Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
~ - l 21 ~60~ ~
This invention relates to solid fuel
products which are substantially exempt of slag
formation during combustion. More specifically, the
present invention relate~ to products such as pellet
5 f uel derived f rom wood wastes and which contain an
additive which prevent the formation of a slag
during combustion. The invention also relates to a
method for manufacturing solid fuel products, such
as pellets from bioma~s which comprises
10 homogeneously blending an additive into the biomass,
which is capable of altering the melt point
temperature of the slag forming ~mpurities which may
be present in the bioma~s.
For the past years, the pellet fuels
15 industry has managed steady controlled growth, and
little has changed in the makeup of the pellet fuel.
Some research work on additives for binding, BTU
enhancement by the sue of polystyrene, and water
resistance using hemicellulose (natural components
20 within the wood) were carried out in the early
1980's, when energy alternatives were the hot topic
of the day.
There is no doubt, mankind will one day run
out of hydrocarbons and go back to a carbon based
25 economy, deriving most of its chemical base feed
~tocks as it did 150 years ago. In like manner,
almost as a precursor to future energy needs, the
pellet stove started out as a solution to an
environmental need in the US Pacif ic Northwest Due
30 to air pollution from the emissions of older stick
stoves, the States of Oregon enacted ~egislation
(known today as the Clean Air Act across the USA)
banning their use, and leading to the advent of the
pellet burner, a clean, dependable non polluting
35 form of residential heating
- 2 _ 21~6l~l 0
In the beginning, pellet fuel manufacturers
did not have a problem with sawdust supplies. Few
board and pulp and paper plants used the discards of
the sawmill operations, and most of the waste
5 sawdust along with the bark was simply burnt or
landfilled. As the pellet industry grew and demana
for the raw materials increased from other
industries, prices began to rise and materials
became a valuable commodity.
Bark on the other hand is still plentiful,
as there are few uses for this high energy,
landfill, piled high, ground water polluting,
renewable resource.
Engineered fuel derivea principally from
15 bark feed stock sources could enjoy many advantages
over existing supplies of sawdust. In the Unitea
States, the average pellet fuel producer is paying
10 . 00 to 25 . 00 dollars US per short ton of bone dry
sawdust. In Canada the price varies according to
20 supply, demand factors related to geographical
location and winter energy demands Prices vary
from lO . 00 to 30 00 CDN per short ton delivered
There are many more perceived benef its to
pellet mill operations, such as decreased dependence
25 on non renewable and imported fossil fuels, the
support of regional economic activities and jobs.
Employment opportunities are created in fuel
handling and transportation, plant operations, ana
supporting services. In regions, expenditures for
30 biomass fuels, and other goods and services,
increase economic activity, and the use of lower
cost fuels helps keep energy costg low
There is thus a need to develop lower cost
feed stock sources for the production of residential
35 pellet fuels, and to address the subject of landfill
_ 3 _ ~15~
diversion of bark mill residues, which are known to
leach and pollute ground waters, lakes and streams.
In recent years technological advances in
board manufacturing, pulping processes and a need
5 for maximizing raw material usage, has led to
greater competition for traditional pellet mill raw
materials such as sawdust and planer shavings.
In 1994-95, the pellet industry produced
over 675, 000 short tons of pellet fuel worth an
10estimated 100,000,000 dollar and 6,000 airect and
indirect jobs.
The pellet fuels produced from over 70
operating plants throughout the United States and
Canada are already experiencing higher prices for
raw materials, as they compete with larger process
industries .
The pellet fuels industry last year
celebrated over 10 years of growth. ~very year in
which 50 to 60, 000 pellet stoves are sold, leads to
an approximate increase in demand of 150,000 to
180, 000 tons of wood pellets . There are over 30
manuf acturers of pellet 3toves in North America
which produced in excess of 60, 000 units for 1993-94
and 70,000 units ~or 1994-95. Of these, over 90
percent of the units being produced are the top feed
type designs, and these are considered to be the
most ash sensitive types. They are sensitive to the
oCcurrence of fusion of inorganic materials in the
burn pot, which degrade the stove perf ormance, by
restricting air flow to fuel As well, they were
designed around less than 1% or less ash content as
is the makeup of sawdust residues in the Pacific
Northwest and require more attention to maintenance
when greater percentages of ash in pellets are
burnt, having been designed with small ash pans.
