- Email: [email protected]
Fuel quality changes in full tree logging residue during storage in roadside slash piles in Northwestern Ontario
b i o m a s s a n d b i o e n e r g y 4 2 ( 2 0 1 2 ) 4 3 e5 0
Available online at www.sciencedirect.com
http://www.elsevier.com/locate/biombioe
Fuel quality changes in full tree logging residue during storage in roadside slash piles in Northwestern Ontario Shuva Gautam*, Reino Pulkki, Chander Shahi, Mathew Leitch Faculty of Natural Resources Management, Lakehead University, 955 Oliver Road, Thunder Bay, Ontario, Canada P7B 5E1
article info
abstract
Article history:
The procurement of logging residue for energy production can be uneconomical due to
Received 29 June 2010
high moisture content and low gross calorific value. High moisture content in biomass
Received in revised form
lowers the energy density and transportation becomes less efficient. Logging residue, from
24 January 2012
full tree logging operations stored in roadside beehive and windrow piles, were examined
Accepted 20 March 2012
to determine the effect of storage method and duration on fuel quality in northwestern
Available online 18 April 2012
Ontario, Canada. The fuel attributes assessed were moisture content, gross calorific value and ash content. Windrows displayed lower moisture content than beehives. Softwoods
Keywords:
generally displayed lower moisture content and higher gross calorific values than hard-
Ash content
woods. Smaller diameter stems displayed higher gross calorific value and ash content than
Bioenergy
larger diameter stems. The moisture content (green basis) reduced from a green state to
Fuel quality
15.1% after two years of storage in roadside slash piles, whereas gross calorific value and
Gross calorific value
ash content did not change significantly with storage time. The gross calorific values
Moisture content
ranged from 19.5 to 23.1 MJ kg1 and the ash content ranged from 0.4% to 8.4% for all
Northwestern Ontario
species, components and storage years. The study demonstrates that the storage regime plays a significant role on the fuel quality of logging residue. Proper storage and drying techniques improve the fuel quality and net energy yield from logging residue biomass, thereby leading to an overall cost reduction of the biomass feedstock. ª 2012 Elsevier Ltd. All rights reserved.
1.
Introduction
In northwestern Ontario, approximately 13%e14% of volume harvested using the full tree to roadside, roundwood to mill system is left on the harvest site as logging residue in the form of slash piles [1,2]. The present options for slash piles are either: i) to leave these on site to decay where they pose a potential fire hazard and occupy valuable space where seedlings could be established [3], or ii) to dispose these slash piles through burning [2,3]. A third option is to use the slash piles for energy production in the light of global warming issues [4,5] and the current forest industry crisis that is partly a result of high energy costs. However, the economic viability
of procuring such low bulk density logging residue, already low in energy density, depends on its moisture content (MC) [7,8]. The fuel quality in bioenergy production is assessed based on the properties that affect the energy yield and costs. MC, gross calorific value and ash content of the logging residue are three properties used to assess the fuel quality in slash piles, as each of these properties determine the viability of biomass procurement for bioenergy production [9]. The gross calorific value of a fuel is a measure of the maximum amount of energy that is released during burning of a given quantity of the fuel [10]. However the net energy released will depend on its MC [6,11,12] and ash content [9]. Furthermore, the overall cost can
* Corresponding author. E-mail address: [email protected] (S. Gautam). 0961-9534/$ e see front matter ª 2012 Elsevier Ltd. All rights reserved. doi:10.1016/j.biombioe.2012.03.015
44
b i o m a s s a n d b i o e n e r g y 4 2 ( 2 0 1 2 ) 4 3 e5 0
160.0 140.0 120.0 100.0 80.0 60.0 40.0
-40.0
Materials and methods
2.1.
Study area and logging description
The study materials were located in harvest compartments in the eastern part of the Crossroute Forest, to the west of Atikokan, Ontario, Canada. A harvest compartment is a unit of land allocated for harvest under the same logging method and system during an annual work schedule (AWS) and is also prescribed a regeneration method in its totality. The AWS outlines all forest management activities, including harvesting that are scheduled for a one-year period extending from 1st April to 31st March of the following year. The mean monthly temperature and precipitation recorded at the Atikokan weather station (Lat. 48 45.6670 N, Long. 91 37.6830 W) during the storage period are displayed in Fig. 1. Average daily maximum temperatures for June, July and August are 21.7 C, 24.7 C and 22.8 C [18]. Average monthly precipitation for June, July and August are 103.3 mm, 97.9 mm and 97.8 mm, and the average annual precipitation is 739.6 mm [18]. The compartments were harvested using a conventional mechanized full tree system [19], utilizing feller bunchers, grapple skidders, stroke delimbers and slashers. As a result the logging residue (branches and tops) were left in roadside
Jul
May
Jan
2008
Mar
Nov
Jul
Sept
May
Jan
Mar
Nov
JUL
Sept
May
Jan
Nov
Jul
Sept
Mar
200 07
2006
MeanTemp(°°C)
2009
TotalPrecip(mm)
Fig. 1 e Mean monthly temperature and total monthly precipitation at the Atikokan weather station. Source: Environment Canada [18].
piles. A total of 36 piles were excavated for the purpose of the study. The shape of the slash piles studied fall either under the general category “half-section of sphere” or “half-ellipsoid” referred to as beehive and windrow, respectively. General shapes of piled woody debris have been outlined in reference [20]. Tree species were divided into the categories of softwoods and hardwoods. The softwoods included balsam fir (Abies balsamea (L.) Miller), white spruce (Picea glauca (Moench) Voss), black spruce (Picea marianana (Miller) B.S.P.) and jack pine (Pinus banksiana Lamb.); and the hardwoods included red maple (Acer rubrum L.), white birch (Betula papyrifera Marshall) and trembling aspen (Populus tremuloides Michx.). The heights of the piles were recorded as being either greater or less than 2 m; the width ranged from 8 m to 16 m. Storage years included logging residue stored for 1, 2 and 3 drying (summer) seasons.
2.2.
2.
May
0.0 -20.0
Jan
20.0 Mar
rise due to extra transport, dumping and handling costs to deal with ash [9,13]. Therefore, information on gross calorific value alone is not sufficient. Studies in Europe have shown that logging residue can be stored in-field to reduce MC and improve the fuel quality [8,15e17]. In the study conducted by Nurmi [15], Norway spruce (Picea abies (L.) Karst.) logging residue in Finland was piled 4.5 m high. It was found that the average MC reduced from 55% (green basis) to approximately 40% after being piled from September 1994 to September 1995. Another treatment of the same study that involved leaving the residue on the cutover from September 1994 to June 1995, then forwarding to the landing until September 1995 resulted in a lower MC of 28%. The form, duration of storage, and the weather conditions affect fuel quality [8]. However, no comparable published research could be found for Canadian continental climate and boreal forest conditions. Knowledge gained from European studies could be applied to northwestern Ontario to a certain extent, but due to differences in logging systems, stored material and weather conditions, this may not be completely valid. This is especially true in areas of continental climate conditions, where there is relatively low summer rainfall and high daily temperatures. The purpose of this study was to determine a storage regime for boreal forest conditions, with a dry continental climate, that enhances the fuel quality of full tree logging residue stored at roadside. The specific objectives were to determine the differences in fuel quality attributes (MC, gross calorific value and ash content) for: (i) storage years (three drying seasons), (ii) pile shapes (beehive and windrow), (iii) pile heights (more than and less than 2 m), (iv) species (softwood and hardwoods), and (v) logging residue’ branch sizes (small with diameter less than or equal to 4 cm and large with diameter greater than 4 cm).
Experimental design and statistical methods
Analysis of variance tests were carried out using the General Linear Model in SPSS to test the null hypotheses of no significant difference in MC, gross calorific value, and ash content for different storage years, pile shapes, pile heights, species, and logging residue diameters. The experimental designs for all the models are full factorial design, with MC, gross calorific value, and ash content as the dependent variables, respectively for each model and storage years, pile shapes, pile heights, species, and logging residue diameters as the independent variables. For gross calorific value and ash content analysis of variance, only two storage years were tested due to financial constraint. Storage years 1 and 2 were tested for ash content but for gross calorific value years 2 and 3 were tested because shorter term studies already exist [8,13,17] with longer term studies lacking. The data for each model was tested for normality and homogeneity of variance before conducting the analysis of variance. Any significant differences in the analysis were analyzed using post-hoc tests.
2.3.
