SiSiC Heat Exchangers for Recuperative Gas …giaki.org/sites/default/files/progetti/SiSiC Heat...energy. Energy consumption in Europe in the industry sector represents about 30% of

SiSiC Heat Exchangers for Recuperative Gas Burnerswith Highly Structured Surface Elements

Alberto Ortona*

SUPSI, Manno, Strada Cantonale 2c 6928 Manno, Switzerland

Dimosthenis Trimis, Volker Uhlig, and Robert Eder

TU Bergakademie Freiberg, Gustav-Zeuner-Str. 7, 09599 Freiberg, Germany

Sandro Gianella

Erbicol, Viale Pereda, 22 6828 Balerna, Switzerland

Paolo Fino and Giuseppe D’Amico

Politecnico di Torino, Corso Duca degli Abruzzi 24, 10129 Torino, Italy

Emmanuel Boulet and Claire Chazelas

Institut Franc�ais du textile et de l’habillement, 93 Chemin des Mouilles, 69130 Ecully, France

Tobias Gr€amer

NOXMAT, Ringstrasse 7, 09569 Oederan, Germany

Enrico Cresci and Joachim G. W€unning

WS W€armeprozesstechnik, Dornierstraße 14, 71272 Renningen, Germany

*[email protected]

© 2013 The American Ceramic Society

Int. J. Appl. Ceram. Technol., 1–11 (2013)DOI:10.1111/ijac.12087

Herwig Altena

Aichelin, Fabriksgasse 1-3, 2340 M€odling, Austria

Franz Beneke and Michel Debier

European Committee of Industrial Furnace and Heating Equipment Associations CECOF,Lyoner Straße 18, 60528 Frankfurt, Germany

Heat exchangers are used in gas burners as recuperators. Their efficiency is proportional to their surface which is usually

limited by the burner length and diameter. Ceramic recuperators used nowadays in industrial burners are tubular. We studiedand developed a new generation of ceramic heat exchanger with the final goal of increasing the efficiency or, at similar effi-ciencies, of reducing their size and weight. A commercial SiSiC heat exchanger component was used to guarantee safety and

robustness. The use of structured textile geometries followed by their conversion into a ceramic is the main innovation of thiswork.

Introduction

Industrial production is a major consumer ofenergy. Energy consumption in Europe in the industrysector represents about 30% of the overall fuel con-sumption in all sectors. One can asses that ca. 60% ofthe industrial fuel consumption is spent for industrialprocess heating, and furthermore, more than 80% ofthe process heat is covered by fossil fuels (Source: Ger-man Federal Ministry of Economics and Technology).An efficiency increase in process heating of only a fewpercent would result in a significant reduction in energyconsumption and carbon footprint.

Industrial branches of high energy consumption canbe found in the metal industry (steel and alloys), glassindustry, and ceramics production and processing.Because most of the relevant processes are operated at ahigh temperature, heat recovery becomes essential interms of efficiency. The most common way to recoverheat is by preheating the air or other gasses with hoteffluent gasses. The so-called recuperative or regenerativeheat exchanger systems,1 which may be integrated in theburner assemblies, are commonly used for this purpose.

The use of burners equipped with recuperatorsresults in significant efficiency increase. However, theirefficiency is proportional to the heat exchanger surfacewhich is limited by the overall volume and burner lengthrestrictions and by the required heat exchanger and itsmaterial limitations, which will be discussed later on.

The highest efficiency can be reached using regen-erative burners. However, the complexity of regenera-tive burner systems is higher, due to the discontinuousalternately mode of operation utilizing several regenera-tor matrices.

In all cases, it can be presumed that a 100 Kdecrease in the flue gas temperature after heat recoveryresults in an estimated reduction in fuel consumptionof about 5% (Fig. 1).

Heat exchangers used nowadays are mostly madeof high temperature alloy steel with corresponding tem-perature limitations. The use of ceramic materialsbrings the following advantages.

• operation and heat recovery at the higher tem-peratures,

• higher corrosion resistance in many harsh envi-ronments.

On the other hand, they present the followingdrawbacks:

• lower amount of heat can be recovered in com-parison with the metal recuperator, due to man-ufacturing restrictions,

• fragile mechanical behavior, which implies thatcomponents are always over-dimensioned,

• restrictions due to dust transported with the gasflows,

• heat recovery devices should be directly inte-grated into the thermal processing facility, to

2 International Journal of Applied Ceramic Technology—Ortona, et al. 2013

minimize thermal losses and utilize the energyin an immediate manner.

