AEAPTMS: Amino-Silane Coupling Agent

• AEAPTMS is a multifunctional amino-silane coupling agent featuring a trimethoxysilane anchoring group and adjacent primary–secondary amines that enable effective interfacial passivation.
• It self-assembles on oxide surfaces to form controlled monolayers, influencing hydrogen bonding, charge dynamics, and defect binding as shown in perovskite studies.
• Its reactive nature facilitates chemical modification of formamidinium perovskites, balancing improved voltage and carrier dynamics against risks of multilayer formation.

Searching arXiv for papers on AEAPTMS and closely related amino-silane surface chemistry. [3-(2-aminoethylamino)propyl]trimethoxysilane (AEAPTMS) is an amino-silane coupling agent built around a trimethoxysilane anchoring group and a polyamine terminus. In the form explicitly examined in recent perovskite studies, it is a trimethoxysilane-bearing primary–secondary diamine, $\mathrm{(MeO)_3Si–(CH_2)_3–NH–CH_2CH_2–NH_2}$, used as a reactive interfacial modifier and passivator; in closely related silica self-assembled-monolayer literature, it is also discussed alongside, and in some contexts treated as essentially equivalent to, DETAS-type triamino trialkoxysilanes for purposes of monolayer formation on oxide surfaces (Huang et al., 3 Sep 2025, Dufil et al., 2020). Across these contexts, AEAPTMS is significant because its silane headgroup enables interfacial anchoring while its amine-rich tail controls defect binding, hydrogen bonding, surface dipoles, and, in formamidinium-containing perovskites, direct chemical reactivity with the lattice’s organic cation (Sood-Goodwin et al., 8 Jul 2026, Huang et al., 3 Sep 2025).

1. Molecular identity and nomenclature

AEAPTMS is identified explicitly as $[3\text{-(2-aminoethylamino)propyl}]$trimethoxysilane, a trimethoxysilane terminated by neighboring primary and secondary amines (Huang et al., 3 Sep 2025). Relative to APTMS, which bears one primary amine on a propyl chain, AEAPTMS introduces a second amine adjacent to the first, and that adjacency is central to its surface chemistry. In the perovskite literature, this neighboring primary–secondary diamine motif is presented as enabling cooperative coordination to undercoordinated $\mathrm{Pb^{2+}}$, in line with prior DFT arguments attributed to Lin et al. (Huang et al., 3 Sep 2025).

A separate oxide-SAM study on (3-trimethoxysilylpropyl)diethylenetriamine (DETAS) writes the molecule as $$[\mathrm{MeO}]_3\text{–Si–C}_3[\mathrm{NH–C}_2]_2\text{–NH}_2$$ and states that AEAPTMS, AEAPTES, and “N-[3-(trimethoxysilyl)propyl]diethylenetriamine” are often used for essentially the same or very closely related triamino-functional trialkoxysilane, such that AEAPTMS and DETAS can be treated as equivalent for SAM formation on silica (Dufil et al., 2020). The literature represented here therefore contains a nomenclature asymmetry: one line of work treats AEAPTMS as a diamine silane, whereas another transfers silica-SAM insights from a closely related triamino analogue. This does not erase the structural distinction, but it explains why AEAPTMS appears in both reactive passivation and hydrogen-bond-stabilized monolayer discussions.

A chemically informative signature of AEAPTMS in formamidinium perovskites is the cationic cyclization product 1-(3-(trimethoxysilyl)propyl)-4,5-dihydro-1H-imidazol-3-ium, detected at $m/z = 233.3$ and yielding a fragment $\mathrm{C_3N_2H_7^+}$ at $m/z = 71.06$ (Huang et al., 3 Sep 2025). That product encodes the molecule’s bifunctional character: a silane tether retained at one end and a newly formed imidazolium heterocycle generated from the amine-rich tail.

