Rate of reaction of nitric oxide and oxygen by metal sulfides
by Kent Moroni Hodgson
A thesis submitted in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE
in Chemical Engineering
Montana State University
© Copyright by Kent Moroni Hodgson (1975)
Abstract:
Nitric oxide has received a great deal of attention as- one of the major air pollutants- with which
environmentalists are concerned. Nitric, oxide is emitted into the atmosphere -where it can then Be
oxidized to NO2 in the presence of sunlight and oxidizing agents. No. commerically acceptable method
for removing NO from exhaust and flue gas has been developed yet.
This research is concerned with the rate of two reactions.
MeS + 4N0 → MeSO4 + 2N2 MeS + 202 -> MeSO4 The rates of reaction at 300°C, 4000C, 500°C for
these two reactions were determined for ten metal sulfides, The rates of reaction were studied using a
Cahn R-100 continuous recording electrobalance. This device was used to measure the weight of a
sample continuously as it hung suspended from one arm of the balance into the reactor. The rate of
reaction was calculated, from the continuously recorded weight increase. The rate of reaction of NO
with metal sulfides was determined using a gas mixture with a composition of 2.5% NO and 97.5% He.
The rate of reaction of O2 with metal sulfides was determined using a gas mixture with a composition
of 2.5% O2 and 97.5% He, In general, the reaction of the metal sulfide with oxygen proceeds faster for
all temperatures tested. The reaction rates with NO ranged from 3.38 x 10“° grams of FeSO4 formed
per minute per gram of FeS to 2.076 x 10^-4 grams of FeSOlt formed per minute per gram of FeS. The
reaction rates with O2 ranged from 5.9 x 10^-6 grams of ZnSO4 formed per minute per gram of ZnS to
6.48 x 10^-4 grams of PbSO4 formed per minute per gram of > PbS8 The sulfides of Ba, Fe, Sr, Cd,
Pb, Ca, Zn and sulfurated potash -reacted with both the O2 and the NO to form the sulfate.
The relative reaction rates of NO with a given amount of ZnS or SrS at 300°C and a given amount of
BaS, SrS or CdS at 400°C for a given time are greater than the relative rates of reaction of O2 with the
same amount of BaS, SrS or CdS at 4000 and the same amount of ZnS or SrS at 300°C for the same
time.
Cupric sulfide, thallium sulfide, manganese sulfide, and molybdenum disulfide could not be used
because they decompose below- 300°C.
Tungsten disulfide when reacting with NO and 02 produced a net weight lost indicating that
undesirable side reactions were controlling, Sulfurated potash produced the fastest reaction rates with
NO and O2. However, since the composition was unknown, rates of reaction of the solid could not be
calculated. STATEMENT OF PERMTSRTOW TQ GOPX
In presenting tills- th.esIs- In partial fulfillment of the requirements
for an advanced degree.at Montana State University, I agree that the
Library shall' make.it freely available'for inspection.
I further agree
that permission for extensive copying of this thesis for scholarly
purposes m a y be granted b y my major professor, or, in his absence, b y the Director of Libraries.
It is unde r s t o o d 'that a n y copying or
publication of this thesis for financial gain shall not be allowed
without m y written permission.
Pa^e
Itfj i97ST
RATE OF REACTION OF" NITRIC OXIDE AMD- OXYGEN
BY METAL SULFIDES
- Dy
KENT MO RONI HODGSON
A th.es.is. siibmitted in p a r t i a l .fulfillment
of the requirements for the degree
of
MASTER OF SCIENCE
in
Chemical Engineering
Approved:
H e a d 7 Major DepartS^nt
C h a irman, Examining Committee
MONTANA STATE UNIVERSITY
Bozeman, Montana
December, 1975
iii
ACKNOWLEDGMENTS
The author wishes to thank the faculty and staff of the Department
of Chemical Engineering for their suggestions and assistance,
"The
author wishes to thank Dr. F,P. M c Candless, his advisor, for his help
and suggestions.
The author wishes to thank his wife for her support
and help in preparing this thesis.
Financial support for this study was provided "by the Environmental
Protection Agency, Grant No. R-800682-03-0„.
iv
TABLE OF CONTENTS
Pase
VITA.
. . . . . . . .
II
ACKNOWLEDGMENTS. ......
LIST OF TABLES.
. . .
LIST OF FIGURES
0 *
ABSTRACT.
•.e.o.o;.,.,.......
. . e e e . . . .
0
0
0
6
0
0
-
0
0
6
0
«*«
*
o
a
• Ci
T
o . a
&
. . . . . . . . . . .
INTRODUCTION.
#**
. .
PREVIOUS WORK.
' vi
vil
* o * o * * a #
a*
. . . . . . . .
- I
o
-
e
«
e
«
e
o
e
e
REDUCTION W I T H METAL SULFIDES.
O B J E C T I V E S . .................. ..
APPARATUS . O • «. • O • •
Ill
k
. .
