Optimizing Deep Geothermal Drilling for Energy Sustainability in the Appalachian Basin

Abstract

This study investigates the geological and geomechanical characteristics of the MIP 1S geothermal well in the Appalachian Basin to optimize drilling and address the wellbore stability issues encountered. Data from well logs, sidewall core analysis, and injection tests were used to derive elastic and rock strength properties, as well as stress and pore pressure profiles. A robust 1D-geomechanical model was developed and validated, correlating strongly with wellbore instability observations. This revealed significant wellbore breakout, widening the diameter from 12 ¼ inches to over 16 inches. Advanced technologies like Cerebro Force™ In-Bit Sensing were used to monitor drilling performance with high accuracy. This technology tracks critical metrics such as bit acceleration, vibration in the x, y, and z directions, Gyro RPM, stick-slip indicators, and bending on the bit. Cerebro Force™ readings identified hole drag caused by poor hole conditions, including friction between the drill string and wellbore walls and the presence of cuttings or debris. This led to higher torque and weight on bit (WOB) readings at the surface compared to downhole measurements, affecting drilling efficiency and wellbore stability. Optimal drilling parameters for future deep geothermal wells were determined based on these findings.

1. Introduction

The Appalachian Basin, covering approximately 175,000 square miles, spans parts of New York, Pennsylvania, Ohio, Maryland, Virginia, West Virginia, and Kentucky [1]. This region is characterized by numerous deep geological structures. In 2010, Southern Methodist University (SMU) reported that it was feasible to access the necessary heat for district heating within a reasonable drilling depth in north-central West Virginia [2,3,4]. A 2017 study by Cornell University [5] confirmed increased heat flows in this area. In 2023, West Virginia University drilled a deep geothermal well and conducted multiple temperature logs for a period of six months, demonstrating the potential for deep geothermal wells to support district heating and cooling (GDHC) and underground thermal energy storage (UTES) systems [6,7,8]. However, drilling and completing deep wells in the Appalachian Basin pose several challenges. Conventional drilling with mud in the Appalachian Basin is typically slow and expensive, particularly as daily rig rates increase, making it economically viable only in specific areas [9,10]. Normally, oil and gas wells in this region are drilled with air to depths of 3000 to 7000 feet. In some cases, both air and mud circulating systems are employed to improve the accuracy of wellbore trajectories [11,12]. As drilling operations reach deeper horizons in the Appalachian Basin, new challenges arise in wellbore construction. These include larger and deeper hole sections, higher temperatures, increased formation pressures, and zones with high gas flow rates [13]. Such conditions make it difficult and unsafe to complete wells using conventional drilling techniques. Consequently, a zonal analysis of drilling deep wells in the Appalachian region is essential to illustrate the potential problems that may be encountered.

Figure 1 presents a stratigraphic column of West Virginia, highlighting potential thermal energy sources in the Point Pleasant and Trenton Group (indicated by red arrows) and storage horizons in the Hampshire Group and Brallier Formation (marked by blue arrows). Figure 2 shows the MIP 1S vertical geothermal science well located on the MIP pad in Morgantown Industrial Park. Additionally, it shows horizontal wells MIP 1, 3, 5, and 6H drilled and completed in the Marcellus Shale.

Figure 3 shows a planned wellbore configuration, detailing each stage of the drilling process. Initially, a 32-inch hole is established with a 26-inch casing set at a depth of 50 feet true vertical depth (TVD), employing an Auger bit and drilled solely with air. This is followed by drilling a 24-inch hole with an 18-5/8-inch casing set at 491 feet TVD to safeguard the coal and the deepest freshwater zone. In this phase, a combination of air and freshwater mist is utilized with a hammer bit. Subsequently, the drilling progresses to a depth of 1910 feet TVD with a 17-1/2-inch borehole and a 13-3/8-inch intermediate casing, utilizing a hammer bit and a mixture of air and processed water. The wellbore extends to a depth of 10,509 feet, with a 12-1/4-inch borehole and a 9-5/8-inch casing set at the same depth. Initially, drilling employs air and processed water mist until 8668 feet, where an HD75FS PDC air bit is utilized, followed by a transition to synthetic oil-based mud starting with 12.0 pounds per gallon (ppg) density and ending at 13 ppg using an HD75F PDC bit. The original plan aimed to set the second intermediate casing at 12,500 feet and proceed to a target depth of 15,000 feet. However, due to various technical challenges encountered during drilling, operations terminated at 10,509 feet. The final segment of the wellbore is equipped with Cerebro Force logging while drilling (LWD) in-bit technology. Subsequently, the well undergoes casing, cementing, and preparation for future drilling operations, contingent upon funding availability.

