**5. Discussion**

Fuel from woody biomass is a common commercial fuel but it has a higher amount of ash, e.g., woody pellets have only 0.15 wt.% of ash but coniferous cuttings could be contaminated by bark or pieces of soil.

From Table 5, it clearly follows that the dominant oxide of the inorganic fraction of SP are aluminum, calcium, and silicon oxide. The second most abundant component is potassium oxide, followed by manganese oxide, silica, ferric oxide, and sodium oxide. This table also clearly shows that phosphates and sulphates form an inconsiderable part of ash matter from biomass trapped in the filter. Chlorides represent only 0.1% by wt. in the monitored sample. A surprising result was the determined concentrations of heavy metals in the analyzed matter, particularly in terms of the toxicity of fly ash and its disposal and storage related to transport etc. This involves mainly copper, chrome, zinc, and manganese oxides.

The determined distribution of particle sizes of solid pollutants trapped by the sleeve filter from the raw generator gas produced by gasifying wood chips by air, demonstrated that their absolute size ranges from approximately 0.4 to 185 μm, whereas 99% by wt. of monitored particles were smaller than 100 μm. The proportion with a diameter up to 1 μm was approximately 1%, and up to 10 μm was approximately 22%.

The distribution of the individual mass fractions from 10% to 80% did not demonstrate any significant differences. The difference in particle size between 99 and 100% cannot be considered decisive due to the genesis of the sample.

Comparison of TG and GC-MS data shows that the results of thermogravimetric analysis do not correspond fully with the results of the performed identification. The sum of the weight of the individual identified tar components was substantially lower than the weight of substances released from the heated filtration cake in a nitrogen atmosphere. This could be caused by chemical sorption of organic substances on the SP, which could not be removed by the solvent.

The performed analyses of the incrust show that this is made up of mainly inorganic components of ash from the raw material gasified in the generator [23]. Its formation is defined significantly by potassium, calcium, and silicon oxides; and also phosphorus, iron, and magnesium oxides in the amorphous part [24].

Many of the identified oxides are typical components of ash matter forming low melting ash during combustion, or wood gasification. The relatively high silica content shows that this component was dosed into the generator with fuel during its preparation. One of its undesirable effects in the generator´s oxidation zone, where relatively high temperatures are present, is the formation of crystalline substances creating incrusts. An example of a K2O-CaO-SiO2 phase diagram is shown in Figure 6, where even the following eutectics can be found [25]:


To prevent similar undesirable conditions during the operation of cogeneration units, it is necessary to substantially limit its presence in the woody matter dosed into the generator. This applies to silica itself contaminating the tree bark or silica present in soil. The thermal conditions of decomposition of clay minerals must be considered: Kaolinite 450–700 ◦C, montmorillonite 600–700 ◦C, 800–900 ◦C, and illite 450–700 ◦C, 850–950 ◦C.

**Figure 6.** Detail of the K2O-CaO-SiO2 ternary system [25].

**Author Contributions:** Conceptualization, J.N. and P.B.; Data curation, J.N.; Funding acquisition, J.F.; Investigation, P.B. and S.S.; Methodology, S.S.; Supervision, J.N. and J.F.; Validation, J.K.; Visualization, V.P.; Writing—original draft, V.P.

**Funding:** This research was funded by Ministry of Education, Youth and Sports of Czech republic, gran<sup>t</sup> number CZ.1.05/2.1.00/19.0389: Research Infrastructure Development of the Centre ENET.

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.
