See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/296814369

Moisture relationships in composting processes

Article in Compost science & utilization · September 2002

CITATIONS READS
115 623

4 authors, including:

Tom L Richard H.V.M. Hamelers

Pennsylvania State University Wetsus
197 PUBLICATIONS 3,651 CITATIONS 218 PUBLICATIONS 14,708 CITATIONS

SEE PROFILE SEE PROFILE

Adrie Veeken
Attero
60 PUBLICATIONS 2,130 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Value from Urine View project

Novel methods for electrochemical capture and conversion of CO2 View project

All content following this page was uploaded by Adrie Veeken on 13 April 2016.

The user has requested enhancement of the downloaded file.

1372 Biotechnol. Prog. 2004, 20, 1372−1381

Air-Filled Porosity and Permeability Relationships during

Solid-State Fermentation
Tom L. Richard,*,† Adrie H. M. Veeken,‡ Vinnie de Wilde,‡ and
H. V. M. (Bert) Hamelers‡
Department of Agricultural and Biosystems Engineering, Iowa State University, Ames, Iowa 50011, and
Department of Environmental Technology, Wageningen University, 6700 EV Wageningen, The Netherlands

An experimental apparatus was constructed to measure the structural parameters of

organic porous media, i.,e. mechanical strength, air-filled porosity, air permeability,
and the Ergun particle size. These parameters are critical to the engineering of aerobic
bioconversion systems and were measured for a straw-manure mixture before and
after 13 days of in-vessel composting. Porosity was measured using air pycnometry
at four (day 0) and five (day 13) moisture levels, with each moisture level tested at a
range of different densities. Tested wet bulk densities varied with moisture level, but
dry bulk densities generally ranged from 100 to 200 kg m-3. At each moisture/density
combination, pressure drop was measured at airflow rates ranging from 0.001 to 0.05
m sec-1, representing the range of airflow rates found in both intensive and extensive
composting. Measured air-filled porosities were accurately predicted from measure-
ments of bulk density, moisture, and organic matter content. Reductions in air-filled
porosity at increasing moisture content were accompanied by an increase in perme-
ability, apparently due to aggregations of fines. This aggregation was quantified by
calculating an effective particle size from the Ergun permeability relationship, which
increased from 0.0002 m at 50% moisture to 0.0021 m at 79% moisture. The range of
airflow velocities reported in composting systems requires consideration of the second-
order drag force term, particularly at velocities approaching 0.05 m s-1 for the higher
moisture treatments tested. Calculated permeabilities for the matrix ranged from 10-10
to 10-7 m2, varying with both air-filled porosity and moisture. Mechanical strength
characterization provided a means to predict the effects of compaction on air-filled
porosity and permeability of porous media beds. The results of this investigation extend
porous media theory to the organic matrices common in solid-state fermentations and
help build a framework for quantitative and mechanistic engineering design.

