Ci a ion: Ne ˇ i o á, L.; Zemánek, D.
S udy o he Mine alogical
Composi ion o an Alumina–Silica
Binde Sys em Fo med by he
Sol–Gel Me hod. Ma e ials 2023,16,
5466. h ps://doi.o g/10.3390/
ma16155466
Academic Edi o : Daxin Li
Recei ed: 30 June 2023
Re ised: 21 July 2023
Accep ed: 24 July 2023
Published: 4 Augus 2023
Copy igh : © 2023 by he au ho s.
Licensee MDPI, Basel, Swi ze land.
This a icle is an open access a icle
dis ibu ed unde he e ms and
condi ions o he C ea i e Commons
A ibu ion (CC BY) license (h ps://
c ea i ecommons.o g/licenses/by/
4.0/).
ma e ials
A icle
S udy o he Mine alogical Composi ion o an Alumina–Silica
Binde Sys em Fo med by he Sol–Gel Me hod
Lenka Ne ˇ i o áand Da id Zemánek *
Facul y o Ci il Enginee ing, B no Uni e si y o Technology, Ve e i 331/95, 602 00 B no, Czech Republic;
[email p o ec ed].cz
*Co espondence: [email p o ec ed].cz; Tel.: +420-60-4341018
Abs ac :
Colloidal bonds a e ealized by sol–gel echnology. The binde sys em o he e ac o y
cas able belongs o he Al
2
O
3
–SiO
2
bina y diag am. Mulli e is he mos he mally s able mine al
in his sys em. This wo k was mo i a ed by an a emp o maximize he mulli e con en in he
NCC binde sys em, because a high con en o mulli e is a gua an ee o he long se ice li e o
e ac o ies. Ini ially, he mine alogical composi ion o he pu e gel was es ed a e d ying and i ing
a empe a u es be ween 1000
◦
C and 1600
◦
C. The beha io o he gel du ing d ying was desc ibed.
Subsequen ly, a me hod o minimizing gel sh inkage du ing d ying was sough . To his aim, ine
ille s (mic o ille s) o alumina and silica we e es ed. In pa icula , he eac i i y o he mic o ille s,
he abili y o he mic o ille s o eac wi h he sol o o m mulli e, and he d ying sh inkage o he
mic o ille -doped gel we e e alua ed. The s udy showed ha he leas sui able sou ce o Al
2
O
3
in
e ms o i s eac i i y is abula co undum, which p oduces he lowes amoun o mulli e. The in e nal
s uc u e o he p epa ed binde sys em when using di e en mic o ille s was desc ibed. Based
on he esul s om he second s age o he wo k, se e al comple e ma ixes o he binde sys em
we e designed and he deg ee o hei mulli iza ion a di e en i ing empe a u es was s udied.
Du ing his s age, i was shown ha he deg ee o mulli iza ion o he binde sys em depends mainly
on he mic osilica con en . In he binde sys em, he maximum mulli e con en eco ded was 76%.
The e ec o amo phous SiO
2
on he bulk densi y and in e nal s uc u e o he binde sys em was
also desc ibed.
Keywo ds: sol–gel me hod; e ac o y; cas able; binde ; mulli e; mine alogical composi ion
1. In oduc ion
The sol–gel bonded no-cemen cas able (SGBNCC) is a ype o e ac o y cas able ha
was de eloped o educe he need o cemen in he manu ac u ing p ocess. The bond
ob ained om cemen and calcium alumina e cemen is connec ed wi h he de elopmen
o high s eng h a empe a u es below 1000
◦
C [
1
,
2
]. A gene al disad an age o his is ha
co esponding linings need o be hea ed up e y sensi i ely, especially du ing he ini ial
hea ing. Because he cemen con ains calcium oxide, he e is also a isk o he o ma ion
o low mel ing phases. The p esence o calcium oxide in he cemen poses a isk o he
eme gence o low-mel ing phases [
3
]. The SGBNCC has many ad an ages o e adi ional
e ac o y ma e ials. As such, i has become inc easingly popula in he e ac o y indus y
and may become a majo playe in u u e applica ions. The use o sol–gel bonding in he
p epa a ion o a e ac o y cas able wi hou cemen is a ela i ely new echnique and is
becoming mo e and mo e impo an in he cons uc ion indus y due o i s ad an ages
o e adi ional me hods.
The sol–gel p ocess is ypically ca ied ou by mixing a p ecu so solu ion, which
con ains nanopa icles ha make up he solid ma e ial and a sol en [
4
,
5
]. The sol can
hen be ans o med in o gel, ei he by e apo a ing he sol en o lea e a solid ma e ial,
o by adding a chemical agen ha p omo es he c osslinking o he pa icles o o m a
Ma e ials 2023,16, 5466. h ps://doi.o g/10.3390/ma16155466 h ps://www.mdpi.com/jou nal/ma e ials
Ma e ials 2023,16, 5466 2 o 14
h ee-dimensional ne wo k. The esul ing solid ma e ial, which is made up o a ne wo k o
in e connec ed pa icles, is known as a gel. This gel can be used as a binde in a no-cemen
e ac o y cas able [6–8].
In o de o p epa e a no-cemen cas able using he sol–gel me hod, se e al impo an
ma e ials a e equi ed. The p ima y inpu s a e e ac o y agg ega es and a sol–gel binde .
The agg ega es used can a y depending on he applica ion and he desi ed p ope ies o
he cas able, bu hey a e ypically made o ma e ials such as alumina, mulli e, bauxi e,
andalusi e, co undum, silicon ca bide, silicon ni ide, o zi conia [
9
]. The binde is mos
commonly colloidal silica o alumina; hei ans o ma ion o gel leads o a “glue e ec ”
whe e agg ega e pa icles a e linked oge he [10].
