- Email: [email protected]
Grass pellet Quality Index: A tool to evaluate suitability of grass pellets for small scale combustion systems
Applied Energy 103 (2013) 679–684
Contents lists available at SciVerse ScienceDirect
Applied Energy journal homepage: www.elsevier.com/locate/apenergy
Grass pellet Quality Index: A tool to evaluate suitability of grass pellets for small scale combustion systems Jerome H. Cherney a,⇑, Vijay Kumar Verma b a b
Department of Crop & Soil Sciences, 503 Bradfield Hall, Cornell University, Ithaca, NY 14853, United States Department of Mechanical Engineering, Vrije Universiteit Brussel, Pleinlaan 2, 1050 Brussel, Belgium
h i g h l i g h t s " A Quality Index ranking grass pellets for combustion potential does not exist. " Grass pellet biomass is extremely variable and therefore quality control is essential. " Proposed system sums qualitatively different parameters into one index value. " Model structure allows for effective evaluation and ranking.
a r t i c l e
i n f o
a b s t r a c t
Article history: Received 16 May 2012 Received in revised form 14 October 2012 Accepted 16 October 2012 Available online 17 November 2012
US renewable fuels policy strongly encourages biomass crop production, which should lead to expansion of biomass heating scenarios. Chemical composition of grass biomass can be extremely variable, depending on species, soil fertility, and harvest management. Biomass quality concerns have hindered the development of grass biomass for residential combustion, with no comprehensive evaluation system for grass pellet quality. Quality labeling will strengthen the fledgling grass biomass heating market, gain consumer confidence, and help to control combustion related emissions. The proposed system sums qualitatively different parameters into one Quality Index for relative evaluation and ranking of grass pellets for residential combustion potential. Parameters were selected and weighted for their relative importance based on available literature. Weighting was accomplished by adjusting the compositional working range for each parameter. A limit also was established for each parameter, beyond which the pellet lot was considered as unacceptable for residential combustion, regardless of the total Quality Index score. The model structure allows for effective evaluation and ranking of grass pellet lots regardless of the specific values ultimately chosen for acceptable limits and working ranges by the industry. Applying the Quality Index to a range of grass pellet types resulted in a reasonable ranking of pellets based on physical characteristics and composition. Ó 2012 Elsevier Ltd. All rights reserved.
Keywords: Grass composition Grass pellets Quality Index Residential combustion Solid biofuel
1. Introduction Rural North America has a tremendous capacity for energy production [1–3]. While the ongoing desire for energy security is the primary driving force for alternative energy development in the USA, environmental issues will sooner or later overshadow energy supply issues. An ideal alternative solid biomass feedstock should be nearly carbon neutral, without significant net increase in atmospheric carbon dioxide. It also needs to be seen as a valuable crop from the farmer’s perspective [4,5].
⇑ Corresponding author. Tel.: +1 607 255 0945; fax: +1 607 255 2644. E-mail addresses: (V.K. Verma).
[email protected]
(J.H.
Cherney),
[email protected]
0306-2619/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.apenergy.2012.10.050
The Northeastern USA has significant space heating demand, and most states in the region have state regulations controlling combustion emissions [6]. New York and the New England states represent approximately 80% of the total heating oil demand in the USA. The Northeast USA has millions of acres of abandoned and underutilized land suitable for grass biomass, without interfering with traditional agricultural crops and without the requirement for establishment [7]. Existing mixed grass stands are appropriate for grass biomass production if the energy conversion process is combustion. The use of such lands for this purpose makes this biomass option relatively immune from the indirect land use change debate [8]. Energy conversion efficiency can be very high with grass combustion [9]. Grass energy farming is a small-scale, low-technology, closedloop energy system that will result in rural jobs and economic
680
J.H. Cherney, V.K. Verma / Applied Energy 103 (2013) 679–684
diversification, absorbing excess production capacity. In general, the desired feedstock for biomass combustion is opposite of that required for ruminant animal forage, with a range in composition among herbaceous plant species [10,11]. While grass breeding for biomass will eventually lead to compositional changes in the feedstock [12–14], major compositional changes can be achieved in the near-term through agronomic and harvest management [15]. 1.1. Biomass composition and combustion Soil type and inherent soil fertility can strongly impact mineral uptake [16]. Plant uptake of potassium (K) and chloride (Cl) are generally well correlated, due to luxury uptake of both elements in excess of plant requirements, and the common practice of fertilizing with KCl [17]. Potassium concentration has been reported as low as 0.6 g kg 1 to as high as 70 g kg 1 in cool-season grasses on a dry matter basis [18]. Concentration of most elements in grass decreases with plant age, making mature plants more desirable for combustion from a compositional standpoint. Harvest management can have a major impact on grass composition, particularly for water soluble components such as K and Cl [19,20]. Delayed baling following mowing, as well as overwintering grass in the field, also will reduce insoluble nitrogen (N) and silica (Si) in grass feedstock due to the preferential loss of inflorescence and leaf blade which are higher in N and Si content than other plant parts [21,22]. Dry matter losses in storage also are possible [23]. Biomass quality issues have hindered commercialization of grass pellets in residential combustion systems [24]. The most serious quality issues with grass feedstock are generally considered to be the alkali and chloride content [17,25–27]. Boiler corrosion and fouling are directly related to alkali and chloride content. Particulate emissions primarily consist of aerosol-forming elements like potassium and chloride, as well as sulfur (S). Chloride also acts as a catalyst, facilitating the movement of iron away from metal surfaces and the deposition of inorganic compounds [28,29]. Potassium is the primary alkali element present in grasses, with typically very low concentration of sodium [20]. Release of K can be minimized by controlling combustion temperature, but this does not prevent the release of Cl [25,26]. Nitrogen content of grass has little impact on combustion efficiency but is undesirable from an environmental standpoint. Nitrogen oxides are the second most important contributor to global acidification from human activities [25]. There are concerns that N release to the atmosphere from biofuel production negates any green house gas reduction benefits [30]. Nitrogen emissions are positively correlated with feedstock N content. Sulfur and Si, in combination with alkali, lead to reactions associated with fouling and slagging in boilers [28,31]. In general, approximately one half of the total ash content of grass is silica. Silica, in combination with K and other elements, affect the ash melting behavior in grasses [32]. The total amount of ash impacts the design of ash handling and storage systems for a given appliance, but ash content per se is not a major drawback to grass combustion within certain limits. Some residential heating appliances currently available are able to handle ash content up to 10%. Of course, total ash content is likely to be positively correlated with concentrations of undesirable elements, particularly Si and K. High total ash and Si content result in lower energy content. Other ash forming elements such as P, Ca, Mg, Al, Fe, and Mn have less impact on combustion, or have a relatively small range in concentration in grasses. Elements such as As, Cd, Co, Cr, Cu, Hg, Mo, Ni, Pb, V, or Zn might be a cause for concern, and German and Austrian wood pellet ENplus quality standards include analysis for heavy metals [33]. Such elements are normally present in grasses in very small quantities, significantly lower amounts than
found in wood [34]. This is one of the few advantages that grass biomass has over woody biomass, as relatively slow growing woody species have the potential to accumulate significant quantities of heavy metals such as mercury. One exception might be with grass grown on fields treated with industrial sewage sludge [35], a relatively rare occurrence in the Northeast USA due to restrictions on such applications. Ash melting behavior [32] is an important parameter, but determination of melting temperatures is not feasible for routine sample analysis. A contributor to elemental contents of grass that is difficult to assess is that of surface soil contamination. Over 100 lots of mature hay bales were sampled in New York in the fall of 2011, and ash content ranged from 38 to 212 g kg 1 (Cherney, JH, unpublished). Hay lots with high ash values also had corresponding very high Al, Fe, and Ti concentrations, indicative of soil contamination. Elemental components of a grass sample analysis originating from surface soil contamination will vary depending on soil type, plant digestion technique, and level of contamination [36]. Plant digestion analysis techniques only partially release elements bound in soil. Level of soil contamination is a function of how rough the soil surface is, soil moisture content, the particular type of mowing, raking and baling equipment used, and whether the grass is baled in the fall, or left standing or windrowed over winter. Soil contamination is also typically highly variable from bale to bale. The plant chemical components most impacted by surface soil contamination are silica and total ash. Soil contamination negatively affects gross energy value of the feedstock on a dry matter basis through dilution effects. 1.2. Quality evaluation Quality standards for grass pellets would encourage their sustainable production and consumption for space heating. A survey of bioenergy experts in the EU showed that a majority of respondents agreed that there was a lack of European standards for bioenergy production, trade, and development [37]. The majority of respondents also agreed that a European-level standard would help to develop a sustainable bioenergy trade and encourage public acceptance of biomass energy, and that certification of bioenergy was necessary. Certification of biomass provides added value through product differentiation, enhancing market competitiveness. A multi-criteria assessment model was used recently to rank biomass pellets for suitability for use in large heat and power generation plants [38]. Technical, environmental and economic factors were assigned weights and evaluated for the quantitative and qualitative criteria. While this model is useful for energy planning, it would have limited value for evaluating specific lots of pellets for compositional parameters. Physical characteristics are the primary basis for evaluating wood pellet quality in North America [39]. Properties included in the fuel quality grade specifications include fines, bulk density,
Table 1 Pellet Fuels Institute fuel grade requirements for wood pellets [39]. Parameter 3
Bulk density, kg m Fines, g kg 1 PDIa, g kg 1 Moisture, g kg 1 Total ashb, g kg 1 Chloride, g kg 1 Diameter, mm Length, % greater than 38.1 mm a b
Premium
Standard
Utility
641–737 65 P965 680 610 60.3 5.84–7.25 61
609–737 610 P950 6100 620 60.3 5.84–7.25 61
609–737 610 P950 6100 660 60.3 5.84–7.25 61
Pellet durability index. Total ash and chloride concentrations are expressed on a dry matter basis.