Field sampling
A list of all harvest compartments in the area west of Atikokan for the annual work schedule years of 2006e07, 2007e08, and
b i o m a s s a n d b i o e n e r g y 4 2 ( 2 0 1 2 ) 4 3 e5 0
2008e09 containing slash piles was obtained from AbitibiBowater Inc., Fort Frances, Ontario. The slash piles in the compartments were categorized into storage years, pile shapes (beehive and windrow), and pile heights (more than and less than 2 m). MC data was collected during the summers of 2008 and 2009. The collection dates extended from 17th July to 30th July in the summer of 2008, and from 18th May to 3rd June in the summer of 2009. MC, gross calorific value and ash content were measured at the surface, as well as inside the slash piles. Ribbons were laid in an east to west orientation over the beehive type slash piles, and along the length of the windrow type slash piles at a randomly selected position. All biomass material that intersected the ribbon at the surface of the pile was cut with a chainsaw at the point of the intersection to expose the complete diameter of the biomass sample. A Protimeter Surveymaster moisture meter was used to measure the MC at various points on the cross section of the cut portion; the MC and species, as well as the diameter of the cross section were recorded. MC in this study is expressed in green weight basis which is the ratio of the weight of water in wood to the weight of green wood (sum of mass of water and mass of dry wood). This procedure was continued for all biomass intersecting the ribbon. Each slash pile was then dissected from the highest point of the pile vertically to a depth of 1.5 m for internal measurements, using a chainsaw. During the excavation process, at least one cross-sectional face was undisturbed and the excavated biomass materials were thrown to the side. A measurement tape was hung from the top of the pile to the depth of 1.5 m and a ribbon was used to divide the 1.5 m depth hole into 25 cm sections. At each 25 cm section, the MC and species were recorded for all exposed biomass throughout the entire depth of the 25 cm section; a total of 3359 MC readings were recorded. Wood residue samples from each section were collected and brought to the laboratory to determine their gross calorific values and ash contents. The species (softwood and hardwood) and logging residue diameters (small with diameter less than or equal to 4 cm and large with diameter greater than 4 cm) of each sample were recorded. Prior to using the moisture meter, an equality test was performed to test the reliability of the readings obtained from the moisture meter. The equality test involved collecting a number of samples of all species used in the study with varying MC values. The samples were measured for MC first using the moisture meter then by the oven-drying (laboratory) method. Paired t-tests were performed to determine if the MC values displayed by the moisture meter were significantly different from those determined through the oven-drying (laboratory) method. Two paired t-tests were performed: one for samples with MC below fibre saturation point (FSP) and the other for samples above the FSP as it is commonly known that electric moisture meters are much more accurate in measuring samples with MC below FSP. For samples below FSP, the MC means for moisture meter (17.4%) and laboratory (17.3%) measured samples were not significantly different (difference ¼ 0.20, SD ¼ 6.7, N ¼ 17, p ¼ 0.909). Also, for samples above FSP, the MC means for moisture meter (31%) and laboratory (29.2%) measured samples were not significantly different (difference ¼ 3.77, SD ¼ 14.6, p ¼ 0.392). In light of the
45
paired t-test results, the average values displayed by the moisture meter were used as estimates of MC. However, the MC values obtained from the moisture meter are only approximate values and should not be considered absolute as there are many variables that affect the readings displayed by the moisture meter.
2.4.
Laboratory methods
A Parr 6200 Bomb Calorimeter was used to determine the gross calorific values of the collected samples. The samples were dried to MC of 10e20%. After achieving the desired moisture level, the samples were ground using a large Wiley Mill followed by a mini Wiley mill. A bone dry sample of known mass (w1 g 0.5 g) in pellet form was placed in the bomb, which was then placed in the calorimeter. The calorimeter pressurizes the bomb with pure oxygen and is submerged under 2 L of distilled water of known temperature. The biomass in the bomb is ignited and the combustion raises the temperature of the bomb and consequently of the water. The calorimeter gives an output of heat value in MJ kg1 based on the water temperature change, taking into consideration the electrical input during ignition. A total of 48 samples were tested for gross calorific value determination. The ash content was determined according to procedures outlined in laboratory analytical procedures manual [21]. The wood biomass samples were ground following the same procedure used for the gross calorific value determination. Oven dry biomass between 0.5 and 2.0 g was weighed and placed in a crucible. Crucibles were labeled and dried thoroughly by placing them in a muffle furnace and a ramping program started as per the directions outlined in laboratory analytical procedures manual [21]. The ramping program gradually elevates the temperature up to 575 C and maintains it for 180 min, burning the biomass in the crucible. The crucible is transferred into a desiccator and allowed to cool. The crucible is weighed and put back into the muffle furnace until a constant weight of 0.3 mg is achieved. The ash is weighed to the nearest 0.1 mg, ash content is determined using Equation (1). A total of 24 samples were tested for ash content determination. % Ash content ¼
Weightcrucible and ash Weightcrucible 100 Oven dry weightsample
3.
Results and discussion
3.1.
Moisture content
[1]
A summary of MC observed in the slash piles is presented in Table 1. The average MC ranged from 12.4% to 26%. The average MC observed in this study is lower than the values observed in the Scandinavian countries [8,15e17]. The lower MC obtained in this study may be attributed to the continental climate in Northern Ontario [22]. Hot summers recorded in the study area (Fig. 1) lead to wood biomass feedstock drier than those presented in studies of Scandinavian countries. The climate is moderate in the Scandinavian countries due to maritime influences and does not reach similar extremes as in
46
b i o m a s s a n d b i o e n e r g y 4 2 ( 2 0 1 2 ) 4 3 e5 0
Table 1 e Average moisture content (% green basis) from top 1.5 m of roadside slash piles. Storage years 1
Shape
Windrow
Windrow Beehive
3
Pile height <2 m (SD)
Beehive 2
Species
Windrow Beehive
Softwood Hardwood Softwood Hardwood Softwood Hardwood Softwood Hardwood Softwood Hardwood Softwood Hardwood
17.3 (2.50) 21.1 (3.64) 21.8 (4.10) 26.0 (7.21) 19.1 (2.23) 14.2 (2.67) 12.4 (1.23) 21.2 (3.26) 15.4 (0.77) 17.9 (2.82) 14.2 (1.97) 22.7 (12.54)
>2 m (SD) 16.9 22.9 18.6 24.5 17.7 18.1 14.5 23.6 15.8 18.4 15.9 24.0
(3.31) (7.24) (1.54) (1.96) (2.65) (4.08) (3.86) (3.73) (6.33) (4.93) (3.99) (4.87)
Northwestern Ontario. The significant treatments and the interaction effects from the analysis of variance are presented in Fig. 2. The results suggests that the average MC of logging residue drops significantly ( p < 0.001) from the first drying season (21.3%) to the second drying season (17.8%), however, there is no significant change between second to third drying season (18.2%). The results from ANOVA are presented with a focus on significant treatments and interaction effects. Once cut, green wood starts to lose moisture quickly at first as it loses free water and then at a slower rate for bound water [23]. The MC gradually approaches an equilibrium state that fluctuates with temperature and relative humidity [24]. It is the hygroscopic nature of cell walls that allows the equilibrium MC to fluctuate in this manner. Our results are similar to the trends found in other studies on pulpwood drying rates [25e27]. The hygroscopicity of the cell wall is due to the presence of hydroxyl groups [28,29]. In the amorphous regions, there is irregularity in the molecular arrangement leaving accessible hydroxyl groups for bonding with water.
The cell is laid down in the presence of water and upon removal of water from cell walls, many of the hydroxyl groups form new hydrogen bonds among themselves that may not be broken when remoistened. Thus fewer sites are available for bonding with water. Esteban [28] has defined this phenomenon of loss of hygroscopic response as hygroscopic aging. The analysis of variance results also reveal that the mean MC value of logging residue of softwood species in slash piles was significantly lower ( p < 0.001) than hardwood species (Fig. 2). The lower MC values displayed by softwood species is best explained by the differences in the chemical composition and anatomical construction between hardwood and softwood species. In hardwoods, hemicelluloses constitute approximately 25e40% as opposed to 20e30% in softwoods [24]. Hemicellulose is the most hygroscopic component of cell wall followed by cellulose and lignin [30]. Consequently, the cell walls of hardwoods will have more potential bonding sites available for water than softwoods. Also, the majority of pits in softwood contain tori that aspirate in the earlywood of sapwood once MC is below FSP impeding transportation [24]. Such pit aspiration does not occur in hardwood sapwood due to the lack of tori in their pits thus hardwoods can be more hygroscopic. An added factor that may have contributed to the lower MC values in softwoods is transpiration drying [31]. The stomata of the leaves and needles are the main pathways for evaporation of moisture from a living tree [32]. Once cut, the stomatas close and transpiration is reduced significantly, especially in conifers [33]. However, it was observed that needles were intact in softwoods for longer durations than leaves on hardwoods. A similar observation was made in references [34,35]; this would allow softwoods a longer time frame for transpiration drying thus a greater overall reduction in moisture. The MC did not vary significantly ( p < 0.001) with slash pile height. However, since the measurements of MC were only performed to a depth of 1.5 m, no conclusions on the material below this depth can be made. In the literature, references
Fig. 2 e Average moisture content (% green basis) of logging residue from roadside slash piles for different storage years, pile shape and species. Bars with no common letters are significantly different from each other. The error bars show the standard deviation.