Introduction substrate, which determine the size, shape, and con-

Solid-state fermentations include a variety of microbial
processes engineered for food and biotechnology applica-
tions (1), feed and forage preservation (2), and solid waste
treatment (3, 4). The gas phase is important in all these
systems, as evidenced by their fundamental categoriza-
tion as aerobic or anaerobic processes on the basis of
oxygen availability. In all anaerobic processes oxygen
must be excluded, while those that produce methane or
hydrogen need mechanisms for gas recovery. Airflow is
essential in aerobic processes, providing mass flow for
oxygen replenishment and moisture and carbon dioxide
removal and serving as a heat transfer fluid for temper-
ature management through energy redistribution and
heat removal. In aerobic systems regulation of airflow
thus provides the primary means to manage and control
biological activity (5). Gas transfer in all of these systems
is strongly affected by the physical properties of the
* To whom correspondence should be addressed. E-mail: tlr@
iastate.edu. Address after October 1, 2004: Department of Agri-
cultural and Biological Engineering, Pennsylvania State Univer-
sity, State College, PA 16802-1909. Phone: (814) 865-7792. Fax:
(814) 863-1031.
† Iowa State University.
‡ Wageningen University.
10.1021/bp0499505 CCC: $27.50 © 2004 American Chemical Society and American Institute of Chemical Engineers
Published on Web 07/31/2004
nectedness of a network of gas-filled pores. A better
understanding of the engineering properties of these
solid-state matrices is important for further advances in
process control and analysis.
Three of the physical parameters of particular impor-
tance are the air-filled porosity, permeability, and me-
chanical strength of the matrix. Air-filled porosity char-
acterizes the volume of gas in the system at any given
time, whose component gas concentrations will change
depending on the rate of biological activity. Replacement
or extraction of these gases typically requires convective
air movement, as diffusion is rarely sufficient in full-scale
biologically active systems (6). Permeability determines
airflow rates at different pressure gradients, and thus
gas transfer and heat exchange under both natural (7,
8) and forced convection conditions (5, 9). Air-filled
porosity and permeability relationships are different for
different substrates, as each substrate will have a dif-
ferent mix of particle sizes and shapes, and these
mixtures can be compacted to various bulk densities (10).
Mechanical strength determines the resistance of the
matrix to compaction, thus determining bulk density, air-
filled porosity, and permeability, as overburden stress
increases with bed depth.
Biotechnol. Prog., 2004, Vol. 20, No. 5
Although these engineering parameters and their
relationships with other matrix properties apply to all
of the solid-state fermentation systems previously men-
tioned, much of the recent effort to characterize these
parameters for organic matrices has focused on compost-
ing systems. These systems often include a wide variety
of feedstocks, including lignocellulosic plant material,
high protein bacterial and fungal biomass, and industrial
fats and grease (5). There are also diverse approaches to
process control, from intensive systems with mechanical
ventilation to more extensive systems that rely on
diffusion, natural convection, and wind for aeration (3,
5). The combination of diverse feedstocks operating under
a wide range of airflow velocities makes composting a
useful model system for analysis, as the results should
have broad applicability to other solid-state fermenta-
tions.
Air-filled porosity (a), also referred to in the compost-
ing literature as free air space (FAS), has been attracting
increasing interest as a key design parameter for com-
posting systems. In recent years researchers have re-
ported the relationship of air-filled porosity with moisture
(11), with packing density and state of decomposition (12,
13), and with moisture and density (14-16). The rela-
tionship between air-filled porosity and biodegradation
rate has also been examined by several investigators
(16-18) and incorporated in kinetic models of the com-
posting process (5, 19).
In full-scale composting systems adequate air-filled
porosity is necessary but not sufficient for good process
control, as air must also flow for the necessary heat and
mass transfer to occur. The resistance to air movement
increases at higher airflow rates and greater path lengths
(20) and is thus normally reported for a specific flow rate
and normalized for a standard path length. Resistance
to airflow can be either reported directly as pressure drop
under a set of specified conditions or transformed into a
permeability value using equations provided later in this
paper. Permeability is considered a property of the
material and thus applies within a much broader range
of flow conditions. However, some caution must be used
in both the calculation and use of permeability values
depending on whether one is operating in laminar,
intermediate, or turbulent flow regimes, as in the latter
two situations it is necessary to consider form drag force
effects (21, 22). Permeability is also only constant when
the porous media remains constant. In solid-state fer-
mentation, important changes in the permeability of the
porous media will occur as a result of biomass growth,
biodegradation of the organic particles, changing mois-
ture content, and compaction. Biomass growth can have
a particularly strong effect in processes that involve fungi
(23), whose hyphae have been shown to fill up to 34% of
the initial air-filled porosity (24).
The effects of compaction on porosity and permeability
have important implications for the design and operation
of solid-state fermentation processes. Current design
procedures generally use rules of thumb or empirical
relationships that require extensive measurement of each
parameter (15, 20). The investigation reported here
examines these relationships by applying porous media
theory to data from a new experimental apparatus. The
apparatus allows measurement of porosity and perme-
ability of a single substrate under various applied
stresses and resulting bulk densities. By varying the
moisture content of a straw-manure substrate before
and after composting, the effects of these variables on
porosity and permeability can readily be observed. The
objective of the study is to develop a better understanding
1373
of the relationships between compaction, porosity, and
permeability, which can provide the framework for more
comprehensive engineering analysis. Such a framework
can address critical design issues such as the effect of
bed depth and moisture on gas transfer phenomena,
blower sizing, and required aeration frequency.
Theoretical Relationships in a Porous Organic
Matrix
Porosity. Air-filled porosity, a, is the volume fraction
of air (usually reported on a percentage basis) in a porous
matrix and can be defined by the following equation:
Vg
a ) (1)
Vg + Vw + Vs
where Vg, Vw, and Vs are the volumes of gas, liquid, and
solids in the matrix respectively, which sum to the total
volume, Vt. Solid-state fermentations can include both
absorbed and free water, both of which are considered
part of the liquid phase Vw in our analysis, facilitating
conversion from mass to volume units since the density
of water is well-known. This conversion assumes the
mass of dissolved compounds in this liquid solution is
small, which is a reasonable approximation for compost-