681
J.H. Cherney, V.K. Verma / Applied Energy 103 (2013) 679–684
diameter, length, pellet durability index, moisture content, ash content, and chloride content (Table 1). Concentration of chlorine is generally expressed as chloride ion. Fines are the percentage of fuel that pass through a screen. Screen size is stated by the specific standard method, it may vary with pellet diameter [40]. Bulk density is the fuel mass per unit volume of fuel, a uniform density ensures steady combustion behavior [41]. Pellet durability index (PDI) is a measure of the ability of fuels to resist degradation due to shipping and handling [40]. Of the available standardized pellet tumbling devices the device with the most accuracy and precision was the tumbler described by ASAE S269.4 [42]. Pellet durability can be evaluated immediately after pelleting (green strength) or after pellets have cured (cured strength). Particle density is included in European standards and is related to bulk density [24]. Fines are rarely mentioned in European literature. The goal of biomass densification is to produce a strong and durable product that will resist breakdown during handling, transportation and storage [43]. Feedstock composition has a major effect on densification, and is also impacted by particle size, feedstock conditioning and densification equipment variables. Herbaceous biomass, particularly overwintered warm-season grasses, can be particularly difficult to pellet consistently. Temperature and chemical composition influenced bonding of particles of wood or straw [44]. Compressive force, particle size and moisture content significantly affected pellet density in switchgrass [45]. Optimum densification conditions for switchgrass were achieved by preheating to greater than 75 °C at a moisture content of 8–15% [46]. Currently in Northeastern USA and Canada, grass pellets are being produced using a wide range of pelleting equipment, some of this equipment is not well suited to grass pelleting. As most equipment currently used for grass pelleting is either designed for pelleting wood or animal feed materials, modifications to allow for grass pelleting are done by trial and error, and result in a range in pellet physical quality. Physical characteristics therefore are an important component of overall grass pellet quality, although chemical composition remains the critical component. Grass pellets are being used on a limited scale as a combustion fuel in Europe and North America. The major issues facing grass pellets for residential combustion are unlike those of wood pellets. Grass is not an ideal combustion fuel, with elevated levels of several problematic elements. The potential range in composition of grass biomass is tremendous. Mature grass harvested in NY in 2010 ranged from 0.1 to 13.4 g kg 1 Cl on a dry basis, depending on species, fertilization, and harvest management (Cherney JH, unpublished). Residential scale combustion equipment is now being designed to specifically address the compositional issues of grass pellets. A residential pellet boiler burning reed canary grass was within acceptable European emissions limits for CO and NOx, and emitted approximately one half of the particulates as wood pellets [47,48]. Also, small scale appliances can be fitted with equipment for exhaust gas after-treatment if necessary [49]. Fuel indexes that characterize a combustion feedstock can form a good basis for evaluation of combustion-related problems [50]. Wood pellet standards have been applied to grass pellets [51], but this system is inadequate for non-wood pellet fuels. Without any system for evaluating and ranking the suitability of grass biomass for residential combustion, the variability in grass composition will seriously hinder the development of a residential grass pellet combustion industry. The objective of this research is to develop a Quality Index for grass pellet evaluation system that weights the relative contributions of major physical and compositional parameters to generate a single value for evaluation and ranking of pellet quality for combustion.
2. Methodological approach 2.1. Parameter selection A set of parameters was chosen for consideration in an index (Table 2), based on available literature and on the typical list of characteristics used to evaluate pellets in North America. Parameters selected for an index should impact pellet durability, handling and feeding, combustion behavior, and/or combustion emissions. Of the physical characteristics used to evaluate wood pellet quality (Table 1), those that appear most appropriate to include in a grass pellet Quality Index include fines, bulk density, and durability index. Mechanical durability and bulk density are relatively independent, uncorrelated measurements [24,41,42]. Pellet length and diameter are not included in this proposed index, pellet diameter varies from 6 to 12 mm, depending on the country [52]. Although bulk density impacts shipping and storage efficiency, fines and durability index are considered more important from a grass pellet quality standpoint. Based on the previous discussion above, some of the compositional features are more important to the success of grass pellet combustion than physical characteristics. Key compositional characteristics essential for evaluating grass pellet quality include Cl, K, N, Si, and S, total ash, moisture content, and gross energy value, listed in relative order of importance. Chloride and K content are critical to the combustion process in small appliances, and the chloride and alkali combustion products were directly related to pellet composition of chloride and alkali [20]. While sodium (Na) is similar to K regarding negative combustion effects, grasses generally contain 50–100 times more K than Na, hence the negligible quantity of Na is not considered here. Nitrogen and silica are also important, but less so compared to Cl or K. Silica is often approximated by commercial labs using the acid-insoluble ash (AIA) procedure. Acid detergent solution quantitatively recovers all silica from biogenic and soil sources, and ashing of the acid detergent residue is a reasonable estimate for total silica [53]. The impact of sulfur content of grass pellet on combustion quality is slightly less than that of N or Si. Release of N, S, and Cl in emissions is generally well correlated with feedstock N, S, and Cl content [50]. Phosphorus is not included because it is much less variable in grasses, compared to some of the other elements, and is not greatly influenced by management issues, such as soil fertility or leaching. Factors that have an impact but are considered less important to overall grass pellet quality include gross energy content, total ash, and moisture content, although moisture content affects all the physical properties of grass pellets [54], as well as net calorific value and combustion efficiency [24]. Many new appliances are equipped with automated ash removal from the combustion chamber, such that total ash has less of a negative impact. Gross energy content of grasses is relatively consistent [55,56], assuming there is not a significant seed component. Although a variety of molar ratios have been developed to estimate ash melting, deposit Table 2 Relative ranking of parameters for a grass pellet Quality Index. Parameter
Rank
References discussing the parameters
Chlorine Potassium Nitrogen Pellet durability index Fines Silica Sulfur Total ash Moisture content Gross calorific value Bulk density
1 2 3 4 5 6 7 8 9 10 11
[20,25–27,29] [20,26,27] [30] [24,42] [52] [28,31] [28,31] [32] [24,56] [24] [39,40]
682
J.H. Cherney, V.K. Verma / Applied Energy 103 (2013) 679–684
buildup, or emission potential [50], inclusion of such ratios here is beyond the scope of this proposed index. 2.2. Identification of a working range for parameters Optimum values for each parameter were identified (Table 3), based on grass data found in the literature and on actual values determined for grass pellet samples over the past 6 years in New York State (Cherney JH, unpublished). Commercial forage analysis laboratory databases containing thousands of grass hay analyses over the past decade in the Northeastern USA also were utilized. Commercial labs in the region with such databases include DairyOne, Ithaca, NY, and Cumberland Valley Analytical Services, Inc., Hagerstown, MD. All optimum values listed in Table 3 have been observed in grass pellet samples in NY, although no one sample has contained all of these values. A maximum or minimum value was selected for each parameter, resulting in a working range of acceptable values for each parameter (Table 3). An upper or lower limit was also selected for each parameter, beyond which the pellet sample is considered as unacceptable for small scale combustion, regardless of the total index score for that lot of pellets. This is a lower limit for bulk density, durability index and gross energy (higher values are preferred), and an upper limit for all other parameters (lower values are preferred). 2.3. Proposed Quality Index For a given grass pellet sample, actual parameter values can be expressed as a proportion of the working range defined in Table 3. For example a total ash content of 46.2 is 0.44 of the working range (20–80 working range). For parameters with a reverse scale (bulk density, pellet durability index, and gross energy) the range is reversed. For example, a bulk density of 0.594 is 0.69 of the working range (0.705–0.545 working range). Summing these calculated values for all parameters generates the Quality Index value. The partial index value for a given parameter is calculated in the same manner, regardless of whether the actual value is inside or outside the working range. For example, a sample with bulk density of 0.737 means that the fraction of the working range is 0.20 (0.705–.545 working range), and the contribution of bulk density to the Quality Index value would be 0.20. Table 3 Selected physical and compositional parameters impacting grass pellet quality. Parameter 3
Bulk density, kg m Fines, g kg 1 PDId, g kg 1 Moisture, g kg 1
Composition (DM basis) Total ash, g kg 1 Gross calorific value, MJ kg S, g kg 1 Si (AIAf), g kg 1 N, g kg 1 K, g kg 1 Cl, g kg 1
1e
Optimuma
Max./min.b
Limitc
705 5.0 990 70
545 25.0 950 120
529 40.0 900 130
20 19.77 0.4 10 2.5 1.0 0.05
80 17.45 3.0 30 15.0 4.0 0.4
100 16.28 4.0 50 20.0 20.0 2.0
a Optimum value for each parameter, observed in at least one lot of grass pellets tested in NY. b Minimum value for working range for bulk density, durability index, and gross energy; maximum value for working range for all other parameters. c Minimum acceptable value for bulk density, durability index, and gross energy; maximum acceptable value for all other parameters. d Pellet durability index, measured after pellet curing. e Gross calorific value is the higher heating value (HHV), assuming the water component is in a liquid state at the end of combustion. f Acid insoluble ash (AIA).
