Fritz Moedinger

Abstract The addition of biomass ashes to a brick feed has been investigated in a full-scale industrial production process over several days. The addition of biomass ashes to the brick feed is part of an ongoing research and development project targeted at substituting or combining quarried raw materials with suitable waste materials. In this chapter some of the early results of the first industrial trial runs with biomass ashes are presented. The main scope of this ongoing research and development project is reduction of production cost, generation of additional revenue from gate fees, and improvement of product characteristics.

9.1 Introduction

Besides other methods of recycling biomass ashes as described in other chapters in this book (see Chap. 1, Knapp and Insam 2011; Chap. 6, Omil et al. 2011; Chap. 11, Ribbing and Bjurstrom 2011), another option is to use the ashes for producing construction materials such as concrete (see Chap. 10, Berra et al. 2011) and brick.

For the production of ceramic bricks, the predominant raw material used is mineral clay. Any good brick clay should have low shrinkage and low swelling characteristics, consistent firing color, and a relatively low firing temperature, but at the same time produce an adequately dry and fire-strength brick. The guiding rule of choice on wastes and by-products must rest on their compatibility with the original (host) raw material being used, whereas they must not degrade the final product by focusing simply on making it a repository for wastes. Thus, it is necessary to

F. Moedinger

University of Staffordshire/Recuperi Industriali S. r.l. Fritz Moedinger, Via Don Bosco 10, 39042

Bressanone, Brixen, Italy

e-mail: fritz. moedinger@gmail. com

H. Insam and B. A. Knapp (eds.), Recycling of Biomass Ashes,

DOI 10.1007/978-3-642-19354-5_9, © Springer-Verlag Berlin Heidelberg 2011 establish a company-and-production — site-specific “tailor-made” quality product standard. In general, the firing of a mineral mass rich in aluminum silicate such as brick clay allows permanent stabilization of almost all heavy metals included except chromium (Anderson and Skerratt 2003).

2.3.1 Effect of Biomass Ashes on P Uptake and Shoot Biomass

In the field experiments positive results of ash supply were found in Rostock on the loamy sand (Table 2.6), but not in Trenthorst on the sandy loam (Table 2.7). In Rostock in 2007 higher barley yields (significant) and higher P uptakes (by trend) were found after SA and RMA application in comparison with the control. For maize in 2008 the best effects were found again after SA supply but also after CA supply (Table 2.6).

The missing effects in the Trenthorst experiment concerning yield and P uptake (Table 2.7) were most probably related to the soil conditions, mainly to the higher pH of this soil (Tables 2.2, 2.10, 2.11). Therefore, the liming effect of biomass ashes did not result in a further adv antage regarding the availability of nutrients, like we expected for sandy soils with lower pH. Furthermore, the soil P content in the Trenthorst soil was higher, which may have masked the P fertilizing effects of the ashes.

Owing to the lower soil volume, the fertilizing effects were higher in the pot experiments than in the field experiments. Significant effects were found for both soils in the 2007 and 2008 experiments.

The crop P uptake increased when P was supplied, independently of whether ash or TSP was added. In 2007, maize showed the highest P uptake of all main crops, with a mean value of 91.3 mg pot-1. In comparison with the control, the maize P uptake rose owing to P supply. The highest increasing rates were found for CA (about 34%) and TSP (about 44%) (Table 2.8). These results are in coherence with the biomass yield of maize (data not shown).

In 2008 on sandy loam, barley, which generated the highest biomass, also had the highest P uptake (especially after RMA supply: 94.6 mg pot-1) (Table 2.9).

Maize and lupin showed notable positive reactions on CA fertilization, with up to 74% more P uptake than in the control. For the catch crops, the highest P uptakes were found for phacelia and buckwheat in both pot experiments in 2007 and 2008.

Crop-specific P utilizations from the P sources were also relevant. In 2007, usually the highest effects were found for TSP and RMA, whereas the effects of RMA were a little smaller than those of TSP. The opposite was found for lupin and

phacelia, with slightly better results due to RMA. In consequence of the lower P concentration, SA application usually resulted in a lower crop P uptake than the other ashes. For oil radish after SA supply, even lower values were found than in the control without P. However, the P uptake of phacelia in the SA treatment was 36% higher than in the control. The SA effect on the P uptake of phacelia was even comparable to the RMA effect (Table 2.8). Differences in P uptake were also found after CA application, with high values for maize and rather low values for barley and lupin.

