WO2008064010A2

WO2008064010A2 - Fuel storage tank monitoring and phase separation detection system

Info

Publication number: WO2008064010A2
Application number: PCT/US2007/084559
Authority: WO
Inventors: Thomas Coleman; James Bruya; Alexei Zyuzin
Original assignee: Dtec Systems Llc
Priority date: 2006-11-13
Filing date: 2007-11-13
Publication date: 2008-05-29
Also published as: WO2008064010A3

Abstract

The present invention is a sensor system capable of detecting the phase separation at its earliest stages of formation as well as measuring the degree of phase separation (that is, the depth or volume of the separate ethanol/water layer which forms at the bottom of the tank). The invention can be used also to improve the accuracy of measuring contemporary capacitance-based level measurement systems which are currently in use for monitoring fuel storage tanks for inventory tracking purposes, leak detection, and water accumulation. Such accuracy and performance improvement may enable the use of capacitive water-level sensors with oxygenated fuels and even in the cases of phase separation in ethanol blends.

Description

"FUEL STORAGE TANK MONITORING AND PHASE SEPARATION DETECTION

SYSTEM"

BACKGROUND OF THE INVENTION

Phase separation is a phenomenon that occurs in fuel blends containing ethanol when even a small amount of water enters a storage tank. Because of ethanol 's high affinity for water, it has a potential to come out of the gasoline solution and form a separate water/ethanol layer at the bottom. DNR Tanks Update, Iowa Department of Natural Resources, Spring 2005. Early detection of the formation of this water/ethanol layer is very important. One of the reasons is to halt the depletion of ethanol in the bulk of gasoline. Otherwise the fuel octane rating will decrease and lowered oxygenate content will cause deterioration of the engine's emission quality. Another reason is the urgency to remove the bottom phase formed during the phase separation, because the water/ethanol mixture at the bottom is very corrosive and could damage the storage tank and associated systems.

Leaking underground storage tank (UST) systems holding petroleum fuels have been recognized as major sources of groundwater contamination, hi recent years, many thousands of USTs have been replaced, closed, or upgraded in order to comply with the federal regulations of 40 CFR part 280. Standards for design and installation of USTs have addressed the primary mode of UST failure, external corrosion. However, internal corrosion remains a significant mode of failure. As external corrosion is brought under control, internal corrosion may become the predominant failure mode. United States Environmental Protection Agency, Causes of Release from UST Systems, EPA 510-R-92-702, Sept. 1987.

The primary cause of internal corrosion hi USTs is water contamination. When condensed, water acts as an electrolyte to enable corrosion of the steel walls and welded seams. Gangadharan, A.C. et al., Leak Prevention and Corrective Action for Underground Storage Tanks, 1988.

Bacterial growth, dependent upon the presence of water, contributes to the corrosion of both steel and fiberglass tanks, and accessory equipment such as leak detectors, mixing valves, turbine pump components, and filters. ASTM, D-6469 Standard Guide to Microbial Contamination in Fuels and Fuel Systems, 2001. As both of these modes of corrosion depend on the presence of condensed water in the UST, obviously the best preventative strategy against internal corrosion would be to keep the UST and its contents as dry as possible. It should be emphasized that both of these corrosion processes are severely aggravated by the presence of ethanol in the fuel. E. Schuller, Stress Corrosion Cracking in Steel Ethanol Aboveground Storage Systems, Think Tank, issue 49, Delaware Department of Natural Resources and Environmental Control Tank Management Branch, 2006.

In practice, operators of USTs allow water to accumulate and form an aqueous layer at the bottom of a tank. When a sufficient quantity of water has formed, usually a few inches in depth, a maintenance contractor will pump it out. However, it is often not possible to remove all of the water, as: tanks may not be level or tilted away from an inserted pump inlet and may have dents or irregularities that trap water in pockets. Fitzgerald, J.H., Corrosion of Underground Storage Tanks, Plant Engineering, July 21, 1983 Worse yet, it may be difficult to remove water from a depression or pit where corrosion is already advanced. If a water layer is allowed to form, the corrosion process has started and may be difficult to stop. With the widespread and increasing use of oxygenated fuel blends this practice became unacceptable because of the risk of phase separation. DNR Tanks Update, Iowa Department of Natural Resources, Spring 2005.