_ 4 _ 2~ o
It is generally accepted in the industry, that the
pellet stove of the future should be user friendly,
and manufacturers have agreed that higher ash fuels
in the 3 to 596 ash range wlll be a reality in the
5 future. It is therefore imperative to lmprove the
pre-processlng technology of hlgh ash feed stocks by
altering the lnorganlc content.
As the pellet stove ls based on the forge
prlnclple, (hlgh alr to fuel ratlos) high
10 temperatures ln the burn pot area cause inorganlc
content of the resldue to fuse, resultlng ln the
formatlon of slag or cllnkers. If the melt polnt of
the inorganlcs could be altered or elevated, then
the lnorganic content of the fuel would simply
15 maintaln lts original form, fall through the burn
pot grate and into the ash pan. On the other hand,
bark ls a natural choice as an alternative pellet
fuel feed stock, but due to the aforementioned
clinker characteristics, it is not used ln the
20 majorlty of resldentlal pellet fuel stoves.
Fusion temperature lnf luences of a slngle
element on a combination of elements are complex
effects that can not be det~rm; n~fl with confldence.
Many factors play a dlrect role ln fuslon occurrence
25 and many more, though mlnor ln detall, can
ultimately have a profound effect from one burn
cycle to another In the case of lnorganlc
materlals, present in bark and to a lesser extent ln
sawdust, there ls a need to develop a successful
30 countermeasure.
A caref ul examlnation as to the makeup of
the feed stock is necessary in order to assess
whether or not addltlonal procedures could be
lmplemented so as to lmprove bark quallty. A
35 careful examlnation regardlng the composltion (l.e.
bark or sawdust percentages), manipulatlon, h;~nfll in~
_ 5 _ 21$61~1 0
from source to pellet mill, seasonal harvesting
practices, pellet appliance combustion conf iguration
and delivery of pellet fuel in relation to
combustion alr systems in combustion chambers, was
5 necessary in order to arrive at the ideal additive
f ormulation .
As opposed to reducing the highest
temperature the ash reaches, it has been decided
that better combustion perfnr~n~ne~ could be
10 achieved by increasing the fusion point temperature
of the inorganic material present in the feedstock,
used in the manufacture of the pellet fuel.
A review of the prior art indicated that it
is generally known to raise the fusing point of
15 materials that may otherwise form a slag when
burning the products on which they may be present.
For example, in U.S. 2,016,821, Nelms
discloses a process for treating coal by uniformly
applying to the surface of the coal a liquid mixture
20 of alumina bearing materlal such as bauxite, the
mixture having an alumina content of three to six
percent o~ the weight of the ash of the coal.
On the other hand, in an article published
by "Hazardous Iqaterials Control Resources Institute
25 (HMCRI)", Schofield et al generally disclose
introducing chemical additives by feeding them along
with materials to be heated, so as to increase the
fusion temperature of a mixture of inorganic
materials However, the general disclosure of
30 fusion temperature increase for slag forming
materials is not suf f icient to Yolve the problem
associated with these materials when present in fuel
pellets .
It is an object of the present invention to
3 5 provide a method which will enable to use f uel
- 6 - 215~Ql~
pellets even when the latter contain an excess of
slag forming impurities
It is another object of the present
invention to provide pellet fuels which do not slag
5 during combustion, thereby enabling to operate a
6tove without interruption.
In accordance with the present invention,
there i8 provided a method for manufacturing solid
fuel products of high bulk density, which are
10 substantially exempt of slag formation following
combustion in a combustion chamber. According to
this method, there is provided a comminuted biomass
material, the latter is treated to give a product of
increased bulk density which is converted into
15 bonded shaped such as by extruding the product into
pellet form having a bulk density of over 30 pounds
per cubi foot At least about 0 . 25 weight percent
of an additive capable of raising the fusion
temperature of slag forming substances that may be
20 present in the biomass material, above the
temperature in the combustion chamber when the
latter is in operation, is substantially
homogeneously blended into the biomass material
bef ore converting the product into bonded shapes
25 such as by extruding the biomass into pellet form.
