Introduction

Red pepper of Capsicum gender represents an indispensable fresh vegetable in Brazil, an important condiment and spices. Capsicum powder is one of the most important spices consumed worldwide [1]. In Brazil, Capsicum are integrated to agribusiness and have an important socioeconomic value, being a source of jobs, encompassing an annual production of 30 million tons representing the third higher worldwide vegetable production and consumption [1,2,3].

Furthermore, fuel supply has been an extensively discussed question and concerns about their future are increasing. There is a growing interest in agricultural wastes used as biomass especially for those countries having such resources [4, 5]. Biomass is a source of organic carbon, which can replace fossil resources by using pyrolysis [6, 7]. Thus, efficient biomass thermal modification technologies worldwide have been target of so many researches.

The efficiency of the biofuel depends on the chemical composition of biomass and the operating process characteristics [8]. It is necessary to know the physicochemical properties of this residue in order to know the thermochemical conversion process into biomass derived fuels [1, 9] to understand the structural changes and to determine the reactions during the process [10,11,12].

The delignification process from biomass waste has emerged as an attractive along with biofuel production [13, 14]. This is probably caused by their recalcitrant nature that prevents the release of monomeric sugars in conversions into components in a biorefinery [13,14,15]. Therefore, a fast, green, low-cost and efficient delignification method can promote better results to biomass reuse [13, 14, 16].

For these reasons, the aim of the present study was to investigate the thermochemical conversion of Capsicum, considering thermal delignification in oxidizing and inert atmosphere and different heating rates which may cause different characteristics at biomass combustion and pyrolysis processes. Heating rates varying may also cause different ignition/burnout temperatures and volatile yield of biomass, so it is thus essential to choose suitable conditions [17].

The minimum temperature that allows an ignites spontaneously of biomass without external source of ignition is defined as ignition temperature [18,19,20]. The ignition temperature of biochar is related to its safety in storage and delivery [21]. The burnout temperature of a biofuel is an indicator of its reaction degree. The higher the burnout temperature, the fewer the combustible components in the biofuel [22].

Different heating rates were used for the kinetic study to obtain the apparent activation energy and pre-exponential factor considering Kissinger–Akahira–Sunose (KAS) and Ozawa–Flynn–Wall (OFW) isoconversional methods. These methods allow this study without knowing the kinetic model of the reaction mechanism. An extensive and accurate knowledge of their characteristics is highly important. However, studies related are scarce. So, it is important to investigate these processes, which can provide the data for the better biofuel production.

Materials and methods

Sample preparation

The biomass of this study was Capsicum spp. Samples were washed and dried by oven drying Solab model SL-100/42 at 105 °C to constant mass [23]. Dried biomass was ground in a Wiley mill, MA048-Marconi, and sieving in a Solotest sieve, NBR# 200 (74 µm).

Ultimate and proximate analyses

Ultimate analysis was used to determine the C, H and N content of the biomass and biochar. This analysis was performed on a PerkinElmer Mark Model 2400 Series II.

Proximate analysis was based on standard ASTM D 1762–84 [24] and presents mass percent of the moisture, volatile matter (VM) content and ash content, whereas the fixed carbon (FC) content was determined by difference. It is the easiest and most widely used method to characterize a biomass fuel.

Calorific value

A HHV was calculated using two equations based on the carbon, hydrogen and nitrogen content [25]. An ordinary least squares regression and a partial least squares regression method are given in Eqs. 1 and 2, respectively:

$$ {\text{HHV}} = 1.87{\text{C}}^{2} - 144{\text{C}} - 2802{\text{H}} + 63.8{\text{CH}} + 129{\text{N}} + 20147 $$
(1)
$$ {\text{HHV}} = 5.22{\text{C}}^{2} - 319{\text{C}} - 1647{\text{H}} + 38.6{\text{CH}} + 133{\text{N}} + 21028 $$
(2)

where (C) is carbon, (H) is hydrogen and (N) is nitrogen content. HHV was calculated using an average from two methods and corrected for ash content, providing a dry ash free basis (daf).

