1. Introduction
Nowadays, there is increasing pressure on the use of renewable sources of fuel in domestic boilers. The primary renewable energy source is plant biomass [
1]. A promising form of biomass is energy crops, which are usually compressed into pellets for combustion [
2]. The number of pellets made of alternative non-wood material, so-called agropellets, is continuously increasing. Agropellets are produced by pressing agricultural commodities, such as energy plants, rapeseed and cereal straw, waste, oilcake, and others [
3]. The combustion of agrofuels generates minimal greenhouse gases and other potentially hazardous emissions under optimal conditions relative to conventional fuels [
4]. Biomass is even considered neutral from the point of view of carbon dioxide production since the amount of carbon dioxide produced by combustion is comparable to the amount consumed by plants as they grow. The amount of these substances released during combustion is influenced by the composition of the fuel, the type of combustion equipment used, the setting of the combustion process, etc. One of the factors that significantly affects the combustion efficiency and potential emissions production is the characteristics of the biomass combusted. In addition to solid biofuels, there are also liquid and gaseous biofuels that are the product of solid biofuel transformation processes; however, this study does not focus on them.
Biomass is composed of organic and inorganic substances containing mainly carbon, hydrogen, and oxygen. In addition to these essential elements, there are also often nitrogen, chlorine, iron, and alkali metals [
5].
On the contrary, sulfur and heavy metals are only present in trace amounts compared to fossil fuels. The more of these elements the biomass contains, the higher the number of harmful substances will be released during its combustion. Moreover, the amount of these elements in biomass is greatly influenced by the type of biomass and the place of cultivation.
Emissions from biomass combustion can then be divided into three main groups:
Pollutants from incomplete combustion: CO, CxHy, tar, soot, unburnt hydrocarbon particles, hydrogen, and incompletely oxygenated nitrogen compounds (HCN, NH3, N2O).
Pollutants from complete combustion: nitrogen oxides (NO, NO2) and CO2.
Pollutants from trace elements of impurities: incombustible dust particles, sulfur, chlorine compounds, and trace metals (Cu, Pb, Zn, Cd) [
5].
The most monitored pollutants are carcinogenic, poisonous, and greenhouse gases. The most important pollutants are characterized in the following passage.
The quality of the combustion process determines the formation of carbon dioxide. The combustion of biomass is characterized by long-flame CO burning. Undesirable cooling results in the release of pure carbon (soot), resulting in significant heat losses. For this reason, the combustion and post-combustion chambers for biomass in the boiler bodies are much larger than for fossil fuels, and secondary or tertiary air is supplied to the flames. This results in improved combustion in terms of the chemistry of the reaction, which leads to a significant reduction in CO and unburned chemicals. In terms of sulfur oxides, biomass is considered ecological fuel compared to fossil fuels because the sulfur content from which sulfur oxides are produced during combustion processes is present only in low concentrations in biomass. Furthermore, the fuel releases large amounts of water vapor and hydrogen, with which sulfur reacts to form hydrogen sulfide (H
2S) [
5,
6].
Usually, about 0.5–5% of nitrogen is present in biomass [
1,
2,
3,
4]. All nitrogen content is converted into NO
x compounds during combustion. At temperatures of 700–800 °C, mainly N
2O is produced, which contributes to the greenhouse effect. At temperatures above 1000 °C the formation of NO prevails, which is unstable and oxidizes to NO
2, which is involved in the creation of photochemical smog, possibly due to a reaction with water to form acid rain (HNO
3) [
7]. Domestic boilers, however, usually do not reach temperatures that lead to the formation of NO to such an extent [
5]. Nevertheless, the values of NO
x emissions produced by the combustion of different biomass types with varying contents of nitrogen show an apparent effect of the increased nitrogen content in non-woody biomass on total NO
x emissions [
8].
Chlorine is present in biomass in the form of inorganic and organic compounds. The fundamental problem caused by these substances in the flue gas is their reactivity and the high ability to corrode the materials they come into contact with. It is released into the environment during the combustion of fuels containing chloride (e.g., coal and some plant materials and wastes). Chlorine reacts with airborne water vapor to form hydrogen chloride. Hydrogen chloride gas is rapidly converted to hydrochloric acid, which contributes to the formation of acid rain [
9].
One of the critical factors in terms of the optimization of the combustion process, construction of the fireplace, and distribution of combustion air distribution into primary, secondary, and possibly tertiary air is the proportion of volatile combustible material [
10]. Increased portions of volatile combustible materials and a lack of secondary or tertiary air will lead to an increase in unburned chemicals and products of incomplete combustion (CO, C
xH
y) [
3,
5,
10,
11].