These effects underline the crop-specific mechanisms (see also Fig. 2.1) which should be considered when planning ash application within the crop rotation. Plant — specific adaptation mechanisms may warrant a sufficient P supply also under conditions of P deficiency in soil. For example, rape may excrete organic acids in the root zone, which is an effective strategy to increase P uptake mainly on soil with higher pH (Hoffland 1992). Buckwheat has been shown to have different P uptake efficiencies depending on soil conditions (Zhu et al. 2002). According to Van Ray and Van Diest (1979), different plant species differ in their behaviour with respect to uptake of cations. Excessive accumulation of cations within the plant can result in net excretion of H+, and in a lowering of pH values in soil, as was found in our experiments after cultivation of phacelia and buckwheat. Besides such kinds of chemical modifications in the rhizosphere, morphological root adaptations of plants may also help to supply plants with P (Fig. 2.1).

High P uptake of catch crops used as green manure provides a high potential for P supply of succeeding crops, when the decomposing catch crop releases P. According to our results, buckwheat and phacelia, which had high P uptake rates when fertilized with ashes, could be suitable in this sense. Furthermore, a combi­nation of green manure crops and ash application can also provide the soil with organic material, which ashes do not contain.

physiological enhanced chemical availability:

• changes in rhizosphere chemistry (pH; redox potential)

•

release of organic acids and enzymes

In most humid tropical and subtropical regions where acid soils prevail, warm and moist conditions result in weathered soil types. These tropical soils are depleted in available phosphorus, and usually the total phosphorous level is low, ranging from

0. 01 to 0.1% (Chen and Ma 2001). Much of the phosphorus is bound to aluminum and iron complexes during pedogenesis (Walker and Syers 1976). Through adsorp­tion and precipitation mechanisms aluminum forms insoluble and stable comple­xes with inorganic and organic phosphates, forming highly insoluble phosphorous compounds; therefore, their solubilization is a prerequisite for phosphorous uptake by plants. Soils suffering from aluminum toxicity are generally associated with phosphorous deficiency. The phosphorous-use efficiency in such soils is around 10-15% in the best situations (Verma et al. 2005). The low phosphorous status of these soils is of great concern because large amounts of phosphorous need to be applied to raise the concentrations of available soil phosphorous to an adequate level (Sanchez and Uehara 1980). Low phosphorous availability is considered to be one of the main limiting factors to plant growth in acid soils, in addition to human wealth in tropical areas (Barber 1995; Sanchez 2002). For temperate soils, ash amendments have been shown to alleviate phosphorous deficiencies (see Chap. 2, Schiemenz et al. 2011).

The fuels in Sweden are solid biofuels as well as different combustible wastes, including municipal solid wastes (MSW). The share of fossil fuels, coal or oil, is very small. The ash content in the fuels differs considerably. Clean heartwood as in wood pellets has an ash content of 0.2-0.5%. Wood chips obtained from forest residues have an ash content of 2-4% as the proportion of bark increases. Waste wood, such as construction and demolition debris, usually has higher ash contents. For peat, a figure of 5% is generally used, although the content can be much higher. Waste materials have comparatively high ash contents: 25% in MSW is quite common and various sludges in the pulp and paper industry have ash contents ranging between 10 and 50%.

The most dramatic effects of wood ash and lime on enchytraeids have been reported in laboratory microcosms containing only a small tree seedling as a primary producer (Liiri et al. 2002c, 2007) or no plants (Pokarzhevskii and Persson 1995). It has long been realized that ash effects on decomposers are stronger in laboratory experiments than in the field, but the reason for this has been unknown (Huhta et al. 1986). In a recent experiment by Nieminen (2008b) the negative effects of wood ash on enchytraeid size and abundance were offset by sucrose addition without any change in pH, supporting the hypothesis that wood ash effects can be alleviated by relaxing carbon limitation to microbes.