Despite the obvious benefits of ceramic heatexchanger components for heat recovery at hightemperature applications, the penetration of such tech-nologies in industrial furnaces is restricted by compara-tively higher prices and larger size of such components.Relatively simpler ceramic heat exchanger geometrieswith large dimensions are used up to now, due to severalmanufacturing and operational obstructions. Increasingtheir efficiency will lead to decrease their dimensionovercoming the above-mentioned economical problems.

Ceramic recuperators used nowadays in industrialburners are tubular. Common recuperator lengths are inthe order of 0.5 m and hence longer than the thickness ofa typical furnace wall. Heat enhancement with structuredsurfaces is limited by the associated ceramic productiontechnologies. Given these technological obstructions andthe dimensional limitations, the level of heat recovery isrestricted. State of the art recuperative burners show an airpreheating and flue gas temperature level after the recuper-ator in the range of 500–700°C resulting in heat lossesthrough the flue gas of about 25–35%.2

Within the CEREXPRO project (§ 5), we studiedand developed a new generation of ceramic heat exchang-ers for high temperature heat recovery having the goal of

increasing the efficiency or, due their size and weightreduction at similar efficiencies, of lowering their price.The use of structured geometries typical of the textileindustry followed by their conversion into a ceramic is themain innovation of this project. Although the principle isnot new, what makes it innovating is the unusual combi-nation of such different production fields which in turnleads to a high design flexibility of the heat exchanger.3

To facilitate the development, this approach wasapplied on an existing robust SiSiC heat exchanger com-ponent successfully applied in industrial furnaces. Thisapproach guaranteed application safety and robustness.Due to the improved efficiency, we proved that a signif-icant size reduction or, alternatively, an increase in theheat recovery level can be achieved because of the higherheat transfer rate of the heat exchange elements.

This paper is divided into four sections. The thermofluid dynamic behavior of the structured surface is studiedin section ‘Design’. Section ‘Processing’ deals with thetextile structure development and production and its con-version process from textile-derived materials into a thermalshock resistant ceramic (e.g., silicon infiltrated siliconcarbide – SiSiC). Finally, section ‘Prototype’ presents a pro-totype of SiSiC heat exchanger for recuperative gas burnerswith SiSiC structured surface elements manufacturing andfield testing.

Design

The design of the highly structured SiSiC surfaceelements must meet the requirements of robustness andadaptability to an existing heat exchanger. It was alsotaken into account what textile and ceramic state of theart technology could provide. An arc (or loop) of 180°was selected as the basic element for the structured sur-face. The aim was to reach a higher heat transfer coeffi-cient by increasing the transferring surface and also byutilizing the bigger dispersion and thermal boundarylayer structures due to the resulting turbulent flowcharacteristics. Figure 2 shows different possiblearrangements of such loops on a flat surface.

Numerical simulations using commercial CFDpackages were performed for geometries in Fig. 2 tocalculate the flow characteristics and analyze theincrease in heat transfer and pressure drop, that is, fric-tion factor coefficient. To reduce computation time, aplanar arrangement was selected as reference geometry.To simulate the fluid flow and heat transfer process, a

0

10

20

30

40

50

60

70

80

90

100

0 200 400 600 800 1'000 1'200

Com

bust

ion

effic

ieny

[%]

Flue gas temperature [°C] before heat recovery

Regenerative burner Recupertative burner

central heat exchanger

Fig. 1. Combustion efficiency with respect to the exhaust gassestemperature before heat recovery; Experimental results for differenttypes of heat recovery 2.

www.ceramics.org/ACT SiSiC Heat Exchangers 3

typical repeating cell was defined. This cell was used torepresent the entire structure by geometrical symmetryto further decrease the computational efforts.

The governing equations were solved using finitevolume methods assuming a forced convection situa-tion. Boundary conditions of the cubic cell were asfollows: constant temperature at the bottom surface (bluein Fig. 2), which must be lower than the inlet tempera-ture of the air flow (red arrow in Fig. 2) to generate aheat flux from the hot air to the cooled heat exchangersurface. A heat transfer coefficient of approximatelya = 82 W/m/K of the flat solution was used as bench-mark for evaluation of the new structured surfaces, whilethe operating conditions were kept constant.

