CH05CH15-Mullins
ARI
21 March 2014
11:36
V I E W
A
Review in Advance first posted online on March 31, 2014. (Changes may still occur before final publication online and in print.)
N
I N
C E
S
R
E
D V A
Annu. Rev. Chem. Biomol. Eng. 2014.5. Downloaded from www.annualreviews.org by Istanbul Universitesi on 04/25/14. For personal use only.
Downhole Fluid Analysis and Asphaltene Science for Petroleum Reservoir Evaluation Oliver C. Mullins,1 Andrew E. Pomerantz,1 Julian Y. Zuo,2 and Chengli Dong3 1
Schlumberger-Doll Research, Cambridge, Massachusetts 02139; email: [email protected]
2
Schlumberger DBR Technology Center, Edmonton, Alberta T6N 1M9, Canada
3
Shell International Exploration and Production Company, Houston, Texas 77079
Annu. Rev. Chem. Biomol. Eng. 2014. 5:325–45
Keywords
The Annual Review of Chemical and Biomolecular Engineering is online at chembioeng.annualreviews.org
asphaltenes, downhole fluid analysis, Yen-Mullins model, Flory-Huggins-Zuo equation of state, reservoir evaluation
This article’s doi: 10.1146/annurev-chembioeng-060713-035923 c 2014 by Annual Reviews. Copyright All rights reserved
Abstract Petroleum reservoirs are enshrouded in mysteries associated with all manner of geologic and fluid complexities that Mother Nature can inspire. Efficient exploitation of petroleum reservoirs mandates elucidation of these complexities; downhole fluid analysis (DFA) has proven to be indispensable for understanding both fluids and reservoir architecture. Crude oil consists of dissolved gases, liquids, and dissolved solids, known as the asphaltenes. These different fluid components exhibit fluid gradients vertically and laterally, which are best revealed by DFA, with its excellent precision and accuracy. Compositional gradient analysis falls within the purview of thermodynamics. Gas-liquid equilibria can be treated with a cubic equation of state (EoS), such as the Peng-Robinson EoS, a modified van der Waals EoS. In contrast, the first EoS for asphaltene gradients, the Flory-Huggins-Zuo (FHZ) EoS, was developed only recently. The resolution of the asphaltene molecular and nanocolloidal species in crude oil, which is codified in the Yen-Mullins model of asphaltenes, enabled the development of this EoS. The combination of DFA characterization of gradients of reservoir crude oil with the cubic EoS and FHZ EoS analyses brings into view wide-ranging reservoir concerns, such as reservoir connectivity, fault-block migration, heavy oil gradients, tar mat formation, huge disequilibrium fluid gradients, and even stochastic variations of reservoir fluids. New petroleum science and DFA technology are helping to offset the increasing costs and technical difficulties of exploiting ever-more-remote petroleum reservoirs.
325
Changes may still occur before final publication online and in print
CH05CH15-Mullins
ARI
21 March 2014
11:36
INTRODUCTION
Annu. Rev. Chem. Biomol. Eng. 2014.5. Downloaded from www.annualreviews.org by Istanbul Universitesi on 04/25/14. For personal use only.
Oil and gas reservoirs exhibit an enormous array of categories, structures, sizes, and shapes, with fluids ranging from dry gas to tar and everything in between, commensurate with the myriad of geophysical and geochemical processes that give rise to these reservoirs. Indeed, the term oil reservoir indicates those oil-bearing subsurface structures that can be exploited economically; thus, it is dependent on the price of oil and gas and therefore is dependent on the ever-improving technology that can be implemented. The understanding of the structure and properties of a given oil reservoir improves dramatically with production from the field (1). However, in many settings, predictions of production are needed prior to any production; this circumstance is especially applicable to deepwater oil production (1, 2). Individual projects can cost billions of dollars, and unexpected shortfalls of production can shift the economics steeply negative. One of the biggest uncertainties of oil reservoirs is the size of the producing unit (1, 2). Oilfields can consist of giant flow units or compartments where permeable formations are well connected vertically and laterally. In contrast, oilfields can consist of a large number of small compartments, where each compartment requires penetration by a well for drainage. Each well in this type of reservoir will access only a small number of the compartments, and total production will be very limited. An industry-wide study of production from 28 oilfields in deepwater Gulf of Mexico determined that 75% of the reservoirs underperformed in both rate and total recovery, primarily owing to unrecognized compartmentalization (3). Nevertheless, in deepwater, it is particularly important to predict production prior to any production taking place. This is because deepwater production facilities are very expensive and must be in place prior to production of the field. Pressure measurements and acquisition of fluid samples are always required to understand various production parameters (2, 4). Pressure connectivity is a necessary but insufficient condition to establish flow connectivity. New methods of reservoir connectivity analysis are mandated.