3. Results and discussion While there are additional parameters that could be included in an index, including P, Mg, Ca, Na, Al, Mn and heavy metals, selection of parameters used in an index should be based on the relative importance of parameters discussed in the literature. Physical parameters of grass pellets are rarely discussed in the literature, but are an essential component, based on wood pellet quality priorities [33,39]. Compositional factors are much more important for grass compared to wood. Inclusion of Cl, K, N, Si and S can be justified based on their potential impact on combustion and/or emissions. Total ash and gross energy content are exceptions to this logic, as neither of these parameters strongly impacts the combustion process. While total ash concentration is often mentioned as a key factor in grass combustion, it is the composition of the ash that impacts combustion. Gross energy content also is cited as a key component, but in reality there is typically relatively minor variation in energy content of grasses (excluding grains), on average slightly lower than that found in premium wood pellets. Any soil contamination of the feedstock will increase ash and decrease gross energy. Total ash and gross energy are included in the Quality Index but are not heavily weighted. The structure of this index is quite flexible, and other factors, such as pellet odor [57], could be easily incorporated into the index if they are identified as important. It is reasonable to set a limit for each parameter that serves as a basis for rejecting a given lot of pellets as unacceptable for residential combustion. Individual parameter limits are somewhat arbitrary and not everyone will agree with the limits proposed in this system. Optimum values for parameters were chosen based on observed values. The working range selected for each parameter was selected based on the relative importance of each parameter to the combustion process. Relative contribution of a given parameter to the total Quality Index value is adjusted by modifying the working range. The smaller the working ranges of a given parameter, the larger the potential contribution of that parameter to the total index value. Eleven different grass pellet lots were evaluated using the Quality Index, four examples are shown in Table 4. The range in Quality Index values for the eleven lots of pellets was 5.25–16.0. The high ash content of Lot D pellets originating in Canada was likely caused by soil contamination during harvest. Bales of fall-harvested switchgrass in NY in both the fall of 2009 and 2010 were under 45 g kg 1 ash (Cherney JH, unpublished). All four of the feedstocks in Table 4 were subject to in-field leaching to some extent, while two of these pellet lots were poorly pelleted. The contribution of physical parameters to the total Quality Index for the four pellet lots in Table 4 was 38%, 17%, 56% and 51%, respectively. Chloride concentration was relatively low in all four pellet lots, and the contribution of Cl to the total Quality Index for the four pellet lots was 14%, 31%, 2%, and 19%, respectively. The relatively small working ranges for Cl and K weight these parameters heavily for the overall index value. For example, Cl content in grass hay analyzed at commercial laboratories in the Northeast USA averages between 6 and 7 g kg 1 Cl, with some samples exceeding 10 g kg 1. A chloride value of 6 g kg 1 generates a partial index value of 17.0 for Cl, exceeding the total Quality Index values for all pellet lots in Table 4. Of course, such a large Cl value would also make this pellet lot unacceptable by exceeding the Cl limit value. Working ranges of parameters were adjusted to reflect the relative importance of parameters. Grass pellets could be sorted into different classes based on Quality Index scores, one example of such classes is provided in Table 5. The minimum number of classes would consist of one class of pellets suitable for residential combustion appliances, one class
683
J.H. Cherney, V.K. Verma / Applied Energy 103 (2013) 679–684 Table 4 Calculated Quality Index values for several grass pellet lots. Parameter
Aa Analysis
A Indexb
B Analysis
B Index
C Analysis
C Index
D Analysis
D Index
Bulk density, kg m3 Fines, g kg 1 PDIc, g kg 1 Moisture, g kg 1
594 11.0 987 115
0.69 0.30 0.70 0.90
698 3.7 979 115
0.04 -0.07 0.27 0.90
647 44.5 916 112
0.36 1.98 1.85 0.84
623 114.0 917 91
0.51 5.45 1.83 0.42
46.2 19.25 0.8 13.2 11.2 3.7 0.3
0.44 0.22 0.15 0.16 0.70 0.93 0.71 5.25
47.1 19.32 1.0 12.3 11.4 6.9 0.8
0.45 0.19 0.23 0.12 0.71 2.00 2.14 6.96
39.0 18.81 1.1 19.5 13.0 5.7 0.1
0.32 0.41 0.27 0.48 0.84 1.57 0.14 9.01
73.3 18.62 0.9 40.9 6.0 5.1 1.1
0.89 0.50 0.19 1.55 0.28 1.40 3.00 16.0
Composition (DM basis) Total ash, g kg 1 Gross calorific value, MJ kg S, g kg 1 Si (AIAd), g kg 1 N, g kg 1 K, g kg 1 Cl, g kg 1 Calculated Quality Index a b c d
1
A = overwintered switchgrass pellets; B = reed canary grass (Phalaris arundinacea L.) pellets; C = mixed cool-season grass pellets; D = fall harvested switchgrass pellets. Individual parameter index values = proportion of the working range. Pellet durability index, measured after pellet curing. Acid insoluble ash (AIA).