Pressure drop, Nusselt number, and heat transfercoefficient were calculated at different Reynolds num-bers. Figure 3 shows the ratio of the expected perfor-mance in comparison with the reference conventional

recuperator burner at nominal conditions average (Rey-nolds number 5200).

These results indicate that the targeted recuperatordesign will result in a significantly higher heat recovery andslightly higher pressure losses, given the same burner size. Onthe other hand, keeping the same heat recovery, the burnerlength, and the pressure losses can be significantly reduced.

Processing

Textile Structure Assembly

Textile refers to any material made of interlacingfibers. Fabric refers to any material made through weav-ing, knitting, spreading, crocheting, or bonding thatmay be used in production of further goods In textileindustry.4 Three processes were analyzed to realize asuitable structure as well as several fiber materials.

Nonwoven An assembly of textile fibersheld together by mechanicalinterlocking in a random web or mat,by fusing of the fibers (in the case ofthermoplastic fibers), or by bondingwith a cementing medium. Initially,the fibers may be oriented in onedirection or may be deposited in arandom manner. Nonwoven wasimmediately abandoned because theproposed design could not beobtained with this technology.

(a)

(b)

Fig. 2. Selected loops design (Radius R = 8 mm and diameterD = 1 mm): (a) Geometry 01 (3600 Loops/m²). Distance betweenLoops: 12 mm in flow direction and 24 mm perpendicular to theflow. (b) Geometry 02 (7200 Loops/m²). Distance between Loops:8 mm in flow direction and 18 mm perpendicular to the flow.

1.0

2.3

8.2

1.0

7.5

17.6

0

4

8

12

16

20

Basic geometry (Noxmat)

Geometry 01 (3600 loops / m²)

Geometry 02 (7200 loops / m²)

Rat

io c

urre

nt v

alue

/ ba

sic

geom

etry

Ratio of heat transfer coefficients

Ratio of friction factors

Fig. 3. Friction factor and heat exchange efficiency calculatedfor the different layouts.


Weaving Weaving is made by interlacing twoyarns of similar materials so that theycross each other at right angles toproduce woven fabric. The warp yarnsrun lengthwise in the fabric, and thefilling threads (weft) run from side toside. Trials were performed, and samplesproduced on rapier weaving looms thatwere adapted to match the specificationof the design (Fig. 2). Two weavedrawings have been set corresponding tothe different densities of loops, andthread count for base fabric hasbeen determined according to thediameter of loops. A rapier loomwas adapted to reproduce a sort ofvelvety texture (Fig. 4a). Velvet processuses several warp beams to balance thedifference of absorbed warp lengthbetween base fabrics and loops andperforms a weft thread or rails tomaintain the loops height.

Knitting Knitting is a way to make a fabric byinterlocking series of loops of one ormore yarns. Two different typesof knitting technologies exist: weft andwarp knitting. Weft Knitting is acommon type of knitting, in whichone continuous thread runs crosswisein the fabric making all of the loops inone course. Different trials showed theinterest to use in the building of thestructure an additional yarn withthermo soluble properties (Fig. 4b).

The best performing material, both for weavingand ceramization, was a multifilament yarn made ofpolyethylene terephthalate (PET) fibers. Textile struc-ture assembly had many objectives to comply with:

• structure geometry should comply with thedesign constraints,

• being the textile structure a scaffold on whichdeposit the ceramic layer, the thermo-physicalbehavior of its material under pyrolysis shouldbe tuned properly,

• elastic modulus as well as yarn thickness andsizing should be sufficient to keep their geome-try while replicating,

• another extremely important requirement isthat, as discussed later, the base fabric in contactwith the SiSiC tube walls should be as looseand light as possible to minimize the formationof porosity onto the SiSiC tube (Fig. 5). Poros-ity is detrimental because it reduces heat transferand acts as a crack initiator during thermalcycling.

The base fabric was assembled using PET yarnsmade of 120 filaments with a dtex 668 (the texexpresses the mass in grams per 1000 meters of a yarn).The loop structure was obtained with PET yarns madeof 660 filaments with dtex 2004. As fibers selection wasalso dictated by the ceramization requirements, furtherdetails will be given in the next section. Indeed, therewere opposite needs for the two technologies, for exam-ple, carbon fibers are interesting because they can beconverted to SiC during LSI, but due to their rigidity,they are difficult to weave.

