Comparison of results of impedance spectroscopy methods with results of impact-echo method in investigation of high-temperaturedegraded concrete

P ocedia Enginee ing 151 ( 2016 ) 265 – 270 A ailable online a www.sciencedi ec .com 1877-7058 © 2016 The Au ho s. Published by Else ie L d. This is an open access a icle unde he CC BY-NC-ND license (h p://c ea i ecommons.o g/licenses/by-nc-nd/4.0/). Pee - e iew unde esponsibili y o he o ganizing commi ee o ICEBMP 2016 doi: 10.1016/j.p oeng.2016.07.372 ScienceDi ec In e na ional Con e ence on Ecology and new Building ma e ials and p oduc s, ICEBMP 2016 Compa ison o esul s o impedance spec oscopy me hods wi h esul s o impac -echo me hod in in es iga ion o high- empe a u e- deg aded conc e e Mi osla Lunak*, I o Kusak, Daniela S e ko a Ins i u e o Physics, Facul y o Ci il Enginee ing, B no Uni e si y o Technology, Ve eĜí 95, 60200 B no, Czech Republic Abs ac The g ow h in he anspo indus y esul s in mo e acciden s, which o en include i e in loca ions wi h he conc e e s uc u es, which also e ec s i s li e ime and i s unc ionali y. A e any case o i e loading i is equi ed o e i y s uc u e’s condi ion. One way how o assess he isks connec ed wi h u he usage o he s uc u es a e being exposed o he i e is he non-des uc i e es ing. This pape is dealing wi h non-des uc i e measu emen o changes o elec ic pa ame e s o he cemen based mo a s subjec ed o he high empe a u es. P isma ic samples wi h dimensions o 40×40×160 mm we e p epa ed om wa e , Po land cemen CEM I 42.5 R and qua z sand (in a a io 1:3), wi h wa e /cemen a io 0.46. Samples we e in en ionally deg aded by high empe a u e in ange o 200–1200 °C (200 °C s ep) and he changes in bulk densi y, lexu al ensile s eng h and elec ic p ope ies (loss ac o an į ( ), he imagina y componen impedance ImZ ( ) and he eal pa o componen impedance ReZ ( ) we e moni o ed. Two me hods we e used o es ing o elec ic p ope ies – Impedance spec oscopy and Impac -echo me hod. The ob ained esul s we e compa ed and sui abili y o used me hods was assessed. © 2016 The Au ho s. Published by Else ie L d. Pee - e iew unde esponsibili y o he o ganizing commi ee o ICEBMP 2016. Keywo ds: Impac -echo me hod; cemen -based composi e ma e ial; high- empe a u e deg ada ion; bulk densi y; lexu al ensile s eng h; impedance spec oscopy; dielec ic losses * Mi osla Lunak. Tel.: +0-420-541-147-659. E-mail add ess: [email p o ec ed] © 2016 The Au ho s. Published by Else ie L d. This is an open access a icle unde he CC BY-NC-ND license (h p://c ea i ecommons.o g/licenses/by-nc-nd/4.0/). Pee - e iew unde esponsibili y o he o ganizing commi ee o ICEBMP 2016 266 Mi osla Lunak e al. / P ocedia Enginee ing 151 ( 2016 ) 265 – 270 1. In oduc ion The in as uc u e is expanding e y quickly in p esen ime and he e o e he e is a big demand on building ma e ials. The g ow h wi hin he au omobile indus y esul s in mo e acciden s, which o en include i e in loca ions wi h he conc e e s uc u es, which also e ec s i s li e ime and i s unc ionali y. A e any case o i e loading i is equi ed o e i y s uc u e’s condi ion. The aim o non-des uc i e es ing is o p edic li e ime o a s uc u e and assessmen o isks connec ed wi h i s u he usage. G ea ad an age o his me hod is ha a e es ing he s uc u e emains undamaged. The impedance spec oscopy is a non-des uc i e es ing me hod employing he impedance cha ac e is ic equency dependence o analyse he p ope ies o he cemen and non-cemen ma e ials. The equipmen se -up designed o s udy he sys em unde in es iga ion includes: a me al-ma e ial-me al composi ion, which is ele an o iden i ying he applica