Table 5 Proposed categories for grass pellet Quality Index values. Class A 66 (suitable for residential combustion in appropriate appliances) Class B >6–12 (may be suitable for residential combustion in appropriate appliances) Class U >12, or limit exceeded for one or more individual parameters (unsuitable for residential combustion)
of questionable quality for residential combustion, and a class consisting of pellets clearly unsuitable for residential combustion appliances. Pellet lots where one or more individual parameters exceed the defined limits fall into the unsuitable class. For the eleven lots of grass pellets evaluated, two were Class A, three were Class B, and six were Class U. However, if all pellet lots tested had been pelleted reasonably well, as Lot B pellets were, the eleven lots of grass pellets would then score out as seven in Class A, four in Class B, with no Class U pellet lots. 4. Conclusions The biomass heating industry in North America is relatively immature compared to Europe, and additional quality standards would encourage grass biomass producers to generate more consistent fuel quality, and help to gain consumer confidence in biomass fuels. Some method of evaluating and ranking grass pellets for both their physical and combustion properties is essential if these pellets are to be considered for possible residential combustion use. Any index needs to be flexible enough to allow inclusion of new parameters and adjustment of their relative ranking, as we move towards a better understanding of herbaceous biomass combustion. The proposed system roughly approximates a crude expert system. That is, the model was adjusted based on available facts and expert opinion, until it generated results for a given set of pellet parameters that might be deemed reasonable by an expert. Selection of parameters to include and their optimum values were based on available literature. Selection of acceptable limits and working ranges for parameters are based on the estimated relative importance of parameters to the residential combustion process. Although the values chosen could be debated by a group of experts, the model structure itself should allow for effective evaluation and ranking of grass pellet lots regardless of the specific values ultimately selected for acceptable limits and working ranges. Eleven different lots of grass pellets were evaluated and the
relative ranking based on the Quality Index appears reasonable. While much more elaborate systems are possible, the proposed system here is a relatively straightforward method of summing qualitatively different parameters into one index for evaluating and ranking grass pellets for residential combustion potential. An extension of existing quality labeling systems to grass pellets would encourage production of a more consistent fuel, and help to win the trust of industry and small combustion appliance owners. References [1] Cherney JH, Johnson KD, Lechtenberg VL, Hertel JM. Biomass yield, fiber composition and persistence of cool-season perennial grasses. Biomass 1986;10:175–86. [2] Lewandowski I, Scurlock JMO, Lindvall E, Christou M. The development and current status of perennial rhizomatous grasses as energy crops in the US and Europe. Biomass Bioenergy 2003;25:335–61. [3] Wright L, Turhollow A. Switchgrass selection as a ‘‘model’’ bioenergy crop: a history of the process. Biomass Bioenergy 2010;34:851–68. [4] Jensen K, Clark CD, Ellis P, English B, Menard J, Walsh M, et al. Farmer willingness to grow switchgrass for energy production. Biomass Bioenergy 2007;31:773–81. [5] Paulrud S, Laitila T. Farmers’ attitudes about growing energy crops: a choice experiment approach. Biomass Bioenergy 2010;34:1770–9. [6] Villeneuve J, Palacios JH, Savoie P, Godbout S. A critical review of emission standards and regulations regarding biomass combustion in small scale units (<3 MW). Bioresource Technol 2012;111:1–11. [7] Woodbury PB, Cherney JH, Wightman J, Duxbury JM, Cox WJ, Mohler CL, et al., Evaluating strategies for biomass fuel production in New York State, In: Third USDA symposium on greenhouse gases and carbon sequestration in agriculture and forestry. 21–24 March, 2005, Baltimore, (MD). USDA; 2005 250. [8] Mathews JA, Tan H. Biofuels and indirect land use change effects: the debate continues. Biofuels Bioprod Biorefin 2009;3:305–17. [9] Samson R, Ho Lem C, Bailey Stamler S, Dooper J. Developing energy crops for thermal applications: optimizing fuel quality energy security and GHG mitigation. In: Pimentel D, editor. Biofuels solar and wind as renewable energy systems: benefits and risks. New York: Springer; 2008 [chapter 16]. [10] Cherney JH, Johnson KD, Volenec JJ, Anliker KS. Chemical composition of herbaceous grass and legume species grown for maximum biomass production. Biomass 1988;17:215–38. [11] Cherney JH, Johnson KD, Volenec JJ, Greene DK. Biomass potential of selected grass and legume crops. Energy Sources 1991;13:283–92. [12] Cassida KA, Muir JP, Hussey MA, Read JC, Venuto BC, Ocumpaugh WR. Biofuel component concentrations and yields of switchgrass in south central US environments. Crop Sci 2005;45(2):682–92. [13] Casler MD, Cherney JH, Brummer EC. Biomass yield of naturalized populations and cultivars of reed canary grass. Bioenergy Res 2009;2:165–73. [14] Casler MD, Stendal CA, Kapich L, Vogel KP. Genetic diversity, plant adaptation regions, and gene pools for switchgrass. Crop Sci 2007;47(6):2261–73. [15] Paappanen T, Lindh T, Karki J, Impola R, Rinne S, and Lotjonen T, et al., The development of production and use of reed canary grass in Finland. In: 18th European Biomass Conference and Exhibition, 3–7 May, 2010, Lyon, p. 1616–20. [16] Burvall J. Influence of harvest time and soil type on fuel quality in reed canary grass. Biomass Bioenergy 1997;12:149–54.
684
J.H. Cherney, V.K. Verma / Applied Energy 103 (2013) 679–684
[17] Sander B. Properties of Danish biofuels and the requirements for power production. Biomass Bioenergy 1997;12:177–83. [18] Cherney JH, Cherney DJR, Bruulesma TW. Potassium management. In: Cherney JH, Cherney DJR, editors. Grass for dairy cattle. CAB Publishing, New York, 1998. p. 137–60 [chapter 6]. [19] Bakker RR, Jenkins BM. Feasibility of collecting naturally leached rice straw for thermal conversion. Biomass Bioenergy 2003;25:597–614. [20] Dayton DC, Jenkins BM, Turn SQ, Bakker RR, Williams RB, Belle-Qudry, et al. Release of inorganic constituents from leached biomass during thermal combustion. Energy Fuels 1999;13:860–70. [21] Samson R, Mani S, Boddey R, Sokhansanj S, Quesada D, Urquiaga S, et al. The potential of C4 perennial grasses for developing a global BIOHEAT industry. Crit Rev Plant Sci 2005;24:461–95. [22] Adler PA, Sanderson MA, Boateng AA, Weimer PJ, Jung HG. Biomass yield and biofuel quality of switchgrass harvested in fall and spring. Agron J 2006;98:1518–25. [23] Sanderson MA, Egg RP, Wiselogel AE. Biomass losses during harvest and storage of switchgrass. Biomass Bioenergy 1997;12(2):107–14. [24] Obernberger I, Thek G. Physical characterization and chemical composition of densified biomass fuels with regard to their combustion behaviour. Biomass Bioenergy 2004;27:653–69. [25] Bjorkman E, Stromberg B. Release of chlorine from biomass at pyrolysis and gasification conditions. Energy Fuels 1997;11:1026–32. [26] Jensen PA, Frandsen FJ, Dam-Johansen K, Sander B. Experimental investigation of the transformation and release to gas phase of potassium and chlorine during straw pyrolysis. Energy Fuels 2000;14:1280–5. [27] Passalacqua F, Zaetta C, Janssone R, Pigaht M, Grassi G, et al. Pellets in southern Europe; the state of the art of pellets utilization in southern Europe: new perspectives of pellets from agri-residues. In: 2nd World conference on biomass for energy, industry and climate protection. Florence (Italy)/Munich (Germany): ETA-Florence/WIP-Munich; 2004. [28] Jenkins BM, Baxter LL, Miles Jr TR, Miles TR. Combustion properties of biomass. Fuel Process Technol 1998;54:17–46. [29] Kaufmann H, Chlorine-compounds in emissions and residues from the combustion of herbaceous biomass. Thesis no. 12429, Swiss Federal Institute of Technology (ETH), Zurich; 1997. [30] Crutzen PJ, Mosier AR, Smith KA, Winiwarter W. N2O release from agro-biofuel production negates global warming reduction by replacing fossil fuels. Atmos Chem Phys Discuss 2007;7:11191–205. [31] Baxter LL, Miles TR, Miles Jr TR, Jenkins BM, Milne T, Dayton D, et al. The behavior of inorganic material in biomass-fired power boilers: field and laboratory experiences. Fuel Process Technol 1998;54:47–78. [32] Paulrud S, Nilsson C, Ohman M. Reed canary-grass ash composition and its melting behavior during combust ion. Fuel 2001;80:1391–8. [33] DEPI, Certification of wood pellets for the end consumer market. Manual for Germany and Austria. Deutsches Pelletinstitut; 2010. [cited 10.05.12]. [34] Obernberger I, Biedermann F, Widmann W, Riedl R. Concentrations of inorganic elements in biomass fuels and recovery in the different ash fractions. Biomass Bioenergy 1997;12(3):211–24. [35] McBride MB, Cherney JH. Molybdenum, sulfur and other trace elements in farm soils and forages after sewage sludge application. Commun Soil Sci Plant Anal 2004;35:517–35. [36] Cherney JH, Robinson DL. A comparison of plant digestion methods for identifying soil contamination by Ti analysis. Agron J 1983;75(1):145–7. [37] Magar SB, Pelkonen P, Tahvanainen L, Toivonen R, Toppinen A. Growing trade of bioenergy in the EU: public acceptability, policy harmonization, European standards and certification needs. Biomass Bioenergy 2011;35:3318–27.