Reservoir Crude Oils Reservoir hydrocarbons exhibit enormous variability, from methane to tar and everything in between. Generally, live crude oils (those at reservoir conditions) contain dissolved hydrocarbon gases, liquids, and dissolved or colloidally suspended solids, known as the asphaltenes (cf. Figure 1). The dissolved gas is described by the gas-oil ratio at specific surface conditions (pressure = 1 atm, temperature = 60◦ F); typical units are standard cubic feet of gas per barrel of oil. The asphaltene fraction is listed as mass fraction (5). Reservoir fluids often vary in composition, principally in terms of the concentrations of the dissolved gas, the gas-oil ratio (GOR), and the asphaltene content, in large part owing to gravity. Dynamical processes involving reservoir fluids that are recent in geologic time can also create vertical and lateral fluid gradients in reservoirs. Gas-liquid equilibrium modeling of reservoir fluids can be accomplished by using cubic equations of state (EoS), which are extensions of the original cubic EoS, the van der Waals equation. For example, the Peng-Robinson EoS provides significant utility for reservoir crude oils (5). Until recently, there had been no EoS for asphaltene gradients. The origin of this shortcoming was a lack of consensus on the molecular size or colloidal structures of asphaltene in crude oils and even in laboratory solvents. Even estimates of the asphaltene molecular weight spanned orders of magnitude (6, 7). This situation has been resolved (7, 8); the molecular and nanocolloidal species found in crude oil have been codified in the Yen-Mullins model (see below) (8, 9). This has led to development of the industry’s first EoS for asphaltene gradients in reservoir fluids, the Flory-Huggins-Zuo (FHZ) EoS (see below) (10, 11). 326
Mullins et al.
Changes may still occur before final publication online and in print
CH05CH15-Mullins
ARI
21 March 2014
11:36
Gas + liquid
Fluid + (suspended) solid
+ Hydrocarbon liquids
Gas
Asphaltenes
Annu. Rev. Chem. Biomol. Eng. 2014.5. Downloaded from www.annualreviews.org by Istanbul Universitesi on 04/25/14. For personal use only.
Figure 1 Reservoir crude oils generally contain dissolved hydrocarbon gas, liquid hydrocarbons, and dissolved or colloidally suspended solids, known as the asphaltenes (2). The ratios of these three phases often change with depth. Measuring and thermodynamic modeling of these dissolved phases are very instructive about both the reservoir fluids and the reservoir architecture.
Downhole Fluid Analysis Reservoir fluid samples are generally acquired in many or all of the wells drilled into a field. The samples are acquired in open hole, immediately after the well is drilled and prior to completing the well with steel casing. Figure 2 shows the sampling tool, the modular formation dynamics tester (MDT) (12). Figure 2a shows the probe, a steel tube that communicates hydraulically with
a
b The MDT
Flowline Exit port Sample bottles 1st optical analyzer
Probe Pump
2nd optical analyzer Flowline Exit port Figure 2 The modular formation dynamics tester (MDT) tool (12) used to perform downhole fluid analysis (DFA) and also to capture samples for further laboratory analysis (2). (a) The probe has a steel tube that extends to establish hydraulic communication with permeable zones, thereby enabling extraction of formation fluids. (b) Different modules of the MDT, including optical fluid analyzers that form the foundation of DFA. www.annualreviews.org • Petroleum Reservoir Evaluation
Changes may still occur before final publication online and in print
327
CH05CH15-Mullins
ARI
21 March 2014
11:36
4 Light oil Medium oil
3
OD 2
Heavy oil
Water Condensate
1
0
OBM filtrate 500
1,000
1,500
2,000
Annu. Rev. Chem. Biomol. Eng. 2014.5. Downloaded from www.annualreviews.org by Istanbul Universitesi on 04/25/14. For personal use only.