Figure 4 provides a comparison between the actual daily drilling activities (shown in blue) and the planned daily drilling activities (shown in orange) for the MIP1S geothermal science well. The graph highlights a delay in the actual drilling compared to the planned schedule in shallow depth. This delay was primarily caused by various drilling issues, including a stuck drill bit and lost equipment at shallow depths. Additionally, Figure 4 indicates a significant reduction in the rate of penetration after switching from air-mist to synthetic oil-based mud at a depth of 8668 feet. This reduction ultimately forced the drilling to end at 10,509 feet, preventing the completion of the planned 15,000 foot vertical well. Moreover, the inability to reach the target depth resulted in missing most sidewall cores, and poor hole conditions led to inaccurate density log measurements. These challenges prompted our research group to revisit the drilling plan, conduct geomechanical modeling, and perform a comparative study with other wells drilled on the same pad. The goal is to optimize the drilling procedure for future deep geothermal wells in the Appalachian Basin.

2. Methodology

A comprehensive dataset, including well logs, Cerebro Force™ drilling records, limited sidewall cores, and a diagnostic fracture injection test (DFIT) from an adjacent well, was utilized to construct and validate a 1D geomechanical model of the MIP 1S well. Figure 5 illustrates the 1D geomechanical modeling workflow employed in this study. This workflow was used to conduct wellbore stability analysis for four wells drilled and completed from the MIP pad, encompassing the MIP 1S geothermal science well and three unconventional shale gas production wells. The workflow incorporates calculations for overburden/vertical stress, pore pressure, and rock strength properties, which are essential for determining minimum and maximum horizontal stress and analyzing wellbore stability.

Calculated vertical stress, also known as overburden stress, uses bulk density measurements (

Table 1. Side wall cores density and porosity measurements.

Sample ID	Core Depth (ft)	As Received Bulk Density (g/cc)	Dry Grain Density (g/cc)	Total Porosity (%)
MIP1S-4	2040.00	2.672	2.810	4.883
MIP1S-5	2116.00	2.485	2.697	7.873
MIP1S-6	2132.00	2.509	2.683	6.500
MIP1S-7	2152.00	2.681	2.781	3.590
MIP1S-8	2200.00	2.489	2.676	7.007
MIP1S-10	2236.00	2.545	2.701	5.760
MIP1S-11	2244.00	2.554	2.725	6.283
MIP1S-12	2717.00	2.631	2.707	2.816
MIP1S-13	2834.00	2.712	2.820	3.816
Onondaga	7550.00	2.664	2.708	1.643
Oriskany	7652.00	2.589	2.658	2.575
P-H2 Utica	10,618.60	2.682	2.817	4.791
P-H56 Pt. Plsnt	10,669.45	2.499	2.732	8.536

) based on the methodology outlined by [14]. Due to the challenging hole conditions encountered in the MIP 1S well, the density log required correction. The corrected density used data from the dipole-sonic log and Gardner’s equation, and further calibration utilized density measurements from sidewall cores, as shown in Figure 6. In Figure 6, the second gray track is the measured density (ZDEN), the blue track is the correction with Gardner’s equation, and the third track shows the calibration of the corrected density log with measured densities from sidewall cores.

Pore pressure was determined using Eaton’s method, which relies on sonic log data. Additionally, a shale discrimination line was established using gamma ray, resistivity, and mineralogy measurements to enhance accuracy. Data from a DFIT in an adjacent well was used to further calibrate the pore pressure gradient obtained through Eaton’s method.

