A review on geothermal wells: Well integrity issues

Geothermal energy is an important potential and a strategic area for developing activities regarding renewable energy and future studies. It involves a great potential and a main role in the worldwide energy sector, particularly electricity generation. Nevertheless, the role of geothermal electricity in today’s world is not dominant in comparison to other renewable energy sources like solar PV and wind power. Though Enhanced Geothermal Systems (EGS) greatly includes the potential for power generation, the EGS technology feasibility should be further studied. Numerous studies exist to investigate the energy production feasibility (heat and electricity) from such resources focusing on drilling processes. Today, the geothermal projects are mainly focused on completion operations and drilling research area, in which it is possible to reduce the costs considerably by suitable design. To drill a geothermal well, the drilling operations are quite similar to the petroleum industry. Though, the different aspects of geothermal drilling will make it more challenging and complex than the oil and gas well drilling. The type of resource and the formation type of resource is the big differences. Normally, fissures and fractured hard volcanic rocks contain geothermal resources, however, porous media of sedimentary formations comprises oil and gas resourses. Petroleum and geothermal well drilling have different geological and physical features including rock types, reservoir pressure, temperature, well and casing design as well as the well completion program. Hence, in this paper, it is tried to present the most common issues in geothermal well drilling operations and to introduce the potential for the future research areas. © 2020 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).

activities regarding renewable energy and future research. Since billions of years ago, the earth's core produces and radiates heat. The earth center has a temperature of 5500 C, the same as the sun's surface temperature (Fig. 1). A fundamentally limitless and renewable source of energy is created by this constant heat flow over the billion years (Blodgett, 2014).
During the centuries, geothermal energy has had various usages. The water from hot spring was used by the Maoris in New Zeeland and native Americans for medical purposes and cooking over thousands of years. Romans and Ancient Greeks possessed geothermally heated spas. People in Pompeii by living near Mount Vesuvius tapped hot water from the earth for heating their buildings. Geothermal waters were utilized by Romans to treat skin and eye diseases. The geothermal spas have been enjoyed by Japanese for centuries" (Nersesian, 2010). Though, the geothermal energy was commercially used for district heating first in Boise, Idaho, the USA in 1892 (US Department of Energy). Later at the 19th century, developing thermodynamics continuously led to the generation of power (electricity) from geothermal heat through transforming the hot steam energy in mechanical energy and then into electricity via generators and turbines. Certainly, the generation of geothermal electricity is related to the Larderello in northern Italy (Stober and Bucher, 2013) with the first geothermal discovery for industrial purposes.

Use of geothermal resource
Based on the Lindal diagram (Blodgett, 2014) (Fig. 2), the potential utilization of geothermal fluids is represented at various temperatures that are valid today as well. Nonetheless, over the temperature of 85 C, a binary cycle plant can generate electricity now (Dickson, Fanellib).

i. Direct heat Use
The most common and oldest utilization of geothermal energy is direct use. District heating, space heating, agricultural and greenhouse applications as well as industrial applications are the most common forms of utilization worldwide (Dickson, Fanellib). In Iceland, space and district heating has been progressed vastly, where geothermal direct use had a share of 66% of primary energy usage in 2011 (Dickson, Fanellib); which has been increased to 96% of heating and cooling energy consumption in 2015 (Breembroek et al., 2013). It is interesting to know that district heating using geothermal energy is saving about 7% of Iceland's GDP. Fig. 3 shows a basic flow diagram of the geothermal district heating system of Reykjavik (Gudmundsson, 1988).
Using geothermal energy for buildings air conditioning has been vastly grown since the 1980s, by the introduction of heat pumps; which allow extracting and utilizing the heat from near-surface with the sources at low temperatures, such as the ground itself, shallow aquifers and ponds (Fig. 4). Although electricity is needed for operating the heat pumps, in favorable climate circumstances and also an optimized design, there will be a positive energy balance always (Dickson, Fanellib) (Fig. 5).
Recently heat pumps became very popular all over the world, especially in the USA, Germany, and Switzerland (Dickson, Fanellib). Tables 1e3 shows the global leaders of the direct use of geothermal energy (John and Boyd, 2016).
The five countries, which are the leaders of geothermal energy direct utilization, use 65.8% of the whole world capacity (John and Boyd, 2016).
Using heat pumps has been vastly grown during the past years. The largest share of geothermal direct usage belongs to geothermal heat pumps with 55.15% of total annual energy usage and 70.9% share of installed capacity in 2015 (John and Boyd, 2016). However, as the International Energy Agency's stated in the latest report on world energy outlook (Agency and International Energy, 2018) only 3% of EU energy demand building for heating is provided by heat pumps.  ii. Electricity Generation Geothermal primary energy can be used to generate electricity as well. Generating electricity is experiencing continuous growth worldwide. In 2011, Bertani forecasted that the installed capacity of geothermal power plants for generating electricity would be raised to 18,500 MWe by 2015 (Zarroukh and Moon, 2014). However according to an updated report released in 2015, worldwide geothermal installed capacity for electricity generation was 12,600 MWe, producing 73,549 GWh of energy in 2014 (Bertani, 2015). Fig. 6 shows the trend of geothermal electricity growth from 1950 to 2015 (2020 predicted) (Bertani, 2015).
During the past years several studies had been carried out to investigate the potential of geothermal energy production. Specially in Europe a recent studiy predicted of a rapid rise in geothermal applications by 2050 (Longa et al., 2020). Fig. 7 shows the potential of power generation by geothermal energy which are constrained to resources in depth below 2 km (Iceland and some other countries with volcanic activities have been excluded). On the other hand, direct heat utilizztion can be reached through depth lower than 2 km (Longa et al., 2020). Moya et al. (2018) has provided the latest development in geothermal industry. The aspects of the development in power plant technology and direct heat utilizations have been carried out in the study. Geothermal Utilization has been introduced as an alternative for current fossil feuls so the emissions can be decreased significantly through the coming years. Also direct heat utilizations could be used to increase the revenue of a geothermal project (Moya et al., 2018).
Nevertheless, despite the great benefits, various issues exist currently impeding a wider uptake of geothermal wells including the aggressive nature of geothermal fluids and high temperatures, the high costs of investment and maintenance of geothermal wells, as well as the high thermal strains on cement and tubular. Instead, the hydrothermal resources are partially available just in some geographical regions. The existing geothermal resources are mostly in the low-temperature class (less than 160 C) (Fig. 8).
Regarding these challenges, this paper is mainly aimed at investigating well integrity issues related to the geothermal wells and establishing knowledge regarding transfer between geothermal energy and petroleum industry. Hence, it is tried to present the most common issues in geothermal well drilling operations and to introduce the potential for the future research areas.

