The bands that define a silicone, read straight off the spectrum.
Cure state, residual cyclics and chain length each show up as a specific, strong Raman band - Si-H at ~2160 cm⁻¹, vinyl C=C at ~1600 cm⁻¹, cyclic-versus-linear siloxane at ~490 cm⁻¹ - read inline, in the reactor itself through an immersion probe, with no sampling and no 60-to-120-minute GC/MS slot. A Gekko feasibility study builds the CNN chemometric model on your silicone matrix and hands you the validated model and the numbers.
A validated measurement on your silicone - before any hardware.
A Gekko feasibility study answers one question: can inline Raman read your silicone - cure state, residual cyclics, chain length - in your chemistry, at the accuracy you need. Inline means the measurement happens in the reactor or line itself, through an immersion probe, with no sampling and no 60-to-120-minute GC/MS slot. Real samples, a CNN chemometric model built on your matrix, and a report that hands you the validated model and the numbers.
The study runs on the Spectrally X1 with the same measurement and model stack that later runs on the Spectrally X1 INLINE at your line - at-line on samples first, then with the immersion probe under near-process conditions. What the report shows is what the analyzer reads in production.
Your silicone, measured against your reference method
A PDMS, LSR, HCR or chlorosilane stream is not a generic matrix. The study correlates the Raman spectrum of your silicone with the methods you run today - GC/MS for residual cyclics, DSC or gel-time for cure, GC for chlorosilane composition - and reports the agreement, the working bands and the validity range on your own chemistry.
- Measured on a Spectrally X1 Raman spectrometer - 785 nm or 1064 nm laser for strongly fluorescing media, 8 cm⁻¹ resolution, 300-3500 cm⁻¹ range
- At-line in vials, in-line with an immersion probe, or on-site at the reactor with a mobile unit - sapphire window option for corrosive chlorosilane and HCl media
- Model metrics reported against your reference values, with the validity range stated
Representative process samples - reactor pulls, LSR/HCR compound, chlorosilane cuts, silicone fluids - or existing spectra and reference datasets on your silicone.
Full-spectrum models - Si-H, vinyl and siloxane bands, CNN core - correlated against your reference method on your own matrix.
Spectra, band assignments, model metrics versus your reference values, and the validated model on your silicone chemistry.
The study delivers the model and the numbers before any instrument decision, so the pilot rests on quantified fact.
The questions that define a silicone sit on three strong bands.
Cure state, residual cyclics and chain length each map onto a specific, intense Raman band. Silicones are strong, specific Raman scatterers, so those bands are read directly and continuously in the reactor - the exact information a cure or a cyclics question needs, and hard to get any other way inline.
Cure state - Si-H ~2160 cm⁻¹ and vinyl C=C ~1600 cm⁻¹
In addition-cure LSR, HCR and RTV, crosslinking is the Si-H plus vinyl reaction. Both bands are read directly: the Si-H band at ~2160 cm⁻¹ and the vinyl C=C band at ~1600 cm⁻¹ fall as the network forms, so the cure is confirmed in-process instead of by gel-time observation or a 30-to-90-minute DSC run on a finished sample.
Residual cyclics D4/D5/D6 - siloxane ~490 cm⁻¹
The cyclic-versus-linear siloxane ratio reads from the ~490 cm⁻¹ backbone band, continuously - the regulated by-product tracked in-process instead of through a 60-to-120-minute GC/MS slot with extraction. NIR does not separate cyclic from linear; the Raman band does.
Chain length and silanol - the molecular-weight property, direct
Siloxane chain length / DP and silanol (Si-OH) content ride on the same spectrum, read as the composition they are rather than inferred from a viscosity proxy. One probe, one spectrum, the properties that set silicone performance.
What an inline Raman deployment returns.
These figures are from the phenol-formaldehyde resin portfolio - the fastest payback in the Spectrally installed base, with a self-cleaning reactor probe reading phenol and formaldehyde inline. They show the class of result an inline deployment delivers once the model is on the line, and batch-to-batch variability there held below 1.5%. Your silicone study builds the validated model and the numbers on your own matrix.
Fastest return in the portfolio, on a phenol-formaldehyde resin reactor.
Documented on the PF-resin deployment, per year.
Lower losses once the endpoint is read inline, not after the batch.
Less off-spec material across the PF-resin campaign.
From a crate of samples to a validated model.
The sequence is the same whether the study runs in the Gekko lab or on-site at your plant with a mobile unit.
Measured against your reference method
Study samples are read in parallel on the Raman system and the method you run today - GC/MS for residual cyclics, DSC or gel-time for cure. The model carries the agreement with those values, and that agreement is what the report shows.
The mode matches your process
At-line in vials for screening, an immersion probe for near-process conditions, or on-site measurement at the reactor with a mobile unit - a sapphire window for corrosive chlorosilane and HCl media. Once the analyzer is on the line, each result lands in 5 s.
Built for curing, filming media
Curing silicone can film the optical window. The self-cleaning retractable probe - retract, rinse, return - keeps the window clean without interrupting measurement, and the 30 mW ATEX laser reads in Zone 0.
What lands on your desk.
Every study closes with the same set of deliverables - the validated model and the numbers to take to a pilot.
Measured spectra with band assignments - Si-H, vinyl and siloxane - model performance against your reference values, and the result on your silicone.
A CNN chemometric model scoped to your chemistry - PDMS, addition-cure LSR/HCR, chlorosilane or medical-grade - built on the Si-H, vinyl and siloxane bands, with its validity range.
Where the measurement sits - immersion probe in the reactor or the circulation line, a sapphire window for corrosive chlorosilane media, ATEX Zone 0 - and the recommended hardware configuration for a pilot.
Raman owns composition and crosslinking - chain length, silanol, vinyl, cyclics, cure. Residual Pt catalyst is quantified by ICP as an element. Each method on what it does best, both feeding one batch record.
The hardware the model runs on
A validated model transfers to the process as Spectrally X1 INLINE with the immersion Spectrally X1 PROBE and its self-cleaning module - the retract, rinse and return cycle keeps the optical window clean where curing silicone would otherwise film it - and Spectrally OS turning the spectrum into cure state, residual cyclics and composition PASS/FAIL for your PLC, DCS and MES. A sapphire window handles corrosive chlorosilane and HCl media, and the 30 mW laser reads in ATEX Zone 0. Same bands, on your line, 24/7, resolving to single ppm, with deployment typically 3 to 5.5 months.
Run the study on your silicone.
Send representative samples or R&D data from your PDMS, LSR/HCR, chlorosilane or medical-grade silicone process. We measure them against your reference method - GC/MS for residual cyclics, DSC for cure - build the CNN chemometric model on your matrix and hand you the validated model and the numbers, ready to take to a pilot.