Dynamic Young’s modulus (E_dyn) and Poisson’s ratio (ν) were derived from dipole sonic log data using Equations (1) and (2).

$$v_{dyn}={\displaystyle \frac{V_p^2-{2V}_s^2}{\left.2(V_p^2-V_s^2\right)}}$$

(1)

$$E_{dyn}={\displaystyle \frac{\rho V_s^2\left({3V}_p^2-{4V}_s^2\right)}{V_p^2-V_s^2}}$$

(2)

where $\rho$ is the bulk density (g/cm³), V_S is the shear wave velocity (ft/sec), and V_P is the compressional wave velocity (ft/sec).

These dynamic measurements were then correlated with triaxial test results from sidewall cores, as presented in

Table 2. Side wall core rock mechanics analysis.

Depth (ft)	FANG (deg)	Poisson’s Ratio 1	Poisson’s Ratio 2	UCS (psi)	E (psi)
2152	43.48	0.20	0.18	7438	5,327,000
7980	30.79	0.36	0.17	5989	7,000,000

. To convert the dynamic Young’s modulus to a static value, a specific correlation was applied as shown in Equation (3) [15]. The static Young’s modulus was then used to calculate the unconfined compressive strength (UCS), which is a critical parameter for assessing rock strength. The tensile strength (TSTR) was assumed to be one-tenth of the UCS, providing an estimate of the rock’s tensile failure threshold. The friction angle (FANG), an important parameter for understanding rock shear strength, was estimated using shale volume and neutron porosity measurements. These estimates were then calibrated against sidewall core measurements to ensure accuracy.

$$\mathrm{Y}\mathrm{M}\mathrm{s}\mathrm{t}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{c}=0.835\times \mathrm{Y}\mathrm{M}\mathrm{d}\mathrm{y}\mathrm{n}\mathrm{a}\mathrm{m}\mathrm{i}\mathrm{c}-0.424$$

(3)

Minimum (${\sigma }_h\mathrm{)}$ and maximum (${\sigma }_H\mathrm{)}$ horizontal stresses were computed by utilizing a poro-elastic equation and considering stress anisotropy in the field using Equations (4) and (5).

$${\sigma }_h={\displaystyle \frac{\nu }{1-\nu }}{\sigma }_v-{\displaystyle \frac{1-2\nu }{1-\nu }}{\alpha P}_p+{\displaystyle \frac{E}{1-{\nu }^2}}{\epsilon }_y+{\displaystyle \frac{E\nu }{1-{\nu }^2}}{\epsilon }_x$$

(4)

$${ \sigma }_H={\displaystyle \frac{\nu }{1-\nu }}{\sigma }_v-{\displaystyle \frac{\nu }{1-\nu }}{\alpha P}_p+{\alpha P}_p+{\sigma }_{tect}$$

(5)

where Pp is the pore pressure (psi), α is the Biot elastic coefficient, ε_y is the vertical strain, ε_x is the lateral strain, and σtect is the tectonic stress (MPa). These calculations incorporate rock elastic properties, vertical stress, pore pressure, Biot’s coefficient, and tectonic strain in the x and y directions [15]. The minimum and maximum horizontal stress profiles were then calibrated using DFIT data from adjacent wells.

3. Analysis and Discussions

Figure 7 presents a comparative study of the elastic and rock strength profiles for the MIP 1S geothermal well alongside the MIP 3H, 4H, and MIP SW wells, all drilled and completed from the same MIP pad using the workflow described in Figure 5. For each well, three tracks are displayed, including their elastic properties (this track includes measurements of Young’s modulus and Poisson’s ratio, which have been calibrated using sidewall core data), pore pressure, and stress measurements (this track includes vertical stress as well as minimum and maximum horizontal stress measurements, calibrated using data from DFIT).