ion limi s o he impedance spec oscopy me hod. The me hod canno be applied o hick-laye low- conduc i i y ma e ials. Rein o ced conc e e p oduc s may se e as an example. The p inciple o he men ioned me hod is based on e alua ion o he dielec ic s udying he dielec ic losses e sus equency plo s. The dielec ic losses o composi e ma e ials and plas ics can assume alues which a e many imes highe han hose o he mos ma e ials commonly used in he building indus y [4]. Analysis o impedance spec a a iances in he an į( ) and ImZ( ) o ReZ( ) o he inhomogeneous ma e ials is a pa o he impedance spec oscopy which is s ill wai ing o i s de elopmen . A p esen , one is no able o de e mine unambiguously he indi idual ma e ial componen con ibu ions o he o al elec ic conduc i i y and pola iza ion a a ious equencies o he exci ing ield. Impac -echo is a me hod o non-des uc i e e alua ion o conc e e s uc u es. The p inciple o he me hod is based on analysing an elas ic impulse-induced mechanical wa e [12]. I is a echnique o law de ec ion in conc e e. The me hod o e comes many o he ba ie s associa ed wi h law de ec ion in conc e e based on ul asonic me hods. A sho -du a ion mechanical impac p oduced by apping a small s eel sphe ical body gi es ise o a low- equency s ess wa es ha p opaga e in o he s uc u e and a e e lec ed by laws and ex e nal su aces. A ansduce eco ds su ace displacemen s caused by e lec ions o hese wa es. Classical impac -echo has ecei e a ached close o impac . Desc ibed me hod has ansmi e in di e en posi ions. This signal desc ibes ansien local ib a ions, which a e caused by he mechanical wa e mul iple e lec ions inside he s uc u e. The dominan equencies o hese ib a ions gi e in o ma ion abou he condi ion o he s uc u e o o de e mine he loca ion o laws, a which he wa es a e ebounded. As a ule, he signal is digi ized by means o a da a p ocessing sys em o be ans e ed in o a compu e memo y [12]. 2. Ma e ial used Mo a samples (o dimensions 40 mm × 40 mm × 160 mm) we e p oduced using CEM I 42,5 R Po land cemen (CƼeskomo a ský Cemen -Heidelbe g Cemen G oup), qua z sand om Fil acƼní písky, s. .o. (in a a io o 1 o 3) and wa e . (wa e /cemen a io 0.46). In compliance wi h CƼSN 721200 s anda d, he specimens we e demolded a e 24 hou s, hen cu ed in wa e o 27 days and inally ai -cu ed o 31 days a labo a o y empe a u e (25±2 °C) and ela i e humidi y o 53±5 %.A e ini ial cu ing; he specimens we e d ied a a empe a u e o 60 °C o wo days. Subsequen ly, he specimens we e subjec ed o g adual hea ing in a u nace a 200 °C, 400 °C, 600 °C, 800 °C, 1 000 °C and 1 200 °C. The empe a u e inc ease a e was 5 °C/min. A dwell o 60 minu es a each empe a u e was p o ided, in o de o ind ou he e ec o he empe a u e on he specimens. A e hea ea men , he specimens we e le o cool down spon aneously a labo a o y condi ions. 3. Expe imen al se up A he Depa men o Physics, Facul y o Ci il Enginee ing, TU B no, he IS-based measu emen s ha e been implemen ed using ollowing ins umen a ion: Agilen 33220A gene a o , Agilen 54645A double-channel oscilloscope, HP 82350 PCI HP-IB In e ace ca d, and a PC. To ope a e he abo e men ioned ins umen s and o p ocess he IS da a acqui ed, a so wa e called IS_alpha has been p epa ed by he i s au ho o his pape . [2, 3] 267 Mi osla Lunak e al. / P ocedia Enginee ing 151 ( 2016 ) 265 – 270 In o de o pe o m impedance analysis, i was necessa y o place he samples be ween b ass elec odes. The samples we e es ed o he equency spec a om 40 Hz o 1 MHz. Moni o ed a iables we e: loss ac o an į ( ), he imagina y componen impedance ImZ ( ) and he eal pa o componen impedance ReZ ( ). In