- Email: [email protected]
Compost convective airflow under passive aeration
Bioresource Technology 86 (2003) 259–266
Compost convective airflow under passive aeration Suzelle Barrington *, Denis Choiniere, Maher Trigui, William Knight Department of Agricultural and Biosystems Engineering, Macdonald Campus of McGill University, 21 111 Lakeshore, Ste Anne de Bellevue, Qu ebec, Canada H9X 3V9 Received 17 August 2001; received in revised form 24 June 2002; accepted 26 June 2002
Abstract For composting, passive aeration can save energy costs while being just as efficient as forced or active aeration. Passive aeration requires the proper design of aeration ducts, and thus, the proper prediction of the convective airflow rates created by the temperature differential between the compost and the ambient air. To establish such relationship, the temperature and convective air flow regimes of composts were investigated using three bulking agents (wood shavings, hay and straw), each at three moisture contents (MC––60%, 65% and 70%) spanning the normal values. All bulking agent and aeration treatments were aerated in duplicate under passive and active regimes. Laboratory vessels of 105 L were used for all treatments. Passive aeration treatments produced temperatures above 57 °C, as did the treatments actively aerated at 4 mg of air s1 kg1 of initial dry compost material. Compost MC had an effect only on the peak compost temperature, occurring between day 2 and 6. After 6 days of composting, MC no longer had any effect on temperature regime because of the loss of moisture by each mixture. A relationship was established between the Grasholf number (Gr––ratio of buoyancy to viscous forces) and the convective airflow rates, to size the aeration ducts for passive aeration. In general, convective airflow rates ranged from 1.5 to 0.7 mg of dry air s1 kg1 of initial compost dry matter, from day 0 to day 20, respectively, and for all compost treatments. This airflow rate sizes the aeration ducts installed under compost piles for passive aeration. As compared to straw where airflow rate dropped over a given level of Gr, wood shavings and hay were found to be more effective as bulking agents, as their airflow rate increased constantly with Gr. Ó 2002 Elsevier Science Ltd. All rights reserved. Keywords: Compost; Passive aeration; Convective airflow; Grasholf number
1. Introduction Composting is a controlled biological process used to stabilize and transform wastes into an aesthetic soil amendment. Aeration is one component of the controlling process as it insures the growth of adequate aerobic microbe populations and the development of stabilizing temperatures. Composts can be aerated by one of three methods: natural or static pile, passive and active or forced. Natural aeration is the cheapest and simplest, as it requires no installations. Natural aeration occurs simply by diffusion and convection, governed by the exposed surfaces and their respective properties (Fernandes et al., 1994). Passive aeration requires the installation of ducts
*
Corresponding author. Tel.: +1-514-398-7776; fax: +1-514-3988387. E-mail address: [email protected] (S. Barrington).
under the compost piles to enhance the convective forces, created by the temperature differences between the composting material and the ambient air (Sartaj et al., 1997). Active aeration requires the installation of ducts under the compost piles and fans pushing air into these ducts and through the compost piles (Haug, 1993). With active composting, thermocouples are often used to control the rate of aeration, as excessive aeration cools the compost and leads to large N losses while inadequate aeration prevents the proper development of stabilizing temperatures (Diaz et al., 1993). Although natural aeration can be rate limiting, passive aeration has proven just as efficient as active aeration, while being less costly. The concept of passive aeration was introduced by McGarry and Strainforth (1978). Mathur et al. (1990) successfully composted sheep, dairy and poultry manures with peat using passive aeration, as temperatures reached 55 °C within 4 days. Zhan et al. (1992) obtained temperatures of 60 and 65 °C, after 3 days of composting poultry manure with
0960-8524/03/$ - see front matter Ó 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 9 6 0 - 8 5 2 4 ( 0 2 ) 0 0 1 5 5 - 4
260
S. Barrington et al. / Bioresource Technology 86 (2003) 259–266
peat. Sartaj et al. (1997) found that passive aeration had a higher composting rate than active aeration, and did not produce adverse cooling effects and high N losses, as with active aeration. Patni et al. (2001) reported that passive aeration was cheaper, but just as effective as active aeration, and could lead to lower N losses especially for poultry manure, as poultry manure is especially rich in N. Passive aeration depends highly on the proper design of the ventilation ducts conveying air under the compost piles. These ducts must provide sufficient airflow, yet offer limited friction losses because of the low pressure differentials created by thermal buoyancy forces created by the warm air mass in the compost. The viscous forces occurring inside the ventilation ducts themselves, expressed in terms of pressure differential, depend on air velocity and duct size (ASHRAE, 2001). Convective airflows occurring inside the compost piles have been less documented, along with the resulting viscous forces or pressure differential developed. Lynch and Cherry (1996) developed a model characterizing the flow profile of air between passive aeration pipes under compost piles. For velocities of 0.1–0.6 mm s1 , and under active aeration regimes, Barrington et al. (in press) have demonstrated that pressure differentials respect the Hagen–Poiseuille equation for laminar flow. For velocities under 0.1 mm s1 , such as those occurring under passive air flow rates, pressure losses drop under 1.0 Pa for compost piles offering a porosity of 0.5–0.7. The objective of the project was to model the natural convective airflow rates developed within the compost, and the temperature differentials existing between the compost and the ambient air. To reach this objective, the compost temperatures under both passive and active aeration were compared over a 15–21 day period. These temperatures also verified that passive aeration was just as effective as active aeration. During the composting process, the convective airflow occurring in the compost under passive aeration was measured to establish a relationship between convective air flow and compost temperature. The Grashof number (Gr), representing the ratio of buoyancy to viscous forces, was used to relate compost temperature to convective airflow. Laboratory composting vessels were used to compost three types of compost (wood shavings, hay and straw with swine manure) at three different MC (60%, 65% and 70%), representing the general range of compost MC.
pressure drop respecting the perfect gas law. The density differential can be defined as (Geankoplis, 1983): b ¼ ðqa qw Þ=ðqðTw Ta ÞÞ ¼ 2=ðTf þ Ta Þ
ð1Þ
where b is a volumetric expansion coefficient, dimensionless; qa and qw are the ambient and hot surface air densities, in kg m3 , and Tw , Ta and Tf are the hot surface, ambient and surface air film temperature in K. Against the warm surface, the lighter warm air rises while creating a natural convective airflow. The same process occurs within the airflow channels of a compost pile. The air within the air flow channels of the compost material is heated, becomes lighter and rises through the air flow channels, while drawing cold air from under the pile. The buoyancy forces creating this natural convective airflow are resisted by the viscous forces of the airflow moving inside the airflow channels. Passive aeration creates such low air velocities inside flow channels that laminar conditions can be assumed (Barrington et al., in press) to evaluate the viscous forces acting inside the compost airflow channels. The Grashof number (Gr) can therefore be used to predict airflow rates through compost piles, assuming a desired temperature profile over time. The Gr is the ratio of buoyancy and viscous forces. Expressed in terms of volume of air over a unit cross-sectional area, the Gr number changes with apparent air velocity (air flow rate per unit cross-sectional area of the compost) through the compost, and the physical properties of the compost material. Such ratio is expressed as: Gr ¼ hAeq2 gðTw Ta Þ=ððTw þ Ta Þ=2Þ=l2
ð2Þ
where Gr is the Grashof number, dimensionless; h is the height of the pile in m; A is a unit cross-sectional area of compost pile expressed as 1.0 m2 ; e is the porosity of the compost, dimensionless; q is the density of the air at ambient temperature in kg m3 ; g is the gravitational constant in m2 s1 ; Tw and Ta are the compost and ambient air temperatures in K; and l is the viscosity of the ambient air in kg m1 s1 . The Grashof number plays a role similar to that of the Reynolds number in forced convection. The Reynolds number is defined as the ratio of viscous and inertia forces. Establishing the relationship between apparent air velocity through the compost and the Gr number can provide some means of modelling passive air flow rate through a compost pile, knowing its temperature, porosity, height and cross-sectional area.
