After conditioning the cigarettes according to ISO 3402 (1999) specifications, physical property determinations were performed using a C2 instrument (Cerulean, Milton Keynes, UK). Five replicate measurements were made on each cigarette variety to determine cigarette length, pressure drop, filter ventilation, filter circumference, rod circumference, filter length, tobacco length, and over wrap length. Air permeability was determined using a PPM1000M instrument (Cerulean, Milton Keynes, UK).

Tobacco filler and cigarette weights were determined manually with calibrated and certified balances.

All quantitative methods were performed in a strict QA/QC environment. Each analytical method below has been full validated and has sufficient dynamic range for all samples. The limits of detections for each method was obtained using Taylor’s method.⁸ All reported data fall between the lowest and highest calibration curve point. Any measured values lower than the lowest calibration point are reported as non-detects. All run QC were checked using a modified Westgard protocol.⁹ Any QC failures were excluded and samples repeated if needed. All data were reviewed by a quality control officer and quality manager prior to release.

Previously unopened packages of cigarettes were used for menthol analysis to minimize substantial evaporative losses. Sample preparation procedures were described previously.¹⁰ Tobacco menthol levels were determine analytically using an Agilent 7890 GC coupled with a 5975 MSD (Agilent Technologies; Newark, DE, USA). The chromatographic separation was achieved using an Ultra-2 capillary column (25m × 0.32mm × 0.52µM, Agilent Technologies; Andover, MA, USA) with research grade helium carrier gas as previously described.¹⁰ Results are the mean and standard deviations of five replicate analyses.

Tobacco filler from individual cigarettes was removed from the paper wrapper and ground to uniformity with a Cuisinart® Spice and Nut grinder. For each cigarette type, five 0.25 g aliquots of the ground filler were measured into 15 mL amber vials and spiked with 100 µL internal standard solution, representing 100 ng ¹³C labeled standard for each analyte. Ten mL of 100 mM aqueous ammonium acetate solution was added to each sample and shaken for 60 minutes at 250 rpm on a Benchmark IncuShaker 10 L (Benchmark Scientific, Edison, NJ, USA) Approximately 0.5 mL of the extract was pipetted into a Thompson Filter Vial and press-filtered at 0.45 µm.

Analytical separation was performed on an Agilent 1200 G1312B binary pump. (Agilent Technologies, Santa Clara, CA) using a 5µm particle size Xterra MS C18 4.6 × 50mm HPLC column (Waters Corporation, Milford, MA, USA) with mobile phase pumped at 1 mL/min at 60°C. Elution began with 95% solution A (5mM aqueous ammonium acetate), 5% solution B (95% acetonitrile, 5% water with 5mM ammonium acetate) mobile phase for 1 min, followed by a 1 min gradient to 65% solution A, 35% solution B. The mobile phase was maintained at 65% solution A, 35% solution B for 3 min, followed by a 2 min gradient to the original mobile phase. Analytes were quantitated with an API 4000 triple quadrupole mass spectrometer (Sciex Corporation, Farmington MA, USA) working in positive electrospray ionization mode as previously described.¹¹ Tandem mass spectrometry data was evaluated and quantitated using Sciex Analyst 1.5 software.

Analyses of alkaloids were performed using an Agilent 7890 GC coupled with a 7000 Triple-Quad detector (Agilent Technologies, Newark, DE, USA). The chromatographic separation was accomplished using a DB-1701 capillary column (30m × 0.250 µM, 0.25 µM, J&W Scientific) with research grade helium used as the carrier gas as previously described.¹² Results are the mean and standard deviations of five replicate analyses (N = 5).

Tobacco filler from individual cigarettes were weighed, spiked with an isotopically labeled13C-PAH mixture as an internal standard, extracted with 10 ml cyclohexane followed by evaporation to 1 ml. Extracts were filtered through a 0.45 µM Nylon syringe filter. The PAHs were subsequently analyzed by SIM GC/MS, normalized to tobacco weight and expressed in nanograms of analyte per gram tobacco. Analytes measured included acenaphthene (ACE), acenaphthylene (ACL), anthracene (ANT), benzo[a]pyrene (BAP), benzo[e]pyrene (BEP), chrysene (CHR), fluoranthene (FLR), fluorene (FLU), naphthalene (NAP), phenanthrene (PHE), pyrene (PYR), benz[a]anthracene (BAA), benzo[b]fluoranthene (BBF), and benzo[k]fluoranthene (BKF). Lowest reportable levels were determined as the concentration of the lowest standards for each analyte. Five replicates were analyzed for each variety.

