8/12/2019 bone char 2

1/13

he rmo ch imzca Acta, 56 1982) 345-357EIsevier Scientific Publishing Company, Amsterdam-Printed m The Netherlands 345

KINETICS QF THE lKERMAL DECOMPOSITION OF ANIMAL BONECHARWAFA I. ABDEL-FATTAH and M.M. SELIMNatronal Research Centre, Dokkl, Carro Egypt)(Received 26 January 1982)

ABSTRACTThe effec: of process vat-tables, i e. tnrttal surface area and rate of heatmg. on the

producuon of carbon-free bone ash, as well as INS echmque, were investigated. Thermogravl-metric analysis up to 1000C indicated that a uniform rate of heatmg of 10C mm- wasmore favourable than 5OC min- _ Also, grindmg to achieve 180 and 98 m2 g- surface areafor the bone char and bone char waste accelerated the combustion process. Dynamic heattreatment resulted m a deflection in the activation energy where behxv 400C it was 4.75 kcalmole- and only 2.07 kcal mole- was obtained between 400 and 700C. Isothermal heattreatment yelded an actlvatlon energy of 4.07 kcal mole- for both samples.

INTRODUCTIONThe kinetic study of the thermal decomposition of solids depends mainlyon the following factors: (i) initial and final temperature of decomposition;(ii) the rate of temperature increase; (iii) the atmosphere surroundmg thesolid sample (inert, oxidizing or reducing); and (iv) the nature or state of thesolid sample itself (size of particles). These factors, in general, greatly affectthe texture, ,;tructure and composition of the final product.In dealing with kinetic data, it should be taken into consideration that theoverall rate of a complex process involving a series of consecutive steps isfixed by the rate of the slowest single step. The one with the lowest energybarrier will be the most rapid, major contributor to the overall process [I].The kinetics of the thermal decomposition of solids are usually studied bymeasuring the fraction ar of material reacted at constant temperature T andpressure P [2]. Another method of analysis is where the residual weightfraction is related to the temperature for a sample heated at a fixed rate [3].In a previous publication [4], the thermal decomposition of animal bonecarbon and bone carbon waste was investigated in air and in vacuum. It wasfound that thermal treatment yielded a direct change in the surface area

which was accompanied by changes in the amount of chemisorbed water.The present work is devoted to a study of the kinetics of the thermal0040-6031/82/oooO-oooO/ O2.75 0 1982 Elsevler Scentfic Publishing Company

8/12/2019 bone char 2

2/13

46

treatment of bone char and bone char waste. A knowledge of the kinetics ofsuch changes accompanying the decomposition is essential in determining itspractical feasibility.

EXPERIMENTAL

M u terra I..Samples of animal bone char(Ad) and its waste(Adw) after the refiningprocess of sugar were supplied by the Sugar Refining Co., Hawarndya,

Egypt, for the present study.Grtndrng and surface area

Both samples were ground in an automatic agate mortar for 40 min. Thespecific surface areas of the two samples, as provided and after grinding,were measured using the BET method and N, as adsorbate.Thern7og ravtlttetrtc anaiysis

The samples were heated contmuously at a constant rate while the loss inweight was recorded. The rate of heating was varied between 5 and 10Cmin- to determine the correct rate. An automatic thermobalance, providedby Gebriider Netzsch, W. Germany, was used.

ynamic heat treatmentKnown weights of the specimens were heated in porcelain crucibles in anelectric upright muffle from room temperature to the required temperatureat a rate of 10C min-I. The specimens were taken out of the kiln at

temperatures in the range 200-700C after various soaking times of 4, 7, 10,60, 120, 180 and 240 mm. Experiments were carried out in duplicate.Isothermal heat treatment

A known weight of each sample was inserted in the muffle at temperaturesin the range 450-650C with 50C intervals. The specimens were withdrawnfrom the muffle after a given time. Prior to the isothermal experiments allthe specimens were fired in a platinum dish at a rate of 10C min- up to4OOOC for 10 min and immediately allowed to cool in a desiccator.

8/12/2019 bone char 2

3/13

347RESULTS AND DISCUSSIONSaiffliediei

The BET surface areas of the two samples as provided by the manufac-turer (Ad and Adw) and after grinding (Adg and Adwg) are:

Ad Adg Adw AdwgS S A {mg-1 112 180 66 98The differences in the final values of S.S.A. of both samples, althoughsimilar weights of the samples were ground with the same speed for a fixedtime, are attributed primarily to the differences in the starting surface areasand the nature of each sample.ff ect of gri ndi ng on the burnr ng process

The DTG curves of both samples heated continuously and uniformly at arate of 10C min- are illustrated in Fig. 1 (A and B) for the Ad and Adw

006-

006

4dw4dwg

Temperature YZ)Fig. 1. Diffcrentlal weight losses for both samples before and after grmding. Heating10C min-. rate.

