1. Introduction
Most studies about gas emissions in solid biofuels are focused on their combustion, concerning both environmental and health issues. Therefore, greenhouse gases produced during biomass combustion and pyrolysis have been widely studied [
1,
2,
3,
4,
5,
6]. However, pollutants can be released by other means, not only by thermal degradation. Some gases, although at lower levels, are also off-gassed during their storage as a result of chemical and/or biological processes, such as auto-oxidation of fatty acids [
7,
8,
9,
10,
11]. Among them, carbon monoxide (CO) should be taken into account due to its harmful characteristics to users and workers that carry out maintenance tasks [
6]. As a matter of fact, certain fatal accidents due to CO inhalation have been reported in maintenance workers [
8].
CO is an odorless and colorless gas (which makes it unnoticeable) that is produced in the combustion process, and also in the decomposition of organic matter, which is a common mechanism during wood storage [
12]. The health risks related to its inhalation are well-known [
7]. CO replaces oxygen, joining hemoglobin. This replacement drastically reduces the oxygen supply to vital organs, with the subsequent fatal effects [
13,
14]. Knowing that CO is produced during wood storage, the existing studies show that CO generation in wood pellets is possibly due to spontaneous auto-oxidation of fatty acids in the wood [
15].
Experimental data shows that even small quantities of stored pellets are enough to generate lethal levels of CO in confined spaces, even though oxygen levels were normal [
8]. Huang et al. connected gas emission during storage with certain variables, such as the oxygen quantity, headspace [
16], temperature, and relative humidity [
17]. In these cases, the influence of these parameters was important to assess and control CO levels. Other authors have studied several kinetic models to predict gas emissions during pellet storage [
18] or shipping [
19,
20] and their possible connection with certain variables in order to thoroughly understand the formation of this compound and avoid its accumulation in confined spaces. Although CO does not contribute directly to the greenhouse effect, it is important from an environmental point of view as it has an influence on OH radicals and its amount affects, indirectly, to the formation of other greenhouse gases, such as methane and the tropospheric ozone.
Most studies about CO generation during pellet storage (aforementioned) have been carried out with commercial products (coniferous woody pellets, mainly), and, therefore, the analysis of other raw materials could be interesting.
Eucalyptus wood is abundant in the southwest of the Iberian Peninsula. It is estimated that there is about 71,000 ha of this kind of tree in the region of Extremadura (Spain) [
21], but it is less and less used due to the low activity of the paper industry in this area. Because of that, the use of eucalyptus wood is becoming more important for densified solid biofuel production. Similarly, waste from cork (biomass from forest industry) is increasingly used for energy purposes, either to produce pellets or directly as biofuel in industrial boilers for water heating or steam generation [
22]. However, there are no studies about CO generation during the storage of these by-products.
In this research work, CO emissions off-gassing from stored pellets made of cork and eucalyptus were analyzed, assessing different influence factors (especially ventilation conditions).
3. Results and Discussion
3.1. Proximate and Ultimate Analysis
The main characteristics of the eucalyptus and cork pellets are shown in
Table 5. The pellets were stored at room temperature, as explained in
Section 2.1.1.
The main characteristics of the two by-products obtained were similar to the results found in the literature for eucalyptus [
35,
36] and cork wastes [
37]. In this research work, the moisture levels were low, especially for the cork pellets. The ash content was slightly high in both cases. This could be due to the fact that the eucalyptus bark was not removed and the cork wastes included the external part, which contains more impurities than the inner part.
The carbon percentage was higher in the cork waste (51.00%) than in the eucalyptus pellets (44.40%). This fact could be important when it comes to CO generation. The N and S levels were low in both cases, so polluting emissions derived from these elements were avoided. It should be pointed out that both durability and bulk density was higher in the cork pellets than in eucalyptus, implying higher quality of the former.
3.2. Fat Content
Table 6 shows the results of fat content for the analyzed samples (both powder and pellet).
The fat content was much higher in the case of the cork waste, especially for the powder. This could imply, therefore, a higher amount of fatty acid for these samples, compared to eucalyptus.
On the other hand, other factors such as moisture content, temperature and the fatty acid profile of each kind of wood could have an influence on the process. Then, the higher concentration of fatty acid in the raw material could influence the whole CO process, depending on its chemical structure and auto-oxidation kinetics. It was observed that the fat content of the pelleted products was lower than the original residues in both cases. This was due to two factors: On the one hand, to the loss of light volatiles caused by the temperature rise in the pelletizing process, and on the other hand, to less effectiveness in the extraction process executed in the experiment, due to the reduction of hexane penetration.
3.3. CO Concentration
The results are shown in the following Figures.
Figure 3 shows CO concentration for each sample during the sample collection (in experiment 1), to determine the maximum CO levels for each headspace in the sample. The sampling time refers to the time that the probe was within the vessels during the measurement.
In the case of the E samples,
Figure 3 shows that the CO concentration increased up to a maximum (in this case at 29 ppm, after 45 h of storage without air renewal) from which it decreased to negligible values. This was due to the measurement system. The CO concentration was diluted due to the air entrance in the container when sampling (avoiding as much as possible this fact). This was the reason why different containers were chosen to measure the rest of the sealed samples, as it was a destructive detection system and the samples could not be reused.