The heological beha io o he ma e ial is also impo an ; i should be able o low
easily and ill small oids in he mold wi hou se ing oo quickly [
11
]. S udies ha e ound
ha he addi ion o su ac an s [
12
,
13
] o mic osilica [
14
] can imp o e lowabili y and
adjus se ing ime.
D ying is an impo an s ep in he p oduc ion o sol–gel bonded cas able e ac o ies, as
i helps o emo e excess wa e om he mix u e, pa icula ly when he gel is o ming s ong
bonds among he pa icles. The emo al o excess wa e om he mix u e is bene icial,
pa icula ly when he gel is o ming s ong bonds among he pa icles.
The d ying p ocess ypically begins wi h a p elimina y d ying s ep, in which he
cas able is le o d y a oom empe a u e o a pe iod o ime. The du a ion o his
s ep depends on he size and shape o he cas able, as well as he ambien humidi y and
empe a u e [15,16].
A e he p elimina y d ying s ep, he cas able is ypically placed in a d ying o en,
whe e i is hea ed a a con olled empe a u e and humidi y. When i is used as sho c e e,
he lining empe a u e inc ease mus be con olled and planned p ope ly. When i is used as
a sho c e e, he empe a u e inc ease in he lining mus be con olled and planned p ope ly.
The empe a u e and humidi y used du ing his s ep depend on he composi ion o he
cas able and he desi ed p ope ies o he inished p oduc . In gene al, he empe a u e o
he d ying o en is kep low, be ween 50 and 150
◦
C, o p e en any chemical eac ion o
sh inkage in he cas able. In gene al, a longe d ying ime is needed o la ge and hicke
pieces o cas able han o smalle and hinne pieces o he cas able [
14
–
16
]. The d ying o
SGBNCC is a c ucial s ep in he p oduc ion p ocess, and ca e mus be aken o ensu e ha
i is pe o med co ec ly and ha he inished p oduc has he desi ed p ope ies.
The du abili y o SGBNCC is ypically be e han ha o he adi ional no-cemen
cas able e ac o ies. This is due o he unique p ope ies o sol–gel binde s which p o ide
se e al ad an ages o e o he ypes o binde s.
One o he main ad an ages o sol–gel binde s is ha hey c ea e e y s ong bonds
among he elemen s o he e ac o y agg ega e. This imp o es he mechanical s eng h o
he cas able e ac o y, making i mo e esis an o ab asion and co osion. Addi ionally,
because sol–gel binde s a e made om pu e oxides, hey a e able o wi hs and high
empe a u es and he mal shock be e han o he ypes o binde s [
17
–
20
]. This means ha
hey a e less likely o expe ience c acking o de e io a ion due o he mal cycling o apid
empe a u e changes.
The du abili y o SGBNCC also depends on he composi ion o he e ac o y mix u e
and he ope a ing condi ions o he u nace o equipmen in which hey a e used. The
quali y o he e ac o y agg ega e, i s pa icle size, i s shape, and i s chemical composi ion
play a c ucial ole in de e mining he du abili y o he cas able.
The mine alogical composi ion o a sol–gel bond can also ha e a signi ican in luence
on i s du abili y [
21
,
22
]. The sol–gel bond p o ides he cas able wi h i s s eng h and
du abili y unde high- empe a u e condi ions. The mine alogical composi ion o he sol–
gel bond can a ec he p ope ies o he cas able in a numbe o ways, including impac ing
i s he mal expansion coe icien , he mal conduc i i y, and co osion esis ance [23].
One key ac o ha can a ec he du abili y o e ac o y cas able is he p esence o
alumina in he sol–gel bond. Alumina is a highly e ac o y mine al ha is able o wi hs and
Ma e ials 2023,16, 5466 3 o 14
e y high empe a u es, and i is o en used in e ac o y cas able o enhance i s he mal
pe o mance. Howe e , i he concen a ion o alumina in he sol–gel bond is oo high, i
can lead o he o ma ion o mic oc acks in he conc e e, which can de e io a e he ma e ial
p ope ies and educe i s du abili y o e ime [24].
Ano he impo an mine al ha can in luence he du abili y o e ac o y conc e e is
silica. Silica, in he o m o silica sol, is added o he sol–gel bond. Amo phous nanopa -
icles can eac wi h alumina pa icles a ele a ed empe a u es o o m mulli e, which
p o ides excellen hea p ope ies o he cas able [
25
,
26
] and i can be ob ained by se e al
echniques [
27
–
30
]. Va ious ypes o colloidal silica sols, alumina, mulli e, spinel, e c. can
be used in he p epa a ion o sol–gel bonded e ac o y cas able. Va ious ions can be used
o s abilize hese sols [31].
In ou esea ch, a silica sol s abilized wi h Na
+
ca ions was selec ed. Al
2
O
3
mic opa -
icles we e added o he mix u e as mic o ille s. Alumina mic opa icles wi h di e en
mine alogical and chemical composi ions can be used o esea ch. The composi ion, p op-
e ies and beha io o he mic opa icles in he ma e ial a e p ima ily in luenced by hei
manu ac u ing p ocess.
The s udy deals wi h a bonding sys em ealized by he sol–gel me hod applicable in a
no-cemen cas able. The mo i a ion o ca ying ou his s udy was he e o o ex end he
li e cycle o a no-cemen cas able by inc easing he mulli e con en o he binde sys em and
he absence o knowledge o he in luence o he ine Al
2
O
3
pa icles on he g een s eng hs
and especially he in luence on he mine alogical composi ion o he ma e ial. A kind o
hyb id syn hesis o mulli e om colloidal SiO2and Al2O3mic opa icles was es ed.
