Assessing the reliability of the NIST library during routine GC-MS analyses: Structure and spectral data corroboration for 5,5-diphenyl-1,3-dioxolan-4-one during a recent OPCW proficiency test

Full text of DOI:10.1002/jms.4073

Assessing the reliability of the NIST library during routine GC-MS analyses: Structure and
spectral data corroboration for 5,5-diphenyl-1,3-dioxolan-4-one during a recent OPCW
proficiency test
Carlos A Valdez
Corresponding author
valdez11@llnl.gov
Lawrence Livermore National Laboratory
Forensic Science Center
Livermore
California
United States
94550-5507
Roald N Leif
leif2@llnl.gov
Lawrence Livermore National Laboratory
Forensic Science Center
Livermore
California
United States
Saphon Hok
hok1@llnl.gov
Lawrence Livermore National Laboratory
Forensic Science Center
Livermore
California
United States
Armando Alcaraz
alcaraz2@llnl.gov
Lawrence Livermore National Laboratory
Forensic Science Center
Livermore
California
United States
This article has been accepted for publication and undergone full peer review but has not
been through the copyediting, typesetting, pagination and proofreading process which may
lead to differences between this version and the Version of Record. Please cite this article as
doi: 10.1002/jms.4073
This article is protected by copyright. All rights reserved.

Dear Editor,
In the field of analytical chemistry, it is the nature of the available instrumentation that makes a
specific technique uniquely suited for answering a specific challenge. Often times, instruments
equipped with ultra-sensitive array of detectors and state-of-the-art separation components have
served to highlight their invaluable role in the field.[1, 2] Nonetheless, it is their ultimate ability to
predict the structure of an unknown material on the basis of mass spectral data that exponentially
magnifies their value. Their efficacy in achieving this prediction is largely based on the size of their
internal mass spectral library that conveniently can be accessed using the software. The library can
thus exhaustively correlate an unknown compound’s mass spectrum to known compounds in a
database and provide the analytical chemist with a probable structure for the unknown analyte.[1]
During routine analysis, it is customary that after the appropriate collection of the compound’s mass
spectrum, the next step involves its comparison to a compound in the instrument’s National Institute
of Standards and Technology (NIST) library.[3] The results are commonly given a probability score
based on the similarity between the mass spectral data of the analyzed compound to the its closest hit
in the library. Part of the algorithms involved in the overall matching process include a direct
comparison of key fragment ions, their individual intensities and their abundance relative to each
other in the spectrum. Thus, compounds bearing fragmentation ions with identical nominal mass
values may not obtain a high similarity score due to the difference in the relative intensities of the
peaks associated with these ions. It is often difficult to reliably depend on the identification given by
the instrument’s spectral library due to the fact that the library’s size is never large enough and
consequently may not contain the unknown compound. Another reason and this one has become
more important as internal NIST libraries grow, is the finding that some errors may exist in the
current spectral libraries that can lead to erroneous association of a structure to a given unknown
compound leading to a complete misidentification of such compound or a derivative thereof.[4] This
becomes an important issue when attempting to unambiguously identify compounds in complex
matrices pertaining to real world scenarios. These scenarios are often encountered and in order to
remain competent when they occur, laboratories around the world, including ours, associated with the
Organisation for the Prohibition of Chemical Weapons (OPCW) take yearly proficiency tests (PTs)
with this sole aim. The PTs require for a competing laboratory to successfully identify reportable
compounds in six separate mixtures, matrices varying in nature every year, amidst a number of
interferences and red herring compounds that are purposely added to make their instrumental analysis
a challenging task.[5, 6] More critically, these PTs enforce a strict 2-week timeline from receipt of the
samples to be completed thus immersing the laboratories seeking certification as close as possible in a
real-case scenario.[7] Although some of the reportable compounds for the PT are commercially
available and very often possess a matching hit in the instrument’s NIST library, there is the
possibility that the lab will encounter a compound that needs to be chemically synthesized for use as a
This article is protected by copyright. All rights reserved.