47
b i o m a s s a n d b i o e n e r g y 4 2 ( 2 0 1 2 ) 4 3 e5 0
[7,8] report that smaller piles lose moisture rapidly when the vapor pressure deficit of the ambient air is high in the summer. However, the reports also state that when the temperature and relative humidity drop, smaller piles regain moisture more rapidly. There was a significant effect ( p < 0.001) of pile shape on MC of logging residue in slash piles. In addition, the interaction between storage year, shape and species (Fig. 2) also significantly ( p < 0.01) affects the MC. For hardwood species, beehives display higher MC than windrows in all storage years. Similar trends are not prevalent in slash piles of softwood species. Windrow piles show lower MC than beehive in storage year 1, whereas beehive piles show lower MC than windrows in years 2 and 3. The lower moisture values displayed by windrows can be attributed to the greater surface area to volume ratio compared to beehive piles. The widths of windrow piles are much smaller compared to beehive piles and smaller width translates to less resistant to airflow, leading to faster drying of the piles. In a study conducted by Hall and Rudolph [26] on jack pine pulpwood piles, the MC fluctuation was much higher in the exposed wood than the inside wood. The significant effect on MC due to interaction between storage year, shape and species is most likely due to the level of compaction in the piles. Hardwood piles had a greater amount of void space compared to softwood piles. The more pronounced branching of hardwoods allows less compaction and hence there is a higher proportion of void space in hardwood piles which leads to lower airflow resistance. Therefore, in hardwood piles, the larger beehive piles show a higher MC than windrow piles. In softwood piles, the reduced void space leads to higher airflow resistance especially in the beehive piles. Consequently, the majority of logging residue in softwood beehive piles is protected from interface with the ambient air and it takes longer for logging residue in these piles to reach equilibrium MC, but once it is achieved, the fluctuations in MC occur at a much slower rate compared to windrow piles.
3.2.
Gross calorific value
A summary of mean gross calorific values and corresponding standard deviations of the samples are presented in Table 2. The values ranged from 19.9 to 23.1 MJ kg1. The highest average gross calorific value was displayed by large diameter softwood species from inside of the pile (1.5 m) in drying season 3 and the lowest value was displayed by smaller diameter hardwood species from the surface of the piles in drying season 2. The results of analysis of variance to test the null hypothesis that logging residue diameter, species, location in pile, and storage years do not have any effect on the gross calorific value found that there is a significant difference ( p < 0.05) in calorific value between species. In addition, the gross calorific values varied significantly ( p < 0.05) for interaction between species, location in pile, and storage years; the interaction effect is depicted in Fig. 3. Softwood species displayed slightly higher gross calorific value than hardwood species with one exception in year 3, where hardwood species at the surface of the slash piles showed higher average gross calorific value than softwood species but the difference was not significantly different. The higher gross calorific value per
Table 2 e Summary of gross calorific values in MJ kgL1 of logging residues in roadside slash piles. Storage years
Locationa
2
Inside pile
Diameterb Small (SD)
Surface 3
Species
Inside pile Surface
Softwood Hardwood Softwood Hardwood Softwood Hardwood Softwood Hardwood
21.1 (0.4) 20.3 (0.6) 21.2 (1.0) 20.4 (0.5) 20.8 (1.0) 20.0 (0.0) 19.9 (0.2) 20.7 (1.0)
Large (SD) 22.4 21.7 22.3 20.9 23.1 21.4 21.7 22.5
(0.5) (0.8) (0.5) (0.9) (0.2) (0.4) (1.2) (1.1)
a Inside pile is logging residue samples from depth of 1.5 m. b Small ¼ 4 cm diameter, Large ¼ >4 cm diameter.
kg displayed by softwoods can be attributed to the higher percentage of lignin and resin present in softwoods compared to hardwoods [13,36]. Lignin and resins have considerably higher gross calorific values than cellulose and hemicellulose [37]. Furthermore, the materials studied being branches, will contain a higher percentage of reaction wood [12]. In softwoods, reaction wood contains a higher percentage of lignin than normal wood, meanwhile the opposite is true for hardwoods [14,36]. Branch size had a significant effect ( p < 0.001) on the mean gross calorific value in roadside slash piles. Smaller diameter branches displayed higher gross calorific values than larger diameter branches in both windrows and beehives slash piles. Similar results were observed in reference [38] in samples collected in Manitoba. In softwood species, the higher gross calorific values in smaller diameter samples can once again be attributed to the presence of compression wood and also a higher percentage of bark. The percentage of bark increases sharply as the branch diameter decreases [13,39]. Bark generally has a higher gross calorific value than stem wood because it is richer in lignin, resin, terpenes and other combustable elements [13]. In hardwoods, although smaller diameter branches contain lower amount of lignin, due to the presence of tension wood, the lignin content in bark is three to four times more compared to softwood bark [12]. The higher percentage of lignin in bark coupled with the fact that there is a greater percentage of bark in smaller diameter branches leads to their higher gross calorific values. The storage duration did not significantly affect the mean calorific values of logging residue in slash piles. Hakkila [13] also states that although the gross calorific value of logging residue changes during storage, the change is insignificant. This can be attributed to low MC, which limits microbial activities and thus decay of wood. The calorific value of logging residue in slash piles does not vary significantly unless the wood is decayed. Decay in wood can be initiated only at MC over 20e25% [40].
3.3.
Ash content
A summary of ash content and corresponding standard deviation values of logging residue from roadside slash piles
48
b i o m a s s a n d b i o e n e r g y 4 2 ( 2 0 1 2 ) 4 3 e5 0
Fig. 3 e Gross calorific values of logging residue from roadside slash piles for different storage years, location in piles and species. “Inside” represent samples from depth of 1.5 m and surface represents samples from the surface of the piles. Bars with no common letters are significantly different from each other. The error bars show the standard deviation.
are presented in Table 3. The values range from 2.7% to 8.4%. The highest average ash content was found in smaller diameter hardwood species in drying season 1 and the lowest value was found in large diameter hardwood species in drying season 1. The results of analysis of variance to test the null hypothesis that logging residue diameter, species, and storage years do not have any effect on the ash content showed that branch size had a significant effect ( p < 0.05) on the mean ash content. Large diameter branches show lower ash content (3.9%) than small diameter branches (6.5%). In addition, there was an interaction effect ( p < 0.05) between branch size and species and is depicted in Fig. 4. In large diameter branches, softwoods show higher ash content (4.2%) but in smaller diameter branches hardwoods show higher ash content (8%) than softwoods. Majority of ash in a tree is concentrated in the bark tissues because of its importance to physiological functions [14]. As discussed earlier, smaller diameter branches have a much higher proportion of bark compared to larger branches and stems [13]. The ash content in softwood bark is reported to be approximately 2% whereas in hardwoods it averages 5%.
Therefore, the proportion of bark in logging residue is closely related to the ash content. This also explains the large difference in the ash content between softwood and hardwood in small branches compared to larger stems. The ash content did not vary significantly with the storage duration. This is most likely because the majority of logging residue was protected from weather factors when stored as slash piles. If fully exposed to environmental factors, there is a greater impact of leaching of elements by rainfall [41,42]. Although not significant, ash content of slash piles from drying season 2 show higher values (5.8%) than samples from drying season 1 (4.6%). Petterson and Nordfjell [8] also report on trials, mainly on samples from the Scandinavian countries, non-significant higher values of ash content for longer drying durations.
Table 3 e Summary of average ash content (%) of logging residue samples from roadside slash piles. Storage years 1 2
Diametera
Species
Large (SD) Softwood Hardwood Softwood Hardwood
Small ¼ 4 cm in diameter. a Large ¼ >4 cm in diameter.
3.4 2.7 4.9 4.6
(0.29) (0.88) (0.63) (0.96)
Small (SD) 3.8 (0.20) 8.4 (0.22) 6.1 (0.59) 7.6 (0.16)
Fig. 4 e Average ash content of logging residue samples from roadside slash piles for different storage years and sample diameter. “Large” represent stems with diameter larger than 4 cm and “small” represents diameter equal to or less than 4 cm. Bars with no common letters are significantly different from each other. The error bars show the standard deviation.
b i o m a s s a n d b i o e n e r g y 4 2 ( 2 0 1 2 ) 4 3 e5 0
4.
Conclusion
The study found that the MC continues to drop after the first year of drying onto the second year. The material did not show further reduction in moisture into the third year. Hardwood stored in windrow piles showed lower MC than beehive piles, however such generalization cannot be made for softwood piles due to higher airflow resistance in softwood piles. Softwood stored in windrow formation demonstrated lower MC in the initial year but as time progressed, beehive piles demonstrated better results. Based on the study, it can be concluded that beehive piles show lower MC for softwood materials stored for lengthy period; however for short term storage, i.e. less than a year, windrow piles demonstrate faster drying. Logging residue quality in terms of gross calorific value and ash content did not change significantly during the entire storage period. Softwood species displayed higher gross calorific values on per unit weight basis than hardwood species; it must be noted that wood density also needs to be taken into account to determine the species that are more suitable for bioenergy production. The study also confirmed that small diameter branches have higher calorific value and ash content than larger diameter stems; logging residue being mainly composed of smaller diameter branches, the higher amount of ash content needs to be dealt with appropriately after combustion. It is clear that managing the storage of logging residue is much more complicated than fossil fuel such as coal. Logging residue being a low value material, a slight change in any quality parameter can make the feedstock uneconomical. However, this study showed that logging residue can be stored in-field to improve quality, consequently increase net energy yield, and reduce cost of transportation. References [43e45] have shown that further improvement in quality can be observed through the use of cover; effectiveness of utilization of cover should be researched in the northwestern Ontario context. Ultimately, forest managers must apply a system to keep track of all piles present in harvest blocks and appropriately plan the procurement activities to ensure economic efficiency in utilizing logging residue.