2. Methods
2.2. Experimental material
2.1. Theoretical calculations
The bulking agents were wood (pine––Pinus) shavings, grass hay, and wheat (Triticum aestivum L.) straw. Liquid swine manure was added to each bulking agent at a wet mass ratio of 3:1. Because all mixtures were designed to have a C/N ratio of 18:1–20:1, some needed
Convective airflow results from cold air coming in contact with a warm surface. The resulting heating process lowers the density of the air, and creates an air
S. Barrington et al. / Bioresource Technology 86 (2003) 259–266
to be corrected using an additional nitrogen source. The wood shavings were corrected using rolled soybeans (Glycine max Merr.) as urea would boost the pH of the mixture above 9.0. The liquid swine manure was obtained from the grower hog barn at the Macdonald Campus Farm of McGill University, where the grower hogs are fed a corn and soybean ration. The pH, MC, total nitrogen (TN) and total carbon (TC) of the experimental materials were measured prior to composting (Table 1). For all experimental material, the particle size distribution was determined by sieving (Fig. 2). Three moisture contents (MCs––60%, 65% and 70%) were selected to represent the standard range used for composting and were tested in duplicate for each bulking agent and only one bulking agent was tested at any one time (Table 2). Tap water was added to the mixture to obtain the desired MC. The 70% MC corresponded to the maximum MC capable of being held by the bulking agent while the 60% MC corresponded to the minimum MC at which composting could be carried out effectively. Each bulking agent and MC combination was aerated in duplicate using active and passive aeration.
261
Table 2 Compost mixtures and initial characteristics Item
Compost mixture
Units
Wood shavings
Hay
Straw
Bulking agent Swine slurry Soybeans
% d.w. % d.w. % d.w.
79 13 8
75 25
74 26
Total
% d.w.
100
100
100
MC
% d.w.
63 67 72 18
60 65 70 20
60 66 71 19
C/N ratio
2.3. Composting vessel The tests were conducted using twelve identical 105-L cylindrical polyethylene vessels, 0.95 m high and 0.40 m in diameter (Fig. 1). Each vessel was insulated with 100mm mineral wool for a thermal resistance (RSI) value of 2.5 m2 °C W1 . For aeration purposes, an air plenum, 100 mm in height, was created at the bottom of each container using a supporting metal wire mesh. The wall of the vessel at the level of the plenum had one perforation measuring 50 mm in diameter. 2.4. Method
Fig. 1. Experimental compost vessels.
Only one bulking agent adjusted at three MC was composted at any one time. For each test, twelve vessels were filled to give three MC levels, each in quadruplicate. Two of the four vessels were aerated passively while the other two were aerated actively. The bulking
agent and swine manure slurry were sampled and analyzed before each test. A 15 kg quantity of bulking agent was mixed with the required amounts of swine slurry, water and amendment
Table 1 Characteristics of experimental material Material
Characteristics a
Swine slurry Wood shavings Hay Straw Soybeans
DM (%)
TN (g kg1 )
pH
16.8 92.4 87.2 86.9 90.0
75.1 (5.54) 0.64 (0.038) 11.0 (1.80) 9.79 (0.25) 6.8 (0.3)
7.2 4.4 5.2 6.3 –
(0.86) (0.33) (1.11) (0.45) (0.5)
(0.16) (0.13) (0.10) (0.60)
Ash (%)
C (%)
Density (kg L1 )
19.8 (1.38) 0.38 (0.13) 6.5 (0.26) 8.96 (0.42) 2.3 (0.2)
43.8 54.4 51.1 49.8 53.4
1.03 0.78 0.66 0.50 –
(0.01) (0.10) (0.09) (0.04)
Note: all analyses are reported on a dry weight (DW) basis; the C content was calculated from fð100% ashð%ÞÞ=183g; the value in parenthesis is the standard deviation. a Average analysis for the swine slurry, as its dry matter content varied, between tests, each test consisting of a different bulking agent.
262
S. Barrington et al. / Bioresource Technology 86 (2003) 259–266
(soybeans for the wood shavings) to obtain a C/N ratio ranging between 18:1 and 20:1 and a MC of either 60%, 65% or 70%. The batches were prepared by weighing all ingredients on a scale with an accuracy of 0.1 kg, dumping the ingredients in a 400 L container and thoroughly mixing all ingredients manually. Then, the mixture was used to fill two experimental vessels, one for the passive aeration test and the other for the active aeration test. The procedure was repeated for all other treatments using the same amount of bulking agent and swine liquid manure, but with different amounts of tap water. Once filled, the twelve vessels were either passively and actively aerated. For the passively aerated vessels, the plenum perforation was simply left open. For the actively aerated vessels, the plenum received compressed air at a rate of 2 L min1 , corresponding to 4 mg of dry air s1 kg1 of initial compost dry matter (Lau et al., 1992). Compost temperature was monitored every 15 min using a thermocouple, calibrated prior to the test and inserted through the wall of the composting vessels, at mid height. A manual thermometer was used to verify the variation in temperature of the compost with depth. The mid depth thermocouple was found to measure the temperature of the compost over its full depth, except for the top 10 mm, which fell in temperature to that of ambient air. The ambient air temperature was also recorded using a calibrated thermocouple installed in the experimental room. The convective air flow rates from the passive composts were measured by placing a sealed cap on each vessel for the duration of the measurement procedure only (5 min day1 ) and measuring the velocity of the air escaping the cap through a central orifice, 5 mm in diameter. The discharge coefficient, Cd , of the orifice was previously measured using an empty vessel. An airflow rate, measured using a ball air flow meter (0–12 L min1 0:1 L min1 ) was pushed into the empty compost vessel while measuring the orifice air velocity, V. This procedure provided a measure of the discharge orifice, Cd : Cd ¼ 1:0 e0:1547V for V under 2:75 m s1
ð3Þ
1
ð4Þ
Cd ¼ 0:64 for V greater than 2:75 m s
compost material cross-sectional area, perpendicular to air flow, in m2 . 2.5. Analytical procedures All composts materials were analyzed using standard methods (APHA et al., 1995). The MC was determined by drying the compost sample at 103 °C for 24 h. The pH was determined with a probe on 5 g of material soaked in 50 ml of distilled water for 24 h. The total kjeldahl nitrogen (TKN) was determined by digesting the material with sulphuric acid at 500 °C and measuring the NH3 -N content at a sample pH of 13, using a NH3 sensitive electrode. The NO 3 -N content was determined by soaking a 5 g sample in 50 ml of distilled water for 24 h and measuring the level of NO 3 -N using a NO3 sensitive electrode. Total nitrogen was calculated by adding TKN to NO 3 -N. The ash content was obtained by burning each dried sample at 500 °C for 4 h. The organic matter content was equated to the volatile portion of the burned samples and converted to C using a factor of 1.83 (Castellanos and Pratt, 1981). 2.6. Statistical analysis For each bulking agent, temperatures developed for both passive and active aeration were compared using the procedure ANOVA. For each ANOVA, a MC and aeration combination consisted of a treatment while each measurement over time was classified as a block. DuncanÕs New Multiple Range Test was used to establish the significance of differences among treatments (Steel and Torrie, 1986). The relationship between the compost Gr number and apparent air velocity was established through a linear regression analysis performed using Microsoft 2000Õs Excel program (Fig. 2).