It has been recently recognized that presence of ethanol in the UST has a potential to aggravate the corrosion process of the tank and its subsystems. In addition, the American Petroleum Institute (API) has recently recognized a condition that can occur in above-ground fuel ethanol storage systems constructed of steel. This condition is called stress corrosion cracking (SCC). This condition can occur at stresses less than normal design levels and significant corrosion is not required to be present for it to occur. If not detected, SCC can result in leaks with the subsequent loss of entire contents of a tank. E. Schuller, Stress Corrosion Cracking in Steel Ethanol Aboveground Storage Systems, Think Tank, issue 49, Delaware Department of Natural Resources and Environmental Control Tank Management Branch, 2006. If the leak occurs, the presence of ethanol can potentially worsen the ecological contamination because of its cosolvency with certain gasoline components. P. Ellis, Ethanol in the Environment, Think Tank, issue 49, Delaware Department of Natural Resources and Environmental Control Tank Management Branch, Spring 2006.

Water may enter a UST by precipitating out of the petroleum fuel or by condensing from the atmosphere during tank venting. In the former case, water may be introduced into a UST with the delivery of petroleum fuels. Water can find its way into gasoline fuels during refinery processing and distribution, present in both free and dissolved forms. Marshall, E.L. and Owen, K., Motor Gasoline, The Royal Society of Chemistry, Oxford, UK.

When gasoline is pumped from a delivery truck on a warm day into a cool underground tank, as the gasoline cools, its capacity to retain dissolved water decreases. Water tolerance is defined as the volume percentage of water that a gasoline blend can retain in solution at a given temperature without phase separation. If the gasoline cools to the extent that its water content exceeds its water tolerance, the excess water will precipitate from the gasoline, creating turbidity or forming a separate aqueous phase at the bottom of the tank. For ethanol blended fuel, ethanol also will partially or fully migrate into the bottom phase. Water may also condense from atmospheric air drawn in to replace product being removed from a tank. The water may condense on the sides of the tank or on the surface of the product, where there is a chance it may dissolve in the fuel. However, if the water tolerance of the product has been exceeded, additional water will contribute to the formation of a separate water or water/ethanol phase.

The actual water tolerance of a gasoline product depends greatly on whether it has been blended with an oxygenating compound. The two most common oxygenates, methyl tert-butyl ether (MTBE) and ethanol, are blended for three main reasons. First, they may be added in small concentrations (1-7 vol%) to improve octane ratings. Second, they may be blended in larger concentrations in order to comply with the U.S. Clean Air Act Amendments of 1990. Briefly, non-attainment areas of the U.S. for ozone air quality standards must use Reformulated Gasoline (RFG), containing at least 2 wt% oxygen. RFG fuels are most commonly achieved by adding either 5 vol% ethanol or 11 vol% MTBE. Non- attainment areas for carbon monoxide standards must use oxygenated fuel, which must contain at least 2.7 wt% oxygen. Oxygenated fuels are commonly achieved by adding 7.5 vol% ethanol or 15 vol% MTBE. Finally, blending ethanol into gasoline is encouraged as part of an effort to increase the use of renewable energy resources and gain independence from foreign oil production. Currently, while other oxygenates may not be blended beyond an oxygen content of 2.7 wt%, ethanol may be blended to an oxygen content of 3.7 wt% (10 vol% ethanol) under a waiver intended to increase its use. Federal excise tax exemptions and state incentives exist to make the use of ethanol more economically attractive, and therefore gasoline blended with 10 vol% ethanol is commonly distributed. Conventional gasoline, containing no oxygenates, can dissolve up to 150 parts per million (ppm) water at 70⁰F. Gasoline oxygenated with ethers such as 15 vol% MTBE can increase this solubility to 600 ppm. Gasolines blended with 10 vol% ethanol have water solubilities up to 7000 ppm. However, in addition to the greatly increased water solubilities found in ethanol-gasoline blends as compared to conventional gasoline and ether-gasoline blends, a qualitatively different behavior is observed in ethanol-gasoline blends when their water tolerance is exceeded. Contacting water- saturated conventional gasoline or ether- gasoline blends with additional water will not affect their properties. Cooling these fuels when they are water-saturated causes some of the water to precipitate as it becomes insoluble. hi the case of ether blends, very little ether will be present in the precipitated water. Owen, K. and Coley, T., Automotive Fuels Reference Book, Society of Automotive Engineers, Inc., 1995.

