PEG–PNIPAM Micelles and Drug Delivery
- PEG–PNIPAM micelles are temperature-responsive assemblies that form core–shell structures with a collapsed PNIPAM core and a hydrated PEG corona above the LCST.
- Micelle morphology, size, and drug encapsulation are tuned by varying the PEG and PNIPAM block lengths, affecting hydration and sustained DOX release.
- Coarse-grained simulations reveal that micelle composition modulates structural parameters and interaction energies, guiding design rules for effective drug delivery.
Searching arXiv for the specified paper to ground the article in the current record. PEG–PNIPAM micelles are temperature-responsive polymeric assemblies formed by poly(ethylene glycol)-poly(N-isopropylacrylamide) block copolymers above the lower critical solution temperature (LCST) of PNIPAM, where PNIPAM becomes effectively hydrophobic and collapses into micellar cores while PEG remains hydrated and forms the corona (Jamirad et al., 16 Aug 2025). In coarse-grained molecular dynamics simulations directed at controlled Doxorubicin (DOX) delivery, these micelles exhibited core–shell-like organization, predominantly ellipsoidal or rod-like morphologies, composition-dependent hydration and stability, and spontaneous drug encapsulation with DOX localized mainly in the PEG-rich shell and the core–shell interface (Jamirad et al., 16 Aug 2025). The central design variable is the hydrophilic–hydrophobic balance set by PEG and PNIPAM block lengths, which modulates micelle size, shell thickness, adsorption thermodynamics, and diffusion-controlled release.
1. Molecular architecture and thermoresponsive self-assembly
The systems examined comprised PEG blocks with 10 or 20 monomers and PNIPAM blocks with 20, 25, 30, or 40 monomers, denoted as PEGmNIPn; a neat PNIPAM control, NIP25, was also studied (Jamirad et al., 16 Aug 2025). Representative 100-chain simulation boxes were reported as 17.4×17.4×17.4 nm for NIP25, 17.7×17.7×17.7 nm for PEG10NIP20, 18.2×18.2×18.2 nm for PEG10NIP25, 19.9×19.9×19.9 nm for PEG10NIP30, 21.7×21.7×21.7 nm for PEG10NIP40, and 19.1×19.1×19.1 nm for PEG20NIP25, with polymer concentration fixed at 10 wt.% and periodic boundary conditions in all dimensions. Self-assembly was studied in MARTINI water with 10% antifreeze beads at 310 K, while single-chain coil–globule behavior was assessed between 280 and 330 K.
The thermo-responsive basis of assembly is the PNIPAM LCST, referenced as approximately $32\,^\circ\mathrm{C}$, and validated in the simulations by a sharp decrease in the single-chain radius of gyration of PEG10NIP20 between 280 and 330 K, with globular conformations above approximately 320–330 K (Jamirad et al., 16 Aug 2025). A coarse-grained thermodynamic scan yielded a PNIPAM glass transition temperature of 393.15 K, close to the experimental value of approximately 406.1 K. Copolymer LCST values were not explicitly measured, but the analysis established that PEG increases hydration and can raise LCST.
Above LCST, PNIPAM collapses into the micellar interior and PEG remains solvated in the exterior shell. The resulting aggregates were described as core–shell-like, with PNIPAM-rich cores and PEG-rich shells (Jamirad et al., 16 Aug 2025). Assembly kinetics were tracked through solvent-accessible surface area (SASA), which dropped markedly by about 200 ns and then plateaued, indicating micelle formation on that timescale; formation was faster at 330 K than at 300 K. In the finite simulation boxes used here, 100 chains aggregated into a single ellipsoidal micelle, while critical micelle concentration (CMC) and critical micelle temperature (CMT) were not explicitly measured.
2. Morphology, size, and structural descriptors
Blank PEG–PNIPAM micelles were non-spherical. Eccentricity ranged from 0.66 to 0.78 and the relative shape anisotropy was approximately 0.03–0.06, consistent with ellipsoidal or rod-like aggregates (Jamirad et al., 16 Aug 2025). Wormlike tendencies emerged with larger PNIPAM content, accompanied by higher SASA. Stability was inferred from the stabilization of 0 over the last 200 ns of the simulations, although PEG10NIP30 and PEG10NIP40 showed larger 1 and stronger temporal fluctuations, indicating more dynamic aggregates.
The principal size metrics for blank micelles at 310 K, averaged over the last 200 ns, are summarized below.
| System | 2 (nm) | 3 (nm) |
|---|---|---|
| NIP25 | 6.56±0.58 | 16.93±1.49 |
| PEG10NIP20 | 7.08±1.19 | 18.29±3.06 |
| PEG10NIP25 | 7.69±0.56 | 19.87±1.45 |
| PEG10NIP30 | 6.46±0.22 | 16.69±0.58 |
| PEG10NIP40 | 9.34±0.38 | 24.11±0.98 |
| PEG20NIP25 | 8.19±0.63 | 21.15±1.63 |
The reported effective diameter was obtained from the radius of gyration using
4
As reported, increasing PNIPAM or PEG length significantly increased micelle size, with PEG10NIP40 and PEG20NIP25 yielding the largest blank aggregates (Jamirad et al., 16 Aug 2025). SASA likewise increased with polymer length, from 771.85±35.83 nm5 for NIP25 to 1641.62±41.06 nm6 for PEG10NIP40, indicating greater solvent exposure in larger and more anisotropic assemblies.