Although any type of biomass material may be
used to produce pellets according to the invention,
wood fibre is preferred, as well as bark, or
mixtures of wood and bark, but may also include
30 refuse derived feedstocks such as cardboard, paper,
sludges, switch grass and agriculture waste such as
bagasse, shells or straws, etc
In accordance with a preferred embodiment,
the disintegrated biomass material is compressed
35 optionally in the presence of steam, until reaching
a desirea increasea bulk density, for example the
_ 7 _ 21 ~010
biomass may be contacted but not necessarily with
steam to provide a mixture with a moisture content
between about 12 and 18 weight percent, or the
biomass may already have this percentage, in which
5 case no addition of steam is necessary.
In accordance with another preferrea
embodiment, the additive may be selected among
suitable metallic compounds, preferably metal
oxides, such as aluminum, magnesium, manganese,
10 calcium, silicon or neodymium oxides. A mixture of
oxides is preferred, and the latter will also
include ammonium nitrate. A preferred additive is a
mixture containing calcium oxide, manganese dioxide,
magnesium oxide, aluminum oxide, barium oxide, iron
15 oxide, potassium oxide, silicates and ammonium
nitrate, for example about 0 . Z to 2 weight percent
calcium oxide, about 8 to 12 weight percent MnO2,
about 30-40 weight percent Nl14NO3, about 8 to 12
weight percent MgO, about 0 . 5 to 2 weight percent
20 aluminum oxide, about 0.1 to 0.5 weight percent
barium oxide, about 1 to 3 weight percent iron
oxide, about 0 . 5 to 1 weight percent potassium oxide
and about 30 to 50 weight percent SiO2.
In accordance with the invention, there is
25 also provided solid products of high bulk density
for combustion in a combustion chamber. The
products comprise a compressed biomass material in
particulate form, and at least about 0 . 25 weight
percent of an additive which is capable of ralsing
30 the fusion temperature of slag forming substances
that may be present in the biomass material, above
the temperature ln the combustion chamber when the
latter is in operation. The additive is
substantially homogeneously blended into the biomass
35 material, prior to forming bonded shapes such as a
pelletized fuel
- 8 - 21 a ~l O
.
Baseline evaluations were conducted on raw
materials containing between 2596-10096 bark in the
final densified form. Species of sawdust ana bark
were derived from various soft and hardwood species,
5 including black and white spruce, red oak and maple.
No binding agents were used at any time in
the manuf acture of the wood pellet samples tested .
Densified pellet fuels derived from bark
residues conf orms to the Pellet Fuel Institute
10 standards classification in all categories. It
identifies bark as a high ash fuel (over 1% ash
content~ usually between 2 and 4% is common with
this residue.
It should be noted that there is a direct
15 relationship between lower ash content pellet fuels
and premium prices per ton FOB the mill.
It should also be noted that recently, many
pellet appliance manufacturers in new models are
designing larger ash pans to deal with higher ash
20 fuels of the future. In so doing, they are
beginning to address the issue of ash volume,
however, the problem of ash fusion phen~ ?n~ still
exists. Increasing ash content of a pellet fuel is
directly proportional to increasing the risk of
clinkering or slagging in the burn pot from fusion.
Many examples can be sighted of low ash pellet
manuf acturers having what is termed a bad batch of
pellet fuel, sighting dirt getting into the
f eedstock .
The combustion tests were carried out on 3
types of pellet burners, top feed, bottom feed and
side feed systems. All units were placed in typical
residential installation situations. Burn cycles
were typical with low, medium and high draft and
feed system settings Carryover effects from one
test batch to another were minimi~ed during each
2l56ol~
g
burn cycle by cleaning grates and ash pans at the
completion of each test run.
Bark pellet f uels or mixtures of bark and
sawdust have been produced for over 12 years.
The pellet fuel industry through the pellet
fuels association has developed two pellet fuel
grade classification, based on volume of residual
ash. They are Premium, less than 1% ash content,
and standard, between 1% and 3~6 ash content grades.
