Optimization of two methods based on ultrasound energy as alternative to European standards for soluble salts extraction from building materials

1 Optimization of two methods based on ultrasound energy as alternative to European standards for soluble salts extraction from building materials N. Prieto-Taboada*, O. Gómez-Laserna, I. Martinez-Arkarazo, M.A. Olazabal and J.M. Madariaga Department of Analytical Chemistry, University of the Basque Country, (UPV/EHU), Barrio Sarriena s/n, 48940, Leioa, Spain. Email*: [email protected] Tel.: +34 94 601 82 94; Fax.: +34 94 601 35 00 Abstract The Italian recommendation NORMAL 13/83, later replaced by the UNI 11087/2003 norm, were used as standard for soluble salts extraction from construction materials. These standards are based on long-time stirring (72 and 2 hours, respectively) of the sample in deionised water. In this work two ultrasound based methods were optimized in order to reduce the extraction time while efficiency is improved. The instrumental variables involved in the extraction assisted by ultrasound bath and focused ultrasounds were optimized by experimental design. As long as it was possible, the same non instrumental parameters values as those of standard methods were used in order to compare the results obtained on a mortar sample showing a black crust by the standards and the optimized methods. The optimal extraction time for the ultrasounds bath was found to be of two hours. Although the extraction time was equal to the standard UNI 11087/2003, the obtained extraction recovery was improved up to 119%. The focused ultrasound system achieved also better recoveries (up to 106%) depending on the analyte in one hour treatment time. The repeatabilities of the proposed ultrasound based methods were comparables to those of the standards. Therefore, the selection of one or the other of the ultrasound based methods will depend on topics such This is the accepted manuscript of the article that appeared in final form in Ultrasonics Sonochemistry 19(6) : 1260-1265 (2012), which has been published in final form at https://doi.org/10.1016/j.ultsonch.2012.03.002. © 2012 Elsevier under CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/) 2 as laboratory facilities or number of samples, and not in aspects related with their quality parameters. 1 INTRODUCTION Soluble salts are considered deterioration compounds formed by reaction among acid gases present in polluted environments and different construction materials. They can also accumulate through infiltration of water charged of salts or they can be part of the original composition. In every way, they are one of the most important deterioration factors of constructions materials, especially in carbonate based stones such as limestone [1-5]. The characterization of salts in construction material is very important because the damage they cause (cracking and material loss) depends on their content and nature. For instance, salts having several hydration states with substantial molecular volume differences such as mirabilite (Na2SO4·10H2O) and thenardite (Na2SO4) are usually found causing this effect. The transition from one to another occurs at a wide range of common temperature and relative humidities with a density change from 1.49 to 2.7 gr/cm3 [6-9]. Furthermore, such salts are even more damaging when forming as subeflorescences because they produce higher material losses by the so called plate damage or salt weathering [10, 11]. Therefore, the establishment of the soluble salts content could be essential when constructive elements of buildings are going to be restored or the construction material is going to be reused [12]. Among construction materials, mortars have shown to be the most susceptible to suffer from the mentioned pathologies due to their porosity that facilitates the penetration of the salts deep in the material. Mortars are common construction material used in buildings not only as 3 binders but also with embellishing or decorative purposes. Therefore, a good conservation state of these materials is essential to preserve the integrity of the building. Nowadays, they are generally composed by concrete, arid (usually sand or milled sandstone), water and additives in some cases. However, in the field of cultural heritage the composition of mortars is limited to the raw materials available in the surroundings and thus, they are commonly composed of calcite or gypsum that cement the arid grains. Mortars suffer a faster degradation when beach sand