Geothermal energy: subsurface
As stated earlier, geothermal energy potentially plays the main role in the energy sector of the world and the reduction of carbon dioxide footprint. Nevertheless, geothermal electricity has now a quite low share in comparison to other renewable energy sources (Fig. 9). It is not caused by the geothermal cost since on a Levelized scale it may even be cheaper than other renewables (Clauser and Ewert, 2018), however, the reason is the availability of hydrothermal reservoirs in the earth.
Instead, EGSs (mostly hot dry rocks) that are more costly compared to the usual hydrothermal resources based on USD/Mwh, exist almost accessible in all places on the earth and are able to offer clean energy for all for hundreds of years.
EGS includes a huge potential to generate power over 70 GWe by 2050. Though, further studies and developments are required for the feasibility of EGS technology. Recently, ENGINE a research group in Europe (The Enhanced Geothermal Innovative Network for Europe) including 35 R&D sub-groups, researches regarding EGS aimed at reducing drilling costs by 20e30%, exploration expenses of a geothermal project by 20%, and enhancing the efficiency of generation of electricity from the heat by 20% by 2020 to increment the popularity of EGS in Europe.
Today, completion operations and drilling research areas are more emphasized in geothermal projects where the costs may be reduced and efficiency significantly increased through good design.

Geothermal drilling
The operations in drilling a geothermal well has high similarities to the drilling operations in the petroleum industry. Nevertheless, based on numerous aspects, geothermal drilling is different making it more challenging and complicated in comparison to drilling the oil and gas wells. The kind of resource and the formation type are huge differences. The fissures and fractured hard volcanic rocks normally contain geothermal resources, however, sedimentary  (Gudmundsson, 1988).

P. Allahvirdizadeh
Journal of Cleaner Production 275 (2020) 124009 formations porous media includes oil and gas recourses. The petroleum and geothermal well drillings are different based on geological and physical features like rock types, reservoir pressure, temperature, casing, and well design, as well as the well completion program (Moya et al., 2018). Today, completion operations and drilling research areas are more emphasized in geothermal projects where the costs may be reduced and efficiency significantly increased through good design. Over numerous decades, the cost of developing a geothermal field was estimated within 4.5 M USD-5.5 M USD per installed MW and 50% is spent approximately on wells drilling and completion (DiPippo, 2016). Here, the main issues are highlighted for a geothermal well drilling in this paper.

Circulation loss
Loss of circulation is the lost part of or, the whole drilling mud in severe circumstances in the drilled formation (Bugbee, 1953). The formations with high permeability including faulted, fractured, and joint formations are the zones potentially with the possible occurrence of loss of circulation. Normally, the circulation loss is

P. Allahvirdizadeh
Journal of Cleaner Production 275 (2020) 124009 expected in geothermal well drillings (Shryock, 1984). Drilling fluid circulation lost is the main and oldest problem in a drilling operation directly affecting the drilling cost (Bugbee, 1953). It will increment the non-productive time resulting in the increased cost. In a study (Rehm et al., 2009), it was indicated that the circulation loss-related costs were around 10% of total non-productive time spent in the Gulf of Mexico within 1993e2003. Since drilling fluid and lost circulation substances are normally expensive (25e40% of overall drilling cost is related to the costs of drilling fluid (Lecolier et al., 2005)), the loss o these materials can significantly affect the costs (Lavrov, 2016). Moreover, it was reported that circulation loss costs are 10e20% of the overall drilling operation