2. Interfacial chemistry on oxide and silica

As a trialkoxysilane, AEAPTMS belongs to the general class $\mathrm{R–Si(OR')_3}$, which undergoes hydrolysis to trisilanol and then condenses either with surface silanols or laterally with neighboring silanes. The governing reactions are written as

$$\mathrm{R–Si(OR')_3 + 3H_2O \rightarrow R–Si(OH)_3 + 3R'OH}$$

$$\mathrm{R–Si(OH)_3 + HO–Si(surface) \rightarrow R–Si–O–Si(surface) + H_2O}$$

and

$[3\text{-(2-aminoethylamino)propyl}]$0

with the competition between 2D monolayer growth and 3D polymerization determining final film quality (Kosovari et al., 2024).

The DETAS silica study provides the most detailed kinetic and organizational picture transferable to AEAPTMS-like amino-silanes on native silicon oxide. There, growth followed a first-order Langmuir-type equation,

$[3\text{-(2-aminoethylamino)propyl}]$1

with $[3\text{-(2-aminoethylamino)propyl}]$2 values of order $[3\text{-(2-aminoethylamino)propyl}]$3–$[3\text{-(2-aminoethylamino)propyl}]$4, saturation thicknesses of $[3\text{-(2-aminoethylamino)propyl}]$5–$[3\text{-(2-aminoethylamino)propyl}]$6 depending on condition, and a fully optimized monolayer reaching $[3\text{-(2-aminoethylamino)propyl}]$7 (Dufil et al., 2020). The same study reported an immediate jump to about $[3\text{-(2-aminoethylamino)propyl}]$8 at the earliest measurable time, interpreted as transfer of a film formed at the solution–air interface, followed by a slow reorganization over $[3\text{-(2-aminoethylamino)propyl}]$9 into an upright, hydrogen-bond-stabilized monolayer.

Hydrogen bonding is central in that framework. ATR-FTIR bands at $\mathrm{Pb^{2+}}$0, $\mathrm{Pb^{2+}}$1, $\mathrm{Pb^{2+}}$2, and $\mathrm{Pb^{2+}}$3, together with deformation modes at $\mathrm{Pb^{2+}}$4 and $\mathrm{Pb^{2+}}$5, were assigned to strongly hydrogen-bonded amines, while the $\mathrm{Pb^{2+}}$6 band evidenced $\mathrm{Pb^{2+}}$7 network formation (Dufil et al., 2020). Under optimized conditions, ellipsometry and AFM gave thicknesses near $\mathrm{Pb^{2+}}$8, RMS roughness around $\mathrm{Pb^{2+}}$9, and water contact angles around $$[\mathrm{MeO}]_3\text{–Si–C}_3[\mathrm{NH–C}_2]_2\text{–NH}_2$$0–$$[\mathrm{MeO}]_3\text{–Si–C}_3[\mathrm{NH–C}_2]_2\text{–NH}_2$$1, consistent with exposed terminal amines.

The same study also showed that monolayer quality is strongly process-dependent. In toluene at relative humidity above $$[\mathrm{MeO}]_3\text{–Si–C}_3[\mathrm{NH–C}_2]_2\text{–NH}_2$$2, water aggregation around the amino-silane produced reverse micelles and oligomer deposition, with thickness increases by roughly two orders of magnitude relative to the SAM; by contrast, alcohol solvents were much less sensitive to humidity, and concentrations around $$[\mathrm{MeO}]_3\text{–Si–C}_3[\mathrm{NH–C}_2]_2\text{–NH}_2$$3 to $$[\mathrm{MeO}]_3\text{–Si–C}_3[\mathrm{NH–C}_2]_2\text{–NH}_2$$4 yielded the best monolayer quality (Dufil et al., 2020). This establishes an important oxide-surface principle for AEAPTMS-like systems: amine-rich silanes can form well-organized monolayers, but only within a narrow hydrolysis/condensation regime.