The Mohr–Coulomb failure criterion, established by Mohr in 1900 [16], is essential in geomechanics for assessing wellbore stability. It defines the conditions under which rock fails due to shear stress, which is crucial for determining the safe mud weight window needed to prevent wellbore collapse or fracturing. By comparing calculated shear and normal stresses to the Mohr–Coulomb failure envelope, potential wellbore instabilities can be identified. To ensure stability and prevent various types of wellbore failures, four conventional analyses are used: kick, shear failure, loss, and tensile failure [17,18] Byrnesa. A kick occurs when formation fluids enter the wellbore due to insufficient mud weight, necessitating a mud weight higher than the formation pore pressure but lower than the fracture pressure to prevent formation fracturing. Shear failure, or breakout, happens when stress around the wellbore exceeds the rock strength, causing the rock to break away from the wellbore wall, leading to an enlarged wellbore diameter and increased risks such as stuck pipe and poor hole cleaning. In the MIP 1S well, due to using air/mist drilling up to 8667 ft MD, the minimum shear failure threshold was not achieved, resulting in significant breakouts between 2000 to 3000 ft MD and 5000 to 7500 ft MD. These events were clearly observed in the multi-arm caliper log, showing considerable wellbore enlargements and, in daily drilling reports, noting pipe sticking, lost drill bits, and casing running and testing issues. Losses refer to the unwanted flow of drilling fluids into the formation, which occur when mud weight exceeds the formation’s fracture pressure, inducing fractures and fluid escape. Tensile failure, or breakdown, happens when mud pressure exceeds the rock’s tensile strength, causing it to split or fracture. Loss and tensile breakdown were not identified between 8667 ft and 10,509 ft, where oil-based mud with 12.4 ppg density was used. This observation was cross-checked with the multi-arm caliper and daily drilling reports. Thus, the safe mud weight window is defined as a range higher than the pore pressure and lower than the fracture pressure, while being sufficient to prevent shear failure, ensuring wellbore stability and preventing various failures. Figure 8 shows the wellbore stability analysis of the MIPS 1S geothermal science well using the kick, shear failure, loss, and breakdown criteria based on the Mohr–Coulomb failure criterion. Since air mist was used as the drilling fluid, the minimum mud weight criteria shown in red were not satisfied, resulting in breakouts in shallow depths (i.e., red spikes on track 3) and significant wellbore compressive failure between 5000 to 7500 ft. These events are also shown in the synthetic wellbore breakout image (Track 5) and are highly correlated with the multi-arm caliper log (Track 6).

Figure 9 presents a comparative analysis of wellbore stability for the MIP 3H, MIP 4H, and MIP SW wells. The MIP 3H well demonstrates a more stable wellbore than MIP 1S, with minor breakouts at 2000, 3000, and 5600 ft MD, as indicated in the synthetic breakout image and confirmed by the caliper log. Similar to the MIP 1S geothermal well, the MIP 4H well exhibits significant breakouts at shallow depths between 2000 to 3000 ft, where the minimum shear failure threshold was not met, resulting in notable breakouts. These incidents are clearly observed in the synthetic breakout images (tracks 2 and 3) and confirmed by the caliper log (track 4). Additionally, the MIP 4H well shows major breakout events in the 5000 to 7500 ft section, although these are less severe than those seen in the MIP 1S well. Conversely, the MIP SW well maintains a stable wellbore throughout the entire pad, with only a minor incident around 7500 ft. It is significant that while the MIP 4H and MIP SW wells were drilled using air mist for the entire measured depth (MD), the MIP 3H well switched to mud at 6050 ft.

Figure 10 presents a comparison of the rate of penetration (ROP) recorded for the MIP 1S geothermal science well and the MIP 3H and MIP SW wells, focusing on the sections drilled using air/mist. The data clearly indicate that the MIP SW well achieves a significantly higher ROP than the MIP 3H and MIP 1S wells in the shallow formations up to 4500 ft (Section I). In Section II, MIP 1S shows a higher ROP, while in Section III, MIP SW once again exceeds the others. However, in Section IV, both the MIP 1S and MIP SW wells experience a significant decline in ROP. In Section I, where MIP SW exhibits high ROP, the formation is predominantly sandstone, except for the Big Lime formation encountered between 995 and 1107 ft. Section II, where MIP SW’s ROP decreases, features sandstone with a higher Poisson’s ratio and lower Young’s modulus compared to shallower formations. Section III includes the Elk sandstone, located at the base of the Brailer Formation. Section IV comprises alternating sequences of shale and limestone.

Figure 11 compares WOB in psi, rotary torque in ft-lb, rotary RPM, and airflow velocity in ft/min across the wells. Compared to other wells, significantly higher WOB and rotary torque were applied at the MIP 1S geothermal science well. However, it utilized considerably lower rotary RPM and airflow velocity than the MIP SW well, which correlates with MIP SW achieving better ROP. A detailed analysis of drilling parameter variations with respect to the stratigraphic column of the wellbore is essential to identify problems and issues encountered during drilling.

