetite JCPDS card 88-0315. As all the observed peaks were assigned to the magnetite phase, no peaks for the potential hematite phase could be seen.
The SEM micrographs of Fe/O-AC are shown in Figure  5. SEM images give an idea about the structure and shape of the material. Activated carbon has was a smooth, plated structure with slits ( Fig. 5a). At magnifi cation equal to 400 000 Fe 3 O 4 , nanoparticles on the activated carbon surface were seen. The size of the nanoparticles was about 10-20 nm.
The detailed morphology and structure of the nanoparticles were further investigated by TEM (Fig. 7). The Fe 3 O 4 nanoparticles had a semispherical shape with an average size of 18 nm. Figure 6 shows the EDX spectrum of Fe/O_AC. The sample contained only carbon and iron. The origin of a very small and bright signal about 1.5 keV is aluminum stubs.
The FTIR spectrum of Fe/O_AC catalyst is shown in Figure 8. The characteristic band at 1628 cm -1 and wi-deband between 3000 and 3750 cm -1 is attributed to the adsorbed water's existence. In specifi c, the broad bands (3000-3750 cm -1 ) can be assigned to OH-stretching vibration mode of adsorbed water and the bending vibration of H 2 O is observed at 1628 cm -1 . The appearance of two well-defi ned peaks between 440 and 640 cm −1 is due to iron-oxygen (Fe-O), which confi rmed that the synthesized nanoparticles are iron oxide 66 .

The alpha-pinene autoxidation
In the fi rst stage of research on the Fe/O_AC catalyst activity, the infl uence of temperature on the course of alpha-pinene autoxidation was checked. The reaction parameters were as follows: temperature in the range of 70-120 o C, amount of catalyst 1 wt.%. in relation to alpha-pinene, reaction time 2 and 24 hours. Figure 9 and Figure 10 show the effect of temperature on alphapinene autoxidation. The highest conversion of alphapinene (37.5 mol%) was obtained after 24h at 90 o C. Above, and below this temperature, the alpha-pinene conversion is lower. The selectivity of alpha-pinene oxide decreases with the progress of the reaction, from 25-30 mol% after 2 hours to 0-15 mol% after 24 hours. This is due to post-reactions where alpha-pinene oxide reacts to form campholenic aldehyde. As the reaction progresses, the selectivity of verbenol and verbenone also increases. Above 110 o C, after 24 hours, a signifi cant decrease in the selectivity of verbenol and the lack of alpha-pinene oxide in the post-reaction mixture can be noticed; it is related to the dimerization and polymerization reactions that take place at higher temperatures. The most preferred reaction temperature is 90 o C due to the highest conversion of alpha-pinene in the studied temperature range.
The studies on the effect of the catalyst amount are presented in Figure 11 and Figure 12. The highest alpha-pinene conversion value (37.5 mol%) for the alpha-pinene oxide selectivity (14.9 mol%) was obtained after 24 hours with the catalyst amount of 1 wt.%. For higher catalyst contents in the post-reaction mixture, a signifi cant decrease in the selectivity of alpha-pinene oxide and a decrease in conversion can be noticed. This is due to the very low amount of oxygen in the reaction mixture, resulting in subsequent reactions taking place

LITERATURE CITED
in the presence of the catalyst in the absence of an adequate amount of oxidant. Taking into account the conversion of alpha-pinene and the selectivity of alpha--pinene oxide after 24 hours, the most preferred catalyst content is 1 wt.%.

CONCLUSIONS
The paper presents an ecological method of processing waste such as orange peels into activated carbon and then the use of the material obtained in this way for the synthesis of a Fe-doped carbon catalyst. A relatively high specifi c surface characterizes the obtained catalyst with a Fe content of 26.67 wt.% and the presence of micropores. The research showed that the obtained Fe/O_AC catalyst was an active catalyst for alpha-pinene autoxidation. The main products of this reaction were: alpha-pinene oxide, verbenone, verbenol, and campfolenic aldehyde. These compounds are used in the perfumery and cosmetics industry due to their fragrance properties. The catalyst allowed to obtain the conversion of