CA5TELL6:" (5I'AII\) iB QUALI , 200 2

RECYCLING OF SOLID WASTER IN THE

PRODUCTION OF CERAMIC FLOOR TILES

M. C. Casagra n de""" ', A. P. No vaes de Oliveira'", D. Hotza ''' ',

A. Ol iveira da Silva" ', L. R . A lexan d re "'"

,'S EN AI! CTC - Ce nt ro d e Tecnolog ia e m Cerarnica, Ru a Ga l. Lauro Sod re, 300
Cx. ['. 3247, 13. Co mercia rio - CE P 8802-330 Criciu rna (SC) - Bra zi l
Em ail: ap no @senai-sc.ind .br
''' 'Unive rsid ad e Fed eral d e Sa n ta Ca tarina, De pa rta ment o d e Engenharia Mecani ca,
Laborat 6 rio d e Mat erials, Caixa Postal 476, Ca m p us Unive rs itario - Trinda d e, 88040-900
Flo rian6polis (SC) Brazil
e-mail: m ar coscasag rande@zipmail.com .br
[''' 'Cecrisa Revestim en tos Ce ra rnicos S.A. Av. Manoel D. d e Freit as, 1001 - B. Prospera
CEP 888 13-900 Criciu m a (SC ) - Brazi l. Home Page: ww w.cecrisa .com.b r

A BSTRACT

Scrap [rom the selection of fi llished ceramic tile product» is a solid waste orisillatillS ill the
nunntjacturi ng. process. The bulk of thcec solid wastes is usually deposited in IllI ldfills. 111 the
traditional double fir iuS process ( W illi tiie), the orisillS ccm mic tile solid was/['s are /101'111011 11
recycled by wei s rilldillS, ivith other raw nuttcrials, fol'llliuS tire cerumic body. Tlii« is possi/Jle
beca use double-fired ceram i c tiles are thinner (fi'0lll 4.0 10 6.5 111111 ) and are fired at icmpcraturc«
up to 1090 "C. Moreoz'er, bellIS higlrly porous, tlrey are 1II0re brittle. 111 the CIlse of solid wIIs tes
generated by lire singtc fi rillg process, recyclillg is 1I0t possi/J!e unthout suitll/J!e prepious
processillS, since the products lire thicker thun Ilrose obtuinct! by double fi rillS (from 6.5 10 11 .0
111 111) 0/111 are harder. Conscqucnt lu. s rilldillg these products becomes diffiCIIII by the process
nornutllu elllployed ill ccrantic tile industries. Will i thi« purpose, the obicctive» of this work lire to
ellable recyclillS solid wastes (scrap arisillS ill tire selection of tire fi lial product) by lire uslllS a
su i table dry III ill i IIg process, alld cstat)! ish tire influ ence of tire addi tion of differellt amounts of these
solid lms tes ill a certnnic body (used fo r lire production of ceramic floor tiles) prepared I,y ll'el-
s rilldillS·

1. INTROD UCTION

The m an u factu re o f u tensils in cera mics is on e of th e oldest hum an ac tiv ities . Clav
d ish wa re has bee n foun d d a ting ba ck to 15000 B.C. a nd m an u facture was we ll d eveloped
in Egyp t ten ce nt uries late r. Th is occ u rred du e to some ve ry simp le p ractical reasons: it is
easily sha pe d or m ou ld ed, the ra w m a te rials a re th e most com m on in the ea rth's crus t a nd
p rod uction d oes no t req ui re adva nced kno w led ge . The need for m a te rial s with be tte r
p ro perties to the ex isting ones has led m anki nd to d evelop cer a m ic tech no logy, tha t is the
t'o-, • J<l
III QUALI /? 2002 CASTELL6N (SPAIN)

scie nce applied to se ve ra l m anufacturing processes o f cera m ic produ cts, including

ce ram ic floor and wall tiles, w hich a re of g rea t interest in th e e ngineer ing a rea an d are
widely u sed for coveri ng th e m ost varied environ me n ts.

In th e ceram ic tile indust ry, th e ra w m at erials are ex tracted fro m natural mines (an d
con ta in sign ifica n t d egrees o f impurities a nd physica l, che mica l a nd m ineral ogi cal
variability) an d p rocessed by m eans of sin te ring, in wh ich b u lk p owders are prepared a nd
phys icall y and therma lly treat ed, giving th em sha pe and m echanical p rope rti es, a nd
consolida ting them'". This p rocess essen tia lly in volves the s tages of ra w mate ria ls
weig h ing, milling, s p ray -d ryin g, pressing, drying, g laz ing, firin g a nd se lectio n o f th e final
prod u ct, as shown in figure 1.