The aim o he s udy was o desc ibe he mine alogical composi ion o he sol–gel
binde sys em and i s beha io du ing d ying and i ing a di e en empe a u es. Ano he
objec i e was o minimize he linea changes in he binde sys em du ing d ying by using
a mic o ille and o cha ac e ize he e ec o he i ing empe a u e and he in luence o a
mic o ille on he binde sys em, on i s in e nal s uc u e and mine alogical composi ion
in pa icula .
2. Expe imen al P ocedu e
2.1. Raw Ma e ials and Mix u es, Sample P epa a ion
Silica sol, an opalescen aqueous dispe sion o amo phous silicon dioxide nanopa -
icles, wi h designa ion K1530KD SChem (Ús ínad Labem, Czech Republic) was used
in his s udy o all es ed samples (pa icle concen a ion—30%, a e age pa icle size—
15 nm, a nega i ely cha ged su ace (anionic) s abilized by a low le el o alkaline (Na
+
),
densi y—BD = 1.203 g·cm−3
, pH = 9, iscosi y—
η
= 5 mPas). A 3.5% solu ion o NH
4
Cl
was used as a gelling agen . To achie e ma ix composi ion, which could be used a el-
e a ed empe a u es, aw ma e ials wi h high pu i y we e used in his s udy. Ma e ials
we e ob ained om hei p oduce s— abula alumina (TA) om Alma is (Ludwigsha en,
Ge many), eac i e alumina (RA) om Nabal ec (Schwando , Ge many), g ound alu-
mina (GA) om Nabal ec (Schwando , Ge many) and silica ume (SF) om RW Silicium
(Pocking, Ge many).
Tabula alumina (TA) is a comple ely sin e ed and he mally s able ma e ial. I has a
well-de eloped
α
-Al
2
O
3
c ys al s uc u e. The con en o Al
2
O
3
is highe han 99%. The
powde ed alumina oxide om he Baye p ocess is g inded in a ball mill while he alumina
powde balls a e made by a disk g anula o . The g ains a e i ed in a sha kiln a e
d ying. The sin e ing empe a u e is 1900–1950
◦
C and ully sin e ed abula co undum
s uc u e is subsequen ly ob ained [
32
]. G ound alumina (GA) is inely g ound abula
alumina. Reac i e alumina (RA) is a ully g ound calcined alumina o which a subs an ial
po ion (20 o 90%) is made o p ima y c ys als o less han 1
µ
m. I he boehmi e is i ed a
app oxima ely 1200–1250 ◦C, he calcined alumina is p oduced [32–34].
In he i s s age, he gel was p epa ed using 500 mL o sol and a gelling agen ,
3.5% NH
4
Cl, wi h a mixing ime o 60 s in a polyp opylene mold. The ans o med gel was
subsequen ly demolded and condi ioned a a labo a o y empe a u e o 21
◦
C. Then, he
Ma e ials 2023,16, 5466 4 o 14
gel was d ied o 24 h a 110
◦
C in a labo a o y d ie , and xe ogel was ob ained. Samples o
analysis we e p epa ed by c ushing he d y xe ogel. Xe ogel samples we e i ed a 600
◦
C,
800
◦
C, 1000
◦
C, 1200
◦
C, 1400
◦
C and 1600
◦
C wi h a soaking ime o 5 h a a maximum
empe a u e and a empe a u e inc ease o 4 ◦C/min.
In he second s age, ine pas es we e p epa ed om he d y aw ma e ials. Chemical
composi ion, mine alogical composi ion and pa icle size o ille s a e p esen ed in Table 1.
Table 1. Raw ma e ial p ope ies.
Chemical Composi ion TA GA RA1 RA2 SF
SiO20.09 0.80 0.16 0.12 98.46
Al2O399.55 98.35 99.45 99.53 0.04
TiO2– 0.03 – – 0.01
Fe2O30.01 0.09 – – 0.07
CaO – 0.22 0.01 0.01 0.62
MgO – 0.23 0.14 0.17 0.04
K2O – 0.03 0.04 0.08 0.55
Na2O 0.45 0.25 0.20 0.09 0.01
Phase composi ion
Qua z – – – – 0.26
Co undum 94.98 98.90 100 100 –
Andalusi e – – – – –
Diaoyudaoi e 5.02 1.10 – – –
Amo phous phase – – – – 99.74
Pa icle size (µm) 0.5–10 1.0–10 0.5–10 1.0–10 0.15–0.2
Raw ma e ials we e p ecisely weighed and mixed wi h silica sol o 15 min wi h he
designed amoun o colloidal silica as p esen ed in Table 2 o ob ain a ine pas e. The
consis ency o he pas e was he same in all cases, which was checked using a Fo d cup.
Then, he addi ion o 1.0% o a gelling agen (NH
4
Cl) was ca ied ou wi h a subsequen
mixing o 1 min. The pas es we e hen cas ed in o silicone molds wi h dimensions o
10 ×10 ×100 mm
, in which hey we e ans o med o gel. Samples we e i ed a empe -
a u es in a ange om 1000
◦
C o 1600
◦
C, wi h esea ch in en ion o use empe a u es
ealis ic in possible applica ions.
Table 2. Sol con en o ine pas es in w . %.
Raw Ma e ial Sol (w . %) Gelling Time (s)
TA 38.7 240
GA 42.2 230
RA1 41.7 180
RA2 39.0 180
SF 120.0 120
In he hi d s age, ine pas es con aining he designed amoun o all ine ille s we e
p epa ed. In he ollowing Figu e 1, he calcula ed heo e ical And easen pa icle dis ibu-
ion (APD) is shown using coe icien q = 0.23 wi h a maximal pa icle size o 1 mm. Six
mix u es we e p epa ed wi h di e en used silica con en (SF10–SF20) in o de o achie e
good e ac o y p ope ies; hus, he maximal limi o SiO
2
sou ce was limi ed o 20%.