The orifice airflow was calculated as: Q ¼ ACd V
ð5Þ
where Q is the airflow rate in m3 s1 and A is the crosssectional area of the orifice in m2 . The apparent air velocity through the compost was computed as: Va ¼ Cd VA=Ac ¼ Q=Ac
ð6Þ
where Va is the apparent air velocity through the compost in m s1 ; Cd V is the average velocity through the cap orifice in m s1 , and; A and Ac are the cap orifice and
Fig. 2. Particle size distribution of compost materials.
S. Barrington et al. / Bioresource Technology 86 (2003) 259–266
3. Results and discussion 3.1. Temperature regime The temperature regime of all composted materials, under passive and active aeration, are illustrated in Fig. 3a–c, respectively, for wood shavings, hay and straw. For the wood shavings, all MC of the actively aerated composts and the high MC of the passively aerated compost peaked at a temperature of 67 °C after 2 days, while the mid and low MC of the passive treatments peaked at 65 °C after 4.5 days (Fig. 3a). From day 2–5, the low and medium MC for the passively aerated compost showed lower temperatures (95% confidence level). Nevertheless, all treatments reached temperatures well above the stabilization level of 55 °C.
263
For the hay, the actively and passively aerated composts demonstrated respective peak temperatures of 70 and 57 °C. The temperature of the actively aerated compost remained 10 °C higher that that passively aerated for most of the duration of the test (95% confidence level (Fig. 3b). For the straw, the high and medium MC, actively aerated and the high MC passively aerated compost demonstrated higher temperatures from day 0 to 5 (95% confidence level). After 5 days, there was no significant difference in temperature regime between the actively and passively aerated compost. All composts reached an average peak temperature ranging between 60 and 70 °C, after 5 days (Fig. 3a). For all compost, the addition of 2 L of water caused a temporary rise in temperature, such as for straw, on day 12. In general, passive aeration leads to more moderate thermophilic temperatures in the range of 55–65 °C, as compared to active aeration where temperatures reached 70 °C. This will later be explained when comparing passive and active airflow rates. Compost MC had a variable effect on temperature regime. With wood shavings, the mid MC (65%) produced the highest temperature regime for both types of aeration. With hay, there was no significant difference between MC and temperature regime. With straw, higher MC led to higher temperatures, for both types of aeration (95% confidence level). After 6 days of composting, MC had no effect on temperature regime because by that time, all compost had lost significant amounts of water by evaporation and developed the same limiting range of MC. 3.2. Convective airflow from passive aeration
Fig. 3. Temperature regime for the wood shavings, hay and straw compost.
The convective air flow rates created through the passively aerated, for mid level MC compost are illustrated as a function of time, in Fig. 4a–c, for the shavings, hay and straw compost, respectively. The temperature of the compost and the temperature difference between the compost and ambient temperature are illustrated along with the convective air flow rates to demonstrate that the air flow rates were a function of both time and compost temperature. The results obtained with the low and high MC compost are not illustrated, as they are quite similar to those obtained with the mid MC compost. Except for the straw compost, the convective airflow rate peaked at the same time as the compost temperature and the temperature difference. The straw compost demonstrated convective airflow rates peaking from day 8 to 15, while the compost and the differential temperatures peaked from day 2 to 6. Large changes in compost temperature lead to small changes in convective airflow rates. This reflects the fact that the buoyancy forces are proportional to the
264
S. Barrington et al. / Bioresource Technology 86 (2003) 259–266
c)
Fig. 5. Compost Gr number and apparent air velocity for the wood shavings, hay and straw compost. All Gr values have been reduced by a factor of 1010 . Fig. 4. Compost temperature, compost and ambient air temperature differential and air flow rate for the wood shavings, hay and straw compost.
temperature differential divided by the compost temperature expressed in K. The temperature differentials remained in the range of 10–50 K while the film temperature was of the order of 283–323 K, for a variation in buoyancy forces of 0.035–0.155. In general, all composts and MC produced a convective airflow rate ranging between 15 and 7 mg of dry air s1 , for day 2–18, respectively. This corresponded to an aeration rate of 1.5–0.7 mg dry air s1 kg1 of dry compost (initial dry weight), while the actively aerated compost received airflow of 4.0 mg dry air s1 kg1 of dry compost. The higher aeration rate developed for the active treatment, as compared to the passive treatment, explains their slightly higher temperatures. The apparent compost natural convective air velocity is compared to the Gr in Fig. 5a–c, for the wood shavings, hay and straw compost respectively. Only the mid MC results are illustrated as the other MC produce similar relationships (Table 3). For the shavings and the hay compost, apparent airflow increased with Gr. For the straw compost, apparent air velocity increased for Gr values up to 1.5, 2.0 and 2.5, for the low, mid and high MC, respectively. For Gr above these values, ap-
Table 3 Apparent air velocity as a function of Gr Compost
Low
Mid
High
Wood shavings
Va ¼ 0:34Gr0 (R2 ¼ 0:43)
Va ¼ 0:46Gr0 (R2 ¼ 0:49)
Va ¼ 0:41Gr0 (R2 ¼ 0:42)
Hay
Va ¼ 0:31Gr0 (R2 ¼ 0:72)
Va ¼ 0:49Gr0 (R2 ¼ 0:72)
Va ¼ 0:33Gr0 (R2 ¼ 0:48)
Straw
Va ¼ 0:18Gr0 (Va < 0:02) (R2 ¼ 0:25)
Va ¼ 0:22Gr0 (Va < 0:0225) (R2 ¼ 0:36)
Va ¼ 0:29Gr0 (Va < 0:0275) (R2 ¼ 0:43)
Straw
Va ¼ 0:64Gr0 (Va > 0:02) (R2 ¼ 0:25)
Va ¼ 1:51Gr0 (Va > 0:0225) (R2 ¼ 0:36)
Va ¼ 2:08Gr0 (Va > 0:0275) (R2 ¼ 0:43)
Note: Va is expressed in mm s1 ; Gr0 is Gr 1010 and is expressed for a unit compost cross-sectional area in m2 .