8/12/2019 bone char 2

4/13

34ssamples. respectively, before and after grinding. The ground samples havecomparatively higher weight loss values below 340C. These correspond tothe dehydration, dehydroxylation and decarbonization. The recorded loss inboth samples is unexpectedly lower upon grinding. This means that part ofthe carbon has

8/12/2019 bone char 2

5/13

349

the grinding process affected much more weight loss, especially for the bonechar sarr& < &j.it is clear that the grinding process promoted the reactions through thenewly formed active surfaces. This activation is related to the stored energyin the ground material as well as the smaller particle size. Upon grinding, theexisting bonds in the material are broken, a process that leads to an Increasein its internal energy. The increase in free energy per unit area of the newlyfformed surface is known as .&c s.ur_f_acen~~,vJ 1). C~nse~uenxly thz pssempromoted reactions are attributed to the driving force provided by the excessffree energdvof ttre newly increased surface. In this connection the ,oartt&size is equally important as the rate of many reactions is mversely propor-tional to the particle size.E f f ect o f r a t e o f hea t i ng

Comparing the data plotted in Figs. 1 and 2 for the continuous uniformrates of heating of 10 and 5C min- , respectively, it is found that theformer rate allows the reactions to proceed through more than one step,independent of initial particie srze, for both samples. The areas under thepeaks are larger in the case of the high temperature gradient, i.e. 10Cmm-, which proves that the slower rate of 5OC min- allowed partialfixation of carbon.Dynan n c h ea t t r ea tm en t

In this technique the samples were subJected to a skpwise increase oftemperature, at a uniform heating rate, for various periods of time. Theweight loss data are given in Table 1. The relative weight changes, R(a t,me%m,n ), are also included.Plots of R against temperature for both samples (Fig. 3) show that Rrelations are higher for the Adg sample. High values of R are obtained whenthe samples are held for longer time periods at low temperatures. Above400C longer incremental times produced similar weight changes irrespectiveof the temperature, i.e. the rate of loss is slowed down. Plots of residualfraction ln( 1 - cy) against time (t) at different temperatures (Figs. 4 and 5)gave straight lines in the region up to 10 min (Fig.6). From these straightlines, the rate constant (k) was determined at each temperature. A plot oflog k vs. the reciprocal of absolute temperature (l/T) gave satisfactorystraight lines with two slopes (Fig. 7). The activation energies arz calculatedfrom these curves using the Arrhenius equationd In k / dT= E /RT2

n k = g I ons,)

8/12/2019 bone char 2

6/13

8/12/2019 bone char 2

7/13

351

Adg dynamic

Time In minutes t 1Fig. 4. Relation between In( I - a) vs. time of the d nanuc treatment at different temperaturesfor the Adg sample.

orAdwg dynamic

- R, ~200 TI 0

-0.20 I I I I I0 60 120 100 240 300TWW in minutes t t 1Fig. 5. Relation between In( l- a) vs. time of the dynamic treatment at different temperaturesfor the Adwg sample.

8/12/2019 bone char 2

8/13

352

t t > K at 200 C2 1 K at 300 T3) K at CO@*C 4-0 15 4)Kat SOOT5) K at 6OOC6) K at 700-C 5

Dynamic-0 20 - P Adg

--0 Adwq I I I I I0 2 4 6 8 IO

Time tn mlnuter (t 1Fig. 6 In I - a) vs. time (up to 10 mm of dynamic treatment at different temperatures forAdg and Adwg samples.2.06 kcal mole- . Two such different energies are obtained for the supenm-posed reactions being dehydration, dehydroxylation and decarbonizationThis proves that the transport mechanism has actually changed. In thepresent system the heterogeneous reactions occurring are obviously con-trolled by two classes of rate controllers. The nucleation reaction has beenshown to be controlled by the phase boundary rate while the oxidationreaction is controlled by the transport rate [l]. The exothermic peak previ-ously recorded for the oxidation of carbon at 380 and 410C for Adw andAd samples [4] means that the process is controlled by heat flow where the

Temperature C 1_3o1 yyo sp 400 300 zooI I

0 E = 4 75 k cal. mol-o E = 2 06 k Cal mol:

1.0 1.2 1.4 16 8 2 0 2.2

Fig. 7. Activation energy of both samples calculated from the relatron between log k vs. l/T.