Sample p epa a ion was simila o ha o he second s age.
Ma e ials 2023,16, 5466 5 o 14
Ma e ials 2023, 16, x FOR PEER REVIEW 5 o 14
o achie e good e ac o y p ope ies; hus, he maximal limi o SiO
2
sou ce was limi ed
o 20%. Sample p epa a ion was simila o ha o he second s age.
Figu e 1. Pa icle size dis ibu ion o p epa ed mix u es.
The inal mix u e composi ions a e shown in Table 3.
Table 3. Fine ma ix composi ion, compounds in w . %.
Componen (w . %) SF10 SF12 SF14 SF16 SF18 SF20
TA 32.6 30.6 28.6 26.6 24.5 20.0
GA 21.1 19.9 17.9 15.9 26.0 20.0
RA1 18.6 17.9 17.9 17.9 16.0 20.0
RA2 17.4 19.4 21.4 23.4 15.5 20.0
SF 10.3 12.3 14.3 16.3 18.0 20.0
2.2. Me hods
Chemical composi ion analysis o he aw ma e ials was pe o med by wa eleng h-
dispe si e X- ay spec oscopy (WDXRF) using he SPECTROSCAN MAKC-GV (Spec on
Company, S . Pe e sbu g, Russia) ins umen equipped wi h he QUANTITATIVE
ANALYSIS so wa e ( e sion 4.0, Spec on Company, S . Pe e sbu g, Russia). The samples
we e analyzed in o ms o used beads. Powde X- ay diff ac ion analysis o he aw
ma e ials and enginee ed agg ega es was conduc ed on a Panaly ical Empy ean
diff ac ome e (Panaly ical B.V., Almelo, The Ne he lands) equipped wi h a Cu anode, 1-
D posi ion-sensi i e de ec o a con en ion B agg–B en ano e lec ion geome y. The
se ing was he ollowing: s ep size—0.013° 2θ, ime pe s ep—188 s, and angula ange
5–80° 2θ. Quan i a i e phase analysis was pe o med ia he Rie eld me hod using
Panaly ical High Sco e 3 plus so wa e ( e sion 4.8, Panaly ical B.V., Almelo, The
Ne he lands). Quan i a i e analysis was pe o med using zinci e (ZnO) as an in e nal
s anda d (10 w . % pe sample).
Scanning elec on mic oscopy wi h X- ay mic oanalysis (SEM/EDS) was conduc ed
on gold-coa ed mechanically b oken specimens ( o mo phological analyses) and on
polished ca bon-coa ed hin sec ions ( o chemical mic oanalyses) using he TESCAN
MIRA 3 ins umen (Tescan O say Holding a.s., B no, Czech Republic) wi h he
accele a ing ol age o 30 kV.
Cold c ushing s eng h (CCS) acco ding o s anda d EN 993-5:2018 [35] (using
machine MEGA 11-600 D-S, B io H anice s. .o., H anice, Czech Republic) and cold
Figu e 1. Pa icle size dis ibu ion o p epa ed mix u es.
The inal mix u e composi ions a e shown in Table 3.
Table 3. Fine ma ix composi ion, compounds in w . %.
Componen (w . %) SF10 SF12 SF14 SF16 SF18 SF20
TA 32.6 30.6 28.6 26.6 24.5 20.0
GA 21.1 19.9 17.9 15.9 26.0 20.0
RA1 18.6 17.9 17.9 17.9 16.0 20.0
RA2 17.4 19.4 21.4 23.4 15.5 20.0
SF 10.3 12.3 14.3 16.3 18.0 20.0
2.2. Me hods
Chemical composi ion analysis o he aw ma e ials was pe o med by wa eleng h-
dispe si e X- ay spec oscopy (WDXRF) using he SPECTROSCAN MAKC-GV (Spec on
Company, S . Pe e sbu g, Russia) ins umen equipped wi h he QUANTITATIVE ANALY-
SIS so wa e ( e sion 4.0, Spec on Company, S . Pe e sbu g, Russia). The samples we e
analyzed in o ms o used beads. Powde X- ay di ac ion analysis o he aw ma e ials
and enginee ed agg ega es was conduc ed on a Panaly ical Empy ean di ac ome e (Pan-
aly ical B.V., Almelo, The Ne he lands) equipped wi h a Cu anode, 1-D posi ion-sensi i e
de ec o a con en ion B agg–B en ano e lec ion geome y. The se ing was he ollowing:
s ep size—0.013
◦
2
θ
, ime pe s ep—188 s, and angula ange 5–80
◦
2
θ
. Quan i a i e phase
analysis was pe o med ia he Rie eld me hod using Panaly ical High Sco e 3 plus so -
wa e ( e sion 4.8, Panaly ical B.V., Almelo, The Ne he lands). Quan i a i e analysis was
pe o med using zinci e (ZnO) as an in e nal s anda d (10 w . % pe sample).
Scanning elec on mic oscopy wi h X- ay mic oanalysis (SEM/EDS) was conduc ed on
gold-coa ed mechanically b oken specimens ( o mo phological analyses) and on polished
ca bon-coa ed hin sec ions ( o chemical mic oanalyses) using he TESCAN MIRA 3
ins umen (Tescan O say Holding a.s., B no, Czech Republic) wi h he accele a ing ol age
o 30 kV.