Wavelength (nm) Figure 3 The visible and near-infrared spectra (2-mm path length) for various crude oils (and an oil-based drilling fluid filtrate) and for water (2). Water and oil are easily resolved using the overtone vibrations. Different crude oils are seen to have widely differing magnitudes of electronic absorption traceable to different asphaltene concentrations. Abbreviations: OBM, oil-based mud; OD, optical density.
the permeable formation, and Figure 2b shows a schematic of the entire tool string. A pump is used to withdraw fluid from the formation, move it through the optical analyzes, and then expel it into the borehole or, when desired, capture the fluids into sample bottles. The MDT enables in situ downhole fluid analysis (DFA) measurements to be made of fluid samples in oil wells as they are pumped through the tool (2). In addition, the MDT allows for samples to be acquired for further laboratory analysis (2). The first DFA measurements are based on visible and near-infrared spectroscopy (2, 13, 14). Figure 3 shows the spectra of various crude oils and of water. Oil and water are readily differentiated by using the respective two-stretch vibrational overtones of different carbon-hydrogen (CH) groups at ∼1,725 nm and oxygen-hydrogen (OH) groups at ∼1,450 nm. In addition, different oil types have very different electronic absorption, with monotonically increasing absorption at shorter wavelengths (2, 7, 8). The coloration is linear in asphaltene content (15–17), providing a simple means to measure the dissolved asphaltene content. The range of colors includes yellow, tan, brown, and black, with corresponding spectra shown in Figure 4a. The green color sometimes observed in crude oils is due in part to fluorescence; some light oils even appear blue owing to fluorescence (18). Crude oils and asphaltenes obey the Urbach formalism (19) in the electronic absorption spectra, as shown in Figure 4b. (20, 21). In the standard Urbach observation, the electronic absorption edge corresponds to thermal excitation of absorbers, in accordance with the Boltzmann distribution. Thus, absorption declines exponentially at increasing wavelengths owing to fewer excited absorbers. For crude oils, the Urbach slope is 10 kT, not the typical kT. This is because for crude oils there is a thermal production of larger chromophores from smaller chromophores. The absorbers in crude oil are polycyclic aromatic hydrocarbons (PAHs) (16, 22, 23). The large PAHs generally absorb longer wavelength light in accord with the quantum particle in a box (16, 22, 23). Graphite, with its infinite two-dimensional sheets of aromatic rings, is the limiting case of this effect. Detailed analysis of the optical spectra of crude oils and asphaltenes, both absorption and emission in both the singlet-state (16, 22, 23) and triplet-state (24) manifolds, shows that asphaltene PAHs predominantly have seven fused rings with a distribution of four to ten rings, and asphaltenes have the largest PAHs of any component 328
Mullins et al.
Changes may still occur before final publication online and in print
b
1
0.1
0.5 3
500
1,000
1,500
2,000
2,500
0.01 5,000 10,000 15,000 20,000 25,000 30,000
Wavelength (nm) Annu. Rev. Chem. Biomol. Eng. 2014.5. Downloaded from www.annualreviews.org by Istanbul Universitesi on 04/25/14. For personal use only.