Gravimetric yield and gravimetric yield factor

The gravimetric yield (GY) was obtained for biomass after pyrolysis process in a muffle furnace (Quimis-Q318S21). Dried biomass was put on porcelain pans during the pyrolysis process and was sustained up until temperature peaks of 250, 350 and 450 °C. The process resulted in biochar BC250, BC350 and BC450. The gravimetric yield of these biochar is determined by Eq. 3 [26]:

$$ GY = \left( {\frac{{m_{\text{f}} }}{{m_{\text{i}} }}} \right)*100 $$
(3)

where GY is the gravimetric yield of biochar (%); mf is the final mass after pyrolysis (g); and mi is the initial mass of biomass (g).

According to [26], gravimetric yield factor (GYF) is defined as the ratio between the gravimetric yield percentage and the fixed carbon rate and was determined for the biochar considering Eq. 4:

$$ GYF = \left( {GY/100} \right)*\left( {FC/100} \right) $$
(4)

where GYF is the gravimetric yield factor and FC is the fixed carbon (%).

Thermal analysis

Thermal analysis was carried out at three different heating rates 7.5, 15 and 20 °C min−1, using a simultaneous DSC-TG equipment, TA Instruments, model SDT Q600, from 25 °C to 800 °C. Air and nitrogen analytical degree was purge gas at a 120 mL min−1 flow rate. In order to eliminate the mass and heat transfer interference, small masses, about 5.0 mg, were considered and analyzed in alumina pans. The TA Instruments software provides the mass loss (TG) and the derivative mass loss (DTG) curves, loaded as a function of temperature and time at heating process.

Energy balance

Solid residues of biomass were analyzed considering the energy yields and dry ash free (daf) mass. These calculations were based on Eqs. 5 and 6 [27]:

$$ Y_{\text{mass}} \left( \% \right) = \left( {\frac{{m_{\text{biomass}} }}{{m_{\text{biochar}} }}} \right)_{\text{daf}} *100 $$
(5)
$$ Y_{\text{energy}} \left( \% \right) = Y_{\text{mass}} \left( {\frac{{HHV_{\text{biomass}} }}{{HHV_{\text{biochar}} }}} \right)_{\text{daf}} $$
(6)

Ignition and burnout temperatures methods

Intersection

The ignition temperature (Ti) is the temperature point with a decrease in the DTG curve [28]. The ignition temperature considered the TG curve and two methods, ‘‘Intersection Method (IM)” (Fig. 1a). The ignition index Di is calculated according to Eq. 7:

Fig. 1
figure 1

Determination of ignition (Ti) and burnout (Tb) temperatures considering the intersection method a and ignition temperature from deviation method b

Full size image
$$ D_{\text{i}} = \frac{{\left( {\frac{{{\text{d}}w}}{{{\text{d}}t}}} \right)_{\hbox{max} } }}{{t_{\text{p}} *t_{\text{i}} }} $$
(7)

where (dw/dt)max is the maximum combustion rate, tp is the corresponding time of the maximum combustion rate and ti is the ignition time.

The burnout temperature (Tb) is the oxidation process complete, and it is observed on DTG [28, 29]. Tb was determined from the single peak of a DTG curve [30, 31]. In Fig. 1a can be observed the Tb calculated using IM.

The burnout index (Db), used to analyze the thermal behavior of the samples, is described in Eq. 8:

$$ D_{b} = \frac{{\left( {\frac{{{\text{d}}w}}{{{\text{d}}t}}} \right)_{\hbox{max} } }}{{\Delta t_{{\left( {\frac{1}{2}} \right)}} *t_{\text{p}} * t_{\text{b}} }} $$
(8)

where (dw/dt)max is the maximum combustion rate, ∆t(1/2) the time zone of (dw/dt)/(dw/dt)max = 1/2, tp the corresponding time of (dw/dt)max and tb the burnout time.

