Admixtures effect on fresh state properties of aerial lime based mortars

Construction and Building Materials 23 (2009) 1147–1153

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier .com/locate /conbui ldmat

Admixtures effect on fresh state properties of aerial lime based mortars

M.P. Seabra a,b,*, H. Paiva a, J.A. Labrincha b, V.M. Ferreira a

a University of Aveiro/CICECO, Department of Civil Engineering, 3810-193 Aveiro, Portugalb University of Aveiro/CICECO, Department of Ceramics and Glass Engineering, 3810-193 Aveiro, Portugal

a r t i c l e i n f o

Article history:Received 29 January 2008Received in revised form 11 June 2008Accepted 16 June 2008Available online 3 August 2008

Keywords:Aerial limeMortarsAdmixturesRheology

0950-0618/$ - see front matter � 2008 Elsevier Ltd. Adoi:10.1016/j.conbuildmat.2008.06.008

* Corresponding author. Address: University of AvCivil Engineering, 3810-193 Aveiro, Portugal. Tel.:234370094.

E-mail address: [email protected] (M.P. Seabra).

a b s t r a c t

Especially in rehabilitation works, the use of lime based mortars have been increasing due to the need ofcompatibility between the old and the new materials. The mortars fresh state properties are extremelyimportant since determines the material workability and also have a great influence on its hardened statecharacteristics.In this work, the fresh state properties of aerial lime pastes and aerial lime based mortars were investi-gated. The torque variation with time and the mortar rheological parameters (relative yield stress andplastic viscosity) were obtained using a rheometer suitable for mortars. The correlation of rheologicaldata with slump and relative density measurements was studied.The mortars workability is affected by several parameters, namely, the binder/aggregate and water/bin-der ratios, the kneading water content, the admixtures type and amount. The admixtures influence (typeand amount) on the fresh state properties of aerial lime based mortars are discussed. The used admix-tures were the most common ones for several mortars, such as a water-retaining agent, a plasticizerand an air-entraining agent.

� 2008 Elsevier Ltd. All rights reserved.

1. Introduction

Until the beginning of the 20th century, lime played a role ofgreat importance; it was the most used binder in mortars prepara-tion. However, due to the emergence of Portland cement, this typeof mortars fell into disuse. The cement has undeniable qualities,such as compositional constancy and hardening speed. Neverthe-less, when used as binder in mortars for the rehabilitation of oldbuildings also present numerous problems [1–4]. Cement providesmortars with features that makes them incompatible with the oldlime based ones; in particular, high mechanical strength and highmodulus of elasticity, insufficient water vapour permeability andthe presence of alkali hydroxides, which can react with the salinesolutions that penetrate by capillary absorption, causing solublesalts. Due to this incompatibility between the new and the mortarsexisting in the old buildings, lime based mortars became interest-ing for the scientific community.

The mortar fresh state last for a very small period of time whencompared to its lifetime, however, the control of its rheologicalcharacteristics is fundamental since it exerts a great influence onapplication and on ultimate hardened properties. The fresh statecharacteristics define its easiness of launch, spreading, levelling,finishing, and also the grip levels, controlling the application pro-

ll rights reserved.

eiro/CICECO, Department of+351 234370258; fax: +351

ductivity (manual or mechanical). There are several rheologicalstudies of cement based mortars [5–8] but, on the other hand,the rheological characterization of lime based mortars is still veryincipient [9,10]. Seabra et al. [9] studied the influence of severalparameters (binder/aggregate ratio, kneading water content,admixtures type and amount), on the rheological properties offresh hydraulic lime based mortars. The effect of ageing, on rheo-logical properties of lime putty, was investigated by Atzeni et al.[10]. Given the increasing use of this type of mortars, it is funda-mental to broad the current knowledge on their behaviour.

In order to improve or adjust certain mortars features/proper-ties, whether in the fresh or hardened state, it is common to useadmixtures, which in general are organic substances used in smallpercentages (<5%). The most used ones are water-retaining, water-reducing and air-entraining agents [11]. Their use in lime mortarshave been scarcely studied compared to common cement basedmortars.