parent air velocities decreased. The R2 values associated with the relationship between Gr and Va range between 0.25 and 0.72 because of the change in compost properties during the composting period, especially for the straw compost. For the wood shavings and hay compost, viscous forces increased in parallel with the buoyancy forces, as air velocity increased in the compost airflow channels. For the straw compost, at respective Gr numbers of 1.5,
S. Barrington et al. / Bioresource Technology 86 (2003) 259–266
2.0 and 2.5, for the low, mid and high MC, viscous forces became more important than buoyancy forces, and air flow rates decreased with increasing temperature differentials. This explained why the straw compost was the only material not showing peak airflow rates with peak compost temperatures. 3.3. Designing aeration ducts for passive composting The relationships developed can be used to predict the maximum airflow developing inside a compost mass and the size of ventilation ducts required. For example, let us consider a centre where swine manure is incorporated into hay to be composted inside a silo measuring 30 m in length, 2.4 m in width and 2.0 m in height. The compost mixture has an initial height of 1.8 m, a moisture content of 65%, a bulk density of 0.25 kg m3 and a porosity of 70%. Assume also that the expected maximum temperature is 65 °C. Three ventilation ducts will be installed under and over the length of the compost mass, at a spacing of 0.8 m. These ventilation ducts will receive air from the centre of the silo, to reduce the air pressure drop. Thus, each duct has a maximum length of 18 m, including the length to the outlet. From Eq. (1), the Gr is 0:45 1010 and the expected apparent air velocity across the compost surface is 0.22 mm s1 (Table 3). Thus, the ventilation ducts under the compost mass must offer an airflow of 8.0 L s1 . If three circular ducts are used under the mass, each 0.8 m apart, then each duct must provide 2.7 L s1 . To finish the design, the ventilation ducts must offer a cross-sectional area leading to a pressure drop over their length which is much lower than that developed by the buoyancy forces. Thus, the pressure drop in the aeration ducts will have no effect on passive airflow. Therefore, designing ducts for a pressure drop equal to 5% of that developed by the apparent air velocity across the compost mass is likely to respect this condition. Pressure drops occurring through the compost mass have been related by Barrington et al. (in press), as a function of apparent air velocity. Using such relationship, the expected pressure drop caused by an apparent air flow rate of 0.22 mm s1 , through the hay and swine manure compost column of 1.8 m, is expected to be of the order of 28.0 Pa. Thus, the pressure drop occurring inside the ventilation ducts should be limited to 5% of 28 Pa or 1.4 Pa, or 0.08 Pa m1 of duct length. A single duct with an inside diameter of 155 mm, offers a pressure drop of 0.01 Pa m1 , or 0.18 Pa at an air flow rate of 10 L s1 (ASHRAE, 2001). Thus, three air ducts each measuring 18 m in length and offering a smooth inside diameter of 155 mm will suffice in creating the required passive ventilation for the silo described above. From this calculation, duct pressure losses can be insignificant as compared to those developed from the viscous forces inside the compost mass.
265
The present calculations were conducted for a compost mass having a depth of 1.8 m while all tests were carried out using a compost depth of 0.8 m. Thus, the present research needs to be repeated for various depths of compost and using various types of bulking agents and wastes. Thus, a full range of relationships between Gr, apparent air velocities and the resulting air pressure drop could be measured.
4. Conclusions Passive aeration was just as effective as active aeration in bringing about stabilizing temperatures of 55 °C, in all three types of composts and at all three MC. Nevertheless, passive aeration produced peak temperatures of the order of 57–65 °C, while active aeration produced peak temperatures reaching 70 °C. Compost MC had an effect only on the peak compost temperature, occurring between days 2 and 6. Compost MC losses after 6 days lead to no difference in temperature regime between initial MC treatments and for the same bulking agent. Convective airflow rates expressed as apparent velocities across the compost mass, were related to the Grashof number representing the ratio between buoyancy and viscous forces. As opposed to the straw compost, the wood shavings and hay compost demonstrated viscous forces increasing in parallel with the buoyancy forces, as air velocity increased in the compost airflow channels. Thus, wood shavings and hay are better bulking agents, offering well formed airflow channels, as compared to straw. As a bulking agent, straw has a softer structure that tends to collapse when wet. The relationship established between apparent air velocity and the air pressure drop across the depth of compost mass provides a mean of designing passive ventilation systems under compost piles.
Acknowledgements This project was financially supported by the Natural Sciences and Engineering Research Council of Canada and the Macdonald Campus Farm of McGill University.
References APHA, WPCF and AWWA, 1995. Standard procedures for the analysis of water and wastewater. American Public Healthy Association, American Waste Water Works Association and Water Pollution Control Federation. Washington, DC. ASHRAE. 2001. Handbook of fundamentals. American Society of Heating, Refrigeration and Air Conditioning Engineering, Atlanta, Georgia, USA. (Chapter 32).
266
S. Barrington et al. / Bioresource Technology 86 (2003) 259–266
Barrington, S., Choiniere, D., Trigui, M., Knight, W. Compost air flow resistance. Journal of Agricultural Engineering Research, in press. Castellanos, J.Z., Pratt, P.F., 1981. Mineralization of manure nitrogen––correlation with laboratory indexes. Soil Science Society of America Journal 45, 354–357. Diaz, L.F., Savage, G.M., Eggerth, L.L., Golueke, C.G., 1993. Composting and Recycling. Lewis Publishers, Boca Raton, FL. Fernandes, L., Zhan, W., Patni, N.K., Jui, P.Y., 1994. Temperature distribution and variation in passively aerated static compost piles. Transactions of the ASAE 48, 257–263. Geankoplis, C.J., 1983. In: Transport Processes and Unit Operations, Second edition. Allyn and Bacon, inc, Boston, USA, pp. 244–246. Haug, R.T., 1993. The Practical Handbook of Composting Engineering. Lewis Publishers, Boca Raton, FL, USA. Lau, A.K., Lo, K.V., Liao, P.H., Yu, J.C., 1992. Aeration experiments for swine waste composting. Bioresource Technology 41, 145–152. Lynch, N.J., Cherry, R.S., 1996. Design of passively aerated compost piles: vertical air velocities between pipes. Biotechnology Progress 12 (5), 624–629.
Mathur, S., Patni, N.K., Levesque, M.P., 1990. Static pile passive aeration composting of manure slurries using peat as a bulking agent. Biological Wastes 34 (4), 323–334. McGarry, M.G., Strainforth, J., 1978. In: Compost fertilizer and biogas production for human and farm wastes in PeopleÕs Republic of China. International Development Research Centre, Ottawa, Canada, pp. 7–13. Patni, N.K., Kannagara, T., Nielsen, G., Dinel, H., 2001. Composting caged-layer manure in passively aerated and turned windrows. ASAE paper no. 012271. American Society of Agricultural Engineers, St. Joseph, Michigan, USA. Sartaj, M., Fernandes, L., Patni, N.K., 1997. Performance of forced, passive and natural aeration methods for composting manure slurries. Transactions of the ASAE 40 (2), 457–463. Steel, R.D.G., Torrie, J.H., 1986. Principles and Procedures of Statistics, a Biometrical Approach, Second ed. McGraw Hill Inc., New York. Zhan, W., Fernandes, L., Patni, N., 1992. Composting of poultry manure slurries. CSAE paper no. 92–515. Canadian Society of Agricultural Engineering, Saskatoon, Canada.