Table 1
The leaders in direct utilization of geothermal energy in the world (such as heat pumps)-2015 (John and Boyd, 2016 Table 3 The leaders in direct use of geothermal energy in the world (excluding heat pumps)-2015 (John and Boyd, 2016  costs (Mansure et al., 2002). Based on a report (Finger and Blankenship, 2010), the circulation loss costs in the USA in geothermal drilling operations is approximately 10% of overall well costs and 20% for exploration wells. In the Hengill area in Iceland, 75% of drilled geothermal wells encountered the problems of circulation loss and wellbore collapse (Sverrir., 2014). These problems can lead to the poor cement job endangering the well integrity. The circulation loss is a complex challenge requiring a multidisciplinary method like well design, mechanical analysis of rock, drilling fluid design such as lost circulation materials and loss prevention, and drilling hydraulics (Whitfill and Hemphil, 2003). Fig. 10 represents a time breakdown of a sample well in Indonesia faced with a lost circulation problem. The entire time spent to fight the circulation loss problem was 194 h in that particular well, 30% of the overall operation time. Goodman (1981) had interviewed petroleum and geothermal operators, drilling service consultants, and companies to evaluate the circulation loss issue and possible solutions to prevent it. The report indicates that implementing conventional petroleum industry methods may or may not overcome the lost circulation problem. Using lost circulation materials (LCM) is usually the first choice. Cement plugs and blind drilling can be implemented in severe low pressure-fractured formations. However, using cement plugs may not be successful either due to unknown downhole temperature to calculate the amount of retarder to be added to the slurry. In some geothermal operations, sodium silicate found to be effective to be pumped ahead of the cement slurry. Also, tie back string is recommended in geothermal drilling operations to overcome the lost circulation problem.
There are several studies investigating the proper cement to plug the loss in geothermal drilling conditions. A summary of the studies is given below in Table 4. Several investigations were performed to investigate the potential of polymers to combat the lost circulation challenge. The following section is a summary of the studies investigated polymers as a potential solution that are reviewed in this paper. Hashmat et al. (2017) studied the composition of polyacrylamide, phenol, and formaldehyde as a gel. It was successfully tested at room temperature and moderately high pressure. Mansour et al. (2017) introduced a novel lost circulation material of a thermoset shape memory polymer, 1 which activates when  1 The shape memory polymer is able to change into a temporary shape and returns back to its original shape by stimulation by an external factor like temperature change or electromagnetic wave (125).

P. Allahvirdizadeh
Journal of Cleaner Production 275 (2020) 124009 exposed to loss zone temperature (Smart LCM). They conducted experiments to investigate the efficiency of the smart LCM in sealing the fractures with the activation temperature of 70 C. Results showed that using such materials can seal the loss zone very effectively. Following this study, Mansour and Dahi (Mansour et al., 2018) introduced another smart LCM made of anionic shape memory polymer and also a numerical model to see the effect of particle and fracture size which can be a solution for lost circulation problem with the ability to seal big fractures and not plugging the drilling tools. Silva et al. (2017) also presented an algorithm modeling the loss of circulation in fractured formations, which can solve mathematical models to analyze possible solutions of differential and analytical equations for fluid flow, formation properties, fracture behavior, and drilling fluid rheology. With this info, one can determine the fracture real-time width to design drilling mud with proper weight and LCMs.
Although there are numerous studies investigating new materials performance in downhole conditions as some have been mentioned above, there still are not many studies in geothermal conditions.
In a petroleum drilling operation, if the loss of circulation could not be solved using lost circulation materials, usually cementing into the loss zone may be the solution. However, cementing the loss zone may not be the best solution in geothermal drilling operations, where usually several cement jobs are needed; which each would take up to 10e24 h (Shryock, 1984). So new methods and procedures need to be followed and documented when drilling into a loss zone with major fractures. Mata and Veiga (2004) reported that cross-linked types of cement (CC) may be useful where conventional LCM could not overcome the lost circulation issue, which can be used up to 121 C. Mofunlewi and Okoto (2016) introduced a new-engineered spacer to be pumped before cement slurry where LSM does not solve the issue, to seal the zone with fractures or high permeability causing severe losses. The slurry contains hydrophobically modified polysaccharides which seal the fracture by forming a bridge at fractures. Fig. 11 shows the process of these micelles, which have hydrophilic heads and hydrophobic tails. Results showed that using this spacer could improve the cementing job quality by sealing the fractures and formation permeability to decrease the losses of circulation significantly. In the case of geothermal drillings where fractures and lost circulation is common, this might be a solution, however, the behavior in geothermal conditions have to be investigated.  -An optimum concentration of polymer-crosslinker exists to obtain a stable and strong gel; -The gel strength will enhance by increasing aging time (6 h and 1 day) and temperature (100 C and 120 C).

P. Allahvirdizadeh
Journal of Cleaner Production 275 (2020) 124009 In some geothermal drilling operations, the reverse circulation technique might be useful to overcome the loss of circulation cementing jobs. Moore et al. (2005) tested the reverse-circulation placement method in a deep well and a shallow geothermal well successfully for cementing a job where the loss of circulation was the issue. Lichang et al. (2006) also reported a successful application of reverse circulation drilling in geothermal drilling operations in China with lost circulation problems.
Controlled and managed pressure drilling (CPD/MPD) can be also a solution where lost circulation occurs. Sammat et al. (2012) described a successful application of CPD in a geothermal field in Germany where wells were suffering from lost circulation and stuck pipe issues. Formation damage significantly was reduced by the application of CPD. Singh et al. (2016) also reported implementing of MPD technique in Vietnam where lost circulation was a major problem in HPHT well cementing. They reported that MPD cementing in a closed-loop had been applied successfully in all wells drilled within the study.
There are several studies addressing different solutions for curing the lost circulation problem in drilling operations, but mostly suitable for normal conditions and very few studies investigated the behavior of such materials under severe geothermal conditions. More research and study in this area is needed.

