Backbond
A product by Optagon LabsETH ZürichUniversity of California, Berkeley
Materials science

Design SuperiorMaterials Fast WithPhysics-Driven Simulation

A design assistant that co-optimizes every requirement at once, finding the right chemistry, build and processing parameters.

Create thousands of digital twins, simulate stiffness, impact, water uptake and colour hold in the weather the part actually lives in, then test only the most promising ones.

See the platform
Today's R&D challenge

Trial and error, balancing many conflicting requirements

Seal it against the rain and it gains weight. Take the weight out and it cracks when somebody drops it. Stabilise it against the sun and the cost goes up. Fix all three and the part is worth more than what it is protecting. Then whatever works still has to come off the tool in minutes, clear the label rules, and be bought at volume.

Weatherproof outer body
for a delivery drone
01

The project

A new R&D project is defined.

TARGET Body weightStiffnessImpact resistanceWater resistanceUV stability
02

The targets

Target properties are defined, and they conflict. Fixing one ruins another.

03

The search

Millions of candidate builds, because every chemistry can be coated, blended and laid up in many ways. With tooling and prototypes, only a few are ever made.

ManufacturabilityRegulatory complianceSupply chain
04

The risk

A material that works is only the start. These three are targets of their own, optimised for last when the build is hardest to change. Any one can end the program.

The current build and test process

(up to several years)

Requirements

Lab coupons

Moulded prototype

Water gets in

Reformulate

Field trial

Cracks on impact

Reformulate

Too costly at volume

Reformulate

Qualification

What Backbond does

Backbond co-optimizes for all the targets simultaneously

The same four steps, answered. Backbond hands the team a short list of builds that already hit every target and already survive the tool, the label rules and the supply chain.

Target
Body weight
Define a Number
Stiffness
Define a Number
Impact resistance
Define a Number
Water resistance
Define a Number
UV stability
Define a Number
01

Define

The product, the targets it must hit, and the parameters to change: fibre and filler mix, core and wall, coating chemistry, cure schedule.

02

Simulate

Backbond simulates all possible combinations and evaluates every one, chemistry and build together.

BuildBody weightStiffnessImpact resistanceWater resistanceUV stabilityCost 1Validate2Confident3Validate4Validate5Confident
03

Rank

The candidate builds are ranked by promise and confidence, and those are not the same thing.

Technical properties checkManufacturabilityRegulatory complianceSupply chain
04

Test

The team proceeds to the coupon and the moulded prototype with the most promising builds. By default, these satisfy all targets and constraints.

The Backbond process

(the redesign loops removed)

Requirements

Simulation + focused coupon and field tests

Qualification

Simulation removes the reformulation loops, and those are where the years go. Ageing still has to be witnessed, and no software makes five years pass faster.

Under the hood

How Backbond simulates a design candidate

Four levels of physics decide what a material actually does, and Backbond reasons about how all four interplay to set every spec.

Design space, comprised of many versions of the material
The candidate, dissected, and where it has to survive
Weather coatfluorine-free, 18 µm, 108° contact anglesheds rain, holds colourMoulded skinwoven glass, 4 plies, 0.9 mmsets stiffnessClosed-cell corePET foam, 3 mm, 60 kg/m³sets impact and weightRain, spray and sunIPX5, 20 m/s, five years outdoorsThe frame it closesgasketed seam, 240 g body budget
What sets the final properties
Layer physicsthe polymer and the fibre inside one layer
Surface physicswhere the weather sits, and where two layers meet
Morphologythe build, and the seams it creates
Environmentthe weather and the years it lives in
Final product properties
Candidate 618
Reasoning chain 1 / 250
Layerthe resin takes up 1.2% water in a wet weekand its glass transition falls with it···
Surfacethe coat sheds at 108° while it is newsun chalks it and the surface starts wetting out···
Morphology3 mm of core doubles the stiffness per gramand the seam it creates is where water gets in···
Environmentfive years of sun, then a monsoon seasonthe coat goes first and the skin drinks second···
Together they predict
Body weight212 g ± 6%
Water uptake, 7 d0.6% ± 20%
Cost per airframe$41 ± 12%
Use cases

A few example use cases

Every one is the same shape: a handful of targets that fight each other, and one material that has to hit them all.

