# Evidence: drug-polymer hydrogen bonding

Parent: drug-polymer interactions · class: molecular interaction · names merged here: drug-polymer hydrogen bonding

9 papers measured or used it, 1 only mentioned it (of 207 kept papers). Papers in year order; each section heading ends with the paper's category / subcategory. Generated by tools/build_evidence.py from registry/concepts.csv and the cards; do not edit.

## 2001 Broman — A comparison of alternative polymer excipients and processing methods for making solid dispersions of a poorly water soluble drug · ASD process / ASD process

[2001-broman-polymer-excipients-processing-probucol-solid-dispersions](../../../cards/2001-broman-polymer-excipients-processing-probucol-solid-dispersions/paper.html) · ASD process / ASD process · measured · as: drug-polymer hydrogen bonding

| id | kind | content | locator |
|---|---|---|---|
| R5 | record, measured | PVP carbonyl Raman band — PVP K30 alone; probucol–PVP by solvent method (P4) and by compression moulding (P3) — text: band broadened and shifted to lower wavenumber in both dispersions, the shift smaller for the compression-moulded sample than for the solvent sample; figure: maxima ≈1673–1674 cm⁻¹ (PVP K30 and compression moulding) and ≈1670 cm⁻¹ (solvent); low-wavenumber half-height ≈1664 (PVP K30), ≈1657 (compression moulding), ≈1651 cm⁻¹ (solvent); high-wavenumber flanks overlap (read from Fig. 4) | Fig. 4, p.146; text p.143-144 |
| R16 | record, measured | drug–PEO hydrogen bonding — probucol–PEO dispersions — no evidence of hydrogen bond interactions found (spectra and method not shown) | p.149 |
| C15 | claim, mechanism, authors: "is consistent with"; "supports" | The PVP carbonyl shift fits a hydrogen bond between the probucol hydroxyl and the PVP carbonyl and supports a one-phase amorphous system [rests on: R5; cited: Taylor & Zografi 1998] | p.143-144 |
| C31 | claim, prior hypothesis, authors: "it has been suggested" | PVP's inhibition of crystallisation from the amorphous state has been linked to raising the Tg and forming hydrogen-bonded complexes [rests on: cited: Yoshioka et al. 1995; Taylor & Zografi 1997] | p.147 |

## 2009 Rumondor — Phase Behavior of Poly(vinylpyrrolidone) Containing Amorphous Solid Dispersions in the Presence of Moisture · Solid state stability / Physical stability

[2009-rumondor-pvp-asd-moisture-induced-immiscibility](../../../cards/2009-rumondor-pvp-asd-moisture-induced-immiscibility/paper.html) · Solid state stability / Physical stability · measured · as: drug-polymer hydrogen bonding

| id | kind | content | locator |
|---|---|---|---|
| R4 | record, measured | IR band positions and assignments (cm⁻¹; sh shoulder, w weak) — PVP, amorphous and crystalline drugs and dry dispersions (P1-P3); indomethacin assignments based on ref 19, ketoprofen partly on ref 39 — sub-table below | Table 1, p.1496 |
| R29 | record, derived (difference of R4 band positions; relation named, not printed) | red shift of the PVP C=O band (1681 cm⁻¹, R4.a) when H-bonded to the drug — dry dispersions (P1, P2) — nifedipine 21 (R4.e), pimozide 20 (R4.i), droperidol 24 (R4.o), indomethacin 42 (R4.aa), ketoprofen 44 (R4.ai) cm⁻¹; text: 42-44 versus 20-24 cm⁻¹ | p.1504; Table 1, p.1496 |
| C47 | claim, mechanism, authors: asserted ("Thus", "This in turn leads to") | The susceptible drugs (nifedipine, felodipine, droperidol, pimozide) H-bond to the PVP carbonyl through an NH group, the others (ketoprofen, indomethacin) through COOH; the COOH hydrogen is a better donor (weaker O-H bond from electron sharing), giving stronger COOH-C=O(PVP) than NH-C=O(PVP) bonds [rests on: R4, R29] | p.1504 |
| C48 | claim, conclusion, authors: asserted ("Evidence for") | The larger red shift of the PVP carbonyl with indomethacin and ketoprofen (42-44 cm⁻¹) than with nifedipine, droperidol and pimozide (20-24 cm⁻¹) is evidence of stronger drug-polymer interactions [rests on: R29] | p.1504 |

## 2011 Eerdenbrugh — An ab initio polymer selection methodology to prevent crystallization in amorphous solid dispersions by application of crystal engineering principles · Solid state stability / Miscibility and drug-polymer interactions

