Combustionand Flume

73

Composite Propellent Catalysts: Copper Chromate and Chromite

G. S. Pearson*
Mini,;try of Technology. Rocket Prqlpulsion Establishment. Weslcou. Aylesbnry. Bucks. England

Tile thermal decom position of copper ebrom~ue aad chromite has been examined by simnltaneotts thernlogruvimetric analysis and dill'crential thermal analysis. The successive slages in the decomposition of the two compounds have been idenlflied. and it has been sbown thnt both ultimately yield cuprous chromite and cuprous oxide. Alternate exposure of the cat.~lyst ~ll 4(10 420 ',U to fuel (ammonia. isobutene, or ethylene) and oxygen showed that after the initkd reduction zl redox cycle occurred between cupric oxide lind copper. Both st[iges of this redox cycle were ~lssociated with exothermic re~ctions, No rcdox cycle was observed when methane was used as u fuel. When percblorie ~lcid was used as oxidizer, un irreversible weighl change occurred, along with the oxidation p:i~i 'ff the redox cycle. The relev~lce of these results to the mechanism of composite propellent catalysis is discussed.

Introduction Modification of the burning rate of composite propellents based on ammonium perchlorate in a matrix of polymeric fuel is frequently achieved by the ~ddition of small amounts (ca. I per cent by weight) of various substances. The :ldditive used to increase the burning rate is often copper chromate in the United Kingdom and copper chromite in the United States. The mechanism by which such catalysts are effective has long been debated. Attempts have been made to elucidate the detailed ml,'chanism by experiments with model mixtures [I,2]. However, the results of such experiments are often ambiguous, and the role of the coppercl:romium compound is subject to ctmjecture. Recent experiments have shown that such catalysts ~.re effective in promoting the ignition
* Presenl address: British Embassy, Washington, D.C.

of (a) gaseous fuel-perchloric acid vapor [3], (b) solid fuel-perchloric acid vapor [4], and (e) solid fuel-oxygen [5] systems. These experiments [4] showed also that perchloric acid decomposes exothermically on copper chromite catalyst, which simultaneously undergoes partial chemical changes [6]. The ability of transition metal oxides to catalyze the decomposition of perchloric ~cid has bccn confirmed by Solymosi 1"7]. In the present work. these copper-chromium compounds have been characterized, and their behavior when heated or exposed to various reac!ive gases is studied. Evidence is presented that it redox cycle is involved in the action of such catalysts in gaseous fuel-oxygen and gaseous fuel-perchloric acid vapor mixt~u'e~."[hc results are discussed in relation to the role of such catalysts in the combustion of ammonium perchlorate propellents.
('r,nl,u,~li,m& Flare,'. la. 73 ~41It~7nl IJrilixhCrown. t'opyrigh]. 1970 Puhllshed hy Anlcricilfl lilsevict Publishing (.'tlrilplin)',Inc.

74 Experimental and Techniques

G, S. Pearson introduced through another heated glass tube of smaller diameter, Temperatures were measured by means of Chromel-Alumel thermocouples with a cold junction in ice.

Materials
Two samples of both copper chromate and copper chromite were used. Supplementary experiments were made with cupric oxide. The catalysts and their sources are given in Table I. Table I. Sourct~: and N~minal I:,;'mulas of tbe Cataly~'t Sumple.. Catalyst Copper cbromatc Source Nominal Formula Propellentgrade CuCrOV2Cu(OH)= Cu as CaO 62: Cr as CrO~28 Copper chromate Copper chromite I-Iopkim~ Williams,Ltd. & CuCrO.t2Cu(OH)= Harsbaw ChemicalCo. Harshaw Catalyst Cu e202p Co as CuO 82; Cr as Cr~O:r17 I~.esearch Inorg.Chem. Co. 99/t, Hopkins & Williams.Ltd, CuO '%nalar'" i~rade

Analytical Methods
The composition of the catalyst samples as supplied was determined as follows: I. Copper was determined iodometrically after the reduction of chromium with sulfur dioxide. 2. Chromium was determined by the titration of chromate with ferrous ion. It was necessary to fuse the copper chromite samples with fusion mixture in order to dissolve them. Copper was determined in the RIC copper chromite by atomic absorption spectrophotometry after solution in perchloric acid.