3. Monolayer order, multilayer formation, and biofunctional interfaces

A biomaterials study on silicon wafers did not directly use AEAPTMS; instead it compared APTES, AUTES, and a phthalimide-protected C11 aminosilane, then translated the resulting design rules to AEAPTMS-like multi-amine silanes (Kosovari et al., 2024). Its central mechanistic conclusion was that free terminal amines strongly promote hydrogen bonding, electrostatic interactions, and silanol–amine interactions during deposition, favoring 3D polymerization and disordered multilayer growth. In that system, the protected precursor Alk-Phtha formed a monolayer of approximately $$[\mathrm{MeO}]_3\text{–Si–C}_3[\mathrm{NH–C}_2]_2\text{–NH}_2$$5, whereas the free-amine silanes showed PM-IRRAS and XPS signatures consistent with thicker, more disordered layers.

These structural differences propagated directly into peptide coupling performance. After coupling SMPB and the cysteine-bearing peptide CG-K(PEG$$[\mathrm{MeO}]_3\text{–Si–C}_3[\mathrm{NH–C}_2]_2\text{–NH}_2$$6-TAMRA)-GGRGDS, PM-IRRAS amide I and amide II intensities increased much more for Alk-NH$$[\mathrm{MeO}]_3\text{–Si–C}_3[\mathrm{NH–C}_2]_2\text{–NH}_2$$7+RGD than for APTES+RGD or AUTES+RGD, despite the thinner underlying film (Kosovari et al., 2024). The interpretation was that fewer but more accessible amines outperformed thicker amino-silane networks containing nominally more nitrogen but more buried or internally engaged functionality.

The biological readout reinforced the same point. In serum-free hBMSC culture for $$[\mathrm{MeO}]_3\text{–Si–C}_3[\mathrm{NH–C}_2]_2\text{–NH}_2$$8, average cell area increased from $$[\mathrm{MeO}]_3\text{–Si–C}_3[\mathrm{NH–C}_2]_2\text{–NH}_2$$9 on bare silicon to $m/z = 233.3$0 on APTES+RGD, $m/z = 233.3$1 on AUTES+RGD, and $m/z = 233.3$2 on Alk-NH$m/z = 233.3$3+RGD, with the protected/deprotected route also giving the strongest F-actin organization and vinculin-positive focal adhesions (Kosovari et al., 2024). Because AEAPTMS was not studied directly, any conclusion about it in this context is inferential; the paper’s stated implication is that “more amines” does not necessarily mean better functionalization, and that AEAPTMS, by virtue of its multiple amines, may be especially prone to disorder unless protection or stringent deposition control is used.

4. Reactive passivation in formamidinium halide perovskites

AEAPTMS is directly studied as a surface passivator for $m/z = 233.3$4, a $m/z = 233.3$5 perovskite, where it is applied by room-temperature vacuum vapor deposition for $m/z = 233.3$6, $m/z = 233.3$7, or $m/z = 233.3$8 at a gauge pressure of about $m/z = 233.3$9 in. Hg relative to atmospheric pressure (Huang et al., 3 Sep 2025). In p-i-n devices it was inserted between the perovskite and the $\mathrm{C_3N_2H_7^+}$0 electron-transport layer in the stack Glass/ITO / MeO-2PACz / perovskite / AEAPTMS / $\mathrm{C_3N_2H_7^+}$1 / BCP / Ag.

Its electronic effect is beneficial but dose-dependent. AEAPTMS increased TRPL lifetimes relative to unpassivated films for all treatment times, and longer deposition times gave successively longer lifetimes, consistent with reduced nonradiative recombination (Huang et al., 3 Sep 2025). Device performance, however, peaked at short exposure: around $\mathrm{C_3N_2H_7^+}$2, AEAPTMS produced the highest overall power conversion efficiency among the tested conditions, with improved $\mathrm{C_3N_2H_7^+}$3 and fill factor; at $\mathrm{C_3N_2H_7^+}$4 and $\mathrm{C_3N_2H_7^+}$5, $\mathrm{C_3N_2H_7^+}$6 continued to increase but $\mathrm{C_3N_2H_7^+}$7, fill factor, and PCE declined, consistent with a thicker insulating interlayer and some perovskite disruption. Relative to APTMS, AEAPTMS showed a wider, more robust processing window.