standard. In this letter, we briefly describe our experience with one such case where an unknown
compound of interest’s mass spectrum did not fully match the one proposed by the NIST library in the
GC-MS instrument prompting us to synthesize it in house and fully characterize it.
Within the space of chemical warfare agents (CWAs), incapacitating agents are a class exclusively
devoted to physically or mentally disable a threat without causing its demise.[8] However, at large
doses or prolonged exposure times, these agents may cause death.[9] One of the most common
incapacitating agents is the Schedule 2 compound 3-quinuclidinyl benzilate (BZ, 1, Fig. 1). Due to its
importance in the CWA field, several detection protocols based on LC-[10] and GC-MS[11] techniques
have been developed for BZ. BZ is prone to hydrolysis in an aqueous environment reason why its
indirect detection is carried out retrospectively by analyzing its degradation products namely, 3-
quinuclidinol (3Q, 2) and benzilic acid (3) (Fig. 1). Analysis of both of these materials, particularly
benzilic acid, is not possible by GC-MS means due to its low volatility and high polarity. Methods
for its detection and identification by GC-MS have involve its derivatization into more volatile
analogs through the formation of methyl and silyl esters.[12]
As part of the 40th OCPW-PT that our laboratory participated in, the analysis of several matrices was
needed. Extraction of the matrices with an organic solvent followed by analysis using GC-MS
initially showed a sample concentrated with BZ (1) and among other components, benzilic acid
methyl ester (4) (Fig. 2). Given that the scenario clearly involved BZ, we focused on identifying not
only this material but degradation products arising from its hydrolysis. Within the rules of an OPCW-
PT, reporting compounds that are derivatives of scheduled materials is crucial and mandated as
neglecting to do so may result in a letter grade deduction or failure of the test leading to the loss of
certification by the participating laboratory. Thus, as benzilic acid is a direct hydrolysis product of
BZ, we found the need to carefully evaluate any other peaks in the chromatogram to eliminate other
potentially reportable derivatives of this acid. However, analysis of the matrix by GC-MS revealed
that in addition to the methyl ester compound 4, another clearly resolved peak hinting at the presence
of yet another benzilic acid derivative. This unknown peak appeared at t = 24.3 min. under our GC
conditions and initial library matching using the instrument’s library yielded the cyclic acetal
compound 5 as a possible identification (Fig. 2). As it is customary during OPCW proficiency tests,
further confirmation of the proposed structures by the instrument’s NIST libraries is carried out by
using a standard. In some cases, the standard may be commercially available while in some instances
its chemical synthesis is required. Often times, chemical synthesis of the standards represents the
fastest way to obtain a material for comparison, as due to the time sensitive nature of the PT, a
participating laboratory does not have the luxury to order it from a vendor and lose 2-3 days for its
analysis. The benzilic acid methyl ester 4 was synthesized in house in one step by treating benzilic
This article is protected by copyright. All rights reserved.