Water-retaining agents are normally utilized to increase themortar viscosity. In fluid systems this enhancement is essentialto reduce the separation of their constituents, so the risk of mate-rial segregation during transport and handling is minimized, lead-ing to a hardened product having a better homogeneity andperformance. These admixtures are high-molecular weight poly-mers (soluble in water) which, by their hydrophilic nature, physi-cally fix water molecules thus increasing the mortar viscosity andcohesion. Attractive bonds, among adjacent polymer chains, canalso be developed blocking the water motion and increasingfurther the apparent viscosity, particularly for low deformation

Table 1Particle size distribution of the aggregate

Sieve aperture (lm) 630 315 160 80% Passing 99.6 49.4 3.2 0.1

1148 M.P. Seabra et al. / Construction and Building Materials 23 (2009) 1147–1153

rates. These bonds can be easily destroyed by agitation and thepolymer chains tend to align themselves in the flow direction,leading to a viscosity reduction with the increase of agitationvelocity [12].

The mortars flow resistance can be significantly diminished bythe addition of a water-reducing agent or plasticizer since it pro-vides better particle dispersion. These type of admixtures arehydrophilic surfactants which, when dissolved in water, promotethe particles deflocculation and dispersion. This effect is causedby adsorption, on the particles surface, of their high-molecularweight anions, leading to mutual repulsion of individual particlesand to a reduction in the inter-particles friction [11]. Since floccu-lation is prevented, a lower water amount is needed to get thesame workability.

The air-entraining agents are constituted by molecules having ahydrophilic polar group and a non-polar hydrophobic groupformed by a long chain of carbon atoms. In the mortars, as they al-ways contain some air dragged by their constituents, the hydrocar-bon chain is oriented into the air within the bubble. The ionisedpolar group becomes orientated outside the air bubble into theaqueous phase leading to the formation, during agitation, of stablemicro-air bubbles since the water tension is reduced. The chargessurrounding each air bubble provoke their mutual repulsion thusavoiding their coalescence and promoting its dispersion in themortar. Basically, these agents act by physically creating smalland stable micro-air bubbles uniformly distributed in the mortarleading to a wet density diminution and a workability increase[11]. These air bubbles can act simultaneously as a fluid and an in-ert. On one hand, it is possible to diminish the kneading water con-tent without changing the material workability; by other, they mayreplace part of the sand (with dimensions less than 1 or 2 mm),with the advantage of having a better form coefficient, be deform-able, elastic and able to slide without friction, thus contributing toimprove the material workability [13]. In the hardened product,the micro-air bubbles presence can improve the freeze-thaw resis-tance [14].

The aim of this work is to study the fresh state characteristics ofaerial lime based mortars, particularly the influence of the admix-ture type and amount. The obtained results can help the optimiza-tion of formulations in order to ensure best working conditions.

2. Experimental

2.1. Materials and formulations

Samples were prepared with a commercial aerial lime (CL 90 –EN 459-1; Lusical H100) that has an apparent density of 0.46 g/cm3. According to X-ray diffraction (XRD), the major crystallinephase is calcium hydroxide [Ca(OH)2 – ICDD 76-0571] and somecalcite is also present [CaCO3 – ICDD 05-0586]. The used aggregatehas an apparent density of 1.57 g/cm3 and the XRD results revealedthat it is a siliceous sand. According to the sieving test with vibra-tion, the aggregate has a maximum particle size of 630 lm and amedium particle size of around 315 lm (Table 1). The used admix-tures were a water-retaining agent (Hydroxypropylmethyl cellu-lose, HPMC – Walocel), a plasticizer or water-reducing agent(Peramin SMF – Perstorp) and an air-entraining agent (Silipon –Aqualon).

In order to study the rheological behaviour of the pure binderpaste, a mixture of aerial lime with water was prepared. For allthe compositions tap water was always used and its content wereadjusted taking in account the mortar workability.

Mortars have a binder/aggregate volume ratio of 1/1, corre-sponding to a weight ratio of 1/3.4, assuming the components den-sity. The kneading water content employed was the minimumamount necessary to perform the rheological measurements in

the most difficult case (mortar without admixtures). Hence, firstits amount was adjusted for the admixture-free formulation; then,the same content was used for the compositions with admixtures.Each admixture was added in three distinct amounts (0.05, 0.10and 0.15 wt%); the water-retaining agent was also added in a high-er proportion (0.20 wt%).