However, because ethanol is miscible with water, but only marginally soluble in hydrocarbons, ethanol-gasoline blends behave quite differently than conventional gasoline or ether-containing blends. Contacting ethanol-gasoline blends with a separate water layer will cause the ethanol to migrate from the gasoline phase to the water phase, for which it has greater affinity, thereby depleting the gasoline of ethanol. At equilibrium, the aqueous phase will contain approximately 75% ethanol and 25% water. Motor Gasolines Technical Review (FTR-I), Chevron Products Company, a division of Chevron U.S.A. Inc., 1996.

Consequently a substantial amount of the ethanol can shift to the aqueous phase, leaving the hydrocarbon phase with reduced ethanol concentration.

The unique behavior of ethanol-gasoline blends with water - high water solubility and nearly complete loss of ethanol upon phase-separation - increases the risk of significant internal corrosion occurring in USTs that store it. The corrosion of steel is accelerated not only by the presence of ethanol in the fuel due to the increased water content, but also due to the organic acids that can be present in commercial gasoline. Owen, K. and Coley, T., Automotive Fuels Reference Book, Society of Automotive Engineers, Inc., 1995 Additionally, bacterial growth may be stimulated by ethanol, which serves as a convenient food source. These hazards are known to the gasoline distribution and retail industry. Retail operators, concerned with the dangers of contacting ethanol-gasoline blends with free water, commonly require product delivery technicians to gauge free water in a UST before deliveries are allowed. When free water is found to be excessive, it may be necessary to postpone deliveries until the water can be removed.

Retailers of ethanol-gasoline blends have seen accelerated corrosion of their gasoline dispensing equipment due to bacterial activity or chemical corrosion, leaving them to wonder what damage remains hidden within their USTs. Industry leaders, mindful of these dangers, are concerned about increasing the use of ethanol-gasoline blends without the further development of their distribution infrastructure, handling practices, and monitoring tools.

With the goal of achieving greater energy self sufficiency, the most recent U.S. energy bill includes provisions that would dramatically increase ethanol production and use in gasoline fuels. Based on mandatory renewable fuels standard (RFS), annual ethanol production was projected to increase from 1.8 to 2.3 billion gallons by 2004 and to 5 billion gallons by 2012, nearly tripling its use. Reuters Business, April 24, 2002.

The actual production is rising even faster than expected, - 3.4 billion gallons was produced is 2004, 3.9 billion gallons - in 2005 and as of May 2006 - 4.8 billion gallons. Kansas Ethanol Homepage, US Ethanol Facts,

<http://www.ksgrains.com/ethanol/useth.html>, accessed on 10/06/2006.

Without adequate tools to monitor the water content of gasoline, this widespread deployment of ethanol-gasoline blends may accelerate corrosion that has already been occurring. It will bring ethanol containing fuels to areas without experience in its handling, leading to deleterious effects. Indeed, leaking USTs often occur at retail facilities at which the management personnel, untrained in petroleum handling, are unaware of the behaviors and conditions that lead to the failure of USTs due to internal corrosion. The need for new tools capable of monitoring UST for leaks and early phase separation detection is increasing.