Well cementing
This operation is aimed at creating an annular barrier in the annular space within wellbore and casing to struggle in definite environmental circumstances and holding the casing in position. It is also responsible to protect the casing from formation fluids (Ostroot and Shryock, 1964). To prevent casing expansion over production in geothermal wells, the casing needs to be completely cemented. In production phase of a geothermal well, where completion components are installed on, the casing's role becomes more critical, since it may lead to the vibration at the surface and even blowouts due to due to expansion of the fluid/steam flow (ew Advanced Method for, 1995).
The most critical issues for the geothermal include (Shryock, Smith;Salim and Amani, 2013): Circulation loss and thick filter cake probably reducing cementformation bond strength; High temperature resulting in strength retrogression of the cement; Presence of corrosive components like CO 2 .
Nevertheless, regarding durability, the cement stability at high temperatures leads to well integrity problems over the well life cycle (Ostroot and Shryock, 1964). The temperature is the most imperative element as a result of a robust impact on cement slurry setting time and its impacts on cement strength over the well's lifetime (Guerrero, 1998).
There exist numerous studies in the literature referring to cement behavior in different well conditions, however just a couple have discussed the geothermal high-temperature conditions and cement behavior in such severe conditions. Some of the cement compositions may show sufficient strength at the early stages of setting but will lose their strength after a while due to continuous exposure to high temperatures (Carter and Smith, 1958;Ludwig and Pence, 1956;Ostroot and Walker, 1961).
Proper cement composition for high-temperature conditions and cementing methods and equipment have been widely studied and promoted in the literature. Table 6 represents a summary of the papers reviewed in this study.
As reviewed above, numerous studies exist to address the problems related to the cementing operations and cement behavior itself; however, there are few studies on the impact of cement curing temperature, corrosive components, and thermal loads on final hardened cement. Moreover, the cement performance at high temperatures was rarely discussed (over 350 C). The effect of rock type on cement performance and durability was not investigated much, though, few articles are available in this regard (Silva and Milestone, 2018). . Working mechanism of Sealbond Spacer: (a) The hydrophobically modified polysaccharides within the aqueous fluid from micelles. A dynamic equilibrium exists between adsorption at the formation and dissolution in the aqueous media. (b) Increasing differential pressure results in increasing number of micelles adsorb and realign along the porous formation. (c) At maximum differential pressure, the adsorbed micelles have turned into a film, completely sealing the pore. (d) When a flow forms in opposite direction easily returns the micelles into solution again (Bottiglieri et al., 2014).

Casing
The casing has a main role in the ultimate well cost. Therefore, selecting the suitable casing type (length, grade, and size) is very important based on the requirements and geological data (Capuano and DiPippo, 2016) (Table 9). The casing has numerous performances in a well including (Applied Drilling Engineer, 1986): Isolation of the well fluids from formation and formation fluids; Preventing of borehole collapse while drilling; Offering a clear pathway to the drilling fluid; Minimization of the damage to the subsurface environment; Normally, in casing design, it is essential to clarify the surface and bottom hole location of the well (Applied Drilling Engineer, 1986). It is also essential to consider the number and sizes of casings/tubing, setting depth, and grades of each casing in the design. The following parameters are important in the geothermal well casing and need to be identified in the casing design (Capuano and DiPippo, 2016): Rock temperature; Resource kind (steam, liquid, or both) and temperature; The necessity of surface equipment or downhole; Preventing directional drilling if possible; Presence of corrosive components.
The following casing properties should be considered to meet the criteria of geothermal casing design: Table 7 presents the American Petroleum Institute (API) grading system for the casing. Table 8 also shows the list of non-API casing grades which are commonly used in the oil and gas industry.

Grade
In most of the geothermal environments, the presence of hydrogen sulfide (H 2 S) is probable which will induce corrosion cracking issues to the casing. Only some of the casing grades are H 2 S resistance, resulting in limiting the casings' designs for geothermal well drilling and completion operations (Capuano and DiPippo, 2016). Later in this paper proper casing selection for these conditions has been discussed.

Tensile strength 4
The minimum tension force needed to damage the casing body permanently (F ten ) can be obtained from the following equation: F ten ¼ s yeild *A S 1 where s yeild is the minimum yield strength of the pipe in psi, and A S is the cross-sectional area of pipe in in 2 (see Fig. 12). A most commonly used safety factor of 1.75 for tension forces should be considered.
So the final equation will be: Always the greatest tension load is in the topmost casing; which can be calculated by multiplying the total length of casings hanging below by its weight per foot (buoyancy factor should not be considered in the calculations).

Burst strength 5
Burst pressure is the minimum pressure needed to damage the pipe permanently by bursting. Fig. 13 shows a free body diagram of the casing section exposed to burst pressure, P br .
If it is considered that there is no external pressure and axial loading, then the force F 1 , resulting from the internal pressure of P br , and acting on the projected area of LdS, can be obtained from the following equation: and the counterforce F 2 , resulting from casing wall strength, s s , over the casing area of tL, will be obtained from the equation below: After solving the above-mentioned equations for static condition (F 1 À 2F 2 ¼ 0), P br will be: For the worst-case scenario with an allowable deviation of 87.5% of nominal casing wall thickness, the burst pressure in casings would be: where d n is the nominal wall thickness of the casing. The most commonly used safety factor in casing burst pressure calculations is 1.25.

