Open Access Article
Jeppe S. Mogensen
a,
Thomas M. Hofmann
b,
Suvenika Pererab,
Yann Liea,
Michael Pittelkow
*a and
Lyle Isaacs
*b
aDepartment of Chemistry, University of Copenhagen, Universitetsparken 5, DK-2100 Copenhagen Ø, Denmark. E-mail: pittel@chem.ku.dk
bDepartment of Chemistry and Biochemistry, University of Maryland, College Park, 8051 Regents Dr, College Park, MD 20742, USA. E-mail: LIsaacs@umd.edu
First published on 21st May 2026
Acyclic CB[n]-type host H4 functions as a solid state sequestrant for pharmaceutical micropollutants (PMs) from water, reaching up to 99.8% removal efficiency for acetaminophen within 5 minutes. A head-to-head study shows H4 performs at least as well as activated charcoal (DARCO®) at sequestering acetaminophen from simulated gastric fluid.
000 emergency room visits, 2600 hospitalizations, and 500 deaths a year.8 Additionally, acetaminophen is metabolized by hepatic cytochrome P450's into N-acetyl-p-benzoquinoneimine (NAPQI) which depletes hepatic glutathione and increases NAPQI-protein adducts leading to significant hepatic injury.8 Acetaminophen toxicity accounts for 50% of acute liver failure and 20% of liver transplants in the United States.8 Activated charcoal is currently used as the primary adsorbent in emergency overdose treatment and waste water treatment plants.9–11 N-Acetyl cysteine – which increases glutathione synthesis – is FDA approved as a specific antidote for acetaminophen toxicity.
Over the years, a variety of methods to remove micropollutants from water have been reported including flocculation, coagulation, filtration, reverse osmosis, photodegradation, chemical oxidation, and adsorption.12 The pioneering work of Dichtel on the use of β-cyclodextrin (CD) derived polymers13,14 for the removal of micropollutants from water stimulated supramolecular chemists to explore the use of other classes of macrocycles including calixarenes, calix[4]pyrroles, pillararenes, cavitands, and naphthotubes in this application.15–21 In this context, macrocyclic cucurbit[n]urils (CB[n]) are particularly attractive due to their high affinity binding, very low aqueous solubility (CB[6]: 6.5 µM, CB[8]: 10 µM), high biocompatibility, and accessibility from inexpensive starting materials.22–29 Over the past 15 years, the Isaacs group has synthesized a variety of water soluble acyclic CB[n]-type hosts, studied their molecular recognition properties, and explored their application as in vivo sequestrants for neuromuscular blockers, anesthetics, and drugs of abuse.30–33 Along the way, we synthesized numerous water insoluble variants including the catechol walled glycoluril tetramer derived host H4 (water solubility: 3.4 µM, Fig. 1a) which acts as a solid state sequestrant for endocrine disrupters (bisphenol A, bisphenol S), dyes (methyl violet), perfluoroalkyl substances, and iodine.25,34 A limitation of H4 in these applications is that a large molar excess is required to achieve >90% removal efficiencies (RE). In this work, we present the use of H4 as a solid state sequestrant for pharmaceutical micropollutants from water at both low (250 µM) and high PM concentration (≤90 mM) to simulate water body contamination and overdose scenarios.
![]() | ||
| Fig. 1 Chemical structure of: (a) acyclic CB[n]-type receptor H4 and (b) five compounds (PMs and phenol) used in this study. | ||
Initially, we investigated the solid state sequestration efficiency of H4 toward phenol and the panel of PMs (Fig. 1b) which are prevalent in water bodies.6,11 Experimentally, we incubated a fixed amount of H4 (2.0 mg) for 2 hours at 25 °C using a ThermoMixerTM with different volumes of each PM (0.25 µM) to create different H4:PM molar ratios. Following centrifugation, the concentration of each PM in the supernatant was determined by UV/vis spectroscopy and the removal efficiency (RE) values were calculated as described previously (SI, eqn (S1)).25 Fig. 2a shows the UV/vis spectra collected for acetaminophen (250 µM) alone and after sequestration with solid H4 (0.5, 1, 2, 4, 8 equivalents). Fig. 2b shows that H4 functioned as a solid-state sequestrant for all five compounds with RE values ranging from 35% (ibuprofen) to 99.8% (acetaminophen) at an 8