[2011-van-eerdenbrugh-ab-initio-polymer-selection-asd-crystallization](../../../cards/2011-van-eerdenbrugh-ab-initio-polymer-selection-asd-crystallization/paper.html) · Solid state stability / Miscibility and drug-polymer interactions · used · as: drug-polymer hydrogen bonding

| id | kind | content | locator |
|---|---|---|---|
| C6 | claim, prior hypothesis, authors: "working hypothesis" | Working hypothesis: polymers inhibit crystallization of a given drug better when more favourable drug-polymer hydrogen bonds can be formed than those present in the crystalline drug; its groundwork is partly the known hydrogen-bond disruption by tailor-made additives [rests on: cited: refs 4-11] | p.6172 |
| C19 | claim, mechanism, authors: "it can be expected" | With very strong donors and medium acceptors, carboxylic acid self-association is expected to be disrupted best by polymers with strong acceptor groups that can compete with the drug acceptor groups [rests on: R1.a–g, R2] | p.6175 |
| C32 | claim, mechanism, authors: "would be expected" | Amide N–H groups are poorer donors than O–H groups and amide carbonyls better acceptors than carboxylic acids, so competitive hydrogen bonding would be expected from strong donors and/or extremely strong acceptors [rests on: R1.m–n] | p.6176 |
| C51 | claim, generalisation, authors: asserted ("clearly indicate") | The results clearly indicate that fundamental principles of hydrogen-bond-directed molecular recognition, adapted from crystal engineering, can be applied to design solids in which crystallization is prevented [rests on: R1, R2, R3, R14] | p.6177 |
| X13 | extractor note | claim-data gap — The acid–base reactions, proton transfers and the 4-biphenylmethanol hydrogen bonding invoked in the discussion rest on ATR-FTIR spectra of melts that are not shown; no drug–polymer hydrogen bond was measured in the spin-coated films — C22, C27, C38, C42, C44, R8, R12, R15, R16, R17 — treat the acid–base and hydrogen-bond explanations as resting on unshown evidence |  |

## 2014 Hsieh — Assessing the Impact of Polymers on the pH-Induced Precipitation Behavior of Poorly Water-Soluble Compounds using Synchrotron Wide Angle X-Ray Scattering · Dissolution / D. Crystallization from solution

[2014-hsieh-polymers-ph-induced-precipitation-waxs](../../../cards/2014-hsieh-polymers-ph-induced-precipitation-waxs/paper.html) · Dissolution / D. Crystallization from solution · used · as: drug-polymer hydrogen bonding

| id | kind | content | locator |
|---|---|---|---|
| C37 | claim, mechanism, authors: "may reflect"; "could" | This effectiveness may reflect hydrogen bonds between drug donors and the very good acceptors of PVP and PVPVA or the moderate acceptors of the cellulose derivatives. Alternatively, polymer donors (cellulose derivatives, Eudragit E100) could bond to drug acceptor groups. [rests on: R4; cited: ref 35] | Discussion, p.10 |
| C38 | claim, caveat, authors: "may not be favorable" | Drug–polymer hydrogen bonding may be unfavourable in water because water solvation competes. [rests on: cited: ref 36] | Discussion, p.10 |

## 2014 Trasi — Factors Influencing Crystal Growth Rates from Undercooled Liquids of Pharmaceutical Compounds · Solid state stability / Crystallization tendency

[2014-trasi-crystal-growth-rates-undercooled-pharmaceutical-liquids](../../../cards/2014-trasi-crystal-growth-rates-undercooled-pharmaceutical-liquids/paper.html) · Solid state stability / Crystallization tendency · used · as: drug-polymer hydrogen bonding

| id | kind | content | locator |
|---|---|---|---|
| C9 | claim, assumption, authors: "would be expected" | The five compounds of the PVP study have quite different bulk growth rates but all contain good hydrogen bond donor groups and would be expected to interact similarly with the hydrogen bond acceptor of PVP [rests on: argument only] | p.9977 |
| C33 | claim, prior hypothesis, authors: "are thought to be" | Specific interactions such as drug-polymer hydrogen bonding are thought to be important for the extent of growth inhibition, and above a certain concentration high-Tg polymers also raise viscosity and reduce growth by lowering mobility [rests on: cited: refs 5, 29] | p.9981 |

## 2015 Purohit — Phase Separation Kinetics in Amorphous Solid Dispersions Upon Exposure to Water · Characterization methods / Solid-state methods

