Optimizing Detection of Leachables To Ensure Drug Safety

Application Note Extractables and Leachables Authors Sofia Nieto, Anastasia Andrianova, Bruce Quimby, and David Weil Agilent Technologies, Inc. Abstract Chemicals that are part of polymeric container closure systems (CCS) and drug delivery systems have the potential to migrate into drug products during manufacturing, storage, transport, and delivery, and must be identified in the final products to ensure their safety. This application note presents a rubber gasket extractables study using a unit mass resolution gas chromatography/mass selective detector (GC/MSD) and a high-resolution gas chromatography/quadrupole time-of-flight (GC/Q-TOF) mass spectrometer to establish a process for identifying GC-amenable extractables and leachable (E&L) compounds. Analysis of Volatile Compounds Identified in Rubber Gasket Extracts Using GC/MSD and High-Resolution GC/Q-TOF 2 Introduction Elastomeric gaskets, plungers, and O-rings are common sources of leachable compounds in the manufacturing, storage, and delivery of drug products. E&Ls derived from elastomeric components may impact the stability and efficacy of small and large molecule drug products1 , and therefore need to be characterized thoroughly. Exposure to some E&L chemicals, such as phthalates and nitrosamines, even at low levels, may cause safety concerns.2 Chemicals derived from the elastomer manufacturing process typically include accelerators, activators, antioxidants, fillers, plasticizers (including phthalates), mold release agents, and other additives3 that may leach into the final product. Some additives present in elastomer packaging materials may also contain polycyclic aromatic hydrocarbons (PAHs)4 and aliphatic hydrocarbons. GC/MS is a commonly used technique for analyzing volatile and semivolatile organic compounds in the E&L space. This study demonstrates the capabilities of GC/MSD to identify GC‑amenable compounds present in a solvent extract of a rubber gasket by leveraging chromatographic deconvolution in combination with retention index (RI)‑based filtering. Adding a high-resolution accurate mass GC/Q‑TOF into the E&L workflow provided a higher number of identified chemicals. It also increased confidence in compound identification and enabled structure elucidation of unknown compounds. The study was performed in the Network Workstation configuration using Agilent OpenLab Electronic Content Management (ECM) XT as the data repository. This configuration enabled tools that facilitate compliance with various national and EU electronic record regulations, including audit trails, user authentication, role-based permission controls, and remote data storage.5 Experimental Sample preparation Rubber syringe gaskets were extracted using tetrahydrofuran (THF) solvent at room temperature for six months. An aliquot of the extracts, along with solvent blanks, were analyzed using GC/MSD and GC/Q-TOF systems. Data acquisition The GC/MS analysis was performed using an Agilent 5977C GC/MSD and an Agilent 7250 GC/Q-TOF system in electron ionization (EI) mode. The GC/Q-TOF was also used in low‑energy EI mode to help identify molecular ions of unknowns. Injection conditions were optimized for a broad range of E&L compound boiling points. Using pulsed splitless injection mode and delaying the purge flow to the split vent for 1 to 2 minutes maximized the response for both low- and high‑boiling compounds (Figure 1). Figure 1. EIC (m/z 57) of a C5 to C40 n-alkane standard analyzed under the starting (top) and optimized (bottom) conditions. C40 C31 C22 C18 C13 C9 ×105 Acquisition time (min) Counts 0 1 2 3 4 5 6 7 ×105 Counts 0 1 2 3 4 5 6 7 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 C40 ×104 26 27 28 29 30 31 32 33 34 35 36 0 2 4 6 8 ×104 0 2 4 6 8 3 Initially, both 30 m × 0.25 mm, 0.25 µm and 20 m × 0.18 mm, 0.18 µm Agilent J&W DB-5ms Ultra Inert columns were evaluated for their chromatographic separation capabilities of the complex E&L extracts, as well as sensitivity after optimization of the carrier gas flow for each column dimension. While the 20 m column provided sharper peaks and greater sensitivity for trace-level compounds, the 30 m column offered better separation, with a higher number of components reliably identified. The 30 m column was therefore selected. All data were acquired in full spectrum acquisition mode using the new Agilent J&W DB-5Q nonpolar low bleed column and the DB-5ms Ultra Inert GC column. The acquisition software operated under a unified compliance environment using OpenLab ECM XT. The typical data acquisition parameters are shown in Table 1. Table 1. Data acquisition parameters. Parameter Value MS Agilent 7250 GC/Q-TOF; Agilent 5977C GC/MSD GC Agilent 8890 GC Column Agilent J&W DB-5Q, 30 m × 0.25 mm, 0.25 µm (p/n 122-5532Q) Inlet Multimode inlet, 4 mm Ultra Inert inlet liner, single taper with wool Injection Volume 1 µL Injection Mode Pulsed splitless (1 min purge, pulse at 40 psi for 1.1 min) Inlet Temperature Program 65 °C for 0.01 min, 300 °C/min to 280 °C Oven Temperature Program 45 °C for 2 min; 12 °C/min to 325 °C, 11 min hold Carrier Gas Helium Column Flow 1 mL/min constant flow Transfer Line Temperature 325 °C Quadrupole Temperature 150 °C Source Temperature 200 °C (Q-TOF)/300 °C (MSD) Electron Energy 70 eV (standard EI MSD, Q-TOF); 15, 12, and 10 eV (low-energy EI, Q-TOF) Emission Current 5 µA (standard EI, Q-TOF); 0.3 µA (low-energy EI, Q-TOF), 35 µA MSD Spectral Acquisition Rate 5 Hz (Q-TOF), 2 Hz (MSD) Mass Range m/z 50 to 1,000 (Q-TOF), 45 to 450 (MSD) Data processing The chromatographic deconvolution and library search were performed in the Agilent MassHunter Unknowns Analysis 12.1 Update 2. The NIST23 library was used to perform the initial compound identification. Structural elucidation was performed using the Agilent Molecular Structure Correlator (MSC) software 8.2. Retention time (RT) locking was used to ensure consistent RTs between the GC/MSD and GC/Q-TOF systems. It also allowed for both RI and RT matching. Results and discussion Advantages of using the new Agilent low-bleed DB-5Q column for E&L applications A beta version of the new Agilent DB-5Q column was evaluated in terms of suitability for E&L studies. Many compounds of interest, including phthalates, antioxidants, UV-absorbers, and stabilizers have high boiling points. The detection of these compounds is therefore more susceptible to interference from column bleed, which is more evident at high oven temperatures. Two different sets of DB-5Q and DB-5ms UI columns were compared and a significantly lower column bleed at high oven temperatures was observed for the DB-5Q columns, compared to the DB-5ms UI. One representative example is shown in Figure 2A. The data were acquired on the GC/Q-TOF using an emission current of 0.3 µA, resulting in similar perfluorotributylamine (PFTBA) abundances. The oven was kept at 325 ˚C while PFTBA and background spectra were recorded. A few high-boiling compounds, such as antioxidants and UV-absorbers, were also analyzed on the two columns