is used as arid, due to its high concentration of chlorides that are very soluble [13, 14]. For all the mentioned, the determination of the soluble salts content is especially important in mortars. In this sense, the most common and hazardous soluble salts found in construction materials, since they affect to their integrity, are sulphates, nitrates, chlorides and carbonates/bicarbonates [15]. Usually, the soluble salts are present in black crusts that often appear covering facades protected from rain washing [16-18]. Although gypsum (CaSO4·2H2O) and particulate matter (soot, metals, organics,…) are the principal components of this kind of formation, nitrates have been also found at considerable concentration levels in recent studies due to the increase of the impact of NOx gases [7, 19, 20]. Despite it is well known that salts excess is completely harmful for the integrity of construction materials there is not a global legislation concerning the safety amounts (in terms of material integrity) in building materials. Only sulphate and chloride concentrations in materials and the water used with construction purposes is controlled. For instance, according to the Structural Concrete Instruction EHE-08, approved by Royal Decree in 2008 in Spain the concentration in the original material should not exceed 0.8% w/w and 0.05% w/w of sulphate and chloride respectively. 4 The analyses related with the soluble salts content in building are carried out by protocols recommended by different European standardization organizations used in different countries as Italy, Germany, Britain, Sweden, Spain or France [21-27]. Besides, due to the importance of the determination of total soluble salt content in such materials, especially in the field of Cultural Heritage protection, a global European standard method related to the “Determination of soluble salts in natural stones and related artificial materials used in cultural heritage” is currently being developed [28]. One of the most classical methods is the Italian NORMAL 13/83 protocol [29] which involves a 72 hour treatment. This norm was later replaced by the UNI 11087/2003 [21]. Both methods, as well as most of the standards, are based in the simply stirring of the sample, but the advantage of the UNI 11087/2003 norm over the NORMAL 13/83 norm was that extraction time is shortened till 2 hours. As alternative, new procedures to monitor in situ soluble salts content in stone materials are being developed [19, 30] and an ultrasound energy based method applied to this field can be found in bibliography [7, 31]. However, the real possibilities of this instrumentation to improve the actual standard norms have not been evaluated in depth. On the contrary, the use of techniques based on ultrasound energy has grown in the last years due to their effectiveness and simplicity. Ultrasound energy is used for many applications as alternative extraction method, for example, to microwave extraction. The ultrasound bath (USB) is one of the most common instruments used nowadays for accelerated extractions due to its simplicity and effectiveness. On the other hand, focused ultrasound (FUS) systems that apply the ultrasound energy through a tip immersed directly in the sample, are becoming more and more popular because, among others, of its higher repeatability [32]. Despite FUS has been used with good 5 results in other fields the use of this technique has not been described for the extraction of soluble salts up to now. The aim of the present study was to optimize the extraction of soluble salts from mortars assisted by ultrasound bath (USB) and focused ultrasound systems (FUS) and subsequently, to compare the efficiency of ultrasound assisted extraction methods with the currently used standards. 2 MATERIAL AND METHODS 2.1 Samples A rendering mortar showing a black crust was used in the present study. The sample belonged to a building located in a highly industrialized neighbourhood of Bilbao which has suffered several floods in the past due to the closeness to the Nerbion-Ibaizabal River. The samples with black crust have a high soluble salts concentration due to the capacity of black crust to trap contaminants [18, 19]. Therefore, the rendering mortar of the building was considered an appropriate sample to evaluate the soluble salts extraction methods. This fact was confirmed by means of Raman spectroscopy measurements. An InnoRam BW&Tek handheld Raman spectrometer was also used in order to identify the original composition of the