Collapse resistance
Collapse occurs when the external pressures exceed the casing wall strength while it is empty (zero fluid pressure inside the casing). Collapse resistance of the casing is more related to geometry and casing wall thickness rather than casing grade (Applied Drilling Engineer, 1986). Usually, collapse resistance is the first design factor in casing design since it is the lowest design factor. Normally a safety factor of 1.15 is considered for collapse strength (Capuano and DiPippo, 2016).
Calculating collapse pressure is much more complicated. To see P. Allahvirdizadeh Journal of Cleaner Production 275 (2020) 124009 Table 6 Reviewed papers with a focus on geothermal cementing related issues.
Gallus et al. (Gallus et al., 1979) Temp. ( C): 204e399 Material/method: API class G and J þ additives (silica flour with a concentration of 40e100 wt %; grinds and silica sand, 60e170 or 20e40 mesh, with a concentration of 40e106 wt %; expanding perlite with different amounts; fly ash, with and without silica) Results: API class G and class J cement with the addition of silica (40e80%) may be applicable to geothermal wells Nelson et al. (Nelson et al., 1981a) Temp. ( C): 204 Material/method: Different cement compositions Results: Using Portland cements of normal density (with the addition of silica flour) for geothermal well cementing operations at 204 C of BHCT would be the best choice. Nelson et al. (Nelson et al., 1981b) Temp. ( C): 204 Material/method: Different cement compositions þ calcium silicate hydrate Results: Truscottite and pectolite were identified as potential binders for geothermal conditions to increase the thermal and chemical degradation resistance. Ostroot  Temp. ( C): 260e371 Material/method: API Class A, E and G cement Results: -Using a strength stabilizing agent, such as silica flour, is necessary to achieve an acceptable compressive strength for cement compositions at all casings cement jobs; -API Class E cements are not necessary for geothermal conditions (high temperature). -Nor low-dose neither high-dose NSS cannot enhance the cement's compressive strength at high temperatures; however, the addition of low dose may prevent strength retrogression at long curing times; -By adding 4e6% of NSS to a slurry of class G cement including 35% of silica flour, compressive strength will significantly improve. Chow and Kalousek (Chow and Kalousek, 1976 Using SiO 2 rich material is necessary for cement design of temperatures above 110 C without which compressive strength due to retrogression decreases significantly. However, at low temperatures, silica flour mostly plays as a filler of the composition, which does not affect the mechanical strength positively. API class G cement þ silica (with a concentration higher than 35% with different solid volume fraction and particle size) Results: Matrix chemical composition, phase composition, porosity, pore size distribution was characterized as critical factors in cement job design in deep wells with high pressure and temperature. Fridriksson (Fridriksson, 2017) Temp. ( C): 200 Material/method: Class G cement þ different additives Results: Adding acid-treated silicone rubber would increase the bond and compressive strength when exposed to high temperature. Won et al. (Won et al., 2015) Material/method: Physical properties (uniaxial compressive strength, profitability, thermal conductivity, bleeding potential) of G-class cement Results: -To obtain acceptable groutability, the water-cement ratio should be very high which will reduce the structural strength of the cement accordingly; -The thermal conductivity of G-type cement will decrease by increasing the water/cement ratio at 20 C and 50 C, which may reduce heat loss during production or injection operations. Sugama (Sugama, 2006) Temp. ( C): 200e300 Material/method: Sodium silicate-activated slag (SSAS) cement at H 2 SO 4 and CO 2 rich environment Results: -At 300 C, an unwanted porous microstructure was created by further growth of well-formed tobermorite and xonotlite crystals generated leading to the retrogression of strength and improving the water permeability; -A 20 wt% sodium silicate solution (Na2O/SiO2 mol. ratio of 3.22) was used as the alkali activator, to autoclave the SSAS cements at temperatures higher than 200 C exhibited a minimum water permeability of less than 3.0 x 10-5 darcy and exceptional compressive strength of more than 80 MPa. Hole (Hole, 2008) Material/method: Comparing cementing techniques, material, and equipment in the petroleum industry and geothermal operations. Results: -API Class A or G cements with appropriate additives are commonly used in geothermal applications; -Class G cement with the proper additive of silica flour (up to 40%) would provide an acceptable strength against retrogression of cement (not appropriate in CO 2 rich environment); -A mixture of Portland Class A cement blended with blast-furnace slag in the ratio of 70:30 provides a high corrosion resistant cement. Ravi et al. (Ravi et al., 2008) Material/method: Primary cement operation in Indonesia Results: -Cause of cement job failure was: a) structural failure of the usual cement sheath, b) inefficient hole cleaning of drilling fluid and cement slurry placement and/or c) poor cutting transportation during the hole cleaning operation; -A new elastic cement system was designed and used combined with Industry-recognized best practices for displacement and hole cleaning to overcome the integrity issues for future wells.
(continued on next page) P. Allahvirdizadeh Journal of Cleaner Production 275 (2020) 124009 more details regarding collapse pressure calculation refer to (Applied Drilling Engineer, 1986).