Acknowledgements The authors acknowledge the Ontario Centres of Excellence, Ontario Ministry of Energy and Infrastructure, Thunder Bay Economic Development Commission and AbitibiBowater Inc. for financial support. The authors also acknowledge the valuable assistance from AbitibiBowater during data collection.
references
[1] Ride K. An investigation of wood fibre recovery and related economics of four harvesting systems common to northwestern Ontario [M.Sc.F. thesis]. Thunder Bay (ON): Lakehead University; 1998. [2] Bradley D. Canada e sustainable forest biomass supply chains. Ottawa, ON: Climate Change Solutions; 2007 October [31 pp. Report for IEA Task 40].
49
[3] Luke A, Scarlett D, Archibald D. Roadside slash pile burning in Northeastern Ontario. Timmins, ON: Northeast Science and Technology; 1993 March [8 pp. TN002]. [4] Schlamadinger B, Spitzer J, Kohlmaier G, Ludeke M. Carbon balance of bioenergy from logging residues. Biomass Bioenergy 1995;8(4):221e34. [5] Eriksson E. The Potential for Forestry to Reduce Net CO2 Emissions [Dissertation], Sweden: Swedish University of Agricultural Sciences; 2006. [6] Gautam S, Pulkki R, Shahi C, Leitch M. Economic and energy efficiency of salvaging biomass from wildfire burnt areas of bioenergy production in northwestern Ontario: a case study. Biomass Bioenergy 2010;34(11):1562e72. [7] Jrijis R. Storage and drying of wood fuel. Biomass Bioenergy 1995;9(1e5):181e90. [8] Pettersson M, Nordfjell T. Fuel quality changes during seasonal storage of compacted logging residues and young tree. Biomass Bioenergy 2007;31(11e12):782e92. [9] McKendry P. Energy production from biomass (part 1): overview of biomass. Bioresour Technol 2002;83(1):37e46. [10] Kryla J. Determination of available heat of combustion data for Canadian woody species. Ottawa: Canadian Forest Service; 1984 [89 pp. ENFOR Project P-256]. [11] Tomsons E, Tuherm H, Dolacis J, Hrols J. Dependence of the wood heat value on moisture content. Ann Warsaw Agric Univ e SGGW-AR, For Wood Technol 2003;53:355e61. [12] Nurmi J. Heating values of the above ground biomass of small-sized trees. Acta Forestalia Fennica 1993;236:30. [13] Hakkila P. Utilization of residual forest biomass. Berlin, Heidelberg, New York: Springer; 1989. [14] Bowyer J, Shmulsky R, Haygreen J. Forest products and wood science: an introduction. Wiley Blackwell Publishing; 2003. [15] Nurmi J. The storage of logging residue for fuel. Biomass Bioenergy 1999;17(1):41e7. [16] Lehtikangas P. Quality properties of pelletised sawdust, logging residues and bark. Biomass Bioenergy 2001;20(5): 351e60. [17] Jirjis R. Effects of particle size and pile height on storage and fuel quality of comminuted Salix vimanalis. Biomass Bioenergy 2005;28(2):193e201. [18] National Climate Data and Information Archive [Internet]. Fredericton, NB: Daily Data Report for Atikokan (AUT). c2006 e c2009. [cited 2009 Aug]. Available from: http://www. climate.weatheroffice.ec.gc.ca/climateData/canada_e.html/. [19] Pulkki R. Minimizing negative environmental impacts of forest harvesting operations. In: Adamowicz W, Burton P, editors. Towards sustainable management of the boreal forest. Ottawa: National Research Council of Canada, NRC Research Press; 2003. p. 581e628. [20] Hardy C. Guidelines for estimating volume, biomass, and smoke production for piled slash. Portland, OR: US. Department of Agriculture, Forest Service, Pacific Northwest Research Station; 1996 [PNW-GTR-364]. [21] Sluiter A, Hames B, Ruiz R, Scarlata C, Sluiter J, Templeton D. Determination of ash in biomass: laboratory analytical procedure (LAP). Golden, Co: National Renewable Energy Laboratory; 2008 January [8 pp. NREL/TP-510-42622]. [22] Flannigan M, Weber M. Influences of climate on Ontario Forests. In: Parera A, Euler D, Thompson I, editors. Ecology of a managed terrestrial landscape: patterns and processes of forest landscapes in Ontario. UBC Press; 2000. p. 103e15. [23] Shelton J, Shapiro A. The wood encyclopedia, an information source of theory, practice and equipment relating to wood as energy. Vermont: Vermont Crossroads Press, Inc; 1977. [24] Siau J. Wood: influence of moisture on physical properties. Blacksburg, VA: Dept. Of Wood Sci.. and Forest Products, Viriginia Polytechnic Institute and State University, ISBN 09622181-0-3; 1995.
50
b i o m a s s a n d b i o e n e r g y 4 2 ( 2 0 1 2 ) 4 3 e5 0
[25] Millet R. Drying rate of pulpwood in the woods and in the open. Pulp Paper Mag Can; 1953 November:168e70. [26] Hall O, Rudolph T. Weight loss of stored jack pine pulpwood. St. Paul, MN: University of Minnesota, School of Forestry; 1957 January [2 pp. Minnesota Forestry Notes. No. 57]. [27] Truman A. Variation of moisture content of pulpwood piled in the bush. Pulp Paper Mag Can; 1959 November:15e23. [28] Esteban L, Gril J, Palacios De Palacios P, Casasus A. Reduction of wood hygroscopicity and associated dimensional response by repeated humidity cycles. Ann for Sci 2005;62:275e84. [29] Walker J. Primary wood processing principles and practice. 2nd ed. Dordrecht, Netherlands: Springer; 2006. [30] Christensen G, Kelsey K. The sorption of water vapour by the constituents of wood. Holz Roh Werkst 1959;17(5):189e202. [31] Angus-Hankin C, Stokes B, Twaddle A. The transportation of fuelwood from forest to facility. Biomass Bioenergy 1995; 9(1e5):191e203. [32] Raven P, Evert R, Eichhorn S. Biology of plants. 6th ed. New York: W.H. Freeman and Company Worth Publishers; 1999. [33] Perem E. Transpiration seasoning of pulpwood: a survey of the literature. Ottawa, ON: Forest Products Laboratory, Canada Department of Forestry and Rural Development; 1968. [34] Rogers K. Preharvest drying of logging residues. For Prod J 1981;31(12):32e6. [35] Simola P, Makela M. Rasiinkaato kokopuiden korjuussa (Leafseasoning method in whole-tree logging). Fol For; 1976:273. [36] Plomion C, Leprovost G, Stokes A. Wood formation in trees. Plant Physiol 2001;127(4):1513e23.
[37] Hermann W, Bosshard P, Hung E, Hunt R, Simon AJ. An assessment of biomass feedstock and conversion research opportunities. Stanford, CA: Global Climate and Energy Project, Standford University; 2005 spring [Technical Assessment Report]. [38] Singh T, Kostecky M. Calorific value variation in components of 10 Canadian tree species. Can J For Res 1986;16:1378e81. [39] Wellwood R. Complete-tree utilization: an analysis of the literature (1970e1978). Ottawa, ON: Canadian Forest Service; 1979 [ENFOR Project P-15]. [40] Hudson H. Fungal biology. Great Britain: Cambridge University Press; 1992. [41] Jenkins B, Baxter L, Miles Jr T, Miles T. Combustion properties of biomass. Fuel Process Technol 1998;54:17e46. [42] Vamvuka D, Zografos D, Alevizos G. Control methods for mitigating biomass ash-related problems in fulidized beds. Bioresour Technol 2008;99(9):3534e44. [43] Jirjis R, Lehtikangas P. Fuel quality and dry matter loss during storage of logging residues in windrows. Uppsala, Sweden: Swedish University of Agricultural Sciences. Department of Forest Products; 1983 [Rapport nr. 236]. [44] Jirjis R, Gardenas S, Hedman G. Storage in covered windrows e logging residues from softwood. Uppsala, Sweden: Swedish University of Agricultural Sciences, Department of Forest Products; 1989 [Uppsats no 167]. [45] Lehtikangas P, Jirjis R. Logging residues in covered windrows influence of precipitation on fuel quality. Uppsala, Sweden: Swedish University of Agricultural Sciences, Department of Forest Products; 1995 [Res notes no 173].