Conventional methods for measuring water content within non-aqueous liquids usually fall into two categories, quantitative methods that require expensive equipment and labor, and simple methods that yield highly qualitative results. The quantitative approaches include analytical laboratory equipment and industry-specific analyzers. While water- cut analyzers have been developed for the crude oil industry, these instruments are designed to measure water present in a separate phase from the crude oil. Simple methods, such as color changing indicator chemicals, may be highly portable and easy to use, but may not provide the information desired. The conventional laboratory-based method for the measurement of water dissolved within non-aqueous liquids is Karl Fischer titration (see ASTM D 1744). While very accurate, the use of the Karl Fischer titration requires an expensive piece of equipment, the Karl Fischer titrator, and a trained technician as operator. In a facility or transport setting, standard analytical equipment is expensive, complicated, fragile, maintenance intensive, and requires trained technicians. Additionally, in a field setting, the instrument machinery may not be sufficiently compact, portable, and automated to permit practical use.

Indicator dye/colorimetric methods are known that use indicator materials that undergo changes in color when water or alcohol is present in a storage tank with petroleum fuels. U.S. Pat. No. 4,699,885 to Melpolder and Victor describes a paste that undergoes a change in color when exposed to a water phase. U.S. Pat. No. 4,604,345 to Felder and Panzer describes a paste that undergoes a change in color when exposed to a phase of alcohol or to petroleum fuels containing dissolved alcohol. Any water dissolved. within the petroleum fuel must be removed by a drying agent for the paste to properly indicate the presence of alcohol. U.S. Pat. No. 5,229,295 to Travis describes colorimetric tests for the presence of water and ethanol, and prescribes a separate step for the volumetric determination of alcohol concentration. Due to the reagent handling and restocking requirements, none of these methods is well-suited for the automated detection of phase separation in the UST. While easy to use by a non-technically trained operator, the information gained by these inventions is very limited.

Patents have been issued to inventions which determine the water content of materials by measuring the electrical properties of the materials and relating these properties to water content. U.S. Pat. No. 4,786,873 to Sherman describes a method to determine the water content of hydrocarbon-containing porous earth formations by measuring the dielectric permittivity of the earth formations. U.S. Pat. No. 3,966,973 to Henry et al. describes a process by which the moisture content of food is obtained by measuring the impedance generated by the food passing through an alternating current field. U.S. Pat. No. 6,388,453 to Greer describes a swept-frequency shunt-mode dielectric sensor system is used to measure complex impedance parameters such as capacitance and/or dielectric loss of particulate materials in order to calculate density and water content. U.S. Pat. No. 6,664,796 to Wang describes a process by which the moisture content of a fuel containing exclusively ethanol, and concentration of ethanol in the fuel, is obtained by measuring the resistance of the fuel. However such system would not be able to detect the phase separation if its formation has started. Such system would be useful only if historical information of the measured fuel composition is taken into account.

Many patents have been issued that employ sensors of dielectric properties to measure the water content associated with hydrocarbon liquids, especially crude oil. U.S. Pat. No. 5,070,725 to Cox et al. describes a water-cut meter which measures the impedance associated with crude oil and water mixtures. The percentage of water may be determined in both water continuous and oil continuous samples. U.S. Pat. No. 5,260,667 to Garcia-Golding et al. describes a method for determining the water content of oil-in- water emulsions by measuring the real part of a sample's specific admittance and by making corrections for the sample temperature.

Methods of determining water content in oil streams also include microwave technologies. U.S. Pat. Nos. 4,862,060 to Scott et al. and 5,389,883 to Harper determine water content from the frequency changes between emitted and received microwave signals caused by the dielectric properties of oil and/or water samples. Unfortunately, microwave technologies are often expensive to implement.

A particular complication found with some non-aqueous liquids, such as gasoline and diesel fuels, is that their chemical composition and resulting physicochemical properties constantly change due to the variability in the raw materials of their manufacture, the variability in processing procedures and parameters, and variability in the type and amounts of any blending chemicals added. Additionally, Federal and state laws may require the manufacture of gasoline that meets specific oxygen levels at different times of the year. Such requirements can be met through the use of various ethers or alcohols. The use of alcohols rather than ethers can have an enormous effect on the ability of the gasoline to dissolve water. K. Owen and T. Coley, "Oxygenated Blend Components for Gasoline," Automotive Fuels Reference Book, pp. 275-281, Society of Automotive Engineers, Inc., 1995.

