🔬 condensed matter

A DNA-encoded recipe to direct multi-stage colloidal assembly

This paper demonstrates that a DNA-encoded recipe, which independently programs the time-dependent binding strength and specificity of subunits via multiple biomolecular reactions, enables the kinetic control necessary to guide colloidal building blocks into diverse, complex, multi-scale core-shell structures that are distinct from equilibrium global minima.

Original authors: Pepijn G. Moerman, Chenghung Chou, Thomas E. Videbæk, W. Benjamin Rogers, Rebecca Schulman

Published 2026-04-17

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Pepijn G. Moerman, Chenghung Chou, Thomas E. Videbæk, W. Benjamin Rogers, Rebecca Schulman

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 are trying to build a complex castle out of LEGO bricks.

The Old Way (Equilibrium Assembly):
Usually, when scientists build with tiny particles (like LEGO bricks), they rely on the bricks naturally snapping together in the most stable, comfortable way possible. Think of it like shaking a box of LEGOs until they settle into a flat, uniform pile. The problem is that you can only build simple, repetitive patterns (like a flat wall or a perfect cube). You can't easily build a castle with a red tower in the middle and a blue wall around it, because the bricks don't "know" to stop building the tower and start building the wall. They just want to be in the most comfortable spot, which usually means mixing everything together.

The New Way (The DNA Recipe):
This paper introduces a brilliant new trick: a DNA-encoded recipe. Instead of just shaking the box, the scientists give the LEGO bricks a set of instructions that change over time. They use a biological "timer" to tell the bricks when they are allowed to stick to each other.

Here is how it works, using a simple analogy:

1. The "Sleeping" Bricks

Imagine you have two types of LEGO bricks: Red and Blue.

Normally, they are "asleep." They have no sticky hands, so they float around in the water and ignore each other.
You also have a third type of brick, the Connector, which is always awake and sticky.

2. The DNA "Wake-Up" Call

The scientists add a special "wake-up" signal (a DNA template) to the mix.

The Red Bricks: They get a "wake-up call" immediately. Within an hour, they are fully awake and start grabbing onto the Connectors.
The Blue Bricks: They get a "wake-up call" that is delayed. They have to wait 5 hours before they become sticky.

3. The Result: A Core-Shell Castle

Because of this timing difference, something magical happens:

Hour 1: The Red bricks rush to the Connectors and form a tight, solid ball in the center. They build the Core.
Hour 5: The Blue bricks finally wake up. They try to grab the Connectors, but the Connectors are already busy holding the Red bricks. So, the Blue bricks have to stick to the outside of the Red ball. They form the Shell.

The Result: You end up with a perfect Red core surrounded by a Blue shell.

Why is this a big deal?

If you had just mixed Red, Blue, and Connectors all at once without the timer, they would have all jumbled together into a messy, mixed-up ball. The structure would be boring and uniform.

But by using this DNA recipe, the scientists can program the assembly process like a movie script:

Scene 1: Red bricks act.
Scene 2: Blue bricks act.
Final Scene: A structured, multi-layered object.

The "Kinetically Trapped" Secret

The paper calls these structures "kinetically trapped." Think of it like a snowball.

If you roll a snowball in the snow, it gets big and round.
If you stop rolling it and freeze it instantly, it stays in that shape.
If you let it sit in the sun (equilibrium), it melts and becomes a puddle (the most stable, lowest-energy state).

The DNA recipe allows the scientists to "freeze" the assembly in a specific, complex shape before it has a chance to melt into a boring, mixed-up puddle. They are essentially tricking the system into staying in a temporary, complex state that it wouldn't naturally choose.

Real-World Applications

Why do we care?

Smart Materials: Imagine creating materials that look like cells (with a nucleus and a membrane) but are made of synthetic particles.
Drug Delivery: You could build a tiny capsule where the inside holds medicine (the core) and the outside is a shield that only opens at a specific time (the shell).
Morphogenesis: This mimics how nature builds things. A human embryo starts as a single cell and grows into a complex body with different organs in specific places. This paper shows how we can use simple chemical "recipes" to tell tiny particles to grow into complex shapes, just like nature does.

In short: The scientists figured out how to write a "schedule" for tiny particles. By telling them exactly when to wake up and stick together, they can build complex, multi-layered structures that nature usually can't make with simple building blocks. It's like conducting an orchestra where every instrument knows exactly when to start playing to create a perfect symphony, rather than just making noise.

1. Problem Statement

Traditional colloidal self-assembly relies on equilibrium thermodynamics, where building blocks spontaneously organize into the global free energy minimum. This approach limits the accessible structures to ordered, periodic lattices (e.g., crystals) with features on the scale of the building blocks themselves.