mortar and the degradation compounds. The equipment is composed of a 785nm laser and a Peltier cooled CCD detector. The signal is transmitted by an optic fiber connected to a probe, which offers the possibility to perform microscopic analysis by using different optical lens (4x, 20x and 50x) in a range of 3000-100 cm-1 (non changeable) with a spectral resolution of 3.5 cm-1. According to the spectra obtained on the mortar, it was mainly composed of calcite, iron oxides and aluminosilicates as shown in figure 1. The suitability of this technique has been 6 widely demonstrated for characterization of original composition and soluble salts in previous works [18, 19, 33]. With regard to the soluble salts, sulphates (mainly gypsum, CaSO4·2H2O) and nitrates were the mayor salts that were identified by Raman (see Figure 1) but also chlorides were found by means of ion chromatography following a procedure described elsewhere [34], with a total soluble salt content of approximately 8% w/w. In order to obtain a homogeneous sample, the mortar was grinded firstly in a ball mill Pulverisette from Fritsch and then with an agate mortar. After, the sample was dried at 60ºC for 24 hours. This treatment allows to achieve sample without humidity and a homogeneous particle size. The sample was kept in a desiccator till the extraction treatment. 2.2. Instruments 2.2.1 Ultrasound based systems An Ultrasons-H bath from P-Selecta with a working frequency of 40 kHz (1000 W) was used to optimize the extraction by the ultrasound bath. The conductivity was monitored with a Crison CM/35 portable conductivity meter from Crison Instruments to define the optimal extraction time as recommended in the NORMAL 13/83 method [29, 35]. The focused ultrasound (FUS) system used for the optimization was an HD 2070 Sonopuls Ultrasonic Homogenizer from Bandelin equipped with a GM 2070 generator (maximum 70W, 20 kHz), an UW 2070 ultrasonic converter a SH 70G horn and a MS 73 titanium probe (3mm). In this case the titanium probe was chosen because it allows power setting till 100% instead of 30% for the glass probe. According to the technical requirements the maximum extraction volume is 50 ml in order to assure a correct ultrasound energy transmission. 7 Finally, in order to accomplish the extraction according to European norms the stirring of the samples was carried out by a P-Selecta Asincro stirrer (30W). 2.2.2. Quantitative analysis by Ion Chromatography With the aim of comparing the results obtained by standard methods and proposed ultrasound based methods, the extracted salts were quantified by a Dionex ICS 2500 ionic chromatograph with an ED50 suppressed conductivity detector. An IonPac AS23 (4×250 mm) column and IonPac AG23 (4×50 mm) precolumn from Vertex were used for the separation of anions (fluoride, chloride, sulphate and nitrate). The quantification of cations (sodium, potassium and calcium) was conducted by using an IonPac CS12A (4×250 mm) column and IonPac CG-12A (4×50 mm) precolumn from Vertex. The chromatographic conditions used in anion quantification were 5 mM Na2CO3/0.8 mM NaHCO3, 25 mA and 1 ml/min as mobile phase, suppression current and flow, respectively. In the case of cations 20 mM CH4SO3 as mobile phase, 59 mA of suppression current and 1 ml/min flow were used. Prior to the analysis the samples were filtered with a 0.45 µm nylon syringe filters and brought to a final volume. 2.3. Optimization of the ultrasound assisted extraction Ultrasound bath and focused ultrasound conditions were optimized to carry out salts extraction. Four replicates of the extraction were carried out at the optimum values and the results were assessed by comparison with those obtained with standards (NORMAL 13/83 and UNI 11087/2003). Sample weight and water volume were kept constant in the optimization: 100 mg of sample and 100 ml of deionised water. That is, only instrumental variables were studied. 8 2.3.1. Ultrasound bath optimization The unique instrumental variable to optimize in the ultrasound bath was the extraction time. The optimal extraction time was established at the maximum constant (conductivity/sample weight) according to the continuous measurements performed on each of the four replicates. 