Color stability in mortars and concretes. Part 1: Study on architectural mortars

Construction and Building Materials 120 (2016) 617–622

Contents lists available at ScienceDirect

Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat

Color stability in mortars and concretes. Part 1: Study on architectural mortars Anahí López a,b,c,⇑, Gastón Alejandro Guzmán d, Alejandro Ramón Di Sarli d a

LEMIT: Laboratorio de Entrenamiento Multidisciplinario para la Investigación Tecnológica (CICPBA), Av. 52 s/n, e/ 121 y 122 – B1900AYB – La Plata, Buenos Aires, Argentina CONICET: Consejo Nacional de Investigaciones Científicas y Técnicas, Argentina c UTN-FRLP: Universidad Tecnológica Nacional-Facultad Regional La Plata, Av. 60 esq. 124 s/n – B1900AYB – La Plata, Buenos Aires, Argentina d CIDEPINT: Centro de Investigación y Desarrollo en Tecnología de Pinturas (CICPBA-CONICET La Plata), Av. 52 s/n, e/ 121 y 122 – B1900AYB – La Plata, Buenos Aires, Argentina b

h i g h l i g h t s
The color stability in architectural mortars was studied.
Color was characterized using the CIELAB color space.
Results from CIEDE1976 and CIEDE2000 color-difference formulas were compared.
The weathering effect was analyzed on the colored mortars.
Loss of initial color was a function of the exposure conditions.

a r t i c l e

i n f o

Article history: Received 28 January 2016 Received in revised form 12 May 2016 Accepted 22 May 2016

Keywords: Color stability Architectural mortars CIELAB color space CIEDE1976 CIEDE2000

a b s t r a c t Architectural mortars are mixtures used in the building industry when aesthetic surface value is required. Cementitious mixtures with colorant agent, meet this requirement but color stability is not easy to ensure. This study determines the levels of color stability in architectural mortars. The main materials studied were grey cement and yellow iron oxide pigment. The CIELAB color space was adopted to define color and the CIEDE1976 color-difference formula was compared to CIEDE2000 to asses color stability. Results revealed loss of color in natural environments in a short time and excellent color stability in a chamber operating under controlled humidity and temperature conditions. Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction Different materials proportions allow developing Portland cement-based mixtures with a wide range of properties either in the hardened or in the fresh states. The possibility that the mortar keeps up its original color depends on internal and external causes capable of changing completely the surface color. The internal causes include typical characteristics of the material structure affecting the mechanisms of fluids transport. The external causes comprise interactions between the mortar and the environmental conditions to which

⇑ Corresponding author at: LEMIT: Laboratorio de Entrenamiento Multidisciplinario para la Investigación Tecnológica (CICPBA), Av. 52 s/n, e/ 121 y 122 – B1900AYB – La Plata, Buenos Aires, Argentina. E-mail addresses: [email protected] (A. López), [email protected]. ar (A.R. Di Sarli). http://dx.doi.org/10.1016/j.conbuildmat.2016.05.133 0950-0618/Ó 2016 Elsevier Ltd. All rights reserved.

it is exposed. When the mortar quality is involved, the material deterioration and subsequent loss of its structural function starts as the deterioration progresses. In the particular case of architectural mortar, damages such as stains, leaching or specifically color variations, which do not affect the structural use, have become an issue of great interest. There is a resistance to mixing pigments and cement due to the few experiences that ensure lasting color. For a pigment to be satisfactorily incorporated into the building industry it is assumed that its response in elements exposed to outdoor weather conditions should be evaluated over a period of 5–7 years. Although there are accelerated aging tests that allow inquiring the color stability, they are not widely accepted, as they have not proved the existence of a correlation between accelerated and real changes. The pigment solidness also depends on its clustering, therefore, it is strongly related to the structural matrix [1].