o de o gene a e he acous ic signal, a hamme o 12 g mass, o iginally suspended oma hange , was eleased o all down on he specimen om a heigh o 4 cm. The impulse is e lec ed by he su ace bu also by mic o-c acks and de ec s o he specimen unde in es iga ion. The esponse was picked up by an MIDI ype piezoelec ic senso . I s ou pu ol age was ed in o a TiePie enginee ing Handyscope HS3, which is a wo-channel, digi al, 16 bi s oscilloscope. The piezoelec ic senso was placed a he end o he beam a he cen e o ans e se side and he hamme hi was ca ied ou on he opposi e side in he di ec ion o he longi udinal axis. The senso was a ached o he su ace o he sample by beeswax. Subsequen ly, he as Fou ie ans o m echnique was used o ans o m he eco ded wa e o m in o he equency domain o each o he ou pu signals. Each measu emen un consis ed o 5 sepa a e measu emen s, om which an a e age was calcula ed. [11,12] 4. Resul s and discussion The e a e 6 cu es o di e en deg ee o hea s ess (0–1000 °C) on he g aph o he dielec ic loss ac o on equency dependence in Fig. 1. Cu e 8-200A (sample a e 200 °C hea s ess) has one maximum a equency 2 kHz and has he highes alues o dielec ic loss ac o on he whole equency ange. The shape o cu e is p o ing o he p esence o pola iza ion losses in he ma e ial [2]. The sample hea ed a 400 °C has a smalle maximum a equencies 70 Hz, hen a 800 Hz and shallow peak appea s a a equency o 30 kHz. All g aph alues d opped signi ican ly compa ed o he p e ious sample (200 °C). Pola iza ion losses educed signi ican ly, which is connec ed wi h he wa e e apo a ion (dehyd a ion) and deg ada ion o some componen s o he sealan o he elease o wa e apo . Fig. 1. (a) Spec um o dissipa ion ac o ; (b) Real pa o impedance e sus equency o elec ic ield. A simila pa e n shape o he spec um eaches 8-600A (600 °C) in he lowe equencies bu pe o ms a local maximum a equency o 90 Hz. A he emaining pa o he spec um loss ac o alues a e only sligh ly lowe han o sample 8-400B. Sample 8-800B annealed a 800 °C shows a p onounced maximum a a equency o 80 Hz, he alue o dissipa ion ac o in he equency 100 Hz each highe alues han p e ious samples. Abo e 100 Hz obse ed signi ican dec ease o all alues o loss ac o . Appa en ly he e is a dec ease pola iza ion loss, bu also o he o ma ion o c acks. Cha ac e o he cu e co esponds o sample 8-1000A (1000 °C) sugges s he dominance o conduc ion losses in he ma e ial (dec easing cha ac e h oughou he equency spec um). The lack o wa e and he p esence o c acks subs an ially inc eased esis ance o he sample in he egion o he spec um. 0 0.4 0.8 1.2 10 1 10 2 10 3 10 4 10 5 10 6 8-M A 8-200A 8-400B 8-600A 8-800B 8-1000A /Hz - an  G 10 2 10 4 10 6 10 8 10 2 10 3 10 4 10 5 10 6 8-M A 8-200A 8-400B 8-600A 8-800B 8-1000A /Hz Re Z/ : 268 Mi osla Lunak e al. / P ocedia Enginee ing 151 ( 2016 ) 265 – 270 Fo e e ence sample 8-M A (no hea s ess), wo local maxima in he spec um o dissipa ion ac o we e obse ed, also a equency o 2 kHz and 20 kHz, bu jus o 20 kHz equency alues each lowe dissipa ion ac o han sample 8-200A. Fig. 2. Imagina y pa o impedance e sus equency spec a. Sample 8-1200A (annealed a 1200 °C) could no be measu ed; he esis ance o he sample is ou side he ange o measu ing ins umen s ( he signal is oo a enua ed). The dependency o eal pa o impedance on he equency (log-log) spec um blends o all six samples a a equency o 10 kHz (Fig. 1). This equency has he highes alue o he eal pa o impedance o sample 8-1000A and g adually logical descen o he lowes alue in sample 8-MFA (no hea s ess). A signi ican po ion o he spec um is in he ange o 40-10 kHz. Fo a equency o 10 kHz eal