Huhta et al. (1983) used quite large pieces (40 cm x 60 cm) of Scots pine forest soil. Nematode populations started to decline in control microcosms 3 weeks after the start of the experiments, and later collembolan and enchytraeid populations declined as well. A mixture of birch ash and superphosphate reduced the populations of enchytraeids, mites and later also collembolans as much as lime. Nematode populations were initially stimulated by the treatments, but the effect turned nega­tive towards the end of the experiment. The experiment of Liiri et al. (2007) lasted for 1 year, and enchytraeids went close to extinction in untreated control pots as well as in ash-treated systems. In the experiment of Liiri et al. (2002c) enchytraeid biomass as well as the abundance of fungal-feeding nematodes declined even in unamended microcosms despite sufficient moisture, suggesting that the experimen­tal ecosystems were severely resource limited. In the same way, the enchytraeid population declined in control microcosms in the experiment of Pokarzhevskii and Persson (1995). Nieminen and Setala (2001) demonstrated carbon limitation in experimental microcosms: fungal-feeding nematodes propagated rapidly after cel­lulose addition to pine microcosms. Thus, carbon availability to microbes seems to be limiting decomposer activity in most laboratory microcosms, and consequently, resource limitation may have emphasized the effects of disturbances such as wood ash and lime observed in laboratory microcosms.

Carbon additions not only provide resources for microbes, they can also increase soil moisture (Szili-Kovacs et al. 2007). Using a simple carbohydrate such as sucrose as the only carbon source has the advantage that it is easier to distinguish nutritional effects from other effects. The heterotrophic microcosms reported in Nieminen (2008b) were maintained relatively moist and because no carbon effect on moisture was detected at the end of the experiment, the carbon effects were probably nutritional in that study. In contrast, the topsoil in the greenhouse experi­ment (Nieminen 2009) was much drier at the end of the experiment. Since quite a large amount of solid sucrose was used (Nieminen 2009), a mulching effect was initially possible. In addition, sucrose is hygroscopic, but other mechanisms are also possible. In any case, the increased moisture in the C1000 sucrose treatment was critical for the persistence of enchytraeids (Nieminen 2009). Since enchytraeids are sensitive to drought, and consequently likely to suffer locally from the increasing frequency of extreme conditions such as drought (Lindberg et al. 2002), it can be suggested that more attention should be paid to the interaction of soil carbon and moisture.

The hierarchical nature of the soil food web is also worth noting. Even though low molecular weight carbon compounds are taken up and utilized on timescales ranging from seconds (Hill et al. 2008) to days (van Hees et al. 2002), the effects of labile carbon additions at higher trophic positions were still evident after one growing season.

8.2.1 Methodology and Starting Points

The starting point is that the whole chain has to be analyzed to assess the environ­mental effects involving closure of the nutrient and mineral cycle. Interviews with representatives from the cacao industry and a literature study were used to define the parameters and to describe the scenarios. The most important part of this chain analysis is the definition of the parameters. The so-called base unit is the most important issue, as it expresses the product under study. The functional unit in this study is fertilization of 1 ha of land in Cote d’Ivoire, on which cacao trees are grown together with shade trees. Further, the effect categories have to be determined, which are the parameters of environmental impact, namely, the nutrient balance and the CO2, NOx, and SO2 emissions.

The starting points for this study are:

— The cacao plantation is situated in Cote d’Ivoire, and the beans are transported to the location of the cacao industry in the Netherlands.

— The cacao shells are a 100% natural residue of the beans and are thermally processed in a stand-alone power plant (so-called bioenergy plant) in the Netherlands.

— The efficiency of the nutrients is comparable to that of the commercial fertilizers which are used on the plantation. Also, the effect of trace elements and

contaminants which are present in the ashes is comparable to that of the commercial fertilizers.

The chemical composition of the ashes is predicted with a modified version of the KEMA Trace Model® (KEMA 2005). The model is an empirical and statistical computer model to predict emissions and ash composition of a dry pulverized coal — fired boiler including biomass co-combustion. A modified version of the model has been developed for fluidized bed boilers, taking into account the lower combustion temperature, the use of inert or reactive bed material, and the distribution of ash and bed material.