for comparison. The DB-5Q column produced less column bleed background in these conditions, as evident from the TIC of the UV absorbers (Figure 2B), and a spectrum of the antioxidant Irgafos 168, extracted without background subtraction (Figure 2C). It is typical for E&L extracts to contain a significant proportion of water; therefore, the DB-5Q column performance was tested before and after 130 injections of E&L extracts with various solvents, including ethanol:water (1:1) and THF. Octafluoronaphthalene (OFN) was injected at 1 pg onto the column before and after 130 extract injections. Peak shape, response, and spectrum integrity were all maintained after injecting water-containing extracts (Figure 3). 4 Figure 2. Agilent DB-5ms and DB-5Q column bleed comparison on the GC/Q-TOF. (A) Background and PFTBA spectra collected at oven temperature 325 °C and emission current 0.3 µA. (B) TIC of UV absorbers. (C) Raw spectra of an antioxidant Irgafos 168 without background subtraction (high boiling compound with an RI of 3,398 and an RT of 27.6 minutes). ×105 ×105 ×107 ×107 Counts Counts Counts Counts Counts Counts 0 2 4 6 207.0325 68.9947 130.9915 281.0515 218.9851 96.0043 147.0656 193.0499 263.9866 118.9923 239.2369 295.1032 0 2 4 6 68.9948 130.9916 218.9853 99.9932 118.9916 147.0657 168.9884 191.0012 239.2372 263.9868 281.0515 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 Column bleed ions PFTBA ions A B ×106 ×106 C 0 0.5 1.0 0 0.2 0.4 0.6 0.8 1.0 1.2 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Tinuvin 328 Tinuvin 320 Irgacure 907 Padimate O Methyl-2-benzoylbenzoate Irgacure 184 Benzophenone 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 *207.0323 281.0512 441.2920 308.2018 355.0699 415.1062 489.1250 549.1611 646.4509 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 207.0322 441.2918 281.0511 147.0654 327.0351 415.1061 489.1249 549.1613 646.4512 150 175 200 225 250 275 300 325 350 375 400 425 450 475 500 525 550 575 600 625 650 (*Saturated) Mass-to-charge (m/z) Mass-to-charge (m/z) Acquisition time (min) Column bleed ions Irgafos 168 ions Agilent DB-5ms UI Agilent DB-5Q Agilent DB-5ms UI Agilent DB-5Q Agilent DB-5ms UI Agilent DB-5Q 5 Figure 3. (A) OFN EIC for m/z 271.9867 ± 20 ppm and (B) OFN spectrum. OFN was injected at 1 pg onto an Agilent DB-5Q column before and after 130 injections. All injections were performed in splitless mode. ×104 ×104 ×104 ×104 Counts Counts Counts Counts A B Mass-to-charge (m/z) Acquisition time (min) 0 0.5 1.0 1.5 5.724 0 0.5 1.0 1.5 5.709 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 5 5.1 5.2 5.3 5.4 5.5 5.6 5.7 5.8 5.9 6 6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8 0 0.5 1.0 1.5 271.9868 240.9883 221.9899 202.9914 135.9933 162.9974 92.9945 116.9947 253.9940 0 0.5 1.0 1.5 271.9867 240.9878 221.9896 202.9908 86.0721 105.0323 135.9935 152.9928 171.9932 260.0392 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 Before 130 injections After 130 injections Before 130 injections After 130 injections 6 The consistency of RTs and RIs between the DB-5Q and the standard DB-5ms UI column was also evaluated. The RT values for n-alkanes in a range of C7 to C39, analyzed using an RT-locked method, were found to be very close when comparing the two columns (Figure 4A). The RIs for 70 compounds of various chemical classes and boiling points had a remarkable consistency between the DB-5Q and DB-5ms columns (with an average delta RI of 0.97 RI units) and were comparable to NIST experimental RI values for the semistandard nonpolar