Anda di halaman 1dari 25

Biomass 18 (1989) 127-151

Mass and Energy Balance Analyses of a Downdraft


Gasifier

S. M. Chern, Walter P. Walawender* & L. T. Fan

Department of Chemical Engineering, College of Engineering, Durland Hall, Kansas


State University, Manhattan, Kansas 66506, USA

(Received 26 July 1988; revised version received 18 December 1988;


accepted 16 January 1989)

ABSTRACT

An empirical stoichiometric equation has been developed for wood chip


gasification in a commercial-scale moving bed downdraft gasifier. The
equation is based on an analysis of experimental overall and elemental
material balance data obtained with the gasifier. A thermodynamic
analysis of the gasifier has also been performed, based on the resultant
empirical stoichiometric equation. The first-law and second-law thermo-
dynamic effieiencies of the gasifier have been evaluated for four different
operating modes at three different output temperatures.
The resultant empirical stoichiometry is in agreement with the means
of the experimental data within one standard deviation. The highest first-
law and second-law thermodynamic effciencies have been obtained when
all products are considered useable; they are 90% and 62%, respectively.
The lowest first-law and second-law efficiencies have been obtained when
cool dry gas is considered as the only useable product," they are 72% and
53%, respectively. The heat loss from the system to the surroundings has
been estimated to be 10% of the energy input; this corresponds to a loss of
7% of the available energy input. The available energy dissipation in the
system, due to the various irreversibilities of the gasification process, has
been evaluated to be 31% of the available energy input. This dissipation is
not recoverable and reflects the nature of the process.

Key words: wood gasification, downdrafl gasifier, material balance,


energy balance, thermodynamic efficiency.

*To whom correspondence should be addressed.


127
Biomass 0144-4565/89/S03.50 -- 1989 Elsevier Science Publishers Ltd, England.
Printed in Great Britain
128 s.M. Chern et al.

INTRODUCTION

Moving bed gasifiers have been used to produce gas from charcoal and
wood for more than a century. The first gas producer was built by Bis-
chof in 1839; later in 1878, it was adapted for power generation by
Dowson (Wyer)) Since then, numerous gas producers have been deve-
loped; for a time, the producer gas industry grew rapidly. Producer gas
was a major source of fuel gas in the USA until natural gas dominated
the market in the 1930s. After that time, almost all the producer gas
plants were closed. However, fuel shortages in World War II prompted
renewed interest in producer gas in Europe during the war. In Sweden,
approximately 75 000 vehicles (40% of the automotive fleet), were con-
verted to producer gas operation within two years. The Swedish experi-
ence was compiled by the Swedish Academy of Engineering, and was
later translated into English by Reed and Jantzen. 2 After the war, the
need for gasifiers dwindled and only a minimum level of research was
maintained. The energy crisis of the past decade prompted a search for
alternative energy sources in the USA, and moving bed gasifiers once
again received attention.
Extensive literature is available, describing various aspects of the
design and operation of different types of downdraft gasifiers. However,
little has been published concerning comprehensive mass and energy
balances on gasifiers. Graham and Huffman 3 investigated a commercial-
scale choke-type gasifier rated at 1000 MJ/h. They have reported com-
plete mass balances, thermal efficiencies, mass conversion efficiencies
and gas compositions with wood species, moisture content, and size and
quality of the feedstock as operating parameters for a limited number of
experiments. Walawender et al. 4 have reported on the performance of a
downdraft gasifier with a working capacity of 320-1600 MJ/h. Com-
plete material balances as well as gas compositions have been deter-
mined by them over a wide working range to examine the influence of
the throughput rate on the gasifier performance. Recently, Graboski and
Brogan 5 have reported limited data for a prototype gasifier with a capac-
ity of 15 800 MJ/h; Walawender et al. 6 have also presented additional
data for various operating parameters.
Only a few studies have been published concerning the thermody-
namic analysis of pyrolysis processes. Otoma and Gotoh 7 have applied
the second-law of thermodynamics to the analysis of a pyrolysis system.
Shieh and Fan 8 have derived formulas for estimating the energy and
available energy contents of structurally complicated materials, such as
coal and biomass. Rigorous mass, energy and available energy balances
have been developed for a conceptual pyrolysis system by Ishimi et al. 9
Mass and energy balance analyses of a downdrafi gasifier 129

So far, however, no detailed thermodynamic analysis has been reported


on the moving bed downdraft gasifier.
This work presents the complete material and energy balances for the
'Buck Rogers' moving bed downdraft gasifier, based on the experimental
data of Walawender et aL 4 for the gasification of mixed hardwood chips.
One objective of this work is to formulate an empirical stoichiometry to
describe the overall gasification process. A second objective is to con-
duct a thermodynamic analysis to determine various efficiencies of the
process. The approach is based on the law of mass conservation and the
first- and second-laws of thermodynamics.