2.2. Experimental procedure

The mortar preparation followed a precise and reproduciblemixing procedure since its rheological behaviour can be influencedby the shear history. First, all the components were weighted andmanually dry mixed inside a plastic bag; then, they were automat-ically mixed with water during 15 s and, after resting for 60 s, inorder to gather all the material in the middle of the bowl, theywere automatically mixed during more 75 s, always at the lowerspeed (60 rpm). The mixer geometry is in accordance with DIN1164 [5].

Slump measurements were performed according to EN 1015-3European standard. The mortar sample spread diameter measuredbefore and after 15 strokes (1 stroke per second) represents slumpvalues. The fresh state mortars relative density was evaluatedthrough the weight determination of a specific mortar volume.

The rheological behaviour study was done by applying a defor-mation rate imposed by a rotation speed of 160 rpm and measur-ing the resulting torque values, which are related to the shearstress. This rotation speed (160 rpm) was chosen after some preli-minary studies. For higher deformation rates the resulting shearstress is too high and the rheometer has a torque limitation of300 N mm above which it stops. A paddle rheometer (Fig. 1 – Visk-omat NT) was used. The cylindrical sample container rotates andthe mortar, flowing through the paddle blades, generates a torquethat is measured by a transducer and continuously registered asthe sample is subjected to a defined speed programme profile con-trolled by the computer. The computer registers torque (T), speed(N), and time (t). In this work, a speed profile named ‘‘dwell” wasused (Fig. 2). The speed was kept constant at 160 rpm during150 min with periodical decreases to zero rpm (in 30 s) each15 min, in order to build the flow curves. By plotting the torque(T) against (N) it is possible to determine the rheological parame-ters according to the Bingham model expressed as

T ¼ g þ hN ð1Þ

where g (N mm) and h (N mm min) are characteristic constants ofthe material that are directly related with yield stress and plasticviscosity, respectively [15]. Usually [5,6,15–18] the rheologicalbehaviour is discussed in terms of these g and h values instead ofconverting them to yield stress and plastic viscosity. This would im-ply a calibration procedure of the rheometer with a standard fluidwith known parameters. The admixtures effect discussion is alsoperformed taking into account its influence on these rheologicalparameters.

3. Results and discussion

The influence of the most common admixtures in fresh andhardened state properties of cement based mortars have beenwidely studied [5–8,19–21]. On the contrary, the authors havefound no references for aerial lime based mortars.

0 30 60 90 120 1500

40

80

120

160

Spee

d (rp

m)

Time (min.)

Fig. 2. Schematic representation of the velocity profile: ‘‘dwell”.

M.P. Seabra et al. / Construction and Building Materials 23 (2009) 1147–1153 1149

3.1. Aerial lime behaviour

To study the rheological behaviour of the aerial lime paste (Fig. 3),it was necessary to use a large kneading water amount – 135 wt% – be-cause this material is constituted by small particles with a high spe-cific surface area, needing a large water amount to obtain a suitableworkability to perform the rheological measurements. The mixturerelative density is 1.29 g/cm3 and the initial and final slump valuesare 120 and 200 mm, respectively.

The aerial lime suspension thickens with agitation time, goingfrom a fluid to a paste-like state. Two stages are observed: a firstone (i) where torque values are quite low and stable followed by(ii) a notorious increase of the mixture flow resistance (afteraround 90 min of agitation – Fig. 3). When the water is added tothe aerial lime it wraps the particles leading to the formation ofwater films that act as lubricant. However, since aerial lime parti-cles are small, they have a great tendency to form agglomerates so,initially, the water involves, not the individual particles but theagglomerates leading to the formation of thick water films if thewater amount is enough as in the present case. The energy sup-plied by the agitation induces the agglomerates gradual break-down, increasing the area exposed to the water leading to adecrease of the water films thickness. This thickness reductionmay have a strong influence on the mixture flow resistance. Untila ‘‘critical” thickness value, its decrease does not have a stronginfluence on torque values; when that ‘‘critical” value is reached(related to agglomerates breakdown raising the surface area ex-posed to the water), it induces a significant increase of the mixtureflow resistance, since lubrication starts to be less efficient. This isclear in Fig. 3, where until around 90 min of agitation (first stage),the water films thickness is enough to promote an efficient lubrica-tion presenting the mortar low and stable torque values. For longeragitations times (>90 min), the ‘‘critical” value of the water filmsthickness is reached exhibiting the mixture a notorious increaseof the flow resistance. The device automatically stopped(�125 min) when the apparatus torque limit of 300 N mm isattained.