pa impedance alues coincides ( o samples 0-800 °C), a sample 8-1000A (1000 °C) shows an unexpec ed d op o as low as 1 kȍ ( o 1 MHz). G aph o he imagina y impedance componen dependency on he equency (log-log) shows ha he samples subjec ed o high empe a u e s ess ImZ alues a e highe in he egion o he spec um (Fig. 2). Cu e o he sample wi hou hea s ess 8-MFA in e sec s cu e o 8-200A in he 40 - 300 Hz. Subsequen ly de ia es. Re e ence sample 8-MFA is in he emaining equency ange o he lowes alues o he imagina y impedance componen . The Fig. 3a) sho es g aph whe e he inc easing alue o eal and imagina y pa o impedance is isible, he e a e alues o all samples o he equency o 1 kHz. Fo samples annealed less han empe a u e o 600 °C, he inc ease in he alues is slow, bu ises sha ply o he samples annealed abo e 800 °C. Fo all samples eached Re(Z) is lowe han he alues o Im(Z). In he empe a u e ange o 400–600 °C, he shape o he cu es is ge ing changed om conca e o con ex. The a ea o his change co esponds o he cu e o u he in es iga ed dielec ic samples a a equency o 1 kHz, whe e in he same a ea has been o e u ned cha ac e o changes. The same shapes and changes has cu e o e y impo an peaks o dominan equencies, o a angemen U1 – S1, o sho way o acous ic impulse o senso o impac echo me hod. The e is small change o alues a a ea I, hen big change and hen small change in a ea III. Pe mi i i y cu e (Fig. 3 b) has shape om con ex o conca e o all spec a. The g aph indica ed h ee a eas in which we y o cap u e he s uc u al ans o ma ion o ma e ials a gi en empe a u es (s ages) o annealing. Sepa a ion indica ed empe a u es up o 400 °C and 573 °C. In his in e al a e also ound in lec ion poin s o all cu es. A signi ican dec ease o he dielec ic pe mi i i y in a ea I occu s by dehyd a ion and decomposi ion o some componen s o he sealan o he elease o s eam o CO2. So mos impo an equencies o impac -echo (dominan peaks) ha e e y simila shape o dec easing o alues. The e is one di e en poin o equency (Imp-Echo me hod) o he highes empe a u e o annealing. 104 105 106 107 108 109 102103104105106 8-M A 8-200A 8-400B 8-600A 8-800B 8-1000A /Hz Im Z / : 269 Mi osla Lunak e al. / P ocedia Enginee ing 151 ( 2016 ) 265 – 270 In a ea II we can see no signi ican changes o pe mi i i y alues, a hese empe a u es, he phase ansi ion o qua z (573 °C) (ȕ-qua z o Į-qua z). Decomposi ion o CSH gel causes comp essi e de o ma ion o he bond and an inc ease in po osi y. In he a ea III dielec ic s eep d op occu s, which may be ela ed o he eme gence o new c ys alline phase and a e y low s eng h ma e ial. Fig. 3. (a) Disc e e alues o Im(Z), Re(Z) o 1kHz o elec omagne ic ield and dominan equencies o impac echo me hod, o a angemen U1-S1(sho way o wa e); o di e en empe a u es o annealing; (b) Values o dielec ic pe mi i i y e sus equency o elec ic ield and dominan equencies o impac echo me hod, o a angemen U0-S0(long way o wa e), o di e en empe a u es o annealing. Du ing hea ing, he densi y o he obse ed samples dec eases. Hea ing o he samples leads o dehyd a ion and decomposi ion o some componen s o he sealan o he elease o s eam o CO2. The la ges dec ease in densi y occu s a he beginning o hea ing ( o he empe a u e o 400 °C). Also he ensile s eng h o he samples dec eases wi h inc easing empe a u e, which is ela ed o dehyd a ion and decomposi ion o CSH gel, bu also wi h he phase ans o ma ion o qua z a 573 °C (ȕ-qua z o Į-qua z). Decomposi ion o CSH gel causes he p essu e connec ed wi h de e io a ion o he bonds wi h subsequen inc ease o po osi y. A empe a u e abo e 1000 °C, c ys alline phase is o med in he specimens. 