01Mono-material laminates

A laminate must become all-polyethylene and keep its barrier. Find the coating, tie layer and process window that hold the oxygen transmission, while keeping seal strength, clarity and cost.

02Downgauging

A film must lose 20% of its thickness and keep its drop performance. Find the resin blend, orientation and layer split that hold the toughness, while keeping seal strength, clarity and cost.

03Barrier without foil

A pack must drop its aluminium foil and still reach twelve months. Find the barrier chemistry, coat weight and cure that hold the shelf life, while keeping recyclability, flex crack resistance and cost.

04Hand feel in bio-based leather

A plant-based leather must feel like the nappa it replaces. Find the binder, plasticiser and emboss that hold the hand, while keeping tensile strength, flex endurance and cost.

05Five years of sun

A surface must hold its colour and its water repellency through five years outdoors. Find the coating chemistry, stabiliser package and cure that resist the chalking, while keeping adhesion, weight and cost.

06Restricted chemistry

A material must drop its fluorochemistry and still resist grease. Find the chemistry, surface texture and cure that hold the repellency, while keeping food-contact compliance, printability and cost.

07Weight against stiffness

A moulded body must lose 15% of its weight and keep its first mode clear of the motors. Find the layup, core thickness and rib pattern that hold the stiffness, while keeping impact resistance and cost.

08Rain at the seams

A housing must pass an IPX5 spray without a gasket that costs more than the panel. Find the seam geometry, sealant and cure that stop the ingress, while keeping weight, serviceability and cost.

09Recycled content

A material must carry 30% recycled content and keep its appearance. Find the blend, filtration and stabiliser package that hold the colour and the strength, while keeping odour, consistency and cost.

10Grade discontinued

A resin must be replaced when its price doubles overnight. Find the substitute, blend and process window that hold every spec, without requalifying the pack.

What it makes possible

The future of the materials R&D organization

How Optagon Labs imagines the materials company of the future, drawn from work with hundreds of R&D teams: one brain coordinating specialised agents that work in parallel, running development end to end across the portfolio.

One central brain

Coordinating work across every function

New products

Designing a material for a market the company has never sold into

a first all-weather housing

Putting a known chemistry into a form it has never been made in

a solvent coating run water-based

Designing for a demand that arrived faster than the lab could

recycle-ready by the next tender
Benchmark matching

Reverse-engineering a competitor’s material and beating it on one axis

a rival’s barrier at lower gauge

Matching an incumbent’s performance from a different chemistry

foil shelf life without the foil

Hitting a customer’s target spec from their brief alone

a film at reference seal and clarity
Appearance and feel

Turning a perceptual target into something the engine can optimise

nappa hand, measured on an instrument

Holding the look when the chemistry underneath changes

the same gloss, no solvent

Matching a reference texture from a different process

a grain that survives embossing
Barrier and life outdoors

Designing for a life the current material cannot reach

five years outdoors, same colour

Protecting a layer from what the pack is filled with

grease at the seal face

Holding properties through the process that damages them

flex cracks after the converting line
Process & scale-up

Designing a material for the line the plant already owns

the same die, a new blend

Adapting a build to a different plant’s equipment

cast line to blown

Correcting a material that runs at lab scale and fails on the line

gauge drift across the web
Redesign under change

Finding a new build that holds every spec after a cost shift

holding barrier after a resin swap

Redesigning fast when a supplier discontinues a grade

life after a grade delisting

Absorbing lot variability without losing spec

melt index drift between lots
Materials restriction

Removing a substance the regulator or the market rejects

PFAS-free grease resistance

Replacing a process chemical at the same performance

a water-based primer

Meeting a restriction across a whole portfolio

one pass, every structure
Sustainability

Designing a material the recycling stream will actually accept

one polymer, one stream

Raising recycled content without losing the spec

thirty percent, same colour

Cutting the footprint of a material already in production

less gauge, same protection
No precedent

Designing where the company has no data and no history

a first bio-based backing

Standing up a new application from the physics and the literature alone

a coating that repels without fluorine