(a)

(b)

Fig. 4. Textile structures (Geometry 1) obtained by (a) weavingand (b) knitting.


Textile Ceramization

In combustion environments, silicon carbideceramics are, for their outstanding thermal andmechanical properties, among the best materials.5 Infact, they are commonly applied in into high tempera-ture furnaces because, if passive oxidation conditionsare met,5 they withstand operating conditions for longtime.6

Three processes were envisaged to fabricate highlystructured SiSiC surface elements:

• Chemical vapor deposition (CVD)

• Polymer impregnation and pyrolysis (PIP)

• Liquid silicon infiltration (LSI)

Several process tests were performed, at the end,because of its industrial maturity and also because thecommercial SiSiC tube is made with a similar technol-ogy (proven to be the best performer in a combustionchamber environment), LSI was adopted as manufactur-ing procedure. A thin (800 mPas) but, stable proprietaryslurry made of a-SiC powders (fine and coarse), a plasticbinder and an organic solvent were prepared. The textilestructure was dipped into the slurry and hanged to drainthe excess liquid. As the solvent was partially evaporatedand the structure still tacky, it was placed on the SiSiCsubstrate as per Fig. 6a and further dried. The plate wasthen placed into a furnace and slowly pyrolyzed up to1000°C, under inert atmosphere. The textile structureexperienced a minor shrinking because of the mass loss

during pyroloysis, and some cracks could be observed.Cracks minimization was worth a long optimizationwork in which textile architecture of the base fabric andamount of slurry employed were carefully adjusted. Afterpyroloysis, the object was placed in contact with siliconflakes and LSI was performed at 1500°C under a 10�2

(mbar) residual pressure. The furnace was quicklybrought to the process temperature, held at that temper-ature for 1 h, and left cooling naturally. A final flat sam-ple is shown in Fig. 6b

A great effort was devoted to the selection of properfibers, further called fugitive (Fig. 7a) or permanent(Fig. 7b), meaning that in one case, they would produce,respectively, a hollow or full space within the loop.

Examples of fugitive fibers are polypropylene (PP)and PET. On the other end, permanent fibers are thosecontaining already a ceramic phase (e.g., C, SiC).

(a)

(b)

Fig. 6. (a) Flat sample preparation: Replicated structures(GEO2) are placed onto a SiSiC plate, (b) Final aspect of theplate (GEO1) after pyroloysis and LSI.

Fig. 5. 3D rendering, obtained by X-ray computed tomography,of highly packed base fabric, impregnated and processed, showinghigh internal porosity.


Fugitive, PET fibers were selected because of their:

• low cost,

• easiness to be weaved,

• similar behavior to the slurry (weight loss, shrink-ing) during pyroloysis resulting in less cracksformation.

Their main drawback is due to their low rigiditywhich could not guaranty to keep the loop shapeduring processing.

Ceramic Characterization

To characterize the skeleton material of the loops,another manufacturing technique was used to obtainsamples with suitable dimensions. Green bulk platescomprised of the same constituent materials were pro-duced by hot pressing, pyrolyzed, and silicon infiltrated.This solution allowed to produce monolithic pieces com-plying with the geometries necessary for characterization.

Elastic modulus was measured using the impulseexcitation technique with a Mk5 apparatus (GrindoSonic Leuven B). Thermal conductivity k was calcu-lated according to the relationship: k = a cp q. Ther-mal diffusivity a together with specific heat capacity cpwas measured with a Flashline 4010 (Anter, NewCastle, DE). Measurements of the coefficient of ther-mal expansion as a function of the temperature, with afixed heating rate of 5°C/min and from RT to1500°C, were performed with a Setsys Evolution TMA(Setaram F). Mechanical properties under bending weremeasured with a Sintech 10 D universal testingmachine (MTS, Eden Prairie, MN). Microhardness wasmeasured with a microdurometer (Leitz Wetzlar D).Samples’ microstructure was analyzed with Zeiss-Evo-50 SEM (Table I).

SEM image in Fig. 8 shows a central void typicalin fugitive loops. The microstructure of the as-produced loop shows only two phases: dark gray zonesand light gray zone. XRD analysis (Fig. 8b) confirmedthat these phases are, respectively, silicon carbide andsilicon.