Limitation: It is difficult to create complex, non-periodic structures with compositional variations on mesoscopic length scales (larger than the particle size but smaller than the system size) using equilibrium methods.
Goal: To engineer non-equilibrium assembly pathways that drive the system into specific, kinetically trapped states. This would allow for the creation of diverse, complex structures (like core-shell aggregates) with tunable features that are inaccessible via equilibrium thermodynamics.

2. Methodology

The authors developed a DNA-encoded recipe using the Primer Exchange Reaction (PER) to control the time-dependent binding strength and specificity of colloidal particles.

Building Blocks:
- Precursors: Polystyrene colloids coated with non-complementary single-stranded DNA (ssDNA). Initially, they cannot bind to other particles.
- Co-assemblers: Colloids coated with a specific ssDNA sequence that serves as the binding partner.
The Mechanism (PER):
- A catalytic DNA template strand directs a DNA polymerase to append new DNA domains onto the precursor particles.
- These newly appended domains are complementary to the co-assemblers.
- Kinetic Control: The rate of this polymerization is controlled by the concentration of the DNA template.
  - High template concentration $\rightarrow$ Fast conversion $\rightarrow$ Short lag time.
  - Low template concentration $\rightarrow$ Slow conversion $\rightarrow$ Long lag time.
Experimental Setup:
- Experiments were conducted in water-in-oil microfluidic droplets to create finite-sized, bowl-shaped assemblies, facilitating the visualization of core-shell structures.
- Two types of precursors (labeled red and blue) were mixed with a single type of co-assembler (unlabeled).
- Different template concentrations were applied to the red and blue precursors to create distinct activation delays.
Simulation:
- The authors used NERDSS (Non-Equilibrium Reaction-Diffusion Self-Assembly Simulator) to model the 3D assembly process, incorporating time-delayed activation and irreversible binding.

3. Key Contributions

DNA-Encoded Kinetic Control: Demonstrated a method to program assembly pathways by encoding "recipes" in DNA template concentrations. This allows independent regulation of when specific subunits become active.
Tunable Lag Time: Established a quantitative relationship between template concentration and the lag time (the delay before aggregation begins). The lag time ( $\delta t$ ) is inversely proportional to the template concentration ( $c_T$ ) and can be tuned from minutes to hours.
Pathway-Dependent Morphogenesis: Showed that the same set of building blocks can form qualitatively different structures (mixed vs. core-shell) solely based on the timing of their activation, not their final chemical properties.
Kinetically Trapped States: Proved that the resulting core-shell structures are metastable. When heated above the dissociation temperature and cooled, the system relaxes into a thermodynamically favored mixed state, losing the DNA-encoded heterogeneity.

4. Key Results

Delayed Aggregation: Precursors coated with non-complementary DNA showed no aggregation for a period determined by the PER reaction rate. Once the threshold number of binding domains ( $\chi_T \approx 0.7\%$ of total strands) was reached, rapid aggregation occurred.
Core-Shell Formation:
- Scenario A (Fast Red, Slow Blue): Red particles activated first formed a core; blue particles activated later formed a shell. Result: Red-core/Blue-shell.
- Scenario B (Slow Red, Fast Blue): Blue particles activated first formed a core; red particles formed a shell. Result: Blue-core/Red-shell.
- Scenario C (Simultaneous Activation): Both activated at similar rates. Result: Homogeneous mixed clusters.
Quantitative Heterogeneity:
- The Jensen-Shannon (JS) divergence was used to measure spatial heterogeneity.
- JS divergence increased as the difference in template concentrations (and thus reaction delays) increased.
- The compositional heterogeneity occurred over length scales approximately 10 times larger than the building block size (~10 µm vs. ~1 µm particles).
Simulation Validation: Simulations confirmed that time delays lead to compositional heterogeneity. Early-arriving particles form the core, while late-arriving particles attach to the exterior. The simulations also predicted that time delays affect the overall cluster compactness (fractal dimension).

5. Significance and Implications

Beyond Equilibrium: This work moves colloidal assembly beyond the constraints of free energy minimization, opening a vast design space for complex, non-periodic materials.
Morphogenesis Analogy: The approach mimics biological morphogenesis, where local variations in reaction rates (analogous to morphogen gradients) dictate the spatial organization of cells. Here, local DNA concentrations dictate the spatial organization of colloids.
Memory and Learning: Because the PER reaction creates covalent changes to the particle surface, the particles retain a "memory" of the local chemical environment they experienced. This differs from non-covalent linker approaches where binding is transient. This memory could enable physical learning and adaptability in colloidal materials.
Future Applications: The framework could be applied to:
- Directing the folding of DNA origami.
- Creating "colloidomers" (colloidal polymers).
- Designing materials with tunable mechanical, optical, or transport properties based on internal structural heterogeneity.

In conclusion, the paper presents a robust strategy for kinetic control in self-assembly, using DNA-encoded recipes to program the temporal sequence of particle activation, thereby enabling the precise engineering of complex, multi-scale structures that are kinetically trapped and distinct from their equilibrium counterparts.