Head-versus-tail 7 values further clarified shell organization. For PEG20NIP25, the PEG (head) 8 was 8.82±0.71 nm and the PNIPAM (tail) 9 was 7.67±0.75 nm, corresponding to an approximate shell-thickness proxy of 1.15 nm, much larger than the values for PEG10-containing systems, which were approximately 0.18–0.29 nm (Jamirad et al., 16 Aug 2025). This identifies PEG length as a direct determinant of corona thickness and suggests a structural basis for differences in hydration and transport.
3. DOX loading, localization, and adsorption thermodynamics
Drug loading was examined primarily at 1 wt.% DOX in systems containing 100 copolymer chains at 10 wt.% polymer; supplementary tests at 3 and 5 wt.% DOX were conducted for PEG20NIP25 (Jamirad et al., 16 Aug 2025). Encapsulation efficiency was not reported as a percentage, but the simulations showed that encapsulation was faster than aggregation of free DOX at the same concentration, consistent with high loading efficacy and with prevention of DOX precipitation.
Radial density analysis placed DOX predominantly in the PEG-rich shell and at the core–shell interface, with limited penetration into the PNIPAM core because of steric and unfavorable interactions (Jamirad et al., 16 Aug 2025). The average DOX radial position spanned 3.91–12.18 Å from the micelle center of mass, and DOX–micelle radial distribution functions showed a primary association peak at approximately 0.50–0.55 nm, with secondary peaks near 1.0 and 1.5 nm. The mechanistic implication is that PEG does not merely solvate the exterior; it provides the dominant local environment for DOX solubilization.
Thermodynamic analysis was based on interaction energies rather than PMF or partition coefficients, which were not computed. Across all systems, DOX–micelle interactions were more favorable than DOX–water interactions, demonstrating spontaneous adsorption (Jamirad et al., 16 Aug 2025). Representative ranges were approximately 0 to 1 kJ mol2 for DOX–DOX, approximately 3 to 4 kJ mol5 for DOX–water, and approximately 6 to 7 kJ mol8 for DOX–micelle interactions. The ordering of DOX–micelle interaction strength, from weaker to stronger adsorption, was
9
This ordering shows that stronger adsorption does not arise from a single compositional variable. A plausible implication is that the shell/interface environment, rather than a purely hydrophobic core, governs the free-volume and contact patterns accessible to DOX in these micelles.
4. Hydration structure, RDF analysis, and interfacial organization
Hydration was quantified through polymer–water and copolymer–water radial distribution functions. PNIPAM–water displayed a first peak at approximately 0.26–0.27 nm, whereas PEG–water displayed a first peak at approximately 0.35 nm (Jamirad et al., 16 Aug 2025). Copolymer–water RDFs showed first and second peaks near 0.5 and 1.0 nm, denoting structured hydration shells. PEG20NIP25 exhibited the highest hydration peak intensity, indicating the most hydrated shell among the copolymers studied.
Qualitatively, PEG ether oxygens engaged water via hydrogen bonding and produced multiple hydration shells, while PNIPAM showed closer first-shell coordination but fewer shells because of hydrophobic core packing above LCST (Jamirad et al., 16 Aug 2025). This distinction is important because the shell hydration state determines both steric stabilization and local transport pathways for solutes. DOX–water hydration was reported to be higher in expanded, high-SASA micelles such as PEG10NIP40 and lower when DOX diffused deeper into shells, as in PEG20NIP25.
SASA analysis reinforced this picture. Blank micelle SASA increased from NIP25 at approximately 772 nm0 to PEG10NIP40 at approximately 1642 nm1, and DOX adsorption further increased micelle SASA (Jamirad et al., 16 Aug 2025). The increase was interpreted as consistent with DOX occupying shell or interfacial regions that increase solvent exposure rather than burying deeply within the hydrophobic core. This interfacial localization distinguishes PEG–PNIPAM micelles from a simplified “hydrophobic core storage” picture and clarifies why PEG-rich architectures can combine high hydration with sustained retention.
Shape descriptors also responded to loading. Blank micelles had ellipticity up to approximately 2.66 for PEG10NIP25, and DOX-induced changes in ellipticity depended on block length: short-block systems such as PEG10NIP20 showed changes up to 22%, whereas PEG20NIP25 changed by approximately 0.3% (Jamirad et al., 16 Aug 2025). This suggests that thicker PEG coronas buffer drug-induced shape perturbations.