Available online at www.sciencedirect.com
http://www.elsevier.com/locate/biombioe
Fuel quality changes in full tree logging residue during storage in roadside slash piles in Northwestern Ontario Shuva Gautam*, Reino Pulkki, Chander Shahi, Mathew Leitch Faculty of Natural Resources Management, Lakehead University, 955 Oliver Road, Thunder Bay, Ontario, Canada P7B 5E1
article info
abstract
Article history:
The procurement of logging residue for energy production can be uneconomical due to
Received 29 June 2010
high moisture content and low gross calorific value. High moisture content in biomass
Received in revised form
lowers the energy density and transportation becomes less efficient. Logging residue, from
24 January 2012
full tree logging operations stored in roadside beehive and windrow piles, were examined
Accepted 20 March 2012
to determine the effect of storage method and duration on fuel quality in northwestern
Available online 18 April 2012
Ontario, Canada. The fuel attributes assessed were moisture content, gross calorific value and ash content. Windrows displayed lower moisture content than beehives. Softwoods
Keywords:
generally displayed lower moisture content and higher gross calorific values than hard-
Ash content
woods. Smaller diameter stems displayed higher gross calorific value and ash content than
Bioenergy
larger diameter stems. The moisture content (green basis) reduced from a green state to
Fuel quality
15.1% after two years of storage in roadside slash piles, whereas gross calorific value and
Gross calorific value
ash content did not change significantly with storage time. The gross calorific values
Moisture content
ranged from 19.5 to 23.1 MJ kg1 and the ash content ranged from 0.4% to 8.4% for all
Northwestern Ontario
species, components and storage years. The study demonstrates that the storage regime plays a significant role on the fuel quality of logging residue. Proper storage and drying techniques improve the fuel quality and net energy yield from logging residue biomass, thereby leading to an overall cost reduction of the biomass feedstock. ª 2012 Elsevier Ltd. All rights reserved.
1.
Introduction
In northwestern Ontario, approximately 13%e14% of volume harvested using the full tree to roadside, roundwood to mill system is left on the harvest site as logging residue in the form of slash piles [1,2]. The present options for slash piles are either: i) to leave these on site to decay where they pose a potential fire hazard and occupy valuable space where seedlings could be established [3], or ii) to dispose these slash piles through burning [2,3]. A third option is to use the slash piles for energy production in the light of global warming issues [4,5] and the current forest industry crisis that is partly a result of high energy costs. However, the economic viability
of procuring such low bulk density logging residue, already low in energy density, depends on its moisture content (MC) [7,8]. The fuel quality in bioenergy production is assessed based on the properties that affect the energy yield and costs. MC, gross calorific value and ash content of the logging residue are three properties used to assess the fuel quality in slash piles, as each of these properties determine the viability of biomass procurement for bioenergy production [9]. The gross calorific value of a fuel is a measure of the maximum amount of energy that is released during burning of a given quantity of the fuel [10]. However the net energy released will depend on its MC [6,11,12] and ash content [9]. Furthermore, the overall cost can
* Corresponding author. E-mail address: [email protected] (S. Gautam). 0961-9534/$ e see front matter ª 2012 Elsevier Ltd. All rights reserved. doi:10.1016/j.biombioe.2012.03.015
44
b i o m a s s a n d b i o e n e r g y 4 2 ( 2 0 1 2 ) 4 3 e5 0
160.0 140.0 120.0 100.0 80.0 60.0 40.0
-40.0
Materials and methods
2.1.
Study area and logging description
The study materials were located in harvest compartments in the eastern part of the Crossroute Forest, to the west of Atikokan, Ontario, Canada. A harvest compartment is a unit of land allocated for harvest under the same logging method and system during an annual work schedule (AWS) and is also prescribed a regeneration method in its totality. The AWS outlines all forest management activities, including harvesting that are scheduled for a one-year period extending from 1st April to 31st March of the following year. The mean monthly temperature and precipitation recorded at the Atikokan weather station (Lat. 48 45.6670 N, Long. 91 37.6830 W) during the storage period are displayed in Fig. 1. Average daily maximum temperatures for June, July and August are 21.7 C, 24.7 C and 22.8 C [18]. Average monthly precipitation for June, July and August are 103.3 mm, 97.9 mm and 97.8 mm, and the average annual precipitation is 739.6 mm [18]. The compartments were harvested using a conventional mechanized full tree system [19], utilizing feller bunchers, grapple skidders, stroke delimbers and slashers. As a result the logging residue (branches and tops) were left in roadside
Jul
May
Jan
2008
Mar
Nov
Jul
Sept
May
Jan
Mar
Nov
JUL
Sept
May
Jan
Nov
Jul
Sept
Mar
200 07
2006
MeanTemp(°°C)
2009
TotalPrecip(mm)
Fig. 1 e Mean monthly temperature and total monthly precipitation at the Atikokan weather station. Source: Environment Canada [18].
piles. A total of 36 piles were excavated for the purpose of the study. The shape of the slash piles studied fall either under the general category “half-section of sphere” or “half-ellipsoid” referred to as beehive and windrow, respectively. General shapes of piled woody debris have been outlined in reference [20]. Tree species were divided into the categories of softwoods and hardwoods. The softwoods included balsam fir (Abies balsamea (L.) Miller), white spruce (Picea glauca (Moench) Voss), black spruce (Picea marianana (Miller) B.S.P.) and jack pine (Pinus banksiana Lamb.); and the hardwoods included red maple (Acer rubrum L.), white birch (Betula papyrifera Marshall) and trembling aspen (Populus tremuloides Michx.). The heights of the piles were recorded as being either greater or less than 2 m; the width ranged from 8 m to 16 m. Storage years included logging residue stored for 1, 2 and 3 drying (summer) seasons.
2.2.
2.
May
0.0 -20.0
Jan
20.0 Mar
rise due to extra transport, dumping and handling costs to deal with ash [9,13]. Therefore, information on gross calorific value alone is not sufficient. Studies in Europe have shown that logging residue can be stored in-field to reduce MC and improve the fuel quality [8,15e17]. In the study conducted by Nurmi [15], Norway spruce (Picea abies (L.) Karst.) logging residue in Finland was piled 4.5 m high. It was found that the average MC reduced from 55% (green basis) to approximately 40% after being piled from September 1994 to September 1995. Another treatment of the same study that involved leaving the residue on the cutover from September 1994 to June 1995, then forwarding to the landing until September 1995 resulted in a lower MC of 28%. The form, duration of storage, and the weather conditions affect fuel quality [8]. However, no comparable published research could be found for Canadian continental climate and boreal forest conditions. Knowledge gained from European studies could be applied to northwestern Ontario to a certain extent, but due to differences in logging systems, stored material and weather conditions, this may not be completely valid. This is especially true in areas of continental climate conditions, where there is relatively low summer rainfall and high daily temperatures. The purpose of this study was to determine a storage regime for boreal forest conditions, with a dry continental climate, that enhances the fuel quality of full tree logging residue stored at roadside. The specific objectives were to determine the differences in fuel quality attributes (MC, gross calorific value and ash content) for: (i) storage years (three drying seasons), (ii) pile shapes (beehive and windrow), (iii) pile heights (more than and less than 2 m), (iv) species (softwood and hardwoods), and (v) logging residue’ branch sizes (small with diameter less than or equal to 4 cm and large with diameter greater than 4 cm).
Experimental design and statistical methods
Analysis of variance tests were carried out using the General Linear Model in SPSS to test the null hypotheses of no significant difference in MC, gross calorific value, and ash content for different storage years, pile shapes, pile heights, species, and logging residue diameters. The experimental designs for all the models are full factorial design, with MC, gross calorific value, and ash content as the dependent variables, respectively for each model and storage years, pile shapes, pile heights, species, and logging residue diameters as the independent variables. For gross calorific value and ash content analysis of variance, only two storage years were tested due to financial constraint. Storage years 1 and 2 were tested for ash content but for gross calorific value years 2 and 3 were tested because shorter term studies already exist [8,13,17] with longer term studies lacking. The data for each model was tested for normality and homogeneity of variance before conducting the analysis of variance. Any significant differences in the analysis were analyzed using post-hoc tests.
2.3.
