Minerals in pyrolysis

From PyroWiki
Jump to: navigation, search

In pyrolysis, the presence of mineral matter has a significant influence on the product distribution. Even though the mineral (ash) content of most biomass streams is limited to just a few weight percent, pyrolysis oil quality and yield can decrease rapidly when the concentration of (certain) minerals increases. Apart from the bio-oil quality and yield issue, it is also important to avoid that minerals end up in the bio-oil; they should preferably all remain in the char by-product.

Minerals in biomass

Elements and classification

Biomass is a general term for a wide variety of materials. In attempt to gain insight in the chemical composition of biomass, Vassilev et al[1] compiled peer-referenced data of 86 biomass species. The quantity and composition of mineral matter in the biomass depend on its genetic variety. In addition, mineral matter can also be introduced during the processing of biomass. The most abundant organic and inorganic elements in biomass are, in decreasing order: C, O, H, N, Ca, K, Si, Mg, Al, S, Fe, P, Cl, Na, Mn and Ti. The interactions between elements my differ for each type of biomass. In Table 1 a general classification of biomass varieties is presented

Table 1: General classification of biomass varieties, adapted from Vassilev et al.[1]
Biomass group Biomass sub-group, varieties and species
1. Wood and woody biomass Coniferous or deciduous; soft or hard wood; stems, branches, foliage, bark, chips, pellets, briquettes, sawdust, and others.
2. Herbaceous and agricultural biomass Annual or perennial biomass, such as grasses, flowers, straws (barley, corn, rice, wheat, ...) and other residues (fruits, shells, husks, pits, grains, kernels, cobs, bagasse, etc.)
3. Aquatic biomass Macroalgae, microalgae, marine or freshwater, seaweed, etc.
4. Animal and human biomass wastes Bones, meat-bone meal, chicken litter, various manures, etc.
5. Contaminated biomass and industrial biomass wastes Municipal solid waste, demolition wood, refuse-derived fuel, sewage sludge, paper-pulp sludge, plywood, etc.
6. Biomass mixtures Blends from the above varieties

For pyrolysis oil production aiming at energy applications especially the wood and woody biomass and the herbaceous and agricultural biomass (as well as mixtures from these) are of interest. If dried sufficiently, all groups of biomass can be converted by pyrolysis. Obviously, drying of aquatic biomass is hardly feasible. With respect to the animal, human and industrial waste streams, high levels of contaminants may occur and cause problems in pyrolysis. For the more relevant biomass streams the total ash content, as well as the most abundant elements in the ash, are listed for each group in Table 2. The group of herbaceous and agricultural biomass is further divided in grasses, straws and other residues because of the specific relevance to fast pyrolysis. Noteworthy is the wide variation in numbers reported for each individual group.

Table 2: Total ash and ash composition for relevant biomass groups[1]
Biomass group mean ash content Most abundant elements in ash (oxygenized, after combustion)
1. Wood and woody biomass 2.7 (0.1-8.4) CaO, SiO2, K2O, MgO, Al2O3.
2. Herbaceous and agricultural biomass
2.1 Grasses 4.3 (0.8-9.4) SiO2, K2O, CaO, P2O5
2.2 Straws 7.8 (4.3-18.6) SiO2, K2O, CaO, MgO, P2O5.
2.3 Other residues 4.4 (0.9-16.1) K2O, SiO2, CaO, P2O5, MgO.

Mineral form

In addition to the total content of certain minerals, the form in which the minerals appear in the biomass is important. Generally minerals are present in biomass as inorganic matter, bound to organic matter or as fluid matter. In Table 3, the phase composition of biomass is presented along with some typical components.

Phase composition of biomass, adapted from Vassilev et al.[1]
Matter State and type Phases and components
Organic matter Solid, non-crystalline Structural ingredients, cellulose, hemicellulose, lignin, extractives, others.
Solid, crystalline Organic minerals such as Ca-Mg-K-Na oxalates, etc.
Inorganic matter Solid, crystalline Mineral species from phosphates, carbonates, silicates, chlorides, sulphates, etc.
Solid, semi-crystalline Poorly crystalized mineraloids of silicates, phosphates, etc.
Solid, amorphous Amorphous phase such as glasses, silicates, etc.
Fluid matter Fluid, liquid, gas Moisture and gas and gas-liquid inclusions associated with both inorganic and organic matter.