Gas condensate
OD Heavy oil
OD 1.5
1 Black oil
2
a
Asphalt
2.5
11:36
Black oil
3
21 March 2014
Asphalt
ARI
Condensate
CH05CH15-Mullins
Wavelength (nm)
Figure 4 Visible near-infrared (NIR) spectra of crude oils. (a) Huge variation of the coloration of crude oils is evident; this is due to the variable concentration of asphaltenes. (b) When the spectra are plotted on a logarithmic scale versus photon energy, all crude oil spectra exhibit linear plots of the same slope, in accordance with the Urbach formalism from solid-state physics (2). This physics simplicity helps for measurement of asphaltene content in situ in oil wells. Abbreviation: OD, optical density.
in crude oil. This work relies on the characterization of the type of aromatic carbon that was obtained from carbon X-ray Raman spectroscopy (25). Dissolved gas content of crude oils can be measured by analyzing the CH overtone regions at ∼1,725 nm. Figure 5a shows the resolution of methane from heptane; the spectra of a solution of methane and n-heptane equals the weight sum of the individual spectra. Linearity of vibrational overtone spectra is expected for the weakly interacting alkanes. Figure 5b shows the significant spectral differences for oils with differing amounts of dissolved gases (14, 26, 27). Figure 5c shows that the spectra can yield GOR (14, 26, 27). More recently, the In situ Fluid Analyzer (IFA), a 36-channel optical analyzer equipped with both a filter and a grating spectrometer, has been introduced globally in the oilfield (28). The interpreted products include C1–C5, C6+, relative asphaltene content, CO2 , and fluorescence (28). Other interpreted products, such as C3 determination and H2 S, are under development. The IFA allows for a closer look at reservoir variations than was previously possible and thereby provides a more stringent test of proposed mechanisms producing these variations.
Monitoring Mud Filtrate Contamination of Crude Oil The first application of DFA has been to monitor the fluid flow stream in the MDT as a function of pumping time. Wells are drilled with water-based or oil-based muds, which are maintained at higher-than-formation pressure to avoid blowouts. When a new permeable zone is first penetrated by the drill bit, wellbore fluids rapidly flow into the permeable zone. Drilling fluids contain suspended clays and other solid materials that cannot migrate into the rock and thereby build a mudcake on the borehole wall, which greatly reduces the fluid loss. Nevertheless, some drilling fluid filtrate flows into the permeable formation, as depicted in Figure 6. The focused sampling probe provides a major improvement in contamination reduction (cf. Figure 6), with a central sampling probe and a second annular guard probe (29). Mud filtrate– contaminated fluid enters the formation essentially in an annulus just inside the formation next www.annualreviews.org • Petroleum Reservoir Evaluation
Changes may still occur before final publication online and in print
329
CH05CH15-Mullins
ARI
21 March 2014
11:36
a 0.8 Methane n-Heptane CH4 + n-C7H16
0.7 0.6 0.5
OD 0.4 0.3 0.2
0 1,550 1,600 1,650 1,700 1,750 1,800 1,850 1,900 1,950
Wavelength (nm)
b
1
1
CH4 GOR = 4,370 GOR = 3,000 GOR = 2,050 Dead crude
0.8
OD
0.6 0.4
c
0.8
OD ratio
Annu. Rev. Chem. Biomol. Eng. 2014.5. Downloaded from www.annualreviews.org by Istanbul Universitesi on 04/25/14. For personal use only.
0.1
0.6
0.4 Live crude oils Binary mixture (*0.85) Theory: eqs. 2,3 (*0.85)
0.2
0.2 0 1,550 1,600 1,650 1,700 1,750 1,800 1,850 1,900
Wavelength (nm)
0 0
10,000
20,000
30,000
GOR (scf/stb)
Figure 5 (a) Near-infrared spectra showing the carbon-hydrogen (CH) two stretch overtones of methane, n-heptane, and a mixture of these hydrocarbons; the spectra add linearly with composition (14, 26). (b) NIR spectra of several crude oils of differing gas-oil ratio (GOR) show corresponding spectral changes associated with increasing methane and decreasing liquid hydrocarbons (14). (c) Interpretation algorithms can then give GOR and some compositional information about the crude oils, especially regarding their light ends (14, 26, 27).
to the borehole wall. The guard probe removes this contaminated fluid, which flows into the probe from points near the borehole wall. The central sample probe preferentially collects fluid from deeper in the formation, which is enriched in crude oil or formation fluid (cf. Figure 6). Contamination levels of
ARI
21 March 2014
11:36
V I E W
A
Review in Advance first posted online on March 31, 2014. (Changes may still occur before final publication online and in print.)