Cold c ushing s eng h (CCS) acco ding o s anda d EN 993-5:2018 [
35
] (using machine
MEGA 11-600 D-S, B io H anice s. .o., H anice, Czech Republic) and cold modulus o
up u e (CMOR) acco ding o s anda d EN 993-6:2018 [
36
] (using machine Tes ome ic
M350–20CT, Tes ome ic Co. L d., Rochdale, UK) we e ca ied ou a e d ying and a e
i ing. The appa en po osi y, wa e abso p ion and bulk densi y we e de e mined by a
acuum wa e abso p ion me hod wi h subsequen hyd os a ic weighing (s anda d EN
993-1:2018) [
37
]. De e mina ion o pe manen change in dimensions on hea ing was es ed
acco ding o s anda d EN 993-10:2018 [38].

Ma e ials 2023,16, 5466 6 o 14
3. Resul s and Discussion
3.1. Fi s S age
The i s s age objec i e o his s udy was o desc ibe he ans o ma ion o gel o
xe ogel and he mine alogical composi ion o xe ogel a e i ing. The p epa ed gel was
demolded a e 24 h and ai -d ied in labo a o y condi ions un il he sample weigh became
s eady. D ying o he gel p oduces xe ogel. The ini ial wa e con en o he ma e ial was
179%. Du ing d ying, he ma e ial beha es e y simila ly o clay ma e ial; Figu e 2. As
he wa e con en dec eases, he ma e ial sh inks, which s ops a a ce ain, c i ical wa e
con en , and d ying con inues wi hou signi ican sh inkage. The c i ical wa e con en
in ou case was 31%. The o al linea change in he ma e ial du ing he ans o ma ion
o he gel in o xe ogel was
−
12.8%. The blue cu e ep esen s he wa e d op du ing he
d ying p ocess. The black cu e desc ibes he linea changes in he ma e ial as a esul o
he educ ion in wa e con en . The g aph shows ha he d ying a e (wa e d op) and
he sh inkage a e (leng h d op) inc eased up o a wa e con en o 115.6%; see bo h blue
and black cu es. F om 115.6% o 31.0%, he wa e d op and leng h d op we e ela i ely
cons an , and in he las s age o d ying (be ween 31% and 0% wa e con en ) he wa e
d op was de ec ed wi hou signi ican linea change in he xe ogel.
Ma e ials 2023, 16, x FOR PEER REVIEW 6 o 14
modulus o up u e (CMOR) acco ding o s anda d EN 993-6:2018 [36] (using machine
Tes ome ic M350–20CT, Tes ome ic Co. L d., Rochdale, UK) we e ca ied ou a e
d ying and a e i ing. The appa en po osi y, wa e abso p ion and bulk densi y we e
de e mined by a acuum wa e abso p ion me hod wi h subsequen hyd os a ic weighing
(s anda d EN 993-1:2018) [37]. De e mina ion o pe manen change in dimensions on
hea ing was es ed acco ding o s anda d EN 993-10:2018 [38].
3. Resul s and Discussion
3.1. Fi s S age
The i s s age objec i e o his s udy was o desc ibe he ans o ma ion o gel o
xe ogel and he mine alogical composi ion o xe ogel a e i ing. The p epa ed gel was
demolded a e 24 h and ai -d ied in labo a o y condi ions un il he sample weigh
became s eady. D ying o he gel p oduces xe ogel. The ini ial wa e con en o he
ma e ial was 179%. Du ing d ying, he ma e ial beha es e y simila ly o clay ma e ial;
Figu e 2. As he wa e con en dec eases, he ma e ial sh inks, which s ops a a ce ain,
c i ical wa e con en , and d ying con inues wi hou signi ican sh inkage. The c i ical
wa e con en in ou case was 31%. The o al linea change in he ma e ial du ing he
ans o ma ion o he gel in o xe ogel was −12.8%. The blue cu e ep esen s he wa e
d op du ing he d ying p ocess. The black cu e desc ibes he linea changes in he
ma e ial as a esul o he educ ion in wa e con en . The g aph shows ha he d ying
a e (wa e d op) and he sh inkage a e (leng h d op) inc eased up o a wa e con en o
115.6%; see bo h blue and black cu es. F om 115.6% o 31.0%, he wa e d op and leng h
d op we e ela i ely cons an , and in he las s age o d ying (be ween 31% and 0% wa e
con en ) he wa e d op was de ec ed wi hou signi ican linea change in he xe ogel.
Figu e 2. Gel o xe ogel ans o ma ion p ocess—linea change du ing he d ying p ocess, effec o
wa e con en on linea change— ed cu e, mois u e d op as a unc ion o linea change—blue
cu e, leng h d op as a unc ion o linea change—black cu e.
Mine alogical composi ion o xe ogel a e d ying and i ing a selec ed empe a u es
is shown in Figu e 3. Fo he quan i a i e phase de e mina ion, zinci e (ZnO) was used as
he inne s anda d. The quan i a i e analysis o he phase composi ion is p esen ed in
Figu e 4.
Figu e 2.
Gel o xe ogel ans o ma ion p ocess—linea change du ing he d ying p ocess, e ec o
wa e con en on linea change— ed cu e, mois u e d op as a unc ion o linea change—blue cu e,
leng h d op as a unc ion o linea change—black cu e.
Mine alogical composi ion o xe ogel a e d ying and i ing a selec ed empe a u es
is shown in Figu e 3. Fo he quan i a i e phase de e mina ion, zinci e (ZnO) was used
as he inne s anda d. The quan i a i e analysis o he phase composi ion is p esen ed in
Figu e 4.