MATERIAL B A L A N C E ANALYSIS

The objective of the material balance analysis is to formulate an empiri-


cal stoichiometric equation to describe the gasification of wood chips in
a downdraft gasifier. The stoichiometry formulation is based on both the
overall material balance and the elemental material balances for the pro-
cess obtained from the experimental data of Walawender et a l ? A t w o -
level adjustment of the data has been performed to force material balance
closures.
To formulate a stoichiometry for the overall gasification reaction, a
material balance closure is required for all chemical elements involved in
the reaction. A set of experimental data with perfect balances for all
elements naturally yields the empirical stoichiometry. Nevertheless, a
complete closure is rarely achieved for a real process, especially one as
complicated as moving bed gasification. The inherent variability of wood
characteristics, measurement errors and system fluctuations all con-
tribute to render the complete closure impossible.
The present analysis resorts to material balance closures, both overall
and elemental, to measure to what extent the material balance has been
achieved. The material balance closure is defined as the mass ratio of
output to input. Nine runs from the work of Walawender et aL 4 have
been selected for a preliminary analysis. All have overall closures within
_+5%. A schematic of the system considered in the analysis is shown in
Fig. 1; the data base is summarized in Table 1. One of the nine runs in
Table 1, Run 101, has been rejected from further consideration due to its
poor elemental closures in the preliminary analysis. Table 2 presents the
overall and the elemental closures for the remaining eight runs. The
closures have been evaluated based on the following assumptions.
(1) The elemental composition of dry chips is the mean of the
measured elemental compositions for the mixed hardwood chips
130 X M. Chern el al.

Wood - Dry Gas

Chips
~ H 2 0
Gasifler
_ Char

Air
- Tar

Fig. 1. Flow diagram of the gasifier system.

TABLE 1
Material Balance Summary

Run no. Inputs (kg/h) Outputs (kg/h) Overall


closure (%)
Wet chips Dry air Water Total Dry gas Char Tar Water Total

98 32.0 43.1 0.5 75.6 66.3 0.9 0.09 7.4 74.7 98.9
910 32.0 45.2 0.5 77.7 68.0 0.9 0.14 7.4 76.4 98.3
920 35.7 62-1 0.4 98.2 92.9 1.4 0.09 7.1 101.5 103-4
101 52.4 76.9 0.8 130.1 116.9 1.7 0.09 13.3 132.0 101.4
929 52.7 74-0 1.0 127.7 113.0 1.8 0.14 12.2 127.1 99.5
106 58.1 76.8 0.5 135.4 117.0 1.2 0.09 10.8 129.1 95.4
1 119 89.1 112.1 1.0 202.2 173.8 3.0 0.27 19.7 196.8 97.3
1117 96.2 140.4 0.7 237.3 218.9 2.5 0.18 22.9 244.5 103.0
1221 140.5 202.1 1.1 343.7 302.1 4.1 0.54 43.3 350.0 101.8

TABLE 2
Mass Balance Closures of Original Data

Run C H 0 N Overall

98 0.96 0.86 1.01 0.99 0'99


910 0.96 0.85 1.00 0"99 0.98
920 1.06 0"99 1"09 0.99 1'03
929 0"92 0.93 1.05 0'99 1"00
106 0.82 0.84 0-99 0.99 0"95
1 119 0.94 0"85 0"98 0'99 0"97
1 117 1.02 0,99 1.09 0"99 1.03
1 221 0.93 1.02 1.10 0.99 1.02

aClosure is defined as the mass ratio of output to input.


Mass and energy balance analyses of a downdrafi gasifier 131

studied by Walawender et aL 4 This composition is given in Table


3.
(2) Dry air consists of oxygen and nitrogen only, with a molar ratio of
21-79. Its composition is included in Table 3 on a weight basis.
(3) The elemental composition of tar is assumed to be the same as
that of dry chips due to the lack of experimental data. Although this
introduces error in the elemental composition of tar, its effect on
the material balances is negligible due to the extremely small
amount of tar compared to the other streams (see Table 1).
(4) The elemental composition of char is the mean of the measured
elemental compositions of the char products obtained by
Walawender et al. 4 The char composition is included in Table 3.

TABLE 3
Elemental Compositions"

C H 0 N Ash

W o o d c h i p s ~' mean 48-11 6-05 44.97 0-13 0.74


o 0.44 0'09 0"51 0"06 0.21
D r y air -- -- 23.30 76.70 --
Char' mean 75.83 0"89 4.74 0.06 18.48
o 8' 15 0"08 -- 0"05 --
Water -- 11.11 88.89 -- --

a W e i g h t %.
bNine samples dry basis.
'Eight samples dry basis.

Table 2 indicates that the carbon (C) closures in the original data
range from 0.82 to 1.06, the hydrogen (H) closures range from 0-84 to
1.02, the oxygen (O) closures range from 0.98 to 1.10, all nitrogen (N)
closures are 0"99, and the overall closures range from 0.95 to 1.03.
Among the elemental balances, the poorest average closure is for
hydrogen and the best is for nitrogen. It should be noted that in the work
of Walawender et aL 4, the nitrogen balance was used to determine the
dry air input rate, thus closing the balance on nitrogen.
The problem remaining is one of adjusting the experimental data to
achieve closed balances without losing their overall representation of the
gasification process. Since averaging the data is not sufficient, alternate
132 S.M. Chern et al.

techniques need to be developed; these are the primary and secondary


balance adjustments, which are delineated in the following sections.