Fig. 1. Rheometer (a) viskomat NT measuring system

The aggregate addition (aerial lime/aggregate volume ratio of 1/1)allow to obtain, using much less water (35 wt%), a mixture havingenough fluidity to perform the rheological measurements. In thebeginning of the rheological measurement (Fig. 3), the higher torquevalues exhibited by the pure binder paste, compared with the mortarones, is related to the aggregate effect on the system morphology butalso to the fact that water/binder ratio is lower for the pure binderpaste (1.35) than for the mortar (1.54).

The mortar torque evolution with agitation time exhibits thesame two stages described above for the pure binder paste. Themain difference is that the mortar first stage is much shorter, beingthe maximum torque value attained also after a smaller agitationtime (after 65 and 125 min, respectively for the mortar and thefor the pure binder paste). Obviously, the presence of the aggregatehas a contribution to the mixture rheological behaviour. On onehand, its presence can help the aerial lime aggregates breakdown,

and (b) paddle used in the experimental work.

0 30 60 90 120 1500

100

200

300 mortar35% H2O

binder paste135% H2OTo

rque

(N m

m)

Time (min.)

Fig. 3. Torque evolution with time of the binder paste (135 wt% of H2O) and of themortar (1/1 aerial lime/aggregate ratio and 35 wt% of H2O).

1150 M.P. Seabra et al. / Construction and Building Materials 23 (2009) 1147–1153

since the structure is not so ‘‘compact”; on the other hand, it canprovoke an increase of the mixture flow resistance because of itshigher friction on sliding.

The plastic viscosity (h) and yield stress (g) evolution with agi-tation time, for the binder paste (135 wt% of H2O) and for the mor-tar (35 wt% of H2O), is presented in Fig. 4. The yield stress variation(Fig. 4b) is very similar to the one observed for the torque (Fig. 3)showing its predominance in the overall mixture behaviour. Theaggregate addition (mortar) induced a decrease of plastic viscosityvalues (Fig. 4a) which is in agreement with the much smaller plas-ticity of the aggregate when compared with the aerial lime plastic-ity. Besides this, as referred above, the water/binder ratio is alsolarger for the mortar.

3.2. Aerial lime based mortars – admixtures effect

Table 2 presents the mortar compositions and some of theirfresh state characteristics, namely relative density and slump val-ues (initial and final).

The water-retaining agent addition, until a certain amount(0.10 wt%), promotes a decrease of slump values; the further

0.00

0.05

0.10

0.15

0.20

binder pastemortar

h (N

mm

min

.)

Time (min.)0 30 60 90 120

a

Fig. 4. Plastic viscosity (a) and yield stress (b) variation with time for th

increase of its amount (0.15 and 0.20 wt%) can lead to a higherslump value (Table 2). Its addition always decreases the relativedensity, being this effect particularly notorious when higheramounts (0.15% and 0.20%) are added. The increase of slump val-ues, observed for these two compositions, probably results fromthe mortar density reduction.

Plasticizer-containing formulations show an increase of the ini-tial and final slump values, which is not very influenced by theadded amount (for the tested quantities – Table 2). Also, the rela-tive density was not appreciably affected by the plasticizer pres-ence or its quantity.

The amount of the air-entraining agent added does not signifi-cantly change the slump values, although, its presence leads to aslight decrease. On the contrary, the relative density becomes low-er when the admixture content increases (Table 2).

3.2.1. Influence of the water-retaining agentRheological studies performed for cement and hydraulic lime

based mortars showed that the use of a water-retaining agent pro-motes a thickening effect related to an increase of yield stress andplastic viscosity values [9,19–21]. Nevertheless, the existence of ayield stress minimum value is reported [9,19,21], being thisamount usually assumed by the industry as optimal, since it facil-itates the mortar application.

Fig. 5 shows the change of torque values with agitation time asa function of the water-retaining agent amount. Initially, its intro-duction leads to an increase of torque, as observed for Portland ce-ment [19–21] and hydraulic lime [9] based mortars. This is theconsequence of free water amount decrease due to the water mol-ecules fixation by the polymeric chains, which is greater for higheradmixture contents. However, for longer agitation times (>30 min)an inversion is observed and formulations containing water-retain-ing agents have lower flow resistances (Fig. 5). In this situation,torque values tend to decrease for higher added amounts. This de-crease, observed after a certain agitation time, can result from twofactors: (i) the introduction of air induced by the admixture, whichcontributes to a better lubrication, and (ii) the alignment of longpolymer chains in the flow direction, improving the flow. As canbe seen in table 2 this admixture induces a decrease of the relativedensity, more notorious for the higher added amounts.