ing but may not hold in all solid-state fermentation
systems. Conversion of the solids mass to the volume Vs
is accomplished using the measured gravimetric dry
matter content (DM) and solid-phase particle densities.
To facilitate these conversions we divide the dry matter
content into two fractions, organic matter (OM) and ash
(1 - OM), differentiated by combustion at 550 °C. On a
decimal basis, the gravimetric water content is 1 - DM.
Air-filled porosity can then be calculated directly on a
theoretical basis from easily measured values of DM and
OM, if the material densities of water, organic, and ash
(Fw, Fom, and Fash) and the total bulk density are known
(12):
( )
(1 - DM) DM‚OM DM(1 - OM)
a ) 1 - Fwb + +
Fw Fom Fash
(2)
In eq 2 the total bulk density (Fwb) and dry matter (DM)
are on a wet basis (w.b.) while organic matter (OM) is
on a dry basis (d.b.).
Using eq 2, a first-order approximation of the air-filled
porosity and maximum bulk density can be made by
assuming the organic fraction (OM) has a particle density
of 1.6 × 103 kg m-3 and the inorganic fraction (1 - OM)
has a density of 2.5 × 103 kg m-3 (12, 16, 25).
Air-filled porosity can also be measured directly using
a gas pycnometer, whose measurement principle is based
on the pressure-volume relationships of Boyle’s law. Gas
pycnometers have been extensively used in measuring
soil porosity (26-28) and have also been applied to other
porous materials (29-31). In this study we compared this
direct measurement technique with theoretical approxi-
mation described above.
Permeability. Permeability is a measure of the ability
of fluids to flow through a multiphase material. For
laminar flow situations Darcy’s law (10) relates the
change in pressure (P) across a matrix to distance (x),
superficial velocity (v), and viscosity (µ) by defining a
matrix permeability (κ), as indicated in eq 3:
κ dP
v)- ( )
µ dx
(3)
1374
Where flow is not fully laminar, pressure drop is
usually expressed with a Dupuit-Forcheimer relation-
ship (10, 22, 32):
dP
- ) Rv + βv2 (4)
dx
This general second-order equation includes a first-
order viscous force term that incorporates Darcy’s perme-
ability, while the second order form drag force term
incorporates the density of air (Fa) and a parameter η
sometimes called the passability (33):
dP µ Fa
- ) v + v2 (5)
dx κ η
The dimensionless Reynolds number is often used to
differentiate the relative importance of these first- and
second-order terms in various flow regimes. However,
Lage and Antohe (22) point out that this transition can
be better identified directly from eq 5, using the ratio of
the viscous and form drag forces (L):
Fa 2
v Faκ
η
L) ) v (6)
µ µη
v compaction is an exponential function of the ratio of
κ applied stress (σ) to the mechanical strength (M) of the
Thus when L ≈ 1 or L > 1 both terms are important,
whereas when L , 1 viscous forces dominate and the
simpler eq 3 can be used. This is similar to the criteria
used with the Reynolds number, where values <1 are
used as an indication that viscous forces are dominant
(10, 21, 34). At the other extreme, where L . 1, the
second-order form drag force term would dominate, as
may occur under high airflow, low permeability situations
such as compacted or saturated solid-state fermentations.
Although some previous investigations of permeability
in composting systems neglected the form drag term,
particularly when examining passive aeration systems
(8, 35), other results indicate second-order terms can be
important (5, 9, 20, 36-39). In this study we compare L
to the permeability-based Reynolds number (Reκ) often
used for porous media analysis (8, 21). Reκ is defined as
Favκ1/2/µ . The Reynolds number can also be calculated
on the basis of particle size, sometimes adjusted for
porosity in the form Re ) Favdp/[µ (1 - R)], which we
use as Ergun (40) did to report the dimensionless
resistance to airflow.
Ergun (40) related the coefficients κ and η to fluid,
matrix, and particle properties, with two constants A and
B that appear applicable over a wide range of materials
and flowrates. Equation 7 results from the Kozeny-
Carmen analysis (41), which predicts permeability as a
function of air-filled porosity and particle size.
dp2 a3
κ) ‚ (7)
A (1 -  )2
a
dp a3
η) ‚ (8)
B (1 - a)
In Ergun’s original analysis A and B were found to be
150 and 1.75, respectively. Using additional experimental
data, Macdonald et al. (42) found a better fit for A of 180,
with B estimated at 1.8 for smooth particles and 4.0 for
rough particles. Macdonald’s estimate of A has been
Biotechnol. Prog., 2004, Vol. 20, No. 5
collaborated by more recent investigators, who estimated
it at 181-182 (33, 34). The particle diameter (dp) is an
effective particle diameter, defined as the weighted
average surface-to-volume ratio (42).
By relating pressure drop and permeability to physical
characteristics, the Ergun equation and related theoreti-
cal relationships can be used as a basis for analyzing heat
and mass transfer in porous media systems. Although
this approach has not to our knowledge been incorporated
in composting system models, this framework has been
used to model heat and mass transfer in other biologically
active porous media where air velocities require consid-
eration of the drag force term (7, 43-45).
Compaction. Compression affects both air-filled po-
rosity and permeability of a biological matrix such as
compost by reducing or ultimately eliminating air chan-
nels between the solid particles. This occurs when exces-
sive external force is applied to a composting pile or when
the pile itself becomes so large that the overburden
compresses the material at the base (12, 20). In this
analysis we have adapted Terzaghi’s theory of primary
compression (46, 47), using wet bulk density instead of
pile height as the independent variable, which allows the
calculation of a maximum bulk density as indicated
below. We then define a mechanical strength term (the
resistance to deformation), so that the resistance to
material. The bulk density under any particular applied
stress (Fwb,σi) can therefore be calculated as a function of
the minimum (σ ) 0) and maximum (a ) 0) bulk
densities (Fwb,0 and Fwb,max, respectively) and the mechan-
ical strength M:
Fwb,σi ) Fwb,max - (Fwb,max - Fwb,0) exp (-σ
M)
(9)
The maximum bulk density can be extrapolated from
experimental results under extreme compaction. Alter-
natively, a theoretical maximum bulk density can be
calculated from material characteristics by rearranging
eq 2 and setting a equal to zero:
1
( )
Fwb,max ) (10)
(1 - DM) DM‚OM DM(1 - OM)
+ +
Fw Fom Fash
The minimum bulk density (Fwb,0) and mechanical
stress can then be obtained from eq 9 by nonlinear
parameter estimation, using experimental observations
of substrate bulk densities at various applied stresses.
The results presented below were estimated using the
Solver function in Microsoft Excel (version 2000).
Materials and Methods
In this study we use a new experimental apparatus to
measure compaction, air-filled porosity, and permeability
on a single composting sample, applying to these mea-
surements the theoretical framework of porous media
analysis described above. With this experimental ap-
paratus the same sample can be subjected to varying
degrees of compaction without repacking, allowing
straightforward investigations of the effect of compaction
on porosity and permeability without the complications
of reproducibility encountered when repacking is re-
quired. Previous investigations of the effect of compaction
on compost matrices have examined either permeability
Biotechnol. Prog., 2004, Vol. 20, No. 5 1375