Copper chromite Cupric nxide

The gaseous reactants were perchloric acid vapor (obtained by vaporization [8] of 72 per cent by weight perehloric acid (Hopkins and Williams, Ltd.), ammonia (IC1, Ltd.), isobutene (Air Products Corp.). ethylene, oxygen, and nitrogen (British Oxygen Co.) and methane (West Middlesex County Council Main Drainage Dept.)

Restdts Before the chemical mechanism by which these catalysts are effective could be elucidated, it was necessary to characterize the particular samples of "copper chromate" and "copper chromite," which were not pure cupric chromate, CuCrO4, or cupric chromite, CuCr20 ,, respectively. This characterization was made by chemical analysis of the copper and the chromium content of each sample and by simultaneous DTA and TGA.

Techniques Simultaneous differential thermal analysis (DTA) and thermogravimetric analysis (TGA) experiments were made on the catalyst samples, using a nitrogen atmosphere in a Stanton thermobalance, type TR-02. The apparatus previously used to study the ignition o f composite propellent fuels [8] was used to study the effect on the catalyst samples of reactive gases. The catalyst was contained in the heated glass tube, and the gases were

Cat,llyst Comt~ositions
The chemical analyses ior each sample are shown in Table 2. It was then possible to calculate empirical formulas by assuming that the copper chromate samples were C u C r O , . x C u (O1"[)2 and that the copper chromite samples were CuCrzO4"yCuO. This was conveniently done graphically by plotting the theoretical percentage compositions of copper and chromium for 'values of.x in CuCrO,~. a'Cu(OH)2 from 0.5

Copper Chromate and Chromite Catalysts

Table 2. Cop~r and Clirnmium Contents and Empirical FQrmnlas of the Catalyst Samples

Catalyst Propellent chromate H & W chromate Harshawchromite RIC chromite Cupric oxide

Copper as Cu 48.4
46.0

Chromium as Cr
13.7

Empirical Formula CuCrO~.1.75 Cu(OH): CuCrO~,1.4 Cu(OH)~ ,CuCrOd7.0 CuO Ct~Cr~O~.0.9 CuO ~)8,2o/, CuO

14.4 12.1 29,9 ...

63,5 33.6 78.5

to 3.1) and o f y in CuCr.~O,~.yCuO from 0.5 to 10.0. The empirical tbrmulas corre~,ponding to the analytical results are also given in Table 2.

The stages are thus tentatively assigned to the following processes: (i) Ambient--425'>C: (it) 425-525~C: (iii) (iv) loss orwater, cupric c h r o m a t e ~ cupric chromite. 700-850C: cupric chromite--> cuprous chromite. 875-1000C: cupric oxide---> cuprous oxide.

DTA and TGA Experiments Samples o f the calalysts were heated at 5C m i n - t from ambient temperature to 1lfi0C in a flow of nitrogen (white spot; 0.2 liter min- t). Percentage weight losses calculated from the T G A results are plotted against temperature in Fig. I. Several weight loss stages can be identified from the known components of the samples,

The simultar~eous DTA results show a series of endotherms and occasionally an exotherm. The positions of the peaks of these heat changes are indicated in Table 3.

i
'; i

CUPRIC CHROMATE(PRQPELLENT)~ " I ~,

, --

2o ~soI

i I 1

I , I

.
~ ~--

.
f

. ,.

. ~ .

. , : I ~

~Eup~acC ~ M A T E (H ~W~

iI _
. . .

/
~7

/
LL.J-;--'='=~ "

t
'

, fcu,>p~.~,~'
I !

y . ........ 7' ,

Figure I. Thermogravimetrieresults ['tom heatingthe catalyst samplesaI 5"C mln- = in nitrogen.

76 Table3. Position of Endothenns and Exotherms in the Differential Thermal Analysis of the Catalyst Samples Sample Propellent chromate H & W chromate Harshawchromite RIC chromite Cupric o~idc Endotherms 356, 4"/9,780, 956 357, 471. 769. 935 742, 984 ~ 120,"346~798 971 E~otherms 497 398 ...