Emissions of particulate matter (PM) are also a significant problem in combustion. The formation of PM during biomass combustion is closely related to the release of inorganic substances and alkali metals from the fuel. These substances are fuel ash, and therefore the formation of PM is closely associated with the composition of fuel ash, specifically and predominantly with the number of alkali metals in the ash [
12]. The polluting particles themselves are usually composed of the K, Cl, and S elements, which form aerosols and alkali metal sulfates, chlorides, and carbonates. The critical element in the composition of the dust particles is potassium, which is usually found in the form of K
2SO
4, KCl, and K
2CO
3 [
13,
14]. PM emissions may also be related to the phosphorus content of the fuel. Combustion of agropellets with a high phosphorus content produces PM consisting of the chlorides mentioned above, carbonates, and sulfates, plus an increased amount of phosphates [
15].
Since all the emissions above and fuel behavior in combustion processes are related to the biomass composition, it is always necessary to know its properties, such as moisture, ash content, elemental composition, or lower heating value to optimize it.
This study aimed to investigate the fuel properties (such as coarse and elemental analysis, or lower heating value) of several new, potentially usable biofuels, such as quinoa, camelina, crambe, safflower, and compare them with some traditional biofuels (wood, straw, sorrel, hay). The obtained data can contribute to the expansion of the biofuel portfolio in energy production.
2. Materials and Methods
The section summarizes the subsections containing the description of tests, procedures of determination, processing of measured data, and formulas used for calculation of the monitored values. Determination of dry matter, water content, ash amount, and loss during annealing, determination of volatile combustible content, elemental analysis (C, H, N, S), determination of calorific value using the calorimetric method, and calculation of the lower heating value were performed.
For determination of the dry matter and water content of solid biofuels, three different gravimetric procedures were used based on standards ČSN EN ISO 18 134-1–3 [
16,
17,
18], which were used depending on available sample amount. ČSN EN ISO 18 134-1 is a reference method that was used when a large amount of sample was available. The method in the calculation also included the so-called buoyancy effect on the hot sheet on which the analyzed sample was dried. The sample was weighed with an accuracy of 0.1 g. The result was calculated using the formula (1):
where:
—mass of empty sheet for sample (g),
—mass of sample sheet before drying (g),
—mass of sample sheet after drying (g),
—reference sheet mass before drying (g), and
—reference sheet mass after drying (g).
ČSN EN ISO 18 134-3 is a method that was used when only a limited amount of sample was available. A smaller sample volume was compensated for in this method by higher weighing accuracy requirements. The weighing was carried out only with wholly cooled samples. Both methods mentioned so far utilized oven drying at 105 °C until there was a constant mass. In the second case, the result was calculated according to Equation (2):
where:
—mass of empty crucible with lid (g),
—mass of crucible with sample and lid before drying (g), and
—mass of crucible with sample and lid after drying (g).
To determine the ash content of solid biofuels and the loss on annealing, a procedure based on the standard ČSN EN ISO 18 122 (Solid biofuels – Determination of ash content) [
19] was used, where the sample was annealed in the furnace at 550 °C until a constant sample mass was reached. The result was then calculated as a percentage for both the raw and the anhydrous sample according to Equations (3) and (4):
Determination of ash content in the anhydrous sample:
where:
—mass of empty dish (g),
—mass of dish with test portion (g),
—mass of dish with ash (g), and
—the water content of the test portion used for the determination (%).
Determination of ash content in the raw sample:
where:
—mass of empty dish (g),
—mass of dish with test portion (g), and
—mass of dish with ash (g).
The determination of the volatile combustible solid biofuels content was performed gravimetrically according to the standard ČSN EN ISO 18 123 (Solid biofuels–Determination of volatile combustible content) when the biofuel sample was annealed at 900 °C for 7 min in a porcelain crucible with a lid inside an oven [
20]. The resulting mass percent of volatile combustible in the sample was calculated using the following Equations (5) and (6):
Determination of volatile combustible content in an anhydrous sample:
where:
—mass of empty crucible with lid (g),
—mass of crucible with sample and lid before heating (g),
—mass of crucible with sample and lid after heating (g), and
—the percentage of the mass of the sample water content (%).
Determination of volatile combustible content in a raw sample:
where:
—mass of empty crucible with lid (g),
—mass of crucible with sample and lid before heating (g),
—mass of crucible with sample and lid after heating (g), and
—the percentage of the mass of the sample water content (%).