* O O # C
I
8
6
O * O
9
PROCEDURE
13
RESULTS AND DISCUSSION
14
.22
CONCLUSIONS . . . . . .
RECOMMENDATIONS
e
LITERATURE CITED
o
o
o
o
o
•
o
o
a
e
o
24
o
o
o
o
o
o
o
a
e
o
o
o
'
a
o
o
o
o
'
25.
•V
LIST OF TABLES
Table
I
■ "Page
. Free E n e r g y CKangeB and Heats of Reaction for the
Reduction of NO Employing Metal Sulfides
7
II
Reaction Rates of NO and O^ w i t h Various Metal S u l f i d e s ,
l6
III
A Summary of the Weight Gained or Lost for the '
Various Sulfides Reacting w i t h NO and
19
IV ■
The Relative Rates at which NO and O^ React, .
■ with the Various Sulfides „
.
21
LIST OF FIGUKES
Figure
Floiff Diagram of Apparatus for Measuring Rate '
of Reduction of H O ...................... .. . . .
Reactor Cross-Section=
3
. . . . ..............
. . .
A Sample Recording from the Cahn R-IOO Continuously
Recording Electrohalance . . ............ .. .. = .
VdLi'
ABSTRACT
Nitric oxide lias received a great deal of attention as- one of the
major air. pollutants- WitIi whicA.. environmentalists:'are concerned.. Nitric,
oxide is emitted into the atmosphere where it can then he.oxidized to NO^
in the presence of sunlight and oxidizing agents. No. commerically
acceptable method for removing NO from exhaust and flue' gas has been
developed yet.
'
This research is concerned with the rate of two reactions.
MeS + 4N0 ->•
MeS + 20,
MeSO l, + 2N2
MeSOlt
The rates of reaction at SOO0C 5 ^OO0C 5 500°C for these two reactions
were determined for ten metal sulfides„
The rates of reaction were studied using a Cahn. R--100 continuousrecording electrobalance. This device was used to. measure the weight of
a sample continuously as it hung suspended from one arm of the balance
into the reactor. The rate of reaction was calculated, from the
continuously recorded weight increase. The rate of reaction of NO with
metal sulfides was determined using a gas mixture with a composition of
2,5% NO and 97.5% Ne'. The rate of reaction of Og with metal sulfides was
determined using a gas mixture with a composition of 2.5% O^ and 97.5% He,
In general, the reaction of the metal sulfide with oxygen proceeds
faster for all temperatures tested. The reaction rates with NO ranged
from 3.38 x 10“° grams of FeSOit formed per minute per gram, of FeS to 2.076
x 10“ ^ grams of FeSOlt formed per minute per .gram of F e S . The reaction
rates with Og ranged from 5 =9 x: 10-6 grams of ZnS0% formed per minute per
gram of ZnS to 6 .48 x 10“^ grams of PbSOll formed per minute per gram of
>P b S 8
The- sulfides of Ba, Fe, Sr, Cd, Pb, Ca, Zn and sulfurated potash
-reacted w i t h b o t h the Op a^d the NO to form the sulfate.
The relative reaction rates of NO with a given amount of ZnS or SrS
at 300°C and a given amount of B a S , SrS or CdS at 400°C for a given time
are greater than the relative rates, of reaction of Op w i t h the same amount
of B a S , SrS or CdS at UOO0 and -the same amount of ZnS. or SrS at .300°C forthe same t i m e ..
"■
- Cupric sulfide, thallium s u lfide, manganese sulfide, and molybdenum,
disulfide could not be. used because they decompose b e l o w 300°C.;
Tungsten disulfide when reacting w i t h NO and Og produced a net weight
lost indicating that undesirable side reactions were controlling,
Sulfurated p o t a s h produced the fastest reaction rates with NO and Op.
h o w e v e r , since the composition was unknown, rates of reaction of the' solid
could not be calculated.
.
.
IETRODUCTIDH
The elimination of EO^. as an air pollutant has. Taeen a major concern
of environmentalists since.it was established in 1952 that EO
in smog reactions /( B a r t o k .e t .a l „ , 1 9 7 1 1 °
participates
EO^ represents- the two oxides
of nitrogen, E O (nitric oxi.deI and EOgCnitrogen d i o x i d e } „
The average
U 0S. urban concentration O i h i t r o g e n oxides in air is 20-25 times the
natural atmospheric c o n d i t i o n ( H o p p e r and Taws, 197*0*
On a nationwide
basis about h a l f the EO^ comes from stationary sources and about half
from mobile sources (Science E e w s 5 1972)„
. Most of these oxides are released In the form o f E O .
The conditions
for the formation, of these two oxides are quite different „
In h i g h
temperature combustion processes EO is almost exclusively formed (Bartok
et a l . , 1971) •
H o w e v e r , at the ambient temperature of the atmosphere, the
equilibrium between E O g ,
and NO highly favors EO^.
PREVIOUS WORK
There has been a great deal of w o r k done concerning oxides of
nitrogen.
The author does not intend to give a comprehensive summary
of all the w o r k done.
