Ci a ion: Misák, P.; Kocáb, D.; Baye ,
P.; Vymazal, T.; Ro naníko á, P.
E ec o De-Icing Chemicals on
Conc e e Scaling: The Role o S o age
Wa e . Ma e ials 2023,16, 4928.
h ps://doi.o g/10.3390/
ma16144928
Academic Edi o : Qing- eng Liu
Recei ed: 20 June 2023
Re ised: 4 July 2023
Accep ed: 5 July 2023
Published: 10 July 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
E ec o De-Icing Chemicals on Conc e e Scaling: The Role
o S o age Wa e
Pe Misák * , Dalibo Kocáb , Pa ik Baye , Tomáš Vymazal and Pa la Ro naníko á
Facul y o Ci il Enginee ing, B no Uni e si y o Technology, Ve eˇ í 331/95, 602 00 B no, Czech Republic;
dalibo .kocab@ u b .cz (D.K.); pa ik.baye @ u b .cz (P.B.); omas. ymazal@ u b .cz (T.V.);
[email p o ec ed].cz (P.R.)
*Co espondence: pe .misak@ u b .cz; Tel.: +420-541-147-831
Abs ac :
This pape deals wi h he e ec o he cha ac e o he wa e used o he wa e s o age o
conc e e es specimens on he esul s o es s o esis ance o de-icing chemicals. Two expe imen s
we e conduc ed o in es iga e he e ec o he con en o ee CO
2
in wa e and leaching o calcium
hyd oxide om conc e e on he es esul s. In he i s expe imen , he esis ance o mo a s o
wa e and de-icing chemicals was in es iga ed. I was ound ha he cha ac e o he wa e s o age,
i.e., esh wa e s. p e iously used wa e , can signi ican ly a ec he es esul s. The second
expe imen ocused on in es iga ing he e ec o he con en o ee CO
2
in wa e on he es esul s.
I was ound ha he con en o ee CO
2
in he wa e can s a is ically signi ican ly in luence he
es esul s. In conclusion, he pape shows ha he cha ac e o he wa e used o wa e s o age o
conc e e es specimens and he con en o ee CO
2
in wa e a e essen ial ac o s ha can signi ican ly
a ec he esul s o conc e e esis ance es s o de-icing chemicals. Fu he esea ch is needed o
unde s and hese in luences and hei po en ial use o imp o e he esis ance o conc e e.
Keywo ds:
conc e e; scaling; de-icing chemicals; wa e s o age; s a is ical analysis; su ace laye es
1. In oduc ion
Conc e e is he mos widely used ma e ial in he cons uc ion indus y. Indeed, i is he
second mos used ma e ial on Ea h a e wa e [
1
]. Howe e , he p oduc ion o conc e e is
associa ed wi h sus ainabili y issues since i is made p ima ily om cemen , agg ega e, and
wa e , and each componen di ec ly impac s he en i onmen [
2
]. Global cemen p oduc ion
in 2021 was app oxima ely 4.1 billion onnes [
3
], while annual conc e e consump ion is
abou se en imes highe [
4
]. The p oduc ion o Po land cemen , un il ecen ly he p ima y
componen o conc e e, is esponsible o abou 7–8% o global CO
2
emissions [
5
]. This
signi ican ca bon oo p in leads o g owing conce ns abou he sus ainabili y o conc e e
s uc u es and he deple ion o na u al esou ces such as agg ega e o
wa e [6–8]
. Sus-
ainabili y assessmen ools include h ee undamen al pilla s o sus ainabili y—economic,
ecological, and socio-cul u al [
9
]—and designing and building du able s uc u es is closely
ela ed o a leas he i s wo pilla s.