Primary balance adjustments

The strategy for the primary adjustment is to make minor adjustments in


the stream rates of the various process inputs and outputs to achieve the
best overall and elemental material balance closures, without altering the
composition of any stream. This approach is based on the assumption
that the material imbalances are primarily the result of minor errors in
the direct measurements and indirect determinations of the various
stream rates. As depicted in Fig. 1, there are two input streams: air (with
moisture) and wood chips (with moisture); and there are four output
streams: dry gas, water, char and tar. There are four elemental species, C,
H, O and N; sulfur has been neglected because its content in wood is
small. Ash is only present in the chips and char.
Since the amounts of tar and char generated by the gasifier were very
small compared to the other streams, they were not altered in the
primary adjustment. Efforts were directed toward adjusting the
four remaining streams to achieve the best overall results for the five
closures. This has given rise to a multi-variable, multi-objective optimiz-
ation problem. To solve it, priority must be assigned to each of the five
closures; the overall, C, H, O and N closures have been selected in
descending order according to their relative importance in the
subsequent thermodynamic analysis. The objective function is the sum of
squares of deviations of individual closures, multiplied by their
corresponding weighting factors. The weighting factors have been
arbitrarily chosen to be 50, 10, 10, 3 and 2 for the overall, C, H, O and N
closures, respectively. The problem has then been solved with an
exhaustive search method.
Table 4 presents the optimum primary adjustment to each stream in
terms of percentages of the stream rates for the eight data sets. The
maximum adjustment is 17%. Table 5 gives the elemental and overall
closures corresponding to the adjustments in Table 4 for the eight runs
after adjustment. As can be seen, all the overall closures are 1-00 (to two
decimal place accuracy). Note that the ranges on closure have been
substantially reduced for C, H and O.
Table 6 presents the mass balance data (on a dry ash-free basis) for the
eight runs after the first adjustment; the data have been normalized such
that the wood chip feed rates are all 100. Note that all water input and
output streams have been combined into a single term, the net water
Mass and energy balance analyses of a downdrafi gasifier 133

TABLE 4
Summary of the Primary Adjustment ~

Run Streams adjusted (%)

Air (weO Chips (wet) Dry gas Wate/'

98 0 - 10 - 5 15
910 0 -5 0 0
920 4 5 0 15
929 4 -8 0 -5
106 4 - 17 0 -5
1 119 0 -7 0 0
1 117 4 2 0 5
1221 4 -5 0 -15

"Adjustment is expressed by the percentages by which the rates of the corresponding


streams were adjusted.
bThis is the vapor in the outlet gas stream.

TABLE 5
Mass Balance Closures ~ After Primary Adjustment

Run C H 0 N Overall

98 1"01 0'98 1-07 0'94 1"00


910 1'00 0'90 1"02 0"99 1'00
920 1"00 0-99 1'06 0"96 1'00
929 0"99 0"98 1'06 0"96 1"00
106 0"99 0"99 1-07 0"96 1"00
1 119 1'00 0"91 1'02 0-99 1"00
1 117 0"99 0"98 1'07 0"96 1"00
1 221 0'97 0"99 1'06 0"96 1-00

"Closure is defined as the mass ratio of input to output.

output. The means and standard deviations (a) for the eight runs are
given at the bottom of the table.

Secondary balance adjustments

The comparison of Tables 2 and 5 indicates that the closures are still not
complete, even though they have been improved after the primary
adjustment. Therefore, secondary adjustments have been conducted.
The adjustment scheme is based on the following considerations.
134 X M. Chern et al.

TABLE 6
Summary of Mass Balance" After Primary Adjustment on Dry Ash Free (DAF) Basis

Run Dry chips + Dry Air = Dry gas + Water + Tar + Char

98 100'00 181.41 265.32 14"05 0'38 2'41


910 100'00 179"67 270'23 7-41 0"54 2"61
920 100"00 184"34 265"42 15"88 0"26 3"20
929 100"00 172"86 273"90 15"68 0"31 3'20
106 100.00 178.65 261-88 14"69 0"20 1'84
1 119 100"00 163'21 253'05 7'74 0'40 3"18
1 117 100"00 168"80 252'98 14"21 0.21 2"11
1 221 100"00 178'12 256"04 18.24 0"46 2"29
mean 100.00 175.88 259.85 13.49 0.35 2.61
a -- 7.06 6.72 3"88 0.12 0"54

"The values were normalized such that all chips are 100.

( 1 ) As many compositions as possible of the six streams are to remain


unchanged. However, all of them cannot remain fixed. The
composition of char has been chosen to be open for adjustment
because (a) the compositions of air and water are fixed, (b) the
compositions of wood chips and dry gas are reliable measured
quantities -- thus, it is preferable to have them fixed due to their
importance in the thermodynamic analysis -- and (c) variation of
the composition of tar is not useful, because the amount of tar is
insignificant.
(2) The amount of dry gas relative to wood chips should be fixed due
to its importance in the thermodynamic analysis. Therefore, the
amount of air, the amount and composition of char, and the net
water output remain to be adjusted.
The secondary adjustments are based on the means for the eight runs,
as presented in Table 6. The first and final stages of these adjustments are
presented in Table 7. In the first stage, the amount of dry air has been
adjusted slightly (for computational purposes) to attain a complete
overall material balance closure to three decimal places (since the
closure obtained in the primary adjustment is based on only two decimal
places). Then each of the first five streams has been broken down into
the elemental constituents, C, H, O and N. Next, the elemental con-
stituents of char are determined by difference to force each elemental
balance to close. Nevertheless, as can be seen from Table 7 (first stage),
the resultant char composition is not realistic with respect to oxygen and
nitrogen.
Mass and energy balance analyses of a downdraft gasifier 135

TABLE 7
First and Final Stages of the Secondary Adjustments

Element Chips + Air = Gas + Water + Tar + Char

First stage
C 48"477 0"000 45"965 0"000 0"168 2-344
H 6'094 0-000 4"215 1"499 0"021 0'359
O 45.306 41.076 79"622 11'991 0"156 - 5'387
N 0-123 135"215 130"050 0"000 0"000 5'288
Sum 100"000 176"291 259-852 13.490 0"345 2"604
Total input = 276.291 Total output = 276.291