Compost convective airflow under passive aeration Suzelle Barrington *, Denis Choiniere, Maher Trigui, William Knight Department of Agricultural and Biosystems Engineering, Macdonald Campus of McGill University, 21 111 Lakeshore, Ste Anne de Bellevue, Qu ebec, Canada H9X 3V9 Received 17 August 2001; received in revised form 24 June 2002; accepted 26 June 2002
Abstract For composting, passive aeration can save energy costs while being just as efficient as forced or active aeration. Passive aeration requires the proper design of aeration ducts, and thus, the proper prediction of the convective airflow rates created by the temperature differential between the compost and the ambient air. To establish such relationship, the temperature and convective air flow regimes of composts were investigated using three bulking agents (wood shavings, hay and straw), each at three moisture contents (MC––60%, 65% and 70%) spanning the normal values. All bulking agent and aeration treatments were aerated in duplicate under passive and active regimes. Laboratory vessels of 105 L were used for all treatments. Passive aeration treatments produced temperatures above 57 °C, as did the treatments actively aerated at 4 mg of air s1 kg1 of initial dry compost material. Compost MC had an effect only on the peak compost temperature, occurring between day 2 and 6. After 6 days of composting, MC no longer had any effect on temperature regime because of the loss of moisture by each mixture. A relationship was established between the Grasholf number (Gr––ratio of buoyancy to viscous forces) and the convective airflow rates, to size the aeration ducts for passive aeration. In general, convective airflow rates ranged from 1.5 to 0.7 mg of dry air s1 kg1 of initial compost dry matter, from day 0 to day 20, respectively, and for all compost treatments. This airflow rate sizes the aeration ducts installed under compost piles for passive aeration. As compared to straw where airflow rate dropped over a given level of Gr, wood shavings and hay were found to be more effective as bulking agents, as their airflow rate increased constantly with Gr. Ó 2002 Elsevier Science Ltd. All rights reserved. Keywords: Compost; Passive aeration; Convective airflow; Grasholf number
1. Introduction Composting is a controlled biological process used to stabilize and transform wastes into an aesthetic soil amendment. Aeration is one component of the controlling process as it insures the growth of adequate aerobic microbe populations and the development of stabilizing temperatures. Composts can be aerated by one of three methods: natural or static pile, passive and active or forced. Natural aeration is the cheapest and simplest, as it requires no installations. Natural aeration occurs simply by diffusion and convection, governed by the exposed surfaces and their respective properties (Fernandes et al., 1994). Passive aeration requires the installation of ducts
*
Corresponding author. Tel.: +1-514-398-7776; fax: +1-514-3988387. E-mail address: [email protected] (S. Barrington).
under the compost piles to enhance the convective forces, created by the temperature differences between the composting material and the ambient air (Sartaj et al., 1997). Active aeration requires the installation of ducts under the compost piles and fans pushing air into these ducts and through the compost piles (Haug, 1993). With active composting, thermocouples are often used to control the rate of aeration, as excessive aeration cools the compost and leads to large N losses while inadequate aeration prevents the proper development of stabilizing temperatures (Diaz et al., 1993). Although natural aeration can be rate limiting, passive aeration has proven just as efficient as active aeration, while being less costly. The concept of passive aeration was introduced by McGarry and Strainforth (1978). Mathur et al. (1990) successfully composted sheep, dairy and poultry manures with peat using passive aeration, as temperatures reached 55 °C within 4 days. Zhan et al. (1992) obtained temperatures of 60 and 65 °C, after 3 days of composting poultry manure with
0960-8524/03/$ - see front matter Ó 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 9 6 0 - 8 5 2 4 ( 0 2 ) 0 0 1 5 5 - 4
260
S. Barrington et al. / Bioresource Technology 86 (2003) 259–266
peat. Sartaj et al. (1997) found that passive aeration had a higher composting rate than active aeration, and did not produce adverse cooling effects and high N losses, as with active aeration. Patni et al. (2001) reported that passive aeration was cheaper, but just as effective as active aeration, and could lead to lower N losses especially for poultry manure, as poultry manure is especially rich in N. Passive aeration depends highly on the proper design of the ventilation ducts conveying air under the compost piles. These ducts must provide sufficient airflow, yet offer limited friction losses because of the low pressure differentials created by thermal buoyancy forces created by the warm air mass in the compost. The viscous forces occurring inside the ventilation ducts themselves, expressed in terms of pressure differential, depend on air velocity and duct size (ASHRAE, 2001). Convective airflows occurring inside the compost piles have been less documented, along with the resulting viscous forces or pressure differential developed. Lynch and Cherry (1996) developed a model characterizing the flow profile of air between passive aeration pipes under compost piles. For velocities of 0.1–0.6 mm s1 , and under active aeration regimes, Barrington et al. (in press) have demonstrated that pressure differentials respect the Hagen–Poiseuille equation for laminar flow. For velocities under 0.1 mm s1 , such as those occurring under passive air flow rates, pressure losses drop under 1.0 Pa for compost piles offering a porosity of 0.5–0.7. The objective of the project was to model the natural convective airflow rates developed within the compost, and the temperature differentials existing between the compost and the ambient air. To reach this objective, the compost temperatures under both passive and active aeration were compared over a 15–21 day period. These temperatures also verified that passive aeration was just as effective as active aeration. During the composting process, the convective airflow occurring in the compost under passive aeration was measured to establish a relationship between convective air flow and compost temperature. The Grashof number (Gr), representing the ratio of buoyancy to viscous forces, was used to relate compost temperature to convective airflow. Laboratory composting vessels were used to compost three types of compost (wood shavings, hay and straw with swine manure) at three different MC (60%, 65% and 70%), representing the general range of compost MC.
pressure drop respecting the perfect gas law. The density differential can be defined as (Geankoplis, 1983): b ¼ ðqa qw Þ=ðqðTw Ta ÞÞ ¼ 2=ðTf þ Ta Þ
ð1Þ
where b is a volumetric expansion coefficient, dimensionless; qa and qw are the ambient and hot surface air densities, in kg m3 , and Tw , Ta and Tf are the hot surface, ambient and surface air film temperature in K. Against the warm surface, the lighter warm air rises while creating a natural convective airflow. The same process occurs within the airflow channels of a compost pile. The air within the air flow channels of the compost material is heated, becomes lighter and rises through the air flow channels, while drawing cold air from under the pile. The buoyancy forces creating this natural convective airflow are resisted by the viscous forces of the airflow moving inside the airflow channels. Passive aeration creates such low air velocities inside flow channels that laminar conditions can be assumed (Barrington et al., in press) to evaluate the viscous forces acting inside the compost airflow channels. The Grashof number (Gr) can therefore be used to predict airflow rates through compost piles, assuming a desired temperature profile over time. The Gr is the ratio of buoyancy and viscous forces. Expressed in terms of volume of air over a unit cross-sectional area, the Gr number changes with apparent air velocity (air flow rate per unit cross-sectional area of the compost) through the compost, and the physical properties of the compost material. Such ratio is expressed as: Gr ¼ hAeq2 gðTw Ta Þ=ððTw þ Ta Þ=2Þ=l2
ð2Þ
where Gr is the Grashof number, dimensionless; h is the height of the pile in m; A is a unit cross-sectional area of compost pile expressed as 1.0 m2 ; e is the porosity of the compost, dimensionless; q is the density of the air at ambient temperature in kg m3 ; g is the gravitational constant in m2 s1 ; Tw and Ta are the compost and ambient air temperatures in K; and l is the viscosity of the ambient air in kg m1 s1 . The Grashof number plays a role similar to that of the Reynolds number in forced convection. The Reynolds number is defined as the ratio of viscous and inertia forces. Establishing the relationship between apparent air velocity through the compost and the Gr number can provide some means of modelling passive air flow rate through a compost pile, knowing its temperature, porosity, height and cross-sectional area.