Table 3. A sectional wellbore discussion of the drilling parameters of MIP1S, 3H, and SW.

Formations	Depth (ft)	Lithology	ROP (ft/hr)			Rotary (RPM)			WOB (Psi)			Air Flow (ft/min)
			MIP 1S	MIP 3H	MIP SW	MIP 1S	MIP 3H	MIP SW	MIP 1S	MIP 3H	MIP SW	MIP 1S	MIP 3H	MIP SW
Gordon	1853	sandstone	57	71	181	34	34	39	137	114	123	7543	8890	14,055
Fourth	2100	sandstone	70	162	231	31	43	56	144	77	86	8563	6847	9112
Warren	2707	sandstone	172	170	255	25	41	59	157	79	80	6912	6646	8070
Speechley	2986	sandstone	210	151	238	29	38	59	190	76	95	7576	6687	8193
Riley	4337	sandstone	152	159	201	24	39	57	176	68	82	7232	6696	8255
Benson	4689	sandstone	152	146	175	26	40	58	155	58	101	7400	6660	8249
Alexander	4939	sandstone	199	150	187	24	40	58	184	54	97	7334	6603	8062
Elk	5482	sandstone	222	149	153	24	41	61	170	53	84	7491	6559	7993
Geneseo/Burkett	7153	grey shale	188	163	209	21	43	57	156	48	73	7463	6722	8090
Tully	7194	limestone	215		151	25		58	132		74	8655		8281
Hamilton	7280	grey shale	196		153	24		59	120		53	9009		8279
Marcellus	7454	black shale	168		142	23		59	118		45	9104		8293
Onondaga	7550	limestone	169		132	21		60	94		53	9097		8452
Oriskany	7672	sandstone	101		141	20		59	116		59	9035		11,745
Salina	8935	salt/dolomite	53			18			209			8361

lists average values of ROP, WOB, RPM, and airflow for the MIP 1S, MIP 3H, and MIP SW wells at different formations and compares them with variations in their lithology.

As detailed in Table 3, the first 4500 ft of the MIP 1S geothermal science well, which comprises various sandstone formations, exhibited the lowest ROP compared to other wells previously drilled on the MIP pad. In this section, the MIP 1S well is operated at lower RPM and significantly reduced airflow velocity, especially compared to MIP SW, despite using substantially higher WOB. It is important to note that the WOB is converted to psi and airflow to ft/min to account for different wellbore sizes across the MIP pad. The reduced airflow in the MIP 1S well, particularly in the shallow “Gordon Sandstone” formation, likely led to the accumulation of drill cuttings and debris in the wellbore. This accumulation necessitated the application of more WOB to maintain the ROP. However, this issue persisted and worsened as drilling continued, resulting in poor hole conditions from the top down to the “Benson” formation, as shown in the caliper log and wellbore stability analysis in Figure 8. The situation improved somewhat in the Benson formation but worsened significantly with notable wellbore breakouts from the “Alexander” sandstone to the top of the “Onondaga”. We believe this was due to the accumulation of debris, resulting in additional friction and wellbore breakouts in the shallower formations, which led to the need for significantly higher WOB during the drilling of the MIP 1S well compared to MIP SW, as detailed in Table 3. These observations are further confirmed with a series of drilling problems such as a broken bit at 1138 ft and a stuck pipe at 2209 ft, 3260 ft, and 7585 ft, recorded in daily drilling reports of MIP 1S.

4. Advanced Downhole Measurement Using Cerebro Technology

To address the slow ROP while drilling the MIP 1S well, advanced downhole measurement and data collection using Cerebro technology were employed to optimize drilling performance and improve wellbore quality. This technology integrates various sensors, including accelerometers, magnetometers, and gyroscopes installed directly at the bit and captures detailed data on drilling dynamics, bit performance, and formation characteristics. Cerebro Force™ in-bit sensors offer direct measurements of weight, torque, bending, vibration, and rotational speed, eliminating issues associated with traditional stabilizers that reduce shock or vibration in the drill string. Notably, the system has no limitations regarding mud weight, flow rate, or lost circulation material (LCM), and it continues to capture data even when the pumps are off. The system operates under hydrostatic pressures up to 20,000 psi and circulating temperatures of 130 °C (266 °F). However, the technology does have potential drawbacks. It is sensitive to extreme temperatures, and inaccuracies in data can stem from inadequate sealing or improper installation. The main risks include battery failure due to water exposure or overheating, which could compromise the system’s performance. Additionally, while the technology captures a high volume of data, like many sensors on the market, the data must be analyzed post-drilling, which can limit real-time decision-making.

