The goal is to develop a revolutionary solid-state gamma-ray detector suitable for use in harsh environment downhole gas and oil exploration.
General Electric Global Research Center, Niskayuna, New York 12309
Gamma detection for down-hole gas and oil exploration is used to determine multiple parameters in the assay of possible petroleum bearing formations. Presently this technology is implemented using scintillation detectors consisting of a scintillation crystal, which produces optical photons when excited by a gamma ray, and a photomultiplier tube (PMT), which detects these photons. The detector must withstand the high vibrations and high temperatures present in down-hole applications. Detector performance and service lifetime under high temperature will limit drilling depth as ambient temperatures rise with increasing drill depth. PMT technology can support useful detector lifetimes in the area of 1,000 hours at 150°C, but only 100–200 hours at 175°C.
The temperature limitation of PMT-based technologies represents the most significant economic motivation for developing an alternative detector. This limitation represents a barrier to discovering reserves at greater depths; exploration companies would like to drill deeper while retaining the ability to use the intelligence from gamma detection. The temperature gradient for drilling into the earth is dependent on geographical location and may vary widely, in the range of 8–50° C/1,000 meters.
In this project, the GE team will fabricate and characterize the essential elements of an APD/scintillator detector system that can meet the system requirements needed for down-hole applications. The known technical requirements for the APD, scintillator, and detector system will be quantitatively defined. Mathematical models of the detection system will be developed and validated to allow the system response to be estimated from component measurements. A silicon carbide (SiC) APD will be designed, fabricated, and characterized. Scintillator materials will be evaluated for efficiency, emission wavelengths, and temperature stability through phosphor powder screening experiments.
Work during the second phase of the program will include: 1) optimization of the final APD design, 2) growth and characterization of full scintillator crystals of the selected composition, 3) arrangement of the APD device packaging, 4) development of the necessary optical coupling between scintillator and APD, and 5) characterization of the combined elements as a full detector system prepared for commercialization.
Results
To date, significant results have been accomplished across a number of tasks. The performers first determined the most critical parameters driving the technical specifications (temperature, reliability and energy resolution). They also refined their system model for estimating performance and evaluating components. Key detector components that have been successfully developed include: avalanche photodiode (APD), scintillator, optical coupler, and electronics. Prototype systems have been developed and tested. The testing showed insufficient system performance to productize the HPHT sensor system at this time. The initial speculation is that the APD electrical leakage was too high and the scintillator crystal's light output was too low resulting in inadequate overall energy resolution. GE will continue to work the optimization of both SiC APDs and the novel scintillator materials through other ongoing GE programs, with the goal of revisiting this system as sufficient improvements are made.
Benefits
The development of a high-temperature, solid-state photodetector fabricated from silicon carbide will circumvent current temperature limitations for gamma detectors as silicon carbide devices can survive and perform at temperatures above 200° C. The new detector will offer two main advantages over current detectors. First, it will be able to operate at higher temperatures, which will allow for deeper drilling and exploration. Second, because the solid-state photodetector has a higher immunity to shock and vibration, it will have a longer life downhole. Both of these advantages will help to reduce the risk and costs for deep drilling.
Summary
The following accomplishments, arranged by general category, have been recorded since the start of the project.
First, the critical parameters driving the technical specifications were determined:
Second, modeling and simulation tools were refined and applied:
categorized under each of the key detector components, include:
Silicon Carbide Avalanche Photodiode Development
Scintillator Development
Optical Coupler
Electronics
System Packaging & Prototype Development
In addition, the performers consulted with oil well drilling customers to collect “voice of the customer” feedback on how and when the sensor system developed under this program might be best implemented.
Recommendations for Improving Tool Performance
The fact that the SiC APD detector did not detect any gamma radiation can be explained by the fact that the detectable minimum of photoelectrons for the SiC APD arrays was 300, while an actual value of absorbed photons was 12. It is thought that by improving the components of the gamma sensor system, namely the SiC APDs as well as the LaX3 scintillator, would lead to better results,
This program has provided the means to understand the system model, the system design elements that have yielded success in construction as well as the overall approach to detect gammas at 200 °C. With that in mind, there are several approaches proposed in order to improve a performance of the sensor. These may be used in combination to arrive at the energy resolution requirements initially determined in GE’s voice of the customer sessions conducted early on in the program. These include:
(February 2008)
The project is complete. The final report is available below under "Additional Information".
$1,319,474
$329,869
NETL – Gary Covatch (gary.covatch@netl.doe.gov or 304-285-4589)
GE Global Research – Peter Sandvik (sandvik@research.ge.com or 518-387-4166)
Final Project Report [PDF-3.40MB]