Ecotoxicity is a key property from a regulatory point of view, H14 in the European Waste Catalogue. Although the EU is expected to issue rules on criteria 2011, today there is no legally binding rule. There is no accepted test procedure for ecotoxicity yet, and an H14 value is computed using knowledge of the chemical composition of the waste and of the ecotoxicity of the pure chemical substances.

There is a difference of opinion in Sweden on the level at which a waste is to be considered ecotoxic. When a chemical product contains more than 2.5% of an ecotoxic substance, it is hazardous enough for organisms to warrant labelling as such. Analyses of ash usually yield the elemental composition, very seldom the concentration of chemical compounds. A study for the Ash Programme suggested a conservative selection of substances for the key trace elements and recommended that this value, 2.5%, be used to determine whether an ash is ecotoxic or not. The environmental authorities are of the opinion that the limit should be 0.25% because this is the limit for aquatic organisms.

The key element in this discussion is zinc, an essential nutrient. To be conserva­tive and not underestimate the ecotoxicity, the above-mentioned study assumed zinc was present as zinc oxide. At approximately the same time, zinc oxide was reclassified as ecotoxic, which has severe consequences for bioash. The zinc content in clean wood ash recommended by the Swedish Forest Agency is up to

7,0 mg/kg, i. e. 0.7%, which is larger than 0.25%. Concentrations of approxi­mately 0.4% are not uncommon for bark ash.

One way of tackling this difficulty is to examine whether zinc oxide is a reasonable choice as a reference substance after all, rather than just a conservative one. Three investigations should be mentioned here:

• Van der Sloot at ECN utilised the database LeachXS on leaching properties of MSWI residues and concluded that zinc is present mostly as willemite, a zinc silicate (H. van der Sloot, oral communication).

• Sjoblom (2007) reexamined the chemical thermodynamics of zinc in combus­tion residues, in particular the solubility equilibria, and concluded that zinc is present probably as franklinite, a mixed iron and zinc oxide, rather than willemite.

• Steenari and Noren (2008) studied the binding distances for zinc in different combustion residues using extended X-ray absorption fine structure spectros­copy, and found that the distances agree with zinc being bound in silicates or aluminates. They also found that in fresh ash, some of the more soluble zinc compounds are also present in minor quantities. When ash is wetted, zinc is rapidly redistributed to insoluble compounds.

Thus, zinc does not occur as oxide in combustion residues, but rather in much more stable compounds. Choosing oxide in the ecotoxicity assessment was overly conservative and significantly overestimates the environmental impact of zinc in ash. Settling for a realistic model compound is not straightforward: it seems more

probable that zinc is found as willemite or a silicate, but there is no official assessment of these compounds. On the other hand, less probable franklinite but with similar properties has been assessed as non-ecotoxic.

At the end of the day, when there are doubts, an ecotoxicity test will clinch the discussion. In a study co-financed by the Ash Programme, the Swedish Waste Management Association and the Swedish Environmental Protection Agency, a battery of tests were performed on different combustion residues (Stiernstrom et al. 2009). Care was taken to choose organisms that naturally occur in brackish waters rather than the freshwater organisms that are sensitive to the salt concentration.

At a liquid-to-solid leaching ratio of 10 (waste test condition to be on the conservative side, but for chemical products the liquid-to-solid leaching ratio should be 10,000 according to OECD good practice) almost all residues were ecotoxic. A more detailed examination revealed that the ecotoxicity is due to the high concentrations of potassium and calcium (nutrients) and of aluminium. This is scarcely a reason to consider the residues as ecotoxic.

Results from tests of bottom ash from MSWI are illustrative. Ash which had been stored and allowed to mature for 3 months was somewhat ecotoxic. Ash which had been aged and exposed to the weather (actual liquid-to-solid leaching ratio of 3.4, rainwater as leachant) for 12 years was not ecotoxic at all to the three organisms tested. And yet, its composition had not changed, except for a lower content of chlorides and probably of sulphates, too.

