at SciVerse ScienceDirect

Applied Thermal Engineering 50 (2013) 159e167

Contents lists available

Applied Thermal Engineering

journal homepage: www.elsevier .com/locate/apthermeng

Study of different grouting materials used in vertical geothermalclosed-loop heat exchangers

Roque Borinaga-Treviño a,*, Pablo Pascual-Muñoz a,1, Daniel Castro-Fresno a,1, Juan José Del Coz-Díaz b,2

aDepartment of Transports, Projects and Processes Technology, University of Cantabria (UC), Avda. de Los Castros s/n, 39005 Santander, Cantabria, SpainbDepartment of Construction and Engineering of Production, University of Oviedo, Edif. Dep. Viesques no7, 33204 Gijón, Asturias, Spain

h i g h l i g h t s

< All the mix proportions satisfied consistency and strength requirements imposed.< All aggregates used improve the neat cement Effective Thermal Conductivity (ETC).< Mix aggregate proportion increase improved the ETC for all but the CDW sand.

a r t i c l e i n f o

Article history:Received 5 March 2012Accepted 20 May 2012Available online 29 May 2012

Keywords:GeothermalMortarGroutConductivityRecycled material

* Corresponding author. Tel.: þ34 942 203 943; faxE-mail addresses: roque.borinaga@unican.es,

(R. Borinaga-Treviño), pablo.pascualm@unican.es (Punican.es (D. Castro-Fresno), juanjo@constru.uniovi.e

URL: http://www.giteco.unican.es/1 Tel.: þ34 942 203 943; fax: þ34 942 201 703.2 Tel.: þ34 985 182 042.

1359-4311/$ e see front matter � 2012 Elsevier Ltd.doi:10.1016/j.applthermaleng.2012.05.029

a b s t r a c t

As a renewable source, low-enthalpy geothermal energy is becoming more relevant in heating andcooling buildings, by using an adapted heat pump to exchange thermal energy with the ground. Verticalclosed-loop geothermal systems exchange heat with the ground through a closed buried pipe system,sealed with a grouting material that ensures the stability and thermal transmission of the borehole.Different grouting materials have been tested recently, which use cement or bentonite as a base material.However, the use of recycled materials, which might contribute to the sustainability of the project, hasnot yet been studied. This paper analyzes the use of different natural and recycled aggregates as mainconstituents in cement-based mortars. Results show that all mixes fulfill the minimum consistency andstrength requirements. The use of any of the aggregates proposed improves the thermal conductivitycompared to the cement mortar on its own, independently of the proportion used. Limestone sand, silicasand and electric arc furnace slag enhance the thermal conductivity of the grout as its proportion of useincreases. However, no satisfactory results have been obtained for Construction and Demolition Waste-based mixes because of their high water requirement.

� 2012 Elsevier Ltd. All rights reserved.

1. Introduction

Due to the increasing fossil fuel prices and global warmingeffect, some leading countries have started to search for newrenewable energy sources. According to Omer [1], 40% of theworldwide energy is consumed in lighting, heating and coolingbuildings. Most heating and cooling systems use fossil fuels or air-based heat pumps, which have low efficiencies and emit CO2 andNOx, thus contributing to the global warming process. Geothermal

: þ34 942 201 703.roque.borinaga@gmail.com

. Pascual-Muñoz), castrod@s (J.J. Del Coz-Díaz).

All rights reserved.

energy has proven to be a clean, efficient source, which has,moreover been declared renewable by the European Union 2009/28/CE directive [2]. A geothermal heat pump system takes advan-tage of the year-round constant ground temperature to obtainhigher efficiencies than conventional air to air or air to water heatpumps. Instead of using ambient air as a heat source or sink,geothermal pumps use the ground as a heat exchanger.

According to Florides [3], ground heat exchangers are dividedinto two main groups, open and closed geothermal systems. Opengeothermal systems use the existing groundwater directly toexchange heat with the ground, extracting it from a shaft orinjecting it to the ground. Closed geothermal systems use a heatcarrying fluid which flows through a buried pipe circuit andexchanges heat indirectly with the ground. There are two mainadvantages that made closed geothermal systems more interestingthan open ones. On the one hand, they do not need to have anaquifer to be operational, widening their field of application. On the

Table 1Specific gravity and water absorption (EN 1097-6) and grain-size distribution of theaggregates (EN 933-1).

Aggregate Specificgravity

Waterabsorption

Sieve size (mm)

4 2 1 0.5 0.25 0.125 0.063

Passing percentage

L 2.71 0.52 100 99 61 40 28 21 16S 2.65 0.16 100 100 80 67 47 26 17EAF 3.82 1.83 100 100 39 11 5 3 2CDW 2.57 5.07 100 100 73 48 29 19 12F 2.753 N/A 100 100 100 100 97 89 75

R. Borinaga-Treviño et al. / Applied Thermal Engineering 50 (2013) 159e167160

other hand, evenwhen an aquifer is available their influence on thegroundwater table is smaller than that of open systems, except forlarge geothermal borefields, where thermal affection must befurther evaluated.