:
1 H4
:
PM molar ratio. Significantly, H4 functioned as a very effective solid state sequestrant for acetaminophen (RE: 95%) even at a 2
:
1 H4
:
acetaminophen molar ratio. Conversely, H4 functioned poorly for ibuprofen sodium (RE = 35% at 8
:
1 molar ratio); this is likely due to unfavorable electrostatic interactions between its anionic carboxylate and the electrostatically negative C
O portals of H4.35 H4 also functioned poorly for citalopram with 39% RE, presumably due to its steric bulk. Conversely, H4 functioned moderately well as a sequestrant for propranolol (RE: 78%) and phenol (RE: 82%) at an 8
:
1 H4
:
PM molar ratio. Interestingly, the RE for propranolol is lower than would be expected based on its structure due to the fact that propranolol increases the solubility of H4 by forming the H4·propranolol complex which can be detected by 1H NMR spectroscopy of the supernatant (SI, Fig. S14). In addition to achieving a high RE value, it is important that a solid sequestrant is recyclable and exhibit rapid adsorption. Previously, we showed that H4 could be recycled and reused for the sequestration of organic micropollutants over five cycles.25 As part of this study, we confirmed that H4 can be recycled by washing with methanol and water and maintains its high RE values for acetaminophen removal over three cycles (SI, Fig. S34). We also performed kinetic studies varying the incubation time of acetaminophen (250 µM) with H4 at an 4
:
1 H4
:
acetaminophen molar ratio (SI, Fig. S10). Under these conditions, H4 achieved a RE value of 93% after only 60 seconds showcasing H4 as a fast acting and effective adsorbent for acetaminophen. For comparison, Dichtel's porous CD polymer exhibits rapid uptake of a variety of micropollutants (seconds to minutes) whereas Sessler's porous calix[4]pyrrole polymer requires 10 minutes to reach 90% RE for methylene blue.13,15 We previously measured the BET surface area of H4 as 25 m2 g−1 which is substantially lower than Dichtel's CD polymer (263 m2 g−1).25 The excellent removal efficiency of H4 for acetaminophen despite its low BET surface area strongly suggests that acetaminophen is included in the cavity of solid H4.
Next, we performed additional experiments to uncover some of the reasons behind the outstanding removal efficiency of H4 toward acetaminophen. First, we used 1H NMR spectroscopy to confirm the high removal efficiency determined by UV/vis spectroscopy. Experimentally, we performed the solid-state sequestration of acetaminophen (0.25 mM or 90 mM) in D2O using 2 equivalents of H4. 1H NMR spectra recorded for the supernatants (SI, Fig. S12 and S13) did not show any resonances for acetaminophen confirming the high RE values determined by UV/vis spectroscopy. Next, high resolution mass spectrometry of the supernatant (SI, Fig. S17) was performed, which revealed the presence of ions corresponding to [H4·acetaminophen + H]+ at m/z 1172.46584, [H4 + Na]+ at m/z 1043.38498, [H4·acetaminophen + 2Na]2+ and [H4 + 2Na]2+ at m/z 533.18716 with ≤ 6 ppm accuracy which confirms the affinity of H4 for acetaminophen in water.
Given the water insolubility of H4 and the H4·acetaminophen complex, we attempted to perform a 1H NMR titration in CF3CO2H (TFA) which is one of the only solvents capable of dissolving H4. Unfortunately, we observed the formation of a new AA’BB’ pattern in the aromatic region of the 1H NMR spectrum; GC-MS established that acetaminophen undergoes trifluoroacetylation under these conditions (SI, Fig. S15).36 We also performed solid state magic angle spinning 13C NMR spectroscopy of the solid obtained upon sequestration of acetaminophen from water using H4 (SI, Fig. S16). The spectra clearly demonstrate the presence of both H4 and acetaminophen in the solid and display small changes in chemical shift relative to uncomplexed acetaminophen. However, we could not use this data to elucidate the geometry of interaction between H4 and acetaminophen in the solid.
Attempts to grow crystals of H4·acetaminophen from TFA by slow evaporation and vapor diffusion methods (antisolvents: MeOH, EtOH, i-PrOH and acetone) were not successful. Therefore, a different crystallization protocol was employed (Fig. 3a). First, H4 was crystalized from TFA by vapor diffusion of i-PrOH. Then, TFA was removed and the crystals were washed with EtOH. Next, the crystals were soaked in a solution of acetaminophen (550 mM) in EtOH which provided crystals suitable for single crystal X-ray diffraction. Unfortunately, the disorder within the cavity of the H4 was too high for accurate structural elucidation of the complex. Therefore, the crystals were crushed and soaked in an ethanolic solution of acetaminophen (550 mM). The structure of the microcrystalline H4·acetaminophen complex was solved by electron diffraction (Fig. 3b).37–39 Acetaminophen is held inside the cavity of H4 by a combination of three H-bonds (a, b, c) and π–π interactions. The three H-bonds possess the following H⋯O distances (a: 2.03, b: 2.01, c: 1.76 Å), O⋯O(N) distances (a: 2.98, b: 3.02, c: 2.71 Å), and (N)O–H⋯O angles (a: 158.95, b: 163.16, c: 156.95°). A fourth H-bond (d) is formed between the tips of the aromatic sidewalls (H⋯O: 1.97 Å, O⋯O: 2.76 Å, O–H⋯O angle: 134.00°). The acetaminophen guest also engages in π−π interactions with one of the aromatic sidewalls. The distance of the phenylene C-atoms of acetaminophen from the mean plane of the C-atoms of the aromatic sidewall average 3.461 Å which is in agreement with the accepted π−π stacking distance of 3.4 Å.40 Fig. 3c shows the packing of four equivalents of the H4·acetaminophen complex roughly along the y-axis in the solid. The convex face of one equivalent of H4·acetaminophen engages in head-to-tail type offset π−π interactions with the adjacent equivalent of H4·acetaminophen. The CH2-bridges of the o-xylylene sidewalls of one molecule of H4 are located above the opposing aromatic sidewall of the adjacent H4. The distance of the C-atoms of one aromatic sidewall to the mean plane of the opposing sidewall averages 3.388 Å.
Given the excellent RE values achieved by H4 toward low [acetaminophen] (250 µM) relevant for environmental pollution of water bodies, we wondered whether H4 would perform as well at high [acetaminophen] (≤90 mM) more relevant in an overdose situation. Fig. 4a and b show the RE values achieved for a series of different acetaminophen concentrations (10, 30, 50, 70, 90 mM) at 1
:
2 and 2
:
1 H4
:
acetaminophen molar ratios upon incubation at 37 °C for 5 minutes in H2O. At a 2
:
1 H4
:
acetaminophen ratio, the RE values are very high whereas at a 1
:
2 ratio roughly half the acetaminophen is removed which provides additional support for our conclusion that acetaminophen is captured inside the cavity of H4. As a comparator, we selected DARCO® activated charcoal (mesh size 12–20) and used equal masses relative to H4 to sequester acetaminophen under identical conditions (37 °C, H2O, 5 min) and the results are reported in Fig. 4a and b. These results show that H4 performs as well as activated charcoal which is used clinically as an antidote for acetaminophen poisoning. To more fully mimic an oral acetaminophen overdose situation, we performed related sequestration experiments at 37 °C using simulated gastric fluid (SGF, Biochemazone BZ175) as the medium. SGF contains NaCl (200 mM) and pepsin (3.2 g L−1) and has pH 1.5 which we thought might interfere with the sequestrant abilities of H4 but not activated charcoal. Fig. 4c and d shows the RE values determined during the sequestration of acetaminophen (10, 30, 50, 70, 81 mM) from SGF at 37 °C by H4 or an equal mass of DARCO® activated charcoal. We were delighted to find H4 performs slightly better than DARCO® activated charcoal under identical conditions. A control experiment performed using SGF lacking pepsin showed slightly higher RE values for both H4 and DARCO® which indicates that pepsin inhibits removal of acetaminophen. In addition, we monitored the removal efficiency of acetaminophen from SGF using H4 and DARCO® activated charcoal as a function of time (SI, Fig. S35 and S36) and find they both exhibit rapid uptake within 60 seconds. As another control experiment, we tested the ability of CB[6] and CB[8] as sequestrants for acetaminophen from SGF at 37 °C at 1
:
2 and 2
:
1 CB[n]
:
acetaminophen molar ratio (SI, Fig. S27–S32). We observed that CB[6] is a very poor sequestrant for acetaminophen whereas CB[8] performs well (RE = 84% at 2
:
1 molar ratio) but is less efficient than H4.
We have shown that H4 is an effective solid state sequestrant for a panel of PMs at environmentally relevant levels (250 µM). The uptake of acetaminophen by H4 is especially impressive reaching a RE of 93% at a 4
:
1 molar ratio after 60 seconds and 99.8% after 2 h. At high acetaminophen concentrations (≤90 mM) relevant in overdose cases, H4 performs as well as activated charcoal and outperformed macrocyclic CB[6] and CB[8]. In conclusion, the work suggests that water insoluble hosts (e.g. H4 and its analogues) hold great promise as solid state sequestrants for detoxification and clean-up of water bodies.
Conceptualization: S. P. and L. I.; funding acquisition, project administration, and supervision: L. I., M. P.; investigation: J. S. M., T. M. H., S. P., Y. L.; writing – original draft: J. S. M. and T. M. H.; writing – review and editing: all authors.
CCDC 2539579 contains the supplementary crystallographic data for this paper.41
| This journal is © The Royal Society of Chemistry 2026 |