N
I N
C E
S
R
E
D V A
Annu. Rev. Chem. Biomol. Eng. 2014.5. Downloaded from www.annualreviews.org by Istanbul Universitesi on 04/25/14. For personal use only.
Downhole Fluid Analysis and Asphaltene Science for Petroleum Reservoir Evaluation Oliver C. Mullins,1 Andrew E. Pomerantz,1 Julian Y. Zuo,2 and Chengli Dong3 1
Schlumberger-Doll Research, Cambridge, Massachusetts 02139; email: [email protected]
2
Schlumberger DBR Technology Center, Edmonton, Alberta T6N 1M9, Canada
3
Shell International Exploration and Production Company, Houston, Texas 77079
Annu. Rev. Chem. Biomol. Eng. 2014. 5:325–45
Keywords
The Annual Review of Chemical and Biomolecular Engineering is online at chembioeng.annualreviews.org
asphaltenes, downhole fluid analysis, Yen-Mullins model, Flory-Huggins-Zuo equation of state, reservoir evaluation
This article’s doi: 10.1146/annurev-chembioeng-060713-035923 c 2014 by Annual Reviews. Copyright All rights reserved
Abstract Petroleum reservoirs are enshrouded in mysteries associated with all manner of geologic and fluid complexities that Mother Nature can inspire. Efficient exploitation of petroleum reservoirs mandates elucidation of these complexities; downhole fluid analysis (DFA) has proven to be indispensable for understanding both fluids and reservoir architecture. Crude oil consists of dissolved gases, liquids, and dissolved solids, known as the asphaltenes. These different fluid components exhibit fluid gradients vertically and laterally, which are best revealed by DFA, with its excellent precision and accuracy. Compositional gradient analysis falls within the purview of thermodynamics. Gas-liquid equilibria can be treated with a cubic equation of state (EoS), such as the Peng-Robinson EoS, a modified van der Waals EoS. In contrast, the first EoS for asphaltene gradients, the Flory-Huggins-Zuo (FHZ) EoS, was developed only recently. The resolution of the asphaltene molecular and nanocolloidal species in crude oil, which is codified in the Yen-Mullins model of asphaltenes, enabled the development of this EoS. The combination of DFA characterization of gradients of reservoir crude oil with the cubic EoS and FHZ EoS analyses brings into view wide-ranging reservoir concerns, such as reservoir connectivity, fault-block migration, heavy oil gradients, tar mat formation, huge disequilibrium fluid gradients, and even stochastic variations of reservoir fluids. New petroleum science and DFA technology are helping to offset the increasing costs and technical difficulties of exploiting ever-more-remote petroleum reservoirs.
325
Changes may still occur before final publication online and in print
CH05CH15-Mullins
ARI
21 March 2014
11:36
INTRODUCTION
Annu. Rev. Chem. Biomol. Eng. 2014.5. Downloaded from www.annualreviews.org by Istanbul Universitesi on 04/25/14. For personal use only.
Oil and gas reservoirs exhibit an enormous array of categories, structures, sizes, and shapes, with fluids ranging from dry gas to tar and everything in between, commensurate with the myriad of geophysical and geochemical processes that give rise to these reservoirs. Indeed, the term oil reservoir indicates those oil-bearing subsurface structures that can be exploited economically; thus, it is dependent on the price of oil and gas and therefore is dependent on the ever-improving technology that can be implemented. The understanding of the structure and properties of a given oil reservoir improves dramatically with production from the field (1). However, in many settings, predictions of production are needed prior to any production; this circumstance is especially applicable to deepwater oil production (1, 2). Individual projects can cost billions of dollars, and unexpected shortfalls of production can shift the economics steeply negative. One of the biggest uncertainties of oil reservoirs is the size of the producing unit (1, 2). Oilfields can consist of giant flow units or compartments where permeable formations are well connected vertically and laterally. In contrast, oilfields can consist of a large number of small compartments, where each compartment requires penetration by a well for drainage. Each well in this type of reservoir will access only a small number of the compartments, and total production will be very limited. An industry-wide study of production from 28 oilfields in deepwater Gulf of Mexico determined that 75% of the reservoirs underperformed in both rate and total recovery, primarily owing to unrecognized compartmentalization (3). Nevertheless, in deepwater, it is particularly important to predict production prior to any production taking place. This is because deepwater production facilities are very expensive and must be in place prior to production of the field. Pressure measurements and acquisition of fluid samples are always required to understand various production parameters (2, 4). Pressure connectivity is a necessary but insufficient condition to establish flow connectivity. New methods of reservoir connectivity analysis are mandated.
