Sustainable Polymers, Biodegradation and Circular Recycling

Learning objective: Evaluate recycling, degradation, bio-based feedstocks and design-for-circularity with material-property and system-boundary checks.

What this unit requires

Sustainable design requires separate ledgers for collected material, process losses, accepted products, useful function and life-cycle inventory. Avoid double-counting recovered mass or displacement credits and state when unverified assumptions dominate the result.

An engineering statement is useful only after its units, system boundary, assumptions and evidentiary status are defined. This unit deliberately distinguishes mathematical identity, conditional model prediction, laboratory observation and verified process claim. Calculated values in these pages are synthetic worked examples, not measured commercial-resin data. Seek grade- and application-specific property evidence before any equipment or product decision.

Four-lesson pathway

1. [Mechanical reprocessing and blend properties](../lessons/unit-08-lesson-01.html) — [numerical model 22](../labs/lab-22.html)
2. [Degradation, weathering and service lifetime](../lessons/unit-08-lesson-02.html) — [numerical model 23](../labs/lab-23.html)
3. [Biopolymers, compostability and material selection](../lessons/unit-08-lesson-03.html) — [numerical model 24](../labs/lab-24.html)
4. [Circularity metrics and life-cycle boundaries](../lessons/unit-08-lesson-04.html) — [numerical model 24](../labs/lab-24.html)

Demonstrate learning

• [Sixteen multi-part assessment questions](../assessments/unit-08.html): numerical reconstruction, equation audit, model challenge and design trade-off.
• [Unit-level engineering project](../projects/unit-08.html): a practical but fictional application with traceable mass, quality and evidence requirements.
• [Interactive offline laboratory manual](../LABORATORY-MANUAL.html): adjustable, source-visible calculations with CSV export.

Completion standard

Explain a material mechanism, reconstruct the featured calculation with consistent dimensions, test one limiting case and specify an experiment that could invalidate the conclusion. An equation that reproduces its own arithmetic has been software-verified; it has not thereby been validated against polymer specimens or factory operation.

Unit 08 · Lesson 01 — Mechanical reprocessing and blend properties

[Unit 08 course map](../units/unit-08.html) · [Interactive simulation 22](../labs/lab-22.html) · [Assessment](../assessments/unit-08.html) · [Applied case](../projects/unit-08.html)

1. The polymer-processing question

Mechanical recycling changes composition and molecular history. Mixed grades, additives, colorants, fillers and chain scission complicate predictions from simple virgin–recycled blend fractions. Linear mixing rules can sometimes screen density or nominal property changes, but they often fail for toughness, interfacial adhesion and environmental-stress cracking. Regrind quality depends on collection, sorting, washing, drying and stabilization. A higher recycled fraction is not automatically better if material losses, contamination and part rejection eliminate the expected benefit.

This lesson belongs to Sustainable polymer systems, whose goal is to evaluate recycling, degradation, bio-based feedstocks and design-for-circularity with material-property and system-boundary checks. The challenge is to connect chemical structure or thermomechanical history to an output that can be measured, rather than reporting attractive numerical results without a clear object of study. Name the material family and the actual specimen state before using a correlation. A number obtained on dry pellets can be inappropriate for a moist, filled, highly oriented or degraded finished article.

Sustainable design requires separate ledgers for collected material, process losses, accepted products, useful function and life-cycle inventory. Avoid double-counting recovered mass or displacement credits and state when unverified assumptions dominate the result.

2. Mathematical description and physical quantities

Starting relationship: Linear property proxy Pblend=wrec Prec+(1−wrec)Pvirgin.

A formula is meaningful only with its variable definitions. Identify the dependent variable, its SI unit, the independent variables and the dimensional basis of every fitted coefficient. For a ratio, inspect positive denominators and the zero-input limit; for a rate law, distinguish a rate per mass from a rate per volume; for a thermal transition, use kelvin if a reciprocal temperature is used. Reconstruct the equation's origin from counting, conservation, constitutive behavior or an explicitly stated empirical fit. Those origins cannot be interchanged. In particular, a pure mass-balance identity may survive changes in polymer chemistry while a viscosity correlation usually does not.

Keep material circularity separate from environmental benefit — A recycling yield can be verified with a mass balance, whereas climate benefit requires measured energy and emissions under matched boundaries. Chemical and mechanical recycling can change polymer structure, additives and quality. Explicitly compare accepted products of equal function; recovered kilograms are an intermediate accounting metric, not proof of circular benefit.

Sketch a control volume, measurement chain, or molecular population appropriate to Mechanical reprocessing and blend properties. Annotate the sketch with the actual controlled settings and the measured outputs, and mark each coefficient as an assumed teaching value, a fitted value or a possible independent measurement. The resulting sketch should make clear what extra measurement would transform a numerical demonstration into an experimentally testable model. Avoid assigning industrial meaning to a symbol solely because its unit appears plausible.

3. Worked numerical investigation

For synthetic stiffness values Erec=800 MPa, Evirgin=1200 MPa and 40% recycled fraction, Eblend≈1040 MPa. The actual two-phase composite response may not obey the straight line.

Independent calculation procedure: Extract the supplied initial conditions into a table and convert all reported units before substituting numbers. Record at least one intermediate result; then recompute the output in a calculator or spreadsheet without copying the answer. Vary the first independent quantity by 10% while all others remain fixed. Compute both the absolute response and percentage response; decide whether a local linear approximation is adequate. Reverse the question by asking whether the underlying measurement has enough resolution to distinguish the two numerical responses.

Constraint check: Identify any physically impossible input: a negative molecule count, mass recovery above the feed, mass fraction exceeding unity, a temperature on an incorrect absolute scale, a stress beyond a known material domain, or a nonfinite denominator. Not all of these appear in this model; choose those that apply. Discuss whether the original example is conservative or optimistic for real material variation. Values here are synthetic to explain algebra and must not be presented as published resin property tables.

4. Model verification is not material validation

Mixing laws can miss phase separation, compatibilization, morphology and impact-strength thresholds; do not make product safety claims from them.