column phase (Figure 4B). For additional information about the new ultralow bleed 5Q columns, see a separate technical note.6 0 5 10 15 20 25 30 35 0 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 RT (min) Alkane RI DB-5ms DB-5Q 900 1,400 1,900 2,400 2,900 3,400 RI DB-5ms DB-5Q NIST 1,501 1,500 1,482 1,525 1,527 1,545 1,536 1,535 1,515 A B Figure 4. A comparison of RTs and RIs between the Agilent DB-5ms UI column and Agilent DB-5Q column. (A) n-Alkane RTs on the DB-5ms UI (blue) and DB-5Q (orange) columns. (B) RI consistency for 70 compounds between the DB-5ms (dark blue) and DB-5Q (orange) columns. NIST23 experimental RIs are shown in green. 7 Identification of semivolatile compounds in rubber gasket extract using GC/MSD and GC/Q-TOF Over 100 compounds were initially identified in the sample using the GC/MSD by searching deconvoluted spectra against the NIST23 library and filtering the results based on RIs. Figure 5 shows an example of an identified compound, eicosyl acetate, in the presence of coeluting components with a high library match score (LMS) and excellent RI matching. To take advantage of the accurate mass, high sensitivity in full spectrum acquisition mode, and MS/MS capabilities beneficial for identification of unknowns, the same rubber gasket extracts were also analyzed using the GC/Q-TOF. Over 80 compounds were identified in common between the GC/MSD and GC/Q-TOF, a few of which are shown in Table 2. Figure 5. TIC of a rubber gasket sample and deconvoluted spectrum for eicosyl acetate with an LMS of 91.8 and RI delta of 1. LMS: 91.8 ∆RI: 1 8 Table 2. Common compounds identified by both GC/MSD and GC/Q-TOF using a library match factor cutoff of 70. RT Compound Name Formula CAS No. 4.48 Butanoic acid C4 H8 O2 107-92-6 5.11 Dipropyl acetal C8 H18O2 105-82-8 5.68 N-Ethylacetamide C4 H9 NO 625-50-3 5.75 Pentanoic acid C5 H10O2 109-52-4 7.13 Hexanoic acid C6 H12O2 142-62-1 7.15 Glycerin C3 H8 O3 56-81-5 7.22 Phenol C6 H6 O 108-95-2 8.04 2-Acetyl-5-methylfuran C7 H8 O2 1193-79-9 8.44 Heptanoic acid C7 H14O2 111-14-8 8.53 Isovaleraldehyde dipropyl acetal C11H24O2 1000431-60-3 8.54 Acetophenone C8 H8 O 98-86-2 8.55 p-Cresol C7 H8 O 106-44-5 8.60 4-Methylbenzaldehyde C8 H8 O 104-87-0 8.79 (1-Methoxypropyl)benzene C10H14O 59588-12-4 9.23 Triacetonamine C9 H17NO 826-36-8 9.63 Benzoic acid C7 H6 O2 65-85-0 9.72 Octanoic acid C8 H16O2 124-07-2 10.95 Nonanoic acid C9 H18O2 112-05-0 11.69 2,3-Dihydro-1H-pyrrolizin-1-one C7 H7 NO 17266-64-7 12.74 Diphenyl ether C12H10O 101-84-8 12.85 n-tert-Butylphenetole C12H18O 17269-94-2 12.93 Longifolene C15H24 475-20-7 13.18 Dimethyl phthalate C10H10O4 131-11-3 13.41 Ethyl 3-phenylpropenoate C11H12O2 103-36-6 13.42 1-Dodecanol C12H26O 112-53-8 13.76 2,4-Di-tert-butylphenol C14H22O 96-76-4 13.78 Butylated hydroxytoluene C15H24O 128-37-0 14.38 (3-Decyl)benzene C16H26 4621-36-7 14.54 Pentyl salicylate C12H16O3 2050-08-0 14.63 Diethyl phthalate C12H14O4 84-66-2 14.79 p-tert-Octylphenol C14H22O 140-66-9 15.12 Tributyl phosphate C12H27O4 P 126-73-8 15.39 (1-Ethylnonyl)benzene C17H28 4536-87-2 RT Compound Name Formula CAS No. 15.56 n-Hexyl salicylate C13H18O3 6259-76-3 15.62 3-Pentadecanone C15H30O 18787-66-1 15.74 4-(1,1-Dimethylheptyl)phenol C15H24O 30784-30-6 15.82 4-(7-Methyloctyl)phenol C15H24O 24518-48-7 15.93 1-Phenyl-1,3,3-trimethylindane C18H20 3910-35-8 16.20 Tetradecanoic acid C14H28O2 544-63-8 16.30 3,5-di-tert-Butyl-4-hydroxybenzaldehyde C15H22O2 1620-98-0 16.67 2,6,10,14-Tetramethylhexadecane (Phytane) C20H42 638-36-8 16.74 3,5-di-tert-Butyl-4-hydroxyacetophenone