It is important to point ou t th at in thi s co ntex t, th e wastes genera ted in the course of

the proces s (shown in the figure 1) a re curren tly reincorporat ed in to the process, bei ng
used as ra w material. The so lid wastes ge ne ra ted during fina l p roduct se lectio n a re
characterized by the presen ce of a d e fect in th e fired product, m a kin g th em unsuitable for
use. These solid wastes are not frequ ently u sed because th ey are h ard m at erial s and ,
cons eq uen tly, diffi cult to mill. At th is moment, th ey are u su ally d eposi ted in land fills.

RAW MATERIALS

I
....-
WEIGHING
TANK LIQUID
WASTES
-...
RAW MATERIALS
(GLAZ E) I
MILLING
I
~
+ SPRAY DRYING

•
W EIGHING
I I
~ PRESSING
MILLING

... J.
DRYING
STORING

...
GLAZING

. I

SINTERING OR
FIRING

I ...
SELECT ION

SOLID WASTES
(LANDFILLS)

Figure 1. Ftouxtmn of till' ceramic jicor tilt' production 1' r O(t'::;5 .

I'us - .ill
C ASTEL L6:'-i (SI'A I:'-il -
~ QUALI

Throughou t all thi s evo lu tio n, ma nkin d became us ed to ta king ad va ntage o f the
natural so urces of the planet and ge nera ting wastes w ith o u t any concern: sources were
abu nd a nt and n ature u sed to accept a ll th e re fus e w ithou t comp lain ing. Accord ing to
Toffler!", af ter the l S'" cen tu ry, w ith th e s p rea d of th e in dust ria l revol u tion, th e
deve lopment model or s tra tegy of th e natio ns has co nsoli da ted th ei r social a nd techn ique
base s. The m a in objecti ve w as sh o rt term econom ic grow th, usi ng ne w p rodu ction
processes and inten se exp lo ra tio n of energy and raw m a te ria ls, w hose so u rces w e re
co nside red lim itles s. Th is mode l genera ted im p ressive excesses o f eco nom ic wea lt h, but it
has als o brou ght g reat soc ia l a nd en vironmental p ro b lems, includ ing w a stes.

In the Brazi lian ce ram ic tile in d u stry the vo lu me of the m a terial di scard ed b v
b reakage repres ents o n a verage 3.0;;;' 01 a ll n a tiona l prod uction . Correla ting th(s
pe rce nt age wi th an e ffective prod u ction in 2000 o f 452.7 m illion square m et ers, acco rding
to ANFACERl JI, we h a ve a pproxima tel y 200 th ou sa nd ton s a yea r of th is was te bein g used
as landfi lls. A nd the feature of co ncen tra ting thi s ma teria l in a speci fic a rea (la nd fill)
constitutes a p all iati ve but also harm ful mea sure. Pa lliati ve because w ith th is quantity o f
rejected materi a l 80 squ are kilometres a re occup ied ann ual ly. Ha rm fu l becau se these
materi a ls poss ess w a te r-solu b le consti tuents in the ir vitreous coa ting, such as Lead,
Ca d m iu m and other ha rmfu l metals, w h ich represent a great danger for hu m an heal th!".

In th is co ntex t, the objectives of thi s work are to enab le recycl ing th ese so lid w as tes
by us ing an appropriate m illing process to ob ta in a powder to be add ed in the m illing ste p
of a w et m illing process in floor tile ma n ufacturin g and to ve rify the inf luence of th is
a d d itio n o n the cha rac te ristics of the final p rod u ct.

2. EXPERIMENTAL PR O CED UR ES

So lid wast es w er e separa ted in s u fficien t amounts to ca rry ou t the tests co ns id ered
on a lab scale. These materials were put through a cru sher mi ll (Hazernag APS-0604B),
and dry milled wi th the aid o f a h am mer m ill un it (Ha ze rna g N ovo ro tor 650 /500 11), both
these facil ities belo n ging to a recycl ing tr ia l pl ant as show n in fig ure 2. The powder
obtained was s ieved u sing a 35 mesh sieve.

1. feed of solid wastes

2. transport belt
3. hammer mill
4 . elevator
5. crusher mill
6. powder storing
7. suppo rt

.. ~: 0)

i
•
\ ..
_
t _ b~~~

~2 1000

Figllrt' 2. f~c cy cli l1X ptant It/,lf (l U t.