Du abili y is a e y gene al e m, as i is no de ined in e ms o ime, o example,
compa ed o he se ice li e o s uc u es. When assessing he du abili y o a ma e ial, i
is necessa y o p ecisely de ine he en i onmen in which i is o be e alua ed [
10
]. The
du abili y o conc e e can be cha ac e ised by i s abili y o esis he ac ion o ex e nal and
in e nal agen s, wi h he mode o ailu e o he conc e e being highly dependen on he
ype and in ensi y o he ac ion. The same conc e e may exhibi a ce ain le el o esis ance
o he ac ion o so (low mine alised) wa e s ( his is co osion caused by he leaching
o calcium hyd oxide; see, e.g., [
11
]) ye a he same ime, a comple ely di e en le el o
esis ance o he ac ion o sulpha es on conc e e ( his in e nal s uc u al damage is caused
by he inc ease in he olume o p oduc s o he co osion eac ion; see, e.g., [
12
]). One o
Ma e ials 2023,16, 4928. h ps://doi.o g/10.3390/ma16144928 h ps://www.mdpi.com/jou nal/ma e ials
Ma e ials 2023,16, 4928 2 o 15
he basic du abili y p ope ies o conc e e in mild clima es is undoub edly i s esis ance o
os exposu e [
13
]. This e e s o he abili y o conc e e o esis he epea ed al e na ion o
posi i e and nega i e empe a u es unde simul aneous exposu e o wa e , high humidi y,
o wa e in combina ion wi h de-icing chemicals. The al e na ion o posi i e and nega i e
empe a u es has no signi ican ad e se e ec on d y conc e e [14].
I a conc e e s uc u e o a pa o i is exposed o wa e and al e na ing posi i e
and nega i e empe a u es, i is necessa y o assess whe he he conc e e su ace is also
a ec ed by any de-icing chemicals—mos o en by sal and he esul ing NaCl solu ion,
which may a ec he conc e e su ace di ec ly o in he o m o sal mis o sp ay. When
a de-icing chemical is applied o conc e e, he deg ada ion p ocesses comple ely di e
om hose in conc e e only exposed o wa e , eezing, and hawing. Wi hou he ac ion o
de-icing chemicals, wa e will en e he po e s uc u e o he conc e e, due o abso p ion,
and will eeze he e a sub-ze o empe a u es. Consequen ly, he e is a isk o mic o-c acks
and hen c acks in he in e nal s uc u e o he conc e e, i.e., he conc e e may become
damaged. On he o he hand, scaling due o sal s is de ined as su ace damage. This is
caused by os and he esul ing c ys allisa ion o he sal solu ion in he su ace laye o he
conc e e [
15
,
16
]. The conc e e scaling mechanism is qui e di e en om ha o eeze– haw,
which causes wa e o c ys allise in he in e nal s uc u e o conc e e, he eby causing i o
ail. In he case o scaling, i is a su ace ac ion only. While his does no a ec he quali y
o he conc e e inside he s uc u e, i makes i sensi i e o he pene a ion o wa e and
o he agg essi e agen s in o i s s uc u e. This educes he o e all esis ance o he conc e e
and, consequen ly, he se ice li e o he s uc u e, as well as he isk o ein o cemen
depassi a ion. Scaling was ex ensi ely s udied and desc ibed many yea s ago, bu is s ill
emains a subjec o many pape s oday; see, e.g., [
17
–
22
]. The mos undamen al indings
in he s udy o conc e e scaling a e:
•
Damage o he su ace laye o conc e e is signi ican ly wo se when he wa e applied
o he conc e e su ace con ains a ce ain amoun o solu e [23].
•
The e ec i eness o he concen a ion is almos independen o he cha ac e o he
de-icing chemical in he solu ion (e.g., sal s, u ea). All hese de-icing chemicals ha e
been ound o ha e simila e ec s on conc e e [24].
• Damage o he su ace laye o conc e e is ini ially e lec ed in he o ma ion o small
chips o lakes o ma e ial ha g adually all o he su ace [25].