Table 1. Characteristics of Fresh and Composted Manure-Straw Mixtures

sampling day, moisture %
0, 79% 0, 77% 0, 74% 0, 69% 13, 77% 13, 73% 13, 69% 13, 58% 13, 50%
total solids (kg kg-1) 0.207 0.232 0.260 0.307 0.225 0.270 0.313 0.421 0.498
organic matter (kg kg-1) 0.852 0.852 0.852 0.852 0.816 0.816 0.826 0.818 0.829
effective particle size (m) 0.00210 0.00147 0.00116 0.00095 0.00214 0.00095 0.00060 0.00025 0.00022
initial wet bulk density (kg m-3) 488 545 416 576 577 277 262 203 194
maximum wet bulk density (kg m-3) 1108 1087 1137 1171 1158 1118 1158 1115 1058
mechanical strength 8437 9577 12917 NA 11 064 26 212 28 809 81 957 44 398
(wet bulk density basis) (m2 N-1)

(12, 20) or air-filled porosity (14, 15) but not both on the
same samples.
Substrate Sample Preparation. The substrate tested
in these trials was a swine manure-straw mixture. The
swine manure was fecal solids collected from a belt-
separator (48), which included a small fraction of straw
bedding (16% of manure dry weight). An additional 61
kg of chopped straw was combined with 777 kg of manure
on a belt conveyor and conveyed through a finger flail
twice to achieve thorough mixing and an initial moisture
content of 69% (wet basis). In all cases in this paper
moisture is reported on a wet basis (w.b.) and was
determined by drying in a 105 °C oven to constant weight
(49). A 650 kg portion of the manure-straw mixture was
remixed with 217 kg of additional water to a moisture
content of 77% and composted for 13 days, during which
time the moisture content dropped to 50%. Organic
matter content decreased from 0.852 ( 0.006 gOM/gDM
to 0.821 ( 0.006 gOM/gDM, as measured by ashing at
550 °C (50). This concentration reduction indicated that
17.9% of the organic matter mass was converted during
the 13 days, as calculated using a constant ash assump-
tion (51). Samples at a range of different moisture
contents were then generated from both the 69% mois-
ture fresh mixture and the 50% moisture composted
mixture by adding additional water to 30-50 kg samples,
soaking in plastic tubs until the moisture was absorbed,
and remixing through the same conveyor apparatus. For
the fresh (day 0) mixture the resulting moisture range
was 69%, 74%, 77%, and 79% moisture (21-31% DM),
Figure 1. Schematic representation of the CPP measurement
whereas for the partially composted (day 13) mix the device (numbers are explained in the text).
resulting range was 50%, 58%, 69%, 73%, and 77%
moisture (23-50% DM). Dry matter and organic matter controller [7] (Brooks Instruments, model 5853S, The
content for each of these samples is presented in Table Netherlands). A load cell [8] (STS 500 C2, Sun Scale Inc,
1. Testing each material in the apparatus required Taichung, Taiwan) on the compression screw indicated
several weeks, so to minimize changes in the fresh or compaction pressure on the compost bed.
composted materials the samples were sealed in plastic Each sample was loaded in the apparatus and com-
bags and stored frozen at -20 °C and then thawed as pacted to six or seven different bulk densities. The
needed with a maximum of 2 weeks refrigerated storage starting measurements were on a sample that nearly
at 5 °C. filled the chamber under minimal compaction, and the
Experimental Setup. The apparatus developed for sample was compressed after each set of porosity and
characterization of the compaction/porosity/permeability pressure drop measurements until the screw completed
relationships combines features of an air pycnometer (15, its travel, which typically occurred at a final applied
16) with those of an air permeameter (20, 52) and is stress of 60-100 kg m-2. The initial dry bulk density
illustrated in Figure 1. The 45-L sample chamber [1] ranged from 81 to 113 kg m-3, and the final dry bulk
includes a perforated bottom plate over an air inlet densities ranged from 151 to 205 kg m-3. On a wet basis,
plenum [2] and a perforated compression plate under a these same bulk densities ranged from initial values of
top air outlet plenum [3]. The depth of the compression 182-563 kg m-3 to final values of 352-918 kg m-3. Any
plate can be adjusted by means of an airtight screw [4] drainage water that was produced during compaction was
to increase the density of the mixture at a known volume removed after each compaction cycle by a drain valve [9]
(VS). Although the apparatus is designed to allow heating (Figure 1).
of the test chambers, in this study all measurements were Air-Filled Porosity Measurement. For free air
made at temperatures between 20 and 24 °C. Pressure space measurements a compressed air chamber (VB) [10]
gauges were used to measure pressure drop across the (Figure 1) was pressurized, the sample chamber inlet and
compost bed [5] and the change in pressure during outlet closed, and free air space calculated from the
porosity measurements [6] (Testo GmbH, models 525 and difference in pressure observed in the compressed air
520, respectively, Lenzkirch, Germany). Airflow was chamber when the valve between it and the sample
controlled and monitored using a 0-15 m3 s-1 mass flow chamber is opened. For a pycnometer with headspace
1376
above the sample within the sample chamber, Boyle’s law
can be used to derive the general equation for the
pressure volume relationships under these two pressure
regimes:
VBPi ) [VA + VB - VS(1 - a)]Pf (11)
where Pi is the initial pressure in the compressed air
chamber (VB), VA is the sample chamber (initially at
atmospheric pressure), Pf is the final pressure in the
connected compressed air and sample chamber, and VS
is the volume of the sample within the sample chamber
A (26, 14). The air-filled porosity, a, is then given by
a )
( )
PiVB
Pf
- VA - VB + VS
(12)
VS
The pycnometer was calibrated by adding known
amounts of water (VS) in approximately 2 L increments,
with the exact volume determined by weight, and then
measuring initial and final pressures as indicated above.
By plotting Pi/Pf versus VS, VB can be determined directly
from the slope and VA from the intercept, which were
done using a least-squares fit of the calibration measure-
ments to the following equation:
Pi 1 (VA + VB)
) - VS + (13)
Pf VB VB
For each compost sample and at each moisture/density
combination, porosity was measured six times: three
before and three after the pressure drop measurements.
Average ratios of Pi/Pf from these six measurements were
used to calculate air-filled porosity using eq 12.
Organic matter and ash content were measured in
triplicate (49, 50), allowing comparison of the measured
air-filled porosity values with the theoretical values from
eq 2.
Permeability Measurement. The superficial velocity
of airflow in composting systems can vary over several
orders of magnitude. In extensive composting systems
the flow velocities can be quite low, with Fogiel et al. (53)
reporting velocities ranging from 0.0002 to 0.0011 m s-1,
whereas Veeken et al. (54) reported velocities ranging
from 0.002 to 0.018 m s-1 in naturally ventilated pilot-
scale static piles. In more intensive, mechanically ven-
tilated systems, velocities can be considerably higher. In
two full-scale agitated bed sewage sludge composting
systems, measured velocities ranged from 0.008 to 0.02
m s-1 (55).
In this study, pressure drop for each sample at each
moisture/density combination was measured at seven
superficial velocities, ranging from 0.001 to 0.05 m s-1.
Pressure drop stabilized within a few seconds after
adjusting flowrates using the mass flow controller, which
had been calibrated prior to the start of the experiments.
Compaction Measurement. Applied stress on the
compost sample was measured at the start of each set of
porosity and permeability measurements at a particular
bulk density and again at the completion of those porosity
and permeability measurements. The semi-elastic nature
of the composting material deformed to absorb some of
the applied stress, so that load cell readings dropped
during the initial part of the 25 to 45 min required to
make the porosity and permeability measurements.
Agnew and Leonard (15) report that under conditions of
constant applied stress (and increasing bulk density), the
Biotechnol. Prog., 2004, Vol. 20, No. 5
Figure 2. Air-filled porosity as a function of wet bulk density.
Percent values in the symbol legends refer to percent moisture
in this and all subsequent figures and tables.
time required to reach equilibrium ranged from 1 h to 3
days. Under our experimental conditions, with applied
stress decreasing under conditions of constant bulk
density, equilibrium was more rapidly achieved, and the
load cell readings were deemed stable by the end of the
25-45 min permeability measurement period.
Results and Discussion
Air-Filled Porosity. Measured air-filled porosity data
at various measured wet bulk densities are presented in
Figure 2. The linear nature of this relationship has been
noted by previous investigators, some basing their analy-
sis on eq 2 (16), whereas others derived empirical
relationships (14, 15). Comparing data for a variety of
organic feedstocks and matrices, Agnew and Leonard (15)
proposed a general relationship of a ) 1.0-0.0009Fwb,
which is shown as the solid line in Figure 2. Our
experimental data and theoretical predictions of a from
eq 2 fit this relationship well, with intercepts of 1.014 (
0.031 and 0.997 ( 0.002 and slopes of -0.00090 (
0.00006 and -0.00087 ( 0.00004, respectively. However,
eq 2 indicates a significant effect of dry matter content
on this relationship, which is not evident in the empirical
analysis. The dashed lines in Figure 2 show this effect
for a hypothetical compost with an OM concentration of
80% (d.b.) and DM contents of 20% and 50% (w.b.).
Although the relationship predicted by eq 2 for a DM
content of 20% is consistent with Agnew and Leonard’s
equation, a significant deviation is predicted as solids
content increases, especially at high bulk densities. Thus
Agnew and Leonard’s equation (15) should only be
applied in the high-moisture and/or low-density region
where their relationship appears applicable.
The theoretical values of air-filled porosity calculated
from eq 2 correlated well with measured values for all
the conditions in this study, with r2 values of 0.988 and
0.987 for days 0 and 13, respectively (Figure 3). Although
there is likely to be some variation in the particle
densities of different composts, the assumed component
densities of 1.6 × 103 kg m-3 for the organic fraction and
2.5 × 103 kg m-3 for the inorganic fraction appear
accurate within the range of measurement error of this
experimental system. Equation 2 thus appears to provide
an accurate means of predicting air-filled porosity for the
full range of solid-state fermentations, subject to verifica-
tion of these component densities for other substrate
matrices.
Pressure Drop and Permeability. Pressure drop
increased with respect to both superficial velocity and
Biotechnol. Prog., 2004, Vol. 20, No. 5 1377