Po-, - -.I I
_II QUAU J2/l. 2002 CASTELL6 K (SI'AIK I

The other raw materials making up the composition we re dried and deaggregated
manu ally, wi th the aid of a mortar, u p to a pa rticle size lower than 2.8 mm (7 mesh ).

The raw materials powd ers and solid wastes were processed acco rd ing to the
flowc ha rt show n in figu re 3. Each powder w as weigh ed in acco rda nce w ith the standa rd
form u lation , conta ining three d istinct clays, two argillites and one talcum. To the sta ndard
form u lation, id enti cal to the one used in a ceram ic tile comp any, we ad ded 1, 3, 5, 10 and
30 w t% rejects or so lid wastes, respectively. The six different resul ting formu lated batches
were se parately milled in a wet process, in a 5,000 cm' ball mill for 40 minutes. The milling
reject was found to var y from 5.8 to 7.2 wt% on a 325 mesh sieve, while den sity ran ged
from 1.66 to 1.69 g / cm'. The fl ow time varied from 25 to 35 seco nd s in a number 4 Ford
cup. Subse que ntly, the ceram ic sus pension (slurry) resulting from the milli ng process was
dri ed in an oven fo r 24 hours at 110 "C (± 10 "C) and humidified to 7 wt%. The we t powder
wa s passed th rou gh a 16 mesh sieve for granu lating and pressing.

Each powd er was compacted in a lab hyd rauli c press (mo del Gabrielli 24 / 11 0A)
with a specific pressu re of 25 Mf'a. The compacted sam ples we re sintered in a kiln (mo del
Scha ly LAB 44) at 11 50 "C, using a heating rate of 10 "C/ min from roo m temperat ure to
1000 "C and 5 "C / min from 1000 "C to 11 50 "C, w ith a hold ing time of 10 minut es at 11 50
"c. After sintering, the samples we re air-que nche d to room temper ature.
STAGES
'..-.A.NAI.- ySIS j
L_ _ _,

RAW MATERIALS
SOLID WASTE
PRE PARATION
PREPARATION

~
I BATCH FORM ULATION

.!. "
I WET MILLING
;.. de nsity, flow
t ime and
m illing rej ect s

,
I SLURRY DRYING
I
I Particle size
distribution

DEAGGREG ATION,
, viscosity
W ETIING AND POWDER
GRANULATION

;... MOR and bulk

f--- - - - - - ...... de nsity

..... MOR, wate r

' -- - - - -- - ...... absorpt ion and
shri nkage

Figure 3. Flowchart of labora tory proCt'ssillg.

C ASTELL6 "! (SPAI"!) !!l QUALI . 2002

Slurry sa m ples o f each milled ba tch were also collected and subjec ted to th e laser
scattering particle size an alyzer (mod el Cila s 1064L) for d eterminati on of the particle size
di str ibution. The flow curve of each sa m ple was determined using a viscomete r (m od el
Bohlin 88 BV). Th e che m ical com position of each powder w as determ ined in a XRF
spectro meter (m odel Phillips PW2 400). The dry com pacted bodies were su bjected to
ben ding strength tests a nd bulk density m easurements accord ing to the Arch im edes
p rin ciple with m ercury immer sion method . Th e sin tered bo d ies were also su bjec ted to
bendi ng strength tests, wate r ab sorption and linear thermal sh rinkage mea surements.

3. RESU LTS AN D DISCUSS ION

Th e sta ndard com position o f the stu d ied floor ceram ic tile was ob tained u sin g three
di fferent kinds o f cla y, tw o argillites and a talcum, accordin g to th e formu lat ion shown in
tab le 1.

Ran' mate rials % \\'t

Argi llites A 30 .0
Argill ite s B 20 .0
Clay I 20 .0
Clay 2 10.0
Clay 3 13.0
Ta lcum 7.0
To tal 100 .0

Table 1. St andard ceram ic body compos ition,

Table II shows th e che m ical com p osi tions of the standard cera m ic bod y, before
firing, and th e so lid waste aft er firing. It can be observed that the main m eaningful
di fference is th e va lues of loss on ignition (L.a .!.), besides th e acceptab le variat ions
inherent to th e ra w m aterials and to the m anufacturing process. A ltho ugh th e solid w aste
has a vitreo us glaze layer, the sim ilarity bet ween the chem ica l com pos itio ns of th e
stand ard ceramic body an d th e solid w as te suggests there sho u ld no t be any interference
in the fin al form u lati ons, after ad d ition o f any am ou nts of solid waste .

