🔬 materials science

High density two-component glasses of organic semiconductors prepared by physical vapor deposition

This study utilizes spectroscopic ellipsometry to demonstrate that co-vapor deposited NPD and TPD organic semiconductor glasses form high-density, thermally stable materials whose properties are governed by substrate temperature and surface equilibration mechanisms, with phase transformations initiating at the free surface and propagating into the bulk.

Original authors: Yejung Lee, Shinian Cheng, M. D. Ediger

Published 2026-07-13

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Yejung Lee, Shinian Cheng, M. D. Ediger

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you're baking a batch of cookies. Usually, if you just dump the dough into a pan and let it cool, it turns into a standard, slightly squishy cookie. But what if you could bake them in a special oven that lets the dough molecules dance around on the surface before they freeze in place? You'd end up with a cookie that's packed tighter, denser, and way tougher to break.

That's essentially what scientists Yejung Lee, Shinian Cheng, and M. D. Ediger discovered, but instead of cookies, they were working with organic semiconductors—the special materials used to make the glowing screens in your phone and TV. Specifically, they mixed two ingredients called NPD and TPD.

The Magic of the "Surface Dance"

When you make these materials using a method called Physical Vapor Deposition (PVD), you're essentially spraying them as a gas onto a cold surface to turn them into a solid glass. The big question was: Does mixing two different molecules mess up the special "super-dense" properties that single-molecule glasses have?

The answer is a resounding no. In fact, mixing them creates materials that equal or exceed the special properties found in single-molecule glasses.

The researchers found that when they deposited these mixed glasses at just the right temperature (specifically between 0.8 and 0.9 of the material's "glass transition temperature," or Tg), they created a material that was:

Up to 1.6% denser than the same material made by just cooling it down from a liquid.
Up to 7% higher in onset temperature compared to the liquid-cooled glass, meaning it takes significantly more heat to make it lose its shape.

Think of it like a crowd of people in a hallway. If they just walk in and sit down randomly (like a liquid cooling down), there are gaps. But if they are allowed to do a little dance on the very top layer before the whole group freezes, they can shuffle into a much tighter, more efficient formation. This "surface dance" is what the scientists call the surface equilibration mechanism.

The Rules of the Game

The paper is very clear about what works and what doesn't. They ruled out the idea that mixing two different molecules would ruin the stability or density. Even though NPD and TPD have slightly different sizes and melting points (a 35 K difference in their glass transition temperatures), they played nice together.

They also ruled out the idea that the mixing had to be a perfect 50:50 split to work. They tested everything from almost pure NPD to almost pure TPD, and the "surface dance" worked perfectly across the entire range. Whether it was a tiny drop of one ingredient in a sea of the other, or a 50:50 mix, the result was the same: a super-dense, super-stable glass.

How Do We Know? (The Proof)

The scientists didn't just guess; they measured everything with a tool called spectroscopic ellipsometry, which is like a super-precise ruler that can see how thick the glass is and how light bends through it.

The Density Test: They heated the glass up and watched it expand. They found that the "as-deposited" (vapor-made) glass was thinner than the "liquid-cooled" glass at the same temperature. Thinner means the molecules are packed tighter. They calculated that the vapor-made glass was 1.6% denser at its peak.
The Stability Test: They watched the temperature at which the glass started to turn back into a liquid. For the vapor-made glass, this "onset temperature" was up to 7% higher than that of the liquid-cooled version. That's a huge jump in stability!
The Transformation: When they heated a super-stable glass to turn it back into a liquid, they saw something cool happen. It didn't melt all at once. Instead, a "front" of liquid started at the very top surface and marched down into the glass at a constant velocity. This is exactly what happens with single-molecule glasses, proving that the mixed glass behaves the same way.

The Orientation Puzzle

There's another cool thing about these glasses: the molecules don't just sit there; they line up. The scientists measured something called birefringence (how the glass bends light differently depending on the direction). They found that the molecules in the mix lined up in a way that was predictable based on the pure ingredients. If you know how NPD lines up and how TPD lines up, you can predict exactly how the mix will line up. It's like knowing how two different types of dancers move, and then knowing exactly how they'll move when they dance together.

Why Does This Matter?

The paper suggests that these findings could be a big deal for the future of electronics.

Better Screens: Because the molecules are packed so tightly, the material might conduct electricity (charge carriers) much better. The authors note that for a similar material, a 1% increase in density led to a 25-fold increase in how well it moves electricity. If that holds true here, a 1.5% density increase could mean a massive boost in performance.
Longer Life: These dense glasses are also more stable, which could mean your phone or TV lasts longer without breaking down.

The scientists are careful to say this is the first time anyone has measured the density of a mixed, vapor-deposited glass. While they are very confident in their numbers for NPD and TPD, they admit they aren't sure yet if every mix of organic semiconductors will behave this way. But for now, they've shown that mixing these two specific ingredients creates a "super-glass" that is denser, stronger, and more stable than anything made by just cooling it down. It's a tiny step in the lab, but a giant leap for making better, brighter, and longer-lasting gadgets.