Final stage
C 48-477 0.000 45.965 0'000 0"168 2.344
H 6.094 0"000 4-215 0.625 0.021 1.233
O 45.306 39.470 79.622 4'998 0.156 0"000
N 0.123 129.927 130.050 0.000 0"000 0.000
Sum 100.000 169-397 259.852 5'553 0.345 3.577
Total input = 269.397 Total output = 269.397

The elemental analysis of char presented in Table 3 indicates that the


oxygen and nitrogen contents of char should be very small or negligible.
Therefore, the assumption of zero oxygen and nitrogen contents for char
has been adopted. Subsequently, the secondary adjustments have been
executed according to the following procedures.
(1) The nitrogen content of char is forced to zero by subtracting the
excess from the nitrogen in the air and the oxygen content of air is
adjusted to maintain the composition of air.
(2) The oxygen content of char is forced to zero by adjusting the
oxygen content of the net water output to compensate for the
changes of oxygen in the air and char.
(3) The hydrogen content of water is adjusted to maintain the com-
position of water.
(4) The hydrogen content of char is adjusted to maintain the material
balance for hydrogen.
The results are presented as the final stage in Table 7.
Converting the results in Table 7 from a weight basis to a molar basis
using the mean gas composition of the experimental data (see Wala-
wender et al. ),4 and expressing the dry gas in terms of its respective con-
stituents yields the empirical stoichiometric equation for the overall
gasification presented in Table 8. Note that the structurally complicated
136 S. M. Chern et al.

TABLE 8
An Empirical Equation for the Gasification of Wood
Chips in a Moving Bed Gasifier

1"0 C h i p s + 1"233 0 2 = 1"435 H 2 + 0"0112 C3H~,


or 5-871 Air)
+ 1"549 COz + 0"0509 C2H 4

+ 0"0077 C2H 6 + 0"0044 N 2


(or 4 ' 6 4 3 )
4- 0"249 C H 4 + 1"878 C O

+ 0"308 H 2 0 + 0"00345 Tar


+ 0"0357 Char
where chips, tar and char all have arbitrary
molecular weight of 100 and their empirical formulas
are
Chips, tar: C4o4H6.0502.83No.o088
Char: C5.45H34.6

materials -- wood chips, tar and char -- all have an arbitrary molecular
weight of 100. The average molar composition of the dry gas can be
evaluated from the quantities of the various components given in Table
8.

T H E R M O D Y N A M I C ANALYSIS

The purpose of this analysis is to evaluate various thermodynamic


efficiencies for the moving bed downdraft gasifier. The approach is
based on the first- and second-laws of thermodynamics. Thermo-
dynamic efficiencies can be used to identify the sources of inefficiency in
the process.

Background

Terminologies, concepts and relationships essential for the present


analysis are defined or delineated in the following (see Shieh and Fan, 8
Ishimi et al., 9 and Szargut and Petela). 1

Reference state
The dead state is chosen as the reference state for the evaluation of
enthalpy and available energy. This is the state in which the system under
Mass and energy balance analyses of a downdrafi gasifier 137

consideration can neither exchange energy with nor perform work on its
surroundings; in other words, this is the state in which materials in the
system are in equilibrium with their environment. For practical purposes,
the dead state of a material species will be the most stable state of that
material species in its surrounding environment. Table 9 presents the
datum level materials, datum level concentrations, specific chemical
enthalpies and specific chemical exergies of the materials involved in the
present gasification process (Shieh and Fan)/ The temperature and
pressure of the dead state are considered to be 298.15 K and 1 atm,
respectively, in this study. Note that the standard state is also at 298.15 K
and 1 atm but with the chemical species in the pure state.

Energy balance
Applyingthe first-law of thermodynamics to the system shown in Fig. 2
and assuming that the changes in potential energy and kinetic energy are
negligible, we have

Z(Zflknk)i+W=Z(Zflknk),+Qi , (1)

TABLE 9
Specific Chemical Enthalpy, Specific Chemical Exergy, Datum Level Material and
Datum Level Concentration

Datum level Datum level fl o~ e oh


material concentration (kcal/mol) (kcal/mol)
(molar fractiopO

C(s) CO2(g ) 0.000302 94-052 98"131


CO(g) CO2(g ) 0.000302 67.630 65.790
CO2(g) CO2(g ) 0.000302 0"0 4.802
CH4(g) CO2(g), H20(I) 0.000302, 1 212.800 152-380
C2H4(g) CO2(g), H20(I) 0.000302, 1 337-240 278.950
C2H6(g) CO2(g), H20(I ) 0.000302, 1 372-820 287.990
C3H6(g) COz(g), H20(I ) 0.000302, 1 491.990 408.990
H2(g) H20(I ) 1 68.320 33-197
H20 (1) H20(I ) 1 0'0 0.0
H20(g) H20(I) 1 10.511 0"0
Nz(g) N2(g) 0.78' 0"0 0-147
O2(g) O2(g) 0"2096 0"0 0'926
Air N2(g), O2(g) 0"78', 0"2096' 0'0 0"0
"Specific chemical enthalpy.
bSpecific chemical exergy.
"If the air composition of N2 = 0.79 and 02 = 0.21, used in the previous material balance
analysis, is used here, e of N 2 will have a slightly different value, 0.140 kcal/mol. How-
ever, this difference has a negligible influence on the subsequent calculations.
138 S.M. Chern et al.