[38] Sultana A, Kumar A. Ranking of biomass pellets by integration of economic, environmental and technical factors. Biomass Bioenergy 2012;39:344–55. [39] Pellet Fuels Institute, Pellet Fuels Institute standard specification for residential/commercial densified fuel. Arlington: Pellet Fuels Institute; 25 October, 2010.. [40] ASAE, Cubes, Pellets, and Crumbles—definitions and methods for determining density, durability, and moisture content. ASAE S269.4. St. Joseph: American Society of Agricultural and Biological Engineers; 1991. [41] Fiedler F. The state of the art of small-scale pellet-based heating systems and relevant regulations in Sweden, Austria, and Germany. Renew Sustain Energy Rev 2004;8:201–21. [42] Temmerman M, Rabier F, Daugbjerg Jensen P, Hartmann H, Bohm T. Comparative study of durability test methods for pellets and briquettes. Biomass Bioenergy 2006;30:964–72. [43] Kaliyan N, Morey RV. Factors affecting strength and durability of densified biomass products. Biomass Bioenergy 2009;33:337–59. [44] Stelte W, Holm JK, Sandi AR, Barsberg S, Ahrenfeldt J, Henriksen UB. A study of bonding and failure mechanisms in fuel pellets from different biomass resources. Biomass Bioenergy 2011;35:910–8. [45] Mani S, Tabil LG, Sokhansanj S. Effects of compressive force, particle size and moisture content on mechanical properties of biomass pellets from grasses. Biomass Bioenergy 2006;30:648–54. [46] Kaliyan N, Morey RV. Densification characteristics of corn stover and switchgrass. Trans ASABE 2009;52(3):907–20. [47] Verma VK, Bram S, Gauthier G, De Ruyck J. Evaluation of the performance of a multi-fuel domestic boiler with respect to the existing European standard and quality labels: Part-1. Biomass Bioenergy 2010;35:80–9. [48] Verma VK, Bram S, Gauthier G, De Ruyck J. Performance of a domestic pellet boiler as a function of operational loads: Part-2. Biomass Bioenergy 2010;35:272–9. [49] Messerer A, Schmatloch V, Poschl U, Niessner R. Combined particle emission reduction and heat recovery from combustion exhaust – a novel approach for small wood-fired appliances. Biomass Bioenergy 2007;21:512–21. [50] Sommersacher P, Brunner T, Obernberger I. Fuel indexes: a novel method for the evaluation of relevant combustion properties of new biomass fuels. Energy Fuels 2011;26:380–90. [51] Vermont Grass Energy Partnership, Technical assessment of grass pellets as boiler fuel in Vermont, Final Report, VT Sustainable Jobs Fund, Univ. of VT extension service, and biomass energy resource center, Montpelier, VT; 12 January, 2011. . [52] Verma VK, Bram S, De Ruyck J. Small scale biomass heating systems: standards, quality labeling and market driving factors – an EU outlook. Biomass Bioenergy 2009;33:1393–402. [53] Van Soest PJ, Robertson JB, Lewis BA. Methods for dietary fiber, neutral detergent fiber, and nonstarch polysaccharides in relation to animal nutrition. J Dairy Sci 1991;74:3583–97. [54] Colley Z, Fasina OO, Bransby D, Lee YY. Moisture effect on the physical characteristics of switchgrass pellets. Trans ASABE 2006;49(6):1845–51. [55] Aravindhakshan SC, Epplin FM, Taliaferro CM. Economics of switchgrass and miscanthus relative to coal as feedstock for generating electricity. Biomass Bioenergy 2010;34:1375–83. [56] Lemus R, Brummer EC, Moore KJ, Molstad NE, Burras CL, Barker MF. Biomass yield and quality of 20 switchgrass populations in southern Iowa. USA Biomass Bioenergy 2002;23:433–42. [57] Stahl M, Wikstrom F. Swedish perspective on wood fuel pellets for household heating: a modified standard for pellets could reduce end-user problems. Biomass Bioenergy 2009;33:803–9.
Contents lists available at SciVerse ScienceDirect
Applied Energy journal homepage: www.elsevier.com/locate/apenergy
Grass pellet Quality Index: A tool to evaluate suitability of grass pellets for small scale combustion systems Jerome H. Cherney a,⇑, Vijay Kumar Verma b a b
Department of Crop & Soil Sciences, 503 Bradfield Hall, Cornell University, Ithaca, NY 14853, United States Department of Mechanical Engineering, Vrije Universiteit Brussel, Pleinlaan 2, 1050 Brussel, Belgium
h i g h l i g h t s " A Quality Index ranking grass pellets for combustion potential does not exist. " Grass pellet biomass is extremely variable and therefore quality control is essential. " Proposed system sums qualitatively different parameters into one index value. " Model structure allows for effective evaluation and ranking.
a r t i c l e
i n f o
a b s t r a c t
Article history: Received 16 May 2012 Received in revised form 14 October 2012 Accepted 16 October 2012 Available online 17 November 2012
US renewable fuels policy strongly encourages biomass crop production, which should lead to expansion of biomass heating scenarios. Chemical composition of grass biomass can be extremely variable, depending on species, soil fertility, and harvest management. Biomass quality concerns have hindered the development of grass biomass for residential combustion, with no comprehensive evaluation system for grass pellet quality. Quality labeling will strengthen the fledgling grass biomass heating market, gain consumer confidence, and help to control combustion related emissions. The proposed system sums qualitatively different parameters into one Quality Index for relative evaluation and ranking of grass pellets for residential combustion potential. Parameters were selected and weighted for their relative importance based on available literature. Weighting was accomplished by adjusting the compositional working range for each parameter. A limit also was established for each parameter, beyond which the pellet lot was considered as unacceptable for residential combustion, regardless of the total Quality Index score. The model structure allows for effective evaluation and ranking of grass pellet lots regardless of the specific values ultimately chosen for acceptable limits and working ranges by the industry. Applying the Quality Index to a range of grass pellet types resulted in a reasonable ranking of pellets based on physical characteristics and composition. Ó 2012 Elsevier Ltd. All rights reserved.
Keywords: Grass composition Grass pellets Quality Index Residential combustion Solid biofuel
1. Introduction Rural North America has a tremendous capacity for energy production [1–3]. While the ongoing desire for energy security is the primary driving force for alternative energy development in the USA, environmental issues will sooner or later overshadow energy supply issues. An ideal alternative solid biomass feedstock should be nearly carbon neutral, without significant net increase in atmospheric carbon dioxide. It also needs to be seen as a valuable crop from the farmer’s perspective [4,5].
⇑ Corresponding author. Tel.: +1 607 255 0945; fax: +1 607 255 2644. E-mail addresses: (V.K. Verma).
[email protected]
(J.H.
Cherney),
[email protected]
0306-2619/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.apenergy.2012.10.050
The Northeastern USA has significant space heating demand, and most states in the region have state regulations controlling combustion emissions [6]. New York and the New England states represent approximately 80% of the total heating oil demand in the USA. The Northeast USA has millions of acres of abandoned and underutilized land suitable for grass biomass, without interfering with traditional agricultural crops and without the requirement for establishment [7]. Existing mixed grass stands are appropriate for grass biomass production if the energy conversion process is combustion. The use of such lands for this purpose makes this biomass option relatively immune from the indirect land use change debate [8]. Energy conversion efficiency can be very high with grass combustion [9]. Grass energy farming is a small-scale, low-technology, closedloop energy system that will result in rural jobs and economic
680
J.H. Cherney, V.K. Verma / Applied Energy 103 (2013) 679–684
diversification, absorbing excess production capacity. In general, the desired feedstock for biomass combustion is opposite of that required for ruminant animal forage, with a range in composition among herbaceous plant species [10,11]. While grass breeding for biomass will eventually lead to compositional changes in the feedstock [12–14], major compositional changes can be achieved in the near-term through agronomic and harvest management [15]. 