Field sampling
A list of all harvest compartments in the area west of Atikokan for the annual work schedule years of 2006e07, 2007e08, and
b i o m a s s a n d b i o e n e r g y 4 2 ( 2 0 1 2 ) 4 3 e5 0
2008e09 containing slash piles was obtained from AbitibiBowater Inc., Fort Frances, Ontario. The slash piles in the compartments were categorized into storage years, pile shapes (beehive and windrow), and pile heights (more than and less than 2 m). MC data was collected during the summers of 2008 and 2009. The collection dates extended from 17th July to 30th July in the summer of 2008, and from 18th May to 3rd June in the summer of 2009. MC, gross calorific value and ash content were measured at the surface, as well as inside the slash piles. Ribbons were laid in an east to west orientation over the beehive type slash piles, and along the length of the windrow type slash piles at a randomly selected position. All biomass material that intersected the ribbon at the surface of the pile was cut with a chainsaw at the point of the intersection to expose the complete diameter of the biomass sample. A Protimeter Surveymaster moisture meter was used to measure the MC at various points on the cross section of the cut portion; the MC and species, as well as the diameter of the cross section were recorded. MC in this study is expressed in green weight basis which is the ratio of the weight of water in wood to the weight of green wood (sum of mass of water and mass of dry wood). This procedure was continued for all biomass intersecting the ribbon. Each slash pile was then dissected from the highest point of the pile vertically to a depth of 1.5 m for internal measurements, using a chainsaw. During the excavation process, at least one cross-sectional face was undisturbed and the excavated biomass materials were thrown to the side. A measurement tape was hung from the top of the pile to the depth of 1.5 m and a ribbon was used to divide the 1.5 m depth hole into 25 cm sections. At each 25 cm section, the MC and species were recorded for all exposed biomass throughout the entire depth of the 25 cm section; a total of 3359 MC readings were recorded. Wood residue samples from each section were collected and brought to the laboratory to determine their gross calorific values and ash contents. The species (softwood and hardwood) and logging residue diameters (small with diameter less than or equal to 4 cm and large with diameter greater than 4 cm) of each sample were recorded. Prior to using the moisture meter, an equality test was performed to test the reliability of the readings obtained from the moisture meter. The equality test involved collecting a number of samples of all species used in the study with varying MC values. The samples were measured for MC first using the moisture meter then by the oven-drying (laboratory) method. Paired t-tests were performed to determine if the MC values displayed by the moisture meter were significantly different from those determined through the oven-drying (laboratory) method. Two paired t-tests were performed: one for samples with MC below fibre saturation point (FSP) and the other for samples above the FSP as it is commonly known that electric moisture meters are much more accurate in measuring samples with MC below FSP. For samples below FSP, the MC means for moisture meter (17.4%) and laboratory (17.3%) measured samples were not significantly different (difference ¼ 0.20, SD ¼ 6.7, N ¼ 17, p ¼ 0.909). Also, for samples above FSP, the MC means for moisture meter (31%) and laboratory (29.2%) measured samples were not significantly different (difference ¼ 3.77, SD ¼ 14.6, p ¼ 0.392). In light of the
45
paired t-test results, the average values displayed by the moisture meter were used as estimates of MC. However, the MC values obtained from the moisture meter are only approximate values and should not be considered absolute as there are many variables that affect the readings displayed by the moisture meter.
2.4.
Laboratory methods
A Parr 6200 Bomb Calorimeter was used to determine the gross calorific values of the collected samples. The samples were dried to MC of 10e20%. After achieving the desired moisture level, the samples were ground using a large Wiley Mill followed by a mini Wiley mill. A bone dry sample of known mass (w1 g 0.5 g) in pellet form was placed in the bomb, which was then placed in the calorimeter. The calorimeter pressurizes the bomb with pure oxygen and is submerged under 2 L of distilled water of known temperature. The biomass in the bomb is ignited and the combustion raises the temperature of the bomb and consequently of the water. The calorimeter gives an output of heat value in MJ kg1 based on the water temperature change, taking into consideration the electrical input during ignition. A total of 48 samples were tested for gross calorific value determination. The ash content was determined according to procedures outlined in laboratory analytical procedures manual [21]. The wood biomass samples were ground following the same procedure used for the gross calorific value determination. Oven dry biomass between 0.5 and 2.0 g was weighed and placed in a crucible. Crucibles were labeled and dried thoroughly by placing them in a muffle furnace and a ramping program started as per the directions outlined in laboratory analytical procedures manual [21]. The ramping program gradually elevates the temperature up to 575 C and maintains it for 180 min, burning the biomass in the crucible. The crucible is transferred into a desiccator and allowed to cool. The crucible is weighed and put back into the muffle furnace until a constant weight of 0.3 mg is achieved. The ash is weighed to the nearest 0.1 mg, ash content is determined using Equation (1). A total of 24 samples were tested for ash content determination. % Ash content ¼
Weightcrucible and ash Weightcrucible 100 Oven dry weightsample
3.
Results and discussion
3.1.
Moisture content
[1]
A summary of MC observed in the slash piles is presented in Table 1. The average MC ranged from 12.4% to 26%. The average MC observed in this study is lower than the values observed in the Scandinavian countries [8,15e17]. The lower MC obtained in this study may be attributed to the continental climate in Northern Ontario [22]. Hot summers recorded in the study area (Fig. 1) lead to wood biomass feedstock drier than those presented in studies of Scandinavian countries. The climate is moderate in the Scandinavian countries due to maritime influences and does not reach similar extremes as in
46
b i o m a s s a n d b i o e n e r g y 4 2 ( 2 0 1 2 ) 4 3 e5 0
Table 1 e Average moisture content (% green basis) from top 1.5 m of roadside slash piles. Storage years 1
Shape
Windrow
Windrow Beehive
3
Pile height <2 m (SD)
Beehive 2
Species
Windrow Beehive
Softwood Hardwood Softwood Hardwood Softwood Hardwood Softwood Hardwood Softwood Hardwood Softwood Hardwood
17.3 (2.50) 21.1 (3.64) 21.8 (4.10) 26.0 (7.21) 19.1 (2.23) 14.2 (2.67) 12.4 (1.23) 21.2 (3.26) 15.4 (0.77) 17.9 (2.82) 14.2 (1.97) 22.7 (12.54)
>2 m (SD) 16.9 22.9 18.6 24.5 17.7 18.1 14.5 23.6 15.8 18.4 15.9 24.0
(3.31) (7.24) (1.54) (1.96) (2.65) (4.08) (3.86) (3.73) (6.33) (4.93) (3.99) (4.87)
Northwestern Ontario. The significant treatments and the interaction effects from the analysis of variance are presented in Fig. 2. The results suggests that the average MC of logging residue drops significantly ( p < 0.001) from the first drying season (21.3%) to the second drying season (17.8%), however, there is no significant change between second to third drying season (18.2%). The results from ANOVA are presented with a focus on significant treatments and interaction effects. Once cut, green wood starts to lose moisture quickly at first as it loses free water and then at a slower rate for bound water [23]. The MC gradually approaches an equilibrium state that fluctuates with temperature and relative humidity [24]. It is the hygroscopic nature of cell walls that allows the equilibrium MC to fluctuate in this manner. Our results are similar to the trends found in other studies on pulpwood drying rates [25e27]. The hygroscopicity of the cell wall is due to the presence of hydroxyl groups [28,29]. In the amorphous regions, there is irregularity in the molecular arrangement leaving accessible hydroxyl groups for bonding with water.
The cell is laid down in the presence of water and upon removal of water from cell walls, many of the hydroxyl groups form new hydrogen bonds among themselves that may not be broken when remoistened. Thus fewer sites are available for bonding with water. Esteban [28] has defined this phenomenon of loss of hygroscopic response as hygroscopic aging. The analysis of variance results also reveal that the mean MC value of logging residue of softwood species in slash piles was significantly lower ( p < 0.001) than hardwood species (Fig. 2). The lower MC values displayed by softwood species is best explained by the differences in the chemical composition and anatomical construction between hardwood and softwood species. In hardwoods, hemicelluloses constitute approximately 25e40% as opposed to 20e30% in softwoods [24]. Hemicellulose is the most hygroscopic component of cell wall followed by cellulose and lignin [30]. Consequently, the cell walls of hardwoods will have more potential bonding sites available for water than softwoods. Also, the majority of pits in softwood contain tori that aspirate in the earlywood of sapwood once MC is below FSP impeding transportation [24]. Such pit aspiration does not occur in hardwood sapwood due to the lack of tori in their pits thus hardwoods can be more hygroscopic. An added factor that may have contributed to the lower MC values in softwoods is transpiration drying [31]. The stomata of the leaves and needles are the main pathways for evaporation of moisture from a living tree [32]. Once cut, the stomatas close and transpiration is reduced significantly, especially in conifers [33]. However, it was observed that needles were intact in softwoods for longer durations than leaves on hardwoods. A similar observation was made in references [34,35]; this would allow softwoods a longer time frame for transpiration drying thus a greater overall reduction in moisture. The MC did not vary significantly ( p < 0.001) with slash pile height. However, since the measurements of MC were only performed to a depth of 1.5 m, no conclusions on the material below this depth can be made. In the literature, references
Fig. 2 e Average moisture content (% green basis) of logging residue from roadside slash piles for different storage years, pile shape and species. Bars with no common letters are significantly different from each other. The error bars show the standard deviation.