At the same time, closed systems are classified in two maingroups, depending on the geometry of the resulting heat exchanger.Horizontal closed-loop systems, which are buried at depths up to5 m, are affected by exterior climate conditions. Because of the lowdepth needed, horizontal systems are easy to install and hence,economical when a land plot is available. However, this is notapplicable in many places where land is scarce. Vertical closed-loopsystems, also called Vertical Ground Loop Heat Exchangers (VGLHE)are a possible option when there is little available land. A closedpipe circuit is introduced into a vertical borehole reaching depths ofup to 200 m. The stability of the borehole walls is ensured byinjecting a grouting material, but at the same time, this materialmust present good thermal properties to transmit heat from thepipes to the ground or vice versa.

Philippacopoulos, Allan et al. [4,5] simulated the temperature ofthe grouting in a simple U shape VGLHE, evaluating the influence ofthe grouting, pipe placement and bonding strength on the groutingtemperature distribution. They concluded that bonding qualitybetween pipe and grout is more significant than that between groutand the existing ground. They also added that gaps diminish theheat flux in the pipe surroundings, decreasing the efficiency of thesystem, thus grouting with high shrinkage when drying should beavoided. Influence of the grout to formation thermal conductivityratio in heat transfer is also analyzed by Smith and Perry [6],concluding that thermal conductivity of the grout should be equalto or greater than the formation’s not to decrease the system effi-ciency. Finally, Chulho Lee et al. [7] proposed a thermal conduc-tivity range of 1.7e2.1 W/(m K) for grouts combined with mostexisting ground types.

According to Allan and Philippacopoulos [5], cement alone andbentonitemixes, with a thermal conductivity of 0.80e0.87W/(m K)and 0.75e0.80 W/(m K), respectively, were predominantly used inthe United States until 2000. However, these values were signifi-cantly reduced by the loss of water, due to the high porosity of thegrouts. Bentonite-based grouts have been improved significantly inthe last decade. Smith and Perry [6] tested the performance ofbentonite-based grouts in a real installation, comparing standardbentonite grouts to the enhanced bentoniteecement sand mix.Carlson [8] compared the performance of the enhanced bentonitegrout proposed by Smith with standard bentonite groutsconcluding that the total borehole length required could bereduced by 10% in the former case. Chulho Lee et al. [7] improvedthe thermal conductivity of the grout either by using silica sand orgraphite. They demonstrated greater improvement when graphitewas used, obtaining up to 3.5 W/(m K) of thermal conductivity forthe resulting grouting materials. Recently, Delaleux et al. [9] usedCompressed Expanded Natural Graphite to enhance bentonitegrouts, obtaining a thermal conductivity of 5 W/(m K), hence thisgrout could be used in most existing ground types. However,Delaleux et al. [9] also warned that for each 10% reduction in watercontent by weight there is a reduction in the thermal conductivityof the grout by 1W/(m K). In dry state, graphite enhanced bentonitegrouts have values ranging from 1.5 to 2 W/(m K), due to thevolume shrinkage of the mix.

The grouting materials studied in this paper are also intended toprovide a preliminary analysis of the grouting materials used forthermally active foundations, which require a compressive strengthsimilar to that of the usual structural concrete. As bentonite-basedgrouts do not fulfill this minimum strength requirement, cement-based grouting materials are proposed instead. Geothermalmortars have not been fully investigated in the last decade. Allan

and Philippacopoulos [10] presented a complete characterization ofdifferent cement-sand based mortars, studying, the influence ofsuperplasticizer and the aggregate type on the rheologic, mechanic,hydraulic and thermal properties. As a consequence, Allan andPhilippacopoulos [5,10,11] presented a superplasticized cement-sand mortar to be used as grouting material, obtaining a thermalconductivity of 2.42 W/(m K) which only reduced to 2.16 W/(m K)when oven dried. However, all the proposed aggregates were silica-sands or quartzite sands. Therefore, the final objective of this paperis to analyze the possibility and/or convenience of using differentaggregates as the main material in geothermal grouting materials.

2. Experimental methodology

2.1. Materials

All the mortars designed are made of water (w), cement (c),superplasticizer (sp) and different types of aggregates (s). CEM II-B(V)/32.5R type cement has been chosen in agreement with EN 197-1 [12]. This cement is a mix of Portland cement and up to 35% ofsiliceous fly ash by weight, valid for the use in foundations. Itscompressive strength is greater than 10 and 32.5 MPa at an age of 2and 28 days, respectively.