G. S. Pearson rises. Repeated exposure to a single gas resulted in no marked heat release or visible glows on the second and subsequent exposures. 4. Exposure of the sample to ambient air between the introduction of fuel and of oxygen caused the temperature rise on introduction of the oxygen to be reduced. 5. The temperature rise was greatest on the first introduction of fuel vapor to a fresh sample of catalyst. 6. Subsequent exposure: to fuel and oxygen was ineffective at 260-270C unless the catalyst was reactivated by heating to about 500C, for example, by simultaneous introduction of fuel and oxygen. 7. At higher temperatures, c a . 400C, the temperature rise was greater for isobutene than for ammonia, which resulted in a greater rise than ethylene. On the other hand. the temperature rise on the subsequent introduction of oxygen was similar for each of these fuel gases. Harshaw copper chromite behaved in a similar way, but it was observed that the order of the tt:mperature rises at 400-420C differed 'with ammonia, isobutylene, and ethylene: the rise was greatest for ethylene and least for ammonia.

" These endotherms were small although definite

The endotherms at 470C, 740-800C, and 950~C can be assigned to processes (ii), (iii). and (iv), respectively, found in the T G A results.

Reaction with Gaseous Fuels and Oxidants The effect of passing either a gaseous fuel or an oxidant over a heated saraple of the catalyst was examined by recording the temperature change o f the sample when the gas was introduced and by gravimetric measurements.

Temperature Changes The temperature of the scruple was monitored by placing a ChromeI-Alumel thermocouple so that the junction was covered by a layer of catalyst. It is apparent that the temperature changes recorded by this method will depend on the precise position of the thermocouPle, and the results are therefore qualitative. The experimental observations for a sample of copper chromate may be summarized as follows: 1. Methane was not reactive either alone or with oxygen. 2. Simultaneous introduction of a fuel (ammonia or isobutene) and oxygen caused the catalyst to glow red hot, and a rapid temperature rise was recorded. The glow was sustained as 10ug as the gas mixture was introduced. 3. Alternate introduction of fuel and oxygen gases produced transient glows and temperature

Gravimetric Results The temperature changes showed that copper chromate and chromite reacted rapidl?, abo~e 350C with gaseous fuels (ammonia, ~sobutene and ethylene) and with oxygen when these were introduced alternately. Several modes of reaction are possible: (i) Activation of fuel molecules by absorption on the surface of the catalys~ (no inte';action with the oxygen ia the catalyst), followed by subs,:quent reaction with oxygen gas, (ii) Reztctinn of fuel molecules with catalyst oxygen in the surface layer, followed by replacement of the catalyst oxygen from the gaseous oxygen.

Copper Chromate and ChromiteCatalysts

pERC~mAC~ INmALS~,UPLS O~ V~OGHT !~'~r-~ C",. 0~. C~,

77

,,

kV

vr

v ......'H'~'

CHROMATE tHL~

Figure 2. Percentageweightchanges observed on alternate exposure of catalyst samples to thcl (ammonia. isobutenc. ethylene, or methane)and to oxygen. (iii~ Reaction of fuel molecules wtth catalyst oxygen throughout the bulk of the catalyst. The chan~o,es in weight of samples of catalyst were determined to differentiate between processes (0 and (ii) or (iii). Different sample weights were used to differentiate between surl'ace (ii) and hulk (iii) reactions. All of the gravimetric experiments were made at 400-420C and followed a standard pattern. Samples of the catalyst, as supplied, we:-e first heated to determine the water content. They were then exposed to fuel and oxidant gases alternately in that order. The percentage weight losses and gains obserwd for copper chromate and copper chromite in ammonia, isobutene, ethylene, methane, or oxygen are shown in Fig. 2. Methane was found to be unreactive, thus confirming the observation that no temperature change occurred on exposure to this gas. Although sample weights of chromate and chromite of 0.16-0.46 and 0.37-0.97 g, respectively, were used, the reproducibility was good, and the percentage weight changes were independent of sample weight, thus showing that a hulk reaction was involved. Furthermore, the percentage weight losses were independent o f the nature of the fuel (ammonia, isohutene, or ethylene), showing that reduction

CHrOMiC .RO~ITe Cu~nl OXJ~C

x.~ / /

\l/
%/

C.~Ma~

\ X, /

t~O~ELLE~T

Figure 3. Percentageweightchanges observedon alternate exposure of catalyst samples to isobutefle and oxidizer (perchloric acid or oxygen).