[2015-purohit-phase-separation-kinetics-asd-water](../../../cards/2015-purohit-phase-separation-kinetics-asd-water/paper.html) · Characterization methods / Solid-state methods · measured · as: drug-polymer hydrogen bonding

| id | kind | content | locator |
|---|---|---|---|
| R17 | record, measured | carbonyl-region IR spectra — 80:20 RTV–PVP film on KRS-5 before and after 97% RH, dried before measurement (P10); amorphous RTV and PVP references; exposure time not stated — text: dry amorphous RTV carbonyl peaks at 1630 and 1717 cm-1; dry PVP carbonyl at 1681 cm-1; unexposed ASD shows a new peak at 1654 cm-1 (arrow); after exposure the 1654 cm-1 feature is absent and peaks at 1630 and 1717 cm-1 and at 1681 cm-1 are present. Figure: exposed ASD spectrum has maxima at ≈1678 and ≈1632 cm-1 with a shoulder near ≈1705 cm-1 and a minimum near ≈1652 cm-1 (read from Fig. 11) | Fig. 11, p.30; text p.14-15 |
| R18 | record, measured | carbonyl-region IR spectra — 50:50 RTV–HPMCAS film before and after 97% RH, dried before measurement (P10); amorphous RTV and HPMCAS references; exposure time not stated — text: new peak at 1733 cm-1 in the ASD; spectrum unchanged after exposure and drying. Figure: HPMCAS carbonyl maximum ≈1743 cm-1; amorphous RTV maxima ≈1717 and ≈1630 cm-1; ASD maximum ≈1733 cm-1 before and after, with a slightly lower after-trace between ≈1690 and ≈1600 cm-1 (read from Fig. 12) | Fig. 12, p.31; text p.15 |
| C17 | claim, conclusion, authors: "This is indicative of"; "correlates well" | In 80:20 RTV–PVP the new 1654 cm-1 band is a hydrogen bond between the PVP carbonyl and RTV donors; after 97% RH the interactions are disrupted by AAPS, cohesive interactions re-form (1630 and 1717 cm-1 for an RTV-rich phase, 1681 cm-1 for a PVP-rich phase); this indicates AAPS and correlates well with the fluorescence data; together with DSC it confirms initial miscibility [rests on: R17; abstract] | p.2, p.14-15 |
| C27 | claim, mechanism, authors: asserted; "consistent with" | Ritonavir has three hydrogen-bond-donating NH groups that bond to the PVP carbonyl (IR); as in other NH-containing systems, sorbed water disrupts the drug–polymer hydrogen bonds, producing a drug-rich phase and IR changes consistent with loss of drug–polymer H-bonding; AAPS of RTV–PVP had been expected from the chemistry of the two components but not reported [rests on: R17; cited: ref 17] | p.18 |

## 2020 Saboo — Congruent release of drug and polymer from amorphous solid dispersions: insights into the role of drug-polymer hydrogen bonding, surface crystallization and glass transition · Dissolution / A. Release mechanism and LoC

[2020-saboo-congruent-release-hbond-crystallization-tg-pvpva-asd](../../../cards/2020-saboo-congruent-release-hbond-crystallization-tg-pvpva-asd/paper.html) · Dissolution / A. Release mechanism and LoC · measured · as: drug-polymer hydrogen bonding

| id | kind | content | locator |
|---|---|---|---|
| R14 | record, measured | IR of IND:PVPVA ASD tablets versus DL — ASD standards 10:90 to 90:10 with amorphous IND and PVPVA (P11) — for 10-70% DL the 1705 cm⁻¹ band is absent and a band at 1634 cm⁻¹ (assigned in ref 22 to hydrogen-bonded PVP carbonyl) appears, its intensity rising from 10 to 70% DL; the 1591 cm⁻¹ band rises continuously with DL | Fig. 5B, p.18 |
| R16 | record, measured | IR of 25:75 tablet surface versus dissolution time — IND:PVPVA 25:75 at 0, 5, 10, 20 min (P11) — successive increases of the 1634 and 1591 cm⁻¹ bands from 0 to 5, 10 and 20 min; 1705 cm⁻¹ dimer band absent in all spectra | Fig. 6, p.19; text p.19 |
| R18 | record, measured | IR of INDester systems — PVPVA, crystalline INDester, amorphous INDester, INDester:PVPVA 25:75 tablet (Fig. S9); 25:75 tablet surface at 0, 5, 20 min (Fig. S10) (P11) — no change of the PVP carbonyl region in the 25:75 ASD; crystalline INDester has a band at 1607 cm⁻¹ that amorphous INDester lacks; no discernible difference between 0 and 5 min; band at 1607 cm⁻¹ present at 20 min | Figs. S9, S10, SI p.9; text p.20 |
| C20 | claim, conclusion, authors: "indicating" | Loss of the 1705 cm⁻¹ band and a new 1634 cm⁻¹ band growing with DL indicate an increasing number of IND-PVPVA hydrogen bonds; the 1591 cm⁻¹ band is a useful indicator of surface drug enrichment [rests on: R13, R14; cited: ref 22] | p.18 |
| C23 | claim, conclusion, authors: asserted; "consistent with" | INDester ASDs show no evidence of drug-polymer hydrogen bonding (PVP carbonyl region unchanged); the 1607 cm⁻¹ band after 20 min is consistent with drug crystallization, in good agreement with PXRD [rests on: R18, R9] | p.20 |
| C48 | claim, conclusion, authors: "clearly demonstrate that our hypothesis appears to be incorrect" | Short-time and low-temperature release data show that the hypothesis appears incorrect: IND ASDs have a much lower LoC than INDester ASDs, and hydrogen bonding in dry IND-PVPVA ASDs did not raise the LoC relative to the analog without hydrogen bonding [rests on: R3, R10] | p.32 |