Silica (S), Limestone (L), Electric Arc Furnace slag (EAF) andConstruction and DemolitionWaste (CDW) sands have been chosenas basic aggregates. Silica and limestone sands are common naturalaggregates in Spain. EAF slag is a by-product of the steel-makingprocess of Global Steel Wire in Santander. Its use as coarse andfine aggregate of concrete has been successfully demonstrated bymany authors [13e15]. However, the CDW aggregate used in thisresearch does not have any application in Spain at this time. Theoriginal CDW comes from a crushed-concrete recycled aggregateplant. The final commercial product is a 0e60 mm recycled coarseaggregate, obtained after a crushing and selection process. The firststep of this process consists of cleaning the raw material, thuseliminating all the surface imperfections before crushing it. Asa result, a 0e6 mm sized recycled-sand aggregate is obtained,which has until now been discarded because the current Spanishstructural concrete standard (EHE08) [16] does not allow the use ofCDW recycled aggregates with a size of less than 4 mm. In thisresearch, maximum grain size is limited to 2 mm in all the aggre-gates to ensure the pumpability of the mix, according to Allan [11].

Grain-size distribution of the aggregates by weight was deter-mined according to EN 933-1 (Table 1) [18]. Both EAF slag and CDWlack fines, hence limestone filler (F) was proposed to correct theirsize distribution. Apparent particle density (rs) and water absorp-tion was also obtained according to EN 1097-6 [17], showingsignificant density differences between aggregates. As a conse-quence, volume-weighted grain-size distribution was evaluated todetermine the replacement factor of EAF and CDW with limestonefiller, resulting in 25% and 10% replacement factor by weight,

Table 2Size distribution by volume of the aggregates studied.

Aggregate Sieve size (mm)

4 2 1 0.5 0.25 0.125 0.063

Passing percentage by volume

L 100 99 61 40 28 20 16S 100 100 79 66 47 27 18EAF (75%)/F (25%)a 100 100 59 39 34 30 25CDW (90%)/F (10%)a 100 100 76 53 36 25 18

a Percentage of the aggregate used by total weight of the resulting aggregate.

R. Borinaga-Treviño et al. / Applied Thermal Engineering 50 (2013) 159e167 161

respectively (Table 2). While the EAF-based aggregate has similarsize distribution to the limestone sand, the CDW-based aggregatelies between the silica and limestone values.

The superplasticizer (sp) used is a Melment F10 provided by themanufacturer, utilized to reduce the water to cement ratio, toimprove the pumpability of the grout and to prevent segregation.The dosage proposed by the manufacturer is between 0.5 and 2% byunit weight of cement. Allan [19] observed a significant improve-ment of the rheology of the cement grouts with the increase of thesp dosage. Therefore, the maximum dosage recommended by themanufacturer was used for all the mixes.

2.2. Fresh mortar characterization

Mortars are the result of mixing cement, aggregates, waterand additives, creating a viscous mix that hardens after a vari-able time. According to Gustafson [20], when mortar is in itsfresh state, it displays a viscous Bingham fluid behavior thatpermits its injection through a pipe to the required distance ordepth. The flow table consistency test (EN 1015-3) [21] is usedto determine the water to cement ratio of the mix proportionsto meet the fluid consistency of 260e320 mm imposed on thegrout, close to the self cast mortars. Fig. 1 shows the flow tableand the mold used, as well as the resulting diameter observed inone of the tests.

To complete the characterization of the fresh mortar, bulkdensity (rm-f) is also determined according to EN 1015-6 [22]. Selfcompacting ability was confirmed by filling the test recipients,since no compaction energy was required for this. All the mixproportions that lead to segregation after a visual inspection of theconsistency test have been directly discarded.

Fig. 1. Flow table and mold used (a)

2.3. Hardened mortar characterization

3 specimens were made for each of the following tests. For 24 hafter making the mix, specimens were cured under ambient labo-ratory conditions, until molds were removed. From then, until thespecified curing age is reached for each of the test specimens, theywere submerged in water at 20 �C to simulate curing in water-saturated ground. The container was, at the same time, ina temperature and moisture controlled chamber at 20 �C and 95%,respectively. Dry bulk density (rm-h) of the hardened mortar isdetermined at an age of 28 days according to EN 1015-10 [23].Based on the results obtained for the determination of (rm-h),water-accessible porosity (n) is calculated (Equation (1)). msat andmdry are the water-saturated and oven dried mass values, respec-tively. Specimen volume (Vmold) is calculated based on the mold’svolume, 256$10�6 m3.

n ¼�msat �mdry

�.ðrw VmoldÞ (1)

Compressive strength of the hardened mortar is determined atan age of 7 and 28 days according to the standard EN 1015-11 [24].Nominimum strength is required for the grouts used in geothermalvertical heat exchangers, unless the grout is proposed for thermallyactive piles. If the grout is considered as a non-structural concrete,a minimum compressive strength of 15 MPa is recommended bythe Spanish EHE08 norm [16].