Xe ogel mine alogical composi ion is p edominan ly composed o a glassy phase,
which is signi ied in XRD di ac og ams by a cu ed backg ound o he 2
θ
in e al 2–30
◦
,
as shown in Figu e 3—110
◦
C, 600
◦
C and 800
◦
C. Mino c ys alline phases o hali e (NaCl)
and sal ammoniac (NH
4
Cl) we e de e mined o sample S110. Thei p esence in xe ogel
ea ed by he d ying p ocess is caused by hei c ys alliza ion om he solu ion (hali e
c ys alized due o eac ion be ween ee alkalis con ained in colloidal silica and ee chlo ide
om sal ammoniac, and sal ammoniac ec ys allized because i was used as a gelling agen ).
A e i ing a 600
◦
C, hese c ys allic p oduc s a e no p esen in he ma e ial due o hei
decomposi ion in a ange o 300–600 ◦C; Figu e 3.
Ma e ials 2023,16, 5466 7 o 14
Figu e 3.
XRD di ac og ams wi h ocus on 15–40
◦
2
θ
sec ion o xe ogel a e d ying and i ing a
di e en empe a u es wi h zoomed a ea 21.5–22.5◦2θ.
Ma e ials 2023, 16, x FOR PEER REVIEW 8 o 14
Figu e 4. Fi ing empe a u e in luence o idymi e, c is obali e and glass phase con en in ce amic
body.
The samples we e examined using SEM, and he esul s a e p esen ed in Figu e 5a–
c. The d ied sample a 110 °C is p esen ed in Figu e 5a and shows an amo phous s uc u e
wi h isible c acks caused by he d ying p ocess. Figu e 5b shows he o iginal s uc u e
o he gel, which is s ill amo phous and wi hou he c ys alline phase con en . As he
empe a u e inc eases, idymi e and c is obali e c ys allize om ha amo phous phase.
The highe he empe a u e , he mo e c is obali e is o med a he expense o he glass
phase and in he empe a u e abo e 1200 °C a he expense o idymi e. The appa en
po osi y o he ce amic body dec eases. The effec o sin e ing on he o iginal gel s uc u e
can be seen in Figu e 5c.
Figu e 5. Fi ing empe a u e in luence o inne s uc u e o he xe ogel.
3.2. Second S age
In he ollowing expe imen , six aw ma e ials we e selec ed and combined wi h
colloidal silica o c ea e a ine pas e. Figu e 6 shows linea changes a e i ing. Resul s
show ha GA, RA1, RA2 samples ha e linea dependence du ing he hea ea men wi h
inal sh inkage o 5.4–7.2% a 1600 °C on a e age. The TA sample, due o i s aw ma e ial
ab ica ion p ocess, pe o ms as expec ed, and i s maximum linea sh inkage peaks a
1600 °C wi h a alue o 2.1%. Tes samples con aining umed silica (SF) showed he highes
sh inkage. The sh inkage was al eady 11.8% when i ed a 1000 °C. The maximum
sh inkage o 22.1% occu ed a 1200 °C. As he empe a u e was inc eased abo e 1200 °C,
seconda y po osi y de eloped, leading o puffing o he ce amic body du ing i ing. A e
i ing a 1600 °C, a sh inkage o 14% was de e mined (see Figu e 6).
(a) 110 °C (b) 800 °C (c) 1600 °C
Figu e 4.
Fi ing empe a u e in luence o idymi e, c is obali e and glass phase con en in ce-
amic body.
Fu he hea ea men leads o he loss o he glassy phase and he c ys alliza ion
o idymi e and c is obali e. The o ma ion o c is obali e om xe ogel was expec ed a
empe a u es abo e 1100
◦
C as men ioned elsewhe e [
39
]. In ou condi ions, c is obali e
was iden i ied a i ing empe a u es as low as 1000
◦
C. As p esen ed in Figu e 4, he
highes c is obali e con en , 84.7%, was de e mined a 1600
◦
C, while he idymi e con en
Ma e ials 2023,16, 5466 8 o 14
dec eased om i s highes con en , 32.6%, a 1200
◦
C o 4.9%. The ollowing Figu e 4shows
c ys al phase de elopmen a he expense o he glass phase.
The samples we e examined using SEM, and he esul s a e p esen ed in Figu e 5a–c.
The d ied sample a 110
◦
C is p esen ed in Figu e 5a and shows an amo phous s uc u e
wi h isible c acks caused by he d ying p ocess. Figu e 5b shows he o iginal s uc u e
o he gel, which is s ill amo phous and wi hou he c ys alline phase con en . As he
empe a u e inc eases, idymi e and c is obali e c ys allize om ha amo phous phase.
The highe he empe a u e, he mo e c is obali e is o med a he expense o he glass
phase and in he empe a u e abo e 1200
◦
C a he expense o idymi e. The appa en
po osi y o he ce amic body dec eases. The e ec o sin e ing on he o iginal gel s uc u e
can be seen in Figu e 5c.
Ma e ials 2023, 16, x FOR PEER REVIEW 8 o 14
Figu e 4. Fi ing empe a u e in luence o idymi e, c is obali e and glass phase con en in ce amic
body.
The samples we e examined using SEM, and he esul s a e p esen ed in Figu e 5a–
c. The d ied sample a 110 °C is p esen ed in Figu e 5a and shows an amo phous s uc u e
wi h isible c acks caused by he d ying p ocess. Figu e 5b shows he o iginal s uc u e
o he gel, which is s ill amo phous and wi hou he c ys alline phase con en . As he
empe a u e inc eases, idymi e and c is obali e c ys allize om ha amo phous phase.
The highe he empe a u e , he mo e c is obali e is o med a he expense o he glass
phase and in he empe a u e abo e 1200 °C a he expense o idymi e. The appa en
po osi y o he ce amic body dec eases. The effec o sin e ing on he o iginal gel s uc u e
can be seen in Figu e 5c.