47
b i o m a s s a n d b i o e n e r g y 4 2 ( 2 0 1 2 ) 4 3 e5 0
[7,8] report that smaller piles lose moisture rapidly when the vapor pressure deficit of the ambient air is high in the summer. However, the reports also state that when the temperature and relative humidity drop, smaller piles regain moisture more rapidly. There was a significant effect ( p < 0.001) of pile shape on MC of logging residue in slash piles. In addition, the interaction between storage year, shape and species (Fig. 2) also significantly ( p < 0.01) affects the MC. For hardwood species, beehives display higher MC than windrows in all storage years. Similar trends are not prevalent in slash piles of softwood species. Windrow piles show lower MC than beehive in storage year 1, whereas beehive piles show lower MC than windrows in years 2 and 3. The lower moisture values displayed by windrows can be attributed to the greater surface area to volume ratio compared to beehive piles. The widths of windrow piles are much smaller compared to beehive piles and smaller width translates to less resistant to airflow, leading to faster drying of the piles. In a study conducted by Hall and Rudolph [26] on jack pine pulpwood piles, the MC fluctuation was much higher in the exposed wood than the inside wood. The significant effect on MC due to interaction between storage year, shape and species is most likely due to the level of compaction in the piles. Hardwood piles had a greater amount of void space compared to softwood piles. The more pronounced branching of hardwoods allows less compaction and hence there is a higher proportion of void space in hardwood piles which leads to lower airflow resistance. Therefore, in hardwood piles, the larger beehive piles show a higher MC than windrow piles. In softwood piles, the reduced void space leads to higher airflow resistance especially in the beehive piles. Consequently, the majority of logging residue in softwood beehive piles is protected from interface with the ambient air and it takes longer for logging residue in these piles to reach equilibrium MC, but once it is achieved, the fluctuations in MC occur at a much slower rate compared to windrow piles.
3.2.
Gross calorific value
A summary of mean gross calorific values and corresponding standard deviations of the samples are presented in Table 2. The values ranged from 19.9 to 23.1 MJ kg1. The highest average gross calorific value was displayed by large diameter softwood species from inside of the pile (1.5 m) in drying season 3 and the lowest value was displayed by smaller diameter hardwood species from the surface of the piles in drying season 2. The results of analysis of variance to test the null hypothesis that logging residue diameter, species, location in pile, and storage years do not have any effect on the gross calorific value found that there is a significant difference ( p < 0.05) in calorific value between species. In addition, the gross calorific values varied significantly ( p < 0.05) for interaction between species, location in pile, and storage years; the interaction effect is depicted in Fig. 3. Softwood species displayed slightly higher gross calorific value than hardwood species with one exception in year 3, where hardwood species at the surface of the slash piles showed higher average gross calorific value than softwood species but the difference was not significantly different. The higher gross calorific value per
Table 2 e Summary of gross calorific values in MJ kgL1 of logging residues in roadside slash piles. Storage years
Locationa
2
Inside pile
Diameterb Small (SD)
Surface 3
Species
Inside pile Surface
Softwood Hardwood Softwood Hardwood Softwood Hardwood Softwood Hardwood
21.1 (0.4) 20.3 (0.6) 21.2 (1.0) 20.4 (0.5) 20.8 (1.0) 20.0 (0.0) 19.9 (0.2) 20.7 (1.0)
Large (SD) 22.4 21.7 22.3 20.9 23.1 21.4 21.7 22.5
(0.5) (0.8) (0.5) (0.9) (0.2) (0.4) (1.2) (1.1)
a Inside pile is logging residue samples from depth of 1.5 m. b Small ¼ 4 cm diameter, Large ¼ >4 cm diameter.
kg displayed by softwoods can be attributed to the higher percentage of lignin and resin present in softwoods compared to hardwoods [13,36]. Lignin and resins have considerably higher gross calorific values than cellulose and hemicellulose [37]. Furthermore, the materials studied being branches, will contain a higher percentage of reaction wood [12]. In softwoods, reaction wood contains a higher percentage of lignin than normal wood, meanwhile the opposite is true for hardwoods [14,36]. Branch size had a significant effect ( p < 0.001) on the mean gross calorific value in roadside slash piles. Smaller diameter branches displayed higher gross calorific values than larger diameter branches in both windrows and beehives slash piles. Similar results were observed in reference [38] in samples collected in Manitoba. In softwood species, the higher gross calorific values in smaller diameter samples can once again be attributed to the presence of compression wood and also a higher percentage of bark. The percentage of bark increases sharply as the branch diameter decreases [13,39]. Bark generally has a higher gross calorific value than stem wood because it is richer in lignin, resin, terpenes and other combustable elements [13]. In hardwoods, although smaller diameter branches contain lower amount of lignin, due to the presence of tension wood, the lignin content in bark is three to four times more compared to softwood bark [12]. The higher percentage of lignin in bark coupled with the fact that there is a greater percentage of bark in smaller diameter branches leads to their higher gross calorific values. The storage duration did not significantly affect the mean calorific values of logging residue in slash piles. Hakkila [13] also states that although the gross calorific value of logging residue changes during storage, the change is insignificant. This can be attributed to low MC, which limits microbial activities and thus decay of wood. The calorific value of logging residue in slash piles does not vary significantly unless the wood is decayed. Decay in wood can be initiated only at MC over 20e25% [40].
3.3.
Ash content
A summary of ash content and corresponding standard deviation values of logging residue from roadside slash piles
48
b i o m a s s a n d b i o e n e r g y 4 2 ( 2 0 1 2 ) 4 3 e5 0
Fig. 3 e Gross calorific values of logging residue from roadside slash piles for different storage years, location in piles and species. “Inside” represent samples from depth of 1.5 m and surface represents samples from the surface of the piles. Bars with no common letters are significantly different from each other. The error bars show the standard deviation.
are presented in Table 3. The values range from 2.7% to 8.4%. The highest average ash content was found in smaller diameter hardwood species in drying season 1 and the lowest value was found in large diameter hardwood species in drying season 1. The results of analysis of variance to test the null hypothesis that logging residue diameter, species, and storage years do not have any effect on the ash content showed that branch size had a significant effect ( p < 0.05) on the mean ash content. Large diameter branches show lower ash content (3.9%) than small diameter branches (6.5%). In addition, there was an interaction effect ( p < 0.05) between branch size and species and is depicted in Fig. 4. In large diameter branches, softwoods show higher ash content (4.2%) but in smaller diameter branches hardwoods show higher ash content (8%) than softwoods. Majority of ash in a tree is concentrated in the bark tissues because of its importance to physiological functions [14]. As discussed earlier, smaller diameter branches have a much higher proportion of bark compared to larger branches and stems [13]. The ash content in softwood bark is reported to be approximately 2% whereas in hardwoods it averages 5%.
Therefore, the proportion of bark in logging residue is closely related to the ash content. This also explains the large difference in the ash content between softwood and hardwood in small branches compared to larger stems. The ash content did not vary significantly with the storage duration. This is most likely because the majority of logging residue was protected from weather factors when stored as slash piles. If fully exposed to environmental factors, there is a greater impact of leaching of elements by rainfall [41,42]. Although not significant, ash content of slash piles from drying season 2 show higher values (5.8%) than samples from drying season 1 (4.6%). Petterson and Nordfjell [8] also report on trials, mainly on samples from the Scandinavian countries, non-significant higher values of ash content for longer drying durations.
Table 3 e Summary of average ash content (%) of logging residue samples from roadside slash piles. Storage years 1 2
Diametera
Species
Large (SD) Softwood Hardwood Softwood Hardwood
Small ¼ 4 cm in diameter. a Large ¼ >4 cm in diameter.
3.4 2.7 4.9 4.6
(0.29) (0.88) (0.63) (0.96)
Small (SD) 3.8 (0.20) 8.4 (0.22) 6.1 (0.59) 7.6 (0.16)
Fig. 4 e Average ash content of logging residue samples from roadside slash piles for different storage years and sample diameter. “Large” represent stems with diameter larger than 4 cm and “small” represents diameter equal to or less than 4 cm. Bars with no common letters are significantly different from each other. The error bars show the standard deviation.
b i o m a s s a n d b i o e n e r g y 4 2 ( 2 0 1 2 ) 4 3 e5 0
4.
Conclusion
The study found that the MC continues to drop after the first year of drying onto the second year. The material did not show further reduction in moisture into the third year. Hardwood stored in windrow piles showed lower MC than beehive piles, however such generalization cannot be made for softwood piles due to higher airflow resistance in softwood piles. Softwood stored in windrow formation demonstrated lower MC in the initial year but as time progressed, beehive piles demonstrated better results. Based on the study, it can be concluded that beehive piles show lower MC for softwood materials stored for lengthy period; however for short term storage, i.e. less than a year, windrow piles demonstrate faster drying. Logging residue quality in terms of gross calorific value and ash content did not change significantly during the entire storage period. Softwood species displayed higher gross calorific values on per unit weight basis than hardwood species; it must be noted that wood density also needs to be taken into account to determine the species that are more suitable for bioenergy production. The study also confirmed that small diameter branches have higher calorific value and ash content than larger diameter stems; logging residue being mainly composed of smaller diameter branches, the higher amount of ash content needs to be dealt with appropriately after combustion. It is clear that managing the storage of logging residue is much more complicated than fossil fuel such as coal. Logging residue being a low value material, a slight change in any quality parameter can make the feedstock uneconomical. However, this study showed that logging residue can be stored in-field to improve quality, consequently increase net energy yield, and reduce cost of transportation. References [43e45] have shown that further improvement in quality can be observed through the use of cover; effectiveness of utilization of cover should be researched in the northwestern Ontario context. Ultimately, forest managers must apply a system to keep track of all piles present in harvest blocks and appropriately plan the procurement activities to ensure economic efficiency in utilizing logging residue.