•
Wi hou a ee liquid wi h de-icing chemicals on he su ace o he conc e e, ac ual
scaling will no occu , e en when he conc e e is sa u a ed wi h wa e and subjec ed
o al e na ing eeze– haw cycles [26].
•
The concen a ion o sal s in he liquid applied o he conc e e su ace is mo e impo -
an han he concen a ion o he liquid in he po es o he in e nal s uc u e o he
conc e e [23].
•
The esis ance o conc e e o scaling is gene ally ela i ely low, bu can be inc eased
signi ican ly by adding an ai -en aining mix u e o he conc e e [24,27–29].
Tes ing he esis ance o a conc e e su ace o scaling is qui e di icul i sel . Se e al
s anda ds and guidelines, bo h in Eu ope and No h Ame ica, de ine he es ing p ocedu e
di e en ly (see Sec ion 2). Many ac o s, some o which a e lis ed abo e, will in luence
he ou come o a scaling es . A he same ime, hese es s a e cha ac e ised by ela i ely
high a iabili y in he esul ing alues, including epea abili y and ep oducibili y [
30
].
The es s deal wi h he leng h and empe a u e o he eeze- haw cycles, he es specimen
su ace a ea, he cha ac e is ics o he de-icing chemicals, e c. S ill, he e ec o he wa e
s o age in which he es specimens a e placed be o e he es has ye o be conside ed. I is
easonable ha when es ing a cu h ough a es specimen (e.g., he p ocedu e acco ding o
s anda d [
31
]), he wa e in he wa e s o age does no in luence he es esul . Howe e , i
is also easonable ha he su ace should be es ed a he han he cu h ough he specimen,
as his is signi ican ly mo e ep esen a i e o he ac ual beha iou o he conc e e in he
s uc u e. In his case, he wa e in which he specimens a e s o ed be o e he es may ha e
an in luence.
Ma e ials 2023,16, 4928 3 o 15
This pape deals wi h his o e looked e ec and p esen s he esul s o wo consecu i e
expe imen s. The pilo expe imen showed ha he le el o wa e sa u a ion wi h calcium
hyd oxide in he wa e s o age a ec s he esul s o conc e e scaling es s. The second, la ge
expe imen u he e ined he esul s o he pilo expe imen and mo e p ecisely de ined
he condi ions unde which conc e e scaling es esul s can be signi ican ly a ec ed.
2. Tes ing Me hods
The basic concep o all me hods ha e alua e he esis ance o conc e e o he ac ion o
de-icing chemicals is simila . The es specimens o which he chemical solu ion is applied
a e placed in a de ice which allows au oma ic cycling be ween posi i e and nega i e
empe a u es. The main pa ame e s and p ocedu es o he a ious es s by which he
esis ance o conc e e o scaling can be e alua ed a e b ie ly desc ibed below.
2.1. P ocedu e Acco ding o SS 13 72 44 [32]
This es , de eloped in Sweden, is called he “Bo ås es me hod” o he “slab
es ” [
19
] and is one o he basic me hods. The es equi es a minimum o ou speci-
mens wi h a es ed a ea g ea e han 0.0225 m
2
. The ecommended size o he specimens is
150 mm ×150 mm ×50 mm
. The es is conduc ed one-dimensionally, he change in em-
pe a u e and humidi y a ec s only he su ace o be es ed, and he o he specimen su aces
a e insula ed. No mo e han 15 min be o e he es , a 3% NaCl solu ion is applied o he
es su ace up o a heigh o 3 mm. The es cycles be ween empe a u es o
−
18
°
C and
+20
°
C, wi h a cycle ime o
24 h
. The es e alua ion is usually conduc ed a e 7, 14, 21, 28,
42, and 56 cycles. The esul is he cumula i e amoun o ma e ial scaled o he specimen
in kg/m2[32].
