METHODS FOR CLASSIFYING PETROLEUM COKE

Abstract

A method of determining the identity of a petroleum coke sample including obtaining a nuclear magnetic resonance (NMR) measurement of the sample, determining a relaxation decay value of a fluid in the sample from the NMR measurement, comparing the relaxation decay value to relaxation decay values of known petroleum coke materials in a reference group to determine whether the petroleum coke is one of the known materials.

Claims

1. A method of determining the identity of a petroleum coke sample comprising: obtaining a nuclear magnetic resonance (NMR) measurement of the sample, determining a relaxation decay value of a fluid in the sample from the NMR measurement, comparing the relaxation decay value to relaxation decay values of known petroleum coke materials in a reference group to determine whether the petroleum coke is one of the known materials.

2. The method of claim 1, wherein the comparing step includes searching a list of known petroleum coke materials and corresponding relaxation decay values.

3. The method of claim 1, wherein the known petroleum coke materials are selected from the group consisting of sponge-type coke, shot coke, transition coke, dense coke and needle coke.

4. The method of claim 1, further comprising outputting the result.

5. The method of claim 1, wherein the result is one of an identification of the substance as a known material in the reference group or determination that the substance is not one of the known materials in the reference group.

6. The method of claim 1, wherein the relaxation decay value is T.sub.1 relaxation time or T.sub.2.relaxation time.

7. The method of claim 6, wherein the T.sub.1 relaxation time is the T.sub.1 log mean and the T.sub.2.relaxation time is the T.sub.2 log mean.

8. The method of claim 1, further comprising determining a porosity value of the sample from the NMR measurement and comparing the porosity value to porosity values of known petroleum coke materials.

9. The method of claim 8, wherein the known petroleum coke materials are selected from the group consisting of sponge-type coke, shot coke, transition coke, dense coke and needle coke.

10. The method of claim 1, wherein the sample comprises at least two different known petroleum coke materials and the step of determining a relaxation decay value comprises determining at least two relaxation decay values of a fluid in the sample from the NMR measurement and comparing the relaxation decay values to relaxation decay values of known petroleum coke materials in a reference group to determine whether the petroleum coke sample comprises one or more of the known materials.

11. The method of claim 10, further comprising determining weightings for the relaxation decay values to determine the percentage of types of known petroleum coke materials in the sample.

12. The method of claim 1, wherein the fluid is an NMR active fluid.

13. The method of claim 12, further comprising prior to the step of obtaining the low field nuclear magnetic resonance (NMR) measurement of the sample, saturating the sample with the NMR active fluid.

14. The method of claim 1, wherein the NMR measurement is a low field NMR measurement.

15. The method of claim 1, wherein the step of comparing the relaxation decay value includes fitting to a mathematical equation using predetermined T.sub.2 distribution of each COKE type to be classified.

16. The method of claim 15, wherein the mathematical equation is
ERROR=measured T.sub.2−(Type A T.sub.2*weight A+Type B T.sub.2*weight B+ . . . )

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0013] FIG. 1 illustrates a conventional NMR pulse diagram of a Carr-Purcell-Meilboom-Gill (CPMG) sequence usable in methods of the present disclosure;

[0014] FIG. 2 illustrates a conventional NMR pulse diagram of an Inversion Recovery (IR) sequence usable in methods of the present disclosure;

[0015] FIG. 3 is a flowchart depicting an example method for classifying coke;

[0016] FIG. 4a is a graph illustrating the T.sub.2 relaxation distribution for a first Type A coke sample;

[0017] FIG. 4b is a graph illustrating the T.sub.2 relaxation distribution for a second Type A coke sample;

[0018] FIG. 4c is a graph illustrating the T.sub.2 relaxation distribution for a first Type C coke sample;

[0019] FIG. 4d is a graph illustrating the T.sub.2 relaxation distribution for a second Type C coke sample;

[0020] FIG. 4e is a graph illustrating the T.sub.2 relaxation distribution for a first Type D coke sample;

[0021] FIG. 4f is a graph illustrating the T.sub.2 relaxation distribution for a second Type D coke sample; and

[0022] FIG. 5 is a plot of porosity versus the T.sub.2 log mean of coke samples.