The distinctive feature of AEAPTMS in this system is not merely Lewis-base coordination but direct chemical reaction with formamidinium. In DMSO-$\mathrm{C_3N_2H_7^+}$8, equimolar AEAPTMS and FAI produced a new $\mathrm{C_3N_2H_7^+}$9 NMR peak at $m/z = 71.06$0 within $m/z = 71.06$1, while the FA$m/z = 71.06$2 methine resonance at $m/z = 71.06$3 disappeared; after $m/z = 71.06$4, $m/z = 71.06$5 NMR showed a new signal at $m/z = 71.06$6, and HSQC connected the $m/z = 71.06$7 proton to that carbon (Huang et al., 3 Sep 2025). The reaction was interpreted as nucleophilic attack on FA$m/z = 71.06$8, intramolecular cyclization, and ammonia elimination: $m/z = 71.06$9

Time-of-flight SIMS indicated that the same chemistry occurs at the perovskite surface. The FA$\mathrm{R–Si(OR')_3}$0 fragment at $\mathrm{R–Si(OR')_3}$1 decreased continuously with increasing AEAPTMS exposure, while the imidazolium-derived fragment $\mathrm{R–Si(OR')_3}$2 at $\mathrm{R–Si(OR')_3}$3 increased concomitantly and by about $\mathrm{R–Si(OR')_3}$4 relative to unpassivated films (Huang et al., 3 Sep 2025). Depth profiles further showed shallower penetration for AEAPTMS than for APTMS. The authors’ interpretation was that rapid formation of a bulky imidazolium-silane product confines AEAPTMS near the surface, which plausibly explains its wider processing window.

5. Surface-potential dynamics and grain-boundary electrostatics

A subsequent nanoscale study examined AEAPTMS-passivated $\mathrm{R–Si(OR')_3}$5 using dual-pass amplitude-modulated Kelvin probe force microscopy in ambient conditions (Sood-Goodwin et al., 8 Jul 2026). In that framework,

$\mathrm{R–Si(OR')_3}$6

and

$\mathrm{R–Si(OR')_3}$7

AEAPTMS was introduced ex situ by vapor-phase exposure in a sealed Petri dish at $\mathrm{R–Si(OR')_3}$8 for $\mathrm{R–Si(OR')_3}$9.

The dark CPD distribution narrowed dramatically after passivation, from a full width at half maximum of $$\mathrm{R–Si(OR')_3 + 3H_2O \rightarrow R–Si(OH)_3 + 3R'OH}$$0 in unpassivated films to $$\mathrm{R–Si(OR')_3 + 3H_2O \rightarrow R–Si(OH)_3 + 3R'OH}$$1 in AEAPTMS-passivated films, without obvious morphological changes in AFM topography (Sood-Goodwin et al., 8 Jul 2026). Because the absolute CPD shift was not calibrated to an external work-function reference, the paper interprets the width reduction, rather than the absolute peak position, as the meaningful indicator. The result was taken as evidence that AEAPTMS homogenizes the energetic landscape.

Under broad-spectrum white light, the steady-state surface photovoltage increased from $$\mathrm{R–Si(OR')_3 + 3H_2O \rightarrow R–Si(OH)_3 + 3R'OH}$$2 to $$\mathrm{R–Si(OR')_3 + 3H_2O \rightarrow R–Si(OH)_3 + 3R'OH}$$3, while the illumination-rise time constant decreased from $$\mathrm{R–Si(OR')_3 + 3H_2O \rightarrow R–Si(OH)_3 + 3R'OH}$$4 to $$\mathrm{R–Si(OR')_3 + 3H_2O \rightarrow R–Si(OH)_3 + 3R'OH}$$5 (Sood-Goodwin et al., 8 Jul 2026). The paper interprets this as simultaneous enhancement of local quasi-Fermi-level splitting and suppression of slow defect- or ion-mediated transients. Wavelength-resolved measurements refined that picture: under $$\mathrm{R–Si(OR')_3 + 3H_2O \rightarrow R–Si(OH)_3 + 3R'OH}$$6 illumination the passivated sample showed a monotonic SPV rise to a higher steady state, whereas the unpassivated sample exhibited an initial drop followed by a slower rise; under $$\mathrm{R–Si(OR')_3 + 3H_2O \rightarrow R–Si(OH)_3 + 3R'OH}$$7 sub-bandgap illumination, the unpassivated film showed a measurable SPV response but the AEAPTMS-passivated film was reduced to the order of dark drift, consistent with reduced sub-bandgap electronic disorder.