78 of the catalyst was involved, rather than an adsorption process in which adsorbed fuel molecules were subsequently oxidized. A more extensive series of experiments in which duplicate samples of each catalyst were used and which included perchloric acid as an oxidizer is shown in Fig. 3. The reproducibility was observed to deteriorate after the samples had been exposed to perchloric acid vapor, suggesting that surface reaction as opposed to bulk reaction had occurred. This was confirmed by visual observation and by the dependence of the percentage weight change on sample weight, shown in Fig. 4. The surface layer of the chromate and chromite samples changed color from black to a mottled reddish brown on exposure to p,,~rchlorie acid, and an alkalin ~ extract of the modified sample was yellow, ihdicat;og chro-

G. S, Pearson

mate ion. No such chromate formation was observed on exposure to oxygen. The cupric oxide samples were effectively deactivated by exposure to perehloric acid vapor, since the percentage weight changes after acid attack were reduced by a factor of 5 or 6 from those previously recoJrded. The cupric oxide samples were also observed to reduce the amount ofperchloric acid fumes, presumably by catalyzing the decomposition of the acid vapor. No parallel effect was observed with the chromate'-chromite samples, suggesting that cupric oxide is a more effective catalyst for the decomposition of perchloric acid vapor.

PERCENTAGE WEIGHT CHANGE 20

CePPER CH~OMA're (~Iw)

te

14

'~

'~N,

.c,o 4

\
X X

12 I0

Discussions Thermal Decomposition of Copper Chromate and Chromite In addition to serving as a ballistic modifier lbr solid propellents, copper chromite has had wide commercial application as a catalyst--for example, in car exhausts [9"1, where it is used to achieve complete oxidation of carbon monoxide and of hydrocarbons to carbon dioxide. Consequently, the nature of the chemical compounds in copper chromite catalyst and the effect thereon of heat have been studied. An early study [10] showed that cupric chromite-cupric oxide mixtures were converted to cuprous chromite on heating at about 900"C: CuCrzO 4'+ CuO-> CuzCr20,~ + -~Oz (I) The cuprous chromite produced was stated to be stable at room temperature, but reaction I could be completely reversed by heating in oxygen for several hours at 600-700C. Pure cupric chromite decomposed at temperatures above 900C to cuprous chromite and chromic oxide: 2CuCrzO4-->Cu2Cr204 + CrzO3 + 02 (2) Reactants and products in both these reactions were identified by X-ray diffraction. More recently the CuO-CrzO3-copper chromite system has been examined by Chareosset

a~ 02 x...15o-$ O--nNn 3

I I
C-I 0"2 0'3 0'4 O'5 0'6 SAMPLE WEIGHT,GRAM

Figure 4, Percentage weight changes with various reactive gases for different weights of copper chromate (H & W).

Copper Chromate and Chroraite Catalysts et al. [11, 12"1, using DTA and analytical methods. The D T A or" a cupric oxide-~,'hromic oxide mixture showed an exotherm at 375C, followed by an endotherm at about 500C. Analysis at 425C showed that chromate was present, and it was eotleluded that the exotherm corresponded to crystallization of the chromic oxide (initially in the ~morphous state)and formation of copper ehronaate. The endotherm was assigned to the decomposition of the chromate to chromite, and this was confirmed by DTA of a sample of copper chromate. X-ray analysis of the CuO-Cr20 s ~ample above 800C showed that the cupric chromite bad been converted to cuprous chromite, confirming the earlier observations [10]. A kinetic study showed that cupric chromite was not obtained from a CuO-CrzO 3 mixture below 550C even on heating for 24 hr. However, the addition of 10 per cent cupric chromite reduced the limiting temperature to below 410C. Thermogravimetric studies of copper chromate at constant temperature showed that decomposition occurred at measurable rates at 400-450C. The reaction is initially fast but then slows down before finally accelerating again. showing that autocatalysis is involved after the initial reaction. Chemical and X-ray analysis indicated that the initial reaction was I~ocupric and chromic oxides: 2CuCrO a ~ 2CuO + CrzO 3 + ~O 2 (3)