Reservoir Crude Oils Reservoir hydrocarbons exhibit enormous variability, from methane to tar and everything in between. Generally, live crude oils (those at reservoir conditions) contain dissolved hydrocarbon gases, liquids, and dissolved or colloidally suspended solids, known as the asphaltenes (cf. Figure 1). The dissolved gas is described by the gas-oil ratio at specific surface conditions (pressure = 1 atm, temperature = 60◦ F); typical units are standard cubic feet of gas per barrel of oil. The asphaltene fraction is listed as mass fraction (5). Reservoir fluids often vary in composition, principally in terms of the concentrations of the dissolved gas, the gas-oil ratio (GOR), and the asphaltene content, in large part owing to gravity. Dynamical processes involving reservoir fluids that are recent in geologic time can also create vertical and lateral fluid gradients in reservoirs. Gas-liquid equilibrium modeling of reservoir fluids can be accomplished by using cubic equations of state (EoS), which are extensions of the original cubic EoS, the van der Waals equation. For example, the Peng-Robinson EoS provides significant utility for reservoir crude oils (5). Until recently, there had been no EoS for asphaltene gradients. The origin of this shortcoming was a lack of consensus on the molecular size or colloidal structures of asphaltene in crude oils and even in laboratory solvents. Even estimates of the asphaltene molecular weight spanned orders of magnitude (6, 7). This situation has been resolved (7, 8); the molecular and nanocolloidal species found in crude oil have been codified in the Yen-Mullins model (see below) (8, 9). This has led to development of the industry’s first EoS for asphaltene gradients in reservoir fluids, the Flory-Huggins-Zuo (FHZ) EoS (see below) (10, 11). 326
Mullins et al.
Changes may still occur before final publication online and in print
CH05CH15-Mullins
ARI
21 March 2014
11:36
Gas + liquid
Fluid + (suspended) solid
+ Hydrocarbon liquids
Gas
Asphaltenes
Annu. Rev. Chem. Biomol. Eng. 2014.5. Downloaded from www.annualreviews.org by Istanbul Universitesi on 04/25/14. For personal use only.
Figure 1 Reservoir crude oils generally contain dissolved hydrocarbon gas, liquid hydrocarbons, and dissolved or colloidally suspended solids, known as the asphaltenes (2). The ratios of these three phases often change with depth. Measuring and thermodynamic modeling of these dissolved phases are very instructive about both the reservoir fluids and the reservoir architecture.
Downhole Fluid Analysis Reservoir fluid samples are generally acquired in many or all of the wells drilled into a field. The samples are acquired in open hole, immediately after the well is drilled and prior to completing the well with steel casing. Figure 2 shows the sampling tool, the modular formation dynamics tester (MDT) (12). Figure 2a shows the probe, a steel tube that communicates hydraulically with
a
b The MDT
Flowline Exit port Sample bottles 1st optical analyzer
Probe Pump
2nd optical analyzer Flowline Exit port Figure 2 The modular formation dynamics tester (MDT) tool (12) used to perform downhole fluid analysis (DFA) and also to capture samples for further laboratory analysis (2). (a) The probe has a steel tube that extends to establish hydraulic communication with permeable zones, thereby enabling extraction of formation fluids. (b) Different modules of the MDT, including optical fluid analyzers that form the foundation of DFA. www.annualreviews.org • Petroleum Reservoir Evaluation
Changes may still occur before final publication online and in print
327
CH05CH15-Mullins
ARI
21 March 2014
11:36
4 Light oil Medium oil
3
OD 2
Heavy oil
Water Condensate
1
0
OBM filtrate 500
1,000
1,500
2,000
Annu. Rev. Chem. Biomol. Eng. 2014.5. Downloaded from www.annualreviews.org by Istanbul Universitesi on 04/25/14. For personal use only.