618

A. López et al. / Construction and Building Materials 120 (2016) 617–622

In the process of deterioration of cementitious mixtures, the development of calcium carbonate (CaCO3) deposits or efflorescence is one of the factors affecting its coloration. When the process occurs in a colored cementitious mixture, the contrast is sometimes more marked. For minimizing the mortars‘ porosity, and as results the efflorescence, the use of materials with low level of soluble salts and in the right proportions is recommended [2,3]. On the other hand, independently of the set of mixtures and their interaction with the molds, the color is a sensorial perception that co-exists with three fundamental elements: the object, the illuminant, and the observer [4]. The CIELAB color space proposed by the Commission Internationale de l’Eclairage (CIE) is one of the most used systems to evaluate color. Represented in cylindrical or polar systems, this space is defined by three variables: lightness (L⁄), and two coordinates (a⁄ and b⁄) in the first system, or (L⁄), saturation (C⁄), and hue (h⁄) in the second one. The cylindrical system includes the vertical axis (L⁄), that indicates clarity or darkness, and a horizontal plane defined by the a⁄ and b⁄ axis. The a⁄ axis represents the variation red-green, being positive (+a⁄) for red and negative (a⁄) for green, while the b⁄ axis represents the variation yellow-blue, being positive (+b⁄) for yellow and negative (b⁄) for blue [5,6]. At the same time, the saturation indicates how intense a color is, and the hue (h⁄) is the angle indicating if the color is red (0°), yellow (90°), green (180°) or blue (270°); Fig. 1 shows the location of these variables. C⁄ is calculated as shown in Eq. (1) and, h⁄ as shown in Eq. (2). In addition to the color, this system has allowed determining the choice of cleaning techniques on facades made with rocks [7], observing the evolution of damages caused by fire in concrete [8] and characterizing the mortar color with natural hydraulics limes (NHL) commercial binders for the preservation of historical architecture [9].  2 1=2

C  ¼ ½ða Þ2 þ ðb Þ  

h ¼ arctg

ð1Þ



b a

ð2Þ

One of the main uncertainties arising from the need of colored mortar is associated with color stability. The stability or solidity of the color indicates the resistance to change of L⁄, a⁄ and b⁄ due to the action of the radiation or variations in the climatic conditions [10]. To assess the discoloration, it is appropriate to use a colordifference formula [5,11]. There are advantages when mortars studies are carried out. Evaluations conducted on concrete interlocking blocks with inorganic pigments revealed denser structures when black pigment was used [12]. The flow of mortars with yellow, red, black or green pigments proportions varying from 3% to 12% was also evaluated; the experimental results showed that with yellow or red pigment, the flow decreased rapidly with increasing pigment content. To

avoid this effect it was necessary to increase the amount of mixing water or to use a superplasticizer to obtain the flow (180 mm). When green or black pigment was used, no significant fluidity changes were observed [13]. The so called ‘‘Okamura” method, which starts the analysis with mortars was developed for dosage of self-compacting concrete [14–16]. It assumes the concrete as a material consisting of two phases: the mortar and the coarse aggregate. This method was verified in self-compacting concrete with fibers and high strength [17] and colored self-compacting concrete [18]. The color analysis as a function of time involves relating the L⁄, ⁄ a , and b⁄ values. Therefore, using a parameter such as the total color-difference (DE), where those variables are included may be simpler than making evaluations of each of them. This paper describes the methodology used to evaluate the color stability on mortars exposed to two environments. To avoid ‘‘color” subjectivities, CIELAB color space was used to define color values. Accordingly, the CIEDE1976 (DE⁄76) and CIEDE2000 (DE00) color-differences formulas were used to evaluate the color stability. 2. Materials and methods 2.1. Materials proportions The mixtures were developed using local materials (La Plata, Province of Buenos Aires, Argentina) with the characteristics recommended in the bibliography to be used in Self-Compacting Concretes (SCC) [19]. According to the IRAM 50000 standard, similar amounts of commercial Portland cement with calcareous filler (defined in this paper as G1) and ordinary ‘‘grey” Portland cement (G2) were used. Chemical compositions, physical and mechanical features of cements and filler are detailed in Table 1. The G1 is equivalent to CEM II/AL 42.5N cement and the G2 is equivalent to CEM I 42.5N cement of European standard. The pigment was yellow Meranol S.A.C.I.Ò, density = 3.80 g/cm3. In order to increase the mixtures flowability and reduce the water contents, an ether polycarboxylated (GLENIUM B 255, solid content about 18%) of BASF Argentina S.A.Ò was included as a superplasticizer. The mixtures were prepared using tap water, and a silica natural sand (fineness modulus: 2.39, density 2.63 g/cm3, and absorption rate 0.5%). Table 2 shows the different materials and proportions used for mortars production. The F group was made with G1 cement, while the N group with the G2 one. Both mortar groups contained a superplasticizer and yellow pigment at 2%, 4%, or 6% relationship in cement weight (pigment/cement or (p/c)). To evaluate the possible effect of the pigment increment on the mortars fluidity and viscosity, the pigment was replaced by the same volume of filler (FC). The solid additive incorporated in the F and N groups was expressed as a cement weight relationship of 0.40% and 0.35%, respectively. A water/cement relationship = 0.50 was used in both groups. 2.2. Environments Before and after 18 months of exposure to: 1) a chamber (C1) operating under controlled humidity (55 ± 5%) and temperature (21 ± 2 °C) conditions; or 2) the natural environment (A1) of La Plata station (urban-industrial area), annual average Table 1 Cement and filler. Chemical composition (%)