Casing weight (lbm/foot)
Casing weight can be calculated by multiplying casing length by casing weight per foot which are available in API tables for each casing grade.
Besides the pre-mentioned criteria for casing design, the increased temperature is the main role player in geothermal drilling operations. A portion of the strength of steel pipes is lost at higher temperatures. For higher grades of casings, this loss is further even (Finger and Blankenship, 2010). For instance, by increasing the temperature from 25 C to 371 C, a reduction occurs from 632 MPa to 484 MPa in the yield strength of an L-80 grade casing. However, at the same temperature range, a yield strength in the K-55 casing will reduce from 388 MPa to 359 MPa (Snyder, 1979). The yield strengths of various casing grades as a function of temperature are represented in Fig. 14 (Larri, 1997).
It is important to take into account the presence of corrosive components in the geothermal project. CO 2 and H 2 S are mainly present in the geothermal reservoirs. Hence, selecting the materials  Gallus et al. (Gallus et al., 1979) Rickard et al. (Rickard et al., 2012) Material/method: Secondary cementation method Results: Top squeeze and top fill methods might be an acceptable solution in geothermal well cementing operations. Berard et al. (Berard et al., 2009) Temp. ( C): 288 Material/method: Foamed cement Results: Using calcium aluminate phosphate cement (CaP) can be resistant in conditions of high carbonic acid ratio. Additionally to obtain a lower density one could add nitrogen or air to the slurry which may be useful to reduce the risk of loss of circulation. Santra et al. (Santra et al., 2009) Temp. ( C): 204 Material/method: Magnesium-based Sorel cement slurries Results: -A new retarder is introduced to create the possibility of pumping Sorel cement based slurries at a downhole temperature of 204 C; -No appreciable strength retrogression has been observed Omosebi et al. (Omosebi et al., 2015) Temp. ( C): <221 C Material/method: Temperature effect on cement compressive strength in an acidic environment of high CO 2 gas concentration (up to 100%) Results: -Temperature plays a major role in dynamic cement degradation in an acidic environment; -Maximum degradation was observed to take place at temperatures between 107 C and 177 C. Philippscopoulos and Berndt (Philippacopoulos and Berndt, 2002) Material/method: Structural performance of the cement by calculating the cement response to pressure and temperature loads Results: A model-based design should be developed with a focus on the structural response of geothermal wells to different loads. Sugama et al. (Sugama et al., 2012a) Material/method: Cement composition of Class F fly ash and sodium silicate under thermal shock Results: -Cement sample was autoclaved at 200 C then heated up to 500 C by air for 24 h and then continuously immersed in 25 C water; -Type G cement with and without Type F fly ash and quartz flour was failed and generated multiple cracks. Sugama et al. (Sugama et al., 2012b) Temp. ( C): 200 Material/method: Corrosion-resistance of modified cement mixture for carbon steels by adding foaming agent and acrylic emulsion as a corrosion inhibitor Results: An optimum concentration of additives is introduced to achieve a desirable slurry density lower than 1:3g=cm 3 , compressive strength higher than 500 psi, and brine caused corrosion rate below 70 mpy of CS. Pyatina et al.  Temp. ( C): 270 Material/method: Self-repairing performance of Thermal Shock Resistant cement (TSRC) and Ordinary Portland cement with SiO 2 blend (OPC) Results: Although OPC showed a higher compressive strength, the recoveries for TSRC was much higher (above 80% while OPC recovery was above 50%). Pyatina and Sugama  Temp. ( C): 90 Material/method: Calcium Aluminate cement/Fly ash F blends in an acidic environment Results: -18 days of sulfuric acid exposures at 90 C to all samples (hydrated for 24 h at 300 C) caused weight and diameter loss in TSRC; -Class G/silica blend gained more mass and increased diameter owing to the creation of a gypsum layer. Vidal et al. (Vidal et al., 2018) Temp. ( C): 300 Material/method: Effect of adding rice husk ash (RHA) instead of silica flour to cement compositions Results: -RHA as an anti-retrogression additive to cement slurry; -Using class G cement with 40% of RHA cured at 300 C resulted in an increment in compressive strength of the composition by 11%. Paiva et al. (Paiva et al., 2018) Material/method: Cement composition with geo-polymers Results: A formulation with a metakaolin-potassium basis with the addition of microsilica, retarder, and mineral fiber was designed which resulted in an increase in tensile strength, higher thermal durability, and accordingly improving mechanical performance in high temperatures. Da Silva et al. (da Silva et al., 2018) Temp. ( C): 300 Material/method: Effect of polyurethane addition on thermomechanical features of Portland class A cement Results: Improved casing and cement sheath thermal mismatch. Teodoriu et al. (Teodoriu et al., 2013) Material/method: Cements exposed to cyclic loading conditions Results: A stress analysis presented using an analytical method and finite element solution to introduce a new cement specimen that is suitable for HPHT conditions. Shaughnessy and Helweg (Shaughnessy and Helweg, 2002) Temp. ( C): 204 Results: Optimized cementing method and procedure.