The implementation of Cerebro technology at the bit revealed significant discrepancies between the surface-applied parameters and those encountered downhole, particularly in WOB and torque, as discussed earlier and presented in Figure 12. Surface torque readings of the MIP 1S well were consistently higher than those recorded downhole, and this discrepancy increased with depth. This indicates that the torque applied at the surface does not directly translate to the downhole conditions due to factors such as friction and mechanical losses along the drill string. Similarly, surface WOB readings were consistently higher than those measured downhole, likely due to hole drag, where friction along the borehole reduces the effective WOB reaching the bit. Understanding these discrepancies is crucial for optimizing drilling performance as they highlight the challenges in accurately translating surface measurements to downhole conditions. These observations are also confirmed with our comparison study presented in Figure 11 and Table 3.

5. Discussion

The wells in our study were drilled from the same pad, experience the same lithology, and are operated by the same company. This consistent operational setting forms a solid basis for our comparison, particularly when assessing the performance of advanced technologies like the Cerebro Force™ In-Bit Sensing system. From an operational perspective, these similarities allow us to isolate drilling performance variables and provide clearer insights into optimization strategies. This technology has had a minimal impact on drilling costs, indicating its economic viability.

Table 4. Optimum drilling parameters in different formations.

Formations	Depth (ft)	Lithology	Optimum ROP (ft/hr)	Optimum Rotary (RPM)	Optimum WOB (Psi)	Optimum Air Flow (ft/min)
Gordon	1853	sandstone	181	39	123	14,055
Fourth	2100	sandstone	231	56	86	9112
Warren	2707	sandstone	255	59	80	8070
Speechley	2986	sandstone	238	59	95	8193
Riley	4337	sandstone	201	57	82	8255
Benson	4689	sandstone	152	26	155	7400
Alexander	4939	sandstone	199	24	184	7334
Elk	5482	sandstone	153	61	84	7993
Geneseo/Burkett	7153	grey shale	209	57	73	8090
Tully	7194	limestone	151	58	74	8281
Hamilton	7280	grey shale	153	59	53	8279
Marcellus	7454	black shale	142	59	45	8293
Onondaga	7550	limestone	132	60	53	8452
Oriskany	7672	sandstone	141	59	59	11,745

outlines the optimal drilling parameters and expected rate of penetration (ROP) for various formations in West Virginia’s stratigraphic column. Most formations align with the MIP SW drilling operation parameters, which is known for higher ROP (Figure 10) and improved wellbore stability (Figure 9). However, the Benson and Alexander sandstones present exceptions. The Alexander and Benson formations are often grouped together, suggesting they share similar characteristics [19]. The Upper Devonian Alexander siltstone is a gas-producing unit in the eastern Appalachian Basin. It is part of the Brallier Formation, primarily composed of silt with occasional shale layers, and its thickness varies from 10 to 140 feet. Three distinct intervals have been identified within the Alexander, with gamma-ray logs showing alternating silt and shale layers that thin upwards towards the basin (Cummings, 2014). Composed primarily of quartz with minor feldspar and clay minerals, these formations exhibit strong cementation and overall stability. In contrast, the Riley and Elk sandstones have more heterogeneity with mixed mineral compositions and higher clay and feldspar content, leading to less stable lithologies with shaly intervals. (Regional subsurface rock correlation diagram Appalachian storage hub project WVGS https://www.wvgs.wvnet.edu (accessed on 15 July 2024)).

The mechanical properties of the Benson and Alexander sandstones reflect higher compressive strength and lower ductility due to their well-cemented structure and quartz dominance. The Benson sand is more prolific in gas production due to its coarser grain size, better sorting, and higher quartz content [20]. On the other hand, the Riley and Elk sandstones have more variable mechanical properties, with higher clay content and shaly layers contributing to lower compressive strength, increased ductility, and inconsistent permeability. As a result, drilling the Benson and Alexander formations requires a higher weight on bit (WOB), lower revolutions per minute (RPM) to save the drill bit, and increased airflow for effective cuttings removal. In contrast, the softer Riley and Elk formations allow for higher ROP, lower WOB, and higher RPM.