Deviation methods

The deviation method (DM) showed the ignition temperature identified from the DTG curves [19], in oxidizing and nitrogen atmosphere, during combustion and pyrolysis process, respectively. Figure 1b shows the determination of the ignition temperature based on DM.

In both methods were considered TG and DTG curves at oxidative atmosphere and three different heating rates, 7.5, 15 and 20 °C min−1.

Kinetic study

Model-free methods are based on an isoconversional process considering activation energy as a function of the conversion degree a chemical process [32,33,34]. In this work, two model-free kinetics were considered aiming to understand the kinetic behavior of Capsicum spp. The one-step global model considers only a single reaction in thermochemical processes as follows [35, 36]:

$$ {\text{Biomass}}\, \mathop \to \limits^{k}\, {\text{Water}} + \left( {{\text{Volatiles}} + {\text{Gases}}} \right) + \left( {\text{Biochar}} \right) $$

where k is the rate constant of reaction and temperature dependence is described by the Arrhenius equation, 9:

$$ k = A{\text{e}}^{\left( - {\text{E}}_{\text{a}} /{\text{RT}} \right)} $$
(9)

where Ea is the apparent activation energy/kJ mol−1; T is the absolute temperature/K; R is the universal gas constant 8.31 J/K mol−1; and A is the pre-exponential factor (s−1).

A kinetic study considering the Arrhenius equation was previously described in [1]. The kinetic study in this paper was calculated using dried Capsicum spp. and Kissinger–Akahira–Sunose (KAS) and Ozawa–Flynn–Wall (OFW) models described below.

KAS model

KAS is an isoconversional model [33, 37] without any assumption concerning the method [32]. It was calculated considering Eq. 10:

$$ \ln \left( {\frac{\beta }{{T^{2} }}} \right) = \ln \left( {\frac{{AE_{\text{a}} }}{g\left( \propto \right)R}} \right) - \frac{{E_{\text{a}} }}{RT} $$
(10)

where Eα is activation energy and A is the pre-exponential factor at a given conversion degree, α, respectively. The temperature T is that the conversion α is reached at a heating rate β. The plot of ln (β/T2) versus 1/T slope gives −Ea/R [32].

OFW model

OFW is one of the most common and widely isoconversional models [33] to study the thermokinetic parameters [38] and was applied considering Eq. 11 [39].

$$ \ln \left( \beta \right) = C_{ \propto } - \frac{{E_{\text{a}} }}{RT} $$
(11)

where β is heating rate; Eα apparent activation energy; Cα function of the conversion degree α; R universal gas constant 8.31 J/K mol−1; and T the absolute temperature (K).

Thermodynamic parameter

Pre-exponential factor (A) is the frequency of collisions between reactant molecules and was calculated according to Arrhenius equation [39, 40], considering an intermediate value of β (i.e. 15 °C min−1) and Eq. 12. A reaction order was not previously considered.

$$ A = \beta E_{\text{a}} \exp \left( {\frac{{{\raise0.7ex\hbox{${E_{\text{a}} }$} \!\mathord{\left/ {\vphantom {{E_{\text{a}} } {RTm}}}\right.\kern-0pt} \!\lower0.7ex\hbox{${RTm}$}}}}{{RT_{\text{m}}^{2} }}} \right) $$
(12)

Results and discussion

Biomass characterization

These results show significant differences caused by the highly variable composition of biomass. There were large variations determined by the proximate and ultimate analyses and can represent the moisture contents and ash yields. It was found that the ultimate analysis provides a more detailed chemical composition and it is the most accurate comparing with proximate data which has low accuracy because it provides only an empirical composition of the biomass and biochar. Figure 2 shows changes in elemental composition at four different pyrolysis temperatures.

Fig. 2
figure 2

Changes in elemental composition as result of pyrolysis treatment at four different final temperatures

Full size image

The pyrolysis process caused a decrease in the C and H, which can be attributed to the dehydration and decarboxylation reactions during the process.