Microscopic characterization confirmed that thebase material of the loops is very similar to that of theSiSiC ERBISIC foams (Erbicol SA, Balerna CH).Indeed, this material applied by replica on reticulatedfoams has proven to withstand long lasting severeoxidative environments typical of the porous burnerstechnology.7

(a)

(b)

Fig. 7. 3D rendering, obtained by X-ray computed tomography,of SiSiC loops. (a) Hollow dark gray region inside the loop for-merly occupied by the. fugitive yarn, (b) full, that is, permanentyarn.

Table I. Material Properties Measured on theSiSiC Material

Property Average value RSD%

Young’s modulus (GPa) 264 6Vickers micro hardnessLoad 500 (g) HV

1729 11

MOR (3-point flexuralstress) (MPa)

203 21

Thermal conductivity (W/m/K)193°C 85.08501°C 59.661004°C 41.681504°C 37.49Coefficient of thermal expansion (per °C)100°C 7 9 10�6

300°C 4.5 9 10�6

1400°C 5 9 10�6


Properties were utilized to refine system design.They also confirmed to be quite similar to other com-mercial SiSiC products.

Prototype

Manufacturing

The manufacturing method developed and opti-mized in section ‘Processing’ was adopted to fabricatetwo complete heat exchanger units with the two

reference geometries. Textiles and SiSiC tubes (SchunkIngenieurkeramik Willich-M€unchheide, D) were com-bined in the ceramization process, and silicon infiltratedto obtain the final article. Textile structure placinginside and outside the tube walls was optimized, giventhe small radiuses of curvature of the tube.

The entire procedure was optimized on cylindricalSiSiC tubes produced by Schunk. It was then appliedon a real recuperator.

The first tubes were cut in shorter units. Textileswere impregnated in the ceramic slurry and cut to

(a) (b)

Fig. 9. Two SiSiC heat exchangers with different loops distribution. (a) Geometry 01, 3600 Loops/m². Loop: R = 8 mm/D = 1 mmand (b) Geometry 02, 7200 Loops/m². Loop: R = 8 mm/D = 1 mm.

(a) (b)

Fig. 8. (a) SEM image of a loop cross-section produce by replicating a PET yarn (hollow space) as per the process described in the previous section,loop cross-section is not round due to slurry draining before solidification (b) X-ray diffractometry analysis of the loop material.


dimension. These patches were placed inside and out-side the tube sections. Several trials were performed toobtain tube sections complying with the design require-ments. The pieces were pyrolized and infiltrated withmolten silicon according to the ceramization processdefined in section “Textile Ceramization”. Once vali-dated the entire procedure, two complete recuperatorsfor an 80 kW burner were produced. Results for thetwo different geometries are shown in Fig. 9.

Field Tests

The recuperators tested in this work at differentpower levels were as follows:

• plain recuperator without loops and with wastegas guiding tube (Fig. 10a)

• structured recuperator with loops (geometry 1)and with waste gas guiding tube (Fig. 10b)

Both recuperators were assembled in the same testburner and operated in indirect heating using a jackettube ø 200 mm. To characterize the burner, many dif-ferent measuring points were assembled at the burnerand the jacket tube. Figure 11 shows the points whereseveral parameters were acquired.

Via these measurements the following parameterscould be determined:

• Average temperatures at inlets and outlets aswell as in the jacket and flame tube

• temperature profiles throughout the recuperator

• waste gas mixture

• firing efficiency

• pressure drops

(a) (b)

Fig. 10. burner with different recuperators (a) plain recuperator and (b) loop recuperator with Geometry 01, 3600 Loops/m². Loop:R = 8 mm/D = 1 mm.

Fig. 11. Overview of all measuring points (red for waste gas, yellow for natural gas, blue for combustion air).


Using the loop recuperator with geometry 1, thefiring efficiency was increased by 7–9% depending onthe waste gas inlet temperature. Although tests havebeen not yet performed, we expect a further increase inthe efficiency using of loop geometry 2 and anoptimizing of the waste gas guiding tube. The firingefficiency at capacities of 40 and 80 kW with different

waste gas inlet temperatures is shown in Fig. 12. Theincrease in the efficiency causes a higher pressure dropat the air and at the waste gas sides (see Fig. 13). Theair preheating (Fig. 14) could be increased by 60–80°Cwith the use of the new recuperator geometry.