7q tures [13] have recently been interpreted to con. firm the various reactions 1-5. The decomposition of basic cupric chromate has also been studied [14, 15] by TGA. The decomposition was found to occur in stages--the first at 450-510C, corresponding to reaction 5; the second at above 850"C. corresponding to reaction I ; and finally at about 1100C the conversion of cupric oxide to cuprous oxide: 2CuO -.->CuzO + ~O z (6)

which was then followed by Ibrmation of cupric chromite: CuO + CrzO 3 ~ C u C r z O ~ (4)

The present work has confirmed that the processes involved in the thermal decomposition of chromate and chromite samples are those of reactions 5, I, and 6, which correspond to stages OiL (iii), and (iv) observed in the T G A experiments described in a previous sectiota. It is further evident from the relative weight losses at the various stages in the T G A that the propellent-grade and the H & W copper chromates are similar and that both contain cupric chromate and cupric hydroxide. On the other hand. the two samples of copper chromite show appreciably different behavior in the TGA experiments. Harshaw chromite clearly has a large content of cupric oxide and relatively little cupric chromite, whereas RIC chromite appears to have no cupric oxide after the cupric chromite to cuprous chromite transition (reaction 1) has occurred, The RIC chromite thus contains I mole or less cupric oxide per mole of cupric chromite. These conclusions are in good agreement with the analytical results for copper and chromium conteut.

Addition of 40 per cent cupric chromite accelerated the decomposition of cupric chromate, and autocatalysis was not observed. The chromite was considered to promote the direct formation of chromite from chromate without the intermediate production of chromic oxide: 2 C u C r O ~ CuCrzO~. CuO + ~O, (5)

X-ray and ESR analyses of CuO-Cr,,O,~ mix-

Reaetioa with Gaseous Fuels and Oxidizers The use of metal oxides as catalysts lbr the oxidation of hydrocarbons has long been known [16.-19]. However, the precise mechanism is still not established since catalyst studies are fi'equently more concerned with the ~'elutivc efficiencies of different catalysts than with the mechanism by which a catalyst is effective [18]. The mechat~ism is usually considered to involve

80 adsorption of reactants 6n the surface of the catalyst,, followed by reaction and desorption from the surface. Support for this view comes from the similarity in the patterns of catalytic activity for a range of metal oxides, both for hydrocarbon oxidation and for oxygen atom recombination There is some agreement, however, that in metal oxide catalyzed oxidations the metal ions in the lattice andergo redox cycles in which the hydrocarbon reacts with the metal ions in their high-valence state ant] the ions are then reoxidized by molecular oxygen The effect of gaseous fuels and oxidizers on copper chromate or chromite has been infrequently reported. An early study [10] of a cupric oxide-cupric chromite catalyst stated that after use in a liquid-phase hydrogenation below 300"C. the cupric chromite was largely reduced to cuprous chromite by reaction I, which competed with the reduction of cupric oxide to cuprous oxide (reaction 6) and to metallic copper The presence of metallic copper in the reduced cutulys( was contirmed by X-ray analysis. It was found that the copper was rapidly reoxidized to cupric oxide but that the cuprous chromite required a more extended exposure (4-6 hr) to oxygen at 650+C to reoxidize it to cupric chromite. A recent paper [20] ascribes the reactivity of such cupric oxide-cupric chromite catalysts to the reduction of cupric oxide to cuprous oxide and copper. The effect of the chromite was not elaborated.
TaMe 4. Experimental and Calculated W e i g h t C h a n g e s