I
il L ; It 1
i2 . [I II 2
I
I
f
I
I
I T
i m-----I e I
I

W Qo
Fig. 2. Schematic diagram of an open flow system.

where n k is the number of moles of species k in a stream. Q0 is the heat


loss from the system to the surroundings, W is the work supplied to the
system by the surroundings, and /~k is the partial molar enthalpy of
species k relative to the dead state. Subscripts i and e refer to the input
and exit streams, respectively, and/3 is defined as
/~ = h - ho (2)
where h is the partial molar enthalpy and h o is the partial molar enthalpy
in the dead state. Equation (2) can be expanded to facilitate its evaluation
as follows:
~= (h - h ) + (h - ho)
= (h - h o ) + (h - h )

=/~"+ (~)f~fi)T ) p d T + ),rdP (3)


T o

= fl" + [~ - T ( O ~ / O T ) p ] d P
pO

Here, the superscript 0 refers to the standard state and the subscript 0 to
the dead state.
When some of the exit streams are discarded to the environment as
wastes, eqn ( 1 ) can be conveniently expressed as
~=He, u+~,d+Hc (4)
where
Mass and energybalance analyses of a downdraft gasifier 139

is the total energy input in the form of enthalpy and work,

e,u le, u

and

are, respectively, the useable and discarded portions of the energy output
in the form of enthalpy, and H c = Q0 represents the energy loss.

Available energy balance


The available energy of a system refers to the maximum amount of work
that the system can perform when it is brought to its dead state. Making
an available energy balance on the system depicted in Fig. 2 gives

where a is the entropy created in the system and gk is the partial molar
exergy of material species k. The partial molar exergy, g, is defined as
g = (h - / ~ . ) - T0(g- g , ) (6)
where g is the partial molar entropy and go is the partial molar entropy in
the dead state (Szargut and Petela] Gaggioli, ~ Riekert, ~2 and Fan and
Shieh). 13 This equation can be expanded to facilitate computation as
follows:
e = ~ - ~,(g - / , , )

= [ B - To(i - / o ) ] + Ii0p/ 1 - ro/r)dr

+
f
pO
[~ - ( T - To)(OtS/OT)p]dP

= go+ (1 - TolT)dT+ [ t S - ( T - To)(a~lOT)p]dP (7)


pO

where
io =/~o _ To(~o-~o)
140 S.M. Chernet al.
is termed the partial molar chemical exergy (Fan and Shieh). 13 When
some of the exit streams are discarded as wastes, eqn (5) can be conveni-
ently expressed as
Ai=Ae, u+ Ae, d+ Adis (8)
where

Ai=~i (~-~k6knk)i+W
is the total available energy input,

e,u / e , tl

and

e,d /e,d

are, respectively, the useable and discarded parts of the available energy
output, and
Adi s = ToO
is the available energy dissipation.

Thermodynamic efficiencies
The thermodynamic efficiency of a process can be defined in various
ways. Only the first-law (conservation) efficiency, r/i, and second-law
(conservation) efficiency, r/2, will be considered in this analysis. Referr-
ing to eqn (4), the first-law efficiency is defined as
rh =He,./H ~ (9)
This efficiency represents the ratio of the enthalpy retained in the
useable output to the total energy input (including enthalpy and work). In
terms of eqn (8), the second-law efficiency can be expressed as
r12=Ae,u/A i (10)
This efficiency represents the ratio of the available energy retained in the
useable output to the total available energy input. These two efficiencies
are concerned with the conservation of energy and available energy,
respectively. Variations of both are possible depending on what is
considered to be a useable output and the conditions under which the
useable products leave the system.
Mass and energy balance analyses of a downdraft gasifier 141

Computation
The gasifier under consideration is illustrated in Fig. 3. There are two
input streams: chips (with moisture and ash) and air (with moisture); and
there are four output streams: dry gas, water vapor, tar and char (with
ash). The enthalpy and exergy of each stream have been evaluated
according to eqns (3) and (7), respectively. The first-law and second-law
efficiencies of the process have been estimated according to eqns (9) and
(10), respectively. All input streams are assumed to enter the system at
298.15 K. The moisture in the wood chips is treated as liquid water,
ignoring the interaction between the water and wood. The amount of ash
is small compared to wood chips (less than 1%; see Table 3). The ash has
been treated as an inert material passing through the system and carrying
away only a small amount of sensible heat. The binding energy between
ash and wood or char was ignored.

-Electric Eye
"---1 Airgitator

-qP
Leveling
Rod I
I
I
I
Ic:l
i ! m

J I Fan M o t o r
- - - - . J r , L- . . . . .

Ceramic
Bat is

Grid Plate

Gear Box -

Fig. 3. Schematic of the commercial downdraft gasifier.