The present invention uses multiple point physicochemical measurements that would allow calibration of capacitance-based level meters. Calibration of the level sensor for changing dielectric permittivity of the fuel would improve their accuracy in measuring oxygenated fuel blends. SUMMARY OF THE INVENTION

The present invention is directed to a sensor system capable of detecting the phase separation at its earliest stages of formation as well as measuring the degree of phase separation {that is, the depth or volume of the separate ethanol/water layer which forms at the bottom of the tank). The invention can be used also to improve the accuracy of measuring contemporary capacitance-based level measurement systems which are currently in use for monitoring fuel storage tanks for inventory tracking purposes, leak detection, and water accumulation. Such accuracy and performance improvement may enable the use of capacitive water-level sensors with oxygenated fuels and even in the cases of phase separation in ethanol blends.

More particularly, the present invention is directed to a method for sensing and detecting phase separation in gasoline/ethanol fuel blends as well as measuring the degree of phase separation (that is the depth or volume of the separate ethanol/water layer which forms at the bottom of the fuel storage tank), the method comprising the steps of: sensing the capacitance, resistance and temperature of the fuel storage tank contents at each of at least two points having different depths within the tank where at least one of the points is near the bottom of the tank within a depth at which a separate ethanol/water layer would be present under conditions at which phase separation would occur, and at least one of said points is above the level at which a separate ethanol/water layer would be present under conditions at which phase separation would occur, and determining if phase separation of a gasoline/ethanol blend has occurred within the storage tank, and/or the extent to which phase separation has occurred within the storage tank, by comparing temperature-adjusted capacitance and resistance values resulting from said sensing step at each at least one point to a plurality of predetermined values of capacitance and resistance for gasoline/ethanol fuel blends of varying concentrations.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention will be described in conjunction with the following Figures: Figure 1 is a depiction of phase separation in a gasoline/ethanol fuel blend due to water accumulation in excess of the water tolerance limit.

Figure 2 is a graphical depiction of measured dielectric permittivity of water/oxygenate mixtures. Figure 3 is a phase separation detection in gasoline/ethanoi fuel blend using two discrete measurement points.

Figure 4 is a depiction of phase separation detection in a gasoline/ethanoi fuel blend using two discrete measurement points and continuous capacitance-based level probe.

Figure 5 is a depiction of phase separation detection in a gasoline/ethanoi fuel blend using two discrete measurement points and one or more intermediate discrete measurement points.

DETAILED DESCRIPTION OF THE PRESENT INVENTION The basis for this invention is the significant difference in the dielectric properties of gasoline, ethanol, and water based on measurements of capacitance and resistance.

The table values of dielectric permittivity for such liquids are 2.0 for gasoline, 24 for ethanol, and 80 for water. The resistivity of these liquids is dependent on the presence of trace-level chemicals and is not generally assumed constant. However, it can be safely assumed that the resistivity of gasoline will be higher than the resistivity of either ethanol or water. It is expected that if phase separation occurs as shown in Figure 1, the bottom ethanol/water layer will have higher dielectric permittivity than the gasoline fuel layer above. Also, the bottom layer will become electrically more conductive. The formation of multiple layers with considerably different dielectric properties makes it impossible to accurately estimate the liquid level using a conventional capacitance level sensor.

It is assumed that the dielectric properties of the related mixtures (such as ethanol/gasoline and ethanol/water) will have a dielectric constant which is proportional to the dielectric permittivities of the constituents and their volume ratios. Figure 2 shows the experimental measurements of the dielectric permittivities of the water/ethanol and water/MTBE mixtures. One complicating factor to this is the migration of select gasoline constituents into the bottom layer based on the strength of their constituents affinities for ethanol and/or water. Such migration may lower the dielectric permittivity (as evidenced by capacitance values) of the bottom layer below that of ethanol and also may significantly lower the resistivity of the bottom layer as was observed experimentally. However this effect is in no way prohibitive to this invention and can even form the basis for the phase separation detection using resistivity values. This invention utilizes measurements of the dielectric properties as evidenced by capacitance and resistance values at a minimum of two separate positions in the fuel storage tank - one near the bottom and one in the middle or near the top, as shown in Figure 3 which depicts the phase separation of gasoline and ethanol/water phases in the same manner as Figure 1. The positions of the measurement points need to be selected in a way that the top and the bottom layers in the case of the phase separation would be measured distinctly. Increasing the number of measurement positions might improve the accuracy of the system for example in the case when there is significant diffusion between the top and the bottom layer. The respective measurements can then be compared to each other, with the relative difference, and/or respective values, indicating the presence or absence, or the extent of, phase separation.

