United States
Environmental Protection
Agency
Air and Energy Engineering
Research Laboratory
Research Triangle Park NC 27711
Research and Development
EPA/600/S7-89/006a Jan.1990
f/EPA Project Summary
Evaluation of FGD Dry Injection
Sorbents and Additives:
Volume 1. Development of High
Reactivity Sorbents
Wojciech Jozewicz and John C. S. Chang
EPA's efforts to develop low cost,
retrofit flue gas cleaning technology
include the development of highly
reactive sorbents. Recent work
addressing lime enhancement by
slurrying with siliceous materials and
testing in a laboratory packed-bed
reactor is discussed in this report
The solids generated from a
furnace ilmestone injection process
were reactivated by slurrying at
elevated temperatures. Compared
with untreated solids, reactivity
toward SO2 was significantly
enhanced by hydration. The SO2
capture by solids Increased with
increasing time and temperature of
hydration. The SO2 capture was
probably enhanced by the calcium
aluminate silicate hydrates formed
during the slurrying process.
In addition to flyash, silica from
alternative sources was reacted with
lime. The dry sorbents produced by
slurrying several dlatomaceous
earths, or montmorillonitlc clays, with
lime were found to be highly reactive
with SO2 The most reactive sorbent
was generated by slurrying silica/lime
at a weight ratio of 1 to 1. The
morphology of the developed
sorbents was characterized.
Several additives were tested to
evaluate their potential to promote
the lime/silica reaction rate and in-
crease sorbent reactivity. Of those
tested, NaOH, Na2HPO • 7H2O,
(NH4)2HPO4, and H3PO4 were found
effective in enhancing the flyash/lime
reaction. The maximum enhancement
effect was obtained by using 4 to 8
mol % additives.
Pressure hydration of flyash and
lime fostered the formation of a
reactive sorbent much quicker, and
used less flyash, than did
atmospheric hydration. The reactivity
of the sorbent in the packed bed
reactor correlated well with B.E.T.
surface area, increasing with
increasing surface area. The
optimum temperature range for the
pressure hydration of flyash with lime
was between 110 and 160°C.
This Project Summary was
developed by EPA's Air and Energy
Engineering Research Laboratory,
Research Triangle Park, NC, to
announce key findings of the research
project that is fully documented In a
separate report of the same title (see
Project Report ordering information at
back).
Introduction
Highly reactive sorbents are needed for
the dry sorbent injection flue gas
desulfurization (FGD) process. The dry
sorbent injection concept is very
attractive, especially for retrofitting
existing power plants, because of its
technical simplicity and low capital cost.
However, due to limited space and the
high velocity of flue gas, an extremely
short gas/solid reaction time is available
for S02 absorption. As a result, very low
-------
utilization (less than 30%) was obtained
by using a conventional sorbent such as
lime. Better sorbent utilization could be
achieved when the sodium based
sorbents, such as nahcolite and trona,
were used. Sodium salts generated from
FGD processes possess much higher
solubility in water (12.5 g of Na2S03 per
100 g of cold water) than do calcium salts
(0.0043 g of CaS03 • 1/2H20 per 100 g
of cold water), and therefore may
constitute the environmental hazard of
leaching salts into ground water if the
waste is disposed of in landfills.
Calcium silicate hydrates produced by
pozzolanic reaction between silica and
lime in water at elevated temperatures
are very attractive calcium based
sorbents for dry injection processes.
When prepared in appropriate forms,
these hydrates are very reactive with S02
and should result in good sorbent
utilization. Previous study has shown that
reactive calcium silicate hydrates can be
produced by slurrying flyash with lime at
temperatures below 100°C (atmospheric
hydration), but relatively high flyash/lime
ratios (greater than 3) and long reaction
times (longer than 8 h) are required. For
commercial application this method might
not be viable. A high flyash/lime ratio
would increase flue gas duct loading and
overload the particulate control device.
The long reaction time would increase
the reactor size and capital cost. Goals of
this work included: evaluation of potential
techniques of reducing the flyash/lime
ratio as well as the reaction time, and
applying the concept of using silica/lime
sorbent to reactivate solids from furnace
limestone injection processes to enhance
overall SO2 removal and promote waste
utilization.
The major experimental apparatus
used was a laboratory-scale packed bed
reactor. The glass reactor (45 mm
diameter, 150 mm long) was packed with
powdered sorbent mixed with 70-mesh
sand to prevent channeling. The
reactivity of various sorbents was
measured by the quantity of SO2
absorbed under typical dry injection flue
gas conditions. Reagent grade Ca(OH)2
was used as the baseline sorbent.