3.3.1. CO Concentration without Air Renewal
Figure 4 shows the maximum CO levels over time, in non-ventilation conditions for each sample (experiment 1).
In the case of the E samples, it could be observed that, at the initial storage stages, the CO concentration was low, but increased considerably as the storage was longer. Thus, a stable value was reached after 185 h. Some authors have pointed out the important role of temperature in CO emissions during storage. Although the pellet samples were stored at room temperature, they reached high temperatures in the manufacturing process. Therefore, a certain amount of light volatiles, including fats, was released. This fact could explain the initial and significant rise in CO [
38,
39]. After 185 h, the CO concentration decreased slowly. The same behavior, at similar times, was observed by other authors [
38]. This could be due to the fact that the stored pellets decomposed their organic matter and were not able to generate more CO, possibly on account of oxygen depletion.
3.3.2. CO Concentration with Air Renewal
Figure 5 shows the maximum CO levels after 72 h, for each sample and its corresponding air renewal (experiment 2).
It could be observed that the trend of CO maximums was similar to the case of the sealed containers. Nevertheless, the values for air renewal were slightly higher. Therefore, air renewal could imply higher CO generation, possibly due to the oxygen supply to auto-oxidation of fatty acids. When ventilation took place, the exhausted atmosphere in the sealed containers could renew air and oxygen, making the CO generation more intense.
For the ET samples (2-L containers for the eucalyptus pellets), the same headspace volume was obtained with a greater number of pellets (that is, the free-volume/mass ratio was reduced). It can be observed that the profiles and trends were similar to the E samples, but the emissions were approximately 45% higher. It could be considered, therefore, that higher amounts of pellets, even though with the same headspace (that is, the free space between the pellets and the top of the container), will generate higher CO concentrations. This could be explained by the fact that auto-oxidation of fatty acids is a balanced reaction and addition of excess reagent tends to shift the reaction balance to the right, that is, to generate more products.
Regarding the cork samples, a similar trend was observed for the eucalyptus sample regarding the CO concentration, as could be observed in the 1-L containers. However, the CO generation was higher. Thus, the raw material composition, especially with regard to fatty acids, could play an important role when it comes to auto-oxidation and the subsequent CO emission during storage [
40]. In this case, the drop in CO generation versus storage time was higher than for the eucalyptus samples, where this decrease was more gradual. In the case of the cork pellets, there seemed to be more intensive CO generation at initial stages, completing the auto-oxidation process in shorter periods compared to eucalyptus pellets. The higher fat content (and possibly the fatty acid content) of cork, compared to eucalyptus, could explain this fact [
41,
42,
43].
3.4. Limit Value for Occupational Exposure
The values obtained exceeded, in any case, the 20 ppm limit, but not 30 ppm in all cases. Consequently, the workers that should carry out maintenance tasks on an ongoing basis in storage silos of pellets that are similar to those studied in this research work must wear some kind of respiratory protection.
The 100 ppm limit was exceeded only in one of the experiments with the cork waste pellets and air renewal. Therefore, maintenance workers performing tasks in storage silos for short periods of time must pay attention.
3.5. Scanning Electron Microscopy
Figure 6 and
Figure 7 show SEM images for the transversal and longitudinal sections (see
Figure 2) of the eucalyptus (a) and cork (b) wastes, respectively. The SEM images showed that the surface of the eucalyptus pellets was less rough than the surface of cork pellets. In
Figure 6, for the transversal section, clear differences can be observed. In the case of cork (
Figure 6b) some cell structures were perceived. This implies that the exposed surface was higher, promoting auto-oxidation.
In
Figure 7, the longitudinal section of pellets can be observed. The eucalyptus pellets (
Figure 7a,b) presented fewer and more superficial crevices compared to cork pellets (
Figure 7c,d). This fact could contribute to the higher surface area in contact with air for the cork pellets.
In such a way, and taking into account that auto-oxidation of fatty acids in wood depends on the availability of fatty and oxygen, the surface area could play an important role in CO production. The higher the surface area is, the more available the fatty acid is. Therefore, oxygen could interact more actively with wood and, subsequently, with its fatty acids. This could explain the higher CO emissions for the cork pellets in the same experimental conditions.
3.6. General Considerations
The surface could play an important role when it comes to CO generation as, according to the literature, CO levels in storage containers are strongly dependent on some physical properties such as bulk density and porosity [
36].
Finally, it was proved that continuous ventilation is highly advisable to avoid high CO concentrations. The CO emissions decreased once ventilation took place, allowing the completion of auto-oxidation. This way, the initial storage stages are especially dangerous, and extremely careful ventilation is needed. Indeed, pre-cooling after pellet manufacture is highly advisable, to avoid an increase in CO levels due to high temperatures during pellet storage. However, auto-oxidation is a long process and safety measurements, such as CO detectors and continuous air renewal, are recommended during the whole storage.
In any case, in order to guarantee the optimum pellet quality, several factors such as humidity or temperature should be taken into account, especially to avoid moisture accumulation, which could worsen the pellet quality.