Hydrogen Index

The hydrogen index (HI) in nuclear magnetic resonance (NMR) logging is defined as the number of hydrogen atoms per unit volume of a fluid divided by the number of hydrogen atoms per unit volume of pure water at surface conditions (approximately 6.69 x 10^22 hydrogen atoms per cubic centimeter), expressed as a dimensionless ratio or fraction that determines how much NMR signal a fluid contributes relative to the same volume of water, with HI of exactly 1.00 for pure water (the calibration reference), HI of approximately 0.85 to 0.98 for most liquid crude oils (slightly less than water due to the lower hydrogen density of longer-chain hydrocarbon molecules compared to water), HI of 0.95 to 1.00 for liquid natural gas condensates (light hydrocarbons near water in hydrogen density), HI of 0.01 to 0.4 for natural gas (dramatically lower than water because gas at reservoir conditions has far fewer hydrogen atoms per unit volume than liquid water, and the HI varies strongly with gas pressure, temperature, and composition), and HI of approximately 0 for solid organics, dry clay minerals, and crystalline minerals (which have negligible mobile hydrogen content); the hydrogen index is a critical parameter in NMR log interpretation because the NMR tool measures the total magnetic polarization of hydrogen nuclei in the formation, and any fluid with HI less than 1 will cause the NMR apparent porosity to underestimate the true porosity of the rock by the factor (1 - HI), a significant problem in gas reservoirs where the low HI of free gas causes the NMR to underread the gas-filled porosity.

Key Takeaways

  • The hydrogen index of natural gas varies strongly with pressure and temperature in a manner that is critical for NMR interpretation in gas reservoirs: at very high pressures (above 7,000 to 10,000 psi) and elevated temperatures, gas becomes supercritical and its density approaches that of a light liquid, with HI rising from 0.02 to 0.05 at shallow low-pressure conditions to 0.4 to 0.7 at deepwater ultra-high-pressure conditions where a cubic centimeter of gas-phase hydrocarbons contains nearly as many hydrogen atoms as a cubic centimeter of liquid oil; the pressure and temperature dependence of gas HI follows from the real-gas equation of state (using Z-factor to relate actual gas molar volume to ideal gas molar volume) combined with the hydrogen atom density of the specific gas composition; in deepwater wells drilled through pressurized gas sands (where reservoir pressure may be 12,000 to 15,000 psi and temperature 150 to 200 degrees Celsius), the gas HI may be 0.3 to 0.5, causing the NMR to underestimate gas porosity by only 50 to 70 percent rather than the 90 to 95 percent underestimate that would occur for the same gas at surface conditions, making the NMR porosity more useful as a gas indicator tool in deepwater reservoirs than in shallow wells.
  • The gas effect on NMR porosity is used as a qualitative gas detection indicator: when NMR-derived porosity (which reflects only the hydrogen-bearing pore fluids, weighted by their HI) is compared to density-derived porosity (which reflects all pore fluids plus the matrix and organic carbon, weighted by their electron density) and neutron-derived porosity (which reflects all hydrogen-bearing materials including clay water, structural water, and pore fluids), the presence of free gas creates a characteristic crossplot response where NMR porosity is lower than density porosity (because gas has lower HI than the water assumed in the density porosity calibration), and density porosity is lower than neutron porosity (the classic gas crossover of the density-neutron crossplot, because gas has lower electron density than water causing density to over-read porosity, and gas has lower hydrogen density than water causing neutron to under-read porosity, creating the crossover); the three-porosity comparison (NMR versus density versus neutron) is more diagnostic of gas than the two-porosity gas crossover alone, and has been applied in deepwater exploration to confirm gas saturation in reservoirs where resistivity-based gas identification is ambiguous due to high-salinity formation water masking the resistivity contrast of the gas.
  • Bitumen and heavy oil hydrogen index effects are the opposite of gas effects: heavy crude oils with API gravity below 20 degrees have hydrogen index values of 0.90 to 0.98 at reservoir conditions, close to but slightly below water HI, which means the NMR slightly underestimates total porosity in heavy oil reservoirs but the error is small compared to the gas HI effect; however, bitumen (the solid to very viscous organic material in oil sands and heavy oil deposits) has a hydrogen index near 1.0 but an extremely short T2 relaxation time (below the instrument dead time of 0.2 to 0.5 milliseconds), meaning that the bitumen hydrogen is present and polarized but relaxes so quickly that it cannot be detected by the NMR tool; the result is that NMR porosity in bitumen-bearing sands underestimates total porosity by the fraction of the pore space occupied by bitumen, underestimates producible fluid porosity because the bitumen pore space is not producible without thermal stimulation, and accurately represents the mobile fluid porosity (water plus free oil) that is producible by primary or cold production methods; recognizing the bitumen dead-time effect versus the gas HI effect versus the reduced-HI light hydrocarbon effect is one of the key interpretation challenges in NMR log analysis of oil sands, heavy oil, and gas reservoirs.
  • NMR tool calibration to hydrogen index uses pure water as the primary reference standard, defining the NMR response per unit volume of 100 percent water-saturated porosity as the baseline against which all other measurements are referenced: before logging, the NMR tool is calibrated in a water-filled test jig or in a downhole calibration mode that measures the NMR response of the borehole fluid (assumed to be fresh or saltwater of known HI) to verify that the tool gain is set correctly; the calibration converts the raw NMR magnetization amplitude to porosity units by dividing by the calibration amplitude per unit porosity, under the assumption that all pore fluids have HI = 1 (water); if the actual pore fluids have HI less than 1, the NMR will read apparent porosity below the true porosity by the factor (1 - volume fraction of non-water fluid times (1 - HI of that fluid)), requiring a hydrogen index correction to the NMR porosity before it is used in reservoir evaluation; the hydrogen index correction for gas or oil requires independent knowledge of the HI (from fluid samples, PVT analysis, or formation temperature and pressure) and the saturation (from resistivity or NMR-derived saturation), making HI-corrected NMR porosity most accurate when all of these inputs are well constrained from calibration to core and fluid data.
  • Source rock hydrogen index in pyrolysis geochemistry (Rock-Eval analysis) uses a completely different definition of hydrogen index than NMR logging: in Rock-Eval pyrolysis, the hydrogen index (HI) is defined as the milligrams of hydrocarbons generated per gram of total organic carbon (TOC) during programmed heating to 550 degrees Celsius (S2 yield divided by TOC, multiplied by 100 to give mg HC / g TOC), quantifying the hydrogen richness of the organic matter and therefore its potential to generate oil and gas during natural burial and maturation; a source rock with HI above 600 mg HC / g TOC contains predominantly oil-prone type I or type II kerogen, while HI below 200 mg HC / g TOC indicates gas-prone type III kerogen derived from terrestrial plant material; the Rock-Eval HI decreases with increasing thermal maturity as the organic hydrogen is progressively converted to hydrocarbons and expelled, making HI versus Tmax crossplots (the van Krevelen diagram in geochemical form) the standard method for classifying kerogen type and maturity in source rock evaluation; this geochemical HI has no mathematical relationship to the NMR HI but is unfortunately also called "hydrogen index" in the petroleum literature, requiring context to determine which definition is intended.