Furthermore, the percentage of carbon, hydrogen, nitrogen, and sulfur in the sample was determined using elemental analysis and the oxygen content was calculated. The elementary analyzer Vario Macro cube CHNS (Elementar company) was used for the analysis, working on the principle of sample combustion in a catalytic tube, separation of different gases from monitored components by adsorption-desorption on columns, and subsequent detection using a thermal conductive detector [
21]. The measured concentrations of individual elements in the original sample were also recalculated for combustible and dry matter according to the following Equations (7)–(9):
Determination of elemental content in a biofuel sample:
From the measured concentration values in the original sample (wt%) of carbon
, hydrogen
, nitrogen
, and sulfur
in the raw sample, the oxygen concentration
was calculated assuming that the elements C, H, N, S, and O constituted all the combustible content in the sample:
where
is the ash content in the original sample (wt%).
When determining the concentrations of C, H, N, S, and O in a combustible content, it was necessary to assume that these elements together made up all the combustible content and water in the original sample. The water in the original sample consisted of only the elements H and O. From the molar masses of H and O, it was possible to determine the mass fraction of the given elements in water (H
2O):
where:
—hydrogen mass content in water (-),
—oxygen mass content in water (-),
—hydrogen molar mass (kg·mol−1), and
—oxygen molar mass (kg·mol−1).
By subtracting water from the original sample, the concentrations of H and O were reduced, while the concentrations of C, N, and S were maintained, as seen in Equations (10) and (11):
where:
—reduced hydrogen concentration (wt%),
—reduced oxygen concentration (wt%),
—oxygen concentration in the original sample (wt%),
—hydrogen concentration in the original sample (wt%),
—percentage by mass of hydrogen content in water (-), and
—percentage by mass of oxygen content in water (-).
Concentrations
,
, and
in the original sample, along with the reduced concentrations
and
, together form real ratios related to the combustible content. These have to be recalculated to make up 100% of the combustible content; for a calculation example, see Equations (12) and (13):
where:
—the concentration of carbon in the combustible content (wt%),
—hydrogen concentration in the combustible material (wt%),
—carbon concentration in the original sample (wt%),
—reduced hydrogen concentration (wt%),
—nitrogen concentration in the original sample (wt%),
—sulfur concentration in the original sample (wt%), and
—reduced oxygen concentration (wt%).
Subsequently, the remaining concentrations were calculated for , , and in the combustible content similarly.
A simple relation was used to convert the concentrations of the elements C, H, N, S, and O in the combustible content to the concentrations of individual elements in the dry matter (only the sample relation for C is described here):
where:
—the concentration of carbon in dry matter (wt%),
—carbon concentration in the combustible content (wt%), and
—combustible content in dry matter (wt%).
The concentrations , , , and in the dry matter were subsequently calculated.
Subsequently, the calorific value of the selected materials was determined using an IKA C 200 calorimeter (IKA company) or a 6100 Compansated Calorimeter (Parr Instrument Company) following the standard ČSN EN ISO 18125 (Solid biofuels – Determination of higher and lower heating values). The principle was to burn the weighed analytical sample in an oxygen atmosphere at high pressure in a calorimeter vessel. The measured higher heating values determined by both calorimetric methods indicate the higher heating value of the original sample
. The following Equations (15) and (16) were used to convert the higher heating value of the original sample
to the higher heating value of dry matter
and the higher heating value of the combustible content
[
22]:
where:
—the higher heating value of the combustible content (kJ·kg−1),
—higher heating value of dry matter (kJ·kg −1),
—higher heating value of the original sample (kJ·kg −1),
—combustible content in dry matter (= loss by annealing in dry matter) (wt%), and
—dry matter content in the sample (wt%).
The lower heating value could then be calculated from the higher heating value using Equation (17). The lower heating value is defined as the higher heating value released by burning 1 kg of fuel minus the condensation heat of the water produced by combustion. In accordance with ČSN EN ISO 18 125 [
23], Equation (17) was chosen to determine the lower heating value of the original sample
:
where the concentration of combustible hydrogen in the original sample
was calculated using Equation (18):
where:
—lower heating value of the original sample (kJ·kg−1),
—higher heating value of the original sample (kJ·kg −1),
—the evaporation heat of water at 20 °C has a value of 2454 (kJ·kg −1),
—concentration of water in the sample (wt%),
—concentration of combustible hydrogen in the original sample (wt%),
—concentration of hydrogen in the original sample (wt%), and
—combustible content in the original sample (wt%).