142 S . M . Chern e t ai.

Work is done on the system by an electric motor powering the gas fan,
the rotating grate and char air lock. The motor is rated at 5 hp, which is
slightly less than 1% of the average total energy input rate for the eight
experimental runs. Thus the mechanical work input has been neglected
in the computations.
A method for evaluating fl0 (specific chemical enthalpy) and e
(specific chemical exergy) of structurally complicated materials, e.g.
biomass, from their elemental compositions has been described by Shieh
and Fan. 8 According to their method,
fl=(1 + 0.15[O]) (14137.20[C] + 61127.145([H] - [O]/8)) (11)
and
e = 14750.7[C] + 9.47[N] + 50311.38[H]- 5724.49[0]
+ 0.1510] (14137-20[C1 + 61127.145[H]- 7640.9[0]) (12)
where [ ] represents the weight fraction of an individual element in the
complex substance; both fl0 and e are expressed in kcal/kg, fl0 and e of
the wood chips, char and tar (on a dry ash-free basis) have been
evaluated from these equations.
Since the system is open to the atmosphere and the pressure variation
through the gasifier is negligible, the pressure of the system is assumed
constant at 1 atm. Furthermore, ideal gas behavior is assumed. Cons-
equently, all partial molar terms involved in the evaluation of the
enthalpy and exergy of the gases reduce to ordinary molar terms and the
pressure dependent term in eqn (3) vanishes due to the ideal gas assump-
tion. The pressure dependent term in eqn (7) is small compared to the
other terms; therefore, it has been neglected. Thus, all the pressure
dependent terms have been dropped in evaluating the enthalpies and
exergies, thereby reducing eqns (3) and (7) to
T

and
/~flo+
f 7-o
cedT (13)

T
g--- e +
fTo
Cp(1- To/T)dT (14)

respectively.
To evaluate the temperature-dependent terms, required in eqns (13)
and (14), the heat capacities of the different materials must be known as
Mass and energy balance analyses of a do wndraft gasifier 143
functions of temperature. Heat capacity correlations of the form
ce=a+bT+cT 2+dT 3 (15)
have been used for H2, C3H6, CO2, C2H4, C2H6, N2, CH4, CO and
H20(g). The coefficients in eqn (15) are available in standard
monographs (Reid et al.). ~4 The sensible heat of crude oil is used to
approximate that of tar due to the lack of better information. The heat
capacity of char is approximated by graphite (C) and that of ash by
silicone dioxide.
The crude product stream comprises dry gas, water, tar and char. In
practice, the entire crude product can be directly utilized for some
purpose, e.g. heating. In other cases, some of its constituents must be
removed before use; in principle, the four constituents of the crude
product can be separated. Therefore, in evaluating the efficiencies of the
process, the following four operating modes are considered.
Mode 1 All output species are recovered as useable products.
Mode 2 Char is discarded, and the rest of the constituents are
recovered as useable products.
Mode 3 Char and tar discarded and the rest of the constituents are
recovered as useable products.
Mode 4 Only dry cooled gas is recovered as useable product.
In the experiments, the temperature of the crude product was
observed to range between 644 K and 700 K. Thus, the efficiencies of
the process have been evaluated at three output temperatures, namely,
298.15 K, 644 K and 700 K. A temperature of 298"15 K represents that
of the cool dry gas output.
Tables 10 and 11 summarize the enthalpies and exergies of the
various constituents of the input and output streams, respectively. The
enthalpies and exergies of each constituent are reported in two parts; one
is related to the chemical energy,/30 and e , and the other is related to the
sensible heat, Ate( =/37- fl0) and ACT( = eT-- e). The latter is evaluated
at two output temperatures, 644 K and 700 K. Also included in the
tables are the percentages of input energy recovered in the various
outputs at each temperature. The evaluation of enthalpy and exergy for
char has been conducted using two elemental compositions, one from
the empirical stoichiometry and the other from the measured composi-
tion given in Table 3. Tables 10 and 11 indicate that the important
energy related constituents in the output are CO, H2, CH4 and char, in
descending order.
Table 12 summarizes the first-law and second-law efficiencies for the
four operating modes at the three output temperatures. The char com-
144 S.M. Chern et al.

IIII

oo

,..0
x~ IIII .
E
0

0
~.~ IIII
0
lb..,

x~ IIII

~o
0
}. ID,.

,~ .~: IIII
.

U
lb.,,

I I I I

I I I I

~.0

0
Mass and energy balance analyses of a downdraft gasifier 145

[1111

oO

x~ IIIII Z

E
0

0
IIIII
.~

E
0

x~ IIIII

IIIII

..,;,

o
...;,
b

IIIII

IIIII ~ ~ ~ ~ I

0 0
~. oo~z~.- ~ ~ ~ o. ~

0
146 S. M. Chern et al.

TABLE 12
Summary of First-Law and Second-Law Efficiencies (%)

Mode Discarded output Output temperature (K)

298 644 700

First-law efficiency
1 None" 81"24 88"30 89"51
2 Char(ash) 74.92 81.89 83"08
3 Char(ash), tar 74.54 81.49 82.67
4 Char(ash), tar, water 72.00 78.24 79.31
Second-law efficiency
1 None" 59.40 61.76 62.40
2 Char(ash) 53.23 55.47 56-09
3 Char(ash), tar 52.85 55.07 55"69
4 Char(ash), tar, water 52.85 54.85 55.40

aNote the evaluation of enthalpy and exergy for char was based on the measured com-
position of char (presented in Table 3).

position used for the evaluation of the efficiencies with all products
considered useable is the measured composition given in Table 3. The
highest first-law and second-law efficiencies are 90% and 62%, respect-
ively, for Mode 1 with an output temperature of 700 K. The lowest first-
law and second-law efficiencies are 72% and 53%, respectively, for
Mode 4 with an output temperature of 298" 15 K (cool dry gas).

DISCUSSION

The empirical stoichiometry is compared to the original experimental


data (Walawender et al.)4 from which the formula is derived. The first-
law and second-law thermodynamic efficiencies are discussed for four
operating modes at various temperatures of the exit streams, thereby
identifying the sources of inefficiency in the process.