acid with one equivalent of trimethyloxonium tetrafluoroborate[13, 14] and the crude mixture was used
in the corroboration of structure of the methyl ester present in the mixture.
We synthesized 5 via an acid-catalyzed condensation reaction between benzilic acid and
dimethoxyformaldehyde (Scheme 1). The cyclic acetal was obtained in 77% yield after column
chromatography. Although compound 5’s retention time matched the one exhibited by the unknown
chemical in the sample between its mass spectral data did not (Fig. 3). Interestingly, upon checking
the original report for the synthesis of 5 [15], we found that back in 1968 the Mörsdorf group
1
synthesized the material as evidenced from the H NMR chemical shift of the methylene protons
(Scheme 1, in gray) from the acetal functionality matched the one obtained in our laboratory ( = 5.50
ppm, CDCl3).
Regarding the noted differences between the mass spectra for the synthesized cyclic acetal 5 and the
one from the instrument’s NIST internal library, one can see that the most striking one lies in the
intensity of the respective molecular ion peak (‡, m/z = 240, Figs. 3A and 3B). Whereas the library
reference spectrum shows a pronounced molecular ion peak, the peak in the spectrum of our cyclic
acetal is barely visible. A second difference lies in the nature of the base peak (†). For the library
reference compound, the base peak is m/z = 105 while for our synthesized standard is m/z = 165 (±).
Intuitively a fragment bearing a m/z = 165 would arise from a dibenzyl radical cation species (Scheme
2, path a), an ionic structure that is expected to exhibit stability due to its doubly benzylic nature. In
the case of the m/z = 105 fragment, this one would arise from the cleavage of the molecule to yield a
benzene carbonyl cationic species that could in part enjoy stabilization (Scheme 2, path b). A third
difference lies in the overall intensity of the peak m/z = 182, arising from path d in Scheme 2, with the
library entry featuring a more intense peak over that one exhibited by our synthesized 5. In addition
to these major difference, two ions that appear very similar in intensities in both spectra are the ones
giving rise to m/z = 77 and m/z = 195 arising from paths c and e respectively in Scheme 2. Other,
more minor differences can be observed in Figure 3. Initially we attributed the discrepancy between
the mass spectral data as the actual analysis of two, distinct isomeric structures. However, this idea
was discarded based on the Mörsdorf group’s synthetic preparation to 5,5-diphenyl-1,3-dioxolan-4-
1
one and the fact that we obtain an exact match by H-NMR analysis. A likely explanation could be
that the spectrum differs significantly mainly due to the type of ion separation and subsequent
detection employed in the original work when compared to ours (TOF). Unfortunately, the original
work does not specify the type of instrumentation and mode of detection employed. Therefore, in this
letter we would like to introduce the revised and corrected mass spectrum for 5,5-diphenyl-1,3-
This article is protected by copyright. All rights reserved.

dioxolan-4-one (5). The structure of the synthesized material as corroborated by multinuclear (1H,
13C) NMR spectroscopy as well as EI-GC-MS.
Experimental section
Synthetic and analytical procedures are described in the Supporting Information section
accompanying this letter.
Acknowledgements
This work performed under the auspices of the U.S. Department of Energy by Lawrence Livermore
National Laboratory under Contract DE-AC52-07NA27344. This document (LLNL-JRNL-745368)
was prepared as an account of work sponsored by an agency of the United States government. Neither
the United States government nor Lawrence Livermore National Security, LLC, nor any of their
employees makes any warranty, expressed or implied, or assumes any legal liability or responsibility
for the accuracy, completeness, or usefulness of any information, apparatus, product, or process
disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any
specific commercial product, process, or service by trade name, trademark, manufacturer, or
otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by
the United States government or Lawrence Livermore National Security, LLC. The views and
opinions of authors expressed herein do not necessarily state or reflect those of the United States
government or Lawrence Livermore National Security, LLC, and shall not be used for advertising or
product endorsement purposes.
Sincerely Yours,
Carlos A. Valdez*,1-3, Roald N. Leif 1-3, Saphon Hok1-3 and Armando Alcaraz1-3
1 Physical and Life Sciences Directorate, Lawrence Livermore National Laboratory, Livermore, CA
94550, USA.
2 Nuclear and Chemical Sciences Division, Lawrence Livermore National Laboratory, Livermore, CA
94550, USA.
3 Forensic Science Center, Lawrence Livermore National Laboratory, Livermore, CA 94550, USA.
*Correspondence to:
Dr. Carlos A. Valdez
Lawrence Livermore National Laboratory,
Physical and Life Sciences Directorate,
Nuclear and Chemical Sciences Division and the
Forensic Science Center, L-091
Livermore, CA 94550, USA
Email: valdez11@llnl.gov
This article is protected by copyright. All rights reserved.