Finally, thermal conductivity of the hardened mortar (lm-h) isdetermined at an age of 14 days with the TCi C-Therm thermalconductivity tester, as seen in Fig. 2. Transient plane Source (TPS)method has been used to determine the thermal conductivity ofmortars since no specific standard has been found for this. TPS wasdeveloped by Gustafsson [25] for the determination of thermalproperties of homogeneous solid materials. It has been successfullyused to determine thermal properties of highly porous buildingmaterials [26], insulation building materials [27] and hydratingcement pastes [28]. For carrying out the test, a 75mm diameter and80 mm high cylindrical specimen was chosen. To improve thethermal contact between the sensor and the specimen, silicon oilhas been used since the software of the equipment has alreadycalibrated its influence on themeasured thermal conductivity. Priorto the realization of the test, specimens were oven dried at 105 �Cfor 48 h and left at 20 �C, the ambient temperature of the labora-tory, for at least 7 days to reach thermal equilibrium. Final thermalconductivity is calculated as the mean of the 9 tests carried out (as

. Result of one of the mixes (b).

Fig. 2. Thermal conductivity test.

R. Borinaga-Treviño et al. / Applied Thermal Engineering 50 (2013) 159e167162

test is non-destructive, 3 non-consecutive test were performed perspecimen, resulting in 9 test result per mix proportion).

2.4. Mix proportions

Mix proportions have been defined by the proportion of thedifferent materials by unit weight of cement used. Aggregate to

Fig. 3. Laboratory mixers: (a) for mort

cement ratios (s/c) of 1, 2 and 3 were set for each of the previouslymentioned aggregate types. sp proportion (sp/c)was fixed to 0.02 forall the batches. Neat cementwith awater to cement (w/c) ratio of 0.3and sp/c of 0.02 is used as referencematerial. Finally, water to cementratio is determined by imposing a 260e320 mm slump diameter inthe flow table test (EN 1015-3) [21], to obtain similar rheologicalproperties for all the mortars. For that purpose, a laboratory mortarmixer with a 4 l capacity was utilized. For making the final validmixture, a horizontal planetarymixerwas used, with a 150 l capacity.Bothmixers are shown in Fig. 3. The final mix proportions expressedby unit weight of cement are shown in Table 3.

3. Results and discussion

3.1. Fresh state properties

Fig. 4 shows fresh mortar bulk densities for different aggregatesand proportions, respectively. All the aggregates, except the CDWone, display higher densities as their proportion of use increases,denoting the influence of the apparent particle density of theaggregates. The decrease observed in the CDW mix is because thehigh water absorption of the aggregate, 5.07%, significantlyincreases thewater to cement proportion required. The influence ofthe aggregate in the fresh bulk density is directly dependent on itstype and proportion of use. L aggregate presents similar values forthe s/c¼ 1 and 2mix proportions, attributed to thew/c ratios of themixes (s/c ¼ 1 mix has a w/c ratio of 0.34 and while s/c ¼ 2 mix’s is0.5). As a consequence, the higher density of the aggregate for the s/c¼ 2mix is compensated by the higher density of the cement pasteof the s/c ¼ 1 mix. For all the cases evaluated, fresh mortars arefound to be denser than the bentoniteewater mix present in mostof the boreholes 1050 kg/m3 [29]. Therefore, it is not expected tohave any difficulties in pumping the bentoniteewater mix usedduring the drilling process out of the borehole.

3.2. Hardened state properties

Compressive strength at 7 and 28 days is shown in Fig. 5.Compressive strength decreases as aggregate to cement ratio

ar (b) horizontal planetary mixer.

Table 3Mix design proportions calculated for the fixed flow table consistency.

Mix s1 s2 s/c s1/c s2/c sp/c w/c

1 0 0 0 0 0 0.02 0.302 L 0 1 1 0 0.02 0.303 L 0 2 2 0 0.02 0.374 L 0 3 3 0 0.02 0.435 EAF F 1 0.75 0.25 0.02 0.306 EAF F 2 1.5 0.5 0.02 0.407 EAF F 3 2.25 0.75 0.02 0.498 CDW F 1 0.9 0.1 0.02 0.409 CDW F 2 1.8 0.2 0.02 0.6210 CDW F 3 2.7 0.3 0.02 0.9311 S 0 1 1 0 0.02 0.3412 S 0 2 2 0 0.02 0.5013 S 0 3 3 0 0.02 0.54

R. Borinaga-Treviño et al. / Applied Thermal Engineering 50 (2013) 159e167 163

increases for all the aggregates evaluated. All the observed valuesare above the minimum of 15 MPa required by the EHE08 norm[16]. Therefore, all the mixes evaluated could be used for usualgeothermal heat exchangers taking into consideration this crite-rion. However, the resistance value of CDW mixes decreasessharply compared to the rest of the aggregates, indicating badbonding strength in the cement paste because of its high waterabsorption. Results also show that the improvement in compressivestrength at 28 days over the results observed for 7 days of curing isinsignificant. Hence, it is sufficient to determine compressivestrength at 7 days of curing for design purposes, shortening thetesting time. L, S and EAF-based grouting materials display suffi-cient compressive strength to be used in thermally activatedfoundations, but further investigation is necessary to determinetheir validity.