P. Allahvirdizadeh
Journal of Cleaner Production 275 (2020) 124009 should be in a way to survive in such circumstances for the well's expected lifetime. CO 2 corrosion in the Imperial Valley of California had the responsibility to corrosion rate of 3 mm of the casing yearly leading to the plugging the well followed by 10e12 years that is much less than estimated (Finger and Blankenship, 2010). Most of the wells in the Imperial Valley of California were now retrofitted or completed with titanium casing to overcome the issue. Although the result was satisfactory, it enforced a very high investment expense (around 3000 USD per 1 m of casing) (Finger and Blankenship, 2010). Snyder (1979) has discussed the casing failure modes in shallow to moderate geothermal wells with temperatures varying from 148 C to 343 C. Following issues were observed to be the most common problems for casings: o Strength loss due to temperature elevation; o Mechanical wear of casing inner side; o Buckling due to thermal stress and pure cementing job; o Corrosion (internal and external) and scaling. Chiotis and Vrellis (1995) presented the failures and issues observed in deep geothermal wells over 15 years in Greece. The reservoir had a temperature of 300 C with high salinity. Following failures were observed in the wells: o Well head damage due to thermal expansion and poor cementing of the casing; o Decoupled casing joints due to thermal stress; o Buckling of the casing at some interval in well; o Leakage in an abandoned well after 18 years due to casing corrosion.
Thermal stress during the production and cooling of the well was mostly responsible in these cases. However, these could be avoided by proper cementing, slow preheating the well before production, and good casing design. Chiotis and Vrellis proposed a  neutral temperature factor for casing design, which would ensure that the selected casing would not face with plastic deformation during the well life. Southon (2005) studied the different failure modes in the casing with a specific focus on the expansion of the trapped fluid in the casing-to-casing annulus. It was indicated that using a tie-back liner might eliminate the issue; however, the tieback system itself may implement other challenges. The study stated that implementing the construction and completion techniques properly will most probably reduce the chance of such issues.
Teodorio and Falcone (Comparing completion desi, 2009) presented an experimental and theoretical model for investigation of low-cycle fatigue (LCF) resistance of an N-80 grade casing with a diameter of 18 5/8 in with buttress thread connection. The results indicate that in a geothermal environment, under severe loads, the LCF resistance of buttress connections in the casing could be very low (10 cycles). However, the temperature variation effect on other casing connection types and grades were not investigated in this study. Torres (2014) discussed how to consider induced loads due to thermal cycling in geothermal wells. He also investigated the casing failure mechanisms by actual and numerical tests. Results showed that casing collapse by the expansion of trapped water due to poor cement job is the most common issue in tieback strings. Also in shallow sections, failures in casing connections and collapse of the casing itself due to the same issue occurred. This issue can be avoided by effectively placing the cement and selecting proper casing centralizers and grades. Liu et al. (2015) also indicated that integrated casing stress analysis with thermally induced stresses provides a perfect casing mechanical analysis. Kaldal et al. (2015) studied the wellhead equipment displacement due to thermal stresses using finite element modeling. The model was validated by data of five different geothermal wells in Iceland. The model can be used further to study the structural analysis of casings exposed to different loads. Using the model, Kaldal et al. (2016) studied the structure of casing in deep geothermal well in Iceland. The results indicated that the thermal stress of cycling has the most effect on the production casing. However, changes in casing thickness and stiffness impose additional stress and strain on neighboring casings. Teodoriu (2015) described casing failure models. The study listed the most issues resulting in casing failure as follows: o Failure of cement exposed to cyclic loads; o Failure of casing material exposed to cyclic loads; o Local bucking in casing connections; o Corrosion. Corrosion Prior to drilling the deep production wells, the geothermal fluids' nature is normally not known (Nogara and Zarrouk, 2018). The nature of the geothermal fluid is a challenge for the geothermal industry, which may usually include dissolved CO 2 and H 2 S, leading to the corrosion in surface and wells facilities (Gunnlaugsson et al., 2014) (Ragnarsdottir et al., 2018). At downhole circumstances, the corrosion rate is usually worse than the well sections close to the surface (Treseder and Wieland, 1977). In geothermal drilling, the issues related to the drill-string corrosion or the nature of the casing are typically the same as the sour oil/gas drilling (Ellis et al., 1983). Pitting, wear, corrosion fatigue, and erosion are the main problems (Fig. 15). The strength of the pipes is reduced by these corrosion-related issues. In geothermal circumstances, it becomes worse owing to the high temperature and typically the existence of H 2 S (Ellis et al., 1983).
Non-metallic casings were investigated for low to intermediate geothermal wells by the National Water Well Association of the US. All the suggested substituents were inert to the corrosive components within geothermal fluids, although their temperature resistance and mechanical strength are lower compared to the steel casings (Ellis et al., 1983) (ASSOCIATION andNATIONAL WATER WELL, 1979).
Another experimental study had been carried out on industrial drill pipes by Bareteri et al. (Barteri et al., 1996) to define selection criteria of steel selection in geothermal environments. Their results showed that the steel cleanliness and mechanical strength of pipes necessarily do not affect stress corrosion sensitivity in simulated geothermal conditions. All of the pipes that were studied, were subjected to stress corrosion cracking in simulated geothermal sour conditions. They showed that adding lime to the drilling fluid could be a very effective solution to reduce the environmental assisted cracking in geothermal conditions. Thomas (Titanium in the geotherma, 2003) proposed using titanium alloys (with molybdenum, palladium, or ruthenium) in the geothermal industry can be a solution for Sulphur oxide corrosion  even in high temperatures (260 C). An application of such material (casing grade 29) in Salton sea, USA, enabled the development of a highly corrosive brine from a geothermal resource. The cost for such casings may be very high and newer technologies in cost reduction of such pipes need to be applied. In this case, the technology may even allow us to inject the seawater into depleted aquifers to produce steam from hot dry rock resources. Babler et al. (2009) studied the corrosion resistance of eight different casings (carbon steel, highly alloyed stainless steel, titanium-based alloy, and nickel-based alloy; 2 of each) with exposure to geothermal fluid. Based on the results, a recommendation was given for material selection for future geothermal sites. Accordingly, in the presence of nitrogen saturated fluid with a temperature of 150 C, results showed that nickel-based and   (Snyder, 1979).