Rearrange the equation to infer one input from the target output. If the rearrangement is ill-conditioned, explain how small measurement error may produce a large inferred-parameter error. Identify a pair of conceptually different mechanisms capable of moving the observed output in the same direction. A sound diagnosis cannot select one mechanism solely from that output; it needs an orthogonal measurement. This is the central protection against overinterpreting a polymer melt curve, a DSC peak, a strength measurement or a recycling-yield statistic.

Compare Mechanical reprocessing and blend properties with Degradation, weathering and service lifetime as a neighboring subject in this unit. Do the two relationships describe the same physical quantity or two independent descriptors? Would their characteristic times, spatial scales, material states or experimental procedures be comparable? Propose one experiment that measures both and write what a disagreement would imply. A second model is useful only if it provides independent information rather than repeating the same assumption in a new notation.

5. Evidence and experimental design

Characterize the feed’s contaminants, water content and molecular weight; measure blended property distributions, not single coupons.

Measurement planning: Define specimens, sampling locations, conditioning, calibration method, measurement resolution and number of independent replicates. Specify one reference material or control and the expected direction of a genuine effect. Organize a data ledger with columns for parameter, unit, instrument, method, timestamp, operator, uncertainty and calibration status. Do not fit and validate a model on the same observations without acknowledging that the validation is not independent. Where the property depends on processing history, preserve the exact prior temperature, deformation and storage conditions.

A robust plan explicitly separates structural uncertainty, parameter uncertainty and observation uncertainty. Structural uncertainty asks whether important physics are absent; parameter uncertainty asks whether fitted numbers are poorly identified; observation uncertainty asks whether instrumentation or sampling is noisy or biased. Repeated sampling can reduce some random errors but cannot repair an incorrect equation or uncalibrated specimen history. For Mechanical reprocessing and blend properties, name one of each, and predict which category will dominate extrapolation into an unmeasured processing regime.

6. Interactive investigation and a falsifiable prediction

Open [Laboratory 22](../labs/lab-22.html). Read its exact equation and domain, then predict the trend with the first slider before viewing the curve. Change only that slider at three positions, record numeric outputs and use CSV export for the complete sweep. Change a second parameter next; compare whether the response retains its direction and curvature. Treat visible invalid-condition warnings as part of the lesson, not software errors to ignore. The simulation operates entirely in the browser with no external requests.

Investigation: Compare a screening stiffness target with separate impact and fatigue requirements.

Create a two-column evidence note: one column lists predictions directly supported by the stated mathematical relationship; the second lists interpretations that require independent polymer measurements. State what observation would most clearly falsify the assumed relation. If no practical experiment is yet proposed, the numerical model is a consistency exercise rather than a validated predictor. Do not infer equipment tolerances, worker-safety margins or compliance from fictional sliders.

7. Design decision with limitations

A process engineer evaluating Mechanical reprocessing and blend properties faces at least two objectives: a desired quantity or geometry and the assurance that material quality remains fit for intended service. Build a trade-off table containing accepted-product quality, throughput or test cost, risk of extrapolation and measurement effort. For each alternative, name the assumption on which the apparent advantage depends. Identify a stop condition, such as failure to reproduce the response with a new batch or detection of a qualitatively different material phase or deformation regime. This makes the decision conditional and auditable.

Read the technical reference titles on the [references page](../REFERENCES.html) and check whether they discuss terminology, measurement method or product specifications. A terminology reference cannot supply experimental constants; a standard's title and abstract do not grant permission to claim full procedural compliance. The book is an editorial draft and a verified industrial decision requires current complete standards, manufacturer information and professional review.

8. Independent lesson problems

1. Derive or reconstruct Linear property proxy Pblend=wrec Prec+(1−wrec)Pvirgin from the most appropriate physical or statistical principle. Define every symbol and every unit, then identify a limiting case.
2. Reproduce this lesson's worked example and change one input by ±10%; report output changes with justified significant figures.
3. Identify two competing physical explanations for a deviation from the model, and one measurement that distinguishes them.
4. Design a sampling plan that tests applicability after a change in material grade, temperature or processing history. Show how you would preserve calibration traceability.
5. Write a six-sentence engineering review: objective, model result, credible restriction, alternative, required experiment and conditional conclusion.

Next: return to the [unit pathway](../units/unit-08.html) or complete the [unit assessment](../assessments/unit-08.html).

Unit 08 · Lesson 02 — Degradation, weathering and service lifetime

[Unit 08 course map](../units/unit-08.html) · [Interactive simulation 23](../labs/lab-23.html) · [Assessment](../assessments/unit-08.html) · [Applied case](../projects/unit-08.html)

1. The polymer-processing question

UV light, oxygen, moisture, heat, mechanical cycling and chemical environments change polymer structure at different rates. Artificial weathering accelerates selected stressors but may change failure mechanism instead of merely compressing real time. A first-order property-retention model illustrates long-term sensitivity; it is not a validated life-expectancy prediction. Service design requires the actual part geometry, environment, stress history and required function, plus uncertainty and accelerated-test correlation. A degradable label also does not prove degradation in any receiving environment.

This lesson belongs to Sustainable polymer systems, whose goal is to evaluate recycling, degradation, bio-based feedstocks and design-for-circularity with material-property and system-boundary checks. The challenge is to connect chemical structure or thermomechanical history to an output that can be measured, rather than reporting attractive numerical results without a clear object of study. Name the material family and the actual specimen state before using a correlation. A number obtained on dry pellets can be inappropriate for a moist, filled, highly oriented or degraded finished article.

Sustainable design requires separate ledgers for collected material, process losses, accepted products, useful function and life-cycle inventory. Avoid double-counting recovered mass or displacement credits and state when unverified assumptions dominate the result.

2. Mathematical description and physical quantities

Starting relationship: Ideal retained property fraction R(t)=exp(−kt).

A formula is meaningful only with its variable definitions. Identify the dependent variable, its SI unit, the independent variables and the dimensional basis of every fitted coefficient. For a ratio, inspect positive denominators and the zero-input limit; for a rate law, distinguish a rate per mass from a rate per volume; for a thermal transition, use kelvin if a reciprocal temperature is used. Reconstruct the equation's origin from counting, conservation, constitutive behavior or an explicitly stated empirical fit. Those origins cannot be interchanged. In particular, a pure mass-balance identity may survive changes in polymer chemistry while a viscosity correlation usually does not.

