Advances in high-resolution Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR MS) enable molecular-level characterization of ultracomplex asphaltene samples. Such analyses most often reveal compounds that are highly aromatic but alkyl-deficient in nature and, thus, support the classical “island” model of asphaltene architecture. However, recent works that combine chromatographic separations with mass spectrometry for the analysis of crude oils have shown that differences in ionization may greatly affect the analysis of complex mixtures (known as the matrix effect). Simply, compounds that ionize with greater efficiency are preferentially observed and mask the detection of poorly ionized compounds. Asphaltenes are not immune to this phenomenon. In the first of this series (10.1021/acs.energyfuels.7b02873), it was demonstrated that asphaltenes generated by different precipitants showed greatly varied monomer ion yields (ionization efficiencies). This work focuses on the development of an extrography fractionation method that selectively targets the removal of asphaltene species that exhibit high monomer ion yields and, thus, restrict mass spectral characterization of less efficiently ionized species. Silica gel was used as the stationary phase, and a unique solvent series separated asphaltenes based on their interaction with the silica surface, which was later determined to depend heavily upon the structure as well as monomer ion yield. The first two solvents (acetone and acetonitrile) isolated compounds that most efficiently produce monomeric asphaltene ions and, thus, cause bias in mass spectrometric analyses of whole asphaltenes. A solvent polarity gradient was then used, with n-heptane, toluene, tetrahydrofuran, and methanol, to separate remnant asphaltene compounds on the basis of polarity and structure. Our results demonstrate that mass spectrometry of whole asphaltenes does not reveal the complete molecular composition but rather preferentially exposes highly aromatic, alkyl-deficient, island-type structures. Early eluting fractions are shown to resemble the composition of the whole asphaltene and are enriched in island structures, whereas the analysis of later-eluting fractions reveals archipelago structural motifs as well as species with atypical asphaltene molecular compositions. We also demonstrate that, as molecular weight increases, the asphaltenes exhibit increased contributions of archipelago structural motifs. Higher mass ions (m/z > 550), even from asphaltene fractions enriched in island structures, exhibit fragmentation pathways that originate from archipelago structures. Thus, positive-ion atmospheric pressure photoionization (APPI) FT-ICR MS provides molecular-level data that suggest that the island model is not the dominant structure of asphaltenes. It coexists with abundant archipelago structures, and the ratios of each are sample-dependent.

For decades, discussion of asphaltene structure focused primarily on molecular weight. Now that it is widely accepted that asphaltene monomers are between ∼250 and 1200 g/mol, disagreement has turned to asphaltene architecture. The classic island model depicts asphaltenes as single core aromatic molecules with peripheral alkyl side chains, whereas the less widely accepted archipelago model, includes multiple aromatic cores that are alkyl-bridged with multiple polar functionalities. Here, we analyze asphaltene samples by positive-ion atmospheric pressure photoionization Fourier transform ion cyclotron resonance mass spectrometry and perform infrared multiphoton dissociation to identify their aromatic core structures to shed light on the abundance of island and archipelago structural motifs. Our results indicate that island and archipelago motifs coexist in petroleum asphaltenes, and unlike readily accessible island motifs, asphaltene purification is required to detect and characterize archipelago species by mass spectrometry. Moreover, we demonstrate that mass spectrometry analysis of asphaltenic samples is biased toward the preferential ionization/detection of island structural motifs and that this bias explains the overwhelming mass spectral support of the island model. We demonstrate that the asphaltene structure is a continuum of island and archipelago motifs and hypothesize that the dominant structure (island or archipelago) depends upon the asphaltene sample.

Asphaltene structure is one of the most controversial topics in petroleum chemistry. The controversy is centered on the organization of aromatic cores within asphaltene molecules (single aromatic core, island and multiple aromatic core, archipelago) and specifically the inconsistency between the island model and the composition of the products derived from asphaltene pyrolysis/thermal cracking. Such products are consistent with the coexistence of island and archipelago asphaltene structural motifs. However, the archipelago model continues to lack the widespread acceptance of the petroleum community, in part due to mass spectrometry results in support of the island model. In the first and second part of this series, we demonstrated that the disproportionally high atmospheric pressure photoionization (APPI) ionization efficiency (monomer ion yield) of island species is due to weak nanoaggregation of large aromatic cores which do not extensively aggregate in toluene, whereas more archipelago-dominant fractions were shown to have low monomer ion yield due to a greater propensity for aggregation. The discrepancy leads to bias toward the selective ionization of island compounds and thus the erroneous mass spectrometry support of the predominance of the island structural model. A separation method based on aggregation trends and therefore the efficiency of monomeric ion production is critical to access archipelago structures. In the work presented herein, we demonstrate that dominance of island or archipelago structural motif is sample dependent. We present the positive-ion APPI Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR MS) characterization of asphaltenes and asphaltene extrography fractions derived from Wyoming Deposit (island dominant) and Athabasca Bitumen (archipelago dominant) C7 asphaltenes. Wyoming Deposit asphaltenes resemble the “classical” island-type asphaltene structure: they exhibit a high concentration of highly aromatic/alkyl-deficient species with a compositional space close to the polycyclic aromatic hydrocarbon (PAH) limit. Fragmentation results from infrared multiphoton dissociation (IRMPD) confirm that island is the dominant structural motif in Wyoming Deposit C7 asphaltenes; the predominant fragmentation pathway for all extrography fractions consists of loss of CH2 units (or dealkylation), without significant loss of aromaticity. Conversely, Athabasca Bitumen C7 asphaltenes exhibit an “atypical” molecular composition. More than 40 wt % of the sample is extracted in the latest extrography fractions, which are composed of difficult-to-ionize species, a fraction of which exhibit atypically low double bond equivalent (DBE = 5–12) and extended homologous series with carbon numbers up to 60. The fragmentation behavior of all Athabasca Bitumen-derived fractions demonstrates a predominant contribution of archipelago motifs. Our results suggest that the Yen-Mullins molecular definition of asphaltenes cannot be used to describe all asphaltene samples. Island and archipelago structura...

Naphthenic (NAP) acids have been previously characterized by negative electrospray ionization (ESI) Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR MS), both directly and from acid extracts. Here, we show that collective characterization of NAP acids with negative-ion ESI, both with and without prior extraction, results in a loss of signal from high mass acids because of ion suppression by low mass acids. We have developed an extraction to simultaneously isolate and fractionate NAP acids into distinct molecular weight ranges, thereby grouping acids into fractions with similar ionization efficiency. A NAP acid fraction was isolated (all acids isolated collectively in one fraction) and compared to NAP acid isolation by molecular weight. Characterization of acid molecular weight fractions extended the observed upper mass limit from ∼850 Da for the collectively isolated acids to ∼1450 Da for the acids isolated by molecular weight range, thereby approximately doubling the observed mass range for NAP acids. Plots of double bond equivalents (DBE = number of rings plus double bonds to carbon) versus carbon number span both higher carbon number and higher DBE values than are accessed by collective acid characterization. The high mass resolving power and mass accuracy of FT-ICR MS are essential for identification and resolution of acidic species across a wide mass range.