## 2021 Yang — Drug Release and Nanodroplet Formation from Amorphous Solid Dispersions: Insight into the Roles of Drug Physicochemical Properties and Polymer Selection · Dissolution / A. Release mechanism and LoC

[2021-yang-asd-nanodroplet-release-drug-properties-polymer-selection](../../../cards/2021-yang-asd-nanodroplet-release-drug-properties-polymer-selection/paper.html) · Dissolution / A. Release mechanism and LoC · used · as: drug-polymer hydrogen bonding

| id | kind | content | locator |
|---|---|---|---|
| C8 | claim, assumption, authors: asserted ("can") | All high-Tg APIs carry H-bond donors and can H-bond with both polymers; the two low-Tg APIs lack donors (no H-bonds with PVPVA) but contain chlorine and can halogen-bond with polymer carbonyls [rests on: Fig. 1; cited: refs 26, 27] | Introduction p.2068 |
| C25 | claim, caveat, authors: may be | All high-Tg drugs here can H-bond with both polymers while the low-Tg drugs lack donors, so the poor release at low DL may be an H-bonding effect, a Tg effect or both [rests on: C8; M1–M6] | Discussion p.2077 |
| C29 | claim, mechanism, authors: "are likely to" | From the indomethacin vs indomethacin-ester LoCs, strong drug–polymer interactions such as H-bonding likely lower the polymer release rate and hence the drug release rate [rests on: R24] | Discussion p.2077 |
| C37 | claim, generalisation, authors: discussion: "suggest"; conclusions: "point to the potential importance" | Both Tg and H-bonding potential matter for dissolution performance; the findings point to drug Tg and/or drug–polymer interactions as notable factors in release vs DL [rests on: C3, C26, R24] | Discussion p.2078; Conclusions p.2079 |

## 2022 Soto — Impact of Surfactants on the Performance of Clopidogrel-Copovidone Amorphous Solid Dispersions: Increased Drug Loading and Stabilization of Nanodroplets · Dissolution / A. Release mechanism and LoC

[2022-correa-soto-surfactants-clopidogrel-copovidone-asd-nanodroplets](../../../cards/2022-correa-soto-surfactants-clopidogrel-copovidone-asd-nanodroplets/paper.html) · Dissolution / A. Release mechanism and LoC · used · as: drug-polymer hydrogen bonding

| id | kind | content | locator |
|---|---|---|---|
| C47 | claim, mechanism, authors: discussion: "likely"; conclusions: "was attributed" | The high LoC of CPD arises from two drug properties: its very low Tg and the absence of specific interactions with PVPVA [rests on: M1, R26, R27; cited: refs 12, 13, 84] | p.183, p.185 |
| C49 | claim, generalisation, authors: "tend to"; "somewhat difficult to deconvolute" | Drugs that hydrogen-bond strongly to PVPVA tend to have low LoCs, although these drugs also had high Tgs, so the two effects are hard to separate; neither CPD nor PVPVA has hydrogen-bond donors [rests on: R28, M1, M2; cited: refs 13, 84] | p.184 |
| C50 | claim, conclusion, authors: asserted ("further exemplified") | The lack of CPD-PVPVA interaction is further shown by less than 5% PVPVA in CPD nanoparticles, unlike ibuprofen, which takes up much PVPVA [rests on: R27, R31] | p.184 |

## Mentioned only

- 2015 Schram — [2015-schram-hpmcas-conformation-felodipine-crystal-growth-inhibition](../../../cards/2015-schram-hpmcas-conformation-felodipine-crystal-growth-inhibition/paper.html) · Dissolution / D. Crystallization from solution (ref 22)