This is only a summary of some of the w o r k that
seems to offer promise of being significant- in the-control of EO .
X•
Shaw (1973) in.studying the reduction of nitrogen oxide emissions
from a gas turbine combustor b y fuel modification fou n d that soluble
■
organo metallic.additives w h i c h became heterogeneous reduction or de­
composition catalysts would reduce the EO^ emmission b y 30%.
A number
of other additives were experimented with; however, the. organo metallic
■
2
additives produced the "best results.
The major.drawback "of these addi­
tives is; that it seems to "be a case of trading one. pollutant for another»
The Bell Lahoratory h a s developed a manganese, rare earths, lead on
a ceramic support catalyst.
This i s m a i n l y for use i n .automobiles.
Good
results have "been obtained, h o w e v e r , a small amount of H H^ is p r o d u c e d .
In experimenting w i t h this catalyst their experience showed that w h e n CO
and hydrocarbon emissions were controlled effectively the NO yield in­
creased.
A n d w h e n the NO was controlled effectively the CO and hydro­
carbon y ield increased..[Scientific American, 1973).
A dual bed catalyst system for the simultaneous reduction of SO^ and
NO has been developed b y Sood and Kittrell' (.197^)»
CO + NO
2C0 + SOg
The reactions:
-*• COg + 1/2 Ng
2C0g + 1/2 Sg
CO + 1/2 Sg ■-»•« COS
2008 + SOg
3/2 Sg + 2C0g
are considered to take place in this system.
To get 90% removal stoichio
metric quantities are necessary and the catalyst, b e d temperature must be
controlled very carefully..
The reduction of nitric oxide with various hydrocarbons- as studied
b y Ault and Ayen (1971). has shown that in general, an. increase in carbon
number in the hydrocarbon.studied resulted in a decrease in the required
temperature for a given nitric oxi.de conversion.
For -a given carbon
number the required temperature for a given nitric oxide conversion
3
decreased Tixtlx degree of saturation„
Th.e catalyst u s e d Tsras a tarIum
promoted copper chromite catalyst <> '
Good results were obtained "by- reducing nitric oxide u s i n g a copper
nickel catalyst.
■
The' results indicated tlxat the catalyst' activity w as
dependent on the' copper-nickel ratio.
w i t h increasing copper content„
In general, activity increases
'■
Activation of the catalyst in CO at
500°C led to the opposite trend in activity; that is activity increased
with an increase in nickel content.(Bauerle et a l , , March, 1974).
Ammonia- has b e e n the only reluctant purported t o show true select­
ivity for the heterogeneous reduction of nitric oxide t o nitrogen in the
presence of excess oxygen„
•
A study b y Bauerle et a l . (December, 1974)
indicates that the catalytic reduction of NO w i t h NH^ on Pt in a simulated
plant exhaust is not strictly selective in the sense that, the NH^-O
reaction has little appreciable effect on NO reduction.
Nitrous oxide
(NgO) is a major reaction product and is produced b y the reduction of
b o t h NO and O^ w i t h NH^.
Regenerative sorption of nitric oxide has been found to w o r k good for small concentrations of NO (.Gidaspow and Onischak, 1973).
The N O is
sorbed w i t h FeSO^.
FeSO1^ + NO
FeSOj^-NO
This reaction can h e reversed with, h e a t .
The main drawback, is that the
flue gas and exhaust gases must be cooled below. IOO0C.
A fluidized be d of catalyst to eatalytically reduce nitric oxide
-
h
has been studied. Using a platium. silica almnnia cataiys,t a conversion
efficiency of over 99% Ttas obtained.
This' w as using a.stream of .pure U O
(.Dieteven et al,, 1973).
'
REDUCTION 'W X T E 'METAL 'STlLi1IXiES
.
It w a s shown b y White (.1973)' i n i tially that m e t a l sulfides will '
reduce nitric oxide.
Complete reduction was attained over a temperature
range of UOO0C to SQO0C,
White also successfully l o w e r e d the tempera­
ture for the reduction of nitric oxide b y the •addition of various chemi­
cals , the temperature range was from UOO0C to 550PC„.
It was also
shown that NO wo u l d b e reduced in the presence of O g .a n d that the
presence of water vapor did not appear to deter the reduction of NO.
White also determined that in the reduction of NO w i t h calcium sulfide
the solid product was 80 weight per cent calcium sulfate.
Erickson (197U) determined that the best support material for CaS
was Harshaw 1602 . 1/8 inch pellets.
He also tested Na l c o 2910-E 1/8
inch, Alcoa T-71 l/U inch and Linde TM-O-IllU pellets®
The Harshaw l602
gave good reduction of NO without forming any HgS or S O g .
In testing NiS as a reducing agent Erickson found that SOg.was
produced along with, the reduction of NO.
In working with, contaminants .
in the NO gas stream it w a s found b y Erickson that E g O produced small
amounts of HgS and decreased the reduction of NO.