G . S. Pearson

The present work shows that alternate exposure of catalyst to fuel and oxygen results in reproducible weight losses and weight gains, respectively. The weight losses are independent of whetL, er the fuel is ammonia, isobutene, or ethylene. This suggests that the fuels react with the oxygen contained in the catalyst, which is simultaneously reduced. The fact that the percentagc weight changes are independent of the weight of the sample of catalyst suggests that this reaction involves the oxygen contained in the bulk of the catalyst and not just that present in the surface layer. Gas chromatographic analysis of the product gases when isobutenc was passed over copper chromate showed thut carbon dioxide was a product. The heat release in the reuction of fuel with oxygen in the catalyst will depend on the nature of the fuel. On the other hand. the effect of oxygen will be to reoxidize the catalyst from its reduced state. The heat release for this reaction should be independent of the nature of tac fuel previously introduced. The weight changes observed experimentully at 400-410"C may be compared with the calculated weight losses corresponding to the Ibrmation of cuprous chromite and copper us the components of the reduced catalyst. Both experimental and calculated percentage weight losses are given in Table 4. It is evident that the total weight loss on exposure to a gaseous fuel does indeed correspond to reduction of me

for Copper Chromate and C h r o m i t e Samples

Calculalcd Purccrlt~l~ Wighl ('hangcs I~lllptrical l+ornltlhl Lo~h ol + 1120 9+0 8.0
......
. . . . . .

Expurimcnlsd I'rcemag Wighl ('ll~lng~h (';llldyM

I l+1t

I IcaiI .it~o i:uel 27,3 25+0

14.0 10.0

SLlbSeqUcnt xYgUll/IdtlUl

f.'~l<_'r O~ ~ (.'IICr:O+ 6.R

7.5

('uCr2() ~ I CuzCrzO+ 2,3

2.5

('tlO ~ Cu20 4,(J

3.5

(.'La:0 ~ Cu 4.0
3.5

'l'olal

Propclk:rlt chroma[ tl & W chromate

I-larshaw chromit~ RI(' chromite

i I.[) 10,0
0.5 3,7

8.(J 6.8
12.0 2.0

CuCrO+, 1.75 Cu(OH) 2 CuCrO~ +1.4 ('u<Oltl;

CuCr20+.7,0 CuO
CtlCr20~'o.gCuO

26,1
25,(J

2.4)
5,1

6.1
0

6,1
0

142
5.1

2.6" Calcul i cd assur ing CuCr+O+ -~ ~ ?u.Cr+ )4 + tCr~Os + IO

2.6

2.6

7.8."

Copper Chromate and Chromite Catalysts catalyst to cuprous chromite and cop!0er. The weight loss on heating is somewhat larger than that expeci, ed for loss of water from the cupric hydroxide. Two possible explanations are that (a) the presence of I or 2 per cent of water adsorbed on the catslyst sample increases the weight loss of the first stage, with only a small effect on the other stages, and (b) the flattening of the T G A curve for the chromate between the loss of water and the onset of the chromate decomposition is not extensive, and it is likely that some decompositioll may already have occurred. The possibility of adsorbed water could well explain the apparent discrepancy in the figures Ibr the samples of RIC chromite. If the decomposition of the cupric chromite was intermediate between the two cases computed--with weight changes for the third, fourth, and filth stages o1"4, I, and I per cent--and about 4 per cent of water was present, then the percentagc weight losses would be in good ~tgreement with the experimental observations. It is further evident that in tile subsequent exposure to oxygen and fuel the redox cycle involved is that of copper-cupric oxide. Samples of Analar cupric oxide were subjected to a similar cycle of events. The total weight loss on laeating + fuel was 20.5, in good agreement to the calculated loss of 20.1 for reduction to copper. Subsequent weight changes on exposure to fuel or oxygen were about 7- I 0 per cenl, but the reproducibility was worse than Ibr the copper chroma~.e or chromite samples. Since the cupric oxide sample was observed to have conglomerated after exposure to fuel and oxygen alternately, it is possible that this is the reason Ibr the decreased weight changes. It also suggests that the role of tile chromate/chromite may in part be related to preventing such conglomeration. The effect of perchloric acid vapor is to give an increased weight change over that from oxygen, but the interpretation of the experimental weight changes is complicated by the acid affecting only the surface layer. Visual obselration of the chrorea:e/chromite samples aft.:r