This ash classification standard for pellet fuels is
based solely on ash quantities post combustion and
its effect on current system design. More
specifically it is based on the quantity of unburned
inorganic content of carbon in each pellet type
An additive consisting of about 0 . 2 to 2
weight percent calcium oxide, about 8 to 12 weight
percent ~nO2, about 30 to 40 weight percent ammonium
nitrate, about 8 to 12 weight percent aluminum
oxide, about 0.1 to 0~5 weight percent barium oxide,
about 1 to 3 weight percent iron oxide, about 0 . 5 to
1 weight percent potassium potassium oxide and about
30 to 40 weight percent SiO2 was initially blended
with industrial grade bark pellets derived from
black and white eastern spruce species. The pellets
were manufactured at the Energex plant at Lac
Megantic, Quebec. The additive wad applied to the
surface area of the pellets at a rate varying from 5
to 20 grams per 18 kg of pellets, blended together
then placed in an 18 kg hopper of an Enviro fire
EFll top feed pellet stove. This particular type of
combustion system was chosen, as it is considered to
be one of the more ash sensitive top feed pellet
burners on the market due to grate design.
The result of this pr~1 im;nilry study was to
evaluate the effectiveness of the additive as an
application onto, rather than into a finished
- 10 - 2l~6olo
product (pellet). Measured quantities of 5, 10, 15
and 20 gram portions of additive were mixed with 12
( 2 per test burn ) test sample of 18 kg of wood
pellets comprised of 100% bark.
~EST 1.
No ~ ;tiYe 100% ~rk r,c~ll~t
The pellet samples containing no additive on
high burn rate produced clinkers ~ solidif ication of
inorganic elements ) . Clinkering or solidif ication
and subsequent adherence to grate surfaces, of
solidified inorganic materials, resulted in
inadequate air to f uel ratios in the burn pot and a
reduction in overall stove performance, i.e. reduced
flame, reduced heat output, higher particulate
matter out venting pipe, and bl~rk~n;ng of stove
door glass. As pellet stove performance is directly
related to a 3 0 to 1 air to f uel ratio, a reduction
in air reaching the burn pot results in poor
perf ormance .
In this example, the only course of action
is higher maintenance by scraping the burner pot.
Steps required to improve stove performance are to
turn the stove of f, open stove door and scrape the
clinker build up off the grate surface. This had to
be done repeatedly, in order to maintain some degree
of stove performance. In addition to the obvious
end user inconvenience, the negative health effects
from indoor smoke spillage can cause serious health
problems. Pellet stove owners purchase the
appliance for the sake of convenience, and frequent
attention of this sort is a frustrating experience
and would result in a negative consumer attitude
towards this type of quality pellet fuel.
- ll - 21 ~ 1 0
TEST Z.
Rnrk ~ellet~ wi~h 5 g~ - of :..lA~v~3
18 kg of Energex wood pellets were blenaed
with 5 grasns of additive, using the same blending
procedure as in test 1 The Envirofire top feed
stove results post combustion were considered to be
the same as in step 1, requiring frequent cleaning
of clinker deposits by scraping the grate surface.
TEST 3.
0 R~rk PellQtS Wi~h 10 ~ ~ of ~ tive
18 kg of Energex wood pellets were blended
with 10 grams of additive, using the same blending
procedure as in previous tests . The Envirof ire top
feed stove results post combustion were considered
to differ from text 2. There was apparently less
clinker on grate surf ace areas, but attention to
grate surf ace areas was still accurate . Results
were almost identical.
TEST 4 .
R;~rk ~ell~t~ with 15 of AA-Iitive
18 kg of Energex wood pellets were blended
with 15 grams of additive, using the same blending
procedure as in previous tests. The result was
almost no clinker on grate surface areas at the end
of the burn cycle. Most of the inorganic silicate
(sand) simply fell through the grate and into the
ash pan. There was no scraping of grate surface
required, until the end of the combustion cycle
although some accumulation was present
It was concluded that there was a positive
impact on clinker formation when a blend of 15 to 20
grams of additive is used on surface areas of bark
wood pellets. The physical characteristics of the
ash are significantly altered. Although surface
applications gave rise to the fact that the additive
has some effect on clinker formation, tests revealed
- 12 - 215~10
varying end results. It is believed that this was
probably the result of uneven distribution of the
additive in the mix, static attachment of additive
on metal and plastic surface areas or a combination
5 of circumstances involving a3h melting due to a hot
spot near the point of air to pellet fuel
combustion .