The presence of E g
seemed to increase the reductions of NO b y CaS .but caused the formation
of H g S .
Natural gas also produced HgS but did not s e e m to effect the'
5
reduction of CaS„
He also found that Og and COg d f d not effect the re­
duction of CaSo
"
M c I n t y r e . (197^1 in studying the reaction
casW
+ Uh o Gg )
CasojiCs) + 2N Gg)
-
found that the global rates for the reduction of C a S o n h i g h alumina
Harshaw pellets increased from «25 x I O ^
at 390°C t o
x I O ^ moles
of CaSO^ formed per hour per gram of pellet at U93°'C.
Linde molecular
sieves; gave g r e a t e r 'average rates„
Rates varied b e t w e e n
.32 x !O'"
I
and .64 % 10 "■ moles CaSO^ formed per hour per gram o f p e l l e t .for three
temperatures between 392° and U38°C.
McIntyre also determined that at
UUO0C and w i t h a flow rate ranging from „12 std. cm^ p e r second to 3.8
3
std. cm
per second..external film diffusion was not important for the
reaction using Harshaw p e l l e t s.
It was also shown b y McIntyre that
the rate of reaction of CaS with Og is greater than the rate of reaction
of CaS w i t h HO.
This work is closely related to the w o r k of vSixts and McIntyre„
The general reactions to be studied are:
MeS(s) + Uno(g)
MeS Cs) + 20g Cg)
-
M eSO^Cs) t 2Hg (g) MeSO^(s)
As can be seen for every mole of MegO^ formed the so l i d w i l l increase in
weight
by 64 grams.
Therefore, b y continually weighing the solid, the
rate of reaction w i t h time can be determined.
White has summarized in Table I the enthalpy and free energy of
6
reactions for these various' sulfides- w i t h NO,
McIntyre has. also shorn
that at least one sulfide (.CaS) reacts very- rapidly with oxygen. - Since
-
Og and NO are commonly found together it Is desirable to k n o w the rates
of reactions of these two compounds w i t h the various "sulfides.
This will
help determine if a metal sulfide can h e profitably used to remove NO
from flue and exhaust gases,
•
Table I Free Energy Changes and Heats of Reaction for
the Reduction of NO employing Metal Sulfides (White, 1973)
General Reaction';. MeS + 4N0 --- - MeSO^ +
Free Energy Change (Kcal/mole)
298°K
500 0K
. i*"ZSiel,
"264.6
Metal Sulfide
Calcium sulfide
Cadmium sulfide
^
Cobalt sulfide
Cupric sulfide
IOOO0K
' Heat of Reaction
(Kcal/mcle)
15000K
-
2 1 5 .7
-
1 6 6 .8
2 98°K
-
3 1 3 .5
-
2 7 3 ,3
-
226.3
-
1 7 9 .3
-
1 3 2 .3
«229.4
-
2 0 9 .5
-
1 6 0 .2
-
1 1 0 .9
-
2 5 8 .8
-
2 2 0 .5
-
1 9 9 .6
-147.7
9 5 .8
-
2 5 1 .4
Cuprous sulfide'
-
2 5 4 .7
-
2 3 5 .2
-
Lead sulfide
-
2 6 1 .5
—241.0
Molybdenum sulfide
-
2 0 8 .7
—190o2 . —144.5
Mercuric sulfide ■
-
2 0 9 .9
Silver sulfide
-
220.5
-
2 0 0 .9
-
1 5 2 .5
Tin sulfide
-
2 2 0 .7
-
2 0 1 .0
-
1 5 2 .0
-
—243•8
-
2 2 4 .9
-
1 7 8 .0
-131.1 ■
Zinc sulfide
;
2 4 5 .3
.- 1 9 0 .9
1 8 7 .0
-190,1
-
•
.
—138« S
-
—283*4'
1 3 9 .2
9 8 .8
-
- 291*8
-
2 3 5 .9
-96.9
-
2 3 7 .9
' - 1 0 4 .1
-
249.3
1 0 3 .0
-
2 4 9 .9
-
271,8
-
-143.9
-
OBJECTIVES
The objective of this research is to determine and compare the
reaction rates of the formation of metal sulfates from NO and 0
various metal sulfides.
Vrith
'
APPARATUS
Figure I is a schematic diagram of the apparatus: u s e d to. s t u d y the
reactions. . The'balance .mechanism is a C a h h R-IOQ continuous recording
electrobalanceo
This device i s used to measure t h e w e i g h t of a. sample' -
continuously as it h a n g s .suspended from one arm of, t h e balance in the
reactoro
The R-IO0. electrobalance has- a 100 gram capa c i t y for sample ■- -
■weight and container.
Tare capacity is 100 grams mechanically and 50
'
milligram w i t h the coarse zero. ' The electrobalance h a s three electrical
weight suppression ranges capable of electronically t a r i n g as little as
10 micrograms or as m u c h as 10 grams.
The readability of the electrobalance is ..5- micrograms arid it has
six weight ranges: 10 grams, I gram, 100 milligrams , I milligram and 100
micrograms, full chart scale.