2.2. P ocedu e Acco ding o CEN/TS 12390-9 [31]
This es me hod is based on he s anda d [
32
] and di e s only in mino de ails— he
es ing is conduc ed on he cu ing su ace o he es specimen, cycling be ween empe a-
u es o
−
20
°
C and +20
°
C. This s anda d [
31
] is also he basis o he s anda ds used in
Li huania, F ance, and he USA.
2.3. P ocedu e Acco ding o RILEM TC 117-FDC/CDF [33]
This es me hod is used in Ge many. I is necessa y o p oduce i e o mo e specimens
wi h a es a ea o mo e han 0.08 m
2
— o example, es cubes (
150 mm ×150 mm ×150 mm
).
The mould o he cube p oduc ion is co e ed wi h Te lon oil in he middle, and no
demoulding agen may be used du ing he p oduc ion. The su ace o be es ed is adjacen
o he Te lon su ace. The specimens a e placed on pads in an au oma ic de ice in o which a
3% NaCl solu ion is pou ed o subme ge 5 mm o he cubes. The minimum es empe a u e
is
−
20
°
C and he maximum is +20
°
C. One cycle akes 12 h, and he esul is he cumula i e
amoun o ma e ial ha has scaled o om he specimen in kg/m
2
a e 14 and 28 cycles
(4 and 6 cycles a e also ecommended [33].
2.4. ASTM C672/C672M-12 P ocedu e [34]
Acco ding o he Ame ican S anda d [
34
], a leas wo specimens wi h a minimum es
a ea o 0.045 m
2
and a heigh g ea e han 75 mm mus be es ed. A 4% CaCl
2
solu ion is
pou ed on he es specimen up o a heigh o 6 mm (3% NaCl can also be used). The es
specimens a e es ed by al e na ing empe a u es be ween
−
18
°
C and +23
°
C; he du a ion
o one cycle is 24 h and a isual inspec ion o he su ace o be es ed is conduc ed a e
e e y i h cycle. The inal e alua ion is pe o med a e 50 cycles [34].
2.5. P ocedu e Acco ding o MTO LS-412 and BNQ NQ 2621-900 [35]
The MTO LS-412 es p ocedu e is used in On a io, Canada. The es is pe o med on
a leas wo specimens wi h dimensions 300 mm
×
300 mm
×
75 mm. A 3% NaCl solu ion
is applied o he es su ace up o a heigh o 6 mm, and he specimen is hen placed in a
Ma e ials 2023,16, 4928 4 o 15
de ice in which i is cyclically ozen a
−
18
°
C and hawed a +25
°
C. One cycle las s 24 h,
and he condi ion o he es a ea is isually assessed a e e e y i h cycle [35]. The BNQ
es p ocedu e is used in he p o ince o Quebec in Canada. The minimum numbe o
specimens o pe o m he es is wo, wi h a minimum es ed a ea o 0.05 m
2
. A 3% NaCl
solu ion is pou ed on o he es specimens, which mus hen be co e ed wi h a wa e p oo
memb ane o p e en e apo a ion. The es specimens a e exposed o cyclic empe a u e
changes om
−
18
°
C o +25
°
C in an au oma ic es chambe o 24 h pe cycle. A e 7,
21, 35, and 56 cycles, he amoun o he ma e ial ha has scaled o om he su ace o he
es ed a ea is de e mined, and he es esul is he cumula i e amoun o he scaled o
ma e ial om he specimen in kg/m2a e 56 cycles [35].
2.6. P ocedu e Acco ding o ˇ
CSN 73 1326 [36]
This es me hod is commonly used in he Czech Republic and a sligh ly modi ied
e sion in Slo akia. The s anda d [
36
] lis s h ee p ocedu es acco ding o which he es
can be pe o med. Only me hod A (DiCh-A me hod) will be desc ibed in de ail he e, as i
was used o he expe imen desc ibed below.