Shallow sandstones like Fourth (2100 ft) and Warren benefit from higher RPM (56–59 RPM), resulting in better ROP. However, deeper formations like Benson and Alexander see a stabilization or decline in RPM due to their challenging drilling conditions. In the original drilling plan of MIP SW, higher RPM and lower WOB was implemented that could have resulted in inefficient drilling in the Benson and Alexander formations.

The relationship between geological characteristics and drilling parameters is crucial for understanding performance. In the Appalachian Basin, mid-depth sandstones like Warren (2707 ft) and Speechley (2986 ft) achieve high ROPs of 255 ft/hr and 238 ft/hr, respectively, due to their stable lithology. However, deeper sandstones like Benson (4689 ft) and Alexander (4939 ft) see a drop in ROP to 152 ft/hr and 199 ft/hr due to increased compaction and hardness. These formations demand higher WOB (up to 184 psi) and lower RPM (24–26 RPM) to maintain drilling efficiency and wellbore stability.

Airflow management is also critical. In harder formations like Benson and Alexander, higher airflow is necessary to clear denser cuttings, while softer formations like Riley and Elk require less airflow. Understanding these relationships is key to optimizing drilling strategies and improving efficiency in geothermal exploration in the Appalachian Basin.

In conclusion, the shallow wellbore instability encountered in MIP 1S can be largely attributed to unoptimized drilling parameters. The solution lies in applying optimized drilling strategies, as demonstrated by the successful performance of MIP SW, where such wellbore instability was effectively mitigated.

6. Conclusions

The study of geomechanical modeling and drilling parameters in the MIP pad wells has highlighted critical insights for optimizing drilling operations, particularly for the MIP 1S geothermal science well. The comparative analysis between the wells drilled in the MIP pad has shown significant variations in wellbore stability, drilling parameters, and ROP, which are key to improving future geothermal drilling projects in the Appalachian Basin. Geomechanical modeling has proven essential for understanding wellbore stability and the challenges encountered during drilling. The model constructed for MIP 1S using well logs, sidewall cores, and DFIT data identified significant instability at shallow depths and a clear reduction in ROP after switching from air mist to synthetic oil-based mud at 8668 feet. This instability was less pronounced in the MIP SW and MIP 3H wells, indicating that maintaining stable drilling conditions in the MIP 1S well is critical for better performance. The comparative study of drilling parameters revealed that the MIP SW well consistently achieved higher ROP than MIP 1S, particularly in the shallow formations up to 4500 feet. This higher performance in the MIP SW well was attributed to the optimized WOB, rotary torque, and air flow velocity. Specifically, the MIP SW well employed higher rotary RPM and airflow velocity, which correlated with its higher ROP. Conversely, the MIP 1S well showed a higher WOB and rotary torque but lower RPM and airflow, suggesting a need for parameter adjustments.

Advanced downhole measurements using Cerebro technology revealed significant discrepancies between surface-applied WOB and torque and those encountered downhole. Surface torque and WOB readings were consistently higher compared to downhole measurements, with the discrepancy increasing with depth. This indicates that non-optimal drilling practices at shallow depths led to poor hole conditions, worsening as drilling progressed. To achieve an optimum ROP in future geothermal wells, it is recommended to increase rotary RPM and airflow velocity to numbers closer to the values used in the MIP SW well at shallower depths (<4500 ft), as presented in Table 3. These recommendations aim to enhance drilling performance and efficiency with a lead to more successful geothermal projects in the Appalachian Basin. From an environmental standpoint, our work aligns with efforts to reduce carbon emissions by promoting geothermal energy as a sustainable alternative. The proposed geothermal system at West Virginia University, which would replace the current natural gas-based heating and cooling system, represents a significant long-term environmental benefit. While drilling activities have associated emissions, the life-cycle analysis clearly favors the adoption of geothermal energy, as it offers substantial reductions in greenhouse gas emissions over time.