11.3 Conclusion

Projects within the Ash Programme have contributed increased knowledge of the properties of combustion residues as well as demonstrated large-scale uses for residues. The Ash Programme has been active since 2002, and it is quite probable that it will be extended for another 3 years at the end of 2011.

With ash from clean solid biofuels, priority should be given to recycling the ash to forest soils as a compensation for the extraction of mineral nutrients in an intensive harvesting, i. e. whole-tree harvesting. If that is not possible, the residues should be used in civil works, e. g. for building roads or hard surfaces. One should put all such ash to use, instead of rejecting particular streams.

The Ash Programme has striven for a common understanding of the level of risk for human health and the environment when using ash in civil works. Boundaries between low risk and not low risk have been suggested. The Swedish Environmen­tal Protection Agency has developed another boundary, below which one does not even need to pay notice to the environmental authorities.

All reports are public and are available from Varmeforsk’s Web site (http:// www. varmeforsk. se). The language is Swedish, but there is always a summary in English.

When looking at the effect of wood ash amendments on tree growth in Nordic countries, Augusto et al. (2008) revealed a considerable site dependency using a meta-analysis approach. Whereas wood ash was not able to improve tree growth on mineral soils, it had a significant effect on trees planted on organic soils. Reviewing different studies from Finland, Sweden and Switzerland with regard to the impact of wood ash applications on tree growth and vitality, Lundstrom et al. (2003) reported neutral or even negative effects of ash fertilisation. Investigating the effect of hardened wood ash application (up to 3 Mg ha-1) on ground vegetation in young Norway spruce stands on mineral soils, Arvidsson et al. (2002) found that biodiver­sity and plant biomass were not affected. In a Swiss forest, fine roots of spruce were influenced by ash application (4 t ha-1) on mineral soil, whereby ash fertilisation enhanced the number of root tips, forks and the root length, but resulted in decreased root diameters (Genenger et al. 2003). In a set of field experiments applying wood ash (1-9 Mg ha-1) on 30-60-year-old Scots pine (Pinus sylvestris L.) and Norway spruce (Picea abies (L.) Karst.) stands on mineral soil in Sweden, stem growth was only promoted when N (150 kg ha-1) was added, whereas wood ash amendments without N did not lead to significant responses (Jacobson 2003). The same was true for combined wood ash and N applications on a Scots pine (Pinus sylvestris) stand on a low-productivity mineral soil, where wood ash plus N positively influenced tree growth even 20 years after application. Nitrogen fertilisation alone only led to a short-term effect (Saarsalmi et al. 2006). Whereas wood ash or sludge application alone did not have any influence on the structure of a commercial willow plantation in central Sweden in a 3-year experiment, harvestable shoot biomass was increased by a combined sludge (2.6 t ha-1) and ash (5.5 t ha-1) treatment and thus gave results comparable to fertilisation with mineral fertiliser corresponding to 14.5 kg P ha-1 year-1, 48 kg K ha-1 year-1 and 100 kg N ha-1 year-1 (Adler et al. 2008). However, this treatment showed negative effects on wood fuel quality concerning P, K and heavy metal concentrations in the bark and wood. Plant growth or biomass production was also not influenced by wood ash fertilisation (10 and 20 Mg ha-1 for 3 years) in a willow plantation on a silt loam soil in the state of New York (Park et al. 2005). In this experiment, wood ash did not have an impact on nutrient concentrations of foliar, litter and stem tissue, whereas the concentrations of soil-extractable P, K, Ca and Mg were significantly higher than in control plots. In contrast, Moilanen et al. (2002) observed a positive effect of ash fertilisation (8 and 16 t ha-1) on tree volume growth even 50 years after amendment on a drained peat mire, being accompanied by elevated nutrient concentrations in the peat.

