Molecular Ion: An Extended, Fully Collective, and More Contemporary Definition

Undeniably, “molecular ion” is one of the most common, important, and widely used terms in mass spectrometry (MS), and its proper definition is therefore essential for the best description of MS data. But unfortunately, as we have recently argued,1 the current “IUPAC Definitions of Terms”2 has poorly defined this pivotal term, that is, in an obsolete, semantically misleading, and segregating fashion, and not as collective as it should be. This definition states that a “molecular ion” is: An ion formed by the removal of one or more electrons from a molecule to form a positive ion or the addition of one or more electrons to a molecule to form a negative ion. The semantically misleading nature of this definition is perceived in the adjective “molecular,” which inherently carries a strong, widely recognized, and intuitively meaning. “Molecular” is readily and universally comprehended by most, if not all English-speaking scientists, to mean “of molecules, relating to molecules, consisting of molecules, or produced by molecules.” Therefore, the original definition of “molecular ion” adopted by IUPAC has imposed a “semantically unexpected” segregation onto the term, making it “biased” toward a specific subclass of ions—radical cations or radical anions—made by a specific method among the many by which molecular ions can currently be formed. Specifically, the definition is “biased towards the addition or removal of electrons.” This bias is further reinforced by the current definition when it continues by stating—to remove any doubt—that: [The term] molecular ion is reserved for the intact ionized molecule with no component added or removed other than electrons. The semantically misleading nature of the IUPAC definition therefore arises, we argue, from the classification of a diverse class of ions as “molecular,” whereas arbitrarily reserving the term exclusively for a single subclass of molecular ions. It is so intuitively semantic that “molecular ions”—by the very nature of the adjective “molecular”—should encompass all subclasses of ions made from any intact molecule M, regardless of the method or the added or removed ionic species, that consequently many new or even skilled MS users have often been inclined to classify as molecular not exclusively the subclasses of M+. or M−. ions, but also other classes of ions that clearly “appear to be” and are, in fact, “molecular,” such as [M + nH]n+, [M − nH]n-, [M + nNa]n+, M+ or M− as prompted by the M in their formula. We have therefore concluded that a new collective definition for “molecular ion” was needed, and that we should also seek in this new definition to liberate the term from its “electron feudalism” by aligning its semantic meaning with its actual description: “a comprehensive collection of molecular ions.” In a letter to this journal,1 we have therefore proposed what we believed to denote a more collective, more contemporary, and more semantically sound definition for this pivotal term in mass spectrometry. Our definition, first targeted to both electron and proton addition (or removal), had the merit, we argue, of unifying other ambiguous terms that are quite often used to replace the term “molecular ion,” such as the even more ambiguous terms of “quasimolecular ion” or “pseudomolecular ion,” or even bizarre terms such as “protonated molecular ion.” We have proposed that “molecular ion” should be described as: An ion such as M+., M−., [M + nH]n+ or [M − nH]n− formed, without further fragmentation, from an intact molecule M by the addition or removal of electrons or protons or a permanently charged “molecular” cation M+ or anion M− which is directly transferred as such from its salts to the gas phase environment of a mass spectrometer. At that time, we were confident that this updated definition would gain broad acceptance. But, prompted by the ensuing discussion, we have realized that our renovated definition still falls short of being as inclusive as needed. We initially expanded it to the addition or removal of protons as well as to even-electron permanently charged ions, but we now recognize that the optimal solution would be to liberate the definition of “molecular ion” from any restrictions at all. We will present and discuss this “fully expanded” definition at the conclusion of this letter. But then, soon after our proposal was published, two responses appeared in this journal. In the first response, we were pleased to learn that Holman and Bristow3 agreed with part of our proposal, that is, with the application of the term “molecular ion” to even-electron permanently charged ions M+ or M−. But subsequently, we were surprised to know that they argued against its use for all the other subclasses of molecular ions (and subsequently in favor of the segregating current definition), noting that the application of our updated definition to ions such as protonated and cationized molecules would “risk ambiguity and confusion where clarity presently exists.” We were surprised with these comments on “clarity” because we were confident that we had clearly demonstrated that, in the current IUPAC definition of a molecular ion, “clarity was what was most lacking.” Our proposal aimed therefore to eliminate the lack of clarity, that is, the ambiguity