C16H24O2 14035-33-7 16.81 Isopropyl myristate C17H34O2 110-27-0 16.98 2,4-Diphenyl-4-methyl-2(E)-pentene C18H20 22768-22-5 17.59 7,9-Di-tert-butyl-1-oxaspiro(4,5)deca-6,9- diene-2,8-dione C17H24O3 82304-66-3 17.60 Farnesyl acetone C18H30O 1117-52-8 17.98 Dibutyl phthalate C16H22O4 84-74-2 17.99 n-Hexadecanoic acid C16H32O2 57-10-3 18.34 18-Norabieta-8,11,13-triene C19H28 1000197-14-1 18.71 N,N-Dimethyltetradecanamide C16H33NO 3015-65-4 19.38 Linoleic acid C18H32O2 60-33-3 19.60 Octadecanoic acid C18H36O2 57-11-4 19.80 n-Pentadecylcyclohexane C21H42 6006-95-7 20.31 N,N-Dimethylpalmitamide C18H37NO 3886-91-7 21.40 Eicosyl acetate C22H44O2 822-24-2 21.46 Antioxidant 2246 C23H32O2 119-47-1 21.56 N,N-Dimethyllinoleamide C20H37NO 2501-33-9 21.60 N,N-Dimethyloleamide C20H39NO 2664-42-8 21.74 Dehydroabietic acid C20H28O2 1740-19-8 22.09 Antioxidant 425 C25H36O2 88-24-4 23.02 Squalane C30H62 111-01-3 23.83 13-Docosenamide, (Z)- C22H43NO 112-84-5 26.81 Chondrillasterol C29H48O 481-17-4 27.37 (24Z)-Ethylidenecholesterol C29H48O 481-14-1 9 To gain higher confidence in E&L compound identification, the accurate mass information was used to either confirm or reject the compound ID with assistance of the ExactMass tool of the MassHunter Unknowns Analysis software. The ExactMass tool automatically assigns fragment ions with formulas that are a subset of the molecular formula of the top library hit, when possible. The library hit can be considered a false positive when most specific fragments do not match the compound formula within a small mass error. Figure 6 provides two such examples. Due to the higher sensitivity in full spectrum acquisition mode and higher data acquisition rate of the GC/Q-TOF, compared to the GC/MSD, a few additional compounds have been identified by GC/Q-TOF (Table 3). These compounds included catalysts, solvents, vulcanization accelerators, plasticizers, antioxidants, and UV stabilizers used in rubber manufacturing. The compound identification was confirmed using accurate mass and RI information. A B Figure 6. Confirmation of compound ID using accurate mass. Fragment formulas are assigned based on accurate mass and the molecular formula of the library hit. The mass error of each prominent fragment ion is then calculated and displayed in the ExactMass table. (A) A confirmed compound identified uniquely by GC/Q-TOF. (B) A false positive, as determined when processing the GC/Q-TOF data based on accurate mass. However, the same compound ID was incorrectly assigned to this spectrum based on the GC/MSD unit mass data with a high library match score of 89. 10 Table 3. Compounds identified uniquely by GC/Q-TOF. RT Compound Name Match Factor Formula Delta RI CAS No. 4.17 Methyl isobutyl ketone 92.8 C6 H12O –29.7 108-10-1 4.61 Acetylacetone 87.7 C5 H8 O2 –19.7 123-54-6 4.63 Dimethylformamide 99.1 C3 H7 NO –21.2 68-12-2 4.86 Hexanal 96.7 C6 H12O –18.9 66-25-1 5.03 Furfural 80.0 C5 H4 O2 1.1 98-01-1 5.80 o-Xylene 96.5 C8 H10 3.3 95-47-6 5.93 2,6-Lutidine (2,6-dimethylpyridine) 82.0 C7 H9 N –14.1 108-48-5 6.02 2-Heptanone 94.6 C7 H14O –9.3 110-43-0 6.21 Heptanal 94.6 C7 H14O –11.7 111-71-7 6.66 3-Hepten-2-one 79.6 C7 H12O –6.2 1119-44-4 6.91 Piperidine, 2,2,6,6-tetramethyl- 91.0 C9 H19N –19.8 768-66-1 7.10 Benzaldehyde 90.9 C7 H6 O –10.8 100-52-7 7.36 α-Methylstyrene 95.6 C9 H10 –4.2 98-83-9 7.63 Octanal 89.1 C8 H16O –5.5 124-13-0 7.96 2-Ethylhexanol 92.6 C8 H18O –1.7 104-76-7 8.11 N-Methyl-α-pyrrolidone 84.7 C5 H9 NO 1.4 872-50-4 8.16 2-(2-Hydroxypropoxy)-1-propanol 82.7 C6 H14O3 