Keep material circularity separate from environmental benefit — A recycling yield can be verified with a mass balance, whereas climate benefit requires measured energy and emissions under matched boundaries. Chemical and mechanical recycling can change polymer structure, additives and quality. Explicitly compare accepted products of equal function; recovered kilograms are an intermediate accounting metric, not proof of circular benefit.

Sketch a control volume, measurement chain, or molecular population appropriate to Degradation, weathering and service lifetime. Annotate the sketch with the actual controlled settings and the measured outputs, and mark each coefficient as an assumed teaching value, a fitted value or a possible independent measurement. The resulting sketch should make clear what extra measurement would transform a numerical demonstration into an experimentally testable model. Avoid assigning industrial meaning to a symbol solely because its unit appears plausible.

3. Worked numerical investigation

If hypothetical k=0.002 day⁻¹, property retention after 180 days is ≈0.698. Real k may depend strongly on climate, additives, strain and specimen thickness.

Independent calculation procedure: Extract the supplied initial conditions into a table and convert all reported units before substituting numbers. Record at least one intermediate result; then recompute the output in a calculator or spreadsheet without copying the answer. Vary the first independent quantity by 10% while all others remain fixed. Compute both the absolute response and percentage response; decide whether a local linear approximation is adequate. Reverse the question by asking whether the underlying measurement has enough resolution to distinguish the two numerical responses.

Constraint check: Identify any physically impossible input: a negative molecule count, mass recovery above the feed, mass fraction exceeding unity, a temperature on an incorrect absolute scale, a stress beyond a known material domain, or a nonfinite denominator. Not all of these appear in this model; choose those that apply. Discuss whether the original example is conservative or optimistic for real material variation. Values here are synthetic to explain algebra and must not be presented as published resin property tables.

4. Model verification is not material validation

One rate coefficient cannot represent oxidation-induction periods, crack growth or diffusion-limited degradation.

Rearrange the equation to infer one input from the target output. If the rearrangement is ill-conditioned, explain how small measurement error may produce a large inferred-parameter error. Identify a pair of conceptually different mechanisms capable of moving the observed output in the same direction. A sound diagnosis cannot select one mechanism solely from that output; it needs an orthogonal measurement. This is the central protection against overinterpreting a polymer melt curve, a DSC peak, a strength measurement or a recycling-yield statistic.

Compare Degradation, weathering and service lifetime with Biopolymers, compostability and material selection as a neighboring subject in this unit. Do the two relationships describe the same physical quantity or two independent descriptors? Would their characteristic times, spatial scales, material states or experimental procedures be comparable? Propose one experiment that measures both and write what a disagreement would imply. A second model is useful only if it provides independent information rather than repeating the same assumption in a new notation.

5. Evidence and experimental design

Use matched control exposures, mechanistic chemical markers and relevant load-bearing measurements over service conditions.

Measurement planning: Define specimens, sampling locations, conditioning, calibration method, measurement resolution and number of independent replicates. Specify one reference material or control and the expected direction of a genuine effect. Organize a data ledger with columns for parameter, unit, instrument, method, timestamp, operator, uncertainty and calibration status. Do not fit and validate a model on the same observations without acknowledging that the validation is not independent. Where the property depends on processing history, preserve the exact prior temperature, deformation and storage conditions.

A robust plan explicitly separates structural uncertainty, parameter uncertainty and observation uncertainty. Structural uncertainty asks whether important physics are absent; parameter uncertainty asks whether fitted numbers are poorly identified; observation uncertainty asks whether instrumentation or sampling is noisy or biased. Repeated sampling can reduce some random errors but cannot repair an incorrect equation or uncalibrated specimen history. For Degradation, weathering and service lifetime, name one of each, and predict which category will dominate extrapolation into an unmeasured processing regime.

6. Interactive investigation and a falsifiable prediction

Open [Laboratory 23](../labs/lab-23.html). Read its exact equation and domain, then predict the trend with the first slider before viewing the curve. Change only that slider at three positions, record numeric outputs and use CSV export for the complete sweep. Change a second parameter next; compare whether the response retains its direction and curvature. Treat visible invalid-condition warnings as part of the lesson, not software errors to ignore. The simulation operates entirely in the browser with no external requests.

Investigation: Distinguish an accelerated lifetime claim from a demonstrated failure-mode equivalence test.

Create a two-column evidence note: one column lists predictions directly supported by the stated mathematical relationship; the second lists interpretations that require independent polymer measurements. State what observation would most clearly falsify the assumed relation. If no practical experiment is yet proposed, the numerical model is a consistency exercise rather than a validated predictor. Do not infer equipment tolerances, worker-safety margins or compliance from fictional sliders.

7. Design decision with limitations

A process engineer evaluating Degradation, weathering and service lifetime faces at least two objectives: a desired quantity or geometry and the assurance that material quality remains fit for intended service. Build a trade-off table containing accepted-product quality, throughput or test cost, risk of extrapolation and measurement effort. For each alternative, name the assumption on which the apparent advantage depends. Identify a stop condition, such as failure to reproduce the response with a new batch or detection of a qualitatively different material phase or deformation regime. This makes the decision conditional and auditable.

Read the technical reference titles on the [references page](../REFERENCES.html) and check whether they discuss terminology, measurement method or product specifications. A terminology reference cannot supply experimental constants; a standard's title and abstract do not grant permission to claim full procedural compliance. The book is an editorial draft and a verified industrial decision requires current complete standards, manufacturer information and professional review.

8. Independent lesson problems

1. Derive or reconstruct Ideal retained property fraction R(t)=exp(−kt) from the most appropriate physical or statistical principle. Define every symbol and every unit, then identify a limiting case.
2. Reproduce this lesson's worked example and change one input by ±10%; report output changes with justified significant figures.
3. Identify two competing physical explanations for a deviation from the model, and one measurement that distinguishes them.
4. Design a sampling plan that tests applicability after a change in material grade, temperature or processing history. Show how you would preserve calibration traceability.
5. Write a six-sentence engineering review: objective, model result, credible restriction, alternative, required experiment and conditional conclusion.