In Table 1 it is possible to observe the atomic ratio of H/C, N/C and O/C molar elements of biomass and biochar, and Fig. 3 represents a plot of the molecular ratios of hydrogen to carbon (H/C) and oxygen to carbon (O/C) based on the Van Krevelen diagram.

Table 1 Atomic ratio of biomass and BC250, BC350 and BC450
Full size table
Fig. 3
figure 3

Van Krevelen diagram for biomass and biochar

Full size image

The O/C and H/C ratio varied with increasing pyrolysis temperature with changes in the elemental composition between biomass and biochar, suggesting the increase in aromaticity during the pyrolysis process. It is possible to observe two separate groups: with higher O/C and H/C ratios due to the presence of hemicellulose and cellulose, representing the biomass and the second group, with lower O/C and H/C ratios, representing the biochar samples. The lower values observed in the biochar samples could be explained by a higher degree of aromaticity due to hydration, decarboxylation and decarbonylation promoting a carbon high content. Based on that, it is possible to conclude that the higher final pyrolysis temperature, the lower the atomic ratios, which can be related to higher degree of aromaticity, its recalcitrance and its stability.

The results for ultimate and proximate analyses, HHV, GY and GYF, obtained for biomass and all biochars, are shown in Table 2.

Table 2 Ultimate, proximate, calorific and gravimetric analyses of biomass and BC250, BC350 and BC450
Full size table

A greater variability was observed in the volatile and ash analysis, with the volatile content decreasing with higher temperatures. The variation in ash content was the opposite, being lowest in biomass and highest in biochar, mainly BC450. The highest HHV was given in BC350 corresponding to the highest values of carbon and fixed carbon.

The decrease in the GY with increasing temperature can be explained by a greater initial decomposition of biomass at higher temperature or to secondary decomposition after biochar formation.

The GYF decreased from 1.75 to 0.96 as the pyrolysis temperature increased from 250 to 450 °C. This may be related to the fast heating, which promotes rapid depolymerization of the solid material to more volatiles compounds. This comparison shows that the initial composition of biomass can vary largely and will be important to determining the final composition of the biofuel.

Thermal analysis

Figure 4 shows TG and DTG curves of Capsicum spp. at oxidizing (A, B and C) and inert atmosphere (D, E, and F) at three different heating rates.

Fig. 4
figure 4

Distributions of TG and DTG curves of biomass in air a 7.5, b 15 and c 20 °C min−1 and nitrogen d 7.5, e 15 and f 20 °C min−1

Full size image

Figure 4 also shows that lower heating rate is the temperature which occurs main thermochemical processes in both atmospheres. Using three heating rates, it was possible to better understand the difference between biomass thermal conversion process.

Different stages of mass loss were observed at the oxidizing atmosphere, and the first stage was due to the moisture loss and release of some light volatiles at temperature below 100 °C. The moisture content in the biomass was 8.92%, 9.58% and 10.10% for the heating rates of 7.5, 15 and 20 °C min−1, respectively, at the oxidizing atmosphere. It is possible to observe under inert atmosphere main steps of the water evaporation, devolatilization of thermally labile and more stable volatiles, and biochar formation (Fig. 4a, b and c).

The moisture content in the biomass was 9.09%, 9.74% and 10.78% for the heating rates of 7.5, 15 and 20 °C min−1, respectively, at inert atmosphere (Fig. 4d, e and f). The second step of thermal conversion is represented by gaseous released, a fraction of the condensed, preceding liquid fraction. Thermal decomposition of lignin is observed in a temperature range from 150 to 550 °C, and the mass loss rate is not as clear as in hemicellulose and cellulose which decomposition process was estimated in the range 160–320 °C with a maximum mass loss rate of about 250 °C characterizing thermal conversion of biopolymers. At temperature range 250–400 °C, it is possibly associated with carbonaceous matter formed earlier.