P. Allahvirdizadeh
Journal of Cleaner Production 275 (2020) 124009 titanium-based alloys are suitable for such conditions and stainless steels are also resistant but with limitations. Where carbon steels proved to not be suitable. Karlsdottir et al. (Karlsdottir and Thorbjornsson, 2012) investigated a corroded casing (grade K-55 and L-53) in the Krafla geothermal site located in Iceland. The study revealed the fact that the liner failed due to hydrogen embrittlement. Results showed that the maximum corrosion took place in a transition zone, where the hot steam containing H 2 S, CO 2 and HCL meets a colder steam/ water in a certain depth (1600 m) creates a very corrosive environment with a high concentration of HCL. The solution could be replacing these transition zones with materials resistant to corrosion as well as sealing the colder fluid coming into well.
Numerous other studies exist focusing on selecting the appropriate material for geothermal systems that mainly lead to the use of nickel-based alloys and titanium-based alloys in some cases. Nevertheless, using such substances is the high cost. The ultimate cost may be reduced by the technology developments as well as the optimal usage of such casings.

Time & cost analysis
Geothermal energy is a renewable source of energy found around the globe. However, due to some constraints, which mainly

P. Allahvirdizadeh
Journal of Cleaner Production 275 (2020) 124009 high upfront costs and risk, has not been much developed and widely spread compared to other renewables for electricity generation (GEO ELEC deliverable no 3.2, 2013) (Fig. 16). Fig. 17 shows a breakdown of capital expenses for a geothermal project (Bronicki and Meyers, 2017). As shown in the figure, about 43% of the capital cost is related to exploration and drilling operations. So decreasing these costs as a major role player of deep geothermal projects can significantly improve the geothermal energy role in the energy section in the future.
Cost-effective drilling, by the possibility of cutting down the drilling and completion costs to develop new well construction solutions, is essential for economic utilization of deep geothermal resources (Reinicke and Ostermeyer, 2015).
Achieving a considerable reduction in these costs is only possible by using new and innovative methods for the construction of a well. Cutting drilling costs can be achieved for example by improving drilling methods (especially on rock destruction), new drilling equipment, and new well construction solutions (Reinicke and Ostermeyer, 2015). Because the nature of the drilling market is competitive, thus confidential, there is not much data published on the breakdown of a geothermal drilling operation cost (106). However, a few data published for some case studies as shows a breakdown of cost for a reference well with a large diameter with a depth of 2235 m, which was studied in Nesjavellir field in the Hengill Geothermal Area, Reykjavik, Iceland (107). The estimation for this reference well was determined in terms of the number of working days for each section of the well.

P. Allahvirdizadeh
Journal of Cleaner Production 275 (2020) 124009 Fig. 18 and Fig. 19 also show the time analysis of two regular diameter wells in Iceland and Kenya (Ong'au, 2012). Table 10 also indicates the percentage of the productive time for each part of well drilling and construction operations (Reinicke and Ostermeyer, 2015). Based on the time analysis of the Kenyan and Icelandic wells, resulted that 45% of the work time is spent on actual drilling.
As stated above, the decrease of drilling time (actual drilling and tripping), as a major role player in terms of time/cost could result in total cost reduction considerably (Reinicke and Ostermeyer, 2015).
The ability to circulate while RIH or POOH operations may also be a possible solution to reduce the tripping time; however further investigations are required to find out that if it is possible and suitable for large diameter holes (Reinicke and Ostermeyer, 2015).
Another factor affecting the drilling time increase is unexpected problems that may occur during the drilling operations. Among which loss of fluid circulation is the most common issue, especially in geothermal well drilling operations (Reinicke and Ostermeyer, 2015), which can be resulted in: a) increasing the drilling time; b) loss of expensive drilling fluid material; c) poor cementing; And this ultimately can also result in well integrity issues due to the poor cementing job as well as possible equipment loos in the hole.  Plugging loss zones in conventional hydrocarbon well drilling operations is usually achieved by cementing the zone with bridging agents (Allan and Kukacka, 1995); but when loss zones include the main fractures, which is very common in geothermal well drilling, cement have to set rapidly and even may not withstand severe geothermal environment (Allan and Kukacka, 1995) and this will increase the drilling time and respectively the cost.

Conclusions
Although currently energy production from geothermal resources is not dominant in energy sectore, it has a huge potential to be a major role player in future energy strategy. EGS is one of these potential which can be reached in almost all over the globe. A majour challenge is drilling operations in such resources. Drilling operations of a geothermal well is quite similar to the drilling operation in petroleum industry. Though, the different aspects of geothermal drilling will make it more challenging and complex than the oil and gas well drilling. The type of resource and the formation type of resource is the big differences. Petroleum and geothermal well drilling have different geological and physical features including rock types, reservoir pressure, temperature, well and casing design as well as the well completion program. Hence, in this paper, it was tried to present the most common issues in geothermal well drilling operations and to introduce the potential for the future research areas. So the following can be concluded: The cost is not the main reason behind which has hold the geothermal energy behind comparing to renewable resources, rather it is the hydrothermal resources availability around the earth; EGS is nearly presented anywhere able to offer energy to all for many of years with huge potential; Research and development of EGS is the key for geothermal energy to make a significant contribution to energy needs, along with other renewables and new technologies for existing fossil fuel systems; A very commen issue in geothermal well drilling operations are related to well integrity problems during the drilling and production phases. Casings, annulus cement and the liners play role in well integrity as the primary components; since a geothermal well should last much longer than a typical hydrocarbon well, the importance of well barriers' role becomes dominant in geothermal drilling; Amongst the challenges for drilling operations of a geothermal well, casing and cementing issues seem to mainly affect the integrity of a well in comparison to others; An important area of future studies could be focused on bonding between well cement and casing/annulus in high-temperature geothermal wells to enhance the knowledge and widen the aspects of the potential issues in this regard.

Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: