Compound characterization data — Bequette et al., 2026

Joseph P. Bequette, Jiashan Mi, Daisy Otundo, Aaron J. Rossini. “Magnetic levitation–based isolation of active pharmaceutical ingredients for solid-state NMR spectroscopy analysis.” ChemRxiv (2026). https://doi.org/10.26434/chemrxiv.15007542/v1

Supporting information is appended to the main paper

Diphenhydramine hydrochloride recovered from Banophen
SEM, average crystal size	>100 μm
¹H–¹H spin-diffusion NMR (400.3 MHz, 50 kHz MAS)	Cross-peaks among broad resonances at approximately 3, 7, and 9 ppm; the lactose autocorrelation at approximately 4 ppm was absent

Diphenhydramine hydrochloride recovered from Benadryl
SEM, average crystal size	<100 μm
¹H NMR (400 MHz, d₄-MeOH)	Minor impurities were evident in the baseline between approximately 1–2 ppm; the authors attributed them to a small amount of magnesium stearate
¹H–¹H spin-diffusion NMR (400.3 MHz, 50 kHz MAS)	Cross-peaks among broad resonances at approximately 3, 7, and 9 ppm
¹H and ¹³C solid-state NMR	Cellulose signals were absent

Diphenhydramine hydrochloride isolated by benchtop MagLev
Experimental ¹H solid-state NMR δiso assigned from peak-maximum intensities (ppm)	site 2 8.5–6.7; site 3 8.5–6.7; site 4 8.5–6.7; site 5 6.1; site 6 3.2; site 7 2.0; site 8 2.5; N–H 10.9
Experimental ¹³C solid-state NMR δiso assigned from peak-maximum intensities (ppm)	site 1 145.8; site 2 129–127; site 3 129–127; site 4 129–127; site 5 83.7; site 6 60.9; site 7 57.6; site 8 43.0

Banophen capsule contents
¹H–¹H spin-diffusion NMR (400.3 MHz, 50 kHz MAS)	Cross-peaks among broad resonances at approximately 3, 7, and 9 ppm; lactose at approximately 4 ppm showed only an autocorrelation
¹H spin-echo NMR (400.3 MHz, 50 kHz MAS), recycle delays	2, 20, 100, and 1200 s; the relative intensity of the lactose resonance at approximately 4 ppm increased with recycle delay

Benadryl tablet
¹H–¹H spin-diffusion NMR (400.3 MHz, 50 kHz MAS)	Only limited correlations were observed; diphenhydramine hydrochloride signals at 10 wt.% were not clearly resolved

Acetaminophen recovered from Excedrin in Gd(acac)3·TOPO solution
¹H–¹H spin-diffusion NMR (400.3 MHz, 50 kHz MAS)	Cross-peaks at approximately 6.7↔1.3 and 6.0↔1.2 ppm; the off-diagonal direct-dimension peak at approximately 9 ppm lacked a mirrored indirect-dimension cross-peak

Acetaminophen recovered from Excedrin in aqueous MnCl2
¹H–¹H spin-diffusion NMR (400.3 MHz, 50 kHz MAS)	Cross-peaks at approximately 6.7↔1.3 and 6.0↔1.2 ppm; the off-diagonal direct-dimension peak at approximately 9 ppm lacked a mirrored indirect-dimension cross-peak

Prepared acetaminophen Form II
FTIR-ATR	The approximately 806 cm⁻¹ feature was absent; a band was present near 770 cm⁻¹

Acetaminophen isolated from Excedrin
Experimental ¹H solid-state NMR δiso assigned from peak-maximum intensities (ppm)	site 1 1.2; site 3 6.7; site 4 6.0; Ph–OH 7.9; N–H 9.2
Experimental ¹³C solid-state NMR δiso assigned from peak-maximum intensities (ppm)	site 1 23.1; site 2 132.5; site 3 120.2; site 4 115.8; site 5 151.9; site 6 169.5

Caffeine recovered from Excedrin in Gd(acac)3·TOPO solution
¹H–¹H spin-diffusion NMR (400.3 MHz, 50 kHz MAS)	Cross-peak at approximately 7↔4 ppm
¹³C solid-state NMR, residual signal	Signal intensity at approximately 75 ppm indicated partial co-isolation of microcrystalline cellulose

Caffeine isolated from Excedrin
Experimental ¹H solid-state NMR δiso assigned from peak-maximum intensities (ppm)	site 1 3.1; site 2 3.5; site 3 4.0; site 6 7.6
Experimental ¹³C solid-state NMR δiso assigned from peak-maximum intensities (ppm)	site 1 28–35; site 2 28–35; site 3 28–35; site 4 105.7; site 5 149.9; site 6 142.4; site 7 149.9; site 8 149.9

Aspirin recovered from Excedrin in aqueous MnCl2
¹H–¹H spin-diffusion NMR (400.3 MHz, 50 kHz MAS)	Cross-peaks at approximately 1.5↔7.5 and 12.5↔7.5 ppm; out-of-phase correlation at approximately 12.5↔1.5 ppm
¹H{¹³C} CP-HETCOR NMR (400.3 and 100.7 MHz, 50 kHz MAS)	A weak cross-peak indicated minor acetaminophen carryover
¹³C solid-state NMR, residual signal	Signal intensity at approximately 75 ppm indicated partial co-isolation of microcrystalline cellulose

Aspirin isolated from Excedrin
Experimental ¹H solid-state NMR δiso assigned from peak-maximum intensities (ppm)	site 1 1.4; site 3 7.1; site 4 7.1; site 5 7.1; site 6 7.1
Experimental ¹³C solid-state NMR δiso assigned from peak-maximum intensities (ppm)	site 1 19.4; site 2 152.4; site 3 126.1; site 4 137.9; site 5 126.1; site 6 133.7; site 7 121.9; site 8 171.6; site 9 171.6

Active pharmaceutical ingredients recovered from Excedrin
SEM, average particle size	Approaching 100 μm; caffeine crystals were smaller than acetaminophen and aspirin crystals

Ground Excedrin tablet
¹H–¹H spin-diffusion NMR (400.3 MHz, 50 kHz MAS)	Cross-peaks at approximately 13↔8 and 7↔2 ppm
¹³C solid-state NMR, broad resonance	Approximately 75 ppm, assigned to microcrystalline cellulose

Microcrystalline cellulose reference
¹H spin-echo NMR (400.3 MHz, 50 kHz MAS)	Broad resonance centered at approximately 4 ppm
¹³C projection from ¹H{¹³C} CP-HETCOR NMR (100.7 MHz, 50 kHz MAS)	Highlighted region at 70–80 ppm; the authors assign the overlapping resonances to C2, C3, and C5

Synthesized Gd(acac)3·TOPO
Evans-method ¹H NMR (9.4 T, chloroform), Δf; F; c; calculated χc	847 Hz; 400 × 10⁶ Hz; 26.36 mol m⁻³; 1.92 × 10⁻⁸ m³ mol⁻¹