Abstract

The description of heptanes and heavier components (C7+) in reservoir fluids can be important for equation of state (EOS) predictions of phase and volumetric behavior. This paper describes a procedure for C7+ characterization of heavy oil based on crude assay data which are typically measured for refining and marketing applications.

C7+ characterization is defined as (1) modeling the molar distribution that quantifies molar (mass) amounts and molecular weights of discrete plus fractions, (2) estimating specific gravity and normal boiling point of plus fractions, and (3) estimating EOS and viscosity-model parameters Tc, pc, ω, Vc, s (volume shift) and binary interaction parameters (BIPs) kij of plus fractions.

From crude assay data, we use the mass fractions and overall sample molecular weight to determine the parameters for gamma molar distribution parameters – shape (α), lower bound (η) and average plus molecular weight (Mo) in the gamma distribution model. The molar distribution model, together with measured assay-cut specific gravity and boiling point data, are used to determine parameters in the Soreide correlation describing specific gravity-molecular weight relationship. The inter-correlation of assay data is also affected by correlation used between molecular weight, boiling point, and specific gravity. We use a parameter, fTwu, where fTwu=1 honors the Twu correlation exactly, while fTwu= 0 honors a pure-paraffinic correlation between molecular weight and boiling point (independent of specific gravity).

Viscosities of each fraction are correlated against measured data and/or standard-pressure liquid viscosity estimates from the Orrick-Erbar correlation. We use the Lorentz-Bray-Clark (LBC) correlation, where fraction critical volumes are adjusted to give consistency between the LBC estimates for each fraction individually. This approach to determining the critical volumes has proven useful in making the LBC more predictive for overall oil mixture viscosities at reservoir conditions.

You can access this article if you purchase or spend a download.