The results presented are attributed to the fact that ash is low in N, which is the main limiting element for plant growth on mineral soils in boreal forests. In contrast, ash fertilisation was considered to be more suitable for peatland forests displaying higher N contents (Hanell and Magnusson 2005) and was found to promote tree growth (height, diameter and biomass) of a young Pseudotsuga menziesii plantation and a Pinus radiata plantation on N-rich mineral soil in

Spain (Solla-GullcSn et al. 2006, 2008). Besides enhanced stem-wood growth of Norway spruce, Rosenberg et al. (2010) also observed increased CO2 evolution rates even 12 years after wood ash applications (6 Mg ha-1) on a N-rich soil, indicating that ash amendments of N-rich sites have to be evaluated carefully regarding their effect on C and N cycling.

Plots on lutites and migmatites were selected for this study in order to compare two plantations of the same age, but located over two different geological materials.

The following figures include the estimated marginal means of the dasymetric variables: total height and normal diameter (corrected in the covariance function, data from the plots before the treatment in 2003 are used as a covariate). A statistical analysis comparing between levels of the factor “treatment” corresponding to each level of the factor “time” and the levels of the repeated means is also included.

Different effects were observed in both types of plots. In the first plot (Fig. 6.4), the statistical analysis revealed significant increases in height and diameter growth as a result of the application of ash and phosphorus, mainly from 2005 onwards. The time-treatment interaction revealed some differences between the treatments, with benefits to both plots. In the second plot (Fig. 6.5) increases in the normal diameter for WA and WAP treatments were significant. The different response of the plantations may be due to the differences in the nutritional state, as indicated by the different site index values (18.7 and 23.1 m in the lutites and migmatites, respectively).

Fig. 6.3 Foliar macronutrients in soils over lutites and migmatites. Control untreated, WA applications of 4.5 Mg wood ash ha-1 for three consecutive years (2003, 2004, 2005), WAP single applications of 4.5 Mg wood ash ha-1 and 0.1 Mg P2O5 ha-1

-j

‘O

Fig. 6.4 Estimated marginal means of Pinus radiata D. Don in plots over lutites. Control untreated, WA application of 4.5 Mg wood ash ha-1 for three consecutive years (2003, 2004, 2005), WAP single applications of 4.5 Mg wood ash ha-1 and 0.1 Mg P2O5 ha-1. In the repeated measures analysis, the sphericity assumption was not valid and therefore values of P-Huynh Feldt were used in univariate tests of their interactions. Ht total height, Dn normal diameter

In compositional terms, good brick clay should feature:

• Grain size distribution (sieve line): A high percentage of larger particles requires higher sintering temperatures, resulting in a greater energy requirement and longer firing times.

• Accessory minerals: Quartz, feldspar, and amphibole have an effect on the sintering behavior and might result in undesired colors.

• Organics: Create voids in the finished product. If the organics have a high sulfur content, this might have effects on color, the fumes, and the kiln atmosphere.

• Sulfur: Pyrite and marcasite release SOx on firing, creating large-diameter pores in the brick that might reduce compressive and flexural strength.

• Carbonate minerals: Calcite, dolomite, and other carbonate minerals, such as ankerite and siderite, do have, if finely dispersed, an effect on the release of low — temperature-carbonization gases owing to the formation of channels and funnels but in larger quantities will reduce the compressive and flexural strength of the final product and might lead to chipping on the surface.

• Alkalis: A low alkaline earth content, magnesium and calcium being the most common, is desirable to avoid firing interactions which could promote discolor­ation of the final product.

• Metal oxides and hydroxides: Goethite and hematite, for example, both contain­ing iron oxide (the chief colorant responsible), which ideally should be in the range 5-12% (for good strong color), are the main origin of the red brick color.

• Natural radioactivity of the raw materials (radon).

Adding foreign substances to influence the physical properties of the finished product or the workability is nothing new. It is important to distinguish between “additions” to the clay body (such as saw dust or paper sludge), and “substitutions” (such as sewage sludge, ash, or treated aluminum salt slag) that replace a part of the original clay body. The distinction between “addition” and “substitution” for extraneous materials is not always straightforward or easy: Any substance that is added to the original clay body without substitution of clay modifying its inherent characteristics might be considered an addition. A substitution, on the other hand, may be viewed as any material that for a required volume of brick reduces the quantity of clay needed to achieve that specific volume target. A substitution can also modify the clay body.