Sustainability of the exploitation of Campi Flegrei geothermal area using a 1 zero-mass extraction device

In this paper, the use of a zero-mass extraction device has been simulated in the volcanic area of 14 Campi Flegrei (Italy), one of the most promising geothermal districts of Italy. The sustainability of the heat extraction has been studied with a coupled model of the geothermal reservoir and the deep 16 borehole heat exchanger. The reservoir model has been built using the SHEMAT software, the heat transfer in the deep borehole heat exchanger has been simulated using GEOPIPE, a pure conductive semi-analytical model. An iterative approach has been used to couple the two simulators. The work 19 has demonstrated that the area of Campi Flegrei is a promising candidate to produce sustainable geothermal energy with a zero-mass extraction device. It is also demonstrated that the coupled 21


24
The area of Campi Flegrei is one of the most promising geothermal districts of Italy. The area is part 25 of the Neapolitan volcanoes district, which includes also Ischia Island and Somma-Vesuvius volcano 26 (Fig. 1).  the CFDD project has been stopped and none pilot plant has been realized in the area until today. 48 The social response to the exploration and utilization of geothermal resources in the area of Campi 49 Flegrei is negative. The caldera is one of the greatest geohazard areas on Earth (Piochi et al., 2014) 50 and a great part of the population living there, perceive the drilling activities and the production 51 and reinjection of fluids as an unacceptable risk. 52 The possibility to produce geothermal energy with a zero-mass extraction device may be the key to 53 increase social acceptance. This solution entails the use of a deep borehole heat exchanger, or 54 WellBore Heat eXchanger (WBHX), as named by Nalla et al. (2005). The heat exchanger is formed of 55 an external steel casing and two internal coaxial tubes, also in steel (Fig.2). The internal tubes are    The results of Mottaghy and Dijkshoorn (2012) also indicate that the presence of a groundwater flow 86 has a positive effect on the deep borehole heat exchanger performance. The common approach to 87 studying the heat extraction with a ground heat exchanger is the use of pure conductive models, 88 whereas the evaluation of the groundwater influence on a coaxial heat exchanger, needs the 89 simulation of the heat transfer into the reservoir and between it and the DBHE. In this case, the most 90 accurate method entails the application of the conservation equation of mass, momentum and 91 5 energy in the heat exchanger and the surrounding rock, by using Multiphysics software, or coupling 92 a numerical reservoir model (i.e. TOUGH2, SHEMAT, FEFLOW) with a DBHE model (analytical, 93 semi-analytical, or numerical). Mottaghy and Dijkshoorn (2012) have highlighted that the great 94 computational time required by that software, can be drastically reduced, without losing accuracy, 95 coupling the reservoir model with a semi-analytical finite difference formulation for the WBHX.

96
In this paper, we simulate the use of a zero-mass extraction device in the volcanic area of Campi 97 Flegrei (Italy) using a coupled model of the geothermal reservoir and the deep borehole heat 98 exchanger. The reservoir model has been built using the SHEMAT software (Clauser, 2003), which 99 can simulate the brine production through wells, but not the production of heat via the WBHX. The 100 heat transfer in the deep borehole heat exchanger has been simulated using GEOPIPE (Alimonti and 101 Soldo, 2016), a pure conductive semi-analytical approach based on thermal resistances and the 102 Fourier equation. An iterative approach has been used to couple the two simulators. Then, the 103 simulation has been carried out using only the pure conductive semi-analytical model.

104
The paper has two main targets: to understand if the pure conductive simulation is a too 105 precautionary condition in presence of high temperature convective structures in the ground and to 106 study the sustainability of heat production via a zero-mass extraction device in the area of Campi    The area of Mofete has been selected for the scope of this paper, having very high geothermal 119 gradients (100−170 ℃ km -1 ) resulting from the AGIP campaign (Fig. 4). The results indicate the 120 presence of three main aquifers, of which the two shallower are the productive ones. The first aquifer 121 is at the depth of 500-1000 m and has 20% in weight of non-condensable gases, the second one has a 122 higher content of vapour (40%) and it is located between 1800 m and 2000 m (Fig. 4), the last aquifer  This important quantity of heat is generated in the reservoir of Campi Flegrei, which contains a hot 129 and saline geothermal system. According to Berrino et al. (1984) and Woo and Kilburn (2010), there 130 is a relatively shallow magma sill at the depth lower than 3−4 km, whereas at the depth of 8−10 km 131 is located the greatest magmatic source (Fig. 5), with a thickness of ~1 km and a diameter equal to 132 that of the caldera (Zollo, 2008). The heat transfer in Campi Flegrei reservoir depends on the 133 permeability: in the first kilometre the fracturing system produces a high hydrothermal circulation, 134 so the advection is the main type of heat transport; between 1000 and 1800 the heat moves driven by   The semi-analytical model of BHE 161 The GEOPIPE simulator is a semi-analytical model based on thermal resistances (see Fig. 6). The The energy balance of the deep borehole heat exchanger is expressed by the following relation: where, ̇ is the total heat exchanged by the working fluid with the ground.

185
The following set of differential equations is numerically solved finding the outlet temperature Regarding the calculation of the ground temperature with depth, GEOPIPE can operate in two 207 modes: the operator can assign a geothermal gradient and the ground temperature at 0 , so the 208 simulator produces the temperature profile; the operator assigns the temperature profile directly.

209
This second option is that one followed for our analysis.       Table 5 shows the grid 301 composition.  Due to the complexity of the geothermal field and the coupling with the WBHX, severe boundary 303 conditions have been implemented (see Fig. 9). The flow boundary conditions specify the mass flux Method B (Fig. 10 (b)) uses a constant thermal power scenario which is more realistic by the   Figure 14 illustrates the progressive enlargement of the thermal disturbance in time. The influence 374 radius after 1 month of operation ( Fig. 14 (a)) is about 10 m, it reaches 25 m after 6 months and it 375 remains stable since 1 year (Fig. 14 (b) and (c)). The thermal radius reaches 50 m after 3 years of heat 376 extraction with a DBHE (Fig. 14 (d)). The discussed results show a massive temperature decrease 377 and a considerable thermal interference that seem inconsistent with the use of deep borehole heat    The graph in Figure 17 indicates that the pure conductive approach produced a steady-state   These two zones are characterized by different values of rock density (see Tab. 4) and so of thermal 426 diffusivity, which affects the temperature gradient between the ground and the fluid in the annulus.

427
As can be seen in Fig. 19 the heat production curve is the reflecting curve of the temperature 428 difference curve. The strong numerical difference of heat production values respect to those of the 429 constant flow rate scenario is explained by the refinement of the mesh: this paradox highlights the 430 complexity of defining the right volume to refer the heat production for SHEMAT input file.   The present work evaluates the production of geothermal energy with a zero-mass extraction device 455 applied in the volcanic area of Campi Flegrei, Italy. The area is characterized by very high thermal 456 gradients, but this potential is underused, and the geothermal sector meets social resistances.

457
Therefore, the possibility to produce geothermal energy, avoiding all the risks related to the brine 458 extraction, seems to be very interesting in this area.

459
The paper is focused on the identification of the most correct simulation method to study the thermal   The second scenario estimates a negligible decrease of temperature, with maximum values of 2 ℃. 476 The results indicate that the heat extraction with a WBHX operating at a fixed thermal power of 850 477 kW does not affect the thermal field of Campi Flegrei reservoir, probably due to the recharge effect 478 generated by the convective structures of the reservoir, so the plant is sustainable in time.

479
The output of the pure conductive model is very different, not simulating any thermal source into