2.3.2. Focused ultrasound optimization The variables to optimize were power setting (P, 0–100%) which allows controlling amplitude of the delivered ultrasound, cycles (c, 1-9) and extraction time (t). Cycles are an instrumental variable that indicates a pulsed mode; for instance, cy = 4 means a 0.4 s active period followed by a 0.6 s passive period [32]. According to instrumental requirements, and in order to avoid recommended limit values, the extractant volume was set at 45 ml. These instrumental requirements far from being negative, could involve to obtain concentrations above detection limits in some cases, since the final volume is reduced. A full factorial design was carried out in order to determine the significant variables from the initial hypothesis. Power (10-90%), cycles (1-9) and extraction time (5-100 min) were studied through quantification by ionic chromatography. The three variables were checked till the maximum fixable value except of the power, for which 90% was set as maximum value in order to preserve the tip from damage. The high level of time was determined by the maximum working time of the instrument. The significance of the cited variables was analysed by a full factorial design accomplished by the UnScrambler v9.2 software [36]. After, a central composite design (CCD) was carried out to determine the optimal conditions with the significant variables. NLREG 6.3 program [37] was also used to determine the response equations for each of the analytes. 9 2.4. Comparison of optimized ultrasound methods with standards The Italian NORMAL 13/83 of 1983 and its actualization so called UNI 11087/2003 were used to extract the soluble salts from the mortar in order to compare their extraction capabilities as well as to assess the suitability of the optimized methods. Both standard methods consist on stirring of 100mg of the sample in 100ml of deionised water. The difference between them is that extraction time was reduced from 72 hours to 2 hours in the UNI 11087/2003 method. The non instrumental parameters were kept constant as far as it was possible in the optimization of the ultrasound based methods. Four replicates of the samples were performed by each of the methods to accomplish the comparison step. 3 RESULTS AND DISCUSSION 3.1 Optimization of the ultrasound assisted extraction 3.1.1. Ultrasound bath optimization The extraction of the salts by the ultrasound bath was optimized by monitoring the conductivity at different extraction times. Besides, the temperature at different positions of the bath was monitored as well. Temperature showed different trends in the bath depending of the position of the baker, which could affect to the repeatability of the analysis and therefore to the election of the optimal time. The temperature increased from room temperature to 51ºC with differences up to 4ºC depending on the position of the baker in the bath. Figure 2 shows that maximum conductivity value was achieved approximately at 100 minutes for some samples. However, 120 minutes was chosen as extraction time in order to ensure constant conductivity values. The most elevated temperatures observed in the different positions in the bath seemed not to be related with maximum conductivity values achieved in the extraction. In addition, re-adsorption phenomena were also occasionally observed as can be seen in Figure 2 (serie A). 16 [13] A. Silva, T. Cruz, M. Paiva, A. Candeias, P. Adriano, N. Schiavon, J. Mirão, Environ. Earth Sci. 63 (2011) 1641-1650. [14] E.A.L. Costa, V.P. Campos, L.C.P. da Silva Filho, H.A. Greven, J. Environ. Manage. 90 (2009) 1060-1068. [15] H. Brocken, T.G. Nijland, Constr. Build. Mater. 18 (2004) 315-323. [16] P. Maravelaki-Kalaitzaki, Anal. Chim. Acta 532 (2005) 187-198. [17] C. Sabbioni, G. Zappia, N. Ghedini, G. Gobbi, O. Favoni, Atmos. Environ. 32 (1998) 215-223. [18] N. Prieto-Taboada, M. Maguregui, I. Martinez-Arkarazo, M. Olazabal, G. Arana, J. Madariaga, Anal. Bioanal. Chem. 399 (2010) 2949-2959. [19] I. Martínez-Arkarazo, M. Angulo, L. Bartolomé, N. Etxebarria, M.A. Olazabal, J.M. Madariaga, Anal. Chim. Acta 584 (2007) 350-359. [20] S.W. Massey, Sci. Total Environ. 