Figure 3. Correlations between theoretical and experimental

(measured) air-filled porosity for days 0 and 13.
increasing bulk density, as illustrated for the day 0, 69%
moisture condition in Figure 4. This relationship was
linear in the lower velocity range, where eq 3 was
generally valid, but as velocities approached 0.5 m s-1,
the form drag influence became evident. In this transi-
tional region it is important to evaluate whether second-
order terms are needed, which we did by calculated the
Lage number L, based on eq 6 (22), as well as the
permeability and porosity based Reynolds numbers (Reκ
and Re, respectively) as previously defined. Because the
Lage number was seen to sometimes exceed 1 under our
experimental conditions, a second-order polynomial least
squares curve fit was used to estimate the R parameter
in eq 4, which was then used to calculate the permeability
κ as indicated in eq 5. For each sampling date and
moisture treatment, the range of results for R, β, L, Reκ,
and Re are presented in Table 2. R2 values are from the
second-order polynomial curve fit of R and β.
As expected, the high end of the range for the dimen-
sionless numbers occurred at the higher velocities tested.
At 0.5 m s-1 the Lage number L exceeded 1 for both day
0 and day 13 samples under high moisture conditions,
while the porosity-based Reynolds number Re was
consistently above 1 and also increased with moisture.
In contrast, Reκ was relatively constant with respect to
moisture, with values consistently less than 1. These
results are consistent with Lage and Antohe’s analysis
and demonstrate the advantage of their approach over
the use of Reκ to evaluate flow regimes in this transitional
region.
Permeability values for selected experimental condi-
tions are indicated in Figure 5. Measured data points for
four of the samples are illustrated, along with the
theoretical curve from eq 7 using the average dp calcu-
lated as described below. Although our experimental data
coincides with the relatively straight central section of
the curves, Ergun’s equation indicates more rapid changes
will occur at the extremes of the porosity range. Perme-
ability increased with increasing air-filled porosity as well
Figure 4. Pressure drop as a function of superficial velocity
at different wet bulk densities for day 0, 69% moisture.
Figure 5. Calculated permeability values for various air-filled
porosities for the different days and sample moisture contents.
Lines indicate theoretical permeabilities across the full range
of air-filled porosities, as predicted by eq 7 based on the effective
particle size in Table 1.
as with increasing moisture content of the matrix.
Because of the inverse relationship between air-filled
porosity and wet bulk density (eq 2), a corollary to this
result is that permeability increases with decreasing wet
bulk density. The increase of permeability with increas-
ing air-filled porosity and decreasing bulk density is
intuitive, with increasing air volume facilitating air
movement, but the effect of moisture warrants additional
discussion.
Holding air-filled porosity constant, increases in volu-
metric moisture content correspond to a decrease in the
solids volume. Because the organic particles are hydro-
philic, capillary forces draw this water into the fine pores,
effectively combining small particles into larger, smoother
aggregates, each composed of both solid and liquid
phases. The increasing size of particle aggregates is
reflected in a corollary increase in pore sizes as well. Thus
for any particular air-filled porosity, the resulting in-
creasing proportion of large pores increases the perme-
ability of the matrix. This phenomenon has previously

Table 2. Range of Matrix Transport Parameters for Each Sample Material

Dupuit-Forcheimer parameters and fit dimensionless numbers
expt R β R2 L Reκ Re
day 0, 69% 883 to 10 144 12 665 to 175 460 0.9994 to 0.9999 0.086 to 0.22 0.003 to 0.51 0.1 to 11
day 0, 74% 1186 to 42 371 15 281 to 230 369 0.9981 to 0.9994 0.014 to 0.10 0.002 to 0.36 0.1 to 12
day 0, 77% 2314 to 16 020 6132 to 223 503 0.9985 to 0.9998 0.003 to 3.70 0.001 to 0.31 0.15 to 10
day 0, 79% 1142 to 12 226 21 781 to 499 406 0.9997 to 0.9999 0.016 to 6.17 0.001 to 0.45 0.2 to 15
day 13, 50% 1419 to 58 748 NA 0.9982 to 0.9995 na 0.001 to 0.40 0.03 to 6
day 13, 58% 1195 to 33 641 5244 to 27 150 0.9987 to 0.9999 0.001 to 0.22 0.002 to 0.44 0.03 to 7
day 13, 69% 1063 to 28 783 970 to 49 196 0.9973 to 0.9996 0.001 to 0.09 0.002 to 0.46 0.06 to 9
day 13, 73% 786 to 20 655 4193 to 105 590 0.9976 to 0.9998 0.031 to 0.06 0.002 to 0.40 0.07 to 12
day 13, 77% 826 to 26 700 6334 to 602 497 0.9982 to 0.9998 0.007 to 1.13 0.002 to 0.53 0.09 to 15
1378
Figure 6. The relationship between dry matter and effective
particle size for day 0 and day 13 samples. The curve is fit to
the combined data set; error bars indicate standard deviations
from mean values.
been observed in other biological materials including
biomass (56) and grain (57). Although this study did not
include measurement of particle or pore size distribu-
tions, an effective particle size can be estimated from eq
7 for each combination of air-filled porosity and perme-
ability data. Figure 6 plots the effective particle size
calculated for each moisture level tested, with the mean
values also presented in Table 1. This effective particle
size increases dramatically near saturation as small
pores fill due to capillary forces, partially compensating
for the effect of reduced air-filled porosity at high
moisture levels.
It is also possible to define a dimensionless resistance
to airflow φ, which can be used to evaluate the consis-
tency of the Ergun equation with the experimental
results (40, 58).
dP dp 3
φ) ‚ 2‚ (14)
dx Fv 1 - 
Figure 7 illustrates φ as a function of Re/(1 - ), with
experimental data from all samples shown. A linear least-
squares fit to this log-transformed data results in the
equation ln(φ) ) 8695 ln[Re/(1 - )], with an r2 value of
0.977. The deviation on the lower right is from the day
0, 79% moisture sample, which had the highest Lage
number. This result is consistent with theory, as the
Ergun equation predicts a horizontal deviation of this
curve as the drag force term begins to dominate (40).
Permeability has a significant impact on the costs of
aeration in solid-state fermentation systems (38), affect-
ing both capital costs in blower sizing as well as operating
power requirements. Thus the ability to predict perme-
ability as a function of matrix and fluid characteristics
can provide important opportunities for more efficient
engineering design.
Mechanical Strength and Compaction. The bed
depth of solid-state fermentation reactors is often limited
by excessive compaction, which can reduce pore space and
restrict permeability. The effects of applied stress on bulk
density of the compost samples are illustrated in Figure
8. Data points indicate the wet bulk density and stress
after equilibration. Solid and dashed lines illustrate the
predicted wet bulk densities at each stress using the
estimated values for mechanical strength (M). Table 1
summarizes the initial and maximum wet bulk densities
and mechanical strength parameters for each sample.
The maximum wet bulk densities were determined from
the x-axis intercepts (a ) 0) extrapolated from the
experimental air-filled porosity/wet bulk density data
illustrated in Figure 2. The initial wet bulk densities and
Biotechnol. Prog., 2004, Vol. 20, No. 5
Figure 7. Comparison of the modified Ergun equation with
all experimental data.
Figure 8. Effect of applied stress on wet bulk density. Points
are experimental data, and lines indicate curves fit to eq 9 using
the mechanical strength values in Table 1.
mechanical strength parameters were then obtained by
nonlinear parameter estimation using eq 9.
There was a significant increase in mechanical strength
with increasing dry matter for the day 0 samples, with
the linear regression having an r2 of 0.94. A similar trend
was evident for the day 13 samples, although here the
regression r2 was only 0.52 due to the high dry matter
treatment results. The day 13, 58% moisture sample
peaked at about 82 000 m2 N-1, over 7 times the
magnitude of the day 13 high moisture treatment.
However, the even drier day 13, 50% moisture sample
was only 44 000 m2 N-1. It is not clear whether one of
these two low moisture points is an outlier, or if there is
a local maximum for mechanical strength, since our day
0 treatments were not tested in this drier range. Com-
posting for 13 days did not appear to affect the mechan-
ical strength of the substrate relative to the day 0 results,
although the number of datapoints are limited.
The increased mechanical strength observed with
increasing dry matter is consistent with other evidence
that lignocellulosic material softens and weakens as it
gets wet (59). For solid-state fermentations, this means
that drier matrices have considerably greater structural
stability, resisting compaction and maintaining porosity
and permeability much better than they do when wet.
Conclusions
A combination of experimental measurement and
application of porous media theory can be used to develop
an analytical framework for the gas transfer phenomena
associated with solid-state fermentation. The air-filled
porosity of composting materials can be predicted from
relatively simple measurements of bulk density, mois-
Biotechnol. Prog., 2004, Vol. 20, No. 5 1379