With two separate dielectric property measurements resulting from measuring capacitance and resistance for the middle and/or top layers and the bottom layer, phase separation can be detected easily. The existence of phase separation would be indicated by a substantial difference between the measurements of capacitance and resistance at the respective measurement points, particularly when compared to standardized capacitance and resistance values for various gasoline/ethanol blends of varying concentrations.

Specifically, it is expected that the measurement at the bottom of the tank will indicate a considerably higher dielectric permittivity as measured by capacitance due to a higher concentration of water and ethanol if phase separation exists. Also, it is expected that the resistivity at the bottom position will be lower than the resistivity of the fuel at an upper portion of the tank.

For automated phase separation detection, the values of dielectric permittivities as measured by capacitance of the top and the bottom layer can be compared (subtracted from one another). If the difference is higher than some predetermined threshold difference value, the phase separation event is identified. Alternatively, the dielectric permittivity of the bottom layer can be compared to a predetermined value. If the permittivity measurement exceeds that predetermined value, the phase separation event is identified. Similarly, the dielectric permittivity of the top layer can be compared to a predetermined value. Once the permittivity of the layer falls below that value (because of ethanol migration into the bottom layer), phase separation is also identified. It should be noted that monitoring the changes in the top layer alone will result in lower detection sensitivity compared to monitoring the dielectric changes in the bottom layer. Other, more complicated algorithms may be utilized employing the combination of measurements of capacitance and resistance of the top and the bottom layers.

There are many types of sensors that can be used to measure the dielectric properties of the liquid phases in the tank as may result from a determination of capacitance and resistance. One of ordinary skill in the art can readily select which sensor would be appropriate to use in the present invention to achieve the desired result. Of course, as the temperature of the contents of the tank are also measured at the same time, one of ordinary skill in the art can also readily determine such temperature by use of conventional temperature sensors.

The simplest sensors include parallel-plate and cylindrical capacitance sensors. Fringing electric field sensors might be utilized as well. It might be possible that in addition to measuring the resistivity and capacitance of the liquid, the fringing electric field sensor might be capable of measuring the degree of cloudiness of the liquid and the properties of the colloidal phase. This might become important for applications where significant mixing of the separated phases occurs. It is also possible that gasoline degree of cloudiness may be an indicator of an onset of phase separation, and if its measurement is combined with multipoint dielectric measurements, a greater phase-separation detection system accuracy can be achieved.

The capacitance and resistance measurements might be performed using many different measurement approaches including spectroscopic and microwave techniques. These and other advanced techniques might improve the accuracy of the detection method, but even the simplest capacitance and resistance measurements techniques should be sufficient for the detection of the phase separation. The choice of the sensor and the measurement technique should depend on the particular application requirements and limitations.

One example is the presence of other capacitive sensors in the tank such as level measurement sensor. In this case it is important to select a different operating frequency for the two phase separation detection sensors (or one sensor which would be mechanically moved between the two measurement positions) in order to avoid the interference from the capacitance-based level sensor.

Since the measurement system described above provides the bulk capacitance and resistance properties of the two layers in the case of phase separation, or the properties of the single gasoline layer if there is no phase separation, it can be used to improve the accuracy of the capacitance-based level measurements by use of a continuous capacitance-based level probe as shown in Figure 4. As mentioned above, the accuracy of the level measurements is negatively affected by any change in the dielectric properties of the liquid being measured. The drawback of a typical capacitance level sensor is that it is 2-ρoint calibrated using "full" and "empty" liquid levels. The intermediate values are simply mapped in between using a linear model.