Next: return to the [unit pathway](../units/unit-08.html) or complete the [unit assessment](../assessments/unit-08.html).

Unit 08 · Lesson 03 — Biopolymers, compostability and material selection

[Unit 08 course map](../units/unit-08.html) · [Interactive simulation 24](../labs/lab-24.html) · [Assessment](../assessments/unit-08.html) · [Applied case](../projects/unit-08.html)

1. The polymer-processing question

Bio-based, biodegradable and compostable describe different attributes: origin of carbon, biological decomposition pathway and performance under specified test conditions. A fully fossil-derived polymer can be biodegradable; a bio-derived polymer can be very persistent. Different receiving systems such as industrial compost, soil, marine water and landfill impose incompatible temperature, moisture, oxygen and microbial conditions. Material selection must also consider barrier properties, moisture susceptibility, processing temperature, cost and end-of-life collection. Avoid converting a laboratory mass-loss observation into an unsupported environmental claim.

This lesson belongs to Sustainable polymer systems, whose goal is to evaluate recycling, degradation, bio-based feedstocks and design-for-circularity with material-property and system-boundary checks. The challenge is to connect chemical structure or thermomechanical history to an output that can be measured, rather than reporting attractive numerical results without a clear object of study. Name the material family and the actual specimen state before using a correlation. A number obtained on dry pellets can be inappropriate for a moist, filled, highly oriented or degraded finished article.

Sustainable design requires separate ledgers for collected material, process losses, accepted products, useful function and life-cycle inventory. Avoid double-counting recovered mass or displacement credits and state when unverified assumptions dominate the result.

2. Mathematical description and physical quantities

Starting relationship: Illustrative mass-loss percentage L=100(m0−mt)/m0, defined on a consistent dry basis.

A formula is meaningful only with its variable definitions. Identify the dependent variable, its SI unit, the independent variables and the dimensional basis of every fitted coefficient. For a ratio, inspect positive denominators and the zero-input limit; for a rate law, distinguish a rate per mass from a rate per volume; for a thermal transition, use kelvin if a reciprocal temperature is used. Reconstruct the equation's origin from counting, conservation, constitutive behavior or an explicitly stated empirical fit. Those origins cannot be interchanged. In particular, a pure mass-balance identity may survive changes in polymer chemistry while a viscosity correlation usually does not.

Keep material circularity separate from environmental benefit — A recycling yield can be verified with a mass balance, whereas climate benefit requires measured energy and emissions under matched boundaries. Chemical and mechanical recycling can change polymer structure, additives and quality. Explicitly compare accepted products of equal function; recovered kilograms are an intermediate accounting metric, not proof of circular benefit.

Sketch a control volume, measurement chain, or molecular population appropriate to Biopolymers, compostability and material selection. Annotate the sketch with the actual controlled settings and the measured outputs, and mark each coefficient as an assumed teaching value, a fitted value or a possible independent measurement. The resulting sketch should make clear what extra measurement would transform a numerical demonstration into an experimentally testable model. Avoid assigning industrial meaning to a symbol solely because its unit appears plausible.

3. Worked numerical investigation

A dry test specimen goes from 10.0 g to 7.5 g: nominal mass loss is 25%, but this does not alone demonstrate mineralization or fulfillment of any compostability standard.

Independent calculation procedure: Extract the supplied initial conditions into a table and convert all reported units before substituting numbers. Record at least one intermediate result; then recompute the output in a calculator or spreadsheet without copying the answer. Vary the first independent quantity by 10% while all others remain fixed. Compute both the absolute response and percentage response; decide whether a local linear approximation is adequate. Reverse the question by asking whether the underlying measurement has enough resolution to distinguish the two numerical responses.

Constraint check: Identify any physically impossible input: a negative molecule count, mass recovery above the feed, mass fraction exceeding unity, a temperature on an incorrect absolute scale, a stress beyond a known material domain, or a nonfinite denominator. Not all of these appear in this model; choose those that apply. Discuss whether the original example is conservative or optimistic for real material variation. Values here are synthetic to explain algebra and must not be presented as published resin property tables.

4. Model verification is not material validation

Fragmentation, additive extraction, moisture changes and specimen handling can give mass loss without true polymer biodegradation.

Rearrange the equation to infer one input from the target output. If the rearrangement is ill-conditioned, explain how small measurement error may produce a large inferred-parameter error. Identify a pair of conceptually different mechanisms capable of moving the observed output in the same direction. A sound diagnosis cannot select one mechanism solely from that output; it needs an orthogonal measurement. This is the central protection against overinterpreting a polymer melt curve, a DSC peak, a strength measurement or a recycling-yield statistic.

Compare Biopolymers, compostability and material selection with Circularity metrics and life-cycle boundaries as a neighboring subject in this unit. Do the two relationships describe the same physical quantity or two independent descriptors? Would their characteristic times, spatial scales, material states or experimental procedures be comparable? Propose one experiment that measures both and write what a disagreement would imply. A second model is useful only if it provides independent information rather than repeating the same assumption in a new notation.

5. Evidence and experimental design

Distinguish residue collection from carbon mineralization, record exposure conditions and use the complete applicable standard rather than selective claims.

Measurement planning: Define specimens, sampling locations, conditioning, calibration method, measurement resolution and number of independent replicates. Specify one reference material or control and the expected direction of a genuine effect. Organize a data ledger with columns for parameter, unit, instrument, method, timestamp, operator, uncertainty and calibration status. Do not fit and validate a model on the same observations without acknowledging that the validation is not independent. Where the property depends on processing history, preserve the exact prior temperature, deformation and storage conditions.

A robust plan explicitly separates structural uncertainty, parameter uncertainty and observation uncertainty. Structural uncertainty asks whether important physics are absent; parameter uncertainty asks whether fitted numbers are poorly identified; observation uncertainty asks whether instrumentation or sampling is noisy or biased. Repeated sampling can reduce some random errors but cannot repair an incorrect equation or uncalibrated specimen history. For Biopolymers, compostability and material selection, name one of each, and predict which category will dominate extrapolation into an unmeasured processing regime.

