Calculation of average molecular descriptions of heavy petroleum hydrocarbons by combined analysis

by quantitative 13C and DEPT-45 NMR experiments John C. Edwards*, A. Ballard Andrews†

* Process NMR Associates, LLC, 87A Sand Pit Rd, Danbury, CT 06810 USA (*e-mail: john@process-nmr.com) † Schlumberger-Doll Research, Cambridge, MA, USA

OverviewMuch debate has centered around the validity and accuracy of NMR measurements toaccurately describe the sample chemistry of heavy petroleum materials. Of particularissue has been the calculated size of aromatic ring systems that in general seem to beunderestimated in size by NMR methods. This underestimation is principally caused byvariance in chemical shift ranges used by researchers to define the aromatic carbon typesobserved in the 13C NMR spectrum, in particular the bridgehead aromatic carbons thatcan be shown to overlap strongly with the protonated aromatic carbons. The ability todiscern between bridgehead aromatic carbons and protonated carbons in the 108-129.5ppm region of the spectrum is key in the derivation of molecular parameters that describethe “molecular average” present in the sample. Utilizing methodologies developed byPugmire and Solum for the solid-state 13C NMR analysis of coals and other carbonaceoussolids [1], we have developed a new liquid-state 13C NMR method that allows the relativequantification of overlapping protonated and bridgehead aromatic carbon signals to bedetermined [2].

The NMR experiments involve the combined analysis of both quantitative 13C single pulseexcitation which observes all carbons quantitatively, and a DEPT45 polarization transferwhich observes only the protonated carbons in the sample. Though the DEPT45 results arenot quantitative across all carbon types (CH, CH2, and CH3) due to polarization transferdifferences, the technique is well enough understood that simple multiplication factorsallow the relative intensities of the different carbons to be determined. The average ringsystem sizes derived from these NMR experiments tend to be several ring systems largerthan has been calculated in previous studies [3]. In heavy petroleum asphaltenes theaverage aromatic ring system is 5-7 rings in size which is in agreement with FTICR-MS andfluorescence measurements [4-7], rather than the 3-4 rings previously reported [3]. Theresults obtained by liquid-state NMR are compared to the carbon-type NMR andmolecular parameters derived from the solid-state 13C analysis of the same samples. Achemometric approach to the derivation of the parameters has been developed so thatautomated output of the analysis results can be routinely provided by technician levelpersonnel. Finally, a quantitative 13C NMR method is described that utilizes internalstandards to quantify the amount of carbon in the sample observed in the NMRexperiment. This 13C quantitation yields information on the degree of aggregation in theasphaltene components.

Experimental13C NMR spectra were obtained on a Varian Mercury 300-MVX at a resonance frequencyof 75.43 MHz. Single Pulse Excitation (SPE) experiments were performed with inversegated 1H decoupling and with 0.025M CrAcAc present in the CDCl3 solvent to facilitate theaccumulation of quantitative NMR data. Samples typically contain around 80-100 mg ofsample dissolved in 1 gram of CDCl3-CrAcAc. DEPT45 experiments were obtained on thesame sample.

DEPT Experiment

13C NMR Analysis Examples

13C qNMR

13C DEPT-45 NMR

Comparison ofQuantitative NMR withDEPT-45 NMR

Integration of respective aliphatic carbon zones for normalization calculation

Carbon

Type

Multiplication

Factor

CH 1.414

CH2 1.000

CH3 0.943

Carbon Chemical and Molecular Parameters Relationship between mole fraction of bridgeheadcarbons χb and aromatic cluster size (C).From Solum, et al. Energy& Fuels, 3, 187, 1989.

Aggregation estimation by qNMR - internal standard technique. PEG acts as a carbon content standard allowing observed carbon to be calculated as well as acting as the DEPT normalization standard.

1) Pugmire et al., Energy & Fuels, 3, 187, 1989. 2) Andrews, Edwards, et al., Energy & Fuels, 2011, 25 (7), pp 3068–30763) Sheremata et al., Energy& Fuels, 6, 414, 2004. 4) Badre et al., Fuel, 85, 1, 2006. 5) Sharmar et al, Energy & Fuels 16, 490, 2002. 6)Groenzin and Mullins, “Asphaltenes, Heavy Oils and Petroleomics, Ch. 2.7) Rodgers and Marshall, “Asphaltenes, Heavy Oils and Petroleomics, Ch. 3.

For more details see: " Comparison of Coal-Derived and Petroleum Asphaltenes by 13C Nuclear Magnetic Resonance, DEPT, and XRS", A. Ballard Andrews, John C. Edwards, Andrew E. Pomerantz, Oliver C. Mullins, Dennis Nordlund, and Koyo Norinaga, Energy Fuels, 2011, 25 (7), pp 3068–3076

13C SPE NMR

S