An automatic range expander automatically,
brings the chart pen hack to zero up to ten times.when a weight change
t a k e s .the weight outside of the recorder weight range.
3
the instrument is ± 10
The precision of-
*
io
■
of the meter and recorder- range and ± 10
of
load and the accuracy is ± 5 x 10
weighings.
"
of mass suppression range ■for absolute
The maximum weight change is 10'grams increase or decrease.
The system shown in Figure I
normally operates- w i t h feed gas pass- ■.
ing through a rotometer and entering the bottom of th e reactor.
gases leave just ahoye the reactor and are yented t o hood.,
Exhaust
A helium
line is .run into the glass hell housing the balance m e c h a n i s m to keep
the hell'purged of exhaust gases, • D u r i n g .startup th e v a l v e arrangement
makes it possible to pass helium through the reactor as w e l l as- over
10
Balance Mechanism
Tare P a n
Exhaust
to
Hood
Sampling
Septum
«#— Rotameter
Furnace
Inlet
Reactor
Rotameter
Inlet
Figure I.
Flow diagram of apparatus for
measuring rate of reduction of NO
11
the balance mechanism.
The reactor is enclosed in a Lindberg 5^-331 h i n g e d tube furnace
during normal operation.
It can be removed to allow access to the
reactor tube before and after a run.
It is controlled b y a Teco
•
TC-IOOO proportional temperature controller (not •s h o w n ).
' The reactor cross-section is shown in Figure 2.
The powdered
sulfide rests on a 39mm diameter circular stainless steel pan
pended
b y a .Imm nickel wire from the balance arm.
sus­
Tlie reactor is
a 57mm d i a meter, 840mm long Flothru V y c o r , 8^75/50, tube with a ground
glass joint at the bottom and a ground glass ball joint at the top..
Attached to the bottom joint of the reactor is a glass connector,
and in this connector is mounted a porous glass plate.
Two thermo­
couples are cemented with epoxy into a hole in the side
of
the glass
connector and extend up the Vycor tube to a point just be l o w the support
pan.
One thermocouple wire is attached to. a proportional controller
and the other is attached to a temperature recorder.
w i t h 40 m e s h Ottawa sand to preheat the feed gas.
The tube is filled
The gas is fed into
the bottom of the reactor, passes over the powdered sulfide and is ex­
hausted out the top
12
Suspension
Wire
Exhaust
Outlet
Vycor
Tube
Thermocouples
Support
Pan
5 — Otta w a Sand
Porous Glass
Plate
Feed
Inlet
Figure 2.
Reactor cross-section
PROCEDURE'
For measuring the rate of reaction for a particular sulfide, the
powdered sulfide was placed on the weighing pan of t he Cahn electro­
balance 0
Eight tenths, of a gram, of the powdered sulfide was used for
each r un and the powdered sulfide was evenly distributed over the
weighing pan.
After the weighing pan had been placed i n the reactor,
the reactor was heated to the operating temperature.
A stream of pure
helium was passed through the purge line in the bell and another stream
of pure helium, was passed through the reactor while h e ating the reactor
and until no further weight change was recorded.
Next, in measuring the rate of reaction wi t h NO, a 2„5% NO,
97°5%
He mixture was fed into the reactor at an approximate rate of .2125 std.
car per second.
The upstream pressure was 15 psig.
T he reaction was
then allowed to proceed for at least one hour or u n t i l it was possibleto determine the rate of reaction from the recorded we ight increase.
This procedure was used for runs at.300°, Ij-QO0 and 5 0 0 ° C .
In determining
the rate of reaction of the sulfide with oxygen, a 2.5% O ^ , 97.5% He
3
mixture was fed at a rate of approximately „2125 std. cm
the reactor.
The upstream pressure was 15 psig.
repeated f o r temperatures of 300°, U00° and 500°C„
per second into
This procedure was also
RESULTS AND DISCUSSION
Figure 3 .shows an example of a typical run.
Th e section marked
"a" represents the time during which t h e 'furnace is h e a t e d and pure
He is fed through the reactor.
settled d o ™
'
This was continued u n t i l the equipment-
and a horizontal line was recorded.
mixture was fed through the reactor.
T h e n the reaction
This is section slL 1i in Figure 3.
This weight increase represents the formation of the m e t a l sulfate.
The reaction wo u l d eventually slow and stop.
This is section ”c!l.
The
time for this was different for the'' different sulfides and was not
investigated.
The rate of reaction that was calculated was from section
11L n and was the initial rate of reaction.
This was determined by the
weight change in the first hour of reaction or sooner i f possible.
The sulfides used either came in a powdered form or were powdered
before using.
No attempt was made to make the particle size of the
different sulfides the same.
The primary concern of this investigation
was not to -compare the reaction' rates of the different sulfides, but
to compare the relative rates of each sulfide w i t h n i tric oxide and
Table II is a summary of the reaction rates determined in this
research-.