Fig. 6 presents the evolution of plastic viscosity (h) and yieldstress (g) values with agitation time and water-retaining agent

0 30 60 90 1200

75

150

225

g (N

mm

)

Time (min.)

b

e binder paste (135 wt% of H2O) and for the mortar (35 wt% H2O).

Table 2Mortars relative density and slump values of the studied formulations (aerial lime/aggregate ratio 1/1 and 35 wt% H2O)

Composition Slump (mm) Relative density (g/cm3)

Admixture type Weight (%)

Water-retaining agent 0 145–230 1.820.05 105–175 1.800.10 100–155 1.790.15 105–175 1.680.20 145–195 1.68

Plasticizer 0.05 170–250 1.800.10 170–250 1.820.15 170–240 1.82

Air-entraining agent 0.05 120–210 1.770.10 120–210 1.740.15 120–200 1.72

0 30 60 90 120 1500

100

200

300

0.2%

0.15%

0.05%

35 wt.% H2O0%

0.1%

Torq

ue (N

mm

)

Time (min.)

Fig. 5. Influence of the water-retaining agent presence and content (0, 0.05, 0.1,0.15 and 0.2 wt%) on torque variation with time of aerial lime based mortars.

0 30 60 90 120 1500

100

200

0.0

0.1

0.2

g (N

mm

)

Time (min.)

0% 0.05% 0.10% 0.15% 0.20%

h (N

mm

min

.)

a

b

Fig. 6. Plastic viscosity (a) and yield stress (b) variation with agitation time formortars having different water-retaining agent contents (0, 0.05, 0.1, 0.15 and0.2 wt%).

M.P. Seabra et al. / Construction and Building Materials 23 (2009) 1147–1153 1151

content. The use of this admixture reduces the plastic viscosity val-ues (Fig. 6a). The yield stress (Fig. 6b) varies in a similar manner tothe torque values, confirming its predominance in the overall rhe-ological behaviour. The presence of this admixture has a strongerinfluence on yield stress values then in plastic viscosity ones. Itsintroduction in the mortar formulations leads to a notorious delayin the g increase with agitation time (Fig. 6b), which is as longer ashigher is the admixture content. This behaviour is related to thefactors explained above for torque variation with time and admix-ture content.

3.2.2. Influence of the plasticizerAccording to the literature, the addition of a plasticizer to Port-

land cement [5,6] or to hydraulic lime [9] based mortars decreasestheir flow resistance due to particles deflocculation and dispersion;it also promotes the release of the water retained inside the flakes.Its influence is particularly notorious in the yield stress valueswhich exhibits a sharp drop off; neither its presence nor its amountsignificantly affects the plastic viscosity.

Fig. 7 shows the change of torque values with agitation time asa function of the plasticizer amount (0.05, 0.1 and 0.15 wt%). Tor-que values strongly decrease by adding this admixture, being thisreduction very dependent of the admixture content. The formula-tion having 0.15 wt% exhibits low and stable torque values duringall the rheological measurement (150 min). The adsorption of the

hydrophobic part of the plasticizer molecules on the particles sur-face promotes its deflocculation and dispersion. Their presence inthe particle surface leads to less bonded water amount at thatsame surfaces, leaving more free water in the surrounding films.That is the reason why the use of this admixture allows to decreasethe kneading water amount, keeping the workability withinacceptable limits. This can improve the mechanical strength ofthe hardened product.

Fig. 8 shows the variation of plastic viscosity (h) and yield stress(g) with agitation time and added amount of plasticizer. The plasticviscosity is not strongly affected by the admixture addition andjust a slight decrease is noticed (Fig. 8a). On the contrary, yieldstress values deeply decrease when plasticizer is added, being thiseffect more severe for higher admixture additions and for longeragitation time. Also in this case the evolution of the yield stresswith time and admixture relative amount (Fig. 8b) is very similarto the one observed for torque values (Fig. 7). The notorious in-crease of g values is observed after longer agitation times. Theaddition of only 0.05 wt% of plasticizer delayed this increase inabout 15 min; adding 0.15 wt% the delay is about four times bigger(60 min).