5. Conclusion The ob ained esul s show ha he equency inspec ion ca ied ou by means o he Impac -echo me hod makes a con enien ool o assess he quali y and li e o hese composi e ma e ials when exposed o ele a ed empe a u e. The acous ic esul s con i med he di e ences in he s uc u e o mo a specimens. The es ed specimens we e mo a s p epa ed om a mix u e o cemen and qua z sand. The specimens we e in en ionally deg aded by applica ion o ele a ed empe a u es o 200 °C o 1 200 °C. A shi o he p edominan equencies and a change in he damping coe icien we e obse ed o occu du ing he deg ada ion p ocess. Specimens we e es ed by impedance measu emen and in e es ing simila i y we e ob ained wi h impac echo es esul s. When dominan equencies we e dec easing, he pe mi i i y was dec easing oo, and by he same end. The de ia ion was obse ed o he highes empe a u e o annealing. Di e ence be ween measu ed esul s o a angemen U1-S1 (sho way) and U0-S0 (long way o wa e) a e in wid h o in e al o equencies and in nume ical ank. Nex discussion is necessa y o unde s anding o p o e o deg ada ion o annealing conc e e. 106 107 108 0400 800 1200 0 3500 7000 10500 14000 / Hz A angemen U1-S1 Im (Z) Re(Z) A e a III A e a I A e a II Tempe a u e o hea s ess /°C ReZ, ImZ / : 0400 800 1200 3500 4000 4500 5000 5500 / Hz 0 50 100 150 H A angemen U0-S0 H A e a III A e a I A e a II Tempe a u e o hea s ess /°C 270 Mi osla Lunak e al. / P ocedia Enginee ing 151 ( 2016 ) 265 – 270 Acknowledgemen s This pape has been wo ked ou unde he p ojec GAýR No. 16-02261S and unde he p ojec No.S-16-2967 suppo ed by Facul y o Ci il Enginee ing BUT. Re e ences [1] I. Kusak, M. Lunak, Compa ison o impedance spec a o conc e e eco ded wi h u ilizing ca bon ansi ion pas e, Ad . Ma . Res. 897 (2014) 131–134. [2] V. Men lik, 'LHOHNWULFN«SUYN DV VW«P BEN – echnická li e a u a, P aha, 2006. [3] M. Lunak, I. Kusak, Z. Chobola, Dielec ic p ope ies o conc e e specimens a e hea s ess, Appl. Mech. Ma e . 446–447 (2014) 1389–1394. [4] M. Cabeza, P. Me ino, A. Mi anda, X.R. No oa, I. Sanchez, Impedance spec oscopy s udy o ha dened Po land cemen pas e, Cemen Conc e e Res. 32 (2002) 881–891. [5] T. Ficke , L. Topola , I. Kusak, Is componen ial s eng h analysis o conc e e possible?, Mag. Conc e e Res. 65 (2013) 1480–1485. [6] L. Pazde a, L. Topola , V. Bilek, J. Smu ny, I. Kusak, M. Lunak, Measu ing o Conc e e P ope ies du ing Ha dening, ESA 2010. 1. CZ, Palacky Uni e si y (2010) 311–318. [7] I. Kusak, M. Lunak, P. Schaue , T acing o conc e e hyd a ion by means o impedance spec oscopy (New ool o building elemen s es ing), Appl. Mech. Ma e . 248 (2013) 370–378. [8] J.R. Macdonald, Impedance spec oscopy, emphasizing solid ma e ials and sys ems, bibliog aphy, Canada, 1987. [9] M. Lunak, I. Kusak, L. Pazde a, L. Topola , V. Bilek, Moni o ing o cemen -based ma e ial solidi ica ion, ocusing on elec ical p ope ies, P oc. o he 48 h In e na ional Scien i ic Con e ence Expe imen al S ess Analysis 2010, 233–240. [10] M. K ause, M. Ba mann, R. F ielinghaus and e al., Compa ison o pulse-echo me hods o es ing conc e e, NDT & E In e na ional, 30 (1997) 195–204. [11] P. Mazal, L. Pazde a, J. D oĜáþek, Applica ion o acous ic emission me hod in con ac damage iden i ica ion, In . J. Ma e . P od. Tech. 41 (2011) 140–152. [12] I. Plsko a, Z. Chobola, M. Ma ysik, Assessmen o ce amic ile os esis ance by means o he equency inspec ion me hod, Ce am.-Silika y 55 (2011) 176–182. [13] J.M. To en s, J. Ronce o, R. Ge u, U iliza ion o impedance spec oscopy o s udying he e a ding e ec o a supe plas icize on he se ing o cemen , Cemen Conc e e Res. 28 (1998) 1325–1333.