'In this table the units are mg of metal sulfate formed per
minute per m g of metal sulfide initially on the weig h i n g pan.
The
negative signs denote competing reactions which are -proceeding faster
than the desired oxidation resulting in a loss of w e i g h t instead, of a
gain in w e i g h t .
place.
This indicates that some unknown reaction is taking-
The "d11 represents temperatures- at wh i c h the m e t a l sulfide
.
15
WEIGHT OF SAMPLE (ma)
5Sreaction stops
'weight change
time
Reaction mixture is fed through, reactor
TIME (minutes)
FIGURE 3«
A sample recording from the Cahn R-IOO
continuously recording electrobalance
16
TABLE I I 0
Reaction Rates of RO and 0,
w i t h Various Metal Sulfides
6
grams of MeSOj formed
minuses initial grams 'of MeS'
Rate x 10
4oo°c
300°C
Sulfides
NO
BaS
-
FeS
-
■ o_
’
0-, 00
NO
500°C
0„
NO
o„
24.00
32.60
874.00
3.38
85.00.
207.60
428.00
18.60
ZnS
80
5.90
0.00
lU.oo.
0.00
'25.10
SrS
7.92
7.65
9.81
18.70 -
13.20
27.70
a
a.
InCuS
CdS
d
3.98
d
d
a
15.50
14.7&
22.80
116.00
0.00
37.00
9=18
648.00
50.30
20.50
191.00
-
a
a
a'
+
+
T .
a
■•a
a
a
a
a
a
CaS
O
O
O
Cu2S
9.79.
5.58
MnS
a
a
FbS
14.80
0.00
MoS,
SouJ^icLem
oioo
d
■ i7.4o
a
a
10.90
■a
a
17
d e c omposes, therefore, no reaction rate was determined. The positive
(+) sign in Table II represents the fact that there was a weight gain,'
H o w e v e r , the exact composition of the sulfide was unknown and the rate
of reaction could not be determined.
Table' II shows that the me t a l
sulfides reacted faster w i t h oxygen in every case except one,
At 300°C
strontium sulfide reacted slightly.faster w i t h nitric oxide than with
oxygen.
From the data presented in Table II,it can b e seen that B a S ,
F e S , S r S , G d S , P b S , CaS and sulfurated
oxygen and nitric oxide.
potash all react both with
The SrS rates for oxygen and nitric oxide
are nearly the same for the various temperatures at w h i c h data was
taken.
The rates of reaction listed in Table II also increase w i t h
temperature with few exceptions.
A t 300°C most of the rates are fairly
small, but at 400°C and 500°C they increase significantly.
M n S , thallium s u l f i d e , CuS and MoS^ decomposed at temperatures
such that no rates of reaction could be determined.
The fact that
they were decomposing was indicated b y the electrobalance.
The electro—
balance wo u l d continuously show a loss of weight, w h e n the reactor and
the hell were being purged by a pure stream of helium.
For cuprous sulfide rates of reaction w i t h nitric oxide and oxygen
were not determined at SOO0C and UOO0C due to the. lack of a sufficient
amounts of the chemical,
At JOO0C cuprous sulfide does not react w i t h
HO,- but it does react with oxygen„
Zinc sulfide also does not react
with HO at U-OQ0C and JOO0C , but does react with O^.
18
It should be noted that the rate of reaction to form PbSO^ from
PbS and HO is slower at 500°C than at ^OO0Co
competing side reaction.
This m a y be. caused b y a
A competing side reaction m a y also be the
reason that at SOO0C the reaction.of PeS with 0
an d HO showed a.weight
loss where at temperatures greater than 400°"C -the sulfate producing
reaction seemed to increase to such a point that 'a fairly rapid, weight
gain was recorded,
negative weight gains or no weight gains are seen for several of I
the sulfides at 300°C„
This .seems to be the temperature at which side
reactions dominate or at w h i c h no reaction takes place.
In the case of sulfurated potash the reaction was very fast as
indicated b y a very rapid weight gain.
The exact chemical composition
of the sulfurated p o t a s h Is- .not 'known so it was impossible to calculate
rates of reaction.
Table III gives a summary of t he weight gained or •
lost per unit time for all the sulfides tested.
Table III shows t h a t .
the greatest increase in weight with time-was recorded for the sulfurated
potash and
at 3 0 0 ° C .
The run reacting sulfurated potash with HO at
500°C was not repeated w i t h oxygen because after the run w i t h HO' at
500°C a loss is the weight of the weighing pan of about
.3 grams led to
the conclusion that some type of reaction with the' stainless steel pan
was taking place.
i
The fact that in general metal sulfides react faster with oxygen
than with nitric oxide may be slightly misleading.
Since the objective
19
TABLE X I X 0
A Srmmary- of the Weight Gained or Lost
for the Various SuXfIdes Reacting with WO and 0 o
bate
x ip 3
ar.^ _ ° 4 i|« ij S L ^ a a 6 s a
300°C
Sulfides
SO
.....