Cement

CaO SiO2 Al2O3 Fe2O3 SO3 MgO K2O Na2O LOI

Admixture

G1

G2

FC

61.9 20.1 3.3 3.2 2.4 0.6 0.9 0.0 7.0

64.1 21.3 3.7 3.6 2.5 0.7 1.0 0.1 2.2

50.2 9.2 1.1 0.7 0.1 0.3 0.4 0.1 37.8

Physical and mechanical features f0 c Density Specific surface Blaine Fig. 1. CIELAB color space.

LOI: loss on ignition.

1d 28 d (g/cm3) (m2/kg)

(MPa) (MPa)

13.8 52.2 3.09 364

24.0 62.2 3.11 337

– – 2.8 –

619

A. López et al. / Construction and Building Materials 120 (2016) 617–622 Table 2 Material proportion in mortars. Mortars

f/c

p/c

Cement

Filler

Pigment

Sand

448 440 433 426

– 9.9 19.8 29.8

1106

398 390 383 376

– 9.9 19.8 29.8

1157

3

(%)

(kg/m )

F F2 F4 F6

0.90 0.89 0.87 0.86

– 2 4 6

495 (G1)

N N2 N4 N6

0.80 0.78 0.77 0.76

– 2 4 6

495 (G2)

In industrial practice, the small color-difference has shown non-uniform effects on the calculated values. In different ranges and directions; changes in the external conditions can modify the perceived magnitude. As a result, adjustments are needed in order to improve this evaluation through the correction of different effects affecting the color-difference. The Eq. (4) shows the CIEDE2000 formula [25,26].

"


DE00 ¼

f/c: filler/cement ratio.

values of the climatic variables: relative humidity (70%), temperature (16 ± 2 °C), rain (1300 mm), the color of mortar samples was measured. Before their initial measurement and transfer to different exposure conditions, all the samples were kept in the C2 chamber at 23 ± 2 °C and 95 ± 5% for 28 days as suggested by the IRAM 1534 standard [20]. 2.3. Samples preparation The materials used for the mortars were premixed, first manually and then mechanically at two speeds. The mixing sequence had to steps: 1) 1/3 of the additive was diluted in water and 2) the cement and additions were aggregated. These components were mechanically mixed at low speed for 2½ min. During this period, the rest of the additive and the sand were added after 30 s and 1 min, respectively. Finally, the mortar was mixed at high speed for 2 min; the volume of each mortar was 2 L. This procedure agrees with that reported in the literature for colored SCC [21]. As a first approach, the prismatic mortar specimens (40  40  160 mm) were obtained using molds of steel without a release agent. With the aim of preparing all the samples following the same procedure to start each color measurement test, they were exposed to the C1 environment for 24 h. The surface dust was removed to minimize possible interferences in the color measurements. After the initial measurements of color, part of the samples was exposed to the A1 environment while the rest was kept in the C1 environment for 18 months.