Acknowledgements The authors acknowledge the Ontario Centres of Excellence, Ontario Ministry of Energy and Infrastructure, Thunder Bay Economic Development Commission and AbitibiBowater Inc. for financial support. The authors also acknowledge the valuable assistance from AbitibiBowater during data collection.
references
[1] Ride K. An investigation of wood fibre recovery and related economics of four harvesting systems common to northwestern Ontario [M.Sc.F. thesis]. Thunder Bay (ON): Lakehead University; 1998. [2] Bradley D. Canada e sustainable forest biomass supply chains. Ottawa, ON: Climate Change Solutions; 2007 October [31 pp. Report for IEA Task 40].
49
[3] Luke A, Scarlett D, Archibald D. Roadside slash pile burning in Northeastern Ontario. Timmins, ON: Northeast Science and Technology; 1993 March [8 pp. TN002]. [4] Schlamadinger B, Spitzer J, Kohlmaier G, Ludeke M. Carbon balance of bioenergy from logging residues. Biomass Bioenergy 1995;8(4):221e34. [5] Eriksson E. The Potential for Forestry to Reduce Net CO2 Emissions [Dissertation], Sweden: Swedish University of Agricultural Sciences; 2006. [6] Gautam S, Pulkki R, Shahi C, Leitch M. Economic and energy efficiency of salvaging biomass from wildfire burnt areas of bioenergy production in northwestern Ontario: a case study. Biomass Bioenergy 2010;34(11):1562e72. [7] Jrijis R. Storage and drying of wood fuel. Biomass Bioenergy 1995;9(1e5):181e90. [8] Pettersson M, Nordfjell T. Fuel quality changes during seasonal storage of compacted logging residues and young tree. Biomass Bioenergy 2007;31(11e12):782e92. [9] McKendry P. Energy production from biomass (part 1): overview of biomass. Bioresour Technol 2002;83(1):37e46. [10] Kryla J. Determination of available heat of combustion data for Canadian woody species. Ottawa: Canadian Forest Service; 1984 [89 pp. ENFOR Project P-256]. [11] Tomsons E, Tuherm H, Dolacis J, Hrols J. Dependence of the wood heat value on moisture content. Ann Warsaw Agric Univ e SGGW-AR, For Wood Technol 2003;53:355e61. [12] Nurmi J. Heating values of the above ground biomass of small-sized trees. Acta Forestalia Fennica 1993;236:30. [13] Hakkila P. Utilization of residual forest biomass. Berlin, Heidelberg, New York: Springer; 1989. [14] Bowyer J, Shmulsky R, Haygreen J. Forest products and wood science: an introduction. Wiley Blackwell Publishing; 2003. [15] Nurmi J. The storage of logging residue for fuel. Biomass Bioenergy 1999;17(1):41e7. [16] Lehtikangas P. Quality properties of pelletised sawdust, logging residues and bark. Biomass Bioenergy 2001;20(5): 351e60. [17] Jirjis R. Effects of particle size and pile height on storage and fuel quality of comminuted Salix vimanalis. Biomass Bioenergy 2005;28(2):193e201. [18] National Climate Data and Information Archive [Internet]. Fredericton, NB: Daily Data Report for Atikokan (AUT). c2006 e c2009. [cited 2009 Aug]. Available from: http://www. climate.weatheroffice.ec.gc.ca/climateData/canada_e.html/. [19] Pulkki R. Minimizing negative environmental impacts of forest harvesting operations. In: Adamowicz W, Burton P, editors. Towards sustainable management of the boreal forest. Ottawa: National Research Council of Canada, NRC Research Press; 2003. p. 581e628. [20] Hardy C. Guidelines for estimating volume, biomass, and smoke production for piled slash. Portland, OR: US. Department of Agriculture, Forest Service, Pacific Northwest Research Station; 1996 [PNW-GTR-364]. [21] Sluiter A, Hames B, Ruiz R, Scarlata C, Sluiter J, Templeton D. Determination of ash in biomass: laboratory analytical procedure (LAP). Golden, Co: National Renewable Energy Laboratory; 2008 January [8 pp. NREL/TP-510-42622]. [22] Flannigan M, Weber M. Influences of climate on Ontario Forests. In: Parera A, Euler D, Thompson I, editors. Ecology of a managed terrestrial landscape: patterns and processes of forest landscapes in Ontario. UBC Press; 2000. p. 103e15. [23] Shelton J, Shapiro A. The wood encyclopedia, an information source of theory, practice and equipment relating to wood as energy. Vermont: Vermont Crossroads Press, Inc; 1977. [24] Siau J. Wood: influence of moisture on physical properties. Blacksburg, VA: Dept. Of Wood Sci.. and Forest Products, Viriginia Polytechnic Institute and State University, ISBN 09622181-0-3; 1995.
50
b i o m a s s a n d b i o e n e r g y 4 2 ( 2 0 1 2 ) 4 3 e5 0
[25] Millet R. Drying rate of pulpwood in the woods and in the open. Pulp Paper Mag Can; 1953 November:168e70. [26] Hall O, Rudolph T. Weight loss of stored jack pine pulpwood. St. Paul, MN: University of Minnesota, School of Forestry; 1957 January [2 pp. Minnesota Forestry Notes. No. 57]. [27] Truman A. Variation of moisture content of pulpwood piled in the bush. Pulp Paper Mag Can; 1959 November:15e23. [28] Esteban L, Gril J, Palacios De Palacios P, Casasus A. Reduction of wood hygroscopicity and associated dimensional response by repeated humidity cycles. Ann for Sci 2005;62:275e84. [29] Walker J. Primary wood processing principles and practice. 2nd ed. Dordrecht, Netherlands: Springer; 2006. [30] Christensen G, Kelsey K. The sorption of water vapour by the constituents of wood. Holz Roh Werkst 1959;17(5):189e202. [31] Angus-Hankin C, Stokes B, Twaddle A. The transportation of fuelwood from forest to facility. Biomass Bioenergy 1995; 9(1e5):191e203. [32] Raven P, Evert R, Eichhorn S. Biology of plants. 6th ed. New York: W.H. Freeman and Company Worth Publishers; 1999. [33] Perem E. Transpiration seasoning of pulpwood: a survey of the literature. Ottawa, ON: Forest Products Laboratory, Canada Department of Forestry and Rural Development; 1968. [34] Rogers K. Preharvest drying of logging residues. For Prod J 1981;31(12):32e6. [35] Simola P, Makela M. Rasiinkaato kokopuiden korjuussa (Leafseasoning method in whole-tree logging). Fol For; 1976:273. [36] Plomion C, Leprovost G, Stokes A. Wood formation in trees. Plant Physiol 2001;127(4):1513e23.
[37] Hermann W, Bosshard P, Hung E, Hunt R, Simon AJ. An assessment of biomass feedstock and conversion research opportunities. Stanford, CA: Global Climate and Energy Project, Standford University; 2005 spring [Technical Assessment Report]. [38] Singh T, Kostecky M. Calorific value variation in components of 10 Canadian tree species. Can J For Res 1986;16:1378e81. [39] Wellwood R. Complete-tree utilization: an analysis of the literature (1970e1978). Ottawa, ON: Canadian Forest Service; 1979 [ENFOR Project P-15]. [40] Hudson H. Fungal biology. Great Britain: Cambridge University Press; 1992. [41] Jenkins B, Baxter L, Miles Jr T, Miles T. Combustion properties of biomass. Fuel Process Technol 1998;54:17e46. [42] Vamvuka D, Zografos D, Alevizos G. Control methods for mitigating biomass ash-related problems in fulidized beds. Bioresour Technol 2008;99(9):3534e44. [43] Jirjis R, Lehtikangas P. Fuel quality and dry matter loss during storage of logging residues in windrows. Uppsala, Sweden: Swedish University of Agricultural Sciences. Department of Forest Products; 1983 [Rapport nr. 236]. [44] Jirjis R, Gardenas S, Hedman G. Storage in covered windrows e logging residues from softwood. Uppsala, Sweden: Swedish University of Agricultural Sciences, Department of Forest Products; 1989 [Uppsats no 167]. [45] Lehtikangas P, Jirjis R. Logging residues in covered windrows influence of precipitation on fuel quality. Uppsala, Sweden: Swedish University of Agricultural Sciences, Department of Forest Products; 1995 [Res notes no 173].