81 exposure showed that the color changed from black to a brownish red. This was particularly evident with the RIC chromite, which contains the least cupric oxide. Exposure to fuel vapor alter the perchlorie acid resulted in a reduced overall weight loss, as compared to the weight loss observed on the exposure of fresh samples 9 f catalyst to fuel vapor. This showed that the perchloric acid had reacted with the catalyst samples irreversibly, as well as by forming the chromate. After this irreversible weight increase from first exposure to perchloric acid vapor no subsequent irreversible weight gains were observed. The most pronounced effect was found with cupric oxide, with a weight increase of about21 percentand th, -eduction oftheweight changes on exposure to fuel and oxidant to about 1 per cent. The weight increases for the othel" catalysts were less pronounced but were largest with the Harshaw chromite. These observations suggest that the perchloric acid reacts with the cupric oxide, forming an unreactive species such as a chloride or oxychloride. The tbrmation of oxychloride and chloride by reaction of cupric oxide with perchloric acid was confirmed by testing aqueous and acid extracts with silver nitrate solution. Oxychloride was present ill apnreciably greater quantity than chloride. These observations are in general accord with those nmde by Wise et al. [23] that copper chromite (Harshuw) changes color on decomposing ammonium perchlorate, showing chromate Ibrmation, and that this color change is not observed ill the presence of :.tmmonia,

Composite Propellent Catalysis The role of chromate or chromite catalysts in the combustion of composite propellents may now be considered. The present work has shown that these catalysts arc effective, at temperatures close to those prevailing on the propellent surface [21 ], in promoting the reaction of ammonia or olefins with perchloric acid or oxygen. Under

82 the same experimental conditions, a paraffin hydrocarbon, methane, W-~tsunaffected. It has also been shown that pronounced heat release is observed both when these catalysts are exposed to fuel vapor and when they are subsequently exposed to perchloric acid or oxygen. Furthermore, under the heterogeneous conditions prevailing on the surface of composite propellents, it is possible that the catalyst partides are participating in a redox cycle similar to that shown in these studies. It is therefore envisaged that the heat re,ease on the catalyst particles on the surface of the propellent will a priorienhance the burning rate as a consequence of the increased heat transfer into the solid propellent. Such surface reaction, which results in the reaction of a part of the fuel and the oxidizer, may also enlaance the burning rate through a second effect in which the gas-phase burning veloe~ty is increased [22]. The catalyst may also be effective in other ways, such as by promoting the reaction o f perehlorie acid vapor with solid fuel [4] or by modifying the pyrolysis mechanism of the solid fuel. Copper chromate (propellent grade) and copper chromite (Harshaw) may be compared on the basis of the amount of oxygen available for initial reaction with the fuel. This is greater 07.0 per cent) for chromate than for chromite (1~4.2 per cent), but the difference is offset to some extent by the water present in the chromate (8.9 per cent). On the other hand, the amount of oxygen involved in the subsequent redox cycle (CuO-CuzO-Cu) is greater with chromite (12.2 per cent) than with chromate (8.0 per cent). Their relative merits as burning rate or ignition catalysts will depend on which factor is the more significant for the practical conditions concerned. It should be emphasized that after the initial reduction the chemical species present from both copper chromate and copper chromite are the same, and the samples then differ only in the relative amounts of copper oxide and chromite.

G.