Tobacco filler from individual cigarettes was removed from the paper wrapper, 5 mL of 2N NaOH was added to 1.0 grams of tobacco filler in a 60 mL amber vial, followed by extraction with 50 mL of an MTBE stock that contained quinoline as an internal standard. A 1 mL aliquot of the extract was placed in a 2 mL amber vial for GC/MS analysis. The sample preparation and nicotine analytical method have been previously described.¹³ Five replicates were analyzed for each variety.

Tobacco filler from individual cigarettes was removed from the paper wrapper, placed into a 60-mL amber bottle, and weighed. Deionized water (30 mL) was added and the sample was sealed with a PTFE-lined cap. The sample was shaken for 30 min at 160 rpm on a Max Q 2000 shaker (Barnstead/Labline, Dubuque, IA, USA). A 1.5mL sample aliquot was centrifuged in a Sorval Pico Biofuge (Thermo Fisher, Waltham, MA, USA) at 13,000 rpm for 5 min to remove tobacco debris. A 10:1 water to sample dilution was made in a 1.5 mL sample vial. The sample was vortexed and placed in a sample tray for analysis. Five replicates were analyzed for each variety.

An ISO Guide 34 1000 mg/L ammonium standard was purchased from Sigma Aldrich (St. Louis, MO, USA). Ultrapure water used in all procedures was from an Aqua Solutions water system (Aqua Solutions, Jasper, GA, USA).

Samples (25 µL injections) were analyzed with a Dionex ICS-3000 chromatography system with Chromeleon version 6.8 software and a 4 mm Ion Pac CS12A cation exchange column (Thermo Scientific, Sunnyvale, CA, USA). Isocratic separation of the ammonium ion was achieved using a 20 mM methanesulfonic acid eluent prepared by the EG40 eluent generator with cation self-regenerating suppressor that required only an ultrapure water source in order to suppress the conductance from the eluent (Thermo Scientific, Sunnyvale, CA USA). The column temperature was maintained at 25°C with a run time of 15 minutes. Ammonium ions were detected using a conductivity detector.

Tobacco filler from 4 individual cigarettes of each variety was removed from the paper wrapper and combined for analysis. The tobacco was dried for 3 hours at 90°C and rendered more homogeneous by grinding for 20 seconds with a Smart Grind coffee grinder (Black and Decker, Middleton, WI, USA). Samples (0.100 to 0.150 g) were digested with a Milestone Ethos microwave system and analyzed with “Triple Quad” Inductively Coupled Plasma Mass Spectrometry as previously described.¹⁴ Dried tobacco was analyzed for mercury using a combustion mercury analyzer as previously described¹⁵ with minor modification.¹⁶ Quality control was maintained throughout using Reference tobacco 1S3 (North Carolina State University, Raleigh, NC, USA) and certified reference material (CRM) CTA-OTL-1 (Instytucie Chemii i Techniki Jadrowej, Warszawa, Poland) as previously described.^14,16 Results are reported as the mean and standard deviations of 5 replicate analyses.

Based on physical properties, the cigarette paper seemed consistent for all SPECTRUM varieties. Cigarette diameter, circumference, tip length (Table 1), and paper porosity differed negligibly from variety to variety. The average paper porosity for all SPECTRUM cigarette varieties was 31.5 ± 3.7 CORESTA units. All measured physical properties are in Tables 1 and 2.

Menthol concentrations varied greatly among the SPECTRUM cigarettes. Twelve mentholated products (NRC101, NRC103, NRC105, NRC201, NRC301, NRC302, NRC401, NRC405, NRC501, NRC601, NRC602, and NRC701) had menthol concentrations ranging from 4970 – 7540 µg menthol/cigarette. Non-mentholated products had measured menthol concentrations ranging from 1.1 – 114.0 µg/cigarette. Both of these ranges were consistent with menthol concentrations recently reported for commercial menthol and non-menthol cigarette brands sold in the United States.¹⁷ Results are reported in Table 3.