BaS
-2.08
FeS
— 2.27
500°C
Uoo0C
0 o ..
id
0
,
-20.80
so
0o
4.09
5 .3
1.172
27.58
so
0,
-gw.
7.22
182.00
70,9
lUU.oo
t.
ZnS
1 .5 0
1..77
SrS
2.22
2.15
CuS
d
CdS
CD
Co
d
3.80
0
2.73
d
3.62
U.53
5.29 d
5.;50
Cu2S
2.53
0
MoSg.
0
SuXfurated
Potash
Thai Ii nm
Sulfide
CaS
MnS
3.68
d
2.U3
0
S'
ro
FbS
0
38.8
d
M
7.81
d ’
27,6
8.07
8.26
3.52
31.70
1.59
108.00
-3.08
-8:33
:'-22.90
- 6 .5U
-25.20
a•
d
185.00
d
•-
.7.97
79.8
88.70
d
0
2.08
d
127.00
d
6.50
3.85
d
32.UO
.
20
of studying these reactions is to remove the NO from the 'atmosphere„
If we look, at the amount
removed we find that
of NO removed compared to the amount of oxygen
for every mole of MeSO^ formed from MeS and NO
four moles of NO are required 0
required.
When O^ reacts only two moles are
This says that even though the oxygen m a y react faster more
nitric oxide maybe consumed inspite of its rate
solid being slower,
■
of reaction with the
' '
From Table IV it can be seen that at SOO0C the'-relative rate of
reaction of NO w i t h a given amount of ZnS or SrS i n a given time is. '
greater than the relative reaction rate of CU.
Also at UOO0C the
relative rate of reaction of NO with a given amount of B a S , SrS or CdS
in a given time is greater than the relative rate of reaction of O^, '
To check the reproducibility of the results recorded in Tables ■
I I 5 I I I 5 and I V 5 three independent determinations of the reaction
rate of CaS w i t h oxygen were made at JOO0C 0 ■ These rates were determined
to be cOOOO .87s ,OOOO 865 5 and „000081 grams CaSO^. formed per minute per .
gram of CaS,
This indicates that the results of this research are
reproducible to within about 6,9%,
21
TABLE I Y
Th.e Relative Rates at which RO and
React with the Various Sulfides
7
RATE x 10
moles of R O reacted
minute i n i t r a T g r a m s of H'eS"
4oo°c
300°C
NO
BaS
—*
FeS
1,19 .
SrS
1,73
CdS
.76
. °2
NO
- ■500°C
«2
CaS
°2
' 0
3.19
2.06
5.59
.74.90
-
.89
11.20
54.80
56.40
.73
0
0
3.11
1.73
2.14
2.04
2.88
1,49
2.82
2.19
1.88
3.02
11.10
3.32
o.
ChigS
PbS
NO
OO
CO
ZnS
moles of O 0 reacted
minute"Ihitxal grams ofTIeS'
1.95
3.32
0
2.70
• 12.60
1.21
1.64
2.56
3.20
42.70
•
5.71
CONCLUSIONS
The reaction
•
MeSU)
+ 202(g ) -
MeS04(s)
proceeds faster than the reaction
MeS(s) +
4- MeSO^^^ + 2 % ^ )
at SOO0C s UOO0C s 500°Co
'
When B a S 3 E e S , S r S , C d S , P b S , CaS and sulfurated potash are
reacted with NO or O^ an increase In weight is observed indicating
that the sulfate is formed.
Strontium sulfide reacts with NO and Og at practically the same
rate at the temperatures 300°C, UOO0C 5 and ^ O O 0C.
W h e n the rate of reaction could be determined it increased with
temperature is almost every case.
N O 5 CdS + NO and PbS + NO.
,The exceptions being ZnS +
This indicates that side reactions
were probably occuring.
Cupric sulfide, thallium sulfide, manganese sulfide and
molybdenum disulfide should not be considered in processes which involve temperatures greater than 300°C because they decompose .
below SOO0C 0 .
Tungsten disulfide should not b e considered f u r t h e r ,
When
tungsten disulfide, is reacted with NO or Og .a weight loss is
recorded.
This indicates that competing side reactions far exceed
the desired reaction
23
7o
Sulfurated potash, reacts the fastest of all the sulfides tested,
w i t h NO and
80
V
.
At BQO0C the relative rate of NO reacting w i t h a given amount of
ZnS or SrS in a given time was greater than the relative reaction
rate of oxygen reacting with the same amount of ZnS or SrS for
the same time,
9o
At IiOO0C the relative rate of NO reacting w i t h a given amount of
BaS a S r S , or CdS in a given time was greater than the relative
reaction rate of oxygen reacting with the same amount of B a S 5 SrS
or CdS for the same time
RECOMtENDATTCWS
Although, there has been quite a bit of xcork done recentlyconcerning the reduction of nitric oxide w i t h metal sulfides: there are
still some areas in w h i c h continued investigation w o u l d be benefical.