DL0 K L SL

2


þ

DC 0 K C SC

2


þ RT

DC 0 K C SC




DH 0 K H SH

#1=2 ð4Þ

where the measured variables L⁄, a⁄, and b⁄ are transformed to L0 , a0 , and b0 to correct the hue angle, chroma and lightness values [6,26]. As well, the formula incorporates specific corrections for no uniformity of CIELAB space such as the so-called weighting functions (SL (Lightness difference), SC (Chroma difference), and SH (Hue difference)) and parameters accounting for the influence of illuminating and viewing conditions in color-difference evaluation (the so-called parametric factors KL, KC, KH). In the present study, the parametric factors were set as 1 because the reference conditions were similar to the usually found in the industrial practice. The inclusion of the rotation term accounts for the interaction between chroma and hue differences in the blue region and a modification of the a⁄ axis of CIELAB, which mainly affects colors with low chroma (neutral colors). In Eqs. (3) and (4), average L⁄, a⁄, and b⁄ values were used. They were calculated from five (5) measurements performed on the sample face exposed to the environment. They are identified in the table as Li, ai and bi. C⁄ and h⁄ are identified as Ci and hi.

3. Results and discussion 3.1. Color stability in mortars Table 3 shows the color parameters values periodically measured on 64 cm2 of each sample before (sub-index 1  age 1) and after (sub-index 2  age 2) 18 months exposure to an industrial (A1) or controlled (C1) environment. The DE76 and DE00 average values calculated by using those parameters are also shown.

Table 3 Color stability in mortars exposed to A1 and C1 environments.

2.4. Color measurements and color stability The color was determined using the CIELAB color space [4,22], from the initial and final L⁄, a⁄, and b⁄ values were obtained using a BYK Gardner spectrophotometer. This instrument has a measuring head containing the integrating sphere combined with a xenon lamp. After the equipment is positioned on the sample to be measured (object) and the flash trigger is activated, a light beam impacts on the sample diffusely. The spectrum reflected by the sample is captured at an angle of 10° and through a fiber glass cable the signal passes through standard color filters (observer) and coded to L⁄, a⁄ and b⁄ values, which are read on the display. Important technical data: the sphere has a diffuse/10° geometry, the standard illuminant is D65, and the white repeatability is 0.10 DE⁄ [23]. To assess the color stability, the total color-difference parameter (DE) was calculated using successive measurements performed at different times. This work compares the results from the CIEDE1976 (DE⁄76) [6,24] and CIEDE2000 (DE00) formulas [6,25,26]. The color-difference DE⁄76 between the a and b points in an object is the Euclidian distance between the color stimulus in both points, and approximately represents the color-difference perceived by the color stimulus in the CIELAB color space. This phenomenon occurs when the objects are seen by an observer adapted to a chromaticity field, which is not much different from the average light in identical surroundings from white to light grey. The Eq. (3) calculates this vector magnitude, and it is specified in the European standard EN 12878 [24].  2 1=2

DE76:ab ¼ ½ðDL Þ þ ðDa Þ2 þ ðDb Þ  2

ð3Þ

where:

Da ¼

ab



 bb

Db ¼

Mor

i

Li

ai

bi

Ci

hi

DE76

DE00

A1

F

1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2

67.2 63.8 65.0 62.5 63.2 61.9 63.0 60.7 66.3 63.5 64.0 61.2 62.5 60.2 62.8 60.4

1.6 1.7 3.3 3.4 4.9 4.3 6.1 6.2 1.5 1.6 6.1 3.6 5.3 5.3 6.8 6.1

7.4 8.3 18.9 19.5 23.4 21.9 28.0 29.4 8.2 9.1 19.8 19.5 25.1 25.1 29.6 27.5

7.6 8.5 19.2 19.8 23.9 22.3 28.6 30.0 8.3 9.2 20.2 19.9 25.7 25.6 30.4 28.1

77.8 78.3 80.0 80.0 78.1 79.0 77.7 78.1 79.6 80.1 79.8 79.5 78.2 78.1 77.1 77.4

3.6

2.92

2.6

2.15

2.1

1.41

2.7

2.03

2.9

2.38

3.8

3.51

2.4

2.05

3.3

2.29

1 2 1 2 1 2 1 2 1 2 1 2 1 2 1 2

73.7 67.8 68.7 67.2 67.8 66.6 68.1 67.3 65.3 64.8 64.7 64.1 64.3 63.8 66.6 65.6

1.0 1.1 2.6 2.8 4.5 4.5 6.1 6.2 0.9 0.9 2.6 2.7 4.0 4.0 5.5 5.7

6.1 8.1 19.3 19.5 26.0 25.6 30.7 30.7 8.4 8.6 19.8 19.6 25.1 24.5 31.1 30.5

6.2 8.2 19.4 19.7 26.4 26.0 31.3 31.3 8.5 5.7 20.0 19.8 25.4 24.8 31.6 31.1

80.4 82.4 82.4 81.8 80.3 80.1 78.8 78.6 83.9 84.1 82.6 82.3 80.9 80.7 79.9 79.4

6.3

4.79

1.5

1.21

1.3

1.04

0.9

0.70

0.6

0.48

0.7

0.56

0.8

0.53

1.2

0.91

F2 F4 F6 N N2 N4 N6 C1

F F2 F4 F6 N

DL ¼ Lb  La 

Env



aa



 ba

From here on, the sub-indexes a and b will be (age 1) and (age 2), according to the comparison of the color measured before and after the exposure, respectively. Also, to dispense with the asterisk in the DE⁄76 nomenclature is shortened to DE76.

N2 N4 N6

Visual assessment of DE76: 0.5–1.5: slight; 1.5–3.0: obvious; 3–6: very obvious. 6– 12: large. Env: environment; Mor: mortar.

620

A. López et al. / Construction and Building Materials 120 (2016) 617–622

Fig. 2. DLi in mortars.

Fig. 3. Dbi in mortars.

Fig. 4. Effect of the pigment content on the Li and Ci parameters in the A1 environment.

With regard to the color stability [22], reported that if DE76 values are >1.5, color-differences in concrete surfaces can be perceived at naked eye. In line with this statement, the DE76 and DE00 values obtained in the present study indicated that color changes were less perceived in the C1 environment (DE76 < 1.5 and DE00 < 1.2) than those obtained in the A1 environment (DE76 > 2.1 and DE00 > 1.4). Such results reveal that in addition to the absence of solar radiation, the exposure under controlled humidity and temperature conditions contributed to the color stability.

The exposure to A1 environment allows stating that the color was slightly involved, since values of DE76 < 3.8 were obtained at a relatively short time. While mortars lightness (Li) reached between 60 and 73 units, and hue (hi) remained near 80° as the yellow pigment amount increased 2%, 4% or 6%, the saturation (Ci) fluctuated from 8 to 20, 25 or 30 units, respectively. As the pigment amount increased, the DE76 levels remained similar in both (A1 and C1) environments, therefore, the initial loss of color is independent of the pigment content. While the DE76 values depended strongly on the group coordinates their differences

A. López et al. / Construction and Building Materials 120 (2016) 617–622

621

Fig. 5. Effect of the pigment content on the Li and Ci parameters in the C1 environment.

Fig. 6. DE between mortars with different pigments content in age 1. A1 left; C1 right.

Fig. 7. DE between mortars with different pigments content in age 2. A1 left; C1 right.

were considered primarily influenced by changes in Li. In the A1 environment such values (DLi) changed between 1.3 and 3.5 units, while in samples exposed to the C1 they changed only one (1) unit. This means that after 18 months of exposure in A1 environment, the samples surface became darker as it is shown by the negative values of differences (DLi), Fig. 2. Moreover, the saturation parameter of F4 samples showed a random behavior since sometimes the Ci value decreased and sometimes it did not. In addition to changes in Li, differences in

the coordinate bi (Dbi) values were also important, and accompanied the Ci changes though to a lesser extent. With regard to this variable, a random behavior in the F group samples exposed to both environments was observed. The N group samples showed a slight decrease in bi (less yellow); Dbi differences are displayed in Fig. 3. On the other hand, there is a pigment content from which there is no reason to use larger quantities because the saturation (i.e. the mixture cannot be more vivid), and the lightness (i.e. the mixture

622

A. López et al. / Construction and Building Materials 120 (2016) 617–622

cannot be clearer or more dark) do not change. Figs. 4 and 5 show the Li and Ci behavior of the F mortars group (F-L1, F-L2, F-C1, F-C2 values) and N (N-L1, N-L2, N-C1, N-C2 values) exposed to A1 or C1 environments, respectively.