and confusion caused by this segregating and semantically confusing definition for “molecular ion,” reintegrating into the definition a collection of arbitrarily “expelled molecular ions.” Clearly, we argue, our definition causes no ambiguity or confusion; rather, it removes them. In this letter, we aim to contest the arguments of Holman and Bristow and more persistently defend our proposal. To argue that our proposal introduces ambiguity and confusion for “the other cases,” Holman and Bristow presented the ESI(+)-MS of sulfaguanidine (M) as an example. In this spectrum, they highlighted the detection of M as its protonated [M + H]+, sodiated [M + Na]+, and potassiated [M + K]+ forms. According to them, our proposed definition could induce us to collectively label these three species simply as “molecular ions,” a description that would “lack discriminative information.” Holman and Bristow further contended that our collective definition suffers from another weakness: “verbosity.” That is, they suggested that the collective definition, to ensure clarity, would require lengthier descriptions, such as “molecular ions of sulfaguanidine were observed as protonated, sodiated, and potassiated molecules.” In contrast, using the already available current IUPAC “subterms” would allow, they argued, for a more concise description of the detected ions: “protonated, sodiated, and potassiated molecules of sulfaguanidine were observed,” a description that would “conveys the same information without unnecessary elaboration.” They also criticized our proposal for another weakness: “vagueness”; noting that simply reporting “three molecular ions were detected” would fail to provide “clear, accurate, and specific data” to the reader. We respectfully disagree! Firstly, we argue that “length of description” should never be used as a criterion to evaluate the quality of a definition, especially when the term being described is as central to a field as “molecular ion.” In science, to prevent confusion and minimize their occurrences, terms must be as semantically accurate and inclusive as possible, rather than focusing on word counting. Definitions should precisely reflect the semantics of the term, without artificially excluding any members from the group, therefore avoiding confusion particularly among non-experts. This care would eliminate the need for a user to consult the definition just to discover that it contradicts the clear semantic of the term. It is within this strong semantic dichotomy that “the risk of confusion and ambiguity resides.” Secondly, we disagree because the choice to describe a collection of detected ions using either the collective term or specific subterms is entirely on the writer, based on the need for more or less details. This flexibility in using either the collective term or subterms is not a weakness but rather an advantage of our proposal. For example, in the spectrum Holman and Bristow have offered as an illustration, there is indeed a “risk of ambiguity and confusion,” but this risk resides not in the use of our definition but otherwise in the current segregating and semantically misleading IUPAC definition of “molecular ion.” Writers might erroneously describe the spectrum as containing “three types of molecular ions,” when none of the detected species [M + H]+, [M + Na]+, and [M + K]+ would qualify as such. Worse still, in desperation, writers may misclassify these three “molecular(M) ions” as either “pseudomolecular,” “quasimolecular” or “protonated (cationized) molecular ions,” terms that are currently deprecated by IUPAC, but are still often incorrectly found in MS literature, even in textbooks. It is precisely this ambiguity, vagueness, verbosity, and confusion that our definition eliminates! To further illustrate our argument, we also like to offer some spectra as examples (Figure 1).4 Let us us consider first the EI(+)-MS of the C60 fullerene (Figure 1a). When describing it, regardless of whether using the current IUPAC definition or our proposed definition, we could state that “three types of molecular ions were detected,” if we would only consider the monoisotopic ions while ignoring their 13C isotopologues, namely, 12C60+. of m/z 720, 12C602+ of m/z 360, and 12C603+. of m/z 240. Using either definition, we could therefore choose to concisely and collectively state that “three molecular ions were detected.” Alternatively, if we judge that more details is necessary, we could specify these ions as “the singly charged molecular ion 12C60+. of m/z 720, the double charged 12C602+ of m/z 360, and the triply charged 12C603+. of m/z 240, and their respective isotopologues.” To be even more precise, we could collectively state that nine molecular ions were detected, describing them also in terms of their isotopologues, such as 12C60+. of m/z 720, 12C5913C1+. of m/z 721, 12C5813C2+. of m/z 722, and 12C5713C3+. of m/z 723, and so on. It is precisely this flexibility that is provided by the availability of both collective and specific, well-defined subterms, and it is entirely up to the writer to choose when to use them. Figure 1b shows the simulated EI(+)-MS of perbrominated C60, denoted as C60Br60, which serves as a notable example where the necessity for “clear, accurate, and specific reporting of data [that would] unambiguously define [all] the detected