The es is conduc ed on h ee es cubes (150 mm
×
150 mm
×
150 mm). A e emo al
om he moulds, he specimens a e placed in wa e s o age a 20
°
C. When he conc e e is
28 days old, an au oma ic cycling es is s a ed, whe e he uppe su ace o he specimen is
es ed wi hou any modi ica ion. The indi idual es specimens a e placed in es con aine s
made o a non-co osi e ma e ial, which allows he es specimens o be imme sed in a 3%
NaCl solu ion. The ma e ial ha scales o om he specimen is cap u ed. The solu ion
is pou ed in o he con aine so ha he es cubes a e imme sed o 5 mm a ound hei
ci cum e ence. One eeze– haw cycle akes 2 h, he minimum empe a u e is
−
15
°
C,
and he maximum is +20
°
C. This is a ela i ely quick es , a leas compa ed o o he
s anda ds. A e e e y 25 h cycle, he amoun o ma e ial ha has scaled o om he
specimen su ace in he d ied s a e is de e mined. The su ace esis ance o cemen conc e e
o wa e and de-icing chemicals is gi en by he amoun o he scaled o ma e ial om he
specimen in g/m
2
a e a ce ain numbe o cycles. Mo e in o ma ion on his me hod can
be ound, o example, in pape s [37].
3. Expe imen al
Two sub-expe imen s we e conduc ed as pa o his s udy. Th ee labo a o ies pa ici-
pa ed in he i s expe imen , which can be called a pilo . The mo i a ion o his expe imen
was he epea ed ailu e o one o he labo a o ies o pa icipa e in in e -labo a o y com-
pa ison p o iciency es ing p og ammes. The labo a o y ied o de e mine he cause o
long- e m highe alues o he amoun s o scaled o ma e ial om specimens wi hin he
DiCh-A es ing me hod. I was assumed ha he eason migh be he wa e whe e he
conc e e specimens we e cu ed be o e he ac ual es ing. This was because he labo a o y
cleaned he wa e s o age anks signi ican ly mo e equen ly han is common p ac ice.
An expe imen was, he e o e, conduc ed whe e one se o es specimens was cu ed in
wa e whe e o he specimens had been p e iously s o ed. A second se o es speci-
mens was placed in esh wa e . The aim was o de e mine whe he he wa e chemis y,
i.e., eshwa e s. p e iously used wa e , could signi ican ly in luence he esul s o he
es s pe o med by he DiCh-A me hod. As was subsequen ly disco e ed, an impo an
ac o ha signi ican ly in luenced he es esul s was ha each labo a o y used wa e om
hei local sou ces.
To in es iga e his e ec , a u he la ge-scale expe imen was designed o es whe he
he con en o ee CO
2
in he wa e s o age could s a is ically signi ican ly a ec he es
esul s. In his case, a mo e ex ensi e se o 150 mm conc e e es cubes had al eady
been p epa ed.
Ma e ials 2023,16, 4928 5 o 15
3.1. Pilo Expe imen —Mo a Tes ing
The pilo expe imen in ol ed 18 es specimens wi h dimensions
40 ×40 ×160 mm
p oduced om mo a wi h EN 196-1 s anda d silica sand [
38
] and CEM I 42.5 R cemen .
Each labo a o y ecei ed six es specimens. Th ee es specimens we e always placed in
wa e whe e o he conc e e es specimens had p e iously been cu ed. I was, he e o e,
wa e which was a sa u a ed solu ion o calcium hyd oxide (used wa e ). The o he h ee
specimens we e placed in esh wa e om hei local sou ces ( esh wa e ).
The esul s we e s a is ically p ocessed a e pe o ming 100 eeze– haw cycles ac-
co ding o he DiCh-A me hod and a e de e mining he amoun o scaled o ma e ial
om he specimen a e e e y 25 cycles. Fi s , he equali y o he means o he esul s o he
es s pe o med by he h ee labo a o ies was es ed a he 0.05 signi icance le el. The one-
ac o analysis o a iance (ANOVA) was used. The es s we e pe o med sepa a ely o
each le el o eeze– haw cycles and he wa e used in he wa e s o age. Based on he
analyses pe o med, he di e ences be ween he es esul s a e s a is ically signi ican a
each expe imen le el. The e o e, he esul s o he es s conduc ed by he h ee labo a o ies
a e s a is ically signi ican ly di e en . This is also illus a ed in a g aphical ep esen a ion
(see Figu e 1).