Fast Facts

The NMR hydrogen index concept was formalized in the context of oilwell logging by Schlumberger's nuclear magnetic resonance research group in the late 1980s and early 1990s, as the development of the CMR (Combinable Magnetic Resonance) tool for formation evaluation required a systematic treatment of how fluids with different hydrogen densities affect the NMR measurement. The Rock-Eval pyrolysis hydrogen index was introduced by Espitalie, Laporte, and Madec at IFP (Institut Francais du Petrole) in 1977, and the unfortunate sharing of the term "hydrogen index" between two completely different measurements in NMR logging and organic geochemistry has caused occasional confusion in well reports and petrophysical evaluations that combine both types of data.

What Is Hydrogen Index?

Hydrogen index (HI) in NMR logging is the ratio of hydrogen atom density in a fluid to the hydrogen atom density of pure water, determining how much NMR signal that fluid contributes per unit volume relative to water. Pure water has HI = 1.0; liquid crude oils have HI of 0.85 to 0.98; natural gas has HI of 0.01 to 0.4 depending on pressure and temperature. Fluids with HI below 1 cause NMR apparent porosity to underestimate true porosity, a significant effect in gas reservoirs where gas HI may be as low as 0.02 at shallow conditions. NMR porosity must be corrected for hydrogen index effects when gas or light hydrocarbon is present in the pore space. In source rock geochemistry, hydrogen index has an unrelated meaning: milligrams of hydrocarbons per gram of TOC from Rock-Eval pyrolysis.

Hydrogen index in NMR logging is also called the HI factor or NMR hydrogen index. Related terms include nuclear magnetic resonance logging (NMR, the wireline or LWD tool that measures the magnetic relaxation of hydrogen nuclei in formation fluids after being polarized by a static magnetic field and tipped by a radio-frequency pulse, providing porosity (from the total NMR signal amplitude, scaled by the hydrogen index of the fluid), pore size distribution (from the T2 relaxation time spectrum), permeability, and fluid typing information that is calibrated to the hydrogen index of the specific fluids present), T2 relaxation (the transverse magnetic relaxation time of hydrogen nuclei in an NMR measurement, determined by the pore surface-to-volume ratio in water-saturated rock (smaller pores relax faster), by the bulk fluid relaxation time (heavy oils have short T2, free water has long T2), and by diffusion in a magnetic field gradient (gas and light oil diffuse faster and have shorter apparent T2 in gradient fields), with the T2 distribution providing the pore size distribution and fluid typing information used in NMR formation evaluation), gas effect (the underestimation of NMR porosity in gas-bearing formations caused by the low hydrogen index of gas relative to water, combined with the fast diffusion-enhanced relaxation of gas in the NMR tool's magnetic field gradient that causes gas to appear at short T2 times indistinguishable from bound water, creating both a porosity underestimate and a fluid typing ambiguity that requires the hydrogen index correction and multi-echo NMR acquisition to resolve), Rock-Eval pyrolysis (a geochemical analysis technique that heats a crushed rock sample at programmed temperatures to measure the yield of free hydrocarbons (S1), generated hydrocarbons (S2), and CO2 (S3), from which the source rock hydrogen index (S2/TOC x 100 in mg HC/g TOC), oxygen index, and maturity parameter Tmax are computed for kerogen type classification and source rock quality assessment), and NMR porosity (the total porosity of a formation measured by nuclear magnetic resonance logging, computed from the total amplitude of the NMR T2 decay signal after calibration to pure water, which underestimates true porosity in formations containing fluids with hydrogen index less than 1 (gas, bitumen, or heavy oil with very short T2) and must be corrected for the fluid hydrogen index before use in pore volume and saturation calculations).