6. Interactive investigation and a falsifiable prediction

Open [Laboratory 24](../labs/lab-24.html). Read its exact equation and domain, then predict the trend with the first slider before viewing the curve. Change only that slider at three positions, record numeric outputs and use CSV export for the complete sweep. Change a second parameter next; compare whether the response retains its direction and curvature. Treat visible invalid-condition warnings as part of the lesson, not software errors to ignore. The simulation operates entirely in the browser with no external requests.

Investigation: Design an evidence register for a packaging claim that avoids conflating feedstock, degradation and certification.

Create a two-column evidence note: one column lists predictions directly supported by the stated mathematical relationship; the second lists interpretations that require independent polymer measurements. State what observation would most clearly falsify the assumed relation. If no practical experiment is yet proposed, the numerical model is a consistency exercise rather than a validated predictor. Do not infer equipment tolerances, worker-safety margins or compliance from fictional sliders.

7. Design decision with limitations

A process engineer evaluating Biopolymers, compostability and material selection faces at least two objectives: a desired quantity or geometry and the assurance that material quality remains fit for intended service. Build a trade-off table containing accepted-product quality, throughput or test cost, risk of extrapolation and measurement effort. For each alternative, name the assumption on which the apparent advantage depends. Identify a stop condition, such as failure to reproduce the response with a new batch or detection of a qualitatively different material phase or deformation regime. This makes the decision conditional and auditable.

Read the technical reference titles on the [references page](../REFERENCES.html) and check whether they discuss terminology, measurement method or product specifications. A terminology reference cannot supply experimental constants; a standard's title and abstract do not grant permission to claim full procedural compliance. The book is an editorial draft and a verified industrial decision requires current complete standards, manufacturer information and professional review.

8. Independent lesson problems

1. Derive or reconstruct Illustrative mass-loss percentage L=100(m0−mt)/m0, defined on a consistent dry basis from the most appropriate physical or statistical principle. Define every symbol and every unit, then identify a limiting case.
2. Reproduce this lesson's worked example and change one input by ±10%; report output changes with justified significant figures.
3. Identify two competing physical explanations for a deviation from the model, and one measurement that distinguishes them.
4. Design a sampling plan that tests applicability after a change in material grade, temperature or processing history. Show how you would preserve calibration traceability.
5. Write a six-sentence engineering review: objective, model result, credible restriction, alternative, required experiment and conditional conclusion.

Next: return to the [unit pathway](../units/unit-08.html) or complete the [unit assessment](../assessments/unit-08.html).

Unit 08 · Lesson 04 — Circularity metrics and life-cycle boundaries

[Unit 08 course map](../units/unit-08.html) · [Interactive simulation 24](../labs/lab-24.html) · [Assessment](../assessments/unit-08.html) · [Applied case](../projects/unit-08.html)

1. The polymer-processing question

Material recovery, recycled content, collection rate and avoided emissions represent different quantities and denominators. A product can contain high recycled content yet be difficult to recover after use; losses occur during sorting, washing, melt filtration and conversion. System boundaries determine environmental results: displacement assumptions, transport, electricity mix and quality equivalence all matter. A mass-based circularity metric does not imply lower carbon intensity without inventory data. Engineering comparisons should use equal functional service, not simply equal kilograms of resin.

This lesson belongs to Sustainable polymer systems, whose goal is to evaluate recycling, degradation, bio-based feedstocks and design-for-circularity with material-property and system-boundary checks. The challenge is to connect chemical structure or thermomechanical history to an output that can be measured, rather than reporting attractive numerical results without a clear object of study. Name the material family and the actual specimen state before using a correlation. A number obtained on dry pellets can be inappropriate for a moist, filled, highly oriented or degraded finished article.

Sustainable design requires separate ledgers for collected material, process losses, accepted products, useful function and life-cycle inventory. Avoid double-counting recovered mass or displacement credits and state when unverified assumptions dominate the result.

2. Mathematical description and physical quantities

Starting relationship: Overall recovered fraction R=Rcollection·Rsorting·Rreprocessing·Rconversion.

A formula is meaningful only with its variable definitions. Identify the dependent variable, its SI unit, the independent variables and the dimensional basis of every fitted coefficient. For a ratio, inspect positive denominators and the zero-input limit; for a rate law, distinguish a rate per mass from a rate per volume; for a thermal transition, use kelvin if a reciprocal temperature is used. Reconstruct the equation's origin from counting, conservation, constitutive behavior or an explicitly stated empirical fit. Those origins cannot be interchanged. In particular, a pure mass-balance identity may survive changes in polymer chemistry while a viscosity correlation usually does not.

Keep material circularity separate from environmental benefit — A recycling yield can be verified with a mass balance, whereas climate benefit requires measured energy and emissions under matched boundaries. Chemical and mechanical recycling can change polymer structure, additives and quality. Explicitly compare accepted products of equal function; recovered kilograms are an intermediate accounting metric, not proof of circular benefit.

Sketch a control volume, measurement chain, or molecular population appropriate to Circularity metrics and life-cycle boundaries. Annotate the sketch with the actual controlled settings and the measured outputs, and mark each coefficient as an assumed teaching value, a fitted value or a possible independent measurement. The resulting sketch should make clear what extra measurement would transform a numerical demonstration into an experimentally testable model. Avoid assigning industrial meaning to a symbol solely because its unit appears plausible.

3. Worked numerical investigation

At 80% collection, 75% sorting, 90% reprocessing and 95% conversion yields, R=0.513, meaning about 51.3% of starting mass reaches accepted product in the declared simplified chain.

Independent calculation procedure: Extract the supplied initial conditions into a table and convert all reported units before substituting numbers. Record at least one intermediate result; then recompute the output in a calculator or spreadsheet without copying the answer. Vary the first independent quantity by 10% while all others remain fixed. Compute both the absolute response and percentage response; decide whether a local linear approximation is adequate. Reverse the question by asking whether the underlying measurement has enough resolution to distinguish the two numerical responses.