(CONICET) of Argentina by the financial support to this research work.

3.2. Comparison of the color-difference values

[1] R.M. Cornell, U. Schwertmann, Introduction, in: G. Buxbaum, G. Pfaff (Eds.), Industrial Inorganic Pigments, WILEY-VCH Verlag GmbH & Co., KgaA, Weinheim, 2006. [2] V.S. Ramachandran, Concrete Admixtures Handbook. Properties, Science, and Technology, Noyes Publications, New Jersey, USA, 1995. [3] E. Püttbach, Pigments for the colouring of concrete-questions of quality, Betonwerk + Fertigteil-Technik/Concr. Precast. Plant Technol. 10 (1992) (Disponible en www.BFT-online.info). [4] R.D. Lozano, El Color Y Su Medición, Américalee S.R.L., Buenos Aires, 1978. [5] R.D. Lozano, A new approach to appearance characterization, Color Res. Appl. 31 (2006) 164–167. [6] CIE 15.3 technical report draft, 3rd ed., Colorimetry, 2004. [7] R.F. Fort, M.C. Mingarro, J. López de Azcona, B. Rodríguez, Chromatic parameters as performance indicator for stones cleaning techniques, Color Res. Appl. 25 (2000) 442–446. [8] E. Annerel, L. Taerwe, Methods to quantify the colour development of concrete exposed to fire, Constr. Build. Mater. 25 (2011) 3989–3997. [9] D. Gulotta, S. Goidanich, C. Tedeschi, T. Nijland, L. Toniolo, Commercial NHLcontaining mortars for the preservation of historical architecture. Part 1: Compositional and mechanical characterisation, Constr. Build. Mater. 38 (2013) 31–42. [10] F. Coelho, Variación Del Color Y Textura Superficial De Hormigones Vistos, Con Adición De Pigmentos Inorgánicos, Sometidos a Distintos Estados De Exposición Ambiental (Tesis doctoral), Universidad Politécnica de Madrid, 2000. [11] J.M. Artigas, P. Capilla, J. Pujol, Tecnología Del Color, Universitat de València, España, 2002. [12] H. Lee, J. Lee, M. Yu, Influence of iron oxide pigments on the properties of concrete interlocking blocks, Cem. Concr. Res. 33 (2003) 1889–1896. [13] H. Lee, J. Lee, M. Yu, Influence of inorganic pigments on the fluidity of cement mortars, Cem. Concr. Res. 35 (2005) 703–710. [14] H. Okamura, Self-compacting high-performance concrete, Concr. Int. 19 (1997) 50–54. [15] H.Y. Okamura, M. Ouchi, Self-compacting concrete, J. Adv. Concr. Technol. 1 (2003) 5–15. [16] K. Ozawa, K. Maekawa, H. Okamura, High performance concrete with high filling capacity, in: E. Vázquez (Ed.), Admixtures for Concrete: Improvement of Properties (Proc. Intern. RILEM Conf.), Chapman and Hall, Londres, 1990, pp. 51–63. [17] J.M. Tobes, Hormigones autocompactante simples y reforzados con fibras: diseño, caracterización y aplicciones (Tesis doctoral), Facultad de Ingeniería, Universidad Nacional de La Plata, 2009. [18] A. López, J.M. Tobes, G. Giaccio, R. Zerbino, Advantages of mortar-based design for coloured self-compacting concrete, Cem. Concr. Compos. 31 (2009) 754– 761. [19] EFNARC, Specification and Guidelines for Self-Compacting Concrete, EFNARC (European Federation of Producers and Applicators of Specialist Products for Structures), 2002. www.efnarc.org. [20] IRAM 1534, Hormigón de Cemento Portland. Preparación y curado de probetas para ensayos en laboratorio, 2004. [21] A. López, Diseño y caracterización del hormigón autocompactante coloreado (Tesis doctoral), Facultad de Ingeniería, Universidad Nacional de La Plata, 2009. [22] G. Teichmann, The use of colorimetric methods in the concrete industry, Betonwerk + Fertigteil-Technik/Concr. Precast. Plant Technol. 10 (1990) 58– 73. [23] BYK Gardner GmbH, Manual of instruments for color measurement, SpectroGuide 45/0 Gloss-Spectro-Guide Gloss, 2009. [24] EN 12878, Pigments for Colouring of Building Materials Based on Cement and/ or Lime – Specification and Methods of Test, 2005. [25] M. Melgosa, R. Huertas, Relative significance of the terms in the CIEDE2000 and CIE94 color-difference formulas, Opt. Soc. Am. 21 (2004) 1–7. [26] G. Sharma, W. Wu, E.N. Dalal, Color-difference formula: implementation notes, supplementary test data, and mathematical observations CIE 2000, Color Res. Appl. 30 (2005) 21–30. [27] E. Valencia, M.S. Millán, Diferencias de color entre dos ejemplares del atlas de color Munsell Colour differences between two individual collections of Munsell colour chart, Óptica Pura y Aplicada 38 (2005) 57–65. [28] M. Melgosa, R. Huertas, J. Romero, Recientes recomendaciones de la Comisión Internacional de Iluminación con respecto a la evaluación industrial de diferencias de color, in: Actas del Séptimo Congreso Argentino del Color, Editorial Nobuko, Buenos Aires, Argentina, 2006, pp. 77–85.