For the calculation of the lower heating value of the dry matter
, Equation (19) was used:
where the concentration of hydrogen in dry matter
was calculated using Equation (20):
where:
—lower heating value of dry matter (kJ·kg−1),
—higher heating value of dry matter (kJ·kg −1),
—the evaporation heat of water at 20 °C has a value of 2454 (kJ·kg −1), and
—concentration of hydrogen in dry matter (wt%).
The following Equation (21) was used for the conversion from the lower heating value of dry matter to the lower heating value of the combustible content:
where:
—the lower heating value of the combustible content (kJ·kg−1),
—the lower heating value of the dry matter (kJ·kg−1), and
—combustible content in the dry matter (wt%).
3. Results and Discussion
Table 1 summarizes the measured water content values
, which was determined using the gravimetric method described in the previous section. The water content is an important parameter that affect fuel quality. Above all, it directly affects its lower heating value by reducing the dry matter content and by consuming heat to evaporate water during combustion [
24]. During combustion, the combustion temperature may fall below the optimum value due to evaporative heat consumption.
Consequently, there is a risk of incomplete combustion of fuel and the generation of above-the-limit emissions [
25]. If the flue gas temperature drops below the dew point, water condensation will occur, leading to an acceleration of the flue gas corrosion of the combustion device [
26]. Ideally, the moisture of the material to be combusted is less than 15% in the case of pellets or less than 20% in the case of loose material. As can be seen from
Table 1, the water content ranged from 3.82% to 11.92%, which meant the materials were suitable for combustion. The pellets had very low moisture contents, which partially caused the pellets to crumble and break. Low moisture in a very narrow range of values is influenced by storage in a dry and warm fuel storage environment. The standard deviation and the confidence interval were calculated for the average water content. From the moisture content, the dry matter content in the sample was found range between 88.08% and 96.18%.
After determining the moisture content and dry matter content, the ash contents of the raw and anhydrous samples were determined, and the loss during annealing and the ballast fraction were calculated from these data. After finding the water content, the ash content is another important parameter that characterizes the examined fuel sample.
Table 2 shows that the lowest ash content of 0.3% was found in a wood pellet sample, which corresponded to the fact that only wood mass was present almost entirely in this sample. By contrast, in the case of agro-materials, the ash content is higher: hay 4.83%, sunflower 3.92%, and safflower 6.6%. An increased content of ballast substances was evident, which also corresponded to the increased value of the calculated ballast portion. The highest ash content was determined in samples with a high percentage of waste sludge present due to the increased occurrence of heavy metals and other hazardous elements contained in the combusted material. This phenomenon is disadvantageous for the material to be burned because the increased ash content during the combustion makes the boiler operation more challenging in terms of removing the ash from the boiler body and faster filling of the ashbin.
The ash content for the selected commodity may also vary depending on the different regions from which it is extracted. In plant and woody materials, the ash content is greatly influenced by the content and composition of substances derived from the soil, whose composition varies in different locations. For this reason, the ash content can only be compared approximately. For example, in Barbanera and Cotana [
27], the ash content in the dry matter of the digestate was 12.38%, whereas in the sample digestate we analyzed, 11.31% ash was found. Similar values were found in safflower seed (3.0%) [
28], sunflower peel (2.7%) [
29], and wheat straw (6.72%) [
30].
Another variable characterizing the fuel is the ballast portion
. As mentioned, it is the proportion of substances reducing the lower heating value of the fuel. The ballast ratio values largely correspond to the ash value. As can be seen in
Table 2, low amounts of ballast were observed in the case of wood material, with increased values found in the analyzed agro-materials and the highest values were reached for the material containing waste sludge. It was precisely in the waste sludge that the non-combustible components were concentrated, which in turn significantly reduced the lower heating value of the material. For this reason, waste sludge is often used in mixed pellets with varying proportions of woody or plant biomass.
The evaluation of the rough analysis of the selected samples is subsequently shown in
Figure 1 and
Table 3. The content of water, ash, and combustible content in the chosen materials varied greatly, as can be seen from the table below.
After carrying out and evaluating the rough analysis of the materials intended for combustion, the determination of the volatile contents in the raw and anhydrous samples was carried out. The volatile content, together with the solids, make up the total combustible content in the samples. An example is given in the following
Table 4.