Once mortar is hardened, part of the water used in the mixnecessary to meet the required consistency is lost, affecting the drybulk density and the resulting mortar porosity. A higher water tocement ratio implies an excess of water that increases porosity andhence decreases dry bulk density. This effect is also present in thiscontribution, as is shown in Fig. 6. Unlike the fresh mortar bulkdensity, dry bulk density has higher values for s/c ¼ 2 than fors/c ¼ 3, indicating a greater loss of water in the latter. For the samereason, a minimum porosity value is obtained at s/c ¼ 2 for all theaggregates analyzed. In the CDW mortars, a special increase of theaccessible porosity is observed, due to the high water to cementratio required by the mix.

Thermal conductivity is the main parameter that defined thethermal behavior of the grouting material. Depending on the

Fig. 4. Fresh mortar densities for diffe

ground conditions at installation, grout can bewater saturated or indry state. Since water is a better thermal conductor than air, drystate thermal conductivity should be considered as a moreconservative value. First of all, influence of aggregate type andproportion of use on the resulting thermal conductivity of the groutare evaluated, as shown in Fig. 7.

Mean value of the thermal conductivity for s/c ¼ 1, 2 and 3 iscalculated for each aggregate type, showing a variationwith respectto the s/c ¼ 1 material of 12.8%, 6.6%, 20.1% and 14.5% is observedbecause of its proportion of use. However, an increment of up to202%, 234%, 150% and 142% is achieved due to the use of L, S, EAFand CDW aggregates, respectively, when compared to the purecement reference material, 0.9 W/(m K). Therefore, the addition ofany aggregate to the grout significantly improves thermalconductivity. Moreover, the type of aggregate used causes a greaterinfluence on the thermal conductivity than its proportion, for theproposed mix proportions.

Mortar can be considered as a composite material whichconsists of aggregate solid particles, grout and air-filled or water-filled voids. Although this relationship is unknown, it can besupposed that when specimens are oven dried, final thermalconductivity of the grout (lm-h) would depend on the thermalconductivity of the air (la), the c-sp-w hardened paste (lp) andthermal conductivity of aggregate particles (ls) as well as on thevolumetric proportion of each of them (Va, Vp and Vs). This volu-metric parameters are explained in the Appendix A. The thermalconductivity of air of 0.026 W/(m K) at 300 K is obtained from theexisting bibliography [30], significantly lower than would be ex-pected for either the solid or the c-sp-w paste. Effective thermalconductivity of the c-sp-w paste is unknown. However, accordingto Allan [10], thermal conductivity of the pure cement groutdecreases as the water to cement ratio increases.

Finally, thermal conductivity of the aggregate particles must beestimated. For that purpose, different models are used to estimatethe effective thermal conductivity of the composite materials[31,32]. Since not all the variables are measurable in mortars, theirvalidity cannot be proven. However, the proposed parameterscould be used to seek tendencies. There is not any standardizedmethod to evaluate this property. The best option would be tocreate a solid specimen from the uncrushed rawmaterial. However,this option was not possible for either the EAF or especially for theCDW, because of their production process. In this contribution,water-saturated aggregate-alone effective thermal conductivity(lsa,eff) is measured to determine the final thermal conductivity ofthe particles, according to the ASTM D-5334 standard [33].

rent aggregates and proportions.

Fig. 5. Compressive strengths at 7 (a) and 28 (b) days of curing age.

Fig. 6. Aggregate proportion dependent hardened dry bulk density (a) and porosity (b).

R. Borinaga-Treviño et al. / Applied Thermal Engineering 50 (2013) 159e167164

Fig. 7. Thermal conductivities for different aggregate types and proportions.

R. Borinaga-Treviño et al. / Applied Thermal Engineering 50 (2013) 159e167 165

According to Zhang [31], series (2), quadratic parallel (3) andgeometry-weighted models (4) are proposed to estimate thethermal conductivity of the solid particles.

lsa;eff ¼ ð1� fwÞls;series þ fwlw (2)

lsa;eff ¼�ð1� fwÞ

�ls;quadratic

�0:5þfwðlwÞ0:5�2

(3)

lsa;eff ¼ �ls;geometry

�ð1�fwÞðlwÞfw (4)

where ls and lw are the thermal conductivities of the aggregateparticles and water, respectively. The thermal conductivity of waterof 0.6 W/(m K) at 294 K is obtained from the existing bibliography[30]. Aggregate porosity (4w) has been calculated according to EN1097-3 [34]. Table 4 shows the results obtained for the test (lsa,eff,and 4w), as well as the estimated particle thermal conductivity forthe different models and aggregates evaluated (ls,series, ls,quadratic,ls, geometry).

Table 4Thermal conductivity of aggregate particles for different prediction models.