Figu e 1. Mo a scaling esis ance es esul s.
The es s pe o med by he A-lab show signi ican ly highe alues o he scaled o
ma e ial in g/m
2
a all expe imen le els, i.e., a 25 eeze– haw cycle in e als. Howe e ,
he e we e mo e in e es ing obse a ions om he poin o iew o he objec i e o he
expe imen pe o med. In he design o he expe imen , i was assumed ha using esh

Ma e ials 2023,16, 4928 6 o 15
wa e o s o e he es specimens would inc ease he esul s o he es s pe o med acco ding
o he DiCh-A me hod. The esul s o labo a o ies B and C showed a sligh inc ease in he
amoun o scaled o ma e ial om he specimens. Howe e , in he case o labo a o y A,
he e ec o he cha ac e o he wa e s o age is exac ly he opposi e. This e ec is mos
e iden a e 100 eeze– haw cycles. The a e age alues ob ained di e by mo e han
500 g/m
2
, co esponding o an almos 50% inc ease in he scaled o ma e ial om he
specimen su ace. By analysing he es esul s, he au ho s concluded ha he opposi e
e ec in he di e en labo a o ies could be caused by he ac ha each labo a o y used
wa e om hei local sou ce o s o e he es specimens. I was disco e ed ha he DiCh-A
me hod’s es esul s could be mainly in luenced by he con en o ee CO
2
in he wa e
used o wa e s o age. As will be shown la e in he pape , also based on he esul s
o mic oscopic analysis, a e y hin CaCO
3
laye o ms on he es specimens a highe
ee CO
2
con en in he wa e , which leads o a ce ain ha dening o hei su ace. This
ha dening does no a ec he s eng h cha ac e is ics o he conc e e bu may a ec he
esul s o conc e e su ace laye es s, such as he amoun o scaled o ma e ial in he
specimen acco ding o he DiCh-A me hod.
3.2. The Second Expe imen —Tes ing Conc e e
The signi ican di e ences om he pilo expe imen we e as ollows. The expe imen
was conduc ed in a p o iciency es ing p og amme in ol ing 23 mos ly acc edi ed es ing
labo a o ies. By pa icipa ing in simila p ecision expe imen s, he labo a o ies ha e
demons a ed hei p o iciency in he es ing p ocedu es unde conside a ion. The o ganise
p o ided nine een labo a o ies wi h h ee cube-shaped es specimens wi h an edge o
150 mm
. The emaining ou labo a o ies ecei ed nine o hose es specimens. All 105 es
specimens we e p oduced om one ba ch, a one ime, in one place, and using he same
mould ype. F esh conc e e o class C 30/37 XF4 S4 D
max
16 was made a he conc e e plan
ina1m3mixe . The composi ion o he esh conc e e is gi en in Table 1.
Table 1. The composi ion o esh conc e e.