DETAILED DESCRIPTION

[0023] While this invention is susceptible of embodiment in many different forms, there is shown in the drawings, and will herein be described hereinafter in detail, some specific embodiments of the instant invention. It should be understood, however, that the present disclosure is to be considered an exemplification of the principles of the invention and is not intended to limit the invention to the specific embodiments or algorithms so described.

[0024] The present invention in one embodiment relates to a method of using NMR to investigate different internal structures (pore and interstitial regions) of different types of coke. The internal structures of a coke sample are filled with water (or other NMR active fluid) and the relaxation decay of the fluid is measured using NMR. The relaxation decay is directly affected by the internal structure of the coke sample.

[0025] The relaxation parameter used can be a T.sub.1 or T.sub.2 relaxation decay. The relaxation parameters T.sub.1 and T.sub.2 can be measured using a conventional NMR measurement pulse sequence. In certain embodiments, parameters of the sequence are selected to measure the range of T.sub.1 or T.sub.2 expected. The raw data is analyzed by fitting a sum of exponential as described by equation (1):

[00001]$\begin{array}{cc}S\left(t\right)=\underset{i=1}{\overset{N}{.Math.}}.Math.A\left(i\right).Math.e^{-\frac{t}{T_2\left(i\right)}}& \left(1\right)\end{array}$

where T.sub.2.sup.(i) is the list of T.sub.2 values that are chosen to be fitted to; t is the time (for example, increments of 2*TAU can be used); A(i) is the amplitude or amount of signal at the T.sub.2.sup.(i) value; i is the current T.sub.2 value to fit to (for example, ˜100 T.sub.2 values log spaced between 0.01 ms and 10 sec can be used); and S(t) is the signal as a function of time.

[0026] This equation can be solved by minimizing equation (2):

[00002]$\begin{array}{cc}\mathrm{Error}.Math.=\sqrt{\underset{t=\mathrm{TE}}{\overset{t=j*\mathrm{TE}}{.Math.}}.Math.{\left(m\left(t\right)-\underset{i=1}{\overset{N}{.Math.}}.Math.A\left(i\right).Math.e^{-\frac{t}{T_2\left(i\right)}}\right)}^2}+\alpha .Math.\sqrt{\underset{i=1}{\stackrel{N}{.Math.}}.Math.{A\left(i\right)}^2}& \left(2\right)\end{array}$

where m(t) is the measured signal as a function of time; J is the echo number and the balance of the variables are as per equation (1).

[0027] The alpha or smoothing factor, α, can be determine by a variety of methods (Butler 1981).

[0028] In one embodiment, a conventional inversion recovery sequence, such as depicted in FIG. 1, is used to measure the T.sub.1 relaxation decay. The inversion recovery sequence is repeated at different recovery times (TI) and the detected signal recovers with the relaxation constant of T.sub.1. A conventional Carr-Purcell-Meilboom-Gill (CPMG) NMR pulse sequence, such as depicted in FIG. 2, can be used to measure T.sub.2. In FIG. 2, the T.sub.2 relaxation decay parameter is the decay of the acquired echoes (marked with x's) over time.

Examples

[0029] Two samples (I and II) of three different types of coke (A, C and D) were tested using the exemplary method 100 illustrated in FIG. 3. In step 1, a coke sample is dried to remove fluid present in the sample. Step 1 is optional. In step 2, the coke sample is saturated with water by placing the sample under vacuum, introducing water to wet pores and interstitial regions of the sample and then releasing the vacuum. The vacuum created inside the sample facilitates the water to enter the pores and interstitial regions of the sample. It will be understood by those skilled in the art that other suitable sample wetting techniques can be used. In step 3, the T.sub.2 relaxation decay of the sample is measured using a conventional NMR apparatus (not shown) programmed to apply a CPMG pulse sequence to samples. In step 4, the T.sub.2 relaxation decay measurement is analyzed to obtain a T.sub.2 relaxation decay parameter value. The T.sub.2 relaxation decay parameter value is compared to known classification parameter values for coke types in a reference group of coke types to determine whether a known coke type is present in the sample tested. At this point, for example, results of the identification can be outputted. Step 4 can be implemented as computer-executable instructions in a computing environment (not shown). This could be as simple as classification based on the average to T.sub.2 value. Or more complicated such as a fit to a mathematical equation using predetermined T.sub.2 distribution of each COKE type. For example, the mathematical equation could be:

ERROR=measured T.sub.2−(Type A T.sub.2*weight A+Type B T.sub.2*weight B+ . . . )

[0030] The “ . . . ” in the above equation is included to indicate that this type * weighting can be extended to the various types of coke which are to be classified. The computing environment can be integrated with the NMR apparatus of step 3 or can be separate.