Constraint check: Identify any physically impossible input: a negative molecule count, mass recovery above the feed, mass fraction exceeding unity, a temperature on an incorrect absolute scale, a stress beyond a known material domain, or a nonfinite denominator. Not all of these appear in this model; choose those that apply. Discuss whether the original example is conservative or optimistic for real material variation. Values here are synthetic to explain algebra and must not be presented as published resin property tables.

4. Model verification is not material validation

Constant recovery factors miss quality-dependent routing, additives and multiple loops; embodied-emissions benefits require verified life-cycle data.

Rearrange the equation to infer one input from the target output. If the rearrangement is ill-conditioned, explain how small measurement error may produce a large inferred-parameter error. Identify a pair of conceptually different mechanisms capable of moving the observed output in the same direction. A sound diagnosis cannot select one mechanism solely from that output; it needs an orthogonal measurement. This is the central protection against overinterpreting a polymer melt curve, a DSC peak, a strength measurement or a recycling-yield statistic.

Compare Circularity metrics and life-cycle boundaries with Mechanical reprocessing and blend properties as a neighboring subject in this unit. Do the two relationships describe the same physical quantity or two independent descriptors? Would their characteristic times, spatial scales, material states or experimental procedures be comparable? Propose one experiment that measures both and write what a disagreement would imply. A second model is useful only if it provides independent information rather than repeating the same assumption in a new notation.

5. Evidence and experimental design

Require measured material flow by unit operation, a consistent functional unit and sensitivity to quality rejection and displacement rates.

Measurement planning: Define specimens, sampling locations, conditioning, calibration method, measurement resolution and number of independent replicates. Specify one reference material or control and the expected direction of a genuine effect. Organize a data ledger with columns for parameter, unit, instrument, method, timestamp, operator, uncertainty and calibration status. Do not fit and validate a model on the same observations without acknowledging that the validation is not independent. Where the property depends on processing history, preserve the exact prior temperature, deformation and storage conditions.

A robust plan explicitly separates structural uncertainty, parameter uncertainty and observation uncertainty. Structural uncertainty asks whether important physics are absent; parameter uncertainty asks whether fitted numbers are poorly identified; observation uncertainty asks whether instrumentation or sampling is noisy or biased. Repeated sampling can reduce some random errors but cannot repair an incorrect equation or uncalibrated specimen history. For Circularity metrics and life-cycle boundaries, name one of each, and predict which category will dominate extrapolation into an unmeasured processing regime.

6. Interactive investigation and a falsifiable prediction

Open [Laboratory 24](../labs/lab-24.html). Read its exact equation and domain, then predict the trend with the first slider before viewing the curve. Change only that slider at three positions, record numeric outputs and use CSV export for the complete sweep. Change a second parameter next; compare whether the response retains its direction and curvature. Treat visible invalid-condition warnings as part of the lesson, not software errors to ignore. The simulation operates entirely in the browser with no external requests.

Investigation: Identify which stage gives the largest marginal improvement in final accepted mass.

Create a two-column evidence note: one column lists predictions directly supported by the stated mathematical relationship; the second lists interpretations that require independent polymer measurements. State what observation would most clearly falsify the assumed relation. If no practical experiment is yet proposed, the numerical model is a consistency exercise rather than a validated predictor. Do not infer equipment tolerances, worker-safety margins or compliance from fictional sliders.

7. Design decision with limitations

A process engineer evaluating Circularity metrics and life-cycle boundaries faces at least two objectives: a desired quantity or geometry and the assurance that material quality remains fit for intended service. Build a trade-off table containing accepted-product quality, throughput or test cost, risk of extrapolation and measurement effort. For each alternative, name the assumption on which the apparent advantage depends. Identify a stop condition, such as failure to reproduce the response with a new batch or detection of a qualitatively different material phase or deformation regime. This makes the decision conditional and auditable.

Read the technical reference titles on the [references page](../REFERENCES.html) and check whether they discuss terminology, measurement method or product specifications. A terminology reference cannot supply experimental constants; a standard's title and abstract do not grant permission to claim full procedural compliance. The book is an editorial draft and a verified industrial decision requires current complete standards, manufacturer information and professional review.

8. Independent lesson problems

1. Derive or reconstruct Overall recovered fraction R=Rcollection·Rsorting·Rreprocessing·Rconversion from the most appropriate physical or statistical principle. Define every symbol and every unit, then identify a limiting case.
2. Reproduce this lesson's worked example and change one input by ±10%; report output changes with justified significant figures.
3. Identify two competing physical explanations for a deviation from the model, and one measurement that distinguishes them.
4. Design a sampling plan that tests applicability after a change in material grade, temperature or processing history. Show how you would preserve calibration traceability.
5. Write a six-sentence engineering review: objective, model result, credible restriction, alternative, required experiment and conditional conclusion.

Next: return to the [unit pathway](../units/unit-08.html) or complete the [unit assessment](../assessments/unit-08.html).

Unit 08 applied engineering design study — Sustainable polymer systems

Case status: fictional scenario. All measurements, market costs, process coefficients and material specifications are synthetic unless properly sourced by the learner. No calculated output authorizes a machine setting, product certification or environmental claim.

Case brief

A hypothetical packaging company seeks 60% recycled content while maintaining accepted-part output. Collection, sorting, pelletizing and conversion losses are incompletely measured, and marketing wants to claim a life-cycle emissions benefit.

The essential discipline is to prevent a model from silently becoming evidence. A design team should draw a boundary, define a function, preserve material and energy units, and identify a measurement capable of changing its conclusion. In polymer systems the same nominal resin can behave differently after drying, heating, oxidation, stretching and reprocessing. Product acceptance must be assessed on the relevant final article and sampling plan.

Technical analyses

Create stagewise material balances, test blend-property linearity as a screening hypothesis, and compare equal-function product boundaries. Explicitly separate recycled content, post-use recovery and verified avoided emissions.