5. Diffusion, retention, and composition-dependent release behavior
Release behavior was inferred from mean square displacement (MSD) and the corresponding diffusion coefficients rather than from direct experimental release curves. The study used
2
and
3
with 4, to extract DOX diffusion coefficients from the CG trajectories (Jamirad et al., 16 Aug 2025). The MSD curves were approximately linear over time for DOX in all systems, consistent with normal diffusion, although sudden late-time increases were attributed to pore or cluster formation in the micelle network, enabling transiently faster transport.
The measured DOX diffusion coefficients at 310 K were:
| System | 5 (cm6 s7) |
|---|---|
| NIP25 | 8 |
| PEG10NIP20 | 9 |
| PEG10NIP25 | 0 |
| PEG10NIP30 | 1 |
| PEG10NIP40 | 2 |
| PEG20NIP25 | 3 |
The mobility ranking was reported as PEG10NIP30 4 PEG10NIP25 5 PEG10NIP40 6 PEG10NIP20 7 PEG20NIP25 8 NIP25 (Jamirad et al., 16 Aug 2025). Lower 9 was interpreted as stronger retention and greater sustained-release propensity. On that basis, PEG20NIP25 and NIP25 exhibited the most controlled transport, while PEG10NIP30 supported the fastest diffusion.
The governing trade-off is explicit. Larger PNIPAM blocks increase hydrophobic core formation and strengthen drug binding, but can produce denser cores that retard release; longer PEG blocks thicken and hydrate the corona, stabilizing the micelle while also slowing outward diffusion (Jamirad et al., 16 Aug 2025). The study additionally invoked the Stokes–Einstein relation conceptually,
0
to frame diffusion–size–viscosity relations, although no explicit kinetic model was fitted. The 700 ns simulation window showed equilibrium assembly and DOX remaining largely confined to shell and interfacial regions.
6. Simulation methodology, design rules, and limitations
The multiscale simulation framework combined all-atom parameterization and coarse-grained assembly simulations. All-atom models used OPLS-AA for PEG and PNIPAM, SPC/E water, and atactic PNIPAM stereochemistry; coarse-graining used the MARTINI framework, with PEG mapped to N0 and terminal P2 beads, PNIPAM mapped to C1, Nda, and C3 beads, and DOX represented by MARTINI drug-like beads including P3, SC4, SNa, SP2, SC5, and SNda (Jamirad et al., 16 Aug 2025). All-atom equilibration comprised steepest descent minimization, NVT at 300 K for 200 ps, NPT at 300 K and 1 bar for 200 ps, and a 1 ns production run with a 2 fs timestep, 1.2 nm cutoff, velocity-rescaling thermostat, and Parrinello–Rahman barostat. Coarse-grained simulations used minimization, NVT at 310 K for 50 ns, NPT at 310 K and 1 bar for 50 ns, and 700 ns production with a 20 fs timestep, velocity-rescaling thermostat, Berendsen barostat, Verlet 1.2 nm cutoff, periodic boundary conditions, and 10% antifreeze water. Results were averaged over the last 200 ns, and each system was repeated 3–5 times to estimate standard errors.
Within this framework, the study proposed concrete composition-dependent design rules for DOX delivery (Jamirad et al., 16 Aug 2025). For sustained release and high stability, PEG20NIP25 combined 1 cm2 s3, strong hydration, modest effective diameter of approximately 21–22 nm, and minimal shape change upon loading. For faster diffusion, PEG10NIP30 yielded the highest 4 cm5 s6 and a blank 7 nm, but exhibited dynamic aggregation behavior. For high loading with controlled diffusion, PEG10NIP40 combined 8–25.77 nm, strong DOX–micelle interaction, moderate 9 cm$32\,^\circ\mathrm{C}$0 s$32\,^\circ\mathrm{C}$1, and high SASA.
Several limitations delimit the scope of these conclusions. Coarse-grained MARTINI mapping smooths atomistic details such as specific hydrogen bonding and electrostatics; interaction energies are system-level aggregates and large in magnitude (Jamirad et al., 16 Aug 2025). CMC and CMT were not measured, assembly was observed in finite boxes at 10 wt.% polymer, and LCST shifts due to salt, pH, or protein corona were discussed conceptually but not simulated. Drug loading in the main simulations was modest at 1 wt.%, with higher-load tests performed only for PEG20NIP25; high payloads enlarged micelles and increased anisotropy, which may risk burst release or toxicity. No in vitro or in vivo validation was provided, and release kinetics were inferred from $32\,^\circ\mathrm{C}$2 and MSD rather than directly measured.
Taken together, these results define PEG–PNIPAM micelles as composition-tunable, thermoresponsive nanocarriers in which shell thickness, hydration structure, and core compactness jointly regulate drug retention and transport (Jamirad et al., 16 Aug 2025). Longer PNIPAM increases binding and micelle size but may slow release and introduce dynamic porosity; longer PEG improves hydration, steric stability, and controlled release by thickening the corona. This suggests that hydrophilic–hydrophobic balance in PEG–PNIPAM systems is best treated as a coupled interfacial design variable rather than a simple core-versus-shell partitioning problem.