At inert atmosphere, above 400 °C, begin the carbonization phase and biochar formation. Cellulose and lignin degradation and decomposition are almost overlapping, and the thermal conversion of biomass is characterized by the mass ratio of its main components superposition. The maximum mass loss was observed at around 300–400 °C with 58% in inert atmosphere and 23% in oxidizing atmosphere, which suggested that biomass and biochar were more stable under inert conditions.

This behavior can be explained by particle thermal exchange influenced by the release of other particles under oxidizing atmosphere. Degradation process presented a pattern of almost samples varying according of atmosphere used. These steps do not occur uniformly or sequentially but are related to nonlinear chemical and thermal processes presented by this biomass. Differences at thermal decomposition and degradation for biomass samples have been related to the variations in the elemental and chemical compositions of the samples at different atmospheres. Total mass loss was higher at smaller heating rates which allow higher volatile matter and lower ash content conversion.

Considering a great number of pretreatment methods available to delignify biomasses, it’s possible to notice that thermochemical conversion can be considered a promising method of removing lignin from natural fibers since removal of lignin is a thermally sensitive process and varies with thermal conditions processes and with the nature of sample [41,42,43]. Generally, the major thermochemical conversion of the biomass occurs about 300 °C and 500 °C.

Considering the biomass pretreated thermally (BC250, BC350 and BC450), it’s possible to notice the difference between thermochemical conversion temperature (250 °C and 450 °C) and this behavior can be explained by previous delignification process. The content of lignin present in biomass can be changed by the effects of pyrolysis time and temperature. Removing lignin content of biomass during pyrolysis/torrefaction process can promote lower and higher values of activation energy and HHV, respectively, and an improvement in the carbon content and reduction in the oxygen content. The kinetic behavior of the biomass is related to these parameters and can explain an improvement in the thermochemical conversion of the biomass.

The overlapping of TG and DTG curves at 7.5 °C min−1 and 15 °C min−1 indicates that higher heating rates can promote mass and heat transfer effects not measurable and are not recommended to kinetic parameters determination.

Energy balance

The mass and energy losses of torrefaction process (250 °C) in nitrogen of Capsicum spp. showed that the fuel contains up to 65% of the original energy content in the solids. Analyzing energy balances for the pyrolyzed biomass, it was possible to observe that with increasing reaction temperature process becomes less endothermic and/or more exothermic. The volatilization of acid and organics has the highest influence on energy balance. Heat process generates a solid fuel with an increase in HHV converting biomass with low-energy into a high-energy feedstock. The thermal treatment not only destructed the fibrous structure and tenacity of biomass but also increased the calorific value. During thermal process, the biomass devolatilizes decreasing in mass, but the initial energy content preserved in the solid product. Thus, the energy density of biochar becomes higher than the biomass. As the energy of biochar is higher compared to biomass, larger transportation distances are possible. Moreover, thermal treatment causes hydrophobic characteristics, promoting storage more attractive because of the decomposition behavior. Figure 5 represents mass and energy variation to three different pyrolysis temperatures.

Fig. 5
figure 5

Mass and energy yields (dry ash free) for biochar at treated at temperatures of 250, 350 and 450 °C. *Samples: 1BC250; 2BC350; 3BC450

Full size image

Ignition and burnout temperatures

The determination of the ignition and burnout temperatures of biomass could be influenced by heating rate variation, biomass characteristics, particle size and volatile matter content. The samples utilized presented a small particle size (74 µm), and it is estimated that its ignition and burnout characteristics are directly related to heating rate increasing. The TG and DTG curves considered for ignition and burnout temperatures determination can be observed in Fig. 6.

Fig. 6
figure 6

Distributions of TG and DTG in terms of (% and  %/ °C) of biomass three heating rates

Full size image

According to IM, the ignition temperature of biomass is between 342 and 355 °C and tends to ascend when the heating rate increases. This is possibly due to the cellulose content in biomass. The ignition temperature based on DM is in the range of 210–250 °C, which is lower than those based on IM. Table 3 shows the different temperatures of the DTG peaks according to various heating rates studied.