C O ~I POUN DS STAN DA RD C E RA M IC SOLID WASTE

BODY ",t %

w t%
Si0 2 68. 3 69 .4
A I, O, 16.4 17.1
FC20 t 3.3 3.5
Ca O 0.6 0.9
1\'a:O 0.8 0.8
K20 2.9 3.3
MnO 0.0 0.0 I
Ti0 2 0.3 0.7 I
MgO 2.4 _) . "1 I
P,O, 0.1 0.1 I
ZnO 0.2 0.5 I
z-o, 0. 1 0.7 I
L.O .1. 4.6 0 .3 I

Table II. Chemical alJ lIly~ i~ .

Po,,; - -B
 CA STEL LO[\; (S['A I[\;)

Table III sho ws a compa riso n of th e particl e size d istribu tion between th e sta ndard
ce ra m ic body and th e so lid wa s te, w h ile tabl e IV s hows a compa rison o f th e pa rticle size
di st ributi on bet ween the s tand ard ceram ic body a nd othe r cera mi c bod ies with solid
wa ste ad d itions.

CUMULATIVE VAL UES STANDARD CERAMIC SOLID WASTE

BODYCuml ( um)
Diameter at 10% 0.93 3.56
Diameter at 50% 7.49 44.60
Diameter at 90% 3 1.54 268.89
Diameter at 100% 75.00 500.00

Table III - SimI/lard ceramic body ami ~(lfjd U'1l~ t l' particle size distri bution,

In tab le III It ca n be observed that th e solid wa ste has a hi gh e r pa rticle size th an th e

s ta nda rd ce ram ic bod y; therefore it must be ad d ed to th e stand ard m ass, together wi th th e
com position' s raw m ateri als in th e milling ste p .

The utilization o f a crusher mill a nd a hamm er mill , ava ilable technologies at a n

accessible cos t, is su fficien t to red uce th e solid waste to a powder w ith an av e rage particle
size lower th an 500 11m (32 mesh). This po wder, when ad de d to th e com pos itio n o f th e
cera m ic body th at origin at ed it, in percentages th at va ry fro m 1 to 30 wt% and su bjected
to th e we t milling process, d oes not signi fican tly affect th e particle size di s tribu tion of th e
res ul ting s us pens ion, as sho wn in tabl e IV.

CUM ULATIVE VALUES STD FI F3 F5 FlO F30 I

Diameter at 10% 0.93 0.97 0.95 0.87 0.98 1.05 I

Diameter at 50% 7.49 7.75 9.88 8.09 8.44 9.65

Diameter at 90% 31.54 32.25 32.04 32.59 33.49 35.68

Table IV - Particle size distributio n of the formulations obtained.

Wh ere:
STD = Sta nda rd cera m ic bod y com position
F1P = Sta ndard ce ram ic body com position w ith solid w aste ad d ition of 1 w t%
F3P = Sta ndard ce ram ic bod y com positio n with solid w aste ad d ition o f 3 wt%
F5P = Stand ard ceram ic body com position with solid waste ad d ition of 5 wt%
FlOP = Standard ceram ic body com pos ition with solid w aste ad d ition of 10 wt%
F30P = Stand ard ceram ic body composition with solid was te ad d ition of 30 wt%

In table IV it can be observed th at th e so lid wa ste w as well incorporat ed in to th e

cera m ic bo dy com positio n, show ing th at th e com m in u tion process was eno ug h to m ak e
th e particle size su ita ble o f th e added solid waste. That happens due to th e fact th at the
1'0"· .\4
 2002

ceram ic ma terials possess hi gh shea r resistance, w h ich lead s to hi gh co mpression

stre ngth, together with no tch se nsi tivi ty and lo w frac ture resist an ce, or they are brittle
mat e rial s!".

Tabl e V, shows the res u lts regard ing the physica l tests for th e studied fo rmu la tions.

Samples Bulk de ns ity !\lOR Green Vtscnslry Linear T hermal water MOR Fired
(glem') Dry Sa mr.)e (Pa. s) Shrinkage Abso rption Sa mple.
(:\' /m m ) \\" t% \\"t% (N / mm' )
ST D 1.9.1 3.9 0 . 15 .1.9 9.0 2X.1 I
FI 1.95 3. 7 0 .1 7 3.9 8.6 2X.6 I