Componen kg/m3
Cemen CEM I 42.5 R (Mo ká cemen plan , CZ) 400
Fine agg ega e 0–4 mm (B a ˇcice, CZ) 807
Coa se agg ega e 8–16 mm (Olb amo ice, CZ) 915
Ai -en aining admix u e LPS A 94 (Sika, CZ) 1
Plas icizing admix u e SVC-4035 (Sika, CZ) 2.2
Wa e 182
A e 24 h om he conc e e mixing, he es specimens we e de-moulded and placed
in di e en wa e s o age anks. Six y-nine specimens we e placed in an emp y and clean
ank ha was illed wi h esh wa e om he wa e supply (Figu e 2a). These specimens
we e used o he main pa o he in e -labo a o y compa ison (he eina e e e ed o
as “ esh wa e ”). The o he 12 es specimens we e placed in a ank wi h p e iously
used wa e o conc e e specimens no in ended o his expe imen . This wa e could be
conside ed a sa u a ed solu ion o calcium hyd oxide (Figu e 2b), and his s o age is e e ed
o as “p e iously used wa e ”. The emaining 12 specimens we e placed in a ank wi h a
cons an in low o wa e om he wa e supply (Figu e 2c), and his s o age is e e ed o
as “slow lowing wa e ”. The wa e low a e was adjus ed so ha sa u a ion wi h conc e e
leacha es could no occu h oughou he ma u a ion p ocess, i.e., o he emaining 27 days.
The aim was o maximise he leaching o calcium hyd oxide om he es specimens.
The di e ences in wa e p ope ies in he wa e s o age we e also moni o ed by measu ing
he pH alue: slow lowing wa e , 6.67; esh wa e , 11.44; p e iously used wa e , 13.33.
The pH measu emen s we e aken wi h an au oma ic empe a u e-compensa ed p obe a
14 days
and hen 21 days a e p oduc ion. The alues a e om he second measu emen . A
Ma e ials 2023,16, 4928 7 o 15
28 days
a e p oduc ion, all es specimens we e w apped immedia ely a e emo al om
wa e s o age in impe meable oil and dis ibu ed o he pa icipan s o he expe imen .
Figu e 2.
A angemen o es specimens in (
a
) esh wa e , (
b
) p e iously used wa e , and (
c
) slow
lowing wa e .
4. Resul s and Discussion
4.1. Valida ion o Tes Resul s
The i s s ep in he analysis o he main pa o he expe imen was o e alua e he
s a is ical pe o mance o he ou labo a o ies ha pa icipa ed in he main pa o he
expe imen . This was o assess whe he he esul s o he labo a o ies we e s a is ically
signi ican ly di e en om he esul s o he es s pe o med by o he labo a o ies in
he la ge-scale in e -labo a o y compa ison (ILC). The esul s o he es s pe o med by
23 labo a o ies
acco ding o ˇ
CSN 73 1326 [
36
] on specimens unde he same condi ions we e
assessed. These specimens we e cu ed in a wa e s o age en i onmen called
“ esh wa e ”.
Be o e he ac ual assessmen o s a is ical pe o mance using he so-called z-sco e, i
was e alua ed whe he any o he es esul s could be conside ed ou lie s. Fo his pu pose,
he Coch an and G ubbs es s we e pe o med a a signi icance le el o 0.01 [
39
]. Based on
exceeding he c i ical alues o he es s, one labo a o y was excluded om he expe imen
due o ou lie esul s a all expe imen le els. Due o he la ge scale o he ILC e alua ion,
only he ou pu s a e 100 eeze– haw cycles a e p esen ed below.
The mean alues and sample s anda d de ia ions o he es esul s o all labo a o ies
a e shown in Figu e 3. As can be obse ed om he g aph, he pe o mance o one o he
Ma e ials 2023,16, 4928 8 o 15
labo a o ies was classi ied as “unsa is ac o y” and ha o one labo a o y as “p oblema ic”
based on he z-sco e [
40
]. The assigned alue, indica ed by he blue line in Figu e 3, was
e alua ed by i e a i e algo i hm A [41]. This algo i hm has he ad an age o being obus
and, hus, less sensi i e o ou lie s han he o dina y a i hme ic mean. Th eshold alues o
|z-sco e|>2 and 3 a e shown in Figu e 3in ed.
Figu e 3. E alua ion o he esul s o he in e -labo a o y compa ison.