227 (1999) 109-121. [21] UNI 11087:2003. Beni culturali - Materiali lapidei naturali ed artificiali. Determinazione del contenuto di sali solubili, Ente Nazionale Italiano di Unificazione (UNI), (2003). [22] WTA 4-5-99: Evaluation of masonry, WTA-Merkblätter, (1999). [23] BS EN 772-5:2001 Methods of test for masonry units. Determination of the active soluble salts content of clay masonry units, The British Standards Institution (BSI), (2002). [24] SIS EN 1015-17/A1:2004 Methods of test for mortar for masonry - Part 17: Determination of water-soluble chloride content of fresh mortars, Swedish Standards Institute (SIS), (2004). [25] DIN 5110: Testing of ceramic raw materials and materials; determination of soluble salts (percolation method), Deutsches Institut Fur Normung E.V. (German National Standard), (1957). [26] UNE EN 1744-5:2006 Tests for chemical properties of aggregates - Part 5: Determination of acid soluble chloride salts, Asociación Española de Normalización y Certificación (AENOR), (2007). [27] NF EN 1744-5: Tests for chemical properties of aggregates - Part 5 : determination of acid soluble chloride salts, Association Française de Normalisation (AFNOR), (2007). [28] European Committee for Standardization (CEN) http://www.cen.eu . [29] Raccomandazioni Normal 13/83: Dosaggio dei Sali solubili (Normal 13/83 recommendation: Determination of Total Amount of Soluble Salts, Consiglio Nazionale delle Ricerche and Istituto Centrale del Restauro (CNR-ICR), (1983). 17 [30] C. Tropea, M.P. Sammartino, G. Visco, Curr. Anal. Chem. 6 (2010) 94-99. [31] A.C. Iñigo, R. Alonso, S. Vicente-Tavera, Ultrason. Sonochem. 8 (2001) 127-130. [32] S. Fdez-Ortiz de Vallejuelo, A. Barrena, G. Arana, A. de Diego, J.M. Madariaga, Talanta 80 (2009) 434-439. [33] M. Pérez-Alonso, K. Castro, I. Martínez-Arkarazo, M. Angulo, M.A. Olazabal, J.M. Madariaga, Anal. Bioanal. Chem. 379 (2004) 42-50. [34] E. Pinto Guerra, Rissanamento di murate umide e degradate, second ed., Dario Flaccovio, Italia, 2011. [35] H. Svahn, Non-destructive field tests in stone conservation, Riksantikvarieämbetet, Stockholm, 2006. [36] Camo Process, 9.2 (2005). [37] P.H. Sherrod, (1992-2005). [38] A.F. White, S.L. Brantley, Chem. Geol. 202 (2003) 479-506. 18 Table 1.- Extraction values of the Italian standard NORMAL 13/83 and the UNI 11087/2003 expressed in mg/kg, as well as the recovery of the analytes of the UNI 11087/2003 standard in comparison to preceding norm. Table 2.- Results in mg/kg obtained by ion chromatography for the experiments of the CCD designed for the FUS method optimization. Table 3.- Summary of the parameter values fixed for the proposed ultrasound methodologies and standards used in the present work. 19 Figure 1.- Raman spectra of the black crusted mortar used in this work. a) calcite, gypsum some nitrates and silicates , b) sulphates and iron oxides c) calcite and shoot, d) gypsum Figure 2.- Results obtained for some of the position (series) in the ultrasound bath during monitoring of the corrected conductivity (conductivity/sample weight) at different extraction times. Figure 3.- Representation of the concentration in mg/kg obtained with the different protocols: 1 NORMAL 13/83, 2 UNI 11087/2003, 3 USB and 4 FUS Figure 4.- Summary of the steps to be followed to accomplish the extraction methods optimized in this work. 1 Table 1.- Extraction values of the Italian standard NORMAL 13/83 and the UNI 11087/2003 expressed in mg/kg, as well as the recovery of the analytes of the UNI 11087/2003 standard in comparison to preceding norm. CHLORIDE NITRATE SULPHATE SODIUM POTASIUM CALCIUM NORMAL 13/83 2284 1945 38386 2043 988 31865 Standard deviation 36 102 333 341 68 5535 RSD 2% 5% 1% 17% 7% 17% UNI 11087/2003 2071 1723 36955 959 770 27200 Standard deviation 15 38 246 50 13 801 RSD 1% 2% 1% 5% 2% 3% Recovery 91% 89% 96% 47% 78% 100% 1 Table 2.- Results in mg/kg obtained for the experiments of the CCD designed for the FUS method optimization. Time (min) Power Cycles Chloride Nitrate Sulphate Sodium Potassium Calcium *L:A-a 4.98 50 5 1637 1255 37699 678 491 23391 *H:A-a 99.98 50 5 1997 1603 38834 826 637 23238 *L:B-a 52.5 10 5 2173 1480 38526 1296 310 24502 *H:B-a 52.5 90 5 1954 1569 38635 807 663 23314 *L:C-a 52.5 50 1 2176 1467 38311 1317 291 21602 *H:C-a 52.5 50 9 2322 1597 39546 1421 426 24087 Cube001a 22.8 25 3 2044 1436 38803 1292 291 22643 Cube002a 82.2 25 3 2164 1536 38650 1251 276 23935 Cube003a 22.8 75 3 1798 1474 39073 751 530 22096 Cube004a 82.2 75 3 2282 1609 39308 1310 353 24046 Cube005a 22.8 25 8 1939 1501 40034 792 589 24266 Cube006a 82.2 25 8 2312 1576 39603 1406 388 22463 Cube007a 22.8 75 8 1952 1531 39676 807 637 23216 Cube008a 82.2 75 8 2103 1672 39209 880 736 23620 Cent-a 52.5 50 5 2070 1652 40472 852 680 24105 Cent-b 52.5 50 5 1991 1579 39228 827 658 23461 Cent-c 52.5 50 5 1942 1572 39007 814 644 23096 1 Table 3.- Summary of the parameter values fixed for the proposed ultrasound methodologies and standards used in the present work. STANDARDS CURRENT WORK NORMAL 13/83 UNI 11087/2003 USB FUS Weight (gr) 0.1 0.1 0.1 0.1 Volume (ml) 100 100 100 45 Time (h) 72 2 2 1 Method Agitation Agitation Ultrasound bath Focused ultrasound