Table 3 Temperature and intensity of DTG peaks at three heating rates of Capsicum spp.
Full size table

All ignition and burnout values can be observed in Table 4.

Table 4 Ignition and burnout temperatures of Capsicum spp. biomass
Full size table

The lower ignition temperature shows the high volatile contents of biomass and easy ignition. The burnout temperature indicates that oxidation of biomass occurred between 470 °C and 503 °C. These higher temperatures reveal that biomass needs a higher temperature during the combustion process, as well as longer residence time. Proximate values of burnout temperatures can be observed in the literature [31].

These temperature variations suggest that cellulose content changes the ignition characteristics and decomposition of lignin since the cellulose has compounds that are easily volatilized. At lower heating rates, it is possible to observe the lowest burnout temperature because of that the volatile matter presents some complex characteristics, being harder to decompose and lower ash content comparing other heating rates. The results also showed that the lower is the carbonaceous material of the biochar shorter is the burnout time.

Figure 7 represents the ignition and burnout temperatures variation and are directly related to heating rate. A higher heating rate presented higher Ti and Tb.

Fig. 7
figure 7

Ignition and burnout temperatures considering the intersection and deviation methods varying according to heating rates

Full size image

Ignition index agrees with the results of ignition temperature, and his higher value is indicative of better ignition performance. The influence of heating rates on the burnout time must be considered, so the burnout time represents burnout performance of biomass. The larger value of indexes represents better combustibility. The ignition and burnout index can be observed in Table 5.

Table 5 Biomass ignition and burnout characteristics
Full size table

Kinetic study

Biomass kinetic study showed complicated and unknown processes during thermochemical conversion, characterizing multi-step biomass degradation.

Figure 8 presents biomass degradation as a function of absolute temperature (Kelvin) at heating rates of 7.5, 15 and 20 °C min−1 and oxidizing atmosphere.

Fig. 8
figure 8

Biomass thermal behavior at different heating rates in air atmosphere

Full size image

The activation energy value can be observed in Table 6, and the highest values of Eα represent a greater difficulty for thermal conversion of biomass into biochar. This can be explained by the presence of mineral elements in biomass. This mineral composition can be a barrier, preventing the diffusion of heat and the release of volatiles during the carbonization process. The activation energy variation also suggests biomass complexity.

Table 6 Kissinger–Akahira–Sunose and Ozawa–Flynn–Wall biomass activation energy and pre-exponential factor
Full size table

At the temperature range of 460–720 K, the activation energy was between 61–193 kJ mol−1 and 72–201 kJ mol−1 for KAS and OFW, respectively. Activation energy variation can also show a difficult capability of ash formation indicating a thermal conversion efficiently. OFW model showed higher values than KAS method. Therefore, it is important to use more than one kinetic model, allowing a greater range of values and helping to understand the thermal behavior of biomass. Both models were considered in the agreement since they presented a coefficient of determination (R2) between 0.97 and 0.99. Figure 9 shows a comparison of activation energy for KAS and OFW.

Fig. 9
figure 9

A comparison of E(α) for KAS and OFW

Full size image

Conclusions

The biomass is an interesting biofuel source, not requiring the exacerbated expenditure of energy and time with the pyrolysis process. The values obtained for biomass suggest that it can be used in the biofuel production process, supplying small agribusiness. The ultimate analysis provides a more detailed chemical composition, and proximate data presented a low accuracy and provided an empirical composition of the biomass and biochar. The pyrolysis process promoted a lower C and H and can be attributed to the dehydration and decarboxylation reactions during the process. The lower ignition temperature shows the high volatile contents of biomass, and the burnout temperature reveals that biomass probably requires longer residence time or higher temperature to complete the combustion process. Biomass kinetic study showed the complexity of the biomass and unknown processes during thermochemical conversion, characterizing the multi-step biomass degradation.