Material balance analysis

The means and standard deviations, reported by Walawender et aL 4


(based on the eight selected runs), for the air input, dry gas output, net
water output and char output relative to the chip feed are, respectively,
1.62 and 0.13, 2.46 and 0.19, 0.13 and 0.054, and 0.032 and 0.0061 kg/
kg dry chips. The empirical stoichiometry derived in this study has
Mass and energy balance analyses of a downdraft gasifier 147

yielded air input, dry gas output, net water output and char output ratios
of 1.69, 2"60, 0"056 and 0.036, respectively (see the bottom line of Table
7). The comparison of the derived stoichiometry with the experimental
data shows that the air input, dry gas output and char output ratios are
within one standard deviation of the means of the experimental data.
The net water output ratio is only slightly more than one standard
deviation from the mean of the experimental data. Therefore, the
derived stoichiometric formula is representative of the gasification
process in terms of the relative amount of each stream.
The air input ratio of 1.62 from the present work is in reasonable
agreement with the ratio of 1.38 (average of four runs) reported by
Graham and Huffman 3 for air gasification of poplar in a forced draft
gasifier. When the data from the two experiments of Graboski and Bro-
gan 5 (also a forced draft system) are adjusted for air used in the combus-
tion of propane in the gasifier head space, air input ratios of 1.66 and
1.74 are obtained. These values are in very good agreement with the pro-
posed stoichiometry.
For the specific procedure employed in the final stage of the secon-
dary adjustment, the compositions of various streams, except that of the
char stream, remain unchanged. Note that the composition of tar has
been arbitrarily assigned, because the amount of tar is negligible.
However, since the char has been selected as the victim to force closure
of the material balance, its composition is distorted. The molar
hydrogen-to-carbon ratio in the forced char composition is 6"3, which is
higher than that of methane. The distortion of the composition of char is
the trade-off for complete material balance closure along with mainten-
ance of the relative amount of each stream.
Alternate procedures may be employed for the final stage adjustment
which result in much more realistic char compositions. However, the
relative amount of each stream and/or the gas composition will be
altered as a consequence. For example, the oxygen deficit in the char in
the first stage of the secondary adjustment (see Table 7) may be forced to
zero by distributing the deficit in various ways between the air and the
chips (all to air, all to chips, or split). This results in an increase in total
nitrogen input since the compositions of air and chips must be main-
tained. The increase in nitrogen input along with the nitrogen in the char
can be added to the dry gas to maintain nitrogen closure and force the
char nitrogen to zero. The gas composition must now be adjusted. This
adjustment can be accomplished by assuming that the mass fraction of
nitrogen in the dry gas increases while the amounts of carbon, hydrogen
and oxygen in the gas remain unchanged. This assumption implies that
the nitrogen-free gas composition remains unchanged; however, the
148 S. M. Chern et al.

mass fraction of nitrogen-free gas decreases. This procedure allows easy


recovery of the amounts of CO, CO2, H2, etc., from the nitrogen-free
gas. Finally, the amounts of char carbon and hydrogen are readily deter-
mined by closing their respective elemental balances. In the above
procedure any oxygen deficit that is absorbed by the chips increases the
amount of chips above 100. Normalization restores the starting basis.
The above adjustment scheme results in an approximate dry ash-flee
char composition of 87-88% wt carbon with the balance hydrogen. The
resulting char composition is relatively insensitive to the split of the
deficit oxygen between the chips and air; however, the char output and
other relative stream rates are influenced by the split. If all of the oxygen
is distributed to the air stream, the respective ratios for air, gas, water
and char are 1.99, 2.83, 0.13 and 0.027, and the gas contains 51% N2 on
a molar basis. If all of the oxygen is distributed to the chip feed stream,
the respective ratios for air, gas, water and char are 1.58, 2.37, 0"12 and
0"082, and the gas contains 48.2% N 2 on a molar basis.
Directing all of the oxygen deficit to the air stream results in higher
than normal values for the air and gas ratios along with a high N2% in the
gas (mean observed gas N2% = 47.2). Directing all of the deficit oxygen
to the chip stream results in low gas and high char ratios with only a
small effect on gas composition. In contrast, the selected adjustment
procedure in Table 7 results in reasonable values for all stream ratios as
well as an unaltered gas composition. Therefore, sacrificing the char
composition results in the least overall distortion of the process stoichio-
metry.

Thermodynamic analysis
The evaluation of the first-law and second-law efficiencies is based on
the stoichiometry derived in the material balance analysis; nevertheless,
one adjustment has been made. The composition of char is based on the
experimental data to provide a more realistic measure of the gasifier
efficiency when char is considered to be a useable product.
The highest first-law efficiency, 90%, implies that the heat loss from
the gasifier to its surroundings is 10% of the energy (enthalpy) input.
The available energy loss due to this heat loss is about 7%. This avail-
able energy loss has been evaluated by assuming that heat is dissipated
from the gasifier at a uniform temperature of 900 K to its surroundings at
298.15 K. Therefore, the use of insulation to reduce the heat loss is
important, especially in terms of available energy.
The lowest first-law efficiency, 72%, corresponds to the cold gas
efficiency; this is the same as the mean reported by Walawender e t al. 4
Mass and energy balance analyses of a downdraft gasifier 149