References
[1]
Mondello L, Tranchida PQ, Dugo P, Dugo G (2008) Comprehensive Two-Dimensional Gas
Chromatography-Mass Spectrometry: A Review. Mass Spectr. Rev. 27:101-124.
[2]
Ackermann BL, Berna MJ, Eckstein JA, Ott LW, Chaudhary AK (2008) Current Applications
of Liquid Chromatography/Mass Spectrometry in Pharmaceutical Discovery After a Decade of
Innovation. Annu. Rev. Anal. Chem. 1:357-396.
[3]
and collision-induced dissociation. J. Chromatogr. A 974:161-168.
[4] Wille SMR, Lambert WEE (2004) Phenmetrazine or ephedrine? Fooled by library search. J.
Chromatogr A 1045:259-262.
Kienhuis PGM, Geerdink RB (2002) A mass spectral library based on chemical ionization
[5]
Mayer BP, Valdez CA, Hok S, Chinn SC, Hart BR (2012) 31P-Edited diffusion-ordered 1H
NMR spectroscopy for the spectral isolation and identification of organophosphorus compounds
related to chemical weapons agents and their degradation products. Anal. Chem. 84:10478-10484.
[6]
Albo RLF, Valdez CA, Leif RN, Mulcahy HA, Koester C (2014) Derivatization of pinacolyl
alcohol with phenyldimethylchlorosilane for enhanced detection by gas chromatography-mass
spectromerty. Anal. Bioanal. Chem. 406:5231-5234.
[7]
Convention. Anal. Bioanal. Chem. 406:5067-5069.
[8] Coupland RM (2003) Incapacitating chemical weapons: a year after the Moscow theatre
siege. Lancet 362:1346.
Blum M-M, Murty Mamidanna RVS (2014) Analytical chemistry and the Chemical Weapons
[9]
Rieder J, Keller C, Hoffmann G, Lirk P (2003) Moscow theatre siege and anaesthetic drugs.
Lancet 361:1131.
[10]
Dolle F, Hinnen F, Vaufrey F, Demphel S, Bramoulle Y, Fournier D, Crouzel C (2001)
Highly efficient synthesis of [11C]Me-QNB, a selective radioligand for the quantification of the
cardiac muscarinic receptors using PET. J. Label. Comp. Radiopharm. 44:337-345.
[11]
Byrd GD, Paule RC, Sander LC, Sniegoski LT, White E, Bausum HT (1992) Determination
of 3-quinuclidinyl benzilate (QNB) and its major metabolites in urine by isotope dilution gas
chromatography/mass spectrometry. J. Anal. Toxicol. 16:182-187.
[12]
Park YK, Chung WY, Kim B, Kye Y-s, Shin M-s, Kim D. (2013) Ion-pair single-drop
microextraction determinations of degradation products of chemical warfare agents in water.
Chromatographia 76:679-685.
[13]
Valdez CA, Leif RN, Alcaraz A (2016) Effective methylation of phosphonic acids related to
chemical warfare agents mediated by trimethyloxonium tetrafluoroborate for their qualitative
detection and identification by gas chromatography-mass spectrometry. Anal. Chim. Acta 933:134-
143.
This article is protected by copyright. All rights reserved.

[14]
Valdez CA, Leif RN, Hok S, Hart BR (2018) Analysis of chemical warfare agents by gas
chromatography-mass spectrometry: methods for their direct detection and derivatization approaches
for the analysis of their degradation products, Rev. Anal. Chem. doi:10.1515/revac-2017-0007.
[15]
Pailer M, Streicher W, Takacs F, Mörsdorf K (1968) Derivatives of 1,3-dioxolan-4-one and
1,3-oxathiolan-5-one. Monatsh. Chem. 99:891-901.
This article is protected by copyright. All rights reserved.

Scheme 1. Synthesis of benzilic acid acetal 5.
Scheme 2. Proposed fragmentation patterns for compound 5 leading to observed spectral peaks.
This article is protected by copyright. All rights reserved.

Figure 1. Hydrolysis of BZ to provide 3-quinuclidinol and benzilic acid.
A
B
Figure 2. A) Full chromatogram of OPCW-PT organic sample and B) section enlargement showing
the peaks for the benzilic acid methyl ester (t = 23.5 min.) and an unknown compound (*) (t = 24.3
min.).
This article is protected by copyright. All rights reserved.

Figure 3. Comparison between LLNL (A) and NIST (B) mass spectra for 5,5-diphenyl-1,3-dioxolan-
4-one.
This article is protected by copyright. All rights reserved.