Aggregate leff 4w ls,series ls,quadratic ls,geometry

L 1.6 28 2.0 2.1 2.4S 2.5 32 3.5 3.9 5.0CDW þ F 1.3 45 1.8 2.0 2.3EAF þ F 1.3 32 1.6 1.7 1.9

l Thermal conductivity, W/(m K).4w Porosity of the aggregate according to UNE-EN 1097-3, %.

Table 5Volumetric and thermal conductivity values for the evaluated mortars.

Mix s1 s2 w/c lm-h Volumetric proportions (%)

Va Vs Vp

2 L 0 0.30 1.6 15 35 503 L 0 0.37 1.8 9 49 424 L 0 0.43 1.8 13 59 285 EAF F 0.30 1.4 17 29 466 EAF F 0.40 1.5 18 42 407 EAF F 0.49 1.5 18 48 348 CDW F 0.40 1.3 21 33 469 CDW F 0.62 1.2 18 44 3810 CDW F 0.93 1.0 36 46 1811 S 0 0.34 1.8 19 34 4712 S 0 0.50 1.8 17 47 3613 S 0 0.54 2.1 26 55 19

lm-h Laboratory measured hardened-mortar apparent thermal conductivity,W/(m K).

For all the prediction models proposed, a higher conductivity isobserved in the CDW than in the EAF particles, which contrasts withthe tendencies observed for each of the resulting mortars. Table 5shows the results obtained for the proposed volumetric andthermal parameters. For the L, S and EAF-based mortars, accessibleporosity has a similar value. As water to cement ratio increases,a decrease of lp is expected, but this phenomenon is compensated bythe higher thermal conductivity and volumetric proportion ofaggregate present in the mix. However, CDW mixes display a sharpincrease in the water to cement ratio with the aggregate proportionand a significantly higher accessible porosity, which causes a sharpdecrease in the thermal conductivity of the c-sp-w paste and air,respectively. The higher aggregate thermal conductivity and itsvolumetric proportion in the mix are not sufficient to compensatethis negative phenomenon, whichwould explain the decrease in thethermal conductivity with the increase in the aggregate proportion.

4. Conclusions

Limestone, silica, Electric Arc Furnace slag and Construction andDemolition Waste sands are proposed as main aggregates formortars used in geothermal closed-loop heat exchangers. Lime-stone filler has been used to correct the grain size distribution ofthe latter two, with a substitution rate of 25% and 10% by weight,respectively. A flow table test diameter of 260e320 mm has beenused for the design of self compactable geothermal groutingmortars. This fact has been verified during the filling of the speci-mens and the realization of the fresh bulk density test. The objec-tives proposed in this paper have led to the following conclusions:

� The proposed mixes fulfill the minimum of 15 MPa ofcompressive strength required for non-structural material in theSpanish concrete standard. However, a maximum proportion ofs/c ¼ 2 is proposed for the Construction and Demolition Wastesand-basedmortars due to the drastic decrease in the values thatthe mortars display as aggregate proportion increases.

� The use of the four aggregates proposed improved the thermalconductivity of the pure cement used as reference material andit is independent of its proportion of use for the range evalu-ated. Silica, Limestone, Electric Arc Furnace slag andConstruction and Demolition Waste sand-based mortars havea maximum thermal conductivity of 2.1 W/(m K), 1.8 W/(m K),1.5 W/(m K) and 1.3 W/(m K), respectively.

� Earth Energy Designer software [35] permits the estimation ofthe borehole thermal resistance by using the Multipole method[36]. For that purpose, a standard double U configuration ischosen, with a borehole diameter of 140 mm and pipe externaland internal diameters of 32 mm and 26 mm, respectively.

R. Borinaga-Treviño et al. / Applied Thermal Engineering 50 (2013) 159e167166

Effective borehole thermal resistances for Neat cement, L, EAF,CDW and S mortars has been estimated in 0.1139 (m K)/W,0.0727 (mK)/W, 0.0808 (mK)/W, 0.0901 (mK) and 0.0665 (mK)/W, respectively. As a consequence, the use of improved mortarspermit a reduction of the total borehole length required of 16%,12%, 9% and 18% for the L, EAF, CDW and S mortars compared tothe one required by the neat cement reference material.

� Effective thermal conductivity increases as aggregate propor-tion increases for Silica sand, Limestone sand and EAF sand,while maintaining similar porosity values. CDW-based mortarsdecrease as the proportion of use increases, due to the highwater content required for the mix. The high porosity observedin the corresponding mixes is attributed to the high waterabsorption of the aggregate.

Acknowledgements

The authors wish to acknowledge the support provided by theGITECO Group of the University of Cantabria and the Department ofConstruction at the University of Oviedo. The research project‘Development of systems for rainwater catchment and storagethrough the use of permeable pavements to assess its non-potableuses as a resource of low-enthalpy geothermal energy (REV)’[BIA2009-08272] of the University of Cantabria is possible thanksto the finance of the Research, Development and Innovation StateSecretary of the Spanish Ministry for Economy and Competitive-ness. Finally, authors wish to acknowledge the financial supportprovided by the research projects FICYT FC-10-EQUIP10-17 andBIA-2008-00058.