As can be seen in
Table 5, the volatile content value ranged from 47.49wt% to 81.30wt% for the raw sample and 51.45wt% to 88.42wt% for the anhydrous sample, with average values of 74.4wt% and 79.4wt%, respectively. The volatile content value may vary within one material, as was noted for the safflower. For whole seeds, the value was 73.4wt%. On the other hand, for peels, the volatile content was higher (79.15wt%). In the safflower pellets after the oil press, the volatile content was 74.3wt%. This pellet contained both the seed and the peels. The values of the determined safflower volatile content approximately corresponded to the 83wt% found in another study [
28]. A similar trend was observed in the case of camelina.
After determination of the volatile content, the elemental analysis was carried out to determine the carbon, hydrogen, nitrogen, and volatile sulfur content of the sample, and the calculation of the oxygen content was added. The measured concentrations of individual elements in the original sample were also converted to the content in the combustible and dry matter. The measured and calculated values of the elemental analysis are summarized in the following
Table 6,
Table 7 and
Table 8.
It is apparent from
Table 6 that wood materials reached very similar values for all monitored elements. The values from the wood samples were close to the measured percentages of elements in the samples of hay and straw, which in terms of elemental analysis, seems to be a suitable fuel that could replace wood pellets. However, a slightly increased sulfur content (up to 0.11wt%) was observed with these samples. The increased sulfur content was also observed for some oilseed samples, such as mustard (1.08wt%), sunflower (up to 0.2wt%), camelina (up to 0.9wt%), and cocoa (0.11wt%). This was similar to information found in other literary sources [
29,
30,
31,
32]. The combustion of sulfur-containing material releases its volatile content, which subsequently reacts with hydrogen to form hydrogen sulfide, or with oxygen to form sulfur dioxide. The low presence of sulfur in the raw material monitored only meant the formation of a negligible amount of these gaseous emissions in the combustion process.
In
Table 7, the contents of the monitored elements in the combustible were recorded. The conversions given in
Section 2 were used to obtain these values. Compared to the elements in the raw sample, a slight increase in nitrogen and carbon content, and a decrease in the amount of hydrogen and oxygen, were observed for the combustible content. The change of these values was influenced by the reduction of water and ash in the combustible content.
In the case of
Table 7, there was a significant decrease in other elements to the detriment of hydrogen and oxygen. However, as already mentioned, the final concentration of sulfur and nitrogen in the material was mainly influenced by the particular soil composition in which the biomass was grown and the use of fertilizers. Higher sulfur concentrations in pellets increase the SO
2 emissions and may also cause corrosion when sulfur compounds condense on the exchanger surfaces of the boiler [
33]. The content of elements in the combustible was converted to the content of elements in the dry matter. The results are summarized in
Table 8 below.
After elementary analysis, the higher heating values of individual materials were determined using the calorimetric method. From the higher heating value of the original
sample, the higher heating value was then calculated in the combustible
and dry matter
. The measurement was performed at least three times, and the mean result was calculated from the measured values. The measured and calculated higher heating values fluctuated in a relatively wide range. The amount of woody mass greatly influenced these values in the sample. It is also evident from
Table 9 that the presence of oily substances in the material had a significant influence on the value of the higher heating value. This was observed, for example, in the case of camelina and safflower samples. In the case of camelina samples obtained from the same sources, the effect of the oil content was noticeable. The whole seeds reached a significantly higher value of the higher heating value of 25.154 MJ·kg
−1, as opposed to the already pressed seeds, which had a value of 20.942 MJ·kg
−1. The tables also show that the higher heating value of the combustible was greater than that of the original sample. In fact, in the case of fuel, the carrier of energy was only combustible. The remaining fuel, ash, and water only reduced this energy of combustible content. This was evident, for example, with digestate having a high higher heating value of combustible and a low higher heating value of the original sample. This significant difference was due to the high content of ballast, i.e., ash and water, in the sample.
The lower heating values of the original sample
, dry matter
, and combustible
were then calculated from the experimentally determined higher heating values. The lower heating value of the original
sample is a quantity that indicates the final energy potential of the sample during real combustion. This is the primary parameter for comparing potential fuel, whatever the type of material. The lower heating values showed a similar trend to the higher heating values, i.e., the lower heating values of wood pellets and some oil-containing pellets, such as mustard (22.363 MJ·kg
−1), camelina (23.280 MJ·kg
−1), and safflower (21.567 MJ·kg
−1) were high. On the other hand, digestate samples and the mixture of sawdust with a high waste sludge content showed a low lower heating value. When comparing the higher heating values and the lover heating values of these samples, it was observed that these values were influenced by the high ballast ratio. The results of the lower heating values of all analyzed samples are summarized in the following
Table 10.