Most of the elements can be present in each of the groups from Table 3, however some major associations are reported by Vassilev et al[2]. • Si-Al-Fe-Na-Ti, typically glass, silicates and oxyhydroxides form; • Ca-Mg-Mn, typically carbonate, oxyhydroxide, glass, silicate, phosphate and sulphate form; • K-P-S-Cl, typically in phosphate, sulphate, chloride, glass form, with some silicate and carbonate forms present.

Influence of minerals on the fast pyrolysis process

Effect on product distribution

Even though it is generally accepted that higher ash content typically decreases the organic liquid yield, hard data is scarce. While extremely relevant for both the oil yield and quality, limited research has been carried out to understand the effect of ash on the pyrolysis reactions[3]. One of the difficulties in performing and interpreting these results lies in the fact different pyrolysis systems are used. Often studies are performed in which biomass is either enriched with certain minerals, or minerals are specifically removed before pyrolysis, to avoid these limitations.

The simples way to investigate pyrolysis behaviour is by thermographic analyzer (TGA)[1]. In this system, biomass is heated, usually not very fast (1-100°C/min), to the desired temperature and the weight loss of the original biomass sample is continuously monitored. This technique supplies information about the decomposition behaviour as function of temperature. Numerous studies are available that show addition of certain minerals decreases the temperature at which the decomposition rate is highest, see for example Hwang et al[4] who impregnated poplar wood with potassium, and Yang et al.[5] who mixed palm oil wastes with K, Cl, Na, Ca, Mg, Fe and Al.

The product distribution as function of total ash content resulting from the pyrolysis of various biomass streams in a twin-screw mixing reactor at a capacity of 10 kg/h has been reported by Tröger et al.[6]. Here, a clear trend of decreasing organic liquid production for increasing ash content is found, see Fig. 1.


Effect on product quality

Due to decreased organic liquid production and increased water production, the moisture content of the pyrolysis oil increases significantly for higher as contents. This increased moisture content results in phase-separation of the pyrolysis oil into an organic phase with usually higher viscosity, and a waterphase with high water content and low viscosity. Phase separation is undesirable for most applications, and should be avoided. In Fig. 2 data from Tröger et al.[7] on the water content in the pyrolysis liquid as function of the biomass ash content is presented. Clearly for a few wt.% ash in the original biomass, the risk of phase separation is apparent.



Information on the exact mechanisms explaining the chemistry behind the influence of minerals on the pyrolysis behaviour are scarce. It is generally accepted the minerals show catalytic activity in the pyrolysis process. Agblevor and Besler[8] state especially potassium and calcium catalyze biomass decomposition and char formation, although no proof is provided. Important in unraveling the mechanism is also the location of the mineral in the process. During pyrolysis the biomass components are vaporized, after which the vapour is rapidly quenched to maximize the pyrolysis oil yield. At the University of Twente[2], detailed investigations were performed to determine the mechanism behind the pyrolysis oil production loss. Initial work[9] showed both homogeneous reactions within the pyrolysis vapour (polymerization caused by high temperature) and heterogeneous reactions between the minerals (Na+K) and the pyrolysis vapour to affect the oil production. Later a wire mesh reactor was designed and constructed with specific aim to investigate the initial pyrolysis reactions[10]. Using the wire mesh reactor, it was shown the char production doubled using KCl impregnated wood compared to the untreated wood.