ture, and organic matter content. With one set of mea- design parameters from relatively simple experimental
surements, the effects of changing moisture and packing results. Future research will apply these methods to
density on air-filled porosity follow well-defined relation- practical design issues of bed configuration and depth
ships. constraints, matrix blending requirements, critical design
Interestingly, the reduction in air-filled porosity that constrains for naturally ventilated systems, and efficient
accompanies increasing moisture content coincides with design of mechanical aeration systems.
an increase in permeability, up to points very near
saturation. This appears to result from changes in Acknowledgment
particle and pore size distributions as moisture facilitates This work was partially supported by Iowa State
the aggregation of fines into fewer larger particles with University and the Dutch E.E.T. (Economy, Ecology and
corresponding larger intraparticle pores. This aggrega- Technology) program from the Ministry of Economic
tion was quantified by calculating an effective particle Affairs, the Ministry of Education, Culture and Sciences,
size from the Ergun permeability relationship, which and the Ministry of Housing, Spatial Planning and the
increased exponentially as moisture increased near satu- Environment.
ration. Under the tested conditions, this effective particle
size increased a full order of magnitude from 0.0002 m Notation
at 50% moisture to 0.0021 m at 79% moisture. A Ergun viscous component constant, dimension-
The effect of moisture on particle and pore size less
distributions has clear physical impacts on permeability B Ergun inertial component constant, dimension-
and may have important biological implications as well. less
The deductive model of biodegradation kinetics proposed db dry basis, presentation of moisture, dry matter,
by Hamelers (19) implies an important role of particle organic matter and ash fractions, kg kg-1
and pore size distributions on biological activity. Under- dp particle diameter, m
standing the physical mechanisms by which moisture DM dry matter fraction, dry basis, kg kg-1
affects solid-state fermentation kinetics can provide
L Lage number, dimensionless
important insights for process design and control (16).
Within the range of airflow velocities reported in M mechanical strength, m2 N-1
composting systems, this study indicates that several OM organic matter fraction, dry basis, kg kg-1
scenarios require consideration of the second order form P pressure, Pa
drag force term. Form drag becomes significant at veloci- Pf final pressure in the connected compressed air
ties approaching 0.05 m s-1, especially under high and sample chambers, Pa
moisture conditions. The dimensionless analysis proposed Pi initial pressure in the compressed air chamber,
by Lage and Antohe (22) provides a direct assessment of Pa
the importance of this second-order term. Permeability Re porosity based Reynolds number, dimensionless
values were between 10-10 and 10-7 m2 for the range of Reκ permeability based Reynolds number, dimen-
matrix conditions tested. sionless
Permeability data plotted using Ergun’s dimensionless v superficial velocity, m s-1
resistance to airflow φ generates a linear relationship on Vg volume of gas in a sample or unit volume of
a log-log plot, permitting estimation of permeability sample, m3
values based on velocity and effective particle size. Vw volume of water in a sample or unit volume of
Although these relationships need to be tested with other sample, m3
organic matrices, this method offers important potential Vs volume of solids in a sample or unit volume of
for aeration system design. sample, m3
This study demonstrated the key role of the effective Vt total volume in a sample or unit volume of
particle size parameter in advanced engineering design sample, m3
of solid-state fermentation systems. Further research is VA volume of sample chamber, m3
needed to develop accurate methods to estimate this
VB volume of compressed air chamber, m3
value from particle size distributions or other simple
tests, without extensive experimental work. Possible VS volume of the sample within the sample cham-
ber, m3
approaches include correlations with particle size distri-
butions (10), using tracer gases to generate breakthrough wb wet basis, presentation of moisture content, kg
curves for displacement analysis (60), and new nonde- kg-1
structive technologies such as nuclear magnetic reso- x depth of bed, m
nance. R Forscheimer viscous term coefficient, kg m-3 s-1
Mechanical strength characterization provides a means β Forscheimer inertial term coefficient, kg m-2
to predict the effects of compaction on air-filled porosity a air-filled porosity of compost matrix, m3 m-3
and permeability of porous media beds. Compaction is φ resistance to airflow, dimensionless
one of the most critical design parameters in full-scale η passibility, m
solid-state fermentations, whether to ensure air move- κ permeability, m2
ment in aerobic systems or to limit it in anaerobic
µ kinematic viscosity, Pa s
designs. Increased moisture appears to reduce the me-
chanical strength of biological matrices, although detailed Fa density of air, kg m-3
understanding of this moisture effect as well as the Fash particle density of mineral fraction, kg m-3