1'.1 1.95 3.X 0 . 17 3.9 S.9 26.7 I

F5 1.9 3 4.3 0.\ 3 3. 7 S.S 25 .2

1'10 1.9.1 ~ .I 0 .1 3 .1X 9. 1 2 ~. 7

F.10 I.X5 2. 1 0 . 10 39 10. 1 20.2 I

Wlll'r t2: ~ 'O R = Mod ul u... o f Ru p tu re

The in fluence of the ad d ition of m illed soli d waste, in a cerami c floor tile bod v, in
perce nt ag es varyi ng fro m 1 to 10 wt j!" d oes n ot a ffec t properties su ch as: slurry vi scosity,
lin ear therma l sh rinkage, water absorpt ion and be ndi ng stren gth (g reen and sin tered ),
Thi s assu res a ve ry good safe ty margin in u tilization in the stan d a rd cera mi c bod y,
wi thou t m ean ingful effec ts on the ch a rac te ristics o f the fina l prod uct. On the oth er hand,
it is an in teres ting way o f using this w aste, sin ce th e cu rren t demand correspo nd s to abo u t
3 wt% o f the produ ction .

By ad d in g 30 wt j;, of the sol id waste , a sig n ifican t increase is noticed in water

ab so rp tio n . Th is ca n be exp lained by a low e r d en si fica tion of th e compa cted bod ies due
to the low p lasticity of the solid wast e. Th is sa me fa ct explains th e sig n ifica nt red uction in
their bendi ng strength . Ho wever, this effec t can easily be revert ed wi th the add ition of
larger a mounts of p lastic raw m ate ria ls (clay ), w ithou t a ny add itiona l cost" !

4. C O N C LUS IO NS

The laborato ry s tud ies dem onstr a te the p e rfect feasibility of re using ce ra m ic floor
tile so lid wa stes as a raw m a te rial source, to be fed back in to the same prod u ction process
in the m illing stage. From an econ om ic poin t of view, th is ma te ria l can p lace the ce ra m ic
flo or tile manufacture r in a strong co m pe titive positio n in the m ar ke t. Mo reove r, it
dimini shes th e ambien t liabi lities of the com pany, at the same time p reservi ng the
environment and the he alth of the pop u lation . For exa m ple, some g las ses co n tai n lea d ;
d epen d ing on the frit u sed , w h ich co u ld be sodi u m -based for exa m p le, th is di sso lves in
th e p rese nce of w a te r an d co n ta m ina tes the soi l.

The so lid w astes, a fte r tr a nsform in g th e m in to powder, can be u sed for ot her specific
ap p lica tio ns curren tly in d e velop m ent by th e a u thors.

The developed re use process has been shown to be techni call y a nd economi ca lly
feasib le for th e compa n ies of th e sector, w hi le th e tech nolog y in volved is compatible with
II QUALI~2002 CASTELL 61': (SPAIN)

cu rrent produ ction de mands and w ith the lay-outs of the ceram ic pl ants, which further
reinforces the feasibility of th is p rocedure.

The basic principle of th e p rocess can be extende d to the handling of solid wa stes in
other sectors, such as civil construction, metallurgic industri es (casting sa nd ) and
thermoelectric plants (fly ashes), amo ng others.

REFERENCES
[I J- !v1AN FREDIK I, T. Estrutura dos materia is ce ramicos Pa rt e - I. Ccrd mica lnforma cao . n.12, ago /set. 2000. pA2-i .
121· TO FFLER, ALV I ~ . A terce ira onda . 25.ed. Sao Pau lo: Record, 2001.
131· AI\' FAC ER - Associaca o d os Fab rica ntcs de Ce ram ics pMa Rev cstimcn to. Desempenho dol in d us tria cerdm ica . 5Jo
Pa ulo, Catalogo informa tivo, 2000. www. anfa cer.org.b r: acessad o em 15 o ut. 2001.
I·l} BEDESCH I, E.; CAPUANO F. and r,,1ED ICI, c. Ri fiuti ce ram ici e terreni con ta m in ati: Pro cedure d i s ma lt ime nto e
tecnologi e d i bonifica. Q UADERNI 0 1 TEC;\sICH E DI PROTEZIO r\ E AtvtBIENTALE. Cc ra mica e a mbie nte ne gli
a nni 90. Bologna : Pitago ra Edi trici. 1994.
[5]- VAN VLACK, L H. Pr in cipi os da cienc ia e tecnolog ia dos materiais . l l .cd . Rio de Janei ro: Ca m pos. I&W-. p.326-7.
[61- OLIV EIRA, A. P. N. et al. D esen volvimento de urn processo integrado para rea pro vei tam ento de re jeitos d o seto r
cerdmico. CTC - Centro de Tecnologia Ce rarnica. Criciu ma: s.cd. 2001. n .p.

Pes - -l.h