Figu e 5. Fi ing empe a u e in luence o inne s uc u e o he xe ogel.
3.2. Second S age
In he ollowing expe imen , six aw ma e ials we e selec ed and combined wi h
colloidal silica o c ea e a ine pas e. Figu e 6 shows linea changes a e i ing. Resul s
show ha GA, RA1, RA2 samples ha e linea dependence du ing he hea ea men wi h
inal sh inkage o 5.4–7.2% a 1600 °C on a e age. The TA sample, due o i s aw ma e ial
ab ica ion p ocess, pe o ms as expec ed, and i s maximum linea sh inkage peaks a
1600 °C wi h a alue o 2.1%. Tes samples con aining umed silica (SF) showed he highes
sh inkage. The sh inkage was al eady 11.8% when i ed a 1000 °C. The maximum
sh inkage o 22.1% occu ed a 1200 °C. As he empe a u e was inc eased abo e 1200 °C,
seconda y po osi y de eloped, leading o puffing o he ce amic body du ing i ing. A e
i ing a 1600 °C, a sh inkage o 14% was de e mined (see Figu e 6).
(a) 110 °C (b) 800 °C (c) 1600 °C
Figu e 5. Fi ing empe a u e in luence o inne s uc u e o he xe ogel.
3.2. Second S age
In he ollowing expe imen , six aw ma e ials we e selec ed and combined wi h
colloidal silica o c ea e a ine pas e. Figu e 6shows linea changes a e i ing. Resul s
show ha GA, RA1, RA2 samples ha e linea dependence du ing he hea ea men
wi h inal sh inkage o 5.4–7.2% a 1600
◦
C on a e age. The TA sample, due o i s aw
ma e ial ab ica ion p ocess, pe o ms as expec ed, and i s maximum linea sh inkage
peaks a 1600
◦
C wi h a alue o 2.1%. Tes samples con aining umed silica (SF) showed he
highes sh inkage. The sh inkage was al eady 11.8% when i ed a 1000
◦
C. The maximum
sh inkage o 22.1% occu ed a 1200
◦
C. As he empe a u e was inc eased abo e 1200
◦
C,
seconda y po osi y de eloped, leading o pu ing o he ce amic body du ing i ing. A e
i ing a 1600 ◦C, a sh inkage o 14% was de e mined (see Figu e 6).
Ma e ials 2023, 16, x FOR PEER REVIEW 9 o 14
Figu e 6. Fi ing empe a u e in luence o linea change in ce amic body.
Figu e 6 demons a es he e olu ion o he mine alogical composi ion o RA1 xe ogel
a e i ing a empe a u es o up o 1600 °C. A 1000 °C, he ma e ial con ains only an α
co undum and a glassy phase. A e i ing a 1100 °C, c is obali e appea s in he
mic os uc u e and c ys allizes om he glass phase. When i ed abo e 1400 °C, mulli e is
o med, which is consis en wi h [40]. Acco ding o B aga, when he p ecu so has a high
deg ee o homogenei y, he empe a u e a which mulli e o ma ion begins is low.
Howe e , when he e is he e ogenei y, he mulli e o ma ion empe a u e is conside ably
inc eased, eaching empe a u es abo e 1400 °C. The SiO
2
nanopa icles, which o igina e
om he silica sol, eac wi h he Al
2
O
3
mic opa icles o o m mulli e. The amoun o
mulli e o med depends on he SiO
2
con en and on he eac i i y o he Al
2
O
3
pa icles.
The SiO
2
con en was de e mined by he amoun o sol used in he mix u e, which was he
same o all ecipes. Figu e 7 shows ha he eac i i y o he used alumina ille a ies.
Tabula alumina is he sou ce o he leas eac i e Al
2
O
3
, while RA1 was ound o be he
mos eac i e. The mulli e con en a e i ing a 1600 °C o RA1 mix u e was 40%, while
o TA i was less han 28%.
Figu e 7. E olu ion o mine alogical composi ion in RA1 ma e ial and he effec o ille on he
mulli e con en o ce amic body a e i ing.
3.3. Thi d S age
I was con i med, in he second s age, ha a empe a u e highe han 1400 °C is
equi ed o c ys alliza ion o mulli e om he mel ; Figu e 7. Fo his eason, i ing
empe a u es o 1200 °C, 1400 °C and 1600 °C we e used o s udy he mine alogy in he
inal pa o he s udy. As he ollowing Figu e 8 demons a es, i was con i med ha
mulli e is o med by he eac ion o SiO
2
wi h Al
2
O
3
only a i ing empe a u es abo e 1400
°C. The sou ce o silica is a sol and used silica. Sou ces o Al
2
O
3
a e inely g ound abula
Figu e 6. Fi ing empe a u e in luence o linea change in ce amic body.
Figu e 6demons a es he e olu ion o he mine alogical composi ion o RA1 xe o-
gel a e i ing a empe a u es o up o 1600
◦
C. A 1000
◦
C, he ma e ial con ains only
Ma e ials 2023,16, 5466 9 o 14
an
α
co undum and a glassy phase. A e i ing a 1100
◦
C, c is obali e appea s in he
mic os uc u e and c ys allizes om he glass phase. When i ed abo e 1400
◦
C, mulli e
is o med, which is consis en wi h [
40
]. Acco ding o B aga, when he p ecu so has a
high deg ee o homogenei y, he empe a u e a which mulli e o ma ion begins is low.
Howe e , when he e is he e ogenei y, he mulli e o ma ion empe a u e is conside ably
inc eased, eaching empe a u es abo e 1400
◦
C. The SiO
2
nanopa icles, which o igina e
om he silica sol, eac wi h he Al
2
O
3
mic opa icles o o m mulli e. The amoun o
mulli e o med depends on he SiO
2
con en and on he eac i i y o he Al
2
O
3
pa icles.