impact of substrate degradation will require additional Fom particle density of organic matter, kg m-3
research. Fw density of water, kg m-3
The results of this investigation extend theoretical Fwb total bulk density of organic matter, wet basis,
porous media analysis to the organic matrices common kg m-3
in solid-state fermentations. The theoretical relationships Fwb,max maximum total bulk density ) density at zero
have been shown to apply and allow prediction of key porosity, wet basis, kg m-3
1380
Fwb,0 total bulk density at zero stress, wet basis, kg
m-3
Fwb,σi total bulk density at any particular stress σi, wet
basis, kg m-3
σ applied stress, N m-2
References and Notes
(1) Doelle, H. W.; Mitchell, D. A.; Rolz, C. E. In Solid Substrate
Cultivation; Elsevier Applied Science: London, 1992.
(2) McDonald, P.; Henderson, A. R.; Heron, S. J. E. In The
Biochemistry of Silage, 2nd ed.; Chalcombe Publications:
Marlow, Bucks, UK, 1991.
(3) Richard, T. L. Municipal solid waste composting: physical
and biological processing. Biomass Bioenergy 1992, 3 (3-4),
163-180.
(4) O’Keefe, D. M.; Chynoweth, D. P. Influence of phase
separation, leachate recycle, and aeration on treatment of
municipal solid waste in simulated landfill cells. Bioresour.
Technol. 2000, 72 (1), 55-66.
(5) Haug, R. T. The Practical Handbook of Compost Engineer-
ing; Lewis Publishers: Boca Raton, FL, 1993.
(6) Nakasaki, K.; Hakano, Y.; Akiyama, T.; Shoda, J.; Kubota,
H. Oxygen diffusion and microbial activity in the composting
of dehydrated sewage sludge cakes. J. Ferment. Technol.
1987, 56 (1), 43-48.
(7) Beukema, K. J.; Bruin, S.; Schenk, J. Three-dimensional
natural convection in a confined porous medium with internal
heat generation. Intl. J. Heat Mass Transfer 1983, 26 (3),
451-458.
(8) Lynch, N. J.; Cherry, R. S. Design of passively aerated
compost piles: vertical air velocities between the pipes.
Biotechnol. Prog. 1996, 12, 624-629.
(9) Keener, H. M.; Marugg, C.; Hansen, R. C.; Hoitink, H. A. J.
Optimizing the efficiency of the composting process. Proceed-
ings of the International Composting Research Symposium;
Hoitink, H. A. J., Keener, H., Eds.; Renaissance Publica-
tions: Columbus, OH, 1992; 59-94.
(10) Bear, J. Dynamics of Fluids in Porous Media; Elsevier:
New York, 1972.
(11) Oppenheimer, J.; Martin, A. G.; Walker, L. P. Measure-
ments of air-filled porosity in unsaturated organic matrices
using a pycnometer. Bioresour. Technol. 1997, 59, 241-247.
(12) van Ginkel, G. T.; Raats, P. A. C.; van Haneghem, I. A.
Bulk density and porosity distributions in a compost pile.
Neth. J. Agric. Sci. 1999, 47, 105-121.
(13) Eftoda, G.; McCartney, D. Correlating bench-scale FAS
tests to full-scale windrow composting performance. In 2002
International Symposium on Composting and Compost Uti-
lization; Michel, F. C., Rynk, R., Hoitink, H. A. J., Eds; J. G.
Press: Emmaus, PA, 2002.
(14) Baker, S. M.; Richard, T. L.; Zhang, Z.; Monteiro da Rocha,
S. Determining the free air space inside compost mixtures
using a gas pycnometer. Proceedings of the International
Meeting of the American Society of Agricultural Engineers;
ASAE: St. Joseph, MI, 1998; ASAE paper no. 984094.
(15) Agnew, J. M.; Leonard, J. J. Using a modified pycnometer
to determine free air space and bulk density of compost
mixtures while simulating compressive loading. In 2002
International Symposium on Composting and Compost Uti-
lization; Michel, F. C., Rynk, R., Hoitink, H. A. J., Eds; J. G.
Press: Emmaus, PA, 2002.
(16) Richard, T. L.; Hamelers, H. V. M.; Veeken, A. H. M.; Silva,
T. Moisture relationships in composting processes. Compost
Sci. Util. 2002, 10 (4), 286-302.
(17) Schulze, K. L. Continuous thermophilic composting. Appl.
Microbiol. 1962, 10 (2), 108-122.
(18) Jeris, J. S.; Regan, R. W. Controlling environmental
parameters for optimum composting. Part II: moisture, free
air space and recycle. Compost Sci. 1973, 14, 8-15.
(19) Hamelers, H. V. M. A mathematical model for composting
kinetics. Ph.D. Dissertation, Wageningen University, 2001.
(20) Das, K.; Keener, H. M. Moisture effect on compaction and
permeability in composts. J. Environ. Eng. 1997, 123 (3),
275-281.
Biotechnol. Prog., 2004, Vol. 20, No. 5
(21) Nield, D. A.; Bejan, A. Convection in Porous Media;
Springer-Verlag: New York, 1999.
(22) Lage, J. L.; Antohe, B. V. Darcy’s experiments and the
deviation to nonlinear flow regime. ASME J. Fluids Eng.
2000, 122, 619-625.
(23) Auria, R.; Revah, S. Pressure drop as a method to evaluate
mold growth in solid-state fermentors. In Advances in Bio-
process Engineering; Galindo, E., Ramirez, O. T., Eds.; Kluwer
Academic Publishers: Dordrecht, The Netherlands; 1994; pp
289-294.
(24) Auria, R.; Ortiz, I.; Villegas, E.; Revah, S. Influence of
growth and high mould concentration on the pressure drop
in solid-state fermentations. Process Biochem. 1995, 30, 751-
756.
(25) Rahman, S. Food Properties Handbook; CRC Press: Boca
Raton, FL, 1995.
(26) Kummer, F. A.; Cooper, A. W. The dynamic properties of
soils IX. Soil porosity determinations with the air pressure
pycnometer as compared with the tension method. Agric. Eng.
1945, 26, 21-23.
(27) Page, J. B. Advantages of the pressure pycnometer for
measuring the pore space in soils. Soil Sci. Soc. Am. Proc.
1947, 12, 81-84.
(28) Pall, R.; Mohsenin, N. N. A soil air pycnometer for