Finally, the comparison among the color-difference values revealed that the DE00 values were lower than the DE76 values. As seen in Table 3, while DE76 < 1.5, differences between DE76 and DE00 ranged from 0.1 to 0.3 units, but if DE76 > 1.5 such differences ranged from 0.3 to 1.5 units. In mortars with different pigment contents, this trend was also observed when the DE evolution for each age in A1 or C1 environments was evaluated. Figs. 6 and 7 display bar graphs of DE vs. age 1 (initial measurement) and age 2 (final measurement), respectively. These differences were even greater (from 1.3 to 4.9 units) when mortars with different pigment contents were compared. For industrial usage, the CIE 31 recommends the CIEDE2000 formula [6,27,28], because it provides a better adjustment to the authenticity of the observed variations. However, in order to obtain a clear tendency in color changes and to know each coordinate influence, it is appropriate to use the DE76, DLi, Dai, Dbi calculus as a last resort. 4. Conclusions This work analyzes the color evolution in mortar samples colored with yellow iron oxide pigment, and exposed to an industrial-urban environments or to another one presenting stable humidity and temperature conditions. In such circumstances, the color was defined according to the CIELAB color space and its evolution characterized by the color-difference CIEDE1976 and CIEDE2000 formulas. The experimental results showed the following: 1) The CIELAB color space allowed an accurate definition of the mortars’ color and also of its evolution evaluation in mixtures made with Portland cement. The color-difference formulas enabled the correct evaluation of the levels at which the color was modified because the safe and accurate measurement of its parameters provided an objective evidence of its behavior. Besides, the analysis of each Li, ai, and bi value allowed detecting the parameter that more influenced for changing the total color-difference value. 2) A very good color stability was obtained when the exposure conditions were stable. 3) Significant color variations took place when the samples were exposed to an urban-industrial environment; and 4) In a short period of exposure the original color is lost but the hue is maintained. In other words, all these results clearly demonstrated that the mortars’ color stability is strongly dependent on the environmental conditions to which they could be exposed. Acknowledgements The authors gratefully acknowledged the Comisión de Investigaciones Científicas de la Provincia de Buenos Aires (CICPBA), and the Consejo Nacional de Investigaciones Científicas y Técnicas

References