The presence of minerals in the biomass clearly affects the process already from the initial devolitalization. To improve the pyrolysis oil yield from biomass streams with increased ash content, the most obvious approach is to de-ash the biomass. Biomass pretreatment before pyrolysis has been researched quite extensively, however mainly (only) on relatively small scale. One of the first studies on the influence of mineral matter on the fast pyrolysis of biomass was performed by Raveendran et al.[11] back in 1995. In a packed-bed pyrolyzer, thirteen biomass species with ash content ranging from 0.7 to 23.5 wt.% were pyrolyzed. In addition, tests were performed in which biomass was de-ashed or impregnated with salts. Reveendran showed the total liquid yield increased in all cases when biomass was de-ashed prior to pyrolysis. Unfortunately no distinction was made in the organic and water part of the total liquid. In addition, the de-ashing process was quite extreme comprising of a first wash using 10% HCl at 60°C for 48 hours under constant stirring, followed by a second wash using 5% NaOH at 90°C for 1 hour. Afterwards, the biomass was washed with demineralized water, filtered and dried. This approach improved the performance, but commercial implementation seems unrealistic. Furthermore, the extreme pretreatment conditions might have influenced the organic part of the biomass as well, making it unsure if the yield improvement can be ascribed (fully) to the removal of minerals. In general more extreme pretreatement conditions (in terms of chemicals, temperature and time) result in more effective removal of the mineral matter. There are however also draw-backs from the pretreatment process. Since the pyrolysis process requires a dry feedstock, the washed biomass needs to be dried to moisture content typically below 10%. In addition, when mineral acids or alkaline compounds are used, they need to be thoroughly rinsed from the biomass to avoid increasing the mineral concentration. Use of organic acids can be an interesting alternative. One approach to use organic acids as well as to optimize the overall value chain is the process scheme proposed by Oudenhoven et al.[12]. They utilize part of the organic acids produced in the pyrolysis process in the pretreatment to demineralize wood prior to pyrolysis and subsequently improve the yield of organics, and especially levoglucosan. The vapours from the fluidized bed reactor were quenched in two consecutive condensors, the first one operating at 80°C to recover most of the heavy components, the second one operated at -5°C and recovered most of the light organic and water. The liquid from the second condensor was used to demineralize the wood. This approach increased the organic liquid yield from 48% to 56%, showing a clear beneficial effect, while avoiding the use of external solvents.

Banks et al[13] investigated the pyrolysis of miscanthus pretreated using deionized water, HCl and a surfactant called Triton X-100. The yield of organic liquid was 54 wt.% for the unwashed miscanthus, and 56 wt.%, 49 wt.% and 69 wt.% for the deionized water, HCl and Triton X-100 respectively, confirming washing with adequate mineral solvent significantly enhances the pyrolysis oil yield.

Minerals in pyrolysis oil

Most of the mineral matter from the biomass (>95%) is retained in the char by-product from the fast pyrolysis process [14] [15]. A small part is however found in the pyrolysis oil, most of these inorganic elements transfer from the reactor to the condensor by entrainment of small solid char particles. The effect of minerals is not limited to the production process only. The presence of trace amounts of minerals are believed to reduce pyrolysis oil stability by enhancing polymerization reactions. It is however not clear if the mineral part of the char, or the char itself decreases stability. In an attempt to reduce char and mineral matter in the pyrolysis oil, various researchers used hot gas/vapour filtration between pyrolysis reactor and condensors [16] [17]. In addition, the presence of minerals in the pyrolysis oil is undesirable for most applications. Where in combustion processes small amounts might be tolerable, catalytic processing (e.g. Upgrading_processes or pyrolysis oil gasification) might even be limited by trace amounts of minerals.

Recovery & re-use of minerals

The form in which ashes are recovered from the pyrolysis process depends on the process execution. Fluidized bed systems that use the pyrolysis gas for internal heat generation recover the char by-product from the process in the un-combusted form. For these systems, the ash contains mainly carbon, and might be suitable for energy generation. In case the char by-product is combusted within the pyrolysis system, the ashes have a very low carbon content, and are similar to ash removed from direct biomass combustion (provided the combustion temperature is lower than the ash melt temperature). In case the ash melt temperature is surpassed, the ash will start to melt and glasify. Re-use of biomass ash depends on the composition, which in turn depends on the original material. In the overview of Vassilev et al[18]. the main potential for biomass ash use seems to be as soil amendment or fertilizer, followed by the use in construction materials and sorbents. Occasionally the production of minerals, ceramics and other materials might be feasible.

Concluding remarks

The role of minerals in the fast pyrolysis process, as well as the effect of minerals in the pyrolysis oil product and the recovery of minerals from the ash remain an important research subject for the coming years. Even though it is clear lower mineral concentrations increase the pyrolysis liquid yield, more information on the chemistry behind the process is required. With the implementation of commercial scale pyrolysis of clean wood, the next step will be shifting towards lower value feedstocks, requiring the development of suitable techniques/processes to deal with the higher mineral concentrations.