Describe where the physical state changes, where material is lost, where energy crosses the boundary and where quality is gained or degraded. If two alternatives differ by material type or final function, normalize them before comparing. State which inputs are experimentally measured, independently sourced, nominal manufacturer data or assumed illustrative coefficients. Use an uncertainty interval for any decision-critical property and do not invent a measured value.

Three interactive investigations

1. [Simulation 22](../labs/lab-22.html): sweep the first parameter and reconstruct the default numerical output from its equation; record an analytical limit.
2. [Simulation 23](../labs/lab-23.html): assess which second physical mechanism the model adds; state whether both calculations use the same material state.
3. [Simulation 24](../labs/lab-24.html): deliberately test parameter sensitivity or an invalid-domain warning and explain why treating a fictitious curve as a specification would be unsafe.

Calculate at least one output independently in a spreadsheet or short code segment, save the CSV curves, and annotate differences between numerical rounding and structural model assumptions. The three simulations may complement one another without forming a complete virtual factory.

Evidence protocol

Design a measurement schedule with sensor or analytical method, instrument range, calibration date, specimen preparation, sampling interval and acceptable data completeness. Define a control sample and a replicate strategy. For process comparisons, separate startup and steady production and keep separate numerator/denominator definitions for material yield, rejected-part rate and energy use. A small systematic bias can outweigh an impressive-looking simulated benefit.

Propose at least two counterhypotheses that could explain the same visible process defect or laboratory trend. Specify the smallest experiment that can discriminate them. State which claim you would withdraw if an independent material or product test contradicts the predicted model curve. When the claim concerns safety, food contact, biodegradation or a regulated end use, require the relevant complete, current standard and qualified expert review.

Deliverables

Provide a mass-flow Sankey table, property-acceptance experimental plan, uncertainty analysis and a prohibited-claims list for conclusions unsupported by LCA inventory.

In addition, submit a one-page management summary: intended function and specification, explicit assumptions, normalized numerical comparison, evidence gaps, the chosen next experiment and the decision that cannot yet be made. Append all formulas, spreadsheet inputs and raw exports so another person can reconstruct the case. A recommendation is not complete until it specifies what new evidence would change it.

Assessment rubric

• 25% correctness of mass, dimensional and mathematical reasoning.
• 20% physically appropriate model scope and identification of omissions.
• 20% reproducible model sweeps and independent numerical checks.
• 20% design of experiments and measurement-system integrity.
• 15% appropriate conditional decision and transparent communication.

Unit 08 assessment — Sustainable polymer systems

Format: 16 extended, multi-part problems, four per lesson. The responses are assessed for correct definitions (20%), conservation or mathematical reasoning (25%), reconstructable numerical work (25%), limitation-aware interpretation (20%) and evidence-based decision (10%). Every example is explicitly synthetic, and a student can correctly reject an unsupported model claim.

Lesson 1 — Mechanical reprocessing and blend properties

Question 01: Numerical reconstruction and sensitivity

Reproduce the worked computation for Mechanical reprocessing and blend properties: For synthetic stiffness values Erec=800 MPa, Evirgin=1200 MPa and 40% recycled fraction, Eblend≈1040 MPa. The actual two-phase composite response may not obey the straight line. Produce a three-column ledger (given, calculated and unit), check the result using a different numerical route, then perturb one input by +15% and −15%. Report whether the effects are symmetric in output space. Explain why arithmetic agreement cannot establish the full measured distribution, property or grade suitability. Include one sign or limit check based on Linear property proxy Pblend=wrec Prec+(1−wrec)Pvirgin.

Question 02: Physical derivation and dimensional audit

Develop the equation used in Mechanical reprocessing and blend properties from an appropriate physical basis: Linear property proxy Pblend=wrec Prec+(1−wrec)Pvirgin. Enumerate the assumptions needed to omit alternative effects, show the dimensional units of every input, and identify at least two plausible conditions under which the expression becomes unsuitable. Compare the idea with a neighboring topic in the same unit; which variable would be shared, and which material or instrument information cannot be transferred? Explain which conservation law or identity remains valid even if the fitted relation fails.

Question 03: Simulation challenge and falsification

Run [Laboratory 22](../labs/lab-22.html) and export its entire 41-point parameter sweep. Select one low, one intermediate and one high value; calculate a finite-difference slope, verify one output independently and test a boundary or failure condition. Use this known limitation: Mixing laws can miss phase separation, compatibilization, morphology and impact-strength thresholds; do not make product safety claims from them. Construct a plausible synthetic dataset that violates the model trend and state whether you would investigate wrong coefficients, missing mechanism or instrument bias first. Design a discriminating test to distinguish the possibilities.

Question 04: Engineering decision and measurement plan

An engineering review proposes changing a material or processing condition because of the following evidence recommendation: Characterize the feed’s contaminants, water content and molecular weight; measure blended property distributions, not single coupons. Write two alternatives, a quality target, a measurement cost estimate expressed as assumptions, and a stop condition. Tie the decision to the explicit lesson investigation: Compare a screening stiffness target with separate impact and fatigue requirements. Describe at least three separate observations needed before adoption. State a condition under which your preferred option would reverse and a limitation that makes extrapolation unacceptable without new data.

Lesson 2 — Degradation, weathering and service lifetime

Question 05: Numerical reconstruction and sensitivity

Reproduce the worked computation for Degradation, weathering and service lifetime: If hypothetical k=0.002 day⁻¹, property retention after 180 days is ≈0.698. Real k may depend strongly on climate, additives, strain and specimen thickness. Produce a three-column ledger (given, calculated and unit), check the result using a different numerical route, then perturb one input by +15% and −15%. Report whether the effects are symmetric in output space. Explain why arithmetic agreement cannot establish the full measured distribution, property or grade suitability. Include one sign or limit check based on Ideal retained property fraction R(t)=exp(−kt).

Question 06: Physical derivation and dimensional audit

Develop the equation used in Degradation, weathering and service lifetime from an appropriate physical basis: Ideal retained property fraction R(t)=exp(−kt). Enumerate the assumptions needed to omit alternative effects, show the dimensional units of every input, and identify at least two plausible conditions under which the expression becomes unsuitable. Compare the idea with a neighboring topic in the same unit; which variable would be shared, and which material or instrument information cannot be transferred? Explain which conservation law or identity remains valid even if the fitted relation fails.