2. Methods
2.2. Experimental material
2.1. Theoretical calculations
The bulking agents were wood (pine––Pinus) shavings, grass hay, and wheat (Triticum aestivum L.) straw. Liquid swine manure was added to each bulking agent at a wet mass ratio of 3:1. Because all mixtures were designed to have a C/N ratio of 18:1–20:1, some needed
Convective airflow results from cold air coming in contact with a warm surface. The resulting heating process lowers the density of the air, and creates an air
S. Barrington et al. / Bioresource Technology 86 (2003) 259–266
to be corrected using an additional nitrogen source. The wood shavings were corrected using rolled soybeans (Glycine max Merr.) as urea would boost the pH of the mixture above 9.0. The liquid swine manure was obtained from the grower hog barn at the Macdonald Campus Farm of McGill University, where the grower hogs are fed a corn and soybean ration. The pH, MC, total nitrogen (TN) and total carbon (TC) of the experimental materials were measured prior to composting (Table 1). For all experimental material, the particle size distribution was determined by sieving (Fig. 2). Three moisture contents (MCs––60%, 65% and 70%) were selected to represent the standard range used for composting and were tested in duplicate for each bulking agent and only one bulking agent was tested at any one time (Table 2). Tap water was added to the mixture to obtain the desired MC. The 70% MC corresponded to the maximum MC capable of being held by the bulking agent while the 60% MC corresponded to the minimum MC at which composting could be carried out effectively. Each bulking agent and MC combination was aerated in duplicate using active and passive aeration.
261
Table 2 Compost mixtures and initial characteristics Item
Compost mixture
Units
Wood shavings
Hay
Straw
Bulking agent Swine slurry Soybeans
% d.w. % d.w. % d.w.
79 13 8
75 25
74 26
Total
% d.w.
100
100
100
MC
% d.w.
63 67 72 18
60 65 70 20
60 66 71 19
C/N ratio
2.3. Composting vessel The tests were conducted using twelve identical 105-L cylindrical polyethylene vessels, 0.95 m high and 0.40 m in diameter (Fig. 1). Each vessel was insulated with 100mm mineral wool for a thermal resistance (RSI) value of 2.5 m2 °C W1 . For aeration purposes, an air plenum, 100 mm in height, was created at the bottom of each container using a supporting metal wire mesh. The wall of the vessel at the level of the plenum had one perforation measuring 50 mm in diameter. 2.4. Method
Fig. 1. Experimental compost vessels.
Only one bulking agent adjusted at three MC was composted at any one time. For each test, twelve vessels were filled to give three MC levels, each in quadruplicate. Two of the four vessels were aerated passively while the other two were aerated actively. The bulking
agent and swine manure slurry were sampled and analyzed before each test. A 15 kg quantity of bulking agent was mixed with the required amounts of swine slurry, water and amendment
Table 1 Characteristics of experimental material Material
Characteristics a
Swine slurry Wood shavings Hay Straw Soybeans
DM (%)
TN (g kg1 )
pH
16.8 92.4 87.2 86.9 90.0
75.1 (5.54) 0.64 (0.038) 11.0 (1.80) 9.79 (0.25) 6.8 (0.3)
7.2 4.4 5.2 6.3 –
(0.86) (0.33) (1.11) (0.45) (0.5)
(0.16) (0.13) (0.10) (0.60)
Ash (%)
C (%)
Density (kg L1 )
19.8 (1.38) 0.38 (0.13) 6.5 (0.26) 8.96 (0.42) 2.3 (0.2)
43.8 54.4 51.1 49.8 53.4
1.03 0.78 0.66 0.50 –
(0.01) (0.10) (0.09) (0.04)
Note: all analyses are reported on a dry weight (DW) basis; the C content was calculated from fð100% ashð%ÞÞ=183g; the value in parenthesis is the standard deviation. a Average analysis for the swine slurry, as its dry matter content varied, between tests, each test consisting of a different bulking agent.
262
S. Barrington et al. / Bioresource Technology 86 (2003) 259–266
(soybeans for the wood shavings) to obtain a C/N ratio ranging between 18:1 and 20:1 and a MC of either 60%, 65% or 70%. The batches were prepared by weighing all ingredients on a scale with an accuracy of 0.1 kg, dumping the ingredients in a 400 L container and thoroughly mixing all ingredients manually. Then, the mixture was used to fill two experimental vessels, one for the passive aeration test and the other for the active aeration test. The procedure was repeated for all other treatments using the same amount of bulking agent and swine liquid manure, but with different amounts of tap water. Once filled, the twelve vessels were either passively and actively aerated. For the passively aerated vessels, the plenum perforation was simply left open. For the actively aerated vessels, the plenum received compressed air at a rate of 2 L min1 , corresponding to 4 mg of dry air s1 kg1 of initial compost dry matter (Lau et al., 1992). Compost temperature was monitored every 15 min using a thermocouple, calibrated prior to the test and inserted through the wall of the composting vessels, at mid height. A manual thermometer was used to verify the variation in temperature of the compost with depth. The mid depth thermocouple was found to measure the temperature of the compost over its full depth, except for the top 10 mm, which fell in temperature to that of ambient air. The ambient air temperature was also recorded using a calibrated thermocouple installed in the experimental room. The convective air flow rates from the passive composts were measured by placing a sealed cap on each vessel for the duration of the measurement procedure only (5 min day1 ) and measuring the velocity of the air escaping the cap through a central orifice, 5 mm in diameter. The discharge coefficient, Cd , of the orifice was previously measured using an empty vessel. An airflow rate, measured using a ball air flow meter (0–12 L min1 0:1 L min1 ) was pushed into the empty compost vessel while measuring the orifice air velocity, V. This procedure provided a measure of the discharge orifice, Cd : Cd ¼ 1:0 e0:1547V for V under 2:75 m s1
ð3Þ
1
ð4Þ
Cd ¼ 0:64 for V greater than 2:75 m s
compost material cross-sectional area, perpendicular to air flow, in m2 . 2.5. Analytical procedures All composts materials were analyzed using standard methods (APHA et al., 1995). The MC was determined by drying the compost sample at 103 °C for 24 h. The pH was determined with a probe on 5 g of material soaked in 50 ml of distilled water for 24 h. The total kjeldahl nitrogen (TKN) was determined by digesting the material with sulphuric acid at 500 °C and measuring the NH3 -N content at a sample pH of 13, using a NH3 sensitive electrode. The NO 3 -N content was determined by soaking a 5 g sample in 50 ml of distilled water for 24 h and measuring the level of NO 3 -N using a NO3 sensitive electrode. Total nitrogen was calculated by adding TKN to NO 3 -N. The ash content was obtained by burning each dried sample at 500 °C for 4 h. The organic matter content was equated to the volatile portion of the burned samples and converted to C using a factor of 1.83 (Castellanos and Pratt, 1981). 2.6. Statistical analysis For each bulking agent, temperatures developed for both passive and active aeration were compared using the procedure ANOVA. For each ANOVA, a MC and aeration combination consisted of a treatment while each measurement over time was classified as a block. DuncanÕs New Multiple Range Test was used to establish the significance of differences among treatments (Steel and Torrie, 1986). The relationship between the compost Gr number and apparent air velocity was established through a linear regression analysis performed using Microsoft 2000Õs Excel program (Fig. 2).