Wavelength (nm) Figure 3 The visible and near-infrared spectra (2-mm path length) for various crude oils (and an oil-based drilling fluid filtrate) and for water (2). Water and oil are easily resolved using the overtone vibrations. Different crude oils are seen to have widely differing magnitudes of electronic absorption traceable to different asphaltene concentrations. Abbreviations: OBM, oil-based mud; OD, optical density.
the permeable formation, and Figure 2b shows a schematic of the entire tool string. A pump is used to withdraw fluid from the formation, move it through the optical analyzes, and then expel it into the borehole or, when desired, capture the fluids into sample bottles. The MDT enables in situ downhole fluid analysis (DFA) measurements to be made of fluid samples in oil wells as they are pumped through the tool (2). In addition, the MDT allows for samples to be acquired for further laboratory analysis (2). The first DFA measurements are based on visible and near-infrared spectroscopy (2, 13, 14). Figure 3 shows the spectra of various crude oils and of water. Oil and water are readily differentiated by using the respective two-stretch vibrational overtones of different carbon-hydrogen (CH) groups at ∼1,725 nm and oxygen-hydrogen (OH) groups at ∼1,450 nm. In addition, different oil types have very different electronic absorption, with monotonically increasing absorption at shorter wavelengths (2, 7, 8). The coloration is linear in asphaltene content (15–17), providing a simple means to measure the dissolved asphaltene content. The range of colors includes yellow, tan, brown, and black, with corresponding spectra shown in Figure 4a. The green color sometimes observed in crude oils is due in part to fluorescence; some light oils even appear blue owing to fluorescence (18). Crude oils and asphaltenes obey the Urbach formalism (19) in the electronic absorption spectra, as shown in Figure 4b. (20, 21). In the standard Urbach observation, the electronic absorption edge corresponds to thermal excitation of absorbers, in accordance with the Boltzmann distribution. Thus, absorption declines exponentially at increasing wavelengths owing to fewer excited absorbers. For crude oils, the Urbach slope is 10 kT, not the typical kT. This is because for crude oils there is a thermal production of larger chromophores from smaller chromophores. The absorbers in crude oil are polycyclic aromatic hydrocarbons (PAHs) (16, 22, 23). The large PAHs generally absorb longer wavelength light in accord with the quantum particle in a box (16, 22, 23). Graphite, with its infinite two-dimensional sheets of aromatic rings, is the limiting case of this effect. Detailed analysis of the optical spectra of crude oils and asphaltenes, both absorption and emission in both the singlet-state (16, 22, 23) and triplet-state (24) manifolds, shows that asphaltene PAHs predominantly have seven fused rings with a distribution of four to ten rings, and asphaltenes have the largest PAHs of any component 328
Mullins et al.
Changes may still occur before final publication online and in print
b
1
0.1
0.5 3
500
1,000
1,500
2,000
2,500
0.01 5,000 10,000 15,000 20,000 25,000 30,000
Wavelength (nm) Annu. Rev. Chem. Biomol. Eng. 2014.5. Downloaded from www.annualreviews.org by Istanbul Universitesi on 04/25/14. For personal use only.