Therefore, any changes in the liquid dielectric properties or formation of separate layers with different properties will create a condition that cannot be estimated using such a simple model and calibration procedure. For example, if the dielectric constant of gasoline is to increase due to higher ethanol concentration, a typical capacitance level sensor will produce a measurement that is higher than the actual liquid level.

By combining the dielectric values obtained with the phase separation detection system with the capacitance of the conventional liquid level probe a much greater accuracy in determining the liquid level can be achieved, hi addition, it might be possible to simultaneously detect the levels of both the top and the bottom layers in the cases when phase separation actually occurs. Such measurement capability may employ an advanced mathematical algorithm. One approach is to adopt the "Maxwell capacitor" model which predicts the capacitance of a multi-layered sample with layers having arbitrary dielectric properties. By adapting such model, a mathematical relationship can be established to convert the three dielectric measurements (two from the top and the bottom sensor of the disclosed system, plus the capacitance of a common liquid level sensor) into a pair of liquid level values - one for the top and one for the bottom layer.

Another possible way of accurately detecting the degree of phase separation is to increase the number of measurement positions as shown in Figure 5. For example, a vertical array of capacitance sensors can be utilized at positions of different heights as depicted in Figure 5. The height of the array should be similar to the maximum expected depth of the bottom layer. The number of sensors in the array should be chosen appropriately for the desired accuracy of the phase separation level measurement. It should be noted that further increase in the number of sensors can improve the system performance only if there is very little mixing of the top and the bottom layers. An algorithms can be applied to each of the sensors in the array. The depth of the bottom water/ethanol layer is then measured by determining the number of sensors that are below the liquid phase boundary. By constructing more advanced algorithms, a described sensor array could also provide an estimate of the water/ethanol layer depth in case of partial mixing of the layers. Such algorithms would need to differentiate and account for the "in-between" dielectric properties that will be detected by the sensors located in the area where the layers are mixed. One of ordinary stall in the art can readily arrive at such desired algorithms.

By way of further explanation, as the noted capacitance and resistance measurements are to be compared to standard capacitance and resistance measurements for gasoline/ethanol blends of varying concentrations, the capacitance and resistance values are first temperature- adjusted so as to permit a direct comparison with such standard (pre-determined) capacitance and resistance values which are determined at a standard temperature. One of ordinary skill in the art is readily capable of accomplishing such temperature-adjustment so that the capacitance and resistance values may be appropriately compared to such pre-determined values. Such temperature adjustment is necessary as the capacitance and resistance are each temperature-dependent values.

It has also been determined that the sensing and comparing of both capacitance and resistance enables advantageous results to be achieved not otherwise obtainable by sensing and comparing only one of capacitance or resistance. While the use of only a single sensing of capacitance or resistance at multiple points within the tank may be of some value in determining the presence or extent of phase separation, it has been found that various water- soluble fuel additives migrate to the ethanol-water phase, thus potentially skewing any resulting capacitance or resistance values that may be obtained in the absence of such additives. The use of both capacitance and resistance measurements thus reduces the potential for error that may result due to measuring only one or the other.

The sensing of capacitance and resistance values at the requisite points in the fuel tank may occur periodically or continuously. It is believed preferable for such sensing to occur continuously so that any trends in the formation of the ethanol-water phase may be continuously observed. Such continuous sensing may occur by connecting the requisite sensors to a computing device which is capable of retrieving and displaying such data on a continuous basis. One of ordinary skill in the art is capable of accomplishing such continuous sensing. Examples

In order to verify that the two liquid layers formed by fuel phase separation have sufficiently different dielectric properties, phase separation conditions were created by adding sufficient amount of 18 MΩ deionized water into Union 76 regular unleaded gasoline (76 ElO). 76 ElO gasoline ethanol concentration is rated up to 10% and was experimentally verified. 1.5 L of gasoline was used in the experiment to collect enough liquid in the bottom layer to allow accurate dielectric measurements.