S. Pearson

Conclusions The thermal decompositions of copper chromate and chromite have been shown by TGAD T A studies to proceed by these successive stages: cupric chromate --, cupric chromite --, cuprous chromite, followed by chpric oxide cuprous oxide as the temperature'is increased from ambient to about 1100C. Alternate exposure o f the catalyst at a temperature close to that on the surface of a burning propellent to fuel (ammonia, isohutene, or ethylene) and to oxygen has shown that, after initial reduction to cuprous chromite and copper, a redox cycle (cupric oxide-cuprous oxidecopper) occurs that is associated with exothermic reactions in both the reduction and the oxidation stages of the cycle. In these experiments methane was ineffective and did not participate in a redox cycle. When perchloric acid was used as oxidizer, an irreversible weight change occurred, along with the oxidation part of the redox cycle. It is concluded that the enhanced heat release from such reactions on catalyst particles on the surface of the propellen(, constitutes one route by which the burning rate i~ enhanced. Thanks are given to Dr. D. Sutton for many helpful discussions, and to Dr. J. C. Wright and Mr. E. J. Gallacher o f E.R.D.E. for the copper and chromium analyses. British Crown Copyright. Reproduced b): permission of the Controller, Her Brittanie Majesty's StationeIT Office. References 1. NADAUD, Cotiibu.l'tion d~Flame, 12. 177 (1968). L.. 2. ROSSER.W. A,. FISI~MAN,N., and WISE,H., AIAA J., 4. 1615(1966), 3. PI!A ':AN.G. S., and SI3'l"roN.D., AL4A J., S. 344 (1967). R. ) 4, I)I!ARSON, S.. and SUT1ON,D,. AI/leI J., ~, 21(11 G, (1967). 5, PI!AgSON. S. and SUTTON, A/A/IJ., 7, 77(I( 1969), G. D.. 6. PEARS(IN. S,, ~tndSUTTON.D,, CtH/Ibttsliotl & F'Jame. G,
13. 33(I (1969). 7. SOLY~IOShF,, BOResrI. S.. and LAZ~.rr,.E,. Combu,~tlon d. Flame, 12. 398 (1968). 8. FE ~RSON. G. S., and SuT'IX~N. D.. A I A A J.. 4, 954 (1966).

Copper Chromate and Chromite Catalysts

9. LEAK. R. J.. BRANDENBURG.J. T., and BEHRI~NS. D.. M, Environ. Sci. TechnoL. 2. 790 0968). IO. STROUI'E.J. D.. J. Am. Chem. Soc.. 71. 569 (1949). ] l, CHARCOSSI!T. H.. TURLIER. P., and TRAMBOUZE.Y., Compt. Rend.. 254, 2990 (1962). 12. CHARCOSSLn',H., TURLIER,p,, .~.rtdTRAMBOUZE, J. Y.. Chim. Phys., 6L 1249 (1964). 13. VON SCHULZ, |., EBERT,[., and SCHEVE,J. Z. Anorg. AIIgem. Chem. 346; 66 (1966). ]4. CI.IARSLEY.E. [... and REDI:ERN.J. P.. Firm [llt{.rntl. tional Thermal A nalysl~"Cmlference. p. 124. M acmillian : London (1965). 15. CHARSLI~.E. L.. P'~'ivatecommunication (1969). 16. MARGOLIS.L. YA.. Advm;. Catalysis. 14. 429 (1963). 17 VIJ(.;~. H. H.. and ADAMS.C. R*. Adl'all. ('~ltal)'si.~. 17. 151 (1967),

83
18. SAMPSON.R. J.. and SIICX)TI~R.D,. Oxidathm & C'omhu.~t. Rev.. I. 223 (1965). 19. CULLIS.C. F.. Ind. Eng. Chem.. 59. 18 (1967). 20. OHrSUKA. H., AOMURA. K*, TOMITA. N., HASHIMOTO. K.. and TAKADA.t'}.. II,kkaidu Daigaku Ko.gak~tbl~Iho. No. 41. 199 ~1o~,o): Chem. Abstr.. 69. 702172 (1968). 21. POWLING. J.. Elevt.nth Sympo~ffum (Internath.:lal) on Combu.vtion. p. 447. The Combustion Institute: Pittsburgh (1967). 22. HALl.. A. R.. and P~ARSON.G. S.. Combu.~tiotl ~ Fl(mle (in press). 23. ROSSER.W. A.. INAMI.S. H.. and WISE. H,, Combustion & Flame. 12. 427 (1968).

( R e c e i v e d M a y 1969; revised J u l y 1969)