The lowest second-law efficiency, 53%, represents the second-law cold


gas efficiency. This indicates that when cool dry gas is regarded as the
only useable product, 28% of the energy (enthalpy) and 47% of the
available energy are lost in the process.
The highest second-law efficiency, 62%, along with the available
energy dissipation due to the heat loss, 7%, implies that 31% of the avail-
able energy input is dissipated in the system due to the various
irreversibilities in the gasification process. In other words, 31% of the
available energy input is lost due to the nature of the process and cannot
be recovered.
The variation in the temperature of the exit streams indicates that the
sensible heat accounts for 7-8% of the energy (enthalpy) and only 2-3%
of the available energy for all four operating modes. This in turn
indicates that the sensible heat is more valuable in terms of enthalpy than
in terms of exergy.
Comparing the efficiencies of the different operating modes shows
that, for all output temperatures, (1) discarding char will lower both
efficiencies by approximately 6%, (2) discarding tar will lower both
efficiencies by less than 1%, and (3) discarding steam will lower the first-
law efficiencies by approximately 3%; nevertheless, it has negligible
influence on the second-law efficiencies. These results indicate that the
utilization of the energy (mainly the chemical energy) stored in char can
improve the efficiency of this process.

CONCLUDING REMARKS

An empirical stoichiometric equation has been developed for wood chip


gasification in a commercial-scale moving bed downdraft gasifier. The
equation is based on an analysis of experimental overall and elemental
material balance data obtained with the gasifier. The resultant empirical
stoichiometry is in agreement with the means of the experimental data
within one standard deviation. However, the composition of char has
been distorted in order to force closure of the material balance.
A thermodynamic analysis of the gasifier has also been performed,
based on the resultant empirical stoichiometric equation. The highest
first-law and second-law thermodynamic efficiencies have been obtained
when all products are considered useable; they are 90% and 63%,
respectively. The lowest first-law and second-law efficiencies have been
obtained when cool dry gas is considered as the only useable product;
they are 72% and 53%, respectively. The heat loss from the system to its
surroundings has been estimated to be 10% of the energy (enthalpy)
150 S.M. Chernet al.

input; this heat loss accounts for a loss of 7% of the available energy
input. However, the available energy dissipation in the system, due to the
various irreversibilities of the gasification process, has been evaluated to
be 31% of the available energy input. This dissipation is not recoverable
and reflects the nature of the process. The sensible heat of the output
streams accounts for 7 - 8 % of the energy (enthalpy) input and only 2 - 3 %
of the available energy input. The results also indicate that the utilization
of the energy stored in char can improve the efficiency of this process by
about 6%.

ACKNOWLEDGMENT

This is contribution no. 88-400-J, Department of Chemical Engineering,


Kansas Agricultural Experiment Station, Kansas State University,
Manhattan.

REFERENCES

1. Wyer, S. S., Catechism on Producer Gas. McGraw, New York, 1906, pp.
26-7.
2. Reed, T. B. & Jantzen, D. E., Generator Gas -- The Swedish Experience
from 1939--45, (translated from the work of the Swedish Academy of
Engineering), SERI/TP-33-239. SERI, Golden, CO, USA, 1980.
3. Graham, R. G. & Huffman, D. R., Gasification of wood in a commercial-
scale downdraft gasifier. In Symposium Papers, Energy from Biomass and
Wastes V. Institute of Gas Technology, Chicago, 1981, pp. 633-50.
4. Walawender, W. P., Chern, S. M. & Fan, L. T., Wood chip gasification in a
commercial downdraft gasifier. In Fundamentals of Thermochemical
Biomass Conversion, ed. R. P. Overend, T. A. Milne & L. K. Mudge.
Elsevier Applied Science, New York, 1985, pp. 911-22.
5. Graboski, M. S. & Brogan, T. R., Development of a downdraft modular skid
mounted biomass/waste gasification system. In Energy from Biomass and
Waste XI, ed. D. L. Klass. Institute of Gas Technology, Chicago, 1987, pp.
447-87.
6. Walawender, W. P., Chee, C. S. & Fan, L. T., Operating parameters influenc-
ing downdraft gasifier performance. In Energy from Biomass and Waste XI,
ed. D. L. Klass. Institute of Gas Technology, Chicago, 1987, pp. 411-15.
7. Otoma, S. & Gotoh, S., Appfication of the exergy concept to a resource
recovery system. Paper presented at Int. Recycling Congr., Berlin,
Germany, I October, 1979.
8. Shieh, J. H. & Fan, L. T., Estimation of energy (enthalpy) and exergy (avail-
ability) contents in structurally complicated materials. Energy Sources, 6
(1982) 1-45.
Mass and energy balance analyses of a downdraft gasifier 151

9. Ishimi, T., Shieh, J. H. & Fan, L. T., Thermodynamic analysis of a biomass


pyrolysis process. In Wood and Agricultural Residues, ed. E. J. Soltes. Academic
Press, New York, 1983, pp. 439-65.
10. Szargut, J. & Petela, R., Egzergia (in Polish). WNT, Warsaw, 1965.
11. Gaggioli, R. A., The concept of available energy. Chem. Engng. Sci., 16
(1961)87-96.
12. Riekert, L., The efficiency of energy-utilization in chemical processes.
Chem. Engng. Sci., 29 (1961) 1613-20.
13. Fan, L. T. & Shieh, J. H., Thermodynamically based analysis and synthesis
of chemical process systems. Energy, 5 (1980) 955-66.
14. Reid, R. C., Prausnitz, J. M. & Sherwood, T. K., The properties of gases and
liquids. 3rd edn, McGraw-Hill, New York, 1976, pp. 629-66.