Results for NNN (N-nitrosonornicotine) and NNK (4-(N-nitrosomethylamino)-1-(3-pyridyl)-1-butanone) were reported in Table 3. NNN concentrations ranged from 963 ng/g to 1732 ng/g, NNK values were 45.2 ng/g to 517 ng/g. As expected, concentrations of NNN in the tobacco filler were higher than that of NNK. These results place the NNN levels of SPECTRUM cigarettes within the lower third of the concentration range reported for commercial cigarette products.¹⁸ The majority of SPECTRUM NNK values were lower than the lowest NNK level (330 ng/g) published by Stepanov et al.¹⁸

Although PAHs are primarily combustion products, difference in tobacco processing, ie, curing practices, can result in PAHs being present in tobacco filler. The results for analyses of PAHs in the tobacco from SPECTRUM cigarettes are reported in Table 4. Of the 14 PAHs analyzed, 8 PAH compounds were present at detectable levels ranging from 0.57 – 349 ng/g. No detectable levels of BAP, ACL, FLU, ANT, BAA, or NAP were found in the SPECTRUM cigarettes. The highest PAH concentrations were found for CHR (193 – 349 ng/g) and PHE (32.7 – 59.7ng/g), which were detected in all of the products analyzed. The compounds PYR (8.47 – 16.6 ng/g) and FLR (7.78 – 11.8 ng/g) had lower levels but were detectable in all of the varieties tested. Little data on PAH concentrations in cigarette filler tobacco has been published since fire-cured tobacco is not commonly used for filler. However, phenanthrene and chrysene, concentrations in all Spectrum variety tobacco fillers were higher than the respective concentrations determined in 3R4F (used as a quality control, 26.3 ng/g and 3.7 ng/g, respectively), whereas fluoranthene, benzo[e]pyrene and benzo[a]pyrene concentrations were lower in all Spectrum variety tobacco fillers than in 3R4F (15.7 ng/g, 2.0 ng/g, and 1.3 ng/g, respectively).

The results of analyses for nicotine and minor alkaloids (nornicotine, myosmine, anabasine, anatabine, and isonicotine are reported in Table 5. Nicotine concentrations ranged from very low to within the range reported for commercial products (0.28 – 25.0 mg/g). The majority of products had nicotine concentrations that were well below the levels of comparable commercial cigarette products,¹⁹ which typically range from 8 – 22 mg/g, indicating a selective reduction of nicotine. Overall concentrations of other alkaloids were higher than those found in commercial products¹² and generally increased with measured nicotine concentration for each SPECTRUM product type. Alkaloid concentration ranges for the SPECTRUM varieties were 472 – 1235 µg/g for nornicotine, 14.8 – 28.7 µg/g for myosmine, 5.9 – 242.4 µg/g for anabasine, 29.7 – 2124 µg/g for anatabine and 3.2 – 64.9 µg/g for isonicotine.

Ammonia levels in the tobacco filler differed little between the cigarette varieties, ranging from 2.2 to 3.7 mg ammonia per gram of conditioned tobacco filler (Table 6). Within-variety replicate measurements showed little variation with relative standard deviations ranging from 1.6 to 5.6 %. This range is in line with previously reported ammonia concentrations in tobacco filler.²⁰

The results for 11 inorganic harmful and potentially harmful constituents of SPECTRUM cigarette tobacco filler showed that concentration ranges for beryllium, nickel, and arsenic in filler tobacco from SPECTRUM cigarettes (Table 6) were lower than the respective ranges of concentrations in filler tobacco previously reported for 50 varieties of cigarettes commercially available in the United States.¹⁵ Chromium, cobalt, and cadmium concentrations in SPECTRUM filler tobacco (Table 6) were below, or corresponded with the lower range of chromium, cobalt, and cadmium concentrations previously reported.¹⁵ Manganese and lead concentrations in SPECTRUM cigarette filler tobacco fell predominantly within the ranges of manganese and lead concentrations previously reported.¹⁵