1.1. Biomass composition and combustion Soil type and inherent soil fertility can strongly impact mineral uptake [16]. Plant uptake of potassium (K) and chloride (Cl) are generally well correlated, due to luxury uptake of both elements in excess of plant requirements, and the common practice of fertilizing with KCl [17]. Potassium concentration has been reported as low as 0.6 g kg 1 to as high as 70 g kg 1 in cool-season grasses on a dry matter basis [18]. Concentration of most elements in grass decreases with plant age, making mature plants more desirable for combustion from a compositional standpoint. Harvest management can have a major impact on grass composition, particularly for water soluble components such as K and Cl [19,20]. Delayed baling following mowing, as well as overwintering grass in the field, also will reduce insoluble nitrogen (N) and silica (Si) in grass feedstock due to the preferential loss of inflorescence and leaf blade which are higher in N and Si content than other plant parts [21,22]. Dry matter losses in storage also are possible [23]. Biomass quality issues have hindered commercialization of grass pellets in residential combustion systems [24]. The most serious quality issues with grass feedstock are generally considered to be the alkali and chloride content [17,25–27]. Boiler corrosion and fouling are directly related to alkali and chloride content. Particulate emissions primarily consist of aerosol-forming elements like potassium and chloride, as well as sulfur (S). Chloride also acts as a catalyst, facilitating the movement of iron away from metal surfaces and the deposition of inorganic compounds [28,29]. Potassium is the primary alkali element present in grasses, with typically very low concentration of sodium [20]. Release of K can be minimized by controlling combustion temperature, but this does not prevent the release of Cl [25,26]. Nitrogen content of grass has little impact on combustion efficiency but is undesirable from an environmental standpoint. Nitrogen oxides are the second most important contributor to global acidification from human activities [25]. There are concerns that N release to the atmosphere from biofuel production negates any green house gas reduction benefits [30]. Nitrogen emissions are positively correlated with feedstock N content. Sulfur and Si, in combination with alkali, lead to reactions associated with fouling and slagging in boilers [28,31]. In general, approximately one half of the total ash content of grass is silica. Silica, in combination with K and other elements, affect the ash melting behavior in grasses [32]. The total amount of ash impacts the design of ash handling and storage systems for a given appliance, but ash content per se is not a major drawback to grass combustion within certain limits. Some residential heating appliances currently available are able to handle ash content up to 10%. Of course, total ash content is likely to be positively correlated with concentrations of undesirable elements, particularly Si and K. High total ash and Si content result in lower energy content. Other ash forming elements such as P, Ca, Mg, Al, Fe, and Mn have less impact on combustion, or have a relatively small range in concentration in grasses. Elements such as As, Cd, Co, Cr, Cu, Hg, Mo, Ni, Pb, V, or Zn might be a cause for concern, and German and Austrian wood pellet ENplus quality standards include analysis for heavy metals [33]. Such elements are normally present in grasses in very small quantities, significantly lower amounts than
found in wood [34]. This is one of the few advantages that grass biomass has over woody biomass, as relatively slow growing woody species have the potential to accumulate significant quantities of heavy metals such as mercury. One exception might be with grass grown on fields treated with industrial sewage sludge [35], a relatively rare occurrence in the Northeast USA due to restrictions on such applications. Ash melting behavior [32] is an important parameter, but determination of melting temperatures is not feasible for routine sample analysis. A contributor to elemental contents of grass that is difficult to assess is that of surface soil contamination. Over 100 lots of mature hay bales were sampled in New York in the fall of 2011, and ash content ranged from 38 to 212 g kg 1 (Cherney, JH, unpublished). Hay lots with high ash values also had corresponding very high Al, Fe, and Ti concentrations, indicative of soil contamination. Elemental components of a grass sample analysis originating from surface soil contamination will vary depending on soil type, plant digestion technique, and level of contamination [36]. Plant digestion analysis techniques only partially release elements bound in soil. Level of soil contamination is a function of how rough the soil surface is, soil moisture content, the particular type of mowing, raking and baling equipment used, and whether the grass is baled in the fall, or left standing or windrowed over winter. Soil contamination is also typically highly variable from bale to bale. The plant chemical components most impacted by surface soil contamination are silica and total ash. Soil contamination negatively affects gross energy value of the feedstock on a dry matter basis through dilution effects. 1.2. Quality evaluation Quality standards for grass pellets would encourage their sustainable production and consumption for space heating. A survey of bioenergy experts in the EU showed that a majority of respondents agreed that there was a lack of European standards for bioenergy production, trade, and development [37]. The majority of respondents also agreed that a European-level standard would help to develop a sustainable bioenergy trade and encourage public acceptance of biomass energy, and that certification of bioenergy was necessary. Certification of biomass provides added value through product differentiation, enhancing market competitiveness. A multi-criteria assessment model was used recently to rank biomass pellets for suitability for use in large heat and power generation plants [38]. Technical, environmental and economic factors were assigned weights and evaluated for the quantitative and qualitative criteria. While this model is useful for energy planning, it would have limited value for evaluating specific lots of pellets for compositional parameters. Physical characteristics are the primary basis for evaluating wood pellet quality in North America [39]. Properties included in the fuel quality grade specifications include fines, bulk density,
Table 1 Pellet Fuels Institute fuel grade requirements for wood pellets [39]. Parameter 3
Bulk density, kg m Fines, g kg 1 PDIa, g kg 1 Moisture, g kg 1 Total ashb, g kg 1 Chloride, g kg 1 Diameter, mm Length, % greater than 38.1 mm a b
Premium
Standard
Utility
641–737 65 P965 680 610 60.3 5.84–7.25 61
609–737 610 P950 6100 620 60.3 5.84–7.25 61
609–737 610 P950 6100 660 60.3 5.84–7.25 61
Pellet durability index. Total ash and chloride concentrations are expressed on a dry matter basis.
681
J.H. Cherney, V.K. Verma / Applied Energy 103 (2013) 679–684
diameter, length, pellet durability index, moisture content, ash content, and chloride content (Table 1). Concentration of chlorine is generally expressed as chloride ion. Fines are the percentage of fuel that pass through a screen. Screen size is stated by the specific standard method, it may vary with pellet diameter [40]. Bulk density is the fuel mass per unit volume of fuel, a uniform density ensures steady combustion behavior [41]. Pellet durability index (PDI) is a measure of the ability of fuels to resist degradation due to shipping and handling [40]. Of the available standardized pellet tumbling devices the device with the most accuracy and precision was the tumbler described by ASAE S269.4 [42]. Pellet durability can be evaluated immediately after pelleting (green strength) or after pellets have cured (cured strength). Particle density is included in European standards and is related to bulk density [24]. Fines are rarely mentioned in European literature. The goal of biomass densification is to produce a strong and durable product that will resist breakdown during handling, transportation and storage [43]. Feedstock composition has a major effect on densification, and is also impacted by particle size, feedstock conditioning and densification equipment variables. Herbaceous biomass, particularly overwintered warm-season grasses, can be particularly difficult to pellet consistently. Temperature and chemical composition influenced bonding of particles of wood or straw [44]. Compressive force, particle size and moisture content significantly affected pellet density in switchgrass [45]. Optimum densification conditions for switchgrass were achieved by preheating to greater than 75 °C at a moisture content of 8–15% [46]. Currently in Northeastern USA and Canada, grass pellets are being produced using a wide range of pelleting equipment, some of this equipment is not well suited to grass pelleting. As most equipment currently used for grass pelleting is either designed for pelleting wood or animal feed materials, modifications to allow for grass pelleting are done by trial and error, and result in a range in pellet physical quality. Physical characteristics therefore are an important component of overall grass pellet quality, although chemical composition remains the critical component. Grass pellets are being used on a limited scale as a combustion fuel in Europe and North America. The major issues facing grass pellets for residential combustion are unlike those of wood pellets. Grass is not an ideal combustion fuel, with elevated levels of several problematic elements. The potential range in composition of grass biomass is tremendous. Mature grass harvested in NY in 2010 ranged from 0.1 to 13.4 g kg 1 Cl on a dry basis, depending on species, fertilization, and harvest management (Cherney JH, unpublished). Residential scale combustion equipment is now being designed to specifically address the compositional issues of grass pellets. A residential pellet boiler burning reed canary grass was within acceptable European emissions limits for CO and NOx, and emitted approximately one half of the particulates as wood pellets [47,48]. Also, small scale appliances can be fitted with equipment for exhaust gas after-treatment if necessary [49]. Fuel indexes that characterize a combustion feedstock can form a good basis for evaluation of combustion-related problems [50]. Wood pellet standards have been applied to grass pellets [51], but this system is inadequate for non-wood pellet fuels. Without any system for evaluating and ranking the suitability of grass biomass for residential combustion, the variability in grass composition will seriously hinder the development of a residential grass pellet combustion industry. The objective of this research is to develop a Quality Index for grass pellet evaluation system that weights the relative contributions of major physical and compositional parameters to generate a single value for evaluation and ranking of pellet quality for combustion.