Calcium silicate hydrates were prepared
by pressure hydration or by slurrying
siliceous materials with lime in a stirred
beaker immersed in a thermostated water
bath. After mixing, the samples were
vacuum filtered and dried before testing
in the reactor. A stainless steel high
pressure 300 ml vessel was used to
prepare the calcium silicate hydrates at
temperatures above 100°C (pressure
hydration). The high pressure vessel was
heated electrically and its temperature
controlled by a thermocouple. Sorbent
reactivity was calculated in terms of SO2
capture and lime conversion. S02
capture was defined as the amount of
S02 absorbed per unit weight of sorbent
(mmol S02/g). Lime conversion was the
percent of lime in the sorbent which
reacted with the absorbed S02 assuming
that CaSO3 was the product.
Reactivation of Boiler
Limestone Injection Solids
The objective of this segment of work
was to apply the pozzolanic reaction
concept to Boiler Limestone Injection
Solids (BLIS). Their potential for
producing reactive sorbent to increase
overall S02 capture and sorbent
utilization was evaluated.
Due to the low sorbent utilization of
boiler limestone injection processes,
BLIS usually contains significant amounts
of unreacted lime. Other ingredients
include calcium sulfate, limestone, and
flyash. Six samples of BLIS, produced in
1981 by boiler limestone experiments,
were tested. The untreated BLIS showed
virtually no S02 removal capability in the
packed bed reactor. Hydrating the BLIS
at elevated temperatures for a prolonged
period of time was tested as a means of
reactivating these solids.
For the BLIS samples hydrated at 65
and 90°C over 1 to 8 h intervals, the S02
capture increased with increased time of
hydration, but leveled off at about 1.8
mmol S02/g. X-ray elemental analysis of
hydrated samples showed Al, S, and Ca
peaks, suggesting that ettringite or other
synthetic calcium aluminate sulfate
crystals were formed during hydration. X--
ray powder diffraction analyses
confirmed the existence of those crystals
and further suggested that they were
hydrated crystals with a high water
content. A significant increase of solid
surface area by hydration was measured
by B.E.T. analyses. The initial value for
untreated BLIS was 2.67 m2/g. The sur-
face area of hydrated BLIS reached as
high as 50.86 m2/g. The surface area
increased with both hydration time and
temperature. A good correlation was
obtained between the developed surface
area and the reactivity of the hydrated
BLIS (see Figure 1).
Screening of Alternative
Sources of Silica
The objective of screening alternative
silica sources was to evaluate the
reactivity of various siliceous materials
with lime. The siliceous materials
evaluated included cryptocrystallim
forms of silica such as diatomaceou
earths, tripolis, and pumices. Severe
grades of naturally occurring bentoniti
clays and kaolins were also tested. Eac
sample of the siliceous materials teste<
was hydrated with lime at 90 °C. Th
reactivity of the hydration product wa
evaluated in the bench-scale packed be
reactor.
The hydrates produced from th
natural grade diatomaceous earths wer
more reactive than those from calcine
earths. The calcination of diatomaceou
earths used in commercial products
especially when sodium salts were use
as an additive, caused by an SO"!
decrease of B.E.T. surface area. Sodiur
based additives used during calcinatic
probably caused sintering at the hig
temperatures (above 900°C) in the kill
producing siliceous slag on the surface i
diatomaceous earth similar to the glas^s
layer on the surface of flyash particles. *
The hydrates produced from naturall
occurring bentonitic clays were als
found to be more reactive than thos
from treated or "activated" clay
Hydrates from kaolins generall
performed more poorly than bentonit
clays.
Parametric tests were conducted wi
the two most promising alternate
siliceous materials-MN-53 (a diati
maceous earth) and Bentonite 149
bentonitic clay). The most reactiv
hydrates were produced by reactir
siliceous materials with lime at
silica/CaO ratio of about 1. Most likel
the calcium silicate was formed durii
the hydration process according to tl
hypothetical reaction:
Ca(OH)2 + Si02
H2O
Assuming the relative reactivity towa
S02 being in the order CaSiO • H20
Ca(OH)2 > SiO2i only at equimolar, cc
dition can the greatest amount of tl
most reactive hydrates be produced.
ray powder diffraction analysis confirm
that Ca(OH)2 and Si02 disappeared afi
hydration. The less unreacted t!
Ca(OH)2 was, the more reactive were t
hydrates.
Effects of Additives on Sorbei
Reactivity
The objective of this segment was
evaluate the effectiveness of additives
enhancing reactivity of hydrat
produced by flyash/lime slurrying. T
additives tested included sodii
hydroxide, sodium phosphate, amn
-------
2.5
2.0
I
I
co
I
CO
7.5
7.0
0.5
8/JS A/0. 7
o
D
A
O
V
Temp.