0.1 106-62-7 9.01 Nonanal 96.3 C9 H18O –3.0 124-19-6 10.08 2,4-Dimethylthiophenol 89.1 C8 H10S 19.0 13616-82-5 10.29 Benzene, 1,3-dibromo- 91.2 C6 H4 Br2 14.1 108-36-1 10.70 Benzothiazole 92.2 C7 H5 NS –9.3 95-16-9 11.44 m-tert-Butylphenol 72.0 C10H14O –2.2 585-34-2 12.35 3-Hydroxy-2,2,4-trimethylpentyl 2-methylpropanoate** 73.2 C12H24O3 –3.7 77-68-9 12.57 p-tert-Pentylphenol 74.3 C11H16O 3.2 80-46-6 13.27 BHT-quinol 84.6 C15H24O2 14.2 10396-80-2 13.54 Dicyclopentyl(dimethoxy)silane 88.3 C12H24O2 Si –11.9 126990-35-0 13.58 3-Tridecanone 83.2 C13H26O 4.6 1534-26-5 13.98 Ethyl 4-ethoxybenzoate 82.8 C11H14O3 –5.7 23676-09-7 14.77 (2-Decyl)benzene 88.2 C16H26 10.0 4537-13-7 15.06 (1-Butylheptyl)benzene 83.8 C17H28 –4.1 4537-15-9 15.08 Fenuron 73.1 C9 H12N2 O –5.2 101-42-8 15.15 Benzophenone 93.4 C13H10O –10.0 119-61-9 15.55 2,4-Ditert-butyl-6-nitrophenol 78.7 C14H21NO3 1.7 20039-94-5 15.89 4-(1,1-Dimethylheptyl)phenol 83.2 C15H24O –25.9 30784-30-6 16.69 Anthracene 86.4 C14H10 –23.5 120-12-7 17.17 Diisobutyl phthalate 88.5 C16H22O4 5.0 84-69-5 17.70 Methyl hexadecanoate 74.6 C17H34O2 1.3 112-39-0 19.01 p-Tolyl disulfide 73.8 C14H14S2 3.4 103-19-5 21.05 Methyl dehydroabietate 79.9 C21H30O2 –17.2 1235-74-1 22.26 Bis(2-ethylhexyl) phthalate (DEHP) 69.6 C24H38O4 0.0 1000377-93-5 25.72 Tinuvin 770 87.1 C28H52N2 O4 130.4* 52829-07-9 * Only predicted RI is available ** Component of texanol 11 Identification of unknown compounds in the rubber gasket extract A few unknowns have been selected for further identification. A typical structure elucidation workflow of unknown compounds requires identification of the molecular ion as the first step. This is challenging when using a standard EI, as the abundance of molecular ions in EI is rarely preserved. Low-energy EI (LE-EI) is a type of soft ionization that could help increase the relative abundance of molecular ions and thus their tentative identification. This technique is enabled by the LE-EI capable source of the 7250 GC/Q-TOF and is complementary to chemical ionization (CI). This technique does not require a reagent gas or a source change and uses the same tune file as a standard EI. Based on LE-EI results, molecular ions of the unknown compounds were proposed and listed in Table 4. Table 4. Molecular ion formulas of unknowns tentatively identified in the LE-EI experiments. RT (min) Tentative m/z of Molecular Ion Formula 5.59 98.0362 C5 H6 O2 6.37 142.0988 C8 H14O2 7.82 155.1067 C9 H15O2 8.44 143.1067 C8 H15O2 10.72 154.0988 C9 H14O2 11.93 166.0988 C10H14O2 12.10 150.1039 C10H14O 13.31 182.0937 C10H14O3 13.89 206.1301 C13H18O2 15.19 250.1927 C16H26O2 An example of how LE-EI can be used for identification or confirmation of molecular ions is shown in Figure 7, where a gradual increase of tentative molecular ion relative abundance at lower electron energies is observed. Figure 7. An example of using LE-EI to identify or confirm molecular ions. The lower the electron energy, the higher the relative abundance of the molecular ion. The tentative molecular ion is outlined in the rectangle. 