ions,” as postulated by Holmes and Bristow, is nearly impossible to fulfill or even inappropriate. Indeed, providing a detailed description of all species detected in a mass spectrum is desirable, but sometimes such description can be impractical or unnecessary. For Figure 1b, whether using the current IUPAC definition or our updated definition and employing “molecular ion” as the (partially) collective term it currently represents, we could succinctly state that the spectrum has detected a “wide range” of C60Br60+. molecular ions centered around m/z 5515. Alternatively, if we fell it would be helpful to be slightly more specific, we could formulate the ions as 12Cx13Cy79Brw81Brz+. (where x + y = 60 and w + z = 60). If hoping for “clear, accurate, and specific reporting of data” by “unambiguously defining all the detected ions,” then we would indeed face the daunting task of comprehensively describing all detected ions. We would need therefore to start by acknowledging that the monoisotopic molecular ion 12C6079Br60+. of m/z 5455 (composed of 12C with an atomic mass of 12.000 Da and 79Br with an atomic mass of 78.9183 Da) is unnoticed in the spectrum due to its too low abundance. Subsequently, we would be compelled to proceed by describing the numerous formulas for a myriad of isotopologue and isobaric molecular ions detected above a certain abundance threshold, let us say 1%. This exhausting task would involve creating a very long and meticulous list. To illustrate, this list would include the pair of the most abundant “isobaric” isotopologue molecular ions: 12C4013C2079Br4081Br20+. of m/z 5515.126 and 12C3813C2279Br4181Br39+. of m/z 5515.135 (among others), which, due to unit mass resolution, were detected as a single ion of m/z 5515. In Figure 1c, which represents an extreme but not improbable scenario for an ESI(+)-MS of a pure pharmaceutical molecule M (paeoniflorin, C23H28O11 of 432 Da) hypothetically assumed to be detected in various forms of its molecular ions, we encounter a similarly challenging situation. If we were to follow the advice of Holmes and Bristow to provide “clear, accurate, and specific reporting of data” to “unambiguously define all the detected ions,” and to describe each type of ion using its specific IUPAC subterm, we would be compelled to list all the 13 different molecular “M”-species seem in that spectrum. This commitment would result in a tedious list of terms, including the monoprotonated molecule [M + H]+, the doubly-protonated molecule [M + 2H]2+, the monosodiated molecule [M + Na]+, the doubly-charged sodiated molecule [M + 2Na]2+, the proton-bound dimer [M---H---M]+, and the doubly-charged monosodiated monopotassiated molecule [M + Na + K]2+, among others. To conserve space in the journal, we will refrain from providing such a list in full. By utilizing our extended definition of a molecular ion (below), we would have therefore the freedom to simply state that “the paeoniflorin molecule (M) was detected in Figure 1c in 13 subclasses of molecular ions, whose types are revealed by the inserted molecular formulas.” It would be as simple as that! If we deem additional details necessary, we could choose to describe the spectrum more elaborately, emphasizing key subclasses of the 13 molecular ions. Indeed, having fully comprehensive collective terms available—such as the one our extended definition for “molecular ion” provides—is beneficial, and describing group constituents by their specific subterms should only be done when additional details are required. Figure 1d provides another illustrative example, which now highlights the value of our extended definition for large biomolecules, showcasing numerous molecular ions—aligned with our proposed definition—that have been detected for a single protein molecule (M). Note that if we would be indeed required to “unambiguously define all the detected ions,” and given the current absence of a collective IUPAC term for multiprotonated molecules, we would be compelled to define each of the numerous [M + nH]n+ species detected. This would include terms such as “the 64-times charged 64-times protonated molecule [M + 64H]+64,” and so on. But, it is far more efficient to have a collective term, as our extended definition provides, enabling us to simply state, for instance, that “the protein has been detected by many [M + nH]n+ molecular ions with charges (+n) ranging from +41 to +78.” In the second response, Kermit Murray5 also advocated for retaining the IUPAC definition of the term “molecular ion,” “segregating” it exclusively for ionized molecules M+. and analogues. But, for other types of ions, Murray proposed adopting the IUPAC deprecated term “quasimolecular ion.” He argued that this term is better suited and already widely accepted in MS literature. In his own words: If a consensus can be established, quasimolecular ion could be used to describe [not ionized molecules but] protonated and deprotonated molecules, adduct ions, and ions formed from noncovalently bound associations of atoms and molecules. Murray also argued that: The term molecular ion would retain its current definition that describes ions formed by gain or loss of an electron by a molecule. Again, we respectfully disagree. Firstly, if we maintain the current IUPAC definition of “molecular ion,” we perpetuate a mistake, that of having