The SiO
2
con en was de e mined by he amoun o sol used in he mix u e, which was
he same o all ecipes. Figu e 7shows ha he eac i i y o he used alumina ille a ies.
Tabula alumina is he sou ce o he leas eac i e Al
2
O
3
, while RA1 was ound o be he
mos eac i e. The mulli e con en a e i ing a 1600
◦
C o RA1 mix u e was 40%, while
o TA i was less han 28%.
Ma e ials 2023, 16, x FOR PEER REVIEW 9 o 14
Figu e 6. Fi ing empe a u e in luence o linea change in ce amic body.
Figu e 6 demons a es he e olu ion o he mine alogical composi ion o RA1 xe ogel
a e i ing a empe a u es o up o 1600 °C. A 1000 °C, he ma e ial con ains only an α
co undum and a glassy phase. A e i ing a 1100 °C, c is obali e appea s in he
mic os uc u e and c ys allizes om he glass phase. When i ed abo e 1400 °C, mulli e is
o med, which is consis en wi h [40]. Acco ding o B aga, when he p ecu so has a high
deg ee o homogenei y, he empe a u e a which mulli e o ma ion begins is low.
Howe e , when he e is he e ogenei y, he mulli e o ma ion empe a u e is conside ably
inc eased, eaching empe a u es abo e 1400 °C. The SiO
2
nanopa icles, which o igina e
om he silica sol, eac wi h he Al
2
O
3
mic opa icles o o m mulli e. The amoun o
mulli e o med depends on he SiO
2
con en and on he eac i i y o he Al
2
O
3
pa icles.
The SiO
2
con en was de e mined by he amoun o sol used in he mix u e, which was he
same o all ecipes. Figu e 7 shows ha he eac i i y o he used alumina ille a ies.
Tabula alumina is he sou ce o he leas eac i e Al
2
O
3
, while RA1 was ound o be he
mos eac i e. The mulli e con en a e i ing a 1600 °C o RA1 mix u e was 40%, while
o TA i was less han 28%.
Figu e 7. E olu ion o mine alogical composi ion in RA1 ma e ial and he effec o ille on he
mulli e con en o ce amic body a e i ing.
3.3. Thi d S age
I was con i med, in he second s age, ha a empe a u e highe han 1400 °C is
equi ed o c ys alliza ion o mulli e om he mel ; Figu e 7. Fo his eason, i ing
empe a u es o 1200 °C, 1400 °C and 1600 °C we e used o s udy he mine alogy in he
inal pa o he s udy. As he ollowing Figu e 8 demons a es, i was con i med ha
mulli e is o med by he eac ion o SiO
2
wi h Al
2
O
3
only a i ing empe a u es abo e 1400
°C. The sou ce o silica is a sol and used silica. Sou ces o Al
2
O
3
a e inely g ound abula
Figu e 7.
E olu ion o mine alogical composi ion in RA1 ma e ial and he e ec o ille on he mulli e
con en o ce amic body a e i ing.
3.3. Thi d S age
I was con i med, in he second s age, ha a empe a u e highe han 1400
◦
C is
equi ed o c ys alliza ion o mulli e om he mel ; Figu e 7. Fo his eason, i ing
empe a u es o 1200
◦
C, 1400
◦
C and 1600
◦
C we e used o s udy he mine alogy in he
inal pa o he s udy. As he ollowing Figu e 8demons a es, i was con i med ha
mulli e is o med by he eac ion o SiO
2
wi h Al
2
O
3
only a i ing empe a u es abo e
1400
◦
C. The sou ce o silica is a sol and used silica. Sou ces o Al
2
O
3
a e inely g ound
abula alumina, eac i e alumina and wo ypes o g ound alumina.
Ma e ials 2023, 16, x FOR PEER REVIEW 10 o 14
alumina, eac i e alumina and wo ypes o g ound alumina.
Figu e 8. The effec o i ing empe a u e on he mulli e, co undum and c is obali e con en o
ce amic body.
The las pa o he esea ch was ocused o he s udy o he effec o SF on he
muli iza ion o he ma e ial.
In he second s age, i was con i med ha TA, as a sou ce o Al
2
O
3
, is he leas eac i e
one o all he aw ma e ials used; see Figu e 9a. The e o e, his aw ma e ial was
p og essi ely eplaced by SF. SF was used as he SiO
2
sou ce. The composi ion o he
mix u es is shown in Table 3. Mulli e was p o ed in he ma e ial when i ed a 1600 °C,
while i s p esence was excluded when i ed a 1400 °C. This co esponds o he esul s in
he second s age.
The p esence o amo phous SF in he mix u e was con i med o ha e a posi i e effec
on mulli e o ma ion. The maximum mulli e con en (78%), maximum mulli iza ion, was
achie ed a a 20% SF con en in he mix u e. Acco ding o he pa e n o dependence o
he mulli e con en on he SF con en , Figu e 9b, i can be assumed ha u he inc ease in
he mulli e con en in he ce amic body canno be achie ed by inc easing he dose o SF in
he mix u e.
Figu e 9. E olu ion o mine alogical composi ion in RA1 ma e ial and he effec o ma ix on he
mulli e con en o ce amic body a e i ing. (a) Tabula alumina, (b) Silica ume.
A u he inc ease in mulli e con en in he ce amic body can only be hypo he ically
achie ed by p olonging he iso he mal soaking ime a 1600 °C o by inc easing he i ing
empe a u e.
(a) (b)
Figu e 8.
The e ec o i ing empe a u e on he mulli e, co undum and c is obali e con en o
ce amic body.