Custom-made stainless steel cylindrical capacitor cell with no dielectric coating was used to measure the dielectric properties of the separated liquids. The dimensions of the sensor are: length - 63.48 mm, inner cylinder o.d. — 15.50 mm, and outer cylinder i.d. - 17.28 mm. Theoretical capacitance in air is 32.5 pF not accounting for the edge effects. The liquid measurements were performed by filling 50 mL graduated cylinder with liquid and immersing the sensor in that liquid completely. Approximately 35 mL of liquid was necessary to accurately perform the measurement. A thermistor was also placed into the graduated cylinder to record the liquid temperature. The cylindrical capacitor was connected to GenRad 1689 precision RLC Digibridge. 1 V sinusoidal excitation was used. In the following experiments, the capacitance was measured at 100 kHz and the resistance at 1 kHz, unless noted otherwise.

Table 1 presents the results of verification measurements that were obtained prior to the phase separation experiment. Dielectric properties of gasoline and a gasoline/ethanol mixture prior to water addition are summarized in Table 2.

Table 1 Verification dielectric measurements

Table 2 Measured gasoline dielectric properties

To create phase separation conditions, 1.5 L of 76 ElO gasoline was poured into 1 gallon bottle with a lid. 18 MΩ deionized water was added to gasoline in three increments. After water was added, the bottle contents were mixed and left to settle for an interval of approximately 5-7 hours. After that, a sample of the bottom layer was pipetted-out into 50 mL graduated cylinder and the cylindrical capacitor sensor was immersed into the liquid. Later, a sample from the top layer was measured similarly. After the measurement, the ethanol-water-gasoline mixture was reconstituted, more water was added and the additional measurements were taken.

The dielectric measurements obtained during the phase separation experiment for three water addition steps are presented in Table 3. It is clear that there is a significant difference in dielectric properties between the bottom and the top layers of gasoline after the phase separation occurred. Furthermore, the dielectric properties of both layers are different from the properties of the gasoline before water was added and phase separation occurred. Additionally, the dielectric properties of both the top and the bottom layers change as additional amounts of water and ethanol separated from gasoline. Table 3

Dielectric measurements of the top and the bottom phases at different stages of the phase separation experiment

The above description is not intended to be limiting of the scope of the invention, but merely exemplary thereof. As such, various modifications and/or changes may be made within the scope of the invention. The invention is described in the attached claims.

Claims

WHAT IS CLAIMED IS:

1. A method for sensing and detecting the existence and/or extent of phase separation in gasoline/ethanol fuel blends, said method comprising the steps of: sensing the capacitance, resistance and temperature of the fuel storage tank contents at each of at least two points having different depths within the tank where at least one of the points is near the bottom of the tank within a depth at which a separate ethanol/water layer would be present under conditions at which phase separation would occur, and at least one of said points is above the level at which a separate ethanol/water layer would be present under conditions at which phase separation would occur, and determining if phase separation of a gasoline/ethanol blend has occurred within the storage tank, and/or the extent to which phase separation has occurred within the storage tank, by comparing temperature-adjusted capacitance and resistance values resulting from said sensing step at each at least one point to predetermined values of capacitance and resistance for gasoline/ethanol fuel blends.

2. The method of claim 1, wherein capacitance, resistance, and temperature are measured at three or more points having different depths within a fuel storage, wherein at least one of these points is near the bottom of the tank within a depth at which a separate ethanol/water layer would be present under conditions at which phase separation would occur, and at least one point is above the level at which a separate ethanol/water layer would be present under conditions which phase separation would occur.

3: The method of claim 1, wherein capacitance, resistance, and temperature are measured at two or more points having different depths within a fuel storage, wherein at least one of these points is near the bottom of the tank within a depth at which a separate ethanol/water layer would be present under conditions at which phase separation would occur, and at least one point is above the level at which a separate ethanol/water layer would be present under conditions which phase separation would occur, and wherein a continuous capacitance based level sensor probe is also present in the fuel storage tank and used together with the other capacitance, resistance, and temperature measurements and compared to a plurality of predetermined values to determine the precise depth of a separate ethanol/water phase which may have formed due to conditions which would have resulted in phase separation.

4: The method of claim 1, wherein capacitance and resistance are measured using parallel-plate sensors, cylindrical sensors, fringing electric field sensors, spectroscopic or microwave techniques.