a “semantically confusing” and segregating—“biased toward electrons”—definition for such crucial MS term. Users—most particularly beginners and non-experts—would therefore continue, misguided by the “incorrect semantic of the adjective molecular,” to be strongly tempted to misuse the term “molecular ion” for “the expelled molecular ions.” Secondly, we disagree because the term “quasimolecular ion” is not as inclusive as our proposed definition is for even-electrons permanently charged ions (M+ or M−). These “true molecular ions” would not fall under the umbrella of “quasimolecular ion,” leaving them “termless.” Thirdly, and perhaps most significantly, the main problem with the term “quasimolecular ion” lies in its very semantic nature, particularly in the inherent meaning of the adjective “quasimolecular.” If “molecular” means “of molecules, involving molecules, composed of molecules, or derived from molecules,” it logically follows that “quasimolecular” means “of quasimolecules, involving quasimolecules, composed of quasimolecules, or derived from quasimolecules.” But then how would we define a “quasimolecule”? How different from a typical molecule would we allow a “quasimolecule” to be? Therefore, we find the concern raised by Bursey in an editorial for Mass Spectrometry Review justified6: Show me a quasimolecule, and then I shall agree you can ionize it. Bursey also stated in his editorial that: The workers who first used the term quasimolecular ion had the best of intentions since they to an ion with the but not the mass of the molecule. Indeed, if its was to an ion with an but not mass of the then the term “quasimolecular ion” would to another This is so because the term would also to ionized molecules as well such as M+., including therefore ions that the current IUPAC definition already that is, as “molecular ions.” For instance, the of M+. species, from the removal of one electron or are but not the mass of the an ionized molecule would also have to be as “quasimolecular ion.” To this consult any of the numerous mass currently which are being used by many of we can agree that the term “quasimolecular ion” is for ions such as M+. or [M − since these species are formed by that are molecules, only an electron or a But that species such as [M + or [M + Na]+, among the molecule M, with removed from it. Therefore, the term “quasimolecular ion” to such ions would be inappropriate. In we strongly to the use of the term “quasimolecular it would only more confusion to the confusion already caused by the current IUPAC we it, when the term “quasimolecular ion” was the was not to to an ion derived from a as the term since the term means but not or or the we to describe ions that are a molecular ion” or ions that are as a molecular ion” or ions that the for being as a molecular ion.” This was particularly because the IUPAC definition artificially “molecular ions” based on electron addition or the term “quasimolecular ion” was to the problem by the segregating definition of “molecular ion.” But unfortunately, the the other the problem by confusion where confusion already We argue that it is far more to the of the segregating definition of a molecular the term to include the “expelled but rather than creating another class of ions to them. In we like to our proposal. In fact, further prompted by the responses we now would like to an more collective definition for the term “molecular ion,” and new or terms and and their for subclasses of molecular ions: ion collective class of ions and regardless of the method or added or removed any intact molecule being formed by or one or more electrons or protons M, or of one or more molecules (M) with one or more cations such as or such as ions also include permanently charged “molecular” cations M+ or molecular which are transferred as such from their salts to the gas phase environment of a mass such as or ions are therefore as M+., M−., [M + nH]n+, [M---H---M]+, [M − [M + [M + nNa]n+, [M + [M + M+ or or other where M their intact or nearly intact molecular terms such as ion, molecular ion or quasimolecular ion are molecule subclass of molecular ion formed by the removal of protons from a molecule M as [M − molecule subclass of molecular ion formed by the addition of protons to a molecule M as [M + ion of molecular ion formed either by the addition or the removal of one or more electrons from a molecule M to form either a cation such as M+. or an anion such as The term used to describe a molecule that or electrons is permanently charged ion subclass of even-electron molecular ions that be by the addition or removal of a include and This collective definition for “molecular ion,” we argue, ambiguity, vagueness, in some verbosity, while also a more for this pivotal MS term. By it, the fully extended definition that we use the term “molecular ion” to the of different subclasses of ions by MS for molecules. The fully expanded definition also a of even-electron permanently charged ions, both and which were in the and have therefore but that can now be readily transferred as ions to mass It also all types of both “molecular” and “quasimolecular” ions under the same collective term, “molecular ion” a well-defined and term that fully its strong semantic and The data that the of this are available on from the The data are not available due to or