The orifice airflow was calculated as: Q ¼ ACd V
ð5Þ
where Q is the airflow rate in m3 s1 and A is the crosssectional area of the orifice in m2 . The apparent air velocity through the compost was computed as: Va ¼ Cd VA=Ac ¼ Q=Ac
ð6Þ
where Va is the apparent air velocity through the compost in m s1 ; Cd V is the average velocity through the cap orifice in m s1 , and; A and Ac are the cap orifice and
Fig. 2. Particle size distribution of compost materials.
S. Barrington et al. / Bioresource Technology 86 (2003) 259–266
3. Results and discussion 3.1. Temperature regime The temperature regime of all composted materials, under passive and active aeration, are illustrated in Fig. 3a–c, respectively, for wood shavings, hay and straw. For the wood shavings, all MC of the actively aerated composts and the high MC of the passively aerated compost peaked at a temperature of 67 °C after 2 days, while the mid and low MC of the passive treatments peaked at 65 °C after 4.5 days (Fig. 3a). From day 2–5, the low and medium MC for the passively aerated compost showed lower temperatures (95% confidence level). Nevertheless, all treatments reached temperatures well above the stabilization level of 55 °C.
263
For the hay, the actively and passively aerated composts demonstrated respective peak temperatures of 70 and 57 °C. The temperature of the actively aerated compost remained 10 °C higher that that passively aerated for most of the duration of the test (95% confidence level (Fig. 3b). For the straw, the high and medium MC, actively aerated and the high MC passively aerated compost demonstrated higher temperatures from day 0 to 5 (95% confidence level). After 5 days, there was no significant difference in temperature regime between the actively and passively aerated compost. All composts reached an average peak temperature ranging between 60 and 70 °C, after 5 days (Fig. 3a). For all compost, the addition of 2 L of water caused a temporary rise in temperature, such as for straw, on day 12. In general, passive aeration leads to more moderate thermophilic temperatures in the range of 55–65 °C, as compared to active aeration where temperatures reached 70 °C. This will later be explained when comparing passive and active airflow rates. Compost MC had a variable effect on temperature regime. With wood shavings, the mid MC (65%) produced the highest temperature regime for both types of aeration. With hay, there was no significant difference between MC and temperature regime. With straw, higher MC led to higher temperatures, for both types of aeration (95% confidence level). After 6 days of composting, MC had no effect on temperature regime because by that time, all compost had lost significant amounts of water by evaporation and developed the same limiting range of MC. 3.2. Convective airflow from passive aeration
Fig. 3. Temperature regime for the wood shavings, hay and straw compost.
The convective air flow rates created through the passively aerated, for mid level MC compost are illustrated as a function of time, in Fig. 4a–c, for the shavings, hay and straw compost, respectively. The temperature of the compost and the temperature difference between the compost and ambient temperature are illustrated along with the convective air flow rates to demonstrate that the air flow rates were a function of both time and compost temperature. The results obtained with the low and high MC compost are not illustrated, as they are quite similar to those obtained with the mid MC compost. Except for the straw compost, the convective airflow rate peaked at the same time as the compost temperature and the temperature difference. The straw compost demonstrated convective airflow rates peaking from day 8 to 15, while the compost and the differential temperatures peaked from day 2 to 6. Large changes in compost temperature lead to small changes in convective airflow rates. This reflects the fact that the buoyancy forces are proportional to the
264
S. Barrington et al. / Bioresource Technology 86 (2003) 259–266
c)
Fig. 5. Compost Gr number and apparent air velocity for the wood shavings, hay and straw compost. All Gr values have been reduced by a factor of 1010 . Fig. 4. Compost temperature, compost and ambient air temperature differential and air flow rate for the wood shavings, hay and straw compost.
temperature differential divided by the compost temperature expressed in K. The temperature differentials remained in the range of 10–50 K while the film temperature was of the order of 283–323 K, for a variation in buoyancy forces of 0.035–0.155. In general, all composts and MC produced a convective airflow rate ranging between 15 and 7 mg of dry air s1 , for day 2–18, respectively. This corresponded to an aeration rate of 1.5–0.7 mg dry air s1 kg1 of dry compost (initial dry weight), while the actively aerated compost received airflow of 4.0 mg dry air s1 kg1 of dry compost. The higher aeration rate developed for the active treatment, as compared to the passive treatment, explains their slightly higher temperatures. The apparent compost natural convective air velocity is compared to the Gr in Fig. 5a–c, for the wood shavings, hay and straw compost respectively. Only the mid MC results are illustrated as the other MC produce similar relationships (Table 3). For the shavings and the hay compost, apparent airflow increased with Gr. For the straw compost, apparent air velocity increased for Gr values up to 1.5, 2.0 and 2.5, for the low, mid and high MC, respectively. For Gr above these values, ap-
Table 3 Apparent air velocity as a function of Gr Compost
Low
Mid
High
Wood shavings
Va ¼ 0:34Gr0 (R2 ¼ 0:43)
Va ¼ 0:46Gr0 (R2 ¼ 0:49)
Va ¼ 0:41Gr0 (R2 ¼ 0:42)
Hay
Va ¼ 0:31Gr0 (R2 ¼ 0:72)
Va ¼ 0:49Gr0 (R2 ¼ 0:72)
Va ¼ 0:33Gr0 (R2 ¼ 0:48)
Straw
Va ¼ 0:18Gr0 (Va < 0:02) (R2 ¼ 0:25)
Va ¼ 0:22Gr0 (Va < 0:0225) (R2 ¼ 0:36)
Va ¼ 0:29Gr0 (Va < 0:0275) (R2 ¼ 0:43)
Straw
Va ¼ 0:64Gr0 (Va > 0:02) (R2 ¼ 0:25)
Va ¼ 1:51Gr0 (Va > 0:0225) (R2 ¼ 0:36)
Va ¼ 2:08Gr0 (Va > 0:0275) (R2 ¼ 0:43)
Note: Va is expressed in mm s1 ; Gr0 is Gr 1010 and is expressed for a unit compost cross-sectional area in m2 .