°C
90
65
45
25
Various
Additive
None
None
None
None
3% (NHJ2 HP04
7 8 9 10 75 20
B.f. 7. Surface Area, m2/g
30
40 50
Figure 1. Correlation between the surface area developed during hydration and S02 capture
by BUS No. 1 in the sand bed.
nium phosphate, and phosphoric acid.
Sodium hydroxide has the potential of
enhancing the reaction rate between
flyash and lime by increasing the
solubility of silica. Phosphates could
promote the flyash/lime reaction by
increasing the flyash dissolution rate
resulting from their ability to attack the
glassy layer on flyash particle surfaces.
Sodium hydroxide tests were
performed at a flyash/Ca(OH)2 ratio of 2.3
to 1. The enhancement effect was
reflected by an increase of lime
conversion from 35 to 50% when
comparing the reactivity of hydrates
prepared with and without NaOH addition.
The maximum enhancement effect was
obtained with a dose of 4 mol % of NaOH
additive (based on Na/Ca ratios). The
higher the NaOH concentration in the
flyash/lime slurry, the more pronounced
e enhancement effect (Figure 2).
Three phosphates, Na2HP04 • 7H20,
(NH4)2HPO4, and H3P04 were tested. All
data showed that the maximum
enhancement occurred when a dose of 4
to 8 mol % phosphate (based on PO4/Ca
ratios) was used to slurry the flyash with
lime. When CaO instead of Ca(OH)2 was
slurried with flyash, the amount of
phosphate needed to reach the same
degree of reactivity was 50 and 70%,
respectively, when (NH4)2HP04 and
Na2HP04 • 7H2O were used. For exam-
ple, a maximum lime conversion of 50%
was reached when 4 and 8 mol % of
(NH4)2HP04 respectively, were added
during the slurrying of flyash with CaO
and Ca(OH)2. The result may stem from
the dispersing properties of phosphates,
present during the hydration of CaO,
which create finer particles in the product
compared to those obtained using
Ca(OH)2.
Pressure Hydration Evaluation
Pressure hydration was evaluated as a
way to produce calcium silicate hydrates
for dry SO2 control. The focus was on the
pressure hydration of flyash or
diatomaceous earth with Ca(OH)2, aimed
at reducing the time of hydration and the
weight ratio of siliceous materials to lime.
Pressure hydration is used
commercially to hydrate dolomitic lime
(CaO + MgO), as MgO usually is difficult
to hydrate completely within a reasonable
period of time under atmospheric
conditions. Pressure hydration allows the
use of high temperatures (above 100°C)
and high pressure (above 1 atm) to
accelerate the reaction rate. For the
silica/lime reaction, the high temperature
of pressure hydration increases the silica
solubility and enhances the pozzolanic
reaction rate.
-------
0.3
0.2
0.1
0.1 ON NaOH
0.2SNNaOH
I
I
I
4 6 8 10 12 14
Amount NaOH added, mol percent
16
18
20
Figure 2. The effect of NaOH addition during slurrying on the reactivity of dry sorbent in the
sand-bed reactor (preparation conditions: slurrying at 90"C for 8 h; sand-bed
reactor conditions: 60% relative humidity, 64.4° C temperature, and 500 ppm SO2
concentration).
Comparing atmospheric hydrates with
pressure hydrates, the data clearly
showed that a significant decrease of
time of hydration is possible when the
hydration temperature is increased. For
example, atmospheric hydration at 90"C
for 8 h was required to enhance the
conversion of Ca(OH)2 up to 35%, when
the weight ratio of flyash to Ca(OH)2 was
2.3 to 1. The same conversion was
achieved for the same reactants and
weight ratio after they had been pressure
hydrated at 150°C for 1 h (Figure 3).
Pressure hydration could also reduce the
flyash/lime weight ratio requirement. At a
flyash/lime ratio of 1, the product of
pressure hydration at 120°C for 2 h
showed 45% lime conversion. The
atmospherically hydrated product at the
same flyash/lime ratio never achieved the
45% lime conversion.
The present work also included
pressure hydration of diatomaceous
earth/CaO slurry at 150 and 230°C for
about 1 h. No significant increase in
reactivity (lime conversion) was achieved
by pressure hydration of diatomaceous
earth with lime.
Parametric tests were conducted to
pressure hydrate the lime with flyash.
There is an optimum hydration time for
each hydration temperature. At 230°C,
the maximum reactivity (lime conversion)
was obtained with about 1 h hydration of
flyash/lime at a ratio of 2.3. The optimum
hydration time became 4 h at 180°C, and
5 h for 150°C. When the flyash/lime ratio
decreased from 2.3 to 1, the optimum
hydration time shifted from 5 h to 1 h at
150°C.