2. Methodological approach 2.1. Parameter selection A set of parameters was chosen for consideration in an index (Table 2), based on available literature and on the typical list of characteristics used to evaluate pellets in North America. Parameters selected for an index should impact pellet durability, handling and feeding, combustion behavior, and/or combustion emissions. Of the physical characteristics used to evaluate wood pellet quality (Table 1), those that appear most appropriate to include in a grass pellet Quality Index include fines, bulk density, and durability index. Mechanical durability and bulk density are relatively independent, uncorrelated measurements [24,41,42]. Pellet length and diameter are not included in this proposed index, pellet diameter varies from 6 to 12 mm, depending on the country [52]. Although bulk density impacts shipping and storage efficiency, fines and durability index are considered more important from a grass pellet quality standpoint. Based on the previous discussion above, some of the compositional features are more important to the success of grass pellet combustion than physical characteristics. Key compositional characteristics essential for evaluating grass pellet quality include Cl, K, N, Si, and S, total ash, moisture content, and gross energy value, listed in relative order of importance. Chloride and K content are critical to the combustion process in small appliances, and the chloride and alkali combustion products were directly related to pellet composition of chloride and alkali [20]. While sodium (Na) is similar to K regarding negative combustion effects, grasses generally contain 50–100 times more K than Na, hence the negligible quantity of Na is not considered here. Nitrogen and silica are also important, but less so compared to Cl or K. Silica is often approximated by commercial labs using the acid-insoluble ash (AIA) procedure. Acid detergent solution quantitatively recovers all silica from biogenic and soil sources, and ashing of the acid detergent residue is a reasonable estimate for total silica [53]. The impact of sulfur content of grass pellet on combustion quality is slightly less than that of N or Si. Release of N, S, and Cl in emissions is generally well correlated with feedstock N, S, and Cl content [50]. Phosphorus is not included because it is much less variable in grasses, compared to some of the other elements, and is not greatly influenced by management issues, such as soil fertility or leaching. Factors that have an impact but are considered less important to overall grass pellet quality include gross energy content, total ash, and moisture content, although moisture content affects all the physical properties of grass pellets [54], as well as net calorific value and combustion efficiency [24]. Many new appliances are equipped with automated ash removal from the combustion chamber, such that total ash has less of a negative impact. Gross energy content of grasses is relatively consistent [55,56], assuming there is not a significant seed component. Although a variety of molar ratios have been developed to estimate ash melting, deposit Table 2 Relative ranking of parameters for a grass pellet Quality Index. Parameter
Rank
References discussing the parameters
Chlorine Potassium Nitrogen Pellet durability index Fines Silica Sulfur Total ash Moisture content Gross calorific value Bulk density
1 2 3 4 5 6 7 8 9 10 11
[20,25–27,29] [20,26,27] [30] [24,42] [52] [28,31] [28,31] [32] [24,56] [24] [39,40]
682
J.H. Cherney, V.K. Verma / Applied Energy 103 (2013) 679–684
buildup, or emission potential [50], inclusion of such ratios here is beyond the scope of this proposed index. 2.2. Identification of a working range for parameters Optimum values for each parameter were identified (Table 3), based on grass data found in the literature and on actual values determined for grass pellet samples over the past 6 years in New York State (Cherney JH, unpublished). Commercial forage analysis laboratory databases containing thousands of grass hay analyses over the past decade in the Northeastern USA also were utilized. Commercial labs in the region with such databases include DairyOne, Ithaca, NY, and Cumberland Valley Analytical Services, Inc., Hagerstown, MD. All optimum values listed in Table 3 have been observed in grass pellet samples in NY, although no one sample has contained all of these values. A maximum or minimum value was selected for each parameter, resulting in a working range of acceptable values for each parameter (Table 3). An upper or lower limit was also selected for each parameter, beyond which the pellet sample is considered as unacceptable for small scale combustion, regardless of the total index score for that lot of pellets. This is a lower limit for bulk density, durability index and gross energy (higher values are preferred), and an upper limit for all other parameters (lower values are preferred). 2.3. Proposed Quality Index For a given grass pellet sample, actual parameter values can be expressed as a proportion of the working range defined in Table 3. For example a total ash content of 46.2 is 0.44 of the working range (20–80 working range). For parameters with a reverse scale (bulk density, pellet durability index, and gross energy) the range is reversed. For example, a bulk density of 0.594 is 0.69 of the working range (0.705–0.545 working range). Summing these calculated values for all parameters generates the Quality Index value. The partial index value for a given parameter is calculated in the same manner, regardless of whether the actual value is inside or outside the working range. For example, a sample with bulk density of 0.737 means that the fraction of the working range is 0.20 (0.705–.545 working range), and the contribution of bulk density to the Quality Index value would be 0.20. Table 3 Selected physical and compositional parameters impacting grass pellet quality. Parameter 3
Bulk density, kg m Fines, g kg 1 PDId, g kg 1 Moisture, g kg 1
Composition (DM basis) Total ash, g kg 1 Gross calorific value, MJ kg S, g kg 1 Si (AIAf), g kg 1 N, g kg 1 K, g kg 1 Cl, g kg 1
1e
Optimuma
Max./min.b
Limitc
705 5.0 990 70
545 25.0 950 120
529 40.0 900 130
20 19.77 0.4 10 2.5 1.0 0.05
80 17.45 3.0 30 15.0 4.0 0.4
100 16.28 4.0 50 20.0 20.0 2.0
a Optimum value for each parameter, observed in at least one lot of grass pellets tested in NY. b Minimum value for working range for bulk density, durability index, and gross energy; maximum value for working range for all other parameters. c Minimum acceptable value for bulk density, durability index, and gross energy; maximum acceptable value for all other parameters. d Pellet durability index, measured after pellet curing. e Gross calorific value is the higher heating value (HHV), assuming the water component is in a liquid state at the end of combustion. f Acid insoluble ash (AIA).
3. Results and discussion While there are additional parameters that could be included in an index, including P, Mg, Ca, Na, Al, Mn and heavy metals, selection of parameters used in an index should be based on the relative importance of parameters discussed in the literature. Physical parameters of grass pellets are rarely discussed in the literature, but are an essential component, based on wood pellet quality priorities [33,39]. Compositional factors are much more important for grass compared to wood. Inclusion of Cl, K, N, Si and S can be justified based on their potential impact on combustion and/or emissions. Total ash and gross energy content are exceptions to this logic, as neither of these parameters strongly impacts the combustion process. While total ash concentration is often mentioned as a key factor in grass combustion, it is the composition of the ash that impacts combustion. Gross energy content also is cited as a key component, but in reality there is typically relatively minor variation in energy content of grasses (excluding grains), on average slightly lower than that found in premium wood pellets. Any soil contamination of the feedstock will increase ash and decrease gross energy. Total ash and gross energy are included in the Quality Index but are not heavily weighted. The structure of this index is quite flexible, and other factors, such as pellet odor [57], could be easily incorporated into the index if they are identified as important. It is reasonable to set a limit for each parameter that serves as a basis for rejecting a given lot of pellets as unacceptable for residential combustion. Individual parameter limits are somewhat arbitrary and not everyone will agree with the limits proposed in this system. Optimum values for parameters were chosen based on observed values. The working range selected for each parameter was selected based on the relative importance of each parameter to the combustion process. Relative contribution of a given parameter to the total Quality Index value is adjusted by modifying the working range. The smaller the working ranges of a given parameter, the larger the potential contribution of that parameter to the total index value. Eleven different grass pellet lots were evaluated using the Quality Index, four examples are shown in Table 4. The range in Quality Index values for the eleven lots of pellets was 5.25–16.0. The high ash content of Lot D pellets originating in Canada was likely caused by soil contamination during harvest. Bales of fall-harvested switchgrass in NY in both the fall of 2009 and 2010 were under 45 g kg 1 ash (Cherney JH, unpublished). All four of the feedstocks in Table 4 were subject to in-field leaching to some extent, while two of these pellet lots were poorly pelleted. The contribution of physical parameters to the total Quality Index for the four pellet lots in Table 4 was 38%, 17%, 56% and 51%, respectively. Chloride concentration was relatively low in all four pellet lots, and the contribution of Cl to the total Quality Index for the four pellet lots was 14%, 31%, 2%, and 19%, respectively. The relatively small working ranges for Cl and K weight these parameters heavily for the overall index value. For example, Cl content in grass hay analyzed at commercial laboratories in the Northeast USA averages between 6 and 7 g kg 1 Cl, with some samples exceeding 10 g kg 1. A chloride value of 6 g kg 1 generates a partial index value of 17.0 for Cl, exceeding the total Quality Index values for all pellet lots in Table 4. Of course, such a large Cl value would also make this pellet lot unacceptable by exceeding the Cl limit value. Working ranges of parameters were adjusted to reflect the relative importance of parameters. Grass pellets could be sorted into different classes based on Quality Index scores, one example of such classes is provided in Table 5. The minimum number of classes would consist of one class of pellets suitable for residential combustion appliances, one class
683
J.H. Cherney, V.K. Verma / Applied Energy 103 (2013) 679–684 Table 4 Calculated Quality Index values for several grass pellet lots. Parameter
Aa Analysis
A Indexb
B Analysis
B Index
C Analysis
C Index
D Analysis
D Index
Bulk density, kg m3 Fines, g kg 1 PDIc, g kg 1 Moisture, g kg 1
594 11.0 987 115
0.69 0.30 0.70 0.90
698 3.7 979 115
0.04 -0.07 0.27 0.90
647 44.5 916 112
0.36 1.98 1.85 0.84
623 114.0 917 91
0.51 5.45 1.83 0.42
46.2 19.25 0.8 13.2 11.2 3.7 0.3
0.44 0.22 0.15 0.16 0.70 0.93 0.71 5.25
47.1 19.32 1.0 12.3 11.4 6.9 0.8
0.45 0.19 0.23 0.12 0.71 2.00 2.14 6.96
39.0 18.81 1.1 19.5 13.0 5.7 0.1
0.32 0.41 0.27 0.48 0.84 1.57 0.14 9.01
73.3 18.62 0.9 40.9 6.0 5.1 1.1
0.89 0.50 0.19 1.55 0.28 1.40 3.00 16.0
Composition (DM basis) Total ash, g kg 1 Gross calorific value, MJ kg S, g kg 1 Si (AIAd), g kg 1 N, g kg 1 K, g kg 1 Cl, g kg 1 Calculated Quality Index a b c d
1
A = overwintered switchgrass pellets; B = reed canary grass (Phalaris arundinacea L.) pellets; C = mixed cool-season grass pellets; D = fall harvested switchgrass pellets. Individual parameter index values = proportion of the working range. Pellet durability index, measured after pellet curing. Acid insoluble ash (AIA).