Surface applications of additive to a
residential pellet fuel is not therefore desirable
10 due to varying end results. It has now been found
that the additive needs to get into the pellet to be
effective in providing greater exposure to inorganic
silicate. A homogeneous mix is therefore necessary
according to the invention and can thus only be
15 accomplished by blending the additive with the
feedstock prior to pelletization. The optimum point
of introduction of the additive in the industrial
manufacturing process of wood pellets is therefore
critical. A homogeneous blend of additive to wood
20 fiber achieved the desired result, as was
demonstrated in further testing.
TE~ 5.
Samples of bark residues were obtained from
the E3trie, Outaouais and Blainville areas of the
25 Province of Quebec, Canada. The samples were from
process industries such as veneer manufacturers and
sawmill operations. The bark samples varied in
concentrations, as the bark had some sawdust
present. Some of the samples were derived from
30 hardwood process industries while others were
derived from softwoods (board manufacturers). Bark
samples from, hardwoods appeared to contain more
inorganic material than the softwoods, probably due
to the uneven nature of the bark as compared to
35 softwood species. Quantities of sample material
were purchased in truck load quantities, as this
- 13 - 21~60
best represents the way in which a typical
pelletizing operation would receive and process the
same kind of material for further processing into
pe 1 let f uel .
Typical moisture content for green sawmill
residues varied, between 35% and 60% moisture
content . The material was received and of f loaded
into separate piles. The veneer bark (derived
principally from hardwood species) was then dried
down to a 10% to 12% moisture content with the use
of a heil rotary drum dryer having an average per hr
drying output of 8, 000, 000 BTU. The bark was then
commuted through a 150 E~P hammer mill (one quarter
inch screen mesh) 80 as to reduce the feedstock down
to a homogeneous size prior to pelletization. It
was at this point (hammer mill) that the additive
was first introduced at a rate of approximately 525
grams per ton of material (bark ) throughput on an
oven dry basis.
Approximately 12 tons of pellets were
produced and random samples taken at alternating
times throughout the production cycle. These
pellets were then burnt in the Envirof ire II pellet
s'cove at a rate of 18 kg per cycle.
The metering device was an agricultural
based f ertilizer metering system . Attached to the
hopper of the feeder was a small industrial vibrator
( so as to prevent caking of the additive ) . The
suction action of the hammer mill and attached
blowers drew the metered powder from the tube
leading f rom the bottom edge of the hopper meter
r-~h~sniqm (which controls the amount dispensed) and
into the commuted wood f iber .
Post combustion results varied, with some
3 5 test samples clinkering and others not clinkering
- 14 - 2~ o
and it was concluded that the quantity of additive
was insuf~icient to have an effect on the bark.
TEST 6.
Test 5 was repeated except that there were
usea 750 to 850 grams of additive per metric ton of
pellets produced. The pellets were burned in an
Earth stove side feed and a Harmon underfeed pellet
stove .
Each pellet burner was tested on low, medium
and high burn settings for 10 burn cycles (189 kg of
fuel each) which was more representative of a one
week burn for an average household, in an average
winter period. The burn levels were also more
representative of a typical day and night cycle heat
demand over a perlod of time, in this case 1 week
with f luctuations in therm demand.
BTU output per stove was in the range of
between 20,000 to 40,000 BTU per hour on the various
settings .
There was virtually no fusion occurrence or
accumulation on grate surfaces. Ash volume was
high, due to an ash level of 2, 25% . But this
quantity was manageable and required 3 cleanings
(removal of ash from the ash pan) for the Envirofire
EF II as opposed to 1 cleaning per week with less
than . 5% ash for the highest grade premium type
f uels .
The Earth and Harmon systems required in
real terms 1.5 cleanings, as ash pan size and area
are considerable in these types.
It has therefore been est~hl i ~ilG~ that when
at least about 0 . û25 weight percellt of additive are
homogeneously blended into the fuel pellets, no
substantial slag formation takes place.
Although the tests were made with an
additive based on MnOz MgO and NE~4NO3 f or
- 15 - 215~
convenienGe, it is understood that any additive
capable of raising the fusion points of impurities
present in biomass material is within the spirit of
the present invention.