Question 07: Simulation challenge and falsification

Run [Laboratory 23](../labs/lab-23.html) and export its entire 41-point parameter sweep. Select one low, one intermediate and one high value; calculate a finite-difference slope, verify one output independently and test a boundary or failure condition. Use this known limitation: One rate coefficient cannot represent oxidation-induction periods, crack growth or diffusion-limited degradation. Construct a plausible synthetic dataset that violates the model trend and state whether you would investigate wrong coefficients, missing mechanism or instrument bias first. Design a discriminating test to distinguish the possibilities.

Question 08: Engineering decision and measurement plan

An engineering review proposes changing a material or processing condition because of the following evidence recommendation: Use matched control exposures, mechanistic chemical markers and relevant load-bearing measurements over service conditions. Write two alternatives, a quality target, a measurement cost estimate expressed as assumptions, and a stop condition. Tie the decision to the explicit lesson investigation: Distinguish an accelerated lifetime claim from a demonstrated failure-mode equivalence test. Describe at least three separate observations needed before adoption. State a condition under which your preferred option would reverse and a limitation that makes extrapolation unacceptable without new data.

Lesson 3 — Biopolymers, compostability and material selection

Question 09: Numerical reconstruction and sensitivity

Reproduce the worked computation for Biopolymers, compostability and material selection: A dry test specimen goes from 10.0 g to 7.5 g: nominal mass loss is 25%, but this does not alone demonstrate mineralization or fulfillment of any compostability standard. Produce a three-column ledger (given, calculated and unit), check the result using a different numerical route, then perturb one input by +15% and −15%. Report whether the effects are symmetric in output space. Explain why arithmetic agreement cannot establish the full measured distribution, property or grade suitability. Include one sign or limit check based on Illustrative mass-loss percentage L=100(m0−mt)/m0, defined on a consistent dry basis.

Question 10: Physical derivation and dimensional audit

Develop the equation used in Biopolymers, compostability and material selection from an appropriate physical basis: Illustrative mass-loss percentage L=100(m0−mt)/m0, defined on a consistent dry basis. Enumerate the assumptions needed to omit alternative effects, show the dimensional units of every input, and identify at least two plausible conditions under which the expression becomes unsuitable. Compare the idea with a neighboring topic in the same unit; which variable would be shared, and which material or instrument information cannot be transferred? Explain which conservation law or identity remains valid even if the fitted relation fails.

Question 11: Simulation challenge and falsification

Run [Laboratory 24](../labs/lab-24.html) and export its entire 41-point parameter sweep. Select one low, one intermediate and one high value; calculate a finite-difference slope, verify one output independently and test a boundary or failure condition. Use this known limitation: Fragmentation, additive extraction, moisture changes and specimen handling can give mass loss without true polymer biodegradation. Construct a plausible synthetic dataset that violates the model trend and state whether you would investigate wrong coefficients, missing mechanism or instrument bias first. Design a discriminating test to distinguish the possibilities.

Question 12: Engineering decision and measurement plan

An engineering review proposes changing a material or processing condition because of the following evidence recommendation: Distinguish residue collection from carbon mineralization, record exposure conditions and use the complete applicable standard rather than selective claims. Write two alternatives, a quality target, a measurement cost estimate expressed as assumptions, and a stop condition. Tie the decision to the explicit lesson investigation: Design an evidence register for a packaging claim that avoids conflating feedstock, degradation and certification. Describe at least three separate observations needed before adoption. State a condition under which your preferred option would reverse and a limitation that makes extrapolation unacceptable without new data.

Lesson 4 — Circularity metrics and life-cycle boundaries

Question 13: Numerical reconstruction and sensitivity

Reproduce the worked computation for Circularity metrics and life-cycle boundaries: At 80% collection, 75% sorting, 90% reprocessing and 95% conversion yields, R=0.513, meaning about 51.3% of starting mass reaches accepted product in the declared simplified chain. Produce a three-column ledger (given, calculated and unit), check the result using a different numerical route, then perturb one input by +15% and −15%. Report whether the effects are symmetric in output space. Explain why arithmetic agreement cannot establish the full measured distribution, property or grade suitability. Include one sign or limit check based on Overall recovered fraction R=Rcollection·Rsorting·Rreprocessing·Rconversion.

Question 14: Physical derivation and dimensional audit

Develop the equation used in Circularity metrics and life-cycle boundaries from an appropriate physical basis: Overall recovered fraction R=Rcollection·Rsorting·Rreprocessing·Rconversion. Enumerate the assumptions needed to omit alternative effects, show the dimensional units of every input, and identify at least two plausible conditions under which the expression becomes unsuitable. Compare the idea with a neighboring topic in the same unit; which variable would be shared, and which material or instrument information cannot be transferred? Explain which conservation law or identity remains valid even if the fitted relation fails.

Question 15: Simulation challenge and falsification

Run [Laboratory 24](../labs/lab-24.html) and export its entire 41-point parameter sweep. Select one low, one intermediate and one high value; calculate a finite-difference slope, verify one output independently and test a boundary or failure condition. Use this known limitation: Constant recovery factors miss quality-dependent routing, additives and multiple loops; embodied-emissions benefits require verified life-cycle data. Construct a plausible synthetic dataset that violates the model trend and state whether you would investigate wrong coefficients, missing mechanism or instrument bias first. Design a discriminating test to distinguish the possibilities.

Question 16: Engineering decision and measurement plan

An engineering review proposes changing a material or processing condition because of the following evidence recommendation: Require measured material flow by unit operation, a consistent functional unit and sensitivity to quality rejection and displacement rates. Write two alternatives, a quality target, a measurement cost estimate expressed as assumptions, and a stop condition. Tie the decision to the explicit lesson investigation: Identify which stage gives the largest marginal improvement in final accepted mass. Describe at least three separate observations needed before adoption. State a condition under which your preferred option would reverse and a limitation that makes extrapolation unacceptable without new data.