8/12/2019 bone char 2

9/13

353heat evolved tends to raise the interface temperature. This may explain thelower activation energy value obtained above 400C.I s ot h er ma l h ea t t r ea tm en t

In this method the preheated samples were kept for certain lengths of timeat constant temperatures and the weight changes were noted. The sampleswere preheated and quenched at 4OOOCat a rate of 10C min- for 10 min.This temperature proved to be a deflection temperature at which theactivation energy of the reaction changed upon dynamic heat treatment. Theloss extent ((IL)as well as the relative weight changes (R) are given in Table 2.Plotting the relative weight changes (R) against temperature (Fig. 8)illustrates higher R at the lower temperatures experienced, being higher forthe Adwg sample. The relative weight changes apparently proceeds to a statewhere further increase in temperature does not cause a substantial decreasein weight. This is to be expected since it has been shown that as thetemperature difference between experiment temperature and equilibriumtemperature decreases, the diffusion coefficient and driving force decreasefor dehydration and decomposition reactions [ 1,5].Plots of residual fraction In(1 - cy) vs. time (t) for the various tempera-tures are illustrated in Fig. 9 (A and B) for the Adg and Adwg samples. ThisFig. exhibits behaviour similar to the dynamic heat treatment in that thefraction remaining after 120 min does not change much with higiler incre-mental times. Also, the first 10 min yielded the highest change. Plots of

Temperature C 1Fig 8 Relative loss extent vs. temperatureorA =Adg and B=Adwg sample.. after isother-ma l tree+ment.

8/12/2019 bone char 2

10/13

TA2

Toeeaedeemreavooeeae1mnebsohmem

Tm Q

Faoreea

1m

6rnn

Ravwgc

R

1m

lCm

2m

R

R

R

R

61

11

11

21

Asmea

4

00

5

00

5

OO6

6

00

6

00

Awsmeb

4

00

5

001

5

00

6

00

6

00

00

00

00

00

I2

14

15

15

00

00

001

00

12

13

14

14

00

00

00

00

12

13

13

13

00

00

00

00

11

12

13

13

00

01

01

01

11

11

12

12

00

00

00

00

14

17

18

18

00

00

001

00

13

13

16

16

00

00

00

00

12

14

15

15

00

00

00

00

11

13

13

13

00

00

00

00

11

12

13

13

Dmcema4OC01

bDmcema4C00

.

_-_-._

.

_

Y

_

-

.

8/12/2019 bone char 2

11/13

Iso-Adg Iso- Adwg

3 -0.06

5 -0 080 L5OC. soo*c-0 10 a 55O Cp BOO Cx 650.C

-0.12

tfme in minutes ( t 1Fig. 9 Relation between In( I- a) vs. time of isothermal treatment at dtfferent temperaturesfor A = Adg and B = Adwg samples

ln(l- a) vs. (t) for the first 10 min, at various temperatures, gave the rateconstants (Fig. 10A and B). The activation energies are obtained by plottingthe rate constants (k) vs. the reciprocal of the absolute temperature (1 /T)

000002o.c40060 00

1) k at 4so.c2) k at 33~3) k at 550~4) k at 6OOt5) k at SS T

0.040.06

2 4 6 0 10Tame in minutes t )

Fig. 10. ln(l - a) VS. time (up to 10 min) of isothermal treatment at different temperatures forA = Adg and B = Adwg samples.

8/12/2019 bone char 2

12/13

356-3 I 450 SOO 550 600 650 Temp C1 I 1 t I

Adg es 4.07 k cat. mof-3.0 Adw o 1407 k cal mol-1

1000 I TFtg 11. Acttvatton energy of both samples calculated from the relation between log k vs.l/T.

(Fig. 1 I). Beth samples gave nearly the same activation energy, indicatingthat material transport proceeded via one and the sapne mechanism.Comparing the activation energies recorded for ihe isothermal alld dy-namic heat treatments at temperatures above 400cC (4.07 and 2.06 kcalmole- . respectively) reflects the different conditions of the two procedures.The lower activation energy in the case of the dynamic treatment is attri-buted to the gradual proceeding of the reactions rather than the isothermalone having its rate dependent on diffusion. The greater activation energy ofthe isothermal treatment means greater sensitivity to thermal changes [6].

CONCLUSIONSGrinding of both samples accelerated the burning process due to thesmaller particle size and newly active exposed surface for reactions.The continuous uniform rate of heating of 10C min- was found to bemore favourable, as undesirable fixation of carbon occurred with the rate of5C min- as tested thermogravimetrically.The loss extent increased with increasing the time of treatment at lowtemperatures, i.e.

8/12/2019 bone char 2

13/13

357mechanism for the two samples with 4.75 kcal mole- up to 400C and 2.06kcal mole- from 409 to 7OOOC.

The reactions in the case of isothermal treatment also prcceeded via thesame mechanism with an activation energy of = 4 kcal mole-.

REFERENCES1 W.D. Kingery, Introductton to Ceramics, Wiley, New York, 1st edn., 1960.2 V. Sgtava and F. SkTtvar, J. Am. Ceram. Sot., 52 (1969) 59 13 W.W. Wendlandt, Thermal Methods of Analysis, Wiley, Chichester, U.K., 1964. pp.29-404 W-1. Abdel-Fattah, T.M. Ghazi and M.M. Selim, Thermochim. Acta. 51 (1981) 287.5 W. Jost, Diffusion m Sohds, Liquids, Gases, Academic Press, New York, 1952.6 Z.D. Jastrzebsk, The Nature and Properties of Engmeenng Matenals, Wiley, New York,

2nd edn., 1976.