Appendix A

The volumetric proportion (Vi) is defined as the percentage-ratio of volume occupied by the medium (i) and the total volumeof the resulting mortar. Considered mediums are air (a), cement-superplasticizer-water hardened paste (p) and aggregate solidparticles (s).

First of all, it is assumed that accessible porosity (n) representsthe air medium of the mortar and hence it determines Va, which iscalculated with Equation (1). Unknown non-accessible porosity islater considered as part of the c-sp-w medium.

For the determination of rm-f y rm-h and n, total volume of thespecimen is defined as the volume of the mold used. As an approxi-mation, it is assumed that the volumetric proportion of aggregateparticles (Vs) is the same for mortar in fresh and hardened state. Therelative weight of aggregate used for the determination of themix infresh state (s/c) and the apparent density of the aggregate particles(rs) permit the determination of the aggregate’s volume. As mixproportionsaredefinedbyunitweightof cementused, totalweightofthe mix proportion (WT) can be calculated by using Equation (A1).Total volume of the mix for the unit weight of cement used is calcu-lated bymultiplyingWT by the bulk density of the freshmortar (rm-f).

WT ¼ 1þ s=cþ sp=cþw=c (A1)

Therefore, the ratio of the aggregate particles volume with thetotal volume determines Vs (Equation (A2)).

Vs ¼ rm�f$ðs=cÞ=ðrs$WTÞ (A2)

Finally, volumetric proportion of the c-sp-w paste is unknown,but considered as the rest of themedium that completes themortartotal volume, which is calculated with Equation (A3).

Vp ¼ 1� Vs � Va (A3)

References

[1] A.M. Omer, Energy, environment and sustainable development, Renewableand Sustainable Energy Reviews 12 (2008) 2265e2300.

[2] European Union, 2009/28/CE: Fomento Del Uso De Energía Procedente DeFuentes Renovables (2009).

[3] G. Florides, S. Kalogirou, Ground heat exchangers-a review of systems, modelsand applications, Renewable Energy 32 (2007) 2461e2478.

[4] A.J. Philippacopoulos, M.L. Berndt, Influence of debonding in ground heatexchangers used with geothermal heat pumps, Geothermics 30 (2001)527e545.

[5] M.L. Allan, A.J. Philippacopoulos, Performance Characteristics and Modelling ofCementitious Grouts for Geothermal Heat Pumps (2000).

[6] M.D. Smith, R.L. Perry, Borehole grouting: field studies and thermal perfor-mance testing, ASHRAE Transactions 105 (1999).

[7] C. Lee, K. Lee, H. Choi, H.- Choi, Characteristics of thermally-enhancedbentonite grouts for geothermal heat exchanger in South Korea, ScienceChina Technological Sciences 53 (2010) 123e128.

[8] S.W. Carlson, GSHP bore field performance comparisons of standard andthermally enhanced grout, ASHRAE Transactions 106 (2000).

[9] F. Delaleux, X. Py, R. Olives, A. Dominguez, Enhancement of geothermalborehole heat exchangers performances by improvement of bentonite groutsconductivity, Applied Thermal Engineering 33e34 (2012) 92e99.

[10] M.L. Allan, and A.J. Philippacopoulos, Thermally conductive cementitious groutsfor geothermal heat pumps. FY 1998 Progress Report, BNL 66103 (1998), 395.

[11] M.L. Allan, Materials characterization of superplasticized cement-sand grout,Cement and Concrete Research 30 (2000) 937e942.

[12] UNE-EN 197-1, Cement. Part 1: Composition, Specifications and ConformityCriteria for Common Cements, Asociación Española de Normalización y Cer-tificación (AENOR), 2000.

[13] J.M. Manso, J.J. Gonzalez, J.A. Polanco, Electric arc furnace slag in concrete,Journal of Materials in Civil Engineering 16 (2004) 639e645.

[14] J.M. Manso, J.A. Polanco, M. Losañez, J.J. González, Durability of concrete madewith EAF slag as aggregate, Cement and Concrete Composites 28 (2006)528e534.

[15] C. Pellegrino, V. Gaddo, Mechanical and durability characteristics of concretecontaining EAF slag as aggregate, Cement and Concrete Composites 31 (2009)663e671.

[16] Ministerio de fomento, Instrucción De Hormigón Estructural. EHE08, Centrode Publicaciones, Madrid, 2008.

[17] UNE-EN 1097-6, Tests for Mechanical and Physical Properties of Aggregates.Part 6: Determination of Particle Density and Water Absorption, AsociaciónEspañola de Normalización y Certificación (AENOR), 2006.