  1. Vassilev SV, Baxter D, Andersen LK, Vassileva CG. An overview of the chemical composition of biomass. Fuel. 2010 May;89(5):913-33.
  2. Vassilev SV, Baxter D, Andersen LK, Vassileva CG. An overview of the composition and application of biomass ash. Part 1. Phase–mineral and chemical composition and classification. Fuel. 2013;105:40-76.
  3. Venderbosch RH, Prins W. Fast pyrolysis technology development. Biofuels, Bioproducts and Biorefining. 2010;4(2):178-208.
  4. Hwang H, Oh S, Cho TS, Choi IG, Choi JW. Fast pyrolysis of potassium impregnated poplar wood and characterization of its influence on the formation as well as properties of pyrolytic products. Bioresour Technol. 2013 Dec;150:359-66.
  5. Yang HP, Yan R, Chen HP, Zheng CG, Lee DH, Liang DT. Influence of mineral matter on pyrolysis of palm oil wastes. Combust Flame. 2006 Sep;146(4):605-11.
  6. Tröger N, Richter D, Stahl R. Effect of feedstock composition on product yields and energy recovery rates of fast pyrolysis products from different straw types. J Anal Appl Pyrol. 2013;100:158-65.
  7. Tröger N, Richter D, Stahl R. Effect of feedstock composition on product yields and energy recovery rates of fast pyrolysis products from different straw types. J Anal Appl Pyrol. 2013;100:158-65.
  8. Agblevor FA, Besler S. Inorganic compounds in biomass feedstocks .1. Effect on the quality of fast pyrolysis oils. Energ Fuel. 1996 Mar-Apr;10(2):293-8.
  9. Hoekstra E, Westerhof RJM, Brilman W, Van Swaaij WPM, Kersten SRA, Hogendoorn KJA, et al. Heterogeneous and homogeneous reactions of pyrolysis vapors from pine wood. Aiche J. 2012;58(9):2830-42.
  10. Hoekstra E, Van Swaaij WPM, Kersten SRA, Hogendoorn KJA. Fast pyrolysis in a novel wire-mesh reactor: Decomposition of pine wood and model compounds. Chem Eng J. 2012;187:172-84.
  11. Raveendran K, Ganesh A, Khilar KC. Influence of Mineral Matter on Biomass Pyrolysis Characteristics. Fuel. 1995 Dec;74(12):1812-22.
  12. Oudenhoven SRG, Westerhof RJM, Aldenkamp N, Brilman DWF, Kersten SRA. Demineralization of wood using wood-derived acid: Towards a selective pyrolysis process for fuel and chemicals production. J Anal Appl Pyrol. 2013;103:112-8.
  13. Banks SW, Nowakowski DJ, Bridgwater AV. Fast pyrolysis processing of surfactant washed Miscanthus. Fuel Process Technol. 2014;128:94-103.
  14. Leijenhorst EJ, Wolters W, van de Beld L, Prins W. Inorganic element transfer from biomass to fast pyrolysis oil: Review and experiments. Fuel Processing Technology (2016) Vol. 149. p.96-111.
  15. Leijenhorst EJ, Van De Beld L, Meers E, Prins W, editors. Fate of Minerals in the Fast Pyrolysis Process. 22nd European Biomass Conference and Exhibition; 2014; Hamburg.
  16. Baldwin RM, Feik CJ. Bio-oil Stabilization and Upgrading by Hot Gas Filtration. Energ Fuel. 2013;27(6):3224-38.
  17. Hoekstra E, Hogendoorn KJA, Wang X, Westerhof RJM, Kersten SRA, van Swaaij WPM, et al. Fast Pyrolysis of Biomass in a Fluidized Bed Reactor: In Situ Filtering of the Vapors. Ind Eng Chem Res. 2009;48(10):4744-56.
  18. Vassilev SV, Baxter D, Andersen LK, Vassileva CG. An overview of the composition and application of biomass ash. Part 1. Phase–mineral and chemical composition and classification. Fuel. 2013;105:40-76.