Zadick (l97l) showed that CaSO^ could b e regenerated almost
completely- and quite economically to CaS.
The economics and whether or
not B a S O ^ 5 Z n S O ^ 5 S r S O ^ 1and CdSO^ can be regenerated to their respective
sulfides should be studied,
Sulfurated potash reacts very rapidly with W O and 0
studied more.
mined.
and should be
The composition of the sulfur at ed po t a s h should b e deter­
Unless- the composition is known the rate of the reaction cannot
be determined.
A detailed study should be made of the sulfides of potassium and
h o w they react with NO and
.
Xf the regeneration of B a S O ^ 9.S r S O ^ 9 ZnSO^ and CdSO^ is economical9
the use of support materials with these sulfides should be studied.
More w o r k should be done in developing a catalyst w h i c h would
increase the rate of reaction of the metal sulfides w i t h N O 9 and w h i c h
wo u l d reduce the rate of reaction of the metal sulfides w i t h oxygen.
No attempt was made to determine the effect of t he concentrations '
of the NO and O^ in the reaction mixtures on the reaction rate.
This
should be studied to better understand the feasibility of using metal
sulfides to eliminate nitric oxide from the atmosphere.
•
LITERATURE CITED
I=
A u l t , T.Wo and R 0J 0 A y e n , "Catalytic Reduction of Nitric Oxidewith Various Hydrocarbons", AICHE Journal, IT (2), pp, 265271, March, 1971.
2„
B a r t o k 5 W oj A 0R 0 Crawford, and A. S k o p p , "Control of NO
Emmissions from Stationary Sources", Chemical Engineering
P r o g r e s s , 67 (2), pp„ 64-72, Fehruary, 1971°
3°
B a u e r l e , GoL„, Lee L 0 C 0 Sorensen, and K 0 H o h e , "Nitric Oxide
Reduction on Copper Nickel Catalysts", Industrial and
Engineering CHemistry Product Research, and Development, 13
Cl)', p p o 6 1 - 6 4 , March, 1974°.
4o
B a u e r l e , G. L 0 , K 0 N o h e , and E„P„ Koutsoukos , "Nitric Oxide
Reduction with. NH-g on Pt in the Presence of Oxygen” ,
Atmospheric Environment, 8 (l2), p p 0 1331-1333, December, 1974=
5o
D i e t e r e n , H 0M 0L 0 , E 0 H a l l i e , and J oJ 0 M o h r e n , "Catalytic
Reduction of Nitric Oxide", Chemistry and Industry, l8,
p p 0 905-906, September 15, 1973°
6.
Editor, "Science and the Citizen", Scientific American, 228 (4), .
PP- 44,- April, 1973«
7°
Editor, "The Status of the EP A and Nitrogen Oxides", Science N e w s ,
102 (I), p p « 7, July I, 1972»
8»
Erickson, R 0J 0, "Reduction of Nitric Oxide b y Calcium Sulfide
and Nickel S u l f i d e " , Master of Science T h e s i s , Department of
.Chemical Engineering, Montana State University, 1974°
9°
G i d a s p o w , D 0 and M 0 Onischak, "Iron Oxide Sorbents for Regenerative
Sorption of NO
Canadian Journal of Chemical Engineering»
51 (3), p p ° 337-344, June, 1973»
10o
Hooper, J 0R 0 and C 0L 0 Y a w s , "Oxides of Nitrogen N p O , NO, NOp",
Chemical Engin e e r i n g , 8l (17 ), pp« 99-106, August 19, 1974»
Il0
M c I n t y r e , J 0C 0 , "Rate of Reaction of Nitric Oxide b y Calcium •
Sulfide on H i g h Surface Area Supports", Master of Science
Thesis, Department of Chemical Engineering, M o ntana State
University, 1974»
26
12o
Shaw, H „ , "Reduction of Nitrogen Oxide Emissions from a Gas
Turbine Combuster b y Fuel Modifications", Journal of
Engineering P o w e r , 95, pp« 301-308, O c t o b e r , ,1973.
13.
Good, A. and J„ R„ K i t t r e l l , "Dual B ed Catalyst for Simultaneous
Reduction of SOg and NO w i t h C O " , Industrial and Engineering
Chemistry Product Research and D evelopmentT"l3 (3), p p c l80185, September, 197
l4o
White, R o H 0 , "Reduction of Nitric Oxide Using Metal Sulfides",
Master of Science Thesis, Department of Chemical Engineering,
Mont a n a State University, 1973.
15=
Zadick, T 0W 0, "Catalytic Reduction of Calciimi S u l f a t e " , Master
of Science Thesis, Department of Chemical E n g i n e e r i n g ,
Montana State University, 1971.
MONTANA STATE UNIVERSITY LIBRARIES
3 I762 1001 4336 9
II6683
cop .2
Hodgson, Kent M
Rate of reaction of
nitric oxide and oxygen
by metal sulfides
ISSUED TO
DATE
qffWJMtAWT IT
v
------------ 1------- ---------
1*1«!
> .T
J—