[0031] FIG. 4 are graphs illustrating T.sub.2 relaxation distributions for two samples of three different types of coke tested in this example. Differences in the porosity and the average T.sub.2 value were observed for the different types of Coke samples. Table 1 lists the peak T.sub.2 relaxation time, the average T.sub.2 relaxation time (log mean), the maximum T.sub.2 value, the NMR porosity, and the NMR porosity of the dry samples. The peak time is the T.sub.2 time with the maximum intensity. The log mean is the T.sub.2 time that divides the distribution into two equal areas (like an average). The NMR porosity is the total area under the distribution which is the total porosity of the water occupied pores (which at 100% saturation is all of them).

TABLE-US-00001 TABLE 1 T.sub.2 Dry NMR T.sub.2 Log T.sub.2 @ NMR Porosity Sample Peak Mean 99% Porosity Contribution Number (ms) (ms) (ms) (p.u.) (p.u) Type A I 2818 2424 5623 15.7 0.062 Type A II 2818 2551 5623 14.9 0.077 Type C I 1000 279 2512 8.3 0.48 Type C II 1258 362 3162 9.6 0.51 Type D I 1995 1184 5012 8.0 0.095 Type D II 1995 1149 5012 6.0 0.14

[0032] FIG. 5 is a plot comparison and differentiation of petroleum coke samples based on porosity versus T.sub.2 log mean. Groupings for each coke type identified are shown.

[0033] In this example, three different types of coke were found to be present in the samples: Type A which was primarily sponge Coke, Type C which was primarily shot Coke and Type D which was primarily a denser Coke that is not that common.

[0034] In one embodiment, coke samples are classified by using a “cut-off” method which uses the T.sub.2 log mean value. A sample is classified based on the range of T.sub.2 log mean values that are measured. In certain embodiments, the T.sub.2 log mean value range for each type of coke can be predetermined. For example, if the measured T.sub.2 log mean is between 800 ms and 1500 ms, it is indicative of Type C coke.

[0035] In another embodiment, the above described “cut-off” method could use a T.sub.1 log mean value instead of the T.sub.2 log mean value.

[0036] In another embodiment, a second NMR parameter may be used in addition to a relaxation decay parameter. In certain embodiments, the second NMR parameter may be porosity. The use of T.sub.2 log mean and porosity values allows for “zones” on a plot of porosity versus T.sub.2 log mean to define the classification.

[0037] In another embodiment, the T.sub.2 response of known classifications is measured and the sum each T.sub.2 distribution with weightings is used to obtain the measured result on the unknown coke samples. This allows for a percentage of each classification (i.e. 50% Type A, 20% Type C and 30% type D).

[0038] In another example, a series petroleum coke samples with masses ranging from approximately 15 to 25 grams were vacuum saturated with 2% w/w KCl in water solution for a minimum of about 2 hours at ambient temperature. The samples were drained of free water and blotted with a paper towel to remove excess moisture. A given sample was loaded into a 40 mm diameter NMR tube and NMR measurements were performed at ambient temperature using an Oxford Instruments Geo Spec 2 NMR spectrometer operating at 2.36 MHz. The NMR measurements consisting of T.sub.2 and T.sub.1-T.sub.2 spectra (bulk measurement) were performed with scan times ranging from 5 minutes to 5 hours depending upon the sample coke morphology and type of NMR measurement. Three types of cokes were tested: sponge, shot, and a dense coke produced from a Heavy Canadian vacuum residue. It was discovered that the log mean T.sub.2 parameter could easily distinguish between the three types of coke morphologies, with sponge coke having a T.sub.2 log mean (LM) of approximately 2500 ms; shot coke T.sub.2 LM of about 320 ms; and dense coke having a T.sub.2 of around 1150 ms.