Technical Summary: High-Density Two-Component Glasses of Organic Semiconductors Prepared by Physical Vapor Deposition

Problem Statement
Physical Vapor Deposition (PVD) is a standard technique for fabricating organic semiconductor devices, such as OLEDs, due to its ability to produce thin films with exceptional properties, including high kinetic stability, low enthalpy, and high density. While extensive research has characterized these properties in single-component vapor-deposited glasses, the fundamental characteristics of multi-component (co-deposited) glasses remain poorly understood. This gap is critical because active layers in OLEDs often consist of mixtures of host and emitter molecules. Previous studies on co-deposited systems have yielded conflicting results regarding stability, and to the authors' knowledge, no prior work had reported the relative density of co-deposited PVD glasses compared to their liquid-cooled counterparts. Furthermore, it was unclear whether the "surface equilibration mechanism"—which explains the superior properties of single-component PVD glasses—applies to mixtures with varying compositions and differing glass transition temperatures ( $T_g$ ).

Methodology
The authors investigated co-deposited glasses of two linear organic semiconductors, NPD and TPD, which have similar molecular sizes but a $35\text{ K}$ difference in $T_g$ . The study utilized the following approaches:

Sample Preparation: NPD and TPD were co-deposited across the entire composition range (from pure TPD to pure NPD) at various substrate temperatures ( $T_{sub}$ ) ranging from $230\text{ K}$ to $330\text{ K}$ .
Bulk Characterization: Differential Scanning Calorimetry (DSC) was performed on bulk physical mixtures to determine melting points, eutectic temperatures, and glass transition temperatures, assessing thermodynamic ideality.
In-situ Spectroscopic Ellipsometry: This was the primary tool for characterizing the vapor-deposited films. It was used to:
- Measure film thickness changes during heating and cooling cycles to determine the onset temperature of transformation ( $T_{onset}$ ), the fictive temperature ( $T_f$ ), and relative density.
- Determine refractive indices and optical birefringence ( $\Delta n$ ) at $650\text{ nm}$ to infer molecular orientation.
Isothermal Annealing: Highly stable co-deposited glasses were annealed isothermally above $T_g$ to observe the transformation mechanism from glass to supercooled liquid.

Key Contributions and Results

First Measurement of Relative Density in Co-deposited PVD Glasses:
The study provides the first evidence that co-deposited NPD:TPD glasses can achieve densities up to $1.6\%$ higher than corresponding liquid-cooled glasses. The density and thermal stability ( $T_{onset}/T_g$ ) exhibit a non-monotonic dependence on the normalized substrate temperature ( $T_{sub}/T_g$ ), with optimal properties observed when deposition occurs between $0.8$ and $0.9$ of the mixture's $T_g$ .
Validation of the Surface Equilibration Mechanism for Mixtures:
The dependence of density, thermal stability, and birefringence on $T_{sub}/T_g$ for the co-deposited mixtures closely mirrors that of single-component glasses. This suggests that the surface equilibration mechanism, where molecules at the free surface equilibrate to low-energy packing configurations before being buried by subsequent layers, governs the formation of these multi-component glasses.
Molecular Orientation and Birefringence:
The films exhibit anisotropy (birefringence) dependent on $T_{sub}/T_g$ . Negative birefringence indicates a "face-on" packing (molecular long axis in the plane), while positive values indicate "end-on" packing. The authors demonstrate that the birefringence of the mixtures can be quantitatively predicted using a linear mixing rule based on the birefringence of the pure components at the same normalized temperature. This holds true for both equimolar ( $50:50$ ) and dilute mixtures.
Transformation Mechanism:
Isothermal annealing experiments revealed that the transformation of a highly stable co-deposited glass into a supercooled liquid initiates at the free surface and propagates into the bulk at a constant velocity. This "front growth" mechanism is consistent with observations in single-component stable glasses and confirms that the high stability is not an artifact of the measurement but a structural feature of the film.
Thermodynamic Ideality:
DSC analysis of bulk mixtures showed a single $T_g$ for all compositions and melting behavior consistent with the Schröder-Van Laar equation for ideal mixing. The calculated activity coefficients were close to unity, indicating that NPD and TPD form a thermodynamically ideal solution. The authors suggest this ideality, combined with similar molecular sizes and shapes, may be a prerequisite for obtaining reliable ellipsometric density data in co-deposited systems.

Significance and Claims
The paper claims to be the first report demonstrating that PVD can generate high-density, mixed glasses of organic semiconductors with enhanced thermal stability across the entire composition range. The authors assert that these findings are significant for several reasons:

Device Performance: Given previous correlations between density and material properties (e.g., a $1\%$ density increase correlating with a $25$-fold increase in charge carrier mobility for NPD, or enhanced photostability), the observed $1.5\%$ density increase in these co-deposited glasses suggests potential for substantial improvements in OLED performance and durability.
Generalizability of Mechanism: The results extend the understanding of the surface equilibration mechanism beyond single-component systems, showing it applies to mixtures with different $T_g$ values and varying compositions.
Predictive Capability: The ability to quantitatively predict the birefringence of mixtures based on pure component data offers a tool for designing molecular orientations in device layers to optimize light outcoupling.

The authors remain modest regarding the universality of these findings, noting that reliable density measurements via ellipsometry were not achieved for other co-deposited pairs (such as TPD/m-MTDATA or DO37/TPD) which exhibited phase separation or non-ideal mixing. They conclude that the success observed here may be specific to systems that are thermodynamically ideal and possess similar molecular geometries.