175.1123 161.0964 190.1356 71.0494 250.1933 133.1013 281.0458 311.0904 341.0203 175.1118 190.1353 161.0961 250.1929 89.0594 281.0505 341.0197 190.1354 250.1931 175.1118 161.0961 71.0491 98.1087 135.0805 221.0839 250.1929 190.1352 175.1115 107.0853 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 70 eV 15 eV 12 eV 10 eV Mass-to-charge (m/z) 12 Tentative molecular ions identified using LE-EI were selected as precursors in MS/MS experiments (Figure 8) to further perform structure elucidation. The target MS/MS was performed by alternating the MS/MS and full spectrum acquisition modes. The accurate m/z of the precursors were entered in the Acquisition software to facilitate correct recognition of the m/z of the molecular ion in the downstream data processing. The collision energy (CE) was optimized for each compound to yield optimal fragmentation, preserving an abundance of high- and mid-range m/z ions in the spectrum, when possible. The structure elucidation was carried out in the MSC software. The molecular formulas were automatically assigned based on the accurate mass ions from the full spectrum data that matched the m/z of the precursor at the same RT. All possible structures for each tentative molecular formula were extracted from the ChemSpider database and evaluated based on fragmentation patterns. A proposed structure for one of the unknowns is shown in Figure 9. This structure could potentially correspond to a degradation product of an antioxidant. Figure 8. The MS/MS spectrum of one of the unknowns, using a tentative molecular ion as a precursor. 190.1354 175.1117 161.0958 250.1927 55.0550 77.0390 91.0541 105.0698 119.0846 135.0801 207.0649 222.9368 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 CE = 10 V Counts Mass-to-charge (m/z) Figure 9. A proposed structure for one of the unknown compounds in rubber gasket extract using MSC. www.agilent.com DE-000968 This information is subject to change without notice. © Agilent Technologies, Inc. 2024 Printed in the USA, September 10, 2024 5994-7777EN Conclusion The GC/MSD is an effective and accessible tool for the analysis of volatile and semivolatile compounds in complex E&L extracts. The established workflow includes deconvolution and an RI-based library search with the data acquisition performed in a compliant environment. The high-resolution Agilent 7250 GC/Q-TOF enabled the identification of additional components with increased confidence, as well as structure elucidation of the unknown compounds. Furthermore, using the novel ultra-low bleed Agilent J&W DB-5Q GC column resulted in a significant decrease in background, which helps in the identification of late‑eluting compounds. References 1. Zhang, F.; Chang, A.; Karaisz, K.; Feng, R.; Cai, J. Structural Identification of Extractables from Rubber Closures Used for Pre-filled Semisolid Drug Applicator by Chromatography, Mass spectrometry, and Organic Synthesis. J. Pharm. Biomed. Anal. 2004, 34, 841–849. DOI: 10.1016/j.jpba.2003.08.003 2. Baneshi, M.; Tonney-Gagne, J.; Halilu, F.; Pilavangan, K.; Abraham, B. S.; Prosser, A.; Marimuthu, N. K.; Kaliaperumal, R.; Britten, A. J.; Mkandawire, M. Unpacking Phthalates from Obscurity in the Environment. Molecules 2023, 29(1), 106. DOI: 10.3390/molecules29010106 3. Taylor, R.; Son, P. N. Encyclopedia of Chemical Technology. Interscience, New York. 1982, 20, pp. 337–365. 4. Bohrer, D.; Viana, C.; Barichello, M. M.; de Moura, J. F.; de Carvalho, L. M.; Nascimento, P. C. Presence of Polycyclic Aromatic Hydrocarbons in Rubber Packaging Materials and in Parenteral Formulations Stored in Bottles With Rubber Stoppers. JPEN J Parenter Enteral Nutr. 2016, 41(6), 1037–1044. DOI: 10.1177/0148607116633801. 5. Support for Title 21 CFR Part 11 and Annex 11 Compliance: Agilent OpenLab Server and OpenLab ECM XT. Agilent Technologies white paper, publication number 5994-7586EN, 2024. 6. How Does Bleed Impact GC/MS Data and How Can It Be Controlled? Agilent Technologies technical overview, publication number 5994-3228EN, 2024.