The reactivity of pressure hydrated
flyash/lime correlated quite well with
B.E.T. surface area. Higher surface area
hydrates, generally produced at medium
temperatures (150-180°C), optimum
hydration time, and relatively high
flyash/lime ratio (e.g., 2.3), usually
possess greater reactivity toward SOa
The crystal morphology of the pressure
hydrated flyash/lime mixtures was
examined by scanning electron
microscope (SEM) and x-ray diffraction.
Compared with B.E.T. surface area and
reactivity data, two factors were found
necessary for the hydrates to readily
react with SOg under conditions
encountered in a dry injection FGD
process: large sorbent surface area and
amorphous surface structure. Usually,
highly reactive hydrates have a large
surface area and a gel-like amorphous
crystal surface. For one hydrate sample
prepared at 230°C, even though the
measured B.E.T. surface area was high
(17 m2/g), the reactivity was poor (16
lime conversion). SEM pictures show
that, instead of gel-like amorpho
materials, a framework of distinct need
shaped crystals was found. This findi
seemed to indicate that the reactivity
hydrates was a function of crys
structure, which could be affected
hydration temperature.
Conclusions
Experimental results from this stu
indicate that the waste solids from
boiler limestone injection process can
reactivated by atmospherically hydrati
them under conditions which permit I
reaction of calcium with silica to fo
large surface area hydrates. New cryst
(calcium/aluminate/silicate/sulfe
hydrates) were found in the reactival
BLIS samples. The reactivity of the
hydrated BLIS samples increases w
the time and temperature used
hydration, and correlates well w
surface area; i.e., increases w
increasing surface area. The reactivity
these hydrated BLIS samples is sensil
to the relative humidity of flue gas, v
the S02 reactivity increasing relat
humidity.
Siliceous materials from sources ot
than flyash were found effective
-------
O
I
0.5
0.4
0.2
m 150°C
• iao°c
+ 230°C
0.5 1
4
Time, h
Figure 3. Combined time and temperature effects of pressure hydration on the reactivity of
sorbent prepared at the weight ratio of flyash to Ca(OH)2 of 2.3:1; sand-bed reactor
conditions: relative humidity 60%, temperature 64"C, SO2 concentration 500 ppm;
pressure hydrator conditions: temperature 150"C, 2h.
enhancing the reactivity of the sorbents
produced by slurrying with lime at
elevated temperatures. Among the
siliceous materials tested, natural grade
diatomaceous earth and bentonitic clay
were found to be the most effective in
producing reactive hydrates with lime.
The most reactive hydrates were
produced by slurrying siliceous materials
with lime at a silica/CaO ratio of about 1.
It is postulated that calcium silicate
hydrates were formed during the
slurrying process. The high reactivity of
the slurrying product can be partially
attributed to the large surface area of
calcium silicate hydrate and its high
water retaining capability.
Additives such as sodium hydroxide,
sodium phosphate, ammonium
phosphate, and phosphoric acid can be
added to the flyash/lime slurry to
enhance the reactivity of the produced
hydrates. The maximum reactivity
enhancement effect was obtained by
using 4 to 8 mol % (versus Ca) additives.
Very reactive sorbents for dry injection
S02 removal from flue gas can be
produced by pressure hydration of flyash
with lime. Pressure hydration fosters the
formation of hydrates from flyash/lime
slurry in much less time than does
atmospheric hydration. Hydration time,
temperature, and flyash/lime ration are
the three major parameters which affect
the reactivity of the hydrates. Matrices of
optimum pressure, hydration time,
temperature, and flyash/lime ratio for dry
injection S02 removal purpose are
needed for commercial application. SEM
and x-ray diffraction studies demon-
strated the formation of hydrates of
different morphologies by using different
flyash/lime ratios and changing the
conditions of pressure hydration. Both
high B.E.T. surface area and amorphous
surface structure are necessary
characteristics for the hydrates to be
reactive with SO2. Although hydrates with
well-defined, needle-shaped crystals
have a large surface area, they are not
reactive with SO2.
-------
W. Jozewicz and J. Chang are with Acurex Corp., Research Triangle Park, NC
27709.
Charles B. Sedman is the EPA Project Officer (see below).
The complete report, entitled "Evaluation of FGD Dry Injection Sorbents and
Additives: Volume 1. Development of High Reactivity Sorbents," (Order No. PB
89-208 920/AS; Cost $23.00, subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Air and Energy Engineering Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
EPA/600/S7-89/006a
US, OF
/ (SC S300
/* *
/•> i! METER
609044^
FICIALMAIL
U.S.POSIAG!
s 0 3 5 =
.
CHICAGO
------- |