parent air velocities decreased. The R2 values associated with the relationship between Gr and Va range between 0.25 and 0.72 because of the change in compost properties during the composting period, especially for the straw compost. For the wood shavings and hay compost, viscous forces increased in parallel with the buoyancy forces, as air velocity increased in the compost airflow channels. For the straw compost, at respective Gr numbers of 1.5,
S. Barrington et al. / Bioresource Technology 86 (2003) 259–266
2.0 and 2.5, for the low, mid and high MC, viscous forces became more important than buoyancy forces, and air flow rates decreased with increasing temperature differentials. This explained why the straw compost was the only material not showing peak airflow rates with peak compost temperatures. 3.3. Designing aeration ducts for passive composting The relationships developed can be used to predict the maximum airflow developing inside a compost mass and the size of ventilation ducts required. For example, let us consider a centre where swine manure is incorporated into hay to be composted inside a silo measuring 30 m in length, 2.4 m in width and 2.0 m in height. The compost mixture has an initial height of 1.8 m, a moisture content of 65%, a bulk density of 0.25 kg m3 and a porosity of 70%. Assume also that the expected maximum temperature is 65 °C. Three ventilation ducts will be installed under and over the length of the compost mass, at a spacing of 0.8 m. These ventilation ducts will receive air from the centre of the silo, to reduce the air pressure drop. Thus, each duct has a maximum length of 18 m, including the length to the outlet. From Eq. (1), the Gr is 0:45 1010 and the expected apparent air velocity across the compost surface is 0.22 mm s1 (Table 3). Thus, the ventilation ducts under the compost mass must offer an airflow of 8.0 L s1 . If three circular ducts are used under the mass, each 0.8 m apart, then each duct must provide 2.7 L s1 . To finish the design, the ventilation ducts must offer a cross-sectional area leading to a pressure drop over their length which is much lower than that developed by the buoyancy forces. Thus, the pressure drop in the aeration ducts will have no effect on passive airflow. Therefore, designing ducts for a pressure drop equal to 5% of that developed by the apparent air velocity across the compost mass is likely to respect this condition. Pressure drops occurring through the compost mass have been related by Barrington et al. (in press), as a function of apparent air velocity. Using such relationship, the expected pressure drop caused by an apparent air flow rate of 0.22 mm s1 , through the hay and swine manure compost column of 1.8 m, is expected to be of the order of 28.0 Pa. Thus, the pressure drop occurring inside the ventilation ducts should be limited to 5% of 28 Pa or 1.4 Pa, or 0.08 Pa m1 of duct length. A single duct with an inside diameter of 155 mm, offers a pressure drop of 0.01 Pa m1 , or 0.18 Pa at an air flow rate of 10 L s1 (ASHRAE, 2001). Thus, three air ducts each measuring 18 m in length and offering a smooth inside diameter of 155 mm will suffice in creating the required passive ventilation for the silo described above. From this calculation, duct pressure losses can be insignificant as compared to those developed from the viscous forces inside the compost mass.
265
The present calculations were conducted for a compost mass having a depth of 1.8 m while all tests were carried out using a compost depth of 0.8 m. Thus, the present research needs to be repeated for various depths of compost and using various types of bulking agents and wastes. Thus, a full range of relationships between Gr, apparent air velocities and the resulting air pressure drop could be measured.
4. Conclusions Passive aeration was just as effective as active aeration in bringing about stabilizing temperatures of 55 °C, in all three types of composts and at all three MC. Nevertheless, passive aeration produced peak temperatures of the order of 57–65 °C, while active aeration produced peak temperatures reaching 70 °C. Compost MC had an effect only on the peak compost temperature, occurring between days 2 and 6. Compost MC losses after 6 days lead to no difference in temperature regime between initial MC treatments and for the same bulking agent. Convective airflow rates expressed as apparent velocities across the compost mass, were related to the Grashof number representing the ratio between buoyancy and viscous forces. As opposed to the straw compost, the wood shavings and hay compost demonstrated viscous forces increasing in parallel with the buoyancy forces, as air velocity increased in the compost airflow channels. Thus, wood shavings and hay are better bulking agents, offering well formed airflow channels, as compared to straw. As a bulking agent, straw has a softer structure that tends to collapse when wet. The relationship established between apparent air velocity and the air pressure drop across the depth of compost mass provides a mean of designing passive ventilation systems under compost piles.
Acknowledgements This project was financially supported by the Natural Sciences and Engineering Research Council of Canada and the Macdonald Campus Farm of McGill University.
References APHA, WPCF and AWWA, 1995. Standard procedures for the analysis of water and wastewater. American Public Healthy Association, American Waste Water Works Association and Water Pollution Control Federation. Washington, DC. ASHRAE. 2001. Handbook of fundamentals. American Society of Heating, Refrigeration and Air Conditioning Engineering, Atlanta, Georgia, USA. (Chapter 32).
266
S. Barrington et al. / Bioresource Technology 86 (2003) 259–266
Barrington, S., Choiniere, D., Trigui, M., Knight, W. Compost air flow resistance. Journal of Agricultural Engineering Research, in press. Castellanos, J.Z., Pratt, P.F., 1981. Mineralization of manure nitrogen––correlation with laboratory indexes. Soil Science Society of America Journal 45, 354–357. Diaz, L.F., Savage, G.M., Eggerth, L.L., Golueke, C.G., 1993. Composting and Recycling. Lewis Publishers, Boca Raton, FL. Fernandes, L., Zhan, W., Patni, N.K., Jui, P.Y., 1994. Temperature distribution and variation in passively aerated static compost piles. Transactions of the ASAE 48, 257–263. Geankoplis, C.J., 1983. In: Transport Processes and Unit Operations, Second edition. Allyn and Bacon, inc, Boston, USA, pp. 244–246. Haug, R.T., 1993. The Practical Handbook of Composting Engineering. Lewis Publishers, Boca Raton, FL, USA. Lau, A.K., Lo, K.V., Liao, P.H., Yu, J.C., 1992. Aeration experiments for swine waste composting. Bioresource Technology 41, 145–152. Lynch, N.J., Cherry, R.S., 1996. Design of passively aerated compost piles: vertical air velocities between pipes. Biotechnology Progress 12 (5), 624–629.
Mathur, S., Patni, N.K., Levesque, M.P., 1990. Static pile passive aeration composting of manure slurries using peat as a bulking agent. Biological Wastes 34 (4), 323–334. McGarry, M.G., Strainforth, J., 1978. In: Compost fertilizer and biogas production for human and farm wastes in PeopleÕs Republic of China. International Development Research Centre, Ottawa, Canada, pp. 7–13. Patni, N.K., Kannagara, T., Nielsen, G., Dinel, H., 2001. Composting caged-layer manure in passively aerated and turned windrows. ASAE paper no. 012271. American Society of Agricultural Engineers, St. Joseph, Michigan, USA. Sartaj, M., Fernandes, L., Patni, N.K., 1997. Performance of forced, passive and natural aeration methods for composting manure slurries. Transactions of the ASAE 40 (2), 457–463. Steel, R.D.G., Torrie, J.H., 1986. Principles and Procedures of Statistics, a Biometrical Approach, Second ed. McGraw Hill Inc., New York. Zhan, W., Fernandes, L., Patni, N., 1992. Composting of poultry manure slurries. CSAE paper no. 92–515. Canadian Society of Agricultural Engineering, Saskatoon, Canada.