3.2.3. Influence of the air-entraining agentAir-entraining agents act physically by incorporating air

micropores uniformly distributed in the mortar, which lead to adecrease of wet mortar density and a better workability. The in-cluded air leads to better insulation against cold and heat, but alsoto a lower strength of the hardened product [22].

0 30 60 90 120 1500

100

200

300

0.15%

0.05%

35 % H2O 0%

0.1%

Torq

ue (N

mm

)

Time (min.)

Fig. 7. Influence of the plasticizer presence and content (0, 0.05, 0.1 and 0.15 wt%)on torque variation with time of aerial lime based mortars.

0 30 60 90 1200

100

200

300

0.15%0.1%

35 % H2O0%

0.05%Torq

ue (N

mm

)

Time (min.)

Fig. 9. Influence of the air-entraining agent presence and content (0, 0.05, 0.1 and0.15 wt%) on torque variation with time of aerial lime based mortars.

1152 M.P. Seabra et al. / Construction and Building Materials 23 (2009) 1147–1153

The plastic viscosity values of Portland cement based mortarsare notoriously affected by the presence of an air-entraining agent[5,6], on the contrary, the yield stress values are not. The additionof an air-entraining agent to hydraulic lime based mortars mainlyinfluences the yield stress values [9].

For aerial lime based mortars its effect is very dependent on theadmixture amount (Fig. 9). By adding 0.05 wt%, the torque evolu-tion is very similar to the one shown by the admixture-free mortar.

0.0

0.1

0.2

0 30 60 90 120 1500

100

200

0% 0.05% 0.10% 0.15%

h (N

mm

min

.)g

(N m

m)

Time (min.)

a

b

Fig. 8. Plastic viscosity (a) and yield stress (b) variation with agitation time formortars having different plasticizer contents (0, 0.05, 0.1 and 0.15 wt%).

By contrast, the use of higher amounts (0.1 and 0.15 wt%) consid-erably affects the rheological behaviour. In these formulationsthe amount of micro-air bubbles introduced by the admixture issufficient to induce a significant decrease of torque values, allow-ing rheological determinations for longer times (Fig. 9). The mortarflow resistance increases with agitation time but at a much slowerrate than observed for the admixture-free and for the 0.05 wt%admixture-containing formulations.

0.0

0.1

0.2

0 30 60 900

100

200

0% 0.05% 0.10% 0.15%

h (N

mm

min

.)g

(N m

m)

Time (min.)

a

b

Fig. 10. Plastic viscosity (a) and yield stress (b) variation with agitation time formortars having different air-entraining agent contents (0, 0.05, 0.1 and 0.15 wt%).

M.P. Seabra et al. / Construction and Building Materials 23 (2009) 1147–1153 1153

Fig. 10 shows the change of plastic viscosity and yield stress val-ues as a function of agitation time and air-entraining agent con-tent. In terms of plastic viscosity (Fig. 10a), compared with theother admixtures, it increases earlier with agitation time. This isprobably related to the fact that this admixture does not interactwith the kneading water in the same way as the other two, whichare water-reducing agents. As observed with the other testedadmixtures, the evolution of yield stress values (Fig. 10b) followsthe change of torque values (Fig. 9), being strongly influenced bythe presence and quantity of air-entraining admixture.

4. Conclusions

Aerial lime pure pastes show a thickening behaviour, goingfrom a typical fluid to a paste-like state. First, the water wrapsup the agglomerates of aerial lime particles leading to the forma-tion of thick water films which are responsible for the mixturelubrication. Their breakdown, due to the agitation, leads to an in-crease of the specific surface area and a decrease of water filmsthickness. Lubrication becomes insufficient and paste flow resis-tance increases (paste-like state). The aerial lime based mortarwithout admixtures has a similar rheological behaviour than thelime pastes.

The use of admixtures (water-retaining agent, plasticizer andair-entraining agent) considerably changes the rheological behav-iour of these mortars.

The water-retaining agent initially exerts a thickening effect fol-lowed, after some agitation time, by a thinning one due to air-entraining in the mortar.

The use of a plasticizer, even in small amounts, diminishes tor-que values due to increase of free water in the system.

The plasticizer and the water-retaining agent also contribute toreducing the water needed to achieve the same workability.

The air-entraining agent addition can reduce the mortar flowresistance, being its effect very dependent of the added amount,since it is directly related to the air bubbles introduced in themortars.