Certainly, the raising of efficiency and air preheat-ing causes an increase in the NOx emissions in the wastegasses. At 40 kW, they raise about 25%, at 80 kWabout 13% (Fig. 15). With specific arrangements

55

60

65

70

75

80

85

90

600 700 800 900 1000 1100 1200 1300

Firin

g Ef

ficie

ncy

[%]

Waste gas inlet temperature [°C]

40 kW, plain recuperator


40 kW, loop recuperator


Fig. 12. Comparison of firing efficiency of plain and loop recu-perator at different capacities.

5.3

10.8

9.3

14.7

0

4

8

12

16

20

waste gas air

Pre

ssur

e dr

op [m

bar]

plain recuperator, with waste gas guiding tube

loop recuperator, with waste gas guiding tube

Fig. 13. Comparison of pressure drop for waste gas and air ofplain and loop recuperator at 80 kW with an inlet temperatureof 1000 °C.

100

200

300

400

500

600

700

800

600 700 800 900 1000 1100 1200 1300





Fig. 14. Comparison of air preheating of plain and loop recu-perator at different capacities.

0

100

200

300

400

500

600

600 700 800 900 1000 1100 1200 1300

NO

x-Em

issi

on b

ased

on

5 %

O2 i

n w

aste

gas

[mg/

m³] 40 kW, plain recuperator




Fig. 15. Comparison of NOx emissions in waste gas of plainand loop recuperator at different capacities.


concerning to the combustion system, a reduction inthe emissions could be achieved. Due to the increasedsurface area of the tube with highly structured surfaceelements, oxidation phenomena will be enhanced. In anycase, if passive oxidation conditions are met, they shouldnot reduce the performance of the components, as previ-ously demonstrated on SiSiC reticulated ceramics.7

As expected for this kind of materials, the Si-SiCloops are very fragile. After multiple assembling, manyloops on the waste gas side were broken and lost.Therefore, a possibility for protecting the loops duringthe assembling has to be developed.

Further work is currently ongoing to study theoverall behavior of these structures along the entire life-span of the component, in particular, to study thedeposition of dust and or soot onto the loops.

Acknowledgments

The research leading to these results has receivedfunding from the European Union Seventh Framework

Programme (FP7/2007-2013) under grant agreementno 227551, Project Ceramic Heat exchangers withenhanced materials properties (CEREXPRo).

References

1. R. K. Shah and D. E. Metzger, American Society of Mechanical Engineers,American Society of Mechanical Engineers, Heat Transfer Division andAmerican Society of Mechanical Engineers, Gas Turbine Power Division.,“Regenerative and Recuperative Heat Exchangers,” ASME, New York, (345East 47th ST., New York 10017), 1981, pp. v, p. 86.

2. J. G. W€unning and A. Milani, “Handbuch der Brennertechnik f€ur Indust-rie€ofen,” Vulkan Verlag, Essen, (2007).

3. D. Trimis, V. Uhlig, R. Eder, A. Ortona, S. Pusterla, and E. Ambrosio,“New Ceramic Heat Exchangers with Enhanced Heat Transfer Propertiesfor Recuperative Gas Burners,” Heat. Process., 9 [2] 183–187 (2011).

4. Wikipedia, “Textile.” in., 2012.5. N. S. Jacobson, “Corrosion of Silicon-Based Ceramics in Combustion Envi-

ronments,” J. Am. Ceram. Soc., 76 [1] 3–28 (1993).6. F. Mach, F. Issendorff, A. Delgado, and A. Ortona, “Experimental Investi-

gation of the Oxidation Behaviour of SiSiC Foams,” Bioceramics and PorousCeramics: Ceramic Engineering and Science Proceedings, eds., R. Narayan andP. Colombo. John Wiley & Sons, Inc. Hoboken, NJ, 29 [7] 299–311(2009).

7. A. Ortona, S. Pusterla, P. Fino, F. Mach, A. Delgado, and S. Biamino,“Aging of reticulated Si-SiC foams in porous burners,” Adv. Appl. Ceram.,109 [4] 246–251 (2010).


Documents

SiSiC Heat Exchangers for Recuperative Gas …giaki.org/sites/default/files/progetti/SiSiC Heat...energy. Energy consumption in Europe in the industry sector represents about 30% of