The ILC e alua ion also de e mined a epea abili y alue o
=
50.1% and ep o-
ducibili y o
R=
202.6% acco ding o [
39
]. These alues indica e how much a iabili y
can be expec ed be ween he esul s o es s pe o med by one labo a o y ( epea abili y)
and mul iple labo a o ies ( ep oducibili y). Acco ding o [
36
], he es p ocedu es show
signi ican ly highe
and
R
alues han o he commonly used conc e e es ing p ocedu es.
This is also e iden om he esul s o o he ILCs, which can be ound, e.g., a [42].
Howe e , he mos impo an ou come o his pa o he expe imen is ha he pe o -
mance o he labo a o ies ma ked A o D, i.e., he labo a o ies pa icipa ing in he main pa
o he expe imen , was ound o be sa is ac o y. The esul ing z-sco es e en lie in he in e al
h−1;1i, indica ing ha he ou pu s o hese labo a o ies can be conside ed sa is ac o y.
4.2. Assessmen o Homogenei y o Tes Resul s
The nex phase o he e alua ion ocused only on he esul s o labo a o ies A o
D, which measu ed es specimens s o ed du ing cu ing in wa e labelled “ esh wa e ”
and he o he ypes o wa e s o age desc ibed abo e. The esul s o he es s, including
he means and sample s anda d de ia ions, a e shown in Figu e 4. Figu e 4p esen s
he es specimens o labo a o y A a e 100 eeze– haw cycles. I is no iceable om he
su ace colou a ion ha he su ace has been damaged di e en ly by de-icing chemicals
(see Figu e 5).
The i s s ep in he s a is ical analysis o he esul s o labo a o ies A o D was o
conside whe he he di e ences be ween he indi idual es esul s exhibi s a is ically
signi ican di e ences. The aim was o de e mine whe he he esul s o he es s conduc ed
each ime on specimens om he same wa e s o age can be mixed and, hus, be conside ed
as a homogeneous s a is ical se sui able o u he e alua ion. A he same ime, hese
es s will demons a e whe he he expe imen al ou pu s can be in luenced by a highe le el
o ep oducibili y (see abo e). The so-called analysis o a iance (ANOVA) was applied,
whose null hypo hesis is he equali y o means o he esul s be ween labo a o ies. Tes ing
was pe o med a he 0.01 le el o signi icance. The esul s o he ANOVA es s, p esen ed
Ma e ials 2023,16, 4928 9 o 15
as p- alues in Table 2, indica e no s a is ically signi ican di e ences be ween he esul s o
he di e en labo a o ies. A he same ime, he ed c oss in Figu e 6indica es indi idual
es esul s ha we e lagged as ou lie s by he G ubbs es a e mixing he alues and we e,
he e o e, excluded om he expe imen . A g aphical ep esen a ion o all non-ou lying
esul s is shown in Figu e 7. The es esul s o one specimen ha showed a signi ican ly
highe amoun o scaled o ma e ial om he specimen a e 50, 75, and 100 eeze– haw
cycles we e excluded.
Table 2. Resul s o s a is ical es s.
Wa e S o age Numbe o Cycles ANOVA p-Value Shapi o Tes p-Value
(No mali y Tes )
Shapi o Tes p-Value
(No mali y Tes ) a e
Excluding Ou lie s
Slow lowing wa e
25 0.141 0.810 -
50 0.169 0.633 -
75 0.065 0.174 -
100 0.038 0.262 -
F esh wa e
25 0.043 0.992 -
50 0.047 0.227 -
75 0.099 0.861 -
100 0.131 0.647 -
P e iously used wa e
25 0.052 0.044 0.204
50 0.045 0.001 0.232
75 0.061 0.001 0.362
100 0.072 0.003 0.654
Figu e 4.
Compa ison o es esul s o he main pa o he expe imen . The es esul s o each
labo a o y a e shown o all mul iples o eeze– haw cycles. F om he le : 25, 50, 75 and 100 F–T
cycles, acco ding o [36]: (a) Slow lowing wa e (b) F esh wa e (c) P e iously used wa e .