[18] UNE-EN 933-1, Tests for Geometrical Properties of Aggregates. Part 1:Determination of Particle Size Distribution. Sieving Method, AsociaciónEspañola de Normalización y Certificación (AENOR), 2005.

[19] M. Allan, Thermal conductivity of cementitious grouts for geothermal heatpumps, FY 1997 Progress Report. BNL, 65 (1997), 129.

[20] G. Gustafson, H. Stille, Prediction of groutability from grout properties andhydrogeological data, Tunnelling and Underground Space Technology 11(1996) 325e332.

[21] UNE-EN 1015-3, Methods of Test for Mortar for Masonry. Part 3: Determi-nation of Consistence of Fresh Mortar (by Flow Table), Asociación Española deNormalización y Certificación (AENOR), 2006.

[22] UNE-EN 1015-6, Methods for Mortars for Masonry. Part 6: Determination ofBulk Density of Fresh Mortar, Asociación Española de Normalización y Cer-tificación (AENOR), 2007.

[23] UNE-EN 1015-10, Methods of Test for Mortar Masonry. Part 10: Determina-tion of Dry Bulk Density of Hardened Mortar, Asociación Española de Nor-malización y Certificación (AENOR), 2007.

[24] UNE-EN 1015-11, Methods of Test for Mortar for Masonry. Part 11: Deter-mination of Flexural and Compressive Strength of Hardened Mortar, Asocia-ción Española de Normalización y Certificación (AENOR), 2007.

[25] S.E. Gustafsson, Transient plane source techniques for thermal conductivityand thermal diffusivity measurements of solid materials, Review of ScientificInstruments 62 (1991) 797e804.

[26] A. Bouguerra, A. Ait-Mokhtar, O. Amiri, M.B. Diop, Measurement of thermalconductivity, thermal diffusivity and heat capacity of highly porous buildingmaterials using transient plane source technique, International Communica-tions in Heat and Mass Transfer 28 (2001) 1065e1078.

[27] S.A. Al-Ajlan, Measurements of thermal properties of insulation materials byusing transient plane source technique, Applied Thermal Engineering 26(2006) 2184e2191.

[28] D.P. Bentz, Transient plane source measurements of the thermal properties ofhydrating cement pastes, Materials and Structures 40 (2007) 1073e1080.

[29] D. Nuñez-Becerra, Capítulo 8: Lodos tixotrópicos en cimentaciones profundas,in: C. López jimeno (Ed.), Ingeniería Del Terreno, Libro 10, E.T.S.I. Minas-Universidad Politécnica de Madrid, Madrid, 2007, p. 261.

[30] J.P. Holman, Transferencia De Calor, Mc Graw Hill, Madrid, 1999.[31] H.-. Zhang, X.-. Ge, H. Ye, D.- Jiao, Heat conduction and heat storage charac-

teristics of soils, Applied Thermal Engineering 27 (2007) 369e373.[32] D. Kumlutas, I.H. Tavman, M. Turhan Çoban, Thermal conductivity of particle

filled polyethylene composite materials, Composites Science and Technology63 (2003) 113e117.

R. Borinaga-Treviño et al. / Applied Thermal Engineering 50 (2013) 159e167 167

[33] ASTM D-5334-08, Standard Test Method for Determination of ThermalConductivity of Soil and Soft Rock by Thermal Needle Procedure, AmericanSociety for Testing and Materials, 2008.

[34] UNE-EN 1097-3, Tests for Mechanical and Physical Properties of Aggregates.Part 3: Determination of Loose Bulk Density and Voids, Asociación Españolade Normalización y Certificación (AENOR), 1998.

[35] G. Hellström, B. Sanner, Software for Dimensioning of Deep Boreholes for HeatExtraction (1994) 195e200.

[36] L. Lamarche, S. Kajl, B. Beauchamp, A review of methods to evaluate boreholethermal resistances in geothermal heat-pump systems, Geothermics 39(2010) 187e200.

Nomenclature

Aggregate typesL: limestone sandS: silica sandEAF: Electric Arc Furnace slag sandCDW: Construction and Demolition Waste sandF: limestone Filler

Design of the mix proportionsw: water kgc: cement kgsp: superplasticizer kgs: aggregate kg

w/c: water to cement ratio kg/kgsp/c: superplasticizer to cement ratio kg/kgs/c: total aggregate to cement ratio kg/kg

Propertiesr: bulk density kg/m3

n: accessible porosity m3/m3

msat: water-saturated mass kgmdry: oven dried mass kgl: thermal conductivity W/(m K)V: volumetric proportion of the hardened mortar m3/m3

4: porosity of the aggregate m3/m3

Subscriptsm-f: mortar in fresh statem-h: mortar in hardened statea: airs: aggregate’s particlesw: waterp: cement-superplasticizer-water hardened pastesa: aggregate-aloneeff: effectiveseries: series composite modelquadratic: quadratic composite modelgeometry: geometry weighted composite model