Grain-resolved analysis defined

$$\mathrm{R–Si(OR')_3 + 3H_2O \rightarrow R–Si(OH)_3 + 3R'OH}$$8

For both passivated and unpassivated films, $$\mathrm{R–Si(OR')_3 + 3H_2O \rightarrow R–Si(OH)_3 + 3R'OH}$$9, but $$\mathrm{R–Si(OH)_3 + HO–Si(surface) \rightarrow R–Si–O–Si(surface) + H_2O}$$0 was systematically smaller after AEAPTMS treatment across grain-boundary mask widths up to $$\mathrm{R–Si(OH)_3 + HO–Si(surface) \rightarrow R–Si–O–Si(surface) + H_2O}$$1 (Sood-Goodwin et al., 8 Jul 2026). The reported implication is that AEAPTMS suppresses grain-boundary potential barriers and smooths the lateral carrier landscape.

6. Design principles, limitations, and research directions

Taken together, the available literature presents AEAPTMS as a molecule whose value lies in coupling interfacial anchoring to chemically active amine functionality, but whose performance is highly context-dependent. On oxide and silica surfaces, closely related amino-silanes can yield smooth $$\mathrm{R–Si(OH)_3 + HO–Si(surface) \rightarrow R–Si–O–Si(surface) + H_2O}$$2 monolayers with exposed amines when solvent polarity, concentration, water activity, and reorganization time are controlled; under humid, non-polar conditions they can instead form oligomeric or micellar multilayers (Dufil et al., 2020). In peptide-functional silicon systems, analogous data suggest that unprotected amino-silanes can suffer from buried functionality and poor accessibility, making protecting-group strategies attractive when monolayer order is the priority (Kosovari et al., 2024).

In formamidinium perovskites, AEAPTMS is best described not as an inert passivant but as a reactive interfacial modifier. It coordinates defects, changes nanoscale electrostatics, and directly consumes FA$$\mathrm{R–Si(OH)_3 + HO–Si(surface) \rightarrow R–Si–O–Si(surface) + H_2O}$$3 to form an imidazolium-derived product at the surface (Huang et al., 3 Sep 2025, Sood-Goodwin et al., 8 Jul 2026). This combination can be advantageous: higher $$\mathrm{R–Si(OH)_3 + HO–Si(surface) \rightarrow R–Si–O–Si(surface) + H_2O}$$4-relevant photovoltage, faster SPV stabilization, reduced sub-bandgap disorder, and a broader processing window than APTMS have all been reported. It also introduces a clear upper bound on dosage, because thicker AEAPTMS-derived interlayers become electrically insulating and can partially disrupt the 3D perovskite phase (Huang et al., 3 Sep 2025).

Several limitations remain explicit in the current record. The biomaterials paper does not study AEAPTMS directly, so its relevance is mechanistic rather than empirical (Kosovari et al., 2024). The KPFM study does not extract absolute work functions or report full photovoltaic device metrics, and it infers the chemical mechanism from electrostatic observables and prior literature rather than direct spectroscopy (Sood-Goodwin et al., 8 Jul 2026). The passivation paper does not present long-term operational stability for AEAPTMS-treated devices (Huang et al., 3 Sep 2025). A plausible implication is that future work will need to integrate direct chemical analysis, thickness control, and device-level stability measurements to distinguish when AEAPTMS functions primarily as a coordinated Lewis base, when it functions as a reactive precursor to interfacial imidazolium phases, and when its multi-amine character instead drives undesirable multilayer growth.