determination of porosity and particle density. Trans. ASAE
1980, 23 (3), 735-741, 745.
(29) Day, C. L. A device for measuring voids in porous materi-
als. Agric. Eng. 1964, 41 (1), 36-37.
(30) Chang, C. S. Measuring density and porosity of grain
kernels using a gas pycnometer. Cereal Chem. 1988, 65 (1),
13-15.
(31) Geddis, A. M.; Guzman, A. G.; Bassett, R. L. Rapid
Estimate of Solid Volume in Large Tuff Cores Using a Gas
Pychnometer. U. S. Nuclear Regulatory Commission: Wash-
ington, DC, 1996; NUREG/CR-6457.
(32) Dullien, F. A. L. Porous Media: Fluid Transport and Pore
Structure, 2nd ed.; Academic Press: San Diego, 1992.
(33) Buchlin, J.-M.; Stubos, A. Phase change phenomena at
liquid saturated self-heated particulate beds. In Modelling
and Applications of Transport Phenomena in Porous Media;
Bear, J., Buchlin, J.-M., Eds.; Kluwer Academic Publishers:
Dordrecht, The Netherlands; 1991; pp 221-276.
(34) Fand, R. M.; Kim, B. Y. K.; Lam, A. C. C.; Phan, R. T.
Resistance to the flow of fluids through simple and complex
porous media whose matrices are composed of randomly
packed spheres. ASME J. Fluids Eng. 1987, 109 (3), 268-
274.
(35) Barrington, S.; Choinière, D.; Trigui, M.; Knight, W.
Compost airflow resistance. Biosystems Eng. 2002, 81 (4),
433-441.
(36) Higgens, A. J.; Chen, S.; Singley, M. E. Airflow resistance
in sewage sludge composting aeration systems. Trans. ASAE
1982, 24 (4), 1010-1014; 1018.
(37) Saint-Joly, C.; Peyre, A.; Bochu, J. L.; N-Dao, O. Pressure
losses during ventilation of manure. Biol. Wastes 1989, 30,
123-132.
(38) Keener, H. M.; Hansen, R. C.; Elwell, D. L. Airflow through
compost: design and cost implications. Appl. Eng. Agric.
1997, 13 (3), 377-384.
(39) McGucken, R. L.; Eiteman, M. A.; Das, K. Pressure drop
through raw food waste compost containing synthetic bulking
agents. Agric. Eng. Res. 1999, 72, 375-384.
(40) Ergun, S. Fluid flow through packed columns. Chem. Eng.
Prog. 1952, 48, 89-94.
(41) Carmen, P. C. Permeability of saturated sands, soils and
clays. J. Agric. Sci. 1939, 29, 262-273.
(42) Macdonald, I. F.; El-Sayed, M. S.; Mow, K.; Dullien, F. A.
L. Flow through porous mediasthe Ergun equation revisited.
Ind. Eng. Chem. Fundam. 1979, 18 (3), 199-208.
(43) Beukema, K. J.; Bruin, S.; Schenk, J. Heat and mass
transfer during cooling and storage of agricultural products.
Chem. Eng. Sci. 1982, 37, 291-298.
(44) Dona, C. L. G.; Stewart, W. E., Jr. Numerical analysis of
natural convection heat transfer in stored high moisture corn.
J. Agric. Eng. Res. 1988, 40, 275-284.
 Biotechnol. Prog., 2004, Vol. 20, No. 5
(45) Morgan-Sagastume, F.; Sleep, B. E.; Allen, D. G. Effects
of biomass growth on gas pressure drop in biofilters. J.
Environ. Eng. 2001, 127 (5), 388-396.
(46) Terzaghi, K.; Peck, R. B.; Mesri, G. Soil Mechanics in
Engineering Practice, 3rd ed.; John Wiley and Sons: New
York, 1996.
(47) Wall, D. K.; Zeiss, C. Municipal landfill biodegradation and
settlement. J. Environ. Eng. 1995, 121 (3), 214-224.
(48) Kroodsma, W.; Ogink, N. W. M.; Satter, I. H. G.; Willers,
H. C. A technique for direct separation of pig excrements
followed by on-farm treatment of the components. Interna-
tional Conference on Agricultural Engineering, 1998; Oslo,
Norway; EURAGENG; pp 213-214.
(49) Methods of Soil Analysis, part 1, physical and mineralogical
methods; Soil Science Society of America: Madison, WI, 1986.
(50) Methods of Soil Analysis, part 3, chemical methods; Soil
Science Society of America: Madison, WI, 1996.
(51) Choi, H. L.; Richard, T. L.; Ahn, H. K. Composting high
moisture materials: biodrying poultry manure in a sequen-
tially fed reactor. Compost Sci. Util. 2001, 9 (4), 303-311.
(52) Loon, W. K. P.; van de Kreuk, M. K.; Hamelers, H. V. M.;
Bot, G. P. A. Water regulated air flow controller for perme-
ability measurements. In Preprints of Second Int. Workshop
on Control Applications in Post-harvest and Processing
Technology, 1998; IFAC/ISHS/CIBR/EURAGENG, pp 47-51.
(53) Fogiel, A. C.; von Bernuth, R. D.; Michel, F. C., Jr.; Louden,
T. L. Experimental verification of the natural convection
transfer of air through a dairy manure compost media.
View publication stats
1381
Proceedings of the International Meeting of the American
Society of Agricultural Engineers; ASAE, St. Joeseph, MI:
1999; ASAE paper no. 994053,.
(54) Veeken, A.; de Wilde, V.; Hamelers, H. V. M. Passively
aerated composting of straw-rich pig manure: Effect of
compost bed porosity. Compost Sci. Util. 2002, 10 (2), 114-
128.
(55) Summary Report: In-Vessel Composting of Municipal
Wastewater Sludge; United States Environmental Protection
Agency: Cincinnati, OH, 1989; EPA/625/8-89/016.
(56) Cooper, S. C.; Sumner, H. R. Airflow resistance of selected
biomass materials. Trans. ASAE 1985, 28 (4), 309-1312.
(57) Giner, S. A.; Denisienia, E. Pressure drop through wheat
as affected by air velocity, moisture content and fines. J.
Agric. Eng. Res. 1996, 63, 73-86.
(58) Bakker-Arkema, F. W.; Patterson, R. J.; Bickert, W. B.
Static pressure-airflow relationships in packed beds of granu-
lar biological materials such as cherry pits. Trans. ASAE
1969, 12 (1), 134-136, 140.
(59) Skaar C. Wood-Water Relations; Springer-Verlag: Berlin,
1988.
(60) Kirkham, K.; Powers, W. L. Advanced Soil Physics; John
Wiley & Sons: New York, 1972.
Accepted for publication May 7, 2004.
BP0499505