In terms of future work, it is expected to study the influence ofthe admixtures type and amount on the hardened state final prop-erties, namely, mechanical properties, capillary absorption coeffi-cient, liquid water and water vapour permeability, in order tocorrelate the properties in the fresh and hardened states.

Acknowledgements

The authors acknowledge the Foundation for Science and Tech-nology (FCT-Portugal) for the financial support (grant SFRH/BPD/

19482/2004) and also to Weber Cimenfix industry for raw materi-als supply and helpful discussions.

References

[1] Degryse P, Elsen J, Waelkens M. Study of ancient mortars from Salassos(Turkey) in view of their conservation. Cem Concr Res 2002;32(9):1457–563.

[2] Moropoulou A, Polikreti K, Ruf V, et al. San Francisco Monastery, Quito,Equador: characterisation of building materials, damage assessment andconservation considerations. J Cult Herit 2003;4(2):101–8.

[3] Callebaut K, Elsen J, Van Balen K, et al. Nineteenth century hydraulicrestoration mortars in the Saint Michael’s Church (Leuven, Belgium) naturalhydraulic lime or cement? Cem Concr Res 2001;31(3):397–403.

[4] Moropoulou A, Cakmak AS, Biscontin G, et al. Advanced Byzantine cement-based composites resisting earthquake stresses: the crushed brick/limemortars of Justinian’s Hagia Sophia. Constr Build Mater 2002;16(7):543–52.

[5] Banfill PFG. The rheology of fresh mortar. Mag Concr Res 1991;154(43):13–21.[6] Banfill PFG. Rheological methods for assessing the flow properties of mortar

and related materials. Constr Build Mater 1994;8(1):43–50.[7] Park CK, Noh MH, Park TH. Rheological properties of cementitious materials

containing mineral admixtures. Cem Concr Res 2005;35(5):842–9.[8] Banfill PFG, Starrs G, Derruau G, et al. Rheology of low carbon fibre content

reinforced cement mortar. Cem Concr Comp 2006;28(9):773–80.[9] Seabra MP, Labrincha JL, Ferreira VM. Rheological behaviour of hydraulic lime

based mortars. J Eur Ceram Soc 2007;27(2–3):1735–41.[10] Atzeni C, Farci A, Floris D, et al. Effect of aging on rheological properties of lime

putty. J Am Ceram Soc 2004;87(9):1764–6.[11] Lea FM. The chemistry of cement and concrete. 4th ed. New York: Chemical

Publishing Co.; 2001.[12] Saric-Coric M, Khayat KH, Tagnit-Hamou A. Performance characteristics of

cement grouts made with various combinations of high-range water reducerand cellulose-based viscosity modifier. Cem Concr Res 2003;33(12):1999–2008.

[13] Rixom R, Mailvaganam N. Chemical admixtures for concrete. London: E&FNSpon; 1999.

[14] Chatterji S. Freezing of air-entrained cement-based materials and specificactions of air-entraining agents. Cem Concr Comp 2003;25(7):759–65.

[15] Banfill PFG. Use of the viscorder to study the rheology of fresh mortar. MagConcr Res 1990;153(42):213–21.

[16] Golaszewski J, Szwabowski J. Influence of superplasticizers on rheologicalbehavior of fresh cement mortars. Cem Concr Res 2004;34(2):235–48.

[17] Banfill PFG. Influence of fine materials in sand on the rheology of fresh mortar.In: Proceedings of the 5th international masonry conference; 1998. p.119–24..

[18] Banfill PFG. Applications of rheology in mortar production. Proc Brit MasonrySoc 1995;7:7–12.

[19] Lachemi M, Hossain KMA, Nkinamubanzi PC, et al. Performance of newviscosity modifying admixtures in enhancing the rheological properties ofcement past. Cem Concr Res 2004;34(2):185–93.

[20] Khayat KH. Viscosity-enhancing admixtures for cement-based materials – anoverview. Cem Concr Comp 1998;20(2–3):171–88.

[21] Paiva H, Silva LM, Labrincha JA, et al. Effects of a water-retaining agent on therheological behaviour of a single-coated render mortar. Cem Concr Res2006;36(7):1257–62.

[22] Bayer R, Lutz H. Dry mortars. Ullmann’s Encycl Indust Chem 2002;9:1–22.