Gas condensate
OD Heavy oil
OD 1.5
1 Black oil
2
a
Asphalt
2.5
11:36
Black oil
3
21 March 2014
Asphalt
ARI
Condensate
CH05CH15-Mullins
Wavelength (nm)
Figure 4 Visible near-infrared (NIR) spectra of crude oils. (a) Huge variation of the coloration of crude oils is evident; this is due to the variable concentration of asphaltenes. (b) When the spectra are plotted on a logarithmic scale versus photon energy, all crude oil spectra exhibit linear plots of the same slope, in accordance with the Urbach formalism from solid-state physics (2). This physics simplicity helps for measurement of asphaltene content in situ in oil wells. Abbreviation: OD, optical density.
in crude oil. This work relies on the characterization of the type of aromatic carbon that was obtained from carbon X-ray Raman spectroscopy (25). Dissolved gas content of crude oils can be measured by analyzing the CH overtone regions at ∼1,725 nm. Figure 5a shows the resolution of methane from heptane; the spectra of a solution of methane and n-heptane equals the weight sum of the individual spectra. Linearity of vibrational overtone spectra is expected for the weakly interacting alkanes. Figure 5b shows the significant spectral differences for oils with differing amounts of dissolved gases (14, 26, 27). Figure 5c shows that the spectra can yield GOR (14, 26, 27). More recently, the In situ Fluid Analyzer (IFA), a 36-channel optical analyzer equipped with both a filter and a grating spectrometer, has been introduced globally in the oilfield (28). The interpreted products include C1–C5, C6+, relative asphaltene content, CO2 , and fluorescence (28). Other interpreted products, such as C3 determination and H2 S, are under development. The IFA allows for a closer look at reservoir variations than was previously possible and thereby provides a more stringent test of proposed mechanisms producing these variations.
Monitoring Mud Filtrate Contamination of Crude Oil The first application of DFA has been to monitor the fluid flow stream in the MDT as a function of pumping time. Wells are drilled with water-based or oil-based muds, which are maintained at higher-than-formation pressure to avoid blowouts. When a new permeable zone is first penetrated by the drill bit, wellbore fluids rapidly flow into the permeable zone. Drilling fluids contain suspended clays and other solid materials that cannot migrate into the rock and thereby build a mudcake on the borehole wall, which greatly reduces the fluid loss. Nevertheless, some drilling fluid filtrate flows into the permeable formation, as depicted in Figure 6. The focused sampling probe provides a major improvement in contamination reduction (cf. Figure 6), with a central sampling probe and a second annular guard probe (29). Mud filtrate– contaminated fluid enters the formation essentially in an annulus just inside the formation next www.annualreviews.org • Petroleum Reservoir Evaluation
Changes may still occur before final publication online and in print
329
CH05CH15-Mullins
ARI
21 March 2014
11:36
a 0.8 Methane n-Heptane CH4 + n-C7H16
0.7 0.6 0.5
OD 0.4 0.3 0.2
0 1,550 1,600 1,650 1,700 1,750 1,800 1,850 1,900 1,950
Wavelength (nm)
b
1
1
CH4 GOR = 4,370 GOR = 3,000 GOR = 2,050 Dead crude
0.8
OD
0.6 0.4
c
0.8
OD ratio
Annu. Rev. Chem. Biomol. Eng. 2014.5. Downloaded from www.annualreviews.org by Istanbul Universitesi on 04/25/14. For personal use only.
0.1
0.6
0.4 Live crude oils Binary mixture (*0.85) Theory: eqs. 2,3 (*0.85)
0.2
0.2 0 1,550 1,600 1,650 1,700 1,750 1,800 1,850 1,900
Wavelength (nm)
0 0
10,000
20,000
30,000
GOR (scf/stb)
Figure 5 (a) Near-infrared spectra showing the carbon-hydrogen (CH) two stretch overtones of methane, n-heptane, and a mixture of these hydrocarbons; the spectra add linearly with composition (14, 26). (b) NIR spectra of several crude oils of differing gas-oil ratio (GOR) show corresponding spectral changes associated with increasing methane and decreasing liquid hydrocarbons (14). (c) Interpretation algorithms can then give GOR and some compositional information about the crude oils, especially regarding their light ends (14, 26, 27).
to the borehole wall. The guard probe removes this contaminated fluid, which flows into the probe from points near the borehole wall. The central sample probe preferentially collects fluid from deeper in the formation, which is enriched in crude oil or formation fluid (cf. Figure 6). Contamination levels of