Table 5 Proposed categories for grass pellet Quality Index values. Class A 66 (suitable for residential combustion in appropriate appliances) Class B >6–12 (may be suitable for residential combustion in appropriate appliances) Class U >12, or limit exceeded for one or more individual parameters (unsuitable for residential combustion)
of questionable quality for residential combustion, and a class consisting of pellets clearly unsuitable for residential combustion appliances. Pellet lots where one or more individual parameters exceed the defined limits fall into the unsuitable class. For the eleven lots of grass pellets evaluated, two were Class A, three were Class B, and six were Class U. However, if all pellet lots tested had been pelleted reasonably well, as Lot B pellets were, the eleven lots of grass pellets would then score out as seven in Class A, four in Class B, with no Class U pellet lots. 4. Conclusions The biomass heating industry in North America is relatively immature compared to Europe, and additional quality standards would encourage grass biomass producers to generate more consistent fuel quality, and help to gain consumer confidence in biomass fuels. Some method of evaluating and ranking grass pellets for both their physical and combustion properties is essential if these pellets are to be considered for possible residential combustion use. Any index needs to be flexible enough to allow inclusion of new parameters and adjustment of their relative ranking, as we move towards a better understanding of herbaceous biomass combustion. The proposed system roughly approximates a crude expert system. That is, the model was adjusted based on available facts and expert opinion, until it generated results for a given set of pellet parameters that might be deemed reasonable by an expert. Selection of parameters to include and their optimum values were based on available literature. Selection of acceptable limits and working ranges for parameters are based on the estimated relative importance of parameters to the residential combustion process. Although the values chosen could be debated by a group of experts, the model structure itself should allow for effective evaluation and ranking of grass pellet lots regardless of the specific values ultimately selected for acceptable limits and working ranges. Eleven different lots of grass pellets were evaluated and the
relative ranking based on the Quality Index appears reasonable. While much more elaborate systems are possible, the proposed system here is a relatively straightforward method of summing qualitatively different parameters into one index for evaluating and ranking grass pellets for residential combustion potential. An extension of existing quality labeling systems to grass pellets would encourage production of a more consistent fuel, and help to win the trust of industry and small combustion appliance owners. References [1] Cherney JH, Johnson KD, Lechtenberg VL, Hertel JM. Biomass yield, fiber composition and persistence of cool-season perennial grasses. Biomass 1986;10:175–86. [2] Lewandowski I, Scurlock JMO, Lindvall E, Christou M. The development and current status of perennial rhizomatous grasses as energy crops in the US and Europe. Biomass Bioenergy 2003;25:335–61. [3] Wright L, Turhollow A. Switchgrass selection as a ‘‘model’’ bioenergy crop: a history of the process. Biomass Bioenergy 2010;34:851–68. [4] Jensen K, Clark CD, Ellis P, English B, Menard J, Walsh M, et al. Farmer willingness to grow switchgrass for energy production. Biomass Bioenergy 2007;31:773–81. [5] Paulrud S, Laitila T. Farmers’ attitudes about growing energy crops: a choice experiment approach. Biomass Bioenergy 2010;34:1770–9. [6] Villeneuve J, Palacios JH, Savoie P, Godbout S. A critical review of emission standards and regulations regarding biomass combustion in small scale units (<3 MW). Bioresource Technol 2012;111:1–11. [7] Woodbury PB, Cherney JH, Wightman J, Duxbury JM, Cox WJ, Mohler CL, et al., Evaluating strategies for biomass fuel production in New York State, In: Third USDA symposium on greenhouse gases and carbon sequestration in agriculture and forestry. 21–24 March, 2005, Baltimore, (MD). USDA; 2005 250. [8] Mathews JA, Tan H. Biofuels and indirect land use change effects: the debate continues. Biofuels Bioprod Biorefin 2009;3:305–17. [9] Samson R, Ho Lem C, Bailey Stamler S, Dooper J. Developing energy crops for thermal applications: optimizing fuel quality energy security and GHG mitigation. In: Pimentel D, editor. Biofuels solar and wind as renewable energy systems: benefits and risks. New York: Springer; 2008 [chapter 16]. [10] Cherney JH, Johnson KD, Volenec JJ, Anliker KS. Chemical composition of herbaceous grass and legume species grown for maximum biomass production. Biomass 1988;17:215–38. [11] Cherney JH, Johnson KD, Volenec JJ, Greene DK. Biomass potential of selected grass and legume crops. Energy Sources 1991;13:283–92. [12] Cassida KA, Muir JP, Hussey MA, Read JC, Venuto BC, Ocumpaugh WR. Biofuel component concentrations and yields of switchgrass in south central US environments. Crop Sci 2005;45(2):682–92. [13] Casler MD, Cherney JH, Brummer EC. Biomass yield of naturalized populations and cultivars of reed canary grass. Bioenergy Res 2009;2:165–73. [14] Casler MD, Stendal CA, Kapich L, Vogel KP. Genetic diversity, plant adaptation regions, and gene pools for switchgrass. Crop Sci 2007;47(6):2261–73. [15] Paappanen T, Lindh T, Karki J, Impola R, Rinne S, and Lotjonen T, et al., The development of production and use of reed canary grass in Finland. In: 18th European Biomass Conference and Exhibition, 3–7 May, 2010, Lyon, p. 1616–20. [16] Burvall J. Influence of harvest time and soil type on fuel quality in reed canary grass. Biomass Bioenergy 1997;12:149–54.
684
J.H. Cherney, V.K. Verma / Applied Energy 103 (2013) 679–684
[17] Sander B. Properties of Danish biofuels and the requirements for power production. Biomass Bioenergy 1997;12:177–83. [18] Cherney JH, Cherney DJR, Bruulesma TW. Potassium management. In: Cherney JH, Cherney DJR, editors. Grass for dairy cattle. CAB Publishing, New York, 1998. p. 137–60 [chapter 6]. [19] Bakker RR, Jenkins BM. Feasibility of collecting naturally leached rice straw for thermal conversion. Biomass Bioenergy 2003;25:597–614. [20] Dayton DC, Jenkins BM, Turn SQ, Bakker RR, Williams RB, Belle-Qudry, et al. Release of inorganic constituents from leached biomass during thermal combustion. Energy Fuels 1999;13:860–70. [21] Samson R, Mani S, Boddey R, Sokhansanj S, Quesada D, Urquiaga S, et al. The potential of C4 perennial grasses for developing a global BIOHEAT industry. Crit Rev Plant Sci 2005;24:461–95. [22] Adler PA, Sanderson MA, Boateng AA, Weimer PJ, Jung HG. Biomass yield and biofuel quality of switchgrass harvested in fall and spring. Agron J 2006;98:1518–25. [23] Sanderson MA, Egg RP, Wiselogel AE. Biomass losses during harvest and storage of switchgrass. Biomass Bioenergy 1997;12(2):107–14. [24] Obernberger I, Thek G. Physical characterization and chemical composition of densified biomass fuels with regard to their combustion behaviour. Biomass Bioenergy 2004;27:653–69. [25] Bjorkman E, Stromberg B. Release of chlorine from biomass at pyrolysis and gasification conditions. Energy Fuels 1997;11:1026–32. [26] Jensen PA, Frandsen FJ, Dam-Johansen K, Sander B. Experimental investigation of the transformation and release to gas phase of potassium and chlorine during straw pyrolysis. Energy Fuels 2000;14:1280–5. [27] Passalacqua F, Zaetta C, Janssone R, Pigaht M, Grassi G, et al. Pellets in southern Europe; the state of the art of pellets utilization in southern Europe: new perspectives of pellets from agri-residues. In: 2nd World conference on biomass for energy, industry and climate protection. Florence (Italy)/Munich (Germany): ETA-Florence/WIP-Munich; 2004. [28] Jenkins BM, Baxter LL, Miles Jr TR, Miles TR. Combustion properties of biomass. Fuel Process Technol 1998;54:17–46. [29] Kaufmann H, Chlorine-compounds in emissions and residues from the combustion of herbaceous biomass. Thesis no. 12429, Swiss Federal Institute of Technology (ETH), Zurich; 1997. [30] Crutzen PJ, Mosier AR, Smith KA, Winiwarter W. N2O release from agro-biofuel production negates global warming reduction by replacing fossil fuels. Atmos Chem Phys Discuss 2007;7:11191–205. [31] Baxter LL, Miles TR, Miles Jr TR, Jenkins BM, Milne T, Dayton D, et al. The behavior of inorganic material in biomass-fired power boilers: field and laboratory experiences. Fuel Process Technol 1998;54:47–78. [32] Paulrud S, Nilsson C, Ohman M. Reed canary-grass ash composition and its melting behavior during combust ion. Fuel 2001;80:1391–8. [33] DEPI, Certification of wood pellets for the end consumer market. Manual for Germany and Austria. Deutsches Pelletinstitut; 2010.
[38] Sultana A, Kumar A. Ranking of biomass pellets by integration of economic, environmental and technical factors. Biomass Bioenergy 2012;39:344–55. [39] Pellet Fuels Institute, Pellet Fuels Institute standard specification for residential/commercial densified fuel. Arlington: Pellet Fuels Institute; 25 October, 2010.